1. For an analysis connecting this episode to the issues raised by the Social Wars, see Dipersia 1975; and see also Lipovsky 1981: 130–32.

2. 8.4.4: ubi pars virium, ibi et imperii pars est,.

3. Cf. Aen. 12.820–40. Indeed, Annius’s concession at 8.5.6, “let us all be called Romans,” closely resembles Juno’s insistence on the abandonment of the Trojan name.

4. 5.54.7, and see ch. 1, sec. IV. Manlius’s reference to Jupiter’s auguratum templum (8.5.8) not only refers explicitly to the inaugural ritual through which the consuls entered office every year; it also recalls the description of the original inauguration of the temple itself in Camillus’s peroratio. Again, an emphasis on the preservation of rituals and traditions highlights the parallel between maintaining continuity with the past and continually reestablishing the vertical bonds linking Rome’s political leaders to the power of the gods.

5. 8.6.1 ff. Levene 1993: 218–20, finds an inconsistency between this act and Annius’s affirmation of the gods’ power elsewhere (e.g., 8.4.6 and 8.5.4). But the episode as a whole reveals how closely the cult of the gods was connected to place and nationality. Annius slights the Roman Jupiter; his piety in the abstract is irrelevant. Indeed, his failure to recognize that the consulship of the Romans is divinely sanctioned and not to be arbitrarily shared with the Latins already amounts to a rejection of the Roman Jupiter. This is Torquatus’s point.

6. 8.6.5: Est Caeleste numen; es, magne Iuppiter; haud frustra te patrem deum hominum hac sede sacravimus. Notice again the insistence on locality.

7. Levene 1993: 222–23 and Lipovsky 1981: 112–15.

8. For the motif of disciplina in this section and its role in establishing a distinction between Romans and Latins, cf. 8.6.15–16: Agitatum etiam in consilio est ut, si quando unquam severo ullum imperio bellum administratum esset, tunc uti disciplina militaris ad priscos redigeretur mores. Curam acuebat quod adversus Latinos bellandum erat, lingua, moribus, armorum genere, institutis ante omnia militaribus congruentes.

9. For an analysis of the Brutus episode as spectacle, see ch. 5; and see also Feldherr 1998.

10. Fin. 2.61.

11. Enn. Ann. 191–94 (6.xii) Skutsch. See, however, Cornell 1986a, who argues that the book to which the fragment belongs actually includes material from much earlier than the Pyrrhic war and that Ennius is more likely to have described one of the two successful devotiones than the abortive attempt of the third Decius.

12. Another difficulty is that the Romans ultimately lost the battle of Asculum, although for their opponent this was to prove a “Pyrrhic” victory indeed. Also, since some sections of the Roman army were more successful than others, the devotio of Decius may have affected only the troops under his command.

13. Dio. 10.43 = Zon. 8.5.

14. Not only of a sacrificing priest (Deubner 1905: 70). See Versnel 1981: 148–49, for evidence and bibliography.

15. Skutsch 1985: 355, regards the prayer formula itself as “a fairly competent antiquarian’s product.”

16. For parallels, see Versnel 1981, esp. 153 ff. and 164 ff.; and see also Burkert 1979: 52 ff.

17. Versnel 1976. The formula for this type of devotio is preserved by Macrobius Sat. 3.9.9 ff.

18. 8.11.1: Haec, etsi omnis divini humanique moris memoria abolevit nova peregrinaque omnia priscis ac patriis praeferendo, haud ab re duxi verbis quoque ipsis, ut tradita nuncupata sunt, referre. The language of the passage, with its doublets, heavy alliteration, and pleonasm, makes the historian’s own description of his task an echo of the consul’s prayer.

19. 10.28.14. Ironically, this second pontifex also bears the name Livius, although with a different praenomen than the historian.

20. See Deubner 1905: 71, and esp. Wagenvoort 1947: 34. The chin, which in many European cultures was regarded as a center of life forces (cf. Onians 1951: 233), was surely not chosen just because it is easily accessible when the head is veiled.

21. Deubner 1905: 71–72.

22. If this spear ever comes into the possession of the enemy, a suovetaurilia must be performed to Mars as an expiation (8.10.14).

