Drinkwater J.F. The Alamanni and Rome 213-496 (Caracalla to Clovis)
Подождите немного. Документ загружается.
Rhine–Weser Germani in the Tauber/Upper-Main–Regnitz areas.131
Generally cooperative with Rome, these must have been more
aVected by early Elbgermanic raiding than the Empire, and so had
every reason to resent their tormentors. The derogatory interpret-
ation of ‘Alamanni’ could have been picked up by Asinius Quadratus
from Roman troops who participated in Caracalla’s campaign.
Indeed, we should never forget the existence of an interested and
powerful third party. There is now strong acceptance of Wenskus’
view that ‘Alamanni’ and ‘Alamannia’ were far more useful to the
Romans, who needed to classify their neighbours on an imperial
scale, than to the people to which they were attached, who, in their
everyday lives, would have seen themselves more as members of small
communities.132 This raises the question as to whether the name,
though Germanic in form and origin, was wholly Germanic in
the manner in which it was deployed. ‘Alamanni’ and ‘Alamannia’
come from imperial sources: again, we have no idea what the south-
western Elbe-Germani called themselves in the third century.133 In
such circumstances, a single name could have been taken up and
manipulated by the one people that needed to come to terms with
the new phenomenon—the Romans.134
There was an important precedent. Much earlier, the Romans had
hit upon the name Germani as a useful catch-all for those living over
the Rhine.135 Three centuries on, from 261/2, imperial oYcials
would have needed a convenient term for territories now occupied,
under suVerance, by outsiders. ‘Former Upper Germany and Former
Raetia’ would have been a mouthful, and an unwelcome reminder of
Roman withdrawal. ‘Marcomanni’ and ‘Marcomannia’ had long
been in existence to act as models.136 It would have required no
more than an awareness of the existence of ‘Alamanni’, of the insult-
ing interpretation of this name by Germanic neighbours, and a
131 Frank (1997); Schach-Do
¨
rges (1997: 95–7 and Fig. 81).
132 Wenskus (1961: 502); Geuenich (1997a: 19), (1997c: 144); Hummer (1998: 16);
Steuer (1998: 276); Pohl (2000: 103).
133 Above 4.
134 Cf. Gillett (2002b: 120) on the role of ‘Roman labelling’ in the case of Goths,
Vandals and Franks.
135 Pohl (2000: 3–4, 50–6).
136 Cf. Okamura (1984: 287): third-century Italians erroneously applied the name
Marcomanni to early Alamanni.
68 Arrival
rudimentary knowledge of early German on the part of a Roman on
the Rhine/upper Danube frontier to come up with ‘Alamannia’ as a
disparaging but accurate description of the lost lands. The most likely
place for this would have been on the Rhine, well away from the less
numerous but more distinctive Iuthungi; otherwise, instead of
‘Alamannia’, we might have had ‘Iuthungia’.137 ‘What is that? Where
are we/they being sent?’ ‘That is we/they are being ordered to,
Alamannia, rabble-land’ (the haunt of many war-bands, not just
those of ‘authentic’ Alamanni).138 ‘Who are they?’ ‘They—all those
between the upper Rhine and upper Danube139—are the Alamanni,
the rabble,’ would have followed immediately after.
Like Francicus (‘Conqueror of the Franks’), Alamannicus (‘Con-
queror of the Alamanni’) took some time to appear as an imperial
title alongside the established Germanicus. The Wrst known bearer of
Alamannicus is Constantine II, no earlier than 331.140 A little later,
in 354, when according to a carefully worded and detailed inscription
it was taken by Constantius II, it was glossed as Germanicus.141 This
must reXect the Roman establishment’s continuing reluctance to
adopt neologisms and especially, in this case, those derived from
military vocabulary. As Elton notes, until the Wfth century, barbarian
tribal names were used by Rome to indicate where her enemies came
from, not precisely who they were, and to clarify and enhance
victories over relatively small groupings.142 Until the fourth decade
of the fourth century, emperors and their advisers may have felt that
in most cases this task was fulWlled perfectly adequately by Germani/
Germanicus.143
137 Cf. above n.125.
138 For the early imperial shift of emphasis from peoples to the lands they
occupied (in line with classical ethnography: Amory (1997: 18)), see Burns (2003:
178–9).
