Drinkwater J.F. The Alamanni and Rome 213-496 (Caracalla to Clovis)

Подождите немного. Документ загружается.

brought them to me . . . [Constantius] had twice betrayed [the people of

Gaul] to the enemy.264

On 1 January 362, in a panegyric to Julian, the Gallic orator

Claudius Mamertinus talked of: ‘the entire barbarian world raised to

arms against the defender of Roman liberty. . . and of peoples recently

conquered and stubbornly resisting the yoke newly placed on their

necks incited by impious inducements to renewed fury’.265 Libanius’

great funeral oration on Julian, of 365, has Constantius resorting:

to the same trick of calling in the barbarians by letter as he had done before,

and begging them as a favour to enslave Roman territory. One out of many

he induced to break his word, and he began to ravage and also to make

merry in the lands he had got as his reward, and he went to dine with the

generals on our side as though he had done nothing wrong at all. This fellow,

who had dared break the treaty, he arrested in his cups. . . . and, crossing over

into his territory, [Julian] inflicted a well-deserved punishment for his

treachery. Those who had bided loyally by their agreements gathered in

alarm, greatly ashamed at such misconduct, and added oath upon oath, and

he mounted a tribunal in the middle of barbarian countr y, gazed down upon

their chieftains who stood as subjects with their hordes of followers, and

after issuing threats and reminders took his leave. . . . The despatches sent by

the cowardly, treacherous Constantius to the barbarians were of the greatest

assistance to him, for as he sailed or marched along, he read these out to the

cities and the garrisons.266

Against these charges is, as a number of commentators have argued,

Vadomarius’ apparent lack of concern in accepting the Roman oVer

of hospitality: unlikely if he had been guilty of major wrong-

doing.267 It is not impossible, of course, that this was an act of

‘barbarian’ bravado. However, although Ammianus clearly dislikes

Vadomarius, calling him deceitful,268 he never accuses him of being

headstrong. In addition, the notion that Vadomarius should pretend

to break the peace ‘from time to time’ (subinde) is meaningless. He

could have shown himself an enemy only once, after which he would

264 Julian, Ep. ad Ath. 286A–87B.

265 Pan. Lat. 11(3).6.1 [trans. Nixon and Rodgers (1994)].

266 Libanius, Orat. 18.107–8, 113; cf. also Orat. 12.62 of 1 Jan. 363.

267 e.g. Barcelo

(1981: 46); Lorenz, S. (1997: 67–9). Cf. den Boeft, den Hengst, and

Teitler (1991: 44); Nixon and Rodgers (1994: 402 n.38).

268 AM 21.3.5.

258 ConXict 356–61

have been condemned by all Roman authorities. This suggests that

there was little hard evidence to link him to the troubles, and suits the

surprisingly mild punishment meted out to him by Julian.269 It

should also be pointed out that Julian’s proposed reoccupation of

the Rhine bridgehead site or sites, in 360, was apparently unopposed

by Vadomarius, and may itself have severely restricted his capacity to

do harm.

Vadomarius was probably not the instigator of attacks on Gaul.270

The accusations of Julian, Mamertinus and Libanius are not dispas-

sionate reporting but court propaganda. The basic strategy of this

propaganda was to accuse Constantius of doing in 361 what he had

done against Magnentius from 351: of calling in the Alamanni. Its

most powerful tactic was constant reference to Julian’s discovery of

letters sent by Constantius to ‘the barbarians’, that is, in the Wrst

instance, to several kings, including Vadomarius. This justiWed

Julian’s removal of Vadomarius and his embarking on civil war. But

this was deception. Constantius had burned his Wngers very badly in

calling upon Alamannic help to deal with the earlier usurper; and

Julian, a close relative and still the heir apparent, was no Magnentius.

