Limes, Economy and Society

in the
Lower Danubian Roman Provinces

Edited by
Lucrețiu Mihailescu-Bîrliba

PEETERS
 LIMES, ECONOMY AND SOCIETY
IN THE
LOWER DANUBIAN ROMAN PROVINCES
 COLLOQUIA ANTIQUA
Supplements to the Journal ANCIENT WEST & EAST

SERIES EDITOR

GOCHA R. TSETSKHLADZE (UK)

EDITORIAL BOARD

A. Avram (Romania/France), Sir John Boardman (UK), J. Hargrave (UK),

M. Kazanski (France), A. Mehl (Germany), A. Podossinov (Russia),
N. Theodossiev (Bulgaria), J. Wiesehöfer (Germany)

ADVISORY BOARD

S. Atasoy (Turkey), L. Ballesteros Pastor (Spain), J. Bouzek (Czech Rep.),

S. Burstein (USA), J. Carter (USA), B. d’Agostino (Italy),
J. de Boer (The Netherlands), A. Domínguez (Spain),
O. Doonan (USA), A. Kuhrt (UK), Sir Fergus Millar (UK),
J.-P. Morel (France), M. Pearce (UK), D. Potts (USA),
A. Rathje (Denmark), R. Rollinger (Austria), A. Snodgrass (UK),
M. Sommer (Germany), D. Stronach (USA),
M. Tiverios (Greece), C. Ulf (Austria), J. Vela Tejada (Spain)

Colloquia Antiqua is a refereed publication

For proposals and editorial and other matters, please contact the Series Editor:
Gocha R. Tsetskhladze
The Gallery
Spa Road
Llandrindod Wells
Powys LD1 5ER
UK
E-mail: [email protected]
 COLLOQUIA ANTIQUA
————— 25 —————

LIMES, ECONOMY AND SOCIETY

IN THE
LOWER DANUBIAN ROMAN PROVINCES

Edited by

LUCREŢIU MIHAILESCU-BÎRLIBA

PEETERS
LEUVEN – PARIS – BRISTOL, CT
2019
A catalogue record for this book is available from the Library of Congress.

ISBN 978-90-429-3812-0
eISBN 978-90-429-3813-7
D/2019/0602/81
© 2019, Peeters, Bondgenotenlaan 153, B-3000 Leuven, Belgium

No part of this book may be reproduced in any form or by any electronic or mechanical means,
including information storage or retrieval devices or systems, without prior written permission
from the publisher, except the quotation of brief passages for review purposes.
 TABLE OF CONTENTS

Series Editor’s Preface – Gocha R. Tsetskhladze . . . . . . . . . . . . . . . . . . VII

Volume Editor’s Introduction – Lucrețiu Mihailescu-Bîrliba . . . . . . . . IX

List of Illustrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XIII

List of Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XVII

CHAPTER 1 Ovid: The Double Face of the Danube

Marius Alexianu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

CHAPTER 2 Code-Switching: Expression Forms of Linguistic Iden-

tity in Moesia Inferior
Roxana-Gabriela Curcă. . . . . . . . . . . . . . . . . . . . . . . . . . 11

CHAPTER 3 Soldats du milieu rural de la Mésie Inférieure recrutés

dans l’armée romaine
Lucrețiu Mihailescu-Bîrliba. . . . . . . . . . . . . . . . . . . . . . . 19

CHAPTER 4 Trade Economy in Riverine Provinces: A Close-Up

Look at Traders’ Networks and Mobility
Rada Varga and Annamária-Izabella Pázsint. . . . . . . . . 31

CHAPTER 5 Facilities and Medical Staff of the Lower Danubian

Roman Army
Dan Aparaschivei. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

CHAPTER 6 Observations on Age-Rounding for Soldiers from the

Lower Danubian Provinces
Valentin Piftor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

CHAPTER 7 Territoria and regiones in the Lower Danubian Prov-

inces
Florian Matei-Popescu . . . . . . . . . . . . . . . . . . . . . . . . . . 95
VI TABLE OF CONTENTS

CHAPTER 8 Glassware from the Lower Danubian Limes (1st–3rd

Centuries AD): A Brief Survey
Sever-Petru Boțan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

CHAPTER 9 Die Göttin Hekate in Sarmizegetusa

Ioan Piso und Csaba Szabó. . . . . . . . . . . . . . . . . . . . . . . 139

CHAPTER 10 Private Initiative and Buildings of Public Worship in

Roman Dacia
Ana Odochiciuc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

CHAPTER 11 Silver-Plated Coins in the Sites of Roman Dacia

Lucian Munteanu and Ștefan Honcu . . . . . . . . . . . . . . . . 173

CHAPTER 12 Römischer Einfluss in Barbaricum zwischen Diploma-

tie, Klientelpolitik und Defensivstrategie: neue Metho-
den des Machtmanagements in der Spätantike
Alexander Rubel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

CHAPTER 13 Terminus huius Imperii: The Frontiers of the Empire in

the Panegyrici Latini (3rd–4th Centuries AD) – Between
Propaganda and Reality
Nelu Zugravu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253

List of Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259

 SERIES EDITOR’S PREFACE

Despite the study of the limes is well established, many questions remain to be
answered. Scholarly work on the limes has focused mainly on politics, the
economy, militarisation and cult practices, and these four strands have received
quite extensive investigation. It is obvious, however, that new approaches are
necessary for making further progress. The papers in the present collection, the
fruits of a conference held in November 2017 at ‘Alexandru Ioan Cuza’ Uni-
versity in Iași, Romania, seek to do just that through case studies formulated
within the geographical framework of the two provinces of Dacia and Moesia
Inferior, which shared some similarities, and linked to the vital artery of the
River Danube and its tributaries. The coming of the Roman was not simply
a military expansion that saw the militarisation of the region and the creation
of the limes, it had positive results: in the cultural sphere through the mutually
beneficial interaction of Romans and locals; in the economic via the enhance-
ment of their trading and commercial relationships; in the religious by borrow-
ings one from another; and in the social development and opportunities
afforded to incomers and locals.
The papers presented here are based on a comprehensive study of the evi-
dence, be it archaeological, epigraphic, philological, numismatic, onomastic,
literary, etc. The combination has produced a fascinating volume, the editor
and authors of which should be congratulated for the prompt presentation of
the materials for publication.
Gocha R. Tsetskhladze
Llandrindod Wells
1 October 2018
 INTRODUCTION

Roman frontiers have been studied with an emphasis on multiple intercultural

dimensions. Scholarship has focused mainly on the political situation (for
example, the emerging of Roman domination and administration in the prov-
inces), the economy (such as trade and traffic between Romans and barbari-
ans), military issues (the role of the army as a peacekeeper and as a bearer of
cultures, etc.) and religious aspects (for example, the mutual impact of reli-
gious habits), etc. This volume aims to broaden the perspective on Roman
riverine frontiers. Rivers embody two – prima facie – irreconcilable functions.
While on the one hand they can constitute natural frontiers between the Roman
empire and the Barbaricum, on the other hand they facilitate the sharing of
goods and, further, stimulate cultural interaction at the borders of the empire.
So far, scholarship has been concerned mainly with the defensive function of
river frontiers. To deepen our understanding of riparian areas on the border
of the empire, the aforementioned dichotomous character of rivers must be
brought into consideration. The studies presented in this volume, focusing on
the provinces Dacia and Moesia Inferior, will investigate how rivers enhanced
or hampered connectivity on frontiers and thus shaped riverine areas as multi-
functional spaces with different functions.
In order to obtain meaningful results, the papers rely on a comparative
approach. The two chosen provinces, Dacia and Moesia Inferior, share a com-
mon core of features that still allows further research. Both are situated at the
northern periphery of the Roman empire, both in south-eastern Europe, and are
fortified by the limes. In geostrategic terms, they are defined by a large river
(the Danube) with several tributaries. As border provinces, they are also char-
acterised to a considerable extent by military forces and infrastructure. In addi-
tion to this, militarisation also had a great impact on economy and society. The
military presence led to the development of economic exchange with the local
population and produced a certain level of security, necessary for regional
trade. So far the evidence for trade and commerce is much better attested and
explored in other frontier provinces of the empire (for instance the Germanic
provinces) than in Dacia and Moesia Inferior. The same can be said for the
incorporation of the local population in Roman auxiliary units, the granting of
citizenship to individuals, and their integration into the Roman bureaucracy.
Hitherto, mutual effects between the local population and the presence of
X INTRODUCTION

Roman people, military and infrastructure are less documented for the two
provinces.
This book represent the proceedings of a conference held by the ‘Alexandru
Ioan Cuza’ University of Iași (8th–9th November 2017). Scholars from Iași,
Bucharest and Cluj-Napoca gathered in order to present not only the recent
results of their work, but to discuss in which ways the river frontier had influ-
enced the economic, social and religious interchange.
Thus, the studies in this volume treat economic, social and religious issues
in these two limes provinces (Dacia and Moesia Inferior), through literary,
linguistic, epigraphic, numismatic and archaeological approaches. Marius
Alexianu analyses the parameters of the Ovidian poetic perceptions and repre-
sentations concerning the Danube. Roxana-Gabriela Curcă proposes a survey
of the epigraphical material of code-switching, as a linguistic identity pattern,
following historical and philological aspects. Lucrețiu Mihailescu-Bîrliba’s
paper concerns the soldiers from rural area of Moesia Inferior recruited into
the Roman army, while Rada Varga and Annamaria-Izábella Pázsint focus
their research on the traders attested epigraphically in the Roman provinces of
Moesia, Dacia and Germania. Dan Aparaschivei’s study deals with the medi-
cal staff in the army of the Lower Danubian provinces. Valentin Piftor analy-
ses the phenomenon of age-rounding in the Roman army of Moesia Inferior,
and Florian Matei-Popescu discusses the territoria and regiones in the Lower
Danubian provinces. Two studies have an archaeological and numismatic
approach: Sever Boțan presents an overview of the distribution and circulation
of Roman glass vessels in the military settlements of the Danubian limes dur-
ing the first three centuries AD; Lucian Munteanu and Ștefan Honcu discuss
the economic issues related to silver plated coins in Roman Dacia. Ioan Piso
and Csaba Szabó present a detailed overview of Hecate’s cult in Sarmizege-
tusa, Ana Odochiciuc tries to identify those particulars who donated de sua
pecunia for the construction, restoration or decoration of public cult buildings.
The two last papers concern late antiquity. Alexander Rubel sets up several
ways of ‘client management’ via subsidies and gifts even in remote regions
behind the limes to influence and control inner-barbarian relations and con-
flicts. Nelu Zugravu proposes an analysis of frontier terminology in the Pan-
egyrici Latini.
The present book therefore proposes several steps to enhance our under-
standing of riparian settings in border regions of the Roman empire. These
riverine border regions are characterised by a significant presence of the
Roman military, extensive economic activity and religious interchange.
 INTRODUCTION XI

This volume is published within project PN-III-ID-PCE-2016-0271, which

is supported by the Romanian National Council for Scientific Research (CNCS
– UEFISCDI).
I should like to thank the editors of Colloquia Antiqua for accepting this
volume in their series, especially Prof. Gocha Tsetskhladze, the Series Editor,
for his comments, advice and for speeding it to publication, and Dr James
Hargrave for additional copy-editing and help with the index.
Lucrețiu Mihailescu-Bîrliba
 LIST OF ILLUSTRATIONS

CHAPTER 4 (Varga and Pászint)

Fig. 1. Representativeness of independent professionals in the Roman Latin

provinces.
Fig. 2. Representativeness of independent professionals in Germania, Dacia
and Moesia.
Fig. 3. The commercial route Ara Agrippinensium–Eburacum.
Fig. 4. Aurelius Aquila’s and Aurelius Flavus’ network.
Fig. 5. The commercial route Aquileia–Sarmizegetusa.
Table 1. Attested age groups of merchants.

CHAPTER 5 (Aparaschivei)

Table 1. Military physicians in the army of Moesia Superior.

Table 2. Medici ocularii in the Lower Danubian provinces.
Table 3. Military physicians in the army of Moesia Inferior.

CHAPTER 6 (Piftor)

Table 1. Moesia Inferior, deceased with very precise ages and their dedicators.
Table 2. Dacia, deceased with very precise ages and their dedicators.
Table 3. Moesia Inferior, Whipple’s index for the male population.
Table 4. Moesia Inferior, Whipple’s index for civilians.
Table 5. Moesia Inferior, Whipple’s index for soldiers.
Table 6. Dacia, Whipple’s index for the male population.
Table 7. Dacia, Whipple’s index for civilians.
Table 8. Dacia, Whipple’s index for soldiers.

CHAPTER 8 (Boțan)

Fig. 1. Map of the Lower Danubian limes with the places of provenance for
the glass finds.
Fig. 2. 1–3. (a–b) Early Roman glass fragments from Durostorum/Ostrov–
Călărași county (photograph: author); 4–5. (a–b) Early Roman glass fragments
XIV LIST OF ILLUSTRATIONS

from Barboși–Galați county (after Boțan 2015); 6. Millefiori glass fragment

from Poșta–Tulcea county (photograph: D. Paraschiv).
Fig. 3. 1–6. Roman glass fragments from Carsium/Hârșova–Constanța county
(photograph: author); 7–8. Roman glass fragments from Carsium/Hârșova–
Constanța county (after Chiriac 1998).
Fig. 4. 1–3. Roman glass fragments from Noviodunum/Isaccea–Tulcea county
(photograph: author); 4. Roman honeycomb cup from Noviodunum/Isaccea–
Tulcea county (after Boțan, Paraschiv and Nuțu 2010).
Fig. 5. 1–2. Roman glass fragments from Novae/Svishtov (after Biernacki
1973); 3. Roman ribbed bowl from Novae/Svishtov (after Billewicz 1975);
4. Roman lotus-bud beaker from Novae/Svishtov (after Belivanova 1999).
Fig. 6. 1–13. Roman glass fragments from Capidava–Constanța county (after
Matei 1987); 14–26. Roman glass fragments from Capidava–Constanța county
(after Matei 1988–89).
Fig. 7. 1–7. Roman glass fragments from Carsium/Hârșova–Constanța county
(after Chiriac 1999).
Fig. 8. 1–3. Roman glass fragments from Carsium/Hârșova–Constanța county
(after Chiriac 1999).
Fig. 9. 1–26. Roman glass fragments from Troesmis/Turcoaia–Tulcea county
(after Alexandrescu and Kainrath 2016); 27. Roman lotus-bud beaker from
Troesmis/Turcoaia–Tulcea county (after Boțan and Mocanu 2012); 28. Roman
unguentarium from Troesmis/Turcoaia–Tulcea county (after Boțan, Paraschiv
and Nuțu 2010).
Fig. 10. 1–14. Roman glass fragments from Oescus/Gigen (after Kabakčieva
2010).
Fig. 11. 1–10. Roman glass fragments from Oescus/Gigen (after Kabakčieva
2010).
Fig. 12. 1–7. Roman glass fragments from Novae/Svishtov (after Gencheva
2002).
Fig. 13. 1–9. Roman glass fragments from Novae/Svishtov (after Gencheva
2002).
Fig. 14. 1–9. Roman glass fragments from Novae/Svishtov (after Gencheva
2002).
Fig. 15. 1–10. Roman glass fragments from Novae/Svishtov (after Gencheva
2002).
Fig. 16. 1–11. Roman glass fragments from Novae/Svishtov (after Gencheva
2002).
 LIST OF ILLUSTRATIONS XV

CHAPTER 9 (Piso and Szabó)

Abb. 1. Die Votivplatte IDR III/2, 365.

Abb. 2. Zeichnung der Inschrift IDR III/2, 365.
Abb. 3. Die Votivplatte IDR III/2, 347.

CHAPTER 11 (Munteanu and Honcu)

Fig. 1. The chronological distribution of the share of denarii subaerati in the

civilian settlements of Dacia, Pannonia Inferior, Pannonia Superior, Noricum,
Germania Superior and Gallia Belgica (AD 98–244).
Fig. 2. The chronological distribution of the share of denarii subaerati in the
military fortifications of Dacia, Pannonia Inferior, Pannonia Superior and Ger-
mania Superior (AD 98–244).
Fig. 3. The chronological distribution of the share of antoniniani subaerati in
the civilian settlements of Dacia, Pannonia Inferior, Pannonia Superior, Nori-
cum and Germania Superior (AD 211–275).
Fig. 4. The chronological distribution of the share of antoniniani subaerati in
the military fortifications of Dacia, Pannonia Inferior, Pannonia Superior and
Germania Superior (AD 211–275).
Table 1. The share of denarii subaerati (D*) in the sites of Dacia.
Table 2. The chronological distribution of the share of denarii subaerati (D*)
in the civilian settlements (Civ) and military fortifications (Mil) of Dacia, Pan-
nonia Inferior, Pannonia Superior, Noricum, Raetia, Germania Superior, Ger-
mania Inferior and Gallia Belgica (AD 98–275).
Table 3. The share of antoniniani subaerati (A*) in the sites of Dacia.
Table 4. The chronological distribution of the share of antoniniani subaerati
(A*) in the civilian settlements (Civ) and military fortifications (Mil) of Dacia,
Pannonia Inferior, Pannonia Superior, Noricum, Raetia, Germania Superior,
Germania Inferior and Gallia Belgica (AD 98–275).
 LIST OF ABBREVIATIONS

AArchHung Acta archaeologica Academiae scientiarum Hungaricae.

AE L’Année Épigraphique.
AGRW R.S. Ascough, P.A. Harland and J.S. Kloppenborg, Associations
in the Greco-Roman World: A Sourcebook (Waco, TX 2012).
ANRW Aufstieg und Niedergang der römischen Welt (Berlin/New York
1970– ).
CCET Corpus cultus equitis Thracii (Études préliminaires aux reli-
gions orientales dans l’Empire romain 74) (Leiden 1979–84).
CIGD L. Ruscu, Corpus inscriptionum Graecarum Dacicarum (Debre-
cen 2003).
CIL Corpus Inscriptionum Latinarum, consilio et auctoritate Aca-
demiae litterarum regiae Borussicae editum (Berlin 1863– ).
CIMRM M.J. Vermaseren, Corpus Inscriptionum et Monumentorum
Religionis Mithriacae (The Hague 1956, 1960).
FMRD Die Fundmünzen der römischen Zeit in Deutschland (Berlin
1960– ).
FMRÖ Die Fundmünzen der römischen Zeit in Österreich (Vienna
1976– ).
FMRSl Die Fundmünzen der römischen Zeit in Slowenien (Berlin 1988– ).
FMRU Die Fundmünzen der römischen Zeit in Ungarn (Budapest/Bonn
1990– ).
IDR I.I. Russu et al., Inscriptiones Daciae Romanae (Bucharest
1975– ).
IDRE C.C. Petolescu, Inscriptions de la Dacie romaine. Inscriptions
externes concernant l’histoire de la Dacie (Ier–IIIe siècles)
(Bucharest 1996–2000).
IFS M. Peter, Augusta Raurica (Inventar der Fundmünzen der
Schweiz 3–4) (Lausanne 1996).
IGB G. Mihailov, Inscriptiones Graecae in Bulgaria repertae (Sofia
1956–97); I2 (Sofia 1970).
IGLN J. Kolendo and V. Božilova, Inscriptions grecques et latines de
Novae (Mésie Inférieure) (Paris 1997).
IGLR E. Popescu, Inscripţiile greceşti şi latine din secolele IV–XIII
descoperite în România (Bucharest 1976).
IGR R. Cagnat, J. Toutain and P. Jouguet, Inscriptiones Graecae ad
res Romanas pertinenetes (Paris 1911–27).
ILB B. Gerov, Inscriptiones Latinae in Bulgaria repertae (Sofia
1989).
ILD C.C. Petolescu, Inscripții latine din Dacia (Bucharest 1995).
ILJug A. and J. Šašel, Inscriptiones Latinae quae in Jugoslavia reper-
tae et editae sunt (Ljubljana 1963– ).
ILS H. Dessau, Inscriptiones Latinae Selectae (Berlin 1892–1916).
XVIII LIST OF ABBREVIATIONS

IMS F. Papazoglou, Inscriptions de la Mésie Supérieure (Belgrade

1976– ).
ISM Inscriptiones Scythiae Minoris Graecae et Latinae.
I. D.M. Pippidi, Histria et vicinia (Bucharest 1983).
II. I. Stoian, Tomis et territorum (Bucharest 1987).
III. A. Avram, Callatis et territorium (Bucharest/Paris 1999).
IV. E. Popescu, Tropaeum–Durostorum–Axiopolis (Bucharest/Paris
2015).
V. E. Doruţiu-Boilă, Capidaua–Troesmis–Noviodunum (Bucharest
1980).
LGPN P.M. Fraser et al., Lexicon of Greek Personal Names (Oxford
1987– ).
MAMA 10 B. Levick and S. Mitchell, Monumenta Asiae Minoris Antiqua
10: Monuments from Appian and the Upper Tembris Valley,
Cotiaeum, Cadi, Synaus, Ancyra, and Tiberiopolis recorded by
C.W.M. Cox, A. Cameron and J. Cullen (Journal of Roman
Studies Monograph 7) (London 1993).
Mich. P. III J.G. Winter, Michigan Papyyi III: Papyri in the University of
Michigan Collection. Miscellaneous Papyri (Ann Arbor 1936).
OPEL B. Lőrincz, Onomasticon Provinciarum Europae Latinarum
(Budapest 1994– ).
PGM K. Preisendanz, Papyri Graecae magicae (Leipzig 1928–41;
2nd ed., Stuttgart 1973–74).
PIR Prosopographia Imperii Romani, 2nd ed. (Berlin/Leipzig 1933–
2015).
RE Pauly-Wissowa, Real Encyclopädie der classichen Altertums-
wissenschaft (Stuttgart 1894–1972).
RGZM Römische Militärdiplome und Entlassungsurkunden in der
Sammlung des Römisch-Germanischen Zentralmuseums.
RICIS L. Bricault, Receuil des inscriptions concernant les cultes
isiaques (Mémoires de l’Académie des Inscriptions et Belles-
Lettres 31) (Paris 2005).
RMD M.M. Roxan and P.A. Holder, Roman Military Diplomas (Lon-
don 1978– ).
SEG Supplementum Epigraphicum Graecum (Lyons/Amsterdam
1923– ).
SIRIS L. Vidman, Sylloge inscriptionum religionis Isiacae et Sarapia-
cae (Religionsgeschichtliche Versuche und Vorarbeiten 28)
(Berlin 1969).
SM R.W. Daniel and F. Maltomini, Supplementum Magicum
(Abhandlungen der Rheinisch-Westfälischen Akademie der
Wissenschaften. Sonderreihe Papyrologica Coloniensia 16)
(Opladen 1990– ).
TAM E. Kalinka et al., Tituli Asiae Minoris (Vienna 1901– ).
TLL Thesaurus Linguae Latinae (Leipzig 1900– ).
ZPE Zeitschrift für Papyrologie und Epigraphik.
 CHAPTER 1

OVID: THE DOUBLE FACE OF THE DANUBE*

Marius ALEXIANU

Abstract
The author sets out to examine the parameters of Ovidian poetic perceptions and rep-
resentations concerning the Danube. He has targeted his research on analysis of the
sources on the Danube preceding Ovid; Ovid’s perceptions and representations con-
cerning the double reality of the Danube, namely its liquid and frozen form; the differ-
ent consequences of the liquid and frozen Danube on the human communities located
on the right bank of the river; and consideration of the historical credibility of Ovid’s
references concerning the Danube.

Relegated to Tomis, Ovid built his entire poetical venture from the Tristia and
the Epistulae ex Ponto so that the banishment order might be annulled1 or at
least to be exiled somewhere else. This is why literary or historical works have
for long dwelled on Ovid’s exaggerations, deformations, falsifications and lies,
even on the historical, geographical, ethnographical, etc. realities of the Istro-
Pontic area.2 In time, this has led to an over-exaggeration of the importance of
these poetic licences, to the point of considering the exile as fictitious.3 Despite
the fundamental critique of this hypothesis,4 this hypercritical attitude, which
refutes any trace of truth in Ovid’s accounts, has witnessed a resurgence dur-
ing recent years.5

*
This paper was written in the framework of project PN–III-P4-ID-PCE-2016-0271, financed
by the Romanian National Council for Scientific Research (CNCS – UEFISCDI): ‘Riverine soci-
eties. Dynamics of interaction between Roman culture and regional identities in Dacia and Moe-
sia inferior’.
1
For more on Ovid’s exile, cf. Bretzigheimer 1991; Ciccareli 2003, 126; McGowan 2009,
20.
2
For example Podossinov 1987; Rădulescu 1990.
3
Hartman 1905; Fitton Brown 1985.
4
Little 1990.
5
Cf., for example, Ezquerra 2010, who states that there is no evidence for or against the
exile; Bérchez Castaño 2015.
2 M. ALEXIANU

In the following pages, I set out to examine the parameters of the Ovidian
poetic perceptions and representations concerning the Danube. The two-fold
scope of this undertaking is to ascertain Ovid’s contribution to the knowledge
of the double-named river (Hister/Danubius)6 under various aspects in relation
with previous accounts, and to establish the degree of ‘reality’ of the claims
made by the great poet with respect to this topic.
To achieve this goal, I have targeted my research on the following: exami-
nation of the sources on the Danube preceding Ovid; Ovid’s perceptions and
representations concerning the double reality of the Danube, namely the liquid
and frozen forms; the different consequences of the liquid and frozen Danube
on the human communities located on the right bank of the river; and analysis
of the historical credibility of Ovid’s references concerning the Danube.

Sources Preceding Ovid

The written sources preceding Ovid concerning the freezing of the Danube are
rare and brief:
῾Ρῆνος καὶ ”Ιστϱος οἱ ποταμοὶ ὑπ᾿ ἄϱκτον ϱ῾έουσιν, ὁ μὲν Γεϱμανούς, ὁ δὲ
Παίονας παϱαμείβων˙ καὶ ϑέϱους μὲν ναυσίποϱον ἔχουσι τὸ ϱ῾εῖϑϱον, τοῦ δὲ
χειμῶνος παγέντες ὑπὸ χϱύους ἐν πεδίου σχήματι καϑιππεύονται (Aristotle
Mir. 168[182]).
Concrescunt subitae currenti in flumine crustae,/undaque iam tergo ferratos sus-
tinet orbes,/puppibus illa prius patulis, nunc hospita plaustris (Virgil Georgics 3.
360–362).
To summarise, Aristotle notes that the two northern rivers, the Rhine and the
Istros, are navigable in summertime and freeze in winter, thus crossable by
mounted men, while Virgil notes with a poetic touch that where once ships
passed, in winter chariots cross.7
For the issue at hand, I also find it opportune to recall Pausanias, who in the
second half of the 2nd century AD, benefiting from the accruation of geo-
graphical knowledge up to his time, distinguished a special category of rivers,
namely those that freeze in winter (among which are the Ister, Rhine and Hyp-
anis), which he calls winter rivers (cheimerious):
Ἴστϱον μέν γε καὶ ‘Ρῆνον, ἔτι δὲ Ὕπανίν τε καὶ Βοϱυσϑένην καὶ ὅσων ἄλλων
ἐν ὥϱᾳ χειμῶνος τὰ ϱ῾εύματα πήγνυται, τούτους μὲν χειμεϱίους κατὰ ἐμὴν
δόξαν ὀϱϑῶς ὀνομάσαι τις ἄν (Pausanias 8. 28. 2).

6
Luisi 2001, 129: Hister – 26 occurrences, Danuvius – three occurrences.
7
Alexianu 2006, 40–41.
 OVID: THE DOUBLE FACE OF THE DANUBE 3

I have deliberately limited the study to the analysis of Ovid’s poetical reflexes
towards the two states in which the Danube is found (sometimes even today),
namely the liquid state (cf. Tristia 3. 10. 8: ille suis liquidus bella repellit
aquis) and the frozen state (cf. 2. 4. 1: Accipe conloquium gelido Nasonis ab
Histro).

Textual Sequences Concerning the Danube in the Warm Seasons

While liquid, the Danube is hardly able to hold back the invading hosts:
Iazyges et Colchi, Metereaque turba Getaeque
Danubii mediis vix prohibentur aquis (Tristia 2. 189–192).
The poet stresses the joy he feels when the vernal sun melts the snow away,
and the unfreezing Danube stops the crossing of the Sarmatian wagons:
Et mihi sentitur nix verno sole soluta,
........................nec ut ante, per Histrum
Stridula Sauromates plaustra bubulcus agit (Tristia 3. 12. 27, 29–30).
The liquid Danube has the huge advantage of allowing the access of Roman
troops relieving the city of Troesmis:
Donec fluminea devecta Vitellius, unda
Intulit, exposito milite, signa Getis (Ponto 4. 7. 27–28).

Textual Sequences Concerning the Frozen Danube

The passages concerning the crossing of the frozen Danube can be divided into
two categories: those concerning the crossing of the Danube in contexts with-
out an expressis verbis connection with pillaging, and those which concern the
crossing of the Danube for this purpose. The first category includes several
passages that speak about the crossing of the Danube by Sarmatian or Iazyges
wagons:
Quaque rates ierant, pedibus nunc itur, et undas
Frigore concretas ungula pulsat equi;
Perque novos pontes, subter labentibus undis
Ducunt Sarmatici barbara plaustra boves (Tristia 3. 10. 31–34).
........................nec ut ante, per Histrum
stridula Sauromates plaustra bubulcus agit (Tristia 3. 12. 29–30).
ipse vides, onerata ferox ut ducat Iazyx
per medias Histri plaustra bubulcus aquas (Ponto 4. 7. 9–10).
In the second category, a key passage reveals how the two states (liquidus and
gelidus) of the river generate two antithetical situations: while it is warm, the
4 M. ALEXIANU

Danube serves as a sort of natural shield8 for the inhabitants of Tomis and the
neighbouring lands,9 holding back the barbarian attacks from the left side of
the river, while the arrival of winter, described at large10 brings the freezing of
the Danube, which allow its crossing by the barbarian invaders:
Dum tamen aura tepet, medio defendimur Histro:
ille suis liquidus bella repellit aquis.
At cum tristis hiems squalentia protulit ora,
terraque marmoreo est candida facta gelu,
+dum prohibet Boreas et nix habitare sub Arco
tum patet, has gentes axe tremente premi (Tristia 3. 10. 5–12).
Quid loquar, ut vincti concrescant frigore rivi,
Deque lacu fragiles effodiantur aquae ?
Ipse, papyrifero qui non angustior amne
Miscetur vasto multa per ora freto,
Caeruleos ventis latices durantibus, Hister
Congelat et tectis in mare serpit aquis (Tristia. 3. 10. 25–30).
The textual sequences concerning the Danube in wintertime are more numer-
ous than those concerning the liquid hypostasis. The poet shows remarkable
inventiveness when designating the frozen Danube. Thus, cold joins the two
shores together (Tristia 3. 10. 25: ut vincti concrescant frigore rivi), the Ister
freezes (3. 10. 29–30: Hister/ congelat), the waves are turned to stone by frost
(3. 10. 31–32: undas/frigore concretas), new ice bridges, under which water
flows, form on the Danube (3. 10. 33: novos pontes subter labentibus undis),
the smoothed Ister (3. 10. 53: protinus aequato siccis aquilonibus Histro), the
Ister froze three times from the cold (5. 10. 1: Ut sumus in Ponto, ter frigore
constitit Hister), the Ister came to a standstill from the cold (Ponto 1. 2. 81:
ubi frigore constitit Hister), the horses of the barbarians run on the river’s hard
back (1. 2. 83: dura meant celeri terga per amnis equo), the Ister turned to ice
(2. 4. 1: gelido…ab Histro), the barbarian Ister clots its cold-stopped waters
(3. 3. 25–26: Huc quoque venisti, pax est ubi tempore nullo/ Et coit astrictis
barbarus Hister aquis?), the icy Ister (Ibis 10. 138: frigidus Hister).
For illustrating as suggestively as possible this situation, Ovid employs vari-
ous categories of lexemes, such as nouns (frigus), adjectives (gelidus, frigi-
dus), verbs (congelare, concrescere, aequare, constituere, astringere, cogere),
at the same time resorting to conjuring metaphors when referring to the ice
sheet covering the river (novi pontes, dura terga).

8
Alexianu 2007, 33.
9
Pippidi 1971; 1984; Franga 1990; Mantzilas 2014.
10
Pippidi 1988; Nemeti 2009.
 OVID: THE DOUBLE FACE OF THE DANUBE 5

The immediate consequence of the freezing of the Danube is the crossing by

the barbarian enemies;11 skilled mounted archers, they lay waste to the neigh-
bouring lands:
Quaeque aliae gentes, ubi frigore constitit Hister,
Dura meant celeri terga par amnis equo (Ponto 1. 81–82).
Sive igitur nimii boreae vis saeva marinas,
Sive redundatas flumine cogit aquas,
Protinus aequato siccis aquilonibus Histro
Invehitur celeri barbarus hostis equo,
Hostis equo pollens longeque volante sagitta
Vicinam late depopulatur humum (Tristia 3. 10. 51–56).
The reactions of the wretched locals are presented with precision and compas-
sion at the same time: some flee, abandoning their fields, and their petty
wealth, their cattle and carts are plundered:
Diffugiunt alii, nullisque tuentibus agros
Incustoditae diripiuntur opes,
Ruris opes parvae, pecus et stridentia plaustra
Et quas divitias incola pauper habet (Tristia 3. 10. 57–60).
This category of denizens is the lucky one, as others are taken prisoners (look-
ing back despondently on their fields and homesteads), while others are killed
by poisoned arrows; what the enemies cannot take with them is set on fire:
Pars agitur vinctis post tergum capta lacertis,
Respiciens frustra rura Laremque suum,
Pars cadit hamatis misere confixa sagittis:
Nam volucri ferro tinctile virus inest.
Quae nequeunt secum ferre aut abducere, perdunt
Et cremat insontes hostica flamma casas (Tristia 3. 10. 61–66).
Starting from these realities, the general mental state of the locals is that of
continuous stress, of fear for the enemy, whether seen or unseen, which makes
them cease agricultural work and thus leaving the lands untilled:
Tunc quoque, cum pax est, trepidant formidine belli,
nec quisquam presso vomere sulcat humum.
Aut videt aut metuit locus hic, quem non videt, hostem;
Cessat iners rigido terra relicta situ (Tristia 3. 10. 67–69).

11
Della Corte 1978.
6 M. ALEXIANU

Credibility Issue

The Credibility of Ovid’s Own Account

Ovid is aware that his accounts concerning the freezing of the Euxine and the
Danube are hard to believe for his readers in Rome, but he stresses that in the
absence of any reward for false testimony, the reader must accept that the wit-
ness is of good faith:
Vix equidem credar, sed, cum sint praemia falsi
nulla, ratam debet testis habere fidem! (Tristia 3. 10. 35–36).

Historical Credibility
This aspect of the Ovidian opera has for long enjoyed attention.12 However,
concerning strictly the topic at hand, we must also consider newer approaches.
In addressing this issue, I find it relevant to perform an analysis in autoptic,
heteroptic and autopatic terms. Starting from a lexical analysis, it was recently
demonstrated that ‘Ovid’s perception of the space he had ben exiled is simul-
taneously autoptic, heteroptic and autopathic’.13 The autoptic perception is
evinced by an entire series of verbal forms, such as: tangam/tactam (to touch),
visere/vidimus/vidisse/video (to see), adspiciat/adspiceres (to look), adest (to
be present); the autopathic perception is illustrated by verbal forms such as:
perpetior (to undergo or experience hardships, sufferings to the full), premor
(to exert a steady or continuous force against, apply pressure to press), pati/
patior (to be subjected to an operation or process, undergo), aeger eram (ill,
unwell, sick), iaceo (to lie), vivere (to be alive, live) and cingunt (to surround,
encircle). Finally, the heteroptic perception is revealed by the use of verbal
forms such as constat (to consist) and dicitur (to talk, to speak).14
This approach is, sine dubio, valid for Ovid’s entire exile opera. But is this
conclusion also valid with a strict reference to the perceptions and representa-
tions of the Danube?
Ovid’s autopathic reaction is conspicuous in a passage in which he bemoans
expressly the cold accompanying the enemy invasions, assuming that the
reader has become habituated with the idea that in this part of the world cold
weather often leads to the freezing of the Danube:
Frigus et incursus omni de parte timendos
Et quod pulsetur murus ab hoste, queror (Ponto 4. 14. 27–28).
dura iubet gelido Parca sub axe mori (Ponto. 4. 15. 36).

12
Syme 1979.
13
Curcă 2014, 376.
14
Curcă 2014, 376–77.
 OVID: THE DOUBLE FACE OF THE DANUBE 7

In respect of Ovid’s autoptic reactions, there is no linguistic clue that can

prove it. Similarly, there are no linguistic clues that could indicate heteroptic
reactions. But a careful textual examination clearly reveals that Ovid had not
witnessed and recorded as a reporter each time the river froze, the barbarians’
invasions, their plundering and the repeated sufferings endured by the locals.
The only way to explain such details is that the poet turned to information
provided by the locals (particularly non-Greek but also Greeks). The capacity
of omniscient narrator that Ovid assumes cannot be explained otherwise.
Thus, as to Ovid’s reactions towards the frozen Danube, they are distinctly
autopathic and heteroptic, just like in the case of the river in its liquid state, as
clearly shown by the aforementioned passages.
In favour of the historical credibility of Ovid’s accounts concerning this
aspect, we can bring forward a passage which, though it belongs to a later
author, describes an erstwhile event, from the age of king Cotis (second half of
the 1st century BC). Relying inevitably on earlier uncited sources, Florus
(2nd century AD) noted that any time the Danube froze, the Dacians had the
habit of crossing it and ravage the neighbouring lands:
Daci montibus inhaerent. Inde Cotisonis regis imperio, quoties concretus gelu
Danuvius inunxerat ripas, decurrere solebant et vicina populari (Florus 2. 28
[4. 12]. 18).
In the same vein, one may also mention an author from the 4th–5th century,
who claimed that the city of Halmyris was occupied by barbarians who crossed
the river:
Ἀλλ’ ἡ μὲν Ἁλμυϱὶς κϱυσταλλωϑέντος τοῦ Ἴστϱου ὑπὸ τῶν διαβάντων αυτὸν
βαϱβάϱων ἁλίσκεται (Philostorgios Eccl. Hist. 10. 6).
The very fall of Rome ensued from the crossing of the frozen Danube by the
Huns:
…παγέντα τὸν ποταμὸν διαβάντες, ἀϑϱόως εἰς τὴν ῾Ρώμην εἰσήλασαν…
(Philostorgios Eccl. Hist. 11. 8).
Evidence in favour of Ovid’s historical credibility continues to mount even
today. For instance, several records not specifically mentioning the Danube are
confirmed by epigraphic15 and archaeological16 evidence, and recently it has
been argued that ‘Ovid is the first evidence to be brought into the discussion
on the Thraciae strategiae in Scythia minor.’17

15
For example, Bărbulescu and Buzoianu 2014; Matei-Popescu 2017.
16
Irimia 2009.
17
Matei-Popescu 2017; 2018.
8 M. ALEXIANU

Conclusions

Ovid too took notice of something for long remarked by ancient authors,
namely that the river froze in wintertime and thus allowed its crossing. But,
summotus ad Histrum (Ponto 2. 4. 91), the poet goes beyond simple mention-
ing this fact, and insists in detail upon the consequences of a human and strategic-
military nature.
Thus, while it is liquidus, the Danube is able to protect, even if barely, the
civilised communities or those under the influence of the Greek civilisation
between the Ister and the Euxine. Conversely, when the river becomes gelidus,
this strategic function is voided. When this occurs, the Danube can be crossed
by aggressive barbarian warriors, who lay waste the rural communities from
the right bank and seriously endanger the city of Tomis. The often-freezing
Danube aggravates the latent conflict between the two antithetical types of
riparian society. One type is defined in the subtext by hellenismos and incipi-
ently by romanitas, while the other type is characterised, in line with the com-
mon Roman mindset, as barbaries/immanitas.18
Even though they belong to a poet who, in order to obtain the annulment of
his exile order or at least removal to another place, resorts to exaggerations or
even lies,19 the Ovidian accounts of the consequences of the two states of the
river evince a clear visual knowledge,20 even if it reflects the previous percep-
tions of the inhabitants of Tomis, as well as of those from the area between
the Danube and the Euxine Sea. This is why I hold that the presentation of the
double strategic functionality of the Danube has a high degree of historical
credibility. The double face of the Danube represents a key constant in the
unfolding of day-to-day life on both banks of the river, a condition intrinsic to
societal evolution in this area, one that oscillated between periods of security
and insecurity.

BIBLIOGRAPHY

Alexianu, M. 2006: ‘Imaginaire et propagande: Virgile et Horace sur les Gètes et les
Daces’. Classica et Christiana 1, 39–50.
—. 2007: ‘Perceptions subjectives du Danube chez les auteurs grecs et latins’. Istros
14, 27–39.
Bărbulescu, M. and Buzoianu, L. 2014: ‘L’espace ouest-pontique sous l’empereur
Tibère à la lumière d’un décret inédit découvert en Dobroudja’. In Cojocaru, V.,

18
Lassandro 2000, 60–63.
19
Luisi 2001, 132: ‘le bugie ovidiane’.
20
Alexianu 2007, 31.
 OVID: THE DOUBLE FACE OF THE DANUBE 9

Coşkun, A. and Dana, M. (eds.), Interconnectivity in the Mediterranean and

Pontic World during the Hellenistic and Roman periods (Cluj-Napoca), 415–34.
Bérchez Castaño, E. 2015: El destierro de Ovidio en Tomis: realidad y ficción
(Valencia).
Bretzigheimer, G. 1991: ‘Exul ludens. Zur Rolle von relegans und relegatus in Ovids
Tristien’. Gymnasium 98, 39–76.
Ciccarelli, I. 2003: Commento al II libro dei Tristia di Ovidio (Bari).
Curcă, R.-G. 2014: ‘Greek and Latin Authors on the Carpathian-Dniestrian Territory:
An Anthropology of Perceptions’. Studia antiqua et archaeologica 20, 369–80.
Della Corte, F. 1978: ‘Ovidio e i barbari danubiani’. In Della Corte, F., Opuscula,
vol. 6 (Genoa), 293–305.
Ezquerra, A.A. 2010: ‘Ovid in Exile: Fact or Fiction?’. Analele Științifice ale
Universității Ovidius Constanța, Seria Filologie 21, 107–26.
Fitton Brown, A.D. 1985: ‘The unreality of Ovid’s Tomitan exile’. Liverpool Classical
Monthly 10.2, 18–22.
Franga, L. 1990: ‘Ovidius și spațiul danubiano-pontic’. Thraco-Dacica 11.1–2,
225–38.
Hartman, J.J. 1905: De Ovidio poeta commentatio (Leyden).
Irimia, M. 2009: ‘Bastarnii și sarmații – realități istorice la Dunărea de Jos – și
percepția lor în opera lui Ovidiu’. In Buzoianu, L. (ed.), Interferențe ovidiene
(Constanța), 111–21.
Lassandro, D. 2000: Sacratissimus imperator: L’immagine del pinceps nell’oratoria
tardoantica (Quaderni di Invigilata lucernis 8) (Bari).
Little, D.A. 1990: ‘Ovid’s last poems: Cry of pain from exile or literary frolic in
Rome’. Prudentia 22, 23–39.
Luisi, A. 2001: ‘Ovidio e il Danubio’. Studia Antiqua et Archaeologica 8, 127–34.
Mantzilas, D. 2014: ‘Le témoignage d’Ovide sur les peuples de la région du Pont-
Euxin’. In Březina, P. (ed.), Pontus Euxinus (Commentarii Pilsnenses) (Pilsen),
15–36.
McGowan, M.M. 2009: Ovid in Exile. Power and Poetic Redress in the Tristia and
Epistulae ex Ponto (Leiden/Boston).
Matei-Popescu, F. 2017: ‘Ovid at Tomis: the Early History of the Left Pontus under
the Roman Rule, Civiltà Romana’. Rivista pluridisciplinare di studi su Roma
antica e le sue interpretazioni 4, 17–25.
—. 2018: ‘The Thracian strategiae in Scythia Minor’. In Boteva-Boyanova, D., Delev,
P. and Tzvetkova, J. (eds.), Society, Kings, Gods: In memoriam professoris
Margariatae Tachevae (Sofia), 107–18.
Nemeti, S., 2009: ‘Scythicum frigus. Repères pour une histoire du climat au Bas-
Danube (Ier siècle ap. JC.)’. In Hermon, E. (ed.), Sociétés et climats dans l’Empire
Romain: pour une perspective historique et systémique de la gestion des res-
sources en eau dans l’Empire romain (Naples), 411–27.
Pippidi, D.M. 1971: I Greci nel Basso Danubio: dall’età arcaica alla conquista
romana (Milan).
—. 1984: ‘Tomis, cité géto-grecque à l’époque d’Ovide?’. In Pippidi, D.M., Parerga:
écrits de philologie, d’épigraphie et d’histoire ancienne (Bucharest/Paris).
—. 1988: ‘În legătură cu iarna “scitică”’. In Pippidi, D.M. (ed.), Studii de istorie și
epigrafie (Bucharest), 52–56.
10 M. ALEXIANU

Podossinov, A.V. 1987: Ovids Dichtung als Quelle für die Geschichte des Schwarz-
meergebiete (Xenia 19) (Konstanz).
Rădulescu, A. 1990: Ovidio nel Ponto Eusino (Sulmona).
Syme, R. 1979: History in Ovid (Oxford).
 CHAPTER 2

CODE-SWITCHING: EXPRESSION FORMS

OF LINGUISTIC IDENTITY IN MOESIA INFERIOR*

Roxana-Gabriela CURCĂ

Abstract
This study offers a survey of the epigraphic material on code-switching, as a linguistic
identity pattern, following historical and philological aspects. The author examines the
taxonomy of code-switching (tag-switching, inter-sentential and intra-sentential),
the determinant factors that trigger this phenomenon, the influential role of stone-
cutters and the social environment in which it appears, seeking to differentiate between
the rural and urban milieux, the civilian and military contexts, and the ethnic origin of
the users.

Introduction

Given the bilingual background of Moesia Inferior, this study undertakes

a survey of the epigraphic material on code-switching, as a linguistic identity
pattern, following historical and philological aspects. The list of epigraphic
data is by no means exhaustive, but representative for both the Greek- and
Latin-speaking areas of the province. I examine the taxonomy of code-switch-
ing evidences (tag-switching, intra-sentential and inter-sentential), the determi-
nant factors that trigger this phenomenon, the social environment in which it
appears – trying to draw out differences between the rural and urban milieux,
the civilian and the military contexts, and the ethnic origin of the users.
Moesia Inferior was a province strongly defined by two crucial coordinates,
Hellenism and Romanism, as well as by the existence of a Thracian substrate.
These combine to yield an ethno-linguistic mosaic.1 The coexistence of two

*
This work was supported by a grant from the Romanian National Council for Scientific
Research (CNCS – UEFISCDI), project PN-III-P4-ID-PCE-2016-0271, contract no. 63/2017,
‘Riverine societies. Dynamics of interaction between Roman culture and regional identities in
Dacia and Moesia inferior’.
1
Alexianu 2004.
12 R.-G. CURCĂ

dominant languages occasioned the inevitable Greek-Latin and Latin-Greek

interferences, a direct consequence of practising bilingualism, more or less
intended. This phenomenon can be observed not only in the extremely scarce
attestations of bilingual inscriptions, but also in the inscriptions written in
Greek or Latin (for example, the insertion of a Latin ending after the influence
of a Greek declension, the attestation of Latinisms and Graecisms as means of
expressing typically Roman/Greek realities, the existence of formulaic equiva-
lencing as demanded by epigraphic habit, etc.). In this bilingual milieu, the
dominance of Greek over Latin and vice versa is a constant switch, observable
at all phono-morpho-lexical and formulaic levels.2

Theoretical Framework

Since it concerns an aspect of bilingualism with multiple implications from

other fields (such as socio- and psycholinguistics, anthropology, etc.), an
extended definition of code-switching has hardly been agreed upon.3 This pre-
dicament is also because mutual intrusions between two or more languages in
the same communication setting cannot always be symmetrically quantifiable
or subjected to a set of clear and precise rules. From among the accepted defi-
nitions of code-switching,4 I will employ that proposed by J.N. Adams, appli-
cable to epigraphic material. He considers the phenomenon as ‘a full-blown
switch from one language into another within one person’s utterance or piece
of writing’.5
A crucial aspect when discussing code-switching is to distinguish it from
the concept of interference. The distinction between the two aspects of bilin-
gualism is difficult to operate, but it has often been remarked that interference
is marked by the lack of intentionality, while code-switching is seen as a delib-
erate act.6 In the taxonomy of this phenomenon, the main categories should be
mentioned:7
1. Tag-switching (insertion of a Greek tag into a Latin text or of a Latin for-
mula into a Greek text);

2
Alexianu 2005; Bechet 2009, 114; Galdi 2008; Destephen 2011; Curcă 2004; 2011; 2012,
135–78.
3
Nilep 2006, 1.
4
I mention those by Myers-Scotton 2006, 239: ‘the use of two language varieties in the same
conversation’; or by Gardner-Chloros 2009, 4: ‘the use of several languages or dialects in the
same conversation or sentence by bilingual people’. About the motivation for occurrences of
code-switching, see Hoffmann 1991, 116; Adams 2003, 356–69.
5
Adams 2003, 19.
6
Adams 2003, 28. See also a detailed analysis of the differences between code-switching,
interferences and borrowing in Adams 2003, 18–29; Mullen 2012, 18–21.
7
Poplack 1980, 594.
 CODE-SWITCHING: EXPRESSION FORMS OF LINGUISTIC IDENTITY 13

2. Intra-sentential (occurs in the limits of the same sentence8).

3. Inter-sentential (one or more sentences are written in Greek and the follow-
ing one(s) are in Latin or vice versa9).

Occurrences in Moesia Inferior

The instances of code-switching in the epigraphic corpus of the province of

Moesia Inferior have been studied sequentially in the specialised literature,10
but without being investigated from the complex perspective of the parameters
listed in the introduction to this paper.
Funerary inscriptions contain the bulk of the attestations. It is important, in
this case, to identify the connection between the dedicator and the deceased,
their social and juridical status, the linguistic affiliation, the onomastics, the
origin of the individual(s), the motivation; these can vary from case to case,
and is suggested to a certain extent both by the ‘matrix–embedded language’
ratio11 and the monolingual or bilingual status of the patron and/or the person(s)
involved. For example, isolated Greek terms, tags, specific endings or even
specific letters inserted into Latin sentences could mark the recognition of the
cultural prestige of the Greek language and, implicitly, of Greek civilisation.
Conversely, Latin insertions into a Greek background could point to an accept-
ance of the state language.

Tag-switching

A bilingual epitaph from Histria from the 2nd century AD12 honours Apollinaris
Dolichenaris, the patron being his brother, Aurelius Marcus. The inscription
presents a typical case of tag-switching, with the established formula Dis
Manibus written in Latin, and the rest of the inscription in Greek. Besides the
age at which Apollinaris Dolichenaris died, namely at 35, and the onomastics
of the two individuals,13 the inscriptions provides no other information.

8
Let me also mention the classification proposed by Muysken (2000) for intra-sentential
code-switching in alternation, insertion and congruent lexicalization. Adams (2003, 71) also
speaks about character-switching as ‘Greek letters intrude into Latin script or vice versa’.
9
For opinions regarding the inclusion of tag-switching in the category of inter-sentential
code-switching, see Hoffmann 1991, 104; Mullen 2012, 18.
10
Galdi 2004, 117; Curcă 2011, 74, 78; 2012, 149. About the morphological code-switching
reflected in anthroponymy, see the important research of Stoev 2017, 166, 167, 172, 181, 200.
11
Myers-Scotton 1993, 20.
12
ISM I, 283.
13
For other attestations of Jupiter Dolichenus in this area, see the commentary on inscription
ISM I, 283.
14 R.-G. CURCĂ

D(is) M(anibus)
ἐνϑάδε κεῖται Ἀπολ-
λινάϱις Δολιχην[άϱ]ις
ζήσας ἔτη λέ· Τὴν
στήλην ἀνέϑηκε
Αυϱ(ήλιος) Μάϱκος ὁ ἀδελ-
φὸς μνείας χάϱιν.
Χαῖϱε παϱοδεῖτα.
In the case of a bilingual dedication ὐ discovered at Dolna Bešovica, tag-
switching is illustrated by the opening formula Ἀγαϑ[ῆι] τύχ[ηι] written in
Greek, while the rest of the inscription is in Latin.14 Aurelius Pudens, serving
as strator consularis, erected a shrine to deities with great dominion in the
rural area, such as Jupiter Optimus Maximus Rector, Juno Regina, Minerva
and Vulcan. The anthroponym Aurelius Pudens is a clue not only for dating
the inscription (3rd century AD), but also betrays his Greek ethnicity.
Ἀγαϑ[ῆι] τύχ[ηι]
Iovi o(ptimo) m(aximo) rector(i),
Iunoni regin(ae),
Miner(vae), Victo(riae),
Volk(ano), Mercur(io),
fatis divinis
Aur(elius)
Pudens
Strat(or) co(n)s(ularis) v(otum) l(ibens) p(osuit).
Likewise from the rural area, from vicus Trullensium, comes another bilingual
inscription from the 2nd–3rd century AD, which opens with a dedication to
Jupiter Optimus Maximus, and from which we find the names of the two mag-
istrates of this vicus, viz. Publius Aelius Attalus and Titus Aelius Secundus.
Tag-switching occurs in this case through the ending formula, Ἀγαϑοκ(λῆς)
ἐπο[ί]ει, which concludes the inscription in Greek.15
I(ovi) o(ptimo) m(aximo)
pro s(alute) Imp(eratoris) vic(ani)
Trullens(es) per
magi(stros) P(ublium) Ael(ium) At-
talum et T(itum) Ael(ium) Secun-
dum.
Ἀγαϑοκ(λῆς) ἐπο[ί]ει.

14
ILB 156.
15
ILB 183.
 CODE-SWITCHING: EXPRESSION FORMS OF LINGUISTIC IDENTITY 15

Intra-Sentential Switching (Lexical Morphological and Graphical)

An interesting case is revealed by a Greek epitaph from the 2nd–3rd century

AD discovered in Tomis, where the only two mentioned anthroponyms are
written in Latin (most probably a Greek-speaking family who named their
daughters Iunia Dometia and Iunia Nike).16 A glance at the Latin inscriptions,
especially from the Hellenophone area, shows a pronounced tendency to insert
Greek endings into Latin words. I will illustrate morphological switching with
a few of the most representative cases,17 such as: -es instead of -is for singular,
Genitive, third declension (for example legiones18); -um instead of -orum for
Genitive, plural, second declension (Deum19); -as instead of -ae for Genitive,
singular/Nominative, plural, first declension (such as Nicandras20). An inter-
esting situation can be found in the written form of the anthroponym Ti(berius)
Claudios Zηnotos, recorded with Latin characters, where the insertion of the
long vowel –η is an instance of character-switching.21

Inter-Sentential Code-Switching

Most of the bilingual epigraphic documents in Moesia Inferior are more or less
equal in terms of content in Greek and Latin. To illustrate the phenomenon of
inter-sentential code-switching, I mention an epitaph from Odessos, in which
the content of the first part, written in Latin, is not found in the Greek part.22
Since the inscription was found in the Greek town of Odessos, the switching
to Greek must not surprise us. In the epitaph, Antistia Firmina23 is commemo-
rated by her husband, Malius Secundus, a b(ene)f(iciarius) co(n)s(ularis).
D(is) M(anibus)
et memoriae Antistiae Firmi-
ne coiugi rarissime, quae vixit

16
ISM VI.2, 520 (with commentary). Taking into account the statement by Gardner-Chloros
(1987, 102) that ‘a Ioan is a code-switch with a full-time job’, Latinisms and Graecisms also fall
into this category: for example, κανδίδατον, ISM III, 1145; Curcă 2004.
17
Also falling into this category are different Greek words inserted into Latin clauses, such
as the Latin transliteration of the article τ| from Greek to the singular, in Dative case ILB 1767;
Galdi 2004, 117.
18
ISM II, 374(210); Galdi 2004, 194.
19
ILB128; Galdi 2004, 141–43.
20
ISM V, 72, Galdi 2004, 24.
21
AE 1985, 762.
22
IGB I2 218.
23
See Curcă and Zugravu 2005, 318 for the occurrence of Antistii originars from Ancyra,
particularly the case of C. Antistius Valens, veteran of the legio V Macedonica; Mihailescu-
Bîrliba and Curcă 2018, 107.
16 R.-G. CURCĂ

mecum ann(os) XXIII. Malius Secundus

b(ene)f(iarius) co(n)s(ularis) maritus fecit me poni.
εἴ τις τούτῳ ἡϱῴῳ, ὃπου κεῖτε
ἡ πϱογεγϱαμμένη, ϑελήσι ἄλλον
ἕτεϱον ϑεῖναι, δώσι τῷ ταμίῳ (δηνάϱια) βφ᾿
καὶ τῇ ᾿Οδησσειτῶν πόλι (δηνάϱια) βφ᾿.

Conclusions

Where, how and, most importantly, why does code-switching occur in inscrip-
tions? These are the main questions that our study tried to answer in order
properly to understand their occurrence in the epigraphic material of Moesia
Inferior.
As can be seen, the explanation for code-switching in the specialist litera-
ture does not go beyond the general idea of a Greek language influence on
Latin and vice versa at different levels of communication, which though accu-
rate in many cases is far from explaining the complexity of the phenomenon.
Most of the code-switching testimonies occur in the Hellenophone area of the
province, as expected, both in the rural (for example vicus Trullensium) and
urban milieu, in the civil and military environment (for example Aurelius
Pudens strat[or] co[n]s[ularis]). The individuals involved in code-switching
either originate from a Greek-speaking area, or, in the case of couples, could
be both Greek speakers.
The inscriptions of Moesia Inferior show code-switching at both the formu-
laic and the morphologic and graphical levels. On the flexional level, the Latin
forms breaking from the norm, identified above, may be due to contaminations
with Greek flexion, normal in a bilingual area, as well as to certain develop-
ments in Latin phonetics, when they cannot be explained away by lapicidal
errors. As to code-switching in Latin inscriptions, it reflects the intentionality
and volitive character of the native Hellenophones to underline certain particu-
larities of the Latin language that were not reflected in the standardised Latin
alphabet. In some cases, it is very difficult to draw a distinction between the
historical phenomena intrinsic to Latin on the one hand, and those resulting
from the contact between Latin and Greek in the case of bilingual speakers on
the other. In the case of tag-switching, we can observe that it occurs both at the
start and at the end of a bilingual inscription, whether it concerns a salutation
formula or one mandated by the typology of the inscription (for example Dis
Manibus). Even if few bilingual inscriptions have been found in this province,
intra-sentential code-switching with morphological implications can be
observed on a much wider sample of Latin inscriptions, bearing witness to the
 CODE-SWITCHING: EXPRESSION FORMS OF LINGUISTIC IDENTITY 17

practice of bilingualism on a considerable scale. The extra-linguistic context

plays a key role for understanding the motivations that led to the epigraphic
records of code-switching: the possibility that certain terms from a language
are not known; the wish to express the feeling of belonging to a certain ethnic-
ity; stressing lingustic, cultural and professional identity; etc. Study of the
epigraphs from this area from the linguistic perspective points to purely Latin
and purely Greek linguistic identities, and also to an emergent one of glottic
interferences. The results of our case study on the code-switching in Moesia
Inferior confirm similar occurences of this phenomenon attested in other areas
of the Roman empire.

BIBLIOGRAPHY

Adams, J.N. 2003: Bilingualism and the Latin Language (Cambridge).

Alexianu, M. 2004: ‘La situation linguistique de la province romaine Scythie Mineure.
Repères d’une recherche’. In Santelia, S. (ed.), Italia e Romania: Storia, Cultura
e Civiltà a confronto (Atti del IV Convegno di Studi italo-romeno, Bari,
21–23 ottobre 2002) (Quaderni di Invigilata Lucernis 21) (Bari), 145–56.
—. 2005: ‘Les inscriptions bilingues privées de Tomi et de Histria’. In Cojocaru, V.
(ed.), Ethnic Contacts and Cultural Exchanges North and West of the Black Sea
from the Greek Colonization to the Ottoman Conquest (Iaşi), 305–12.
Bechet, F. 2009: ‘Les colonies grecques du Pont Gauche: histoire et langue’. In
Vottéro, G. (ed.), Le grec du monde colonial antique I: Le N. et N.-O. de la Mer
Noire (Actes de la Table Ronde de Nancy, 28–29 septembre 2007) (Études
anciennes 42) (Nancy/Paris), 81–118.
Curcă, R.-G. 2004: ‘Les latinismes dans les inscriptions grecques de la Scythie
Mineure’. In Santelia, S. (ed.), Italia e Romania: Storia, Cultura e Civiltà a con-
fronto (Atti del IV Convegno di Studi italo-romeno, Bari, 21–23 ottobre 2002)
(Quaderni di Invigilata Lucernis 21) (Bari), 247–51.
—. 2011: ‘The bilingual inscriptions of Moesia Inferior: the historiographic frame-
work’. Classica et Christiana 6.1, 71–80.
—. 2012: Elenism și romanitate în Moesia Inferior: interferențe etnice și lingvistice
(Iași).
Curcă, R. and Zugravu, N. 2005: ‘“Orientaux” dans la Dobroudja romaine. Une
approche onomastique’. In Cojocaru, V. (ed.), Ethnic Contacts and Cultural
Exchanges North and West of the Black Sea from the Greek Colonization to the
Ottoman Conquest (Iaşi), 313–29.
Destephen, S. 2011: ‘La coexistence du grec et du latin en Illyricum (Ier–VIe siècle)’.
In Ruiz Darasse, C. and Luján, E.R. (eds.), Contacts linguistiques dans l’Occident
méditeranéen antique (Collection de la Casa de Velázquez 126) (Madrid),
129–44.
Galdi, G. 2004, Grammatica delle iscrizioni latine dell’impero (province orientali).
Morfosintassi nominale (Rome).
18 R.-G. CURCĂ

—. 2008: ‘Aspects du bilinguisme gréco-latin dans la province de la Mésie Inférieure’.

In Biville, F., Decourt, J.-C. and Rougemont, G. (eds.), Bilinguisme gréco-latin et
épigraphie (Actes du colloque organisé à l’Université Lumière-Lyon 2. Maison
de l’Orient et de la Méditerranée-Jean Pouilloux, UMR 5189 Hisoma et JE 2409
Romanitas, les 17, 18 et 19 mai 2004) (Lyons), 141–54.
Gardner-Chloros, P. 1987: ‘Code-Switching in Relation to Language Contact and Con-
vergence’. In Lüdi, G. (ed.), Devenir bilingue – parler bilingue (Actes du 2e col-
loque sur le bilinguisme, Université de Neuchâtel, 20–22 septembre 1984)
(Linguistische Arbeiten 169) (Tübingen), 99–113.
—. 2009: Code-Switching (Cambridge).
Hoffmann, C. 1991: An Introduction to Bilingualism (London/New York).
Jacobson, R. 1998: Codeswitching Worldwide (Berlin/New York).
Mihailescu-Bîrliba, L. and Curcă, R.-G. 2018: ‘Military Presences in Bilingual Inscrip-
tions from Moesia Inferior’. In Fink, S., Lang, M. and Schretter, M. (eds.), Mehr-
sprachigkeit. Vom Alten Orient bis zum Esperanto (Münster), 105–12.
Mullen, A. 2012: ‘Introduction: Multiple Languages, Multiple Identities’. In Mullen,
A. and James, P. (eds.), Multilingualism in the Graeco-Roman World (Cam-
bridge), 1–35.
Muysken, P. 2000: Bilingual Speech: A Typology of Code-Mixing (Cambridge).
Myers‐Scotton, C. 1993: Duelling Languages: Grammatical Structure in Codeswitch-
ing (Oxford).
—. 2006: Multiple Voices: An Introduction to Bilingualism (Oxford).
Nilep, C. 2006: ‘“Code Switching” in Sociocultural Linguistics’. Colorado Research
in Linguistics 19 (https://philpapers.org/archive/NILCSI.pdf).
Poplack, S. 1980: ‘Sometimes I’ll start a sentence in Spanish y termino espanol:
Toward a Typology of Code-Switching’. Linguistics 18, 581–618.
Stoev, K. 2017: ‘The Hereditary Nomenclature in Moesia Inferior and its Value as a
Source for the Study of Identities’. In Gavrielatos, A. (ed.), Self-Presentation and
Identity in the Roman World (Cambridge), 160–206.
 CHAPTER 3

SOLDATS DU MILIEU RURAL

DE LA MÉSIE INFÉRIEURE
RECRUTÉS DANS L’ARMÉE ROMAINE*

Lucreţiu MIHAILESCU-BÎRLIBA

Abstract
The author identifies inscriptions that mention soldiers originating from the rural
milieu of the Roman province Moesia Inferior. These military were more numerous
than we might imagine. The time of enlistment is also important in order to establish
the main reasons for recruiting soldiers from the countryside of this province.

1. Introduction

J’ai abordé le problème des militaires recrutés du milieu rural de Mésie Infé-
rieure à une autre occasion, lorsque j’ai mentionné quatre militaires originaires
des villages situés dans les territoires de Nicopolis ad Istrum et de Novae qui
rentraient chez eux après la fin de leur service.1 Les soldats sont certainement
plus nombreux; j’ai expliqué qu’une des raisons pour lesquelles je n’ai discuté
ce problème qu’en parlant uniquement de vétérans est l’incertitude concernant
le lieu de provenance des diplômes militaires, les sources principales pour ce
genre d’enquêtes. Bien que ces diplômes aient été trouvés, fort probablement,
sur le territoire des anciennes provinces de Mésie Inférieure et de Thrace, il est
encore difficile d’en faire une distinction nette. Les bénéficiaires sont des
Thraces et des Daces. Si, en ce qui concerne les Daces, on peut être plus sûr
qu’ils ont habité la Mésie Inférieure (même si, après 102, la Dacie ne saurait
être exclue), pour les Thraces, il n’y a aucune certitude quant à leur lieu d’ins-
tallation, sauf dans les cas où ceux-ci sont explicitement mentionnés. Même la

* Cet article a été réalisé dans le cadre du projet PN-III-P4-ID-PCE-2016-0271 de la Conseil

National de la Recherche Scientifique roumain (CNCS – UEFISCDI).
1
Mihailescu-Bîrliba et Răileanu 2014. Les sources: CIL XVI, 83, 143; RMD II, 132; RMD
IV, 311.
20 L. MIHAILESCU-BÎRLIBA

désignation comme Bessi, dont on sait qu’ils ont habité surtout la Mésie Infé-
rieure à partir de la fin du Ier siècle, ne peut renvoyer à cette province, car une
partie des Bessi a continué à habiter en Thrace. Pourtant, la liste de soldats
d’origine thrace et dace est immense. Même si, pour la plupart, les diplômes
ne mentionne pas le lieu d’origine du vétéran, on peut supposer de manière
raisonnable qu’il s’agit du territoire rural, car c’est là que sont attestés les indi-
gènes. Mon intention n’est pas de passer en revue toute la liste des soldats du
milieu rural de Mésie Inférieure qui sont attestés dans l’armée romaine, car
cela pourrait constituer le sujet d’un livre. Je commencerai mon enquête en
traitant des textes où il est question de soldats sûrement originaires du milieu
rural de Mésie Inférieure; je mentionnerai ensuite quelques sources qui me
semblent plus importantes du point de vue de l’information et je suivrai les
moments de recrutement dans la province, en cherchant à distinguer (dans la
mesure du possible) certains aspects qui ont déterminé les recrutements.

2. Les villages mentionnés comme sources de recrutement

Les villages sont mentionnés surtout dans les diplômes militaires comme lieu
d’origine des soldats qui reçoivent la honesta missio. Le territoire rural le
mieux documenté est, comme je l’ai déjà suggéré, Nicopolis ad Istrum. Les
textes datent du IIIe siècle, mais cela ne veut pas dire qu’il n’y a pas eu un
recrutement dans le milieu rural avant cette période. Je reviendrai plus tard
avec d’autres textes. Un premier soldat mentionné dans ce territoire est
M. Aurelius Bassus, fils de Derus, originaire de Nicopolis (ex Moesia Infe-
riore), issu d’un uicus nommé Bres[---], qui a servi dans la flotte de Ravenne
et qui a été libéré en 221.2
Un deuxième texte, datant de 224, atteste M. Aurelius Victor, fils de Sporus,
ayant l’agnomen Drubius, qui a également servi dans la flotte de Ravenne.3
J’ai remarqué que le nom de Drubius, même s’il n’est pas attesté ailleurs, est
d’origine thrace: voir, par exemple, le uicus Dizerpera, duquel était originaire
Aurelius Victor4 La formule cui et, rencontrée aussi dans l’inscription men-
tionnant Aurelius Bassus (sans que, dans son cas, l’agnomen soit conservé),
apparaît à cette époque, comme le remarque P. Weiß.5 En fait, il est clair que
la formule était utile pour les bénéficiaires des diplômes lorsqu’ils rentraient

2
RMD IV, 317 = RMD V, 457 = RGZM 54; AE 2001, 2165.
3
RMD V, 463.
4
Mihailescu-Bîrliba et Răileanu 2014, 194–95.
5
Weiß 2000, 279–80.
 SOLDATS DU MILIEU RURAL DE LA MÉSIE INFÉRIEURE 21

chez eux. Les textes mentionnaient leur nom de citoyen mais aussi leur ancien
nom indigène, pouvant ainsi mieux servir comme pièces d’identité.6
Le troisième texte évoque un personnage que j’ai discuté ailleurs,
M. Aurelius Statianus, aussi nommé Apta, ancien marin de la même flotte.7
Il est originaire du village de Zinesdina Maior, du territoire de Nicopolis ad
Istrum, mais on le rencontre ensuite comme actor dans les environs de Novae.
On voit ainsi que, même s’il n’est pas rentré chez lui, il revient dans la pro-
vince et s’installe toujours à la campagne.
Le quatrième texte sur un ancien marin du territoire de Nicopolis a été
trouvé en Bétique.8 Je ne reprendrai pas la discussion concernant la restitution
d’un groupe de lettres ([---]tsitsi);9 il semble que ce mot désignait le village
d’origine de l’ancien marin, sesquiplicarius dans la flotte de Ravenne.
Comme je l’ai déjà dit, le territoire de Nicopolis ad Istrum a été une source
de recrues pour l’armée romaine. Même si dans plusieurs diplômes militaires
le territoire n’est pas explicitement mentionné, les noms indigènes indiquent
que c’était de là plutôt que de la ville même que provenaient les recrues. C’est
le cas des soldats recrutés de Nicopolis ad Istrum pour les cohortes préto-
riennes: L. Septimius Purula,10 C. Valerius Bassus,11 libérés en 208; L. Marius
Maximus, ayant reçu la honesta missio en 212,12 et T. Flavius Mucianus, libéré
en 225;13 C. Iulius Gaianus,14 M. Aurelius Marcus (qui rentre à la maison),15
ayant fini leur service en 226; M. Aurelius Secundus (qui rentre chez lui),
devenu vétéran en 228,16 et M. Aurelius Aulutralis, libéré en 231;17 P. Camu-
rius [---], qui a terminé son service en 233,18 et M. Aurelius Mucianus, vétéran
en 248.19 Il est remarquable que les textes attestent un recrutement du territoire
rural de cette cité à partir de Septime Sévère, lorsque la cité faisait partie de la
province de Mésie Inférieure.
Mais les recrutements dans cette province ont commencé encore plus tôt.
Les villages d’origine des vétérans sont mentionnés plus récemment dans les

6
Mihailescu-Bîrliba et Răileanu 2014, 195.
7
RMD IV, 311.
8
Eck et Fernández 1991; RMD II, 201.
9
Eck et Fernández 1991, 215–16; RMD III, 201, sub numero; Weiß 2000, 281–82.
10
RGZM 49.
11
RGZM 50.
12
RMD V, 455.
13
RMD IV, 310.
14
RGZM 54.
15
CIL XVI, 143.
16
RMD II, 132.
17
RGZM 61.
18
RGZM 62.
19
RMD V, 474.
22 L. MIHAILESCU-BÎRLIBA

inscriptions. À Cius, l’ancien stator du préfet de l’ala II Arauacorum, G. Vale-

rius Herculanus, est originaire d’un uicus nomme Rami[---].20 Il provenait sans
doute de la même région. À Callatis, l’ancien prétorien Aurelius Dalenis
a comme lieu d’origine le uicus Amlaidina,21 dont la localisation demeure
sujette à caution. Valerius Rufus, ancien soldat, fait ériger un monument funé-
raire pour lui-même et pour sa femme, laquelle porte un nom indigène (Zura-
turme); il est originaire du uicus Vorouum Minor.22 Un diplôme de 223 évoque
un anonyme, fils d’un certain Mucatralis, originaire d’un uicus Thiuri[---].23
Les inscriptions attestent par conséquent plusieurs uici d’où les militaires
ont été recrutés. On voit que les textes proviennent pour la plupart de Nicopo-
lis ad Istrum et datent du IIIe siècle, alors que l’inscription de Cius est du
IIe siècle. Pourtant, les inscriptions que je présenterai ci-dessous montreront
que les recrutements ont commencé beaucoup plus tôt.

