Drinkwater J.F. The Alamanni and Rome 213-496 (Caracalla to Clovis)
Подождите немного. Документ загружается.
reduction in the Rhine garrison. Over 30 years, the number of legions
stationed there was cut from six to four.96
Peace, or lack of major activity, persisted under Trajan (98–117),
Hadrian (117–38) and Antoninus Pius (138–61). It ended under
Marcus Aurelius (161–80). Again, we hear of border warfare on the
Rhine and upper Danube with, in 161 and 162, raiding by Chatti and,
in 170–5, by Chatti and Chauci.97 However, the greatest conXicts of
the period were the Marcomannic Wars. This is the name given to the
wars fought by the emperor Marcus Aurelius along the Danubian
frontier in the period 168–80. The reconstruction of the sequence of
these events and their precise chronology is contentious.98 In what
follows, I rely on A. R. Birley.99
On succeeding Antoninus Pius, Marcus Aurelius took as his co-
emperor his step-brother, Lucius Verus. Almost immediately there
was trouble in the east, and Lucius led a successful Parthian war from
162–6. Even before this war was over, Marcus was preparing a new
campaign on the Danube. This began in 168, but was interrupted by
the death of Lucius early in 169. It recommenced in 170, when a major
Roman operation provoked barbarian incursions deep into the
Empire: Quadi and Marcomanni reached northern Italy and Costoboci
attacked Eleusis. Marcus and his generals acted swiftly to drive out the
invaders, and then attacked them on their home territory. From 171–
5 there was Wghting against Marcomanni, Quadi and, further east,
Iazyges, in which the Romans did well. However, this had to be given
up on news of rebellion in the east, led by Avidius Cassius. After
Cassius’ defeat, Marcus returned to the Danube in 178, accompanied
by his son and heir, Commodus. He repeated his earlier success. The
period 179–80 saw the defeat of the Quadi, the submission of the
Iazyges and the wearing down of the Marcomanni. Final victory
seemed in sight; but early in 180 Marcus fell ill and died. Commodus
made peace with the barbarians and ended the Wghting.
The traditional interpretation of the Marcomannic Wars is heavily
inXuenced by our sources’ positive assessment of Marcus Aurelius,
96 Drinkwater (1983a: 60, 62).
97 Historia Augusta, V. Marci 8.7–8; V. Didii Iuliani 1.6–8. Drinkwater (1983a: 76).
98 Garzetti (1974: 484).
99 Birley (1987: 140–210, 249–55).
28 Prelude
and by our knowledge of what was to happen to the Roman Empire
from the third century. In short, Marcus is seen as the Wrst Roman
emperor to face the rising tide of Germanic migration, and the one
who strengthened imperial defences just enough to hold it back for
another couple of generations.100 Much stress is laid on the attack on
Italy of 170: the Wrst by foreigners since the Cimbri in 102 bc, and a
harbinger of the Iuthungian and Alamannic attacks of 260,101 and of
Alaric’s sack of Rome, the very embodiment of the ‘Germanic threat’,
in 410.102 However, it is possible to paint a diVerent picture. A
principled man set on doing his public duty can be just as dangerous
to ordinary people as a self-seeking rogue.
Marcus came to power relatively old (he was born in 121), and with
no militar y experience. Indeed, as Birley observes, it is likely that until
168 he had never left Italy.103 He delegated the eastern war to the
younger Lucius, and Lucius did well. Had he been an ordinary
Roman em per or, Marcus would have felt obliged to emulate his col-
league’s success; but Marcus was more than that. Apart from his own
personal standing, he will have been conscious of the demands of duty:
