Proceedings of the Virgil Society 21 (1993) 89-110 ©1993
Hopelessly Devoted to You
Traces of the Decii in Virgil's Aeneid1
The construction of a self-consciously national epic implies the blending of two worlds and forms. Virgil, at once, both writes within the conven tions of Homeric epic form, producing characters spiritually recognis able as Homeric heroes, and introduces customs, situations and patterns of action which his audience can define as distinctively Roman. Our job is less to question the strictly Roman credentials of this or that custom or action than to observe the simple fact of someone trying to construct a national identity. However, the supreme status granted to Virgil by later Roman writers, and particularly epicists, cannot fail to strike us. What is endorsed is not just the Aeneid in itself but also the meditation on previous versions of national literature which it inevitably contains. In this paper, I wish to examine some examples of the blending of Homeric epic with Roman historiography and Roman cultural experi ence. There will also be considerable room given to the mass of Greek literature intervening between Homer and Virgil. I wish to focus on the traces of devotio and of the Decii visible in the Aeneid, and to consider their presence in the light of the military ideology of the Roman state. This will finally involve offering a substantially secular interpretation of devotio, placing the custom in a much broader context than has lately been the fashion. It is surely fair to say that the anxieties produced in Turnus by the figure of Drances combine those felt by Hector when faced with Polydamas, those of Agamemnon before Thersites and those of a Roman aristocrat assaulted by the optimates' worst caricature of a novus homo, popularis politician.2 When Turnus goes out to face Aeneas on his own, the primary model for his situation, inevitably, is that of Hector facing Achilles in Iliad 22. On the other hand, Virgil appears to have plundered the historiographical tradition for an alternative model of representation
89 MATTHEW LEIGH unus pro omnibus, when, at Aeneid 11.440-2, Turnus protests to his audience that:
vobis animam hanc soceroque Latino Turnus ego, haud ulli veterum uirtute secundus, devovi.
Many scholars have claimed here to discern a reference to the heroic self-sacrifice of two or three generations of Decii, alone charging the line of the enemy in order both to die and to ensure victory for the Roman army.3 This view is supported by the repeated reference to the single- combat of Turnus as expiation or sacrifice at Aeneid 12.16, 12.229 £, 12.234-7 and 12.694 f, a theme which Philip Hardie has lately exam ined in terms of the Girardian sacrificial crisis.4 Further, there is a clear sense in which the lone resistance of Turnus does conform to the spirit of devotio. However, the presumption of such a devotio has been vigorously challenged by C. Bennet Pascal.5 Pascal observes that the references to devotio and to blood sacrifices are only ever made by characters in dramatic situations, thus do not come with the seal of approval of the authoritative primary narrator; that the devotio of the Decii in Livy is to the gods, while Turnus claims to have devoted himself to his people; finally, that Turnus hopes to survive, frequently flees death, and gener ally acts in a manner unworthy of so exalted a figure as the devotus. Pascal's emphasis on the religious aspect of the devotio is almost demanded by the Livian account of the deaths of the first two genera tions of Decian devoti.6 The first narrative contains a long description of the ritual gone through prior to the charge against the lines, and in cludes the prayer formula uttered by the general. The ritual has, of course, been very closely analysed by Roman religionists, most notably by H.S. Versnel in a 1976 Mnemosyne article and in his 1981 contribu tion to the Fondation Hardt volume on Sacrifice in Antiquity.1 He com pares the Livian devotio ducis with the devotio hostium formula found at Macrobius, Saturnalia 3.9.9, and concludes that, while the latter con forms to the proper pattern for a votum, in that it offers payment to the gods in the form of the enemy's land and goods, should victory be granted, the devotio ducis is to be understood as a consecratio in that Decius transfers himself directly from the human to the sacral sphere without waiting for his prayers to be fulfilled. Pascal, however, points to
90 HOPELESSLY DEVOTED his own discovery of epigraphic evidence for the payment of a votum in anticipation of the success of the prayer,8 and therefore argues that the Decian self-devotion is "not a self-curse nor an unquestioning squander ing of life nor an assumption of collective guilt nor a victory-death wager with the gods or enemy, but a straight 'quid pro quo' bargain".9 He is also very concerned to deny the element of the scapegoat and the magical10 in the devotio ducis, an attitude for which there seems little justification when Livy can describe his hero as apiaculum. Pascal, it seems, is as much concerned to define the exact religious character of the Livian devotio as to elucidate the text of Virgil. As we have seen, his definition of the devotio ducis is, in itself, rather reductive, but even more so is a critical procedure which demands the simple transference of a strict analysis of explicit devotio ritual in Livy to the shadow of a devotio in Virgil. This attitude leads him, for instance, to contrast "the citizen-soldier's certainty of his proper relations with the gods" and "the reckless Rutulian's invocation of an uncertain fate", and to conclude that, since the devotio of Turnus is clearly not an authentic devotio of the sort performed by the Decii, it cannot properly be de scribed as such except in the self-serving rhetoric of Turnus himself.11 It is to this line of argument that I wish to respond. Further, I wish to respond in two different ways. First, I wish to examine the devotio ducis as ritual, and to suggest that, while the Aeneid never gives us a formal reproduction of the ritual outlined in the Livian narrative, elements of that