EIRENE

S T UDI A
GR AECA
ET
L A T I NA

LV / 2019 / I–II

Centre for Classical Studies

Institute of Philosophy of the Czech Academy of Sciences, Prague
 E I R E N E . S T U DI A GR A EC A E T L AT I NA , LV, 2019, 231–257

CONSTANTINE’S MILITARY OPER ATIONS

AGAINST THE GOTHS AND THE SAR MATIANS
IN 332 AND 334
STANISLAV DOLEŽAL

In 332, the Roman Emperor Constantine inflicted a crushing defeat upon the
Goths. Allegedly, he had come to the rescue of the Sarmatians which were hard-
pressed by the Goths at the time. Two years later, however, he may have defeated
the very Sarmatians whom he had previously rescued from the Goths. Both
campaigns took place beyond the lower Danube, on the barbarian territory.
This is the gist of the story, told by a number of scholarly publications, such as
The Cambridge Ancient History: “In his later years Constantine had also resumed
military operations, first in 332 against the Goths and then in 334 against the
Sarmatians, notwithstanding the fact that it was they who had called in the Ro-
mans in 332.”1 Some, but by no means all, Constantinian scholars narrate an
aftermath of this episode: that Constantine not only defeated the enemy but
also “transferred the majority of the Sarmatians into Roman territory.”2
There are some questions regarding these apparently successful military
operations of Constantine: questions, which are frequently left unaswered in
scholarly monographs – if they are raised at all. For example, why did Constan-
tine attack the Goths in the first place? Why did he change his mind two years

1
CAMERON 2008, 105. Unfortunately, not every scholarly work concurs even with this sum-
mary. Cf. JONES 1972, 194–195: “In 331 the Goths broke through the Danube frontier, but the
Roman armies, led by the Caesar Constantine, the Emperor’s eldest surviving son, soundly
defeated them in the following year and chased them back across the frontier.” First, we do
not know of any Gothic invasion in 331 and the campaign of 332 seems to have been fought
solely on barbarian soil.
2
ODAHL 2013, 256. The Constantinian scholars mentioning the campaigns but ignoring the
transfer of the Sarmatians include POTTER 2013, 285, POHLSANDER 2004, 78, or BARNES 2014, 165.

231
STANISLAV DOLEŽAL

later, attacking his former allies? Did the Roman military successes against the
Goths and Sarmatians result in a reconquest of some areas north of the lower
Danube? Were the Sarmatians forcibly resettled in the Empire or did they seek
asylum on the Roman territory? What was their fate and did they leave any
traces? I believe that these questions can only be answered after thorough ex-
amination of all available sources.

Sources

Speaking about the sources, we possess a sufficient number of them for these
episodes. From a chronological point of view, they form three separate groups:
1) Near contemporary sources, mostly written shortly after Constantine’s
death: the anonymous Consularia Constantinopolitana,3 Eusebius’ Vita Constan-
tini (which was left unfinished by its author when he died in May 339),4 and
the anonymous Origo Constantini imperatoris.5
2) Sources from the 4th century, written in or after 360: Liber de Caesaribus
by Aurelius Victor, Caesares by the Emperor Julian, Breviarium by Eutropius,
Jerome’s Chronicon and Ammianus Marcellinus’ Res gestae. Except for Ammi-
anus and, to a lesser degree, Jerome, these sources provide little information
about the Sarmatian transfer, although they do speak of Constantine’s Gothic
campaign in 332.6
3) The 5th, 6th or 7th century sources, such as Rufinus, Orosius, Socrates
Scholasticus, Sozomen, Zosimus or Jordanes, add basically nothing new; either

3
Although it was compiled much later as a whole, its entries covering Constantine’s reign
were probably written in late 337 or early 338, see BURGESS 1993, 193, and BLECKMANN 2007,
22 (“shortly after the death of Constantine”).
4
CAMERON – HALL 1999, 3.
5
According to BARNES 2014, 27, it is “a brief, sober and reliable account of the emperor
written very shortly after his death in 337, which is the most accurate and well-informed non-
-Christian source for Constantine that survives”. For differing views, cf. BLECKMANN 2007, 26,
who argues that the Origo “could very easily have been written at the end of the fourth cen-
tury”, KÖNIG (1987, 19) who simply maintains that the work was written sometime in 337–414,
or ODAHL 2013, 3–4, who suggests that the author was a former soldier in Constantinian
armies of the mid-fourth century, and wrote a short political biography of Constantine in his
retirement in the genre of the epitomes but with a more positive tone to Constantine.
6
EUTROPIUS, Breviarium ab Urbe condita, X,7 (Nam etiam Gothos post civile bellum varie profligavit,
pace iis postremum data, ingentemque apud barbaras gentes memoriae gratiam conlocavit), cf. AURELIUS
VICTOR , Liber de caesaribus, 41,13 and 41,18.

232
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

they shortly summarise Constantine’s victories over the Goths and Sarmatians7
or bring confusion into the matter.8
Unfortunately, the events of 332 and 334 are not covered by any extant
panegyric. If any speeches in the period 324–337 were written, “they failed to be
included in the collection of the Gallic panegyrics”, that is to say, the Panegyrici
Latini.9 Not even Libanius or Themistius authored such speeches, and we find
only one reference to Constantine’s victories on the Danube in Julian’s Panegyric
in honour of Constantius.10
Let us examine closely the first group of sources. The Origo Constantini im-
peratoris not only gives a brief account of both campaigns but it is the only
source that states the number of the Sarmatians that were removed from their
homeland to the Empire in 334: “Then he took up arms against the Goths,
and gave assistance to the Sarmatians, who have begged for it. Thus, through
Constantine Caesar, nearly 100,000 died of starvation and cold. Then he ac-
cepted hostages, among whom was a son of the king Ariaric. Thus, when peace
had been confirmed, he turned against the Sarmatians, who had proved to be
of doubtful loyalty. But all the slaves of the Sarmatians rebelled against their
masters, and when the latter had been expelled, Constantine willingly accepted
them, and distributed more than 300,000 people of all ages and both sexes
throughout Thrace, Scythia, Macedonia and Italy.”11 The original text is so im-
portant that it must be quoted in full:

Deinde adversum Gothos bellum suscepit et implorantibus Sarmatis auxilium tulit. Ita
per Constantinum Caesarem centum prope milia fame et frigore extincta sunt. Tunc et
obsides accepit inter quos Ariarici regis filium. Sic cum his pace firmata in Sarmatas
versus est, qui dubiae fidei probabantur. Sed servi Sarmatarum adversum omnes domi-

7
For example, RUFINUS, Historia ecclesiastica, X,8 (Interea Constantinus, pietate fretus, Sarmatas,
Gothos, aliasque barbaras nationes, nisi quae vel amicitiis vel deditione sui pacem praevenerant, in solo
proprio armis edomuit) or OROSIUS, Historia adversus paganos, VII,28 (Praeterea multas gentes diversis
praeliis subegit … Mox Gothorum fortissimas et copiosissimas gentes in ipso barbarici soli sinu, hoc est
in Sarmatarum regione, delevit). Cf. SOCRATES SCHOLASTICUS, Historia ecclesiastica, I,18; SOZOMEN,
Historia ecclesiastica, I,8,8 and II,34,4.
8
ZOSIMUS, Historia nova, II,31,2; JORDANES, Getica, 111–112; ISIDORE OF SEVILLE , Historia de
regibus Gothorum, Vandalorum et Sueborum, 5.
9
BLECKMANN 2007, 24.
10
See note 45.
11
The Origin of Constantine, translated by J. Stevenson, in: LIEU – MONTSERRAT 1996, 48. There
is an obvious mistake (“Ariaric, son of the king”), which I emended.

233
STANISLAV DOLEŽAL

nos rebellarunt, quos pulsos Constantinus libenter accepit et amplius trecenta milia
hominum mixtae aetatis et sexus per Thraciam, Scythiam, Macedoniam, Italiamque
divisit.12

We will return to these statements in due course, now let us focus on the other
contemporary sources.13
Although Eusebius is rather vague on Constantine’s Danubian campaigns,
he still provides valuable information. According to him, Constantine “was the
first to subjugate the Gothic and Sarmatian tribes which had never before learnt
to serve”, subduing them sometimes militarily and sometimes “pacifying the rest
by reasonable negotiations”. The campaign of 334 is narrated in more detail:

As to the Sarmatians, it was God himself who thrust them under the feet of
Constantine, defeating men who gloried in their barbaric mentality in the fol-
lowing way. When the Goths attacked them, the masters armed their servants to
repel their enemies. But when the slaves had won, they turned their arms against
their masters and drove them all from their own land. The masters found no
other safe refuge than Constantine alone. He knew the meaning of rescue, and
received them all as subjects in Roman territory. Those who were suitable he
enrolled in his own forces; to the rest he apportioned land for cultivation of the
means of subsistence, so that they acknowledged that the disaster had turned
out good for them in that they enjoyed Roman liberty instead of barbaric bes-
tiality. Thus God bestowed upon him victories over all the nations, so that of
their own accord all sorts of barbarian tribes were willing to submit to him.14

The third source close to Constantine’s reign, the anonymous Consularia Con-
stantinopolitana, is very succinct yet basically preserves the same information
for the year 332: “Pacatianus and Hilarianus being the consuls, the Goths were
defeated by the Roman army in the lands of the Sarmatians on 20 April” (Pa-
catiano et Hilariano. His consulibus victi Gothi ab exercitu Romano in terris Sarma-
tarum die XII k. Mai.), as well as the year 334: “Optatus and Paulinus being the
consuls, the whole nation of the slaves of the Sarmatians expelled their masters
into the lands of the Romans” (Optato et Paulino. His consulibus Sarmatae servi

12
Origo Constantini imperatoris, 6,30–32.
13
Cf. KULIKOWSKI 2007a, 101, who, referring to the Origo, puts the number of Sarmatian
refugees to 30,000. The same error is found in KULIKOWSKI 2007b, 362.
14
EUSEBIUS, Vita Constantini, IV,6 (translation: CAMERON – HALL 1999, 155).

