C. Velius Rufus

C. VELIUS By D. RUFUS KENNEDY ROM the moment of its first publication in 19031the career inscription of C. Velius Rufus has been the subject of a great deal of attention and scholarly debate.2Why this should have been so is clear enough: the text records the career of a man who, beginning as a centurion (probably not in the ranks), not on rse to the coveted primipilate but went on to important independent field commands and an administrative career which culminated in F major procuratorial governorships. More important still for historians of the period, the career included special commands which shed vital new light on the otherwise often obscure warfare of the reign of Domitian. The interpretation and chronology of the career as it appears in standard textbook treatments of the period may be said to be essentially that of Ritterling as modified and refined by a succession of eminent scholars, notably Syme, Pflaum and Dobson.3 This mainstream interpretation was summarized most recently by the last of these in I978. From the outset, however, and in parallel with the development of this standard view, there have been a number of dissenters: von Domaszewski disagreed with Ritterling on a crucial point and, since then, there have been challenges to various aspects from, for example, McElderry, Hanslik, Saxer and Visy.4 Indeed, in the same year as Dobson presented a refined restatement of the orthodox interpretation, Visy coincidentally, offered a major re-examination. These earlier challenges, however, tend often to disagree with one another and, just as many to-day will be unconvinced by Visy's proposed revisions so, in most cases, his forerunners have found little support. This should not, however, obscure an important point, namely that there has been persistent dissatisfaction with the standard view from the outset. Nor should the implausibility or weakness of the proposed alternatives detract from the very serious nature of some of the objections raised. While the present discussion also offers a re-interpretation of the career, my particular concern has been to emphasize that the standard view is flawed - that at best explanations are required, at worst it must be modified in a number of particulars. Consequently, the treatment 1T. Mommsen, 'Inschrift aus Baalbek', Sitzungbericht der Akademie der Wissenschaft in Berlin, (1903), 817-24. 2 The origin of this article lies in an appendix of my unpublished doctoral thesis, The Auxilia and Numeri Raised in the Roman Province of Syria (Oxford, 1980), and I am grateful to my examiners (Dr A. K. Bowman and Mr M. W. C. Hassall) for suggestions made then, and to Dr B. Dobson, Prof. S. S. Frere, Drs V. Maxfield and M. Roxan who read a draft for the thesis. Since then, it has undergone many changes and has in its most recent form benefited in particular from the valuable criticism of Prof. A. R. Birley and Dr M. Roxan, and advice from my colleagues Derek Mosley and John Drinkwater. None, of course, are responsible for what has finally emerged, nor do they necessarily agree with any or all of it. 3 E. Ritterling, JOAI, vii (1904), Bbl., 23-38; R. Syme, JRS, xviii (1928), 41-55; H.-G. Pflaum, Les Carrieres Procuratoriennes Equestres (Paris, 1960), I, no. 50, 114-17 and III, 966; B. Dobson, Die Primipilares (Bonn, 1978), no. 94, 2i6-7; cf. the discussion under IGLS, VI, 2796, 113-7. 4 A. von Domaszewski, Philologus, lxvi (1907), 164-7; R. K. McElderry, JRS, x (1920), 68-78, esp. 74-8; R. Hanslik, R.-E., VIII.A. Sup. (1955), 629-31; R. Saxer, Untersuchungen zu den Vexillationen des rdmischen Kaiserheeres von Augustus bis Diokletian (Koln, 1967), 22-3; Z. Visy, Acta Archaeologia Academiae Scientiarum Hungaricae, xxx (1978), 37-6o. 183 184 D. KENNEDY here collects, restates and elaborates on some of the objections of earlier commentators, as well as bringing forward others which have arisen in the course of my own examination of the text. The text itself, from Heliopolis/Baalbek, presents no serious problem, being well-preserved and checkable against a photograph:5 C. Velio Salvi f. Rufo, p(rimo)p(ilo) leg(ionis) XII fulm(inatae), praef(ecto) vexillariorum leg(ionum) VIIII. I adiut(ricis), II adiut(ricis), II Aug(ustae), VIII Aug(ustae), VIIII Hisp(anae), XIIII gem(inae), XX vic(tricis), XXI rapac(is), trib(uno) coh(ortis) XIII urb(anae), duci exercitus Africi et Mauretanici ad nationes quae sunt in Mauretania comprimendas,donis donato ab imp(eratore) Vespasianoet imp(eratore)Tito bello ludaico corona vallar(i), torquibus,fa[le]ris, armillis, item donis donato coronamurali,hastis duabus, vexillis duobus et bello MarcomannorumQuadorum Sarmatarum adversus quos expeditionem fecit per regnum Decebali regis Dacorum corona murali, hastis duabus, vexillis duobus; proc(uratori) imp(eratoris) Caesaris Aug(usti) Germanici provinciae Pannoniae et Dalmatiae, item proc(uratori) provinciae Raetiae ius gladi. Hic missus in Parthiam Epiphanemet Callicinum,regis Antiochifilios, ad imp(eratorem)Vespasianumcum ampla manu tributariorumreduxit. M. Afius M. f. Fab(ia) Olympiacus aquilife(r) vet(eranus) leg(ionis) XV Apollinar(is). IGLS VI 2796 = ILS 9200 Most scholars accept that the items are divided into compartments. Retaining the order of the text, its components can be tabulated as follows: I. primuspilus of leg. XII Fulminata praefectus of legionary vexillations tribune of coh. XIII urbana dux of an army drawn from Africa and Mauretania 2. decorated by Vespasian and Titus in the Jewish War decorated a second time decorated in a war against the Marcomanni, Quadi and Sarmatae 3. procuratorial governor of Pannonia and Dalmatia procuratorial governor of Raetia ambassador of Vespasian into Parthia Thus we have military posts (omitting those before the primipilate); military decorations (beginning with a set won before the primipilate); civil administrativeposts; and the 'exploit'. Again, it is widely accepted that within each group, the items themselves are arranged in chronological order from lowest to highest. Finally the standard interpretation holds that, leaving aside the embassy which was not easily categorized, and a single aside (see below), the items from primipilate to procuratorship of Raetia are in fact in correct chronological 6 IGLS, VI, pl. viii. The text given here is that presented by Dobson, op. cit. (note 3), 2I6. C. VELIUS RUFUS I85 sequence. The career of Velius Rufus, together with dates, as may be tabulated from Dobson, is set out below: Prior to 71 72 72-82 82 83 after 83 but before 