ritual are activated in the Aeneid. Second, and perhaps more importantly, I wish to argue that the constant narrowing down of the terms of reference under which we can consider the meaning of devotio, only exacerbates the real problem. This, I shall assert, lies in our failure to broaden our field of reference sufficiently to locate the specifically religious devotio of the Decii within a range of other more secular ideological factors. Both my points can be illustrated with reference to the texts of Livy and of Virgil. It is a commonplace to assert that the repetition of whole or half-lines within the Aeneid cannot be explained in terms of the compositional techniques of the Homeric oral poet, nor even entirely as a simple imitation of the sound and style of Homeric epic. We expect a repetition in Virgil to suggest some relationship between the two passages, we expect the repetition to demand our interpretation. However, one re peated formula has hitherto received only the slightest attention, and by
91 MATTHEW LEIGH its neglect we have missed something very interesting. I refer to the repetition of the death-rush of Coroebus at 2.408 f.:
non tulit hanc speciem furiata mente Coroebus et sese medium iniecit periturus in agmen, with that of Priam at 2.509-11:
arma diu senior desueta trementibus aevo circumdat nequiquam umeris et inutile ferrum cingitur ac densos fertur moriturus in hostis,12 with the resolution of Aeneas at 2.353 f.:
... moriamur et in media arma ruamus. una salus victis nullam sperare salutem, with the dilemma of Nisus at 9.399 £f:
quid faciat? Qua vi iuvenem, quibus audeat armis eripere? An sese medios moriturus in enses inferat et pulchram properet per vulnera mortem, and its resolution at 9.438, when:
at Nisus ruit in medios, slays Volcens and embraces his partner in death. Further, this is the same solution as is taken by Helenor, who is compared to a lion leaping over the huntsmen's spears, and then, Aen. 9.554 f.:
...haut aliter iuvenis medios moriturus in hostis inruit et qua tela videt densissima tendit
Again, we see something similar in the upbraiding of Lausus by Aeneas at 10.811 f.:
quo moriture ruis maioraque viribus audes?
92 HOPELESSLY DEVOTED
fallit te incautum pietas tua
This is taken up in the death-rush of Mezentius at 10.867-70:
dixit, et exceptus tergo consueta locavit membra manusque ambas iaculis oneravit acutis, aere caput fulgens cristaque hirsutus equina. sic cursum in medios rapidus dedit, and in the angry speech of the king at 10.878 f£:
...quid me erepto, saevissime, nato terres? Haec via sola fuit qua perdere posses: nee mortem horremus nee divum parcimus ulli. desine, nam venio moriturus et haec tibi porto dona prius.
Finally, at 11.741 ff., Tarchon finishes his paraceleusis to the troops and leads the fightback against Camilla:
haec effatus equum in medios, moriturus et ipse, concitat et Venulo adversum se turbidus infert, dereptumque ab equo dextra complectitur hostem et gremium ante suum multa vi concitus aufert.
Various remarks should be made regarding the death-rush formula exemplified above. The first is that it has no place in the world of the Iliadic warrior, who is not accustomed to charge the line of the enemy with the deliberate intention of getting killed. Moreover, it is this inten tional pursuit of death which is noted when the passages are discussed, Conington, Quinn, Austin and Mackail all emphasising this aspect of the future participle moriturus, and observing that Priam is killed only later and Tarchon not at all.13 However, it has not yet been stressed to what extent the insistence on the intentional pursuit of death marks these passages as radically different from Homer. Nor has anyone
93 MATTHEW LEIGH discussed the manner in which this intentionality distinguishes these lines from other passages in the Aeneid in which a hero charges the enemy lines, rushes to battle and so on.14 The second observation is that the formula does recall the Livian account of the actions of both the Decii subsequent to their vows and ritual. Thus, of P. Decius Mus, Livy records that Ipse incinctus cinctu Gabino, armatus in equum insiluit ac se in medios hostes immisit, conspectus ab utraque acie, aliquanto augustior humano visu, sicut caelo missus piaculum omnis deorum irae, qui pestem ab suis aversam in hostes ferret,15 and of his son that ...haec exsecratus in se hostesque, qua confertissimam cernebat Gallorum aciem concitat equum inferensque se ipse infestis telis est interfectus.16 Further, it becomes apparent at Livy 9.4.10 f. that the formula is held to be the essential characteristic of the Decian devotio ducis, when the brave lieutenant comments: Equidem mortem pro patria praeclaram esse fateor et me vel devovere pro populo legionibusque vel in medios immittere hostes paratus sum; sed hie patriam video, hie quidquid Romanarum legionum est, quae nisi pro se ipsis ad mortem ruere volunt, quid habent quod morte sua servent?11 The third observation is that the apparent formula for the death-rush to the enemy considerably ante-dates either Virgil or Livy in that it appears in Rhetorica ad Herennium 4.57, which, giving a specimen tractatio of the theme Sapiens nullum pro re publica periculum vitabit, cites Decius as an exemplum: Quod mihi bene videtur Decius intellexisse, qui se devovisse dicitur et pro legionibus in hostes immisisse medios. Similarly, at Tusculan Disputations 1.48, Cicero, discussing Greek devoti, describes Codrus as qui se in medios immisit hostes.18 Further, the formula figures once again in Sallust's description of the death of Catiline, a model of ferocia, not pietas, but somewhat redeemed by the manner of his going, Catilina postquam fusas copias seque cum paucis relictum videt, memor generis atque pristinae suae dignitatis in confertissumos hostis incurrit, ibique pugnans confoditur.19 This relationship of this passage to the Livian account of the Decii, it will be noted, is recognised by Vretska.20 It will also be noted that the death-rush formula which we have isolated is not regarded by Versnel as a simple form of words, but as an essential component of the act, in that it embodies the voluntary and ardent aspect of the sacrifice.21 The fourth observation which we would make is that Coroebus, Priam, Aeneas, Nisus, Helenor, Lausus, Mezentius and Tarchon all seek to
