¹ Wiseman, T.P., ‘The publication of De Bello Gallico’, in Welch, K. and Powell, A. (edd.), Julius Caesar as Artful Reporter: The War Commentaries as Political Instruments (London, 1998), 1–9Google Scholar; for earlier contributions to the composition debate, see Rambaud, M., L'art de la déformation dans les commentaires de César (Paris, 1966)Google Scholar; Gesche, H., Caesar (Darmstadt, 1976), 83–7Google Scholar; Schadee, H., ‘Caesar's construction of northern Europe: inquiry, contact and corruption in De Bello Gallico’, CQ 58 (2008), 158–80CrossRefGoogle Scholar, at 158 nn. 2–4; and Krebs, C.B., ‘Caesar and the dates of De Rerum Natura and the Commentarii’, CQ 63 (2013), 772–9CrossRefGoogle Scholar, at 775 n. 21.

² The similarity between Caesar's two German ethnographies has long been seen as evidence that the ethnographic sections of the Gallic War did not originate with Caesar, but were interpolations by later authors. Klotz, A. (Caesarstudien [Leipzig, 1910], 26–7)Google Scholar first suggested that the ethnographic excursuses were interpolations, Beckmann, F. (Geographie und Ethnographie in Caesars Bellum Gallicum [Dortmund, 1930])Google Scholar, through painstaking philological examination, the opposite. Norden, E., ‘Der Germanenexkurs in Caesars Bellum Gallicum. Die ethnographischen Abschnitte Caesars über Suebi und Germani’, in Rasmussen, D. (ed.), Caesar (Wege der Forschung 43) (Darmstadt, 1967), 116–37Google Scholar, at 118–25 (reprinted from Norden, E., Die Germanische Urgeschichte in Tacitus Germania [Berlin, 1920], 84–104Google Scholar) mostly sided with Klotz (this note), but also argued that Caesar added the Gallic and German ethnographies of Book 6 during final editing of the Gallic War in 52/51 b.c. Jachmann, G., ‘Caesartext und Caesarinterpolation’, RhM 89 (1940), 161–88Google Scholar used an approach similar to Beckmann's, but argued that Books 1, 2 and 4 of the Gallic War were mostly written by late antique interpolators. Berres, T., ‘Die geographischen Interpolationen in Caesars Bellum Gallicum’, Hermes 98 (1970), 154–77Google Scholar, at 155–63, 164–70, 172–5 believed that the number of instances of repetition in the Gallic War was evidence for Klotz's theory, and that Caesar's initial ethnography of Gaul and his descriptions of the Rhine expedition, the British expedition and the Hercynian Forest are all the ‘pseudo-Caesarlike’ additions of another writer, whom Berres believed was a member of Caesar's staff in Gaul. Gesche, who collected the literature this controversy produced (Gesche [n. 1], 83–7, 259–60), wonders about Berres's theory: is it feasible to speak of ‘pseudo-Caesarlike’ repetitions in Caesar's writing, when those very repetitions seem to form a pattern that could be a stylistic trait of Caesar himself (Gesche [n. 1], 86)? I (like most recent students, explicitly or implicitly) side with those skeptical about interpolation, since the abrupt shifts from narrative to ethnographic and geographic excursuses in Caesar's work, while jarring, are a common feature of the Greek historical tradition, and thus no proof in themselves of interpolation. Also, as I argue, one of the primary functions of these excursuses is creating, maintaining and expanding the ethnographic framework; as such, the noticeable difference between them and the rest of the Gallic War only proves that they served a special purpose in Caesar's telling of his campaigns.

³ Lendon, J.E., ‘The rhetoric of combat: Greek theory and Roman culture in Julius Caesar's battle descriptions’, ClAnt 18 (1999), 273–329Google Scholar; Riggsby, A.M., Caesar in Gaul and Rome: War in Words (Austin, 2006), 73–105Google Scholar.

⁴ On Caesar's depiction of Gallic temperament and war aims, see Gardner, J.F., ‘The “Gallic menace” in Caesar's propaganda’, G&R 30 (1983), 181–9Google Scholar, at 185–7; Rawlings, L., ‘Caesar's portrayal of Gauls as warriors’, in Welch, K. and Powell, A. (edd.), Julius Caesar as Artful Reporter: The War Commentaries as Political Instruments (London, 1998), 179–92Google Scholar, at 176–7.

