Drinkwater J.F. The Alamanni and Rome 213-496 (Caracalla to Clovis)
Подождите немного. Документ загружается.
were too scattered. He was then w ithdrawn, to share the consulship
with Valentinian’s young son, Gratian (born 18 April 359 and there-
fore only around seven years old).14 Dagalaifus was succeeded in
his command against the Alamanni by Jovinus, magister equitum.
He surprised and massacred two Alamannic groups, at Scarponne
and at an unnamed location by a river respectively.15 Jovinus then
tracked down and defeated a third group in a pitched battle near
Cha
ˆ
lons-sur-Marne. The Alamanni lost 6,000 dead and 4,000
wounded, the Romans 1,200 and 200.16 The following day he pursued
the survivors, but in vain. He was also distressed to discover that
Roman troops had killed a captured Alamannic king and some of his
followers.17 Jovinus returned to Paris where he was welcomed by
Valentinian and received the consulship for the coming year. ‘At
about the same time’, Valentinian received the head of the vanquished
Procopius.18 Ammianus closes his account by noticing that: ‘Besides
these battles, many others less worthy of mention were fought in
various parts of Gaul, which it would be superXuous to describe,
both because their results led to nothing worthwhile, and because it is
not Wtting to spin out a history with insigniWcant details.’19
Ammianus regularly refers to the consuls of each year, to occur-
rences that are dateable from other sources, and to speciWc calendar
dates. On this basis, the alienation of the Alamanni took place in
summer 365 and their raiding began in October of that year, when
Valentinian was already in Paris. This was also the date of Dagalaifus’
Wrst campaign against them.20 They attacked again early in 366, when
Dagalaifus went against them for the second time, only to be with-
drawn and replaced by Jovinus.21 The precise dating of Jovinus’
victories is important. Ammianus puts them just before Valentinian’s
receipt of Procopius’ head. Lorenz, allowing equal weight to the time
it would have taken for this grisly trophy to reach the west following
14 AM 27.2.1. Kienast (1996: 333).
15 AM 27.2.1–3.
16 AM 27.2.4–7.
17 AM 27.2.8–9.
18 AM 27.2.10: eisdem diebus.
19 AM 27.2.11.
20 Lorenz, S. (1997: 74).
21 For the plausibility of Dagalaifus’ taking the consulship with Gratian as late as
March 366 see Lorenz, S. (1997: 79–80).
268 ConXict 365–94
Procopius’ execution on 27 May 366 and to Ammianus’ indications
of variable temperatures at the time, proposes that Jovinus crushed
the Alamanni in the course of May 366.22 However, given the impre-
cision of weather reporting, it is probably better to concentrate on
what happened to Procopius’ head. Taking into account that, as
Ammianus explicitly tells us occurred at Philippopolis,23 its trans-
port would have been interrupted by public displays, its journey
westwards would have taken signiWcantly longer than a normal
urgent despatch: at least six weeks and possibly more.24 This dates
its arrival in northern Gaul to early/mid-July, putting Jovinus’ suc-
cesses in mid/late June. It allows them to have been won after the
receipt of news of Procopius’ unexpected defeat on 27 May, racing in
advance of his head and arriving in Gaul in early/mid-June.25 It also
allows the associated issuing of Valentinian’s law on postliminium on
20 June 366, though with little time to spare.26
We know that Ammianus errs in reporting that Valentinian was
still en route to Paris by late October/early November, since the
Codex Theodosianus shows him resident in the city by mid-October;
he remained there until January 366, when he moved to Reims.27 It
was probably in Reims, not Paris, that Jovinus was received by
Valentinian following his defeat of the three Alamannic groups in
mid-366.28 The relative proximity of Jovinus’ Wrst and third victories
suggests that the second must have been in the same area. The river
22 Lorenz, S. (1997: 82–3).
23 AM 26.10.6. Cf. Matthews (1989: 201) on this macabre ‘tableau’; also now
Lenski (2002: 81).