94 HOPELESSLY DEVOTED sacrifice themselves out of a sense of responsibility to others. Priam and Aeneas are the leaders of Troy, and their actions are a direct response to the destruction of their city.22 In this, they conform to the conditions identified by Versnel23 as necessary for patriotic self-sacrifice, namely "an all-pervading crisis challenging the continued existence of society as a whole". Further, they themselves fit perfectly the status of the actor required, that is the leader or best thing possessed by the endangered society,24 as do Mezentius and Tarchon in their capacity as generals and kings. However, we will also note that the substitutional character of the devotio is frequently transferred to the private sphere, a wife, for instance, dying that her husband may live, a phenomenon which Versnel catalogues at length.25 It is in this sense that we can best understand the sacrifice offered by Coroebus for Cassandra,26 Lausus for Mezentius or by Nisus for Euryalus. This is surely also the sense in which we can understand the sacrifice of the older Helenor, whose heroic death-rush to the enemy briefly allows the younger Lycus to slip out through their ranks. Further, it is perfectly in the spirit of Mezentius the contemptor divum that he should insist on repaying his son's sacrifice with his own, while disavowing any divine reward for his deed (nee divum parcimus ulli)?1 My fifth observation, the last regarding the specifically ritualistic aspect, is that the depiction of Mezentius and Tarchon charging to their deaths on their warhorses reflects the sacrifice of Decius in an important manner. Livy 8.9.9 records that ipse ineinctus cinctu Gabino armatus in equum insiluit ac se in medios hostis inmisit. Versnel here observes acutely that "A general is a 'man on horseback' as nineteenth century imagery shows and the same appears to be true for royalty and heads of state. No wonder that the way Decius seeks his death is totally ignored in modern literature. Yet, on the other hand, it receives ample emphasis in the ancient texts, where the horse is rarely lacking, surprisingly enough if its presence were so self-evident."28 Both the Decii in Livy ride to their deaths, so, as we shall see later, does the parallel figure of Hasdrubal, while in Lucan the botched devotio of Pompey sees the general ride away from his death on horseback.29 Why? Versnel relates the use of the horse to the equally famous devotio of Curtius at Livy 7.6.5, who equo...exornato insidentem armatum se in specum inmisisse, and notes other examples of sacrifice or execution involving the simulta neous death of the horse. These tend to follow the Curtius theme of
95 MATTHEW LEIGH direct dispatch to the gods below, for instance by the dropping of the traitor Philumenus and his horse down a well at Livy 27.16.4. In the absence of the chance to send one's self down, se inmittere in medios hostis replicates on the horizontal plane the direct self-consecration of the devotus to the gods below.30 For Mezentius and Tarchon, determined to die, this interpretation is clearly appropriate. So, however, is the second reason given by Versnel for the presence of the horse, namely "the element of furor, ecstasy and rapture realised by means of the uncheckable impulse of the frantic animal". This is clear from the case of the second Decius, who concitat equum,31 just as do the first in Valerius (concitato equo),32 Hasdrubal (concitato equo),33 and, in his own special way, Pompey (concitus... sonipes)3i This particular turn of phrase is not present with reference to Mezentius, but, once on horseback, Sic cursum in medios rapidus dedit. Further, it is in just these terms (equum...concitat)35 that Tarchon is described, and nowhere else in the Aeneid is concitare used in relation to a horse.36 My analysis should also cast some new light on the position of Rhaebus at this point in the Aeneid. We have grown accustomed to see the speech to Rhaebus as a simply Homeric element, recalling the conversation of Achilles with Xanthus at Iliad 19.400 ff., the words of Zeus to the horses of Achilles at Iliad 17.437 ff., and those of Polyphemus to his favourite ram at Odyssey 9.447 ff.37 Comparing Rhaebus to Achilles and Xanthus, that is reading him as a purely epic animal, we see rather less of the pathetic fallacy in that Rhaebus, though addressed does not reply. However, if we read him as the traditional steed of the historians' devotio narrative, if we stress the death-rush formula at 10.867-70, we see a definite Steigerung des Pathos, as the animal of history becomes the good companion of epic. I find this process rather touching.
We have left Turnus and his devotio somewhat in mid-air. The argu ment against him was that he never came close to replicating the ritual to which he referred, and, indeed, that much of his backsliding in the eleventh and twelfth books was entirely alien to the spirit of the devotus. My approach has been to take a less 'all-or-nothing' attitude to devotio than Pascal, and to examine how one central element of the devotio
96 HOPELESSLY DEVOTED narrative, the 'death-rush formula', recurred repeatedly at moments of personal sacrifice for state, son or friend in the Aeneid. The question now is whether we can identify anything similar in the actions of Turnus. I believe that we can. The passage to which I refer is that at which Turnus finally accepts his fate and goes out to meet Aeneas for the last time. He has allowed himself to be led away by Juturna, but now Saces comes up (Aen. 12.649 f.), as it happens on a hard-pressed horse (equo spumante) and at a rush (ruit). He now fulfils one of the vital conditions for a devotio by informing Turnus that he is needed in order to avert the final disaster from the Latins, Aen. 12.653-6:
Turne, in te suprema salus: miserere tuorum. fulminat Aeneas armis, summasque minatur deiecturum arces Italum excidioque daturum; iamque faces ad tecta volant.
