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

Подождите немного. Документ загружается.

found at Wyhl, some 8 km (5 statute miles) north-east of the bridge-

head fort at Jechtingen-Sponeck and 17 km (10 statute miles) north-

east of Breisach. Here it appears that for several decades after c.400, to

beyond the middle of the Wfth century, a crossing over the Rhine was

guarded on its right bank by an Alamannic warrior-band under

‘princely’ leaders. Such a presence is comparable to that at Wyhlen

(sic), over the Rhine from Kaiseraugst. On the left bank a process

more in line with that at Krefeld-Gellep may be visible at Jechtingen-

Sponeck where, from the later fourth century, burials in the military

cemetery are indistinguishable from those in the neighbouring

region. It is tempting to suppose that this indicates local service, in

other words that, even before 400, local Alamannic males were being

recruited to serve in the Roman fort, and were accompanied by their

dependants.46 Generally, it now appears that there was some sor t of

imperial military presence along the length of the Rhine until around

455, after which the forts concerned fell under independent local

Germanic leaders.47

This raises three issues. First, though the overall direction of these

strongpoints will have borne very little resemblance to that of the

fourth century, some sort of organization must have existed.48 If

there was a Rhine command, or commands, there must have been

a commander, or commanders. In other words, these discoveries add

weight to the idea that the local generals mentioned in the Notitia

Dignitatum may have been in post after 406/7.49 Oldenstein suggests

that Alzey was under the command of the dux Mogontiacensis.50

Second, what was the strategic purpose of such installations?

They no longer oV ered the sophisticated, naval-based defence of

Valentinian I, which disappeared in the troubles following 406/7.51

And in any case, with barbarian and dissident imperial communities

far to the west of the Rhine and with imperial attention being drawn

increasingly to Constantinople and Africa, the Rhine-line could

46 Fingerlin (1990: 110, 123–31), (1993: 75–9), (1997a: 108); Quast (1997: 172);

cker (1999: 218–19). QualiWed above 175.

47 Brulet (1993: 136–7).

48 Bo

hme (1998: 52–4); Kaiser (1998: 39–40).

49 Above 321.

50 Oldenstein (1993: 125–7).

51 Above 297; Ho

ckmann (1986: 403, 415).

328 The Fifth Century

hardly have been much to do with external defence: it probably

looked inwards. The suggestion has been made that its function

was to check neighbouring left-bank barbarians, in particular

Burgundians.52 However, it may have had a wider role. Its existence

cannot have been planned. It was there because it had always been

there: a 400-year-old system of pay and provisioning (latterly,

presumably, from levies on local communities53) was impossible to

destroy overnight. This gave the Late Empire a valuable resource, an

anomalous projection of Roman military strength from northern

Italy to the North Sea. It could be used to police local barbarians;

but it could also serve as a reservoir of men and supplies for Ae

tius’

Weld army: vital not just for Gaul but for the whole of the area

administered or claimed by Ravenna.54 If so, Roman Gaul may be

seen as having resumed something of its ancient role as a military

holding base.55

Finally, it may be held that the development of left-bank Alamannic

garrisons, probably under their own Romanized leaders,56 and with

their dependants, represented Alamannic expansion of a sort. Thus,

in satisfying the demands of the most ambitious bands, it may help

explain the lack of more aggressive Alamannic movement in this

period.57 It should be noted that the eastern Raetian frontier, along

the Iller and upper Danube, appears to have been held until around

430 though there is as yet no evidence for Alamannic garrisons.58

There are hints of a weakening in the imperial grasp on the Rhine

frontier from the 440s,59 but real decline began in the 450s. The

relative stability of the W rst half of the Wfth century was destroyed

by the Hunnic invasion of Gaul in 451, the murder of Ae

tius by

Valentinian III in 454, and Valentinian’s assassination in 455. There

followed a period of intense political Xux and military defeat (Rome

52 Bayard (1993: 226); cf. Kaiser (1998: 39).

53 For the Wscal importance of the civitates, see Drinkwater (1998a: 295–6), less

positive than here. On the continuity of military dependence on taxation into the

Merovingian period, see Martin (1993: 461, from Durliat).

54 Cf. Elton (1992: 170).

55 Drinkwater (1983a: 120); cf. above 14.

56 Cf. Bo

hme (1998: 54–5), on similar developments among the Franks.

