Published online by Cambridge University Press: 11 February 2009
In his discussion of Roman military institutions Polybius described how the desire for fame might inspire Roman soldiers to heroic feats of bravery, including single combat: (6.54.3–4) τ⋯ δ⋯ μέγιστον, οἱ νέοι παρορμ⋯νται πρ⋯ς τ⋯ π⋯ν ὑπομένειν ὑπ⋯ρ τ⋯ν κοιν⋯ν πραγμάτων χάριν το⋯ τυχεῖν τ⋯ς συνακολουθούσης τοῖς ⋯γαθοῖς τ⋯ν ⋯νδρ⋯ν εὐκλείας. πίστιν δ' ἔχει τ⋯ λεγόμενον ⋯κ τούτων. πολλο⋯ μ⋯ν γ⋯ρ ⋯μονο-μάχησαν ⋯κουσίως Ῥωμαίων ὑπ⋯ρ τ⋯ς τ⋯ν ὅλων κρίσεως κτλ. Modern scholars, however, have taken little notice of this remark and some have tried to belittle the importance of single combat at Rome. Thus G. Dumézil alleged that the Romans fought few single combats and that this was significant for their outlook upon war, while R. Bloch described the duels in the seventh book of Livy as ‘un mode de combat absolument étranger à la tradition romaine, mail auquel les Romains ont été contraints par les habitudes et par les défis des Celtes’. W. V. Harris is the only scholar to have understood the importance of monomachy in the Roman Republic, but even he has not assembled all the evidence necessary for an accurate assessment of the phenomenon. This essay is intended to provide a full treatment and thus to make some contribution in a limited but interesting area to our understanding of Roman attitudes to warfare. I have included a list and discussion of all instances of single combat from the Roman Republic which I have discovered and have argued that the custom continued from prehistoric times at least to 45 b.c.
1 La Religion Romaine Archaïque (1966), 212 = Archaic Roman Religion [trans. Krapp], P. (1970), i. 210Google Scholar ‘single combats are exceptional’.
3 War and Imperialism in Republican Rome 327–70 B.C. (1979), 38–9.
6 Loc. cit. above.
7 1.24.1 ‘tamen in re tam clara nominum error mallet, utrius populi Horatii, utrius Curiatii fuerint’.
8 On the Horatii and Curiatii see in particular Münzer, F., RE iv. 1830–1Google Scholar and viii. 2321–7 and also Ogilvie on Liv. 1.24.1–26.14. Apart from Livy there is a long narrative at Dion. Hal. 3.13.1–22.10.
9 See DH 10.37.3, Val. Max. 3.2.24, Plin. nat. 7.101, Gell. 2.11.3 and Fest. 208 L.; also Münzer, , RE ii A.2189–90Google Scholar and the discussion below.
10 See Diod. 12.64.3, Liv. 4.29.5–6, Gell. 1.13.7, 17.21.17; also Snodgrass, A. M., JHS 95 (1965), 119–20Google Scholar.
11 The principal sources for the exploit of Manlius are Quadr. fr. 10 (Peter), Liv. 7.9.6–10.4, DH 14.12, Zon. 7.24. Others are listed by Broughton, T. R. S., The Magistrates of the Roman Republic [henceforth MRR] (1951), i. 119Google Scholar. On the duels of Torquatus and Valerius Corvus with their respective Gauls see Bayet–Bloch, op. cit. (n. 2), 108–17 (to be used with caution) and Neraudeau, J.-P. in Mélanges … J. Heurgon (1976), 685–94Google Scholar, who discusses Manlius in connection with the iuuentus in the Early Republic. The accounts of Livy and Quadrigarius have often been compared; in addition to Neraudeau see for example Heinze, R., Die Augusteische Kultur (1930), 97–102Google Scholar = Burck, E. (ed.), Wege zu Livius (1967), 378–9Google Scholar, Büchner, K., Römische Literaturgeschichte 3 (1962), 360–5Google Scholar = Burck, op. cit., 380–2, Luce, T. J., Livy: The Composition of his History (1977), 224–6Google Scholar and Lipovsky, J. P., A Historiographical Study of Livy Books Six to Ten (1981), 95Google Scholar.