Turnus listens to the further elaboration of Latium's woes, looks back at the city, and addresses Juturna, Aen. 12.676 ff.:
iam iam fata, soror, superant; absiste morari; quo deus et quo dura vocat Fortuna, sequamur. stat conferre manum Aeneae, stat quidquid acerbi est morte pati; neque me indecorem, germana, videbis amplius. Hunc, oro, sine me furere ante furorem.
Turnus has now fulfilled two further conditions. He is fully aware of the inevitability of his death, a fact he has already acknowledged in a previous speech to Juturna at 12.646 £, and he is giving way to a state of ecstasy. Crucial now are the actions which follow, Aen. 12.681-3:
dixit, et e curru saltum dedit ocius arvis, perque hostes, per tela ruit maestamque sororem deserit ac rapido cursu media agmina rumpit.
Turnus does not travel by horse. He cannot properly do so, for he now faces single combat on foot with Aeneas. On the other hand, he does travel at an enormous rush. Moreover, this rush is not just the frenzied
97 MATTHEW LEIGH activity of a warrior-hero, such as that of Mezentius at 10.729, for, like those cited above, this is explicitly a rush to death. Finally, we will note that Turnus does not rush into the enemy ranks so much as through them. This paradox does not so much deny the devotio as hyperbolise it, pointing to the existence of an even greater challenge on the other side. Turnus makes clear at 12.694 f. that he knows what that challenge is, and that this time he will not shirk it:
...me verius unum pro vobis foedus luere et decernere ferro.
Turnus will meet the challenge, will meet Aeneas, and will lose.
C*^fe/1
I have so far examined the traces of the religious ritual of devotio ducis in the Aeneid. In doing so, I have conformed to the meaning which religionists such as Pascal and Versnel would apply to such a narrative. I have not attempted to suggest that there is any one pedigree, pure- breed devotio in the Aeneid, but I have demonstrated that aspects of the ritual meaning are present in a series of connected narratives. The foreshadowing role of the minor devotiones for the major devotio of Turnus should be clear already, but I will make my position absolutely clear in my final remarks. For now, however, I wish to return to the second initial task which I set myself, and circumvent the tendency to solely religious readings of devotio narrative by locating the ritual itself within much broader political and ideological pressures. In doing so, I will again wish to lean heavily on the centrality to devotio narrative of the formula for the rush to death. I will again conduct my argument in the form of a series of observations. My first observation is that traces of the devotio narrative regularly appear in episodes where there is no apparent religious content. This was demonstrated earlier when I cited the case of Sallust's Catiline, a figure who acts out of a sense of family dignity, and who appeals to the virtus of his troops. Another example is furnished by Livy himself in his account of the death of Hasdrubal, who ne superstes tanto exercitui suum nomen secuto esset, concitato equo se in cohortem Romanam inmisit. Ibi,
98 HOPELESSLY DEVOTED ut patre Hamilcare et Hannibale fratre dignum erat, pugnans cecidit.3S In this, he shows the anxiety of the good general, suum nomen secuto recalling the military oath of loyalty from soldier to general, an oath which itself placed certain implicit obligations on the commander.39 Similarly, this use of the honourable practice of the Decii by the enemies of Rome is further paralleled in the death of Tacfarinas in Tacitus, who ...ruendo in tela captivitatem haud inulta morte effugit.40 His motiva tion would seem, like that of Horace's Cleopatra, to be the fear of slavery. My second observation would be that what is true of Livy, Sallust and Tacitus, is equally true of Virgil. For instance, the resolve of Nisus to die by the death-rush coheres with a second group of deaths in Virgil, those characterised by the concept of mors pulchra and of the mors pulchra per vulnera. Thus, Aen. 9.400 f.:
...An sese medios moriturus in enses inferat et pulchram properet per vulnera mortem.
Similarly, when Aeneas resolves on the death-rush at 2.353 f., we should note the evidence provided by 2.317 that his mental state is very similar to that of Nisus: pulchrum...mori succurrit in armis. Moreover, both these references clearly stand together with the description of the bees at G. 4.217 f., who ...corpora hello I obiectant pulchramque petunt per vulnera mortem, the latter phrase repeated exactly at Aen. 11.647. Pulchra mors is the subject of a brief article by Alfonsi41 and of a short discussion in Raabe.42 Further useful citations are offered by Paolo Serra Zanetti in his article on 'Morte' in the Enciclopedia Virgiliana. For the early Greek evidence, there are also more substantial pieces by Muller, Loraux and Vernant. All of these emphasize the fairness of a death suffered for one's country, a notion first propounded by Hector at Iliad 15.494-9 (oi) ol detK^s* a\ivvo\x€vq) uepl TrdTpris' / reQva\xev), given far greater prominence in the archaic poets (Alcaeus 400, Tyrtaeus fr. 10 West, Kallinos 1.6 f.), and famously translated at Horace, Odes 3.2.13 (dulce et decorum est pro patria mori). However, less emphasis is placed on another Homeric conception, that which argues that a wound to the front is beautiful and honourable, one to the back the mark of coward ice.45 This idea, which first appears at Iliad 8.93 f. in the angry challenge of Diomedes to Odysseus46 and in the praise of Meriones by Idomeneus
99 MATTHEW LEIGH at 13.275 ff.47 takes on a far greater prominence in the hoplite poetry of Tyrtaeus.48 In Rome, the opposition of the front and the back, of the pectus and the tergum becomes a commonplace. One's enemies were one's adversi, a good wound would be that received facing the opposition, a bad one suffered in flight. The standard terms for retreat are terga uertere or tergum dare. This opposition of pectus and tergum becomes a recurrent theme of epic, largely, I maintain, because it reflected the vivid realities of the legionary battle-formation. It is also prominent in one eccentric feature of Roman society, that is the public enumeration of scars and wounds to the front of the body. I do not here have time or space fully to state the evidence for this as a particular characteristic of Rome—I have set it out in full elsewhere49—but would like merely to assert that this is the full significance of the mors pulchra per uulnera. Anyone inclined entirely to disbelieve this assertion might note that at Ovid, Metamorphoses 13.262 ff., Odysseus, cruelly exploiting the legen dary invulnerability of Ajax, tears off his shirt and protests:
...sunt et mihi vulnera, cives, ipso pulchra loco..., while, at Silius Italicus, Punica 5.594 f., Hannibal expresses his admira tion for the dead Sychaeus by observing:
cerno.. .adverso pulchrum sub pectore vulnus cuspidis Iliacae.