3. Soldats provenant du milieu rural de la Mésie Inférieure? Probable-

ment oui

Un autre groupe de textes atteste des soldats portant des noms indigènes. Dans
beaucoup de cas, le lieu de provenance des diplômes n’est pas indiqué, pour
autant qu’ils nous soient parvenus par l’intermédiaire du marché noir. Dans
d’autres cas, même si on a la certitude que les documents proviennent de cette
province et bien que les bénéficiaires des diplômes portent des noms thraces,
leur lieu d’origine est passé sous silence. Théoriquement, ils peuvent être ori-
ginaires de Thrace, et pas forcément du milieu rural. Je pense pourtant que,
dans la plupart des cas (si l’on suit la statistique par provinces), les diplômes
proviennent des campagnes de Mésie Inférieure, ne serait-ce qu’à en juger
d’après les noms indigènes de leurs titulaires.
Les mentions les plus anciennes datent de l’époque des Julio-Claudiens. Un
texte évoque Sparticus, fils de Diuzenus, soldat dans la flotte de Misène, libéré
en 52.24 Le texte le désigne comme Bessus, mais comme je l’ai rappelé lus
haut, cela ne renvoie qu’à son origine thrace, sans que nous ayons la certitude
qu’il était né Mésie Inférieure. Romaesta est libéré en 54, après avoir servi
dans l’ala Gallorum et Thracum.25 On a déjà évoqué Tarsa, ancien

20
ISM V, 117.
21
ISM III, 237.
22
Conrad 2004, 260, no. 503.
23
RMD V, 462.
24
CIL XVI, 1.
25
ISM IV, 1.
 SOLDATS DU MILIEU RURAL DE LA MÉSIE INFÉRIEURE 23

tessserarius dans la flotte de Ravenne, libéré en 71,26 et Cotus, fils de Tharsa,

qui a reçu la honesta missio en 79, après avoir servi dans l’ala I Thracum vic-
trix.27 Il faut également mentionner le texte de 70, où il est question d’un
Thrace libéré de la legio I Adiutrix. Il s’agit d’un diplôme accordé par
Vespasien à Dules, fils de Datus.28 Cette légion a été constituée à partir de la
flotte impériale, comme expression de reconnaissance de Vespasien envers les
marins qui l’avaient soutenu pendant la guerre civile (voir Tacite Hist. 4, 68;
Dion Cassius 55, 24, 3). Le recrutement avait eu lieu en 44–45, au moment où
Claude commençait la campagne contre les Thraces ayant abouti à l’organisa-
tion de la province de Thrace (Tacite Ann. 12, 63). Un autre Thrace, Cersus,
fils de Denturasadus, est libéré par Vespasien en 71, après avoir servi dans
l’ala I Brittonum.29
Les recrutements ont continué sous les Julio-Claudiens, car il y a des
diplômes accordés par Titus: sauf celui pour Cotus, fils de Tharsa, il y a aussi,
la constitutio pour Gusula, fils de Doques en 79,30 pour Soius, fils de Muscel-
lus en 80,31 pour Durises, fils de Bithus en 8032 et pour un anonyme en 80/84.33
Les dernières années de Néron ont également représenté des moments impor-
tants pour les recrutements des Thraces dans les unités auxiliaires. Il s’agit des
soldats comme Turcus Doian[---] f. (libéré en 83),34 Bithus Seuthi f.,35 Gisu-
setes Heptasi,36 Bithus Soi[---]iae f.37 (libérés en 88), Mucaporus Eptacenti
f.,38 Thasis Casiporis f.39 (qui ont reçu la honesta missio en 90), Quelse Dolae
f.,40 [---] Genimoli f.41 (libérés en 91).
À partir des Flaviens, les recrutements des indigènes sont devenus, peu
à peu, une habitude. La plus grande quantité de diplômes militaires provient du
temps de Domitien, puis des Antonins. On peut seulement mentionner une
partie des soldats portant des noms thraces ou daces qui ont probablement été

26
Chiriac, Mihailescu-Bîrliba et Matei 2004.
27
RGZM 3.
28
CIL XVI, 10.
29
RMD V, 324.
30
Weiß 2004.
31
CIL XVI, 26.
32
CIL XVI, 158.
33
RMD V, 326.
34
RMD IV, 210.
35
CIL XVI, 35.
36
RMD V, 329.
37
RMD V, 330.
38
CIL XVI, 36.
39
RMD V, 333.
40
RMD I, 4.
41
RMD IV, 214.
24 L. MIHAILESCU-BÎRLIBA

recrutés dans le milieu rural de Mésie Inférieure (pendant les années de la

honesta missio): Dolens Sublusi f.42 (96), Lucius, qui a un fils nommé
Mucaseis43 (97), Meticus Solae f.44 (99), un anonyme avec la femme Dasia45
(99–110), Hebrenus Bithi f.46 (107), Sitralis Cultra[---] f. (109),47 un Ti.
Claudius [---] ayant un fils Dizala et l’autre Torcus48 (113–114), anonyme Bes-
sus d’origine49 (Hadrien), Aulusenus50 (118), [---]oli f., Bessus51 (122), [---]
Isesi f., Bessus52 (125), Ulpius Valens, Bessus53 (126). Bithus Solae f.54 (140),
Aulenus Her[---] f.55 (144), Thidi Thi[---] f.56 (144–146), [---] Bithi f. (157–
158),57 Heptaporus Isi f.58 (158), [---] Tarsae f.59 (159–160), Aulutralis
Rebocenthis60 (158), Deses Dasoni f.61 (159) un anonyme avec un fils nommé
Dolatus62 (147–160), Mucatralis Bithi f.63 (164) etc.
Les changements dans les unités prétoriennes et dans la flotte64 ont déter-
miné Septime Sévère à recruter encore plus de provinciaux, ce qui explique
entre autres pourquoi des individus originaires de Mésie Inférieure sont sou-
vent mentionnés dans les diplômes militaires. Certains d’entre eux ont été,
certes, recrutés sous Marc Aurèle, mais on retrouve beaucoup de soldats qui
ont commencé leur service militaire sous Septime Sévère et ses descendants.
Je ne dresserai pas une liste complète de ces personnages, mais je rappellerai
certains noms et l’année de leur libération, afin d’avoir une image sur les
recrutements dans cette province: Iulius Iulianus de Noviodunum65 (202),

42
RMD I, 6.
43
RMD V, 337.
44
CIL XVI, 45.
45
RMD IV, 221.
46
Eck et Pangerl 2009, 514–19.
47
RMD II, 84.
48
RMD IV, 225.
49
RMD I, 19.
50
RMD V, 348.
51
RMD V, 361.
52
RMD IV, 235.
53
RMD IV, 236.
54
RMD I, 39.
55
CIL XVI, 90.
56
RMD V, 402.
57
RMD V, 421.
58
CIL XVI, 108.
59
RMD V, 423.
60
Eck et Pangerl 2007.
61
RMD V, 424.
62
RMD II, 106.
63
RMD I, 163.
64
Voir Mihailescu-Bîrliba et Răileanu 2014, 203.
65
RMD V, 449.
 SOLDATS DU MILIEU RURAL DE LA MÉSIE INFÉRIEURE 25

L. Marius Maximus de Nicopolis ad Istrum66 (212), M. Aurelius Valens de

Marcianopolis67 (226), G. Valerius Gaius de Nicopolis ad Istrum68 (227),
M. Aurelius Bithus de Marcianopolis69 (230), Fl. Iulius Iulianus de Nicopolis
ad Istrum70 (232) etc. Il y a aussi des mentions de prétoriens à Rome, comme
Aurelius Mucco du uicus Pereprus,71 Flavius Proclianus, Aurelius Mucianus,
Claudius Valerianus, Valerius Maximus du uicus Agatapara appartenant pro-
bablement à la ciuitas Ausdecensium.72 On voit, par conséquent, qu’il y a des
citoyens provenant des cités (ou de leurs territoires ruraux) qui ne gardent plus
(sauf Bithus) les cognomina indigènes.
On remarque l’absence des diplômes au temps de Marc Aurèle. W. Eck
a interprété cette absence comme une conséquence de la peste: plusieurs auxi-
liares sont morts à cause de l’épidémie. Un laterculus de Viminacium de 195
constitue, dans l’opinion de W. Eck, une preuve qu’à l’époque des guerres
marcomanes on recrutait plus que la moyenne dans une légion: c’était aussi,
à son avis, un effet de la peste. Eck offre aussi une explication secondaire: la
fabrication des diplômes en bronze a cessé, à cause de l’utilisation du métal
pour les besoins de la guerre, ce qui signifie que les diplômes ont été confec-
tionnés de matériaux comme le bois ou la cire.73 Je commenterai désormais
cette situation dans le cadre de mon analyse sur les moments du recrutement.

4. Moments du recrutement

Les soldats mentionnés, bien que la liste soit incomplète, témoignent du grand
réservoir de recrues du milieu rural dans la province. Le monde romain connaît
d’ailleurs de nombreux exemples de soldats provenant du milieu rural provin-
cial. La plupart des documents datent de l’époque des Sévères. Mentionnons
entre autres M. Herennius Papaio, Isaurien, du uicus Callosus,74 C. Iulius
Montanus, fils de Bargadas, du uicus Araba en Syrie,75 M. Aurelius Bithus,
d’un village appartenant au territoire de Philippopolis, soit de la zone balka-
nique,76 M. Aurelius Valens, d’un établissement rural appartenant à la cité de

66
Weiß 2002, 505–12.
67
RMD V, 466.
68
RMD V, 467.
69
RMD V, 469.
70
RMD V, 471a.
71
CIL VI, 2736.
72
CIL VI, 2807.
73
Eck 2012, 66–71.
74
RMD II, 131.
75
RMD IV, 307.
76
RMD V, 459.
26 L. MIHAILESCU-BÎRLIBA

Cibalae en Pannonie Inférieure,77 T. Domitius Domitianus, du uicus Vindenia

du territoire de Claudiopolis en Cilicie,78 tous anciens marins; Antonius Pate-
rio, du uicus C[.]nisco appartenant au territoire de Ratiaria (Mésie Supé-
rieure),79 Aurelius Marcus, du vicus Statuis en Thrace,80 Aurelius Valerius,
d’un village inconnu,81 Aurelius Aurelianus, [---] Epicadus et Flavius Fuscus,
du uicus Perdica et du uicus Titis en Dardanie,82 Aurelius Verus, d’un village
nommé Budalia en Pannonie Inférieure,83 Aurelius Abitus, du uicus Magaris
près de Serdica (Thrace),84 prétoriens. Aux prétoriens, s’ajoute une liste des
soldats provenant de plusieurs uici du territoire de Philippopolis dans une ins-
cription votive de Rome.85 Les militaires originaires des campagnes, recrutés
dans les autres unités auxiliaires, sont, il est vrai, moins nombreux dans les
inscriptions, sans pour autant être absents: rappelons, par exemple, Silvanus,
fils de Tescoris, du uicus Cuetro, en Pannonie Inférieure.86
Pour revenir à la Mésie (à partir de 86, Mésie Inférieure), on distingue des
recrutements à chaque époque. Les diplômes des Julio-Claudiens sont plus
rares, mais on voit que les premiers recrutements ont commencé à l’époque de
Tibère, plus précisément après la révolte des Thraces du sud du Danube contre
les Romains. Tacite nous informe que Poppaeus Sabinus a réprimé une émeute
en Thrace ca. 25–26 (Tacite Ann. 4, 46–51); l’information a été reprise par
Suétone d’une manière plus laconique (Suétone Tib. 41). Les Thraces se
seraient soulevés, puisqu’ils ne voulaient plus combattre dans l’armée romaine
(Tacite Ann. 4, 46, 1–2). Cette révolte a été précédée par d’autres mouve-
ments, en 19 (sous Rhascuporis : Tacite Ann. 2, 64–67; Suétone Tib. 37) et
en 21 (Tacite Ann. 3, 38–39).87 Cela prouve que les recrutements, au moins
ceux au sud du Danube, avaient eu lieu encore à l’époque de Tibère, sinon
d’Auguste. Les recrutements des Thraces dans la flotte et dans les unités auxi-
liaires ont fait suite à ces émeutes, et de ce fait, ils sont en mesure de montrer
que la situation était devenue plus stable. Il convient aussi d’ajouter que
dans le diplôme accordé à Romaesta (54), les témoins sont des Bessi qui

77
RMD III, 194.
78
RMD II, 133.
79
CIL VI, 2730.
80
CIL VI, 2797.
81
CIL VI, 2818.
82
CIL VI, 2845.
83
CIL VI, 37213.
84
CIL X, 1754.
85
CIL VI, 2799.
86
AE 2002, 1183.
87
Voir surtout 3, 38, 3–4 sur leur refus de combattre dans l’armée romaine.
 SOLDATS DU MILIEU RURAL DE LA MÉSIE INFÉRIEURE 27

bénéficiaient du droit de cité après avoir servi dans la flotte impériale.88 Ces
recrutements ont continué sous Claude et Néron, surtout après la réorganisa-
tion de la province de Thrace en 45, et confirment les conclusions de D. Dana
et de F. Matei-Popescu quant au recrutement des Daces au sud du Danube.89
Le recrutement de Tarsa, libéré en 71, et dont le diplôme a été trouvé à Mihai
Bravu,90 prouve que le territoire nord-danubien a aussi commencé à fournir
des recrues vers la seconde moitié du règne de Claude et au début du règne de
Néron. Les années 55 et suivantes font preuve d’un recrutement des gens pro-
venant des campagnes de la province dans les unités auxiliaires terrestres;
c’est le moment, à mon avis, où ces personnes commencent à combattre de
plus en plus dans ces unités.91 Pourtant, la flotte reste encore un corps d’armée
privilégié pour le recrutement de la population indigène des villages de Mésie
(puis Mésie Inférieure).
Le milieu rural de Mésie Inférieure fournit, semble-t-il, un grand nombre de
recrues sous les Antonins, même si les recrutements avaient continué aussi
sous les Flaviens. La division de la Mésie semble être un moment où les recru-
tements pour les troupes de la province nouvellement créée de Mésie Infé-
rieure s’intensifient. Après cet épisode, les campagnes de Trajan et la guerre
d’Hadrien en Judée constituent encore des moments susceptibles d’avoir
entraîné le recrutement d’un nombre important de soldats dans les milieux
ruraux de Mésie. Ce n’est toutefois qu’à partir de Trajan que ces recrutements
sont devenus un processus habituel, qui s’est maintenu jusqu’à l’époque des
Sévères. Le temps de Marc Aurèle, plus précisément après la diffusion de la
peste en Europe et le début des guerres marcomanes (environ 167–180), sont
des périodes pour lesquelles l’émission de diplômes manque. Des recrutements
ont pourtant eu lieu sous Antonin le Pieux. J’ai déjà mentionné l’opinion de
W. Eck à propos de ce sujet: les effets de la peste et la fabrication des diplômes
en bois ou en cire, afin de remplacer le métal qui devait servir pour l’armée.92
La deuxième explication semble plus logique, du moins en ce qui concerne la
Mésie Inférieure, pour autant que dans cette province les traces d’une épidé-
mie ne sont pas visibles. La legio V Macedonica est déplacée en Dacie, pour
combattre contre les Sarmates: c’est un signe que la situation dans la province
était stable. D’ailleurs, les inscriptions du temps de Marc-Aurèle, milieu rural
compris,93 ne dévoilent pas les traces d’une épidémie. Il est vrai, que l’absence

88
CIL XVI, 3. Voir aussi Mihailescu-Bîrliba et Dumitrache 2012, 11.
89
Dana et Matei-Popescu 2009, 236.
90
Chiriac, Mihailescu-Bîrliba et Matei 2004.
91
Voir aussi Mihailescu-Bîrliba et Dumitrache 2012, 15.
92
Eck 2012, 66–67.
93
ISM I, 328, 330–332; ISM V, 63–64.
28 L. MIHAILESCU-BÎRLIBA

de la legio V Macedonica facilite l’invasion des Costoboces en 170,94 mais il

ne s’agit que d’une expédition à la recherche du butin qui avance jusqu’en
Grèce.95
L’époque des Sévères marque un autre moment de recrutements massifs
dans les provinces balkaniques et danubiennes. Les changements dans les uni-
tés prétoriennes ont favorisé ce phénomène. Les diplômes militaires sont, pour
la plupart, pour les soldats de Mésie Inférieure et de Thrace et, au-delà des
textes où les noms des uici sont mentionnés, il est à supposer, à on avis, que
les militaires, ayant servi soit dans les cohortes prétoriennes, soit dans la flotte,
provenaient majoritairement des villages. Les cohortes changent de composi-
tion, les provinciaux étant préférés aux Italiens, et cela en particulier pour des
raisons de fidélité. Les marins recrutés dans les villages de la Mésie Inférieure
m’ont déterminé, en travaillant avec ma collègue V.-M. Răileanu, à avancer
une hypothèse à propos du recrutement dans la flotte à partir de Septime
Sévère. Malgré la règle établie sous Vespasien, selon laquelle l’empereur
accordait la citoyenneté à ceux qui entraient dans la flotte, Septime Sévère
a recruté des pérégrins dans la flotte, ce qui correspondait à ses buts politiques
et militaires et à la nécessité d’augmenter les effectifs militaires. Ces soldats
auraient ensuite obtenu la citoyenneté, en 212 ou peu après, ce qui explique
pourquoi ils tous des Marci Aurelii. Sur les diplômes sont inscrits leurs anciens
noms de pérégrins, accompagnés par la formule cui et. La mention du village
et de leur nom de pérégrin s’explique par des raisons pratiques: ils rentraient
en fait chez eux, là où ils étaient plutôt connus sous leur ancien nom de
pérégrin.96

5. Conclusions

Les soldats recrutés dans le milieu rural de la Mésie Inférieure sont, on l’a vu,
très nombreux. Tous les corps d’armée sont représentés dans les textes: les
ailes et les cohortes, la flotte, les unités prétoriennes. Quant aux légions, il y
a certainement des soldats en provenance des campagnes. Les recrutements
ont existé tout au long de la présence des Romains aux bouches du Danube.
Sous Tibère, d’abord dans la flotte et dans des unités auxiliaires de Thraces, le
recrutement a eu lieu surtout au sud du Danube. À partir de Claude, les recru-
tements s’achèvent partout dans la province de Mésie, puis de Mésie Infé-
rieure, en devenant un phénomène habituel. L’intensité a varié selon les

94
ISM IV, 49–50.
95
Robertson-Brown 2011, 80–82.
96
Mihailescu-Bîrliba et Răileanu 2014, 202–03.
 SOLDATS DU MILIEU RURAL DE LA MÉSIE INFÉRIEURE 29

époques; on constate une ampleur sous Trajan et Hadrien (due aux expéditions
en Dacie, contre les Parthes et en Judée). Sous Marc Aurèle, je ne pense pas
qu’il y ait eu une chute: les sources sont plus silencieuses, car elles ne se sont
pas conservées. D’ailleurs, les témoignages épigraphiques sous Marc Aurèle
en Mésie Inférieure ne révèlent pas une situation de crise ou une épidémie.
Enfin, sous les Sévères, grâce à la nouvelle situation politique et militaire, les
recrutements dans la région sont massifs et la population indigène du milieu
rural de la province constitue un contingent nombreux dans les unités préto-
riennes et dans la flotte.
Une autre chose me semble très importante: ces recrutements du milieu
rural représentent une preuve que l’autorité romaine a été en contact direct et
étroit avec ce milieu. Le principal intermédiaire de ce contact a été l’armée. La
présence romaine et latinophone a contribué, en fait, aux recrutements massifs
de ce milieu; l’acquis du droit de cité à la fin du service, les avantages finan-
ciers ont constitué également un enjeu décisif pour les membres de la commu-
nauté rurale. Mais une fois encore, cette permanence des contacts des autorités
(par leurs représentants) avec la population de la campagne a été une prémisse
importante de la romanisation profonde de cette espace.

BIBLIOGRAPHIE

Chiriac, C., Mihailescu-Bîrliba, L. et Matei, I. 2004: ‘Ein neues Militärdiplom aus

Moesien’. ZPE 150, 265–69.
Conrad, C. 2004: Die Grabstelen aus Moesia Inferior. Untersuchungen zu Chronolo-
gie, Typologie und Ikonographie (Leipzig).
Dana, D. et Matei-Popescu, F. 2009: ‘Soldats d’origine dace dans les diplômes mili-
taires’. Chiron 39, 209–56.
Eck, W. 2012: ‘Die Seuche unter Mark Aurel: Ihre Auswirkungen auf das Heer’. Dans
Lo Cascio, E. (éd.), L’impatto della “peste Antonina” (Bari), 63–78.
Eck, W. et Fernández, F. 1991: ‘Ein Militärdiplomfragment aus der Betica’. ZPE 85,
209–16.
Eck, W. et Pangerl, A. 2007. ‘Eine Konstitution für die Hilfstruppen von Syria Palaes-
tina vom 6. Februar 158 n. Chr.’. ZPE 159, 283–90.
—. 2009: ‘Moesia und seine Truppen II. Neue Diplome für Moesia, Moesia inferior
und Moesia superior’. Chiron 39, 505–89.
Mihailescu-Bîrliba, L. et Dumitrache, I. 2012: ‘Les Thraces dans l’armée romaine
d’après les diplômes militaires. I. Les diplômes de Claude à Domitien’. Dialogues
d’Histoire Ancienne 38.2, 9–16.
Mihailescu-Bîrliba, L. et Răileanu, V.-M 2014: ‘Les territoires ruraux comme source
de recrutement pour la flotte. Le cas de Nicopolis ad Istrum (Mésie Inférieure) et
une nouvelle hypothèse à propos de la politique de recrutement sous Septime
Sévère’. Dialogues d’Histoire Ancienne 40.2, 193–205.
30 L. MIHAILESCU-BÎRLIBA

Robertson-Brown, A. 2011: ‘Banditry or Catastrophe? History, Archaeology and

Barbarian Raids in Roman Greece’. Dans Mathisen, R.W. et Shanzer, D. (éd.),
Romans, Barbarians, and the Transformation of the Roman World: Cultural
Interaction and the Creation of Identity in Late Antiquity (Farnham), 79–96.
Weiß, P. 2000: ‘Zu Vicusangaben und qui-et-Namen auf Flottendiplomen des 3. Jh.’.
ZPE 130, 279–85.
—. 2002: ‘Ausgewählte neue Militärdiplome. Seltene Provinzen (Africa, Mauretania
Caesarensis), späte Urkunden für die Praetorianer (Caracalla, Philippus)’. Chiron
32, 491–543.
—. 2004: ‘Zwei vollständige Konstitutionen für die Truppen in Noricum (8. Sept. 79)
und Pannonia inferior (27. Sept. 154)’. ZPE 146, 239–46.
 CHAPTER 4

TRADE ECONOMY IN RIVERINE PROVINCES:

A CLOSE-UP LOOK AT TRADERS’ NETWORKS
AND MOBILITY*

Rada VARGA and Annamária-Izabella PÁZSINT

Abstract
The current research focuses on the traders attested epigraphically in the Roman prov-
inces of Moesia, Dacia and Germania. By choosing the extremities of the European
Roman empire, we wanted to highlight the specificities, as well as the unity, of traders’
epigraphical habits and manifestations. Besides epigraphic patterns, we focused on
traders’ mobility and the overwhelming importance of personal networks in the devel-
opment of Roman medium- and long-distance trade.

The present paper deals with commercial workers from the geographical
extremities of Roman Europe, having in sight the provinces of Germania,
Moesia and Dacia. The fact that the spatial positioning of these provinces
makes them gatekeepers of riverine, maritime and terrestrial trade alike justi-
fies the choice of the area. Equally, with this research we situate ourselves on
the extremities of Roman Europe’s riverine limes, which makes it interesting
for highlighting trade and commercial patterns and models.
We are seeking to stress the specificities of traders for each area, to high-
light the differences and similarities of expression, but most of all we will be
focusing on the mobility of traders and their families, on their personal and
socio-professional networks, and how they influenced Roman provincial eco-
nomic life. Thus, we will deal with the raw data thematically, not geographi-
cally, following aspects such as the nature of the trade, the movement of fami-
lies and patterns of epigraphic expression.

*
This work was supported by a grant from the Romanian National Council for Scientific
Research (CNCS – UEFISCDI), project PN-III-P4-ID-PCE-2016-0271, within PNCDI III.
32 R. VARGA AND A.-I. PÁZSINT

State of the Art

Merchants (terrestrial and maritime) and associated workers in trade have

aroused interest because of their great mobility, (relative) wealth, heterogene-
ous cultural background and, implicitly, for the multitude of information their
inscriptions can deliver. As a curiosity, we will only stress the fact that the first
published monographs (of which we are aware) dedicated to Roman traders
(Johann August Ernesti)1 and ancient trade and navigation (Pierre-Daniel
Huet)2 date from the 18th century; for the eastern area, their efforts were con-
tinued by Vasile Pârvan in the early 20th century.3 Of course, these works are
scientifically obsolete, but they remain relevant in outlining the evolution of
research themes and interests. Coming to modern historiography, while Italy
and Rome are a focus of renewed interest, the provincial environment has
enjoyed less attention and fewer ample studies. Thus, Gabrielle Wesch-Klein’s4
article on private trade initiatives in North Africa, as well as Julie Vélissaro-
poulos’s work on the ναύκληροι,5 or Octavian Bounegru’s6 books on the mar-
itime merchants from the Black Sea come as rarities. The last important con-
tribution on traders is Wim Broekaert’s7 from 2013, which is a very valuable
attempt at prosopographical reconstruction. Although, in our opinion, he con-
structs his catalogue of sources on rather subjective criteria, it is a step forward
in the research concerning navicularii and negotiatores, as well as in the
advance of prosopographical studies for the Roman non-elites. Other works of
reference published in recent decades present various aspects of the general
trading phenomena: Peter Garnsey focuses on the complexity of trade, 8 build-
ing on the idea that it played a greater role in the ancient economy than claimed
by Moses Finley, Kerstin Droß-Krüpe’s9 volume centres upon the textile trade,
while Andrew Wilson and Miko Flohr deal among other matters with urban
craftsmen and traders.10
For Moesia Inferior, the works of Bounegru are representative; he has writ-
ten several studies on topics that include the merchants from Moesia11 (and

1
Ernesti 1737.
2
Huet 1763.
3
Pârvan 1909.
4
Wesch-Klein 1989.
5
Vélissaropoulos 1980.
6
Bounegru 2006; 2013.
7
Broekaert 2013.
8
Garnsey and Hopkins 1983.
9
Droß-Krüpe 2014.
10
Wilson and Flohr 2016.
11
Bounegru 1983; 2000; 2006, 317–26; 2010.
 TRADE ECONOMY IN RIVERINE PROVINCES 33

Thrace12), their organisation,13 the ships that were used,14 as well as issues
related to transport.15 Others, such as Lietta De Salvo16 and Lucrețiu Mihailescu-
Bîrliba,17 have focused on fluvial and maritime commerce and circulation in
the whole Danube area, bringing a further contribution to the topic.

General Overview

Among the Roman-era ‘independent’ professionals, the best represented

group18 is that of sales workers (Fig. 1); this was most probably due to their
relative wealth and to the open horizons implied by the nature of their occupa-
tion. In this category we include the self-proclaimed merchants (negotiator,
mercator, ἔμπορος, κάπηλος), the navigators and/or ship owners who might
or might not have been traders themselves (ναυκλήροι; their Latin equivalent,
the navicularii, are not present in the provinces within our focus), as well as
sellers of specific produces (seplasiarius, thurarius, coriarius, etc.).
At a provincial level (Fig. 2), sales workers are predominant in Germania
Inferior, Dacia and Moesia Inferior, but are almost non-existent in Moesia
Superior. But as Upper Moesia’s sample is the least numerically significant,
we cannot draw relevant or representative conclusions upon this matter – it is
probably only a question of hazard, not of true epigraphic representativeness.
In Germania Superior, the province with the largest sample of attested medical
personnel, while the number of sales workers is still high, they are surpassed
by medici and general health workers.19 Even if the samples are numerically
small, they are exhaustive and even if we can doubt them as a faithful reflec-
tion of Roman-era epigraphy, we cannot disregard the quantitative realities all
together.
In Moesia, the cities which provide evidence for professionals engaged in
trade20 are, not surprisingly, the ones which are part of the terrestrial,21 fluvial22
or maritime route networks.23 In the Roman trade economy, the provinces of

12
Bounegru 2012.
13
Bounegru 1995b; 2007; 2014.
14
Bounegru et Zahariade 1994; Bounegru 1995; 2008a; 2008b.
15
Bounegru 1997.
16
De Salvo 2010.
17
Mihailescu-Bîrliba 2015.
18
On the classification of Roman professions, see Varga 2017.
19
Varga 2018.
20
Axiopolis, Callatis, Histria, Novae, Odessos, Oescus, Tomis and Viminacium.
21
The Via Danubia, Via Moesica, with further connections to the Via Militaris, Via Egnatia,
Via Valeria, etc.
22
Istros, Hebros, Axios.
23
Tomis–Alexandria, Tomis–Nicomedia, etc.
34 R. VARGA AND A.-I. PÁZSINT

Fig. 1. Representativeness of independent professionals

in the Roman Latin provinces.
Professional, technical and related workers (trained in ‘hard’ sciences) 18;
Professional, technical and related workers (trained in arts and sports-related
fields) 35; Administrative and managerial workers 12; Clerical and related
workers 10; Sales workers 65; Service workers 20; Agricultural, animal husbandry
and forestry workers, fishermen and hunters 4; Processors of raw materials 13;
Craftsmen 50.
 TRADE ECONOMY IN RIVERINE PROVINCES

Fig. 2: Representativeness of independent professionals in Germania, Dacia and Moesia. Professional, technical and related
workers (trained in ‘hard’ sciences): 10 (MI), 1 (MS), 13 (GS), 7 (GI); Professional, technical and related workers (trained
in arts and sports-related fields): 22 (MI), 1 (D), 5 (GS), 1 (GI); Administrative and managerial workers: 3 (MI), 2 (MS), 4 (D);
Clerical and related workers: 1 (MI); Sales workers: 8 (MI), 6 (D), 8 (GS), 16 (GI); Service workers: 0; Agricultural, animal
husbandry and forestry workers, fishermen and hunters: 1 (D); Processors of raw materials: 2 (MI), 1 (MS), 1 (GS);
35

Craftsmen: 6 (MI), 3 (MS), 4 (D), 6 (GS), 2 (GI).

36 R. VARGA AND A.-I. PÁZSINT

Moesia Inferior and Superior were not only areas of local consumption, but
also of further transport of the imported goods, these coming mostly from Asia
Minor. This route is due to the fact that the western shore of the Black Sea
represents a point of contact with Asia Minor,24 favouring not only the trans-
port of goods, people, but also of ideas.25 Further on from this territory, the
terrestrial and fluvial routes mediate the advancement of goods towards the
rest of the empire.

Nature of the Trade

In all the provinces that concern us, there were land and water trade routes.
The geographical coordinates of the provinces, as well as the (needed)
resources and the sales markets are reflected by the general occupational titles:
negotiator/negotians, coriarius, ἔμπορος, ναύκληρος/nauclerus, nauta (espe-
cially in the West, the men designated as such might not have been the main
traders themselves),26 οἰνέμπορος, vestiarius. All these are rather general
terms which sometimes describe more complex realities, as in the case of the
thurarius, or the coriarius for example, who could have been the producers as
well as the sellers; we believe that when negotiator is present, it is clear that
the person was only the trader. On the same side, the normal separation
between production and long-distance trade in the Roman world has been long
noted.27
Sometimes, the nature of trade is not explicit in the text of the inscription,
but the adjoined relief offers clues; such is the case of Martialis,28 named
negotiator in the text and represented graphically along with grain(?) bags and
wine/oil caskets loaded on a boat. In Moesia Inferior, we have representations
of σύμβολον τῆς τέχνης, as in the case of the stele of Apollonios, son of
Epistratos,29 from Callatis. The funerary monument depicts two individuals
near a ship, one of whom wears a hat typical for fishermen. For the first editor
of the inscription,30 the two characters are the children of the deceased, who
are waiting in vain for their fisherman father to return. The ship depicted near
them represents, in her opinion, both the σύμβολον τῆς τέχνης of the father,

24
Bounegru 2017, 167.
25
Bounegru 2017, 166.
26
The case of Blassus Atusiri f. (CIL XIII, 7067), a nauta on whose gravestone we have the
representation of a trading ship, illustrates beautifully the naval, in this case river-based trade
practised in the area (Mihailescu-Bîrliba 2015).
27
Tchernia 2011, 23–26.
28
CIL XIII, 7068.
29
ISM III, 165.
30
Bordenache 1960, 497.
 TRADE ECONOMY IN RIVERINE PROVINCES 37

and the reason behind his death.31 Although this might be the case, we should
also take into consideration the position of Alexandru Avram,32 who sees in
the individual wearing a hat the deceased himself, next to his ship. We can see
in the individual a professional of the sea, who either traded goods, or pro-
vided the goods (fish). In another instance we might identify a possible nego-
tiator vinarius from the iconography of the limestone stele, which depicts two
barrels,33 while in the case of Aurelius Sozomenos from Byzantium34 we can
assume that he was a transporter who transported the goods from the farmland
to the city, as his monument depicts a loaded carriage pulled by oxen.
In other cases, the nature of the trade is more important to the person or more
relevant for the community, and thus registered explicitly in the text of the
inscription. Most frequent are pottery dealers (negotiatores (artis) cretariae)
and salt dealers (negotiatores salarii), as there was salt exploitation in both
Germanic provinces. Desideratus Curmillus is negotiator (artis) cretariae;35
Germania’s most important quarry was at Brohltal, near Cologne, whence the
monument originated. From the sanctuary of Nehalennia36 in Ganventa37 come
two fish sauce merchants (negotiatores alecarii),38 which is interesting both
because of their location and because of the act of devotion itself. Regardless
of the ‘ethnic’ deity, both men have Roman names: Catullinius and Gratus,
the former being identified as a civis Trever, a citizen of Augusta Treverorum.
Other specific trading occupational titles attested in the West are: negotiator
frumentarius, negotiator lanius, negotiator lignarius, negotiator nummularius,
negotiator pistoricius, negotiator seplasiarius, negotiator vinarius, negotiator
gladiarius, negotiator caudicarius, negotiator in ferro, negotiator artis sutoriae,
thurarius, manticularius, materiarius. As compared with the West, an addi-
tional term to indicate what type of merchandise the individuals traded appears
in only two cases from Moesia Inferior: an ἔμπορος specialised in wine trade39
and an ἔμπορος βυρσέων,40 specialised in selling leather goods.

31
Bordenache 1960, 497.
32
ISM III, P. 493.
33
ILB 320 = AE 1966, 350 = CIL III, 7442 = IGLN 100.
34
ISM II, 257.
35
CIL XIII, 8352.
36
A total of 13 traders are attested at the spot.
37
Stuart and Bogaers 2001; Cooley et al. 2007, 235–36; from the same sanctuary we have
dedications of merchants thanking for the safe arrival at destination of their cargoes.
38
In this context, we remark the lack of piscatores or any other fishmongers from the
Germanias.
39
ISM II, 463 = IGLR 28 = Bounegru 2000, 115 = Broekaert 2013, 264, no. 467.
40
ISM II, 320 = Bounegru 2000, 115 = Bounegru 2006, 64 = Broekaert 2013, 265, no. 469.
38 R. VARGA AND A.-I. PÁZSINT

Besides specific epigraphic attestations, and iconographic interpretations,

holding a particular local citizenship could also help determine the nature of
commerce, at least in the case of two merchants: Marcus Liberius Victor41 and
Tertinius Secundus,42 cives Nervii; the territorium of the Nervii, situated
between Schamble and Scheldt,43 is a very fertile area, therefore the possibility
for our men to trade in grains, much needed for the garrison in Nijmegen and
the civilians from the area is not to be excluded. At this point, we must men-
tion the special status of the trading branch which supplied grain for the army,
as private and state capital were intertwined, and the possible implication of
the praefectus annonae’s office in managing this problem.44
From all these trade-related professions, that of lixa appears only once, in
the case of the freedman Lucius Freius Faustus, son of Lucius, lixa legionis V
Macedonica.45 The overall number of individuals who are attested as lixa in
the Roman empire is small,46 and their profession is thought to have included
small traders, sutlers, hand workers, chefs, bakers, even various performers,47
all of whom lived in the vicinity of the military, in order to offer them their
products and/or services.
All these epigraphic examples validate the archaeological evidence, which
points to the fact that trade was being carried out both by land and water (sea,
river).

Mobility

The inscriptions concerning traders bring forth several aspects related to their
private and professional life, among which we may mention: a high degree of
geographical mobility, an increased exposure to natural and human dangers,
economic and personal uncertainty and risk, as well as a need for developing
professional and personal networks. Nonetheless, one of the most fascinating
and individualising traits of traders and merchants is the personal and trade
networks they built. In an age where communications were relatively low and
scarce, personal connections and endorsements were crucial for traders’ safety
and success.

41
CIL XIII, 8725.
42
CIL XIII, 8338.
43
Broekaert 2013, 78.
44
Mitthof 2001 for an overview.
45
AE 1990, 862 = AE 1996, 1336 = Ivanov 1990.
46
Ivanov 1990, 133–35.
47
Ivanov 1990, 133.
 TRADE ECONOMY IN RIVERINE PROVINCES 39

Directly connected to safety and success are the dedications from Dea Neha-
lenia’s sanctuary at Ganventa, which also provide us with interesting examples
of mobility. The first case we will discuss is that of Placidus Viduci f., nego-
tiator Britannicianus and cives Veliocassinius.48 He makes a dedication to the
goddess at Ganventa around AD 200, without mentioning anything else about
his trade or himself on the monument. If this were the only source on Placidus,
it would have still been highly interesting to see a man from Gallia Lugdunen-
sis, with a local, probably native, citizenship operating on the Germania Inferior–
Britannia trade route. Nonetheless, the trader is also attested on a second dedi-
cation from AD 221, from Eburacum.49 This is a construction inscription,
dedicated to the genius loci and to the numina of the emperors, for a vault and
a passageway erected by our merchant. While the citizenship and the quality
of trader on Britannic routes are expressed in a more or less similar fashion to
the previous inscription, the name of the dedicator is Lucius Viducius Placidus,
with the possible presence of the patronymic between nomen and cognomen.50
There are several possible interpretations for the name mismatching: one
implies that he became a citizen through the Constitutio Antoniniana or in
other circumstance between the erection of the two monuments and chose to
follow the Germanic tradition of assuming the father’s name as a cognomen.
The other hypothesis is that he had citizenship at the time of the Dea Neha-
lennia dedication as well, but being a more private act of devotion, did not feel
the need to write his whole, official name on the monument. The father–son
hypothesis, though alluring at first sight,51 was rightly discarded by Broekaert.52
Another matter raised by Placidus’ inscriptions is that of his real residence.
The euergetism of a stranger is a rare act in Britannia53 – and in the whole
western empire – and along with the indication of negotiator Britannicianus,
might suggest that the cives from Lugdunensis had moved, lived and worked
in Eburacum. As a ‘newcomer’, it was important to establish status, thus the
self-financed construction activity for the use of the community.
But Placidus is not the only one whose trade was oriented towards Britan-
nia. Caius Aurelius Verus, freedman of Caius and negotiator Britannicianus,
is attested on two inscriptions,54 dated in the first half of the 3rd century AD.

48
AE 1975, 651 = AE 1982, 724.
49
AE 1977, 512.
50
For commentaries, see Broekaert 2013, 220–22.
51
De Salvo 2006, 782.
52
Broekaert 2013, 221.
53
Bargfeldt 2014, 25–26.
54
AE 1983, 722 and AE 1893, 31 = AE 2004, 967.
40 R. VARGA AND A.-I. PÁZSINT

One of the monuments55 is a dedication to the same Dea Nehalennia, simple in

form and content, but the other gives more of a glimpse into our merchant’s
status and life. It is a statue for Apollo, dedicated in Cologne on a place given
through decurional decree. Most likely, this was his place of residence and the
connection to Britannia is, in this particular case, only a professional one;56
of course, we can also assume a birth origin and subsequent relocation to Ger-
mania Inferior or a simple act of euergetism meant to attract the benevolence
of important trading partners.57
Another negotiator Britannicianus known from the Dea Nehalennia dedica-
tions is Marcus Secundinius Silvanus,58 negotiator cretarius, who lived during
the first half of the 3rd century AD as well. He is one of the few persons erect-
ing more than one altar,59 indication of wealth and piety alike – on one of the
monuments he actually says ob merces rectes conservatas. As we have already
mentioned, the pottery dealers are quite frequent, some of them also known
from the Ganventa sanctuary. Given the contemporaneity of the inscriptions,
the Britannic connections, the similarities of trade (explicit in some cases), as
well as the presumably relatively high economic and social status of some of
the merchants attested here, we can easily imply and define a network – whose
complete ramifications remain, of course, unknown, but with visible nodes and
clear cut utility.
In the absence of personal identification papers and the possibility of real
control, we can assume that impersonal long-distance trade was virtually non-
existent.60 A model produced using ORBIS mapping61 implies that the trip
from colonia Ara Agrippinensium to Eburacum (Fig. 3) would have taken
13 days in June, using the fastest route (which might have not been always
possible) and covered 1186 km. The costs were between 3.39 and 3.79 denarii
per kilogramme of wheat – meaning roughly 350 denarii for the transport of
100 kg. The amount of money involved represents a serious investment, as the
annual stipendium of a legionary miles during the time of Caracalla’s reign
was around 225 denarii.62 This calculation is meant to stress the importance of
personal networks and connections as a safety measure for one’s investment.

55
The second referenced above.
56
Broekaert 2013, 42.
57
The same hypothesis could work in the case of Placidus as well, but, in both situations, we
believe this to be the least likely scenario.
58
CIL XIII, 8793 and AE 1973, 370.
59
Broekaert 2013, 97.
60
Terpstra 2013, 52.
61
The Stanford Geospatial Network Model of the Roman World: http://orbis.Stanford.edu.
62
Speidel 1992, 87–106; 2014, 53–62.
 TRADE ECONOMY IN RIVERINE PROVINCES 41

Fig. 3. The commercial route Ara Agrippinensium–Eburacum.

While in the above cases we are not necessarily talking about Britannic
origins, the only case of ex provincia Britannia comes from Mainz,63 where
Fufidius’ funerary stone was discovered. He was most probably involved in
trade with the legion,64 but there is a lacuna in the text right after negotiator,
so the nature of his trade remains unknown. What we know is that he was
a native of Britannia who got to live and trade in the Mogontiacum area.
Another example of long-distance mobility comes from Mogontiacum as
well – Tiberius Ulpius Iulianus65 is from Tium, ex provincia Ponto Bithinia.
The monument is dedicated by his fratres from an association and his freed-
man. Iulianus might have been a dealer in iron,66 but regardless his trade, we
must note that it led him to the west of the empire, into a province with differ-
ent inhabitants, cultures and languages. It is hard to tell if he lived here, or was
just trading occasionally – we would go for the latter hypothesis, as the absence
of any family members on the monument, at the age of 62, might indicate. The
freedman, Chrysogonus, might have been a business partner as well, going
with his patron in common trading trips. In this case we see that the commer-
cial route from Asia Minor towards the western part of the empire was in use,

63
CIL XIII, 7300.
64
Broekaert 2013, 64.
65
CIL XIII, 6851.
66
If the inscription can be read as i(n) f(erro), though Broekaert (2013, 114) is prudent in this
regard.
42 R. VARGA AND A.-I. PÁZSINT

the traders being either intermediaries who settled in the area, or professionals
who were willing to engage in long-distance trade.
Regarding the geographical relocation of these professionals, in Dacia most
of the attested negotiatores67 are Suri – we thus have four out of seven men
denominating themselves as negotiator Surus. Though they mention their eth-
nic origin, only the cognomina of two of them bear some Greek-Oriental influ-
ence (Alexander68 and Apollophanes69 – definitely not Semitic, but Greek
names). Nonetheless, Aurelius Flavus belongs to one of the most interesting
personal traders’ networks we have attested. Besides the mentioned dedica-
tion, we also have the man’s funerary monument, coming from Salona70 and
dedicated by his amicus, Aurelius Aquila. About the latter we know that he
was also a trader and an ex-decurio of Potaissa.71 The great mobility of mer-
chants and traders72 comes as no surprise, but this case exemplifies how trad-
ing connections (and, here, personal relationships) were transmuted from one
place to the other. In the network illustrated by Fig. 4, we can see much more
than the direct connection, but the indirect existing connection between
Aurelius Alexander, the merchant dedicating alongside Flavus in Dacia and
Aurelius Lucianus from Salona.73
In Moesia Superior we are aware of the origin of the traders in only one
case, coming from a recently published inscription from Viminacium, which
names a negotiator, Aurelius Maxim[us] Surus.74 Even though there are other
Syrians attested at Viminacium,75 their profession remains unknown, as the
origo is not always a certain indicator of profession. A connection with
the characters attested in Dacia is tempting and – of course – possible, but the
evidence available at this point does not allow us to deduce more.
Several personal stories are recorded in Moesia Inferior, stories which stress
the shipper’s mobility and on the mechanisms through which their networks
were built. We begin with the family of Epiphania, whose father and
husband(s?) were ναύκληροι.76 From the inscription we find that her father

67
Those who might have come to the province due to trading activities (such as Titus Flavius
Ibliomarus from Augusta Treverorum), but where a profession is not explicitly mentioned, have
already been discussed (Matei-Popescu 2012, 86–87) and do not form the subject of this paper.
68
CIL III, 7761 = IDR III/5, 218.
69
CIL III, 7915 = IDR III/2, 203.
70
CIL III, 2006.
71
CIL III, 2086.
72
De Salvo 2006, 777.
73
Varga 2016.
74
Ferjančić et al. 2017.
75
IMS II, 211, 213, 169.
76
ISM II, 375.
 TRADE ECONOMY IN RIVERINE PROVINCES 43

Fig. 4. Aurelius Aquila’s and Aurelius Flavus’ network.

was from Hermione (Achaia), and took her with him on his sea journeys. The
name of her first husband, as well as his origin, is unknown to us, but we know
that he was a ναύκληρος like his father-in-law. Both of them died, and the
editor of the inscription believes that pirates probably killed them.77 However
this hypothesis cannot be proved, since there is no evidence for it, and since
Epiphania buried them. Epiphania, who was exposed to a ‘cosmopolitan’ envi-
ronment78 due to the profession of her father, remarried after the death of her
first husband with a certain Hermogenes, who describes himself as a citizen of
Ancyra and of Tomis. We may assume that he probably did business in the
vibrant city of Tomis, and was rich enough to do the city the right kind of
services in order to get himself rewarded with local citizenship and be enrolled
in the Oinopes tribe. What his exact profession was remains unknown to us,
but being himself a (prosperous) ναύκληρος is a possibility we cannot exclude
if we take into consideration the milieu in which Epiphania lived,79 but also
three telling cases, two from neighbouring Histria and one local.

77
Sauciuc-Săveanu 1964.
78
The text describes her as Ἐν μούσεσ Ἐφύην σοφίησ τε μετέσχον (I was born among the
muses, and I enjoyed wisdom’s goods), pointing to the fact that she was an educated woman.
79
Dana 2011, 255.
44 R. VARGA AND A.-I. PÁZSINT

The first testimony is that of the Nicomedian Asklepiades, son of Meno-

philos, who before settling at Histria was also granted the citizenship of Aiza-
noi (Phrygia) for unnamed reasons, but probably connected to his profession.80
To this we add the testimony of Nilos, son of such, from Tyras, who among
others ‘cautioned the ἔμποροι of Tomis who sailed towards Olbia, and inter-
vened for them in case of need’, and as a result was granted local citizenship
and other rights.81 This individual could have been a trader himself, since
Tyras is a point of connection between the western and the northern shores of
the Black Sea,82 and since the citizens of Tyras were granted fiscal advantages
by the Romans.83 Certainly, if being a ναύκληρος or an ἔμπορος did not
always equate with economic success, being the owner of the ship, who leased
it to ναύκληροι or who acted through intermediaries (slaves, freedmen), is
both a sign of high social status and financial power.84
Coming back to the three examples referred above, they point to the need of
ἔμποροι and ναύκληροι to build and rely on professional networks. These
types of professional connections certainly entwined with personal ones, as in
the case of Epiphania – it generally being difficult to tell which generated or
preceded which. On the one hand, family relations can constitute the base for
professional connections as in the case of Asklepiades, son of Menophilos’
family:85 he and his brother (Menophilos, son of Menophilos) were foreign
ἔμποροι coming from Nicomedia, and probably established at Histria for some
years, since the text mentions that Menophilos had also a son named after his
brother, Asklepiades. The common personal name of these two siblings made
Broekaert86 suggest that there might be a kinship connection between these
family members attested at Histria, and an individual named Herakleides, son
of Asklepiades87 from Tomis, who was also an ἔμπορος in roughly the same
period. Even though it is tempting to see such a connection, the fragmentary
state of the inscription from Tomis, and most importantly the common patro-
nymic makes it an unlikely hypothesis.

80
Robert 1978, 424.
81
ISM II, 5.
82
Dana 2011, 258.
83
Dana 2011, 259.
84
Pleket 1983, 136.
85
ISM I, 356 = SEG 49, 1009 = Bounegru 2000, 115 = Broekaert 2013, 263, no. 463. The
first editors believed that the deceased had two personal names, one Greek and one Roman,
Ba[ss]os, but no patronymic. However, Rigsby (1999, 175) rightly reads the text, and restores the
personal name of his father as being also Menophilos, which is much more logical, considering
that his brother has the same patronymic, and that no one from the family has a Roman name.
86
Broekaert 2013, 263, no. 462.
87
ISM II, 403 = Bounegru 2000, 115 = Broekaert 2013, 263, no. 462.
 TRADE ECONOMY IN RIVERINE PROVINCES 45

Having dual citizenship, as mentioned in the previous cases, represented

both recognition of the professional success of these individuals, and a boost-
ing of their commercial ventures due to the advantages granted.88 Cases of
local/dual citizenships are attested also in Germania, but here they are not an
indication of mobility per se, as we are dealing with cives from Augusta
Treverorum89 and the Nervii;90 given the short distance, we do not consider
this genuine mobility.
Coming back to the mobility of the traders, other scantier examples point to
the fact that the individuals who did business in Moesia Inferior come from
port cities of great commercial importance, both near (Byzantium, Nicomedia,
Prusias ad Hypium) and far (Alexandria, Carthage and Hermione). Through
the analysis of the individuals’ origo we can also supplement our sample of
traders: it is the case of Theon, son of Potamos, Alexandrinos,91 a possible
trader, who died at Callatis.92 As suggested by Alexandru Avram93 one possi-
ble scenario derived from this inscription would be that, even though in gen-
eral the goods which were traded by the Alexandrians were distributed through
the help of intermediaries, he might have accompanied his goods up to the
destination, where he passed away. In the Western provinces, we do not have
numerous explicit occurrences of origo; the only exceptions are the already
mentioned Fufidius from Britannia94 and Iulianus from Pontus.95
On the other hand, we have an attestation of a negotiator from Moesia Infe-
rior who settled in Rome: Aurelius Diza, negotianti [---] natus ex provincia
Moesia Inferiore regione Nicopolitane vico Saprisara (Aurelius Diza mer-
chant [---] born in the province of Moesia Inferior, in the region of Nicopolis,
vicus Saprisara).96 He established himself with his wife (and son?) at Rome,
where he continued to practise his profession, in undetermined conditions.
Aurelius Herculanus, the individual who dedicated the funerary monument to
the couple, was part of the civic guard in Rome, therefore, if he was their son
it could mean that they managed to adapt quite well to life in Rome. From

88
Dana 2012, 253, 266; 2011, 45–86; Güney 2014, 609.
89
AE 1973, 375, 362.
90
CIL XIII, 8725, 8338.
91
ISM III, 155.
92
The reason behind including him in this category lies in the fact that the few Alexandrians
attested in the area are known as traders, as for example [---]pios, son of Seppon, who was a wine
trader in the neighbouring city of Tomis, but in a different period. ISM II, 463 = IGLR 28 =
Bounegru 2000, 115 = Broekaert 2013, 264, no. 467.
93
ISM III, 487.
94
CIL XIII, 7300.
95
CIL XIII, 6851.
96
CIL VI, 2933 = Broekaert 2013, 37, no. 22.
46 R. VARGA AND A.-I. PÁZSINT

Fig. 5. The commercial route Aquileia–Sarmizegetusa.

Dacia, we have the relocated Aurelius Aquila and Aurelius Flavus. A complex
case comes from Aquileia,97 where M. Secundus Genialis, originating from
colonia Agrippinensis, identifies himself as a negotiator Dascicus; obviously,
the man, born in Germania Inferior and living in Italy, was involved in direct
commerce with the province of Dacia. As seen on Fig. 5, his trade route was
long and the transport costs were very high. Another very interesting case,
connected to northern Dacia, comes from Augusta Traiana in Thracia.98 Here,
Aurelius Sabinus, a priest of Zeus Dolichenus and a negotiator vinarius Dacis-
cus, dedicates along Aelius Primus, decurio of Porolissum. Most probably,
Sabinus imported wines from the region to Dacia and maybe even beyond, as
Porolissum was a customs point between Dacia and Barbaricum, and the Stara
Zagora region is rich in wine even today, producing over 90 types of wine.
If in the previous cases we have seen how traders and navigators relied on
their personal and familial networks for their business, we will now add a third
manner of ensuring the integration of these professionals to the commercial
environment and to the market: joining or forming private associations.
In Moesia Inferior specific associations are recorded at Tomis99 and
Axiopolis;100 we may assume that they were well organised, and helped trad-
ers and navigators both to make their way into the city, and in practical matters

97
CIL V, 1047.
98
IGB III/2, 1590.
99
ISM II, 60 = IGR I, 610 = AGRW 81; ISM II, 132 = CIL III, 7532 = ILS 4069 = CCET IV,
48; ISM II, 153 = IGR I, 604 = RICIS 618/1005 = SEG 47, 1040 = AGRW 82. In this latter case
what leads us to think that we might be dealing with an association of nαύκληροι is the origin of
the members (Alexandria); they probably either had a common profession associated with the
world of trade and commerce, or they had different professions, but from the same field.
100
CIL III, 7485 = ISM IV, 217: nautae universi Danuvii.
 TRADE ECONOMY IN RIVERINE PROVINCES 47

connected to their profession. What is distinctive about the associations at

Tomis is the nomenclature οἶκος which is recorded, besides at Tomis,101 only
at Amastris102 and Nicomedia.103 This common associative model is probably
the result of the professional network system developed between these cities,
as the connections with Nicomedia are strong. At Axiopolis, these are attested
the nautae universi Danuvii104 – in this case the civilian naval merchants/ship-
pers who could have been organised into a collegium, having Axiopolis (due
to its convenient position, on the Danube and close to the Black Sea) as head-
quarters. The denomination of this group does not ensure a clear vision on who
was part of it, as the members might have been the traders/the owners of the
ships from all over the Danube area105 (but in this case, one would expect more
epigraphic evidence), or they might be only locals.
In Moesia Superior the evidence comes from Viminacium, and it confirms
the fact that the Danube fluvial route was in use, as a shipper (nauclerus)106 and
a collegium nautarum,107 were attested. In Dacia, we do not have explicit trad-
ers’ associations attested, but one can imagine that attested professional groups
such as the uticularii,108 centonarii109 or the very inclusive fabri110 also had
a commercial component. Especially in the case of the uticularii, even if the
exact nature of their transport services is not completely clear, we can safely
assume that they generally specialised in transporting goods. The ‘ethnic’ col-
legia, such as those of the Asiani,111 Galatae112 or Ponto Bithyni113 surely pro-
moted and defended the economic interests of the members, as already dis-
cussed above. In the Inferior Rhine province, the collegia of the fabri
(tignariorum),114 peregrini living in Forum Hadriani115 and pistoriorum116 prob-
ably involved trading operated by their members. Unlike the cases stated above,
in Germania Superior, at Basilia, a collegium negotiatorum Cisalpinourum et

101
See n. 99.
102
Mendel 1901, 36, no. 184 = AGRW 94.
103
TAM IV.1, 22 = IGR III, 4; TAM IV.1, 33 = Robert and Robert 1974, 294–95, no. 572.
104
Pârvan 1909, 185–86.
105
Tocilescu 1884, 2, no. 4, compares these nautae with the nautae Averuci Rhodanici
from Gallia.
106
CIL III, 13804 = AE 1894, 104.
107
IMS II, 61 = AE 1905, 153 = AE 1907, 40 = AE 2012, 1249.
108
CIL III, 944, 1547.
109
CIL III, 1174.
110
CIL III, 975, 984, 1043, 1494, 1553, etc.
111
CIL III, 870.
112
CIL III, 860, 1394.
113
IDR III/5, 153.
114
CIL XII, 8344.
115
CIL XIII, 8808.
116
CIL XIII, 8255.
48 R. VARGA AND A.-I. PÁZSINT

Transalpinorum is attested.117 Collegia fabrum118 (tignariorum)119 and pere-

grinorum (in Marbach am Nektar).120
The cives Romani consistentes that grouped individuals who had a common
origin and a juridical status, are considered to be mostly active as traders.121
In Moesia Inferior, they are attested in a peregrine city like Callatis, near camps
(canabae/vicus), and in rural settlements.122 In all of these examples we are deal-
ing with inscriptions dedicated to the emperors (the guarantors of their rights)
and to the official divinities (Iupiter Optimus Maximus, Iunona Regina, Minerva,
Hercules Invictus), but their actual interests and roles in the local communities
are not stated, therefore we might be slightly reluctant in classifying them as
trade-oriented communities, except for the case from Callatis.123 Certainly, their
presence in a peregrine-city, as well as near the camps, and the involvement of
veterans could lead us to believe in possible commercial activity.

Epigraphic Habits

Regarding the nature of the merchants’ monuments, we have votive and funer-
ary stones in almost equal numbers. Besides the already mentioned construc-
tion dedication and statue base, a few other monuments are as well worth
noting. Thus, Caius Gentilius Victor,124 veteran from the XXII Primigenia –
and bearing a typical military name, rather common on the Rhenish limes –
gives through his testament payment for a statue for the emperor Commodus
and Fortuna Redux, in the interval AD 185–192.125 He deals in swords, some-
thing he most probably has knowledge about from his soldier years – negotia-
tor gladiarius. The statue is worth 8000 sesterces (around 2000 denarii),
which does not say much about the man’s wealth, except that it was above
average;126 if this was all he left, it is not extreme as far as purchasing power
goes. In comparison with the price paid for the statue, the contribution to the
restoration of the temple of Neptune, by the quinquennalis of a collegium nau-
tarum from Viminacium, seems to be rather superficial, as the sum paid was
2000 sesterces (500 denarii).127

117
CIL XIII, 5303; Kolb and Ott 1988.
118
CIL XIII, 7065.
119
CIL XIII, 7371.
120
CIL XIII, 6453.
121
Hasebroeck 1923, 40.
122
Avram 2007, 91.
123
ISM III, 83; Avram 2007, 92.
124
CIL XII, 6677.
125
Mihailescu-Bîrliba 2015, 187.
126
A slave or a horse cost roughly 2500 sesterces.
127
IMS II, 61 = AE 1905, 153 = AE 1907, 40.
 TRADE ECONOMY IN RIVERINE PROVINCES 49

Regarding titles and nomenclature, in the Western provinces there are not
many nuances or doubts. Nonetheless, an interesting case could be mentioned:
Lucius Septimius Fidelis,128 dedicating for a local form of the Matres in colo-
nia Claudia Ara Agrippinensium, names himself a negotiator commerciator
infectorius. Infectus refers to metals and means129 crude, unwrought, so our
character was a dealer in raw materials. As this is the only inscription where
the occupational title is redoubled, we cannot argue for an epigraphic habit,
but rather for a particularity, connected with a personal understanding of the
nature of the job, or with a specific linguistic/expression pattern.
For Moesia Inferior, epigraphy seems to indicate that the use of the terms
was not as strict as we might imagine: theoretically, the ἔμποροι (which are
more or less the Greek equivalent for negotiatores),130 are supposed to be
richer and holding a higher status than the κάπηλοι131 (which are the Greek
equivalent of mercatores),132 and they were involved in a rather large scale
trade.133 Even so, we have at least one inscription, belonging to an ἔμπορος,
which looks terribly poor and shabby.134 In reverse, we also have traders
attested whose businesses were probably of lesser amplitude, but who have
decent monuments and definitely identified themselves – with pride, we
assume – with their occupational status.135
To these particularities we add a linguistic peculiarity: the use of the agno-
men Basileus136 in naming two ναύκληροι137 at Tomis. The term was interpreted

128
AE 2010, 1005.
129
Oxford Latin Dictionary, ed. 1968, 895.
130
Bounegru 2006, 59.
131
A possible, but problematic attestation is that coming from Odessos: IGB V, 5386 = SEG
28, 604. The first editor of the inscription proposes the following restoration of the text:
[Κά]πηλοσ(?) Ἰνγένο[υα] [τὸ]ν Ἑρεμῆν (merchant Ingenua <to> Hermes). This would mean
that we would be dealing with a woman-merchant, being the only similar case that we know of.
We would like to thank Prof. Stephen Mitchell for suggesting to us the following interpretation
of the text: [Ὁ Δεῖνα Κά]πηλοσ Ἰνγένου[οσ] Τὸν Ἑρ{ε}Μῆν (such, merchant, freeborn, <to>
Hermes). In this case we would have an individual whose name is now lost, who was indeed a
merchant, and a freeman who erected a statue of Hermes. The iconography might represent For-
tuna and Hermes.
132
Bounegru 2006, 59.
133
An ἔμποροσ travels between two or more cities in order to distribute his goods, unlike a
κάπελοσ, which can carry out his activities within a single city (Plato Soph. 223d; Broekaert
2013, 257). The former category is not attested in our provinces.
134
ISM II, 463 = IGLR 28; ISM II, 463 - [---]pion son of Seppon, wine trader, with a pitiful
stone; marks: palm branch and a pentagram – it was assumed that it might be a Jewish, Pythago-
rean or gnostic symbol, but in truth we cannot know exactly what it stood for.
135
AE 1990, 862 – Lucius Treius Lucii libertus Faustus, lixa Legionis V Macedonica (Lucius
Treius Faustus, freedman of Lucius, lixa of the legio V Macedonica).
136
In Moesia Inferior it is used also as personal name/cognomen: ISM II, 22; IGLN 107 =
ILB 325 = IGB V, 5262 = SEG 47, 1153; IGB I2, 47.
137
ISM II, 186, 291.
50 R. VARGA AND A.-I. PÁZSINT

in similar and complementary ways: on the one hand, it is thought to be related

to the profession of the individuals;138 on the other, it was seen as a marker of
social status,139 pointing to the success of the individuals in their business
ventures.
In general, in the funerary inscriptions dedicated for individuals attested in
the Lower Danubian provinces, we frequently encounter the age of the
deceased,140 and this epigraphic habit is also reflected in the professional group
of interest to us. Out of the 14 funerary monuments within our focus, ten
record the age of the deceased. Age rounding is practised, but we also encoun-
ter a limited number of cases (two) where the precise age is given in years,
months and days. An interesting correlation of the age with the iconography of
the monument might come from the funerary stele of [---]n son of Stratokles,
an ἔμπορος, attested at Tomis, but originating from Prusias ad Hypium.141 The
iconography of the marble stele depicts the deceased with a volumen in the left
hand, stance which is used to represent literate individuals.142 In this case the
volumen might point to the fact that the individual was known for his book-
keeping activities.143 Ages are less frequent in Latin-language provinces, as it
is visible in Table 1. Without coming as a general rule, most of the epitaphs
indicating ages of death were erected by family members, rather than associ-
ates or peers.

Table 1: Attested age groups of merchants.

Moesia Moesia Germania

Age group Dacia
Inferior Superior Superior
21–40 yrs. 4 0 4 0
41–60 yrs. 4 1 0 2
70–90 yrs. 1 0 1 0

138
Bărbulescu and Buzoianu 2009, 395. Bounegru 1983 gives the term the meaning of owner
of the ship, and captain.
139
Curcă 2008, 283.
140
In Dacia, we have 604 ages registered and in Moesia Superior – 473. The data comes from
the Romans1by1 database and was processed in Varga et al. 2018. If that presents the general
coordinates related to the topic, and provides much-needed statistics, other articles, such as Piftor
2007–08, Mihailescu-Bîrliba and Curcă 2010, and Mihailescu-Bîrliba and Piftor 2011 tackle
issues such as life expectancy, age structure, size and composition of the population, etc.
141
ISM II, 248 = SEG 30, 845 = SEG 39, 670 = Slușanschi 1988, 307, no. III = Bounegru
2000, 115 = Broekaert 2013, 265, no. 468.
142
Dana 2007, 207–09.
143
We would like to thank Stephen Mitchell for this interpretation.
 TRADE ECONOMY IN RIVERINE PROVINCES 51

A Dangerous Profession

Piracy and brigandage was a reality for the professionals from this domain.
Travel implied of course the dangers of the sea/routes, dangers which could
have been connected not only to natural disasters (i.e. shipwrecks due to harsh
weather, or inappropriate travel conditions), but also to the human factor.144
Even though the inscriptions within our focus do not record the dangers that
traders faced due to their profession, several attest the danger of being killed
by latrones, as is the case with Flavius Capito,145 and Valerius Marcus146 from
Moesia Superior. If we take into account an inscription from Callatis,147 which
mentions the legislation on maritime responsibility, we can infer the existence
of sea-related dangers and the legislative pillars meant to deal with their out-
come. Precisely after a dangerous sea voyage (periculo maris) an individual
named Aurelius Statianus,148 an actor from Novae, managed to survive; in
consequence he made a dedication to Deus Aeternus, and restored a temple.
Another monument,149 in honour of the Domus Divina and the Casses,150
puts forth an interesting detail explaining the reason of the dedication: quod
post summersam. Having survived a shipwreck must have felt like divine pro-
tection for anyone and thus Lucius Licinius Divixus’ dedication is fully under-
standable. From an economic point of view, we do not know whether the
wrecked ship was his own property, or if he relied on third-party transport
services for his trade.151
Another poorly known set of data concerns legal matters and responsibili-
ties connected to trade; in this context, two inscriptions from Callatis are
worth mentioning due to the fact that there is reference to the institutional and
legislative pillars connected to the world of trade and commerce. The first
inscription is a decree152 which records a maritime affair, introduced to the
court153 by the εἰσαγωγεῖς. Due to the fragmentary state of the inscription we
cannot tell much on its content but the text mentions the ‘merchandise’ of the
ἔμποροι, fines, and most importantly it attests the existence of maritime

144
Avram 1997.
145
AE 1934, 209: qui casu viminacium dasmini a latronibus atrocissimam mortem perpessus
est (he was atrociously killed by brigands while travelling from Viminacium to Dasminium).
146
AE 1901, 19 = CIL III, 14587 = ILS 8504 = IMS III, 293: a latronibus interfectus (killed
by brigands).
147
ISM III, 36.
148
IGLN 8 = AE 1989, 635.
149
AE 1969–70, 436.
150
Mihailescu-Bîrliba 2015, 192.
151
Broekaert 2013, 78.
152
ISM III, 39; SEG 45, 908; Sauciuc-Săveanu 1927–32, 422–24; Avram 1995, 17–19.
153
The existence of maritime tribunals is also attested in ISM III, 38 = SEG 45, 905.
52 R. VARGA AND A.-I. PÁZSINT

tribunals. Furthermore, another inscription,154 from the Hellenistic period, pro-

vides us with a glimpse on the existing legislation on maritime responsibility:
it emerges that the maritime responsibility of the transporters was limited in
cases of force majeure (natural and human perils), cases in which the respon-
sibility belonged to the creditors, and not to the transporters.