to undertake himself what he asked of subordinates; and, above all, to
do right by the Empire over which he ruled.104 There is also his state of
mind. His private me morandum book, now kno wn as his ‘Medit a-
tions’, shows that he was now absorbed with issues of mortality, and by
the conviction that his own death was at hand.105 Such reXections are
likely to have been stimulated by chronic illness and more recent
incidents such as the arrival of the plague and the deaths of Lucius
and his own young son in 169. In such a state, he may have seen war not
just as a means of demonstrating his abilities as leader but also as a test,
an act of self-sacriWce, through which he might expiate whatever it was
that was causing Heaven to inXict such suVering on his Empire. This
would explain the attention that was given, once hostilities had begun,
to various ‘miracles’ which showed the return of divine favour.106 It is
100 Garzetti (1974: 506).
101 Below 52, 70–2.
102 Cf. Birley (1987: 249).
103 Birley (1987: 156).
104 Cf. Birley (1987: 214, 216); Lendon (1997: 115–16).
105 Cf. Birley (1987: 185, 214, 218–20).
106 Cf. Garzetti (1974: 493); Birley (1987: 252–3).
Prelude 29
no surprise, therefore, that Marcus was preparing a Germanic war even
before the end of th e Parthian war. There is no ne ed to suppose that th is
war was a response to a major new threat. Germanic raiding was a
feature of frontier life.107 It could never have been diYcult for any
Roman ruler looking for a war to Wnd a justiWcation for action in one of
these incursions. But in 170 things went badly wrong, and instead of
Romans attacking barbarians, barbarians attacked the Empire.108
The basic cause of this trouble has usually been sought in the
beginnings of the ‘Vo
¨
lkerwanderungszeit’: the ‘Period of Migration’.
As Goths moved from Jutland to the Black Sea, they pressured
peoples to their south and west, who in turn pressured peoples on
the borders of the Roman Empire, taking them, as Birley puts it, to
‘breaking point’ and forcing them to cross the imperial border.109
Local factors have also been taken into account in explaining the Wnal
push of Marcomanni, Quadi, etc. into the Empire, such as rising
population, caused by the more sedentary life created by their being
held at the imperial frontier.110 However, the domino eVect remains
the major explanation: for a spell, the Empire was swamped by the
rising Germanic tide. However, what happened was short-term raid-
ing, not invasion and conquest.111 The supposed discovery of female
warriors among the barbarian dead in northern Italy should be
treated as sensationalism, not good history, and should not lead us
to conclude that whole peoples were on the march in a search for
land.112 The Empire’s ‘enemies’ at this time consisted of communities
which had long lived close to it in relative harmony. If their young
men did turn to what amounted to intense banditry in this period,
the cause may have been much more close to hand than pressure
from the interior. Given the close relationship between the Danube
communities and the Roman army, the mass movement of troops to
107 e.g. below 121, 207.
108 Again, chronology of this is much disputed: see, e.g. Garzetti (1974: 486, 490)
gives ad 167, possibly 169; Birley (1987: 250–2) gives ad 170.
109 Garzetti (1974: 483–4); Birley (1987: 148 (quotation), 155, 253–5). Cf. the
criticisms of Burns (2003: 235).
110 Birley (1987: 148–9).
111 Garzetti (1974: 486).
112 Thus, contra Birley (1987: 169). Cf. Zonaras 12.23: the discovery of women’s
corpses among Persian dead in 260.