24 It is c.2,150 km (1,290 statute miles) from Nakoleia in Asia Minor, where Pro-
copius was defeated, to Reims. Increased by one-third, to take into account the
physical vagaries of the route, this Wgure becomes 2,866 km (1,720 miles). On this
basis, important news would have taken about 17 days to go from east to west;
cf. Drinkwater (1994: 571). The four weeks or so that it took for the news of
Procopius’ revolt to reach Valentinian in late 365 cannot—as Lorenz, S. (1997: 83)
does—be taken as a guide given uncertainty in Illyricum (AM 26.5.9) which had
disappeared by early 366 (AM 27.10.4). I believe that the head itself travelled at no
more than one-third this speed. Cf. Lorenz, S. (1997: 83).
25 For the implications of this see below 277.
26 Below 278.
27 CT 11.1.13 [18 October 365]; CT 7.7.1 [28 January 366]. Seeck (1919: 226);
Lorenz, S. (1997: 76 and n.24, 78–9, 83–4).
28 Lorenz, S. (1997: 83), from Tomlin.
ConXict 365–94 269
concerned could be the Moselle,29 but given the direction in which
the Wghting was going it could just as easily be the Marne.
According to Ammianus, 365–6 saw a revival of the Alamannic
menace.30 All recent commentators have accepted the idea of a
renascent ‘Germanic threat’ to explain Valentinian’s decision, as
senior emperor, to take the west as his sphere of responsibility and
then refuse aid to Valens. 31 If one accepts that there was no such
threat, another explanation is possible.
However one may interpret their later activities, Alamanni did not
become a problem immediately after Julian’s death.32 It was not until
mid-365, when Valentinian had been in Milan for several months,
that we Wrst hear of the Alamanni becoming hostile. Furthermore,
their agitation was generated by the Roman decision to change the
terms of a long-standing agreement. Thus, when Valentinian chose
the west, he did not do so because he knew that it was currently
enduring or would shortly face major trouble from over the Rhine.
The reasons for his decision to leave the east must be sought else-
where. Imperial politics were potentially Wssile; and all emperors
needed the loyalty of the western armies. This had been demon-
strated recently in Jovian’s moves to secure northern Italy and Gaul,
in which Valentinian had played a leading role.33 However, such
considerations cannot wholly explain Valentinian’s decision to
become western emperor. He could, like Jovian, have dealt with the
west through lieutenants or, if he felt he had to travel there in person,
he could have returned to the east once the job had been done,
putting Valens in charge.
The answer may be found in Raimondi’s recent study of
Valentinian’s choice of the west. This is not her main proposal: that
Valentinian chose the west because he felt that it was endangered by
barbarians. The crucial point against this is the lateness of his arrival
29 So Lorenz, S. (1997: 80).
30 AM 27.1.1.
31 Lorenz, S. (1997: 84 ‘Germanengefahr’, cf. 78, 80, 177); Raimondi (2001: 91–3,
101, 129 ‘le regioni galliche minacciate dai barbari’).
32 Drinkwater (1997: 9); cf. Lorenz, S. (1997: 76–7). Contra, e.g. Jullian (1926:
234); Demougeot (1979: 105); Demandt (1989: 112). The idea of such trouble comes
from an erroneous notice in Zosimus 4.3.4.