⁵ There is some speculation as to whether or not Pytheas’ lost work contained a German ethnography in addition to his better-attested British one (Hawkes, C.F.C., Pytheas: Europe and the Greek Explorers [Oxford, 1977], 26Google Scholar), but there is no evidence to support this claim.

⁶ Williams, J.H.C., Beyond the Rubicon: Romans and Gauls in Republican Italy (Oxford, 2001), 78–9Google Scholar; Woolf, G., Tales of the Barbarians: Ethnography and Empire in the Roman West (London, 2011), 60CrossRefGoogle Scholar.

⁷ On Polybius and ethnography, see Champion, C., Cultural Politics in Polybius's Histories (Berkeley and Los Angeles, 2004), 30–66CrossRefGoogle Scholar, 231–54.

⁸ The intellectual relationship between Cato and Polybius is hotly contested. See Nicolet, C., ‘Polybe et les institutions romaines’, in Polybe (Entretiens sur l'Antiquité Classique 20) (Geneva, 1974), 209–58Google Scholar, at 243–55; Garbarino, G., Roma e la filosofia greca dalle origini alla fine del II secolo a.C.: Raccolta di testi con introduzione e commento (Turin, 1973), 343–8Google Scholar; Musti, D., Polibio e l'imperialismo romano (Naples, 1978), 128–9Google Scholar.

⁹ The Gauls referred to here are called Celts by the Roman Cato, the Greek Polybius and many other Greek ethnographers, but Caesar consistently calls them Gauls—although he does explain at the beginning of his work that they are called Gauls by the Romans and Celts in their own tongue (BGall. 1.1.1). Because Caesar is the focus of this paper, my usage will be consistent with his.

¹⁰ Williams (n. 6), 70–9.

¹¹ For discussion of Posidonius’ place in first-century b.c. historiography, see Yarrow, L.M., Historiography at the End of the Republic: Provincial Perspectives on Roman Rule (Oxford, 2006), 101–18CrossRefGoogle Scholar, 161–6.

¹² Hawkes (n. 4), 25–32.

¹³ Nash, D., ‘Reconstructing Poseidonios’ Celtic ethnography: some considerations’, Britannia 7 (1976), 111–26CrossRefGoogle Scholar, at 112. Tierney argued that all four authors are important sources for reconstructing Posidonius’ writings (Tierney, J.J., ‘The Celtic ethnography of Posidonius’, Proceedings of the Royal Irish Academy 60 [1960], 189–275Google Scholar, at 198).

¹⁴ Nash (n. 13), 115. Norden (n. 2) first argued that Caesar borrowed extensively from Posidonius, whom Norden saw as the main source of northern European ethnography for Caesar, Tacitus and other later authors. Tierney agreed, and believed that Posidonius’ work was the principal source on the Gauls for the rest of antiquity (Tierney [n. 13], 198, 201, 203). Rasmussen believed that Caesar ‘grafted’ the works of other authors like Posidonius into his ethnographic excursuses, blending his own material with theirs to such a degree that his ethnographies became stylistically distinct from theirs (Rasmussen, D., ‘Das Autonomwerden des geographisch-ethnographischen Elements in den Exkursen’, in id. [ed.], Caesar [Wege der Forschung 43] [Darmstadt, 1967], 339–71Google Scholar; repr. from id., Caesars Commentarii. Stil und Stilwandel am Beispiel der Direkten Rede [Göttingen, 1963], 79–104). Nash was highly critical of this approach, observing that the exact contents of Posidonius’ writings ‘are only available through textual reconstruction from derivative authors’ and that most of the supposed references and quotations attributed to him ‘are not acknowledged and can be ascribed to him only by inference’ (Nash [n. 13], 111). She argued rather that no concrete passages from Posidonius exist to prove that Caesar's ethnographic passages are effectively plagiarized, and that Caesar's work was more likely unique to Caesar, the result of his extensive experience in Gaul and beyond (Nash [n. 13], 115–19).