139 Below 83, 125.
140 CIL 3.352¼7000; ILS 6091; MAMA 7.305; Quellen 6, Inscriptions no. 61.
Francicus appears Wrst in 361: below 252.
141 Germanicus Alamamnicus (sic) Maximus, Germ. Max. etc. (note the repetition
of Germanicus). CIL 3.3705; ILS 732; Quellen 6, Inscriptions no. 64. Cf. Zotz (1998:
390) and Bleckmann (2002: 168 and n. 94) on the odd mixing of titles.
142 Elton (1996a: 37).
143 See Bleckmann (2002: 167–9, esp. 167–8) on the likely politico-dynastic
reasons for the innovation: Constantine I wanted to honour Constantine II, but to
reserve Germanicus to himself. Cf. below 198–9.
Arrival 69
In the period 261/2, therefore, as the result of the partition of
the western Roman Empire between Postumus and Gallienus, the
Upper German/Raetian limes was abandoned.144 Elbgermanic war-
rior-bands were able to move into former imperial land east of the
Rhine and north of the Danube and were tempted to cross these
rivers and plunder Gaul and Italy. During the next decade or so such
raiding was a chronic problem. For the most part it must have been
on a relatively small scale, but when conditions were favourable
(usually due to imperial political strife), such groups could amal-
gamate and undertake more daring attacks. Though probably not
involving huge numbers of warriors, these could, thanks to the
availability of the imperial road network and the lack of defences in
the interior, be wide-ranging and must have caused great panic.145
Literary sources tell of a Frankish incursion deep into Spain.146
Archaeological evidence for Alamannic attacks into Gaul, one at
least of which reached southern Aquitania, is provided by Wnds of
Roman metalware at Hagenbach and Neupotz, near Speyer. These are
interpreted as booty lost by raiders while attempting to cross the
Rhine.147
In the military and political chaos which followed the capture of
Valerian, two large groups made their way into Italy, one through
Gaul and the other by way of Raetia.148 Neither was a ‘tribe’ or even
an ‘army’. Those who took the Raetian route appear to have called
themselves Iuthungi. Later generations called the others Alamanni.
Unorganized and disorganized, both groups would have faced diY-
culties arising out of success. With no commissariat and carrying
quantities of booty, both inanimate and animate, they were vulner-
able to counter-attack.149 Probably, as was to happen again in 268
and 271, neither went far into the peninsula.150 Few probably got
144 Cf. Witschel (1999: 248 n.56), though, at 211–12, he sees the process as
gradual, taking from 260 to 265.
145 Witschel (1999: 205, cf. 214–15).
146 Above 52.
147 Hiernard (1997); Nuber (1997: 64 and Figs 42–3); Carroll (2001: 138 and
plate 23).
148 i.e. following Jehne (1996: 189–92, 196–201), though with a revised chron-
ology: cf. above 54–5.
149 Okamura (1984: 273, 289).
150 Below 71, 76.
70 Arrival
beyond the Po valley. Zosimus’ report of the Senate’s raising a militia
to defend Rome suggests reaction to panic rather than to assault.151
There is no sign of serious Wghting near the city, and it is likely that
the main body of attackers began to withdraw irrespective of any
challenge from this quarter. Gallienus returned from putting down
usurpation on the Danube to deal with the Alamanni at Milan.
Imperial propaganda made this a major set-piece battle, with
10,000 Romans supposedly overcoming 300,000 barbarians.152 As
Okamura observes, such Wgures are ‘impossibly high’. The Alamanni
who fell to Gallienus probably amounted to no more than several
thousand. Given the Wssile nature of their alliance, a similar number
probably simply avoided battle and Xed homewards.153 This is prob-
ably what also happened w ith the Iuthungi: evading the imperial
forces, they dispersed and headed north. One group was caught and
destroyed the following year by Genialis.