12 See MRR i. 129. The main sources are Liv. 7.26.1–10, DH 15.1.1–4, Zon. 7.25 and Gell. 9.11.1–10 = Quadr. fr. 12 P (which, however, should not be ascribed to Quadrigarius). Terzaghi, N., St. Etr. 8 (1934), 157–64Google Scholar, argued that in earlier versions of the story the bird was not made to peck the Gaul and that the references to this in the fragment ascribed wrongly to Quadrigarius are interpolations by Gellius from a later account. Since there are no inconsistencies to be explained away this is all special pleading. Still more incredible is his attempt to link Rutilius Rufus with the formation of the story.
13 See Diod. 5.30.2 κράνη δ⋯ χαλκ⋯ περιτίθενται μεγάλας ⋯ξοχ⋯ς ⋯ξ ⋯αυτ⋯ν ἔχοντα κα⋯ παμμεγέθη ɸαντασίαν ⋯πιɸέροντα τοῖς χρωμένοις. τοῖς μ⋯ν γ⋯ρ πρόσκειται συμɸυ⋯ κέρατα, τοῖς δ⋯ ⋯ρνέων ἢ τετραπόδων ζῴων ⋯κτετυπωμέναι προτομαί. Sil. 5.132–6 describes animals on a Gallic helmet (on this passage see Nicol, J., The Historical and Geographical Sources Used by Silius Italicus , 156Google Scholar). Note also Bloch, R., REL 47 bis (1969), 165–72Google Scholar and Keppie, L. J. F., Colonisation and Veteran Settlement in Italy 47–14 B.C. (1983), 30Google Scholar.
15 Nos. 74–232 and 75–509; illustrations in Bayet-Bloch, op. cit. (n. 2), figs 1 and 2 and Terzaghi, op. cit. (n. 12), 163. For further discussion in the context of the legend of Aeneas and Turnus, see Small, J. P., AJA 78 (1974), 49–54CrossRefGoogle Scholar and Studies related to the Theban Cycle on Late Etruscan Urns (1981), 116–21Google Scholar.
16 Rightly rejected by Terzaghi; the idea was developed by Bloch, op. cit. (n. 2), 113–17.
19 Marc. 2.1 Μάρκελλος δ⋯ πρ⋯ς οὐδ⋯ν μ⋯ν ἦν μάχης εἶδος ⋯ργ⋯ς οὐδ⋯ ⋯νάσκητος, αὐτ⋯ς δ' ⋯αυτο⋯ κράτιστος ⋯ν τῷ μονομαχεῖν γενόμενος οὐδεμίαν πρόκλησιν ἔɸνγε, πάντας δ⋯ τοὺς προκαλεσαμένους ⋯πέκτεινεν.
21 See Liv. 45.39.16, Plut. Aem. 31.2, Crawford, M. H., Roman Republican Coinage (1974), i. 289Google Scholar.
22 4.264–310. Note also 5.137–9 (referring to the helmet of Flaminius): ‘nobile Gargeni spolium, quod rege superbus | Boiorum caeso capiti illacerabile uictor | aptarat, pugnasque decus portabat in omnes’. This may provide further evidence for Silius' understanding of the nature of the Gallic wars.
23 However, the view of Nicol, op. cit. (n. 13), 155 that the abuse spoken by Crixus at 279–81 represents Silius' ethnographical research (in the passage cited below Diodorus records this Gallic boasting and it was probably to be found in Poseidonius) seems fanciful, since such invective is regular in epic. Does the reply of Scipio (286–8) come from ethnographical research?
25 22.44.4, 25.9.7 and 28.33.3 are somewhat less close.
26 See Liv. 23.46.12–47.8, Sil. 13.142–78 and App. Han. 161 (who dates the episode to 211).
27 56 P.