234
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

universa gens dominos suos in Romaniam expulerunt).15 Of the sources from 360s
(Aurelius Victor, Julian and Eutropius), only Victor makes a passing remark
about the Goths and Sarmatians being defeated – not removed from their lands
but defeated – and on top of that, he wrongly places these events to the time
when Constantine elevated his son Constans to the rank of Caesar (25 Decem-
ber 333).16 Jerome, who compiled his Chronicon in around 380, actually made a
short remark about the episode which, interestingly, includes the names of the
two Sarmatian tribes involved: “The Sarmatians called Limigantes, having gath-
ered a force, expelled their masters, who are now called the Argaragantes, onto
Roman soil” (Sarmatae Limigantes, dominos suos, qui nunc Argaragantes vocantur,
facta manu, in Romanum solum expulerunt).17 Although no other source records
the name Argaragantes, the name Limigantes is actually confirmed by our last
source to speak about this episode, Ammianus Marcellinus.18
Although the portion of Ammianus’ Res gestae covering Constantine’s reign
is not extant, Ammianus occasionally mentions Constantine19 and once even
refers to his own previous narration about Constantine’s eastern policy, presum-
ably about his planned Persian campaign.20 Thus it is all the more surprising
that the story about the transfer of Sarmatians in 334, which is recounted by
Ammianus in retrospect, does not even contain Constantine’s name. Ammi-
anus includes the story in his narration of events of 358, when Constantius II
defeated the Sarmatians and the Quadi on the Danube, primarily in the vicin-
ity of Pannonia Valeria and Pannonia secunda, but also near Pannonia prima and
Moesia prima.21 However, Ammianus’ narration is fundamentally different from
what we hear from sources already quoted above:22

15
BURGESS 1993, 236.
16
AURELIUS VICTOR , Liber de caesaribus, 41,13 (Et interea Gothorum Sarmatarumque stratae gentes,
filiusque cunctorum minor, Constans nomine, Caesar fit).
17
JEROME , Chronicon, I,28. In the manuscripts of Chronicon, there are also found variants
Arcaragantes and Ardaragantes.
18
A MMIANUS M ARCELLINUS, Res gestae, XVII,13,1 (Limigantes, Sarmatas servos).
19
A MMIANUS M ARCELLINUS, Res gestae, XIV,1,2; XV,5,33 and elsewhere.
20
A MMIANUS MARCELLINUS, Res gestae, XXV,4,23 (non Julianum sed Constantinum ardores Parthicos
succendisse… ut dudum rettulimus plene).
21
A MMIANUS M ARCELLINUS, Res gestae, XVII,12–13.
22
A MMIANUS M ARCELLINUS, Res gestae, XVII,12,17–20: Quibus ordinatis translata est in Sarmatas
cura, miseratione dignos potius quam simultate. Quibus incredibile quantum prosperitatis haec attulit
causa: ut verum illud aestimaretur, quod opinantur quidam, fatum vinci principis potestate vel fie-
ri. 18 Potentes olim ac nobiles erant huius indigenae regni, sed coniuratio clandestina servos armavit in
facinus. Atque ut barbaris esse omne ius in viribus adsuevit, vicerunt dominos ferocia pares, sed numero

235
STANISLAV DOLEŽAL

These affairs once set in order, his attention was turned to the Sarmatians, who
were deserving rather of pity than of anger: and to them this situation brought
an incredible degree of prosperity; so that the opinion of some might well be
deemed true, that fortune is either mastered or made by the power of a prince.
The natives of this realm were once powerful and noble, but a secret conspiracy
armed their slaves for rebellion; and since with savages all right is commonly
might, they vanquished their masters, being their equals in courage and far su-
perior in number. The defeated, since fear prevented deliberation, fled to the
Victohali, who dwelt afar off, thinking that to submit to protectors (considering
their evil plight) was preferable to serving slaves. Bewailing this situation, after
they had gained pardon and been assured of protection they asked that their
freedom be guaranteed; whereupon the emperor, deeply moved by the injustice
of their condition, in the presence of the whole army called them together, and
addressing them in gracious terms, bade them yield obedience to none save
himself and the Roman generals. And to give them restoration to freedom an
increase of dignity, he set over them as their king Zizais, a man even then surely
suited for the insignia of a conspicuous fortune and (as the event showed) loyal;
but no one was allowed, after these glorious achievements, to leave the place,
until (as had been agreed) the Roman prisoners should come back.23

This is a very interesting moment. Up to Ammianus, all the sources – with the
possible exception of Aurelius Victor – are in agreement that there had been an
expulsion of Sarmatians in 334 from their lands and that Constantine had ac-
cepted the refugees in the Roman territory. Now Ammianus asserts that these
refugees, in fact, “fled to the Victohali”, and knows nothing of Constantine’s
clemency. He did not even deem to tell his readers about the year of expul-
sion: it simply happened some time in the past (olim). And yet we can be sure
that Ammianus describes the same event as Jerome or the Origo. His story has
all the essential ingredients: he speaks of the Sarmatians, of the rebellion of
their slaves (whom he calls Limigantes and even distinguishes two branches of
them, the Amicenses and Picenses), and of the expulsion of the Sarmatian “mas-

praeminentes. 19 Qui confundente metu consilia ad Victohalos discretos longius confugerunt, obsequi de-
fensoribus (ut in malis) optabile quam servire suis mancipiis arbitrati: quae deplorantes post impetratam
veniam recepti in fidem poscebant praesidia libertati eosque iniquitate rei permotus inspectante omni
exercitu convocatos allocutus verbis mollioribus imperator nulli nisi sibi ducibusque Romanis parere
praecepit. 20 Atque ut restitutio libertatis haberet dignitatis augmentum, Zizaim regem isdem praefecit,
conspicuae fortunae tum insignibus aptum profecto, ut res docuit, et fidelem, nec discedere quisquam post
haec gloriose gesta permissus est antequam, ut placuerat, remearent nostri captivi.
23
Translation: ROLFE 1935a, 379.

236
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

ters” (dominos).24 However, he still differs substantially from the Origo, Eusebius,
Consularia Constantinopolitana, and Jerome himself. Why?
For one thing, Ammianus was born around the time of Constantine’s opera-
tions beyond the Danube and may not even have heard of the correct sequence
of events. Perhaps he did not know that there were scores of Sarmatians set-
tled throughout the Empire. In 358, after 24 years, these multitudes may have
become indistinguishable from the other provincials of, as the Origo tells us,
Thrace, Scythia, Macedonia and Italy. And in around 390, when Ammianus was
writing his Res gestae, the very identity and memory of these Sarmatians may
have simply faded away. If, for all practical purposes, these scattered Sarmatian
groups have disintegrated and ceased to exist as coherent ethnic groups, Ammi-
anus may have not realized at all that there was a massive transfer of Sarmatian
refugees into the Empire in 334.
Even so, the other sources seem to be equally unaware of any runaway Sar-
matians who had been received by the Victohali. Quite the contrary: Eusebius
stressed that Constantine received all of the Sarmatian “masters”. This may be
one of his panegyrical exaggerations but it does, at least, suggest that the Sar-
matians who fled to the Victohali were not very numerous. Either they escaped
the attention of our sources entirely or they were deemed as not worthy of any
mention at all.25
Who actually were these Victohali? Apart from Ammianus, only Eutropius
and Historia Augusta mention this tribe, but as a mere name in a list of peoples.
Eutropius dryly remarked that Trajan “conquered Dacia, once he had defeated
Decebalus, and a province beyond the Danube was created in that territory,
which now belongs to the Taifali, Victoali and Tervingi”.26 The name slightly
changed to “Victoali”, and the tribe apparently lived in Dacia around the time
when Eutropius was writing his book (c. 369). Historia Augusta27 is more eloquent,

24
In contrast to Jerome, Ammianus knows no Argaragantes; he only speaks of “free Sarma-
tians” (A MMIANUS M ARCELLINUS, Res gestae, XVII,13,19: “Liberi Sarmatae”) to distinguish them
from the Limigantes.
25
Scholarly works usually follow this discrepancy and tend to disregard the division of
Argaragantes into the two groups. One of few exceptions is MÓCSY 2014, 288 (“In the war of
332–4 Constantine had been satisfied with receiving into the empire some of the Argaragantes
who had been expelled by the Limigantes, and letting the rest of them disappear from Roman
sight”).
26
EUTROPIUS, Breviarium, VIII,2,2 (Daciam Decebalo victo subegit, provincia trans Danubium facta
in his agris, quos nunc Taifali Victoali et Tervingi habent). Manuscripts also have variants Victuali
and Victohali (see Eutropii Breviarium ab Urbe condita, ed. H. DROYSEN, MGH AA II, Berlin 1879,
p. 136).
27
Historia Augusta, Vita Marci, 14,1 (profecti tamen sunt paludati ambo imperatores et Victualis et

237
STANISLAV DOLEŽAL

yet more confusing: “Clad in the military cloak the two emperors (i.e., Marcus
Aurelius and Lucius Verus) finally set forth, for now not only were the Victuali
and Marcomanni throwing everything into confusion, but other tribes, who had
been driven on by the more distant barbarians and had retreated before them,
were ready to attack Italy if not peaceably received.”28 Another passage from the
same biography29 is even more obfuscatory: “Then, from the borders of Illyricum
even into Gaul, all the nations banded together against us – the Marcomanni,
Varistae, Hermunduri and Quadi, the Suebians, Sarmatians, Lacringes and Buri,
these and certain others together with the Victuali, namely, Osi, Bessi, Cobotes,
Roxolani, Bastarnae, Alani, Peucini, and finally, the Costoboci.”30 The name in
now spelled “Victuali”, and the tribe allegedly lived somewhere in the area of
the middle Danube, at the start of the Marcommanic Wars, c. 168 CE. However,
Ammianus’ assertion about the Sarmatian refugees having fled to the Victohali
could still be broadly compatible with Eutropius and even with Historia Augusta.
Allowing for its migration to the east in the course of two hundred years, the
tribe may have lived in the second half of the fourth century in West Dacia –
perhaps in a region east of the middle Tisza and at any rate not very far from
Pannonia.31 One more question remains: should we consider these Victohali to
be a Germanic tribe? Is it probable that the expelled Sarmatians sought refuge
with a Germanic people? It is not impossible, especially when Ammianus says
that “fear prevented deliberation”.32