¹⁵ Generally, on the degree of Caesar's engagement with the ethnographic tradition: Allen-Hornblower, E., ‘Beasts and barbarians in Caesar's Bellum Gallicum 6.21–8’, CQ 64 (2014), 682–93CrossRefGoogle Scholar, at 685 n. 20; Nash (n. 13), 124; Woolf (n. 6), 80. On ethnographers correcting and criticizing their predecessors: Woolf (n. 6), 37; Romm, J., The Edges of the Earth in Ancient Thought: Geography, Exploration, and Fiction (Princeton, 1992), 5Google Scholar. On the potentially subversive uses of ethnography in Caesar—points closer to my own—see Rawlings (n. 4), 173; Riggsby (n. 3), 58.

¹⁶ Hinds, S., Allusion and Intertext: Dynamics of Appropriation in Roman Poetry (Cambridge, 1998), 108–9, 131Google Scholar; Woolf (n. 6), 37.

¹⁷ Bellen, H., Metus Gallicus – Metus Punicus: Zum Furchtmotiv in der römischen Republik (Stuttgart, 1985)Google Scholar argues for the historical power of the metus Gallicus from early times by citing the tumultus, a special declaration by the Senate that allowed for all military-age men—even priests and the elderly—to be levied for defense against a dire threat, usually an approaching Gallic host (9–15). He also notes a correlation between crises and the practice of human sacrifice, which the Romans performed only in times of great duress in 226, 216 and 113 b.c., twice (in 226 and 113 b.c.) when threatened by the Gauls and once (in 216 b.c.) after Cannae (43).

¹⁸ Schadee (n. 1), 170–1; Rawlings (n. 4), 180–1; Riggsby (n. 3), 101–2.

¹⁹ Wells, P.S., The Barbarians Speak: How the Conquered Peoples Shaped Roman Europe (Princeton, 1999), 100–1Google Scholar; Allen-Hornblower (n. 15), 686–7; Romm (n. 15), 47; Woolf (n. 6), 54.

²⁰ All translations are my own, although they tend to draw upon those in the Loeb edition of H.J. Edwards. The Latin is drawn from Du Pontet's Oxford Classical Text. Gallia est omnis diuisa in partes tres, quarum unam incolunt Belgae, aliam Aquitani, tertiam qui ipsorum lingua Celtae, nostra Galli appellantur … horum omnium fortissimi sunt Belgae, propterea quod a cultu atque humanitate prouinciae longissime absunt, minimeque ad eos mercatores saepe commeant atque ea quae ad effeminandos animos pertinent important, proximique sunt Germanis, qui trans Rhenum incolunt, quibuscum continenter bellum gerunt.

²¹ Schadee (n. 1), 170; Krebs, C.B., ‘“Imaginary geography” in Caesar's Bellum Gallicum’, AJPh 127 (2006), 111–36Google Scholar, at 113–14; Rankin, H.D., Celts and the Classical World (London, 1987), 2–4Google Scholar; Riggsby (n. 3), 60; Wells (n. 19), 104; Woolf (n. 6), 83–6.

²² Here the weakened and corrupted Gauls live within a clearly defined and measured space with no further room for inquiry or discovery, a feature that Caesar uses—Schadee (n. 1), 162 has argued—to indicate that Gaul and its people are well known to the Romans.

²³ Naturally, then, Caesar emphasizes the great preparations when his legate P. Crassus is sent against them (BGall. 3.20). Schadee (n. 1), 160 also mentions the confusion Caesar creates with his tripartite division of Gaul and its peoples, and infers from his clearer description of the Gauls’ borders that they are the principal focus of this treatment and that the territories of the Belgae and Aquitani exist ‘only in relation to the Galli’.

²⁴ Helvetii quoque reliquos Gallos uirtute praecedunt, quod fere cotidianis proeliis cum Germanis contendunt, cum aut suis finibus eos prohibent, aut ipsi in eorum finibus bellum gerunt.