57 See further below 358.

58 See Mackensen (1999: 239).

59 Brulet (1993: 228), from study of Argonne ware.

The Fifth Century 329

was sacked by Geiseric in 455) and, within 20 years, the end of the

western Empire. Attila led a large, mixed force west from Pannonia.

We do not know his exact route, but since he broke into Gaul on the

middle Rhine it is likely that he went through Alamannia, perhaps

down the Main.60 Ae

tius and his Gallo-Germanic allies repulsed

Attila near Troyes in midsummer 451. We do not know how the

Hunnic incursion and withdrawal aVected the Alamanni, or to what

extent Alamannic communities resisted Attila. Almost certainly, like

Burgundians and Franks, Alamanni were to be found on both sides.61

Beyond this we may guess that the Huns had a signiWcant impact on

Alamannic societ y, but what this was we cannot say exactly.62

One result, v isible in the archaeology of the period, was the

appearance of skull deformation effected in infancy, a Central

Asiatic custom. However, in Alamannia this appears to have been

a restricted and temporary phenomenon, found among individual

adult (mainly non-Mongolian) immigrants, not thr oughout Alamannic

society as a whole.63

In political and military terms, the best indication of change

among the Alamanni in the troubles that followed the Hunnic

invasion and withdrawal are the Wrst clear signs of their expansion.64

In this they were not unique. All barbarian communities in Gaul—

Visigoths, Burgundians and Franks—exploited Roman weakness to

increase their power. Around 457 Franks took Cologne, and at

around the same time Childeric, father of Clovis, began to lay the

foundations of Merovingian greatness.65 What is unusual is that,

for the Wrst time since Jovinus, Alamannic warriors Wnd mention

in the literary sources.

60 Maenchen-Helfen (1973: 129), though on the grounds that, since the Alamanni

are not explicitly named among Attila’s allies, he must have taken a route around

their territory; Geuenich (1997a: 68). Sidonius Apollinaris’ reference to Attila’s

Frankish allies as living by the Neckar (Carm. 7.325) is not helpful here (contra

Anton (1984: 16–18)).

61 Geuenich (1997a: 68); cf. Sidonius Apollinaris, Carm. 7.320–5.

62 Martin (1997b: 166).

63 Quast (1997: 178); Wahl, Wittwer-Backofen, and Kunter (1997: 341–2). For an

alternative, but equally commonplace, explanation see Steuer (1998: 309–11): skull

deformation not as the result of invasion, but as the diffusion of a female fashion.

64 Martin (1997b: 166).

65 Runde (1998: 677). Following Springer (1998: esp. 305), I avoid identifying

these speciWcally as ‘Ripuarian’ or ‘Rhineland’ Franks.

330 The Fifth Century

Our best evidence for expansion comes from Sidonius Apollinaris,

an imperial politician and subsequently bishop of Clermont in the

Auvergne. His father-in-law, and fellow Gaul, Eparchius Avitus, was

the short-lived (9 July 455–17 October 456) successor of Petronius

Maximus, who had engineered the assassination of Valentinian III. In

his panegyric to Avitus, delivered on 1 January 456, Sidonius

describes how Saxons and Franks had exploited the recent troubles to

raid Gaul, and admonishes ‘the bold Alaman’ for ‘drinking the Rhine

from the Roman bank, and proudly lording it over both sides, a

citizen or a conqueror’.66 At a slightly later date, in a second pan-

egyric that, awkwardly, Sidonius had to address to Avitus’ successor,

Majorian (28 December 457–2 August 461), he recounts Majorian’s

activities as magister militum, in the interval between the fall of

Avitus and his succession. Amongst these was his destruction of

an Alamannic war-band. This band, 900-strong, attacked over the

Raetian Alps into the Campi Canini,67 but was defeated by Majorian’s

subordinate, Burco, with a small force.68

As good a source as Sidonius is the sixth-century hagiographer,

Eugippius, who in his ‘Life’ of St Severinus records an Alamannic

king, Gibuldus, active on the border of Raetia and Noricum in the

period 470/6. Though Gibuldus and his Alamanni subjected the

town of Passau to regular attack, he greatly respected Severinus and

eventually yielded to the saint’s demands that he end his raids and

free his Roman captives.69 Eugippius later mentions other serious

Alamannic raiding against Passau and against Ku

nzing (to its west)

and, further east, into northern and southern Noricum.70 This

brought Alamanni into contact with Rugi, pushing westwards.71

Eugippius elsewhere mentions that Passau was subject to attack by

one Hunumundus and his men. Hunumundus is generally identiWed

as the daring ‘Suavian’ (that is Suebian) king Hunimundus,

described by Jordanes, in his sixth-century ‘History of the Goths’, who,

66 Sidonius Apollinaris, Carm. 7.369–75: Rhenumque, ferox Alamanne, bibebas

Romani ripis et utroque superbus in agro.