Similarly, the third major oration of pseudo-Quintilian turns on the horror of the admirable disfigurements of the manly miles Marianus actually attracting the sexual attentions of his commanding officer, while at Elder Seneca, Controversiae 1.8.4-5, the ter fortis reminds his father that ...ex acie redeuntis vulnera osculabaris. A wound is not just pulchrum, but is pulchrum in carefully defined senses. It figures as the aspiration of Roman martial virtus and as the focus of anxiety as to one's masculinity. The concept of the mors pulchra per vulnera, meanwhile, though described in terms of the death-rush of the devotio ducis, and sharing its aspect of patriotic self-sacrifice (it is still mors), carries with it aspirations which transcend the religious rite and refer to the far broader military ideology and spirit of emulation of the society.
100 HOPELESSLY DEVOTED
My third observation would therefore be to return to the point that Priam, Aeneas, Mezentius, Tarchon and Turnus are Trojan or Italian generals. Their status is important, when we consider that the Decii were regarded as models of exemplary virtue in a general at Rome,50 while the death of Hasdrubal was given the same exemplary status in the Histories of Polybius.51 The attitude of Hasdrubal makes him agios' cmordCTeios' etvcu icai CnXou, while other generals alaxp&s \ikv e-rro'iTiaai/ T&S" firms' and Karfaoxwav 8k T&S" irp6 TO{JTOU upd^is, £Troi>eL8ioTov 8e afyioi T6V KaTa\eiTr6p.ei'oi/ eTrotriow ploy. Moreover, after completing his praise of Hasdrubal, Polybius concludes ravra \ikv dbv TIJJLXV eip^aQia TTepl TUV kv TTpdy^.aaii' a.va 101 MATTHEW LEIGH can offer a number of examples of Roman generals acting according to these instructions, and in the spirit of Alexander, the most famous of which being Catiline at Sallust B.C. 60.4: Interea Catilina cum expeditis in prima acie vorsari, laborantibus succurrere, integros pro sauciis arcessere, omnia providere, multum ipse pugnare, saepe hostem ferire: strenui militis et boni imperatoris officia simul exequebatur. Similarly, at B.G. 5.33 ff., Caesar contrasts the good general Cotta with the bad general Titurius. The latter does not foresee the disastrous consequences of his plan and is thrown into confusion when they come, his failings contrasted with the Alexander-like qualities of Cotta: At Cotta, qui cogitasset haec posse in itinere accideratque ob earn causam profectionis auctor non fuisset, nulla in re communi saluti deerat et in appellandis cohortandisque militibus imperatoris et in pugna militis officia praestabat. Further, Cotta is struck while urging on his men,56 stead fastly refuses to join Titurius in negotiating for surrender,57 and finally pugnans interficitur. His colleague is surrounded and killed as he at tempts to parley.58 More particularly still, we will want to note that various generals are recorded as having turned the course of a battle by the impact of their heroic lone charge on their troops. This is most famously the case with the Decii themselves, but also extends to the charge of the Nervii by Caesar at B.G. 2.25 and to that of the Fabii at Livy 2.46 ff. Most importantly, a series of such actions is recorded in book 3 of the Exempla of Valerius Maximus, and in every case the spectacular quality of the deed is emphasised.59 In this case, one is aware of the function of the audience within the narrative as a surrogate for the audience doing the reading, and of the desired replication of the response of the former in that (given time) of the latter. It is thus striking that Virgil's Tarchon should go nomine quemque vocans {Aen. 11.731), and that Virgil should stress the sight provided (Aen. 11.745 f., ...cunctique Latini I convertere oculos) and the exemplary impact of his deed {Aen. 11.758 f., ...ducis exemplum eventumque secuti I Maeonidae incurrunt).60 My fifth and final observation is that we must maintain a sense of genuine historical experience at the back of all of this. When generals live by or are described according to tags culled from Homer and propa gated by the scarcely less mythopoeic Alexander, our categories of fic tional epic and actual real life (as communicated by the historians) become rather blurred.61 The tendency of Livy to Homericise his narrative 102 HOPELESSLY DEVOTED only renders our categories more unclear, and many of the injunctions placed on the Roman general are prefigured in Hellenistic military manuals. On the other hand, we should hold on to the observation that Homeric heroes did not engage in the death-rush exemplified above, and should be confident in asserting that certain Roman social tensions are reflected in the presence of this formula in the Aeneid. Moreover, a clearer idea of those tensions is granted us by an illuminating new study, Nathan Rosenstein's Imperatores Victi,62 which analyses the re sources open to defeated Roman Republican generals to overcome the shame of military reversal and survive in aristocratic political competi tion. The