33 AM 25.8.8, 9.8, 10.6. Raimondi (2001: 41–60).
270 ConXict 365–94
in what would have been the most threatened areas. He did not hurry
to the Rhine frontier, but moved to nor thern Italy, where he
remained for around 10 months.34 On the other hand, there is
much in Raimondi’s underlying idea that Valentinian acted as he
did aware that the emperor against whom he was bound to be
measured was Julian.35 If Valentinian had made plain that his sphere
of activity would be the east, he would have aroused expectations of
imminent major campaigning in Mesopotamia. Julian had waged
glorious if unsuccessful war against Persia. The treaty that Jovian had
made with the Persians to extricate his army from their soil was
regarded as shameful, and might be seen as having become void on
his death. There would have been enormous pressure on Valentinian
to show himself the worthy successor of Julian by recovering the
territories ceded by Jovian. With the west under Valens, there would
have been little chance of Valentinian’s being able to claim that he
was needed elsewhere; and, faced with the likelihood of attack, the
Persians may anyway have launched their own invasion. All this was
high risk and, as he was to show in Gaul, Valentinian was no great
risk taker.36 The easier course was to defy tradition, give the east to
his junior colleague and prove himself in the west, where there was
no great threat.
This helps explain Valentinian’s revision of the Alamannic gifts.
From the historic readiness of Alamannic leaders to deal directly with
Roman emperors and oYcials, from their sensitivit y to their sup-
posed ill-treatment, and from the trouble they caused on their return,
I take Ammianus’ ‘envoys’ (legati) to have comprised at least some
important reges. The description of the gifts they came to collect as
‘customary’ suggests that these were the same tokens of personal
esteem that Hortarius had expected of Julian. Their composition
was ‘stipulated and Wxed’ (certa et praestituta) by long tradition,
that is determined by no particular Roman ruler, and continued by
those who wanted peace with bordering Alamannic communities.37
34 Seeck (1919: 218–26).
35 Raimondi (2001: 98–100). Cf. above 265.
36 Seeck (1921: 5.12).
37 Above 243; contra Lorenz, S. (1997: 74): they had been established by Julian.
Cf. Gutmann (1991: 11).
ConXict 365–94 271
The Alamannic expectation was probably related to the recent change
of western emperor. Both sides appear to have accepted that gift-
giving, though not strictly part of the process, accompanied the
automatic renewal of treaties between Alamannic leaders and
Rome. It was an important piece of imperial theatre. (The re-dating
of these events, from New Year to summer 365, undermines the
assumption that the giving of such presents was par t of an annual
routine, and that it therefore represented Rome’s buying peace from
the Alamanni.38)
Valentinian had probably already concluded that the Alamanni
would be his chief concern in the west, the core activity of his reign,39
since this was now the only way in which he could emulate Julian.
However, his attitude to the gifts should not be seen as marking a
major shift in policy. Valentinian bent tradition, but did not break it.
He reduced, but did not refuse, Roman gifts to the Alamanni. The
wholesale alienation of border communities so soon after his acces-
sion and arrival in the west would have been needlessly risky.40 What
we see is Valentinian acting like Julian in both cowing and provoking
the western enemy of choice. He made his presence felt by signalling
that all previous agreements were subject to review—though not
major revision or immediate cancellation, which would have been
courting disaster.41 ( The lapse of treaties on the deaths of either of
the contracting parties cut both ways.42) Most Alamannic commu-
nities would fall into line but, given the nature of their society,
Roman action would inevitably goad some hotheads to Wght. These
could be picked oV once Valentinian had reached the Rhine; but
other events prevented this.
Valentinian did not expect serious trouble as a result of his actions
and he did not get it. There was still no major Alamannic attack. This
explains his late and apparently unhurried arrival in Gaul in the
autumn of 365, perhaps in anticipation of Alamannic unrest but
38 Routine: cf. Heather (2001: 25); ‘Danegeld’: Asche (1983: 98); Lorenz, S. (1997:
177).
39 Lorenz, S. (1997: 75–6).
40 Pabst (1989: 305 n.16).
41 Cf. Gutmann (1991: 11–12).
42 See Lorenz, S. (1997: 74 n.12).
272 ConXict 365–94
more likely because it was time he visited the region anyway.43 It also
explains why he headed not for the Rhineland or even Reims (Trier
was still being restored after the damage caused by Magnentius’
rebellion44), but for Paris, Julian’s old residence, on the margin of
the region most likely to be aVected by barbarian raids. His interest
in the city may have been political: Valentinian wanted to appear
as emperor at the site of Julian’s proclamation of 360.45 The
situation changed in October 365 when Valentinian heard of the
Wrst Alamannic raids and his own advisers and representatives of
the Gallic communities purportedly expressed worries about these so
grave that the emperor felt obliged to deny help to his brother.