67 For the location of this area see above 209.

68 Sidonius Apollinaris, Carm. 5.373–80.

69 Eugippius, V. Severini 19 (¼Quellen 2, 73–4). Date: Runde (1998: 678).

70 Eugippius, V. Severini 27.1–2; 25; 31.4 (¼Quellen 2, 74–6).

71 Eugippius, V. Severini 31, 1–5 (¼Quellen 2, 75–6).

The Fifth Century 331

together with his Alamannic allies, had been thrashed by the

Ostrogothic king, Theodemer, in a battle fought over the Danube

around 470.72 The anonymous author of the ‘Life’ of bishop Lupus of

Troyes, also writing in the late Wfth/early sixth century, records that at

about the same time an Alamannic king called Gebavultus yielded to

Lupus’ plea to free war captives in his diocese.73 Finally, Gregory of

Tours relates how, in the 15th year of his reign, Clovis went to war

against the Alamanni. Hard-pressed in battle, Clovis vowed to Christ

that if he were successful he would abandon his own gods and

become a Christian. Thereupon the enemy turned tail, their king

was killed and the Alamanni y ielded unconditionally to Clovis.74

Gregory’s narrative is traditionally supplemented by a further refer-

ence in his text. This is his remark that Sigibert the Lame, king of the

Franks on the Rhine, received the crippling wound that gave him his

name in a battle against Alamanni at Zu

lpich.75 These two elements

are usually combined to reconstruct a major Alamannic advance

northwards, down the Rhine, towards Cologne, in the late 490s. To

counter the threat, king Sigibert had to call upon his allies, including

Clovis, and the combined Frankish army fought a decisive battle at

lpich, near Euskirchen just south of Cologne, in 496 or 497.76

Mapping of these events appears to produce a zone of Alamannic

expansion and control at its greatest extent c.750 km (420 statute

miles) long, east to west, and c.650 km (390 statute miles) wide,

north-west to south-east: Geuenich’s ‘Aktionsradius’77 (Fig. 25).

The existence of such an area is questionable. Sidonius’ Rhineland

Alaman is impudent, and on its left bank gulps down the river

following some military success. On the other hand, he is a much

less mobile and dangerous a Wgure than Sidonius’ Saxon pirate or his

Frankish bane of the provinces of Germania I and Belgica II.

Sidonius’ complaint may concern no more than Alamannic military

72 Eugippius, V. Severini 22; Jordanes, Get. 277–80; PLRE 2.57, 1069. Castritius

(2005: 199): probably 469.

73 V. Lupi 10 (¼Quellen 4, 13).

74 Gregory of Tours, Dec. Lib. Hist. 2.30.

75 Gregory of Tours, Dec. Lib. Hist. 2.37.

76 e.g. Koch (1997a: 191).

77 Geuenich (1997a: 70–1). I follow Geuenich here in eschewing the diYcult and

suspect ev idence of the ‘Ravenna Cosmography’ (¼Quellen 4, 9–12); cf. Runde (1998:

679 n.118) concerning Alamannic expansion.

332 The Fifth Century

Rhône

Saône

Seine

DALMATIA

PANNONIA II

Julian Alps

ANNONARIA

Cottian AlpsCottian Alps

Lugunensis

Lugdunensis

GERMANIA

⬘ Aktionsradius⬘

NORICUM

PANNONIA I

Fig. 25 The north-western provinces, Wfth century.