²⁵ Book 1 alone also forms a neat unit of composition, given Caesar's remark that he ‘concluded two great wars in one summer’ and sent his forces into winter quarters earlier than required (BGall. 1.54.2), but the Belgae, who are identified in 1.1 along with the Helvetii and Germans as fierce opponents, have not yet been defeated. Furthermore, the language found at the end of Book 2 after the Belgae have been subdued is even more conclusive: ‘These deeds pacified all of Gaul, and such a report of this war was carried about to the barbarians that legates were sent to Caesar from those tribes who dwelt across the Rhine, promising that they would give hostages and carry out his commands … and for these achievements reported by Caesar's letters fifteen days of thanksgiving were decreed, an honour which before that time had fallen to no one’ (BGall. 2.35.1, 4).

²⁶ Broughton, T.R.S., The Magistrates of the Roman Republic, 3 vols. (New York, 1951–86), 2.194Google Scholar.

²⁷ The schedule of the dissemination of the parts of the Gallic War proposed here accords with earlier scholars’ observation of the contradiction between Caesar's statement that the Nervii had been wiped out in 57 b.c. (BGall. 2.28.1) and the fact that they are found besieging Cicero's camp with 60,000 men by 54 b.c. (BGall. 5.49.1). The puzzle can hardly be solved, but the contradiction is more easily excused if Books 1 and 5 were disseminated separately and some years apart. Gesche (n. 1), 257–8 and Rambaud (n. 1), 9–12, 403–5 collect the literature on this and other contradictions.

²⁸ For more on the comparison between Gauls and Germans in Book 6, see Allen-Hornblower (n. 15), 682–93.

²⁹ Gardner (n. 4), 182; Woolf (n. 6), 88.

³⁰ Riggsby (n. 3), 70. Romm (n. 15), 69–78 collects many examples of thaumata.

³¹ See n. 14 above.

³² It is unclear from Strabo's account whether or not wicker men are original either to Posidonius or to Caesar (Edelstein, L. and Kidd, I.G. [edd.], Posidonius: Fragments Vol. I [Cambridge, 1972], 239 fr. 274 KGoogle Scholar). As with many allegations of human sacrifice and cannibalism in all eras, supporting evidence for the practice is sparse, and it is possible that the idea was simply a literary embellishment (Wells [n. 19], 59–60).

³³ See also Potter, D., ‘Empty areas and Roman frontier policy’, AJPh 113 (1992), 269–74Google Scholar. On Caesar's depiction of uirtus among the barbarians of northern Europe, see Lendon, J.E., ‘Julius Caesar, thinking about battle and foreign relations’, Histos 9 (2015), 1–28Google Scholar, at 15–19; for uirtus among the Romans, McDonnell, M., Roman Manliness: Virtus and the Roman Republic (Cambridge, 2006)Google Scholar, especially 290–300.

³⁴ Krebs (n. 21), 113–14; Schadee (n. 1), 170, 178.

³⁵ Such as a unicorn (BGall. 6.26), a peculiar jointless elk (BGall. 6.27) and the huge aurochs ‘somewhat smaller than an elephant’ (BGall. 6.28). Although the aurochs was an actual species known to have inhabited that region of northern Europe, its size in Caesar is likely an exaggeration, making it a fitting companion to the two other animals.

³⁶ Allen-Hornblower (n. 15), 687–93.

³⁷ See n. 2 above.

³⁸ Allen-Hornblower (n. 15), 683–4 also discusses contrasts between the ethnographic excursuses in Books 4 and 6.

³⁹ On this non-encounter between Caesar and the Suebi, and the similar event in Book 6, see Lendon (n. 33), 18–19. Schadee (n. 1), 170 sees the expedition into Germany as primarily an exploratory one, and posits that Caesar had neither the intention nor the need to fight the Suebi, although she does concede the possibility that he had intended to fight them and, having failed to do so, wrote his description of the journey in a way that made it seem less focussed on conquest.

⁴⁰ Riggsby (n. 3), 58–60 argues that Caesar imitates Herodotus’ depiction of the nomadic Scythians for his German ethnography. The milk-and-meat diet is a key indicator of a nomadic lifestyle, he asserts, and of the peoples whom Herodotus places on the outside edges of the known world, who are characterized by ‘mobility and fluidity’ (58). Nevertheless, aspects of Caesar's German ethnography militate against this interpretation: Caesar's German nations have fixed boundaries (even if they keep them barren; BGall. 4.3.2, 6.23.1–4) and their land-tenancy rotation system implies clear long-lasting boundaries within each tribal area (BGall. 4.1.5–9, 6.22.3–4).