86 86 88/9 88/9-92 c. 92 TABLEI: Dobson's Chronology centurion in an eastern legion and decorated in the Jewish War embassy into Parthia to escort back the Commagenian princes centurion in an eastern legion primus pilus of the legio XII Fulminata praefectus of 9 vexillations in the Chattan War tribune of the XIIIth Urban Cohort at Carthage and, in that capacity, appointed dux of the forces of Africa and Mauretania in a war in the latter area - decorated for the second time conducts XIIIth Urban Cohort to the Danube expedition across Dacia with his cohort in Domitian's war against the Germans and Sarmatians active on the Danube - decorated for the third time two procuratorships in succession It would be unproductive at this stage to rehearse in any detail the alternative views on offer, the more so since they are conveniently summarizedby Visy.6 It is worth noting, however, that one may see the second group of items, the decorations, not as a continuation of the military posts but as a commentary on the military career just recorded. Thus Visy takes the the tribunate to the ties and tewar posts in strict succession against the Marcomanni which results in the African command succeedingthe expedition across Dacia. Moreover, the dating of the vexillationary command has been disputed by von Domaszewski, Homo and Hanslik, who all place it in c. 77/8 and Vespasian's war against the Bructeri. Visy's dating leads him to assign the second set of decorations to the prefecture and he believes the second procuratorship may well have come under Nerva. Before proceeding to consider the individual items it may be useful to set out the areas in which I believe the standard view glosseser ovr or fails to offer convincing explanations for problems: As noted above, Dobson believes that, excluding the embassy, the items are in overall chronological sequence. Thus, the early military offices are uncontroversial; however, having reached the mention of the Mauretanian command, the dedicant wished to note Rufus' decorations but first inserted those already won in the Jewish War but not previously mentioned because the text only began with the primipilate. With this interruption out of the way, the career sequence is resumed with the expedition across Dacia in which Rufus is still tribune of the XIIIth cohort. The impression given in this standard interpretation is the comfortable one of a well-ordered and carefully thought out text with a single explicable break to mention an early honour. Whether or not that is so, but especially if it is, it immediately raises two serious problems: Why, when the vexillations themselves are individually recorded, is their location and purpose omitted? It does not happen elsewhere. Second, Rufus' expedition across Dacia is located and explained but why is his force not named? The prevailing view, that it was cohors XIII urbana, is absurd for so hazardous and potentially important a move and even Hanslik's view, that the cohort was the nucleus of the force, begs the question - what was that force? There is also the important historical point which has led scholars to baulk at placing the vexillationary command in 83: in the very year of Mons Graupius is it likely that Domitian 6 Visy, op. cit. (note 4), 50-4. I86 D. KENNEDY withdrew vexillations from all of Agricola's legions ?7 Though it is in fact a widespread belief that the emperor did withdraw large numbers of troops from Britain8 evidence exists only in the case of one, IX Hispana.9There are numerous smaller but important problems which require answers - e.g. why was the primus pilus of a Cappadocian legion appointed to a vexillationary command on the Rhine?; when did he become primus pilus?; why and when was he chosen to go to Parthia? and why was he appointed in Africa ? None of these problems can be tackled in isolation but as a preliminaryto wider discussion it is necessary to first treat some specific questions. I THE CHARACTEROF THE INSCRIPTION The desire to look for order and symmetry in the layout and content is understandable and, of course, not at all uncommon in career inscriptions. It may nevertheless be imposing modern concepts on the ancient world; and, in this case, the oriental part of it where these supposed virtues are less highly prized. Too many career inscriptions seem only to be explicable if one assumes 'stone-cutter's error' or a 'mistake' in the draft. In the case of this inscription it may be readily seen that the work would have been expensive and cost would not have been an important consideration. Indeed, a prominent characteristicof the items recorded is the wealth of detail provided: the individual legions from which the vexillations come are listed; the embassy into Parthia is fully explained and even the princes named. In terms of content and layout, the omission of his posts prior to the primipilate is not surprising: it was enough to begin by noting this vital and prestigious springboardto great things. The question of chronological order can only be a matter for debate: suffice it to say, that a careful draft by the dedicant aiming at overall chronological order and which was preparedto note decorations in their correct positions would surely have begun with the legionary centurionate in which he was first decorated and in which apparently he performed such an important embassy for the emperor ratherthan placing the one in the middle as an aside and the other tacked onto the end. Once one accepts that any one item is out of chronological order, one has to concede the possibility of more chronological dissonance. Indeed, was the dedicant likely to have known the correct sequence and the precise details ?10In short, there may be considerable confusion in the text and/or it may not have been drafted as carefully as we like to believe such texts were. II THE JEWISH WAR AND THE ANNEXATIONOF COMMAGENE In August A.D.70 the great Jewish revolt, begun in 66, was brought to an end by Titus with the capture and destruction of Jerusalem. Mopping up continued for some time after but the prince himself and most of the additional troops in the war zone were soon on their way home. Titus himself held victory celebrations when, according to Josephus, he acted with considerable generosity: 'He put gold crowns on their heads, gave them gold torques, miniature gold spears and standards made of silver, and promoted every man to a higher rank.' The level of decorations recorded in our inscription shows that at the time Rufus was a centurion; the scope for 7 McElderry, op. cit. (note 4), 75; P. Salway, Roman Britain (Oxford, I981), 38 ff. significantly perhaps makes no mention of withdrawals. cf. Hanslik, op. cit. (note 4), 630 and Saxer, op. cit. (note 4), 23. 