The Fifth Century 333

leaders on the left bank claiming as their own territory over which

they had long since exercised de facto control under nominal imper-

ial suzerainty:78 there was no massive Alamannic ‘invasion’ of

Alsace.79 Indeed, further aWeld in Germania I any Alamannic inter-

lopers would have been in competition with Sidonius’ expanding

Franks.80

Sidonius’ account of the success of Burco makes plain that he

encountered no more than a war-party. At 900-strong, this is in

size little diVerent from those proposed for the third century,81

and calls to mind the war-bands that Wgure in Eugippius’ ‘Life’ of

Severinus. The young bloods of Alamannic warrior society were, as

ever, exploiting imperial weakness to show their mettle. SigniWcant in

this respect is the diVerence between them and Feletheus, king of the

Rugi, who realized that if he was to control an area and its people he

had to do so through its towns.82 ‘King’ Gibuldus, in contrast, was no

more than a war-leader, who had no desire to take or hold land.83

‘King’ Gebavultus appears to have been another war-leader—if what

we are told about him can be trusted.84

An enigmatic and controverted statement by Gregory of Tours—

to the eVect that in the late 460s one ‘Odovacrius’ (clearly the

Odovacer who deposed the last western emperor in 476) allied with

Childeric against Alamanni who had penetrated Italy85—could be

further evidence for such raiding. Current debate, in which some

have amended ‘Odovacrius’ to ‘Adovacrius’ (a Saxon leader in Gaul)

and ‘Alamanni’ to (Gallic-based) ‘Alani’, concerns not whether

78 For the image of barbarians drinking from particular rivers as a poetic means of

indicating their geographical location cf. Nemesianus, Cynegetica 67–8 (¼Quellen 2,

118). Cf. Goetz and Welwei (1995: 1.272 n.57) for Veleda as being honoured by

Rhenopotai.

79 See Martin (1997b: 167) and Wieczorek (1997: 241–2), for the lack of archaeo-

logical evidence for Alamannic settlement in Alsace. Cf. Siegmund (2000: 287–92),

for a likely partition of the area between Franks and Alamanni, but not until the

period 530–670.

80 Martin (1997b: 168); Wieczorek (1997: 242, 246–7).

81 Above 49, 61.

82 Eugippius, V. Severini 31.1, 4 (¼Quellen 2, 75–6).

83 On the weakness of Gibuldus see Pohl (1998a: 642).

84 Below 336.

85 Gregory of Tours, Dec. Lib. Hist. 2.19.

334 The Fifth Century

Alamanni were capable of such attacks, but whether an early

Merovingian king possessed such international contacts.86

Finally, assessment of the battle of Zu

lpich is not straightforward. It

is impossible to establish whether Clovis’ great Alamannic victory and

Sigibert’s wounding relate to the same event. And, anyway, it is now

realized that for Gregory the victory itself was less important than the

Catholic conv ersion an d baptism that it is supposed to have accelerat ed,

and that this conversion may have inXuenced Gregory to exaggerate

Clovis’ military success. There was probably no huge Alamannic force,

and no conclusive victory. Other sources suggest that the process by

which Clovis absorbed the Alamanni lasted several years, and involved

the defeat of a number of lo cal lead ers. All this weakens the likelihood of

a massive Alamannic push north to Cologne.87

Overall, therefore, instead of an area of concerted Alamannic

expansion we should envisage one of marginal settlement and heavy

raiding.88 Where the latter occurred it led to terrorization of civilian

populations, not permanent and regular control.89 Short-term

expansion by Danubian Suebi into Noricum, checked by Theodemer

and ending in the Xight of Hunimundus and his followers, was

probably the occasion of their absorption by Alamanni already

touched upon.90

We return to the question of the Great King.91 In his account of

Clovis’ victory, Gregory of Tours appears to pitch the Frankish king

against a single, supreme Alamannic ruler, leading his army into a

mighty battle which he seems well able to win. His initial success

forces Clovis to extreme measures; and his death precipitates the

surrender of his people. He is a powerful Wgure. Some who believe in

his existence have sought support for the idea that by the end of the

86 PLRE 2.791–2; James (1988: 69); Wood (1994: 39); MacGeorge (2002: 102 n. 88).

87 On ‘Zulpich’seeGeuenich (1997a: 78–82, 85–6), (1998); with Wood (1994: 45–6);

Runde (1998: 680 n.122); Pohl (2000: 32); and below 344–6.

88 See Martin (1997b: 167–9 and Figs 170a–b), for archaeological evidence to

suggest the movement of high-status Alamannic families down the Middle Rhine and

into the Cologne Bight. However, as Martin remarks, these were clearly few in

number and may well have been political refugees.

89 Thus contra Keller (1993: 101–2). Cf. Pohl (2000: 31).

90 Above 48, 168; Keller (1993: 92–3); Geuenich (1997a: 14), (1997c: 145); Pohl

(2000: 103–4); Lotter, Bratoz

, and Castritius (2003: 111); Castritius (2005: 196–9):

these Danubian Suebi were descended from Marcomanni.