Marcomannis cuncta turbantibus, aliis etiam gentibus, quae pulsae a superioribus barbaris fugerant, nisi
reciperentur, bellum inferentibus).
28
Translation: M AGIE 1991, 169.
29
Historia Augusta, Vita Marci, 22,1 (gentes omnes ab Illyrici limite usque in Galliam conspirave-
rant, ut Marcomanni Varistae Hermunduri et Quadi Suebi Sarmatae Lacringes et Buri hi aliique cum
Victualis Osi Bessi Cobotes Roxolani Bastarnae Alani Peucini Costoboci).
30
Translation: M AGIE 1991, 187. A small variation is found in E. Hohl’s edition of Historia
Augusta (Scriptores Historiae Augustae, I: Vita Marci Antonini Philosophi Iuli Capitolini, Leipzig
1965), namely, Sosibes, Sicobotes instead of Osi, Bessi, Cobotes.
31
Cf. SCHWENDEMANN 1923, 83, who placed them at the Vistula (“Die Victuali scheinen weiter
östlich gegen die Weichsel gesessen zu haben”).
32
SCHÖNFELD 1911, 262, considered the Victohali to be a part of Vandals (“wahrscheinlich
ein vandalischer Stamm”) but he also admitted that their name could be Celtic. WOLFRAM
(1990, 63), by asserting that most of the Sarmatian refugees “were admitted into the Empire
by Constantine, while a minority sought refuge with the Vandals”, thus implicitly identifies
Victohali with the Vandals, too.

238
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

The Campaign of 332

Let us now consider the military aspects of the Gothic and Sarmatian campaigns
in 332 and 334. It is obvious that Constantine’s campaigns against the Goths
and the Sarmatians required large-scale invasions of barbarian territory, at least
in the case of the Goths. However, we do not know of any military necessity
for these invasions. For whatever reason, Constantine decided that a clear dem-
onstration of Roman power beyond the Danube was desirable. Therefore, he
must have made the necessary preparations all the more thoroughly, to avoid
a setback. Especially he needed to muster a sufficient number of men. Presum-
ably, he had ample time to do this; but how large was Constantine’s army at
that time and how many troops could he safely earmark for the invasion of
Gothia (as the regions roughly corresponding to the old Roman Dacia may
have been known in Constantine’s day)33 without compromising the Empire’s
security elsewhere? Warren Treadgold, relying heavily on A. H. M. Jones,34 esti-
mated that around (or indeed, after)35 324, the Roman troops stationed in the
eastern portion of the Empire alone may have numbered around 282,000 men
in total, of which some 83,000 men belonged to the elite field troops.36 That
should have been enough to crush either the Goths or the Sarmatians. How-
ever, Constantine could not have committed a major part of these elite units
to the campaign beyond the Danube unless the eastern frontier with Persia was
secured. And it appears that it indeed was. “Licinius may have fought a cam-
paign on the eastern front in 313 or 314; from then on, as far as we know, there
was a standstill in fighting until near the end of Constantine’s reign.”37 In 336,

33
According to OROSIUS (Historia adversus paganos, I,2,52–53), Europe stretches ad Danuvium
quem et Histrum vocant, qui est a meridie et ad orientem directus Ponto accipitur. Ab oriente Alania est,
in medio Dacia ubi et Gothia, deinde Germania est.
34
JONES 1964, 679–686.
35
The author of this article asked Professor Treadgold personally about his vague expression
“around 324” and this is the correction. The difference is very important, given the casualties
the Roman army as a whole must have suffered in 324 in the civil war between Constantine
and Licinius.
36
TREADGOLD 1995, 58–59. It should be noted that the size of the Roman army in the fourth
century is still a grave problem. Some scholars departed from Jones’ and Treadgold’s estimate
of about 600,000 for the whole Empire at the end of the fourth century. LEE (2007, 77) argued
for 500,000, and perhaps most radically, POTTER (2004, 457) reasoned that “the army of the
fourth century was either smaller than, or about the same size as, the Severan army” (i.e.,
between 240,000 and 360,000 men).
37
MILLAR 2001, 207.

239
STANISLAV DOLEŽAL

Constantine had already responded to an invasion of eastern provinces by the

Persians and was assembling a huge army to invade the Sassanid Empire in 337
when he suddenly died on 22 May near Nicomedia. But in the years 324–336
he was free to handle matters in Gothia and Sarmatia as he wished. And Con-
stantine seems to have made thorough preparations. In 328, he opened a new
stone bridge across the Danube, apparently a marvel of Roman engineering.
It linked the Roman city of Oescus (Gigen, Bulgaria) with Sucidava (Celeiu,
Romania) in Gothia.38 About that time, Constantine also built a fortress called
Daphne in the barbarian territory further downstream.39 These measures clearly
point to a future active policy towards the former Roman Dacia. What was
Constantine’s intention?
We must seriously consider the possibility that Constantine dreamt of cam-
paigns which would equal the exploits of Trajan: to reconquer Dacia which
Trajan had conquered, and to annex Mesopotamia which Trajan, albeit briefly,
had occupied. When in 336 Constantine dismissed the envoys of the Persian
king, he did so possibly for the same reason for which he had built the bridge
across the Danube: he may have wished to be remembered as one of the great
conquerors.40 Or perhaps his motives were less ambitious: securing the borders
against possible runaways from an inner war in Dacia and showing the northern
barbarians that he is still in charge and to be reckoned with.41 Nine years earlier,
in 323, on the verge of the civil war with Licinius, Constantine had encroached
his rival’s territory not only to destroy a barbarian king called Rausimod and
to defy Licinius, but also to show the world that he was now in control of the
whole Danubian frontier.42 In 332, this reason could still be valid.
Be that as it may, the sequence of events was as follows. After Constan-
tinople was dedicated on 11 May 330, Constantine used to reside there quite

38
AURELIUS VICTOR , Liber de caesaribus, 41,18 (pons per Danubium ductus; castra castellaque pluribus
locis commode posita); Epitome de caesaribus, 41,13 (pontem in Danubio construxit); Chronicon Paschale,
a. 328.
39
Daphne is mentioned by A MMIANUS M ARCELLINUS (Res gestae, XXVII,5,2) but see DEN BOEFT
– DRIJVERS – DEN HENGST – TEITLER 2009, 107–108, for debate on the location of Daphne.
40
See ODAHL 2013, 261. After all, Constantine’s nephew, the Emperor Julian, in his pamphlet
Caesares compared his uncle with Trajan (see note 103). For further details and discussion see
BLECKMANN 1995, 50–53.
41
KÖNIG 1987, 173–174: “Die Möglichkeit eines Eindringens flüchtiger Sarmaten in römisches
Reichsgebiet oder nur eine Verschiebung des völkischen Gleichgewichtes in dieser Region war
Grund genug, einzugreifen.”
42
For this campaign, see DOLEŽAL 2018.

240
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

often.43 In early 332, however, he travelled towards the Danube. He was in

Marcianopolis on 12 April44 and may have stayed there during the subsequent
Gothic campaign. The Origo Constantini imperatoris clearly says that it was his
son, Constantine II, who was in charge of the military operation beyond the
Danube.45 However, Timothy Barnes did not believe that this actually hap-
pened and argued that the source must refer to the Emperor, not his son.46 The
present author is, too, inclined to believe that it was not the 15-year-old Con-
stantine II, but Constantine himself or his generals who led the armies beyond
the Danube.47 We should bear in mind that it was Constantine himself who in
334 presided over the transfer of the Sarmatians (or at least none of his sons
is named by our sources). The Origo perhaps should be interpreted in that way
that Constantine II was personally present on the territory of the Goths, but
he was heavily guarded by elite troops and that the bulk of the invading army
was actually led by either Constantine or his generals.
On 20 April, there followed the defeat of the Goths on the Sarmatian terri-
tory (in terris Sarmatarum).48 Both Kienast and Barnes agree with the date,49 but
not Otto Seeck who apparently read the source differently and came to the con-
clusion that it must have been February 18.50 Seeck probably trusted the Origo

43
He is attested there several times in 331, see BARNES 1982, 79.
44
Codex Theodosianus, III,5,4; BARNES 1982, 79.
45
Origo Constantini imperatoris, 6,30–31: Deinde adversum Gothos bellum suscepit (i.e., Constan-
tine) et implorantibus Sarmatis auxilium tulit. Ita per Constantinum Caesarem centum prope milia
fame et frigore extincta sunt. The Emperor Julian also asserted (Oratio I,9D) that Constantinus II
“conquered the Getae and secured for us a lasting peace with them”.
46
BARNES 1982, 79. As for Constantine II, Barnes puts him to Cologne, but only for 27 July
332 (BARNES 1982, 85) and even this is disputable because the laws to which Barnes refers (Codex
Theodosianus, II,19,3 and IV,10,1) were clearly issued by Constantine I (Constantinus Augustus),
see PHARR – DAVIDSON – PHARR 1952, 54 and 91. But cf. SEECK 1984, 161, who dated both laws
to 313.
47
Although our sources indicate that Constantine II was born shortly before 1 March 317
(ZOSIMUS, II,20; Epitome de Caesaribus, 41,4; cf. JONES – M ARTINDALE – MORRIS 1971, 223), some
scholars assert that he was actually born in August 316 (BARNES 2014, 102; ODAHL 2013, 263).
However it may be, Constantine II was still 15 in April 332. In spite of that, LENSKI (2016, 43)
opined that “Constantine left combat operations to Constantine II while he himself followed
the war from the frontier city of Marcianopolis”.
48
Consularia Constantinopolitana, s. a. 332: “Pacatiano et Hilariano. His consulibus victi Gothi ab
exercitu Romano in terris Sarmatarum die XII k. Mai” (BURGESS 1993, 236).
49
K IENAST 1996, 300: “Gotensieg des Constantinus Iunior”; cf. BARNES 1982, 79: “victory
over the Goths.”
50
SEECK 1984, 181: “Sieg des Constantinus Caesar über die Gothen.” Cf. SEECK 1911, 4–5.

