¹ T.N. Habinek, The Politics of Latin Literature (Princeton, 1998), 15–33 provides a history of scholarship on the issue; cf. J.E.G. Zetzel, ‘Plato with pillows: Cicero on the uses of Greek culture’, in D. Braund and C. Gill (edd.), Myth, History, and Culture in Republican Rome (Exeter, 2003), 119–38, at 121–2. Mommsen's account in T. Mommsen (trans. W.P. Dickson), The History of Rome, vol. 5 (Cambridge, 1908) serves as an (in)famous example of Cicero's fortunes in this period: ‘By nature a journalist in the worst sense of that term—abounding, as he himself says, in words, poor beyond all conception in ideas—there was no department in which he could not with the help of a few books have rapidly got up by translation or compilation a readable essay.’ (505)

² See e.g. Habinek, T.N., ‘Ideology for an empire in the prefaces to Cicero's dialogues’, Ramus 23 (1994), 55–67 Google Scholar and Habinek (n. 1), 60–8; M. Jaeger, Archimedes and the Roman Imagination (Ann Arbor, 2008), 32–68; Y. Baraz, A Written Republic: Cicero's Philosophical Politics (Princeton, 2012). Horsfall, N., ‘Empty shelves on the Palatine’, G&R 40 (1993), 58–67 Google Scholar collects many of Cicero's pronouncements on Greek culture at 59–60. Zetzel (n. 1) remains the best sustained treatment. For the cultural context in which this attitude emerged, see E.S. Gruen, Culture and Identity in Republican Rome (Ithaca, 1992), especially 223–71.

³ This is in line with recent scholarship on the poem, which has emphasized how skilfully Cicero translates Aratus' Hellenistic leptotēs into a Latin idiom. See in particular Kubiak, D.P., ‘The Orion episode of Cicero's Aratea ’, CJ 77 (1981), 12–22 Google Scholar; id., ‘Cicero and the poetry of nature’, SIFC 8 (1990), 198–214 Google Scholar; Clausen, W., ‘Cicero and the new poetry’, HSPh 90 (1986), 159–70Google Scholar; Gee, E., ‘Cicero's astronomy’, CQ 51 (2001), 520–36Google Scholar; and D.M. Possanza, Translating the Heavens: Aratus, Germanicus, and the Poetics of Latin Translation (New York/Washington, DC, 2004), 39–40. Hurka, F., ‘Ein Akrostichon in Ciceros Aratea’, WJA 30 (2006), 87–91 Google Scholar has even identified an acrostic likely inspired by the famous ΛΕΠΤΗ: lines 317–20 of the Aratea spell out ZONA, a title for the Zodiac used by Vitruvius and Manilius. E. Gee, in ‘Cicero's Poetry’, in C. Steel (ed.), The Cambridge Companion to Cicero (Cambridge, 2013[a]), 88–106, and in Aratus and the Astronomical Tradition (Oxford, 2013[b]), 60–6, makes a compelling case for the importance of the Aratea in the larger intellectual milieu of the mid first century.

⁴ On this, note the observation of J. Soubiran, Cicéron. Aratea, fragments poetiques (Paris, 1972), 88: ‘Un détail, du reste, montre que Cicéron pretend donner à ses compatriotes, plutôt qu'une image exacte du poème aratéen, un nouvel Aratos en latin : ajoutant évidemment à son modèle, il commente plusieurs fois des différences de terminologie astronomique entre Grecs et Romains.’

⁵ As J. Soubiran, ‘L'astronomie á Rome’, in G. Aujac and J. Soubiran (edd.), L'Astronomie dans l'antiquité classique (Paris, 1979) notes, there was not much of a scientific astronomical tradition at Rome prior to the adoption of Hellenistic models. For comparisons of Cicero's Aratea with the original, see Possanza (n. 3), passim, esp. 10–12 and 38–44, and P. Toohey, Epic Lessons: An Introduction to Ancient Didactic Poetry (London, 1996), 78–87; Lewis, A.-M., ‘Rearrangement of motif in Latin translation: the emergence of a Roman Phaenomena ’, Studies in Latin Literature and Roman History 4 (1994), 210–33Google Scholar compares Aratus with all three of his Latin translators. On Cicero's translation practices more generally, see J.G.F. Powell, ‘Cicero's translations from Greek’, in J.G.F. Powell (ed.), Cicero the Philosopher: Twelve Papers (Oxford, 1995), 273–300; Büchner, K., ‘M. Tullius Cicero: Fragmente der Dichtungen’, RE 7A.1, 1236–67Google Scholar, at 1240–2; Morford, M.P.O., ‘Ancient and modern in Cicero's poetry’, CPh 62 (1967), 112–16Google Scholar; and Jones, D.M., ‘Cicero as a translator’, BICS 6 (1959), 22–34 Google Scholar. At De or. 1.154-5 Cicero recommends both the reading of poetry and the translation from Greek into Latin to hone rhetorical fluency, and the large number of translations he produced in his lifetime was likely connected to this.