No Alamannic chief is likely to have been called on to repeat the

experiment. When Ammianus came to look into the claim, he prob-

ably found little concrete evidence to substantiate it. The best he was

able to come up with was mention of a letter from Vadomarius to

Constantius that was highly disrespectful of Julian but not materially

dangerous and hardly surprising in conWdential correspondence

between two allies.271 Ammianus’ ‘rumour’ is simply what was trans-

mitted by the ‘Julianic’ sources. Libanius’ declaration that Julian had

Vadomarius flogged like a common criminal,272 which is out of step

with Ammianus’ account of his immediate treatment and his subse-

quent career, should also be regarded as part of this deception.

Libanius, needing to justify the shocking fact that it was Julian, not

269 Cf. Geuenich (1997a: 52).

270 Cf. Drinkwater (1996: 24 and n.22).

271 Cf. den Boeft, den Hengst, and Teitler (1991: 37).

272 Libanius, Orat. 13.35 [July 362]: ‘And when many tribes had been summoned

to oppose you by enormous payments, you captured their leader without diYculty

and inflicted upon him the disgrace reserved for felons, teaching him, by the lash

upon his ribs, not to make money in this way.’

ConXict 356–61 259

Constantius, who took the Empire to the brink of civil war, was

given to many flights of fancy in respect of Vadomarius and the

Alamanni.273 This undermines the historicity of the Alamannic

‘durbar’ which, according to Libanius, Julian held at the close of his

expedition.

With Vadomarius out of the way, Julian was able to move east-

wards. His short Alamannic campaign enabled him, legitimately, to

concentrate his forces at Kaiseraugst in preparation for the coming

war.274 It also allowed him to give his troops one more Germanic

victory before they faced a more redoubtable foe. The minor

campaign, therefore, served a number of purposes, among which

must have been Julian’s assertion of his authority in the area after

Vadomarius’ deposition and, as part of this, his establishment of

Vadomarius’ son, Vithicabius, as king in his place: the position in

which we Wnd him in 368.275

However, what Julian and his circle subsequently made of Libino’s

disaster distances us from what actually happened earlier in 361. This

was no minor incident. Celtae, Petulantes and comes together formed

a Wrst-rate Wghting unit—a ‘brigade’ of the imperial Weld army—

comprising paired infantry and cavalry regiments, in total about

1,200 strong, under a single commander.276 The deployment of the

brigade outside the normal campaigning season in an area that Julian

had recently gone out of his way to secure suggests a response to a

serious raid—made all the more serious by the defeat of the Roman

force and the death of Libino. The action was a signiWcant loss of face

to the army and to Julian who, unused to failures of this kind and on

the verge of the greatest military and political gamble of his career,

needed to believe that the gods were still with him.277 He could not,

in the Wrst instance, have welcomed it as an excuse to deal with

Vadomarius. Certainly, there is no case for supposing that it was

Julian who engineered the aVair.278 His last Alamannic campaign

thus served also to avenge Libino and expunge the shame of his loss.

273 Cf. den Boeft, den Hengst, and Teitler (1991: 47).

274 Szidat (1981: 88).

275 AM 27.10.3; 30.7.7. PLRE 1, 971. Below 281.

276 Elton (1996a: 89, 91).

277 Cf. Drinkwater (1983b: 370).

278 Nixon and Rodgers (1994: 402 n.38).

260 ConXict 356–61

The defeat would have been a major element in immediate

Romano-Alamannic relations, whoever was to blame. Vadomarius’

presence on Roman soil may therefore have been not entirely fortuit-

ous. The tone of both Ammianus and Libanius implies that social

intercourse between barbarians and Romans was not unusual. This

hints at a fair degree of integration.279 However, the dinner following

which Vadomarius was taken may have been more than mere end-of-

day entertainment, but the climax of a meeting (or series of meetings)

intended to resolve a crisis. It may be signiWcant that Julian sent

Philagrius into the area clearly aware that he might well encounter

Vadomarius within the Empire.280 It is not impossible that among

Philagrius’ many other important duties (which may have included

arrangements for Julian’s coming move eastwards)281 was liaison with

local military commanders concerning the recent defeat. It may well

have been in this sense—from his point of view, honest legitimate

imperial business—that the senior oYcer involved in Vadomarius’ fall

invited him over on imperial orders.282 This interpretation runs

contrary to Ammianus’ claim that Vadomarius crossed the Rhine

fearing nothing, ‘as was natural in time of profound peace, and

pretending to know of no irregular doings’,283 but, given the calamity

that had only just occurred, this rings untrue. One suspects that

Vadomarius felt at ease because he knew he was not to blame.