fourth chapter of this work examines the importance of demon strating nobility even amidst defeat. If a general did what he could to hold the line and exposed himself to the risks which his troops were fleeing, his virtus, even in defeat, could win high praise. The actions of the Decii may now be seen as the extreme response to social pressures which every defeated general faced. To quote Rosenstein, "...A general was expected to display courage and self-control when things were falling apart around him—a willingness to fight hard, take risks, and, if necessary, meet death fighting bravely... Generals died in battle fre quently enough during the last two and a half centuries of the Republic to indicate that crisis often led commanders to put themselves in harm's way, and those who survived took pains to advertise the fact."63 c\»S*^ To close, therefore, when Turnus speaks of his devotio, it is in response to the accusation of Drances that the people of Latium are all dying for him. He now reverses the position, referring his situation to the arche typal self-sacrifice of the general for his men. When he speaks of a devotio, he does not so much imply the exact replication of the religious practices of Livy's Decii, as compare the pressures on him to those felt by them and by generations of other Roman generals. It is appropriate that the assault on the virtus of Turnus should come from the caricature popularis, Drances, their meeting thus offering us the shadow of a dispute which had marked much of the political conflict of the last century of the Republic.64 It is the measure of Drances' despicability that, unlike a Marius, he has no virtus of his own, no deeds, only words 103 MATTHEW LEIGH on which to rely. It is a mark of the good Italian in Turnus that, debating with Drances, he should lament the failure of virtus on his side, Aen. 11.415: quamquam o, si solitae quicquam virtutis adesset! and that it should be with his mind firmly on virtus that he speaks of devotio at 11.440-4: ...vobis animam hanc soceroque Latino Turnus ego, haud ulli veterum virtute secundus, devovi. 'solum Aeneas vocat:' et vocet oro; nee Drances potius, sive est haec ira deorum, morte luat, sive est virtus et gloria, tollat. Nor should we be surprised that his reactions to Saces at 12.666 ff. should centre on the same quality: .. .aestuat ingens uno in corde pudor mixtoque insania luctu et furiis agitatus amor et conscia virtus. Turnus faces his own epic High Noon. He does not face it with complete equanimity or freedom from fear, but his actions in the twelfth book would be less sympathetic, and, for me, considerably less heroic, had he not that very fear to conquer. The use of the death-rush formula, itself eminently Decian, offers us a succession of prototypes, foreshadowings, for the pressures to which Turnus will succumb. The response of Priam and of Nisus is both noble for its utter propriety and pathetic for its impropriety: the age of the former, the youth of the latter, each highlights the inequality of the burden shouldered. The response of Aeneas, repeatedly expressed, and, of necessity, repeatedly repressed, is both eminently proper to his relation to Troy, to his age and vigour, and yet improper because at odds with the longer plan which he cannot possibly perceive. Turnus enjoys no such dispensation. When Virgil allows Turnus to describe his terrible responsibility in terms of devotio, we should not dismiss this as just his self-serving rhetoric. Fundamentally, we should not do so because he is too 104 HOPELESSLY DEVOTED sympathetic a character to be dismissed out of hand as a braggart or trickster. What we see him doing in giving his own inevitably inexact and yet substantial sense to the devotio is not so strange. It is what successions of Roman writers did when they described a heroic self- sacrifice with the formula for the rush to death. When they did so, they understood the devotio ducis less as a ritual, a votum with payment in anticipation of success, than as the ultimate expression of a military code which expected courage from a general, and heroism in defeat. Virgil is striking because he seems to understand the devotio as ritual, and occasionally to involve elements from that sphere, but also to be able to relate it to a range of secular military values: the responsibility of the general to his men; the pursuit of the mors pulchra per vulnera; the challenge to the aristocrat's virtus. These are only some of the obses sions which form the broad field of Roman military aspiration and expectation. It is within this field that we should locate the devotio ducis and assess the traces which it leaves throughout Latin epic. It cannot help the study of Roman religion if we refuse to recognise its phenomena save in their purest form. St Hugh's College, Oxford MATTHEW LEIGH NOTES 1. Versions of this paper were given to the Virgil Society in London and at the Department of Classics of Yale University. I would like to thank both my audiences for their very helpful comments, also Don Fowler, Oliver Lyne and Michael Clarke, who read early drafts and made many useful suggestions. 