8 See e.g., S. S. Frere, Britannia (London, 1978), 128; D. Breeze, The Northern Frontiers of Britain (London, 1982), 49; P. A. Holder, The Roman Army in Britain (London, 1982), I6; M. Todd, Roman Britain (Glasgow, I98I), io8. 9 ILS 2719 records a tribune of II Adiutrix in bello Suebico it[em Sa]rmatico which must be the war of I025; 89 (below). 10Since there is no need to assume that Velius Rufus' primipilate was in the same legion as his earlier career, it is possible that he and the dedicant had served together in XV Apollinaris. C. VELIUS RUFUS I87 promotion during and after so bloody a war must have been considerable - rapid in the case of the more able as Rufus apparently was. In late 70, Titus was in northern Syria around Antioch, even visiting Zeugma; he may not have left the area until early 71 or later. About that time, too, the new governor L. Caesennius Paetus, sent out to replace Mucianus, must have arrived. According to Josephus, 'in the 4th year of Vespasian's reign', Paetus denounced Antiochus IV Epiphanes, client-king of Commagene, alleging an intended revolt and an understanding with Parthia. With the emperor's approval, Paetus struck first and seized Samosata. The royal princes, (Antiochus) Epiphanes and Callinicus, put up a spirited, if doomed, resistance, but the king himself, rather pathetically, if more realistically, travelled to Tarsus in Cilicia and delivered himself into Roman hands. The campaign over, Antiochus was put in chains and despatched to Rome while his sons fled to the court of Vologeses of Parthia: But Vespasianrefusedto let the king be broughtto him in this condition, thinkingit betterto honour their old friendshipthan to make the war a pretextfor relentlessanimosity.He gave instructionsthat while he was still on the way he should be freedfrom his fetters,and insteadof continuinghis journeyto Rome shouldremainfor the presentin Sparta.He furtherassignedhim a large revenue,enablinghim to live not only in luxurybut in royal state. When news of this reachedEpiphanesand his brother,who had hithertobeen extremelyanxious about Antiochus, it was a verygreatload off theirminds.They had hoped that Caesarwould be reconciledto them as Vologeses had intervenedon their behalf; for though they were living in comfort they were most unwillingto live outsidethe Roman Empire.Caesargraciouslypromisedthemsafe conduct, and they travelledto Rome wheretheirfatherat oncejoined themfromSparta.Theyweretreated with all honour and settled in the City. Vespasian's fourth year was I July 72-30 June 73. Vespasian's subsequent generous treatment of both Antiochus and his sons (whose descendants attained consular rank) shows how little he had believed in the alleged treachery of his till-recently-loyal ally in the Jewish War. Indeed the annexation cannot be seen, or dated, in isolation but has to be placed in the context of Vespasian's major reconstruction of the Euphrates frontier culminating, after the annexation of Armenia Minor and Commagene, in the creation of a new two-legion consular province of Cappadocia. The process was begun when Titus sent the legion XII Fulminatasupposedly as a punishment for its poor performance in the war - to the new fortress of Melitene on the Euphrates above Commagene. The annexation of Armenia apparently came next: Aristobulus of Chalcis who was in support of Paetus in Commagene is usually identified as the Aristobulus erstwhile rule of Armenia Minor, but now removed and transferredto his ancestral homelands in Syria. The precise date of this annexation cannot be fixed but has been shown to have occurred in 72.11 As for Commagene, to avoid too tight a chronology for events on the Upper Euphrates, it is tempting to take Josephus literally and see Paetus as initiating in 72 the process which led to invasion in, perhaps, spring 73. However, since Samosata itself later dated by an era beginning in 72,12we must also place the annexation of Commagene as a whole in mid/late 72. While Paet-is himself, with the recollection of his former ignominious surrender in the same region a generation earlier, would probably not have tarried in sending news of the success to the emperor,13it is not necessaryto see the subsequent events happening with the same rapidity. Given the considerable distances over which messages had to be carried, at the very least it 1 A. Bosworth, Antichthon, x (1976), 66 and n. 24. B. V. Head, Historia Numorum, 2 (Oxford, 19I I), 776. 13Paetus' messenger travelled swiftly enough for Vespasian to have Antiochus intercepted and diverted to Sparta. 12 I88 D. KENNEDY would have been many months before the princes set out from Parthia to return to the Roman Empire. Vespasian responded swiftly enough to the news of Antiochus' humiliating approach to have been able to have him freed and diverted to Sparta. It would have been rather longer before he heard of the presence of the princes at the court of Vologeses; indeed, we are told that Vologeses interceded for them with the emperor. At the very earliest, and matters may well have been much more leisurely, the returnjourney probably did not commence until early 74; the entire episode and the delivery of the princes to Rome may not have been completed until 75 or later.14 IV THE STATUS AND CHOICEOF VELIUSRUFUS As we have seen from the inscription, the envoy entrusted with the task of conducting the princes to Vespasian was C. Velius Rufus. For long it was assumed that he performed this function while primuspilus of XII Fulminataat Melitene.l5 Dobson, however, has pointed out on his chronology of over was held for a single year only and that the he that the primipilate gapr6 t1 years before his (next) command over the legionary vexillations was too long to be plausible. From the evidence available, it is clear that the men chosen for such embassies were usually centurions, legionary, if not praetorian. Thus, in A.D. I, Tiberius sent a centurion as envoy to the feuding Thracian kings;17 during the civil wars of 68-70, Vitellius sent centurions to win over the Moorish chiefs;l8 a little later, the troops of Hordeonius and Vocula sent centurions to the Gallic cities.19Closer to our period and area, in 6i, Corbulo in Armenia sent the centurion Casperius to Vologeses who was encamped on campaign at Nisibis.20 More usefully, in 54, both Corbulo and Ummidius