But who was responsible for the attack? The raiders may have been

a band of unruly warriors284 from Vadomarius’ territory simply out

for mischief (stung, perhaps, by Julian’s high-handed treatment of

the previous year), and therefore out of their chief ’s control. Alter-

natively, the leader of a local war-band, knowing Vadomarius’ atti-

tude to Julian, might have wrong ly believed that the king had given

him some sort of permission to carry out an unoYcial raid. Or

Vadomarius himself, wrongly anticipating Constantius’ instructions,

279 Above 157.

280 AM 21.4.2.

281 AM 21.4.2. Szidat (1981: 89, 95).

282 Cf. Browning (1975: 112); Lorenz, S. (1997: 68).

283 AM 21.4.3: ut nihil in profunda metuens pace nihilque secus gestorum simulans

scire.

284 Cf. above 122, 222.

ConXict 356–61 261

may indeed have encouraged such a sortie. Against the last of these is

the case already presentedand Vadomarius’ knowledge of Constantius’

method of working, and his need to cover his own back: he would

never have initiated serious action without written instructions.

Against the other two are the numbers and experience a war-party

would have required to overcome Libino’s brigade. This leaves a large

war-band from the interior.

Although Ammianus says that the raiding Alamanni came from

Vadomarius’ territory, implying that the king was responsible for

their misdeeds, he nowhere says that they were actually of it.285

Population movement from the east was a constant feature of

Alamannic development, dangerous to settled communities on the

imperial frontier and to the Empire itself.286 It is possible that more

distant Alamannic communities were now being disturbed by Julian’s

preparations. When, in the spring of 361, Julian reached the navig-

able Danube he ‘conveniently’ found suYcient boats to hand.287 The

force of Ammianus’ statement is debated, but it has been widely

interpreted as indicating that Julian’s administration prepared a

fleet for him to use.288 Such a fleet had to be Wtted out or constructed

on the Danube in Alamannic territory, upstream from Raetia.289

Indeed, given the possibility that this river could have taken shipping

upstream from Ulm, Julian may even have embarked his troops

here. This was in an area which, to judge from the coin-Wnds,

otherwise saw little Roman presence.290 However, the operation

was, presumably, founded on the combination of Roman expertise

and Alamannic muscle and mate

riel that was commonplace on the

Rhine.291 A raiding-group, excited by Roman activ ity, might have

moved into Vadomarius’ territory before attacking the Empire over

the Rhine. That it was an alien group that Julian dealt with later is

285 den Boeft, den Hengst, and Teitler (1991: 48).

286 Cf. above 106, 238–9.

287 AM 21.9.2: quos opportune fors dederat plurimos. For the date see Barnes (1993:

228); Lorenz, S. (1997: 70).

288 den Boeft, den Hengst, and Teitler (1991: 115); Szidat (1996: 71).

289 Szidat (1996: 72).

290 Above 129–30 and Fig. 16.

291 Above 234.

262 ConXict 356–61

hinted at in Ammianus’ account of their fate—a somewhat detached

story that suggests that Libino’s killers (who, if they had been

local, and especially after the removal of Vadomarius, would have

scattered) were still operating as a single group.292

But there is another possibility. If, as Julian claimed,293 Constantius

had put Raetia on a war footing, and was collecting military supplies

there, the province would have been a tempting target. Is it possible that

Julian encouraged raiding on western Raetia, which then got out of

hand? This would Wt Ammianus’ initial indistinct location of the

troubles, on the Gallo-Raetian border, distant from Vadomarius’ terri-

tory, which has occasioned debate.294 It would also ac c ommodate

Szidat’s suggestion that Libino was sent to negotiate a settlement with

the raiders.295 Overall, therefore, 361 began with a nasty surprise for J ulian

but, by demonizing Alamanni, he was able to turn this to his advantage.