The next major incursion involving Alamanni that we know of
began in 268. This again occurred in strained political circumstances,
following the assassination of Gallienus and his replacement by
Claudius II. There was a battle near Lake Garda, fought probably
in 269.154 Claudius may have left its direction to one of his generals,
possibly Aurelian.155 The eastern location of the conXict suggests
Iuthungian involvement.156 However, this is unlikely since Dexippus
does not make Aurelian refer to this battle when he reminds the
Iuthungi of recent Roman successes at the start of his reign.157
Claudius did not last long as emperor. He died of disease late in
the summer of 270 and his death was, as usual for the time, followed
by civil war. His brother, Quintillus, succeeded him but was soon
challenged by his senior general, Aurelian, and quickly overthrown.
Aurelian was emperor by the end of 270. The reign of Aurelian
returns us to the fragment from Dexippus recording this emperor’s
dealings with the Iuthungi.
151 Zosimus 1.37.2. Cf. Jehne (1996: 190–1): those who threatened Rome were
Iuthungi. On the basic reliability of Zosimus’ report see Bleckmann (2002: 166).
152 Zonaras 12.24. See now Bleckmann (2002: 164).
153 Cf. Okamura (1984: 273): ‘Gallienus . . . intercepted one large group . . . .’
154 Epitome 34.2. Okamura (1984: 275–6); Watson (1999: 43).
155 Watson (1999: 43).
156 Okamura (1984: 283); Watson (1999: 220).
157 Below 74.
Arrival 71
The chronology, location, course and even number of Aurelian’s
barbarian wars are very diYcult to reconstruct from the available
sources, which are obscure and, in the case of the Historia Augusta’s
‘Life’ of Aurelian, suspect. The question of his Iuthungian wars has
recently been reviewed by Saunders and Watson.158 Both note that
pioneering scholarship had Aurelian wage two Iuthungian wars, for
the Wrst of which Dexippus’ fragment 6 was regarded as sole evi-
dence.159 However, the communis opinio of the last 60 years or so has
been that there was only one such conXict, of which Dexippus’
fragment 6 describes the Wnal stage, on the Danube.160 But modern
thinking appears to be returning to the older version of events, either
in its traditional form (Saunders) or with modiWcations (Watson).
Watson proposes that one Iuthungian invasion, undertaken jointly
with Alamanni, took place under Claudius II and was defeated by the
emperor and Aurelian on Lake Garda in 269. It was this war, and the
defeat they suVered while working with Alamanni, that later pro-
duced the Iuthungian ambassadors’ deprecating assessment of the
value of ‘mixed’ forces. Watson’s second, far more serious, incursion
happened under Aurelian, in 271, while the emperor was absent
Wghting Vandals in Pannonia. The Iuthungi invaded alone, through
Raetia and into Italy. Dexippus’ fragment 6 describes events in the
middle of the war, which occurred on the Metaurus not the Danube.
In support of his interpretation of the 271 invasion, Watson adduces
the Augsburg inscription. He dates the events it describes to 261, and
uses the fact of a Iuthungian invasion of Italy in this year to argue
that the Iuthungi were capable of mounting a second major attack on
Italy in 271.
Problems arise as to how much credence can be given to Dexippus’
account of the encounter. Fragment 6 consists for the most part of
two speeches (one Iuthungian, one Roman) and therefore inevitably
falls under the suspicion of having resulted more from Dexippus’
Thucydidean reconstruction of what ought to have been said in the
circumstances than from what was actually spoken. In this respect, it
has to be conceded that Aurelian’s response to the Iuthungi, though
158 Saunders (1992); Watson (1999: 216–20 (with 43, 50–2)).
159 See, e.g. Homo (1904: 61–4, 73–6).
160 See, e.g. Alfo
¨
ldi (1939: 156–7).