Quadratus (governor of Syria) had sent envoys to Vologeses advocating peace and recommending the giving of hostages as a token of his good faith. Persuaded, Vologeses handed over the hostages to a centurion, Insteius Capito, who was already there on earlier business for Quadratus. Corbulo intervened and sent his own envoy to receive them. The latter, Arrius Varus, is said to be prefect of a cohort.21He appears again in the pages of Tacitus accompanying Antonius Primus in 69 - he is said then to have 'gained a reputation for vigour, thanks to service under Corbulo .. But there was a rumour that he had criticized Corbulo's ability in the course of private conversations with Nero ... and so obtained promotion to primus pilus . ..' The following year saw him as praetorian prefect, soon to be given praetorian insignia, though later to be praefectus annonae when demoted by Mucianus. Capito, too, appears again in history:22 three years later during Corbulo's campaigns in Armenia, the latter sent off two sub-groups to undertake separate action, one was placed under a legionary legate the other was given to Capito who is described as praefectus castrorum,an office usually held only after the primipilate.23In short, while the status of the original envoys sent to Vologeses by Corbulo on this occasion is not given, we may see that very senior officers were involved in the actual receipt of the hostages. Under the circumstances, therefore, I 14 For several months of every year journeys of any kind would be unwise on land, and at sea neither messengers nor passengers would travel in the hazardous winter months. 15 e.g. IGLS, VI, 2796 (p. I I5); Pflaum, op. cit. (note 3), i, no. 50 (p. 15). 16 Pflaum, op. cit. (note 3), , no. 50 (p. 996), citing Dobson and correcting his own earlier interpretation in vol. I. 17 Tacitus, Annals 11.65. 18 Tacitus, Histories 11.58. 19ibid., IV, 37. It was a centurion too who conducted Antiochus to Sparta. 20 Tacitus, Annals XV.5. 21 ibid., XIII.9; Histories 111.6, IV.2, IV.4, IV.39, IV.68. Arrius Varus = Dobson, op. cit. (note 3), no. 74; H. Devijver, Prosopographia Militiarium Equestrium, ab Augusto ad Gallienum (Leuven, 1976-80), AI64. 22 Tacitus, Annals XIII.39; Dobson, op. cit. (note 3), no. 63. 23 B. Dobson, ANRW, II.I (Berlin-New York, 1974), 412 f. C. VELIUS RUFUS I89 cannot conceive of the officer sent to receive the Commagenian princes and their followers and conducting them to the emperor himself being anything other than very senior: if not a primus pilus at least of the primi ordines and already a man marked out for his ability. How did he come to be chosen? Ability and connections would have been important but is his eastern origin and his recent involvement in the Jewish War no more than coincidence? After all, the same war had seen the prince Epiphanes distinguish himself in Roman service.24 That they were known to one another seems not at all improbable. Who selected him? Surely not on the sole initiative of the governor of Syria, though one might envisage instructions being sent from Rome ordering the appointment of a suitable individual. Rather more plausible for so delicate a task and from the point of view of efficiency, one should perhaps see Vespasian, his decision made, selecting an appropriateenvoy in Rome and his being despatched from there direct. He could have been in Rome for any one of a number of reasons: as the messenger of Paetus; or transferred to the centurionate in the new Praetorian Guard;25 or, indeed, in the numerusof primipilaresat Rome !26 V THE PREFECTUREOF VEXILLATIONS The inscription reads VIIII but the legions listed immediately after are only eight in number. A simple error on the part of the stone-cutter seemed obvious until Ritterling27drew in some tile-stamps published over 20 years before by Mommsen.28 Discovered at Mirebeau-surBeze, 22 km N.E. of Dijon in the territory of the Lingones of Upper Germany, they recorded construction by a number of legionary vexillations and seemed to locate the activity of Velius Rufus' command and provide the name of the missing legion. The relevant variants are: I. VEXIL L*EGIONVM I *VII XI* XIIII *XXI 3. VEXILL[ GIONV [ I *VII[ 2. VEXIL · LEGION[VM VIII * XI * XIIII X[XI 4. VEXILLEGIOIH[ ]VG · VII[ The first two pose no problems; the latter are traditionally restored to read: 3. Vexil. L[e]/gionum/II, VII[II, XX] 4. Vexil. legion[um]/II Ad., II A]ug. VII[II, XX]. In short, those named or restored are the eight legions named in Rufus' command with the addition of XI Claudia.Could they be dated? The legiones Adiutriceshad only become iustae legiones under Galba in 69; at the other extreme, XXI Rapax was destroyed under Domitian, possibly in 92. A precise date for the brigading together of these legions was seen in the campaign against the Bructeri beyond the Lower Rhine in 77/8 and harmonized well with the then supposed dating of Velius Rufus' recent activities in the East.29Moreover, it seemed unthink24 Josephus, BJ V.46. Epiphanes had already been wounded earlier in Roman service, at the First Battle of Bedriacum cheering on the forces of Otho. 25 B. Dobson and D. Breeze, Epigraphische Studien, viii (1969), I02. 26 A. von Domaszewski, (B. Dobson), Die Rangordnung des romischen Heeres2 (Koln-Graz, 1967), I I6; cf. Saxer, op. cit. (note 4), 23 and Dobson, op. cit. (note 23). 27 Ritterling, op. cit. (note 3), 25 f. 28 Mommsen, op. cit. (note 2), 437-4I. 29 Domaszewski, op. cit. (note 4), 166; cf. McElderry, op. cit. (note 4), 74 f. and Hanslik, op. cit. (note 4), 629. 190 D. KENNEDY able that vexillations would have been taken from Agricola in Britain at the time of his great advances c. 78-83; instead the early date allowed them to be returned before his campaigns began. However, Ritterling, elaborated on by Syme, rejected this date, pointing rather to the e Rhine as the occasion for bringing these Chattan War of Domitian in 83 beyond tethe pper vexillations together. The scene of war fitted better with their presence in Germania Superior and indeed a more telling case could be made: tile-stamps seemed to show that in 83, and for a very short time only, Germania Superior had 5 not 4 legions - XXI Rapax, till recently in the lower province, having been released by the newly created I Minervia.30Thus, this large group of vexillations could be seen as a force drawn from two provinces only: the four British legions and the five of Upper Germany. Further pieces could be added. Vexillations withdrawn from Agricola's army seemed borne out by Tacitus