One might expect the new western emperor, who had already
sought to establish his dominance over Alamanni and who needed
to conWrm his credentials as a soldier, to have taken immediate
personal action. But Valentinian sent out Dagalaifus. Moreover,
when Dagalaifus then, and again in 366, proved unable to bring the
Alamanni to heel and was recalled, he did not suVer for his failure
but, on the contrary, was given the immense honour of being
appointed consul with no less a colleague than Valentinian’s son
and heir, Gratian.46 After Dagalaifus, Valentinian again failed to go
himself to Wght the Alamanni but, in 366, sent another of Julian’s old
generals, Jovinus, who Wnally defeated them and was also honoured
with a consulship. Jovinus’ victories may be dated after the west had
received news of Procopius’ defeat. It is odd that Alamanni who were
slippery and invincible during the rebellion were relatively easy to
suppress afterwards.
The explanation for this strange run of events surely lies in its
political context. Valentinian made much of Alamanni in announ-
cing that he would not help Valens against Procopius. He is supposed
to have been persuaded by his counsellors and deputations from the
Gallic communities not to expose Gaul to barbarian attack by leaving
the country, and to have reconciled himself to this diYcult decision
by appreciating that, while Procopius was merely the enemy of
43 Cf. Lorenz, S. (1997: 76).
44 Drinkwater (1997: 5–6).
45 Raimondi (2001: 122–4).
46 See Lorenz, S. (1997: 79 and n.42), for Valentinian’s trust in Dagalaifus.
ConXict 365–94 273
himself and his brother, the Alamanni were the enemies of the whole
world.47 This is very suspect.
Valentinian’s prioritizing of the interests of the state over those of his
family resembles the excuse given by Claudius II for not proceeding
against the usurper Postumus in 268, and may be regarded as a
convenient piece of moralizing.48 Symmachus, in a speech delivered
early in 369, seems defensive about Valentinian’s early activities in
Gaul, making him reach the Rhine quickly and (as Ammianus was to
do in his turn) presenting the emperor’s refusal to help Valens as the
sacriWcing of personal pride to the public good.49 This suggests that the
western court, realizing the sensitivity of the situation, took pains to
put into circulation an old but reliable justiWcation for Valentinian’s
refusal to accept the obligations traditionally due to a blood relative.
Valentinian had no intention of returning east; and to avoid such a
return he and his counsellors exaggerated a series of minor frontier
skirmishes.50 Again, this was made possible not because of what the
Alamanni had recently done, or what they might do, but because of the
image of them created in the 260s and 270s, and the 350s.51 Contrary to
what he claimed, it is likely that the emperor’s mind was made up by his
appreciation of the gravity of internal, not external, dangers.
Valentinian had to keep the western armies on his side. The revolt
of Procopius will have shaken the new and untested dynasty. We need
to visualize Procopius’ rebellion as it may have appeared when it was
enjoying its early successes: not, as Ammianus would have it, the last
47 AM 26.5.12–13.
48 Drinkwater (1987: 33).
49 Orat. 1.14, 17–19, 23. Cf. below 289. Contra Raimondi (2001: 95), the date of
this speech prevents its being used as direct evidence for the state of the Empire and
for Valentinian’s plans at the start of his reign.