Conclusions

The available epigraphic data confirms that the geographical location of the
provinces, as well as the geographical characteristics of the environment, led
to the development of a trade economy in the area from the earliest times.
With the integration of the areas in the Roman empire, transport and com-
merce became easier.
While generally Greek epigraphy is richer in personal details, benefitting
from a tradition of relatively sophisticated expression, we can observe that
Latin inscriptions put up by traders bring forth generally the same quantity and
quality of data. The inscriptions mentioning traders are usually rich in infor-
mation, allowing identification by both ancient and modern readers, providing
proof of their status and position. The first thing revealed by traders’ monu-
ments is the importance of their profession within communities: the general
over-representation of traders, the figurative depictions and/or the detailed
explanation of the nature of the trade all stand for this. This is rather normal in
the context of a profession more perilous than the average (travelling, regard-
less the means, was risky; travelling with goods held, of course, an even higher
degree of risk) and which brought relative prosperity. A second fact is the
traders’ need to insert themselves in the communities with which they devel-
oped business. This was done through acts of euergetism and through estab-
lishing relationships with the local elites. Coming to the last important point,
which modelled merchants’ ways of expression, we have the personal relation-
ships that built on family, ethnicity and religion in order to create networks
and webs of trust. The act of building personal networks in this domain is
visible especially in the examples coming from Moesia Inferior and Dacia.
While subjected to local particularities and epigraphic habits, the most strik-
ing feature revealed by this comparative study is that merchants and traders
manifest more common features than differences imprinted by geographical
area. At the two ends of the Danubian limes, we find commerce workers iden-
tifying with their profession, trying to establish a respectable place in society,
and striving to consolidate relationships in the same ways.

154
ISM III, 36.
 TRADE ECONOMY IN RIVERINE PROVINCES 53

BIBLIOGRAPHY

Avram, A. 1995: ‘Kallatiana’. Studii și cercetări de istorie veche și arheologie 46,

17–34.
—. 1997: ‘Date cu privire la pirateria de pe coasta de vest a Mării Negre’. In Cihó, M.
(ed.), Timpul Istoriei. In Honorem Ligiae Bârzu (Bucharest), 114–22.
—. 2007: ‘Les Cives Romani Consistentes de Scythie Mineure. État de la question’.
In Compatangelor-Soussignan, R. and Schwentzel, C.-G. (eds.), Étrangers dans
la cité romaine (Actes du colloque de Valenciennes, 14–15 octobre 2005, Habiter
une autre patrie, des incolae de la République aux peuples fédérés du Bas-Empire)
(Rennes), 91–109.
Bărbulescu, M. and Buzoianu, L. 2009: ‘Inscriptions inédites et révisées de la collec-
tion du Musée d’Histoire Nationale et d’Archéologie de Constantza. I’. Pontica
42, 389–407.
Bargfeld, N. 2014: ‘Newly Invented Tradition. The Individual and the Community at
the Northern Frontier of the Roman Empire’. In Fejfer, J., Moltesen, M. and
Rathje, A. (eds.), Tradition: Transmission of Culture in the Ancient World (Acta
Hyperborea 14) (Copenhagen), 17–41.
Bordenache, G. 1960: ‘Attività edilizia a Tomi nel II secolo dell’e.n.’. Dacia n.s. 4,
255–72.
Bounegru, O. 1983: ‘Considerații asupra negustorilor din Dobrogea secolelor I–III
e.n.’. Studii Clasice 21, 59–65.
—. 1995a: ‘Représentations de naves actuariae sur la côte ouest de la Mer Noire’.
Studia Antiqua et Archaeologica 2, 147–60.
—. 1995b: ‘Observations sur l’organisation et l’activité des associations de naviga-
teurs de Dacie et de Mésie’. Studia Antiqua et Archaeologica 2, 161–70.
—. 1997: Considérations concernant le transport des marchandises et le ravitaillement
de l’armée sur le limes du Bas-Danube. In Groenman-Van Waateringe, W. et al.
(eds.), Roman Frontier Studies 1995 (Proceedings of the XVIth International
Congress of Roman Frontier Studies) (Oxbow Monograph 91) (Oxford), 311–16.
—. 2000: ‘Der westliche Pontosraum und seine Handelsbeziehungen in der Römischen
Kaiserzeit’. Studia Antiqua et Archaeologica 7, 121–39.
—. 2006: Trafiquants et navigateurs sur le Bas Danube et dans le Pont Gauche
à l’époque romaine (Philippika 9) (Wiesbaden).
—. 2007: ‘Oikos Ton Naukleron. The Shipowners Organisation in the Pontic and
Aegean Area’. In Mayer Olivé, M., Baratta, G. and Guzmán Almagro, A. (eds.),
XII Congressus Internationalis Epigraphiae Graecae et Latinae: provinciae
Imperii Romani inscriptionibus descriptae (Barcelona), 191–97.
—. 2008a: ‘Naves Actuariae. Seeschiffe für den Amphorentransport in römischer
Zeit? Eine ikonographische und historische Untersuchung’. Peuce n.s. 6,
277–82.
—. 2008b: Comerț și navigatori la Pontul Stîng și Dunărea de Jos (sec. 1 –3 p. Chr.)
(Iași).
—. 2010: ‘Armateurs et marchands de Nicomédie dans la Méditerranée à l’époque
romaine’. Classica et Christiana 5.2, 287–98.
—. 2012: ‘Marchands Thraces dans les sources épigraphiques des provinces romai-
nes de Mésie et Thrace’. In Boteva-Boyanova, D., Mihailescu-Bîrliba, L. and
54 R. VARGA AND A.-I. PÁZSINT

Bounegru, O. (eds.), Pax Romana: Kulturaustausch und Wirtschaftsbeziehungen

in den Donauprovinzen des Römischen Kaiserreichs (Akten der Tagung in Varna
und Tulcea, 1.–7. September 2008) (Antiquitas 1) (Kaiserlautern), 53–56.
—. 2013: Mercator: Studien zur antiken Wirtschaft im Pontosgebiet und in der Ägäis
(Antiquitas 2) (Kaiserslautern).
—. 2014: ‘Emporoi et nauclères dans les ports de la Méditerranée orientale. Structures
institutionnelles et images publiques des commerçants et des armateurs’. In
Eck, W. and Funke, P. (eds.), Öffentlichkeit–Monument–Text (XIV Congressus
Internationalis Epigraphiae Graecae et Latinae, 27–31 Augusti MMXII – Akten)
(Corpus Inscriptionum Latinarum. Auctarium, n.s. 4) (Berlin/Boston), 433–35.
—. 2017: ‘Le Pont Gauche et le commerce de Rome: traditions hellénistiques et
modèles commerciaux Romains’. In Mihailescu-Bîrliba, L. (ed.), Migration,
Kolonisierung, Akkulturation im Balkanraum und im Osten des Mittelmeerraumes
(Konstanz), 161–69.
Bounegru, O. and Zahariade, M. 1994. ‘Roman Ships on the Lower Danube (1st–
6th Centuries). Types and Functions’. In Westerdahl, C. (ed.), Crossroads in
Ancient Shipbuilding (Proceedings of the Sixth International Symposium on Boat
and Archaeology, Roskilde, 2–5 September 1991) (Oxbow Monograph 40)
(Oxford), 35–41.
Broekaert, W. 2013: Navicularii et Negotiantes: A Prosopographical Study of Roman
Merchant and Shippers (Pharos 28) (Rahden).
Cooley, A., Mitchell, S. and Salway, B. 2007: ‘Roman Inscriptions 2001–2005’.
Journal of Roman Studies 97, 176–263.
Curcă, R. 2008: ‘Agnomia în inscripțiile grecești de la Tomis’. Tyragetia 2.1,
281–86.
Dana, M. 2007: ‘Education et culture à Istros. Nouvelles considérations’. Dacia n.s.
51, 185–209.
—. 2011: Culture et mobilité dans le Pont-Euxin: Approche régionale de la vie
culturelle des cités grecques (Scripta Antiqua 37) (Bordeaux).
—. 2012: ‘Pontiques et étrangers dans les cités de la Mer Noire: le rôle des citoyen-
netés multiples dans l’essor d’une culture régionale’. In Heller, A. and Pont, A.-V.
(eds.), Patrie d’origine et patries électives: Les citoyennetés multiples dans le
monde grec d’époque romaine (Actes du colloque international de Tours,
6–7 Novembre 2009) (Scripta Antiqua 40) (Bordeaux), 249–66.
De Salvo, L. 2006: ‘Mobilità di mercanti nell’Occidente Romano’. In Akerraz, A. et
al. (eds.), L’Africa Romana: Mobilità delle persone e dei popoli, dinamiche
migratorie, emigrazioni ed immigrazioni nelle province occidentali dell’Impero
Romano (Atti del 16 Convegno di Studio, Rabat, 15–19 dicembre 2004) (Collana
del Dipartimento di storia dell’Università degli studi di Sassari n.s. 31) (Rome),
773–90.
—. 2010: ‘Circolazione e commercio per via d’acqua nelle province danubiane’.
In Zerbini, L. (ed.), Roma e le province del Danubio (Atti del convegno inter-
nazionale, Ferrara, Cento, 15–17 ottobre 2009) (Pubblicazioni del LAD 1)
(Soveria Mannelli), 79–94.
Droß-Krüpe, K. (ed.) 2014: Textile Trade and Distribution in Antiquity (Philippika 73)
(Wiesbaden).
Ernesti, J.A. 1737: De Negotiatoribus Romanis Diputatiuncula (Leipzig).
 TRADE ECONOMY IN RIVERINE PROVINCES 55

Ferjančić, S., Korać, M. and Ricl, M. 2017: ‘New Greek and Latin Inscriptions from
Viminacium’. ZPE 203, 235–49.
Garnsey, P. and Hopkins, K. (eds.) 1983: Trade in the Ancient Economy (London).
Güney, H. 2014: ‘The Economic Activities of Roman Nicomedia and Connectivity
between the Propontic and the Pontic World’. In Cojocaru, V., Coșkun, A. and
Dana, M. (eds.), Interconnectivity in the Mediterranean and Pontic World during
the Hellenistic and Roman Periods (Cluj-Napoca), 605–24.
Hasebroeck, J. 1923: ‘Die Betriebsformen des griechischen Handels im IV. Jh.’.
Hermes 58, 393–425.
Huet, P.-D. 1763: Histoire du commerce et de la navigation des Anciens (Lyons).
Ivanov, R. 1990: ‘Lixa Legionis V Macedonicae Aus Oescus (Moesia Inferior)’. ZPE
80, 131–36.
Kolb, A. and Ott, J. 1988: ‘Ein “Collegium negotiatorum Cisalpinorum et Transal-
pinorum” in Augusta Rauricorum’. ZPE 73, 107–10.
Matei-Popescu, F. 2012: ‘The Origins of the Tradesmen in Dacia’. In Boteva-
Boyanova, D., Mihailescu-Bîrliba, L. and Bounegru, O. (eds.), Pax Romana:
Kulturaustausch und Wirtschaftsbeziehungen in den Donauprovinzen des
Römischen Kaiserreichs (Akten der Tagung in Varna und Tulcea, 1.–7.
September 2008) (Antiquitas 1) (Kaiserlautern), 85–98.
Mendel, G. 1901: ‘Inscriptions de Bithynie II’. Bulletin de Correspondance Hellé-
nique 25, 5–92.
Mihailescu-Bîrliba, L. 2015: ‘Marchands et trafiquants en Germanie Supérieure.
Origine et raisons de leur activité’. In Mihailescu-Bîrliba, L. (ed.), Colonisation
and Romanization in Moesia Inferior. Premises of a Contrastive Approach
(Antiquitas 3) (Kaiserlautern), 185–98.
Mihailescu-Bîrliba, L. and Curcă, R. 2010: ‘New Perspectives on the Demography of
the Roman Province of Moesia Inferior’. The Mankind Quarterly 51.2, 139–53.
Mihailescu-Bîrliba, L. and Piftor, V. 2011: ‘L’espérance de vie, la structure d’âge et la
mortalité en Mésie Inférieure et en Scythie Mineure. Un regard comparatif’.
In Nikolov, V., Bacvarov, K. and Popov, H. (eds.), Interdiziplinäre Forschung
zum Kulturerbe auf dem Balkanhalbinsel (Sofia), 419–51.
Mitthof, F. 2011: Annona militaris: Die Heeresversorgung im spätantiken Ägypten.
Ein Beitrag zur Verwaltungs- und Heeresgeschichte des Römischen Reiches im
3. bis 6. Jh. n. Chr. (Papyrologica Florentina 32) (Florence).
Pârvan, V. 1909: Die Nationalität der Kaufleute im Römischen Kaiserreiche (Breslau)
(apud N. Zugravu [ed.], Scrieri de istorie romană [Iași 2008], 107–269).
Piftor, V. 2007–08: ‘L’espérance de vie et la structure d’âge de la population féminine
en Mésie Inférieure (Ier–IIIe siècles apr. J.-Chr.)’. Studia Antiqua et Archaeo-
logica 13–14, 46–57.
Pleket, H.W. 1983: ‘Urban Elites and Business in the Greek Part of the Roman
Empire’. In Garnsey and Hopkins 1983, 131–44.
Rigsby, K.J. 1999: ‘Two Danubian Epitaphs’. ZPE 126, 175–76.
Robert, J. and Robert, L. 1974: ‘Bulletin Épigraphique’. Revue des Études Grecques
87.441, 186–340.
Robert, L. 1978: ‘Documents d’Asie Mineure’. Bulletin de Correspondance Hellé-
nique 102.1, 395–543.
Sauciuc-Săveanu, T. 1927–32: ‘Callatis’. Dacia 3–4, 411–82.
56 R. VARGA AND A.-I. PÁZSINT

—. 1964: ‘Pe marginea unei inscripții funerare din Tomis’. Studii și Cercetări de Isto-
rie Veche 15.1, 137–38.
Slușanschi, D. 1988–89: ‘Tomitana Graeca’. Pontica 21–22, 305–11.
Speidel, M.A. 1992: ‘Roman Army Pay Scales’. Journal of Roman Studies 82, 87–106.
—. 2014: ‘Roman Army Pay Scales Revisited. Responses and Answers’. In Reddé, M.
(ed.), De l’or pour les braves! Soldes, armée et circulation monétaire dans le
monde romain (Actes de la table ronde… 12–13 septembre 2013) (Scripta
Antiqua 69) (Bordeaux), 53–62.
Stuart, P. and Bogaers, J.E. 2001: Nehalennia: Römische Steindenkmäler aus der
Oosterschelde bei Colijnsplaat (Rijksmuseum van Oudheden te Leiden 11;
Corpus signorum imperii romani. Nederland 2) (Leiden).
Tchernia, A. 2011: Les Romains et le commerce (Centre Jean Bérard, Études 8)
(Naples).
Terpstra, T.T. 2013: Trading Communities in the Roman World: A Micro-Economic
and Institutional Perspective (Columbia Studies in the Classical Tradition 37)
(Boston/Leiden).
Tocilescu, G. 1884: ‘Neue Inschriften aus der Dobrudscha und Rumänien’. Archäolo-
gisch-Epigraphische Mitteilungen aus Österreich-Ungarn 8, 1–34.
Varga, R. 2016: ‘Aurelius Aquila, Negotiator ex Provincia Dacia. A Prosopographic
Reconstruction’. In Ardevan, R. and Beu-Dachin, E. (eds.), Mensa Rotunda
Epigraphica Napocensis (Cluj-Napoca), 27–34.
—. 2017: ‘The Professionals of the Latin West. Encoding the Occupational Titles’.
In Cupcea, G. and Varga, R. (eds.), Social Interactions and Status Markers in the
Roman World (Archaeopress Roman Archaeology 37) (Oxford), 9–21.
—. 2018: ‘Epigraphic Manifestations of the Medical Personnel. A Comparative
Approach on the Epigraphy of “Military” and “Civilian” Health Workers’.
In Proceedings of the 23rd International Limes Congress (Mainz), 1106–07.
Varga, R., Pázsint, A., Boda, I. and Deac, D. 2018: ‘Romans1by1. Overview of
a Research Project’. Digital Classics Online 4.2, 12–31.
Vélissaropoulos, J. 1980: Les nauclères Grecs: Recherches sur les institutions mari-
times en Grèce et dans l’Orient hellénisé (Hautes études du monde gréco-romain
9) (Geneva/Paris).
Wesch-Klein, G. 1989: ‘Private Handelsförderung im römischen Nordafrika’. Mun-
stersche Beitrage zür Antiken Handelsgeschichte 7.1, 29–38.
Wilson, A. and Flohr, M. 2016: Urban Craftsmen and Traders in the Roman World
(Oxford).
 CHAPTER 5

FACILITIES AND MEDICAL STAFF

OF THE LOWER DANUBIAN ROMAN ARMY

Dan APARASCHIVEI

Abstract
The rigorous organisation of the Roman army during the Principate brought forward
the issue of caring for the health of its soldiers. The state’s intention to treat wounded
soldiers as soon as possible and have them in optimal physical and mental condition
led to the construction of military hospitals within castra, as well as the training of
specialist medical staff. The objective here is to identify the period during which this
system was implemented on the Lower Danubian limes, as well as the medical facili-
ties available to the armies of the provinces of Moesia Superior, Moesia Inferior and
Dacia. Besides the valetudinarium of Novae, we bring arguments for the existence of
other such constructions in legionary or auxiliary castra in the region, as well as a pos-
sible connection with the general strategy of the army in its deployment in critical
areas of the empire. The density of military units garrisoned on the Lower Danubian
limes allows us to outline the interesting diversity of the physicians who served both
the legions and the auxiliary and naval units. Based on the epigraphic sources, physi-
cians with different appellations were identified, some of these being unique within the
empire.

The Roman army is the first army in the ancient world in which the treatment
of illnesses and war injuries was carried out by well-trained and well-organised
personnel, with established roles and hierarchy. In this regard, we possess rel-
evant information from both contemporary authors and archaeological
research.1 The development of this service is based on several measures taken
for the first time by Caesar. He granted the right of citizenship to practitioners
of the medical arts.2 Octavian Augustus was the one who, aware of the

1
See particularly targeted studies in Scarborough 1968; Davies 1969a; Nutton 1969; 1970;
Wilmanns 1995; Salazar 2000; Baker 2004; Aparaschivei 2012a; Israelowich 2015, 88–109.
2
Suetonius Jul. 42: Omnes professos, et liberalium artium doctores, quo libentius et ipsi
urbem incolerent, et coeteri appeterent, civitate donavit. There are opinions, however, that the
turning point in the evolution of Roman medicine was the transformation of Greece into a Roman
58 D. APARASCHIVEI

importance of the effective healthcare of military personnel, set up perma-

nently a body of medical professionals/doctors within military units. Later on,
from Vespasian and Hadrian, the privileges of practitioners of the liberal arts
multiplied.3 Positive repercussions also affected the military physicians. More-
over, the acknowledgment of the importance of medical services by civilian
decision-makers is shown by the remuneration of the specialists and the build-
ing of military hospitals from public funds. This is also confirmed by the rela-
tively large number of medici who practised mainly in the border regions dur-
ing the Principate.
It is obvious, judging from the sources, that physicians treated both violent
traumas and illnesses which also occurred in peacetime.4 They decided whether
certain soldiers were to be discharged on medical reasons, to be treated in
hospitals, or to be allowed to continue their convalescence in the barracks.5
Since their emergence as a separate corps, physicians were attached to various
military units. They were immunes, which exempted them from the daily
duties of ordinary soldiers.6 The epigraphic sources mostly mention the appel-
lative medicus, which is complemented by designations, mainly confirming
the affiliation with a military unit (medicus alae, medicus cohortis, medicus
legionis, medicus duplicarius etc.), his specialty (medicus chirurgus), as well
as medicus castrensis or castrorum, medicus ordinarius or miles medicus.7
However, the absolutely original initiative with very beneficial effects in
the treatment of the soldiers was the construction of military hospitals,

province in 146 BC and, implicitly, the easier access thence to Greek culture and civilisation
(Salazar 2000, 77).
3
Suetonius Aug. 59; Cassius Dio 53. 30; André 1987, 86–87. For the decrees granted by
Vespasian in the year AD 74, see comments at Oliver 1989, 121–23 and no. 38. The doctors
approved by the Roman authorities were largely of Greek origin, and their training was based on
principles highly developed during the Hellenistic period.
4
For example, one of the Vindolanda tablets presents the report made by one auxiliary unit,
cohors Tungrorum, in AD 90. It says that out of the 296 troops, 265 were in good condition,
while 31 were in need of medical attention. Out of these, 15 suffered from various illnesses
(aegri), six were wounded (volnerati), but not necessarily in combat, while ten suffered from eye
diseases (lippientes) (Bowman and Thomas 1994, no. 154; see also Allason-Jones 1999, 136).
5
According to the regulations of the time, the doctors were responsible both for keeping the
soldiers active and for discharging them, depending on their state of health (Codex Iustinianus
12. 36. 6). For example, a whole unit in Vindobona was disbanded because of disease. See also
in Baker 2004, 41. Apparently, Trajan even issued a rescriptum according to which those who
were physically unfit could be honourably discharged (Arrius Menander Dig. 49. 16. 4. Details
in Israelowich 2015, 90).
6
Tarruntenus Paternus Dig. 50. 6. 7: Optio valetudinarii, medici, capsarii, et artifices et qui
fossam faciunt, veterinarii, specularii, fabri, sagittarii, aerarii et qui aegris praesto sunt omnes
inter immunes habentur.
7
For details about each of these categories of doctors, as well as for their presence through-
out the empire, with bibliography, see Aparaschivei 2012b, 88–89 and Bader 2014, table 1.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 59

valetudinaria. Particularly in the hot regions of the empire, commonly on the

frontier, ancient sources and archaeological research confirm that such well-
equipped hospitals were built, which would provide for wounded, as well as
sick soldiers from the legionary or auxiliary castra.8 There is no doubt that
the employment of doctors and construction of established military hospitals
had a pragmatic purpose: to return to active duty as many soldiers as possible,
healed both physically and mentally. The presence of the ‘specialist’ divini-
ties, Asclepius and Hygieia, in the valetudinaria, also had a consistent role in
increasing the soldiers’ confidence and implicitly in enhancing physical form.
The emergence of these healthcare facilities did not only enable physical
healing, but also gave a significant psychological boost to the soldiers. This is
why the emperors agreed and encouraged the construction of these edifices,
which included implicitly temples or altars dedicated to the health-protecting
gods. We are not just talking about establishments where doctors with scien-
tific studies were active, but also where ‘doctor’ Asclepius played an impor-
tant role.9
It is certain that the refinement of the military healthcare system was a pro-
cess that developed over time, for the entire Roman army. However, we have
good reason to believe that for some sensitive areas of the empire, as well as
in certain historical contexts, special measures for the provision of medical
services were being implemented, within the overall strategy of the emperor
and the state.
In the general context outlined above, my objective is to study the medical
facilities as well as the specialist staff in the provinces of Moesia Superior,
Moesia Inferior and Dacia. Of course, this approach is limited by the informa-
tion available from current sources, while other research based on new discov-
eries can alter certain conclusions.

Valetudinaria in Moesia Superior, Moesia Inferior and Dacia

Once the province of Moesia was established, probably in the early years of
the Christian era, the need for a better organisation of the army in this region
was acknowledged by the authorities. Following Domitian’s administrative

8
Hyginus Gromaticus Mun. Castr. 4. Archaeologically there are about 30 such edifices
attested on the Rhenish-Danubian limes, with a particularly large number in Britannia (Dycek
2004, 159; Baker 2004, 90, who express also reservations about the attribution of such structures
in the castra). The necessity of such facilities in the army is argued for even by contemporary
sources, which remark that, often, the number of soldiers unavailable due to illness equalled that
of the soldiers killed in battles (Herodian 3. 9. 6; cf. Jackson 1996, 2229).
9
Dyczek 1999; Künzl 2005, 55–61.
60 D. APARASCHIVEI

reorganisation, the two provinces, Moesia Superior and Moesia Inferior,

became the staging base for the confrontation with the Dacians. Therefore, the
deployment of additional troops and the thorough organisation of the limes
were high priorities in the imperial strategy as a whole. Trajan continued such
preparations in the two Moesias prior to the decisive confrontation with Dece-
balus, so that in AD 101 the vast majority of the troops staged south of the
Danube were involved in this war.10 The need to use at maximum capacity
the available human resources led the Roman authorities to the thorough
organisation of the army, which included its medical corps. Consequently, the
foundation of medical hospitals in parallel with field medical facilities, organ-
ised in tents,11 as well as the thorough organisation of the body of profession-
als who serviced the hospitals, were the most important measures to be taken.
For our area of interest, the most relevant example for assessing the impor-
tance the Imperial authority attached to the care of wounded soldiers is the
valetudinarium of Novae, in Moesia Inferior. Here is located one of the best
preserved army hospitals in the Roman world.12 It was uncovered within the
praetentura of the castrum of the legio I Italica in the course of archaeological
excavations carried out by a Polish-Bulgarian team in the 1970s. This spec-
tacular building was erected in the reign of during Trajan, by soldiers of the
legio I Italica, with the assistance of other troops detached in the area, during
the war with the Dacians.13 Interestingly, the land on which the hospital was
built was occupied before that by the Flavian thermae, a compound also in
connection with hygiene and healthcare, which was one of the necessary meas-
ures recommended by ancient authors to be implemented inside castra.14 The
81.90 × 72.90 m rectangular structure, with an interior court, consists of two
rows of rooms, separated by a central corridor.15 The importance of hygiene

10
Three legiones of Moesia Inferior, along with five alae and 11 cohortes, participated in the
Dacian War (Matei-Popescu 2010, 267). In Moesia Superior, in AD 100, besides the legions
there are documented three alae and 21 cohortes (Matei-Popescu and Ţentea 2006, 76).
11
Caesar De Bello Gallico 6. 38. The oldest valetudinarium attested, in Haltern, follows
a ground plan that seems to imitate the layout of a cluster of tents, tentoria (Nutton 1969, 26; cf.
von Schnurbein 1974).
12
Press 1985; 1987; 1990; Dyczek 1997, 43–44; 2000; 2004.
13
A large number of tiles bearing the stamp of the Italic legion have been uncovered in the
hospital area, and there is also sufficient evidence for the active presence of detachments from
the legio XI Claudia Pia Fidelis (Matei-Popescu 2010, 133) and the legio I Minervia Pia Fidelis
(Dycek 2002, 686).
14
The location of the military camps, the maintenance of hygiene and a set of general rules
for communal living were among the measures of prevention. Ancient authors recommended
certain sites as adequate for setting up the camp. Vegetius Mil. 1. 22: loci salubritas eligatur.
15
About the dimensions and plans of these structures in the empire, see details in Baker 2004,
97–100.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 61

and the organised spacing of the sick-hall, elements found in all such construc-
tions throughout the empire, were confirmed by the archaeological discoveries.16
The valetudinarium of Novae functioned until the time of Caracalla.17 Most
likely, this hospital served military units in the conflict zone and was the place
of convalescence for Roman soldiers wounded in the battle theatre. It is there-
fore obvious that the building of such an edifice within this Danubian fortress,
probably prior to the start of the Dacian War, was an action part of the general
strategy established by Trajan. Supporting this opinion is the well-known
information that one of Trajan’s advisers during the Dacian War was Titus
Statilius Crito, the head of emperor’s personal medical staff and imperial
procurator.18 From such a position he undoubtedly had foreseen the conse-
quences of the conflict; therefore the construction of military hospitals in
certain strategic points was part of the opening strategy of this confronta-
tion. The historian-physician Crito is often mentioned in relation to the
Dacian expeditiones, mainly in connection with his lost historical treatise
known as Getica.19 However, one should note that Statilius Crito was a physi-
cian and author of some important pharmacology treatises.20 From his position
of member of the General Staff he had, though, significant influence on the
emperor’s decisions. In fact, the influence of imperial physicians over their
masters is well known, for as early as Antonius Musa, the physician of
Augustus.21 Certainly the military strategy of the emperors encompassed the

16
Dycek 2004, 160–62. As an example, an Egyptian papyrus reveals significant informa-
tion regarding the purchase of blankets of white soft wool for such a valetudinarium (Baker
2004, 86).
17
However, it was concluded, based on numismatic finds, that between the reigns of Elaga-
balus and Gordian III the hospital might have been used only partially (Dyczek 2003; 2005,
231–32).
18
Nutton 1977, 221, no. 40. Details in Scarborough 1985, 387–89; Sammama 2003, 365,
no. 246, nn. 5–6; Barbara 2014, 35.
19
The work in which he narrates the war between Romans and Dacians, which is very likely
to be a major source for Trajan in writing his De bello dacico, also inspired later authors who
gave some information about this conflict. See Scarborough 1985, 388–92; cf. Russu 1972.
20
Crito of Heracleea is known and very appreciated for his work Kosmetika, but we also
know parts of a work on the composition of drugs, Pharmakitides bibloi: see Galen De Compo-
sitione medicamentorum secundum locos 1. 1, K. 12, p. 401 and 1. 3, K. X12, p. 446; De anti-
dotis 1, 17, K. 14, p. 103; cf. Scarborough 1985, 394–96.
21
The importance of the personal physician of the Imperial family in the making of military
and political decision has been best proven by the career of Galen, perhaps the most famous
physician of antiquity, who became the close adviser of emperor Marcus Aurelius. For example,
in De libris propriis, 18–19 Galen asserted that Asclepius advised him to not accompany the
emperor against the Germans. Details on the influence of the court physicians on Roman emper-
ors, but also about the honours received by such professionals in the Imperial court (see Scarbor-
ough 1985, 389–90; Barbara 2014). Some other personal physicians of Roman emperors are well
known from the sources: Tiberius’ (Caricles) – Tacitus Ann. (6. 50), Nero’s (Andromachus of
62 D. APARASCHIVEI

healthcare component, and counselling on this area could best be given by the
chief physician of the imperial house. So Crito’s acknowledged roles as coun-
sellor, friend of the emperor and historian of the Dacian War, overshadowed
his occupation as physician. Therefore, it is normal to think that his influence
on the development of a healthcare system for the Roman army in the prov-
inces of the Lower Danube, at the end of the 1st century and the beginning of
the 2nd century, was quite significant.
Regarding Moesia Superior, we have only indirect evidence for the exist-
ence of military hospitals in the legionary fortress of this province.
Some Serbian scholars are convinced, however, of the existence of a valetu-
dinarium in Singidunum, the headquarters of the legio IV Flavia. This hypoth-
esis is supported by the discovery of an inscription of a praefectus castro-
rum.22 Vegetius writes that the praefectus castrorum was responsible for the
proper running of the physicians’ work, of the provision of medical supplies
and equipment as well as for the care of the patients, as an overseer.23 How-
ever, there is no proven connection between the presence of this praefectus
and the existence of a hospital within the Singidunum fortress. It is attested
though the existence of qualified medical personnel within the legio IV Flavia,
which was stationed in this Moesic centre for quite a long time.24 It confirms
that an inscription mentioning a medicus legionis of this military unit, discov-
ered in Aquincum, where units of the legion were temporarily detached in the
second part of the 2nd century AD.25 This important settlement of Pannonia
Inferior is also the find-spot of a votive monument of a veteran of the legio IV
Flavia, ex optione valetudinarii.26 The monument was raised to honour Teles-
phorus, the son of Asclepius, a divinity whose cult starts in the Danube area in

Crete) – Galen De theriaca ad Pisonem 14, Claudius’ (Quintus Stertinius Xenophon of Kos, fol-
lowed by Gaius Stertinius Xenophon and Scribonius Largus) – Tacitus Ann. 12. 61 and 67.
22
This is a certain Nassius Dexter, praefectus castrorum (Mirković 2000, 7–10, pls. 1–2;
Krunić 2004, 48). Another praefectus castrorum of the Singidunum legion fortress is mentioned
in IMS I, 16.
23
Vegetius Mil. 2. 10: ʻsick soldiers, the doctors who tended them and payments belonged to
his (praefect of the camp) duties’ – translation of Milner 2001. However, the actual management
of the hospital was the responsibility of an optio valetudinarii (see details in Aparaschivei 2012b,
85, n. 239).
24
The legio IV Flavia Felix was garrisoned in Singidunum by Domitian and returned there in
AD 117, after taking an active part in the expeditiones Dacicae followed by a short stationing
in Dacia (Opreanu 2006). Mócsy (1974, 86) believes that is possible that the legion could have
been first stationed in Viminacium or even in Ratiaria. See Aparaschivei 2010, 62 for discussion,
literature and hypotheses.
25
CIL III, 3537; Wilmanns 1995, 216, no. 64; Zimonyi 2014, appendix, no. 10.
26
Wilmanns 1995, 217, no. 65.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 63

the mid-2nd century AD.27 Thus, even if the legio IV Flavia was sometime in
the 2nd century stationed also in Aquincum,28 it is not impossible that Titus
Flavius Priscus, the donor, worked as optio valetudinarii in Singidunum.
The location of a military hospital in Singidunum, possibly erected during
Trajan’s reign, seems logical, and must have been part of a larger project,
towards preparing the Dacian expeditiones. As evidence of a significant medi-
cal activity in the castrum of Singidunum and its area of influence, there is the
large number of archaeological finds of medical instruments.29
In the same logic must be regarded the castrum of Viminacium. It was here
that the legio VII Claudia was stationed, as early as the second half of the
1st century AD.30 Temporarily, the legio IV Flavia Felix was also stationed
here, around the beginning of Trajan’s Dacian War.31 Viminacium gained
a crucial role precisely during and around the time of the war with Dacians,
hence the existence of a military hospital in this strategic point seems highly
appropriate.32
Exceptional information regarding the situation of military hospitals on the
level of the empire originates precisely in Moesia Superior. An inscription,
mentioning a valetudinarium of the cohors II Aurelia nova milliaria equitata
in the fortress of Stojnik,33 confirms that this type of construction was a requi-
site not only for legionary castra but for the auxiliary ones as well.34 It is also

27
For an overview of this cult, see Deonna 1955; while for its spread in the area of Danube
provinces, see Aparaschivei 2012b, 162–79; Antal 2014; Varga 2016.
28
Such a valetudinarium is already documented in Aquincum, mainly through the two
optiones valetudinarii identified here: Titus Flavius Priscus, a veteran of the legio IV Flavia and
Titus Venusius Aper, whose service with a certain legion is quite doubtful (as a hypothesis one
may include the legio II Adiutrix, while IV Flavia or even X Gemina are not to be excluded as
options) (Wilmanns 1995, 214, no. 62).
29
Krunić 1992, 10; 2004, 48.
30
In AD 69 the legion is already mentioned in Moesia from where it is transferred to Italy to
take part in the Civil War, although we have no certainty that Viminacium was its first garrison
in the province (Tacitus Hist. 2. 85; cf. Aparaschivei 2010, 61, n. 38).
31
At least five funeral stelae of members of this legion have been uncovered in Viminacium
(IMS II, 86–89, 93).
32
Spasić-Đurić 2002, 63. However, despite numerous finds connected to a medical activity
carried out in this centre (five tombs of physicians, complete medical kits or isolated parts, oph-
thalmologists’ cases, etc.), including in a military environment, the remains of the military hos-
pital could not yet be identified archaeologically.
33
This fortified settlement, located in an important mining area of Moesia Superior, gains
also special military significance, at least since 2nd century AD (IMS I, 112). The auxiliary mili-
tary unit documented as stationed here watched over the Kosmaj mountain mining area (cf.
Wagner 1938, 91–92; Ţentea and Matei-Popescu 2002–03, 281).
34
CIL III, 14537 = ILS 9174: Valetu/dinarium / coh(ortis) II Aur(eliae) / nov(ae) (miliariae)
equit(atae) / c(ivium) R(omanorum) T(itus) Bebeni/us Iustus prae(fectus) Imp(eratore)
C [[ommodo]] / II et Vero II co(n)s(ulibus).
64 D. APARASCHIVEI

true that within the empire, apart from the one just mentioned, there are only
a few other examples, many of these unconfirmed and even contested by
scholars.35
One should note that this valetudinarium of Stojnik is dated in the reign of
Commodus, in a period when he was still engaged in the final stage of the
Marcomannic Wars, and that the region was a sensitive one.36 One can assume,
again, an immediate military interest as the reason for erecting this building
dedicated to the care of soldiers, part of a complex regional strategy.37
This locational variability of the military hospitals, as well as the meticu-
lousness with which they were built, the adaptation to environmental condi-
tions and compliance with hygiene and location rules demonstrate the serious-
ness with which the Romans regarded medical care, the military included.
Despite the fact that the reality of such structures is challenged, sometimes
with good arguments,38 we believe that these structures were perceived as
a necessity in the context of the development of a war strategy that included
all the elements, including the healthcare of soldiers.
Regarding the territory of Dacia, the data concerning the medical facilities
is unfortunately very scarce, both archaeological and other types. This is quite
intriguing, given the high level of army presence in the region even after the
founding of the province. Several structures were proposed through the years

35
There is another piece of epigraphic evidence, found in Beroea (Aleppo), in Syria, on the
Cappadocian limes: AE 1987, 952 – Imp(eratori) Neruae Traiano Caes(ari) | Aug(usto) Ger-
manico Dacico | per L(ucium) Fabi[um] Iustum \ leg(atum) [Aug]usli [p]r(o) pr(aetore) \
[ualetu]dinar(ium) a novo fact(um) / [-- / --] coh(ortis) IIII Luce[nsium]. A papyrus fragment
from Egypt mentions also a valetudinarium in Cappadocia (Baker 2004, 86). The word valetudi-
narium also appears on a tablet from the fort of Vindolanda, in connection to a fabrica: Bowman
and Thomas 1994, 155. One should include here, in the same context, the inscriptions mentioning
the optiones valetudinarii, of Africa, Italia, Gallia Lugdunensis, Germania Inferior and Pannonia
Inferior (CIL VIII, 2553, 2563; CIL VI, 175, 31145, AE 1973, 53; CIL IX, 1617; CIL XIII,
8011; AE 1937, 181; Wilmanns 1995, 217, no. 65). Add the archaeological evidence in Cor-
bridge, Housesteads, Wallsend Hodd Hill and Pen Llystyn, Fendoch (Britannia), Valkenburg
(Germania Inferior), Wiesbaden (Germania Superior) Oberstimm and Künzing (Raetia); cf. Press
1994, 95–96; Allason-Jones 1999; Baker 2002, 71. The debate on the authenticity of these hos-
pital buildings is still intense. The hypothetical plans of construction consist either of a corridor
or an interior courtyard around which the treatment rooms would be arranged. Some researchers,
however, think that these plans would correspond rather to fabricae or other buildings that were
built inside fortresses (see Baker 2004, 95–101 for discussions).
36
For the final stages of the Marcomannic Wars, including the beginning of peace negotia-
tions, see Dietz 1994, 10; 2000, 138–39. However, Commodus already started the reorganisation
of the defensive system of this part of the empire (details in Kovács 2009, 261).
37
In this area a physician tomb was also uncovered, which grave-goods include a medicine
case (Krunić 2001, 127–29, fig. 9).
38
Baker 2002. For the argument against, see Künzl 2005, 55–61.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 65

as possible army hospitals. One such is in the Ilişua castrum, the garrison of
ala I Tungrorum Frontoniana.39 Here, though, the building with the proposed
medical purpose does not seem to possess the basic layout required of a hos-
pital.40 Then, in the case of Buciumi and Porolissum-Pomet there were several
buildings that were also included in this functional category.41 However, the
latter case lacks also a sufficient degree of certitude, although both the layout
and the location of the buildings, close to the praetorium, would qualify them
for this purpose.42
Even though there were other hypotheses put forward in the literature about
Dacia, currently there are no buildings sufficiently documented.43 Why is this?
The existence of a specialist medical body and medical facilities, maybe
organised in tents or barracks as tabernae medicinae, is certain, despite the
fact that historical sources are not very generous with such information. Dacia
was a militarised province, but it is possible that the provincial authorities may
have regarded the building of army hospitals in the Dacian castra as, inessen-
tial after the establishment of the province. In the context we have specified
above, it is possible that the reason for such a decision was the exclusion of
this province from any major military strategy that would require such invest-
ment. However, at least the military units near the Danube could have bene-
fited from the services of military hospitals located in the castra south of the
Danube.
Accordingly, we believe that the Imperial strategy influenced the building
of military hospitals on the Danube line. Thus, the well-documented valetudi-
narium of Novae, the possible army hospitals of Singidunum and Viminacium
and, perhaps, the other hospitals, located in the auxiliary castra, may have
been part of a strategy elaborated by emperors together with their war-planning
staff. The intense activity in the provinces of Moesia Inferior and Moesia
Superior, close to the beginning of Trajan’s war with the Dacians, caused also
the construction of several military hospitals, employing significant funds
from the state, the training of medical personnel and of an entire medical corps
for these facilities in locations vital to the Roman army. This process could
only result from the implementation of a coherent healthcare strategy in the

39
Protase and Gaiu 1999, 418.
40
Marcu 2006, 465–66; Gui 1997, 119.
41
Gudea 1997a, 29, pl. 15; 1997b, 25.
42
Hyginus Gromaticus (Mun. Castr. 4) gives the reason for this location, the quiet that could
be provided to the patients: ut valetudinarium quietum esse convalescentibus posset.
43
Marcu 2006, 461; Gui 2011, 119.
66 D. APARASCHIVEI

Roman army. The famous scene on Trajan’s Column showing a soldier-

physician tending to the wounds of a comrade works as a symbol for the care
of the emperor for his soldiers, realised in the implementation of a set of meas-
ures of medical nature, as part of a well-elaborated strategy.44

The Medical Personnel in the Roman Army on the Lower Danube

From the edicts issued by various 2nd-century emperors, we know that physi-
cians enjoyed several immunities, including optional exemption from military
service.45 Yet they could also join the army for the whole duration of service
as military doctors, by their own will, after practising as civilians. Then, as
civilian doctors, they could temporarily engage in the army. Another possibil-
ity was for simple soldiers to become practising physicians, after training dur-
ing standard military service, probably instructed also by physicians with for-
mer civilian training and practice.46 Regarding the training of these military
doctors and the time allocated for their training, there is plenty of debate,
including in contemporary sources.47 It is certain that the need for military
doctors was real and perceived as such, at least from the 1st century AD.48
Some recent calculations based on the approximate number of available
Roman soldiers in the mid-2nd century AD indicate that the number of physi-
cians required to meet the needs of the army at this stage of the existence of
the Roman empire was approximately 600–800.49
As far as our area of interest is concerned, the calculations for the number
of soldiers, for example only in Moesia Inferior, would amount to 20,000–
22,000 at the end of the 1st century AD and the first half of the 2nd century
AD.50 Therefore, at least 30 doctors should have worked in the three legions of

44
Comments at Aparaschivei 2017.
45
Modestinus Dig. 27. 1. 6. 8; see Oliver 1989, 589–90.
46
Wilmanns 1995, 70–73; Bader 2014, 45–47.
47
Thessalus of Tralles, Nero’s personal physician, said he could acquire enough knowledge
of medicine in six months (Phdr. 1. 14). He was immediately challenged (Pigeaud 1993). Galen
himself, the most famous physician of the Roman period, went to study for many years between
AD 145 and 157. See also Boudon 1994, 1425–29; Ellis Hanson 1989, n. 1; Allason-Jones 1999,
134; also Aparaschivei 2012b, 89–90.
48
The first information about doctors in the Roman army is provided by Cicero, in a context
in which his aim was to show the courage of the soldiers rather than the usefulness of these pro-
fessionals. However, it is clear that military doctors have had some role in the frontline units of
Rome since the 1st century BC (Cicero Tusc. 2. 16. 38; cf. Salazar 2000, 78).
49
Wilmanns 1995, 70. For calculations regarding to the number of soldiers in the Roman
army, in different periods of the empire, see at 33–36.
50
Duch 2015, 236.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 67

the province, the I Italica, the IX Claudia and the V Macedonica, to which one
should add, of course, other medical professionals, associated with numerous
auxiliary units51 and with the Classis Flavia Moesica.52 There were two legions
in Moesia Superior during the first part of the 2nd century AD, at Viminacium
and Singidunum,53 and two alae, ten cohortes and one numerus.54 In Dacia,
the number of troops enforcing the province’s security was even larger, and
varied after the Dacian War.55
In the context of these numbers, obviously relative, I aim here at identifying
the situation of the medical staff in the three provinces that are in our sphere
of interest.
Regarding the medical corps serving in Moesia Superior, only three or four
military physicians are known (Table 1). A certain Claudius Magnus, medicus,
was attached to the cohors I Aurelia Dardanorum equitata,56 garrisoned in
Naissus.57 Also a medicus cohortis, Aelius Martialis, is attested by an inscrip-
tion, at Timacus Minus, the headquarters of cohors II Aurelia Dardanorum
milliaria equitata.58 One may include among the doctors of military units
headquartered in Moesia Superior another medicus legionis, whose funeral
stone was discovered in Aquincum, Pannonia Inferior.59 Detachments of this
legion were also stationed in Aquincum under Marcus Aurelius and Septimius
Severus, at least temporarily.60

51
Regarding the spread of auxiliary troops in Moesia Inferior, see Matei-Popescu 2010,
167–244.
52
For details about the fleet of Moesia, see Matei-Popescu 2010, 245–55.
53
Mócsy 1974, 96–99.
54
Gudea 1977, 881–82.
55
Piso 2000. For the auxiliary troops in the Dacian provinces, see Matei-Popescu 2015. Had-
rian withdrew some military units and implemented the administrative division of Dacia. The
garrisoning of the legio V Macedonica in Potaissa during Marcus Aurelius’ reign, in the context
of the Marcomannic Wars, raised the number of troops in the province.
56
Wagner 1938, 130–31; Ţentea and Matei-Popescu 2002–03, 280.
57
Krunić 2004, 50; Grbić 2013, 61–63.
58
AE 1903, 290; cf. Wagner 1938, 131–32; Ţentea and Matei-Popescu 2002–03, 280–81.
59
Wilmanns 1995, 216; Zimonyi 2014, appendix, no. 10. The doctor’s name is subject to
debate, due to the poor state of preservation of the stone.
60
Mócsy (1974, 183) believes that this military unit was sent to the province of Pannonia
whenever needed, as a reserve legion. Most likely, however, these were detachments of the
legion that were temporarily redeployed.
68 D. APARASCHIVEI

Table 1: Military physicians in the army of Moesia Superior.

Place of
Army
No. Name Position discovery/ Date References Comments
unit
province
1. Claudius cohors I medicus Naissus / AD 202 Grbić 2013,
Magnus Darda- Moesia 61–63
norum Superior
2. T. Aelius cohors II medicus Timacus AE 1903, 290;
Martialis Darda- cohortis Minus / IMS III/2, 49
norum Moesia
Inferior
3. M. Valerius legio VII medicus Drobeta / 2nd Benea 1974–75; Military doctor
Longinus Claudia legionis Dacia century IDR II, 42 who practised in
AD the castrum of
Viminacium
(Moesia Superior)
4. [--Cum] legio IV medicus Aquincum / AD CIL III, 3537; units of the legio
mia(?) Flavia legionis Pannonia 197–202 Wilmanns 1995, IV Flavia were
Inferior 216, no. 64; temporarily de-
Zimonyi 2014, tached here from
appendix, no. 10. Singidunum
(Moesia Superior)

Valerius Longinus, medicus legionis, served in the legio VII Claudia, head-
quartered in Viminacium during the Marcomannic Wars. He received orna-
menta decurionalia from the city of Drobeta, in Dacia, probably for services
rendered to the city.61 The relationship between military doctors and civilians,
and vice versa, between civilian physicians and the military was a normal one.
We have clear evidence in the Graeco-Roman world.62 Besides Longinus, we
have other examples from our area of interest. In Moesia Inferior, two doctors,
probably civilians, were identified in inscriptions uncovered in the military

61
The tombstone was discovered in Drobeta and is the only mention of a military doctor on
the territory of Dacia, even though he actually served in a legion pertaining to the military organ-
isation of Moesia Superior (Benea 1974–75; IDR II, 42; Wilmanns 1995, 222–24, no. 70; Apara-
schivei 2012b, 194–95).
62
One interesting case is the physician Marcus Ulpius Telesp(h?)orus, who was medicus alae
Indiana Gallorum in Germania Superior and then served with ala III Asturum, in Mauretania
Tingitana, to eventually arrive to be a civilian physician, medicus salariarius civitatis Ferentien-
sium (Wilmanns 1995, 194–96; Bader 2014, 54). Then, there are several examples of physicians
within the Greek world who cared patients both military and civilian, in times of war (Samama
2003, nos. 126, 367). For more details, see Salazar 2000, 69; Aparaschivei 2012b, 194–96.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 69

hospital of Novae.63 Then we have the very interesting case of Asclepiades,

mentioned on a funeral stone found in Odessos as archiatros, the chief of doc-
tors in the city.64 Even though it is obvious that at the time of death he served
the civilian community of Odessos, within the inferior register of the funeral
monument we find pieces of the military equipment of a Roman soldier. They
may suggest that at some point he served in the army, most likely also as
a physician.65
For Dacia, besides Longinus, which was more related to the community of
Drobeta, there are no other military doctors known. Instead, we have two
names of ophthalmologists on the signacula medicorum oculariorum. These
are P. Cornelius Colonus66 and T. Attius Divixtus (Table 2, nos. 1–2).67

Table 2: Medici ocularii in the Lower Danubian provinces.

Place
No. Name of discovery/ Date References Comments
province
1. P. Cornelius Colonus Gârbou/Dacia second half of AE 1982, 837 discovered near the
2nd–first part of northern limes of
3rd century AD Dacia
2. T. Attius Divixtus Apulum/Dacia second half of IDR III/6, 431
2nd–first part of
3rd century AD
3. C. Iulius Iunianus Viminacium/ second half of IMS II, 225;
Moesia Superior 2nd century AD Mirković 1986,
218
4. Aelius Polyctetus Singidunum/ second half of Mirković 1986
Moesia Superior 2nd century AD

This mention of the two does not necessarily imply a direct connection with
the military. However, it is noteworthy that the two seals with collyria were
discovered in Gârbou, near the northern limes of Dacia, respectively in Apu-
lum, an important town, and also a castrum of the legio XIII Gemina

63
AE 1998, no. 1134; IGLN 176. For other examples in neighbouring regions, see Zimonyi
2014, 195–98.
64
IGB I2, 150.
65
Aparaschivei 2017, 73, pl. 26, fig. 1.
66
AE 1982, 837, found in Gârbou. For the spread of cognomen Colonus, see OPEL II, 69.
67
IDR III/6, 431, from Apulum; see also Gui 2011, 116. One should note the Celtic origin of
the name (OPEL II, 103), a fact that argues for the idea that the ophthalmologists’ stamps have
clearly a higher rate of occurrence in the Celtic influence regions.
70 D. APARASCHIVEI

throughout the existence of the province of Dacia. It is already well know from
literature that eye diseases were very dangerous for soldiers, but also very
common. Specific diseases, which were mentioned quite often, were the lip-
pitudo and aspritudo, along with about 25 other ailments recorded on the oph-
thalmologists’ stamps. These diseases were facilitated by conditions of
hygiene, dust, smoke, diet and, in general, by the daily habits of the troops.68
Such conditions afflicted both the sick soldier, but they also threatened his
comrades through their infectious potential.69 Also, a soldier with poor eye-
sight was a danger to himself and to his comrades during the fight. In the
Vindolanda tablets, soldiers with eye problems are described as being unfit to
carry out their tasks.70 The treatments (collyria) for eye ailments indicated on
these signacula were certainly in use at that time, including for treating very
common eye diseases in the military environment. So, it is supposed that oph-
thalmologists were in great demand in the army.
From the same region, in Moesia Superior, we have information about two
other ophthalmology specialists (Table 2, nos. 3–4). One of them, C. Iulius
Iunianus, worked in the powerful military centre of Viminacium, sometime in
the 2nd century AD.71 The other one, Aelius Polyctetus, was an ophthalmolo-
gist practising in Singidunum.72 The latter was identified on a stamp found
right near the legionary castrum, which may suggest a direct connection with
the army’s needs for such specialists. Moreover, at the level of the whole
empire, many of these oculists’ stamps have been found near military units.
For this reason, it was hypothesised that these pieces, which could also have
the role of amulets, would have been used particularly by military doctors or
civilian physicians working in the army.73
It is worth mentioning the cremation grave of a medicus ocularius found
in the territory of Viminacium, dated to the end of 1st century AD and the

68
For details about eye afflictions and the specialists in ophthalmology, see Jackson 1996,
2228–51.
69
Jackson 1996, 2229–30.
70
Bowman, and Thomas 1994, 90–98; Birley 1992. There are other contemporary sources
that include testimonies about soldiers particularly investigated for possible eye diseases. Of
interest is the information from an Egyptian Oxyrhynchos papyrus, P. Oxy. 1. 39, where a certain
soldier Tryphon from Alexandria was discharged after the consultation of three doctors because
he had poor vision (after Davies 1969a, 92; 1969b, 211 and 222; Baker 2004, 37).
71
IMS II, 225; Mirković 1986, 218, pl. 10, nos. 2a–2b.
72
Mirković 1986, 217–18, pl. 10, nos. 1a–1d.
73
Nielsen 1974, passim; Allason-Jones 1999, 136–37; Pérez-Cambrodí et al. 2013, 90–92.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 71

beginning of the next one.74 The grave-goods included, among other things,
instruments used for performing cataract or trachoma surgery, as well as spe-
cific drugs.75
In my opinion, military doctors were most likely focused on the healing of
wounds and less specialised in ophthalmology or dentistry, for example.76 The
oculists mentioned in collyria stamps were not military doctors, but they cer-
tainly served various military units, with which they could have more or less
official contractual relations as specialists. As far as we know, there is only
one clear mention of the service of a medicus ocularius in a military unit, in
the fleet, which information comes from a literary source and not from an
inscription.77 Moreover, military doctors, unlike civilians, had the privilege of
a fixed salary, being spared the need to search actively for paying patients.
However, the dangers they were exposed to were obviously greater. As eye
afflictions were widespread, the doctors believed logically that there would be
no shortage of clients, so the temptation to join the army as military doctors
was less. We can assume that the most advantageous manner of working with
the army, which in turn needed such specialists, was to conclude fixed-term
contracts. Therefore, their titles did not include the military units with which
they had worked, but they undoubtedly offered their services in exchange for
material benefits.
The epigraphic sources pertaining to Moesia Inferior provide the most accu-
rate evidence for the existence of a complex medical corps within the Roman
army (Table 3).

74
Korać 1986; Krunić 2001, 125–27. About other discoveries of tombs of doctors in the
Viminacium area, see Spasić-Đurić 2002, 62–64, who also postulates the existence of a valetudi-
narium; see also Spasić-Đurić 2005. These finds are clear evidence of the evolution of medical
science to a level comparable to what is happening in other corners of the Empire, but cannot be
linked to the military environment.
75
Among the grave-goods it is worth mentioning a case (Korać 1986, pl. 4.17) which con-
tained, in several compartments, nine surgical instruments (four scalpels, two sharp hooks, one
needle, one tweezer [vulsella], pl. 1, 1 [A–I] and pl. 4.1–9; as well as drugs in the form of pas-
tilles: pl. 4.10–13; see also Spasić-Đurić 2002, 63).
76
Allason-Jones 1999, 138.
77
Axius, medicus ocularius classis Britannicae, is mentioned by Galen De comp. med. 4. 8.
Generally, there are no mentions of specialist doctors within the military. However, there is the
exception of one medicus chirurgus cohortis in Italy (AE 1945, 62).
72 D. APARASCHIVEI

Table 3: Military physicians in the army of Moesia Inferior.

Place
No. Name Army unit Position of discovery/ Date References Comments
province
1. Aurel(ius) vexillatio medicus Montana/ AD 155 CIL III, 7449
Artemo legionis XI Moesia Inferior
Claudia
2. Q(uintus) cohors medicus Carsium/ 2nd cen- CIL III, 7490;
Erucius ignota cohortis Moesia Inferior tury AD ISM V, 103
Victor
3. Ael(ius) ala ignota medicus Troesmis/ late 2nd CIL III, 6205;
Aur[elia]nus alae Moesia Inferior century ISM V, 170
AD
4. [V]etu[ri]us legio I medicus Barboşi (dep. late 2nd– CIL III, 7517;
Italica legionis Galaţi)/Moesia early 3rd ISM V, 299
Inferior (?) century
AD
5. L(ucius) legio I medicus Tyras (Bilhorod- late 2nd– Karyshkovskii
Pa[piri]us Italica vexil- Dnistrovskyy) early 3rd 1987, 53; AE
Olymphicus lationis Moesia Inferior (?) century 1995, 1350;
AD Sarnowski
1995, 326
6. N(umerius) classis medicus Tyras (Bilhorod- late 2nd– Karyshkovskii
Seius Flavia dupli- Dnistrovskyy) early 3rd 1987, 53; AE
Ga[rgil?]ius Moesica carius Moesia Inferior (?) century 1995, 1350;
AD Sarnowski
1995, 326
7. Ae(lius) legio I medicus Novae/ second Kolendo (civilian
Macedo Italica (?) Moesia Inferior half of 1998, 62–64; physician in
2nd cen- AE 1998, military
tury AD 1134 service?)
8. Διό[δω]ρος legio I medicus Novae/ after AD IGLNovae (civilian
Italica (?) (?) Moesia Inferior 212 176; Kolendo physician in
1998, 64 military
service?)

Aurelius Artemo is a physician mentioned in a 2nd-century AD inscription.78

His unit, the legio XI Claudia, was headquartered in that period at Durosto-
rum, but a vexillatio of this legion was garrisoned at Montana, where the
inscription was discovered.79 An interesting fact is that, around one century

78
CIL III, 7449; Gummerus 1932, no. 431; Aparaschivei 2012b, 90–91.
79
Velkov and Alexandrov 1988, 271–77; Matei-Popescu 2010, 136–37.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 73

before, another medicus of the legio XI Claudia is mentioned in Burnum, in

the province of Dalmatia, where that legion was posted at the time.80 Another
medicus of Moesia Inferior is mentioned in a funerary inscription of Carsi-
um.81 Quintus Erucius Victor, medicus cohortis, was active at some time dur-
ing the 2nd century AD, most probably in an auxiliary cohort about whose
place of origin there is no information. Troesmis is the origin of a medicus
alae.82 This medicus served with the ala that covered the region from which
the legio V Macedonica was withdrawn. At Barboşi, in a castrum located on
the left bank of the Danube guarding the north-eastern border of Moesia Infe-
rior, a certain Veturius, medicus legionis I Italicae, is mentioned.83 Two other
army physicians, who attended the units garrisoned in this centre at the mouths
of the Dniester, are represented on a monument from Tyras dedicated to Ascle-
pius and Hygieia.84 Lucius Papirius Olymphicus, medicus vexillationis, appears
together with Numerius Seius Ga[rgil?]ius, medicus duplicarius classis Flaviae
Moesiacae as dedicators of this votive monument. For the first one the desig-
nation of medicus vexillationis is the only occurrence known so far within the
territory of the empire. It is worth noting that the Roman medical system was
flexible, adapting to the military needs of the moment. So, for a vexillatio,
a temporary ‘task force’ combining detachments from regular units, a special
medicus was assigned, probably chosen from among all the doctors of those
legions from which the detached soldiers came.85
The second medicus, N. Seius Ga[rgil?]ius, medicus duplicarius, was
attached to a unit of the Classis Flavia Moesica, headquartered at Novio-
dunum. As confirmed by the cases of other naval units, medical services were
provided by medici duplicarii.86

80
This is a certain Varius Aristo, [me]dicus [leg(ionis)] XI, who was active around the mid-
dle of 1st century AD with the legion, which was stationed at the time in Dalmatia (Cesarik 2014,
739–44).
81
ISM V, 103.
82
The name of this physician, Aelius Aurelianus, is hypothetical (ISM V, 170).
83
ISM V, 299.
84
Karyshkovskii 1987, 53; AE 1995, 1350; Sarnowski 1995, 326; Aparaschivei 2012b,
93–95.
85
It is interesting, however, that Aurelius Artemo, the doctor of a vexillatio of the legio XI
Claudia (see above), posted in Montana, is mentioned only as medicus. Had he possibly had
another position in that unit than the one of medicus vexillationis, as Papirius Olymphicus was
mentioned? Could he have been a civilian physician temporarily employed in the army? In the
case of another physician who served with a vexillatio of the legio I Minervia, in Germania Infe-
rior, the title is miles medicus (Wilmanns 1995, 185–86, no. 38).
86
Medici duplicarii are attested in the naval units stationed in Italia, classis praetoria Mis-
ennsis and classis praetoria Ravennas (CIL VI, 3910, 32769; CIL X, 3441, 3442 [medicus dupli-
carius triere], 3443, 3444; CIL XI, 29 [medicus duplicarius nave], 6944; AE 1984, 337; see
Nutton 1970.
74 D. APARASCHIVEI

If for these six doctors we are certain that each of them served in a military
unit, there are, however, two interesting mentions from the same province.
Two inscriptions were found within the valetudinarium of Novae, which men-
tion Aelius Macedo87 and Diodoros.88 Most probably, the two were civilian
physicians employed in the military hospital in the castrum of Novae.
The temporary commissioning of civilian physicians as army physicians was
a usage that was quite probable, as well as documented throughout the empire.89
Finally, the healthcare activities in one or another of the provinces could be
inferred from the density and diverseness of the medical instruments uncov-
ered through archaeological excavations. In all three provinces under scrutiny,
a large number of such artefacts were uncovered in various civilian centres,
as well as in castra, where they could have been used by army medical
personnel.
For Moesia Superior, impressive finds of surgical instruments and objects
employed in medicine and pharmacy were found at Singidunum (Belgrade),
Viminacium (Kostolac), Timacus Minus (Knjaževac), Kosmaj, Transdierna
(Tekija) and Diana (Karataš).90 Likewise, the territory of Dacia produced
a large number of archaeological finds of medical instruments from several
castra (Porolissum, Bologa, Buciumi, Râşnov, Gilău, etc.), in addition to finds
from urban settlements (civilian).91 In Moesia Inferior, the quantity of medical
instruments uncovered by archaeology is comparable to other provinces,
although the find context makes it hard to ascertain whether their use was mili-
tary or civilian.92

Some Conclusions

The Roman state was definitely involved, either at the level of prevention or at
that of caring for wounded soldiers, particularly in times of conflict. More and
more believable is the hypothesis that the Roman military was provided with
a strategy of healthcare. But when was this well-organised system for preven-
tion and treatment for Roman army personnel implemented on the Lower
Danube?
Around the time the Dacian War began, as well as between the two expedi-
tiones, apart from the strategic plans relating to the disposition and

87
Kolendo 1998, 62–64; AE 1998, 1134.
88
IGLN 176; Kolendo 1998, 64.
89
See discussions in Aparaschivei 2012b, 68–69.
90
Krunić 1992, 51–78; 2000.
91
Gui 2011, 119–25, with all the literature from the province; add Varga 2015.
92
Kirova 2010, 23–79 and appendices; Aparaschivei and Vasilache 2012.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 75

manoeuvre of troops, the strengthening of castra, the construction of bridges

and roads, etc.,93 Trajan and his military staff also considered a strategy for
the recovery of soldiers to be employed during the war. According to the
available information on the healthcare facilities of the Lower Danubian
region, especially about the valetudinarium of Novae, and identified medical
staff, it is only at the end of the 1st century AD, and especially during the
2nd century AD, that we may speak of a strengthening of the healthcare sys-
tem within the Roman army. Thus, even if Trajan launched the strategy in the
context of the Dacian War, the system he introduced functioned ever better
after his reign.
Eleven possible military doctors from the two Moesian provinces were
active during 2nd century AD and possibly at the beginning of the 3rd. The
valetudinarium in Stojnik is recorded in the second part of the 2nd century.
Many medical instruments identified in the fortresses correspond to the
2nd–3rd-century AD levels of these fortifications. On the other hand, after
the end of 3rd century, there is no longer substantial information about how
the healthcare system of the military worked. Most likely, it was not so struc-
tured, and the many problems faced by the army and the empire in general led
to a degradation of these services.
According to contemporary sources, any legionary castrum would have had
a valetudinarium. And yet the archaeological research does not confirm the
implementation of this recommendation. It is obvious that military bases where
such hospitals with qualified staff were located, either legions, auxiliary troops
or navy, were true focal nodes where patients from other areas could be
brought. The level of connectivity between the castra, as well as the well-
established supply system allowed for a good linkage between the different
fortifications on the limes.94 Therefore, some of the military units may not
have permanent healthcare facilities and trained medical personnel, but such
a hospital or medical unit would have been within reach.
In the case of the Danubian limes, the array of fortifications along the river
was designed based on strategic considerations, but also aiming at most effi-
cient supply.95 This is argued by the fact that after the transformation of Dacia
into a province, some of these forts continued to function, even if there were

93
See Mirković 1996, 35–39.
94
Israelowich 2015, 87.
95
Whittaker 2008, 317; Matei-Popescu 2010, 30.
76 D. APARASCHIVEI

various permutations of troops.96 Thus, establishing hospitals on this line, at

least at strategic and easily supplied hub locations, was a rational process.97
The presence of valetudinaria attested in the area implies automatically the
existence of a body of medical specialists and other medical personnel. An
analysis of the affiliation of medical personnel in the region shows that they
were integrated within legiones (I Italica, VII Claudia, XI Claudia, IV Flavia)
or vexillationes, within Classis Flavia Moesica, as well as within auxiliary
units, alae and cohorts. We have, therefore, evidence on military doctors or
civilian doctors temporarily working within the army. There is also every rea-
son to believe that the oculists recorded in both Dacia and Moesia Superior
were employed in the healing of soldiers affected by eye diseases.
Even though the medical system in the Roman army should have been uni-
tary, the diversity and complexity of medical personnel in Moesia Inferior is
evident, and in contrast with the total absence of military medics in Dacia.
This is the result, I believe, notwithstanding any statistical deficiencies and
shortcomings in our state of knowledge, of a major advantage that the south
Danubian province had over the north Danubian one. The Greek and Hellenis-
tic healthcare culture and tradition in this province had resulted in the forma-
tion of a pool of civilian physicians who could also be utilised by the
military.98

BIBLIOGRAPHY

Allason-Jones, L. 1999: ʻHealth Care in the Roman North’. Britannia 30, 133–46.
André, J. 1987: Être médecin à Rome (Paris).
Antal, A. 2014: ʻA God of Convalescence. Telesphorus/Genius Cucullatus in Roman
Dacia’. Acta Musei Napocensis 51.1, 195–206.

96
For the hypothesis of deserted forts on the Moesian limes, after the conquest of Dacia, see
Mirković 2002, 757–63.
97
See the details in Lemke 2016. One should note that, for example, significant quantities of
lamps and tableware have been discovered in the hospital of Novae in special rooms at hallways’
ends (Dycek 2002, 686). Thus, in addition to what was necessary for the medical act itself, main-
tenance of such a structure involved a much larger quantity of supplies: food, ceramic, glass or
metal vessels, lamp oil, wine and fish products used for medical purposes. An interesting 2nd-
century AD inscription found in the legionary hospital of Aquincum confirms the tax exemption
for wine used in military hospitals: immune in r(ationem) val(etudinarii) leg(ionis) II Adi(utricis)
– AE 1933, 120; AE 1996, 1260. The use of wine and various fish sauces for therapeutic pur-
poses is confirmed by other sources of the time (cf. Davies 1970, 92; Curtis 1991, 1–4, 176).
98
In this regard, the mentions of medical colleges in Hellenistic and Roman inscriptions in
Greek cities such as Histria, Dionysopolis and Odessos, which we know from contemporary
sources, is a good argument (see Aparaschivei 2012b, 71–72, 74–75).
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 77

Aparaschivei, D. 2010: Oraşele romane de la Dunărea Inferioară (secolele I–III p.