30 Prelude
the east to Wght the Parthian war and their return infected with
disease could well have spelled social and economic disaster.113 Raid-
ing was an alternative means of sur vival, and would have been made
easier by the depletion of the frontier garrisons. But, in the mid-160s,
raiding drew imperial attention and resulted in a build-up of imper-
ial forces clearly intent on aggression. No wonder that some Quadi
and Marcomanni panicked and attacked Wrst. And no wonder that
Roman forces, expecting relatively easy victory, were surprised and
let them through. The invasions of Italy and Greece should never
have happened. But the basic weakness of the Empire’s Germanic
enemies is shown in how quickly Rome was able to restore the
situation, and take the oVensive. Again, Rome was the aggressor
here: Marcus Aurelius was set on annexing new territories.114
The Wrst phase of the Marcomannic Wars, from 168–75, was
therefore not fought by an Empire on the defensive. The decisive
incident was the ‘invasion’ of Italy—humiliating in itself and raising
memories of the Cimbri, and even of Brennus. The second phase,
from 178–80, followed exactly the same pattern, as Marcus, obstin-
ately fulWlling his duty, returned to Wnish what had been interrupted
by revolt. Now, however, there was an additional, more conventional,
factor in play. The young (he was born in 161) Commodus’ formal
acceptance as heir had been accelerated to meet the challenge of
Avidius Cassius, and as part of this process he needed to be accepted
by the army. The second phase, like the Wrst, was characterized by
Roman aggression and military superiority.115 Birley describes the
conXict as ‘a grim and sordid necessity’; but it was necessary only
because Marcus regarded it as such. There appear to have been
numerous opportunities for a negotiated peace, but these he chose
to ignore.116 Again as in 16 bc, there was a ‘threat’ on the Danube in
the 170s but it was Roman, not Germanic.
Following Commodus’ departure from the northern frontier, there
was peace with the Germani for several decades. Even prolonged
Roman political instability and civil war, prompted by the assassin-
113 Burns (2003: 229–35); cf. below 49, n. 36.
114 Birley (1987: 142, 163, 183, 207, 209, 253); Isaac (1992: 390–1).
115 See Birley (1987: 207–9).
116 See, e.g. Birley (1987: 155–6, 169–70).
Prelude 31
ation of Commodus at the end of 192 and leading to the accession of
Septimius Severus (193–211), did not provoke signiWcant raiding. In
213, Severus’ successor, his son, Caracalla (211–17), made contact
with ‘Alamanni’ on the middle Main, which is when the following
chapter begins.
The preceding account of Roman dealings with western Germani is
based on the literary sources. It can be supplemented from archaeo-
logical research. The course of the Augustan and Tiberian campaigns
can be reconstructed through the study of right-bank forts and
marching-camps, etc.117 However, these were temporary and, with
the exception of certain sites (in particular, that of Waldgirmes), will
not be considered further here. More important is what came later—
above all the development of the security line across the re-entrant
angle of the upper Rhine and upper Danube, now termed the ‘Upper
German/Raetian limes’. 118
Although imperial troops w ithdrew to the left bank of the Rhine
after the Varus disaster, it is unlikely that they abandoned the right
bank for any length of time. As early as Tiberius, a permanent Roman
bridgehead was established in Chattian territory at Hofheim, oppos-
ite Mainz; and the Roman military presence in this region appears to
have been strengthened under Gaius.119 Under Claudius, this area
saw the opening of a silver mine near Wiesbaden, beginning imperial
exploitation of transrhenish resources that continued into the
fourth century.120 The main advance, however, came under the
Flavians. Vespasian moved into the area north of the lower Main,
the Wetterau, and into the Neckar valley, and drove a road from the
Rhine to the Danube. In the course of his conXict with the Chatti,
Domitian built a set of defended roads in the Wetterau. He consoli-
dated Roman presence on the Neckar, and threw open the land
117 The classic study remains Wells (1972).
118 Cf. below 50. I recognize the great problems involved in using limes as a
synonym for defended frontier line (see Isaac (1992: 408–9)) but, like other equally
questionable terms (such as ‘Romanization’), it is too useful a piece of historical
shorthand to give up.
119 Drinkwater (1983a: 36); Barrett (1989: 131). Schnurbein (2003: 104) suggests
the re-establishment of a Roman presence in the area before the death of Augustus.
120 Tacitus, Annales 11.20.4: in agro Mattiaci. Cf. below 133.
32 Prelude
between the Rhine and the Neckar, the Agri Decumates, to colonists
from Gaul.121 By 100, the protection of Roman territory south of the
Main was centred on the Neckar and the Swabian Alp.