91 Above 123.

The Fifth Century 335

Wfth century the Alamanni were capable of producing paramount

kings in the identi Wcation of Gibuldus of Passau and Gebavultus of

Troyes as one and the same person. As Childeric was consolidating

Frankish power in northern Gaul, the writ of an Alamannic ruler ran

from Bavaria to Champagne. As Clovis outdid Childeric, so might

another Alamannic leader have outperformed Gibuldus/Gebavultus

in Wghting in 396/7 or, as part of a subsequent campaign of resistance

to the Franks, c.500. The current communis opinio is, however, with

Geuenich, against the notion of an Alamannic Great King. The case for

a common identity for Gibuldus and Gebavultus is weakened by the

strong possibility that the exploits of the second may be no more than

a doublet of those of the Wrst.92 And the more that Gregory’s account

of Clovis’ defeat of the Alamanni in 496/7 is seen to relate to religious

aVairs the less important becomes the battle itself and the Alamannic

leader who W gured in it. He becomes one of many such leaders in the

long process of Alamannic defeat and decline.93 From beginning to

end, independent Alamannia was politically fragmented.

However, the idea of some sort of consolidation of Alamannic

power structures should not be dismissed entirely out of hand. The

Alamanni of the Wfth century were not the same as those of the

fourth. Since 406/7 and, especially, 455, much had happened

(including the emergence of a Thuringian kingdom94), and we should

allow that Alamannic society and politics may have changed in

response.95 As Geuenich says, unlike those of other Germanic

peoples of the Roman west, we know nothing of the pedigrees and

marriage alliances of Wfth-century Alamannic royal families; and we

have uncovered no great Alamannic royal palace or graves.96 How-

ever, this last is suspect as an argumentum ex silentio. The sumptuous

burial of Childeric, found at Tournai in 1653, is unique and was

uncovered only by accident.97 To be airbrushed out of history is the

92 Quellen 4, 13 n.1; Geuenich (1997a: 74).

93 For the debate see Geuenich (1997a: 72–5, 83–5), (1997c: 144–8), (1998: 423–

30). Contra: Martin (1997b: 167); Lotter, Bratoz

, and Castritius (2003: 119–21).

94 See Pohl (2000: 40–1).

95 Keller (1998: 585–91); below 343.

96 Geuenich (1997a: 85–6), (1997c: 148), (1998: 431).

97 James (1988: 58–64); Brulet (1997: 164–70); Mu

ller-Wille (1997: 206–11); Pe

rin

and Kazanski (1997).

336 The Fifth Century

frequent fate of the loser. Clovis’ success was unexpected: without it

there would have been no great Merovingian kingdom, and Gregory

of Tours would never have had occasion to seek out details of Clovis’

ancestors. More substantive is that if we accept a protracted process

of Frankish conquest of Alamannia in the later Wfth and early sixth

centuries we must accept that the various Alamannic leaders with

whom Clovis contended were no mean Wgures. They held out against

one of the ablest generals of the age for a considerable period of time,

and must therefore have been able to draw on signiWcant resources.

This indicates a fair degree of political sophistication. Indeed, the

authority eventually exercised over the Alamanni by Clovis, and else-

where by the Ostrogothic king, Theoderic, suggests the availability of a

signiWcant level of local control.98 The Alamanni had produced major

leaders in the shape of Chnodomarius and, especially, Macrianus.

These had been broken by Rome, but after the middle of the Wfth

century Roman control had vanished.99 As Wolfram says, again noting

the case of Franks under Clovis, w ith the right talents and in the right

circumstances, in this period a Great King could emerge out of

nowhere;100 so why not from the Alamanni? I propose that in this

period the Alamanni may have had not one Great King but several

‘greater kings’,101 some possibly from a new centre of power on the

Neckar.102

Given the nature of the literary sources, likely major economic and

social developments are accessible only through archaeology.

Though, as ever, diYcult, this yields intriguing signs of continuity

and change.

The late fourth and Wfth centuries were the great age of the

‘Ho

hensiedlungen’. These continued and Xourished in their role as

regional centres down to the middle of the Wfth century. 103 Their

existence may have operated against the emergence of super-regional

power bases, that is against the emergence of a Great, or even greater

kings, down to around 455.104

98 Hummer (1998: 16); for Theoderic see below 344.

99 Above 330–1.

100 Wolfram (1998: 614–15).

101 Cf. Keller (1998: 591).

102 Below 343.

103 Steuer (1990: 196, 202); Pohl (2002a: 174).

104 Cf. above 124; below 344.

The Fifth Century 337