29 See Liv. 25.18.4–15. The story is told differently by Val. Max. 5.1.3.
30 Han. 160.
31 Lib. 188–9.
32 For the sources see Walbank on Pol. 35.5.1–2 and Astin, A. E., Scipio Aemilianus (1967), 46 n. 4Google Scholar.
33 See Val. Max. 3.2.21 and Liv. ep. Oxy. 53–4 (slightly divergent) and, for discussion, Kornemann, E., Die neue Livius-Epitome (1904), 59–60Google Scholar and Münzer, , RE xvii. 1763Google Scholar. There is, however, no need to believe that these stories about Occius were invented under the influence of the career of Siccius Dentatus (no. 2): such influence is more probable from the historical figure to the legendary.
34 See e.g. Liv. 25.18.9 ‘si parum publicis foederibus ruptis dirempta simul et priuata iura esse putet, Badium Campanum T. Quinctio Crispino Romano palam duobus exercitibus audientibus renuntiare hospitium’.
35 See Kinsella, op. cit. (n. 79), 130 and 181 ‘Then they bitterly reproached each other and they broke off their friendship’.
38 I am grateful to Mr G. Herman, who is writing a book on guest-friendship, for discussing this material with me.
39 Mar. 3.2.
40 Frontin. strat. 4.7.5.
41 Ampel. 22.4 ‘Lucius Opimius sub Lutatio Catulo consule in saltu Tridentino prouocatorem Cimbrum interfecit’.
42 Civ. 1.50.219–20.
44 Plut. Sert. 13.3–4.
45 See Münzer, F., RE vii A. 241Google Scholar, for the sources, which stem from Varro, who served with Pompey. Pliny describes the incident thus: (nat. 7.81) ‘atque etiam hostem ab eo ex prouocatione dimicantem inermi dextera superatum et postremo correptum uno digito in castra tralatum’.
46 Per. 8. By far the best treatment of this incident is to be found in the unpublished Oxford doctoral thesis of Henderson, J. G. W., Anecdote and Satire in Phaedrus (1976), 370–85Google Scholar. This is also the best treatment of the Latin literary sources for single combat, and I owe much to it. Henderson has no difficulty in demolishing on pp. 372–3 the fantasy of Havet, L., RPh 22 (1898), 177–8Google Scholar, that this tale is based upon the story of the son of Tritanus.
47 Hirt. Gall. 8.48.1–7.
48 [Caes.] Hisp. 25.3–5.
49 [Quint.] decl. min. 317 (pp. 246–9 Ritter).
50 Liv. 4.19.1–6 (almost certainly annalistic reconstruction). However, Propertius (4.10.23–38) describes the exploit of Cossus in terms of a formal duel and Valerius Maximus does the same for Romulus (3.2.3; contrast Liv. 1.10.4 and DH 2.33.2).
52 See the graphic description at Plut. Marc. 6.1–8.6, which may well go back to an eye–witness account.
54 Diod. 36.10.1 and 37.23; see also Vogt, J., Ancient Slavery and Modern Ideology (1974), 77–8Google Scholar.
56 See below n. 141.
59 Cf. also Men. 184–8, Poen. 470–3 and Truc. 621–9.
58 Op. cit. (n. 3), 37–8 and 252.
59 The Campanian evidence is discussed below.
63 This evidence is assembled and used to illuminate Campanian attitudes to warfare by Nicolet, C., MEFRA 74 (1962), 463–517CrossRefGoogle Scholar, but he was wrong to argue that the artistic evidence pointed to a lost tradition of epic poetry. For a more sophisticated treatment of the Campanian cavalry placing them in their social setting, see Frederiksen, M. W., DdA 2 (1968), 3–31Google Scholar (and cf. Campania , 143–8Google Scholar).
64 Davie, M. R., The Evolution of War: A Study of its Role in Early Societies (1929), 176–95Google Scholar (especially 177–9). The following paragraphs draw heavily upon this work.
65 See Lea, H. C. in Bohannan, P. (ed.) Law and Warfare: Studies in the Anthropology of Conflict (1967), 233–53Google Scholar.