⁶ On the origins of Roman etymology, see E. Rawson, Intellectual Life in the Late Roman Republic (Baltimore, 1985), 117–31, and J.J. O'Hara, True Names: Vergil and the Alexandrian Tradition of Etymological Wordplay (Ann Arbor, 1996), 42–50.

⁷ Aelius Stilo was quite interested in Latin etymology, and likely devoted an entire book to the subject. See GRF 59–70; Gruen (n. 2), 234–5; Rawson (n. 6), 120; and P. Dietrich, ‘De Ciceronis ratione etymologica’ (Diss., Friedrich-Schiller-Universität, 1911), 32. For Cicero's tutelage under Aelius, see Brut. 56 and 205–7.

⁸ Text of the Aratea on its own is from Soubiran (n. 4). When discussing lines that appear in De Natura Deorum, I will follow the text printed by A.S. Pease, M. Tulli Ciceronis De Natura Deorum: Libri Secundus et Tertius (Cambridge, 1958), while providing cross-references to Soubiran. There are minor variants between the two, but this is not surprising, given that Cicero's interlocutor Balbus is in theory (and Cicero may be in fact) quoting from memory; a far likelier hypothesis for the differences than the suggestion that Cicero published an updated edition of the poem. On this see Soubiran (n. 4), 12–15.

⁹ Variants of καλέω are used frequently by Aratus to indicate etymologically significant star names; see e.g. Phaen. 27, 36, 66, 164, 261, 315 and 331. Translations of Aratus are adapted from D.A. Kidd, Aratus Phaenomena (Cambridge, 1997).

¹⁰ O'Hara, J.J., ‘Naming the stars at Georgics 1.137-38 and Fasti 5.163-82’, AJPh 113 (1992), 47–61 Google Scholar, at 48. Cf. O'Hara (n. 6), 10–11.

¹¹ signum in the sense of a statue is found as early as Plaut. Rud. 560; it is also used to describe the constellation Arcturus in the same play (Rud. 4). Its usage in astronomical terminology is discussed by A. Le Boeuffle, Les noms latins d'astres et de constellations (Paris, 1977), 23–31, who notes rightly that it implies an intelligent creator, making it particularly appropriate for Cicero's translation of Aratus.

¹² I thank one of my readers for bringing this point to my attention.

¹³ This use of the future tense can be read as a virtual imperative, carrying overtones of Cicero's authorial control over Latin technical vocabulary; cf. R. Risselada, Imperatives and Other Directive Expressions in Latin (Amsterdam, 1993), 169–78: ‘future indicative directives are predominantly used in situations in which the speaker has reason to be confident that the addressee will comply.’ Perhaps Cicero takes such cues from his model; Cusset, C., ‘Poétique et onomastique dans les Phénomènes d'Aratos,’ Pallas 59 (2002), 187–96Google Scholar, at 190 argues for a similar resonance in Aratus' use of the future (κατακείσεται) at Phaen. 180.

¹⁴ One of the most attentive readers of Cicero's Aratea was Lucretius, as Gee (n. 3 [2013(b)]) has shown. But even if we just use constellation terminology as a marker, we will see that Virgil and Ovid also look to the Aratea, a pattern consistent with what Gee argues (n. 3 [2013(a)]), and what was already noted, albeit in a more qualified way, by Soubiran (n. 4), 69–81.