72 Arrival
much longer than their speech, is far less substantial. It is just a
reaction to the points they raise, heavily rhetorical and couched in
terms of the usual Roman disparagement of barbarian impudence
and unreliability. It has to be used with great caution. Elsewhere, on
the other hand, rich circumstantial detail oVers hope of some degree
of authenticity. For example, Dexippus’ account of Aurelian’s theat-
rical staging of the meeting rings true. In other respects, such as the
walling of Rome, his great triumph and his building of a magniWcent
temple to the Sun, Aurelian demonstrated a keen understanding of
the power of display.161
As far as the speeches are concerned, the Iuthungian envoys’
insistence on the distinction between the people they represented
and other Germanic raiders, however interpreted, and its apparent
acceptance by the imperial authorities, also rings true. Likewise, the
envoys’ recollection of recent cash payments from the Empire and
their claim that the renewal of these subsidies would bring about a
mutually beneWcial peace (6.5, 7) appear to be more than rhetoric.
These points are substantial and unexpected, give the Iuthungi
remarkable courage in dealing with Aurelian, and excite the emperor’s
most barbed comments. It was probably because of them that the
passage was excerpted from Dexippus’ Scythica for inclusion in a
tenth-century compilation of extracts ‘On foreign embassies to the
Romans’ (De legationibus gentium ad Romanos).162 In addition, the
possibility of discrete groups of Elbgermanic warriors serving in
third-century imperial forces has been strengthened by Wnds of
Roman artefacts and gold coins in rich burials in Thuringia (the
so-called Haßleben–Leuna grave-group, from type-sites near Erfurt).
These have been interpreted as indicating payment by Postumus and
his successors to leaders of war-bands.163 Other groups could have
agreed to serve the Central Empire. The Iuthungian claim therefore
161 Cf. Watson (1999: 176–7).
162 Saunders (1992: 320).
163 Drinkwater (1987: 225), from Werner (1973); Steuer (1997a: 150–1), (1998:
290–1). See also Martin-Kilcher (1993), concerning the rich burial of a Germanic
oYcer in Cologne c.300, and others elsewhere along the Rhine and Danube, which
she associates with the Haßleben-Leuna graves; likewise Pirling (1993: 109), for
‘Frankish’ graves at the fort at Gellep, between Cologne and Xanten, from the third
century. Cf. below 145.
Arrival 73
deserves to be examined closely; but Wrst it must be set within a
precise chronological and historical context.
The events described in fragment 6 must come relatively early in
Aurelian’s reign. This derives from what Aurelian is made to say to
the Iuthungi. The emperor makes no allusion to any earlier victory of
his over Iuthungi. As recent examples of the success of Roman
eYciency over barbarian brute strength, he cites only defeats suVered
by ‘Scythians’ and ‘Galmionoi’(?) (6.11: the text appears corrupt). I
follow Saunders in interpreting these as victories against Goths and
Alamanni won by Claudius II.164 I also follow Saunders in placing
Aurelian’s Wrst brush with the Iuthungi as emperor before his war
with the Vandals (as described in fragment 7), and so accept that at
the start of Aurelian’s reign there were two Iuthungian ‘wars’, the
second of which followed the Vandal campaign.165 Inverted commas
are important here. In assessing the strength of Alamanni and
Iuthungi, one must establish what is meant by ‘war’: full-scale inva-
sion or a series of irritating raids? Taking seriously the idea of some
earlier alliance with the Empire, I propose that the state of hostility
described in fragment 6 was not a major attack on Italy. The Iuthungi
refer to destructive incursions into Raetia and deep into Italy as a
boast and as a threat, but not as their current activity. In other words,
contrary to the usual interpretation, they should not be seen as
claiming that they had just penetrated Italy with 40,000 Wrst-class
cavalrymen, but rather that this is what they had achieved on a
previous occasion.166 Their point is not that they currently dispose
of such men. On the contrary, at 6.6–7, they concede the weakness of
their position: having divided their strength and having been
defeated in a major battle they now face a larger force conWdent of
victory. Rather, the large numbers that they throw about represent
the avenging force that Aurelian will bring down on himself if he
destroys them. Such numbers are, of course, grossly inXated.167
Much of this may have been put into the Iuthungian envoys’ mouths
by Dexippus to explain the apparent eVrontery that follows.