Chr.) (Iaşi).
—. 2012a: ʻPhysicians and Medicine in the Roman Army of Moesia Inferior’. Dacia
n.s. 56, 99–118.
—. 2012b: Healthcare and Medicine in Moesia Inferior (Iaşi).
—. 2017: ʻMedical Care for the Roman Army on Trajan’s Column?’. In Mitthof, F.
and Schörner, G. (eds.), Columna Traiani – Traianssäule Siegesmonument und
Kriegsbericht in Bildern (Beiträge der Tagung in Wien anlässlich des 1900.
Jahrestages der Einweihung, 9.–12. Mai 2013) (Vienna), 71–76.
Aparaschivei, D. and Vasilache, V. 2012: ʻInstrumentarul medical şi/sau de cosmetică
din Moesia Inferior’. Pontica 45, 279–310.
Bader, P. 2014: ʻThe Identity, Legal Status and origin of the Roman Army’s Medical
Staff in the Imperial Age’. In Maire, B. (ed.), Greek and Roman in Latin Medical
Texts: Studies in Cultural Change and Exchange in Ancient Medicine (Studies in
Ancient Medicine 42) (Leiden/Boston), 43–59.
Baker, P.A. 2002: ʻThe Roman Valetudinaria: Fact or Fiction?’. In Arnott, R. (ed.),
The Archaeology of the Medicine (Papers given at a session of the Annual Confer-
ence of the Theoretical Archaeology Group held at the University of Birmingham
on 20 December 1998) (BAR International Series 1046) (Oxford), 69–79.
—. 2004: Medical Care for the Roman Army on the Rhine, Danube and British Fron-
tier in the First, Second and Third Centuries AD (BAR International Series 1286)
(Oxford).
Barbara, S. 2014: ʻ“Memorial” Strategies of Court Physicians in the Imperial Period’.
In Maire, B. (ed.), Greek and Roman in Latin Medical Texts: Studies in Cultural
Change and Exchange in Ancient Medicine (Studies in Ancient Medicine 42)
(Leiden/Boston), 25–42.
Benea, D. 1974–75: ʻUn medic al legiunii a VII-a Claudia la Drobeta în timpul
războaielor marcomanice’. Sargetia 11–12, 303–06.
Birley, A.R. 1992: ʻA case of eye disease (Lippitudo) on the Roman frontier in
Britain’. Documenta Ophthalmologica 81.1, 111–19.
Boudon, V. 1994: ʻLes œuvres de Galien pour les débutants (“De sectis”, “De pulsi-
bus as tirones”, “De ossibus ad tirones”, “Ad Glauconem de methodo modendi”
et “Ars medica”): médicine et pédagogie au IIe s. ap. J.-C.’. ANRW II 37.2,
1421–67.
Bowman, A.K. and Thomas, J.D. 1994: The Vindolanda Writing Tablets (Tabulae
Vindolandenses II) (London).
Cesarik, N. 2014: ʻThe Inscription of Medicus of the XIth Legion from Burnum’.
Collegium Antropologicum 38.2, 739–44.
Curtis, R. 1991: Garum and Salsamenta: Production and Commerce in materia
medica (Studies in Ancient Medicine 3) (Leiden).
Davies, R.W. 1969a: ʻThe Medici of the Roman Armed Forces’. In Epigraphische
Studien vol. 8 (Düsseldorf), 83–99.
—. 1969b: ʻJoining the Roman Army’. Bonner Jahrbücher 169, 208–32.
—. 1970: ʻThe Roman military medical service’. Saalburg-Jahrbuch 27, 84–104.
Deonna, W. 1955: De Télesphore « au Moine Bourru »: Dieux, Génies et Démons
encapuchonnés (Brussels).
Dietz, K. 1994: ʻZum Ende der Markomannenkriege: die expeditio Germanica tertia’.
In Friesinger, H., Tejral, J. and Stuppner, A. (eds.), Markomannenkriege:
78 D. APARASCHIVEI

Ursachen und Wirkungen (Spisy Archeologického ústavu av Čr Brno 1) (Brno),

7–15.
—. 2000: ʻLegio III Italica’. In Le Bohec, Y. and Wolff, C. (eds.), Les légions de Rome
sous le Haut-Empire (Actes du congrès de Lyon, 17–19 septembre 1998) (Collec-
tion du Centre d’études romaines et gallo-romaines n.s. 20) (Lyons), 133–43.
Duch, M. 2015: ‘The Impact of Roman Army on Trade and Production in Lower
Moesia (Moesia Inferior)’. Studia Europaea Gnesnensia 11, 235–60.
Dyczek, P. 1997: ʻNovae – western sector, 1996’. Archeologia 48, 43–49.
—. 1999: ʻA sacellum Aesculapii in the valetudinarium at Novae’. In Gudea, N. (ed.),
Roman Frontier Studies (Proceedings of the XVIIth International Congress of the
Roman Frontier Studies) (Zalău), 495–500.
—. 2000: ʻNovae – western sector (section IV), 1997–1998’. Archeologia 51, 89–103.
—. 2002: ʻRemarks on supply of the Roman army from the point of view of the
valetudinarium at Novae (Moesia Inferior)’. In Freeman, P., Bennett, J.,
Fiema, Z.T. and Hoffmann, B. (eds.), Limes XVIII (Proceedings of the XVIIIth
International Congress of the Roman Frontier Studies held in Amman, Jordan,
September 2000) (BAR International Series 1084) (Oxford), 685–94.
—. 2003: ʻForty years of excavations at Novae’. Novaensia 14, 7–20.
—. 2004: ʻWhy the legionary hospital in Novae is unique?’. Ephemeris Dacoromana
12, 159–69.
—. 2005: ʻThe Site of the Valetudinarium in Novae in the Third Century A.D.:
Remodelling the Architecture’. In Mirkovic, M. (ed.), Römische Städte und
Festungen an der Donau (Akten der regionalen Konferenz. Organisiert von Alexan-
der von Humboldt-Stiftung, Beograd, 16.–19. Oktober, 2003) (Belgrade), 231–38.
Ellis Hanson, A. 1989: ʻDoctors and diseases’. Journal of Roman Archaeology 2,
299–304.
Grbić, D. 2013: ʻVojni lekari u Gornjoj Meziji’. Etno-kulturološki zborni 17, 61–66.
Gudea, N. 1977: ʻDer Limes Dakiens und die Verteidigung der obermoesischen
Donaulinie von Trajan bis Aurelian’. ANRW II 6, 851–87.
—. 1997a: Das Römergrenzkastell von Buciumi/ Castrul roman de la Buciumi (Zalău).
—. 1997b: Das Römergrenzkastell von Moigrad-Pomet = Porolissum1/ Castrul roman
de la Moigrad-Pomet = Porolissum 1 (Zalău).
Gui, M. 2011: ʻEvidence for Medical and Personal Care in the Case of the Roman
Army in Dacia’. Ephemeris Napocensis 21, 115–30.
Gummerus, H. 1932: Der Ärztesstand im römischen Reiche nach dem Inschriften
(Helsinki).
Israelowich, I. 2015: Patients and Healers in the High Roman Empire (Baltimore).
Jackson, R.P.J. 1996: ʻEye Medicine in the Roman Empire’. ANRW II 37.3, 2228–51.
Karyshkovskii, P.O. 1987: ʻPosvyatitelnaya nadpis’ rimsckikh vrachei iz Tiry. Novye
issledovaniya po arkheologii Severnogo Prichernomor’ya’. Sbornik naucnykh
trudov, 52–56.
Kirova, N. 2010: Medicinata v Provincite Dolne Mezia e Trakia (I–Krai na III v.) (Sofia).
Kolendo, J. 1998: ʻInscriptions en l’honeur d’Esculape et d’Hygia du Valetudinarium
de Novae’. Archeologia 49, 55–70.
Korać, M. 1986: ʻMedicus et chirurgus ocularius iz Viminacijuma’. Starinar 37,
53–71.
Kovács, P. 2009: Marcus Aurelius’ Rain Miracle and the Marcomanic Wars
(Mnemosyne Suppl. 308) (Leiden/Boston).
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 79

Krunić, S. 1992: Rimski medicinski i farmaceutski instrumenti iz Singidunuma i oko-

line (Belgrade).
—. 2000: Rimski medicinski, farmaceutski i kozmetički instrumenti na teritoriji Gornje
Mezije (Belgrade).
—. 2001: ʻLekarske antichke kutije tipa loculamentum’. Zbornik Narodnog Muzeja
17.1, 121–34.
—. 2004: ʻOrganizacija medicinske sluzhbe u Gornjoj Meziji svetlu nalaza iz
Singidunuma’. Godishnjak grada Beograda 51, 41–56.
Künzl, E. 2005: ʻAesculapius im Valetudinarium’. Archäologisches Korrespondenz-
blatt: Urgeschichte, Römerzeit, Frühmittelalter 35, 55–64.
Lemke, M. 2016: ʻDanube, limes and logistics. Some thoughts on Roman army supply
in Moesia Inferior’. Novaensia 27, 9–38.
Marcu, F. 2006: ʻThe Valetudinarium at Ilişua’. In Gazdac, C. and Gaiu, C. (eds.),
Fontes historiae. Studia in honorem Demetrii Protase (Bistriţa), 461–70.
Matei-Popescu, F. 2010: The Roman Army in Moesia Inferior (Bucharest).
—. 2015: ʻThe Auxilary Units in Moesia Superior and Dacia. A Review and an
Update’. In Vagalinski, L. and Sharankov, N. (eds.), Limes XXII (Proceedings of
the 22nd International Congress of Roman Frontier Studies, Ruse, Bulgaria,
September 2012) (Sofia), 407–18.
Matei-Popescu, F. and Ţentea, O. 2006: ʻParticiparea trupelor auxiliare din Moesia
Superior şi Moesia Inferior la cucerirea Daciei’. In Teodor, E.S. and Ţentea, O.
(eds.), Dacia Augusti Provincia. Crearea provinciei (Actele simpozionului
desfăşurat în 13–14 octombrie 2006 la Muzeul Naţional de Istorie a României)
(Bucharest), 75–120.
Milner, N.P. (ed. and transl.) 2001: Vegetius: Epitome of Military Science (Liverpool).
Mirković, M. 1986: ʻZwei neue Stempel von Augenärzten aus Obermösien’. ZPE 64,
217–18.
—. 1996: ʻThe Iron Gates (Ɖerdap) and the Roman Policy on the Moesian Limes AD
33–117’. In Petrović, P. (ed.), Roman Limes on the Middle and Lower Danube
(Belgrade), 27–40.
—. 2000: ʻEin praefectus castrorum (?) legionis IV Flaviae in Singidunum’. In
Popovic, M. (ed.), Singidunum, vol. 2 (Belgrade), 7–10.
—. 2002: ʻDeserted forts – the Moesian limes after the conquest of Dacia’. In Freeman,
P., Bennett, J., Fiema, Z.T. and Hoffmann, B. (eds.), Limes XVIII (Proceedings of
the XVIIIth International Congress of the Roman Frontier Studies held in Amman,
Jordan, September 2000) (BAR International Series 1084) (Oxford), 757–63.
Mócsy, A. 1974: Pannonia and Upper Moesia: A History of the Middle Danube
Provinces of the Roman Empire (London).
Nielsen, H. 1974: Ancient Ophtalmological Agents: A Pharmaco-Historical Study of
the Collyria and Seals for Collyria used during Roman Antiquity, as well as of the
Most Frequent Components of the Collyria (Odense).
Nutton, V. 1969: ʻMedicine and the Roman Army. A Further Reconsideration’.
Medical History 13, 260–70.
—. 1970: ʻThe Doctors of the Roman Navy’. Epigraphica 32, 66–71.
—. 1977: ʻArhiatri and the Medical Profesion in Antiquity’. Papers of the British
School at Rome 45, 191–226.
Oliver, J.H. 1989: Greek Constitutions of Early Roman Emperors from Inscriptions and
Papyri (Memoirs of the American Philosophical Society 178) (Philadelphia).
80 D. APARASCHIVEI

Opreanu, C. 2006: ʻActivitatea constructivă a legiunii IIII Flavia Felix la nordul

Dunării, în anii 101–117 d. Chr.’. In Teodor, E.S. and Ţentea, O. (eds.), Dacia
Augusti Provincia. Crearea provinciei (Actele simpozionului desfăşurat în 13–14
octombrie 2006 la Muzeul Naţional de Istorie a României) (Bucharest), 50–74.
Pérez-Cambrodí, R.J., Pinẽro, D.P., Cerviño, A., Brautaset, R. and Murube del Castillo,
J. 2013: ʻCollyria Seals in the Roman Empire’. Acta Medico-Historica Adriatica
11.1, 89–100.
Pigeaud, J. 1993: ʻL’introduction du Méthodisme à Rome’. ANRW II 37.1, 565–99.
Piso, I. 2000: ʻLes légions dans la province de Dacie’. In Le Bohec, Y. and Wolff, C.
(eds.), Les légions de Rome sous le Haut-Empire (Actes du congrès de Lyon,
17–19 septembre 1998) (Collection du Centre d’études romaines et gallo-romaines
n.s. 20) (Lyons), 205–25.
Press, L. 1985: ʻThe valetudinarium and the Portico Building in Novae’. In Fol, A.,
Tacheva, M. and Nedjalkov, N. (eds.), Terra Antiqua Balcanica II: Studia in
honorem C.M. Danov (Sofia), 367–71.
—. 1987: ʻThe valetudinarium at Novae after four seasons of archaeological excava-
tions’. Ratiarensia 3–4, 171–84.
—. 1990: ʻValetudinarium of Legio I Italica at Novae’. Studia i Prace 15, 327–34.
—. 1994: ʻThe legionary valetudinarium at Novae in comparation with other Danubian
Hospitals’. In Susini, G.C. (ed.), Limes (Studia di storia 5) (Bologna), 93–100.
Protase, D. and Gaiu, C. 1999: ʻLe camp romain et l’établissement civil d’Ilişua. Les
résultats des fouilles archéologiques effectuées dans le courant des années 1978–
1995’. In Gudea, N. (ed.), Roman Frontier Studies (Proceedings of the XVIIth
International Congress of the Roman Frontier Studies) (Zalău), 415–29.
Russu, I.I. 1972: ʻGetica lui Statilius Crito’. Studii Clasice 14, 111–28.
Salazar, C. 2000: The Treatment of War Wounds in Graeco-Roman Antiquity (Studies
in Ancient Medicine 21) (Leiden/Boston/Cologne).
Samama, E. 2003: Les médecins dans le monde grec. Sources épigraphiques sur la
naissance d’un corps médical (Hautes études du monde gréco-romain 31)
(Geneva).
Sarnowski, T. 1995: ʻL’organisation hiérarchique des vexillationes Ponticae au miroir
des trouvailles épigraphiques récentes’. In Le Bohec, Y. (ed.), La Hiérarchie
(Rangordnung) de l’armée Romaine sous le Haut Empire (Actes du Congrès de
Lyon, 15–18 septembre 1994) (Paris), 323–28.
Scarborough, J. 1968: ʻRoman Medicine and the Legions: a Reconsideration’. Medical
History 12.3, 254–61.
—. 1985: ʻCriton, Physician to Trajan: Historian and Pharmacist’. In Eadie, J.W. and
Ober, J. (eds.), The Craft of the Ancient Historian: Essays in Honor of Chester
G. Starr (Lanham, MD), 387–405.
von Schnurbein, S. 1974: Die römischen Militäranlagen bei Haltern: Bericht über die
Forschungen seit 1899 (Bodenaltertümer Westfalens 14) (Münster).
Spasić-Đurić, D. 2002: Viminacium, the Capital of the Roman Province of Upper
Moesia (Požarevac).
—. 2005: ʻGrob lekara iz Viminaciuma’. Glasnik Srpskog Arheoloshkog Drushtva 21,
281–92.
Ţentea, O. and Matei-Popescu, F. 2002–03: ʻAlae et Cohortes Daciae et Moesiae.
A review and updating of J. Spaul’Ala and Cohors’. Acta Musei Napocensis
39–40, 259–96.
 FACILITIES AND MEDICAL STAFF OF THE ROMAN ARMY 81

Varga, T. 2015: ʻMedical Instruments in Roman Dacia: a survey beyond typology and
functionality’. Acta Musei Napocensis 52, 183–204.
—. 2016: ʻTelesphorus. A healing child god in Roman Dacia’. Studia Universitatis
Babeș-Bolyai. Historia 60.1, 17–40.
Velkov, V. and Alexandrov, G. 1994: Montana: Epigrafski pametnici ot Montana i
rajona (Sofia).
Wagner, W. 1938: Die Dislokation der römischen Auxiliarformationen in den
Provinzen Noricum, Pannonien, Moesien und Dakien von Augustus bis Gallienus
(Neue deutsche Forschungen 203) (Berlin).
Whittaker, C.R. 2008: ʻThe Frontiers’. In Bowman, A.K. et al. (eds.), The Cambridge
Ancient History 11: The High Empire, AD 70–192, 2nd ed. (Cambridge),
293–319.
Wilmanns, J. 1995: Der Sanitätsdienst im römischen Reich: Eine sozialgeschichtliche
Studie zum römischen Militärsanitätswesen nebst einer Prosopographie des
Sanitätspersonals (Medizin der Antike 2) (Hildesheim/Zürich/New York).
Zimonyi, A. 2014: ʻPhysicians and Health Care in Roman Pannonia’. In Németh, G.
and Bajnok, D. (eds.), Miscellanea Historiae Antiquitatis (Proceedings of the
First Croatian-Hungarian PhD Conference on Ancient History) (Hungarian Polis
Studies 21) (Budapest), 189–208.
 CHAPTER 6

OBSERVATIONS ON AGE-ROUNDING FOR SOLDIERS

FROM THE LOWER DANUBIAN PROVINCES

Valentin PIFTOR

Abstract
The aim of my paper is to research the age-rounding process of soldiers from the
Roman provinces of Moesia Inferior and Dacia during the 1st–3rd centuries AD. To
this end, I will use Whipple’s index and I will compare the values of the military popu-
lation (comprising soldiers and veterans) and those of the civilian population. I will try
to provide a series of explanations for the lower degree of age-rounding among
deceased soldier compared with civilians.

The age-rounding process has been a constant of societies since time immemo-
rial, until the emergence and mandatory character of civil status documents.
Roman antiquity is no exception to this rule.
I will follow the age-rounding process for samples of soldiers and veterans
within the provinces of Dacia and Moesia Inferior, both compared with the
male civilian population and between the two provinces.1
In order to calculate age-rounding, I will apply Whipple’s index, also used
by Duncan-Jones in his study Age-rounding, Illiteracy and Social Differentia-
tion in the Roman Empire.2 It applies to the interval 23–62 years old, eliminat-
ing young ages, better known by the parents, and older ages, less known. This
interval is divided into four decades. The index is calculated as follows: out of
the percentage of ages ending in a multiple of 5 within a decade, we subtract
20 and multiply by 1.25. If there are only ages ending in a multiple of 5, the
result is 100, and if they account for 20%, then the result is 0.

1
For Moesia Inferior, I used the information within my doctoral thesis, Piftor 2012 (plus ISM
IV), while for Dacia I used the database within the work penned by Mihailescu-Bîrliba 2004,
updated using the information within the Romanian Epigraphic Chronicle until 2011.
2
Duncan-Jones 1977.
84 V. PIFTOR

Walter Scheidel3 points out that Whipple’s index starts from two false pre-
sumptions. The first is that the number of ages within all decades must be
equal, which is not even valid for contemporary societies. The second is that
within each decade, the division by digits must be equal (each figure must
represent 10% of the sample by decade). For the first decade, he believes there
should be an equal number of persons in each series, birth rate should be con-
stant (in pre-industrial societies, epidemics, famine and wars make birth rate
fluctuate quite a lot) and life expectancy should be quite high, given that this
situation is seldom encountered even in contemporary societies. As to the sec-
ond presumption, it is hard to believe that the same number of persons die age
23 and age 32, and this situation is also valid for the other decade. Thus, if the
life expectancy of a sample is between 20 and 30 years old, most of the sample
would end up not being included when calculating Whipple’s index. Hence,
Scheidel’s idea of extending the sample from 23 to 62 years old to a sample
from 10 to 69 years old, thus including a greater number of persons. Scheidel
proposed a model where each figure within each decade is calculated. The
samples in Dacia and in Moesia Inferior are rather small in order for us to
calculate the value of each figure within the decade.
In his study, Duncan-Jones used around 40,000 ages at death on the funer-
ary inscriptions within the Latin-speaking areas of the Roman empire. He cal-
culates age-rounding by social class, by the sex of a person, by the discovery
region and by the province where it was discovered. In the study about digit
preference, Scheidel only studied the Roman Egypt; however, it has a sample
of 2136 persons for the category 10 to 69 years old, a sample hard to find in
any area of the empire except perhaps for Rome. I will calculate Whipple’s
index for the sample proposed by Duncan-Jones, but not for the extended one
proposed by Scheidel, because the samples of soldiers within the two prov-
inces start from the age of 19.
First of all, I analyse comparatively the samples of the male populations in
the two provinces. Moesia Inferior has 285 individuals and Dacia 321. I will
use four categories: 0 – ages ending in 0, 5 – ages ending in 5; another digit
– ages ending in any other digit except for 0 and 5; and precise – ages ending
in any digit stating the months and/or the days.

3
Scheidel 1996, 54.
 OBSERVATIONS ON AGE-ROUNDING FOR SOLDIERS 85

There is certain symmetry of age percentages ending in 0 and of precise ages,

the difference being only a greater number of persons with an age ending in 5
in Dacia, and in another digit in Moesia Inferior, as seen in the charts above.
If we eliminate from the sample the ages up to 19, the following situation
emerges:

In Dacia, it may be observed that the preponderance of precise ages has rather
high percentages in young ages, unlike in Moesia Inferior; also in Dacia, it
may be observed that 30% of the sample has precise ages and other digits,
while in Moesia Inferior the percentage is higher: 35%.
86 V. PIFTOR

Upon comparing the male civilian population over 19 (above) within Moesia
Inferior with the military one (soldiers and veterans), it may be observed that
a higher percentage of ages ends in 0 among the civilian population than in
case of soldiers. They have a higher percentage of ages ending in other digits,
the precise ages are equal in percentages, and for ages ending in 5, the percent-
age is close. Therefore, it may be pointed out that the dedicators of soldier
inscriptions in Moesia Inferior tend to mention a more precise age of the
deceased or they know it better.

In the case of Dacia it is worth noting a better representation of precise ages in

case of soldiers than of civilians. Civilians have more ages ending in 0 and
fewer ages ending in another digit. The soldiers have a slightly higher percent-
age among ages ending in 5. Therefore, in case of Dacia, I point out a higher
rounding trend among the civilian population than in case of soldiers.
The ages mentioned precisely within the sample of the male population over
18 in Moesia Inferior are as follows: in the military setting, […] Aemilius,
librarius legati in the legio V Macedonica, lived 18 years and 8 months,4 […]
Claudianus, a former speculator, died at 44 years, 5 months and 10 days,5
Caius Veturius Verus, a soldier in the legio V Macedonica, died at 19 years,
8 months and 24 days,6 Valerius Septimius, a veteran, lived 63 years, 6 months
and 16 days,7 Gaius Iulius Aemilianus, veteran in the legio VII Claudia, lived
60 years, 10 months and 4 days.8 In the civilian setting, Theocritos, the son of
Theocritos, a peregrinus, lived 22 years, 9 months,9 an anonymous man lived

4
ISM II, 184.
5
ISM II, 211.
6
Epigraphica, p. 205.
7
Epigraphica, p. 217.
8
CIL III, 7421.
9
ISM II, 186.
 OBSERVATIONS ON AGE-ROUNDING FOR SOLDIERS 87

23 years and 11 months,10 an anonymous citizen died at 46 years, 7 months

and 16 days,11 Aurelius Hermes, a citizen, died at 70 years and 11 days,12
Apriorianus, a citizen, died at 19 years, 2 months and 6 days.13

Table 1: Moesia Inferior,

deceased with very precise ages and their dedicators.

Legal/occupational Name Legal/occupational Kinship

Name
status of dedicators status degree
[…] Aemilius Librarius legati leg. Aemilius Citizen ? The father?
V Macedonica
[…] Claudianus Ex speculatoribus – – –
Caius Veturius Soldier of the leg. V Caius Veturius Citizen The father
Verus Macedonica
Valerius Septimius Veteran Aelia Domna Female citizen The wife
Valerius Valerianus Citizen The son
Valerius Vitalis Citizen The son
Valerius Martialis Citizen The son
Gaius Iulius Veteran leg. VII Aemilius Vitalis A soldier in a prae- The son
Aemilianus Claudia torian cohort
Dudus Citizen The brother
Aemilius Victor Citizen The son
Aemilius Iulius Citizen The son
Aemilius Aelius Citizen The son
Theocritos, the son Peregrinus Rufina Iasonos Female The mother
of Theocritos peregrinus
Anonymous – The parents – –
Anonymous Citizen Antonia Severa Female citizen The wife
Titius Marcianus Citizen The son
Titia Marciola Female citizen The daughter
Aurelius Hermes Citizen (paganus) 6 sons Citizens The sons
and a daughter
Apriorianus Citizen Aurelius Aprio Decurio coloniae The father

10
ISM II, 305.
11
ISM II, 349.
12
ISM V, 42.
13
ILB 82.
88 V. PIFTOR

Whereas in case of Claudianus the dedicators were not preserved, for the two
soldiers who died young, the dedicators are their fathers, who are citizens, and
in the case of the two veterans, the dedicators are the family, the husbands or the
brother along with the children. With the civilian population, the dedicators are
still their parents or their children. Upon studying their legal status, it is worth
noting that, except for Rufina Iasonis, the female peregrinus who set up the
inscription for her son who died rather young, the rest of the dedicators are citi-
zens; there is a citizen with a magistrate and a praetorian soldier function. This
stands to show that, among citizens, there was a trend of stating as accurately as
possible the age of the deceased or of knowing ages as well as possible.
The highly precise ages from the sample of the male population in Dacia
over 19 are as follows: in the military setting, Aelius Bassus, veteran of the
legio V Macedonica, lived 66 years, 10 months and 16 days,14 Marcus Aurelius
Antoninus, librarius in the legio XIII Gemina, lived 22 years, 11 months and
2 days,15 Aurelius Fabius, signifer alae Siliana, lived 29 years and 11 months16.
In the civilian setting, an anonymous person lived 32 years and 2 months,17
Marcus Valerianus, a citizen, lived 60 years and one month,18 Aurelius Aulucmus,
former peregrinus, currently citizen, lived 35 years, 6 months and one day.19

Table 2: Dacia,
deceased with very precise ages and their dedicators.

Legal/occupa- Name Legal/occupa- Kinship

Name
tional status of dedicators tional status degree
Aelius Bassus Veteran leg. V Aelius Helpizon Freedman
Macedonica
Marcus Aurelius librarius leg. XIII Aurelius Marcianus Citizen The father
Antoninus Gemina
Valeria Valentina Female citizen The mother
Aurelius Fabius signifer alae Siliana Aurelius Reburus Veteran The father
Fabia Female citizen? The mother
Anonymous – The wife? – The wife
Marcus Valerianus Citizen Menonianus Slave?

14
IDR II, 40.
15
IDR III/3, 344.
16
CIL III, 847.
17
IDR III/1, 18.
18
IDR III/3, 369.
19
CIL III, 918.
 OBSERVATIONS ON AGE-ROUNDING FOR SOLDIERS 89

Legal/occupa- Name Legal/occupa- Kinship

Name
tional status of dedicators tional status degree
Aurelius Aulucmus former peregrinus, Aurelius Basianus Citizen The father
currently citizen
Aurelia Dasius Citizen The mother

The inscriptions for the two soldiers were set up by the parents, while that of
the veteran by a freedman (perhaps he no longer had a family). There is also
a man featured in a fragmentary inscription that seems to have been set up by
his wife. As to the two other citizens, the inscription was probably set up by
a slave or by the parents. Concerning Dacia, whereas there is no deceased
peregrinus, the dedicators pertain to the close family or to the domus, the slave
or the freedman. In Dacia, there is also a preponderance of precise ages also
related to the citizens.
In what regards the differences between the soldiers within the legions and
those within auxiliary troops, in Moesia Inferior the following situation
emerges: there are 58 soldiers or veterans from legions and only 26 soldiers or
veterans from auxiliary troops.

It is worth highlighting that dedicators were more concerned with the age of
the deceased in case of legionary soldiers, which leads to a higher percentage
of ages ending in another digit, unlike the auxiliary troops, where the ages
ending in 0 are dominant. While there are fewer ages ending in 5 compared
with the legions, the percentage of those ending in 0 and 5 is much higher than
for legions.
In Dacia, the situation is quite different. There are only 35 soldiers and vet-
erans from legions and 47 soldiers and veterans from auxiliary troops.
90 V. PIFTOR

In Dacia, the situation is far more balanced, as shown above. There are more
legionary soldiers with precise ages, but also with an age ending in 5. Soldiers
from auxiliary troops record fewer ages ending in 0 and 5 – 61% than legion-
ary soldiers – 69%.
By applying Whipple’s index to the male population in Moesia Inferior, the
following values emerge:

Table 3: Moesia Inferior,

Whipple’s index for the male population.

Ages ending in Whipple’s

Age groups Total
a multiple of 5 index
23-32 23 46 37.5
33-42 26 33 73.48
43-52 37 47 73.40
53-62 31 42 67.26
23-62 117 168 62.05
Mean of the four decades 62.91

Whereas the very low value of the group 23–32 years old is understandable
because the memory is fresher for younger persons, the very high value of the
group 33–42 is quite surprising; this value is lower than the one of the follow-
ing group. The last decade has a slightly lower value, probably due to the high
number of soldiers and veterans within this decade.
 OBSERVATIONS ON AGE-ROUNDING FOR SOLDIERS 91

Table 4: Moesia Inferior,

Whipple’s index for civilians.

Ages ending in Whipple’s

Age groups Total
a multiple of 5 index
23-32 23 46 37.5
33-42 26 33 73.48
43-52 37 47 73.40
53-62 31 42 67.26
23-62 117 168 62.05
Mean of the four decades 62.91

Concerning the sample of soldiers, I note the lower values of the index, but
also an increasing trend with each decade. Hence, there is a higher accuracy of
ages in case of the deceased soldiers.

Table 5: Moesia Inferior,

Whipple’s index for soldiers.

Ages ending in Whipple’s

Age groups Total
a multiple of 5 index
23-32 23 46 37.5
33-42 26 33 73.48
43-52 37 47 73.40
53-62 31 42 67.26
23-62 117 168 62.05
Mean of the four decades 62.91

As for the civilians, the fluctuation noted for the entire male population in
Moesia Inferior is more present. The second decade has a significantly higher
value than the third, while the fourth decade has a lower value than the second
and the third decades. Civilians feature much higher values than the sample
comprising soldiers.
In case of Dacia, the following situation emerges:
92 V. PIFTOR

Table 6: Dacia,
Whipple’s index for the male population.

Ages ending in Whipple’s

Age groups Total
a multiple of 5 index
23-32 23 46 37.5
33-42 26 33 73.48
43-52 37 47 73.40
53-62 31 42 67.26
23-62 117 168 62.05
Mean of the four decades 62.91

At the level of the entire male population, it is worth outlining that the first
decade has a much higher value than the one in Moesia Inferior; it is actually
higher than the second decade of the sample. This may also be explained by
a higher preference for the digit 5 in the Dacia sample than in the Moesia Infe-
rior sample. Overall, however, the values of the group 23–62 and the mean of
the four decades are lower than those in Moesia Inferior. At the level of the
entire sample, the rounding trend is lower.

Table 7: Dacia,
Whipple’s index for civilians.

Ages ending in Whipple’s

Age groups Total
a multiple of 5 index
23-32 23 46 37.5
33-42 26 33 73.48
43-52 37 47 73.40
53-62 31 42 67.26
23-62 117 168 62.05
Mean of the four decades 62.91

The values for soldiers are higher than those in Moesia Inferior, but they still
feature an increasing trend across the four decades.
 OBSERVATIONS ON AGE-ROUNDING FOR SOLDIERS 93

Table 8: Dacia,
Whipple’s index for soldiers.

Ages ending in Whipple’s

Age groups Total
a multiple of 5 index
23-32 23 46 37.5
33-42 26 33 73.48
43-52 37 47 73.40
53-62 31 42 67.26
23-62 117 168 62.05
Mean of the four decades 62.91

The values of the civilian population are equal for the two first decades and
they are both very high, much higher than those of soldiers. In the fourth
decade, however, the value is lower than in case of soldiers. The values of
Whipple’s index for soldiers and for the civilian population in Dacia are
greater than in Moesia Inferior.
Consequently, there is a lower age-rounding trend for deceased soldiers than
for deceased civilians. The highly precise ages are especially related to the
citizens through dedicators, while deceased soldiers represent quite an impor-
tant part for the sample of males over 19.
The higher age-rounding percentages for the military setting in Dacia com-
pared with Moesia Inferior can be explained using the statement made by
Lucrețiu Mihailescu-Bîrliba20 in his study concerning the mortality of legion-
ary soldiers in Moesia Inferior: the presence of several legions for a longer
period in Moesia Inferior, as well as the fact that many legionary veterans
settled in the province where they had activated.
This age-rounding process may also be related to literacy, as I have previ-
ously shown.21 The citizens with magistrate functions in Moesia Superior (this
province comprises a greater number of ex-magistrate citizens with funerary
inscriptions mentioning the age) have the lowest values, followed by deceased
soldiers, then by citizens and finally by the peregrini or the servile population.
Another explanation for the deceased soldiers may be related to the strict
record of the enrolment period by the army; dedicators could use these data to
approximate the age, as Mihailescu-Bîrliba pointed out.

20
Mihailescu-Bîrliba 2014.
21
Piftor 2012; 2013.
94 V. PIFTOR

Another hypothesis related to the high number of ages ending in 0 or 5 may

be that ages were seen as a threshold.22 Hence, the age of 60 can be related to
the candidacy for a magistrate function (at 60, Romans were no longer allowed
to be candidates); 25 was the lower limit for the same office. The correlation
between certain round ages and several life stages in the Roman world is also
argued by Valerie Hope in Remembering to Mourn.23
In order to be able to choose one of the aforementioned hypotheses,
a more thorough research concerning the dedicators of these funerary
inscriptions is necessary, the purpose of which would be to get a better
insight into the age-rounding phenomenon from the perspective of those
who set up the inscriptions.

BIBLIOGRAPHY

Duncan-Jones, R.P. 1977: ‘Age-rounding, Illiteracy and Social Differentiation in the

Roman Empire’. Chiron 7, 333–53.
Hope, V.M. 2011: ‘Remembering to Mourn: Personal Mementos of the Dead in
Ancient Rome’. In Hope, V.M. and Huskinson, J. (eds.), Memory and Mourning:
Studies on Roman Death (Oxford/Oakville, CT), 176–95.
Mihailescu-Bîrliba, L. 2004: Individu et société en Dacie romaine. Études de démo-
graphie historique (Wiesbaden).
—. 2014: ‘La mortalité des légionnaires en Mésie Inférieure’. Studia antiqua et
archaeologica 20, 171–83.
Piftor, V. 2012: Structuri demografice din Moesia Inferior şi Moesia Superior (secolele
I–III p.Chr.) (Dissertation, Iași).
—. 2013: ‘Age-rounding and social status in Moesia Inferior’. Studia antiqua et
archaeologica 21.1, 87–114.
Scheidel, W. 1996: Measuring Sex, Age and Death in the Roman Empire:
Explorations in Ancient Demography (Journal of Roman Archaeology Suppl. 21)
(Ann Arbor).

22
Ioan Piso pointed out this hypothesis to me, and I use this occasion to express my
gratitude.
23
Hope 2011.
 CHAPTER 7

TERRITORIA AND REGIONES

IN THE LOWER DANUBIAN PROVINCES

Florian MATEI-POPESCU

Abstract
The author discusses the problem of territoria and regiones in the provinces of Dacia
and Moesia Inferior. Without entirely excluding the possibility of synonymy between
regio and territorium, especially in the case of records from Rome, any pattern is hard
to establish in this state of research. The texts have to be interpreted as the particular
situation indicates.

In 2008, an inscribed funerary stele was found close to the village of Dülük
(10 km north of the city of Gaziantep), Turkey, where in antiquity lay the city
of Doliche, the origin of the Iupiter Optimus Maximus Dolichenus. The stele
is divided into two registers, the upper register with the image of the deceased,
and the inscription field in the lower register. It was published in 2011 by
Margheritta Facella and Michael Speidel.1 Here is the inscription text, as
established by the two editors:
D. M. / secur(itati) (a)e<t>e(rnae) (?) / et [quie]ti / + A[eliu]s / Vitalis / ex
provinc/ia Dacia sup(eriore) / ter(r)i(torio) Bassia/na(e) vixit an/nis XXIII /
Aur(elius) Valeri/anus dupla(rius) leg(ionis) / XIII c(o)ho(rtis) III frat(ri) / caris-
simo / [f(aciendum) vel p(onendum )] c(uravit)
Therefore, we are dealing with a funerary stele erected through the care of
Aurelius Valerianus, duplarius legionis XIII Geminae, a legionary stationed at
Apulum, in Dacia Superior (Apulensis),2 to his comrade Aelius Vitalis, obvi-
ously still from the legio XIII Gemina, even though this is not explicitly men-
tioned in the inscription. Based on internal analysis (iconography of the
deceased; duplarius is a rank introduced during the period of the Severan

1
Facella and Speidel 2011.
2
Moga 1985, passim.
96 F. MATEI-POPESCU

dynasty), the inscription may date to the 3rd century AD, thus the authors sup-
posed that we were dealing with a detachment of the legion involved in one of
the Eastern campaigns of Septimius Severus or Caracalla. Obviously, a later
dating, possibly during the Eastern campaign of Gordian III, cannot be
excluded.
Still, the most important mention of the inscription is the place of origin of
Aelius Vitalis – ex Dacia superiore, territorio Bassiana(e). The authors chose
a complex solution to a rather simple problem. Since in Pannonia two places
named Bassiana are recorded, one in Pannonia Superior (on the road between
Savaria and Arrabona), and the other in Pannonia Inferior (on the road between
Sirmium and Singidunum), it was argued that Aelius Vitalis came from one of
the two towns in Pannonia, while the mention ex Dacia superiore refers to the
place from where the legionary detachment came to the north of Syria.3 Nev-
ertheless, I believe that in this case, the simplest solution is also correct: ter-
ritorium Bassianae should have been located somewhere in Dacia Superior
and not in Pannonia Superior or Inferior. The name is common, and two
records in a single province prove it was widespread. The fact that this com-
munity was unrecorded in Dacia Superior until then, i.e. there was no mention
in any of the literary or cartographic sources for Dacia, is not surprising at all:
not all settlements in the province are recorded by such sources (for example,
Ad Vatabos/Batavos,4 Ranisstorum5 or Darnithithi6). Hence, this inscription
bears a special importance for the history of the rural communities from
Roman Dacia. Sometime in the 3rd century in Dacia Superior there was a ter-
ritorium Bassianae, an autonomous community, with the centre at Bassiana. In
order to understand better the character and legal status of this community, we

3
Facella and Speidel 2011, 210: ‘Unless we are dealing with an outright mistake or misun-
derstanding by the stonecutter, the reference to Dacia Superior was surely intended to mean that
Vitalis had come from Dacia Superior to northern Syria as a soldier of the same expeditionary
army in which his comrade Valerianus served.’
4
AE 1996, 1540 = IDRE II, 411: natus in Dacia ad Vatabos. The inscription was discovered
at Apamea, in Syria. This is likely to be the fort of ala I Batavorum milliaria from Războieni,
Dacia Superior, although one may not exclude the variant of the stationing location of cohors I
Batavorum milliaria from Dacia Porolissensis. See Ad Pannonios, Teregova, recorded by carto-
graphic sources (Tabula Peutingeriana 7. 4; Ravenna Cosmography 4. 14), evidently the station-
ing location of an auxiliary troop composed of Pannonii (referencing the two provinces, and not
the Pannoni – meaning Pannonii – Pannonius, Pannonii).
5
AE 1969–70, 583 = IDRE II, 362. The inscription was discovered at Grammeni, in the prov-
ince of Macedonia and mentions the place where Ti. Claudius Maximus brought emperor Trajan
the head of Decebalus, obviously somewhere in Dacia: quod cepisset Decebalu(m) et caput eius
pertulisset ei (scil. Traiano) Ranisstoro.
6
IDR I, Dipl I. This is the copy of an imperial constitutio awarded by Trajan when in Dacia,
to the members of cohors I Brittonum millliaria Ulpia torquata p. f. c. R. The place is rendered
in the Locative, Darnithithi.
 TERRITORIA AND REGIONES IN THE LOWER DANUBIAN PROVINCES 97

must review the territoria records in Dacia and the area of the Lower Danu-
bian provinces. In addition, we should examine the relations of these territoria
with other types of rural communities of larger spread, for instance regiones.
Before going further, a short comment on Dacia Superior in the 3rd century
AD. It seems that the usual name, starting in AD 168, was actually Dacia Apu-
lensis.7 Nevertheless, in the military milieu the name of Dacia Superior contin-
ued to be in use, also in official documents like the military diploma from
Drobeta, copied after an imperial constitution granted to the auxiliary units
from Dacia Superior on 1 April AD 179 (et sunt in Dacia superiore sub Helvio
Pertinace leg.).8 This inscription is the second example. If for the first docu-
ment we take into consideration that at the moment of recruitment, in around
AD 154, the soldiers were about to serve in the units of Dacia Superior, this
second document it is to be dated without hesitation to the 3rd century AD. It
is a proof that the old name of the province was in use. In fact, as already
highlighted, M. Claudius Fronto’s honorary inscription from Rome is the only
one where Dacia Apulensis appears as an area of command for a legatus
Auguti pro praetore (from that moment onwards only legati Augusti pro prae-
tore trium Daciarum appear), the other inscriptions being related only with the
procuratores provinciae Daciae Apulensis.9
Two other territoria are epigraphically recorded in Dacia: territorium
Arcobarense,10 near the fort of ala I Tungrorum Frontoniana from Ilişua,
and territorium Sucidavense.11 Both records may be dated to the mid-3rd cen-
tury (AD 246, the exact date for territorium Arcobarense; possibly, the
joint reign of Philip the Arab and Philip Caesar, given that the inscription
is dedicated pro salute Augustorum). If territorium Arcobarense was regarded
as a rural autonomous territory of the military vicus at Ilişua,12 territorium

7
For the reorganisation of the Dacian provinces in the time of Marcus Aurelius, see Petolescu
1987.
8
RMD II, 123.
9
CIL VI, 1377 = 31640 = ILS 1098 = IDRE I, 10; see also CIL III, 1457 = ILS 1097 = IDR
III/2, 90; Macrea 1966, 146; Petolescu 1987, 126.
10
Nemeti and Bărbulescu 2006–07: [Geni]o terri/[t(orii) A]rcoba(rensis) / Aur(elius)
Sal(vius) / [et Ae]l(ius) No/[mu]s magistri (duo) / [Pude]nte et Ael/[iano] co(n)s(ulibus). See
also Nemeti and Bărbulescu 2010. The reading of the last row was improved by Piso 2008, com-
pleting the name of the 246 pair of consuls: [Praes]ente et Al/[bin(o) cos.] – C. Bruttius Praesens
and C. All(?) Albinus. The name of the civil settlement from Ilişua should have been Arcobara
and not Arcobadara, The main argument is the settlement Zourobara which appears also on the
list of the poleis from Dacia, -bara should have been the Daco-Moesian equivalent for -para,
Dana and Nemeti 2012. The first part of the name ark- should be found in toponyms like Arci-
dava, ISM I 358, in the nearby of the Greek city of Histria.
11
IDR II, 190: [D]eae Nemesi, / pro salute Augg(ustorum), / curial(es) territ(orii)
Σuc(idavensis) / [te]mplum a solo / restituerunt.
12
Nemeti and Bărbulescu 2010, 448.
98 F. MATEI-POPESCU

Sucidavense was understood as the territory of a yet unrecorded municipium

Sucidavense.13
Beside these records of territoria, in Dacia are also recorded two regiones:
regio Ansamensium, close to Samum (Căşei), under the supervision of benefi-
ciarii consularis,14 and regio Scodrihe(n)sis, geographically unlocated.15
As mentioned, the relation of territorium to regio is still not entirely clear in
the specialty literature, the tendency being to consider the two terms rather
synonyms.16 I believe, however, that the two administrative terms are not syn-
onyms and equivalents, at least not in every case. Each record should be
approached separately. Circumstances found in the Moesian provinces show
that the issue is far from being resolved.
A significant number of territoria of certain communities without municipal
status are recorded in Moesia Inferior,17 without taking into account the terri-
toria of the Greek cities located on the western coast of the Black Sea (see
below): territorium Troesmense,18 territorium Capidavense,19 territorium
Montanense,20 territorium Dianensium (unknown identification and localisa-
tion; it is excluded that the name could be close to that of Dimum),21 territo-
rium Abritanorum,22 territorium civitatis Ausdecensium,23 territorium Aegys-
sense24 and most likely a territorium at Barboşi (Şendreni).25
Circumstances at Troesmis are the most clear, since the duality canabae –
civil community – is recorded all over the Danubian provinces. The canabae

13
Petolescu 2011, no. 13.
14
CIL III, 822; CIL III, 827 = 7633; AE 1957, 326. See also Ott 1995, 106–07. A vicus
Samum and a regio Ansamensium are recorded.
15
CIL VI, 2698 = Nemeti and Bărbulescu 2006–07, 114–15, no. 8: D. M. / Aur(elius) Passar
/ mil(es) coh(ortis) VIIII / pr(aeroriae) (centuria) Hilariani / vix(it) an(nis) / XXX mil(itavit)
an(nis) VIIII domo / Daciae regione Scodri/he(n)se Tullius Lupus / et Masculinus Ter/tullinus et
Claudius / Longinus et Aur(elius) / Longius commani/puli b(ene)m(e)r(enti) f(e)c(e)r(unt).
I believe that the inscription may be dated, alike many of the similar inscriptions of the praetori-
ans, rather in the first half of the 3rd century AD. Under such circumstances, it definitely refer-
ences the province of Dacia from north of the Danube.
16
See the discussion in Nemeti and Bărbulescu 2006–07, 112–13, where the equivalence
regio – territorium is supported. See also Mócsy 1992, 133–34.
17
See a preliminary discussion in Matei-Popescu 2010b.
18
ISM V, 135 = AE 1980, 818. See also Vulpe 1976, 290–91.
19
CIL III, 12491 = ILS 7181 = ISM V, 77.
20
AE 1987, 811: ag(ens) t(erritorii) M(ontanensium).
21
ILB 223 (198–209).
22
AE 1985, 765: Imp(eratori) Caes(ari) / M. Iulio / Philippo / Pio Fil(ici) (!) invi(cto) /
Aug(usto) et M. / Iulio Philipo (!) nob(ilissimo) Caesa/res (!) per / ter(ritorium) Abri(tanorum) /
m(ille) p(assuum) I.
23
CIL III, 144372 = CIL III, 14438 = IDRE II, 338 = AE 1957, 333.
24
AE 2004, 1281.
25
ISM V, 296: q(uin)q(uennalis) … ex de(creto) or(dinis).
 TERRITORIA AND REGIONES IN THE LOWER DANUBIAN PROVINCES 99

lay nearby fortresses, while the civil communities (vici, civitates) developed
at a distance of 1.5–2.5 km from the fortification, hence extra leugam.26
Thus, both the canabae of the legio V Macedonica and a civil community27
are recorded. Therefore, territorium Troesmense refers without a shadow of
a doubt to the civil community and not the canabae legionis V Macedonicae.
The same situation is found at Durostorum, where beside the canabae Aeliae28
a civil settlement is recorded, which later became municipium Aurelium
Durostorum.29
Similarly clear is the situation territorium civitatis Ausdecensium, a rural
territory of a civitas inhabited by the Ausdecenses, a Thracian-origin popula-
tion. This territorium bordered the territory of a rural community of Dacians.30
Less clear though is the situation of the territories recorded around the for-
tifications of auxiliary troops (Montana, Abritus, Capidava, Aegyssus,
Barboşi). It is generally agreed that near auxiliary forts developed only vici
militares, in the immediate vicinity of the fortification.31 Still, how spread
might have been the territory of such a military vicus? The inscription record-
ing a quinquennalis territorii Capidavensium was found at Ulmetum.32
It would be hard to believe that the territory of cohors I Ubiorum33 or cohors I
Germanorum,34 which also comprised the military vicus, might have spread
that far to the east. At Abritus, under Philip the Arab, a territorium Abritano-
rum is recorded, yet an inscription found in the same area mentions
a prata publica, whose limits were traced by the prefect of a cohort stationed
at Abritus.35 Unfortunately, the inscription cannot be dated with accuracy;

26
Vittinghoff 1994, 89–105. For the distinction intra et extra leugam, see Piso 2005, 151–93.
At Aquincum there were for instance three communities: canbae, the civil vicus, turned muni-
cipium under Hadrian, and civitas Eraviscorum (Mócsy 1992, 136–38).
27
CIL III, 6167 = ISM V, 157: c(ives) R(omani) Tr[oesmi consist(entes)]; AE 1960, 337 =
ISM V, 158: q(uin)q(uennalis) [c]anab(ensium) et dec(urio) Troesm(ensium).
28
CIL III, 7474 = ILS 2475.
29
ISM I 302; Bujor 1960, 145–48, no. 2. See also Piso 2005, 172–73.
30
CIL III, 144372 = CIL III, 14438 = IDRE II, 338 = AE 1957, 333; Bărbulescu 2001, 193,
215–16.
31
Vittinghoff 1994, 126: ‘Parallel verläuft die Entwicklung bei einigen Auxiliarkastellen. Da
ihre Besatzungssollstärke nur 500–1000 Mann betrug, finden wir bei ihnen nicht zwei getrennte
Siedlungen, sondern in der nächsten Nachbarschaft vor den Toren der Festung entsprechend den
canabae der Legionen einen Kastellvicus.’
32
CIL III, 12491 = ILS 7181 = ISM V, 77, reused in the gate of the Late Roman fort: loci
princeps et quinquennalis territorii Capidavensis.
33
Matei-Popescu 2010a, 235–36, no. 42.
34
Matei-Popescu 2010a, 213–15, no. 28.
35
CIL III, 13726: Apronius / Maximus / trib(unus) coh(ortis) / prat(a) pub(lica) / circu[m]
c[l]usi[t]. See complete discussion in Vittinghoff 1994, 128, 134, quoting an example from His-
pania of the territory limits of cohors IIII Gallorum: ex auctoritate Ti. Claudi Caesaris Augusti
Germanici imperatoris terminus pratorum cohortis IIII Gallorum inter cohortem et civitatem
100 F. MATEI-POPESCU

however, there is a record of a territory under the use of the cohort, a prata
publica.36 The civil settlement at Abritus might have been called castellum
Abritanorum, as shown on two inscriptions.37 It is likely that this community
and not the military vicus was in possession of that territorium Abritanorum
recorded under Philip the Arab. The situation at Montana is more complex,
since a regio Montanensium38 is also recorded in the area, an imperial estate
under the surveillance of centuriones regionarii and beneficiarii consularis.
Anyhow, by the mid-3rd century AD at Montana both castrenses, most likely
soldiers in garrison, as well as cives Montanenses are recorded.39 This record
might suggest a dual military vicus–civil community. Within this context might
be mentioned the example provided by an inscription from Sexaginta Prista of
around AD 100 which records cives Roman[i consistentes] Sexsaginta (sic!)
Pri[st(is)].40 It is hard to state though whether this is the community of Roman
citizens from the military vicus, developed near the fort of either cohors II Mat-
tiacorum or II Flavia Brittonum41 (still, both units are recorded at Sexaginta
Prista by mid-2nd century – II Mattiacorum in AD 144, and II Flavia Britto-
num after AD 155. At the moment when the inscription was placed, Sexaginta
Prista might still theoretically have been the stationing location of classis Fla-
via Moesica, moved by Trajan at Noviodunum, most likely in AD 103–10542)
– or was it an autonomous civil settlement? The use of the name Sexaginta
Prista (the name obviously references the period when classis Flavia Moesica
was stationed there) and the omission of the veterans might suggest a civil set-
tlement, different from the military vicus; one cannot be certain.43

Beduniensium. Stationed at Abritus until around AD 136, when it was transferred to Thracia
at Cabyle, was cohors II Lucensium equitata (Matei-Popescu 2010a, 219–20, no. 31). Being
a cohors quingenaria is was led by a praefectus cohortis, but it could have been led in certain
circumstances by a tribunus cohortis. It cannot be excluded that after 136 another cohort, this
time milliaria, was stationed at Abritus.
36
Ulpian Dig. 50. 16. 31: pratum est, in quo ad fructum percipiendum falce dumtaxat opus
est (Vittinghoff 1994, 134).
37
CIL V, 942 = ILS 2670; AE 1957, 97. See complete discussion in Matei-Popescu 2010b.
See also Mócsy 1992, 133: ‘Castella, turres usw. können auch Bennenungen nichtmilitärische
Orte gewessen sein’.
38
CIL III, 12385; AE 1969–70, 577; AE 1957, 342. Cîrjan 2010, 90–92.
39
CIL III, 12376, inscription dated to AD 256: [burgum constitui iussit] un[de latrunculos]
observare[nt pro]pter tutela[m ca]stre(n)sium et [ci]vium Montanensium; Vulpe 1976, 294–96.
40
Velkov 1965; AE 1966, 356.
41
Matei-Popescu 2010a, 198–99, no. 18 (II Flavia Brittonum); 222–24, no. 33 (II Mattiaco-
rum). Other two cohorts whose names did not preserve are recorded at Sexaginta Prista in AD 76
(AE 1957, 307).
42
Matei-Popescu 2010a, 246.
43
See also the examples from Pannonia, where at Matrica an inscription was placed
pro salute civeromanorum (sic!) territ(orii) Matricensium in AD 267–68 (AE 1980, 712),
and at Vetus Salina by an altar: [Signu et ar]am Genii civibus R. [et consistentibus terri]torii
 TERRITORIA AND REGIONES IN THE LOWER DANUBIAN PROVINCES 101

On the territories of Aegyssus (reading is yet uncertain: dec(urio) terri(torii)

A[eg(ysensis)]44) and Barboşi we have no information to discuss, the same
applying to territorium Dianensium (it was not located geographically either).
Here we should insert the example provided by the centre at Noviodunum.
From the period between Trajan’s two Dacian expeditions until the end of the
3rd century AD classis Flavia Moesica45 was stationed there. Concurrently,
a municipium Noviodunense is recorded, without being able to specify the
moment when the municipal status was granted, most likely by early 3rd cen-
tury AD. The latter is recorded by an inscription found at Dinogetia, obviously
reused in the erection of the fortification of Romano-Byzantine period.46 But
which of the communities at Noviodunum became a municipium, since two
seem to be recorded? Thus, a vicus Nov(iodunum) is mentioned by an inscrip-
tion found most likely at Noviodunum, dating to AD 178.47 In parallel, several
inscriptions reused in the late fortification at Halmyris evidence a vicus clas-
sicorum, a sailors’ village, where cives Romani consistentes are recorded.
Most likely, the inscriptions were in fact brought from nearby Noviodunum,
the stationing location of classis Flavia Moesica. The inscriptions are distrib-
uted chronologically over several decades from AD 136 until AD 200.48 Like
the legionary fortresses, vicus Noviodunum and not vicus classicorum was
awarded municipal status. We may agree on that year is a terminus post quem
for the grant of municipal status to the community at Noviodunum.49 Given
that, for instance, Troesmis became a municipium under the joint reign of
Marcus Aurelius and Commodus (AD 177–180), Noviodunum was most prob-
ably raised at a later date, under Commodus or even Septimius Severus.
The example provided by the circumstances at Noviodunum, together with
the partial information regarding other centres (Montana, Capidava, Abritus
and Sexaginta Prista), allow us to argue that the existence of two communities,
a military vicus and a civil community nearby a fortification are not the

Vetuss(alinensium), CIL III, 10305 = ILS 7126. Neither in this case was it established with preci-
sion whether these are military vici or civil communities, Mócsy 1992, 161–73. Similarly, this
could also be the case of the centres near Micia and Tibiscum. Pagus Miciensis had obviously
nothing to do with the military vicus near the fortification, and municipium Tibiscum did not
necessarily develop from the military vicus (Petolescu 2011, no. 4).
44
AE 2004, 1281.
45
Matei-Popescu 2010a, 245–55.
46
Barnea 1988; AE 1990, 867. Another inscription records a quinquennalis: ISM V, 268.
47
CIL III, 14448 = ISM V, 233. If the inscription were identified at Noviodunum, then the
issue is solved: we are dealing with a vicus Noviodunum and not vicus Novus.
48
Suceveanu and Zahariade 1986; AE 1988, 986–90.
49
In AD 176, a sailor was taken from Noviodunum into the fleet at Ravenna and discharged
in AD 202. The origins of C. Iulius Valentis f. Iulianus is entered simply: Novi(o)d(uno) ex
Moesia, AE 2001, 2161 = RGZM 45.
102 F. MATEI-POPESCU

necessarily specific to legionary centres, at least not as far as Moesia Inferior

is concerned. Evidently, circumstances in Moesia Inferior might be particular,
since there existed civitates (see for instance the record of a praefectus
civitatium Moesiae et Treballiae50) as early as the nearby Thracian kingdom,
at which military fortifications were later erected, but evidence from other
provinces (Pannonia and Dacia) may indicate that under certain circumstances
two independent communities could also develop with the fortification of an
auxiliary troop.
What is the relationship of territorium to regio, if the situation regio Mon-
tanensium–territorium Montanensium (theoretically, as seen above, territo-
rium Montanensium refers to the civil settlement and not the military vicus,
whilst regio Montanensium was explained as an organism directly under the
control of the Roman army, since the area was of special strategic impor-
tance51) is insufficiently clear? The record of a regio Histriae52 separate from
the territory of the town at Histria (fines Histrianorum in the Horothesia of
AD 10053), which theoretically spread further to the west, at least up to vicus
Ulmetum,54 allows us to exclude the possibility of an automatic equivalence.
It is possible that regionarii55 were also recorded in regio Histriae. In AD 241,
a regio Dime(n)sis from Moesia Inferior56 is recorded at Rome. Immediate
connection was made between this regio and Dimum (Belene), an important
post of portorium Illyrici et ripae Thraciae.57 The Horothesia of Histria58 also
record for year the canabae Dimensium, which would account for the station-
ing of a legion in this area, sometime in the 1st century AD. When the legion
left, the canabae continued to exist under the same organisational form, as

50
CIL VI, 1838, 1839 = ILS 1349.
51
Rankov 1983, passim and 58: ‘the term regio Montanensium regularly refers to an under
special military control and possibly even to a region of imperial estates; this excludes its use for
a municipal territory’. See also, Gerov 1988, 101–06.
52
CIL III, 12489 = ILS 7182 = ISM I, 373; ISM I, 329, 343; ISM V, 123, 124. The last two
inscriptions were found at Cius (Gârliciu) on the Danubian limes, reused in the enclosure erected
in AD 369. If the regio Histriae spread westwards to the Danube, or the building material was
brought inland from the Dobrudja is hard to say. For a complete discussion on regio Histriae, see
Avram 1982; Bărbulescu 2001, 34–35.
53
ISM I, 68, r. 2.
54
ISM V, 62: cives Romani et Bessi consistentes vico Ulmeto, from AD 140 and E. Doruţiu-
Boilă’s comment: ‘thus, it is not excluded that Ulmetum depended on one of the coastal Greek
cities, most likely Histria’. Still, vicus Ulmetum was most likely part of regio Histriae. For other
records of the toponym, see ISM V, 63, 64 and 69.
55
Sarnowski 1988, 101–02, no. 5.
56
CIL VI, 32549.
57
ILB 237, Memor Aug(ustorum) n(ostrorum trium) ser(vus) (contra)sc(criptor) stat(ionis)
Dim(ensis); Gerov 1988, 111–12.
58
ISM I, 67–68, ll. 71–72: a finibus canabarum Dimensium.
 TERRITORIA AND REGIONES IN THE LOWER DANUBIAN PROVINCES 103

they did not receive the municipal status. Unfortunately, the history of the
canabae from Dimum can no longer be traced in the 2nd–3rd centuries AD;
information is lacking. It might be possible that this community had also had
a regio, recorded in the 3rd century AD. Circumstances regarding regio Nico-
politana, still confirmed in the 3rd century59 are complicated. Should one
understand that vicus Saprisara, recorded by this inscription in regio Nicopoli-
tana, directly depended on the authorities of the city at Nicopolis ad Istrum?
Probably not. Like Histria, we are dealing with an autonomous territory where
numerous rural settlements developed, among them this vicus. However, given
that the record is from the 3rd century AD and comes from Rome, like other
records of the type in the province of Thracia (see below), this is still city
territory.
Three regiones are recorded in Moesia Superior, regio Ratiarensis,60 regio
Viminacensis61 and r(egio) Aq(uensis).62 The first two records date to the
3rd century AD. Since in both cases they are praetorians recruited from these
areas, one would have expected a simple mention of their origin, either from
colonia Ulpia Ratiaria63 or municipium Aelium Viminacium (colonia begin-
ning with Gordian III64 – as the inscription is dated most likely to the mid-3rd
century AD, the canabae of the legio VII Claudia, still recorded under
Caracalla,65 had already been incorporated in the municipium/colonia66), if by
regiones one may understand the territoria of the relevant towns. An inscrip-
tion from Narona shows that Roman colonies could concede part of their ter-
ritory to certain rural communities, in this case with the status of pagus.67
This is so for the colony of Ulpia Traiana Sarmizegetusa in Dacia, which early
under Roman control possessed a large territory, gradually lost once with the
emergence of other urban communities out of rural ones.68 It is thus possible
that the regiones in question here were autonomous rural communities

59
CIL VI, 2933.
60
CIL VI, 2730.
61
CIL III, 195.
62
AE 1981, 724 = AE 1982, 841.
63
Mócsy 1970, 101–09. Like the case of 42 soldiers discharged at Viminacium in AD 195,
where origin is indicated as simply Rat(iaria): CIL III, 14507 = IMS II, 53.
64
Mócsy 1970, 145–58.
65
CIL III, 14509 = ILS 9105 = IMS 55: cana[bas refec]erunt leg(ionis) VII Cl(audiae)
[A]nt(oninianae).
66
This might have been recorded if respective praetorian had been born in the canabae of the
legion, as shown by the example of an eques singularis from Brigetio: natus in Pannonia inferi-
ore domo Brigetione at legione prima atiutri(cem): CIL VI, 32783; Mócsy 1992, 90.
67
AE 1950, 44 = ILJug 114: Divo Augusto et / Ti. Caesari Aug. f. Aug(usto) / sacrum /
veterani pagi Scunastic(i) / quibus colonia Naronit(arum) / agros dedit.
68
Piso 1995.
104 F. MATEI-POPESCU

developed on the territory of the colony of Ratiaria and municipium Vimina-

cium. Evidently, pagi and vici are to be sought in these regiones, which housed
communities of veterans and Roman citizens, but also likely peregrini. The
two praetorians recorded at Rome were raised from amongst them. The third
record concerns a vilicus, freedman of the emperor Severus Alexander, of sta-
tio Poetoviensis and r(egio) Aq(uensis) within portorium Illyrici et ripae Thra-
ciae, in AD 225.
In Thracia, a series of inscriptions dated under emperor Antoninus Pius
mention three regiones, all located within city boundaries: Serdica, Marciano-
polis and Augusta Traiana (per fines civitatis … regione …). The names of
two regiones are known: regio Dyptensium69 and regio Gelegetiorum.70 In this
case, since the inscription phrase seems quite clear, these are parts of city ter-
ritoria. This division of the territory in several regiones is likely a tradition
preserved as early as the Thracian kingdom.
Regarding the regiones from Dacia and the Moesias,71 two groups of records
may be established: one group composed of regio Ansamensium, regio Mon-
tanensium and regio Histriae, all recorded prior to the mid-3rd century AD
and placed under military Roman control (regionarii and beneficiarii consula-
ris) and a second group composed of regio Scodri(h)ensis, regio Viminacensis,
regio Ratiarensis, regio Dimensis and regio Nicopolitana, which emerge only
in the 3rd century AD, rather its middle, in the praetorian inscriptions from
Rome. It is possible in this case we are dealing with equivalence between ter-
ritorium and regio, under the circumstances of the gradual disappearance of
the importance of legal status differentiations of various communities. In the
late period (the process yet began by the mid-3rd century AD), all urban com-
munities, regardless of their status are named civitates, without distinction.72

69
AE 1957, 279 = ILB 211 = AE 2000, 1291: praesidia / et burgos ob tutelam provinc(iae) /
Thraciae fecit curante C. Gallonio / Fron(to)ne Q. M(a)rcio Turbone, leg. / A[u]g. pr. pr., per
fines civitatis / Serd(ic)ensium regione Dyptens(ium) / praesidia n. III, burgi n. XII, phruri /
n. CIX.
70
AE 2000, 1268: Turbone leg. A[ug. pr. pr.] / per fines civita[tis Mar]/cianopolitano[rum
re]/gion[e] Gelegetio[rum in propin]/quo phruri n. […].
71
Many more regiones are recorded in Thracia, but there too records might be divided into
those of the 2nd century AD (see above), when obviously regiones are part of territoria civita-
tium, and those of the 3rd century in the inscriptions of soldiers in units stationed in Rome,
mentioning for instance regio Serdica / Serdicensis (CIL VI, 2605 = ILS 2041 – the inscription
may be dated after the establishment of the province of Dacia south the Danube; CIL VI, 2638,
2742; CIL X, 1754 = ILS 2043), regio Marcianopolitanis (CIL VI, 32583) or regio Pautaliensis
(CIL VI, 32660), all the latter with obvious reference to the territory of the respective towns. On
the rather large number of vici that might have lain within the territory of a city, see another
inscription from Rome put up by praetorians recruited from Philippopolis: CIL VI, 32543 = ILS
2094.
72
Vittinghoff 1994, 210–12.
 TERRITORIA AND REGIONES IN THE LOWER DANUBIAN PROVINCES 105

In conclusion, without entirely excluding the possibility of a synonymy

between regio and territorium, especially in the case of the records from
Rome, we specify that any pattern is hard to establish in this state of research.
Returning to Dacia’s case, after having reviewed so many examples, one
may note that territorium Arcobarense does not necessarily refer to the mili-
tary vicus developed near the fort of ala I Tungrorum Frontoniana, rather to
the civil settlement nearby.73 Similarly, until new finds surface, the existence
of a territorium Sucidavense, where curiales appear, does not undeniably
prove that the community at Sucidava was municipalised. Thereafter, one may
conclude that the territorium Bassianae (the name is obviously of Roman ori-
gin and may add to the toponyms of the type: Blandiana, Optatiana or Largi-
ana) is most likely a rural community with its centre at Bassiana. The com-
munity never reached municipal status. Without doubt, both veterans and
Roman citizens in this territory resided (veterani et cives Romani consistentes),
among whom Aelius Vitalis was recruited to the legio XIII Gemina. Unfortu-
nately, there is no evidence on the possible location in Dacia Superior of this
settlement; however, it most probably lay not far from Apulum, headquarters
of the legio XIII Gemina. Obviously, one raises the question whether such ter-
ritoria had emerged early under the Roman control. Unfortunately, in this state
of research, the question remains unanswered, but we should note that all three
records date from the 3rd century AD. In the context of poor epigraphic evi-
dence on the rural communities in Dacia, the inscription from Doliche is an
extremely important document for understating the organisation of the rural
territory in this province.

BIBLIOGRAPHY

Bărbulescu, M. 2001: Viaţa rurală în Dobrogea romană (sec. I–III p. Chr.) (Constanţa).
Barnea, A. 1988: ʻMunicipium Noviodunum. Nouvelles données épigraphiques’. Dacia
n.s. 32, 53–61.
Bujor, E. 1960: ‘Cu privire la unele monumente de la Durostorum’. Studii și cercetări
de istorie veche 11.1, 141–54.
Cîrjan, R. 2010: Statute citadine privilegiate în provinciile dunărene ale Imperiului
roman (sec. I–III p. Chr.) (Cluj-Napoca).
Dana, D. and Nemeti, S. 2012: ʻPtolémée et la toponymie de la Dacie (I)’. Classica et
Christiana 7.2, 431–37.
Facella, M. and Speidel, M.A. 2011: ʻFrom Dacia to Doliche (and back). A New
Grave Stone for a Roman Soldier’. In Winter, E. (ed.), Von Kummuh nach

73
Petolescu 2010, 312, no. 1373. The new reading of the name of the settlement, Arcobara
(Dana and Nemeti 2012), is in favour of its civil and partially native character.
106 F. MATEI-POPESCU

Telouch: Historische und archäologische Untersuchungen in Kommagene (Doli-

chener und Kommagenische Forschungen 4) (Bonn), 207–15.
Gerov, B. 1988: Landownership in Roman Thracia and Moesia (1st –3rd Century)
(Amsterdam).
Macrea, M. 1966: ʻOrganizarea provinciei Dacia’. Acta Musei Napocensis 3, 121–51.
Matei-Popescu, F. 2010a: The Roman Army in Moesia Inferior (Bucharest).
—. 2010b: ʻCastellum Abritanorum’. Studia Antiqua et Archaeologica 16, 61–67.
Mócsy, A. 1970: Gesellschaft und Romanisation in der römischen Provinz Moesia
Superior (Budapest).
—. 1992: Pannonien und das römische Heer. Ausgewählte Aufsätze (Mavors Roman
Army Researches 7) (Stuttgart).
Moga, V. 1985: Din istoria militară a Daciei romane: Legiunea XIII Gemina
(Cluj-Napoca).
Nemeti, S. and Bărbulescu, M. 2006–07: ʻTerritorium Arcobadarense’. Ephemeris
Napocensis 16–17, 107–18.
—. 2010: ʻArcobadara’. Latomus 69, 446–55.
Ott, J. 1995: Die Beneficiarier: Untersuchungen zu ihrer Stellung innerhalb der
Randordnung des römischen Heeres und zu ihrer Funktion (Historia Einzelschrif-
ten 92) (Stuttgart).
Petolescu, C.C. 1987: ‘Die Reorganisierung Dakiens unter Marcus Aurelius’.
Germania 65.1, 123–34.
—. 2010: ‘Cronica epigrafică a României’. Studii și cercetări de istorie veche și
arheologie 61.3–4, 307–30.
—. 2011: ‘Villes de la Dacie romaine’. Dacia n.s. 55, 83–109.
Piso, I. 2005: An der Nordgrenze des Römischen Reiches: Ausgewählte Studien
(1972–2003) (Bochumer historische Studien. Alte Geschichte; Heidelberger alt-
historische Beiträge und epigraphische Studien 41) (Stuttgart).
Rankov, N.B. 1983: ‘A Contribution to the Military and Administrative History of
Montana’. In Poulter, A.G. (ed.), Ancient Bulgaria (Papers presented to the Inter-
national Symposium on the Ancient History and Archaeology of Bulgaria,
University of Nottingham, 1981), vol. 2 (Nottingham), 40–73.
Sarnowski, T. 1988: ‘Quellenkritische Bemerkungen zu den Polizeikräften in Nieder-
moesien’. Eos 76, 99–104.
Suceveanu, A. and Zahariade, M. 1986: ‘Un nouveau vicus sur le territoire de la
Dobroudja romaine’. Dacia n.s. 30, 109–20.
Velkov, V. 1965: ‘Eine neue Inschrift über Laberius Maximus und ihre Bedeutung für
die ältere Geschichte der Provinz Moesia Inferior’. Epigraphica 27, 90–109.
Vittinghoff, F. 1994: Civitas Romana: Stadt und politisch-soziale Integration im
Imperium Romanum der Kaiserzeit, ed. W. Eck (Stuttgart).
Vulpe, R. 1976: ‘Colonies et municipes de la Mesie Inférieure’. In Vulpe, R., Studia
Thracologica (Bucharest), 289–314.
 CHAPTER 8

GLASSWARE FROM THE LOWER DANUBIAN LIMES

(1ST–3RD CENTURIES AD): A BRIEF SURVEY*

Sever-Petru BOŢAN

Abstract
The present study aims to present an overview of the distribution and circulation of
Roman glass vessels in the military settlements of the Danubian limes during the first
three centuries AD. Analysing both published and unpublished archaeological material
from important sites, such as Oescus, Novae, Durostorum, Capidava, Carsium, Beroe,
Troesmis and Noviodunum, the main types of drinking, serving, storage and toilet ves-
sels discovered here are considered. The paper also focuses on typological evolution in
relation to the manufacturing techniques and the ‘fashion of the time’, distribution and
the identification of trade routes that linked the province of Moesia Inferior with other
areas of the Roman empire.