23. Wagenvoort 1947: 32, defines the devotio as “a religious capitis diminutio maxima,” referring to the legal term used to designate the complete loss of citizen rights. He bases this definition partly on the references to the caput of the devotus as the object of consecration in other Latin authors (Val. Max. 5.6.5, Curt. 8.6.28, Flor. 1.17.7, and Ps. Quint. Decl. mai. 12.11 and 12.18). Livy does not use the term explicitly of either Decius (unless the idea of the loss of caput is somehow conveyed through the gesture of veiling the head), but one of the signs by which Decius learns that he will be required to devote himself is that the liver of the victim he has sacrificed, although healthy in every other respect, has had the “caput cut from the pars familiaris”(8.9.1). On the devotio as a rite of separation, see also Versnel 1981: 148–52.

24. For the many levels of ritual substitution involved in the devotio, see Versnel 1981: 159. Palinurus, whose life is demanded by Neptune in return for the safe arrival of the Trojan fleet in Italy (unum pro multis dabitur caput [Aen. 5.815]), provides the clearest example of the logic of this kind of substitution in Roman literature.

25. 10.28.17: contacturum funebribus diris signa tela arma hostium.

26. 8.9.10: sicut caelo missus piaculum omnis deorum irae.

27. 8.9.11: ita omnis terror pavorque cum illo latus signa primo Latinorum turbavit, deinde in totam penitus acies pervasit. For the connections between contact and “contagion”, see Wagenvoort, 1947, esp. 175–78.

28. 8.10.12: haud secus quam pestifero sidere icti pavebant. Another uncanny aspect of the consul’s charge, one that makes it inexplicable on purely rational terms, is that while we might expect the enemy to be terrified by the sight of such a terrific figure charging toward them, the actual death of the consul, rather than restoring the enemy’s courage, only completes their devastation. Cf., too, the instantaneous demoralization, indeed insanity, that afflicts the Gauls after the death of the younger Decius (10.29.2).

29. 8.9.12–14. Evidentissimum is emphatically placed at the beginning of the sentence.

30. Indeed, even the language with which the consequences of devotio are described can be compared to terms applied to the effects of imperium. The Romans, after the death of the consul, are said to “charge into battle as if the signal had just been given for the first time” (8.9.13). Giving the signal for battle, a procedure we saw emphasized especially in Livy’s narrative of the battle of Aquilonia, may be regarded as one of the essential acts of the imperator. So, too, even the report of the devotio of the younger Decius inspires the Roman soldiers who have not actually seen it (ibi auditur eventus P. Deci, ingens hortamen ad omnia pro re publica audenda [10.29.5]). The enemy, on the other hand, are said to “grow sluggish” (torpere [10.29.2]) as a result of the charge of the younger Decius. This sluggishness often appears among troops exposed to imperium that is flawed or ineffectual, or opposed by a conquering enemy, as for example the Romans before the disaster of the Caudine Forks: sistunt inde gradum sine ullius imperio stuporque onmium animos ac velut torpor quidam insolitus membra tenet (9.2.10).

31. Wagenvoort 1947: 122, n. 2, suggests a connection between the physical amplification of the devotus through contact and the large size of the statue that must be buried as a piaculum if the devotus survives the battle (8.10.12).

32. Versnel 1981: 149.

33. The soldier’s description of the wolf as integer and intactus is also interesting in this regard. Although the Romans will not be “untouched” in the battle of Sentinum (cf. 10.29.18), the same adjectives are used repeatedly to describe the “restoration” of the Roman troops after the devotio of the first Decius at the battle of the Veseris (cf. 8.10.4–6).

34. 8.10.4: memores consulis pro vestra victoria morte occubantis. Cf. 10.29.19–20.

35. 7.10.2–4 and 7.26.2.

36. For the recovery of the bodies of the devoti and their funeral, which is in each case conducted by the devotus’s partner in imperium, see 8.10.10 and 10.29.19–20; for the praise and celebration with which the victorious champion is received by his fellows, cf. 7.10.12 and 7.26.10.

37. 7.26.8: adeo duorum militum eventum, inter quos pugnatum erat, utraque acies animis praeceperat.