¹⁵ For broader accounts of etymology in Aratus, see Kidd (n. 9), passim; O'Hara (n. 10), 48–9; O'Hara (n. 6), 35–6; Cusset (n. 13); and Pendergraft, M.L.B., ‘Euphony and etymology: Aratus’ Phaenomena ’, SyllClass 6 (1995), 43–67 Google Scholar. The large number of loci similes between the Aratea and the Aratean scholia indicates that Cicero had some form of Aratean exegesis to hand when he translated the poem; see C. Atzert, ‘De Cicerone interprete Graecorum’ (Diss., Georg-August-Universität, 1907), 3–11, and M. Goetz, ‘De scholiastis Graecis poetarum Romanorum auctoribus quaestiones selectae’ (Diss., Friedrich-Schiller-Universität, 1918), 12–16.

¹⁶ The conjunction of these constellations (sometimes only the first three, sometimes with the Bear/Wagon included) was common in ancient literature: Il. 18.486 is repeated verbatim at Od. 5.273 and Op. 615, while all four recur in the longer passage at Op. 609–21 and in Euripides' Ion (1152–7). The tradition probably dates back even further; G. Ferrari, Alcman and the Cosmos of Sparta (Chicago, 2008), 36 n. 43 adduces the parallel of the Babylonian MUL.APIN., which contains the sequence Stars (Pleiades), Bull of Heaven (Hyades) and True Shepherd of Anu (Orion).

¹⁷ ἐπίκλησις appears in conjunction with etymological glosses elsewhere in the poem (Il. 7.138-41, 22.506-7).

¹⁸ On this passage, see also Pendergraft (n. 15), 56. Aratus emphasizes the etymological connection between the two with his emphatic τὸ δή. For any reader still in doubt of his meaning here, Aratus' scholiast clarifies with the simple statement διὰ τοῦτο [sc. ἅμα τροχόωσιν] γὰρ ἅμαξαι ὠνομάσθησαν (Schol. Arat. on 26–7). The text of Aratus' scholia is that of J. Martin, Scholia in Aratum Vetera (Stuttgart, 1974).

¹⁹ E. Gee, Ovid, Aratus, and Augustus: Astronomy in Ovid's Fasti (Cambridge, 2000), 177 also sees the Bears as programmatic for the rest of the star catalogue. The only constellation from Homer's Shield not etymologized in any way is Orion. But while the names of the Pleiades, Hyades and Wagon may seem opaque and thus in need of explanation, this is not the case for Orion: the name clearly refers to a mythological figure, catasterized upon his death, and indeed in Aratus the focus is on the mythological Orion, particularly in the extended account of his death at Phaen. 636–46. None the less, Aratus does make a point of noting Orion's usefulness as a weather sign (Phaen. 728–32). As for the constellation's name, the best-known etymology in antiquity related it to οὖρον (urine); see e.g. Fast. 5.493-544. Kidd (n. 9), 303–4 considers how the star group came to be associated with the Boeotian hunter.

²⁰ See e.g. Phaen. 10–14: αὐτὸς γὰρ τά γε σήματ’ ἐν οὐρανῷ ἐστήριξεν | ἄστρα διακρίνας, ἐσκέψατο δ’ εἰς ἐνιαυτὸν | ἀστέρας οἵ κε μάλιστα τετυγμένα σημαίνοιεν | ἀνδράσιν ὡράων, ὄφρ’ ἔμπεδα πάντα φύωνται. | τῷ μιν ἀεὶ πρῶτόν τε καὶ ὕστατον ἱλάσκονται. Aratus' focus on etymology may also point towards the poem's Stoic leanings (and was certainly thought to do so by its ancient readers). On the relationship between etymology and Stoicism in Aratus, see E. Gee (n. 19), 70–90; Kidd (n. 9), 10–12; and R. Hunter, ‘Written in the stars: poetry and philosophy in the Phaenomena of Aratus', Arachnion 1.2 (1995) [www.cisi.unito.it/arachne/num2/hunter.html]. Pendergraft (n. 15) and Cusset (n. 13) show that, in addition to serving a Stoic purpose, Aratus' interest in etymology was related to his program of leptotēs; Volk, K., ‘Letters in the sky: reading the signs in Aratus’ Phaenomena ’, AJPh 133 (2012), 209–40Google Scholar, at 221–3 points to the way in which Aratus portrays the ‘real names’ of the constellations as inscribed markers in the sky to be ‘read’ by his audience.