241
STANISLAV DOLEŽAL

which explicitly says that “cold and starvation killed almost one hundred thou-
sand of the Goths” (centum prope milia fame et frigore extincta sunt) and emended
the Consularia Constantinopolitana accordingly: the incident happened not die
XII k. Mai. but die XII k. Mart. Of course, it is hard to imagine how excessive
cold could be a factor responsible for mass deaths in southeastern Romania in
late April. But Constantine would scarcely have sent a major military force into
barbarian territory in February. Perhaps fame et frigore is nothing more than an
alliteration or perhaps the author of the Origo did not have precise chronologi-
cal information of the decisive battle on the barbarian territory. There is no
reason, however, why we should disregard the Consularia Constantinopolitana on
this point (even if we know that it is otherwise capable of chronogical errors).51
Because of this grave chronological problem, there is no consensus among
scholars about the exact time of the Gothic campaign.52
And yet there is a way, I believe, to reconcile the Origo with the Consularia
Constantinopolitana. I propose this scenario. Constantine carefully planned the
invasion, coordinating his actions with the Sarmatians who had implored his
aid. Then the Emperor with Constantine II and the greater part of the mobile
army (say, 50,000, mostly cavalry) unexpectedly invaded Gothia in late March
or early April 332, using the new stone bridge at Oescus. The main body of
Gothic warriors, led by their king Ariaric, had meanwhile campaigned or wintered
on the Sarmatian territory.53 As the Romans encountered no major resistance,
they swiftly traversed the Gothic territory to the northwest where they linked
up with the Sarmatians, effectively cornering the Goths. The mobile army, as it
moved through Gothia, burned Gothic villages, captured livestock and emptied
granaries. This both helped ease the burden of logistics for the Roman army
and added to the misery of the hapless Goths who hastily retreated to the Sar-

51
According to this source, for example, the battle of Cibalae took place on 8 October 314
(Consularia Constantinopolitana, s. a. 314: Volusiano II et Anniano – his consulibus bellum cibalense
fuit die VIII id. Oct.; BURGESS 1993, 235). Seeck, again, followed this date (1984, 163). Today,
most scholars disregard this date and put the battle to the year 316 instead, see LENSKI 2007,
73–74; CAMERON 2008, 90; POTTER , 2004, 378; POHLSANDER 2004, 41; ODAHL 2013, 162; BARNES
2014, 103; CORCORAN 2000, 7. But cf. JONES 1972, 127, or K IENAST 1996, 299, who believed in
314.
52
Barnes asserted that “Constantine took the field in the late winter of 332” (BARNES 1981,
250), then dated the “victory over the Goths” to April 20, 332 (BARNES 1982, 79); elsewhere he
gives no precise date for this campaign (BARNES 2014, 165).
53
We can only guess what Ariaric was doing on the Sarmatian territory at the time of Roman
invasion. STEPHENSON (2010, 225) opines that he “had sought to expand his territory once again,
now at the expense of his Sarmatian neighbours further west”. It is possible; but a simple raid
seems more likely, given the great distance between the Tisza Sarmatians and the core of the
Tervingian / Visigothic territory. See WOLFRAM 1990, 61; cf. ODAHL 2013, 253.

242
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

matian territory. Bereft of their supplies, suffering from increasing hunger, and
perhaps from unusually cold April weather, they at last offered desperate resist-
ance and were decisively beaten on the Sarmatian territory on 20 April. Sev-
eral thousands may have died in battle and several tens of thousands perished
of starvation afterwards (although hardly “almost 100,000”, as boasted by the
Origo). The situation of the Goths became so desperate that Ariaric had to yield
to the Romans. Constantine accepted his son as a hostage together with many
other noble Gothic youths (tunc et obsides accepit inter quos Ariarici regis filium)
and secured both peace and foedus with the Goths (sic cum his pace firmata).54 He
then returned safely to the Roman territory and probably spent the rest of the
year in Constantinople.55 In memory of his Gothic victory, he had a column
erected there “to Fortune that has returned, on the occasion of the complete
rout of the Goths”56 and initiated ludi Gothici, “the Gothic Games”, which were
usually celebrated in February.57
The numismatic evidence is quite limited. There is a medallion worth one
and a half solidus, depicting the Emperor on reverse with the inscription debel-
latori gentium barbararum; above the mint mark, there is a simple inscription
Gothia.58 Further, there is a solidus with the same depiction and description but
with a captive kneeling between the Emperor and a soldier.59 Two multiples
(worth 3 solidi and 2 solidi, respectively) depict a young prince, presumably Con-
stantine II, with inscriptions principia iuventutis and Sarmatia.60 All these coins
are dated to 332–333 and probably refer to the victory over the Goths in Sar-
matia in 332. In contrast, there seems to be virtually no numismatic evidence
for events in Sarmatia in 334.61

54
The treaty of 332 need not concern us here, but see LENSKI 2002, 122–127 for an useful
overview of scholarly discussion.
55
He is attested there on 17 October, see BARNES 1982, 79. CLAUSS (1996, 52) asserts that on
this occasion the Goths were obliged to deliver 40,000 foederati which were to be settled on
the Roman soil. But the only source of this information is Jordanes (Getica, 111–112) who, as
Kulikowski rightly observed (2007a, 85–86), imposed on the treaty “an anachronistic Byzantine
interpretation”. While it is quite possible that the Goths in 332 supplied some recruits for the
Roman army, no fourth-century source explicitly says that they did.
56
ILS 820 (Fortunae reduci ob devictos Gothos).
57
WOLFRAM 1990, 62.
58
BRUUN 1966, 215 (RIC VII, Trier 531).
59
BRUUN 1966, 216 (RIC VII, Trier 534).
60
BRUUN 1966, 215 (RIC VII, Trier 532 and 533).
61
There are coins struck in Siscia or elsewhere in 334 bearing the rather vague legends Vic-
toria Constantini Avg. and Victor omnium gentium but there are no coins pertaining specifically
to the Sarmatian affairs in 334. See STEPHENSON 2010, 227, and BRUUN 1966, 413.

243
STANISLAV DOLEŽAL

The Campaign of 334

Also in contrast with the Gothic campaign, no source offers any chronology of
events of 334 and accordingly, no scholar has, to my knowledge, tried to con-
strue one. Even descriptions of these events vary greatly in scholarly literature.62
It is primarily because we lack the answer to the main question: was the massive
transfer of the Sarmatians in 334 a forced resettlement or rather an immigra-
tion directed by Roman authorities? The Origo indicates a brewing war between
Constantine and the free Sarmatians in perhaps 332 or 333 when it states that
Constantine “turned against the Sarmatians, who had proved to be of doubtful
loyalty”. Was Constantine planning a punitive expedition? If so, his reasons are
unknown to us. However, the expulsion of the free Sarmatians by their “slaves”
changed everything: former allies (and then perhaps enemies) suddenly became
refugees. As we saw, Eusebius and Jerome say much the same. Aurelius Victor
mentions only a defeat of the Sarmatians and no expulsion but that is perhaps
a mere summary of events as he understood them some thirty years later. Or
perhaps he meant that Constantine defeated the Sarmatian “slaves” to save the
free Sarmatians from total annihilation. If that was the case, Constantine prob-
ably assembled his army in similar fashion as he did in 332, and quickly traversed
Gothia, now perhaps a Roman protectorate, to defeat the Limigantes and to pro-
vide a safe passage for 300,000 Sarmatian refugees to the Empire. But that does
not seem likely. The Emperor’s movements suggest another course of events.
Constantine is attested on 17 June at Constantinople, on 5 July at Singi-
dunum, on 4 August at Viminacium and on 25 August at Naissus.63 It appears
that he very quickly moved from Constantinople to Singidunum and then
stayed for about a month in this important border city. Afterwards, he slowly
retreated to Viminacium and then to Naissus. His next movements are unknown:
he may have stayed in Naissus for some time or headed for his usual residence
Serdica or returned to Constantinople (where he is attested in early 335). It is
hard to escape the impression that waves of Sarmatian immigrants were coming
over to the Empire through the area near Singidunum in early July, and that it
was there where the Emperor directed separated groups of refugees to Thrace,

62
K IENAST 1996, 300: “Vertreibung der Sarmaten aus dem Banat“; SEECK 1984, 182: “Die Sar-
maten werden über die Grenze gedrängt”; BARNES 1982, 79: “campaign against the Sarmatians.”
63
Codex Theodosianus, I,22,2 (17 June), Codex Theodosianus, X,15,2 (5 July), Codex Theodosianus,
XII,1,21 (4 August) and Codex Theodosianus, XI,39,3 (25 August). For these dates, see BARNES
1982, 79; cf. STEPHENSON 2010, 227.