38. Aul. Gell. NA 9.13.7–19=Claudius Quadrigarius fr. 10b Peter. For the argument that Quadrigarius is in fact Livy’s source for the duel, despite the differences in his treatment, see Luce 1977: 224–27.

39. Aul. Gell. NA 9.11=Claudius Quadrigarius fr. 12 Peter.

40. See Holford-Strevens 1990: 179 ff.

41. In spite of Gellius’s assertion (NA 9.11.2) that the story is told in libris annalibus, the way the quoted fragment begins suggests that its immediate source was not an annalistic text but a collection of exempla. The beginning of the true excerpt from Quadrigarius on the duel of Torquatus, cum interim (9.13.7), shows that it forms part of a continuous narrative of the battle. It could fit in precisely at the point where 7.9.8 begins in Livy’s version. The Corvus excerpt however was not taken from a narrative of the events in the ager Pomptinus, which are summarized in the second sentence of the selection (9.11.4). The consular dating is included; this would be unnecessary in an annalistic source, where that information would have been given at the beginning of the account of that year. The link to what preceded the excerpt in its source is provided by the phrase adulescens tali genere editus. This would seem to derive either from some parenthetical description of the deeds of famous noble men or more probably from a collection of narratives arranged by subject (e.g., “How Noble Families Won Their Names”; indeed, Wiseman 1986: 98, with n. 58, notes that one later collector of exempla, Aurelius Victor, had a special interest in the origins of noble cognomina). Talis can perform a similar linking function in the work of Valerius Maximus (cf. 3.3.4: talis patientiae aemulus Anaxarchus). Therefore for all its archaizing tendencies of language (Holford-Strevens 1990: 179, n. 10), it is possible that this account, far from being a source for Livy’s text, may have even been retailored to conform to Livy’s version.

42. A point also observed by Fries 1985: 99–100, and Walsh 1961a:71.

43. Oakley 1985: 392, provides a summary of previous scholarly opinions and acknowledges his debt to Harris 1979: 39, n. 1, as the only other scholar to draw attention to the prevalence of single combat at Rome. The opposite notion, decisively refuted by Oakley, that the institution of dueling was associated with a lack of discipline and therefore an essentially alien practice, belonging above all to Rome’s Celtic opponents, will be found, e.g., in Bayet and Bloch 1968: 109–11.

44. Oakley 1985: 397.

45. Holford-Strevens 1984: 148, suggests that the story was worked up by Valerius Antias for the purpose of glorifying the Valerii. But he does not think it was actually invented by Antias. According to Suetonius Tib. 3.2 (= Oakley ex. 9 [1985: 394]), the Livii Drusi also claimed that their cognomen derived from Drausus, the name of an enemy chief killed in battle. However, nothing in the language of the passage allows us to be certain that the killing took place in the context of a formal duel.

46. Oakley 1985: 404–7 contrasts his approach with the conclusions esp. of Neraudeau 1979: 249–58, that dueling, as a definitively un-Roman practice, was associated particularly with the impulsiveness and energy of the iuvenes, who had to be held in check by the disciplina of their elders.

47. Oakley’s interest in recovering the attitudes of the Early and Middle Republic has perhaps led him to underestimate Livy’s own role in making the duelist’s request for his commander’s permission a canonical element in duel narratives. Oakley 1985: 406, declares that this theme “is regular in the literary sources,” but of the seven passages he cites, five come from Livy himself and the other two are of a later date and quite possibly influenced by Livy’s practices. So, too, he sees Livy’s addition of such a scene to Quadrigarius’s narrative as evidence that the emphasis on discipline had become “a topos of single combat.” One could equally say that it reflects Livy’s desire to construct such a topos.

48. Ampel. 22.4 (= Oakley ex. 24 [1985: 396]). Münzer 1931 suggests that this L. Opimius was the son of the consul in 121 B.C.E. and also that he fought the duel in order to restore the reputation of his family.

49. Plutarch Sert. 13.3–4 (= Oakley ex. 26 [1985: 396]). Plutarch’s defense of this refusal, based on Theophrastus’s dictum that a general should die like a general and not a foot soldier, represents his own opinion, not Metellus’s.