Before moving to examine the conflict between Alamanni and

Julian’s successors, it is useful to review the relationship between

Rome and the Alamanni as it had developed from around 260 to

361. Frequent reference has been made to the Roman need for

the Rhine–Danube re-entrant to become an adjacent ‘living-room’

rather than an ‘invasion-corridor’.296 Rome constructed this room by

encouraging the development of settled Alamannic chiefdoms along

the border.297 In the late third and early fourth centuries, any danger

to the Empire came not from these ‘kingdoms’, which were small and

divided, with weak economies and low populations,298 but from

292 AM 21.4.7. Cf. den Boeft, den Hengst, and Teitler (1991: 48).

293 Julian, Ep. ad Ath. 286B: ‘On the frontiers of Gaul in the cities near by he

ordered to be got ready three million bushels of wheat which had been ground in

Brigantia.’ Against den Boeft, den Hengst, and Teitler (1991: 119) and Szidat (1996:

72), I follow Wright (1913: 287) in identifying ‘Brigantia’ as Bregenz/Brigantium, not

Brianc¸on/Brigantio. As Szidat says, Constantius’ main fear was that Julian would seize

Italy and, as we have seen (above 209), Bregenz was a major supply centre on the

military highroad between the High Rhine and the Italian sub-capital at Milan.

Benedetti-Martig (1993: 354) identiWes the place as Szo

ny/Brigetio, which must

surely be too distant.

294 e.g. Szidat (1981: 91–2, 94); den Boeft, den Hengst, and Teitler (1991: 34–5);

Lorenz, S. (1997: 67 and n.261).

295 Szidat (1981: 92).

296 e.g. above 80–1, 99, 163, 189.

297 e.g. above 75, 81, 119, 122.

298 Above 86, 91, 93, 105, 124, 143–4.

ConXict 356–61 263

more distant Elbgermanic raiding-bands, which might suddenly

appear and run riot in neighbouring provinces.299 Rome wanted no

repeat of the troubles of the 260s and 270s. This worked well. There

was some raiding, in particular that leading to the battle of

Langres,300 but the Elbgermanic threat was generally met; and local

Alamannic chiefs played a useful part in preventing Burgundian

expansion down the Main.301 Rome for the most par t left the

Alamanni alone, but they could be exploited by rulers needing a

successful campaign.302 Then, border stability was sacriWced for

political gain, but any such sacriWce was localized and short-term.

Roman superiority made victory inevitable and speedy, and followed

by new agreements and renewed stability.

In the 350s came the usurpation of Magnentius and the Alamannic

‘invasions’. These shook Roman conWdence. Memories of the third

century and general prejudice against barbarians obscured the fact

that Alamannic aggression had been encouraged by close ties with

the Empire, and that Alamannic communities remained weak. This

allowed Julian to exploit the Alamanni as others had done before

him, but on a much greater scale, making them Wgures of fear and

hatred and picking Wghts with them for his own ends.303 However,

the decision to expel Alamanni recently settled within the Empire

was taken by Constantius; and Julian, as others before him, was

careful to restore stability once Wghting was over. After the excite-

ment of 357, Julian appears to have returned to Constantius’ policy

of cooperation with Alamanni, creating a buVer zone of imperial

allies that would have been useful throughout his reign, had he

lived.304

However, important in understanding what follows is, as in the

case of the ‘exclusion’ of high-ranking Alamanni in the Roman

army,305 that this does not make Valentinian the villain of the

299 Above 106, 189, 262.

300 Above 187–9.

301 Above 190.

302 Above 200.

303 Drinkwater (1997: 8); cf. the criticism of Martin (1998: 421).

304 Barcelo

(1981: 47): from 360; Lorenz, S. (1997: 66): from 358.