244
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

Scythia, Macedonia and Italy. Constantine alone seems to be in charge of this

operation and we have no hint that any of his sons was involved.
No matter how this transfer of the Sarmatians was carried out, it certainly
was one of the most massive resettlements of barbarian groups into the Em-
pire. The Origo boasts that Constantine received onto Roman soil “more than
300,000 people of all ages and both sexes”. This number is usually accepted
without doubts.64 Is it realistic, though? Although we are not able to verify it,
the usual accuracy of the Origo, the wide range of the geographic area over which
the Sarmatians were settled, and the large number of other confederations press-
ing upon the late Roman Empire may make the figure possible.65 Besides, there
were many precedents for such a huge transfer of barbarians.
In around 1 CE, Sextus Aelius Catus transferred 50,000 Getae from beyond
the Danube into Moesia.66 Under Augustus, 40,000 German captives were re-
settled to Gaul.67 In around 64 CE, Tiberius Plautius Silvanus Aelianus, legatus
pro praetore Moesiae, transferred more than 100,000 “transdanubian” barbar-
ians onto the Roman soil.68 The notoriously unreliable Historia Augusta tells us
that in around 280, the Emperor Probus settled 100,000 Bastarnae in Thrace.69
Luckily, Historia Augusta can be corroborated by Zosimus in this case, even if
he gives no such number.70 Although this transfer probably was a massive one,

64
ODAHL 2013, 256; BARNES 1981, 250; SOUTHERN 2004, 200.
65
Cf. KÖNIG 1987, 14 who did not believe the figure at all, on the basis that Constantine
would have encountered much stronger resistance from the Sarmatians (“da man so mit ca.
50 bis 60,000 Wehrfähigen zu rechnen hätte”). But Constantine did not have to fight the
Sarmatians; they were clearly dediticii and so they were just transferred into the Empire and
split into many groups, as our sources suggest.
66
STRABO, Geography, VII,3,10 (translation: JONES 1924, 209): “In our own times, Aelius
Catus transplanted from the country on the far side of the Ister into Thrace fifty thousand
persons from among the Getae, a tribe with the same tongue as the Thracians. And they live
there in Thrace now and are called Moesi.”
67
EUTROPIUS, Breviarium, VII,9 (XL captivorum milia ex Germania transtulit et supra ripam Rheni
in Gallia collocavit).
68
ILS I, 986 (legatus pro praetore Moesie, in qua plura quam centum milia ex numero Transdanuvia-
norum ad praestanda tributa cum coniugibus ac liberis et principibus aut regibus suis transduxit).
69
Historia Augusta, Probus, 18,1 (Facta igitur pace cum Persis ad Thracias rediit et centum milia
Bastarnarum in solo Romano constituit, qui omnes fidem servarunt). But cf. Historia Augusta, Probus,
13,7, where the anonymous author asserts that Probus killed “almost 400,000 Germans” in
Gaul. His other assertion that Probus gained 16,000 recruits from the German tribes on the
Rhine (14,7) is perhaps more realistic.
70
ZOSIMUS, I,71,1.

245
STANISLAV DOLEŽAL

not all Bastarnae were moved south; a portion of them remained behind and
was transferred onto the Roman soil somewhat later by Diocletian and Galerius.
In addition, Diocletian finished the task that was started by Aurelian: the trans-
fer of the Carpi. Our sources give us the impression that the whole nations of
Carpi and Bastarnae were simply resettled on the Roman soil by the tetrarchs,
and substantial numbers of Sarmatian captives, too.71 However, we must be cir-
cumspect. We do know, for example, that not all of the Carpi were resettled.72
Several Emperors assumed the victory title Carpicus maximus even after 305.73
Some of the Carpi, who were resettled by Aurelian or Diocletian, were still eth-
nically distinct from the other provincial population in 368.74 Those who did
remain outside the Empire were called “Carpo–Dacians”, evidently to be distin-
guished from those Carpi who were settled within the Empire.75 Similarly, not
all the Tisza-Sarmatians were resettled in 334 onto Roman soil.

The Fate of the Roman Sar matians

To return to the Sarmatian refugees. From the Sarmatians that the Origo report-
ed were distributed throughout Thrace, Scythia, Macedonia and Italy, Eusebius
claimed that those suitable for military service were enrolled in Roman units
and the rest used as farmers. Constantine followed an established practice when
he “selected some of their best fighting men, and enrolled them in divisions of

71
OROSIUS, Historia adversus paganos, VII,25,12 (Postea per eosdem duces [i.e., per Galerium et
Diocletianum] strenue adversus Carpos Basternasque pugnatum est. Sarmatas deinde vicerunt: quo-
rum copiosissimam captivam multitudinem per Romanorum finium dispersere praesidia); EUTROPIUS,
Breviarium, IX,25 (Varia deinceps et simul et viritim bella gesserunt [i.e., Galerius et Diocletianus]
Carpis et Basternis subactis, Sarmatis victis, quarum nationum ingentes captivorum copias in Romanis
finibus locaverunt); AURELIUS VICTOR , Liber de caesaribus, 39,43 (Carporumque natio translata omnis
in nostrum solum, cuius fere pars iam tum ab Aureliano erat); Historia Augusta, Aurelianus, 30,4 (Au-
relianus … Carporum copias adflixit); A MMIANUS M ARCELLINUS, Res gestae, XXVIII,1,5 (Maximinus
regens quondam Romae vicariam praefecturam, apud Sopianas Valeriae oppidum obscurissime natus est,
patre tabulario praesidialis officii, orto a posteritate Carporum, quos antiquis excitos sedibus Diocletianus
transtulit in Pannoniam); JORDANES, Getica, 91 (quos [sc. Carpos] … Galerius Maximinus Caesar
devicit et rei publicae Romanae subegit).
72
NIXON – RODGERS 1994, 116, note 17: “The Carpi were far from being annihilated or in-
corporated within the Roman Empire!”
73
Galerius was Carpicus maximus in 296, 301, 302, 303, 304 and 308/309 (BARNES 1982,
256–257); Constantine himself was hailed as Carpicus maximus in 315 or later (ILS I, 696).
74
During his war with the Goths, Valens and his army established a camp “near a village of
the Carpi” (prope Carporum vicum), A MMIANUS M ARCELLINUS, Res gestae, XXVII,5,5.
75
ZOSIMUS IV,34,6.

246
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

his army along the frontiers; and settled the rest of them as farmers on deserted
or inadequately cultivated lands in provinces through Thrace, Macedonia, and
Italy.”76 This he could very well have done; but how exactly were these Sarma-
tians distributed, especially the fresh recruits, who were to fill the ranks of the
Roman army? Perhaps Constantine followed rules specifically put in place by
another Emperor.
In 175 CE, the Roman Emperor Marcus Aurelius sent 5,500 Sarmatian
horsemen to Britain to serve in the Roman army.77 In the third century, there
was still a numerus of about 500 Sarmatian cavalry stationed at Bremetennacum
(today Ribchester near Lancaster).78 Its original name – numerus equitum Sarma-
tarum Bremetennacum or Bremetennacensium – was changed to cuneus Sarmatarum
in the fourth century.79 Of course, we possess no confirmation whether these
Sarmatae were in fact descendants of those sent to Britain by Marcus Aurelius.80
Nor can we be sure whether they became Sarmatian in name only within one
generation.81 With localized recruitment, the Sarmatians at Ribchester may have
simply disappeared over time. The fate of the unit itself is not known but both
fort and settlement at Ribchester seem to have been abandoned by the end of
the fourth century.82 However, some have argued that at least some members
of this community may have attained their ethnic identity well into the 5th (or
even 6th) century and ultimately laid the foundation for the Arthurian legends.83
If only one cavalry unit was created, what became of the rest? Jarrett presumed

76
ODAHL 2013, 256.
77
DIO CASSIUS LXXII,16; MCLYNN 2009, 368; SULIMIRSKI 1970, 175–176. This seems to be
Marcus Aurelius’ common policy towards the captured barbarians, see Historia Augusta, Vita
Marci, 22,2 (accepitque in deditionem Marcomannos plurimis in Italiam traductis); cf. 24,3 (infinitos
ex gentibus in Romano solo collocavit).
78
R ICHMOND 1945, 15–29.
79
Notitia Dignitatum, Oc., XL,54 (cuneus Sarmatarum, Bremetenraco); for the original name, see
CIL VII 218.
80
BREEZE – DOBSON 1985, 13: “It is not known whether these troops were genuinely all
required in Britain, or were merely sent to a conveniently isolated part of the empire. Nor is
it known where they were stationed; the ala Sarmatarum at Ribchester may have been formed
from this force, but it may have come to the island at a different time.”
81
WADGE 1987, 212: “While auxiliary units in the Roman army retained their ethnic titles,
they did not retain their ethnic make-up. (…) Since many Celtic tribes were noted horsemen
they could easily provide suitable recruits for the Cuneus Sarmatarum for as long as it remained
a cavalry regiment.”
82
WADGE 1987, 213.
83
LITTLETON – THOMAS 1978, 521–522: “Almost certainly they and their descendants married
local Brythonic women… However, the community´s close association with the Roman mi-

247
STANISLAV DOLEŽAL

that they “were distributed between existing alae and cohortes equitatae. This pre-
sents problems; for the total cavalry strength of the army in Britain was of the
order of 13,000. An accession of 5,000 men implies that all units were seriously
depleted or that our judgement of establishment is far from correct. No better
explanation has been published, and the present author has none to offer.”84
Of course, two other logical explanations present themselves: either most Sar-
matians sent to Britain fought on foot or the number given by Dio is far too
high. But if we really need to accommodate additional 5,000 horsemen, then we
may assume that up to ten additional alae may have been created and assigned
to the legions stationed in Britain at the time: legio VI Victrix, legio II Augusta
and legio XX Valeria Victrix. However, these left no trace. All we can say further
is that these Sarmatian cavalry units probably were not kept together, for obvi-
ous security reasons.85 How does this relate to our case?
Luckily, the same document that provides us with useful information about
the lone Sarmatian unit in the Roman Britain has much to tell us about other
Sarmatian units in the Roman Empire. There exists a broad consensus that the
Notitia Dignitatum “provides a fairly comprehensive picture of the organiza-
tion of the military and administrative establishments in the eastern half of the
empire at the end of the fourth century, and somewhat later, and more hap-
hazardly, for the west.”86 As we have seen, auxiliary units in the Roman army
could have a surprisingly long life and there is no reason to reject the infor-
mation supplied by the Notitia Dignitatum solely on the ground that whoever
compiled it did it some 50 years after Constantine died. On the contrary: on
the basis of the Notitia Dignitatum, there is a good reason to believe that those
Sarmatian men deemed fit for military service in 334 were massively enrolled
in the system of laeti.
A. H. M. Jones asserted that the system of laeti “was, to our knowledge, con-
fined to Gaul and Italy” and that it worked in the manner that “the government
set aside certain lands, the terrae laeticae, for the settlement of the barbarians who
sought refuge in the empire”.87 This fits nicely with our Sarmatians moved to the

litary establishment… probably helped it to maintain its ethnic identity and to resist total
assimilation into the indigenous Celtic population.”
84
JARRETT 1994, 76.
85
However, SOUTHERN (2004, 202) opines that “they may have been kept together, as the
Britons on the German frontier seem to have been, since there is some evidence from a later
period of Sarmatian units in Britain.”
86
K ELLY 2004, 40.
87
JONES 1964, 620 (cf. Codex Theodosianus, XIII,11,10). For more information about the system