who tells us that the IXth legion was the weak column in Agricola's army and thus selected for a Caledonian attack,31while an inscription records a tribune of a vexillation of IX Hispana decorated by Domitian in expeditione Germanica.32Finally, some time around 85/6 I Adiutrix moved to the Danube, followed a few years later by XXI Rapax, and later still XI Claudia and XIV Gemina.33 The objections to this later date can be briefly stated: (i) the supposed removal of so many troops from Agricola at that juncture is still worrying; a grave emergency seems unlikely, if they could become engaged in building; (ii) the 'German expedition' may be either that against Antonius Saturninus in 89 or, more plausible, the war against the German Marcomanni and Quadi later still ;34 (iii) Tacitus does not actually say IX Hispanawas understrength (though that is certainly one meaning) merely that the column of which it formed part was regarded as the weakest of the three;35 (iv) whatever the campaign in which the tribune was decorated, his vexillation seems to be alone and he, a senatorial tribune, in command.36 To these we may add some other more general objections. First, as Syme observed in 1928, Ritterling's tile restorations were 'daring':3 it is not at all clear that there is room on nos. 3 and 4 to enter legions II Adiutrix and XX Valeria. Next, building on a worry expressed by McElderry one may ask why a force of this sort, intended for use on the Upper German frontier, should include vexillations from the Upper German legions,38 and why draft in from a distant province a primipilaris- was there no-one at hand worthy of command?39 A force based in Mirebeau, about as far from the frontier as one can get in Upper Germany without crossing into Lugdunensis,40would seem rather to be mustering for a campaign somewhere other than the provinces from which the legions came; Lower Germany or the Danube are the obvious places. Even for those who prefer an early date of c. 77/8 or see Rufus' primipilate as 30 Syme, op. cit. (note 3), 42. Hanslik seems not to know of Syme's article and still prefers to see the vexillations at Mirebeau for use against the Bructeri. 31 Tacitus, Agricola 26. 32 ILS I025. 33 Syme, op. cit. (note 3), 41 f. 34 cf. Q. Vilanius Nepos (ILS 2127) decorated in succession in a Dacian War, a German War and a Dacian War again. 35 Whether or not troops had been removed from Britain there was no need for Agricola to have had his army in equal divisions, legionaries and auxiliaries together. Cf. the implications for size of Corbulo's 3 columns in Tacitus, Annals XIII.39. 36 McElderry, op. cit. (note 4), 74 f. points out that it is doubtful if a senatorial tribune would be expected to serve under an eques. 37 Syme, op. cit. (note 3), 43; cf. McElderry, op. cit. (note 4), 75 f. observing that the restoring of the names of II Adiutrix and IX Hispana was 'arbitrary'. 38 cf. Syme, op. cit. (note 3), 47 f. making a similar point with regard to ILS 2719. 39 McElderry, op. cit. (note 4), 75. 40 Syme, op. cit. (note 3), 43 n. 3 recalls that Tacitus, Histories 1.59 places 8 cohorts of Batavians amongst the Lingones in 69. C. VELIUS RUFUS I9I immediately preceding an appointment in 83 it leaves unanswered the awkward question as to why Domitian chose to appoint an officer from a Cappadocian legion whose experience (as far as we know) had all been in the East, to take charge of a force on the Rhine. A very large force at that: the strength of one or two legions depending on the size of vexillations. A force which should have gone to a senator or at least someone of a senior equestrian rank, even a praetorian prefect.41If true, an astonishing appointment. Perhaps it would be more plausible, however, if some intermediate posts were available, and if the command came rather later when the severity of the wars pointed to the need for professional soldiers in charge and when his own fears of the Senate predisposed the emperor to by-pass senators for the eerreatorsmore important commands. If not dated to 83 what then becomes of the homogeneous components of the force? After c. 85/6 the Rhine lost three legions to the Danube - I Adiutrix, XIIII Geminaand XXI Rapax; Britain lost all of II Adiutrix which was transferredto the Danube not later than 92 - in fact, commonly placed several years earlier.42In short, the vexillations under the command of Velius Rufus could have been made up from a number of provinces: 3 from Britain, 4 recently on the Rhine but now on the Danube, plus VIII Augusta and, perhaps, XI Claudia, the legionary garrison of Upper Germany it would seem for a short time in the early go's before the arrival of XXII Primigenia again made it a three-legion province.43 There are no easy answers to the question of what 7, perhaps, 9, vexillations were doing building at Mirebeau, or even when. That the intended or recent scene of operations was not the Lower Rhine is certain in view of the distance the British, and even Upper German Units, had been brought; I have also argued that the Upper Rhine is little less plausible. Building activity suggests either early mustering for a campaign, perhaps the next spring, or the exploitation of a battle-group prior to its dispersal at the conclusion of a campaign. Either way, there is no need to believe that this force was actually commanded at Mirebeau by Velius Rufus: he could have commanded his vexillationary group before or after the Mirebeau posting. The vexillations could have been taken individually or in a variety of groups from their parent forces at different times over a number of years: the prolonged and widespread wars could easily have led to individual vexillations being separated from the main body for several years. VI THE WAR IN MAURETANIA A brief literary reference of Dio preserved by Zonaras gives us an account of a campaign c. 85/6 to the south of Numidia conducted by the imperial legate (Cn. Suellius?) Flaccus and resulting in the extermination of the entire tribe of the Nasamones.44The focus of Velius Rufus' activity is, however, clearly Mauretania and seems to have involved the forces (excluding presumably the legion in Numidia) of both Africa and Mauretania, potentially a very large force. Presumably appointed, too, at some time other than - probably after - the appointment of an imperial legate over both provinces in place of the usual equestrian procurators. This legate, Sex. Sentius Caecilianus, is attested in office in 75.45 Rufus' command of the coh. XIII urbana at Carthage and his position as dux of a joint African-Mauretanian force strongly suggest that he held them in succession; indeed, he may 41 Saxer, op. cit. (note 4), 23 makesthe same point but cites a second example;his no. 47 (p. 27) = ILS 2726, which enables him to reject Domaszewski's belief that these were 'march commands' only. 42 e.g. Frere, op. cit. (note 8), 140, n. 20. 