164 Saunders (1992: 322–3).
165 Contra Watson (1999: 320–1), I accept Saunders’ argument that fragments
6 and 7 are probably in chronological order and should not be transposed.
166 e.g. Okamura (1984: 297); Saunders (1992: 311, 327); Watson (1999: 51, 218).
167 So Saunders (1992: 324 n. 64).
74 Arrival
Here we have the substance of what was said. What was really
happening is to be found in the envoys’ justiWcation of their recent
actions. These, they say, did not amount to raiding, but to foraging.
This they regarded as legitimate: they were not aware that they were
acting wrongly until they found themselves attacked by Aurelian
(6.5). What they say suggests movement into Raetia and just over
the provincial border into Italia Transpadana, but no further. It
appears that the imperial government had earlier come to terms
with the Iuthungi after a major incursion by them into Italy, agreeing
to pay them to keep the peace and to act as allies if called upon. This
agreement had been kept, until payments were ended by the Romans,
without consulting the Iuthungi, who then simply helped themselves.
For the Iuthungi to get used to such subsidies, the major invasion
and consequent agreement must have come much earlier. I propose
that this incursion was that of 260. An agreement with Iuthungi who
had escaped Roman retribution was made by Gallienus, who is
known to have favoured such arrangements.168 This was maintained
by Claudius but collapsed in the turmoil follow ing his death in 270.169
Aurelian probably dealt with the ensuing trouble immediately on
coming to power, that is without Wrst going to Rome.170 This need
not have taken long, since disturbance was conWned to Raetia.
Indeed, if, as subsequently happened in the case of the Vandals,
Iuthungian cavalry had been recruited directly into Roman service,
some of the unrest described in fragment 6 may have begun on
imperial soil. In other words, if the cessation in the supply of
money aVected pay, not subsidies, what occurred may have been
more in the nature of a mutiny than an attack.
168 Aurelius Victor, Caesares 33.6; Epitome 33.1; Zosimus 1.30.2f. Cf. Witschel
(1999: 218 and n.195).
169 Cf. Okamura (1984: 297): Gallienus or Claudius. Castritius (1998: 353, 356
n.28), following A. Radno
´
ti ((1967), Die germanischen Verbu
¨
ndeten der Ro
¨
mer
(Deutsch–Italienische Vereinigung Heft 3), Frankfurt/Main: 1–20) and the implica-
tions of the Augsburg inscription, suggests that the Iuthungian claim to a Roman
alliance was based on an understanding with Postumus and his successors as ‘Gallic’
emperors, i.e. that the Iuthungi may be equated with the Haßleben-Leuna warriors
(above 73). I consider this unlikely, but since I have not been able to obtain a copy of
Radno
´
ti’s publication I cannot be certain in this respect.
170 Cf. Saunders (1992: 319): from Sirmium.
Arrival 75
Having dealt with the Iuthungi, and established his own position,
Aurelian moved to Pannonia to face raiding Vandals. As Saunders
says, the statement in Dexippus’ fragment 7.4, that he had to inter-
rupt this campaign to return to Italy because the Iuthungi were
attacking ‘again’, marks the beginning of the second ‘war’. However,
this, too, may have been no major conXict. The sole trustworthy
account says simply that Aurelian was victorious in three battles in
Italy, at Piacenza, on the river Metaurus just beyond Fanum
Fortunae, and Wnally near Pavia.171 Such an itinerary is odd, involv-
ing an engagement on the upper Po, a dash south-eastwards along
the Via Aemilia and Via Flaminia down to the Umbrian coast, and
another chase back along the same roads to a Wnal confrontation just
west of the site of the Wrst engagement. Even if Aurelian had won the
initial battle at Piacenza he had not won the war. News of barbarians
heading south, albeit still on the northern side of the Apennines,
must have shaken the peninsula, and may have caused a wave of
opposition to Aurelian in Rome.172 The highly imaginative writer of
the Histor ia Augusta developed this series of events to produce an
exciting but confused tale of Roman defeat, despair and Wnal vic-
tory.173 This reconstruction has been developed and reinforced by
those who take Dexippus’ fragment 6 as evidence for a massive
contemporary Iuthungian invasion of Italy and locate its events in
the course of a single great ‘Iuthungian war’.