⁴¹ Allen-Hornblower (n. 15), 683 n. 4 provides bibliography for and agrees with the traditional view that the Gallic and German excursuses are intended to excuse Caesar's failure against the Germans rather than to emphasize his victory gained by scaring them away (my view).

⁴² Because of allusions to Lucretius which Krebs (n. 1), 779 finds in Book 5 of the Gallic War, he argues that Caesar could not have written Book 5 before late 54 b.c.

⁴³ The tenth legion's beachhead battle with the Britons has the distinction of featuring the first instance of direct speech in the Gallic War (BGall. 4.25.3), a stylistic oddity of Caesar's writing mildly supportive of the idea that he published the work serially. Rasmussen, D., Caesars Commentarii. Stil und Stilwandel am Beispiel der Direkten Rede (Göttingen, 1963), 20–54, 170–1Google Scholar, who, however, argues the opposite, examines every instance of direct speech in the work and collects the literature.

⁴⁴ A subset of what modern scholars call geographic determinism: Woolf (n. 6), 37, 70–3; Hinds (n. 16), 108–9, 131; Hippocratic climatology theorized that climate had a direct effect on human development, suggesting, for example, that people like the Gauls, who lived in cold northern climes, were tough, large-bodied and fierce, while those of hot southern regions were shorter, darker-skinned and more prone to weakness in body and spirit: Schadee (n. 1), 163; Woolf (n. 6), 50.

⁴⁵ Gardner (n. 4), 181–2; Rawlings (n. 4), 173, 180–1; Riggsby (n. 3), 101–2.

⁴⁶ On these qualities among the Romans and their assimilation by the Gauls, see Riggsby (n. 3), 73–105.

⁴⁷ Eisenhut, W., Virtus Romana (Munich, 1973), 17, 21, 30–6Google Scholar; McDonnell (n. 33), 72–134, 320–56.

⁴⁸ Ambiorix's revolt in Book 5 poses the greatest challenge to Caesar before Vercingetorix, but it should be remembered that the Eburones Ambiorix leads are Belgae (BGall. 5.24.2–4), who are categorized as strong opponents by the ethnographic framework. Caesar also notes two separate times the dangers presented by the Morini and Menapii, other Belgic tribes, even if they turn out not to be very difficult to defeat in fact (BGall. 3.28, 4.37–8).

⁴⁹ Both Rawlings (n. 4), 180–1 and Riggsby (n. 3), 97–100 argue that Vercingetorix's leadership and tactical thinking are distinctly Roman in flavour, and cite his Romanitas as the main cause of his temporary success. For Riggsby (n. 3), 101–2 Caesar's depicting Gallic assimilation of Roman technology and culture is intended to increase the impressiveness of his victory over Vercingetorix's Gauls.

⁵⁰ This is in keeping with one of the excuses adduced by Rosenstein, N.S. (Imperatores Victi: Military Defeat and Aristocratic Competition in the Middle and Late Republic [Berkeley and Los Angeles, 1990], 94–7)Google Scholar, whereby defeated Roman generals would regularly shift blame for a defeat from themselves to the common soldiery.

⁵¹ Cunliffe, B., Greeks, Romans, and Barbarians: Spheres of Interaction (New York, 1988), 120Google Scholar.

⁵² Rawlings (n. 4), 180.

⁵³ For reasoning of this sort, Lendon (n. 3), 313.

⁵⁴ Rawlings (n. 4), 181.

⁵⁵ Rawlings (n. 4), 176–7 observes that the Gauls appear to have conceived of warfare and its purpose in a manner quite different from the Romans, and that their primary interest in plunder explains why they retreat with relative ease and frequency. A similar argument might be made about the Germans’ war-aims in Books 4 and 6, but the importance of the point highlighted here is that Caesar happily interprets barbarian opportunism as cowardice, particularly when it makes him appear a more successful general.