248
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

Empire by Constantine in 334. The Origo tells us that one of the areas selected
for settlement of these refugees was Italy. According to the Notitia Dignitatum,
there were fifteen praefecti Sarmatarum gentilium in Italia.88 Six of them were in
northwest Italy, in today Piedmont: Foro Fulviensi, Taurinis, Pollentia in Liguria,
Aquis sive Tertona, Novariae, Vercellis. Only two praefecti were in the northeast, in
the province of Venetia et Histria (Opitergii, Patavio); one was in Cremona and one
in Bononia. Compared to these ten praefecti of the north, only two were in the
south, in Italy’s two southernmost provinces (Apuliae et Calabriae, per Bruttios et
Lucaniam); no exact location is given. One praefectus was in Samnium (regionis
Samnitis), one place name is unknown (Quadratis et Eporizio) and one is missing.
The total would thus be 15 units.89 Assuming that each praefectus commanded
one unit of 500 men, we arrive at 7,500 men. However, we do not know the size
of all the new units created under Constantine. The prefects of the laeti could
thus have commanded up to 1,000 men each.90 Thus we would count up to
15,000 men. But that figure appears too high and we would have heard of this
substantial Sarmatian contribution to the Western comitatus from our sources,
especially Ammianus Marcellinus. We should therefore suppose that the size of
the newly created Sarmatian auxilary units was 500 men.91
The second area where we find units of laeti is Gaul. The Origo does not say
that Sarmatian warriors were posted to any of the Gallic provinces but perhaps
they were sent there at some later time, in 340s or 350s (assuming that the Origo
was composed shortly after Constantine’s death). The Notitia Dignitatum lists
following praefecti Sarmatarum gentilium in Gallia: “a Chora Parisios usque” (near
Paris), “inter Renos et Tambianos provinciae Belgicae secundae” (Tambiani should

of laeti, e.g., LEE 2007, 81–82. But cf. DRINKWATER (2007, 166) who, while admitting that “the
general issue of Gallic laeti is difficult and controversial”, nevertheless opines that “they were
obliged to maintain themselves and supply recruits to the Roman army. However, such recruits
were not formed into their own regiments but posted in the usual way to existing units.”
88
Notitia Dignitatum, Oc., XLII,45–63.
89
BACHRACH 1973, 37 (“many of the truly Sarmatian colonies in Italy, if indeed not all of
them, had very probably been founded during the reign of Constantine the Great early in the
fourth century”).
90
For the debate of the size of new army units under Diocletian and Constantine, see JONES
1964, 680–682; ELTON 2007, 333; CAMPBELL 2008, 127.
91
Perhaps the famous general Victor was one of the Sarmatian soldiers stationed in Italy in
334. He had a splendid career in the Roman army in 360s and 370s (see JONES – M ARTINDALE
– MORRIS 1971, 957–959) and he lived to early 380s, so he may have been about 18 in 334 –
or, if he was younger, he may have joined the army somewhat later. At any rate, he served
Constantius II (A MMIANUS M ARCELLINUS, Res gestae, XXV,5,2: Victor et e palatio Constanti residui;
XXXI,12,6: Victor nomine magister equitum Sarmata sed cunctator et cautus).

249
STANISLAV DOLEŽAL

probably be emended to Ambiani, i.e., Amiens), “per tractum Rodunensem et Alau-

norum” (Rennes?) and “Lingonas” (Langres); one place name is missing and the
list abruptly ends here, leaving us wondering how many such units there actu-
ally might have been. Furthermore, there was one praefectus Sarmatarum et Tai-
falorum gentilium in Poitiers (Pictavis), indicating that some Sarmatians, mixed
with the Taifali, were probably recruited in Dacia.92 These six units could have
numbered 3,000 men in total.
If we take the Origo seriously and believe that Constantine moved to the
Empire some 300,000 men, women and children, then we can assume that some
60,000 from this total, or one fifth, were men either fit for military service or
for farming. We have managed to accommodate no more than 10,500 elite war-
riors so far.93 The rest – about 50,000 men –therefore may have become coloni
in the Eastern provinces throughout Thrace, Scythia, Macedonia, and Italy (per
Thraciam, Scythiam, Macedoniam, Italiamque, according to the Origo). Whoever
compiled the Origo had probably in mind the diocese of Thracia, the diocese
of Macedonia, and the two dioceses of Italy. What about Scythia? Scythia minor,
itself a part of the diocese of Thrace, is a tiny province. Why did the author
point specifically to this corner of the Empire?
A law issued by Theodosius II in 409, concerning captured Scirians on the
Danube, can perhaps provide some useful guidance.94 When an invading force,
composed of the Huns and the Scirians and led by Uldin, the king of the Huns,
was defeated in Thrace in 408, the king escaped only with difficulty back across
the river.95 His men were not so lucky. Many of his Scirians were killed, others
were taken prisoners and shipped to Constantinople. Some of them were sold
to slavery, while the remaining, as Sozomen testifies, “were ordered to settle in

92
Notitia Dignitatum, Oc., XLII,64–70 (the editor, Otto Seeck, remarked at the end of the list
that “excidit in archetypo folium unum”).
93
We would expect at least some traces of Sarmatian warriors in Thrace, Scythia, and Ma-
cedonia but in the Notitia Dignitatum, there is only one Sarmatian unit in the whole East, and
it is in Egypt (Or., XXVIII,26: sub dispositione viri spectabilis, comitis rei militaris per Aegyptum: ala
septima Sarmatarum, Scenas Mandrorum). It appears that the bulk of Sarmatian warriors served
in Italy (and possibly Gaul), while it is possible that some smaller groups of Sarmatians were
used to reinforce military units elsewhere.
94
CODEX THEODOSIANUS, V,6,3.
95
SOZOMEN (Historia ecclesiastica, IX,5) says that much, although he also says that Uldin took
by treachery “a city of Moesia, called Castra Martis”, making incursions from there “to the
rest of Thrace”. This town obviously lies in Dacia ripensis, which is not in Thrace. However,
“the commander of the Thracian armies” (probably magister militum per Thracias) was involved
in the following peace talks, so we may assume that the main action took place in Thrace.
See also M AENCHEN-HELFEN 1973, 63–66.

250
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

different places”. Actually, they were distributed as a supplement to the work-

force of landowners, because the law in question specifies that “We grant to all
persons the opportunity to supply their own fields with men of the aforesaid
race” (i.e., the Scirians).96 Equally important is the caveat that these new coloni
were not to be settled in Thrace or Illyricum (that is, close to their homeland),
but only “across the sea”, which can only mean Asia Minor. And Sozomen in
fact testifies that because of security reasons, these coloni were distributed there:
“I have seen many in Bithynia, near Mount Olympus, living apart from one
another, and cultivating the hills and valleys of that region.” Being dispersed,
they were never to be heard of again. Sozomen himself may have seen them
in his childhood (in 410s); but after a generation or two, their ethnic identity
simply vanished.
More examples can easily be shown. In 386, Theodosius’ magister militum
Promotus defeated a very large group of invading Greuthungi / Ostrogoths,
who were led across the Danube by a chieftain Odotheus.97 Many prisoners
were taken. Some were enrolled in the Roman army (and revolted in 399 under
their leaders Gainas and Tribigild), while others, probably those deemed unsuit-
able for military service, were settled in Phrygia in Asia Minor as coloni.98 In
another case, some time after 365, Ausonius briefly mentions Sarmatian coloni
who were “recently” settled between Trier and Bingen – a very long distance
from their homes.99
It is, therefore, understandable that Constantine settled Sarmatian farmers
purposely in the most distant Balkan provinces relative to their homeland: not
in the dioceses of Pannonia and Dacia but rather in all the Thracian and Mac-
edonian provinces. We can also suppose that those Sarmatian farmers settled
in Italy were those male members of Sarmatian families that were not deemed
suitable for military service. Just as Eusebius wrote: “Those who were suitable
he enrolled in his own forces; to the rest he apportioned land for cultivation of
the means of subsistence.” In a grand total, we may guess that all members of
families included, Italy received some 17,500 Sarmatians, Gaul around 15,000,
and the great majority of the refugees – more than 250,000 – were settled in
the Balkan peninsula.

96
PHARR – DAVIDSON – PHARR 1952, 108.
97
CLAUDIANUS, De IV consulatu Honorii, 626ff.; ZOSIMUS IV,35,1 and IV,38–39; M ARCELLINUS
COMES, s.a. 386; HYDATIUS 39,8.
98
CLAUDIANUS, In Eutropium, II,153–154 (Ostrogothis colitur mixtisque Gruthungis Phryx ager) and
196–197 (bene rura Gruthungus excolet et certo disponet sidere vites).
99
AUSONIUS, Mosella, V,9 (arvaque Sauromatum nuper metata colonis).

251
STANISLAV DOLEŽAL

Dacia restituta?