2. Best argued in A. La Penna, 'Spunti Sociologici Per L'Interpretazione dell' Eneide in Virgiliana', in Recherches sur Virgile, ed. H. Bardon and R. Verdiere (Leiden 1971) 283-93, reprinted in Fra Teatro, Poesia e Politica Romana (1979) 153-65. For a strictly Homeric reading, which, however, makes excellent use of the confrontation of Paris by Hector and Antenor, see Paul F. Burke Jr., 'Drances Infensus: A Study in Vergilian Character Portrayal', TAPhA 108 (1978) 15-20. 3. For a survey of such opinions, see C. Bennet Pascal, The Dubious Devotio of Turnus', TAPhA 120 (1990) 251-68. To this, one should add W.R. Johnson, Darkness Visible (Berkeley, Los Angeles, London 1976) 117-19. 4. See P. Hardie, The Epic Successors of Virgil (Cambridge 1992) Chap. 2, 'Sacrifice and Substitution', 19-56. 5. Pascal [n. 3 above]. 6. For the first two Decii, see Livy 8.8.19—8.11.1 and 10.28.12—10.30.10. For the 105 MATTHEW LEIGH third Decius, see Cicero De Fin. 2.60 ff. and Tusc. 1.89. A full list of references to all three Decii is offered at 46 ff. of H.W. Litchfield, 'National Exempla Virtutis in Roman Literature', HSCP 25, 1-74, 7. See H.S. Versnel, Two Types of Roman Devotio', Mnem. s. 4,29 (1976) 365-40, and H.S. Versnel, 'Self-Sacrifice, Compensation, Anonymous Gods', Entretiens de la Fondation Hardt 27 (1981) 135-94. Useful comparative material for the Decii is also to be found in W. Burkert, Structure and History in Greek Mythology and Ritual (Berkeley, Los Angeles, London 1979) 59-77. 8. C.B. Pascal, 'Do ut Des', Epigraphica 14 (1982) 12-13 and 15. 9. Pascal (1990) 265. I would say of this view that the sense of a bargain is greatly enforced in the case of P. Decius Mus by his assurance in a dream that Rome's success depends on his sacrifice. However, it will be seen from the evidence cited here that non- Livian accounts of this episode did not always stress the assurance-aspect, and that such confidence is far from typical of the broader field of devotio sacrifice in which his deeds should be located. 10. Pascal (1990) 260 n. 29 offers W.W. Fowler, The Religious Experience of the Roman People (London 1911) 208, and G. Stubler, 'Die Religiositat des Livius', Tiibinger Beitrage zur Altertumswissenschaft 35 (1941) 175 and 181. 11. Pascal (1990) 252: "The clinching argument against the notion of a devotio on the part of Turnus is that neither his hanc animam devovi nor the circumstances of his death conform to the formal contractual terms of a Republican devotio." 12 . The comment of Donatus ad loc. is enlightening: Visum est ei arma capere, non ut peteret pugnam aut certando victoriam quaereret, sed ut in armis positus pro integro et forti procumberet. 13. Conington and Mackail make the point ad 11.741, Austin ad 2.408, while only K. Quinn, Virgil's Aeneid, A Critical Description, London (1968) 12 n. 1 cites the entire list. He does not take the point any farther. Conington's note on 11.741 reports discussion as to whether the et ipse moriturus of Tarchon indicates an intention or willingness to die as well as those he slays (Servius), as well as his men, not just exhorting, but also setting the example, or as well as those who have fallen already. As will be seen from this article, I, like Conington, have some sympathy with the second view, though I also regard et ipse moriturus as characterising Tarchon intratextually by reference to the other figures cited above, and intertextually, by reference to the Decii. 14. e.g. Virgil Aen. 1.476, 9.369 ff., 9.799 f., 10.20-2, 10.379, 10.575 f., 10.729, 12.477 f., 12.497 ff. 15. Livy 8.9.9 ff. cf. Valerius Maximus 5.6.5-6: P. Decius Mus, qui consulatum in familiam suam primus intulit, cum Latino bello Romanam aciem inclinatam et paene iam prostratam videret, caput suumpro salute reipublicae devovit ac protinus concitato equo in medium hostium agmen patriae salutem, sibi mortem petens inrupit factaque ingenti strage plurimis telis obrutus super corruit. Ex cuius vulneribus et sanguine insperata victoria emersit. Further instances of the formula are to be found at Orosius, Hist. 3.9.3, in confertissimos hostis sponte prolapsus occubuit and at Seneca, E.M. 67.9 (referring to the second Decius) in aciem confertissimam incurrit, de hoc sollicitus tantum ut litaret. 16. Livy 10.28.12 ff. 106 HOPELESSLY DEVOTED 17. One should not be confused by the suggestion of two alternative actions via vel...vel. Weissenborn ad loc. comments that "...statt der beiden korrespondierenden vel erwartete man ein blosses et an der zweiten Stelle", observing that the second action is just the consequence of the first. 18. I owe this example to a note in David Vessey's discussion of Menoeceus, a Greek devotus celebrated in most Livian terms by Statius. See D.W.T. Vessey, 'Menoeceus in the Thebaid of Statius', CP 66 (1971) 236-43. 