305 Above 159.

264 ConXict 356–61

piece, arbitrarily renewing attacks on Alamanni simply to conWrm

his position as new emperor. Julian had set a precedent. Though he

perished in the east, his reputation as the scourge of the Alamanni

lived on. If Valentinian wanted to be emperor on the Rhine frontier,

at some stage he had to show himself to be the equal, if not the

superior, of the great Julian and wage war on Alamanni.306

306 Cf. Raimondi (2001: 98–100); below 271.

ConXict 356–61 265

ConXict 365–94

Julian left Gaul in spring 361 to face Constantius II. In the event he

achieved sole control of the Empire without a Wght as a result of his

cousin’s unexpected death in October or November of the same year.1

Julian himself perished in June 363 during his invasion of Persia.

There followed the short reign of Jovian and then the accession of

Valentinian I and his brother, Valens, as joint Augusti. After dividing

the military resources of the Empire between them, Valentinian and

Valens parted company, the former moving west to Milan where he

celebrated his Wrst consulship on 1 January 365.2

Ammianus remains our main source down to 378, that is for the

whole of Valentinian’s reign and for the Wrst few years of that of his

western successor, Gratian. The absence of an extensive and coherent

near-contemporary narrative for the 380s and 390s is regrettable and

must, as in the case of the period 285–354, restrict our understanding

of events. However, again like 285–354, there is enough evidence to

be conWdent that what appear to be the main developments in the

relationship between the Rhine-Germani and Rome are real historical

phenomena and not artefacts of the source pattern.

For the Wrst part of Valentinian’s reign, Ammianus continues to

describe imperial dealings with the Alamanni year by year, though in

less detail.3 There are problems in understanding the events of 364–6

because of Ammianus’ inclusion of what is now generally regarded as

no more than an introductory summary of all the foreign wars of the

1 November: Barnes (1993: 224).

2 Lorenz, S. (1997: 73).

3 Elliott (1983: 159–60).

joint reign,4 and, more confusing, because of the diYculties which, as

he himself realized, he faced in dealing with simultaneous happen-

ings in the western and eastern halves of the Empire.5 As far as those

in the west are concerned, Lorenz has recently assessed the current

state of research and has proposed an ordering of events which I

adopt here.6

According to Ammianus, the Alamanni raided Gaul fairly early in

the reign of Valentinian. This was because representatives they had

sent to Milan to receive the ‘usual’ gifts reported that what they had

been oVered had been of unacceptable quality and that they had then

been treated very badly by the mag ister oYciorum, Ursacius.7 Raiding

began around the same time as Procopius’ usurpation against Valens.

Valentinian was given news of both on the same day, around 1

November. He was in Gaul, journeying to Paris.8

Having arrived in Paris, he sent his magister militum, Dagalaifus,

against the Alamanni. These, however, after devastating the frontier

region, had already withdrawn.9 In the meantime, agonizing as to

whether he himself should march against Procopius, he Wnally

decided to follow the advice he was being given and to stay in Gaul,

and moved to Reims.10 In January of the following year various

formations of Alamanni attacked again.11 The leading group was

intercepted by Charietto (comes per utramque Germaniam) and

Severianus, another military comes and commander of the garrison

at Chalon-sur-Sao

ne.12 The Roman force was broken, with Severianus

severely wounded, Charietto killed and a regimental standard

almost lost.13 Valentinian again sent Dagalaifus out from Paris against

the Alamanni, but he failed to engage them, claiming that they

4 AM 26.4.5–6. Tomlin (1979); Lorenz, S. (1997: 77 and n.32).

5 AM 26.5.15.

6 Lorenz, S. (1997: 84–91): basically, contra Seeck and Baynes, that Ammianus’

Alamannic attack ‘immediately after 1 January’ (27.1.1) belongs where he puts it, in

366, and should not be seen as a doublet of events of 365.

7 AM 26.5.7: ex more.

8 AM 26.5.8.

9 AM 26.5.9.

10 AM 26.5.9–14.

11 AM 27.1.1.

12 Charietto: PLRE 1.200 (his oYce is otherwise unknown).

13 AM 27.1.2–6.

ConXict 365–94 267