This paper is an overview of the most important shapes of glass vessels found
in military centres on the Lower Danube. The available finds come from major
centres such as Oescus, Novae, Durostorum, Capidava, Carsium, Beroe, Troes-
mis or Noviodunum (see Fig. 1). For the most part they are published or in
progress towards publication. Although the paper does not claim to be a com-
prehensive, given that full publication of the glass finds from these sites is far
from complete, it aims to depict a unitary perspective on the most common
types of glassware used in everyday activities.
Research on issues regarding the import, circulation and production of
glassware in the Dobrudja and the Black Sea area is no longer a novelty.1
In recent years, in addition to publication of the finds themselves in their
archaeological context, several general studies have appeared related to the

* This paper was elaborated with the help of a documentation stage generously provided by
RGK Frankfurt, 1–30 September 2017 and within project PN-III-P4-ID-PCE-2016-0271 of the
Romanian National Council for Scientific Research (CNCS – UEFISCDI).
1
For an overview of the main directions of research on the glassware in the Daco-Moesian
hinterland, see Popa 2009, 205–11.
108 S.-P. BOŢAN

dynamics of the distribution and circulation of glassware both in the Dobrudja

and, more broadly, in the Black Sea basin.2
The general evolution of glassware in Roman times can be understood by in
the context of the scheme developed by Stuart Fleming: 1. Historical events
that defined that period; 2. Discoveries and technical innovations that have
allowed the development of new forms and new decorative motifs; and
3. Fashion and tastes of the time.3
The end of the 1st century BC and the beginning of the next mark the grad-
ual advance and consolidation of Roman rule on the Danube through the crea-
tion of the province of Moesia in 12 BC.4 In subsequent years the strategic
significance of the region is indicated by the important number of legions and
auxiliary units garrisoned in the Lower Danubian castra.5 The presence of
three legions (V Macedonica, I Italica and IX Claudia6) at different times was
a key incentive for the development of urban, economic and commercial life
in the region,7 by introducing large quantities of imported goods from other
areas such as Asia Minor, Italia or the Levant. Within the castra and adjacent
settlements, the import of a large number of ceramic, metal and glass vessels
and other artefacts became a vector of Romanisation.
In the history of the evolution of glassmaking, the end of the 1st century BC
was a turning point, marked by the invention and refinement of the glass-
blowing process.8 The immediate effects of this discovery can be examined
and analysed on several levels. On the one hand, the discovery of glass-
blowing techniques allowed artisans to increase their production as well as to
diversify the number of shapes and their functionality.9 On the other, it led
to a rapid dissemination of the new technique in the entire Mediterranean basin
and especially in peninsular Italy, where Sidonian artisans migrated massively
in search of more convenient trade outlets.10

2
Chiriac and Boțan 2013, covering the Hellenistic and Roman periods, for the whole Black
Sea basin; 2014, for the western Black Sea region in the 1st–3rd centuries AD; Boțan and
Chiriac 2016; Popa 2016.
3
Fleming 1999, 3, fig. E.I.
4
Ivanov 2012, 23.
5
Matei-Popescu 2010.
6
Matei-Popescu 2010, 276.
7
For an overview of the evolution of urban settlements on the Lower Danube, see Apara-
schivei 2010.
8
Grose 1977, 9.
9
Strabo (16. 2. 25) and Petronius Sat. (50–51) mention in their writings that glass vessels are
cheap and are preferred by aristocrats because of their transparency and elegance. See also Stern
2015, 88–89, for glassmaking innovations brought by Italic workshops.
10
Chiriac and Boțan 2014, 528.
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 109

In this way, a large variety of vessel types were produced and exported
throughout the Roman empire and, implicitly, in the area covered by the pre-
sent study. From this whole assortment 17 representative types of vessel
have been selected, dating from the beginning of the 1st century to the 3rd cen-
tury AD. 11
Among the most refined and most appreciated glass objects are the mille-
fiori-type vessels made in a technique that appeared as early as the end of
Hellenistic era.12 The most common forms are carinated bowls and plates, but
occasionally other forms of vessel are also found.13 In the absence of clearer
archaeological contexts, on the basis of their frequent occurrence in the west
of the empire, David Grose postulates an Italic origin for them, which can also
be supported by the concentration of luxury goods during the Early Imperial
age (i.e. cameo glass).14
In the region of the Moesia Inferior limes two fragments of millefiori ves-
sels were discovered, probably parts of small bowls. That from Ostrov15 is
decorated with white-opaque, yellow and green floral motifs on a blue back-
ground (Fig. 2.1). Unfortunately, the conditions of its discovery are unknown.
The second piece was found in the castrum of Oescus, and it was regarded by
the author as belonging to an Isings type 1 bowl of Alexandrian or Italic origin
dating back to the first half of the 1st century AD (Fig. 10.1).16 Another frag-
ment was discovered at Cotul Celicului near the village of Poşta (Tulcea
county) in an Early Roman layer (Fig. 2.6).17 It is decorated with white-opaque
and yellow motifs on a dark brown background and was dated to the second
half of the 1st century/beginning of the 2nd century AD.
Finally, we should include here a fragment of a hemispherical bowl discov-
ered in Barboşi, made of purple glass and decorated with yellow-green floral
motifs and red stamens (Fig. 2.4). As a clear discovery context is lacking, the

11
Since many glassware types have remained in use for many centuries, it is difficult to
elaborate a chronological evolution. The end of the 3rd century AD, however, marks a transition
period for the glass industry and a time of profound transformation.
12
For the production technique, see Lierke 2009, 105–06.
13
Grose 1989, 256–61, attributes them to the fourth family of Early Imperial tableware.
14
Grose 1989, 257.
15
The find is now kept at the Lower Danube Museum in Calarasi and is part of the V. Culică
collection. I thank Dan Elefterescu for his kindness in providing the material and related
information.
16
Kabakčieva 2000, 40–41, fig. V.1. The find may be chronologically linked to the first years
of presence of the legio V Macedonica in this Danubian castrum.
17
The find is preserved by ICEM Tulcea. Gratitude to Dr Dorel Paraschiv for this
information.
110 S.-P. BOŢAN

dating of this piece is uncertain, given that these motifs are encountered as late
as the 4th century AD.18
Another type of Roman luxury ware consists of ‘marbled’ bowls or plates
made of coloured glass with white-opaque inlays. Three fragments of a large,
violet-colour plate from Ostrov are decorated with white-opaque motifs in the
shape of ‘peacock eyes’ (Fig. 2.3).19 The vessels decorated in this manner are
representative of the early 1st-century AD Italic glass industry as is shown by
similar discoveries.20 A ribbed bowl of blue-cobalt glass paste, with white-
opaque inlays, discovered in Novae (Fig. 12.1),21 as well as a purple one from
Oescus (Fig. 10.2) may also be included in this category. They are both part of
the early series of such pieces, as we will show below.22
The bowls decorated with vertical ribs are the most common product of the
1st-century AD glass industry, the type being the connecting link between the
Hellenistic and Early Roman period in terms of both decoration and manufac-
ture.23 These bowls, with convex or hemispherical shape and variable rim diam-
eters, were manufactured until the beginning of the 2nd century AD in various
chromatic versions (I shall not dwell on this here). Marbled bowls like the item
from Novae, or the rarer white-opaque bowls with faïence appearance (which
are conversely more frequent in the East Carpathian region24) are a feature of
early 1st-century AD glassmaking. However, in the Flavian age, coloured glass-
ware was displaced by greenish or greenish-blue. Ribbed bowls of this type
are frequently discovered and are mentioned in all important centres of the
Danubian limes: Noviodunum (Fig. 4.1–2),25 Troesmis (Fig. 9.1–2),26 Carsium

18
Boțan 2015, 103–04.
19
These fragments are similar both in shape and decoration with a specimen discovered at
Barboşi and dated by analogy at the beginning of the 1st century AD (Fig. 2.5) (Boțan 2015, 131,
no. 238, pl. LIX.2).
20
See Cool 2016, 121, no. 170, from Pompeii in a Tiberian-Neronian layer; Mandruzzato and
Marcante 2005, 108, no. 304, from Aquileia.
21
Gencheva 2002, 47, type 3a, fig. XLVIII.1; Belivanova 1999, 36. A large number of
mosaic ribbed bowls were found in the Dacian East Carpathian area at Poiana (15 specimens),
but also in a Sarmatian tomb from Iași (Boțan 2015, 106, with bibliography and comments).
22
Kabakčieva 1996, 104, no. 1, fig. 7.1, found in the Forum Temple area, along with a frag-
ment of terra sigillata with the stamp GELLIVS.
23
See Lierke 2009, 109–10.
24
Boțan 2015, 107.
25
Boțan 2007, 53–54. Unpublished object discovered in 2001 in the Cetate-Turnul Mare area,
light blue with violet iridescence.
26
Alexandrescu and Kainrath 2016, 238, fig. 48.G1–G2, of blue-green glass.
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 111

(Fig. 7.1–3),27 Capidava (Fig. 6.4–5, 14–15),28 Durostorum,29 Novae (Figs. 5.1,
3; 12.2–4)30 and Oescus (Fig. 10.8–10).31
To support the refinement of the timeline are specimens discovered in
Tomis, one of which appeared in a grave dated by a coin of Vespasian,32 which
argues for the idea that these blue-green ribbed bowls are specific to the sec-
ond half of the 1st century AD.
It is difficult to identify an area of origin for them, given their wide distribu-
tion, but we can consider a possible Italic or Dalmatian provenance, with
wares transported along the Danube.33
In this category of ribbed bowls we may include vessels with thin ribbed
bulging bodies, made of coloured glass and decorated with white-opaque
impressions.34 With the ribbed decoration inspired by the cast bowls, these
ribbed bowls with marvered threads are one of the earliest known examples of
mould-blown glassware, in the first half of the 1st century AD.35 Such vessels
appear frequently in northern Italy, Switzerland and on the Dalmatian coast,
but are also found in quite significant numbers in the north of the Black Sea,
in Cyprus and on the west coast of Asia Minor.36 Most researchers consider
them to be products of North Italic workshops (possibly Aquileia), but their
manufacture in some other centres is not excluded.37
Within the settlements on the Danubian limes, these bowls were identified
at Noviodunum,38 Carsium (Fig. 8.2)39 and Oescus (Fig. 10.3),40 in contexts
dated between the middle and the last quarter of the 1st century AD.

27
Chiriac 1999, 67–71, nos. 1–3, pl. III.1–3.
28
Matei 1988–89, 125, fig. 3.1, 2 – two fragments from the port area of the settlement.
29
Unpublished finds. Study in progress.
30
Gencheva 2002, 47, type 3b, fig. XLVIII.2–4; Biernacki 1973, 131, fig. X.1; Billewicz
1975, 132, fig. VI.4.
31
Kabakčieva 2000, 42, fig. V.8–10.
32
Bucovală 1968, 37, nos. 28 and 29.
33
See also Belivanova 1999, 38, which takes over the conclusions of Mira Ružić on the
pieces discovered in Serbia.
34
These artefacts are known as ‘Zarte Rippenschalen’ (Isings type 17).
35
Whitehouse 2001, 202.
36
Boțan 2015, 112.
37
Hayes 1975, 153; Lazar 2003, 71.
38
Boțan 2007, 54, in the Cetate-Turnul Mare area (2000), of natural greenish colour.
39
Chiriac 1999, 73, no. 9, pl. IV.2. Similar to the one found in Noviodunum.
40
Kabakčieva 1996, 104, fig. 7.2, of yellowish colour, discovered in the Temple area of the
Forum and dated to the second half of the 1st century AD. Also in the northern region of Bul-
garia, in Telerig (Dobrich county), another fragment was uncovered, similar to the Oescus find
(Belivanova 1999, 39, fig. 16).
112 S.-P. BOŢAN

A singular presence is the glass kylix with over-raised handles, discovered

in Novae (Fig. 12.7).41 Inspired by metallic dishes of Hellenistic tradition, the
piece was discovered in scamnum tribunorum, in an archaeological context
dating back to the second half of the 1st century AD and attributed to the
North Italic workshops.42
The category of drinking vessels also includes small bowls known as
Hofheim Cups, with almost cylindrical bodies narrowing towards the rim and
decorated on the outside with wheel-cut lines.43 These pieces can be dated
approximately to the first three quarters of the 1st century AD and are frequent
finds in military castra, therefore not lacking in our region.
Specimens of this type were reported at Carsium (Fig. 8.1),44 Capidava
(Fig. 6.18),45 Durostorum,46 Novae (Fig. 13.1, 4)47 and Oescus (Fig. 10.4).48
We may also include here two intact specimens uncovered at Tomis in funer-
ary contexts, and dated to the 1st–2nd centuries AD.49 In a wider perspective,
the Isings type 12 bowl is well represented in the Black Sea basin, with a large
number of painted vessels uncovered in settlements on the north Black Sea
coast.50
A special category of glassware is represented by mould-blown vessels dec-
orated with various relief patterns. These are generally regarded as products of
Sidonian officinae, and were very popular throughout the empire in the 1st–
4th centuries AD. Some of these vessels were signed by their makers, while
others bear various wishes or sporting scenes, or were decorated with various
stylised vegetal motifs.51
Five fragments of such mould-blown vessels were found within the perim-
eter of the ancient fortification of Carsium/Hârşova, dating from the 1st cen-
tury AD.52

41
Gencheva 2002, 47–48, fig. XLVIII.7.
42
Gencheva 2002, 48; Belivanova 1999, 36.
43
Isings 1957, 27–30, type 12. The name comes from the military castrum on the Rhine
where a large number of specimens were uncovered (see Price 1998, 72).
44
Chiriac 1999, 71, no. 8, pl. IV.1. The item is blueish-green and pertain to the later variants
of the type.
45
Matei 1988–89, 125, fig. 3.5.
46
The find is part of the Călărași Museum collection. It is made of cobalt-blue good quality
glass and can be ascribed to the earlier variants of the type.
47
Gencheva 2002, 48, fig. IL.4; Belivanova 1999, 40.
48
Kabakčieva 1996, 104, fig. 7.3, of cobalt-blue glass, dated to mid-1st century AD.
49
Bucovală 1968, 38–39, nos. 30, 31.
50
Rostovtzeff 1914, 1–26.
51
For the most complete study on mould-blown glassware, see Stern 1995.
52
Chiriac 1998, 223–26; 1999, 71, nos. 4-6, pl. III.4-6.
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 113

The first fragment is part of a small flask of semi-transparent blue-violet

glass paste.53 These bulbous bottles with a band of lozenges (Fig. 3.7), inspired
by the wares made by glassmaker Ennion of Sidon, functioned as aryballoi,
containing various scented oils used after gymnastic exercise.54 Good analo-
gies are to be found in the collection of the Toledo Museum of Art (USA),
dated, like our item, to the second half of the 1st century AD.55
The second piece is a fragment of a small date-shaped flask (Fig. 3.8),
amber-yellow with slight iridescence and exfoliation.56 These containers seem
to originate in workshops on the Syrian-Palestinian coast and appear mainly in
funerary contexts dating from mid-1st century AD to the beginning of the
2nd century.57 Like the flasks above, these vessels were designed to store oils
and could be offered as New Year’s gifts.58
The following three fragments are parts of high quality thin-walled glass
cups, decorated with vegetal or geometric motifs (Fig. 7.4–6). These vessels
can also be ascribed to Syrian-Palestinian manufacture and fall chronologi-
cally within the 1st century AD.59
Glass beakers decorated with simple or stratified oval protuberances resem-
bling the ‘lotus buds’ are a strong presence in Roman glass drinking ware from
the second half of the 1st century AD.60 The significance of these protuber-
ances is not known but they have been associated by many scholars with the
club carried by the Greek hero Heracles, as a vessel in the Ernesto Wolf col-
lection suggests.61 Depending on the shape and distribution of the decoration,
several classifications were made, which point to various production areas,
both in the East and in the Western empire.62 A distribution map of all finds
known by 2012 shows that beakers decorated with lotus buds are particularly
concentrated along the limes in military environments, being probably

53
Chiriac 1998, 223, pl. I.1.
54
Stern 1995, 151.
55
Stern 1995, 150–51, nos. 53, 54.
56
Chiriac 1998, 225, pl. II.1.
57
Stern 1995, 93, with information on the origins, dating and distribution of the type. In the
neighbouring regions, a similar specimen was discovered in Varna, in a tomb dated to the second
half of the 1st century AD (Belivanova 1999, 38).
58
Stern 1995, 94.
59
Chiriac 1999, 71, nos. 4-6, pl. III.4-6.
60
For an exhaustive study of these beakers, see Boțan and Mocanu 2012.
61
Stern 2001, 124, no. 53.
62
Boțan and Mocanu 2012, 145–46. It was discovered that the decoration with simple protu-
berances is a characteristic of Italic workshops, the decoration of stratified protuberances is
a characteristic of Levantine ones, while the protuberances framed in a rhomboid network would
be a characteristic of Rhenish workshops.
114 S.-P. BOŢAN

preferred by soldiers because of their association with the cult of the Greek
hero.63
The Lower Danube is no exception to this rule, as such glass is reported at
Troesmis (Fig. 9.27),64 Carsium,65 Capidava (Fig. 6.1, 13),66 Durostorum,67
Novae (Fig. 5.4)68 and Oescus (Fig. 10.12).69 From the territory of Moesia
Inferior, we can also mention the finds of Isings type 31 beakers from Neptun/
Mangalia North,70 Ibida-Slava Rusa71 and the sanctuary in Telerig (Bulgaria).72
The chromatic and decorative variety of these specimens indicates that they
come from several officinae, both from the Eastern empire and the Italic or
Dalmatian regions, once again emphasising the popularity of this form within
the drinking ware group.
The category of mould-blown glasses includes also conical beakers deco-
rated with thin relief ribs. These Isings type 33 vessels are rarer, and concen-
trated mainly in present-day northern Italy and southern Switzerland, being
probably manufactured in the workshops of Aquileia in the second half of the
1st century AD.73 In our area such pieces were discovered at Carsium
(Fig. 7.7)74 and Oescus (Fig. 10.13),75 to which can be added a complete exam-
ple found in a cremation grave in Tomis.76
Beakers and cups with ‘honeycomb’ type multi-facetted decoration, of
rhomboid or hexagonal shape, reflect a changing fashion, from colourful to
colourless glassware, by the end of the 1st century AD. Although it was long
believed that such decoration was obtained by cold cutting by specialised
craftsmen (called diatretarii), some more recent research has shown that these

63
Boțan and Mocanu 2012, 156, map 2.
64
Chiriac and Bounegru 1975, 99, pl. III.1; Boțan and Mocanu 2012, 147, no. 2, pl. II.2,
uncovered on a terrace situated near the Roman-Byzantine wall in the south-east of the castle,
together with mixed material, dating from the 1st–4th centuries AD.
65
Chiriac 2017, 198, n. 20, unpublished.
66
Matei 1987, 98, fig. 3.4, 16, beaker fragments with simple protuberances, of transparent,
very thin glass.
67
Unpublished finds. Study in progress.
68
Belivanova 1999, 38, no. 13.
69
Kabakčieva 1996, 104, fig. 7.11, of blueish glass, with non-stratified buds, framed by cir-
cular protuberances.
70
Bucovală 1968, 52, no. 57, in a 1st-century AD grave.
71
Boțan and Mocanu 2012, 147–48, nos. 5–7, from G Curtain and Tower T10 extramuros
sectors. See also Chiriac 2017, 198–99, nos. 2–3, pl. IV.
72
Belivanova 1999, 38, no. 14.
73
Isings 1957, 47–48.
74
Chiriac 1999, 77, no. 7, pl. III.7.
75
Kabakčieva 1996, 104, no. 7, fig. 7.12.
76
Apud Chiriac 1999, 77, dated to the second half of the 1st century AD.
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 115

vessels were actually produced by mould-pressing.77 They usually have a coni-

cal shape and are classified into two groups, depending on the separation of
the decor from the undecorated surface (by embossing the decorated part or by
means of ledges).78
It is not possible to determine accurately the origin of such objects; they
were most probably made all across the empire. As evidence of the popularity
they enjoyed, particularly in the military environment, the multi-facetted
glasses were reproduced in both ceramic and silver. The dominant conclusion
reached is that this is a case of transfer of decorative fashions from glass to
ceramics and metal and not vice versa.79
A beautiful specimen was found in an on-site cremation grave in Novio-
dunum (Movila Popii – M2), along with a varied inventory dated to the second
half of the 2nd century AD (Fig. 4.4).80 Some other fragmentary items, of dif-
ferent sizes and decorative variations, were found in Durostorum,81 while in
Bulgaria such pieces can be mentioned at Oescus (Figs. 10.14, 11.1–2).82
Isings type 85b cups with a coil base were extremely popular within
the everyday drinking ware of the second half of the 2nd century AD and the
beginning of the following century. These are very frequent finds particularly
in the western half of the Roman empire83 and occasionally they could be
decorated by painting or engraving.84 Such specimens have been discovered in
Troesmis (Fig. 9.6)85 and Capidava (Fig. 6.23),86 in contexts dated to the 2nd
and 3rd centuries AD.
The last drinking ware type presented in this paper is the flat-footed beaker
with interior umbo. These vessels, with a tall cylindrical body and massive
foot, remained in use from the end of the 1st up to the 4th century AD.87 They
were certainly manufactured in almost all glassmaking workshops within the
Roman empire and were one of the most common and more popular types of
glass as early as the 2nd century AD. In the military castra on the Lower

77
Lierke 2009, 117.
78
See Oliver 1984, with examples.
79
Oliver 1984, 42.
80
Boțan et al. 2010, 231–32.
81
Unpublished finds. Study in progress.
82
Kabakčieva 1996, 104, fig. 7.10, 12–14, uncovered in the Temple area of the Forum, dated
to the Flavian period.
83
Price 1998, 99–100, who states that the type is frequently encountered in the British
settlements.
84
Isings 1957, 102.
85
Alexandrescu and Kainrath 2016, 238, fig. 48.G6.
86
Matei 1988–89, 127, fig. 3.10.
87
Isings 1957, 48, type 34.
116 S.-P. BOŢAN

Danube, they have been found in Troesmis (Fig. 9.16–18),88 Carsium (Fig. 3.2–
4)89 and Novae (Fig. 14.4–5),90 although they are also present on the western
Black Sea coast in Tomis91 and Odessos.92
The category of serving dishes includes an interesting dish of thick, colour-
less glass, with out-turned and folded rim, decorated with applied glass fila-
ment. The find from Troesmis (Fig. 9.3)93 is dated to the second half of the
1st and the following century AD, by analogy with similar specimens from
Augusta Rauricorum.94
A well represented category within military settlements is storage and trans-
port vessels, among which prismatic bottles stand out. Square or cylindrical in
shape, they vary in size from 6 to 40 cm, with many differences regarding
the shape of rim, body or handles.95 Some of the items have their bases deco-
rated with various motifs or even with the name of the manufacturer; therefore
their distribution may be tracked across large areas. In terms of dating, pris-
matic bottles appear as early as the first quarter of the 1st century AD and
remained in use as late as the 3rd century AD.96
As one may expect, such vessels are also present within the Lower Danu-
bian provinces, being recorded at Noviodunum,97 Troesmis (Fig. 9.9–11),98
Carsium (Fig. 3.5–6),99 Capidava (Fig. 6.11),100 Durostorum,101 Novae
(Fig. 15.1–3)102 and Oescus (Fig. 11.6).103
In addition to tableware and transport containers, another category is repre-
sented by toilet containers. Standing out among these are small vessels with
a globular body and two loop-handles, known as aryballoi. Used to preserve
the ointments and aromatic oils necessary in public baths, this type of vessel is
widespread throughout the Roman empire, being manufactured in many

88
Alexandrescu and Kainrath 2016, 239, fig. 48.G16–G18.
89
Boțan 2007, 56, four fragments of glasses discovered in 1995 on the fortress promontory.
90
Gencheva 2002, 48–49, fig. L.4, 5.
91
Bucovală 1968, 49, nos. 50 and 51, in 1st–3rd-century AD contexts.
92
Belivanova 1999, 40, no. 30.
93
Alexandrescu and Kainrath 2016, 238, fig. 48.G3.
94
Rütti 1991, 86, type AR 84.
95
For an overview, see Charlesworth 1966.
96
Isings 1957, 67–69, types 50 and 51.
97
Boțan 2007, 58, Cetate-Turnul Mare sector, 2001.
98
Alexandrescu and Kainrath 2016, 238, fig. 48.G9–G11.
99
Boțan 2007, 58, seven fragments of prismatic vessels uncovered on the fortress promontory
in 1995.
100
Matei 1987, 98, fig. 3.10.
101
Unpublished finds. Study in progress.
102
Gencheva 2002, 46, fig. XLVII.1–10; Biernacki 1973, 131, fig. X.2.
103
Kabakčieva 1996, 106, fig. 7.15, uncovered in the Temple area of the Forum, with an
Isings type 31 beaker and a ribbed bowl, dated to the Flavian dynasty period.
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 117

production centres.104 In the Danubian limes, such artefacts appear at Novio-

dunum (Fig. 4.3),105 Durostorum106 and Novae (Fig. 14.7–8),107 but also in
many settlements from Moesia Inferior,108 especially in contexts dated within
the first two centuries AD.
A large category, frequently encountered in all settlements of the empire,
consists of unguentaria of various shapes and sizes, which were used to pre-
serve the various aromatic essences. A large number of such containers appear
also in funerary inventories, being deposited as grave-goods or ritually broken
on the funeral pyre. Some of the most common types are the candlestick
unguentaria109 uncovered in Noviodunum in funerary contexts,110 as too in
Troesmis (Fig. 9.28),111 the large and tall bell-shaped unguentaria of Isings
type 82 A1 in Oescus,112 or Isings type 28B unguentaria in Noviodunum113 and
Novae (Fig. 16.6).114
The last type presented here is the round-shaped game pieces made of
ceramic, bone, horn, ivory, wood or glass. Game playing is an extremely old
activity, with origins in Egypt or Mesopotamia, perpetuated and adapted well
into the Roman age.115 Board games such as Ludus latrunculorum, Ludus
Duodecim Scripta or other Mills-type games were widespread in Roman
society,116 especially in military camps.117 Within the latter, glass game pieces
predominate, with diameters between 17 and 21 mm. Virgil Mihailescu-Bîrliba
argues that a standardised production of such pieces could have existed.118

104
Based on a large group discovered in several northern Black Sea settlements, Nina
Sorokina defines three distinct production areas: the eastern Mediterranean, Asia Minor and the
Western empire (Sorokina 1987, 40-45).
105
Boțan 2007, 59, Cetate-Turnul Mare sector, 2001.
106
Unpublished finds. Study in progress.
107
Gencheva 2002, 50, fig. L.7, 8.
108
Bucovală 1968, 81–83, nos. 129–135; Belivanova 1999, 43–44.
109
Isings 1957, 98–99, type 82B1, which can be dated to the 1st–3rd centuries AD.
110
Boțan et al. 2010, 228–29, nos. 9–12.
111
Boțan et al. 2010, 228–29, no. 19.
112
Belivanova 1999, 45, fig. 59, mainly dated to the 2nd–3rd centuries AD.
113
Boțan et al. 2010, 230–31, no. 16.
114
Gencheva 2002, 50, fig. LI.6.
115
Mihailescu-Bîrliba 2016, 36.
116
Nuțu and Boțan 2009, 145–47; Mihailescu-Bîrliba 2016, 36.
117
Mihailescu-Bîrliba 2016, 37, makes an overview on all known finds in the castra of the
Rhenish and Danubian limes.
118
Mihailescu-Bîrliba 2016, 38, a finding that is also valid for the discoveries on the Moesia
Inferior limes.
118 S.-P. BOŢAN

Such glass game pieces are thus present in Noviodunum, Beroe,119

Troesmis (Fig. 9.23–26),120 Carsium,121 Durostorum,122 Novae123 and Oescus
(Fig. 11.7–10).124
***
At the end of this exhaustive presentation, a series of explanations is required
regarding the typology and circulation of glass vessels in the Lower Danubian
region during the first three centuries AD. Integrated early into the mecha-
nisms of Roman politics as an area of high economic and strategic signifi-
cance, this region was connected to the economic and commercial network of
the empire. The relatively heavy military presence had obviously a decisive
role in the introduction of a variety of imports as early as the mid-1st century
AD. Beside glassware, a large number of luxury ceramic wares, such as terra
sigillata, metal vessels, artefacts and jewellery were imported into this region
as part of a centralised supply system for the military. With the advantage of
opening both to the Danube and to the sea ports on the west coast of the Black
Sea, the region of Moesia Inferior benefited from strong trade links with
the West and the rich regions of the East, such as Asia Minor, Egypt and the
Syrian-Palestinian area.
The glassware of this period was almost entirely imported, while local or
regional production cannot be ascertained earlier than the second half of the
3rd century AD. The high statistical weight of glass tableware and drinking
ware (bowls, cups, glasses, platters) is evident, which is due to their role in
everyday life. The presence of millefiori and marbled tableware, at a time
when it was still quite expensive, argues for the high financial potential of the
local elite.
Another noteworthy aspect is that many types of drinking ware (marbled
bowls, thin-ribbed bowls, Isings type 12 cups, relief-ribbed beakers) seem to
have an Italic or Dalmatian origin, which may suggest, at least for the 1st cen-
tury AD, an orientation of supply routes westwards, using the Danube as a
connecting artery.125 However, the products of Levantine workshops are also

119
Apud Nuțu and Boțan 2009, 149.
120
Alexandrescu and Kainrath 2016, 239, fig. 48.G23, G24.
121
Apud Mihailescu-Bîrliba 2016, 38.
122
Unpublished finds. Study in progress.
123
Gencheva 2002, 51, fig. LII.11.
124
Kabakčieva 1996, 106, fig. 7.8.
125
See also the more recent Mocanu 2016 on the presence of Italian terra sigillata in the
western Black Sea region. Based on the analysis of ceramic material, the author points out the
gradual diffusion of imported goods on the Danube line, arguing for the dominance of commer-
cial processes over military supply.
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 119

present, and some of the best examples are the mould-blown vessels decorated
with various vegetal or geometric motifs.
In the larger perspective, the typological distribution of glassware in the
centres discussed indicates its dominant role in many aspects of everyday life,
from meals to body care and even recreational activities.
The proper completion of this framework which I have outlined in this study
requires time, as well as the proper publication and dissemination of the glass-
ware finds made in the Danube–Black Sea region.

BIBLIOGRAPHY

Alexandrescu, C.-G. and Kainrath, B. 2016: ‘Glasfunde’. In Alexandrescu, C.-G.,

Gugl, C. and Kainrath, B. (eds.), Troesmis I: Die Forschungen von 2010–2014
(Cluj-Napoca), 238–39.
Aparaschivei, D. 2010: Oraşele romane de la Dunărea Inferioară (secolele I–III AD)
(Iaşi).
Belivanova, A. 1999: ‘Early Roman Glass from Bulgaria (1st–the first half of the 2nd
century AD)’. Archaologia Bulgarica 3.1, 35–49.
Biernacki, A. 1973: ‘Przedmioty szklane’. In Parnicki-Pudełko, S. (ed.), Novae-Sektor
Zachodni 1970 (Poznan), 131–38.
Billewicz, E. 1975: ‘Wyroby szklane’. In Parnicki-Pudełko, S. (ed.), Novae-Sektor
Zachodni 1972 (Poznan), 131–47.
Boțan, S.P. 2007: Vase de sticlă din Dobrogea romană. Studiu de caz: Carsium, Ibida,
Noviodunum (Dissertation, Iași).
—. 2015: Vase de sticlă în spaţiul dintre Carpaţi şi Prut (secolele II a. Chr.–II AD)
(Cluj-Napoca).
Boţan, S.P. and Chiriac, C. 2016: ‘State of the Art and Prospective Research Direc-
tions on Hellenistic and Roman Glass from the Pontus Euxinus’. In Cojocaru, V.
and Rubel, A. (eds.), Mobility in Research on the Black Sea Region (Cluj-Napoca),
101–15.
Boţan, S.P. and Mocanu, M. 2012: ‘Despre paharele decorate cu „muguri de lotus”
(tip Isings 31) descoperite în Dobrogea romană’. Arheologia Moldovei 35,
137–56.
Boţan, S.P., Paraschiv, D. and Nuţu, G. 2010: ‘Vase de sticlă romane şi romano-
bizantine descoperite în nordul Dobrogei’. Arheologia Moldovei 33, 217–42.
Bucovală, M. 1968: Vase antice de sticlă la Tomis (Constanţa).
Charlesworth, D. 1966: ‘Roman Square Bottles’. Journal of Glass Studies 8, 26–40.
Chiriac, C. 1998: ‘Vase de sticlă de epocă romană descoperite la Carsium (I)’.
Arheologia Moldovei 21, 223–26.
—. 1999: ‘Vase de sticlă de epocă romană descoperite la Carsium (II)’. Arheologia
Moldovei 22, 67–83.
—. 2017: ‘Despre câteva fragmente vitrice de epocă romană timpurie de la Ibida
(Slava Rusă), județul Tulcea’. Arheologia Moldovei 40, 197–208.
Chiriac, C. and Boţan, S.P. 2013: ‘Sticlăria elenistică şi romană din Pontul Euxin. Între
producţie şi import’. In Panait-Bîrzescu, F., Bîrzescu, I., Matei-Popescu, F. and
120 S.-P. BOŢAN

Robu, A. (eds.), Poleis în Marea Neagră: Relaţii interpontice şi producţii locale

(Buharest), 278–318.
—. 2014: ‘Roman Glass Vessels in the Western Pontic Area (1st –3rd centuries CE).
General Remarks’. In Cojocaru, V., Coșkun, A. and Dana, M. (eds.), Inter-
connectivity in the Mediterranean and Pontic World during the Hellenistic and
Roman Periods (Cluj-Napoca), 525–53.
Chiriac, C. and Bounegru, O. 1975: ‘Noi descoperiri arheologice şi numismatice la
Troesmis’. Peuce 4, 97–108.
Cool, H.E.M. 2016: The Small Finds and Vessel Glass from Insula VI.1 Pompeii:
Excavations 1995–2001 (Archaeopress Roman Archaeology 17) (Oxford).
Fleming, S. 1999: Roman Glass: Reflections on Cultural Change (Philadelphia).
Gencheva, E. 2002: Părviyat voenen lager v Novae, provinciya Mizija (Severna
Bălgariya) (Sofia/Warsaw).
Grose, D.F. 1977: ‘Early Blown Glass: The Western Evidence’. Journal of Glass
Studies 19, 9–30.
—. 1989: Early Ancient Glass. Core-Formed, Rod-Formed and Cast Vessels and
Objects from the Late Bronze Age to the Early Roman Empire, 1600 BC to AD 50
(Toledo).
Hayes, J. 1975: Roman and Pre-Roman Glass in the Royal Ontario Museum: A
Catalogue (Toronto).
Isings, C. 1957: Roman Glass from Dated Finds (Groningen/Batavia).
Ivanov, R. 2012: ‘The Roman Limes in Bulgaria (1st–6th c. AD)’. In Vagalinski, L.,
Sharankov, N. and Torbatov, S. (eds.), The Lower Danube Roman Limes (1st–6th
c. AD) (Sofia), 23–42.
Kabakčieva, G. 1996: ‘Frührömische Militärlager in Oescus (Nordbulgarien). Ergeb-
nisse der Ausgrabungen 1989–1993’. Germania 74.1, 95–117.
—. 2000: Oescus: Castra Oescensia. Rannorimski voenen lager pri ustieto na Isk’r
(Sofia).
Lazar, I. 2003: The Roman Glass from Slovenia (Ljubljana).
Lierke, R. 2009: Die nicht-geblasenen antiken Glasgefäße (Offenbach).
Mandruzzato, L. and Marcante, A. 2005: Vetri Antichi del Museo Archeologico
Nazionale di Aquileia. Il vasellame da mensa (Udine).
Matei, C. 1987: ‘Cercetările arheologice în zona instalaţiei portuare antice de la
Capidava (I)’. Cultură şi civilizaţie la Dunărea de Jos 3–4, 95–101.
—. 1988–89: ‘Cercetările arheologice în zona instalaţiei portuare antice de la Capidava
(II)’. Cultură şi civilizaţie la Dunărea de Jos 5–7, 121–41.
Matei-Popescu, F. 2010: The Roman Army in Moesia Inferior (Bucharest).
Mihailescu-Bîrliba, V. 2016: ‘Games and Gamers in Dacia’. Arheologia Moldovei 39,
33–56.
Mocanu, M. 2016: ‘Importuri de terra sigillata italică în spaţiul vest-pontic’. Peuce n.s.
14, 119–28.
Nuţu, G. and Boţan, S.P. 2009: ‘Roman Board Game Pieces in Northern Dobroudja’.
Peuce n.s. 7, 145–57.
Oliver, A. 1984: ‘Early Roman Faceted Glass’. Journal of Glass Studies 26, 35–58.
Popa, A. 2009: ‘Câteva consemnări privind stadiul cercetării vaselor de sticlă din
hinterlandul daco-moesic’. Cercetări arheologice 16, 205–11.
—. 2016: ‘Untersuchungen zum Fundbestand und zur Verbreitung der kaiszerzeit-
lichen Glasgefäße jenseits der römischen Provinzgrenzen von Dacia und Moesia
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 121

Inferior’. In Cojocaru, V. and Rubel, A. (eds.), Mobility in Research on the Black

Sea Region (Cluj-Napoca), 483–530.
Price, J. 1998: Romano-British Glass Vessels: A Handbook (York).
Rostovtzeff, M.I. 1914: ‘Steklyannye raspisnye vazy pozdneellinisticheskogo vremeni
i istoriya dekorativnoi zhivopisi’. Izvestiya Imperatorskoi Arkheologicheskoi
Kommisii 54 (St Petersburg), 1–26.
Rütti, B. 1991: Die römischen Gläser aus Augst und Kaiseraugst, 2 vols. (Augst).
Sorokina, N. 1987: ‘Glass Aryballoi (First–Third Centuries A.D.) from the Northern
Black Sea Region’. Journal of Glass Studies 29, 40–45.
Stern, E.M. 1995: Roman Mold-Blown Glass. The First through Sixth Centuries
(Rome).
—. 2001: Roman, Byzantine and Early Medieval Glass. Ernesto Wolf Collection
(Ostfildern-Ruit).
—. 2015: ‘Roman Glass from East to West’. In Bayley, J., Freestone, I. and Jackson,
C. (eds.), Glass of the Roman World (Oxford/Philadelphia), 77–94.
Whitehouse, D. 2001: Roman Glass in the Corning Museum of Glass, vol. 2 (Corning,
NY).
 122
S.-P. BOŢAN

Fig. 1. Map of the Lower Danubian limes with the places of provenance for the glass finds.
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 123

Fig. 2. 1–3. (a–b) Early Roman glass fragments from Durostorum/Ostrov–Călărași county
(photograph: author); 4–5. (a–b) Early Roman glass fragments from Barboși–Galați county
(after Boțan 2015); 6. Millefiori glass fragment from Poșta–Tulcea county
(photograph: D. Paraschiv).
124 S.-P. BOŢAN

Fig. 3. 1–6. Roman glass fragments from Carsium/Hârșova–Constanța county

(photograph: author); 7–8. Roman glass fragments from Carsium/Hârșova–
Constanța county (after Chiriac 1998).
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 125

Fig. 4. 1–3. Roman glass fragments from Noviodunum/Isaccea–Tulcea county

(photograph: author); 4. Roman honeycomb cup from Noviodunum/
Isaccea–Tulcea county (after Boțan, Paraschiv and Nuțu 2010).
126 S.-P. BOŢAN

Fig. 5. 1–2. Roman glass fragments from Novae/Svishtov

(after Biernacki 1973); 3. Roman ribbed bowl from Novae/Svishtov
(after Billewicz 1975); 4. Roman lotus-bud beaker from Novae/Svishtov
(after Belivanova 1999).
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 127

Fig. 6. 1–13. Roman glass fragments from Capidava–Constanța

county (after Matei 1987); 14–26. Roman glass fragments from
Capidava–Constanța county (after Matei 1988–89).
128 S.-P. BOŢAN

Fig. 7. 1–7. Roman glass fragments from Carsium/Hârșova–Constanța county

(after Chiriac 1999).
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 129

Fig. 8. 1–3. Roman glass fragments from Carsium/Hârșova–Constanța county

(after Chiriac 1999).
 130 S.-P. BOŢAN

Fig. 9. 1–26. Roman glass fragments from Troesmis/Turcoaia–Tulcea county

(after Alexandrescu and Kainrath 2016); 27. Roman lotus-bud beaker from Troesmis/
Turcoaia–Tulcea county (after Boțan and Mocanu 2012); 28. Roman unguentarium from
Troesmis/Turcoaia–Tulcea county (after Boțan, Paraschiv and Nuțu 2010).
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 131

Fig. 10. 1–14. Roman glass fragments from Oescus/Gigen

(after Kabakčieva 2010).
132 S.-P. BOŢAN

Fig. 11. 1–10. Roman glass fragments from Oescus/Gigen

(after Kabakčieva 2010).
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 133

Fig. 12. 1–7. Roman glass fragments from Novae/Svishtov

(after Gencheva 2002).
134 S.-P. BOŢAN

Fig. 13. 1–9. Roman glass fragments from Novae/Svishtov

(after Gencheva 2002).
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 135

Fig. 14. 1–9. Roman glass fragments from Novae/Svishtov

(after Gencheva 2002).
136 S.-P. BOŢAN

Fig. 15. 1–10. Roman glass fragments from Novae/Svishtov

(after Gencheva 2002).
 GLASSWARE FROM THE LOWER DANUBIAN LIMES 137

Fig. 16. 1–11. Roman glass fragments from Novae/Svishtov

(after Gencheva 2002).
 CHAPTER 9

DIE GÖTTIN HEKATE IN SARMIZEGETUSA*

Ioan PISO und Csaba SZABÓ

Abstract
This paper presents a relief of Hekate Triformis from colonia Sarmizegetusa, known in
the literature since the early 20th century (IDR III/2, 365), in the meantime lost and
recently identified in a Calvinist church in Cluj-Napoca. The authors give a revised
reading of the inscription and a broader interpretation of the unusual epithet Dea
placida of the divinity. The paper analyses the connections that may exist in Sarmizege-
tusa between the cults of Hecate (IDR III/2, 220) and Persephone (AE 1983, 840) and,
finally, a very curious relief (IDR III/2, 347) representing a combination between a dog
and a goddess, in fact a hypostasis of Hecate.

Der Kult der Hekate in Dakien war bisher Gegenstand mehrerer Studien.1 Was
uns jedoch veranlasst hat, das Thema, wenn auch in einem beschränkten Maß,
wieder aufzunehmen, ist die Wiederentdeckung eines Reliefs aus Sarmizege-
tusa, das als verloren galt (Nr. 1). Vielleicht ermöglicht uns eine kleine Aus-
wahl von Denkmälern eine tiefere Einsicht in einen Kult und in Mentalitäten
zu gewinnen, die in der römischen Gesellschaft und im privaten Leben eine
nicht unbedeutende Rolle gespielt haben.

Ein wiederentdecktes Relief

1912 begab sich Béla Jánó, der Leiter des Kalvinistischen Kuun Kollegiums
aus Orăștie (Broos/Szászváros) nach Grădiște (Várhely, heute Sarmizegetusa),
wie sich das Dorf, das sich heute auf der Stelle des antiken Sarmizegetusa

* Der vorliegende Aufsatz wurde im Rahmen des Projektes CNCS - UEFISCDI, Projekt-
nummer PN-III-P4-ID-PCE-2016-0271 des Rumänischen Ministeriums für Forschung und Inno-
vation finanziert.
Für die kritische Durchlesung des Manuskriptes sind wir Werner Eck zu Dank verbunden.
1
Siehe, zum Beispiel, Bodor 1989, 1126–27; Ștefănescu 2002–03; Stoian 2004; 2006;
Carbó García 2010, 306–07, 942–43.
140 I. PISO UND C. SZABÓ

befindet, damals noch nannte. Beim Bürgermeister besichtigte Béla Jánó eine
Sammlung von römischen Antiquitäten, darunter auch das Relief, von dem
hier die Rede ist. Kurz darauf wurde ein Teil dieser Gegenstände von der
Schule in Orăștie gekauft. Nach dem ersten Weltkrieg wurde das Kuun Kolle-
gium geschlossen und die Sammlung geriet teilweise in das Museum von
Deva, teilweise, darunter auch das hier besprochene Relief, schien verloren
gegangen zu sein.2
Erhalten war die Beschreibung von B. Jánó,3 übernommen von G. Finály.4
Beim zweiten Autor lesen wir folgendes: „Várhely, Marmorrelief, Frau mit
Fackeln in beiden Händen mit der sonderbaren gut lesbaren Inschrift:
HILAREVOCATA DEPPIAGIDEEXVOTOPOSVIT.
Der Kommentar von Finály: „Ob die von mir vorgeschlagene Lesung
Hilare vocata Deppia Gide etc. richtig ist, bleibt zweifelhaft“. In dieser Hin-
sicht hatte er Recht.
Das Relief wurde von Dorin Alicu, Constantin Pop und Volker Wollmann
in ihr Buch mit der richtigen Bemerkung übernommen, dass die Fackeln auf
Hekate hinweisen könnten.5 Dass aber auf dem Relief zwei Frauen dargestellt
wären, beruht auf einem Irrtum. Der Text von B. Jánó und G. Finály wurde in
IDR III/2, 365 von I. I. Russu mit dem Kommentar übernommen, dass er
„wertloser Unsinn“ wäre. Es wurde aber kein neuer Vorschlag gemacht,
obwohl der Text dafür einige brauchbare Elemente enthält.
Durch einen Zufall hat man 2017 entdeckt, dass fünf Gegenstände der
Sammlung aus Orăștie schon 1926–1927 nach Cluj in das Mädchengymnasium
der Kalvinistischen Kirche gebracht worden waren. Vier Gegenstände befin-
den sich jetzt in der Sammlung des Gymnasiums Apáczai und das Relief in der
Kalvinistischen Kirche (str. M. Kogălniceanu).
Die Marmorplatte (Abb. 1) liegt in einer Holzkiste, wobei der Raum zwi-
schen der Platte und dem Kistenrand mit Gips gefüllt wurde. Das Relief misst
45 × 30 × 2 cm. Es fehlen die linke obere Ecke sowie rechts unten ein Teil des
Rahmens. Die Platte ist senkrecht gespalten, an einigen Stellen ist die Ober-
fläche abgeblättert. Das Relief ist von einem Rahmen umfasst, der auf drei
Seiten ungefähr 2 cm und auf der unteren, beschrifteten, 7 cm mißt. Es stellt
zweifellos die Hekate triformis dar.

2
Über das Schicksal der Schulsammlung von Orăștie siehe Szabó, in Vorbereitung.
3
Jánó 1912, 52.
4
Finály 1912, 531.
5
Alicu, Pop und Wollmann 1979, 75, Nr. 34.
DIE GÖTTIN HEKATE IN SARMIZEGETUSA 141

Abb. 1. Die Votivplatte IDR III/2, 365.

142 I. PISO UND C. SZABÓ

Die Göttin ist auf eine sehr ungeschickte Weise dargestellt.6 Außer der fron-
talen Gestalt sind noch zwei Köpfe im Halbprofil und der linke, bzw. der
rechte Fuß der zwei übrigen Gestalten nebst den vier Armen bemerkbar. Sie
trägt einen Chiton mit Ärmeln bis zu den Ellenbogen und mit übergegürtetem
Apoptygma.7 Auf jeder Schulter ist das Gewand mit Fibeln in Form von Halb-
monden befestigt, so wie es auch bei einer Statue (Hekataion) zu sehen ist, die
ebenfalls aus Sarmizegetusa stammt.8 Das Mondsymbol ist auf die enge Bezie-
hung zwischen Hekate und Selene, die manchmal bis zur Identifizierung geht,
zurückzuführen.9 Auf anderen Reliefs aus den Provinzen an der unteren Donau
trägt die Gottheit den Halbmond auf dem Kopf,10 auf diesem Relief dagegen
einen polos (πόλος) oder kalathos (κάλαθος).11 Ansätze desselben Kopf-
schmuckes sind auch bei den zwei anderen Köpfen zu bemerken.
Die Hauptgestalt hält in ihren Händen zwei nach oben gerichtete, brennende
Fackeln. Wie jede chthonische Gottheit ist Hekate Fackelträgerin (δᾳδοῦχος).12
Dieses Attribut fehlt Hekate fast nie,13 es spielt auch eine wesentliche Rolle in
der Magie.14 Die linke Gestalt hält in der linken Hand und die rechte in der
rechten Hand je eine nach unten gerichtete brennende Fackel. Nur angedeutet
sind die Gegenstände, welche die linke Gestalt in ihrer rechten Hand und die
rechte Gestalt in ihrer linken Hand halten. Wahrscheinlich ist es im ersten Fall
ein Schwert (ξίφος) oder ein Dolch (ἐγχειρίδιον) mit dreieckiger Schneide,15

6
Für die dreileibige und dreiköpfige Darstellung siehe Kraus 1960, 84–165; Kehl 1988, 327–
28; Werth 2006, 243–69.
7
Siehe vor Allem die kompetenten Beschreibungen von Bordenache 1969, 55–59; für die
Kleidung siehe noch Sarian 1992, 999.
8
Daicoviciu und Alicu 1984, 225 = Alicu, Pop und Wollmann 1979, 75, Nr. 33 = Diaconescu
2004 II, 144, Nr. 11.
9
Bezeichnend ist zum Beispiel PGM IV, 2785–2870 (Ἐυχὴ πρὸς Σελήνην); siehe
Heckenbach 1912, 2778–79; Nilsson 1948, 72; Johnston 1990, 29-38; Sarian 1992, 986;
Lautwein 2009, 186–89.
10
Das Relief bei Bordenache 1969, 56, Nr. 95 = Metropoulou 1978, 38, Nr. 24 = Ștefănescu
2002–03, 135, Nr. 7; Bordenache 1969, 56, Nr. 96 = Ștefănescu 2002–03, 136, Nr. 14; für den
Halbmond siehe noch Sarian 1992, 1007.
11
Siehe dazu Euseb. Praep. ad Evang. 3, 11, 32: Ἑκάτη δὲ ἡ Σελήνη - - - ὁ δὲ κάλαθος, ὃν
ἐπὶ τοῖς μετεώροις φέρει, τῆς τῶν καρπῶν κατεργασίας, οὓς ἀνατρέφει κατὰ τὴν τοῦ φωτὸς
παραύξεσιν - - -; siehe Hopfner 1943, 169; Werth 2006, 153–65; Carboni 2011, 33.
12
PGM IV, 2559, 2718: δᾳδοῦχε; PGM IV, 2119, 2120: Ἑκάτη τριπρόσωπος ἑξάχειρ κρα-
τοῦσα ἐν ταῖς χερσὶν λαμπάδας - - -; PGM IV, 2799, 2800: ἡ χέρας ὁπλίζουσα κελαιναῖς
λαμπάσι δειναῖς; siehe Hopfner 1943, 169–70; Carboni 2011, 53.
13
Werth 2006, 150–51.
14
Hopfner 1943, 170; Werth 2006, 156, 161; Lautwein 2009, 190.
15
Wie auf den Reliefs a und c aus Sarmizegetusa (Anm. 29); dazu noch Bordenache 1969,
55–56, Nr. 94–96; siehe Werth 2006, 197–99; Hopfner 1943, 170.
 DIE GÖTTIN HEKATE IN SARMIZEGETUSA 143

im zweiten der Griff einer Geißel (μάστιξ).16 Das Schwert war für die Rache,17
die Geißel für die Bändigung der Dämonen bestimmt.18 Beide Symbole sind
spät eingetreten und gehören zum magischen Bereich.
Die Fußbekleidung ist schematisch dargestellt. Man kann aber sicher sein,
dass die Göttin Sandalen trägt. Sie besaßen eine tiefe Symbolik, die die prak-
tische Notwendigkeit weit überragte. Denn sie symbolisieren die beiden Seiten
der Göttin: den Fruchtbarkeitsaspekt, dann bestanden sie aus Gold, und die
unterweltliche Hypostase, dann aus Erz.19 A. Zografou folgert daraus, dass
„les sandales d’Hécate - - - concrétisent le pouvoir possédé par cette déesse de
poser le pied à différents niveaux cosmiques“.20
Auf dem unteren Rahmen wurde in zwei Zeilen die Inschrift geschrieben,
ohne Trennungszeichen und nicht sehr sorgfältig. Die Buchstaben messen
1,5–2 cm. Die erste Zeile beginnt nicht mit einem H, sondern mit einem sehr
klaren F, gefolgt von einer Hasta, die wegen einer kleinen Verlängerung nach
rechts unten als L identifiziert werden kann. Auf FL folgen die Zeichen IA,
was wohl als Fl(av)ia verstanden werden kann. Dass nach einem Gentile ein
Cognomen folgt, ist zwingend; es lautet zweifellos Revocata. Man begegnet
diesem Namen (Revocatus und Revocata) zumeist in Pannonien,21 aber ein-
mal, als Agnomen, auch im dakischen Sarmizegetusa.22 Gewöhnlich erhalten

16
Wie bei Bordenache 1969, 55–56, Nr. 94, Nr. 96.
17
PGM IV, 2249, 2250: πρὶν τοὺς ξιφήρεις ἀναλάβῃς σου κονδύλους; PGM IV, 2481,
2482: ξιφηφόρον; Euseb. Praep. Evang. 5, 14, 2: ξίφος τὸ ποίνιμον; siehe Hopfner 1943, 170;
Carboni 2011, 34.
18
Audollent 1904, 242, 39-41: Ἑκάτης τριμόρφου μαστειγοφόρου; Euseb. Praep. Evang.
5, 14, 2: τὸ δαιμόνων κράτος μάστιγος ψόφος πολὺς; siehe Hopfner 1943, 170; Werth 2006,
207–08; Lautwein 2009, 63. Geißel trägt auch eine der Gestalten eines Hekataions aus
Sarmizegetusa (Anm. 93).
19
Euseb. Praep. ad Evang. 3, 11, 32: Ἑκάτη δὲ ἡ Σελήνη - - - τῆς μὲν νουμηνίας φέρουσα
τὴν λευχείμονα καὶ χρυσοσάνδαλον καὶ τὰς λάμπαδας ἡμμήνας· - - - τῆς δ᾽ἆυ πανσελήνου
ἡ χαλκοσάνδαλος σύμβολον; SM 1, 49, 57–61: <κύων, δράκαινα> [στέμ]μα, κλ[είς, κη]
ρύκ[ιο]ν, τῆς Ταρταρούχου χάλ<κ>εον τὸ σὰνδαλον, τῆς δε[…]που χρύσεον τὸ σάνδαλον·
ἰδὼν δὲ ἐγὼ ἔφυγον τ[ὴν σιδη]ροσάνδαλον ἔβην δὲ ἐπ᾽ἴχνεσι χρυ<σ>οσανδάλου Κόρη[ς·
σῶσον] με, σωσίκοσμε, Δήμητρος κόρη; PGM IV, 2333–2336: χάλκεον τὸ σάνδαλον τῆς
Ταρταρούχου, στέμμα, κλεῖς, κηρύκιον, ῥόμβος σιδηροῦς καὶ κύων κυάνεος; vgl. P. Mich.
III, 154: Ἐρεσχιγὰλ παρθένε, κύων, δράκαινα, στέμμα, κλεῖς, κηρύκειον, [τ]ῆς Ταρταρού-
χου χρύσεον το σάνδαλον; Audollent 1904, 242, 41: χρυσοσάνδαλος; PGM IV, 2119: σάν-
δαλα ὑποδεδεμένη; Siehe Lautwein 2009, 192–94; Zografou 2015, 139–44.
20
Zografou 2015, 142.
21
OPEL III, 28.
22
CIL III, 1471 = IDR III/2, 366: Septimia Septimina quae et Revocata als Tochter eines
P. Ael(ius) Sept(imius) Audeo, Centurio palmyrenischer Abstammung, und einer Cornelia
Antonia.
144 I. PISO UND C. SZABÓ

solche Namen Leute, die ihren Lebenskurs oder ihre religiöse Gemeinde
gewechselt haben.23
Die folgenden Zeichen sind De(ae) p(- - -) placi|d(a)e zu lesen. Das P, das
dazwischen steht, ist vermutlich p(raesentissimae) aufzulösen. Dasselbe Epi-
theton wurde in Sarmizegetusa Core (Persephone) zuteil,24 einer Göttin, die
mit Hekate viele gemeinsame Züge hatte.25 Praesentissma deutet nicht nur
darauf hin, dass Hekate bereit war, dem um Hilfe Bittenden zu helfen,26 son-
dern weit mehr, dass sie, laut Hesiod27, auf der Erde, im Himmel und in Was-
ser, das heißt im gesamten Universum, allgegenwärtig war. Gerade diese Uni-
versalität soll durch die dreifache Gestalt der Gottheit ausgedrückt worden
sein.28
Die Inschrift endet mit einer bekannten Formel: ex voto posuit. Der gesamte
Text lautet somit (Abb. 2):
Fl(av)ia Revocata de(ae) p(raesentissimae) placi-
d(a)e ex voto posuit.
Ähnliche Reliefs aus Sarmizegetusa wurden in das 3. Jahrhundert datiert. Es
gibt deren noch drei29 und dazu zwei Hekataia, das heißt Statuen, die die drei-
gestaltige Hekate darstellen und gewöhnlich an Wegekreuzungen gestellt
waren.30 Die Inschriften, auf deren die Epitheta praesens oder praesen-
tissima erscheinen,31 sind entweder zwischen 230–23532 oder zwischen

23
Siehe Kajanto 1965, 135, 356.
24
AE 1983, 840 = ILD 263. Ebenfalls in Sarmizegetusa wurde eine Weihung Apollini Granno
et Sironae dis praesentibus (AE 1971, 376 = IDR III/2, 191) und eine andere Θεῷ Γράννῳ
Ἀπόλλωνι αἰεὶ καὶ πανταχοῦ ἐπηκόῳ (ΑΕ 1983, 833 = SEG 35, 589 = CIGD 109 = ILD 257)
gemacht. Die Formel αἰεὶ καὶ πανταχοῦ überträgt ganz genau ins Griechische den lateinischen
praesentissimus.
25
Siehe unten.
26
Hes. Theog. 429: ᾯ δ᾽ἐθέλῃ, μεγάλως παραγίνεται ἠδ᾽ὀνίνησιν.
27
Hes. Theog. 426–28: Οὐδ᾽, ὅτι μουνογενής, ἧσσον θεὰ ἔμμορε τιμῆς / καὶ γεράων γαίῃ
τε καὶ οὐρανῷ ἠδὲ θαλάσσῃ / ἀλλ᾽ἔτι καὶ πολὺ μᾶλλον, ἐπεὶ Ζεὺς τίεται αὐτήν; Orph. Hymn.
1, 2: οὐρανίαν χθονίαν τε καὶ εἰναλίαν; siehe Carboni 2011, 21.
28
Siehe Kehl 1988, 331–33; Zografou 1999, 57–70.
29
a – Buday 1916, 92–93, Nr. 1 = Bordenache 1969, 58, Nr. 100 = Metropoulou 1978, 45,
Nr. 53; Alicu, Pop und Wollmann 1979, 74, Nr. 30 = Ștefănescu 2002–03, 134, Nr. 1 = Stoian
2004, 178 = Werth 2006, 390–91, Nr. 213 = Carboni 2011, 153, II.12.11; b – Buday 1916, 93,
Nr. 2 = Alicu, Pop und Wollmann 1979, 74, Nr. 31 = Ștefănescu 2002–03, 134, Nr. 2; c – Buday
1916, 93–94, Nr. 3 = Alicu, Pop und Wollmann 1979, 74, Nr. 32. Nicht Sarmizegetusa zuzu-
weisen, wie Bodor 1989, Taf. VII.18 und Ștefănescu 2002–03, 135 glauben, ist das Relief
Bordenache 1969, 56, Nr. 95.
30
d – Buday 1916, 95, Nr. 4 = Ferri 1933, 346 = Alicu, Pop und Wollmann 1979, 75, Nr. 33
= Daicoviciu und Alicu 1984, 225 = Ștefănescu 2002–03, 134, Nr. 4 = Stoian 2004, 177;
e – Alicu, Pop und Wollmann 1979, 75, Nr. 35.
31
Siehe oben, Anm. 24.
32
Siehe Piso 2013, 221–26.
 DIE GÖTTIN HEKATE IN SARMIZEGETUSA 145

Abb. 2. Zeichnung der Inschrift IDR III/2, 365.

236–23833 datiert. Das ergibt mit einiger Wahrscheinlichkeit auch die Datie-
rung unseres Reliefs.

Dea Placida

Das wiederum belegte Epitheton placida bedarf einer eingehenden Diskussion.

Es erscheint bisher in den folgenden Inschriften:
1. – Tudor 1942, SE 251 = Tudor 1958, SE 303 = AE 1959, 333 = Borde-
nache 1965 = AE 1965, 277 = Gostar 1965, 239, Nr. 3 = Tudor 1968, SE 385
= Bordenache 1969, 57–58, Nr. 99 = Sanie 1974, 111–14 = IDR II, 144 =
Metropoulou 1978, 38, Nr. 25 = Ștefănescu 2002–03, 132 = Stoian 2004, 178
= Werth 2006, 393, Nr. 217 = Bordenache Battaglia 1986, 665 = Carbó García
2010, 942 = Carboni 2011, 146–47 (Cioroiu Nou): Domna placi|da Val(- - -)
MEXY | votum.
2. – CIL III, 1590 a = 8029 = Gostar 1965, 239, Nr. 1 = Tudor 1968, 388 =
SIRIS 295, Nr. 680 = Sanie 1974, 111–14 = IDR II, 338 = Carboni 2011, 147
(Romula): Placidae | reginae | eq(uites) v(otum) l(ibentes) p(osuerunt) | per
Proculo |5 princ(ipe) et | [G]aio opt(ione).
3. – Tudor 1942, SE 118 = Tudor 1958, 145 = AE 1959, 321 = Gostar 1965,
239, Nr. 2 = Tudor 1968, SE 191 = IDR II, 198 (Sucidava): Deae placid(ae) |
pro salute | Marcianae e[t] | Quintilian[i] | fil[iorum] | - - -?].34
4. – Eck und Ivanov 2010, 203, Nr. 2 (Ratiaria) = AE 2010, 1393 (Ratiaria):
Deae Pla|cidae Ael(ius) | Hercula|nus lapida(rius) | o(rnatus) o(rnamentis)
d(ecurionalibus) vo(to) | p(osuit).

33
Siehe Piso 2013, 227–35.
34
Mit einer kleinen Berichtigung in Z. 1, wo ganz am Rande des Inschriftenfeldes placid[ae]
gelesen wurde.
146 I. PISO UND C. SZABÓ

5. – Gerov 1964 = AE 1964, 224 = Gerov 1980, 113–18 = ILB 281 = IGLN
39 (Novae): De(a)e Sanct(a)e | Placid(a)e | Iulius | Statilis |5 Augustalis |
m(unicipii) N(ovensium) ex vo(to) posu|it n(uminis) m(onitu).
6. – Kolendo 1969 = IGLN 40 (Novae): [P]lacidae [A]u[r(elius) | S]aturni-
nus | [p]raef(ectus) A[ug(usti)] leg(ionis) | [I Ita]l(icae) et Aur[elia |5 Q]uieta
S[- - - | - - - e]t coniux | [- - -]PE[- - -].
7. – Bunsch und Kolendo 2011, 42–45, Nr. 1 = AE 2011, 1126 (Novae):
Nu(mini) Plac(idae) | Aur(elius) Val(- - -) S|FN.35
Es ist von vornherein anzunehmen, dass in diesen Inschriften, ungeachtet
der Termini dea, domna, numen oder regina, dieselbe Göttin bezeichnet wurde,
deren sanftes Gemüt mit dem Epitheton placida gekennzeichnet ist.36 Der
Ausgangspunkt der Diskussion war immer und soll auch jetzt das Relief Nr. 1
aus Cioroiu Nou bleiben.37 Die dreigestaltige Göttin, mit Polos und in einem
Chiton mit kurzen Ärmeln und Apoptygma, erscheint ohne ihre furchterregen-
den Symbole,38 wohl aber in einer magischen Geste mit zwei gestreckten, dem
Zeigefinger und dem kleinen Finger über zwei hohen Altären. Der Name der
Gottheit steht im Nominativ: Domna placi|da, nicht etwa im Dativ.39 Fraglich
bleibt die Deutung der Buchstaben MEXY, worin alle Autoren einen griechi-
schen Namen gesehen haben. Ein Name wie Μείξις oder Μειξίας, wie sich
N. Gostar vorstellte,40 ist aber sehr unwahrscheinlich, denn man würde den
Diphthong -ei- erwarten.41 Dann sollte man eine Lösung akzeptieren, die von
G. Bordenache entschieden zurückgewiesen wurde, nämlich die Formel ex
v(oto). Der Dedikant oder die Dedikantin, ein Val(erius) M(- - -) oder eine
Val(eria) M(- - -), werden ihre Namen abgekürzt haben, weil sie im Umkreis
gut bekannt waren und weil die Reliefplatte wahrscheinlich an der eigenen Tür
befestigt war (siehe weiter unten). Dass die Schlussformel ex v(oto) durch
votum verdoppelt wurde dürfte bei einem in allen Hinsichten primitiven

35
Könnte auch Nu(mini) plac(ido) gelesen werden.
36
So Bordenache 1965, 316.
37
Bordenache 1965, Abb. 1.
38
Wahrscheinlich trägt sie nicht einmal Fackeln.
39
Domna[e] placi|da[e] bei Bordenache und Carboni, falsch. Manchmal kann der Name der
angebetenen Gottheit auch im Nominativ stehen, wie zum Beispiel Diana cons(e)rvatrix in einer
Inschrift aus Potaissa; falsch gelesen von Bărbulescu 1972, 205; dom[i]na placi|da bei Carbó
García, ohne Rechtfertigung. Tudor hat mit einer Dedikantin mit dem Namen Domna Placida
gerechnet; vgl. die richtige Deutung bei Bordenache 1965, 316.
40
Gostar 1965, 239.
41
Siehe LGPN I, 302; II, 301; III A, 292; III B, 273, 286: Μειξίας, Μεῖξις, Μειξιάδης,
Μειξίδημος, Μείξυλλος, Μειχύλος, Μιξίδαμος usw.; keiner dieser Namen erröffnet eine
Perspektive.
 DIE GÖTTIN HEKATE IN SARMIZEGETUSA 147

Produkt nicht allzu stören.42 Vielleicht hat man votum deswegen hinzugefügt,
weil die vorige Formel nicht genügend klar war.
Im Mittelpunkt der Diskussion steht das Beiwort placida. G. Bordenache
sprach sich mehrmals und ganz entschieden dafür aus, dass wir es mit Hekate
zu tun haben.43 D. Tudor entgegnete, dass ein solches Beiwort sich der Natur
der Hekate widersetze44 und schien geneigt, es der Göttin Venus zuzuschrei-
ben.45 Ihrerseits identifizierten Vidman46 und Sanie47 die Göttin mit Isis. Kei-
ner dieser Autoren schien die Darstellungsart der Göttin auf dem Relief von
Cioroiu Mare zu berücksichtigen. Vor allem verkannte man die ambivalente
Natur der Hekate.
Diese Göttin konnte zwar die Dämonen gegen die Oberwelt entfesseln; sie
besaß aber auch die Macht, sie zu bändigen.48 Sie war mit ihrem Gefolge
besonders an Straßen und an Dreiwegen berüchtigt49 und dennoch, oder eher
deswegen, errichtete man ihr Bild als Schutzgöttin an der Straße und weihte
ihr dort Kapellen und Hekataia, ebenso an Straßenkreuzungen und vor den
Stadtmauern; ferner sprach man sie als προπυλαία, als προθυραία oder als
πρόδομος an.50 Es sind Mysterien der Hekate bekannt, an denen man zum
Zweck der Heilung teilnahm.51 Als Schicksalsgöttin, die das zyklische Werden
und Vergehen des Lebens beherrschte, war Hekate eine ambivalente Göttin.52
Das heißt, dass sie auch als apotropäische Göttin verehrt wurde.53 Um sie
gnädig zu stimmen, kam man ihr mit schmeichelnden Epitheta entgegen. In
einer solchen Hypostase waren für sie Epitheta wie ἐπήκοος, ἐραννή,
εὐκολίνη, εὐμενής, θεράπαινα, καλλίστη, κουροτρόφος, μελινόη,

42
Eine ähnliche Lösung in EDCS-11200728: Dom(i)na Pla|cida Val(erius) M() ex {V} |
vot<o=U>{}m.
43
Bordenache 1965, 316–18; 1969, 57; 1986, 665, gefolgt von Bodor 1989, 1176, Ștefănescu
2002–03, 132 und Carboni 2011, 146–47. Kurz nach der Beweisführung von Bordenache identi-
fizierte Kolendo (1970, 77–84) die Dea placida von Novae (Nr. 6) mit Hekate; siehe auch den
Kommentar von Kolendo und Božilova zu IGLN 39–40 (Nr. 5–6); zustimmend Kehl 1988, 318;
im selben Sinn Bunsch und Kolendo 2011, 44.
44
Tudor 1968, 388.
45
Tudor 1965, 54; vgl. CIL VI, 783 = Dessau 3167: Veneri placidae sacrum.
46
SIRIS 600.
47
Sanie 1974, 113; so, mit einiger Vorsicht, auch Carbó García 2010, 943.
48
Heckenbach 1912, 2778.
49
PGM IV, 2819–26 (Gebet an Selene): - - - τετραοδῖτι - - - τριοδῖτι - - - τριόδων μεδέεις.
Über die symbolische Bedeutung aller Durchgänge siehe Heckenbach 1912, 2775; Johnston
1990, 23–28; 1991.
50
Bruchmann 1893, 98; siehe auch Zografou 1997, 176.
51
Heckenbach 1912, 2777.
52
Siehe Kehl 1988, 317–18; Lautwein 60, 68.
53
Treffend Burkert 1977, 309: „Dabei handelt es sich offenbar um eine Polarität, in der das
eine ohne das andere nicht sein kann und jeweils erst vom anderen her seinen vollen Sinn
empfängt“.
148 I. PISO UND C. SZABÓ

πρόδομος, προθυραία, προπυλαία, σωτείρη, φυλακή durchaus passend.54

Alle diese Epitheta entstammen dem griechischen Milieu. Im römischen Milieu
aus Dakien und Moesien hat sich für die wohlwollende und hilfeleistende
Hekate eben das Epitheton placida herausgebildet, dem ἐραννή vielleicht am
nächsten kommt.55 Eine gute Parallele dazu bildet das in Brixia56 und in Sar-
mizegetusa57 belegte Epitheton mitis für Volcanus. Es sollte die gute Seite
dieses Gottes, als Beschützer vor der Feuergefahr, unterstreichen, damit man
ihn dadurch bewegte, entsprechend zu agieren.58
Die wiederentdeckte Reliefplatte ist der endgültige Beweis dafür, dass
G. Bordenache die dea placida richtig mit Hekate identifiziert hat.