What were the consequences of Constantine’s campaigning beyond the Dan-

ube in 332 and 334? Eusebius says that Constantine “annexed the whole Scyth-
ian population, which was in the far north divided into numerous barbarian
tribes”.100 It is, of course, a panegyrist’s expression, but Constantine became
Gothicus maximus (for the second time) in 332, Sarmaticus maximus (also for the
second time) in 334, and Dacicus maximus in 336 (or 335).101 Most Constantin-
ian scholars therefore assume that at least a part of the former Roman Dacia
was recaptured by 336 (without stating when it was abandoned subsequently).102
However, our evidence for this supposed reconquest is meagre.103 In 297, an
anonymous rhetorician asserted that Dacia was restored to the Roman rule, and
there is no doubt that he meant the province Dacia ripensis.104 If Constantine
conquered a significant portion of Gothia by 336, why are our sources silent
about such a huge success? And if the Emperor managed to acquire the territory
of (at least south) Romania, why the Sarmatian refugees were not settled there?
Be that as it may, the Origo expressly states that the caesar Dalmatius was
charged by Constantine with protection of the bank of the Danube – very pos-

100
EUSEBIUS, Vita Constantini, I,8,2 (translation: CAMERON – HALL 1999, 70).
101
BARNES 1976, 149–155 (year 336); BARNES 1982, 258 (year 336); BARNES 2014, 165 (“335 or
336”). In addition, the “Sarmatian games” (ludi Sarmatici) that were probably introduced after
Constantine’s victory over the Sarmatians in 323, were extended to seven days, from November
25 to December 1 (WIENAND 2012, 336; SALZMAN 1990, 138).
102
BARNES 1981, 250 (“at least a partial reconquest of the Dacia”); BARNES 1982, 80; BARNES
2014, 165–166; CAMERON 2008, 105 (“some renewal of Roman control in Dacia”), POHLSANDER
2004, 78 (“the partial and temporary recovery of Dacia”), POTTER 2013, 285 (“some territorial
conquest north of the Danube in the area that Aurelian had abandoned in 271”); ELLIOTT 1996,
255 (“he reconquered part of Dacia”); cf. ODAHL 2013, 261. KULIKOWSKI (2007a, 102) opined
that “the Carpathian lands of the old province were certainly not reannexed and subjected
to Roman administration” but he conceded that some new garrisons and small forts beyond
the Danube may have been established by that time; cf. LENSKI 2002, 122: more likely, he
established only a toehold across the Danube.” But cf. a more cautious approach by WILKES
2005, 161, who only speaks of earthworks in Dacia whose purpose was “to mark territory
regarded as being under Roman protection” from that of the German groups or the Empire’s
Sarmatian allies.
103
Basically, the evidence rests on two sources: the Emperor Julian summed up the achie-
vements of his uncle Constantine against the barbarians in his pamphlet Caesares (329c) by
simply saying that Constantine recovered those parts of the Empire which Trajan conquered.
FESTUS (Breviarium, 26) only speaks of a “recent victory” over the Goths when Constantine
was assembling a huge army to invade the Sassanid Empire in 337 (recenti de Gothis victoria).
104
Panegyrici Latini, VIII(4),3,3 (Dacia restituta), cf. NIXON – RODGERS (1994, 113, note 11) and
MILLAR 2001, 189, note 138.

252
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

sibly both of the banks.105 He took up residence in Naissus (probably in 335)106

and he may very well have been tasked not only with securing the Danubian
border and managing the newly acquired land in Gothia, but also with distribu-
tion of the Sarmatian refugees in the Balkan peninsula. This process certainly
took some time and may not have been completed by the end of 334. Also, as
we have seen, Naissus was the city where Constantine himself spent some time
in 334, so it is conceivable that Dalmatius was charged with finishing the mon-
strous task of settling tens of thousands of Sarmatians in the Balkans.
If the Romans were able to hold at least a part of Gothia, for how long?
We know little of the Roman frontier on the Danube in the years following
Constantine’s death – or the death of Dalmatius, for that matter.107 In 362 the
Emperor Julian dismissed the urging of members of his inner circle who “tried
to persuade him to attack the neighbouring Goths, who were often deceitful
and treacherous”. But Ammianus at the same time stresses Julian’s carefulness
about the border fortifications in Thrace and about “the troops posted along
the banks of Danube, who, as he heard, were meeting inroads of the savages
with watchfulness and valour”.108 It is clear that whatever gains Constantine
made in Gothia, all were long lost or abandoned by 362.109
The Gothic victory of Constantine in 332 may have been spectacular, even
grandiose, but its strategic implications were limited: Dacia was not reconquered.
Countless dead Goths in 332 may have helped keep the peace on the lower
Danube until 360s, secure the lasting foedus with Rome, allow for mutual com-
merce, which benefitted both sides, and perhaps discourage future invasions of
other barbarians. But it is doubtful whether an inch of land was acquired. At
best, a Roman protectorate may have been temporarily established along the
Danube but we do not even know where.

105
Origo Constantini imperatoris, 6,35 (ripam Gothicam tuebatur). Cf. CHRYSOS 2001, 69–72.
106
BARNES 1982, 87; ODAHL 2013, 264. However, he is attested there no sooner than in 337.
107
In about 338, for example, Constantius II or his generals (namely, Sappo, dux limitis Scy-
thiae), seems to be involved in some border clashes, possibly with the Sarmatians, in the pro-
vince of Scythia Minor, near Troesmis (ILS I, 724 = CIL III, 12483). See BARNES 2001, 224–225;
cf. JONES – M ARTINDALE – MORRIS 1971, 803.
108
A MMIANUS M ARCELLINUS, Res gestae, XXII,7,8 (suadentibus proximis, ut adgrederetur propinquos
Gothos saepe fallaces et perfidos, hostes quaerere se meliores aiebat: illis enim sufficere mercatores Galatas,
per quos ubique sine condicionis discrimine venundantur); XXII,7,7 (quos per supercilia Histri dispersos,
excursibusque barbarorum oppositos agere vigilanter audiebat et fortiter). Translation: ROLFE 1935b,
211.
109
For summary of events in 332–362, see KULIKOWSKI 2007a, 100–106. For an overview of
the Danubian frontier, see WILKES 2008, 233.

253
STANISLAV DOLEŽAL

The greatest value of the Sarmatian campaign in 334 were the Sarmatians
themselves: not land, but its people became an important asset. And it appears
that this valuable manpower was used well by the Romans. However, Constan-
tine’s plans with Gothia probably went not as the aging Emperor wished and in
the end, death prevented Constantine from showing the world that he could
stand comparison with Trajan.

Bibliography

BACHRACH, B. S. 1973, A History of the Alans in the West, Minneapolis.

BARNES, T. D. 1976, “The Victories of Constantine”, Zeitschrift für Papyrologie und
Epigraphik, 20, pp. 149–155.
BARNES, T. D. 1981, Constantine and Eusebius, Cambridge (Mass.) – London.
BARNES, T. D. 1982, The New Empire of Diocletian and Constantine, Cambridge
(Mass.) – London.
BARNES, T. D. 2001, Athanasius and Constantius. Theology and Politics in the Con-
stantinian Empire, Cambridge (Mass.) – London.
BARNES, T. D. 2014, Constantine. Dynasty, Religion and Power in the Later Roman
Empire, Chichester.
BLECKMANN, B. 1995, “Constantin und die Donaubarbaren. Ideologische Aus-
einandersetzungen um die Sieghaftigkeit Constantins”, Jahrbuch für Antike
und Christentum, 38, pp. 38–66.
BLECKMANN, B. 2007, “Sources for the History of Constantine”, in: LENSKI 2007b,
pp. 14–34.
DEN BOEFT, J. – DRIJVERS, J. W. – DEN HENGST, D. – TEITLER, H. C. 2009, Philo-
logical and Historical Commentary on Ammianus Marcellinus XXVII, Leiden –
Boston.
BOWMAN, A. K. – GARNSEY, P. (eds.) 2008, The Cambridge Ancient History, XII: The
Crisis of the Empire A.D. 193–337, Cambridge
BREEZE, D. J. – DOBSON, B. 1985, “Roman Military Deployment in North Eng-
land”, Britannia, 16, pp. 1–19.
BRUUN, P. M. 1966, The Roman Imperial Coinage, VII: Constantine and Licinius
A.D. 313–337, London.
BURGESS, R. W. (ed.) 1993, The Chronicle of Hydatius and the Consularia Constan-
tinopolitana, Oxford.
CAMERON, A. 2008, “The Reign of Constantine, A. D. 306–337”, in: BOWMAN –
GARNSEY 2008, pp. 90–109.
CAMERON, A. – HALL, S. G. (eds.) 1999, Eusebius: Life of Constantine, Oxford.