The evidence of the Historia Augusta is best ignored. Aurelian’s
swift movement from north to south and back again may be inter-
preted as his having to deal with relatively small bands of highly
mobile opponents—troops of horsemen—able to do a fair degree of
damage and diYcult to bring decisively to battle.174 On the other
hand, the fact that the Iuthungi were unable to cross the Apennines
to the south or take a route other than the one they had come in by to
the north hints that Aurelian had the overall situation under control.
171 Epitome 35.2: Iste in Italia tribus proeliis victor fuit, apud Placentiam, iuxta
amnem Metaurum ac fanum Fortunae, postremo Ticinensibus campis.
172 Zosimus 1.49.2; Saunders (1992: 319).
173 Historia Augusta, V. Aureliani 18.3–21.4. Cf. Saunders (1992: 314): that the
Historia Augusta author ‘extensively reworked’ his material at this point (though
without accusing him of outright fraud).
174 Cf. Okamura (1984: 290).
76 Arrival
It seems that, as a demonstration of Wrmness at the start of his reign
and as a punishment for making trouble, he kept to the threat that he
would not renew earlier agreements which he had made to the
Iuthungi on his Wrst confrontation with them; and this was still
causing unrest. The wider Alamanni were not involved. Zosimus’
statement that in the second war Aurelian faced ‘the Alamanni and
their neighbours’ and defeated them decisively on the Danube
appears to derive from ignorance and carelessness: speciWcally, the
common confusion between Alamanni and Iuthungi and the trans-
posing of events of the W rst war to the second.175 Much the same may
be said of Aurelius Victor’s laconic observation that Aurelian
returned to Italy to deal with raiding Alamanni after campaigns
against Palmyra and Persia and before his defeat of Tetricus in
Gaul, in which events of 271 are transferred to 273.176 It is likely
that, following his victory at Pavia, Aurelian moved directly to deal
with the troubled situation in Rome.177
The threat posed to Italy by Elbgermanic peoples was small. By
Aurelian’s reign the Alamanni had ceased their raiding and the harm
done by the Iuthungi was relatively minor and conWned to the north.
After Aurelian’s victories, the peninsula suVered no further barbarian
incursions until the early Wfth century. This security resulted from
the dedication and skill of individual emperors, and from the enor-
mous underlying strength of the Roman Empire.178 However, it is
also a measure of the nature of the Germanic menace. The raiders
were not migrating tribesmen but bands of young warriors exploiting
current Roman weaknesses to win fame and booty: not ‘the
Alamanni’ or ‘the Iuthungi’ but ‘Alamanni’ and ‘Iuthungi’. In chaotic
conditions they enjoyed success but against coordinated Roman
resistance they were helpless. That Aurelian regarded his Iuthungian
victories as small beer may be reXected in the fact that though he
took the title Germanicus—‘Conqueror of the Germans’—he did not
advertise it widely on his coinage.179 However, such diYdence may
also be interpreted as embarrassment. Though Alamanni and
175 Zosimus 1.49.1; cf. Saunders (1992: 318).
176 Aurelius Victor, Caesares 35.1–3; cf. Dufraigne (1975: 169–70).
177 Watson (1999: 218–19).
178 Drinkwater (2005: 62–3).
179 Okamura (1984: 300).
Arrival 77