43 Syme, op. cit. (note 3), 45 disputes the reduction to two proposed by H. M. D. Parker, The Roman Legions (Oxford, 1928), ii5. 44 Dio 67.4.6; cf. M. Rachet, Romeet les Berberes (Brussels,1970), 156. 45 Cos. des. in 75 (AE (1941) 79; cf. Pflaum,op. cit. (note 3), I 6 n. 7). D. KENNEDY I92 have been appointed dux while still tribune, perhaps even with the Urban Cohort as part of his force. Dobson has already pointed out that the urban cohorts at Lyons and Carthage were effectively independent commands of considerable prestige, their tribunes being excused, apparently, the normal succession of posts in the Rome cohorts and the need for a primipilate iterumbefore entering an equestrianprocuratorialcareer.46In such circumstances,the appointment of a tribune as dux would seem less surprising. When did it happen? We know that part at least of the cohort served in Europe under Domitian: a centurion, Q. Vilanius Nepos, was decorated a Domitiano in, successively, a Dacian War, a German War, then a Dacian War again, i.e. c. 85-c. 89.47It would make most sense for Rufus to have held his command either before or after this period though his involvement may have overlapped the two geographical spheres, i.e. he could have commanded it in Africa and subsequently conducted it to the Danube to take part in the wars. The campaign is in fact traditionally dated to the period before the Danube wars called the cohort away from Carthage.48 Some guidance may be obtained from an examination of the units in the province of Tingitana. EXCURSUS: THE DIPLOMASOF MAURETANIATINGITANA In the first volume of CIL, XVI there were no diplomas yet discovered for Mauretania Tingitana and their evidence was, consequently, not yet available to most such early commentators on Rufus as Ritterlingand Syme.49To-day, we have no less than 29 whole or fragmentary diplomas for Tingitana. The earliest, and amongst the best preserved, are those for 88, I09 (three identical diplomas) and II4/7.50 The diploma for 88 records 10 regiments, none of which bear any honorary epithet. The evidence from this diploma and from those others of Trajan's reign is laid out in TABLEII. By the close of Trajan's reign 3, probably 4, alae, and 7, TABLE II CIL, XVI, I59 CIL, XVI, I6i; I62; Arheologija, CIL, XVI, 165 CIL, XVI, I69; I70 (88) 21 (1979) 4-4 (II4/7) (i22) [I] Aug. c.R. I Hamiorumsag. III Asturump.f. c.R. Gemel[liana...] [Gall. Taurian]atorquatavictrix. Augusta c.R. I Ham. Syr. sag. III Astur. c.R. p.f. Gemelliana[... Gall. Taur. torq. vict. c.R. [............] I Hamior. Syror. [III As]tur. Gemellianac.R. Gallor. Taurian[.... [II milliariasagi]ttarior.c.R. V Delmatarum I Lemavorum II mil. sag. [V Delmatarumc.R.] [I Lemavorumc.R.] [II] Syror.sagit. oo V Delm. c.R. I Lemav. c.R. IIII Gallorumc.R. I Itur. c.R. II Hisp. c.R. II Hispana c.R. I Astur. et Callaec. I Celt. c.R. III Astur. c.R. IIII [Gall. c.R.] I Itur. c.R. [II Hisp. c.R.] II Hispana c.R. [I Astur. et. Ca]llaec. I Celt. [c.R.] [III] Astur. c.R. IIII Gallor. c.R. I Ituraeor.c.R. II [Hisp]anor.c.R. II Hispa[na c.R.] I Astur. et Call[... (109) alae I Augusta I Hamiorum III Asturum Gemelliana Tauriana cohortes II mil. sag. V Delmatarum I Lemavorum I Bracarorum IIII Gallorum 46 Dobson, op. cit. (note 3), III Ast. c.R. 217. 47 ILS 2127. 48 Visy believesthe cohort to have still been in Italy until Domitian sent it to the Danube as attestedon ILS 2127 for part or all of 3 wars, before, in the go's, it was moved to CarthagewhereVelius Rufus was its 49 M. Roxan, Roman Military Diplomas (London, 1978), I9-23. 50 CIL, XVI, 159; I6I; 162; V. Ljubenova, Arheologija, 21.4 (I979), 4I-4; CIL, XVI, 165. tribune. C. VELIUS RUFUS 193 probably 9, cohorts, in Tingitana carry the title civium Romanorum. Furthermore, one is additionally torquata victrix.51Most of these titles have appeared on the diplomas as early as 109; putting it another way, 80 per cent of the units on the diploma for 88 are certainly c.R. a 09. Again, five new units appear on the diplomas by 114/7, and probably at least as early as of Tingitana in I09 for the first time, all designated c.R. and all probably relatively recent arrivals.52 It must be said at once that there is no evidence whatsoever to suggest that any unit from Tingitana, much less I , fought in Domitian's or Trajan's wars in Europe where these titles could have been won. On the other hand, it cannot be categorically stated that the diploma of 88 is not simply failing to give regiments their full styles - in short, 88 may not be a secure terminuspost quem.Indeed, the ala GallorumTaurianawas certainly already torquatavictrix c.R. under Vespasian and before it moved to Mauretania but is not given the title on the diploma of 88.53Again, the ala I Hamiorumis not given its descriptive epithet sag(ittaria). However, no other unit can be shown to have been c.R. and, while this may be due to the scarcity of evidence from Tingitana, there is the possibility that most, if not all, of the other units which became c.R., including those which seem to have arrived as a reinforcement in the same period, won their block enfranchisement between 88 and 109. Indeed, one might suppose that the absence on of c.R. from all of the units ns the o diploma of 88 was because only one of them bore the title and in that case it was of some antiquity already; it is less probable that the c.R. would have been omitted if almost all of them had recently been so signally honoured. One might note, finally, that the diploma issued for Syria only a few months later, does give the title c.R. to three of the units listed on it. Even if we assume that the forces involved in the war to which these honours refer were no larger than the combined strength of the units rewarded, (i.e. all of them but the ala Tauriana) we still have a very large force of c. 1,500 cavalry and c. 5,000 infantry. Furthermore, such widely distributed honours imply considerable success. The focus of the campaign seems, too, to have been Tingitana; the known auxilia in Africa and Mauretania Caesariensisshow no such pattern of rewards and, if employed at all, must have been used either on a minor scale or in support.54With regard to these two provinces, it must be borne in mind that we have only one relevant diploma, for Caesariensis, and that new discoveries may yet reveal units from either or both playing a much larger part.55 Finally, there is the question of the identity of the cohors I Ituraeorumc.R. on the diploma of 109. The matter is not as yet provable but I regardit as most likely to be the cohorsI Ituraeorum on the Syrian diploma of 88 but not found there in later years.56The inference from this, too, 51Thep.f. attributed to the ala III Asturum cannot, as P. Holder (The Auxilia from Augustus to Trajan (Oxford, 1980), 39) has suggested, be anything to do with the revolt of Saturninus. In any case p.f. is a loyalty award not a battle decoration. (I am grateful to Margaret Roxan for this point.) 