This remained the case until Antoninus Pius, when major change
came with the projection of the Upper German border c.30 km to the
east, its integration with the Raetian border, and the combined fron-
tier being delineated by a palisade. This line, from Miltenberg-Ost, via
Lorch, to the Danube near Eining, was given its Wnal shape w ith the
addition of a rampart and ditch backing onto the palisade in Upper
Germany, and the replacement of the palisade with a stone wall in
Raetia122 (Fig. 2). The Upper German/Raetian limes was now in
existence. Its most signiWcant aspect is its vulnerability. Laid along
two intersecting straight lines, it totally disregarded tactical topo-
graphy, and appears to have been created more for show than for use.
It could have oVer ed little direct resistance to a large force of determined
attackers.123 As a military installation it could meet, at best, in
Luttwak’s terms, ‘low intensity’ threats posed by ‘transborder inWltra-
tions and peripheral incursions’, that is raiding, banditry.124 In short,
it appears to reXect a lack of danger from the Germani whom it faced.
How does such an assessment of the Upper German/Raetian limes
square with what we can say about these Germani? Roman militar y
activity may be reconstructed from both texts and artefacts. Ger-
manic settlement is accessible only through archaeology, and early
Germanic archaeology is, as ever, very diYcult.125 However, it has
long been thought that, although there was a strip of Germanic
settlements along the right bank of the Lower Rhine, their main
concentrations lay much further in the interior.126 Of special signi-
Wcance for Alamannic studies is the fact that surprisingly few such
settlements have been found skirting the Upper German/Raetian
limes. To explain this, it has been suggested that the Romans simply
121 Drinkwater (1983a: 57–60); below 126.
122 For further details see Baatz (1975: 16, 58–62); Drinkwater (1983a: 60–3);
Carroll (2001: 38–9).
123 Cf. Isaac (1992: 414); Whittaker (1994: 84); Witschel (1999: 204).
124 Luttwak (1976: 66); cf. Baatz (1975: 44); Drinkwater (1983a: 64). For low-level
raiding as banditry, see Elton (1996a: 62).
125 See Schnurbein (2000: 51–3); cf. above 4, and below 112.
126 e.g. Baatz (1975: 49–50 and Fig. 29) (¼Whittaker (1994: 89 and Fig. 22)).
Prelude 33
appropriated all the best land in the region.127 Thanks to recent
research, we are now in a position to check both the general settle-
ment pattern and the thinking it has generated.
A stretch of early Germanic settlement along the northern bank of
the middle Lahn has recently been subjected to close scrutiny128
(Fig. 3). Like the neighbouring Wetterau,129 this region appears to
have been subject to continuous settlement from the Celtic Late Iron
Age into the Germanic period. It was not inferior land. Indeed, it has
been suggested that Wnds of local non-Roman coinages at major
Roman sites (in particular, Waldgirmes) built in the region under
Augustus indicate the presence of relatively sophisticated Germanic
Area of box
Fig. 3 The Lahn and Wetterau regions [after Abegg-Wigg, Walter, and
Biegert (2000: 57 Fig. 1)].
127 Whittaker (1994: 89).
128 For this,and the bulk of what follows, seeA. Wigg (1997)and Abegg-Wigg, Walter,
and Biegert (2000). For a useful English summary of the work, see A. Wigg (1999).
129 Lindenthal and Rupp (2000); Scholz (1997); Stobbe (2000).
34 Prelude
communities that had adopted important aspects of Celtic economic
and social behaviour.130 The earliest Germanic presence was short-
lived—disappearing early in the Wrst century ad after the Varus
disaster, along with the Roman garrisons. Users of Germanic
‘Rhine–Weser’-style artefacts returned at the end of the same century,
that is at the same time as the development of the Wetterau limes.
The key archaeological sites are at Naunheim, Wettenberg and
Lahnau. Ceramic remains indicate that their inhabitants imported
goods from the Empire, whose border lay, at its closest point, just
15 km (9 statute miles) to the south.