²¹ The Pleiades' morning rising in May marked the beginning of sailing season and the harvest, and their morning setting in November indicated the end of the sailing season and the beginning of the ploughing season. The Hyades had their morning setting slightly later in November, which marked the beginning of the rainy season.

²² Accounts of earlier treatments of these constellations can be found in Fisher, R.S. and Lewis, A.-M., ‘Agamemnon, Troy, and the Pleiades’, RBPh 62 (1984), 5–15 CrossRefGoogle Scholar and Ferrari (n. 16). Both star clusters are now considered part of Taurus, and the distance between them is only about ten degrees; in antiquity, the connection between them was further strengthened by making both groups the daughters of Atlas. See Kidd (n. 9), passim, esp. 279–80.

²³ See also J. Martin, Aratos: Phénomènes (Paris, 1998), LVI-LVII, 173.

²⁴ Aratus plays here, as elsewhere, with the idea of Zeus as both sky-god and the sky itself; a third level of signification, where Zeus represents the providential Stoic godhead, is also likely.

²⁵ This is likely an allusion to Euripides, for whom this adjective was a favourite in describing the constellation: see IA 7–8, Or. 1005 and Rhes. 529–30. For more on Euripides' astronomy, see Ferrari (n. 16).

²⁶ See Op. 614–20, G. 1.137-8, Schol. Arat. 254–5, Kidd (n. 9), 275–6, and O'Hara (n. 10), 52–3.

²⁷ In Babylonian astronomy, Taurus was particularly important, since it was the sign that marked the sun's position in the vernal equinox, the start of the year, a tradition reflected in Virgil (G. 1.217-18). See also Kidd (n. 9), 244.

²⁸ Schol. Arat. (S scholia) on 172: κληθῆναι δὲ οὕτως δι’ ἣν προείπομεν αἰτίαν. ἢ ὅτι παραπλησίως τῷ Υ στοιχείῳ κεῖνται, ἢ ὅτι δύνουσαι ποιοῦσιν ὑληθῆ, ἢ κατ’ ἔλλειψιν τοῦ Θ Θυάδας. The rest of the scholia (MDΔVUA) contain a more compressed version: οἳ δὲ ὅτι δυόμεναι αἴτιαι ὑετοῦ γίνονται. See further Schol. Arat. 171–2; O'Hara (n. 10), 55; Kidd (n. 9), 246; and Pease (n. 8), 822–3, who notes that ὕειν was the preferred etymology among most ancient authors.

²⁹ Further accounts of Cicero's etymological wordplay can be found in Possanza (n. 3), 39–40 and 70 n. 40; O'Hara (n. 6), 47–8; and Gee (n. 3 [2013(b)]), passim.

³⁰ Temo was used in this sense at least as early as Ennius; see Varro, Ling. 7.73-5. For the other Latin names of the constellation (which included Plaustrum, Temo, Arcera and Axis), see Le Boeuffle (n. 11), 82–92. As R. Caldini Montanari, ‘Le Stelle dell'Orsa Maggiore (“Septem Triones”)’, in C. Santini, L. Zurli and L. Cardinali (edd.), Concentus ex dissonis: scritti in onore di Aldo Setaioli (Naples, 2006), 123–36 points out, the name Septem Triones must at first have been applied only to the seven stars of the Big Dipper or Ursa Major, i.e. Helice.

³¹ NA 2.21. The belief that large parts of a word held no significance was not uncommon in ancient etymology; see Baratin, M., ‘Remarques sur l'absence de signification dans les textes theoretiques Latins’, Lingua Latina 5 (1999), 117–26Google Scholar. E. Gunderson, Nox Philologiae: Aulus Gellius and the Fantasy of the Roman Library (Madison, 2009), 153–5 provides an analysis of Gellius' discussion. For more on the constellation, see W. Gundel, De stellarum appellatione et religione Romana (Giessen, 1907), 59–83.