52 M. Roxan, Latomus, xxxii (1973), 845, n. i6; cf. n. 60 and n. 76. 53 Holder, op. cit. (note 51), 32, 36 and 39. 54 The ala I Hamiorum may have been sent to Tingitana from Syria between 83/4 and 88. E. Birley in Corolla Memoriae Erich Swoboda Dedicata (Graz, 1966), 56 n. 4 and Holder, op. cit. (note 51), 15 have proposed to see the ala and cohors Nervianae, in Caesariensis on the diploma of 107 (CIL, XVI, 56) as the ala I Flavia milliaria and the cohors I Augusta, both previously known in Syria. The similar names they bear in Mauretania suggest that they may have arrived simultaneously. The terminus post quem for one of them is AE (1925) 121 for 76. 55 We may note too the cohors Musulamiorum in Syria in 88 (CIL, XVI, 35) and a cohors I Flavia Musulamiorum in Caesariensis in Io07(CIL, XVI, 56, cf. CIL, VIII, 4878 from Numidia naming the prefect of cohor. I Musulamior. and dated early second century; also AE (1913) 157 has it nearby in I I9 - Devijver C235 and M. Jarrett, Epigraphische Studien, (1972), 174, no. 49). Holder, op. cit. (note 51), 233 differentiates between them, but they could easily be the same regiment. 56 Kennedy, op. cit. (note i), 155-8. D. KENNEDY I94 is both that Tingitana received a reinforcement and that it came after 88. The Syrian diploma is dated 7 November 88 so that the unit could not have reached Tingitana until at least mid-89. The tentative implications of this evidence from the diplomas is that there was a substantial and successful war in Mauretania, probably Tingitana, at some date between 88 and 109 for which at least 2 units, including 5 or 6 which were brought in as reinforcements,were signally honoured. Do we in fact have two wars, one in the early to mid 8o's, another sometime between 88 and 109; one in which Velius Rufus had command and another for which we have as yet no indication of commander? Warfare with the tribes on the desert's edge was certainly not uncommon in North Africa, but we must consider the possibility that it is simply a single war the evidence for which has been mis-dated in one case. VII THE EXPEDITIONACROSSTHE DACIAN KINGDOM Velius Rufus was decorated in bello MarcomannorumQuadorumSarmatorumadversos quos expeditionemfecitper regnumDecebali regis Dacorum.There can bee little doubt that this could only have taken place after peace and terms had been agreed with the Dacians.57 Equally, the expedition of which Velius had command must have been sustantial. What force did he have? Why is the composition of so important a force omitted? Why is his primipilate named, his tribunate of the urban cohort and his command of the army in Mauretania, even the legions which contributed vexillations for his prefecture,but not his force in this case, here baldly cited as expeditio? The orthodox answer as we have seen is that he is still tribune of the XIII Urban Cohort which he has now led to Europe. Surely this cannot be the case ? It cannot seriously be entertained either that a man who had held such important commands already should have been left now, in the midst of an extensive war, in command of no more than a single cohort,58 or that that cohort, the paramilitarypolice force of Carthage, would have been anything other than laughable for the enterpriseenvisaged. The plan was a bold one, relying on the good faith of an erstwhile enemy59and placing a body of soldiers far beyond the imperial frontiers. Such a force must have been substantial and would probably have consisted of detachments from elsewhere rather than the regular garrison of Moesia. Once again we are left to enquire what this force is and why it is not mentioned. VIII THE DECORATIONS Three sets of decorations are recorded, the first and third of which are specifically attributed to occasions: the Jewish War and the expedition across Dacia. Thanks now to the recent important re-examination by Valerie Maxfield of Roman military decorations we may readily see that the first set are at the level appropriate to a centurion while the second and third, which are identical, are the appropriate scale for someone of the status of a tribune of a Rome cohort. 60 The first set of decorations confirm, therefore, that at the end of the Jewish War, Rufus was as yet still a legionary centurion. Until his appointment to equestrian procuratorships only two other posts (as opposed to special commands) are recorded: primus pilus and tribunus cohortis XIII urbanae.It seems clear that both sets must have been awarded while he held the latter rank: the traditional view being that they were awarded for either the war in Africa or Dio, 67.7.1I; A. Mocsy, Pannonia and Upper Moesia (London, 1976), 83 f. Hanslik, op. cit. (note 4), 630 suggests it was the nucleus of his force. 59 Dio 67.7.2 makes it clear that although Decebalus was unable to continue the war after Tapae in 88 the formal peace only came after the outbreak of war with the German tribes of the upper Danube. 60 V. Maxfield, Roman Military Decorations (London, I98I). 