50. Front. Strat. 4.7.5 (= Oakley ex. 23, [1985: 396]): C. Marius Teutono provocanti eum et postulanti ut prodiret, respondit, si cupidus mortis esset, laqueo posse eum vitam finire; cum deinde instaret, gladiatorem contemptae staturae et prope exactae aetatis obiecit ei dixitque, si eum superasset, cum victore congressurum. However, Marius too was subject to radically different interpretations. Thus when Plutarch in the early part of his biography celebrates Marius’s rustic origins and essential valor in a manner reminiscent of the glorification of the early heroes of Rome, he notes that Marius himself defeated an enemy in single combat at the battle of Numantia. This brave action contrasts with the general corruption and laxity of the rest of the army. See Plut. Mar. 3.1–2 (= Oakley ex. 22 [1985: 396]).

51. Jos. Bell. Iud. 6.168–76, cited by Holford-Strevens 1984: 148; see Oakley 1985: 410.

52. Bell. Hisp. 25.3–5 (= Oakley ex. 30 [1985: 396]).

53. Even if it is an accident of transmission, it should be born in mind that with the exception of the excerpts preserved in Gellius, one of which may well be most immediately derived from an imperial compilation, all of our accounts of duels date from after the end of single combat as a Roman military institution.

54. See also Fries 1985: 17–18, with bibliography.

55. Nicolaus of Damascus, FGrH 90 F 111 = Stobaeus Ecl. 3.10. 69.

56. Gernet 1981: 216–39. The source for the vindicatio episode is Gaius Inst. 4.16.

57. Gernet (ibid.) here refers to Greek law, but the oath was also a common and powerful feature of Roman legal practice. The Roman oath, which is much more specifically linked to the power of contactus, whether expressed by touching an altar or by raising the hands to heaven, made the link between god and swearer all the more explicit. See Wagenvoort 1947: 50 ff.

58. If Holford-Strevens 1984: 148, is correct in this suggestion, then it is possible that Livy was in fact the first historian to include accounts of both duels in his narrative. The elements of the story itself however may have a much earlier provenance. Bayet and Bloch 1968: 114 ff., have suggested that the legend of the fighting raven in fact betrays the influence of Celtic beliefs in a battle goddess who could appear in this form. For more on the history and development of the narrative, see Köves-Zulauf 1984 and Fries 1985: 146–51.

59. For the contrary interpretation, see Fries 1985: 149: “Der zeitlich geringe Abstand, die Ähnlichkeit der Situation der Torquatus- und Corvus-Episode stellen Livius vor die Aufgabe, den Eindruck einer Doublette zu vermeiden und die Besonderheit der jeweiligen Vorgänge zu betonen.”

60. Naturally, of course, Torquatus’s action was itself inspired by the earlier deed of his father, and Corvus’s action is also meant to be an exemplum for others.

61. The stylistic comparison between Livy and Quadrigarius has been undertaken countless times. My reading owes most to von Albrecht 1989: 86–102, esp. pp. 90–92, who provides a full bibliography. See also Fries 1985: 99–105. Livy’s portrayal of the Torquatus duel as spectacle is explicitly noted by Borzsák 1973: 59–60. For an analysis of the duel as a self-contained “dramatic” incident see Pauw 1991: 36.

In light of the attention that has been focused on the visual interest of Livy’s version in contrast to Quadrigarius’s, it is interesting to note that it is precisely the quality of visual realism that is responsible for Gellius’s interest in Quadrigarius’s narrative. “The philosopher Favorinus used to say that when he read this passage, his mind was shaken and affected by emotions no less than he would have experienced if he himself had actually seen their combat [quam si ipse coram depugnantes eos spectaret]” (Aul. Gell. NA 9.13.5). See Borszák 1973: 60.

62. 7.9.7: Pons in medio erat.

63. Quadrigarius’s Gaul is naked except for his torque.

64. 7.10.6; cf. aestimantibus.

65. 7.10.6: et duo in medio armati spectaculi magis more quam lege belli destituuntur.