66 E. A. Hoebel in Bohannan, op. cit., 195–201.
68 See Best, E., Journal of the Polynesian Society (henceforth JPS) 12 (1903), 37–9, 13 (1904), 75Google Scholar, The Children of the Mist [Memoirs of the Polynesian Society 6] (1925), 334–5Google Scholar, Tregear, E., The Maori Race (1904), 368–70Google Scholar, Downes, T. W., JPS 38 (1929), 164Google Scholar, Buck, P. H., The Coming of the Maori (1950), 399Google Scholar, Kelly, L. G., Tainui: The Story of Hoturoa and his Descendants [Memoirs of the Polynesian Society 25] (1949), 157–8Google Scholar, Vayda, A. P., Maori Warfare [Polynesian Society Monographs 2] (1960), 65–7Google Scholar = Bohannan, op. cit. (n. 65), 370–1.
70 Downes, loc. cit., believed this to be the general rule.
71 See Best quoted below.
72 Kelly, loc. cit. (n. 68).
73 Buck, loc. cit (n. 68), has a description of one such incident.
75 See Bancroft, H. H., The Native Races of the Pacific States of North America (1875), i. 105–6Google Scholar.
76 Davie, op. cit. (n. 64), 178.
78 I am aware that the historicity of events described in these sagas is the subject of debate. In using the sagas as evidence for the ancient Celtic attitude to monomachy I have adopted a position close to that of e.g. Jackson, K. H., The Oldest Irish Tradition: A Window on the Iron Age (1964)Google Scholar, especially 30–1, and MacCana, P., Études Celtiques 13 (1972/1973), 61–119Google Scholar.
79 See pages 4–5, 27, 28, 32, 37–8, 40–5, 73, 88–101, 111–12, 114, 119–42, 164–7, 168–205 in the translation of Kinsella, T. entitled The Tain (1969)Google Scholar.
80 Tac. Germ. 10.3.
81 Biblical scholars have tended to concentrate their energies on elucidating the complexities and problems of the tale rather than in considering the function of single combat in Jewish society at the time of Saul and David. For discussion of the story of David and Goliath see e.g. Hertzberg, H. W., I and II Samuel: A Commentary [trans. Bowden, J. W.] (1964), 142–55Google Scholar, De Vries, S. J., Journal of Biblical Literature 92 (1973), 23–36CrossRefGoogle Scholar, Jason, H., Biblica 60 (1979), 36–70Google Scholar. A notable exception is R. de Vaux, who has discussed single combat in the Old Testament in an article variously reprinted: Biblica 40 (1959), 495–508Google Scholar = Bible et Orient (1967), 217–30Google Scholar = The Bible and the Ancient Near East [trans. McHugh, D.] (1972), 122–35Google Scholar. I cite the English translation, to which I owe much.
82 Quotations from the New English Bible (1970).
83 Cf. I Chron. 20.4–8.
84 Cf. I Chron. 11.21–5.
86 II Sam. 8.18, 20.23.
87 See de Vaux, op. cit. (n. 81), 127–9.
88 Op. cit. (n. 81), 129.
89 There is no synoptic treatment of single combat in Ancient Greece. Note, however, Armstrong, A. M., GR 19 (1950), 73–9Google Scholar, Glück, J. J., AC 7 (1964), 25–31Google Scholar, Fenik, B., Typical Battle Scenes in the Iliad [Historia Einzelschriften 21] (1968)Google Scholar, Ilari, V., Guerra e diritto nel mondo antico (1980), i. 54–5Google Scholar. Hellenistic examples are well discussed by Hornblower, J., Hieronymus of Cardia (1981), 194–6Google Scholar.
91 Op. cit. (n. 89).
92 See Hes. frr. 23 and 176 M-W, Hdt. 9.26.3–5, Paus. 1.41.2, 44.10, 4.30.1, 8.5.1, 45.3 and 53.10.
93 Strab. 8.3.33, Paus. 5.4.2–3.
94 Strab. 9.1.7, Frontin. strat. 2.5.41, Paus. 2.18.9, Polyaen. 1.19.
95 Hdt. 5.94.1–2, Strab. 13.1.38, Plut. mor. 858a–b, Diog. Laert. 1.74, Polyaen. 1.25.
97 In 420 b.c.; see Thuc. 5.41.2. The original contest should be placed at some point in the middle of the sixth century.