254
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

CAMPBELL, B. 2008, “The Army”, in: BOWMAN – GARNSEY 2008, pp. 110–130.
CHRYSOS, E. 2001, “Ripa Gothica and Litus Saxonicum”, in: W. POHL – I. WOOD
– H. REIMITZ (eds.), Transformation of Frontiers. From Late Antiquity to the Caro-
lingians, Leiden – Boston, pp. 69–72.
CLAUSS, M. 1996, Konstantin der Grosse und seine Zeit, München.
CORCORAN, S. 2000, The Empire of the Tetrarchs. Imperial Pronouncements and Gov-
ernment AD 284–324, Oxford.
DESSAU, H. (ed.) 1892, Inscriptiones Latinae Selectae, I, Berlin.
DOLEŽAL, S. 2018, “Where Did Rausimod Come from?”, Eirene. Studia Graeca et
Latina, 54, pp. 155–167.
DRINKWATER, J. F. 2007, The Alamanni and Rome 213–496 (Caracalla to Clovis),
Oxford.
ELLIOTT, T. G. 1996, The Christianity of Constantine the Great, Scranton.
ELTON, H. 2007, “Warfare and Military”, in: LENSKI 2007b, pp. 325–346.
ILS I = Inscriptiones Latinae Selectae, I, ed. H. DESSAU, Berlin 1892.
JARRETT, M. G. 1994, “Non-Legionary Troops in Roman Britain: Part One, the
Units”, Britannia, 25, pp. 35–77.
JONES, A. H. M. 1964, The Later Roman Empire 284–602, II, Oxford.
JONES, A. H. M. 1972, Constantine and the Conversion of Europe, London.
JONES, A. H. M – MARTINDALE, J. R. – MORRIS, J. 1971, The Prosopography of the
Later Roman Empire, I: A.D. 260–395, Cambridge.
JONES, H. L. 1924, Strabo, Geography, III, Cambridge (Mass) – London (Loeb
Classical Library).
KELLY, CH. 2004, Ruling the Later Roman Empire, Cambridge (Mass.) – London.
KIENAST, D. 1996, Römische Kaisertabelle: Grundzüge einer römischen Kaiserchronolo-
gie, Darmstadt.
KÖNIG, I. (ed.) 1987, Origo Constantini. Anonymus Valesianus, I: Text und Kom-
mentar, Trier.
KULIKOWSKI, M. 2007a, Rome’s Gothic Wars, Cambridge.
KULIKOWSKI, M. 2007b, “Constantine and the Northern Barbarians”, in: LENSKI
2007b, pp. 347–376.
LEE, A. D. 2007, War in Late Antiquity. A Social History, Oxford.
LENSKI, N. 2002, The Failure of Empire. Valens and the Roman State in the Fourth
Century A.D., Berkeley – Los Angeles – London.
LENSKI, N. 2007, “The Reign of Constantine”, in: LENSKI 2007b, pp. 59–90.
LENSKI, N. (ed.) 2007b, The Cambridge Companion to the Age of Constantine, Cam-
bridge.
LENSKI, N. 2016, Constantine and the Cities, Philadelphia.
LIEU, S. N. C. – MONTSERRAT, D. 1996, From Constantine to Julian: Pagan and Byz-
antine Views. A Source History, London – New York.

255
STANISLAV DOLEŽAL

LITTLETON, C. S. – THOMAS, A. C. 1978, “The Sarmatian Connection: New Light

on the Origin of the Arthurian and Holy Grail Legends”, The Journal of Americ-
can Folklore, 91, No. 359, pp. 513–527.
MAENCHEN-HELFEN, O. J. 1973, The World of the Huns, Berkeley – Los Angeles –
London.
MAGIE, D. 1991, Scriptores Historiae Augustae, I, Cambridge (Mass.) – London
(Loeb Classical Library).
MCLYNN, F. 2009, Marcus Aurelius. A Life, Cambridge (Mass.).
MILLAR, F. 2001, The Roman Near East 31 BC – AD 337, Cambridge (Mass.) –
London.
MÓCSY, A. 2014, Pannonia and Upper Moesia: A History of the Middle Danube Prov-
inces of the Roman Empire, Oxford – New York (reprint of the 1974 edition).
NIXON, C. E. V. – RODGERS, B. S. 1994, In Praise of Later Roman Emperors: The
Panegyrici Latini, Berkeley.
ODAHL, C. M. 2013, Constantine and the Christian Empire, London – New York.
PHARR, C. – DAVIDSON, T. S. – PHARR, M. B. 1952, The Theodosian Code and Nov-
els and the Sirmondian Constitutions. A Translation with Commentary, Glossary,
and Bibliography, Princeton.
POHLSANDER, H. A. 2004, The Emperor Constantine, London – New York.
POTTER, D. S. 2004, The Roman Empire at Bay AD 180–395, London – New York.
POTTER, D. S. 2013, Constantine the Emperor, Oxford.
RICHMOND, I. A. 1945, “The Sarmatae, Bremetennacvm Veteranorvm and the
Regio Bremetennacensis”, Journal of Roman Studies, 35, No. 1–2, pp. 15–29.
ROLFE, J. C. 1935a, Ammianus Marcellinus, History, I: Books 14–19, Cambridge
(Mass.) – London (Loeb Classical Library).
ROLFE, J. C. 1935b, Ammianus Marcellinus, History, II: Books 20–26, Cambridge
(Mass.) – London (Loeb Classical Library).
SALZMAN, M. R. 1990, On Roman Time. The Codex-Calendar of 354 and the Rhythms
of Urban Life in Late Antiquity, Berkeley.
SCHÖNFELD, M. 1911, Wörterbuch der altgermanischen Personen- und Völkernamen,
Heidelberg.
SCHWENDEMANN, J. 1923, Der historische Wert der Vita Marci bei den Scriptores His-
toriae Augustae, Heidelberg.
SEECK, O. 1911, Geschichte des Untergangs der antiken Welt, IV, Stuttgart.
SEECK, O. 1984, Regesten der Kaiser und Päpste für die Jahre 311 bis 476 n. Chr.,
Frankfurt am Main.
SOUTHERN, P. 2004, The Roman Empire from Severus to Constantine, London –
New York.
STEPHENSON, P. 2010, Constantine: Roman Emperor, Christian Victor, New York.
SULIMIRSKI, T. 1970, The Sarmatians, London.

256
CONSTANTINE’S MILITARY OPER ATIONS AGAINST THE GOTHS AND THE SAR MATIANS

TREADGOLD, W. 1995, Byzantium and Its Army 284–1081, Stanford.

WADGE, R. 1987, “King Arthur: A British or Sarmatian Tradition?”, Folklore, 98,
No. 2, pp. 204–215.
WIENAND, J. 2012, Der Kaiser als Sieger. Metamorphosen triumphaler Herrschaft un-
ter Constantin I., Berlin.
WILKES, J. J. 2005, “The Roman Danube: An Archaeological Survey”, Journal of
Roman Studies, 95, pp. 124–225.
WILKES, J. J. 2008, “Provinces and Frontiers”, in: BOWMAN – GARNSEY 2008, pp.
212–268.
WOLFRAM, H. 1990, The Goths, trans. by T. J. DUNLAP, Berkeley – Los Angeles –
London.

Summary

This article attempts to construe the chronology of Constantine’s campaigns

against the Goths and Sarmatians in 332 and 334, respectively, and to assess
the extent of Roman control of Gothia (the former Roman Dacia) in the fol-
lowing years. The article also argues that there is a link between Constantine‘s
military operations on the lower Danube in 334 and colonies or military units
of Sarmatians in Italy, possibly in Gaul, and elsewhere in the Empire, which are
attested by a number of sources, primarily the Notitia Dignitatum.

Keywords: Constantine the Great; Sarmatians; Goths

STANISLAV DOLEŽAL
Institute of History,
Faculty of Philosophy,
University of South Bohemia,
Branišovská 1645/31a,
370 05 České Budějovice,
Czech Republic,
dolezal@ff.jcu.cz

257
 EIRENE
STUDIA GR AECA ET LATINA
ISSN 0046-1628
Founded 1960

Eirene. Studia Graeca et Latina is an international refereed scholarly journal of classics

which is published by the Centre for Classical Studies at the Institute of Philosophy
of the Czech Academy of Sciences, Prague. It welcomes and publishes original research
on classics, reception of Antiquity and classical traditions. It also brings up-to-date
reviews of scholarly literature on these subjects. ■ The journal accepts submissions
in English, German, French and Italian. All contributions (except for reviews) are
sent anonymously for peer-review. ■ Eirene. Studia Graeca et Latina is abstracted /
indexed in following scientific databases: L’année philologique; The Central European
Journal of Social Sciences and Humanities (CEJSH); European Reference Index for
Humanities (ERIH PLUS); Modern Language Association International Bibliography;
Scopus; Web of Science; EBSCO. ■ For manuscript submission guidelines, contents
of previously published issues, and for possibility to publish selected articles in an
Open Access mode, please visit the journal’s homepage: http://www.ics.cas.cz/en/
journals/eirene. ■ All article submissions and subscription / exchange orders (back
issues are also available for purchase) are to be sent by e-mail to: eirene@ics.cas.cz.
Books for review and other correspondence should be mailed to:

Eirene. Studia Graeca et Latina

Centre for Classical Studies at the Institute of Philosophy of the
Czech Academy of Sciences
Na Florenci 3
110 00 Prague 1
Czech Republic
tel.: +420 234 612 330
fax: +420 222 828 305
 EIRENE
STUDIA GR AECA ET LATINA
LV / 2019 / I–II

© Centre for Classical Studies at the Institute of Philosophy of the

Czech Academy of Sciences, Prague

Prague 2019
ISSN 0046-1628

Ed itor-i n - ch ief
PETR KITZLER
Centre for Classical Studies,
Czech Academy of Sciences, Prague

A ssista nt Ed itors
NEIL ADKIN (University of North Carolina at Chapel Hill) – JAN BAŽANT (Centre
for Classical Studies, Czech Academy of Sciences, Prague) – JAN BOUZEK (Charles
University, Prague) – JAN N. BREMMER (University of Groningen) – KATHLEEN
M. COLEMAN (Harvard University) – SIEGMAR DÖPP (University of Göttingen)
– HERMANN HARRAUER (University of Vienna) – STEPHEN HARRISON
(Corpus Christi College, University of Oxford) – HERBERT HEFTNER (University
of Vienna) – BRAD INWOOD (Yale University) – IRENE J. F. DE JONG (University
of Amsterdam) – DAVID KONSTAN (New York University) – WALTER LAPINI
(University of Genova) – GLENN W. MOST (Scuola Normale Superiore di Pisa /
University of Chicago) – PAVEL OLIVA (Centre for Classical Studies, Czech Academy
of Sciences, Prague, emeritus) – JIŘÍ PAVLÍK (Charles University, Prague) – ROSARIO
PINTAUDI (University of Messina) – ILARIA L. E. RAMELLI (Angelicum – Sacred
Heart University – Oxford) – ALAN SOMMERSTEIN (University of Nottingham) –
DMITRY VL. TRUBOTCHKIN (Russian University of Theater Arts – GITIS, Moscow)

Ma nag i ng Ed itor
JAKUB ČECHVALA
Centre for Classical Studies, Czech Academy of Sciences, Prague

cover and graphic design © Markéta Jelenová

typesetting © Jana Andrlová
print © Karolinum Publishers
English proofreading Pavel Nývlt