Persephone und Hekate

Kore/Persephone (Κόρη/Περσεφόνη) wurde vom Publikum in zwei Hyposta-

sen aufgefasst. Als Tochter der Demeter, die von Hades geraubt wurde, und
als Gattin desselben wies sie Züge aus, die sie von Hekate unterschied; in der
ganzen Geschichte des Raubes spielte die Letztere nur eine Nebenrolle.59
In dieser ersten Hypostase erscheint Core/Persephone in Sarmizegetusa unter
dem Namen Proserpina auf zwei Votivreliefs neben Dis Pater.60
Letzten Endes galt aber Persephone als eine unterirdische Göttin, was sie im
Zauberwesen der Hekate immer mehr annäherte.61 Als Zaubergöttin erscheint
Persephone als mit Hekate identifizierbar, manchmal innerhalb der Triade
Artemis – Hekate – Persephone, oder auch innerhalb der Triade Demeter –
Persephone – Hekate.62 In Beschwörungen von Totenseelen, besonders von
Seelen gewaltsam Gestorbener (βίαιοι), trugen Hekate und Persephone die
gemeinsamen Epiklesen babylonisch-assyrischer Herkunft Ἐρεσχιγάλ und

54
Siehe Roscher 1886–90, 1891–93; Bruchmann 1893, 96–99; Heckenbach 1912, 2776–77;
Kraus 1960, 65; Johnston 1990, 25–28.
55
Orph. Hymn. 1, 1: Εἰνοδίαν Ἑκάτην κλήιζω, τριοδῖτιν, ἐράννην. In den literarischen
Quellen findet man die Beiworte placidus und placida für Amor, Apollo, Iupiter, Liber, Nym-
phae, Pax, Venus und sogar für Manes; siehe Carter 1902, 140.
56
CIL V, 4295.
57
AE 1983, 827 = ILD 251.
58
Siehe Piso 1983, 236.
59
Bräuninger 1937; Lautwein 2009, 69–76; Carboni 2011, 51; Dimou 2016, 29–42.
60
IDR III/2, 199: Dito Patri et Proserpi|nae sacr(um) | C(aius) Val(eriu) Zeno dec(urio) et
IIvir col(oniae) v(otum) s(olvit) | l(ibens) m(erito); CIL III, 12583 = IDR III/2, 200: Papiria Zoe
ex voto | posuit.
61
Siehe die zahlreichen Anrufungen an Core bei Audollent 1904, Indices, S. 461–63.
62
Hopfner 1939, 125–26; Nilsson 1948, 72–73; Carboni 2011, 31, 50–57.
 DIE GÖTTIN HEKATE IN SARMIZEGETUSA 149

Νεβουτοσουαλήθ.63 In einer bekannten Stelle bei Apuleius unterscheidet sich

Proserpina von Hekate weder durch ihr Aussehen, noch durch ihre Tätigkeit
oder Funktion.64
Hekate und Core erscheinen in Sarmizegetusa auf zwei bekannten Inschrif-
ten auf Stein:
A. – AE 1913, 51 = ILS 9515 = IDR III/2, 220 = Piso 2013, 201, Nr. 1:
Deae Aechatae (sic) | pro salutae (sic) | Aeliae Satur/ninae coniu/5gis suae
Aerennius (sic) Gemelli|nus tribunus | statum con|iugis redemit | ex visu et |10
v(otum) s(olvit) l(ibens) m(erito).
B. – Piso 1983, 246, Nr. 15 = AE 1983, 840 = Piso 2013, 222, Nr. 8: Deae
praesen|tissimae Core | M(arcus) Lucceius Felix | proc(urator) Aug(usti)
n(ostri) et |5 Hostilia Fausti|na eius.
In keinem der zwei Fälle kann man feststellen, ob wir es mit einem Altar
oder mit einer Statuenbasis zu tun haben. Wenn man ein Opfer darbringen
will, kann man auf jede Basis einen mobilen focus stellen, während eine Statue
durch ihr eigenes Gewicht und ohne Bauklammer darauf stehen kann.
Als er den Stein mit der Inschrift A aufstellte, bekleidete Herennius Gemel-
linus65 eher den Tribunat der cohors I sagittariorum milliaria oder der cohors
I Vindelicorum milliaria in Tibiscum als den Legionstribunat in Apulum.66
Etwa zehn bis fünfzehn Jahre später, um 198–209, finden wir ihn wieder in
Sarmizegetusa, diesmal amtlich, als proc(urator) Auggg(ustorum), das heißt
centenarer Finanzprokurator von Dacia Apulensis.67 In dieser Eigenschaft ist
er in Sarmizegetusa durch zwei Votivsäulen bekannt: eine wurde von seiner
Frau Aelia Saturnina einer unbekannten Gottheit für das Wohl ihres Gatten,68
die zweite von ihm selbst für das Wohl seiner Frau und seiner vier Kinder an
Deus Aeternus69 geweiht.

63
PGM IV, 2745–50: - - - ὦ Ἑκάτη - - - Περσεφόνα - - - Ἐρεσχιγάλ Νεβουτοσουαλήθ;
siehe Hopfner 1931, 333–34.
64
Apul. Met. XI 2, 5: seu nocturnis ululatibus horrenda Proserpina triformi facie larvales
impetus comprimens terraeque claustra cohibens lucos diversos inerrans vario cultu propitiaris
- - -; siehe auch PGM IV, 2522–28: Ἄρτεμι, Περσεφόνη, ἐλαφηβόλη, νυκτοφάνεια, τρίκτυπε,
τρίφθογγε, τρικάρανε Σελήνη, θρινακία, τριπρόσωπε, τριαύχενε καὶ τριοδῖτι, - - -; siehe
Heckenbach 1912, 2773; Hopfner 1939, 125–26; Lautwein 2009, 136.
65
Siehe zu diesem Ritter Stein 1944, 61–62, 80; Pflaum 1960–61 II, 688, Nr. 254; III, 1066;
Devijver 1976, I, H 14; Thomasson 2009 156, Nr. 50; Piso 2013, 201–04.
66
Piso 2013, 202.
67
Auggg. wurde oft schon seit 198, dem Jahr der Ausrufung Getas zum Caesar, verwendet.
68
CIL III, 1625 (+ S. 1018) = IDR II, 640 = IDR III/2, 342.
69
IDR III/2, 188.
150 I. PISO UND C. SZABÓ

Die Worte statum coniugis redemit ex visu et v(otum) s(olvit) l(ibens)

m(erito) ergeben den Grund für die Weihung des Denkmals. Man versteht dar-
unter, dass infolge eines Traumes Herennius Gemellinus den (kritischen)
Zustand seiner Frau losgekauft hatte70 und deshalb sein Gelübde Hekate
gegenüber eingelöst hat. Die Ursache des schlechten Zustandes der Aelia
Saturnina konnte sowohl eine Krankheit als eine Zauberei gewesen sein.71 Für
beides war Hekate zuständig. Beides konnte aber auch miteinander verbunden
sein, denn bestimmte Krankheiten wurden nach der Vorstellung der Zeit durch
Zaubereien bewirkt. Die Gunst der Göttin wird man durch spezifische Opfer
gewonnen haben.72
Die Quellenlage für M. Lucceius Felix,73 Finanzprokurator von Dacia Apu-
lensis 230–235, ist verschieden. Mit einer Ausnahme74 wurden die Altäre oder
Statuenbasen in der Area sacra des Prätoriums des Prokurators gefunden.75
Keine dieser dreizehn Inschriften enthalten eine Formel wie votum solvit; sie
wurden also nicht infolge von Gelübden, sondern im Rahmen der Dienstoblie-
genheiten des Prokurators geweiht.76 Deswegen wurden sie mit einer einzigen
Ausnahme nur von Prokuratoren unmittelbar und ohne spezifisches Ziel (?)
geweiht, nie aber für das Wohl eines Familienmitgliedes. Die Ausnahme ist
das Denkmal für Core (B); es wurde von M. Lucceius Felix zusammen mit
seiner Frau errichtet. Der Grund dafür ist, dass die Beziehung zu dieser Gott-
heit intimer war. Somit dürfte wie für die Inschrift A Krankheit, Zauberei oder
beides der Anlass gewesen sein. Der Weihung nach zu urteilen, waren der
Prokurator und seine Frau mit den Diensten, die ihnen Core/Persephone erwie-
sen hatte, zufrieden.

70
In diesem Sinn auch die rumänische Übersetzung in IDR III/2, 220; siehe auch Bărbulescu
2003, 238.
71
Siehe Piso 2013, 202.
72
Für die Opfer an Hekate siehe Heckenbach 1912, 2781; Lautwein 2009, 97–104; Carboni
2011, 27.
73
Siehe für diesen Ritter Petersen, PIR2 L 357; Piso 2013, 221–26.
74
CIMRM II, 323, Nr. 2149, 2150 = Alicu, Pop und Wollmann 243 = IDR III/2, 286 – für
Mithras.
75
AE 1998, 1096 = ILD 273 = RICIS 616/0207 – für Serapis; AE 1998, 1091 = ILD 268 =
RICIS 616/0208 – wahrscheinlich für Serapis; AE 1983, 835 = ILD 258 – für Apollo sanctus; AE
1983, 835 = ILD 259 – für Diana sancta; AE 1983, 840 = ILD 263 – für Core; AE 1983, 837 =
ILD 260 – für Aesculapius et Salus; AE 1998, 1092, 1094 = ILD 271, AE 1983, 838 = ILD 261,
AE 1983, 839 = ILD 262, AE 1998, 1098 = ILD 275, AE 1998, 1097 = ILD 274 – für unbekannte
Gottheiten; vielleicht auch AE 1993, 1345 = ILD 282 – für Caelestis.
76
Siehe für diesen Aspekt Piso 1983, 234–35. Die Inschrift, in welcher die Formel [- - - ei]
us votum erscheint, ist jene für Mithras (Anm. 74) und stammt wahrscheinlich aus dem
Mithreum.
 DIE GÖTTIN HEKATE IN SARMIZEGETUSA 151

Hekate und Hund

Die Beziehung der Hekate zum Hund ist ein intensiv besprochenes Thema.
Der Hund war der Hekate heilig, weil er von Natur aus über telepathische
Begabung verfügt und damit „geister- und dämonensichtig“ war.77 In ihrer
wilden Jagd über die Erde und über den Himmel fehlten nie die Hunde.78 Die
Göttin war ebenso schwarz (μέλαινα) wie ihre Hunde;79 besser gesagt, war sie
in schwarz angezogen.80 An Kreuzwegen wurden ihr Hunde geopfert.81 Diese
innige Beziehung hatte die Folge, dass die Göttin manche Hundeeigenschaften
zuschrieb. Sie benahm sich wie ein Hund, sie bellte82 und konnte als dreige-
staltige Göttin mit Tierköpfen dargestellt werden: mit Pferd-, Schlange- und
Hundekopf (Orph. Argon. 977–980), oder mit den Köpfen von Jungfrau, Kuh
und Hund,83 von Frau, Rind und Hund (Lukian. Philops. 14), oder Stier, Löwin
und Hund (Porphyr. De abstin. 3, 17, 2).84 Es ging so weit, dass sie gerade-
wegs als Hund angerufen war.85
Eine einzigartige Votivplatte (Abb. 3)86 wurde 1971 etwa 2 km nord-west-
lich von Sarmizegetusa, auf der Anhöhe Dealul Selei, entdeckt. Die Maße lau-
ten: 14 × 25 × 4 cm. Das Relief ist von einem Rahmen umgeben, der auf der
unteren, beschrifteten Seite, breiter ist. Buchstaben: etwa 1 cm. Die Schrift ist
unsorgfältig, das erste E wurde wie ein I geschrieben, aber der Text ist klar:
Euhemer(us) pro salute Antonini
fili(i) sui.

77
So Lautwein 2009, 102; siehe noch dazu Carboni 2011, 28–30.
78
PGM IV, 2721, 2722: σκυλακάγεια; Orph. Hymn. 1, 5: σκυλακῖτιν; PGM IV, 2528,
2529: τρισὶ μορφαῖσι καὶ φλέγμασι καὶ σκυλακέσσι; siehe Rohde 1898 II, 83–84, 411–13;
Heckenbach 1912, 2776–77; Hopfner 1939, 134–36.
79
Haspels 1971, 324, Nr. 63: - - - Ἑκάτης μελαίνης περι[πέ]σοιτο δαίμοσιν; MAMA X, 52,
Nr. 165; siehe Carboni 2011, 30.
80
PGM IV, 2551: μελανείμων.
81
Paus. 3, 14, 9: θύουσι γὰρ οἱ Κολοφώνιοι μέλαιναν τῇ Ἐνοδίῳ σκύλακα; Plut. Quaest.
Rom. 52: Ὥσπερ οὖν οἱ Ἕλληνες τῇ Ἑκάτῃ κύνα οἱ Ῥωμαῖοι θύουσιν; 111: Ὀλυμπίων μὲν
γὰρ οὐδενὶ θεῶν καθιέρωται [ὁ κύων], χθονίᾳ δὲ δεῖπνον Ἑκάτῃ πεμπόμενος εῖς τριόδους
ἀποτροπαίων καὶ καθαρσίων ἐπέχει μοῖραν; siehe Rohde 1898 II, 407; 1912, 2781; Lautwein
2009, 100; Carboni 2011, 26–27.
82
Seneca Medea 840–841: ter latratus audax Hecate dedit; PGM IV, 2810: ἔχεις σκυλακώ-
δεα φωνήν (an Selene); PGM IV, 2549: κυνολύγματε; siehe Nilsson 1948, 72–73.
83
PGM IV, 2119.
84
Siehe für derartige Erscheinungen Hopfner 1939, 136; Lautwein 2009, 60; Carboni 2011, 30.
85
P. Mich. III, 154: Ἐρεσχιγὰλ παρθένε, κύων, δράκαινα, - - -; PGM IV, 2336: κύων
κυάνεος; PGM IV, 2281: σώτειρα, πανγαίη, κυνώ (an Selene); PGM IV, 2550: λύκαινα; PGM
IV, 1432: Ἑκάτη κυρία, μέλαινα, τ᾽ἐνοδία κύων; siehe Zografou 1997, 177–80.
86
Bărbulescu und Nemeș 1974 = Alicu, Pop und Wollmann 1979, 145, Nr. 387 = IDR III/2,
347.
152 I. PISO UND C. SZABÓ

Abb. 3. Die Votivplatte IDR III/2, 347.

Wir haben es mit Leuten aus den unteren Schichten der Gesellschaft, aber
nicht unbedingt mit einem Sklaven und mit seinem Sohn zu tun.
Weniger klar ist die dargestellte Szene. Dreiviertel des Relieffeldes ist von
einem riesigen Hund eingenommen, der mit geöffnetem Rachen und mit dem
nach oben gewundenem Schwanz nach rechts läuft. Problematisch bleibt das
linke Viertel des Reliefs. Die Editoren, Bărbulescu und Nemeș, erkannten in
der linken unteren Ecke ganz richtig einen Altar. Links vom Hund sei die
Gestalt eines Mannes oder einer Frau zu erkennen, deren Haare mit einer
Schleier bedeckt worden sein sollen. Die Gestalt, deren unterer Teil fehlt, soll
in einer unnatürlichen Stellung, stark gebückt, ihre Hände gegen den Altar
ausstrecken. Die eigenartige Darstellung der Gestalt sei auf die Ungeschick-
lichheit des Handwerkers zurückzuführen. Nach diesen Editoren sollen in die-
ser Szene entweder die Attribute einer Gottheit, die vielleicht mit Hekate zu
identifizieren sei, oder eine Opferszene, in welcher der Hund das Opfer wäre,
dargestellt worden sein.87
Die aggressive Haltung des Hundes spricht klar gegen die zweite Vermu-
tung. Was die eigenartige Gestalt betrifft, ist es zu bemerken, dass man von
vornherein nur die obere Hälfte darstellen wollte und dass sich diese und der
Hund auf derselben waagerechten Linie befinden. Es ist so, als ob beide ein

87
Bărbulescu und Nemeș 1974, 588–90; ähnlich Alicu, Pop und Wollmann 1979, 145; IDR
III/2, ad 347; Bărbulescu 2003, 173.
 DIE GÖTTIN HEKATE IN SARMIZEGETUSA 153

einziges Geschöpf bildeten. Wir haben es höchstwahrscheinlich mit einer dop-

pelgestaltigen Hekate zu tun, als Frau und Hund dargestellt. In einer Zauber-
papyrus wird Hekate ἰσοπάρθενος κύων genannt, was von Preisendanz „Hün-
din in Frauengestalt„ übersetzt wurde.88 Mit demselben Recht kann die Stelle
auf eine Kombination wie in Sarmizegetusa hinweisen. Dass die Göttin selbst
Opfer auf Altären bringt, ist aus vielen Reliefdarstellungen bekannt.89 Die
Annahme von Bărbulescu und Nemeș, man hätte es hier mit einem ritus Roma-
nus zu tun, ist unzutreffend. Opfer an unterirdische Götter werden bei Nacht
und in einem intimen Kreise ausgeführt und sind völlig unkonventionell.90
Was die zwei Editoren als Schleier ansahen, sind wohl die Haare der Göttin.
Da die Votivreliefs gewöhnlich bemalt wurden, kann man ruhig annehmen,
dass die Farbe der doppelten Hekate schwarz war.
Zwischen den gruseligen Schilderungen in den Texten und den, wenn nicht
immer schmeichelhaften, doch zumindest dezenten Darstellungen in der Kunst
besteht ein großer Gegensatz. In Sarmizegetusa haben wir es höchstwahr-
scheinlich mit einer Szene zu tun, welche die Dualität Hekate – Hund, die in
der Literatur und in der öffentlichen Meinung stark vertreten war, am besten
widerspiegelt. Die Form, in welcher erschreckende Bildgedanken, wie wir sie
aus den Zauberpapyri oder aus der Literatur kennen, in Kunst übertragen
waren, hing von der Einbildungskraft und vom Mut des Künstlers ab. Am
einfachsten wäre es gewesen, solche Szenen zu malen, was wahrscheinlich
auch geschehen ist; auch dafür hätte man einen Hieronymus Bosch gebraucht.
In der Skulptur hatte sich darüber noch kein Muster herausgebildet. In Sarmi-
zegetusa scheint man einen solchen Versuch gewagt zu haben.

Einige Schlussfolgerungen

Im spirituellen Universum der Bevölkerung war der Kult der Hekate vermut-
lich viel stärker vertreten, als die wenigen epigraphischen und archäologischen
Quellen aussagen können. Die Handlungen waren für die Öffentlichkeit
zumeist unauffällig, sogar geheim, und materielle Belege sind selten und
bescheiden.

88
PGM IV, 2252.
89
Siehe Bordenache 1969, 55–59; Carboni 2011, 141–49.
90
Für die Vorschriften bezüglich der Kleidung jener, die den unterirdischen Göttern Opfer
darbringen, siehe Hopfner 1928, 362–63; für die allgemeinen Unterschiede zwischen den Opfern
an olympische Götter und den „magischen“ Opfern siehe Scheid 2007, 263–64. Worin die Opfer
für Hekate bestanden, liest man bei Rohde 1898 II, 85, Anm. 1; Heckenbach 1912, 2780–81;
Lautwein 2009, 97–100.
154 I. PISO UND C. SZABÓ

Es gab auch einen offiziellen oder halboffiziellen Kult und er bestand

zumeist aus der Aufstellung von dreigestaltigen Hekataia an Wegekreuzun-
gen.91 Angesichts der kleinen Dimensionen der widerentdeckten Reliefplatte,
ist es unwahrscheinlich, dass sie an einer Wegkreuzung aufgestellt war;92 man
hätte sie kaum sehen oder umgehen können. Sie dürfte unmittelbar neben dem
Hauseingang als Schutzmaßnahme gegen die bösen Geister befestigt worden
sein. Einige Elemente dieser Gattung von Reliefplatten aus Sarmizegetusa
erlauben die Behauptung, dass einige nicht nur in derselben Werkstätte, son-
dern auch von derselben Hand hergestellt wurden. So ist die Art, in welcher
die Göttin die Fackel hält, im wiederentdeckten Relief und im Relief b93 eigen-
artig und identisch.
Die Reliefdarstellungen der Hekate aus Sarmizegetusa und Cioroiu Nou
wurden von Werth einem thrakischen Typus zugeschrieben,94 was insofern
ungerechtfertigt ist, als solche Produkte in großer Zahl auch in Dakien und in
Untermoesien vertreten sind. Andererseits ist das Beiwort placida als eine
Besonderheit gerade des römischen Milieus aus Dakien und Moesien anzuse-
hen. Man kann also höchstens von einer gemeinsamen Erscheinung an der
unteren Donau und im benachbarten Thrakien sprechen.
Die Annahme von Stoian,95 das in Dakien Hekate etwas mit Libera zu tun
hätte, trifft nicht zu. Der Irrtum basiert auf einer falschen Lesung seitens
Mommsen einer Marmorplatte aus Apulum: Trif[ormi] Liberae.96 Die richtige
Lesung der Z. 1 dieser Inschrift, die Stoian unbekannt geblieben ist, lautet aber
[Libero p]atri et Liberae,97 womit sich jede Diskussion erübrigt.
Eine dakische Herkunft der Domna oder Dea placida, wie von Gostar,
selbst mit Vorsicht, angenommen,98 ist unannehmbar.99 Sowohl die

91
Kraus 1960, 103; Johnston 1991, 219–20. Man kennt in Sarmizegetusa zwei Hekataia:
oben, Anm. 8; Alicu, Pop und Wollmann 1979, 75, Nr. 35. Das weitaus berühmteste Hekataion
aus Dakien stammt aus Salinae und befindet sich im Hermannstädter Museum; siehe Petersen
1881, 193–202 = Ferri 1933, 351–53 = Bordenache 1969, 56, Nr. 95 = Römer in Rumänien, 200,
F 40 (Bordenache; David- Țeposu) = Bodor 1989, 1176–77 = Diaconescu 2004 II, 172–76,
Nr. 48 = Carboni 2011, 147–49, II.12.9. Solche Hekataia sind auf eine berühmte Statue des
Alkamenes zurückzuführen: Paus. 2, 30, 2: Ἀλκαμένης δὲ, ἐμοὶ δοκεῖν, πρῶτος ἀγάλματα τρία
ἐποίησε προσεχόμενα ἀλλήλοις; siehe dazu Kraus 1960, 95–118; Carboni 2011, 33; für
Hekataia aus römischer Zeit siehe Kraus 1960, 153–65.
92
Das meint aber Ștefănescu 2002–03, 133.
93
Anm. 29.
94
Werth 2006, 391, 393, siehe auch 378–79, 385–88, 401.
95
Stoian 2004, 184; 2006, 250.
96
CIL III, 1095 = ILS 3268b.
97
IDR III/5, 241.
98
Gostar 1965, 240–41.
99
Zweifel äußerte schon Bordenache 1986, 665; berechtigte Skepsis über die so genannte
interpretatio Romana von dakischen Gottheiten bei Nemeti 2013.
 DIE GÖTTIN HEKATE IN SARMIZEGETUSA 155

Reliefplatte Nr. 1 aus Cioroiu Nou, als auch die neugefundene Reliefplatte aus
Sarmizegetusa stellen ohne jeden Zweifel die wohlbekannte Hekate dar, wäh-
rend das Beiwort placida nicht nur in Dakien, sondern auch in Obermoesien
und in Untermoesien belegt ist.

BIBLIOGRAPHIE

Alicu, D., Pop, C. und Wollmann, V. 1979: Figured Monuments from Ulpia Traiana
Sarmizegetusa (BAR International Series 55) (Oxford).
Audollent, A. 1904: Defixionum tabellae quotquot innotuerunt tam in Graecis Orientis
quam in totius Occidentis partibus praeter Atticas in corpore inscriptionum
Atticarum editas (Paris).
Bărbulescu, M. 1972: ‘Der Dianakult im römischen Dazien’. Dacia n.s. 16, 203–22.
—. 2003: Interferențe spirituale în Dacia romană, 2. Aufl. (Cluj-Napoca).
Bărbulescu, M. und Nemeș, E. 1974: ‘Un relief cu inscripție de la Ulpia Traiana
Sarmizegetusa’. Apulum 12, 587–91.
Bodor, A. 1989: ‘Die griechisch-römischen Kulte in der Provinz Dacia und das Nach-
wirken der einheimischen Traditionen’. ANRW II 18.2, 1077–1164.
Bordenache [Bataglia], G. 1965: ‘Domna Placida’. Studii Clasice 7, 315–18.
—. 1969: Sculture greche e romane del Museo Nazionale di Antichità di Bucarest 1:
Statue e rilievi di culto, elementi architettonici e decorativi (Bukarest).
—. 1986: ‘Domna Placida’. Lexikon Iconographicum Mithologiae Classicae 3.1, 665.
Bräuninger, F. 1937: ‘Persephone’. RE 19, 944–72.
Bruchmann, C.F.H. 1893: Ausführliches Lexikon der griechischen und römischen
Mythologie, Supplement 1: Epitheta deorum quae apud poetas Graecos leguntur
(Leipzig).
Buday, Á. 1916: ‘Monuments de l’époque romaine trouvés en différents endroits de
Transylvanie’. Dolgozatok/Travaux 7.1, 92–133.
Bunsch, E. und Kolendo, J. 2011: ‘Czternaście nieopublikowanych inskripcji z Novae
(Moesia Inferior)’. Novensia 22, 39–66.
Burkert, W. 1977: Griechische Religion der archaischen und klassischen Epoche, 2.
Aufl. (Stuttgart).
Carbó García, J.R. 2010: Los cultos orientales en la Dacia Romana: Formas de
difusión, integración y control social y ideológico (Salamanca).
Carboni, R. 2011: Dea in limine: Culto, immagine e sincretismi di Ecate nel mondo
greco e microasiatico (Tübinger Archäologische Forschungen 17) (Rahden).
Carter, I.B. 1902: Ausführliches Lexikon der griechischen und römischen Mythologie,
Supplement 2: Epitheta deorum quae apud poetas Latinos leguntur (Leipzig).
Daicoviciu, C. und Alicu, D. 1984: Colonia Ulpia Traiana Augusta Dacica
Sarmizegetusa (Bukarest).
Devijver, H. 1976: Prosopographia militiarum equestrium quae fuerunt ab Augusto ad
Gallienum I (Symbolae Facultatis Litterarum et Philosophiae Lovaniensis. Ser. A,
3) (Löwen).
Diaconescu, A. 2004: Statuaria majoră în Dacia romană, 2 Bd. (Cluj-Napoca) (auf
CD).
156 I. PISO UND C. SZABÓ

Dimou, A. 2016: La déesse Korè-Perséphone: mythe, culte et magie en Attique

(Recherches sur les rhétoriques religieuses 18) (Turnhout).
Eck, W. und Ivanov, R. 2010: ‘Zwei Votivinschriften aus Ratiaria in der Provinz
Moesia Superior’. ZPE 174, 201–05.
Ferri, S. 1933: Arte romana sul Danubio (Milan).
Finály, G. 1912: ‘Archäologische Funde im Jahre 1911 – Ungarn’. Archäologischer
Anzeiger, 526–46.
Gerov, B. 1964: ‘Die Rechtstellung der untermösichen Stadt Novae’. In Akten des
IV. Internationalen Kongresses für griechische und lateinische Epigraphik (Wien),
128–33.
—. 1980: Beiträge zur Geschichte der römischen Provinzen Moesien und Thrakien.
Gesammelte Aufsätze (Amsterdam).
Gostar, N. 1965: ‘Cultele autohtone în Dacia romană’. Anuarul Institutului de Istorie
și Arheologie Iași 2, 237–54.
Haspels, H.E. 1971: The Highlands of Phrygia: Sites and Monuments, Bd. 1
(Princeton).
Heckenbach, J. 1912: ‘Hekate’. RE 7.2, 2769–82.
Hopfner, T. 1928: ‘Mageia’. RE 14.1, 301–93.
—. 1931: ‘Orientalisch-Religionsgeschichtliches aus den griechischen Zauberpapyri
Aegyptens’. Archiv Orientálni 3, 327–58.
—. 1939: ‘Hekate – Selene – Artemis und Verwandte in den griechischen Zauber-
papyri und auf den Fluchtafeln’. In Klauser, T. und Rücker, A. (Hrsg.), Pisciculi:
Studien zur Religion und Kultur des Altertums. Franz Joseph Dölger zum sechzig-
sten Geburtstage dargeboten (Münster), 125–45.
—. 1943: ‘Hekate – Selene – Artemis und Verwandte in den griechisch-ägyptischen
Zauberpapyri und auf den Fluchtafeln’. Archiv Orientálni 13.3–4, 167–200.
Jánó, B. 1912: ‘Római emlékek Hunyadvármegyében’. Archeologiai Értesitő 32,
49–57.
Johnston, S.I. 1990: Hekate Soteira: A Study of Hekate’s Roles in the Chaldean
Oracles and Related Literature (American Classical Studies 21) (Atlanta).
—. 1991: ‘Crossroads’. ZPE 88, 217–24.
Kajanto, I. 1965: The Latin Cognomina (Societas Scientiarum Fennica, Commenta-
tiones Humanarum Litterarum 36.2) (Helsinki).
Kehl, A. 1988: ‘Hekate’. Reallexikon für Antike und Christentum 14, 310–38.
Kolendo, J. 1969: ‘Dea Placida à Novae et le culte d’Hécte la Bonne déesse’. Archeo-
logia 20, 77–84.
Kraus, T. 1960: Hekate. Studien zu Wesen und Bild der Göttin in Kleinasien
und Griechenland (Heidelberger kunstgeschichtliche Abhandlungen n.F. 5)
(Heidelberg).
Lautwein, T. 2009: Hekate, die dunkle Göttin. Geschichte und Gegenwart
(Rudolstadt).
Metropoulou, E. 1978: Triple Hekate mainly on Votive Reliefs, Coins, Gems and
Amulets (Athen).
Nemeti, S. 2013: ‘La religione dei Daci in età romana’. In Tauffer, M. (Hrsg.), Sguardi
interdisciplinari sulla religiosità dei Geto-Daci (Freiburg im Breisgau), 137–55.
Nilsson, M.P. 1948: ‘Die Religion in den griechischen Zauberpapyri’. Bulletin de la
Société Royale des Lettres de Lund, 59–93.
Petersen, E. 1881: ‘Der Reliefschmuck der Hekate von Hermannstadt’. Archäologisch-
Epigraphische Mitteilungen 5, 1–84, 193–202.
 DIE GÖTTIN HEKATE IN SARMIZEGETUSA 157

Pflaum, H.-G. 1960–61: Les carrières procuratoriennes équestres sous le Haut-Empire

romain, 3 Bd. (Paris).
Piso, I. 1983: ‘Inschriften von Prokuratoren aus Sarmizegetusa (I)’. ZPE 50, 233–51.
—. 2013: Fasti provinciae Daciae 2: Die ritterlichen Amtsträger (Antiquitas Reihe 1,
Abhandlungen zur alten Geschichte 60) (Bonn).
Rohde, E. 1898: Psyche: Seelenkult und Unsterblichkeitsglaube der Griechen, 2 Bd.
(Freiburg im Breisgau).
Römer in Rumänien 1969: Römer in Rumänien. Ausstellung des Römisch-Germanischen
Museums Köln und des Historischen Museums Cluj (Köln).
Roscher, W.H. 1886–90: ‘Hekate’. In Ausführliches Lexikon der griechischen und
römischen Mythologie Bd. 1.2 (Leipzig), 1888–1910.
Sanie, S. 1974: ‘Dea placida’. Acta Musei Napocensis 11, 111–14.
Sarian, H. 1992: ‘Hekate’. Lexikon Iconographicum Mythologiae Classicae 6.1–2,
985–1018.
Scheid, J. 2007: ‘Sacrifices for Gods and Ancestors’. In Rüpke, J. (Hrsg.), A Compan-
ion to Roman Religion (Malden, MA/Oxford), 263–71.
Ștefănescu, A. 2002–03: ‘Cultul zeiței Hecate în Dacia romană’. Analele Banatului
10–11, 131–40.
Stein, A. 1944: Die Reichsbeamten von Dazien (Budapest).
Stoian [Symonds], C. 2004: ‘Particularités iconographiques de la déesse Hécate dans
la région du Bas Danube’. In Cojocaru, V. und Spinei, V. (Hrsg.), Aspects of
Spiritual Life in South East Europe from Prehistory to the Middle Ages (Iași),
173–93.
—. 2006: ‘Interférences artistiques dans les représentations de la déesse Hécate dans
les provinces romaines du Bas-Danube’. In Mihailescu-Bîrliba, L. und Bounegru,
O. (Hrsg.), Studia historiae et religionis daco-romanae. In honorem Silvii Sanie
(Bukarest), 243–57.
Szabó, C. in Vorbereitung: ‘Romans in the School. Notes on the Archaeological Col-
lection of the Kuun College from Szászváros (Orăștie)’, Ephemeris Napocencis
28.
Thomasson, B.E. 2009: Laterculi praesidum I: ex parte retractatum, 2. Aufl.
(Gothenburg).
Tudor, D. 1942: Oltenia romană (Bukarest).
—. 1958: Oltenia romană, 2. Aufl. (Bukarest).
—. 1968: Oltenia romană, 3. Aufl. (Bukarest).
—. 1965: Sucidava. Une cité daco-romaine et byzantine en Dacie (Brüssel).
—. 1978: Oltenia romană, 4. Aufl. (Bukarest).
Werth, N. 2006: Hekate: Untersuchungen zur dreigestaltigen Göttin (Hamburg).
Zografou, A. 1997: ‘Hécate et Hermès. Passages et vols de chiens’. Uranie. Mythes et
littératures 7, 173–91.
—. 1999: ‘L’énigme de la triple Hécate. De l’entre-deux à la triplicité’. In Batsch, C.
Egelhaaf-Gaiser, U. und Stepper, R. (Hrsg.), Zwischen Krise und Alltag: Antike
Religionen im Mittelmeerraum (Potsdamer altertumswissenschaftliche Beiträge 1)
(Stuttgart), 57–79.
—. 2015: ‘Hécate des rues dans les „papyrus magiques grecs“: des enfers aux
mystères: P. Mich. III, 154 = PGM LXX, 4-19’. In Suárez E., Blanco, M. und
Chronopoulou, E. (Hrsg.), Los papiros mágicos griegos entro lo sublime y lo
cotidiano (Madrid), 135–56.
 CHAPTER 10

PRIVATE INITIATIVE AND BUILDINGS

OF PUBLIC WORSHIP IN ROMAN DACIA*

Ana ODOCHICIUC

Abstract
This paper aims to identify those who donated de sua pecunia for the construction,
restoration or decoration of public cult buildings. It focuses on the epigraphic evidence
mentioning public cult buildings that was discovered in the cities of Dacia.

The art of spatial design and building construction is a field where Romans
have excelled. Today, we are still impressed by the grandeur and majesty of
ruins that symbolise the maiestas imperii. The grandeur of Rome lies not only
in the perfection of the planning and execution of a central policy, but also in
its capacity to encourage the provincial elite to make donations destined to the
construction of public buildings.
This paper seeks to identify those particular individuals who donated de sua
pecunia for the construction, restoration or decoration of public cult buildings
(or of elements thereof), and to propose a hypothesis regarding the underlying
reason for this urban planning initiative. The paper considers settlements
within the province of Dacia, which is also the origin for certain inscriptions
referring to donative acts. Epigraphs mentioning buildings dedicated to private
cults were excluded.

Ampelum

Only one text talks about a private initiative in urban planning at Ampelum.1
The munificent act was that of a clerk within the mining administration

* This work was supported by a grant from the Romanian National Council for Scientific
Research (CNCS – UEFISCDI), project PN-III-P4-ID-PCE-2016-0271, within PNCDI III.
1
IDR III/3, 280.
160 A. ODOCHICIUC

(adiutor tabularii aurariarum Daciarum),2 based in Ampelum. The name of

the benefactor was hypothetically completed as Eutyches. He was an imperial
freedman who erected a temple for the gods Aesculap and Hygia(?). The tem-
ple was probably located where the inscription was found, at Chilia.3

Apulum

The character who founded the construction of the sanctuary dedicated to god-
dess Fortuna Supera remains unknown;4 we do know, however, where it was
found: Partoș.5
At Apulum, a temple was reconstructed by Vibidia Papinia and (?)Tetia
Vibidia, along with a third person.6 The inscription contains the phrase pecu-
nia sua, which stands to show that the restoration of the edifice was done
without using public funds, whereas the epigraph was set up in the honour of
a magistrate. The character to whom the inscription is dedicated can be identi-
fied with one of the decurions who provided support during the city planning
works or, if not, a family member of those who dedicated the inscription.
The decurion and patron of the collegium fabrum at Apulum, P. Aelius
Rufinus, along with his family built a long portico of 12 m.7 He probably took
this initiative while he was patron of the college, given that he insists on men-
tioning this status, but also because the capacity of decurion was added to the
text later.8 Another portico with a length of 9 m was built by the Augustan
P. Aelius Syrus along with his wife Valeria Severa and his daughter Aelia
Syra.9 Another Augustan, M. Gallius Epictetus, made the third portico, which
measured 11 m in length. It probably came from the same construction as the
two preceding ones.10 All three porticos may have belonged to the Asklepieion
near the colony of Apulensis.11 The size of these buildings proves that this
temple would have had considerable proportions.12 The building was probably

2
IDR III/3, 280.
3
IDR III/3, 280.
4
IDR III/5, 1, 78.
5
Rusu-Pescaru and Alicu 2000, 127.
6
IDR III/5, 1, 398.
7
IDR III/5, 1, 6.
8
IDR III/5, 1, 6.
9
IDR III/5, 1, 7.
10
IDR III/5, 1, 13.
11
IDR III/5, 1, 7, 13.
12
Crișan 1971, 342.
 PRIVATE INITIATIVE AND BUILDINGS OF PUBLIC WORSHIP 161

constructed at the end of the 2nd century AD, and the period of peak develop-
ment corresponds to the first half of the subsequent century.13
Another portico was built by the centurion of the legio XIII Gemina,
C. Iulius Lysias, who dedicated the construction to the god Apollo.14 The
inscription was discovered between the city and Partoș.15 The building to
which the monument pertains (probably the sanctuary of the god) has not been
investigated archaeologically, but it is considered part of the two Apulensis
cities. The consensus of opinion is that both the soldiers of the military camp
and the population within civilian settlements used the temple.16
The soldier M. Ulpius Mucianus built a temple with a clock, which he dedi-
cated to I.O.M. and to Juno.17 Scholars have interpreted it in various ways.
D. Tudor believed that the soldier erected only a chapel that featured the
images of the Capitoline gods on the inside and the clock on the outside, and
that this building was placed in a central point of the city.18 D. Alicu argued
that the temple with the clock was placed in the canabae of the legion.19 This
is probably the temple with clocks of the military camp, constructed for the
soldiers of the legion and situated near the castrum.20 R. Etienne has shown
that, in most cases, the setting up of clocks was attributed to official charac-
ters, thus demonstrating the importance of the monument and its relation with
the official or political world.21 The donation took place in AD 193, but the
inscription dates to AD 212–217, thus 20 years later.22 However, M. Ulpius
Mucianus may have changed the inscription when the emperor Caracalla vis-
ited Dacia or when it was dedicated, namely 20 years later.
A temple was erected for the goddess Nemesis by the beneficiarius Teren-
tius Marcianus.23 The inscription that mentions the construction and dedication
of it was discovered in the praetorium of the governor. A. Rusu-Pescaru and
D. Alicu posit the idea of a hall’s having been rearranged into a temple, but
there is no evidence in the entire empire attesting this.

13
Crișan 1971, 346.
14
IDR III/5, 1, 32.
15
Rusu-Pescaru and Alicu 2000, 125.
16
Rusu-Pescaru and Alicu 2000, 125.
17
IDR III/5, 1, 193.
18
Tudor 1964, 294.
19
Alicu 2007, 39.
20
Macrea 1969, 185.
21
Etienne 1992, 361.
22
IDR III/5, 1, 193.
23
IDR III/5, 1, 295.
162 A. ODOCHICIUC

P. Aelius Crescens, duplicarius, built a temple,24 but the fragmentary inscrip-

tion fails to provide further information regarding the god to which the build-
ing was dedicated. The soldier in question is also known from other inscrip-
tions where he worships Mars and Victoria,25 for which reason I believe that
the temple may have actually been built in honour of these two divinities or of
one of them.26
Several inscriptions mention the existence of certain temples dedicated to
Jupiter. An epigraphic item attesting the existence of such a sanctuary belongs
to the Augustan M. Iulius Quirinus, who built the frame and restored the doors
of the entrance to the temple of Jupiter Optimus Maximus.27 This euergetic act
can be included among the ob honorem patronatus, as the inscription contains
this phrase. The temple to be restored by the Augustan was in the colony of
Aurelia.28
Ti. Claudius Anicetus, Augustan of the colonies of Sarmizegetusa and Apu-
lum, built a temple dedicated to the gods Jupiter and Juno,29 The temple is
different from the one within the colony of Aurelia; in this case, D. Alicu and
Rusu-Pescaru place the temple either in the Septimium municipium or in the
canabae of the legion.30

Drobeta

C. Iulius Sabinus, twice second vir of the municipality of Drobeta, restored

and built at his own expense a crypta in honour of his nephew C. Iulius Nae-
sius Sabinus, who obtained the title of flamen.31 The urban planning implica-
tion can be included in the category of ob honorem. The crypta that the inscrip-
tion of Drobeta mentions was a corridor covered by a dome, sometimes
associated with the portico, built underground or semi-underground and that
served as a passage, lit by small windows.32 C. Petolescu ascribes the origin of
the dedicated temple crypta to the Imperial cult.33 Such a construction is also
attested at Apulum as an integral part of the Mithraeum.34 The nephew of the

24
IDR III/5, 1, 375.
25
IDR III/5, 1, 248, 249.
26
Rusu-Pescaru and Alicu 2000, 160.
27
IDR III/5, 1, 152.
28
Rusu-Pescaru and Alicu 2000, 128.
29
IDR III/5, 1, 191.
30
Rusu-Pescaru and Alicu 2000, 129.
31
ILD 54.
32
Petolescu 1983, 69.
33
Petolescu 1983, 69.
34
IDR III/5, 1, 242.
 PRIVATE INITIATIVE AND BUILDINGS OF PUBLIC WORSHIP 163

dedicator is probably the adoptive son of his uncle. This inscription can be
considered proof of the conveying of local power over several generations.
Iulia Maximila fulfilled a vow to Cybele by building a portico pecunia
sua.35 The portico erected by the female worshipper of the goddess Cybele
probably belonged to the temple of the divinity, which was placed on the bank
of the Danube, where the ceremonies took place.36

Micia

C. Iulius Zoticus offered a column to the cult building called ‘La Hotară’,
probably a temple dedicated to Jupiter Optimus Maximus.
Domitia, Vettia and Aux…? built together an iseum. It was considered that
Domitia was the wife of Varenius Pudens, also known at Sarmizegetusa. The
two other dedicants featured along with Domitia do not specify their status,
and the sources remain uncertain. The building was constructed sometime in
the late 2nd or early 3rd century AD.37 We cannot be certain that the three
female benefactors act independently from a legal and financial perspective,
but we can admit their implication in the cult of the goddess.

Samum

Cassius Erotianus, beneficiarius consularis, restored a sacrarium of Neme-

sis.38 The chapel was probably where the dedicant led a statio, being agens in
munere stationis.

Sarmizegetusa Pons Augusti

Aelius Diogenes and Silia Valeria built a temple to the goddess Nemesis pro
salute sua et filiorum suorum.39 The building was leased by the college of
utriclarii, which might have been based at Marga, where the consecrating epi-
graph was discovered, or at Sarmizegetusa (if the inscription had been brought
from there).40 If we assume that the inscription dedicated to the goddess
Nemesis by the two spouses – Aelius Diogenes and Silia Valeria – had come
from Sarmizegetusa, but then it migrated to Marga; in this case, we would

35
IDR II, 26.
36
Rusu-Pescaru and Alicu 2000, 135.
37
Rusu-Pescaru and Alicu 2000, 140.
38
ILD 776.
39
IDR III/1, 272.
40
IDR III/1, 272.
164 A. ODOCHICIUC

have an epigraphic confirmation of a second construction phase of the temple

of Nemesis, east from the amphitheatre41 in the area sacra.42 The building
could have been used as headquarters for this professional association.43 The
divinity was also probably the protector of this college.44 The positioning of
such a place near the amphitheatre seems quite appropriate.45 At Porolissum,
we find the same temple placement.46 The second phase of the building set up
by Aelius Diogenes and his wife Silia Valeria was dated during the reign of
Antoninus Pius.47 Aelius Diogenes had consistent wealth, probably due to sail-
ing and trading with utriclarii, but he does not seem to have had any magis-
tracy at Sarmizegetusa. Regarding his wife Silia Valeria, we only deduce her
involvement alongside her husband in the administration of the familial pro-
fessional association.48

Sarmizegetusa

L. Apuleius Marcus, quaestor and decurion of the colony, restored the temple
of Liber Pater after the invasion of the tribes of hostes.49 I. Piso posits that
when the building was repaired, the magistrate held the position of quaestor.50
In the Roman system there is no proof of payment of summa honoraria by the
holder of such a position; quaestura was not considered a honos.51 Piso pro-
poses that we understand the term scribatus in this inscription as the position
of head of the chancellery.52 The dedicant replaced the name of the magistracy
with his positions, or more probably with his field of activity.53 His euergetic
act was done, as the inscription states, ob scribatum. Whereas there is no evi-
dence of a payment of the legitimate amount for the quaestura, the initiative
can be perceived as a munus,54 manifested permittente ordine. The temple of

41
Rusu-Pescaru and Alicu 2000, 65.
42
Byros 2011, 129.
43
Byros 2011, 130.
44
Macrea 1969, 157.
45
Byros 2011, 130.
46
Rusu-Pescaru and Alicu 2000, 58–61.
47
Rusu-Pescaru and Alicu 2000, 65.
48
Byros 2011, 130.
49
IDR III/2, 11.
50
Daicoviciu and Piso 1975, 161.
51
Dig. 50. 4. 18. 2: Et quaestura in aliqua ciuitate inter honores non habetur, sed personale
munus est. Jacques 1984, 466.
52
Daicoviciu and Piso 1975, 161.
53
Daicoviciu and Piso 1975, 161.
54
Daicoviciu and Piso 1975, 162.
 PRIVATE INITIATIVE AND BUILDINGS OF PUBLIC WORSHIP 165

the god Liber Pater was approximately 300 m north-east of the amphitheatre,55
in the so-called area sacra. L. Apuleius Marcus restored – according to the
inscription – the cubicles and the portico. The first phase of temple construc-
tion dates sometime between the reigns of the emperors Trajan and Hadrian,
while the second phase – also mentioned in the inscription – dates after the
year AD 170.56
The discovery of an inscription fragment makes us believe that Flavius
Graecinus – the patron of the fourth decuria of the college of artisans – took
an urban planning initiative, thus contributing to the temple of Liber Pater.57
The reconstruction of the text remains however a hypothesis.
The freedman Felix – who held the function of adiutor tabularii – built,
restored or decorated a temple that he dedicated to the goddess Diana Augusta,
a hypothesis argued by most specialists.58 Employment of the phrase ob hon-
orem may represent a case of such type of euergetic act, done upon obtaining
a function. The probability of the existence of a temple dedicated to Diana
Augusta that the inscription mentions can be confirmed by other inscriptions
dedicated to her59 and by archaeological research.60
M. Cominius Quintus built a temple along with his wife Antonia Valentina
for the health of Claudia Valentina,61 probably his mother-in-law. The building
is dedicated to a deity that he calls the Queen. This can be identified with Isis,
Juno or Nemesis, but most specialists tend to identify her with the goddess
Isis.62 M. Cominius Quintus is also known from other inscription;63 he was
a notable person in the province: praefectus quinquennalis, pontifex, sacerdos
arae Augusti,64 but also patron of the college of fabri. An equestrian statue
offered by order of the Sarmizegetusa colony to Cominius Quintus contains
the phrase ob merita, which points to the fact that he proved to be generous
towards the city, by ordering the construction of public buildings.65 This
hypothesis is also supported by the discovery of a stamp bearing his name in
the amphitheatre.66 The euergetic act was done in mid-career, probably ob

55
Rusu-Pescaru and Alicu 2000, 57.
56
Rusu-Pescaru and Alicu 2000, 58.
57
IDR III/2, 254.
58
IDR III/2, 198.
59
IDR III/2, 196, 197.
60
Rusu-Pescaru and Alicu 2000, 145.
61
IDR III/2, 19.
62
Rusu-Pescaru and Alicu 2000, 140.
63
IDR III/2, 371, 108.
64
IDR III/2, 108.
65
Diaconescu 2010, 137.
66
Diaconescu 2010, 137.
166 A. ODOCHICIUC

honorem or as a legitimate amount. It was later observed that he exercised

once again the quinquennalitas, and he also became sacerdos arae Augusti.
His urban planning initiative definitely contributed to his obtaining these hon-
ours and functions.
The temple of Aesculap and Hygia at Sarmizegetusa was restored by a cer-
tain Ti. Gr---.67 At Sarmizegetusa, there were several sanctuaries dedicated to
Aesculap and Hygia. They had three construction phases and they were sur-
rounded by an enclosure wall. The four fanum sanctuaries are independent and
they functioned until the Romans’ withdrawal from Dacia.68

Sucidava

Two curiales Sucidavenses restored the temple dedicated to Nemesis. The

inscription does not mention the funds used for restoring the temple, hence we
cannot exclude the possible used of public money through the function he
held. The temple was probably restored in the period AD 248–249. Another
benefactor called Tiberius did work for the temple dedicated to the same
goddess.
What was the reason for the manifestation of such initiatives by all these
provincials: the payment of summa honoraria or euergetic acts ob honorem/ob
liberalitatem?
Whereas there are no accurate data regarding the payment of summa hono-
raria in Dacia, all across the empire donations were usually bigger than the
amount imposed by municipal law.69 This amount was not mentioned in
inscriptions when the final donation was greater than the minimum required;
thus, people actually omitted that part of the construction was done under legal
obligation.70
The ob honorem euergetic acts, voluntary at the beginning, soon became
a moral obligation. The formula had no legal connotation; it was only an
expression suggesting the following idea: for the honos received.71 M. Navarro
Caballero believes that, whereas the notables were morally obligated to make
a contribution for the honours they had received, sometimes the mandatory
character of it was omitted in order to make the epigraphic message nobler.72

67
IDR III/2, 329.
68
Pop 1994, 5.
69
Cébeillac-Gervasoni 1990, 700.
70
Cébeillac-Gervasoni 1990, 701.
71
Navarro Caballero 1997, 130.
72
Navarro Caballero 1997, 130.
 PRIVATE INITIATIVE AND BUILDINGS OF PUBLIC WORSHIP 167

A building engagement taken upon following a promise (pollicitatio) had to

be executed during the year. Starting with the reign of the emperor Trajan,
these pollicitationes became mandatory by law, and their delay led to the
application of an adiectio.73 As shown by F. Jacques, in Africa, executions
related to failure to fulfil a promise in time are comparably more numerous
than those anno suo,74 which demonstrates the existence of a gap between the
law and current reality.75
We cannot make a precise list of the number of ob liberalitatem euergetic
acts in Dacia, mostly because most benefactors were honorati, thus were able
to show off their munificence in a legally mandatory context. We may include
among liberalitas euergetic acts the donations made outside an office or
priesthood.
For freedmen, liberalitas was an excellent means to obtain benefits or a way
to enter certain careers, if not for them, at least for their children. This demon-
strates that slaves/ freedman tended to overcome their humble condition, in
a desire to mitigate their stigma associated with it.76 The euergetic acts of
imperial slaves or freedmen were an expression of wealth and of the function
that they held, given that they were the emperor’s representatives.77
The Augustales are among the most numerous benefactors of the province.
The official or quasi-official character of the college of the Augustales differ-
entiated this institution from the other religious colleges or professional asso-
ciations.78 Upon admission to the college, they paid summa honoraria, to be
used by the municipality.79 Other interventions were sponsored by donations
ob honorem augustalitas.80 Furthermore, the Augustales were all freedmen; on
one hand, they show off their financial potency by setting up monuments,
while on the other they tend to demonstrate their devotion to the Imperial cult,
to the city and to the political power.
Military benefactors were either officers or even miles, and their munificent
acts do not appear to be the results of a mandatory munus, but more likely an
investment for a future career. Their urban planning initiatives must be under-
stood as a wish to be promoted locally, while to obtain a magistracy or flami-
natus perpetuus, a person had to undertake tremendous expenses. Inclusion in

73
Jacques 1984, 746.
74
Jacques 1984, 741.
75
Jacques 1984, 754.
76
Mihailescu-Bîrliba 2004, 78.
77
Mihailescu-Bîrliba 2004, 57.
78
Duthoy 1978, 1281.
79
Duthoy 1978, 1281.
80
AE 1922, 120; AE 1965, 113.
168 A. ODOCHICIUC

the higher social layer also implied an adhesion to its values. Thus, euergetic
acts were a type of behaviour to acquire. Religious euergetic acts can also
demonstrate manifestations of religious devotion. Soldiers and veterans were
an active component of the religious life within the province, mostly concern-
ing the diffusion of Eastern cults. The donations made in the civilian part of
military settlements (canabae) guaranteed a mutually beneficial good relation-
ship between the military camp and the civilian core.81
Euergetic acts by patrons or members of colleges can be ascribed to
a summa honoraria82 paid by these members, just like their civilian counter-
parts, but also to the obtaining of patronage.
Numerous inscriptions mention benefactor priests. Priesthood was a career,
a mixture of honor and munera. The will to become a priest determined
a euergetic act, and often it was interpreted as a morally mandatory munera.83
Upon entering the priesthood, they had to pay summa honoraria or undertake
other euergetic acts for the city.84 Furthermore, priests were also in charge of
temple maintenance.85
Several examples attest the personal contribution of women to the enrich-
ment of the religious architectural setting, independently from their spouses or
from other male persons within their families. Some female donors were the
wives of dignitaries or simple women wishing to open a path for their sons, for
the latter to have entry to certain functions that brought prestige. Women were
not excluded from Eastern cults,86 and in some cases they proved to be wor-
shippers of the deities for which they built temples. The euergetism of women
in Dacia confirms the general trend within the empire by their presence in the
civic and religious life of the province.87
The terms used in epigraphs for cult buildings are classic ones – templum,
fanum, sacrarium and aedes in case of more modest constructions. The temple
serves as the house of the divinity and it is defined by specific architecture,
being protected by Roman law. A sacrarium was a smaller building, a trans-
formed house or chamber used as a sacred place for the divinities of the vari-
ous ethnic groups or small religious groups.88 It is natural to find this public
building in any settlement in Dacia. This construction provides the possibility

81
Pintado 2000, 119.
82
CIL VI, 29691.
83
Dignas 2006, 71.
84
Hemelrijk 2015, 39.
85
Hemelrijk 2015, 89.
86
Sanie 1978, 1099.
87
Byros 2011, 140.
88
Szabó 2015, 124.
 PRIVATE INITIATIVE AND BUILDINGS OF PUBLIC WORSHIP 169

of people gathering around it or for creating unanimitas because the inhabit-

ants are first of all worshippers of the gods.89 Sponsors wished to ensure divine
benevolence, peace and prosperity for the inhabitants of the settlement, as well
as wishing to make known the presence of local deities by erecting temples
dedicated to them. By founding a temple on private property and by dedicating
it to a public cult, the benefactor marked his devotion to the state, and at the
same time, he highlighted the powerful influence of the local elite in the con-
struction of temples dedicated to public cults. The intervention of individuals
in the religious field proved their attachment to the deity; their act was dic-
tated by personal piety, but also by the constraints due to the functions they
held.
The decision to fund various types of buildings does not seem to have been
imposed from the outside, but it adapted to local needs. The types of interven-
tion include works of construction, restoration, decoration or, in certain cases,
works that could not be defined specifically. However, benefactors tended
to build edifices rather than restore them. Reconstruction of a building
was expensive and less attractive: Roman law stated clearly that, in case of
a reconstruction, the name of the previous benefactors would be preserved
(Dig. 50. 10. 7. 1).
We observe mention of the budgetary source – pecunia sua, de suo, etc –
rather than details of the actual amount. Only two inscriptions provide precise
data regarding the costs of these interventions.90 Some prices mentioned in
Africa, Baetica or in Rome – an aedes dedicated to Fortuna Augusta at Tha-
mugadi cost 4400 sestertii, while a temple of Apollo and Diana in Arucci,
Baetica, cost 200,000 sestertii;91 100,000 sestertii was the cost of restoring the
temple of Tiberius92 – are not relevant because they have an exceptional char-
acter and because those involved wished to show off their generosity.
The funding of buildings dedicated to public cults by private persons was
determined by the mandatory character of the fee for the function held (summa
legitima), by promises, by a reward for the honos obtained, as well as by the
attachment to the deity in question. Donations were also initiated due to self-
representational desires and to cultural or cult demands with which the donors
were in contact. They also made donations to maintain their privileged posi-
tion.93 The characters holding the highest magistracies donate both in order to
express their loyalty to their city and in order to open a path towards a brilliant

89
Simion 2009, 100.
90
IDR III/2, 129; IDR III/5, 2, 444.
91
Hemelrijk 2015, 118.
92
Hemelrijk 2015, 119.
93
Mancini 2010, 339.
170 A. ODOCHICIUC

career in the empire.94 Modest euergetic acts can be interpreted as a wish to

imitate social behaviour, to convey the building ideal to future generations,
and to obtain limited notoriety among the members of the public involved
(family, freedmen, clients and loved ones).

BIBLIOGRAPHY

Alicu, D. 2007: ‘Temples attesteés dans les inscriptions de la Dacie romaineʼ. In

Maier, M., Baratta, O. and Almagro-Guzmán, G. (eds.), Acta XII Congressus
Internationalis Epigraphiae Greacae et Latinae (Barcelona), 37–41.
Byros, G.M. 2011: Reconstructing Identities in Roman Dacia: Evidence from Religion
(Dissertation, Yale University).
Cébeillac-Gervasoni, M. 1990: ‘L’évergétisme des magistrats du Latium et de la
Campanie des Gracques à Auguste à travers les témoignages épigraphiquesʼ.
Mélanges de l’École Françiase de Rome. Antiquité 102.2, 699–722.
Crişan, I.H. 1971: ‘Asclepeionul roman de la Apulumʼ. Apulum 9, 341–46.
Daicoviciu, H. and Piso, I. 1975: ‘Sarmizegetusa şi războaiele marcomaniceʼ. Acta
Musei Napocensis 12, 159–63.
Diaconescu, A. 2010. Forurile Sarmizegetusei: O plimbare imaginară prin centrul
politico-administrativ al micii rome de la poalele Retezatului (Cluj-Napoca).
Dignas, B. 2006: ‘Benefitting Benefactors: Greek Priests and Euergetismʼ. L’Antiquité
Classique, 75, 71–84.
Duthoy, R. 1978: ‘Les Augustalesʼ. ANRW II 16.2, 1254–1309.
Étienne, R. 1992: ‘L’horloge de la Civitas Igaeditanorum et la création de la province
de Lusitanieʼ. Revue des Études Anciennes 94.3–4, 355–62
Hemelrijk, E. 2015: Hidden Lives, Public Personae: Women and Civic Life in the
Roman West (Oxford).
Jacques, F. 1984: Le privilège de la liberté. Politique impériale et autonomie munici-
pale dans les cités de l’Occident romain (161–244) (Rome).
Macrea, M. 1969: Viața în Dacia Romană (Bucharest).
Mancini, E. 2010: ‘L’evergetismo municipale in Daciaʼ. In Zerbini, L. (ed.), Roma e le
province del Danubio (Atti del I Convegno Internazionale Ferrara – Cento,
15–17 Ottobre 2009) (Rome), 331–41.
Mihailescu-Bîrliba, L. 2004: Sclavi şi liberţi imperiali în provinciile romane din
Illyricum (Dalmatia, Pannonia, Dacia şi Moesia) (Iaşi).
Navarro Caballero, M. 1997: ‘Les dépenses publiques des notables des cités en
Hispania citerior sous le Haut-Empireʼ. Revue des Études Anciennes 99.1,
109–40.
Petolescu, C.C. 1983: ‘Municipium Hadrianum Drobeta într-o inscripție recent
publicatăʼ. Apulum 21, 67–70.
Pintado, J.A. 2000: ‘El comportamiento munificente de las élites hispanoromanas en
materia religiosa: la construcción de templos por iniciativa privada en „Hispania”ʼ.
Iberia. Revista de la Antigüedad 3, 111–28.

94
Pont 2010, 413.
 PRIVATE INITIATIVE AND BUILDINGS OF PUBLIC WORSHIP 171

Pont, A.V. 2010: Orner la cité: les enjeux culturels, sociaux et politiques de la
construction et de la restauration des monuments publics dans les cités d’Asie et
de Pont-Bithynie, du Ier au IVème siècle ap. J.-C. (Scripta Antiqua 24)
(Bordeaux).
Rusu-Pescaru, A. and Alicu, D. 2000: Templele romane din Dacia (I) (Deva).
Sanie, S. 1978: ‘Deus aeternus et Theos Hypsistos en Dacie romaineʼ. In de Boer, M.
and Edridge, T.A. (eds.), Hommages à Maarten J. Vermaseren: Recueil d’études
offert … à l’occasion de son soixantième anniversaire le 7 avril 1978 (Études
préliminaires aux religions orientales dans l’Empire romain 68), vol. 3 (Leiden),
1092–1115.
Simion, M. 2009: Evergetism și politică edilitară în provinciile romane de la Dunărea
de Jos (sec. I–III p. Chr.) (Dissertation, Bucharest).
Tudor, D. 1964: ‘Les constructions publiques de la Dacie romaine d’après les
inscriptionsʼ. Latomus 23.2, 271–301.
 CHAPTER 11

SILVER-PLATED COINS
IN THE SITES OF ROMAN DACIA*

Lucian MUNTEANU and Ștefan HONCU

Abstract
The paper deals with a particular category of Roman silver coins. Plated denarii and
antoniniani are frequently encountered in monetary circulation in Dacia. The number
of sites containing such coins is significant and their distribution appears relatively
uniform throughout the province. They occur successively in all periods and their
chronological trend is similar to that of the genuine issues. The situation of the plated
silver coins found in other Roman provinces of Danube and Rhine areas appears to be
different. They were found in smaller quantities, and seem to be concentrated in certain
sites (usually where there is solid evidence for the functioning of local minting work-
shops) and in certain periods. The comparison highlights the particularity of silver-
plated coins in the province of Dacia. Both types of silver coins (plated and genuine)
appear consistently in the same archaeological contexts, being accepted, used and lost
in a similar manner. We believe it possible that a great deal of plated denarii and
antoniniani in Dacia originated in the official workshops of the empire.

Among the Roman finds at sites within Roman Dacia, silver-plated coins
(denarii and antoninianii) are frequently encountered. The presence of these
particular denominations in the monetary circulation in this province attracted
the attention of scholars as early as the beginning of 20th century.1 For a long
time, most of the numismatic works treated this subject in a very general man-
ner. Rigorous sequential studies that propose solutions in the context of the
Roman world are quite recent.2

∗
This work was supported by a grant from the Romanian National Council for Scientific
Research (CNCS – UEFISCDI), project PN–III–P4–ID–PCE–2016–0669.
1
Macrea 1932, 130.
2
Ardevan 2002; Găzdac 2007; 2009; Găzdac and Zăgreanu 2008; Găzdac and Gaiu 2011;
Găzdac and Cosma 2014; Pîslaru 2009, 51–74; Gaspar 2015.
174 L. MUNTEANU AND Ş. HONCU

Plated silver coins in Dacian sites consist of two kinds of counterfeit pieces.
The most numerous are the subaeratae coins, made of base metal and dressed
in a coat of precious metal, made mainly by striking or by a ‘silver bath’.3
Another category is the casting technique, which is rarely encountered, being
reported in Napoca, Gilău and Potaissa.4 A particular situation is mention of
the so-called ‘billon’ denarii, which suggests issues of a very low silver con-
tent. When such mentions appeared, we had reservations in regarding them as
counterfeits, although some of them may represent the bronze core of some
plated coins.5 The overall appearance of copy issues rarely allows for differen-
tiation from original models. In a few cases, there appear novel coinage types,
hybrid coins or signs of ‘barbarisation’ (awkward style, inscription errors).6

Plated Denarii

At the present publication stage of numismatic finds, the number of plated

denarii registered in Dacian sites is very high. But every reassessment of older
finds seems to offer surprises by identifying large quantities of such coins.7
We will briefly analyse the finds of plated denarii from Dacia’s civilian and
military sites (see Appendix, I), following the share of such finds within the
total number of denarius finds registered for these sites, as well as their
chronology.
Concerning the settlements, almost a quarter of uncovered denarii are coun-
terfeits. The fact that such issues are present in nearly 90% of the civil sites
selected for analysis (Table 1) is quite significant. The differences between
urban and rural environments are irrelevant. Most such discoveries originate in
Porolissum (municipium), Apulum (municipium and colonia), Micăsasa,
Samum and Arcobadara. Only the lists of numismatic finds of Cristeşti,
Bumbeşti and the villages of Dacia Malvensis are free of copy coinage.8

3
A distinction between the two techniques is made by Pîslaru 2009, 52, 355–58.
4
The identification of cast copies at Potaissa is based on rigorous metallographic analyses
(Pîslaru 2009, 66, 363–65). Also, the hoard ‘Apulum VI’ consists entirely on cast denarii of
bronze core (Găzdac, Oargă and Alföldy-Găzdac 2015).
5
Pîslaru 2009, 54; many such coins (made of ‘billon’ and of ‘potin’) were found in the sites
of Tibiscum (Matei 2015, 110–11, 117–18, 132, 136, 140–41) and Porolissum (‘Sub Ferice’)
(Opreanu and Lăzărescu 2015, 84, 90, 93, 108, 110–11, 114–15, 117–18, 122, 132–33); some of
the plated denarii discovered at Gârla Mare were made by casting (Oberländer-Târnoveanu 2008,
253, 261).
6
Chirilă 1991, 170; Ardevan 2002, 41; Pîslaru 2009, 61–64; Gaspar 2015, 68–70.
7
Găzdac 2007, 598–99; Găzdac and Zăgreanu 2008; Găzdac, Gaiu and Alföldy-Găzdac
2015.
8
Until the recent publication of the numismatic corpora of Tibiscum (Matei 2015) and
Drobeta (Găzdac et al. 2015), both could be included in this category.
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 175

Apparently, one may assert a high frequency of such issues for settlements
located in the northern half of the province, within the Carpathian arch. How-
ever, one must keep in mind that most of the sites in this area have benefited
from systematic excavation and publication. When the same became true for
southern settlements (for example Orlea), the number of copy coinage finds
increased significantly.
In the fortifications in Dacia, the overall weight of plated denarii is higher
compared to the one, calculated above, of the settlements (Table 1).

Table 1: The share of denarii subaerati (D*) in the sites of Dacia.

Most of these coins were found in the military sites of Porolissum (‘Pomet’
and statio), Drobeta and Apulum. The finds in their associated civilian sites
have provided similar values, except for Drobeta. In this latter case, even in
the context of a recent re-evaluation of the finds, one should take into account
that the sample of fortification finds is smaller. The only find-lists free of such
issues are those of Resculum and Răcari. As in the case of settlements, the
number of plated denarius finds seems to be much larger in castra located in
the northern half of the province, as these sites are much better excavated and
reported.
Based exclusively on the qualitatively unequal samples available, the distri-
bution of Dacia’s civilian and military sites with denarii subaerati can be
regarded as quasi-homogenous within the province. This distribution seem to
be indifferent to the status and size of the sites, their geographical location, the
proximity of the army or the quantity of genuine denarius finds.
There are some interesting findings made by studying the chronology of the
plated denarii found in Dacian sites.9 Coins of this kind occur in all

9
We employed a chronological scheme divided into 17 periods, based mainly on the reigns
of the major emperors.
176 L. MUNTEANU AND Ş. HONCU

chronological periods (of denarius circulation), including the one prior to

Roman conquest (Table 2, Figs. 1–2).

Table 2: The chronological distribution of the share of denarii subaerati (D*)

in the civilian settlements (Civ) and military fortifications (Mil) of Dacia,
Pannonia Inferior, Pannonia Superior, Noricum, Raetia, Germania Superior,
Germania Inferior and Gallia Belgica (AD 98–275).

Fig. 1. The chronological distribution of the share of denarii subaerati

in the civilian settlements of Dacia, Pannonia Inferior, Pannonia Superior,
Noricum, Germania Superior and Gallia Belgica (AD 98–244).
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 177

Fig. 2. The chronological distribution of the share of denarii subaerati

in the military fortifications of Dacia, Pannonia Inferior, Pannonia Superior
and Germania Superior (AD 98–244).