98 5.1.1–2.1; the date is uncertain but must be before the campaign of Melabazus against the Perinthians.
99 Op. cit. (n. 89), 78.
100 Hdt. 6.92.2–3.
101 Plut. Pyrrh. 7.4–5, 24.1–4;16.8–10 and 30.5–6, though they illustrate Pyrrhus' temperament, are somewhat different.
102 Op. cit. (n. 89).
103 See Iust. 23.4.12 ‘Denique aduersus prouocatores saepe pugnauit semperque uictoriam reportauit’.
104 For an interesting discussion of this tension in Spartan society, see Hodkinson, S., Chiron 13 (1983), 239–81Google Scholar.
105 Horace et les Curiaces (1942), 11–33.
107 See above.
108 See ‘L'Exploit de Titus Manlius Torquatus’ in Mélanges…J. Heurgon (1976), 685–94Google Scholar and La Jeunesse dans la Littérature et les Institutions de la Rome Républicaine (1979), especially 249–58.
109 (1979), 249.
110 (1979), 249–50.
111 8.12.1 ‘cui uenienti seniores tantum obuiam exisse constat, iuuentutem et tunc et omni uita deinde auersatam eum exsecratamque’. See (1976), 688.
112 (1976), passim. Note 7.10.1 ‘diu inter primores iuuenum Romanorum silentium fuit’, 10.5 ‘armant inde iuuenem aequales’, 10.12 ‘Romani alacres ab statione obuiam militi suo progressi…ad dictatorem perducunt’.
114 See (1976), 692: ‘ensuite l'arrivée des Gaulois suscite des héroismes et le retour à des pratiques guerrières fondées sur le furor et la magie’.
115 See (1976), 692–4 and (1979), 255–6.
116 See e.g. DH 3.19.1–2.
117 See Lipovsky, op. cit. (n. 11), 102–31 for this theme in Livy Books 6–10.
118 On deuotio see now the wide-ranging and suggestive essay on substitutionary sacrifice by Versnel, H. S. in Le Sacrifice dans L'Antiquité [Fondation Hardt Entretiens Tome 27] (1981), 135–94Google Scholar.
119 Enemy forces would sometimes flee after the death of their champion; see Liv. 7.11.1 and amongst other societies the practice of the Maoris (for which see Best, E., JPS 12 , 37Google Scholar, L. G. Kelly, op. cit. [n. 68], 158 ‘The enemy became panic stricken at the unexpected and sudden end of their leader, and took to flight’, Vayda, op. cit. [n. 68], 367), the flight of the Philistines after Goliath's death (I Sam. 17.51) and also e.g. Plut. Pyrrh. 24.4, Arr. an. 4.24.4–5.
120 There is a good collection of passages illustrating the Roman view of their military discipline by Fiebiger, at RE v. 1176–83Google Scholar; see also Lind, L. R., TAPhA 102 (1972), 252–3Google Scholar and note amongst the obiter dicta of ancient authors e.g. Cic. Tusc. 1.2 ‘quid loquar de re militari? in qua cum uirtute nostri multum ualuerunt, turn plus etiam disciplina’, Liv. 2.44.10, Val. Max. 2.7 praef. ‘uenio nunc ad praecipuum decus et ad stabilimentum Romani imperii, salutari perseuerantia ad hoc tempus sincere et incolume seruatum, militaris disciplinae tenacissimum uinculum’.
121 See Liv. 7.10.2–4, 26.2, 23.47.1, 25.18.12, Phaedr. Per. 8.18–25 (where Henderson aptly compares the ‘licet’ of Phaedrus with Liv. 23.47.1 ‘percontaretur liceretne extra ordinem in prouocantem hostem pugnare’), Sil. 13.153–6. Liv. 25.18.12 ‘itaque tantum moratus dum imperatores consuleret permitterentne sibi extra ordinem in prouocantem hostem pugnare’ is remarkably similar in phrasing to 23.47.1. This shows that Livy thought of these scenes in a formulaic manner.