³² NA 2.21.8 = GRF 42: sed ego quidem cum L. Aelio et M. Varrone sentio, qui ‘triones’ rustico uocabulo boues appellatos scribunt quasi quosdam ‘terriones’, hoc est arandae colendaeque terrae idoneos. Varro's account of the etymology can be found at Ling. 7.74-5; cf. Vitruvius 9.4.1 and Festus 339. It is unclear when the name was adopted, but it is already attested in Plautus (Amph. 273). Modern accounts generally assume an etymology from terere ‘to tread’, thus ‘seven threshing oxen’. See A. Scherer, Gestirnnamen bei den indogermanischen Völkern (Heidelberg, 1953), 136; Le Boeuffle (n. 11), 87–9; and Caldini Montanari (n. 30), 131.

³³ See further Gundel (n. 31), 59–60. Aelius Stilo's influence on the Aratea was likely considerable; Kubiak (n. 3 [1990]), 203–4 points to another Aelian etymology (lepus and leuipes, Varro, Rust. 2.12.6 = GRF 12) at Arat. 121.

³⁴ Both poets also use Triones on its own (Aen. 1.744, 3.516; Met. 2.171, 10.446), an even stronger argument for the phrase's semantic relevance. The significance of the constellation for Ovid more generally is discussed in Gee (n. 19), 175–87.

³⁵ It is quite possible that the source of Cicero's recognition was a scholium on the line; all of the modern scholia (MDΔVUAS, Schol. Arat. on 257) note a Euripidean parallel (Or. 1005–6). On this see Goetz (n. 15), 15.

³⁶ Interestingly, in this case Cicero's collocation did not have the same influence that his use of signifer and Septem … Triones did. Vergiliae seems to have had rustic non-poetic connotations: it is used extensively in Varro's Rust., Columella and Plin. HN 18, but in verse only by Cicero, Plautus (Amph. 275) and Propertius (1.8.10); Virgil, Ovid, Manilius and Germanicus use Pleiades instead. See also Gundel (n. 31), 92–101, and Kidd (n. 9), 275. Yet, as Gee (n. 3 [2013(b)]), 202 notes, this passage from the Aratea was highly significant for Lucretius, who alludes to it several times to question received wisdom on various topics.

³⁷ See e.g. Hyg. Poet. astr. 2.21, Servius, on G. 1.138, and Festus 510.15 (dictae quod earum ortu uer finem facit). For modern etymologies, see Le Boeuffle (n. 11), 122–4 and Scherer (n. 32), 222.

³⁸ While this is almost a direct translation of Aratus, it is impressive that Cicero can imbue his version with an etymological significance not present in the original.

³⁹ Cicero burnishes Balbus' Stoic credentials at Nat. D. 1.15-16 and 2.88, where his relationships with Posidonius of Apamea and Antiochus of Ascalon are emphasized. Of course, while Cicero goes out of his way to give Balbus his own set of characteristics and motivations for quoting the Aratea, in an obvious sense, there is no ‘Balbus’ independent from Cicero, who is using the character to perform an act of self-interpretation. Therefore, though I will be discussing Balbus as if he were a ‘real’ person, it is important to remember that Cicero lies behind him: if Cicero can make this easy to forget, it only goes to show how successful he was in distancing himself from this act of self-quotation.

⁴⁰ Cf. Nat. D. 1.40 and 3.62-3, Diog. Laert. 7.147, SVF 2.1061-100. Further discussion of Stoic theories of language can be found in P. Struck, The Birth of the Symbol (Princeton, 2004), 111–41; A.A. Long, ‘Stoic linguistics, Plato's Cratylus, and Augustine's De Dialectica’, in D. Frede and B. Inwood (edd.), Language and Learning: Philosophy of Language in the Hellenistic Age (Cambridge, 2005), 36–55; R.D. Lamberton, Homer the Theologian: Neoplatonist Allegorical Reading and the Growth of the Epic Tradition (Berkeley, 1986), 38–41; and O'Hara (n. 6), 19–21.

⁴¹ I follow Pease (n. 8), 689 for the translation of multae aliae naturae deorum as ‘many other kinds of gods’.

⁴² Nat. D. 2.63: … fabulas poetis suppeditauerunt, hominum autem uitam superstitione omni referserunt.

⁴³ As Dietrich (n. 7), 33–5 points out, many Latin etymologies (particularly the earlier ones) seem to be imitations or translations of Greek etymologies, and these names in particular bear this out. See also Pease (n. 8), 710–11, 720.