19. Sallust, B.C. 60.7. Here, cf. B.C. 61.2: Catilina vero longe a suis inter hostium cadavera repertus est, paululum etiam spirans ferociamque animi, quam habuerat vivos, in voltu retinens. 20. K Vretska, Comm. on Bell. Cat., ad loc. observes of this scene that "Die Erzahlung wiederholt sich in breiterem Ausmass bei Liv. 8.9.10 iiber P. Decius Mus bis in die Einzelzuge." 21. Versnel (1981) 145-8. 22. This is most clearly the case with Priam, whose actions are preceded by 2.507 f., Urbis uti captae casum convulsaque vidit I limina tectorum et medium in penetralibus hostem. 23. Versnel (1981) 143. 24. Versnel (1981) 143-5. 25. Versnel (1981) 163-7. 26. Quinn, op. eit. 11 f. notes that Coroebus has just come to Troy because infatuated with Cassandra (Aen. 2.342 f.) and dies at the altar at which she tried to take refuge (Aen. 2.424-6). 27. Cf. 10.773-6, where Mezentius talks of his right hand and his spears as his gods and impiously vows to render his son Lausus a trophy by dressing him in the arms of Aeneas. 28. Versnel (1981) 152-6. 29. For the botched devotio of Pompey as a positive action, see M. Rambaud, 'LApologie de Pompee par Lucain au Livre VII de la Pharsale', REL 33 (1955) 258-96; J. Brisset, Les lde.es Politiques de Lucain, (Paris 1964) esp. 75 f. and 114-27; E. Narducci, La Provvidenza Crudele: Lucano e la Distruzione dei Miti Augustei (Pisa 1979). However, a sense of the inadequacy of Pompe/s actions emerges at P. Hardie (1992) 54-6, and is given far stronger emphasis in my forthcoming Oxford D.Phil.,Venientia Fata, Non Transmissa: A Study of Lucan's Narrative Techniques. 30. One should here note that the verb of rushing, ruere, also has a clear sense of 'falling down'. 31. Livy 10.28.18. 32. Valerius Maximus 5.6.5-6. Here, cf. Cicero, De Fin. 2.61 equo admisso. 33. Livy 27.49.4. 34. Lucan, Pharsalia 7.677 f. 35. Virgil, Aen. 11.741 f. 36. This coincidence of terms is striking. However, there are plenty of examples of concitare for spurring a horse with no apparent devotio context. See TLL Vol.4, 65.1 do not regard it as impossible that concitare equum should have become typical of devotio narrative from an early time. 37. See J. Glenn, 'Mezentius and Polyphemus', AJP 92 (1971) 131 ff; G. Thome, 107 MATTHEW LEIGH Gestalt und Funktion des Mezentius bei Virgil (Frankfurt 1979) 139-51; S.J. Harrison, Comm. on Virgil Aeneid 10 fOxford 1991) ad loc. 38. Livy 27.49.4. 39. At Aen. 10.672, Tumus frets over the men qui me meaque arma secuti, recalling the self-description of Aeneas' men at Aen. 3.156 as nos te Dardania ineensa tuaque arma secuti. S.J. Harrison (1991) ad loc. points to Walbank on Polybius 6.21.1-3, who cites Dionysius of Halicarnassus 10.18, TOV aTpaTiomK6v 8picov, aKoXovfrfjo-eii' Tot? inrdTois'... and 11.43, 8 TE ydp Bpicosr 6 orpaTicoTLKd?, ov dTrdv-rw udXiora en-TreSoiio-i , Pwudioi, rtiis CTTpaTTiyoXs' dKoXou0eTv KeXefei TOU? aTpaTeuo|iivous J Brroi TTOT' &V dycocnv. 40. Tacitus Annals 4, 25. 41. L. Alfonsi, 'Pulchra Mors', Latomus 22 (1963) 85 f. 42. H. Raabe, Plurima Mortis Imago, Zetemata 59 (Munich 1974) 207 f. 43. Enciclopedia Virgiliana (Rome 1987) 590. 44. C.W. Muller, 'Der Schone Tod des Polisburgers oder Ehrenvoll ist es fur den Vaterland zu Sterben', Gymnasium 96 (1989); cf. N. Loraux, 'La "belle mort" spartiate', Ktema II (1977) 105-20; cf. J.P. Vernant, 'La Belle Mort et le Cadavre Outrage', 45-76 of La Mort, Les Morts dans Les Societes Anciennes, ed. G. Gnoli and J.P. Vernant. 45. Of the writers quoted in n. 44, only Loraux comments on Tyrtaeus' inclusion in the category of Tpioavres, those who have suffered a wound in the back, speaking of an "attitude maximaliste que ne suivent pas ses successeurs". Perhaps even more striking is the absence of comment on specific parts of the body in the passages I have cited in the otherwise excellent commentary of Carlo Prato, Tyrtaeus (Rome 1968). He breaks his silence only at fr. 11 West, 17 ff. (dpyaXeov ydp SmaGe ueTd 108 HOPELESSLY DEVOTED 53. e.g. Onasander 33: ...et \xev ydp kv S TOU cr6\nTavros- f| aa>TT|pi.a oTpareiJuaTds' ecmv, otiTo? obSev el Te9W}{;eTca Tre 109 MATTHEW LEIGH Satire 8, for which see the introduction to Courtney's commentary and the excellent discussion and citations offered by Susanna Braund in Chapter 3 of Beyond Anger (Cambridge 19S8). Cf. Cicero, De LegeAgraria 2.200 for the distinction between those attaining glory in cunabulis and those in campo. Cicero notes that as a nouns homo, he has nulli...maiores, nutti...auctores generis and nullae...imagines to commend him. For other Ciceronian reflections on the theme, cf. Pro Murena 17, In Verrem 2.3.7 and 2.5.180; and Pro Sest. 136. Detailed discussions of virtus and the ideology of novitas in H. RolofT, Maiores bei Cicero (Unpubl. diss., Giittingen 1938) 10 ff; J. HellegouarcTi.Lc Vocabulaire Latin des Relations et des Partis Politiques sous La Republique IParis 1957) 476-83; A.J. Woodman, Comm. on Veil. Pat. 2.94-131 (Cambridge 1977) 255-63; T.P. Wiseman, New Men in the Roman Senate 139BC-AD 14 (Oxford 1971) 107 ff.