On almost every occasion the percentage of finds in military sites is higher

than within settlements. For the period of the Antonine dynasty, plated denarii
average between 15% (settlements) and 25% (castra) of the total value of
these silver denominations. The proportions grow, significantly, even dou-
bling, during the reign of the first emperors of the Severan dinasty. Most plated
denarii discovered in Dacia bear the effigies of Septimius Severus and Cara-
calla (and their families). A smaller number of them are issues of Elagabalus
and Severus Alexander. Even the few denarii issued by Gordian III were imi-
tated on a smaller scale. It is interesting to note that, in all the periods studied,
the chronological distribution of plated denarii follows closely the trend of
genuine issues.10
Consequently, we attempted to discern the extent to which the issue of
denarii subaerati in Dacia is a peculiar one specific to this area or whether
there is a connection to similar situations in other parts of the empire. To this
end, we compared the discovery of plated denarii on Dacian sites to those in
other parts of the Roman world. In terms of comparison, we considered the

10
See also Găzdac 2010, 149, pl. Q3.
178 L. MUNTEANU AND Ş. HONCU

Danubian and Rhenish provinces: Pannonia Inferior, Pannonia Superior, Nori-

cum, Raetia, Germania Superior, Germania Inferior and Gallia Belgica (only
the part related to the Limes Germanicus) (Appendix, II–VIII). The results
produced by the analysis of plated denarii uncovered on the sites of these
provinces seem to be different from the data obtained in the case of Dacia.
These differences are apparent, especially in the quantified form (figures). The
average values, calculated for each province, indicate, in an undeniable man-
ner, that the largest amount of plated denarii originate in the area north of the
Lower Danube. The differences are much deeper and must be identified at
the level of particular elements that make up the general picture. In the case of
Dacia, we have noticed a somewhat consistent spatial distribution of denarii
subaerati, which were found in almost in all the sites in the province. The
image of the discoveries in the other provinces seems to be a different one.
In these cases, one can notice that the lists of numismatic finds that do not
include such silver-plated pieces are more numerous. Also, the distribution of
copy coinage within these provinces seems to be uneven. Most plated denarii
are visibly concentrated on certain sites, while others have provided only a few
finds. In some cases, the concentrations of counterfeit coins appear in centres
where we have solid evidence of the local production of such coinage. It is at
least surprising that the share of such pieces in the monetary circulation in
some of these sites is inferior to the similar values calculated for several
Dacian sites, well documented from a numismatic point of view.11
The chronological distribution of plated denarii of the analysed samples is
very diverse one (Table 2, Figs. 1–2). For almost all the periods, the position
of the Dacian sites is higher. The differences are even substantial in the case
of military sites. Each of the groups of plated denarii pertaining to the six
studied provinces has particular features, making it difficult to identify com-
mon traits. To a certain extent, it is possible to accept a grouping of indicated
values for the Severanic period. This concentration is more visible in the Pan-
nonian and Dacian sites. But, as pertinently suggested elsewhere, it is possible
that the comparative chart does not indicate the real fluctuations of these

11
The data of such a comparison is as follows: Apulum – castra (46% of of the total number
of denarius finds are plated), canabae (39%), colonia (41%), municipium (41%), Porolissum –
castra (48%), municipium (57%) (Dacia); Augusta Raurica (38%), Carnuntum – castra (55%),
canabae (33%), colonia (23%), Flavia Solva (15%), Mogontiacum (13%), Augusta Treverorum
(14%) (extra-Daciae). One should note that the supply of legal tender in most of these sites out-
side of Dacia was relatively (re)assessed and (re)published, according to solid scientific stand-
ards, so that the methodological deficiencies of the research cannot be invoked (as in the case of
older publications pertaining to Dacia).
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 179

monetary issues, due to a presumptive chronological gap between prototypes

and their copies.12
The comparison results highlighted the particular situation of plated denarii
in the province north of the Lower Danube. Most data currently available seem
to document the premise of an unusually intense phenomenon in this area. One
could assume only its temporal and spatial coordinates. The main clue that can
be used to date these pieces is their chronological distribution. These pieces
are present in all periods, as later as Gordian III. For this whole timeframe,
their chronological trend consistently follows the one of the coins they imitate,
a situation that has proved singular. A possible gap between the two kinds of
issues would have generated disproportion, which would also been apparent in
other situations as well. The absence of these disproportions would suggest
that original and copy coinage penetrated the provincial monetary circulation
at very close or even simultaneous dates. Analysis of the archaeological con-
texts in which the plated denarii were discovered provides the verification of
such a possibility. Thus, in the auxiliary fort of Arcobadara, a denarius subae-
ratus, with the effigy of Trajan, was uncovered in a stratigraphic position that
supports its loss in the first half of the 2nd century AD.13 A whole series of
early monetary issues of the same kind (from Vespasian to Antoninus Pius)
was assigned to the first building phase (from the ground up) of the Porolis-
sum customs house, dated in the middle of the 2nd century.14 Finally, similar
pieces, from Trajan, are present in the earliest habitation level of the Buciumi
fort.15 We believe that these archaeological situations, which were documented
only in military environments, are relevant to all categories of sites in the
province.
Generally speaking, the phenomenon of copying the legal tender (especially
by casting and, more seldom, by striking) during the 1st and 3rd centuries AD
was regarded as being characteristic of the western Roman border provinces.
Their issuance would have occurred during times of crisis, which affected the
regular supply of official currency. These copies produced, apparently, with
the tacit approval of the authorities, were accepted in the (local?) monetary

12
Găzdac 2002, 69, 84; 2010, 149, 190. The examples provided by the author, of the contexts
of Augusta Raurica, Rumst and Saint Mard, are significant arguments in this respect.
13
Ardevan 2002, 42, n. 25; Găzdac, Gaiu and Marchiș 2011, 18, 72, no. 73 (denarius subae-
ratus of Trajan was discovered at -0.70 m and pertain to the second phase of the fort, when the
earth construction was enlarged during the reign of Hadrian).
14
Gudea 1996, 41.
15
Gudea 1997, 26–27, 35; Găzdac and Pripon 2012, 19, 26–27, 44, no. 128, 52, no. 185.
Phase I was identified at a depth of -1.20 to -0.90 m and was dated between AD 106 and 114/5
(N. Gudea) or, more plausibly, between AD 106 and the beginning of Hadrian’s reign (C. Găzdac,
E. Pripon).
180 L. MUNTEANU AND Ş. HONCU

circulation until the resumption of normal monetary flow.16 Such a particular

episode was identified in the beginning of the 3rd century. During this period
of crisis, large-scale copying of the main bronze and silver denominations,
especially in the Danubian and Rhenish provinces. A large number of finds are
evidence for the functioning of local minting workshops in these areas. Most
of the materials identified here – clay casting moulds, dies, scrapped pieces,
etc. – bear effigies of the Severan emperors and their families.17
In the Romanian literature, supposed counterfeiting workshops for silver
(and bronze) coinage were located in Dacia, in Porolissum,18 Arcobadara,19
Apulum20 and Potaissa.21 The selection of these sites was made mainly on the
basis of the relative weight of copy issues, of different quality, documented by
the archaeological inventories. The most consistent argument is recent and is
based on finds from the legionary fortress of Potaissa. It consists in the identi-
fication of five possible scrapped pieces and three pondera examinata, together
with evidence of bronze-casting activities.22 Each of these three cases is ques-
tionable and does not prove the certain existence of a workshop within this
military site. Moreover, in the present stage of knowledge, we believe that
there is no clear discovery throughout the northern Lower Danube territory,
which would undoubtedly prove the local production of copies of Roman
denarii.
In these circumstances, it is likely that the explanations of the significant
volume of plated denarii in Dacia must be sought in other areas. The chrono-
logical distribution and quasi-consistent diffusion within the province of these
issues indicate that they circulated in parallel with the original pieces for the
whole duration of Roman rule. These two monetary categories appear consist-
ently in the same archaeological contexts, being accepted, used and lost in
a similar manner.23 In the absence of clear evidence of local minting, one
could safely assume that at least some of the copy coinage originate in official

16
For the main discussions on this subject, see Aubin 2003; Boon 1974; King 1996; Kunisz
1980; 1987; Peter 1990; 2001.
17
King 1996, 259–63, maps 1, 2; Lallemand 1994, 162–73, fig. 7.
18
Winkler 1964, 220–21; Winkler and Ivănescu 1977, 88–93.
19
Ardevan 2002.
20
Suciu 2000, 328; Găzdac, Oargă and Alföldy-Găzdac 2015, 15 (this presumed workshop
might have been located in the vicinity of the discovery site of the hoard, which, unfortunately,
has not been located with accuracy).
21
Pîslaru 2009, 71–74.
22
Pîslaru 2009, 71–74.
23
This important finding, based on a careful study of the actual archaeological contexts in
Potaissa, was made by M. Pîslaru (Pîslaru 2009, 60).
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 181

workshops.24 Such a possibility would explain most of the similarities that

characterise the behaviour of the two monetary species (genuine and copied),
at least in the Dacian sites. The technique of plating (by striking) in which the
copies were made involves high costs, difficult to be covered by provincial or
municipal authorities. The casting alternative was much more accessible,
which was why it was widely employed in local copying within the Western
provinces.
These considerations concern only the territory of Dacia, as the most repre-
sentative finds originate here. Plated denarii appear also on Pannonian sites,
without clear indications of the existence of local minting workshops in which
such coinage might have been produced. These copies may have been aimed
particularly at the frontier provinces of the empire, with the purpose of sup-
plementing the amount of silver money. In fact, they would have been difficult
to distinguish from genuine coinage, especially when the latter was devalued.25
A recent hypothesis ascertains that plated silver coins would have been used to
pay troops in Dacia (especially auxiliary troops).26 But the large number of
such pieces uncovered within the two castra legionis (Potaissa – 39%, Apulum
– 46%) seems to challenge this assertion. Moreover, one should build a solid
argument for the exclusive association of these issues with a military environ-
ment, given that the calculated values for settlements and fortifications are
quite close. Denarii subaerati are well represented in two civilian sites, where
the military influence is minimal: Sarmizegetusa (26%) and Napoca (37%).

Plated Antoniniani

The presence of these issues, produced by means of the plating technique, is

well documented in the civil and military sites of Dacia. These represent, on
average, ca. 20% of all finds of radiate silver coins. It is difficult to appreciate
the extent to which this percentage could be increased, in light of a reassess-
ment of older finds, as in the case of denarii.27 Moreover, as a result of the

24
See the situation of plated denarii present in the late Republican and early Imperial Roman
hoards in the region east of Carpathians, which were regarded as products of the Roman official
mint, circulating in the Roman world, whence they later arrived in the Barbaricum (Mihailescu-
Bîrliba 1980, 38–39; 1990, 108–10).
25
Katsari 2011, 161.
26
Winkler 1964, 220; Găzdac 2007, 602; 2009, 1493 (for auxiliary troops); 2010, 200;
Găzdac, Oargă and Alföldy-Găzdac 2015, 14. For a similar assumption, used as an analogy, see
Augusta Raurica (Peter 1990, 74–77).
27
The recent reassessment of the numismatic finds from the urban settlement of Drobeta led
to the identification of only one plated radiate specimen (out of a total of 116 antoniniani)
(Găzdac et al. 2015, 154, no. 173).
182 L. MUNTEANU AND Ş. HONCU

accelerated depreciation of this nominal, as of Valerian’s’ reign,28 the profits

made by copying these issues would have been insignificant. More than half
of the documented sites in Dacia contain plated antoniniani, in various per-
centages (Table 3).

Table 3: The share of antoniniani subaerati (A*) in the sites of Dacia.

The largest groups come from the province’s urban centres and castra located
within the Carpathian arch (Porolissum – castra and municipium, Apulum –
colonia and municipium, Potaissa – castra and colonia, as well as Sarmizege-
tusa). In many of these intra-Carpathian sites (such as Porolissum ‘Sub Ferice’,
Certiae, Samum, Gherla, etc.), the very large percentages of plated antoniniani
represent in fact a rather small number of such coins. In the extra-Carpathian
areas of the province the situation seems to be different. Here the radiate silver
coins are very numerous, especially in the numismatic inventories of the sites
located in the vicinity of the Danube. However, only a very small part of these
finds are plated. There must be plausible explanations for this quasi-absence of
antoniniani subaerati in southern Dacia. It is possible that those who have
published these finds have not properly identified them due to their poor state
of preservation and the resemblance to genuine coins of poor quality.29 One
should also consider the possibility that the southern part of the province has
been supplied with later coinage from a different source.

28
For a presentation of title and the weight reduction of these denominations, up to Aure-
lian’s reform, see Harl 1996, 130, tabl. 6.2; Carson 1990, 94, 105, 117.
29
This discrimination is regarded as a very difficult process, even in the case of well-
preserved pieces within the composition of the hoard (Ardevan, Suciu and Ciugudean 2003, 36).
In the Sucidavan coin inventory, all late antoniniani are recorded with aes, only in three cases
their silver plating is mentioned (Poenaru Bordea 1975, 71, 88). Similar pieces from Mediaş were
considered ‘small bronzes’, only the specialist catalogue reference allowed their correct identifi-
cation (Chirilă et al. 1967, 458, nos. 29–30).
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 183

The oldest plated antoniniani appear in a military environment and are con-
temporaries of plated denarius finds of Potaissa and Porolissum inventories,
dated to AD 211–222. However, most such coins are concentrated in the reigns
of Gordian III and Philip the Arab (Table 4, Figs. 3–4).

Table 4: The chronological distribution of the share of antoniniani subaerati (A*)

in the civilian settlements (Civ) and military fortifications (Mil) of Dacia,
Pannonia Inferior, Pannonia Superior, Noricum, Raetia, Germania Superior,
Germania Inferior and Gallia Belgica (AD 98–275).

Fig. 3. The chronological distribution of the share of antoniniani subaerati in the

civilian settlements of Dacia, Pannonia Inferior, Pannonia Superior, Noricum and
Germania Superior (AD 211–275).
184 L. MUNTEANU AND Ş. HONCU

Fig. 4. The chronological distribution of the share of antoniniani subaerati in the

military fortifications of Dacia, Pannonia Inferior, Pannonia Superior and Germania
Superior (AD 211–275).

For the first of these intervals (AD 238–244), the percentage calculated for
fortifications (40%) exceeds by several times the corresponding one for settle-
ments (13%). In the next chronological range (AD 244–249), the differences
between military and civilian environments decreases (33% vs 20%). After
this date, the two sets of values progressively decrease, reaching zero for the
reign of Claudius II. The very high percentage (50%) for the reign of Aurelian,
in military environment is based on only one plated coin.30 The appearance of
such a deformed image occurs in the context of the drastic lack of numerical
data.
The chronological trend of the plated antoniniani overlaps that of the genu-
ine radiates only for the first half of the 3rd century AD. Consequently, the
two separate, following divergent tendencies. Explanations for this situation
must be sought in the particular historical (and monetary) evolution of Dacia
during the last period of Roman rule north of the Lower Danube. During the
sixth decade most of the fortifications and settlements in the province’s terri-
tory ceased to exist. During this period the antoniniani appear abundantly

30
Pîslaru 2009, 220, no. 749 (about this find we do not with certainty if it is plated, nor if, at
least, it is an issue of this emperor).
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 185

within Danubian sites (especially Drobeta and Sucidava), but the relative
weight of the plated pieces among them is insignificant, as of the current state
of publication. Outside of this southern area, the radiate silver denomination
dated in this period appear, in much smaller quantities, in the major urban
centres of the province, along the imperial road (linking Drobeta to
Porolissum).31 However, the number of plated coins is very small within these
assemblages.
Quite recently, a large hoard was found in Apulum (municipium), (‘Apulum
VII’),32 which included a large number of plated antoniniani. We have drawn
attention to the chronological structure of this hoard. With the exception of
Aurelian, the plated radiates appear in all periods, defining an ascending
trend.33 Its values overlap, without exception, the chart of the chronological
evolution of this denomination within the sites of Dacia.34 We believe that this
discovery could confirm that the antoniniani subaerati existed in the province
even in this period at the end of Roman rule, but the particular political and
economic context prevented their distribution.35
The finds of plated antoniniani in the Danubian and Rhenish provinces are
presented in a special manner (Table 4, Figs. 3–4). The overall share of such
coins, when appearing, is very low in all provinces. Only a small number of
civilian and military sites in both Pannoniae (6), in Noricum (4) and Germania
Superior (8) include plated radiates in their inventories of numismatic finds.36
As of the chronological distribution of such coins, this seems to fluctuate in
certain periods from one area to another. A relative concentration could be
identified between AD 238 and 253, overlapping most of the Dacian scheme.
Only in very few sites, all newly republished (Carnuntum – colonia, castra
and cannabae and Flavia Solva), the plated antoniniani are represented in
almost all chronological ranges, although in quite limited quantities. In the

31
Ardevan, Suciu and Ciugudean 2003, 43, 160, fig. 3.
32
Ardevan, Suciu and Ciugudean 2003, 49–124.
33
See, for comparison, the ascending evolution of the relative weight of plated antoniniani,
dated AD 238–282, from a consistent sample of 65 hoards discovered in Roman provinces (Estiot
2004 I, fig. 3).
34
Ardevan, Suciu and Ciugudean 2003, 127, annexe 3; for comparison, see Găzdac 2010,
pl. Q4.
35
R. Ardevan speaks of a relative immobility of the coin in society, in the context of distur-
bance to the province’s internal economic circuits. In this period, the imperial currency is distrib-
uted only in the big centres, without reaching the general populace (Ardevan, Suciu and
Ciugudean 2003, 43–44).
36
The recent reassessment of the numismatic finds of Carnuntum led to the reconfiguration
of the relative weight of plated antoniniani (FMRÖ III/1 – 3.6% of the total antoniniani were
plated, FMRÖ III/2 – the percentage reached close to 6%), which situation could apply also for
other sites in this area.
186 L. MUNTEANU AND Ş. HONCU

comparative chart of the provinces, the values of Dacia are, for most periods,
placed at the upper limit.
The situation of the plated antoniniani in the province of Dacia seems to be
a particular one within the Roman world (at least in the Danubian and Rhenish
areas). Their overall share in the circulation of silver coinage is distinct. The
number of sites in Dacia containing such coins is significant and their distribu-
tion appears relatively uniform throughout the province. For a wider period,
their chronological trend is similar to that of the original coins. The differences
that emerge in the final period of Roman rule are explicable by the particular
monetary context of this era. However, the general behaviour of the plated
antoniniani discovered in civilian and military sites in Dacia is similar to that
of the original issues. One should not rule out that both genuine and copy can
originate in the same Imperial minting workshops. Such a hypothesis, quite
plausible, has recently been proposed, on different bases, by the publishers of
the Apulum hoard (‘Apulum VII’).37
There are many similarities between plated denarii and antoniniani of the
civil and military sites of Dacia. Both types of copy coins are well represented
within the numismatic finds of almost the whole province and show a chrono-
logical distribution similar to that of the genuine issues. It seems that plated
and genuine coinage circulated together, being accepted, used and lost in
a similar manner. It is possible that a great deal of plated denarii and antoni-
niani in Dacia originate in the official workshops of the empire.

APPENDIX

List of sites used in numismatic research (all sites have more than 40 coins
registered).
I. Dacia: 1. civilian settlements: a) cities: Alba Iulia (Apulum – colonia and muni-
cipium), Cluj-Napoca (Napoca), Corabia (Sucidava), Drobeta-Turnu Severin (Drobeta),
Jupa-Iaz (Tibiscum), Moigrad (Porolissum), Orşova (Dierna), Reşca (Romula),
Sarmizegetusa (Ulpia Traiana Sarmizegetusa), Turda (Potaissa); b) rural settlements:
Cioroiu Nou (Aquae [?]), Cristeşti, Gârla Mare, Micăsasa, Orlea, Group of rural sites
(19) from Dacia Porolissensis, Dacia Apulensis (45 sites) and Dacia Malvensis
(29 sites); c) military settlements: Alba Iulia (Apulum), Bumbeşti-Jiu, Căşeiu
(Samum), Gherla, Ilişua (Arcobadara), Moigrad ‘Sub Ferice/Drumul Vacilor’ (Porolis-
sum), Romita (Certiae), Stolniceni (Buridava), Veţel (Micia); 2. military fortifications:
Alba Iulia (Apulum), Bologa (Resculum), Buciumi, Căşeiu (Samum), Drobeta-Turnu
Severin (Drobeta), Gherla, Gilău, Ilişua (Arcobadara), Jupa (Tibiscum), Moigrad
(Porolissum – ‘Pomet’ and statio portorii), Răcarii de Jos, Slăveni, Turda (Potaissa).

37
Ardevan, Suciu and Ciugudean 2003, 37–40.
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 187

Bibliography (select): Ardevan 1991; 1993; 2005; Chirilă 1991; Chiţescu and Poe-
naru Bordea 1983; Dudău 2006; Găzdac 2002; 2003; 2010; Găzdac and Cociş 2004;
Găzdac, Gaiu and Marchiş 2011; Găzdac and Gudea 2006; Găzdac and Isac 2007;
Găzdac and Pripon 2012; Găzdac, Suciu and Alföldy-Găzdac 2009; Găzdac, Voişian
and Cociş 2003; Găzdac et al. 2015; Matei 2015; Mitrofan and Ardevan 1997; Ober-
länder-Târnoveanu 2008; Opreanu and Lăzărescu 2015; Petac 2011; Pîslaru 2010;
Poenaru Bordea 1975; 1998; Stîngă 1998; Winkler 1964; 1965; 1975; Winkler and
Băloi 1971; 1973; Winkler and Hopârtean 1973.
II. Pannonia Inferior: 1. civilian settlements: Budapest (Aquincum – colonia and
canabae), Dunaújváros (Intercisa), Pécs (Sopianae), Százhalombatta (Matrica), Tác
(Gorsium – municipium and vicus); 2. military fortifications: Budapest (Aquincum),
Dunaújváros (Intercisa), Százhalombatta (Matrica).
Bibliography: Bíró-Sey and Lányi 1980; 1981; 1982; 1983; 1985; 1986; 1989;
Bíró-Sey, Lányi and Torbágy 1993; Fitz 1978; FMRU I; Fülep 1984; Găzdac 2002;
2003; 2010; Pekáry 1953; Thúry 2000.
III. Pannonia Superior: 1. civilian settlements: Árpás (Mursella), Drnovo (Nevio-
dvnvm), Eisenstadt, Epöl-Kőkút, Halbturn, Hegykő-Téglástó, Héreg, Illmitz, Marz,
Müllendorf, Neckenmarkt, Neusiedl am See, Oslip, Petronell (Carnuntum – colonia
and canabae), Pöttsching, Ptuj (Poetovio), Schützen am Gebirge, Sopron (Scarbantia),
Strebersdorf, Szombathely (Savaria), Szőny (Brigetio), Tokod, Wien (Vindobona –
colonia and canabae), Winden am See; 2. military fortifications: Ács Bum-Bum Kút
(Ad Mures), Esztergom (Solva), Győr (Arrabona), Petronell (Carnuntum), Vienna
(Vindobona).
Bibliography: Bíró-Sey 1977; Fitz 1978; FMRSl I, II, III, IV, V, VI; FMRÖ I.2,
III.1, III.2, IX; FMRU II, III; Găzdac 2002; 2003; 2010.
IV. Noricum: 1. civilian settlements: Adriach, Celje (Celeia), Gleisdorf, Kalsdorf,
Lorch-Enns (Lauriacum), Lupitsch, Rattenberg, St Peter in Holz (Teurnia), Wagna
(Flavia Solva), Wels (Ovilavis), Zollfeld (Virunum).
Bibliography: FMRÖ II.3, IV.1, VI; FMRSl II, III, IV, V, VI; von Prokisch 1989.
V. Raetia: 1. civilian settlements: Augsburg (Augusta Videlicum), Eining (Abusina),
Epfach (Abodiacum), Gauting (Bratananium), Heiden Heim (Aquileia), Hüfingen-
Mühlöschle, Karlstein, Kempten (Cambiodunum), Nassenfels (Vicus Scuttarensium),
Regensburg (Castra Regina), Riegel, Schierenhof, Steppberg; 2. military fortifications:
Aalen, Alkofen, Burghöfe (Summuntorium), Dambach, Eining (Abusina), Faimingen
(Phoebiana), Gnotzheim (Mediana), Günzburg (Guntia), Kosching (Germanicum),
Kumpfmühl, Künzing (Quintana), Pförring (Celeusum), Pfünz (Vetoniana), Regens-
burg (Castra Regina), Risstissen, Schierenhof, Seebruck (Bedaium), Straubing (Sorvio-
durum), Weissenburg (Biriciana).
Bibliography: FMRD I.1, I.2, I.3–4, I.5. I.7, II.2. II.2 N1, II.3, II.3 N1, II.4, II.4 N1.
VI. Germania Superior: 1. civilian settlements: Alzey (Altiaia), Augst (Augusta
Raurica), Bad Dürkheim, Bad Wimpfen (Alisinensium), Baden-Baden (Aquae), Baden
Weiler, Butzbach, Contwig, Dieburg (Med…), Eisenberg, Frankfurt-am-Main (Nida),
Friedberg, Köngen (Grinario), Ladenburg (Lopodunum), Mainz (Mogontiacum),
Mainz-Weisenav, Pforzheim (Portus), Rheinzabern (Tabernae), Rottenburg (Sumelo-
cenna), Rottweil (Arae Flaviae), Saalburg, Speyer (Noviomagus), Stettfeld, Sulz,
188 L. MUNTEANU AND Ş. HONCU

Wiesbaden (Aquae Mattiacorum), Worms (Borbetomagus), Zugmantel; 2. military for-

tifications: Alteburg-Heftrich, Bad Ems, Bingen (Bingium), Butzbach, Echzell, Gerns-
heim, Gross-Gerau (‘Auf Esch’), Heidelberg-Neuenheim, Heldenbergen, Holzhausen,
Jagsthausen, Kleiner Feldberg, Köngen (Grinario), Langenhain, Mainz (Mogontia-
cum), Mainz-Kastel (Castellum Mattiacorum), Miltenberg-Altstadt, Ober-Florstadt,
Ober-Rosbach (Kapersburg), Obernburg-am-Main, Öhringen, Osterburken, Rück-
ingen, Saalburg, Stockstadt-am-Main, Stuttgart-Bad Cannstatt, Walheim, Walldrün,
Zugmantel.
Bibliography: IFS 3, 4; FMRD I.6, II.1, II.1 N1, II.2, II.2 N1, II.3, II.3 N1, II.4, II.4
N1, IV.1, IV.1 N1, IV.2, IV.5, V.1.1, V.1.2, V.2.1, V.2.2.
VII. Germania Inferior: 1. civilian settlements: Altdorf, Düren (Mariaweiler)
(vicus Marcodurum[?]), Jülich (vicus Iuliacum), Kastell Alteburg, Kirchberg, Köln
(colonia Claudia Ara Agrippinensium), Neubourheim (Wackersmühle), Neuss (Novae-
sium), Növernich-(Alt-)Oberbolheim (vicus Narboniacum[?]); 2. military fortifica-
tions: Froitzheim, Kastell Alteburg, Neuss (Novaesium castra I and II), Reckberg.
Bibliography: FMRD VI.1.1. VI.2.1, VI.3.2.
VIII. Gallia Belgica: 1. civilian settlements: Trier (Augusta Treverorum), Ensheim,
Pachten (vicus Contiomagus), Saarbrücken (St Johann), Schwarzenacker.
Bibliography: FMRD III, IV.3.1, IV.3.2, IV.3.4, IV.3.5.

BIBLIOGRAPHY

Ardevan, R. 1991: ‘Monetary Circulation in the Ancient Settlement from Gherla (Cluj
County)ʼ. Eos 79, 223–35.
—. 1993: ‘Aperçu sur la monnaie d’or dans la Dacie romaineʼ. In Hackens, T. and
Moucharte, G. (eds.), Actes du XIe Congrès International de Numismatique 2:
Monnaies celtiques et romaines (Organisé à l’occasion du 150e anniversaire de la
Société Royale de Numismatique de Belgique, Bruxelles, 8–13 septembre 1991)
(Louvain-la-Neuve), 347–53.
—. 2002: ‘Officina falsaria din aşezarea romană de la Ilişuaʼ. Buletinul Societăţii
Numismatice Române 90–91.144–145, 39–43.
—. 2005: ‘Monetary circulation and provincial society in the Roman settlement of
Ilişua (Dacia)ʼ. In Alfaro, C., Marcos, C. and Otero, P. (eds.), Actas XIII
Congreso Internacional de Numismática, Madrid, 2003, vol. 1 (Madrid), 663–71.
Ardevan, R., Suciu, V. and Ciugudean, D. 2003: Tezaurul monetar roman ‘Apulum
VII’ (Alba Iulia).
Aubin, G. 2003: ‘Les moules monétaires en terre cuite du IIIe siècle: chronologie et
géographieʼ. Revue Numismatique s. 6, 159, 125–62.
Bíró-Sey, K. 1977: Coins from Identified Sites of Brigetio and the Question of Local
Currency (Budapest).
Bíró-Sey, K. and Lányi, V. 1980: ‘Fundmünzenbericht 1977ʼ. AArchHung 32.1–4,
455–66.
—. 1981: ‘Fundmünzenbericht 1978ʼ. AArchHung 33.1–4, 347–64.
—. 1982: ‘Fundmünzenbericht 1979ʼ. AArchHung 34.1–4, 363–86.
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 189

—. 1983: ‘Fundmünzenbericht 1980ʼ. AArchHung 35.1–4, 345–71.

—. 1985: ‘Fundmünzenbericht 1981ʼ. AArchHung 37.1–2, 161–76.
—. 1986: ‘Fundmünzenbericht 1982ʼ. AArchHung 38.3–4, 299–312.
—. 1987: ‘Fundmünzenbericht 1983ʼ. AArchHung 39.1–2, 69–92.
—. 1989: ‘Fundmünzenbericht 1984ʼ. AArchHung 41.3–4, 509–26.
Bíró-Sey, K., Lányi, V. and Torbágy, M. 1993: ʻFundmünzenbericht 1987–1988ʼ.
AArchHung 45.1–4, 75-89.
Boon, G. 1974: ‘Counterfeit coins in Roman Britainʼ. In Casey, P.J. and Reece, R.
(eds.), Coins and the Archaeologist (British Archaeological Reports 4) (Oxford),
95–172.
Carson, R.A.G. 1990: Coins of the Roman Empire (London/New York).
Chirilă, E. 1991: ‘Câteva date privind circulaţia monetară la Porolissumʼ. Acta Musei
Porolissensis 14–15, 153–74.
Chirilă, E., Lucăcel, V., Pepelea, V. and Togan, G. 1967: ‘Descoperiri monetare antice
şi bizantine în Transilvaniaʼ. Acta Musei Napocensis 4, 457–59.
Chiţescu, M. and Poenaru Bordea, G. 1983: ʻContribuţii la istoria Diernei în lumina
descoperirilor monetare din săpăturile arheologice din 1967ʼ. Buletinul Societăţii
Numismatice Române 75–76.129–130, 169–208.
Dudău, O. 2006: Circulaţia monetară în castrele de trupe auxilare din provincia
Dacia (Timişoara).
Estiot, S. 2004: Catalogue des monnaies de l’Empire romain 12.1: D’Aurélien à
Florien (270–276 après J.-C.), 2 vols. (Paris).
Fitz, J. 1978: Der Geldumlauf der römischen Provinzen im Donaugebiet Mitte des
3. Jahrhunderts (Budapest).
Fülep, F. 1984: Sopinae: The History of Pécs during the Roman Era and the Problem
of the Continuity of the Late Roman Population (Budapest).
Gaspar, R.B. 2015: ‘Counterfeiting Roman coins in the Roman Empire 1st–3rd A.D.
Study on the Roman provinces of Dacia and Pannoniaʼ. Journal of Ancient
History and Archaeology 2.4, 31–74.
Găzdac, C. 1997: ‘A Lead Mould from Ulpia Traiana Sarmizegetusaʼ. Pontica 30,
351–53.
—. 2002: Circulaţia monetară în Dacia şi provinciile învecinate de la Traian la
Constantin I (Cluj-Napoca).
—. 2003: Monetary Circulation in Dacia and the Provinces from the Middle and
Lower Danube from Trajan to Constantine I (A.D. 106–337) (Cluj-Napoca).
—. 2007: ‘Descoperirile de monedă de argint contrafăcută în castre din Dacia Romană.
Fraudă sau politică monetară romană?ʼ. In Nemeti, S., Fodorean, F., Nemeth, E.,
Cociş, S., Nemeti, I. and Pâslaru, M. (eds.), Dacia Felix. Studia Michaeli
Barbulescu oblata (Cluj-Napoca), 595–605.
—. 2009: ‘The distribution of silver counterfeited coins in the forts of Roman Dacia:
fraud or monetary policy?ʼ. In Morillo, A., Hanel, N. and Martín, E. (eds.), Limes
XX (XX Congreso internacional de estudios sobre la frontera Romana León,
España, Septiembre 2006), vol. 3 (Madrid), 1487–98.
—. 2010: Monetary Circulation in Dacia and the Provinces from the Lower Danube
from Trajan to Constantine I (AD 106–337), 2nd ed. (Cluj-Napoca).
Găzdac, C., Alföldy-Găzdac, Á., Neagoe, M. and Neagoe, O. 2015: Drobeta, the
Never Abandoned City of Roman Dacia (Cluj-Napoca).
190 L. MUNTEANU AND Ş. HONCU

Găzdac, C. and Cociş, S. 2004: Ulpia Traiana Sarmizegetusa (Cluj-Napoca).

Găzdac, C. and Cosma, C. 2014: ‘A Counterfeiter’s Fingerprint on a Forged Denarius
of Marcus Aureliusʼ. Numismatic Chronicle 174, 125–28.
Găzdac, C. and Gaiu, C. 2011: ‘Capcanele evidenţei numismatice. Cazul castrului
auxiliar Arcobadara (Ilişua, judeţul Bistriţa-Năsăud, România)ʼ. Revista Bistriţei
25, 192–96.
Găzdac, C., Gaiu, C. and Alföldy-Găzdac, Á. 2015: ‘The pitfalls of the numismatic
evidence. The auxiliary fort of Arcobadara (Ilişua, Bistriţa-Năsăud County)ʼ.
Journal of Ancient History and Archaeology 2.3, 25–27.
Găzdac, C., Gaiu, C. and Marchiş, E. 2011: Arcobadara (Ilişua) (Cluj-Napoca).
Găzdac, C. and Gudea, N. 2006: Porolissum (Cluj-Napoca).
Găzdac, C. and Isac, D. 2007: The Auxiliary forts from Samum (Căşeiu) and Gilău
(Cluj-Napoca).
Găzdac, C., Oargă, O. and Alföldy-Găzdac, Á. 2015: It was Supposed to be Silver!
The Scrap Coin ʽHoardʼ Apulum VI (Cluj-Napoca).
Găzdac, C. and Pripon, E. 2012: The Roman Auxiliary Fort at Buciumi (Roman Dacia,
Romania): Coins in Archaeological Context (BAR International Series 2381)
(Oxford).
Găzdac, C., Suciu, V. and Alföldy-Găzdac, Á. 2009: Apulum (Cluj-Napoca).
Găzdac, C., Voişian, V. and Cociş, S. 2003: ‘Coin finds from Napoca – The excava-
tions on Deleu stʼ. Revista Bistriţei 17, 77–96.
Găzdac, C. and Zăgreanu, R. 2008: ‘Reinterpreting numismatic evidence. A methodo-
logical case study: the auxiliary fort from Gherla (Cluj county, Romania)ʼ.
Numismatic Chronicle 14, 237–41.
Gudea, N. 1996: Porolissum. Un complex daco-roman la marginea de nord a Imperiu-
lui Roman. II. Vama romană. Monografie arheologică (Cluj-Napoca).
—. 1997: Castrul Roman de la Buciumi/Das Römergrenzkastell von Buciumi (Zalău).
Harl, K.H. 1996: Coinage in the Roman Economy, 300 B.C. to A.D. 700 (Baltimore/
London).
Katsari, C. 2011: The Roman Monetary System: the Eastern Provinces from the First
to the Third Century AD (Cambridge).
King, C.E. 1996: ‘Roman copiesʼ. In King, C.E. and Wigg, D.G. (eds.), Coin Finds
and Coin Use in the Roman World (The Thirteenth Oxford Symposium on
Coinage and Monetary History, 25–27.3.1993) (Studien zu Fundmünzen der
Antike 10) (Berlin), 237-263.
Kunisz, A. 1980: ‘La monnaie de nécessité dans les provinces rhénanes et danubiennes
de l’Empire Romain dans la première moitié du IIIeʼ. In Vallet, G. (ed.), Les
‘dévaluations’ à Rome: époque républicaine et impériale 2 (Gdansk, 19–21 octo-
bre 1980) (Collection de l’École française de Rome 37) (Rome), 129–39.
—. 1987: ‘La monnaie de nécessité à l’époque du haut-empire romain: problèmes et
controversesʼ. In Depeyrot, G., Hackens, T. and Moucharte, G. (eds.), Rythmes de
la production monétaire, de l’antiquité à nos jours (Actes du Colloque inter-
national organisé à Paris du 10 au 12 janvier 1986) (Numismatica Lovaniensia 7)
(Louvain-la-Neuve), 257–65.
Lallemand, J. 1994: ‘Les moules monétaires de Saint-Mard (Virton, Belgique) et les
moules de monnaies impériales en Europe: essai de répertoireʼ. In Remy, H. and
 SILVER-PLATED COINS IN THE SITES OF ROMAN DACIA 191

Gillet-Mignot, P. (eds.), Un quartier de l’agglomération Gallo-Romaine de Saint-

Mard (Virton) (Namur), 141–77.
Macrea, M. 1932: ‘Contribuţii la un repertoriu numismatic al Dacieiʼ. Anuarul Institu-
tului de studii clasice pe anii 1928–1931 1,127–39.
Matei, C. 2015: Circulația monetară romană pe teritoriul anticului Tibiscum (sec. I–
IV p.Chr), vol. I (Szeged).
Mihailescu-Bîrliba, V. 1980: La monnaie romaine chez les Daces orientaux
(Bucharest).
—. 1990: Dacia răsăriteană în secolele VI–I î. e. n. Economie şi monedă (Iaşi).
Mitrofan, I. and Ardevan, R. 1997: ‘Découvertes monétaires dans l’établissement
romain de Micăsasaʼ. Studii şi Cercetări de Numismatică 11, 119–33.
Oberländer-Târnoveanu, E. 2008: ‘Monedele descoperite în complexul arheologic
roman din secolele II–III de la Gârla Mare (com. Gârla Mare, jud. Mehedinți) –
Un studiu de caz asupra economiei monetare în spațiul rural al provinciei Daciaʼ.
Cercetări Numismatice 14, 243–71.
Opreanu, C.H. and Lăzărescu, V.-A. 2015: A Roman Frontier Marketplace at
Porolissum in the Light of Numismatic Evidence. Contribution to the Knowledge
of the Roman Limes Economy (Cluj-Napoca).
Pekáry, T. 1953: ‘Aquincum pénzforgalmaʼ. Archaeologiai Értesítő 80.1–2, 106–14.
Petac, E. 2011: Aspecte ale circulației monetare în Dacia romană (105–275 p.Chr.)
(Collection Moneta 115) (Wetteren).
Peter, M. 1990: Eine Werkstätte zur Herstellung von subaeraten Denaren in Augusta
Raurica (Studien zu Fundmünzen der Antike 7) (Mainz).
—. 2001: Untersuchungen zu den Fundmünzen aus Augst und Kaiseraugst (Studien zu
Fundmünzen der Antike 17) (Berlin).
Pîslaru, M. 2009: The Roman Coins from Potaissa, Legionary Fortress and Ancient
Town (Cluj-Napoca).
Poenaru Bordea, G. 1975: ‘Descoperiri monetare din cetăţuia romano-bizantină de la
Sucidava cu specială privire asupra perioadei 260–328. Campaniile 1966–1971ʼ.
Studii şi Cercetări de Numismatică 6, 69–106.
—. 1998: ‘Monede din vremea Imperiului roman din colecţia şcolii din cartierul Celeiu
al oraşului Corabia, jud. Oltʼ. Studii şi Cercetări de Numismatică 12, 41–81.
von Prokisch, K. 1989: ‘Antike Fundmünzen aus Lauriacum. Die Sammlung Spatt/
Ennsʼ. Jahrbuch des OberÖ Musealvereines – Gesellschaft für Landeskunde,
134.1, 13–61.
Stîngă, I. 1998: Viaţa economică la Drobeta în secolele II–VI p. Ch. (Bucharest).
Suciu, V. 2000: ‘Quelques considérations sur les trésors monétaires de la Dacie
romaineʼ. In Ciugudean, H. and Moga, V. (eds.), Army and Urban Development
in the Danubian Provinces of the Roman Empire (Alba Iulia), 323–32.
Thúry, L. 2000: ‘Money circulationʼ. In Kovács, P. (ed.), Matrica: Excavations in the
Roman Fort at Százhalombatta, 1993–1997 (Budapest), 105–19.
Winkler, I. 1964: ‘Despre circulaţia monetară la Porolissumʼ. Acta Musei Napocensis
1, 215–25.
—. 1965: ‘Circulaţia monetară la Apulumʼ. Acta Musei Napocensis 2, 215–56.
—. 1975: ‘Descoperiri monetare în Ulpia Traiana Sarmizegetusaʼ. Sargetia 11–12,
117–36.
192 L. MUNTEANU AND Ş. HONCU

Winkler, I. and Băloi, C. 1971: ‘Circulaţia monetară în aşezările antice de pe teritoriul

comunei Orlea (I)ʼ. Acta Musei Napocensis 8, 161–72.
—. 1973: ‘Circulaţia monetară în aşezările antice de pe teritoriul comunei Orlea (II)ʼ.
Acta Musei Napocensis 10, 181–212.
Winkler, I. and Hopârtean, A. 1973: Moneda antică la Potaissa (Cluj-Napoca).
Winkler, I. and Ivănescu, I. 1977: ‘Imitaţiile de denari descoperite la Porolissumʼ. Acta
Musei Porolissensis 1, 83–96.
 CHAPTER 12

RÖMISCHER EINFLUSS IM BARBARICUM

ZWISCHEN DIPLOMATIE, KLIENTELPOLITIK
UND DEFENSIVSTRATEGIE:
NEUE METHODEN DES MACHTMANAGEMENTS
IN DER SPÄTANTIKE*

Alexander RUBEL

Abstract
This paper proposes a reassessment of some major aspects of late antiquity. Based
on recent archaeological research in Germany and Romania new light can be shed on
modes of ‘client management’ via subsidies and gifts even in remote regions behind
the limes to influence and control inner-barbarian relations and conflicts. This new
archaeological data, as well as the development of a sophisticated diplomacy and an
archaeologically traceable new construction phase along the Danube from the Tetrar-
chy onwards, using a new design of fortified settlements in the hinterland of the limes,
can give us some clues towards making a re-evaluation of the character of this era.
This paper takes up a stance on late antiquity as an age of major rational political and
military developments, intended to cope with new challenges. Further research, using
recent archaeological data, should concentrate on the impressive administrative perfor-
mance of the later Roman empire, which was not at all doomed in the 4th century AD,
and could develop a rather impressive real ‘strategy’ in terms of a pre-modern
society.

Die „Reichskrise“ des 3. Jhts. hatte das Imperium vor eine enorme Bewäh-
rungsprobe gestellt, die nur wir heute aus der rückschauenden Perspektive als
solche begreifen.1 Die entscheidenden Veränderungen der Rahmenbedingungen,

*
Dieser Beitrag wurde ermöglicht durch eine Projektförderung der rumänischen Forschungs-
förderungsbehörde im Erziehungs- und Wissenschaftsministerium UEFISCDI, Projektnummer
PN-III-P4-ID-PCE-2016-0669, Projekttitel: „Beyond the Fringes of Empire. Roman Influence
and Power north of the Danube and east of the Rhine“.
1
Ein Krisenbewusstsein oder gar eine das ganze Reich gleichmäßig erfassende Krise sind
nach neueren Studien nicht wirklich nachweisbar, deswegen die Anführungsstriche. Die Verän-
derungen der Bedingungen am Rande des Imperiums und der bis zum Ende des 2. Jhts. gültigen,
194 A. RUBEL

welche die Zeit der Stabilität der pax Romana seit Augustus nachhaltig been-
dete lassen sich schon während der Markomannenkriege (166–180) erkennen,
die ihrerseits durch Bevölkerungsverschiebungen und sich neu formierende
Machtzentren in den von den Römern nicht bzw. nur unzureichend kontrollier-
ten entfernteren Gebieten der Germania vollzogen. Die tiefgreifenden Verän-
derungen im Barbaricum wurden recht eigentlich erst nach dem Friedens-
schluss offenbar und bestimmten in der Folge die unübersichtlicher werdenden
Auseinandersetzungen und generierten neue Herausforderungen. Aktuelle
archäologische Forschungen verweisen auf tiefgreifende Wandlungen in den
heute als „germanisch“ angesprochenen Gesellschaften in den Gebieten des
heutigen östlichen Deutschland, Polens und der Slowakei in dieser Zeit.2 Die
folgende sogenannte „Soldatenkaiserzeit“ war dann von massiven Einfällen
unübersichtlicher werdender Koalitionen mobiler Fremdvölker, dem Versagen
der linearen römischen Grenzverteidigung (Limes) bei gleichzeitiger Auf-
nahme größerer Gruppen ins Reich und zunehmender Rekrutierung barbari-
scher Verbände geprägt.3 Die gemeinhin als Niedergangsphänomen interpre-
tierte Abfolge einer Vielzahl von „nur“ militärisch begabten Anführern auf
dem Kaiserthron, die von den jeweiligen Truppenteilen teilweise gegeneinan-
der proklamiert wurden, war gewissermaßen Reflex der neuen Gegebenheiten
und des Primats militärischer Führung. Die notwendigen Reformen, die den
Veränderungen im Reich und an seinem Rande Rechnung trugen, wurden
dann von Diokletian ab 284 und in der Folge besonders auch von Constantin I.
systematisch durchgeführt.4 Dennoch hält sich in der Geschichtsschreibung
weitgehend das Bild vom bereits seit der Soldatenkaiserzeit dem Untergang
geweihten Imperium, das gewissermaßen wie ein geköpftes Huhn nur noch zu
instinktiven und eben kopflosen Reaktionen fähig ist. Zwar spricht man nicht
mehr wie zur Zeit Gibbons in apokalyptischen Bildern vom Untergang, doch
wird die grundlegende Veränderung der sozialen und politischen Rahmen-
bedingungen als Einschnitt gewertet und das Ende des Imperiums als eine
zwar einer sehr komplexen Motivlage geschuldete, aber folgerichtige
und aufgrund der Defizite der Verwaltungs- und Militärstruktur letztlich

vergleichsweise stabilen Verhältnisse bei den angrenzenden „Barbaren“ wirken nur aus der his-
torischen Perspektive als Auslöser einer umfassenden Krise. Siehe bes. Strobel 1993, vgl. die
Darstellung der Forschungsgeschichte bei Sommer 2015. Zum nicht ganz eindeutigen Barbaren-
begriff Rubel 2016.
2
Hierzu zuletzt einleuchtend Voß 2017.
3
Umfassend Johne 2008.
4
Strobel 2007 konnte zeigen, dass die Anpassungen an die neuen Gegebenheiten bes. im
Bereich des Heerwesens schon während der „Soldatenkaiserzeit“ schrittweise erfolgten und
Diokletian und Constantin I. einen bereits eingeschlagenen Weg weiter verfolgten.
 RÖMISCHER EINFLUSS IM BARBARICUM 195

hausgemachte Entwicklung angesehen.5 Aus der Kenntnis des weiteren Ver-

laufs der Geschichte, in der sich die Herrschaftsformen und Lebensweisen
(zum Beispiel Ende der Stadtkultur) in Europa nachhaltig ändern, ganz unab-
hängig davon, ob man das Ende der antiken Staatlichkeit nun eher traditionell
als „Untergang“ begreift, oder zuletzt dominierend eher als „Transformation“,6
wird bis heute gerne geschlossen, dass das Imperium keine geeigneten Ant-
worten auf die neuen Herausforderungen liefern konnte und taumelnd in den
Abgrund der Geschichte stürzte; eine weitere „Krise ohne Alternative“, um
das bekannte Wort von Ch. Meier zu zitieren.
In diesem Beitrag geht es – natürlich in völlig unzureichender Kürze7 – um
den Versuch einer Neubewertung dieser Zusammenhänge unter drei Leit-
aspekten: 1. Die herausragende strategische Bedeutung der Pufferzonen und
„Klientelstaaten“ am Rande des Imperiums wird seit langem diskutiert.8 Hinzu
kommt die zunehmende Bedeutung von auswärtigen Söldnern im Heer, deren
Relevanz für die Gesamtstruktur der römischen Armee in der Spätantike nicht
zu unterschätzen ist, gerade auch bezüglich der verwirrenden Ambivalenz,
dass die gleichen Gruppen sowohl als Verbündete oder foederati, wie auch als
Feinde auftreten können.9 Nun verdichten sich nach neueren Forschungen
besonders im germanischen Raum im heutigen Südskandinavien, in Nord- und
Ostdeutschland die (archäologischen) Hinweise darauf, dass die Römer (auch
noch nach dem 2. Jht.) eine direkte und massive Politik der Einflussnahme
betrieben haben, die nur unzureichend in den Quellen abgebildet ist. Das
Fundmaterial (Waffenopfer, „Fürstengräber“, generell „Importe“10) lässt

5
Das Thema des „Untergangs des römischen Reiches“ und der diesem zugrundeliegenden
Motive ist natürlich ein vielbehandeltes Kernthema der Altertumswissenschaften und kann hier
nicht ausführlich angegangen werden. In seiner Übersicht über die Forschungsgeschichte nennt
Demandt 2014 über 200 in der Forschung diskutierte mögliche Motive für den Untergang des
römischen Reichs. Siehe zuletzt die vielversprechende Einbeziehung der Klima- und Umwelt-
geschichte durch Harper 2017.
6
Die Diskussion bewegt sich etwa zwischen Antipoden wie Ward-Perkins 2005, der das
Ende der Antike als Zäsur und Untergang betrachtet und Brown 1971, der eine viel beachtete und
heute dominierende (etwa Goffart 2006) neue Sichtweise unter dem Zeichen der Transformation
vorgeschlagen hat. Aber selbst „konservative“ Gelehrte wie Ward-Perkins oder Heather 2011,
die die Dramatik des Wegfalls staatlicher Ordnung und steuerfinanzierten öffentlichen Mehr-
werts erkennen, bemühen nicht mehr die frühere Dekadenzmetaphorik.
7
Mit einer Forschergruppe am Institut für Archäologie Iaşi (gefördert durch eine rumänische
Projektförderung, sie erste Anmerkung) und im Rahmen unserer Partnerschaft mit der RGK
Frankfurt versuchen wir diese Aspekte umfassender und systematischer mit einer Reihe von Pub-
likationen zu behandeln.
8
Kornemann 1934, zum nicht ganz treffenden aber etablierten Klientelbegriff im Zusammen-
hang mit Roms auswärtigen Beziehungen Kehne 2001 mit weiterer Literatur.
9
Stickler 2007.
10
Auf die Problematik des Begriffs der römischen „Importe“ kann hier nicht detaillierter
eingegangen werden. Wichtig ist an dieser Stelle lediglich, dass die Rolle des Eindringens
196 A. RUBEL

zunehmend den Schluss zu, dass die Römer in Germanien mittels komplexer
Allianzbildungen (stimuliert durch Zahlungen und Geschenke) und möglicher-
weise auch dem Einsatz von „Militärberatern“ und einer „fünften Kolonne“
von den jeweiligen ethnischen Gruppen zugehörigen Auxiliarveteranen regel-
rechte Stellvertreterkriege initiierten und große militärische Kampanien (wie
die von Maximinius Thrax 235) durch Bündnisse absicherten.11 Entsprechend
muss gefragt werden, ob römische Außenpolitik nicht auch in der Zeit für die
wir nur wenige Quellenbelege haben und die allgemein – beginnend mit der
sog. „Reichskrise“ – als von zunehmendem Verlust der Handlungsfähigkeit
geprägte Krisenzeit gesehen wird, aktiver als häufig angenommen ausgerichtet
war.
2. Eine aufwändige Baupolitik zur Befestigung der Städte war nicht erst
eine letzte Reformmaßnahme des Justinian, sondern lässt sich gerade an der
Donau konstant bis zum Ende des 4. Jhts. bereits seit der Tetrarchie fassen.12
Entsprechend sollte die Frage gestellt werden, ob die neueren Ergebnisse der
Archäologie an der mittleren und unteren Donau es nicht erforderlich machen,
die vielfache Kritik am von E. Luttwak 1976 präsentierten Ansatz einer „grand
strategy“ des Imperium Romanum zu relativieren.13 Mit Verweis auf die schon
länger erforschten „Innenbefestigungen“ in Pannonien14 sollen kurz die Fes-
tungen im Limeshinterland in Scythia minor vorgestellt und hinsichtlich ihrer
Funktion im Rahmen einer durchdachten und nachhaltigen Defensivstrategie
im 4. Jht. interpretiert werden.
3. Zur Vergegenwärtigung des Instrumentariums, welches dem spätrömi-
schen Staat hinsichtlich der Gestaltung seiner Außenbeziehungen zur Verfü-
gung stand, soll der Fall des Fürsten der Bucinobanten, Makrian, genauer
betrachtet werden. Um diesen als störenden Machtfaktor in den Rheinlanden
bei Mogontiacum (Mainz) auszuschalten, zog Kaiser Valentinian I. um 370
alle Register. An diesem abschließenden Beispiel wird die Flexibilität der
römischen Außenpolitik im Umgang mit „Barbaren“ an den Grenzen des Rei-
ches deutlich werden.

römischer Luxusgüter ins Barbaricum durch Militärdienst (Söldnerwesen und Auxilia) sowie
diplomatische Maßnahmen (Geschenke, Subsidien), neben den als Handelsware und Beutegut
eingeführten „Importen“ betont werden muss. Die wichtigsten Einordnungen mit weiterer Fach-
literatur bei Wolters, Erdrich und Voß 2003.
11
Hierzu unten ausführlicher, nur zum Beispiel Voß 2017; Grane 2007; Rau 2012.
12
Mit weiterer Literatur Rubel 2018b.
13
Luttwak 1976, bes. 127–90, vehementeste Kritik bei Mann 1979, Zusammenfassung der
Diskussion um die „grand strategy“ bei Wheeler 1993a, 8–9.
14
Heinrich-Tamáska 2011; 2017a.
 RÖMISCHER EINFLUSS IM BARBARICUM 197

1. Veränderungen im machtpolitischen Gefüge im nördlichen

Barbaricum

Zunächst aber zu den Auswirkungen der Markomannekriege auf die „geopoli-

tischen“ Verhältnisse im nördlichen Barbaricum.15 Neuste archäologische und
numismatische Forschungen etwa im Bereich der Ueckermark und im Thürin-
gischen Raum, sowie neue Auswertungen der in norddeutschen und südskan-
dinavischen Mooren verborgenen Kriegsbeuteopfer erlauben eine vorsichtige
Neueinschätzung der römischen Einflussnahme in diesen Gebieten, über die
aus den Quellen kaum etwas bekannt war. Die wenigen Angaben, etwa über
den Feldzug des Maximinius Thrax tief ins Innere Germaniens im Jahre 235
(Historia Augusta, Vita Maximini duo 12, 1, vgl. Herodian 7, 2, 5–9) wurden
angesichts der krisenhaften Situation, in welcher sich das Reich befand, als
unglaubwürdig zurückgewiesen.16 Erst die Entdeckung des Schlachtfelds am
Harzhorn im östlichen Niedersachsen und die seit 2008 erfolgende Auswer-
tung der Funde ließen eine Neuinterpretation notwendig erscheinen.17 Das
„Harzhorn-Ereignis“ bezeugt entsprechend nach allgemeinem Dafürhalten die
noch ungebrochenen „imperialistischen“ Kapazitäten des Reiches, das im
3. Jht., trotz vermeintlicher „Krise“, weit davon entfernt war, außenpolitisch
eine Wagenburgstrategie zu entwickeln. Münzfunde und Verbreitungskarten
römischer Importe weisen nach neuesten Forschungen darauf hin, dass nach
der grundlegenden Veränderung der Subsidienpolitik durch die Severer (Export
von Denaren in die Germania magna endet praktisch mit Commodus – übri-
gens auch weiter östlich im „freien Dakien“) im ersten Drittel des 3. Jhts.
neue Allianzen mit germanischen Gruppen im thüringischen Becken entstan-
den. Das Ausbleiben der Unterstützungszahlungen durch das Imperium brachte
offenbar das bereits durch die Kriege betroffene Stammesgefüge im heutigen
Mitteldeutschland, Böhmen und Polen nachhaltig durcheinander, wobei bereits
in der ersten Hälfte des 3. Jhts. wieder „Importe“ und Münzen auftauchen,
aber jetzt vermehrt in anderen Gebieten (vorzüglich im heutigen Thüringen
und Sachsen-Anhalt).18 Diese durch „Importe“ und Münzen ausgewiese-
nen neuen Interessensgebiete der Römer sicherten – so die überzeugende

15
Der Begriff Barbaricum wird hier und im Folgenden als historiographischer terminus tech-
nicus für Gebiete und Kontaktzonen außerhalb des römischen Reiches nördlich der Donau und
östlich des Rheins verwendet, wenngleich die ersten Belege für eine solche Verwendung des
Begriffs in der Antike rar sind und erst ins frühe 3. Jht. gehören: Sarnowski 1991.
16
Hose 2013. Dass den Zeitgenossen der „Reichskrise“ weitgehend ein Krisenbewusstsein
fehlte, konnte Strobel 1993 zeigen. Krisenhafte Aspekte traten in Teilen des Reichs, jedoch nie
flächendeckend auf.
17
Pöppelmann et al. 2013.
18
Voß und Wigg-Wolf 2017, Verbreitungskarte 119.
198 A. RUBEL

Schlussfolgerung der Forscher an der RGK Frankfurt – die Flanke der neuen
Operationen im Inneren Germaniens, die nach Maßgabe der Größenordnung
dieses Feldzugs 235/236 eine umfassende logistische und diplomatische Vor-
bereitung erforderten.19
Bereits für die Folgezeit der Markomannenkriege mehren sich die Anzei-
chen, dass militärische Auseinandersetzungen im Inneren Germaniens, wo sich
durch die Markomannenkriege und die anhaltende Attraktivität des Reichs-
gebiets und seiner Kontaktzonen recht unmittelbar östlich des Rheins und
nördlich der Donau Verschiebungen des Siedlungsgefüges ergeben hatten,
eine Neusortierung der innergermanischen Verhältnisse auslösten.20 So lässt
sich ein 2008 bei Pasewalk in Mecklenburg-Vorpommern gefundenes hoch-
wertiges römisches Schwert mit inkrustierter Marsdarstellung, das in den Kon-
text einer auf die ersten Jahrzehnte nach den Markomannenkriegen datierten
massiven kriegerischen Auseinandersetzung am Flusslauf der Uecker gehört
(bereits Ende der 1980er Jahre waren dort zwei weitere römische Schwerter,
sowie Militaria entdeckt worden, zuletzt auch zwei Skelette, die massive
Gewalteinwirkung aufwiesen), im Sinne einer sich abzeichnenden Neuausrich-
tung römischer Einflusspolitik in Germanien interpretieren, die innergermani-
sche Konflikte schürte und „Stellvertreterkriege“ begünstigte: „Steht doch zu
vermuten, dass die während der Markomannenkriege von der römischen Mili-
tärführung gewonnenen Erfahrungen und Erkenntnisse über die Herkunft der
als Gegner wie auch als Verbündete kämpfenden Germanen zu der Einsicht
geführt hatten, innergermanische Konflikte möglichst in Regionen fernab der
Reichsgrenzen zu verlagern oder dort gar zu initiieren“.21
Generell zeichnet sich für den germanischen Raum ein neues Bild ab, dass
Voß und Wigg-Wolff anhand des neueren im CRFB-Projekt gesammelten
Materials und anhand der jüngsten Literatur zusammenfassend beschrieben
haben.22 Demnach verschiebt sich die Interessenssphäre römischer Einfluss-
politik im Zusammenhang mit den Markomannenkriegen in den mitteldeut-
schen Raum, wobei die sich neu formierenden Gruppen enge Verbindungen
zum südskandinavischen Raum aufweisen.23 Die durch Geldzahlungen, diplo-
matische Geschenke und „Importe“ beförderte Bildung eines Elitenetz-
werkes in der Germania diente zunächst den konkreten Interessen römischer

19
Ausführlich mit weiterer Literatur Voß und Wigg-Wolf 2017.
20
Voß 2017.
21
Schmitt und Voß 2017, 221.
22
Voß und Wigg-Wolf 2017. Zum zwischen 2001 und 2008 von der DFG geförderten, aber
unabhängig weiter verfolgten Projekt CRFB (Corpus der römischen Funde im Barbaricum) siehe
https://fallback.dainst.org/projekt/-/project-display/31796.
23
Zum skandinavischen Komplex: Rau 2012; Grane 2007.
 RÖMISCHER EINFLUSS IM BARBARICUM 199

Außenpolitik, legte aber den Grundstein für die Bildung nachhaltiger Macht-
strukturen im Barbaricum und zur Entstehung und Konsolidierung eines
Selbstbewusstseins der germanischen Eliten, das zur Emanzipation vom Impe-
rium führte und letztlich auch die Voraussetzung für die Bildung germanischer
Machtzentren und stabiler Herrschaftsverbände in der Spätantike bildete.24
In diesen wahrscheinlichen Kontext einer Politik von divide et impera sei-
tens der Römer gehören auch die eindrucksvollen Waffendeponierungen in
Norddeutschland und Südskandinavien. In den Opfermooren von Thorsberg,
Vimose, Illerup Ådal, Ejsbøl oder Nydam hat sich aufgrund der physikalischen
Verhältnisse eine Vielzahl von Ausrüstungsgegenständen erhalten, die nach
Kampfhandlungen von den Siegern rituell im Moor als Opfergabe verborgen
wurden.25 Nach intensiver Auswertung des Fundmaterials hat sich in den letz-
ten Jahren ein klareres Bild ergeben, nach dem es sich bei den beteiligten
Gruppen, deren Ausrüstung in den Mooren versenkt worden war, keineswegs
um unbedeutende Räuberbanden handelte (in Ejsbøl umfasste die Krieger-
gruppe mindestens 150 Männer, in Vimose wurden Ausrüstungen von mehre-
ren Hundert Mann versenkt), sondern, dass hier seit dem 3. Jht. Gruppen in
Konflikten, in denen es um „alles oder nichts“ ging, engagiert waren. Die
Opferfunde sind Ausweis „finaler Stadien machtpolitischer Auseinander-
setzungen“26 zwischen hierarchisch gut organisierten großen Gruppen, deren
„Offiziere“ über besonders wertvolle Ausrüstungsgegenstände verfügten.
Gerade die römischen Schwerter, die in den Fundzusammenhängen in den
skandinavischen Opfermooren entdeckt wurden, weisen auf die Existenz von
professionellen germanischen Kriegern hin, die mit oder gegen Rom ihr Hand-
werk gelernt hatten.
Diese Kriegereliten waren nun offenbar in massive Auseinandersetzungen
weitab von der Reichsgrenze involviert. Voß, dem ich in dieser Einschätzung
folge, vermutet hinter diesen massiven Auseinandersetzungen in hochmilitari-
sierten Gesellschaften im nördlichen Germanien die Auswirkungen einer römi-
schen Einflusspolitik, die durch das Prinzip der discordia hostium mögliche
Konflikte weit im Hinterland des eigenen Grenzgebiets unterstützte, eventuell
sogar inszenierte.27 Wenngleich diese Schlussfolgerungen spekulativ bleiben

24
Voß und Wigg-Wolf 2017, 121. Dabei erfolgte der weitere Austausch und die Verbreitung
der römischen „Importe“ (v.a. auch nach Norden) im Rahmen dieser Netzwerke, Voß 2017, 338;
vgl. Rau 2012.
25
Grundlegend mit der Spezialliteratur Rau und von Carnap-Bornheim 2012. Übersicht bei
Rubel 2018. Als Beispiel für neuere Fundanalysen und Publikationen siehe zum Fundplatz
Ejsbøl, Nørgård Jørgensen und Andersen 2014.
26
Rau und von Carnap-Bornheim 2012, 529.
27
Schmitt und Voß 2017, 221–22; vgl. Voß 2017; Voß und Wigg-Wolf 2017.
200 A. RUBEL

müssen, weil sie lediglich auf einer Interpretation des Fundmaterials beruhen,
wird die Möglichkeit, dass auch in der Zeit der „Reichskrise“ und während
der Spätantike eine strategische Ausrichtung der reichsweiten Verteidigungs-
politik erfolgte, doch um einiges wahrscheinlicher.

2. Die „Innenbefestigungen“ an der Donau und die Scythia minor28

Angeregt von den bahnbrechenden Arbeiten Sopronis aus den 1970er Jahren,
die wiederum auf Ansätzen von Radnóti gründen, hat Orsolya Heinrich-
Tamáska in den letzten Jahren die Diskussion um die sog. pannonischen
„Innenbefestigungen“ weiter geführt.29 Diese befestigten Siedlungen im Hin-
terland des Limes (etwa Környe, Tác/Gorsium, Keszthely-Fenékpuszta,
Alsóheténpuszta/Iovia) bildeten zusammen mit den älteren und weiter befes-
tigten Städten wie Sopianae, Sirmium oder Bassianae (etc.) eine innere Vertei-
digungslinie. Dabei ist auch bei den gut erforschten Innenbefestigungen
(zuletzt umfassend publiziert: Keszthely-Fenékpuszta30) letztlich nicht genau
zu bestimmen, ob diese spätantiken Festungsbauten eher in zivilem oder rein
militärischem Kontext gesehen werden müssen. Sie sind, wie Poulter für die
bulgarischen Pendants feststellte, wohl eher eine hybride Art „military-cum-
civilian sites“,31 bei der zivile Nutzung und militärischer (zumindest zeitweise)
Lagercharakter miteinander einhergehen. Sie erscheinen entsprechend als
semiurbane Befestigungen, die zeitweise größere Truppenkontingente aufneh-
men konnten und offenbar auch wirtschaftlich relevante Funktionen (auch im
Bereich der Versorgung und Bevorratung) übernahmen.
Die Datierungen sind ebenfalls immer noch unklar und umstritten. Während
man an den ungarischen Befunden ursprünglich die hufeisenförmigen Türme
dieser Festungen und befestigten Siedlungen etwa in die constantinische Zeit
datieren und die typischerweise bei den Innenbefestigungen verwendeten run-
den Türme einer späteren Phase (valentinianisch und später) zuordnen wollte,32
sind seit einiger Zeit Zweifel an der Datierung auf Basis der Turmformen
gewachsen.33 Aufgrund der für diese Anlagen typischen und im Befund in den
meisten Fällen abgebildeten großen Vorratsgebäuden (horrea) wird mit einiger
Wahrscheinlichkeit angenommen, dass diese Plätze eine zentrale Rolle bei der

28
In diesem Abschnitt folge ich meiner Darstellung in Rubel 2019.
29
Zum Beispiel Heinrich-Tamáska 2011; 2015; 2017a; 2017b. Vgl. auch Tóth 1985; 2009.
30
Heinrich-Tamáska 2013, sowie weitere hier nicht relevante Bände zum Gräberfeld etc. in
der gleichen Reihe.
31
Poulter 2007, 36.
32
Soproni 1978, 140.
33
Ausführlich Tóth 1985.
 RÖMISCHER EINFLUSS IM BARBARICUM 201

Lagerung und Verteilung (eventuell auch bei der Produktion) der annona
gespielt haben. Weitere typische Merkmale dieser Innenbefestigungen in Pan-
nonien sind nach den Ergebnissen der ungarischen Archäologie der unmittel-
bare Zugang zum Wasser, wie etwa in Alsóhetény, wo ein Flusslauf durch die
Festung verläuft und nach Belieben im Inneren angestaut werden konnte,
sowie ihre Lage an wichtigen Fernwegen oder Kreuzungen.34
Die nicht immer eindeutigen Datierungen der spätantiken Innenbefestigun-
gen und des etwa zeitgleichen Ausbaus der Städte deuten frühestens auf die
Zeit der Tetrarchie als Beginn einer strategischen Neuorientierung, die mit den
von Constantin I. dann fortgeführten Militärreformen des Diokletian in Ver-
bindung gebracht werden.35 Aus den Quellen erfahren wir allerdings wenig
über mit den Militär- und Provinzreformen einhergehende Baumaßnahmen.
In diesem Zusammenhang bleiben die archäologischen Forschungen, die in
jüngerer Zeit neben Ungarn auch im heutigen Serbien, Bulgarien und Rumä-
nien vielversprechende Ergebnisse zutage förderten, die maßgeblichen Indika-
toren für eine gesamtstrategische und nachhaltige Ausrichtung der tetrarchi-
schen Reformen. Ein Großteil der Bautätigkeit in Pannonien, so zeigen die
Ergebnisse der ungarischen Forschung, die hinsichtlich der Innenbefestigun-
gen und der spätantiken Verteidigungsstrategie seit Sopronis Vorstoß in den
1970er Jahren am weitesten fortgeschritten ist, ist auf die Zeit Constantins des
Großen und seine unmittelbaren Nachfolger zurückzuführen. Die neueren aber
noch unzureichend ausgewerteten Befunde aus den Ländern am weiteren Lauf
der Donau, deuten in die gleiche Richtung.36
Eines der Verdienste von Heinrich-Tamáska ist es in diesem Zusammen-
hang, die Diskussion um die am Festungsbau ersichtliche spätantike Verteidi-
gungsstrategie aus dem mittlerweile zu engen Pannonien herausgeführt zu
haben, um die Frage aufzuwerfen, ob nicht die gesamte Donaulinie bis ins
Delta Teil eines organisierten Verteidigungssystems gewesen sein könnte.37
Bulgarische Befunde deuten in diese Richtung und im Folgenden soll gezeigt
werden, dass auch der heute rumänische Teil Thrakiens, die Scythia minor, mit
einiger Wahrscheinlichkeit seit constantinischer Zeit in eine umfassende

34
Heinrich-Tamáska 2011, 579. Vgl. Tóth 1987–88, hier 60–61.
35
Zu den Reformen und zur neuen strategischen Ausrichtung etwa: Luttwak 1976, 127–90;
Tomlin 1987; Isaac 1990, 161–218; Campbell 2005. Siehe auch Strobel 2007.
36
Weitere Hinweise mit neuerer Literatur bei Heinrich-Tamáska 2017a. Auch in den Nord-
provinzen lässt sich eine massive Bautätigkeit v.a. im 4. Jht. nachweisen. Vgl. von Petrikovits
1971. Zu den Donauprovinzen weiter etwa Dintchev 2001; 2007; Torbatov 2002; vgl. auch
Poulter 2007.
37
Heinrich-Tamáska 2017a.
202 A. RUBEL

Donauverteidigungsstrategie eingebunden war, wie besonders die Befunde von

Slava Rusă nahe legen.
Das moderne Dorf Slava Rusă wurde bereits Anfang des 20. Jhts. mit der
bei Prokop genannten Siedlung Ibida identifiziert. Die 24 ha umfassende
befestigte Siedlung mit Mauerwerk aus der Spätantike wird seit 2001 systema-
tisch ausgegraben und die jüngsten Befunde zeigen nun an (Festungsbau lässt
sich anhand von Münzfunden und Stratigraphie auf das zweite Drittel des
4. Jhts. eingrenzen), dass die in der ersten Hälfte/Mitte des 4. Jhts. massiv aus-
gebaute Verteidigungsanlage im Kontext der neuen Bau- und Verteidigungspo-
litik gesehen werden muss, die entlang der Donau seit der Tetrarchie und
besonders von Constantin und seinen Söhnen massiv verfolgt worden war.38
In dieses Bild gestaffelter Hinterlandsverteidigung passen auch die Befunde
aus der zur Festungsstadt ausgebauten Siedlung Tropaeum Traiani (Adamc-
lissi) und dem ehemaligen vicus Ulmetum, der im 4. Jht. zur Festung ausge-
baut wurde.39 Damit lassen sich die Befestigungen im Hinterland der Donau in
Scythia minor zwischen dem Fluß und dem Schwarzem Meer gut in die Reihe
der Innenbefestigungen an der Donau einordnen, die eine neue, gestaffelte
Verteidigungslinie markierten und Möglichkeiten für taktische Rückzüge,
sowie zur Aufbewahrung der annona boten.
Erneut muss in diesem Zusammenhang die Frage gestellt werden, die dann
noch im abschließenden Teil dieses Aufsatzes im Zusammenhang genauer
erörter wird, ob angesichts dieser Befunde nicht doch eine nachhaltige Strate-
gie hinter dem über mehrere Generationen erfolgenden Ausbau des Donau-
limes mit einer neuen Staffelung der Befestigungen auch im Hinterland ange-
nommen werden muss, und nicht nur intuitive ad hoc Reaktionen auf aktuelle
Bedrohungslagen hinter den einzelnen Maßnahmen stehen.