⁴⁴ Cicero is the first extant author to connect Jupiter to iuuans pater, though at Ling. 5.67 Varro alludes to it in deriving Juno (which Cicero also derives from iuuans) from iuuat una. The tradition of deriving Ceres from gero, however, can already be found in Ennius' Epicharmus (in Varro, Ling. 5.64): quae quod gerit fruges, Ceres.

⁴⁵ On this, see Dietrich (n. 7), 32–41.

⁴⁶ See Leg. 2.59 (= GRF 13), where Cicero says Aelius thought the archaic word lessus must signify lamentation, because that was what it sounded like. While Balbus does not actually cite Aelius Stilo for any of his etymologies, it is unproblematic to assume an Aelian source, given Cicero's use of his etymologies elsewhere (cf. Top. 10).

⁴⁷ D. Lehoux, What Did the Romans Know? An Inquiry into Science and Worldmaking (Chicago, 2012), 42–6 discusses how Cicero uses observation of the natural world (and the description of such observation) as philosophical proof. As Gee (n. 3 [2001]), 530–1 points out, the poem is not used just for persuasive effect: its beautiful and rational language parallels Balbus' description of the beauty and rationality of the universe.

⁴⁸ These are frr. XII and XIV respectively; while in the Phaenomena Cicero transliterates the name Ophiuchus (fr. XIV), Balbus refers to the constellation with the calque Anguitenens (Nat. D. 2.108). The original passage of Aratus, which contains a similar gloss, is discussed at Pendergraft (n. 15), 55.

⁴⁹ Fr. XVI, looking back to Phaen. 91–3: ἐξόπιθεν δ’ Ἑλίκης φέρεται ἐλάοντι ἐοικὼς | Ἀρκτοφύλαξ, τόν ῥ’ ἄνδρες ἐπικλείουσι Βοώτην, | οὕνεχ’ ἁμαξαίης ἐπαφώμενος εἴδεται Ἄρκτου. Note the use of a variant of ἐπίκλησις. For more on this wordplay, see Possanza (n. 3), 70 n. 40; for Aratus' version, see Pendergraft (n. 15), 56–7.

⁵⁰ See Kidd (n. 9), 279–80.

⁵¹ See e.g. Plin. HN 18.247; further evidence from Gellius and Tiro will be discussed below. In fact, as Le Boeuffle (n. 11), 155–9 notes, the etymology from pigs is far likelier to be accurate. For more on the Roman name, see Gundel (n. 31), 101–7. One does wonder if Ovid (Fast. 3.100–15), who is certainly drawing on the Aratean tradition (see Gee [n. 19], 175–6), might have Balbus' complaints (or perhaps comments from the lost passage of the Aratea) in mind.

⁵² Though Tiro's work is lost, and Gellius is writing much later, it is worthwhile to consider this passage as, at the very least, provisional comparative evidence, especially given its association with Balbus' comments. No matter its provenance, it is still a revealing look at Roman attitudes towards Greek, and earlier Roman, culture.

⁵³ I say ‘happen to be’, yet I doubt this is mere happenstance on Gellius' part; surely he recognized the confluence between this passage and Balbus' comments in De Natura Deorum. This would fit with Tiro's larger role within Noctes Atticae: Gellius quotes Tiro more than any other source, and Gunderson (n. 31), 186–90 points out that Tiro serves as a doublet in the work for both Cicero and Gellius himself. The Pandectae makes him a Gellian figure, while his employment makes him a Ciceronian one; he can thus serve as an intermediary between the two.

⁵⁴ Pease (n. 8), 822–3.

⁵⁵ McDermott, W., ‘M. Cicero and M. Tiro’, Historia 21 (1973), 259–86Google Scholar, at 262: ‘Tiro surely had a large part in the composition [of the philosophica]: taking down passages at dictation, checking references [emphasis mine], perhaps even revising passages.’

⁵⁶ Compare e.g. the comments from the scholia on this etymology quoted above in n. 28.

⁵⁷ The potential pun in suerunt is supported by the fact that the verb itself is quite rare in Cicero's corpus; as Pease (n. 8), 823 notes, it is only seen again at Fam. 15.8, while elsewhere Cicero prefers the compound consuesco.