57 58 C. VELIUS RUFUS I95 the Dacian campaign and then the named expedition against the Marcomanni, Quadi and Sarmatae. IX THE PROCURATORSHIPS The text explicitly attributes the first procuratorship to Domitian (in the colourless formula employed on inscriptions after his death). On balance it would seem likely that the second too was given by Domitian: the career seems to halt there as would not be unnatural under the new regime passing over those elevated by the discredited dynasty. It has been pointed out that the posts given to Rufus are key ones in the go's with the wars in hand against the tribes of the Upper Danube; appropriate, too, to his recent experience. Finally one might note the reference in Martial to a certain Velius close confidant of the emperor, who made an offering of a silver goose in recognition of Rome's success/salvation in the recent wars on the Danube.61 He has been identified as C. Velius Rufus but the matter, though attractive, is by no means certain. If true, however, it would tend to reinforce the need to find a substantial force for him to have commanded in his last expedition. X DISCUSSION Some of the problems involved in interpreting this text seem to me insuperable as matters stand and I cannot claim that the chronology offered below is anything other than tentative. The way forward, however, may lie in re-emphasizingthat Rufus holds only two military posts which are here named, the primipilate and the tribunate; all the other military items are special commands not part of either the equestrian or even senatorial cursus. The ducate in Africa is clearly dependent on the tribunate and, as we have seen, the decorations reveal that it was at that level that he conducted the expedition across Dacia. The resolution of the major problems of the purpose of his vexillationary command and of his force in the expedition across Dacia may be achieved by seeing these items as complementary rather than distinct: in short, Velius Rufus was given charge of the legionary vexillations as his force for the venture against the flank of the Marcomanni, Quadi and Sarmatae. The naming of the primipilate first should not be taken as an indication that subsequent items are given in ascending order but simply a statement at the outset of his antecedents, the much prized post which made his astonishing subsequent career possible. The next items are in fact given in reverse order: prefect of vexillations, preceded by his tribunate of the XIIIth Urban Cohort in which office he was appointed dux of a field-force in Mauretania. The decorations - a commentary on what has immediately preceded rather than a continuation - are listed next in ascending order, the opportunity being taken to explain the first and third; the second may have been passed over either because the occasion of their award was unknown to the dedicant or, more likely, because he had just mentioned it, the ducate in Africa. To summarize: in 70, Velius Rufus, centurion of one of the legions in Iudaea for the war, was decorated and promoted. A few years later, by then a senior centurion, arguably even at Rome in the Praetorian Guard, he was despatched to Parthia to conduct back the refugee Commagenian princes. Promotion followed, again, possibly in the Guard at Rome, and, probably in the later 70's, he achieved the primipilate in legio XII Fulminata, possibly the legion in which he had been a legionary centurion. After his year in office he may have passed into the numerusof primipilares at Rome. Early in the 80's he was appointed tribune of the XIIIth Urban Cohort at Carthage and, while in that office, or subsequently, put in charge of an 61 Martial, Epigramata, IX.3I; cf. Hanslik, op. cit. (note 4), 631; A. N. Sherwin-White, The Letters of Pliny (Oxford), 64I suggests that Velius Paulus and Velius Rufus may be the same man; B. W. Jones, Domitian and the Senate, (Philadelphia, 1979), I20; IGLS, VI, p. I 6. I96 C. VELIUS RUFUS army drawn from the garrisons of Africa and both Mauretanias for a campaign in the latter. For his success there, in the mid 8o's, perhaps he was again decorated. Within a short time the tleoDomitian call upon even the Urban Cohort at Carthage gravity of warfare on the Danube led for a part at least of its strength, which then served in the Dacian campaigns and the intervening Germano-Sarmatian campaign. Rufus probably led his cohort to Europe but may soon have been given the more important command, in the light of his by then proven ability, of a vexillationary group (which need not have included the Urban Cohort). In that capacity he subsequently led a daring expedition, in 89 probably,62across Dacia against the new threat on the middle and upper Danube for which he was again decorated. Rufus' Carthage post qualified him for rapid promotion to the equestrian procuratorial career. In the early go's, and certainly before Domitian's death in 96, he was appointed to two strategically important procuratorial governorships. Had Domitian survived, Rufus might, like M. Valerius Maximianus some 80 years later have been adlected into the Senate to enable him to be given legionary commands: he would certainly, like Cornelius Fuscus and Vibius Maximus, previous procurators of Pannonia and Dalmatia, have been a suitable candidate for a praetorian prefecture.63In the event, his descendants at least did achieve senatorial and consular status. Two points remain. First there is the war in Mauretania. My analysis of the diplomas of the province of Tingitana suggest a major campaign after 88, while the above discussion of Velius Rufus' career points to a date before 89 for his ducate and tribunate. The diplomas are of course not a wholly reliable guide since it is essentially an argument from silence. Moreover, Mauretania frequently saw fighting and there is no reason why there could not have been a campaign in the 8's involving Rufus and a second,, sometime between 88 and 9, in which several regiments won block grants of citizenship. Finally, the wider consequences of dating Velius Rufus' vexillationary command to 89. As noted above, the evidence for large-scale troop withdrawals from Agricola's army even before Mons Graupius rested largely on the dating of Rufus's career and the attribution of the 'German expedition' in which L. Roscius Celer commanded a detachment of IX Hispana to the Chattan War, rather than the equally likely 'German 'war of 89. Without either of these one has to look to a rather later occasion for the withdrawal of the vexillations: the disastrous defeat and death of Oppius Sabinus at the hands of the Dacians in 85 and of Cornelius Fuscus the next year would suggest to me that the vexillations were only withdrawn in 85 or 86, the parent body of II Adiutrix following soon after as the magnitude of the crisis on the Danube sank home. The five German legions on the Mirebeau tiles have been dated to c. 83 and for a short time thereafter, when all five named are believed to have been together in the upper province (see above). Pressing the date forward to 85/6 should not strain that dating. Universityof Sheffield. 62 The victory at Tapae in 88 is usually regarded as having freed Domitian to deal with the Marcomanni, Quadi and Sarmatae (M6csy, op. cit. (note 57), 1976, 87 f.). 63 cf. Syme,JRS lxx (I980), 65.