50 Crump (1975: 49) came close to this interpretation, but followed the current
communis opinio in believing in the authenticity of the Germanic threat. Cf. Gutmann
(1991: 12–13); Raimondi (2001: 121 and n.142). I Wnd a striking resemblance
between the basic sentiment, if not precise expression, of Ammianus’ reports of
Valentinian’s apology at Paris (26.5.13: hostem suum fratrisque solius esse Procopium,
Alamannos vero totius orbis Romani) and Dagalaifus’ earlier advising him not to
choose Valens as his colleague at Nicomedia (26.4.1: si tuos amas . . . habes fratrem, si
rem publicam, quaere quem vestias). It may be that Ammianus followed the oYcial
line on the decision of 365, which showed Valentinian in a good light, and transposed
this back to 363, when, as events were to prove, the new emperor ignored the public
weal. Cf. Paschoud (1992: 74).
51 Cf. above 264.
274 ConXict 365–94
gamble of a pathetic and unworthy successor to Julian, but the
promising debut of the true heir, through his relationship with Julian
and his association with Constantius II’s widow and daughter, to the
throne of Constantine.52 It is no wonder that, as soon as was decently
possible after the suppression of Procopius, Valentinian had Gratian
marry Constantius II’s only child.53 In November 365, he must have
thought twice about taking troops to the east, to be directly reminded
of their Flavian loyalties.
In addition, he cannot have felt completely conWdent about his
own position in the west, where memories of Julian and his achieve-
ments will have been especially strong. The western establishment’s
ambivalence towards the new dynasty was to manifest itself in 367,
when Valentinian’s illness provoked intense speculation as to a suc-
cessor and led to the unprecedented promotion of Gratian directly to
the rank of Augustus.54 In 365, precipitate departure from the west
might compound the uncertainty caused by Procopius’ rebellion by
signifying Valentinian’s lack of conWdence in Valens’ ability to deal
with the usurper, and so create a political vacuum which others,
acting for Procopius or themselves, might be eager to Wll. There were
also practical considerations. Events in the east were moving so
quickly that Valentinian was unable to form a clear picture of what
was happening: a reckless campaign might lead to his own defeat, or
leave him the irreconcilable enemy of a renascent Flavian dynasty.55
His best course was to stay put and hold Wrm: to take limited
measures in Africa and Illyricum to prevent the revolt from spread-
ing, but otherw ise to trust in his brother.56 On the other hand,
despite what he said about the Alamannic threat, in such a state of
Xux it would have been unwise for him to lead a local campaign.
Valentinian’s eyes were on the east, and so he remained, aware of but
apart from the military aVairs of the west, in Paris.57
52 Drinkwater (1997: 11); cf. Lenski (2002: 73–4, 82–4, 98–101).
53 In 374, when Constantia was still only about 12 years old: AM 21.15.6; 29.6.7.
54 AM 27.6.
55 Matthews (1989: 198); Seeck (1921: 5.51).
56 Africa: AM 26.5.14; Illyricum: AM 26.5.11, 7.11–12. On the not inconsiderable
value to Valens of the securing of Illyricum see Gutmann (1991: 15–16).
57 So already, though rather less positively, Seeck (1921: 5.23–4); and now Lenski
(2002: 76): ‘Ultimately, then, Valentinian did little more than secure his own position.’