However, the middle Lahn remains one of only three known areas
of Germanic settlement on the northern half of the Upper German/
Raetian limes (the c.300-km (180-mile) stretch between Rheinbrohl
and Jagsthausen).131 Of the rest, the two nearest to the study area lie
20 km (12 statute miles) to the north-west and 40 km (24 statute
miles) to the north-east respectively. In other words, the most recent
archaeological work appears to conWrm sparsity of Germanic settle-
ment close to the limes. Next to be noted is that importation of
Roman-style artefacts took around 50 years, deep into the second
century, to become signiWcant. And, even then, the proportion of
such artefacts found on the Lahn sites is small, with little metal and
glass and relatively small amounts of pottery. There is no sign of
‘technology transfer’, that is styles and methods of production of
Roman artefact forms being taken up by Germani. Further, analysis
of Roman and indigenous wares suggests that though Germani may
have imported some exotic foodstuVs, they made no fundamental
changes to the content, preparation and presentation of their meals.
This is conWrmed by study of the animal and plant remains, which
indicates direct continuation of Late Iron Age agricultural patterns—
quite diVerent from the neighbouring Wetterau where, for example,
cattle gave way to pigs as Romanized settlers from Gaul further
developed the land.132 Finally, despite their proximity to the Empire,
there is no indication that the middle-Lahn Germani were par t of the
imperial monetary economy.
130 Wigg (2003). For reconstruction of the evolution of Wigg’s ‘Germanic’ coinage
from Celtic models, and its destruction by Drusus’ campaigns, see Heinrichs (2003).
131 Abegg-Wigg et al. (2000: 56).
132 Stobbe (2000: 216).
Prelude 35
Similar mixed results have come from study of the middle Main,
centred on the prehistoric settlement of Gauko
¨
nigshofen, south of
Wu
¨
rzburg133 (Fig. 4). Again we see continuity of settlement from the
Celtic Late Iron Age into the early Germanic period, with the Wrst
Germanic settlement being contemporary with Augustan expansion
into Germania, and close contact between these earliest Germani and
the Roman army. Again, too, there was a break at the beginning of the
Wrst century, followed in the mid-second century by the arrival of
Rhine–Weser artefact-users at the same time as the Wnal advancing of
the limes line. The same period also saw Germanic settlement on the
Tauber.134 Roman products—represented by ceramics—were
imported from Upper Germany and Raetia, though more from
the former than the latter. Roman inXuence has been detected in
the possible use of coins in trade, and in local house-building tech-
niques.135 There is also a case for believing that local Germani
introduced Roman cattle/oxen as draught animals. And this is one
of the areas where, in the third century, there was production of
Roman-style wheel-thrown pottery (‘the Haarhausen–Eßleben ex-
periments’).136 On the other hand, the imported pottery was
restricted to certain main lines and, oddly, in the light of the later
strategic importance of the river, appears to have come by land, from
the limes line, via Osterburken or Jagsthausen, not by water, up the
Main.137 Again there is little sign of such inXuence on the preparation
and serving of food and drink (everyday and as part of rituals). And
the pottery experiment was exceptional and short-lived, probably
because it was dependent on Roman captives.138 Further aWeld across
Franconia, Germanic sites of the period of the Early and High Roman
Empire generally show no long-term adoption of Celto-Roman
agricultural methods. Even at Gauko
¨
nigshofen, the introduction of
Roman stock produced no permanent improvement in local breeds.139
133 See, generally, for what follows Steidl (1997), (2000).
134 Frank (1997); Steidl (2000: 107, Fig. 9).
135 Coins: Steidl (1997: 76). Cf. below 135, for the continuation of coin use into
the fourth century. House-building: Steidl (2000: 104–5).
136 Below 92.
137 Steidl (2000: 106); cf. below 99, 113–15, 131–2, 190.
138 Steidl (2000: 106, 109). Below 92–3.
139 Benecke (2000: 253).
36 Prelude
Fig. 4 Alamannia.
Prelude 37