66. 7.10.10: totoque corpori interior periculo volneris factus insinuasset se inter corpus armaque. The phrase inter corpus armaque and the colorful verb insinuasset represent Livy’s additions to Quadrigarius’s description.

67. Heinze 1933: 101 f., sees Torquatus’s refusal to cut off his opponent’s head, as he does in Quadrigarius’s version, as a sign of Livy’s attempt to soften the cruelty of the earlier version. But while Livy does not draw attention to this act, the Celtic torque was not a simple necklace but a heavy band of twisted metal, which would perhaps have been most expeditiously removed through decapitation.

68. 7.10.8: pectus animorum iraeque tacitae plenum.

69. 7.10.3: volo ego illi beluae ostendere…me ex ea familia ortum quae Gallorum agmen ex rupe Tarpeia deiecit. The verb ostendere responds to the Gaul’s own use of ostendat, “to show which nation is best in war” (7.9.8). The entire career of Torquatus illustrates the process by which family loyalty is harnessed to and eventually superseded by patriotism. We first meet him as a young man whom his father has banished from the city because of his slowness of speech. When a tribune attempts to use this mistreatment against the father, the young man roughly threatens him and forces him to swear an oath at knifepoint to withdraw the accusations—as Livy says, a deed praiseworthy for its piety, even if hardly a model of civic behavior (7.4–5). Yet this is the same man who in the interest of preserving state order will be prepared to sacrifice his own son.

70. 7.10.4: macte virtute ac pietate in patrem patriamque, T. Manli, esto. perge et nomen Romanum invictum, iuvantibus dis, praesta.

71. As such the phrase often formed part of a prayer and was addressed to the god who was “enriched” by sacrifice. See Wagenvoort 1947: 46, n. 3, who defines mactare as meaning both “to strengthen (the gods) by sacrifice” and “to strengthen a sacrifice for the gods,” and Fowler 1911: 182–83.

72. 7.10.12: Romani alacres ab statione…progressi. Cf. 7.10.2: T. Manlius…ex statione ad dictatorem pergit.

73. 7.26.4. Later it is a prodigium (7.26.5).

74. 7.26.1: magnitudine atque armis insignis. Cf. the similar transfer of the adjective insignis from the conspicuous Samnites to the Roman commander who defeats them in Livy’s acount of the battle of Aquilonia discussed in ch. 2, sec. I.

75. 7.26.7: insignem spoliis tribunum.

76. 7.26.7: laetum militem victoria tribuni, laetum tam praesentibus ac secundis dis.

77. Ibid.: hunc imitare.

78. Praef. 10: inde…quod imitere capias.

79. 8.7.5–7: Aderunt (sc. exercitus consulares) in tempore,…et cum illis aderit Iuppiter ipse, foederum a vobis violatorum testis, qui plus potest polletque. Si ad Regillum Lacum ad satietatem vestram pugnavimus, hic quoque efficiemus profecto ne nimis acies vobis et conlata signa nobiscum cordi sint.

80. 8.7.7. Again we recognize the use of single combat not only to distinguish between two combatants but to establish a permanently valid assessment of the two peoples represented.

81. For the motif of the repeated throws, cf., e.g., Il. 3.346–60, 7.247–50, 22.273–90; Aen. 10.776–86; for the death of the horse, cf. Il. 8.83–86, 16.467; Aen. 10.892 (again the Mezentius-Lausus episode, which offers an interesting foil to Livy’s treatment of fathers and sons in this narrative). See also the analysis of the passage’s vividness in Pauw 1991: 37 and 47, n. 24, and Fries 1985: 154–65.

82. Rawson 1991: 582–98.

83. 8.12.1: iuventutem et tunc et omni vita deinde aversatam eum exsecratamque.

84. Lipovsky 1981: 112–15, supported by Levene 1993: 222–23, a reading that seems to me at odds with the sympathetic focus on the consul’s own internal ambivalence and anguish in the speech Livy composes for him.

85. 8.7.21: spoliisque contectum iuvenis corpus.

86. Val. Max. 2.7.6. Sacrificial overtones are also present in Livy’s text, particularly in the consul’s comment his son’s death will “sanctify” the imperia of the consuls (8.7.19).