ConXict 365–94 275
But if Alamanni were no threat, why did the situation take a
turn for the worse early in 366, with the defeat of Charietto and
Severianus, and how was it put right? It must be understood that this
defeat was not caused by a large-scale, concerted invasion—after 357,
a second attempt to settle eastern Gaul and so destroy the work of
Julian—or marked the onset of a developing crisis.58 There was never
any ‘Wrst’ invasion. The Alamannic leaders of 357 thought that they
had been invited into the Empire. Like them, those of 366 must have
realized that settlement without imperial permission was pointless.59
Major invasion is not suggested by Ammianus’ account. His descrip-
tion of Jovinus’ Wrst two victories, over a ‘great throng’ and ‘another
mass’ of Alamanni respectively,60 does not suggest engagement with
their main host. This happened at Cha
ˆ
lons,61 where he implies the
presence of more than 10,000 Alamanni. Taking into account all
three conXicts, a total of, say, even 20,000 is far less than the 35,000
he proposed for the barbarian army at Strasbourg. And, as with the
latter, even this Wgure cannot be taken at face value. And if there was
no invasion, there is no need to suppose a sing le leader, with a
speciWc strategy.62
Events happened probably very much like those around the defeat
of Libino in 361. Supposed Alamannic humiliation at the hands of
Valentinian—a tale that grew in the telling63—provoked a run of
raids by local hotheads and, perhaps, a band or bands from further
aWeld. As those had implicated Vadomarius, these implicated
Vithicabius.64 Imperial disinclination to put an immediate stop to
this predation would only have encouraged it. Then came something
more serious, though still not invasion. One set of raiders, probably
no more than a few thousand strong, broke through the regular
58 Contra Lorenz, S. (1997: 84) and Gutmann (1991: 17). Cf. Gutmann’s (1991: 20)
subsequent qualiWcation of the danger.
59 Above 202, 215, 237.
60 AM 27.2.1–2: maiorem . . . plebem; alterius globi.
61 So Elliott (1983: 159).
62 Contra Lorenz, S. (1997: 69, 80–1): that the Alamanni were probably under the
overall command of a single rex, perhaps Vithicabius of the Brisigavi, and attempted
an organized advance on Reims and Paris.
63 AM 26.5.7.
64 Cf. above 259, 262, below 287. Cf. Gutmann (1991: 16–17): that the raiding of
366 (and 365) was probably carried out by Alamanni from the Upper Rhine.
276 ConXict 365–94
frontier troops and became the responsibility of the regional Weld
army under Charietto and Severianus. These, like Libino, were
comites commanding Wrst-class units. However, their involvement
was still a matter of routine, and the situation was not considered bad
enough to call upon the main Weld army at Reims.65 The disaster
which followed again appears to have been caused by Roman over-
conWdence, which turned into panic following stiV Alamannic resist-
ance and Severianus’ accident. Our sources make much of military
shame,66 but very little of casualties, which suggests that Roman
losses were not considerable—again very similar to what we know
happened to Libino’s force. Astonishing Alamannic success made the
incursion more than an ordinary raid. It committed both parties to
irrevocable action—the Romans to seek revenge67 and the Alamanni
to do the most damage they could before, as they knew they would,
they faced Roman main strength.
Overall, however, the Alamanni had played into Valentinian’s
hands. Early in 366, the situation in the east was still very uncertain.
It continued to be in Valentinian’s interests to remain in Gaul, and
Roman defeat allowed him to do so until news of the usurper’s fall
arrived in early/mid-June. Jovinus appears to have worked his way
behind the Alamanni, bunching them up and cutting oV their retreat.
When the time was right he struck from the rear. Given the smaller
numbers involved, and Jovinus’ long experience as general, his vic-
tory at Cha
ˆ
lons cannot be regarded as the equal of Julian’s at
Strasbourg,68 but he did well, and there is more than a hint in
Ammianus’ account that his third encounter w ith them was no
walkover. Roman casualty Wgures, which are likely to have been
less distorted than barbarian ones,69 especially the 1,200 dead, are
unusually high. This may explain why after the battle the unnamed
65 For the operation of the system see Elton (1996a: 179).
66 As Lorenz, S. (1997: 88) proposes, Zosimus 4.9.2–4, reporting Valentinian’s
ceremonial humiliation of the Batavian regiment following their Xight after a Roman
defeat, must also refer to the events of early 366.
67 Ammianus (27.2.1) states explicitly that Dagalaifus was sent out to ‘make good’
the defeat: correcturus sequius.
68 Contra Lorenz, S. (1997: 82), following Rosen (1970: 128); Elliott (1983: 159).
69 Cf. above 238.
ConXict 365–94 277