132 For Livy's view of the matter see further 24.8.3–6.
133 Note especially Caes. Gall. 2.30.4 ‘nam plerumque hominibus Gallis prae magnitudine corporum suorum breuitas nostra contemptui est’. Further illustrative material is collected by Goodyear on Tac. ann. 1.64.2 and by Sherwin-White, A. N., Racial Prejudice in Imperial Rome (1967), 57–8Google Scholar; add Sil. 5.110–13 and Veg. mil. 1.1.
135 Marc. 7.1.
136 Plut. Pyrrh. 24.1–4.
137 Quoted and discussed by Dawson, R. M. in Medieval Literature and Folklore Studies [Essays…F. L. Utley] (1970), 305–21, especially 310–21Google Scholar.
138 Lüthi, M., Journal of the Folklore Institute 4 (1967), 3–16CrossRefGoogle Scholar. Also analogous are the tales of ill-equipped brigands defying the might of the Roman Emperor; see Shaw, B. D., Past and Present 105 (1984), 3–52CrossRefGoogle Scholar (esp. 43–52) for discussion and analysis from a different point of view.
139 See Harris, op. cit. (n. 3), 9–53 and Hopkins, M. K., Conquerors and Slaves (1978), 25–37Google Scholar for recent discussion.
140 See the thorough survey of Maxfield, V. A., The Military Decorations of the Roman Army (1981)Google Scholar.
141 The four descriptions of the career of Siccius are remarkably similar. Pliny allows Dentatus thirty-four spolia as opposed to Valerius, who recorded thirty-six (though one set of manuscripts may be corrupt), and also records that he took ten prisoners and twenty-one oxen on one occasion and that he was rewarded with a bag of money. Gellius agrees exactly with Valerius. The account of DH, however, is much more extensive. In it one finds, for instance, that on one day Siccius received twelve wounds (37.2), but DH disagrees with Valerius in making Siccius fight nine duels and not eight. Perhaps all four depend ultimately upon Varro, who was cited by Valerius and was interested in such matters (cf. the son of Tritanus, no. 27), though Gellius cites the annalists as his source; see also Maxfield, op. cit. (n. 140), 43–5 for discussion of sources.
142 Similarly commanders might keep a count of the number of battles in which they had fought; see Plin. nat. 7.92 (Julius Caesar compared with the great Marcellus).
144 Compare e.g. Liv. 6.20.8 and Plin. nat. 7.103.
145 Honourable scars might be of use to a Roman both in his political career and if he was in danger of conviction. Note in particular Cic. 2 Verr. 5.3 ‘ipse [sc. M. Antonius] arripuit M.' Aquilium constituitque in conspectu omnium tunicamque eius a pectore abscidit ut cicatrices populus Romans iudicesque aspicerent aduerso corpore exceptas; simul et de illo uulnere quod ille in capite ab hostium duce acceperat multa dixit eoque adduxit eos qui erant iudicaturi, uehementer ut uererentur ne quem uirum fortuna ex hostium telis eripuisset cum sibi ipse non pepercisset, hic non ad populi Romani laudem sed ad iudicum crudelitatem uideretur esse seruatus’ (Antonius' speech was famous; cf. especially in this context Cic. or. 2.124, Liv. per. 70), Liv. 6.20.8 ‘nudasse pectus insigne cicatricibus bello acceptis’ (cf. Plin. nat. 7.103 ‘XXIII cicatrices aduerso corpore exceperat’), 45.39.16–17 ‘“insigne corpus honestis cicatricibus omnibus aduerso corpore exceptis habeo”. Nudasse deinde se dicitur’ and Ter. Eun. 482–3 ‘neque pugnas narrat neque cicatrices suas | ostentat’. Marius contrasted his scars with the imagines of the nobles (Plut. Mar. 9.2). For other passages mentioning scars see e.g. Liv. 2.23.4, 27.2, 4.58.13, 6.14.6, Val. Max. 6.2.8 and 7.7.1. On the mutilated M. Sergius see Plin. nat. 7.104–5.
146 Nos. 10 and 11 above.