3. Valentinian I. und Makrian: Strategie zwischen Diplomatie und

Entführungskomplott

Die direkte und indirekte Kontrolle der Gebiete unmittelbar außerhalb der
Reichsgrenze war für die Römer von vitalem Interesse. Hierbei verfügten die
Römer über eine ganze Reihe von Möglichkeiten von Aufklärung und nach-
richtendienstlicher Informationsbeschaffung40 bis zu massiver militärischer

38
Alle weiteren Daten zum Fundplatz, zum möglichen antiken Siedlungsnamen und zu den
stratigraphischen Verhältnissen zusammenfassend bei Rubel 2019.
39
Die fast exklusiv rumänische Literatur zu diesen Fundplätzen bei Rubel 2019. Zu Tropaeum
Traiani auch Born 2012, 111–30. Zur Bautätigkeit im 4. Jht. in Scythia minor Barnea 1990.
40
Aufgrund der wenigen Quellen ein unterschätzter Bereich. Siehe aber die mannigfachen
Beispiele bei Lee 1993 sowie Austin und Rankov 1995; Sheldon 2005.
 RÖMISCHER EINFLUSS IM BARBARICUM 203

Intervention. Die wichtigste und effektivste, zugleich auch günstigste Methode

der Einflussnahme in den grenznahen Gebieten des Barbaricums war indes ein
avanciertes „client management“, wie Heather die differenzierte Gestaltung
der Beziehungen zu den „Stämmen“ und Königtümern am nördlichen Rand
des Imperium nennt.41 Zwischen geschickter Diplomatie, Bündnispolitik und
militärischem Druck oszillierte auch noch im 4. Jht. die Außenpolitik des
römischen Reiches. In der Spätantike wandelt sich der Umgang mit den
„Klientelrandstaaten“ und ihren Anführern indes zunehmend hin zu einer
Vielfalt von möglichen Instrumenten zur Einflussnahme im Barbaricum, wäh-
rend in der Kaiserzeit militärische Intervention (oder gar Eroberung) und das
Einsetzen oder Dulden von „Klientelkönigen“ dominierte (man denke an die
herodianische Dynastie in Judäa).42 Die Erweiterung der Handlungsoptionen
laßt sich am eindrücklichsten an einem Beispiel erläutern. Nach 350 musste
sich das Imperium auf den Konflikt mit dem Perserreich konzentrieren und
den Usurpator Magnentius in Gallien niederringen und ließ die Zügel an der
Rheingrenze etwas schleifen, wodurch alemannische Lokalfürsten mit ihren
Verbänden eindringen konnten. Julian, dem als Caesar Gallien zugeteilt wor-
den war, konnte zwar erfolgreich gegen die Alemannen bei Straßburg vorge-
hen (357, siehe Amm. 16, 12), musste sich aber dann aber im Konflikt mit
Constantius II. anderen Schauplätzen zuwenden und als sich Augustus ab 361
dann den Reichsgeschäften widmen und später gegen die Perser ziehen. Als
Valentian I. dann ab 365 am Rhein für Ordnung sorgen musste, war ihm der
mächtig gewordene Kleinkönig Macrianus, Anführer des alamannischen Stam-
mes der Bucinobanten, bald ein Dorn im Auge, obwohl dieser zuvor mit Julian
im Einvernehmen gestanden hatte.43 Das Problem mit den von Rom gedulde-
ten „Klientelkönigen“ war natürlich immer, dass diese eigentlich nützlichen

41
Heather 2001. Dieser Artikel beschreibt ausführlich die strategische Dimension dieses
„client management“ der Römer, ein Thema, das hier nur am Rande angeschnitten werden kann,
weswegen nachdrücklich auf den umfassenden Beitrag Heathers verwiesen wird. Hinsichtlich der
Beziehungen zu den östlichen Nachbarn, dem zivilisatorisch und militärisch gleichrangigen Par-
ther- und dann dem Sassanidenreich lagen die Dinge natürlich ganz anders (hierzu u.a. die ein-
schlägigen Beiträge in Baltrusch und Winkler 2015). Der in der römischen Sozialgeschichte aus-
gebildete Begriff vom auf ein gegenseitiges Vertrauensverhältnis gegründeten „Klientelwesen“
wurde nicht ganz glücklich auf die römische Außenpolitik übertragen, sodass sich die Rede von
„Klientelstaaten“ und „Klientelkönigen“ fest eingebürgert hat, hierzu Kehne 2001 mit weiterer
Literatur.
42
Grundlegend Kornemann 1934. Jetzt umfassend zur Kaiserzeit der Sammelband von Balt-
rusch und Winkler 2015. Zum Klientelkönigtum in der Spätantike siehe Winter 1952. Zu den
Verhältnissen in Germanien während der Kaiserzeit Johne 2015.
43
Die Ereignisse schildert Amm. 28, 5; 29, 4; 30, 3. Frieden mit Julian 359/360: Amm. 18,
2, 15–18. Zu Macrianus siehe v.a. Castritius 2001 und Drinkwater 2007 (bes. 304–09), sowie ad
loc. in den Kommentarbänden von den Boeft et al. 2011; 2013; 2015. Vgl. Heather 2001, 45–46.
204 A. RUBEL

Verbündeten oft zur Gefahr wurden, wenn sie (gerade auch aufgrund ihres
durch die Kontakte zu Rom gestiegenen Prestiges) zu mächtig wurden und in
Versuchung kamen, die durch anderweitige außenpolitische Verpflichtungen
des Reichs sich oft ergebenden Gelegenheiten zu nutzen und schlecht
geschützte Grenzgebiete zu überfallen. Dies war offenbar auch beim Unruhe-
stifter (turbarum rex artifex, Amm. 30, 3, 6) Macrianus der Fall, welchem das
Handwerk zu legen zur außenpolitischen Priorität für Valentinian I. wurde.44
Macrianus konnte seine herausragende Position innerhalb seiner peer-group
und sein „Mini-Imperium“45 nur durch sein früheres Bündnis mit den Römern
erlangen, was die potentielle Gefahr derartiger Arrangements für die römische
Seite unterstreicht.46
Valentinians I. erster Versuch, den ungeliebten Anführer der Bucinobanten
zu beseitigen, bestand im Jahr 369/370 in der denkbar effektivsten und ökono-
mischsten Methode, die discordia hostium auszunutzen und in einem Bündnis
mit den zu diesem Zeitpunkt weiter östlich siedelnden und mit den Alamannen
verfeindeten Burgundern gegen Macrianus vorzugehen (Amm. 28, 5, 8–13).
Dabei war es offenbar von Anbeginn an Valentinians Plan gewesen, die Bur-
gunder und die Bucinobanten aufeinander loszulassen, um in diesem Stellver-
treterkrieg keine eigenen Ressourcen engagieren zu müssen.47 Die Burgunder
weigerten sich allerdings, als sie erkannten, dass die Römer selbst nicht gewillt
waren, sich an Kampfhandlungen zu beteiligen, alleine gegen Macrianus vor-
zugehen und zogen mit ihrer großen Streitmacht verärgert von dannen.
Als dieser Versuch fehlgeschlagen war, versuchte Valentinian ein Jahr spä-
ter (371) den unliebsamen Gegner mittels einer gezielten Kommandomission
zu entführen. Als er Nachricht von der Anwesenheit des Macrianus in seiner
Residenz bei Wiesbaden erhalten hatte (offenbar konnte sich Macrianus frei
und entspannt im Grenzgebiet bewegen), schickte er eine Abteilung über den

44
Übergriffe und Aufsässigkeit: Amm. 29, 4, 1–2. „Curbing his power became Valentinians
I’s chief foreign policy aim on the upper Rhine“. Heather 2001, 45. Die Darstellung bei Ammian
ist jedoch leicht übertrieben. Es gibt keine Berichte, dass Macrianus selbst den Rhein in kriege-
rischer Absicht überschritten hätte. Wahrscheinlich erkannte Valentinian die Verwundbarkeit der
wichtigen Stadt Mainz und wollte deshalb die gegenüber, bei Wiesbaden siedelnden Bucinoban-
ten als potentielle Gefahr ausschalten. Drinkwater 2007, 305.
45
Drinkwater 2007, 304.
46
Drinkwater 2007, 249: „Macrianus’ alliance with Julian was the foundation of his Bucino-
bantian ‘empire’“.
47
Heather 2001, 45; Drinkwater 2007, 107; den Boeft et al. 2011, 248. Ammian begründet
die Zurückhaltung der Römer, an der Seite der Burgunder gegen die Bucinobanten vorzugehen,
mit der Angst der Römer angesichts der unerwartet großen Heeresstärke der Burgunder, die
ihrerseits zu einer Gefahr an der Rheingrenze werden konnten. Außerdem sei die Bereitstellung
der Truppen wegen der arbeitsaufwändigen Festungsausbesserungen durch Valentinian noch
nicht abgeschlossen gewesen. Amm. 28, 5, 11.
 RÖMISCHER EINFLUSS IM BARBARICUM 205

Rhein, die allerdings zu viel Lärm machte und so die Gegner aufschreckte, die
ihren Anführer rasch warnen und in letzter Minute in Sicherheit bringen konn-
ten (Amm. 29, 4, 1–5).
Unmittelbar danach griffen die Römer zu einem weiteren Mittel: Sie setzten
einen gewissen Fraomarius als König der Bucinobanten ein, um den sich wei-
ter dem römischen Zugriff entziehenden Macrianus zu entmachten. Jener
konnte sich aber nicht gegen den bei den Römern so wenig gelittenen, bei
seinen eigenen Leuten aber umso beliebteren „Unruhestifter“ durchsetzen (er
musste wohl 372 oder 373 wieder seinem Vorgänger weichen) und wurde in
der Folge von seinen Dienstherren als Kommandeur einer alamannischen
Hilfstruppe nach Britannien geschickt (Amm. 29, 4, 7).48
Als Valentinian 374 einen Feldzug gegen die Quaden vorbereiten musste,
entschied er sich dafür, seinem rechtsrheinischen Gegner Frieden anzubieten,
um an dieser Front Ruhe zu haben. Dieser zelebrierte das auf Treffen mit dem
Augustus und kostete seinen Prestigegewinn weidlich aus.49 Macrianus tauchte
am Treffpunkt bei Mainz am rechten Rheinufer mit wahrhaft königlichem
Gehabe auf: Strotzend vor Stolz präsentierte er sich als der eigentliche Frie-
densstifter (Amm 30, 3, 4: et venit immane quo quantoque flatu distentus ut
futurus arbiter superior pacis). Während der Kaiser auf dem Rhein von einem
Schiff aus verhandelte, war Macrianus wohl aus Sicherheitsgründen50 an „sei-
nem Rheinufer“ geblieben und diskutierte mit stolz erhobenem Haupt (caput
altius erigens) die Bedingungen mit dem Kaiser von Gleich zu Gleich, nicht
ohne dabei auch eine „Show“ abzuziehen, indem er seine Männer wild lär-
mend auf ihre Schilde schlagen ließ. Die beschworene Übereinkunft in beider-
seitigem Einvernehmen sah wahrscheinlich die Garantie seiner rechtsrheini-
schen Besitzungen gegen Bündnistreue vor.51 Interessanterweise entpuppte sich
Macrianus von da an bis zu seinem Tod bei einem Hinterhalt (um 380) im Land
der Franken, in das er wohl in römischem Auftrag eingefallen war, als treuer
Bundesgenosse, der keinerlei Probleme mehr bereitete (Amm. 30, 3, 6–7).52
Mit dem Ablauf der Ereignisse an der Rheingrenze in dieser kurzen Zeit-
spanne lassen sich exemplarisch die unterschiedlichen Mittel erkennen, mit
welchen die Römer ihre Außenpolitik an den Reichsgrenzen konkret gestalten
konnten: 1. Der Stellvertreterkrieg unter Nutzung der Devise divide et impera.

48
Zur Chronologie der Ereignisse Drinkwater 2007, 285–86.
49
Darstellung bei Amm. 30, 3, 4–6.
50
Ob das Treffen auf dem/am Rhein die Gleichrangigkeit der beiden Verhandlungspartner
dokumentieren sollte, oder dem gegenseitigen Misstrauen geschuldet war, bleibt umstritten, Cas-
tritius 2001, 92.
51
Drinkwater 2007, 309.
52
Castritius 2001, 92.
206 A. RUBEL

Die Förderung innergermanischer Konflikte und der gezielte Einsatz von

bestimmten Verbündeten zum Ressourcen schonenden Kampf gegen ausge-
machte Feinde ist die effizienteste Form von „client management“. Selten in
den Quellen beschrieben (aber bereits von Caesar meisterhaft umgesetzt) deu-
tet einiges darauf hin, dass die archäologischen Spuren massiver Konflikte im
Inneren der Germania ab dem 3. Jht. Resultate einer solchen Politik gewesen
sein könnten.53 2. Entführung (oder Ermordung) unliebsamer gegnerischer
Anführer. Besonders die Spätantike ist geprägt von dieser Form der Beseiti-
gung von potentiellen und aktuellen Gefahren durch barbarische Nachbarn.
Das Vorbild für Valentinian war zweifellos Julian gewesen, der den alamanni-
schen Gaukönig Vadomarius entführte, als dieser nichtsahnend einer Einla-
dung zu einem Gastmahl beim Caesar folgte (Amm. 21, 4). Ein weiteres
gefährliches Gastmahl wurde 377 den Gotenführern Fritigern und Alavivus
bereitet, die mit vielen Landsleuten über die Donau nach Scythia minor gelangt
waren und ihrem Gastgeber Lupicinus dauerhafte Probleme bereiteten (Amm.
31, 5). Der halbherzige Mordversuch, dem nur die Leibwachen zum Opfer
fielen, löste bekanntlich den Aufstand der gotischen Flüchtlinge aus, der nach
der Niederlage von Adrianopel (378) und der kompakten Ansiedlung der
Goten unter eigenem Recht (382) enorme Folgen für die innere Verfasstheit
des Reiches haben sollte (Stichwort foederati). Ein weiteres berühmtes Bei-
spiel ist die 449 versuchte Ermordung des Hunnenführers Attila durch einen
verräterischen Verwandten, die zum großen Unglück für das Reich vereitelt
wurde (Priscus fr. 11,1; 15). Eine Vielzahl weiterer Beispiele hat Lee gesam-
melt.54 Dieses durchaus perfide Mittel, das oftmals unter eklatantem Miss-
brauch des Gastrechts angewandt wurde, bringt ebenfalls den Vorteil maxima-
ler Effizienz mit sich, weil die sich um erfolgreiche Anführer versammelnden
Personenverbände der Barbaren, die weitgehend ohne Institutionen auskamen,
durch derartige, gleichsam „chirurgische“ Maßnahmen unmittelbar neutrali-
siert werden konnten. 3. Einsetzung eines Klientelkönigs von Roms Gnaden.
Das Ersetzen unbotmäßiger Herrscher an der Peripherie durch einen Präten-
denten Roms ist seit republikanischen Zeiten ein probates Mittel der Außenpo-
litik. Das funktioniert entweder direkt (wie im nicht ganz glücklichen Fall des
Fraomar) oder durch mittelbare Unterstützung eines Usurpators oder eines
unzufriedenen Verwandten des zu ersetzenden Anführers.55 4. Der klassische

53
Siehe oben sowie zum Beispiel Voß 2017.
54
Lee 2009.
55
Material bei Winter 1952. Germanische Fürsten wurden von den Römern eher mit gewis-
sen Insignien ausgezeichnet, die keine Königswürde im hellenistischen Sinne begründeten. In
diesen Zusammenhang gehört letztlich auch die seit republikanischer Zeit gepflegte Praxis, von
den Vertragspartnern bzw. den Besiegten Geiselgestellung zu verlangen. Die Nachkommen
 RÖMISCHER EINFLUSS IM BARBARICUM 207

zweiseitige Friedensvertrag (mit oder ohne Allianzbestimmungen). Vertragli-

che Einigungen bringen für beide Seiten Verpflichtungen mit sich und bedeu-
ten immer einen Kompromiss, bieten aber den Vorteil maximaler Ressourcen-
schonung, was besonders in Zeiten von umfassendem militärischem
Engagement an mehreren Fronten oft das Gebot der Stunde war. Hierbei kam
es besonders seit dem 5. Jht. zu bedeutenden Verschiebungen, da die Subsidi-
enzahlungen, die zuvor eher Formen der Stimulenz des ehemaligen Gegners
und jetzigen Verbündeten (wichtig auch für das Prestige der von den Zahlun-
gen am meisten profitierenden Anführer) oder fast Mittel zum „Wiederauf-
bau“ oder „Entwicklungshilfe“ waren,56 zunehmend (v.a. seit Attila) zu
erpressten Stillhaltegebühren wurden. Entgegen dem oft kolportierten Bild von
Subsidien als moralisch fragwürdigen Bestechungs- oder Schutzgeldern, waren
die Zahlungen und Geschenke an Barbarenfürsten von herausragender Bedeu-
tung für beide Seiten. Einerseits garantierten sie den Römern eine von grund-
sätzlich in römischem Sinne agierenden „freundlichen Königen“57 geschützte
Pufferzone, andererseits waren die germanischen Anführer auf die Geschenke
angewiesen, weil sie nur durch Umverteilung ihres von außen gewährleisteten
Reichtums an ihre Gefolgschaften ihre Machtposition erhalten konnten.58
Grundsätzlich gehören Subsidien und Geschenke somit zum klassischen Arse-
nal diplomatischer Einflussnahme und waren im Wesentlichen auf drei politi-
sche Ziele zentriert: Bezahlung für Allianzen gegen einen mächtigeren Feind,
das „Erkaufen“ von Frieden (Stillhalteabkommen) sowie Zwietracht und zwi-
schen den benachbarten Barbarengruppen zu schüren und diese zu kriegeri-
schen Auseinandersetzungen anzustiften.59
Dies führt zu den im Kontext unseres Themas anzureißenden weiteren Fra-
gen, die weit ausführlicher behandelt werden müssten, als es hier möglich ist:
Die Bedeutung der sog. römischen „Importe“ (bes. Prestigegüter) und die
abschließende Frage nach den Möglichkeiten langfristiger und konkreter stra-
tegischer Planung seitens des römischen Reichs seit dem 3. Jht. Hinsichtlich
der „Importe“ wird es immer schwierig bleiben, die genauen Wege und

hochrangiger Angehöriger der Eliten (oftmals Königssöhne) wurden in Rom ehrenvoll behandelt
und hatten an einer römischen Erziehung teil. Damit bargen die bereits das Saatkorn für weitere
positive Beziehungen zu Rom bei der Rückkehr in ihre Heimat und waren von Beginn an echte
amici des römischen Volkes. Dieser Aspekt kann hier nicht weiter verfolgt werden, dazu mehr
bei Allen 2006.
56
Heather 2011, 88–90. Vgl. Heather 2001.
57
Braund 1984, passim.
58
Braund 1989, 18; Heather 2011, passim. Allgemein zu Subsidien weiter der grundlegende
Artikel von Gordon 1949, der seine schwer zugängliche Dissertation in Michigan 1948 zusam-
menfasst (Mikrofiche).
59
Gordon 1949, 60.
208 A. RUBEL

historischen Kontexte ihrer Verbringung ins Barbaricum zu verfolgen. Nach-

dem man anfänglich besonders an den Handel als Ursache für die Präsenz
römischer Waren in den nördlichen ans Reich grenzenden Gebiete und weit
darüber hinaus60 gedacht hatte (daher auch der eingebürgerte Importbegriff),
hat sich im 20. Jht. und zu Beginn des neuen Jahrtausends eine neue communis
opinio herausgebildet, die mehrere Faktoren für die Verbringung solcher
Güter ins Barbaricum als maßgeblich ansieht, wobei deren Gewichtung durch-
aus unterschiedlich gesehen werden kann.61 Neben Handel sind dabei vor
allem Kriegsbeute oder Raubgut, sowie für Dienstleistungen erhaltene
Geschenke und Subsidien. In unserem Zusammenhang wird es auch darum
gehen, zukünftig die umstrittene Frage erneut aufzuwerfen, ob nicht ein gehö-
riger Teil v.a. der Buntmetallimporte nicht einem diplomatischen Kontext
zugeordnet werden müsste.62 Spektakuläre Funde aus Raubzügen in Krisen-
zeiten, wie der sog. „Barbarenschatz“ von Neupotz, der von einem missglück-
ten Beutezug einer alamannischen Bande (um 260) zeugt, die ihr Raubgut bei
der Flucht über den Rhein verloren hatte, mögen davon ablenken, dass über
einen langen Zeitraum der Großteil des römischen „Imports“ im Rahmen
eines friedlichen Austauschs Eingang ins Barbaricum fand.63 Dass gerade in
der Spätantike hierbei diplomatische Geschenke und Ehren- sowie Rang-
abzeichen eine besonders wichtige Rolle auch im Sozialgefüge der barbari-
schen Gesellschaften spielten, könnte möglicherweise zu einer leichten Ver-
schiebung bei den Einschätzungen zu den „Importen“ v.a. auch im unteren
Donauraum führen.64
Im Weiteren wird auch zu fragen sein, ob die Vielzahl neuen Fundmaterials
und neue Erkenntnisse (gerade auch bezüglich des Festungsbaus, Stichwort
Innenbefestigungen) nicht dazu Anlass geben könnten, die Frage nach der
Möglichkeit einer nachhaltigen römischen Strategie im 4. Jht. erneut zu stellen
und mit den notwendigen Einschränkungen affirmativ zu beantworten. Seit
Luttwaks aus der Perspektive der modernen Sicherheitspolitik im Kalten Krieg

60
Zu den römischen Funden in weiter von der Reichsgrenze entfernten Gebieten siehe nur
zum Beispiel Grane 2007 für Skandinavien oder Hrnčiarek 2013 für die heutige Slowakei, sowie
die einschlägigen Bände des CRFB.
61
Aus der Vielzahl von einschlägigen Publikationen nenne ich hier nur einige, etwa die wich-
tigste Überblicksdarstellung von Wolters, Erdrich und Voß 2003, sowie Wolters 1990, 1991 und
Stupperich 1993.
62
So hatte schon Erdrich 2001 aber anhand unzureichenden Fundmaterials argumentiert. Hef-
tig zurückgewiesen von Kehne 2003.
63
Neupotz: Der Barbarenschatz 2006. Friedlicher Austauch (v.a. durch Handel): Ruffing
2008.
64
Vielfalt von Material bei Schmauder 2002. Quellen zu diplomatischen Geschenken und
Prestigeobjekten bei Nechaeva 2014.
 RÖMISCHER EINFLUSS IM BARBARICUM 209

geschriebenes Buch die Altertumswissenschaften durcheinanderwirbelte, sind

die Meinungen darüber geteilt, ob das römische Reich seit der spätrepublika-
nischen Expansion und der Kaiserzeit jemals in der Lage gewesen sei, eine
durchdachte Verteidigungsstrategie zu entwickeln, die ohnehin in den Quellen
nicht thematisiert wird.65 Luttwak glaubte im ausgehenden 3. und im 4. Jht.
Anzeichen für eine neue militärische Verteidigungsstrategie zu erkennen, die
die Tiefe des Raumes berücksichtigte und (nach den Reformen des Diokletian
und des Constantin) bei Barbareneinfällen Rückzugsmöglichkeiten im Hinter-
land vorsah, die dann durch strategische mobile Einheiten entsetzt werden
konnten und die er „defence-in-depth“ nannte. Gegen Luttwaks Thesen und
für das Fehlen eines strategischen Verteidigungskonzepts wurden in erster
Linie das nicht ausgebildete (bzw. uns nur in Umrissen bekannte – das ist ein
Unterschied) Nachrichtenwesen der Römer sowie das Fehlen von einschlägi-
gen literarischen Quellen, die auf strategische Pläne verwiesen, ins Feld
geführt. In apodiktischem Ton urteilten die von einem Außenseiter provozier-
ten Fachleute, dass das Imperium weder die Mittel und Kommunikationstech-
niken besessen habe, ein in die Tiefe gestaffeltes Verteidigungssystem zu eta-
blieren, noch den Willen und die ideologische Grundlage für ein solches
Vorgehen gehabt habe, da kriegerischer Erfolg und aggressiver Imperialismus
während der gesamten Existenz des Reiches die ideologische Grundlage und
damit auch die Handelsmaxime für die Kaiser bildete, deren erfolgreiche
Selbstdarstellung von der Mehrung des Reiches und damit vom militärischen
Erfolg abhing.66
Betrachtet man nun die jüngsten Forschungsergebnisse zu den Innenbefesti-
gungen entlang der Donau im Zusammenhang, so darf man vielleicht doch mit
aller Vorsicht die grundsätzliche Frage, ob das Imperium nicht doch ansatz-
weise eine Form einer „grand strategy“ entwickelt haben könnte, erneut auf-
werfen. Das komplexe Thema kann hier nur in wenigen Sätzen angerissen
werden, jedoch sollte man sich die Frage stellen, ob hier nicht der archäologi-
sche Befund Hinweise bereithält, die man vielleicht in ein neues Gesamtbild

65
Luttwak 1976. Die Positionen zu Luttwaks Strategiediskussion sind mit weiteren Literatur-
angaben erörtert bei Rubel 2019 (ich folge hier weitgehend dieser Darstellung). Besonders kri-
tisch war Mann 1979. Eine knappe Übersicht der Positionen bei Wheeler 2007, 237–38.
66
Besonders deutliche Kritik bei Isaac 1990 und Whittaker 1994. Die Diskussion ist bei
Wheeler 1993a; 1993b. zusammengefasst, der seinerseits für eine strategische Ausrichtung des
römischen Militärs argumentiert und zu den wenigen gehört, die Luttwaks Thesen positiv aufge-
nommen haben; aktualisiert in Wheeler 2007, 237–38. Siehe auch Kagan 2006, 335–48. Millar
1982, 9–10, betont die langen Verzögerungen bei der Nachrichtenübermittlung im Reich und die
daraus entstehenden Defizite für zeitnahe und adäquate Entscheidungsfindung. Zum (für vor-
moderne Gesellschaften) durchaus eindrucksvollen Nachrichtenwesen der Römer siehe dennoch
Lee 1993; Austin und Rankov 1995; Sheldon 2005.
210 A. RUBEL

einordnen muss. Schon mit der Tetrarchie wird begonnen, wichtige Zentren
wieder aufzubauen, bzw. zu renovieren (im Falle der Dobrudscha ist das vor
allem die Provinzhauptstadt Tomis).67 Entlang der Donau entsteht spätestens
seit Constantin dem Großen ein Netzwerk von in regelmäßigen Abständen an
wichtigen Binnenverkehrsadern im unmittelbaren Limeshinterland gelegenen
Festungen, bzw. existierende Städte und Ortschaften werden zu stark befestig-
ten Siedlungen ausgebaut, die gleichzeitig zivilen und militärischen Charakter
aufweisen. Diese Plätze dienen offenbar auch der Truppenversorgung im Hin-
terland (annona) und können mindestens zeitweise Militäreinheiten aufneh-
men. Dieses Netzwerk von im Schnitt rund 20-50 km vom Limes entfernten
Stützpunkten wird stetig ausgebaut erfährt zu verschiedenen Zeiten und je
nach militärischer Lage (Zerstörungen nach Einfällen seit der 2. Hälfte des
4. Jhts.) mehr oder weniger umfassende Reparaturen und Renovierungen. Die
nachhaltige Sorge, mit welcher diese Festungen und Siedlungen seitens der
Zentralverwaltung bedacht wurden, verweist m.E. auf eine über isolierte Maß-
nahmen einzelner Kaiser hinausgehende Verteidigungsstrategie, die ein flexib-
les und kurzfristiges Reagieren auf das Eindringen von Barbarengruppen
ermöglichte.
In den 1970er Jahren konnte man die strukturellen Gemeinsamkeiten dieser
über einen weiten Raum verteilten und in ihrer Grundkonzeption und Anlage
vergleichbaren Bauten noch nicht ermessen. Vor dem Hintergrund eines sich
wandelnden Gesamtbildes der Spätantike, die heute nicht mehr einseitig als
Epoche eines zwangsläufigen und unaufhaltbaren Niedergangs betrachtet wird,
bilden neue Erkenntnisse, wie etwa die archäologische Bestätigung des von
der Forschung für unmöglich gehaltenen Eroberungsfeldzugs des Maximinius
Thrax ins Innere Germaniens im Jahre 235, der im sog. „Harzhornereignis“
seinen nachweislichen Niederschlag gefunden hat, die Grundlage für neue
Denkansätze.68 Aus heutiger Sicht und in Kenntnis der historischen Entwick-
lung ist es leicht zu konstatieren, dass das römische Reich seit dem 3. Jht. dem
Untergang geweiht gewesen sei. Dennoch scheinen sich Anhaltspunkte für
eine differenzierte Sichtweise zu mehren, nach der der viel geschmähte spätan-
tike „Zwangsstaat“ keineswegs hilflos oder gar kopflos und mit dem Rücken
zur Wand agierte, sondern versuchte, den Herausforderungen adäquat und mit
neuen Konzepten zu begegnen.
Dass wir über die Defensivstrategie an der Donau nicht aus den antiken
Schriften informiert werden, ist dabei nicht zwangsläufig ein Beleg für die
Inexistenz einer solchen Strategie. Über die ökonomischen Verhältnisse im

67
Barnea 1990, 286.
68
Pöppelmann et al. 2013.
 RÖMISCHER EINFLUSS IM BARBARICUM 211

Reich und über die Dimensionen des Handels erfahren wir schlechterdings
überhaupt nichts in den antiken Schriften. Gleichwohl wissen wir natürlich seit
Rostovtzeff, welche enormen Ausmaße der ökonomische Austausch seit dem
Hellenismus in der Mittelmeerwelt angenommen hat. Vielleicht trauen wir dem
spätantiken Staat nicht genug zu, wenn wir sein Potential zur strategischen Pla-
nung als zu gering für eine „grand strategy“ veranschlagen. Entsprechend
erscheint die „Primitivisierung“ des römischen Militärwesens als eine ähnlich
eindimensionale Fehleinschätzung, wie die sozialgeschichtliche Verengung
ökonomischer Komplexität durch die „Primitivisten“ in der Wirtschaftsge-
schichte.69 Möglicherweise sind die Vorstellungen von einer elaborierten
„defence-in-depth“-Strategie und die moderne Terminologie aus dem Bereich
der Sicherheitspolitik des 20. und 21. Jhts. nicht die geeigneten Kategorien zur
Beschreibung der militärischen Sachverhalte in der Spätantike, und auch grund-
legende Kritik am Ansatz Luttwaks mag gerechtfertigt sein.70 K. Kagan resü-
mierte die Diskussion um Luttwaks Werk deshalb folgendermaßen:
Luttwak performed an invaluable service to the field of Roman history by provok-
ing an important and intriguing series of controversies. In the present state of
these studies, however, it is clear that his fundamental assumptions, arguments,
and conclusions cannot stand. However grand strategy in the Roman Empire
worked, it did not work as Luttwak has described it.71
Aus diesem Grund ist man in der Altertumswissenschaft weitgehend einig dar-
über, dass die Erforschung einer eventuellen „grand strategy“ im römischen
Reich anachronistisch und dem Gegenstand nicht angemessen sei, weshalb
derartige Studien, abgesehen von solchen, die das Fehlen jedweder Strategie
konstatierten, nach Luttwak weitgehend ausblieben.72 Dennoch bleibt als Desi-
derat für zukünftige Überlegungen die angemessene Bestimmung der strategi-
schen Dimension kaiserlicher Entscheidungen und ihrer Nachhaltigkeit sowie
der Auswirkungen von verteidigungspolitischen Maßnahmen in der longue
durée. Kagan schlug eine andere Herangehensweise vor, die berücksichtigt,
dass wir aufgrund der Quellenlage (die Ziele und Pläne der römischen Ent-
scheider sind uns völlig unbekannt, nur deren Auswirkungen in Truppenbewe-
gungen und -stationierungen etc. lassen sich eventuell rekonstruieren) im Falle
des Imperiums nicht in einer ähnlichen Weise vorgehen können, wie beim
Erforschen moderner Staaten und ihrer strategischen Ziele. Wie F. Millar
betont auch Kagan den methodischen Vorrang der wenigen durch literarische

69
Wheeler 2007, 237–38.
70
Heather 2001 nimmt eine diesbezüglich vermittelnde Position ein.
71
Kagan 2006, 346.
72
Kagan 2006, 348.
212 A. RUBEL

Quellen belegbaren Hinweise auf strategische Entscheidungen vor den wackli-

geren argumenta ex silentio und betont die Aussagekraft der zielgerichteten
Truppendislokationen besonders während des Prinzipats, welche allein die
Basis für die Analyse einer eventuellen Strategie des Imperiums bildeten.73
Dennoch weist der hier kursorisch vorgestellte archäologische Befund
zusammen mit unseren Kenntnissen über die militärischen Reformen unter
Diokletian und Constantin dem Großen sowie die Entwicklung der römisch-
byzantinischen Diplomatie zu einer Kunstform74 auf grundlegendere Verände-
rungen im Verteidigungsbereich hin, die planmäßig und nachhaltig über meh-
rere Generationen von der Führung in Rom und dann in Konstantinopel
betrieben wurden, auch wenn wir über einen gezielten Ausbau und Umbau der
Verteidigungsstruktur an der Donau aus den Quellen nichts erfahren, sowie
kaum Kenntnisse über die internen Wege der Entscheidungsfindung (Berater-
stäbe, „Hof“ etc.) haben.75 Dass es sich hierbei um eine zweifellos „strategi-
sche“ Antwort auf neue Herausforderungen handelte, die allerdings nicht nur
mit äußeren Bedrohungen durch Barbareneinfälle sondern auch mit internen
Schieflagen (Usurpationsproblematik, Fiskalregime etc.) in Verbindung stan-
den, wird man nicht bestreiten können. Aus Sicht der Gegebenheiten des
4. Jhts. waren diese Maßnahmen mehr als sinnvoll und überaus wirksam und
darüber hinaus komplementär zu den regelmäßigen Strafexpeditionen, in deren
Folge die zur Raison gebrachten Klientelstaaten mit Subsidien und Handels-
recht belohnt und handzahm gemacht wurden;76 mit Attila und dem Wegfall
eines guten Teils des Steueraufkommens durch Gebietsverluste im 5. Jht.
konnte niemand rechnen. Wir wissen zu wenig über die geographischen und
kartographischen Kenntnisse der römischen Führung, oder achten diese für
gering, und militärische Aufklärung sowie die nachrichtendienstliche Informa-
tionsgewinnung der Militärführung und der Grenztruppen werden in den Quel-
len nicht so prominent beschrieben, wie man sich dies als Historiker wünschen
würde.77 Aus dem Fehlen einer einschlägigen Überlieferung auf einen

73
Kagan 2006, 353; Millar 1982, 3–4.
74
Einschlägig Nachaeva 2014.
75
Hierzu ausführlich Millar 1982.
76
Eine Politik im Stil von „Zuckerbrot und Peitsche“, am deutlichsten sichtbar in Constan-
tins des Großen Kampagne gegen die Goten, die mit dem langanhaltenden Frieden von 332
endete, erkennbar, siehe dazu Heather 2001, 25, 29–30 und 2011, 51–52, sowie Kulikowski
2009, 88–92.
77
Unsere Kenntnisse über die exploratores, frumentarii oder agentes in rebus sind begrenzt,
bieten aber dennoch Stoff genug für monographische Arbeiten: zum Beispiel Sheldon 2005;
Austin und Rankov 1995; Heather 2001, 31–32 hält die durch Spione und den wirtschaftlichen
Austausch gewonnen Informationen, über die das Imperium verfügte, für erheblich.
 RÖMISCHER EINFLUSS IM BARBARICUM 213

unzureichenden Organisationsgrad zu schließen, erscheint indes fahrlässig.78

Die organisatorischen und verwaltungstechnischen Leistungen des römischen
Reichs sowie ein vergleichbarer zivilisatorischer Entwicklungsstand wurden in
Europa erst wieder im 19. Jht. erreicht.79 Nur die archäologisch fundierte Han-
delsgeschichte oder das umfangreiche Material aus dem römischen Ägypten
können uns einen (unzureichenden) Blick durchs Schlüsselloch auf komplexe
Verwaltungs- und Organisationsprozesse gewähren, die man vormodernen
Gesellschaften eigentlich nicht zutrauen würde.80 Entsprechend einseitig wäre
es, die Möglichkeit komplexerer strategischer Planungen für den römischen
Kaiserhof schon arbeitshypothetisch auszuschließen. Argumente für eine über
ad-hoc Maßnahmen hinausgehende, nachhaltige Verteidigungsstrategie, die
möglicherweise besonders während der Herrschaft Constantins detaillierter
ausgearbeitet und auch von seinen Nachfolgern weiter verfolgt wurde und auf
Diplomatie, Einflussnahme, dem Prinzip divide et impera sowie auf strategi-
schen Baumaßnahmen beruhte, habe ich hier in der gebotenen Kürze
vorgebracht.

BIBLIOGRAPHIE

Allen, J. 2006: Hostages and Hostage-Taking in the Roman Empire (Cambridge).

Austin, N.J.E. und Rankov, N.B. 1995: Exploratio: Military and Political Intelligence
in the Roman World from the Second Punic War to the Battle of Adrianople
(London).
Baltrusch, E. und Winkler, J. (Hrsg.) 2015: Amici–socii–clientes? Abhängige Herr-
schaft im Imperium Romanum (Topoi: Berlin Studies of the Ancient World 31)
(Berlin).
Der Barbarenschatz 2006: Der Barbarenschatz: Geraubt und im Rhein versunken
(Speyer).
Barnea, A. 1990: ‘Einige Bemerkungen zur Chronologie des Limes an der Unteren
Donau in Spätrömischer Zeit’. Dacia n.s. 34, 285–90.
den Boeft, J., Drijvers, J.W., den Hengst, D. und Teitler, H. 2011: Philological and
Historical Commentary on Ammianus Marcellinus XXVIII (Leiden).
—. 2013: Philological and Historical Commentary on Ammianus Marcellinus XXIX
(Leiden).
—. 2015: Philological and Historical Commentary on Ammianus Marcellinus XXX
(Leiden).

78
Whittaker 1994, 64–72.
79
Morris 2011, 158–66.
80
Nur zum Beispiel die komplexen Hintergründe der Truppenversorgung in Ägypten: Mitt-
hof 2001. Siehe auch die aus den Papyri gewonnen Informationen zum spätantiken Heerwesen,
die Palme 2004 zusammenfasst.
214 A. RUBEL

Born, R. 2012: Die Christianisierung der Städte der Provinz Scythia Minor: Ein Bei-
trag zum spätantiken Urbanismus auf dem Balkan (Spätantike, frühes Christen-
tum, Byzanz. Reihe B, Studien und Perspektiven 36) (Wiesbaden).
Braund, D. 1984: Rome and the Friendly King: The Character of the Client Kingship
(London/New York).
—. 1989: ‘Ideology, Subsidies and Trade. The King on the Northern Frontier Revisi-
ted’. In Barrett, J.C., Fitzpatrick A.P. und Macinnes, L. (Hrsg.), Barbarians and
Romans in North-West Europe from the Later Republic to Late Antiquity (BAR
International Series 471) (Oxford), 14–26.
Brown, P. 1971: The World of Late Antiquity: From Marcus Aurelius to Muhammad
(London).
Campbell, B. 2005: ‘The Army II: The Military Reforms of Diocletian and Constan-
tine’. In Bowman, A. et al. (Hrsg.), The Cambridge Ancient History 12: The
Crisis of Empire, AD 193–337, 2. Aufl. (Cambridge), 120–30.
Castritius, H. 2001: ‘Macrianus’. Reallexikon der Germanischen Altertumskunde 19,
90–92.
Demandt, A. 2014: Der Fall Rom: Die Auflösung des römischen Reiches im Urteil der
Nachwelt, 2. Aufl. (München).
Dintchev, V. 2001: ‘Gesellschaftliche Komponenten der Struktur frühbyzantinischer
befestigter Siedlungen auf dem Territorium des heutigen Bulgarien’. In Wendel,
M. (Hrsg.), Karasura: Untersuchungen zur Geschichte und Kultur des alten
Thrakien. 1: 15 Jahre Ausgrabungen in Karasura (Schriften des Zentrum für
Archäologie und Kulturgeschichte des Schwarzmeerraumes 1) (Langenweißbach),
217–35.
—. 2007: ‘The Fortresses of Thrace and Dacia in the Early Byzantine Period’. In
Poulter, A.G. (Hrsg.), The Transition to Late Antiquity on the Danube and Beyond
(Proceedings of the British Academy 141) (Oxford), 479–546.
Drinkwater, J.F. 2007: The Alamanni and Rome, 213–496 (Caracalla to Clovis)
(Oxford).
Erdrich, M. 2001: Rom und die Barbaren: Das Verhältnis zwischen dem Imperium
Romanum und den germanischen Stämmen vor seiner Nordwestgrenze von der
späten römischen Republik bis zum Gallischen Sonderreich (Römisch-germanische
Forschungen 58) (Mainz).
Goffart, W. 2006: Barbarian Tides: The Migration Age and the Later Roman Empire
(Philadelphia).
Gordon C.D. 1949: ‘Subsidies in Roman Imperial Defence’. Phoenix 3, 60–69.
Grane, T. 2007: The Roman Empire and Southern Scandinavia – a Northern Connec-
tion! A Re-evaluation of Military-Political Relations between the Roman Empire
and the Barbaricum in the First Three Centuries AD with a Special Emphasis on
Southern Scandinavia (Dissertation, Kopenhagen).
Harper, K. 2017: The Fate of Rome: Climate, Disease, and the End of an Empire
(Princeton).
Heather, P. 2001: ‘The Late Roman Art of Client Management: Imperial Defence in
the Fourth Century West’. In Pohl, W. und Wood, I. (Hrsg.), The Transformation
of Frontiers from Late Antiquity to the Carolingians (Transformations of the
Roman World 10) (Leiden/Boston), 15–68.
—. 2011: Invasion der Barbaren: Die Entstehung Europas im ersten Jahrtausend
nach Christus (Stuttgart).
 RÖMISCHER EINFLUSS IM BARBARICUM 215

Heinrich-Tamáska, O. 2011: ‘Pannonische Innenbefestigungen und römische Konti-

nuität: Forschungsstand und -perspektiven’. In Konrad, M. und Witschel, C.
(Hrsg.), Römische Legionslager in den Rhein- und Donauprovinzen – Nuclei
frühmittelalterlichen Lebens? (Bayerische Akademie der Wissenschaften,
Philosophisch-Historische Klasse, Abhandlungen n.f. 138) (München), 517–34.
—. (Hrsg.) 2013: Keszthely-Fenékpuszta: Katalog der Befunde und ausgewählter
Funde sowie neue Forschungsergebnisse (Castellum Pannonicum Pelsonense 3)
(Rahden).
—. 2015: ‘Inner Fortifications and the Late Roman Defensive System in Pannonia
(4th/5th c. AD) – Some Remarks’. In Vagalinski, L. und Sharankov, N. (Hrsg.),
Limes XXII (Proceedings of the 22nd International Congress of Roman Frontier
Studies, Ruse, Bulgaria, September 2012) (Sofia), 119–25.
—. 2017a: ‘Castra and Towns in the Hinterland of the Limes during Late Antiquity:
Pannonia and the Provinces along the Lower Danube in Comparison’. Acta
Archaeologica Carpathica 52, 205–40.
—. 2017b: ‘Castellum – Castrum – Civitas? L’évolution fonctionnelle des nouveaux
établissements de l’Antiquité tardive en Pannonie et en Mésie seconde: une étude
comparative’. In Rizos, E. (Hrsg.), New Cities in Late Antiquity: Documents and
Archaeology (Bibliothèque de l’Antiquité Tardive 35) (Turnhout), 39–55.
Hose, M. 2013: ‘Ausgelöschte Geschichte. Der Feldzug des Maximinus Thrax in
das Innere Germaniens 235/236 n. Chr. in der historischen Überlieferung’. In
Pöppelmann et al. 2013, 111–15.
Hrnčiarek, E. 2013: Römisches Kulturgut in der Slowakei: Herstellung, Funktion und
Export römischer Manufakturerzeugnisse aus den Provinzen in der Slowakei,
2 Bd. (Universitätsforschungen zur prähistorischen Archäologie 222) (Bonn).
Isaac, B. 1990: The Limits of Empire: The Roman Army in the East (Oxford).
Johne, K.-P. (Hrsg.) 2008: Die Zeit der Soldatenkaiser: Krise und Transformation
des Römischen Reiches im 3. Jahrhundert n. Chr. (235–284), 2 Bd. (Berlin),
235–84.
—. 2015: ‘Klienten, Klientelstaaten und Klientelkönige bei den Germanen’. In
Baltrusch und Winkler 2015, 225–42.
Kagan, K. 2006: ‘Redefining Roman Grand Strategy’. The Journal of Military History
70, 333–62.
Kehne, P. 2001: ‘Klientelrandstaaten’. Reallexikon der Germanischen Altertumskunde
17, 11–13.
—. 2003: Rezension zu Erdrich 2001. Gnomon 75, 323–27.
Kornemann, E. 1934: Die unsichtbaren Grenzen des römischen Kaiserreiches
(Veröffentlichungen des Ungarischen nationalen Ausschusses für internationale
geistige Zusammenarbeit 2) (Budapest).
Kulikowski, M. 2009: Die Goten vor Rom (Darmstadt).
Lee, A.D. 1993: Information and Frontiers. Roman Foreign Relations in Late Antiquity
(Cambridge).
—. 2009: ‘Abduction and Assassination: The Clandestine Face of Roman Diplomacy
in Late Antiquity’. The International History Review 31, 1–23.
Luttwak, E.N. 1976: The Grand Strategy of the Roman Empire from the First Century
A.D. to the Third (Baltimore/London).
Mann, J.C. 1979: ‘Power, Force and the Frontiers of the Empire’. Journal of Roman
Studies 69, 175–83.
216 A. RUBEL

Millar, F. 1982: ‘Emperors, Frontiers and Foreign Relations, 31. B.C. to A.D. 378’.
Britannia 13, 1–23.
Mitthof, F. 2001: Annona militaris: Die Heeresversorgung im spätantiken Ägypten.
Ein Beitrag zur Reichs- und Verwaltungsgeschichte des Römischen Reiches im
3. bis 6. Jh. n. Chr. (Papyrologica Florentina 32) (Florenz).
Morris, I. 2011: Wer regiert die Welt? Warum Zivilisationen herrschen oder beherrscht
werden (Frankfurt).
Nechaeva, E. 2014: Embassies–Negotiations–Gifts: Systems of East Roman Diplo-
macy in Late Antiquity (Geographica Historica 30) (Stuttgart).
Nørgård Jørgensen, A. und Andersen, H.C.H. 2014: Ejsbøl Mose. Die Kriegsbeuteop-
fer im Moor von Ejsbøl aus dem späten 1. Jh. v. Chr. bis zum frühen 5. Jh. n. Chr.
(Jutland Archaeological Society Publications 80) (Højbjerg).
Palme, B. 2004: ‘Die römische Armee von Diokletian bis Valentinian I.: Die papyro-
logische Evidenz’. In Le Bohec, Y. und Wolff, C. (Hrsg.), L’armée romaine de
Dioclétien à Valentinien Ier (Lyon), 101–15.
von Petrikovits, H. 1971: ‘Fortifications in the North-Western Roman Empire from the
Third to the Fifth centuries A.D.’. Journal of Roman Studies 61, 178–218.
Pöppelmann, H., Deppmeyer, K. und Steinmetz, W.-D. (Hrsg.) 2013: Roms vergesse-
ner Feldzug: Die Schlacht am Harzhorn. Katalog zur Niedersächsischen Landes-
ausstellung (Stuttgart).
Poulter, A. 2007: ‘The Transition to Late Antiquity’. In Poulter, A.G. (Hrsg.), The
Transition to Late Antiquity on the Danube and Beyond (Proceedings of the
British Academy 141) (Oxford), 1–50.
Rau, A. 2012: ‘Das nördliche Barbaricum zur Zeit der Krise des 3. Jahrhunderts n.
Chr. Einige kritische Anmerkungen zur Diskussion über provinzialrömischnord-
europäische Verbindungen’. In Fischer, T. (Hrsg.): Die Krise des 3. Jahrhunderts
n. Chr. und das Gallische Sonderreich (Akten des Interdisziplinären Kolloquiums
Xanten 26. bis 28. Februar 2009) (Zakmira 8) (Wiesbaden), 343–430.
Rau, A. und von Carnap-Bornheim, C. 2012: ‘Die kaiserzeitlichen Heeresausrüstungs-
opfer Südskandinaviens – Überlegungen zu Schlüsselfunden archäologisch-
historischer Interpretationsmuster in der kaiserzeitlichen Archäologie’. In Beck, H.,
Geuenich, D. und Steuer, H. (Hrg.), Altertumskunde–Altertumswissenschaft–
Kulturwissenschaft. Erträge und Perspektiven nach 40 Jahren Reallexikon der
Germanischen Altertumskunde (Reallexikon der Germanischen Altertumskunde
Ergänzungsbände 77) (Berlin/Boston), 515–40.
Rubel, A. 2016: ‘Überlegungen zum Barbarenbegriff der Römer. Geten, Thraker und
Daker in den Augen der Römer’. In Rubel, A. (Hrsg.), Die Barbaren Roms:
Inklusion, Exklusion und Identität im römischen Reich und im Barbaricum (1.–
3. Jht. n. Chr.) (Studien zu Archäologie und Geschichte des Altertums 2)
(Konstanz), 11–41.
—. 2018: ‘Gewalt für die Götter: Waffen- und Menschenopfer bei den Germanen’.
In Nemeth, E. (Hrsg.), Violence in Prehistory and Antiquity (Antiquitas 6)
(Kaiserslautern/Mehlingen), 267–78.
—. 2019: ‘Die befestigte Siedlung (L)Ibida/Slava Rusă (Landkreis Tulcea, Rumänien)
im Kontext der Innenbefestigungen an der Donau’. In Heinrich-Tamáska, O.
(Hrsg.), Neue Forschungen in Keszthely-Fenékpuszta 1: Landschaftsarchäologie,
Naturwissenschaften und Vergleichsbeispiele (Castellum Pannonicum Pelsonense.
7) (Rahden).
 RÖMISCHER EINFLUSS IM BARBARICUM 217

Ruffing, K. 2008: ‘Friedliche Beziehungen. Der Handel zwischen den römischen

Provinzen und Germanien’. In Schneider, H. (Hrsg.), Feindliche Nachbarn: Rom
und die Germanen (Köln/Wien), 153–66.
Sarnowski, T. 1991: ‘Barbaricum und ein Bellum Bosporanum in einer Inschrift aus
Preslav’. ZPE 87, 137–44.
Schmauder, M. 2002: Oberschichtsgräber und Verwahrfunde in Südosteuropa im
4. und 5. Jahrhundert. Zum Verhältnis zwischen dem spätantiken Reich und der
barbarischen Oberschicht aufgrund der archäologischen Quellen (Bukarest).
Schmitt, J.-P. und Voß, H.-U. 2017: ‘Mars an der Uecker. Römische Schwerter und
germanische Krieger an der unteren Oder’. In Eriksen, B.V., Abegg-Wigg, A.,
Bleile, R. und Ickerodt, U. (Hrsg.), Interaktion ohne Grenzen Beispiele
archäologischer Forschungen am Beginn des 21. Jahrhunderts, Bd. 1 Schleswig),
209–25.
Sheldon, R.M. 2005: Intelligence Activities in Ancient Rome: Trust in the Gods, but
Verify (London).
Sommer, M. 2015: ‘“A Vast Scene of Confusion”. Die Krise des 3. Jahrhunderts in
der Forschung’. In Babusiaux, U. und Kolb, A. (Hrsg.), Das Recht der „Soldaten-
kaiser“. Rechtliche Stabilität in Zeiten politischen Umbruchs (Berlin/Boston),
15–30.
Soproni, S. 1978: Der spätrömische Limes zwischen Esztergom und Szentendre: Das
Verteidigungssystem der Provinz Valeria im 4. Jahrhundert (Budapest).
Stickler, T. 2007: ‘The Foederati’. In Erdkamp, P. (Hrsg.), A Companion to the Roman
Army (Malden, MA/Oxford), 495–514.
Strobel, K. 1993: Das Imperium Romanum im „3. Jahrhundert“. Modell einer histo-
rischen Krise? Zur Frage mentaler Strukturen breiterer Bevölkerungsschichten in
der Zeit von Marc Aurel bis zum Ausgang des 3. Jh. n. Chr. (Historia Einzel-
schriften 75) (Stuttgart).
—. 2007: ‘Strategy and Army Structure between Septimius Severus and Constantine
the Great’. In Erdkamp, P. (Hrsg.), A Companion to the Roman Army (Malden,
MA/Oxford), 267–85.
Stupperich, R. 1993: ‘Bemerkungen zum römischen Import im Freien Germanien’. In
Franzius, G. (Hrsg.), Aspekte römisch-germanischer Beziehungen in der frühen
Kaiserzeit (Vortragsreihe zur Sonderausstellung „Kalkriese – Römer im
Osnabrücker Land“ 1993 in Osnabrück) (Quellen und Schrifttum zur Kultur-
geschichte des Wiehengebirgsraumes. Reihe B, 1; Schriftenreihe Kulturregion
Osnabrück des Landschaftsverbandes Osnabrück e.V. 6) (Espelkamp), 45–98.
Tomlin, R.S.O. 1987: ‘The Army of the Late Empire’. In Wacher, J. (Hrsg.), The
Roman World, Bd. 1 (London), 107–20.
Torbatov, S.B. 2002: Ukrepitelnata system na provintsiya Skitiya (kraya na III–VII v.)
(Sofia).
Tóth, E. 1985: ‘Zur Chronologie der militärischen Bautätigkeiten des 4. Jh. Pannonien’.
Mitteilungen des archäologischen Instituts der ungarischen Akademie der
Wissenschaften 14, 121–36.
—. 1987–88: ‘Vorbericht über die Ausgrabung der Festung und des Gräberfeldes von
Alsóhetény 1981–1986. Ergebnisse und umstrittene Fragen (Ungarisch mit dt.
Zusammenfassung)’. Archaeologiai Étesítö 114–15, 22–61.
—. 2009: ‘Die spätrömische Militärarchitektur in Transdanubien’. Archaeologiai
Étesítö 134, 31–61.
218 A. RUBEL

Voß, H.-U. 2017: ‘Die Beziehungsgeflechte germanischer Eliten vor und nach den
Markomannenkriegen’. Slovenská Archaeológia 65, 321–42.
Voß, H.-U. und Wigg-Wolf, D. 2017: ‘Romans and Roman Finds in the Central
European Barbaricum: A New View on Romano-Germanic Relations?’ In
González Sánchez, S. und Guglielmi, A. (Hrsg.), Romans and Barbarians beyond
the Frontiers: Archaeology, Ideology and Identities in the North (Themes in
Roman Archaeology 1) (Oxford/Philadelphia), 105–24.
Ward-Perkins, B. 2005: The Fall of Rome and the End of Civilization (Oxford).
Wheeler, E.L. 1993a: ‘Methodological Limits and the Mirage of Roman Strategy:
part 1’. Journal of Military History 57.1, 7–41.
—. 1993b: ‘Methodological Limits and the Mirage of Roman Strategy: part 2’. Journal
of Military History 57.2, 215–40.
—. 2007: ‘The Army and the Limes in the East’. In Erdkamp, P. (Hrsg.), A Companion
to the Roman Army (Malden, MA/Oxford), 235–66.
Whittaker, C.R. 1994: Frontiers of the Roman Empire: A Social and Economic Study
(Baltimore).
Winter, O.F. 1952: ‘Klientelkönige im römischen und byzantinischen Reich’. Jahrbuch
der Österreichischen Byzantinischen Gesellschaft 2, 35–50.
Wolters, R. 1990: ‘Zum Waren- und Dienstleistungsaustausch zwischen dem römi-
schen Reich und dem freien Germanien in der Zeit des Prinzipats. Eine Bestands-
aufnahme, Teil 1’. Münstersche Beiträge zur Antiken Handelsgeschichte 9, 14–44.
—. 1991: ‘Zum Waren- und Dienstleistungsaustausch zwischen dem römischen Reich
und dem freien Germanien in der Zeit des Prinzipats. Eine Bestandsaufnahme,
Teil 2’. Münstersche Beiträge zur Antiken Handelsgeschichte 10, 78–132.
Wolters, R., Erdrich, M. und Voß, H.-U. 2003: ‘Römischer Import’. Reallexikon der
Germanischen Altertumskunde 25, 138–58.
 CHAPTER 13

TERMINI HUIUS IMPERII: THE FRONTIERS

OF THE EMPIRE IN THE PANEGYRICI LATINI
(3RD–4TH CENTURIES AD) –
BETWEEN PROPAGANDA AND REALITY

Nelu ZUGRAVU

Abstract
The references to the frontiers of the Roman empire in the Panegyrici Latini provide
important elements of imperial ideology and propaganda and, at the same time, neces-
sary data for historical reconstruction; this confirms the observation that they are set at
the interference between history and rhetoric, between the ‘informative and the con-
ventional registers of stereotypy’.

Introduction

For a long time deemed irrelevant for purposes of historical reconstruction on

account of their encomiastic style and intentional altering of facts,1 the dis-
courses of the Gallo-Roman orators of the 3rd/4th centuries AD have been
gradually re-evaluated, especially following Édouard Galletier’s publication in
the middle of the last century of the first edition of the Panegyrici Latini cor-
pus.2 The growing fascination with these writings has led, on the one hand, to

1
For those aspects that were accepted, see Malosse 1998–99; Ware 2017.
2
Galletier 1949; 1952; 1955. The corpus contains 12 orations, edited and collected by Lati-
nus Pacatus Drepanius, himself the author, in AD 389, of a discourse ‒ the last of the collection
‒ dedicated to Theodosius I; the florilegium was presented to Theodosius after the victory over
the usurper Maximus during ceremonies that took place in Rome in August 389 (Nixon et al.
1994, 437‒41; Turcan-Verkerk 2003). The first of these is the gratiarum actio delivered by
Pliny the Younger in honour of Trajan on the occasion of his appointment as consul suffectus in
the year 100 (de Trizio 2009, 28; Roche 2011; Balbo 2017a); this laudatio was the model fol-
lowed by the Gallo-Roman panegyrists (see below). The other 11 discourses, delivered in for-
mal and solemn circumstances by Gallic orators (of which some are anonymous) in both their
own cities, such as Augustodunum (Autun), Augusta Treverorum (Treveri), and at Rome or
Constantinople, at various points in time ranging from AD 289 to 389, are disposed in a non-
chronological order (that is why their numbering system differs from one editor to another);
220 N. ZUGRAVU

the production, in various historiographic traditions, of bilingual critical edi-

tions and annotated translations,3 and, on the other hand, to the ‘derhetorised’
valorisation of the information contained in various texts dealing with the cri-
sis of the empire,4 the Diocletian-Maximian diarchy,5 the first tetrarchy,6 the
Constantinian dynasty7 and the reign of Theodosius.8 It has been noted that,
notwithstanding the exaggerations, conventions and stereotypes characteristic
of the literary genre they belong to9 and the Plinian model they claim to
emulate,10 and despite their propagandistic nature and their role in entertaining
the ‘political liturgy’,11 they are important and credible pieces of evidence for

according to the Lassandro and Micunco 2000 edition, these are: 2 [10]: Claudius Mamertinus’
discourse delivered on 21 April 289 at Augusta Treveroroum in honour of Diocletianus Iovius
and Maximianus Herculius; 3 [11]: genethliacus Maximiani Augusti delivered by the same orator
on 21 July (?) 291 also at Augusta Treverorum; 4 [8]: a discourse written by an anonymous
author from Augustodunum, but probably delivered at Augusta Treverorum on 1 March 297 on
the occasion of the celebration of the Quinquennalia by Constantius Caesar; 5 [9]: oratio pro
instaurandis scholis of Eumenius delivered in the spring of 298 at Augustodunum in front of the
governor (vir perfectissimus) of the province of Gallia Lugdunensis; 6 [7]: panegyricus dictus by
an anonymous author on 31 March 307, probably at Augusta Treverorum, on the occasion of the
marriage between Constantine and Fausta, daughter of Maximian; 7 [6]: a discourse by an anon-
ymous orator from Augustodunum delivered at the end of July 310 at Augusta Treverorum in
honour of Constantinus Imperator; 8 [5]: gratiarum actio delivered in 312 at Augusta Trevero-
rum by an anonymous orator from Flavia Aedorum on the occasion of the celebration of the
Quinquennalia by Constantine; 9 [13]: a discourse by an anonymous orator from Augustodunum
delivered in 313 at Augusta Treverorum in the presence of Constantine, in which he evokes pri-
marily the campaign against Maxentius; 10 [4]: panegyric of Nazarius of Burdigala delivered on
1 March 321 at Rome in front of Crispus Caesar and Constantine (II) Caesar; 11 [3]: gratiarum
actio delivered to Julian Augustus on 1 January 362 at Constantinople by Mamertinus, formerly
comes sacrorum largitionum, praefectus praetorio Illyrici, on the occasion of obtaining the con-
sulship; 12 [2]: discoursed delivered by Latinus Pacatus Drepanius of Burdigala in 389 in front
of the Roman senate and the emperor Theodosius, in which he praised the emperor’s civil virtues
and military qualities, dedicating an ample space to the usurper Maximus’ uprising in Gallia,
defeat in Pannonia and death at Aquileia.
3
Morgan and Lieu 1989; Müller-Rettig 1990; Lassandro 1992; Lassandro and Micunco
2000; Müller-Rettig 2008; Nixon et al. 1994; García Ruiz 2006; de Trizio 2009; Laudani 2014;
Bucci 2015. See also Lassandro 2000b, 41‒43.
4
Nixon et al. 1994, 60, n. 19, 88‒89, n. 30, 122‒24, nn. 30‒34, 221, n. 7, 270‒71, nn. 21‒23,
307, n. 54; Hartmann 2008, 39‒40.
5
Nixon et al. 1994, 76‒80, 241, n. 73; Roberto 2014; Casella 2017.
6
Nixon et al. 1994, 104‒08; Lassandro 2000b, 49‒50.
7
Müller-Rettig 1990; Grünewald 1990; Nixon et al. 1994, 178‒90, 211‒17, 254‒63, 289‒93,
338‒42, 386‒92; Lassandro 2000b, 50‒52; Ronning 2007, 189‒380; Warmington 2012;
Blockley 2012; Donciu 2012, 15‒18; Barbero 2016, 25‒67.
8
Nixon et al. 1994, 43‒51, 441‒47; Lassandro 2000b, 52; Lippold 2012.
9
Pernot 1993; Nixon et al. 1994, 21‒26; de Trizio 2006; Bucci 2012; Abbatepaolo 2012.
10
Nixon et al. 1994, 18; Malosse 1998‒99; Ronning 2007, 24‒151; García Ruiz 2013;
Laudani 2014, 42‒44.
11
MacCormack 1981, 1‒14; 2012; Sabbah 1984; Rodríguez Gervás 1991; L’Huiller 1992;
Nixon et al. 1994, 26‒33; Orihuela Sancho 1996; Whitby 1998; Enenkel 2000; Ronning 2007;
Borgognoni 2010; Lopetegui 2015, 72‒79.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 221

discerning various political-military, social and religious features particular to

the late 3rd- and 4th-century empire. By corroboration with data offered by
epigraphy, coins, artistic depictions and archaeological finds, the Panegyrici
Latini have contributed to a more convincing reconstruction and a better
understanding of numerous events (official ceremonies,12 military cam-
paigns13), of ideological and religious aspects (the theology of the diarchy and
the tetrarchy,14 Constantine’s ‘slide’ toward monotheism,15 the imperial
virtues,16 tyrannus,17 heredity as a means of legitimising power,18 etc.19), of
social and institutional circumstances (the movement of the Bagaudae [rebel-
lio Bagaudica],20 urban life,21 taxation [obsequia, census, tributi, stipendii
provinciarum],22 education,23 legal issues,24 the strengthening of frontiers25), of
the relations with the barbarians and the Persian state,26 of the mechanisms
of the empire’s evolution,27 and so on.28 Of all these, I would like to single out
and focus on the frontiers of the Roman state during the age in question, which
are discussed only in a few, if important, studies.29 Hence, in the present paper
I aim to offer: an inventory of the terms used by the Gallo-Roman panegyrists
to designate the frontier; an analysis of the borders and of the imperial space

12
Nixon et al. 1994, 51‒52, 96, n. 65, 114, n. 14, 265, n. 6.
13
Nixon et al. 1994, 26, 43‒44, 61‒62, n. 23, 64‒66, nn. 29‒31, 71, n. 39, 79, 86, n. 25, 89,
nn. 33‒38, 92, nn. 48‒49, 106‒08, 110‒11, n. 6, 113, n. 12, 114‒17, nn. 14‒20, 118‒19,
nn. 22‒24, 120‒22, nn. 26‒29, 124‒26, n. 35, 127‒31, nn. 39‒47, 136, nn. 58‒59, 138, n. 63,
172‒77, nn. 79‒85, 224‒27, nn. 21, 24‒28, 232, n. 44, 233, n. 45, 235, n. 54, 326‒27, nn. 138‒139,
328‒29, nn. 143‒147, 332, n. 158, 346, nn. 14, 17, 362, nn. 76‒77, 363, n. 79, 396, n. 18, 398,
n. 24, 399, n. 27, 401‒02, n. 38; Lassandro 2000b, 28‒32; Lee 2007; Ronning 2007, 371‒72;
Sancho Gómez 2011, 47‒49; Whately 2013a, 130.
14
Béranger 1970; Rodgers 1986; 2012; Nixon et al. 1994, 34, 79‒80; Lassandro 2000a;
2006; de Trizio 2007; Giuliese 2007; La Bua 2009; Rocco 2017.
15
Müller-Rettig 1990; Nixon et al. 1994, 211‒17, 248‒51, nn. 91‒93, 291‒93, 296, n. 11,
341‒42, 344, n. 7, 357‒58, n. 61, 359, n. 63; Cañizar Palacios 2009; Long 2009; Lassandro
2015; Tommasi Moreschini 2016; Hostein 2016; Rocco 2017.
16
Seager 1984; Lassandro 2000a; Giuliese 2005; García Ruiz 2008a.
17
L’Huiller 1992, 263‒75; Lassandro 2000b, 16‒19; Mouchová 2002; Bucci 2010.
18
Müller-Rettig 1990, 52‒60, 66, 79‒84; Marotta 2010.
19
Lassandro 2000b, 24‒28; Guzmán Armario 2010.
20
Nixon et al. 1994, 60‒61, n. 21, 154‒55, n. 12, 270‒71, n. 21; Lassandro 2000a, 105‒44;
2000b, 19‒24; 2013; de Trizio 2009, 80‒81; Bianchi 2016; Bucci 2015, 105‒08.
21
Béranger 1970; Pérez Sánchez 2003; Hostein 2012; Kasprzyk 2016.
22
Nixon et al. 1994, 257‒63, 272‒77, nn. 27‒37, 280‒84, nn. 46‒49, 53, 408, n. 60.
23
Nixon et al. 1994, 145‒50, 254; Roberto 2017.
24
Cañizar Palacios 2013.
25
Sancho Gómez 2011, 47‒49; Sarantis and Christie 2013b.
26
Lassandro 2000b, 28‒36.
27
Chauvot 2016, 365‒76.
28
Rousselle 1976; Nixon et al. 1994, 26‒35; Lassandro 2000b, 15‒38, 43‒58; Rees 2002;
García Ruiz 2008b; Ryan 2009; Balbo 2017b, 32‒35.
29
Lassandro 1987; 1993; 1998; 2000b, 28‒29, 32‒34; 2001; Piacente 2004, 175‒76; Graham
2006, 43‒45, 65; Chauvot 2016, 59‒87.
222 N. ZUGRAVU

delimited by them; finally, a review of the meanings that the orators assigned
to the frontier. I consider such an exploration to be an interesting and useful
one, supporting both the general interest in the investigation of the imperial
frontiers in late antiquity30 and the specific concern with the study of the limes
as reflected in contemporary texts.31 I should add here that I chose to exclude
from the group of the 11 Gallo-Roman orations of the 3rd‒4th centuries the
one delivered in the summer of 389 by Latinus Pacatus Drepanius due to it
providing evidence of a political, territorial and ethnic reality much different
from that witnessed by the earlier texts, in that it was affected by the official
settlement of the Goths within the empire.

Frontier Terminology and its Meaning

From the point of view of frequency of occurrence, the terms used for the
frontier in the Panegyrici Latini are limes, terminus, ripa and finis; we do not
encounter confinium, modus, meta.
1) Limes is attested most frequently: 22 times. In chronological order, we
encounter four in the text of 21 April 289,32 three in the genethliacus of
21 April(?) 291,33 four in the panegyricus dictus of 1 March 297,34 one in
Eumenius’ oratio pro instaurandis scholis delivered in the spring of 298,35 one

30
Lee 1993; Mathisen and Sivan 1996; Pohl et al. 2001; Mazza 2005, 11‒114; Graham
2006; Cañizar Palacios 2017; and see below.
31
Schwartz 1970; Drijvers 2011.
32
Pan. Lat. 2 [10]. 2. 4: An quemadmodum educatus institutusque sis praedicabo in illo
limite, illa fortissimarum sede legionum…?; 2. 6: Ibo scilicet virtutis tuae vestigiis colligendis
per totum Histri limitem…; 7. 3: Atqui Rhenum antea videbatur ipsa sic Natura duxisse, ut eo
limite Romanae provinciae ab immanitate barbariae vindicarentur; 9. 1: Ingressus est nuper
illam quae Raetiae est obiecta Germaniam similique virtute Romanum limitem victoria
protulit…
33
Pan. Lat. 3 [11]. 3. 9: Non enim in otiosa aliqua deliciisque corrupta parte terrarum nati
institutique estis, sed in his provinciis quas ad infatigabilem consuetudinem laboris atque patien-
tiae fracto licet oppositus hosti, armis tamen semper instructus limes exercet, in quibus omnis
vita militia est, quarum etiam feminae ceterarum gentium viris fortiores sunt; 5. 4: transeo
limitem Raetiae repentina hostium clade promotum; 13. 4: vos tantae rei publicae adminis-
tratione suscepta, quos huc atque illuc tot urbes tot castra tot limites tot circumiecta Romano
imperio flumina montes litora vocant, tantum animis ac fortuna valetis ut in unum convenire
possitis, nihilominus orbe securo.
34
Pan. Lat. 4 [8]. 1. 4: <tot> prolati limites; 3. 3: porrectis usque ad Danubii caput Germa-
niae Raetiaeque limitibus; 13. 3: Tu enim ipse, tu domine Maximiane, imperator aeterne, novo
itineris compendio adventum divinitatis tuae accelerare dignatus repente Rheno institisti,
omnemque illum limitem non equestribus neque pedestribus copiis sed praesentiae tuae terrore
tutatus es…; 18. 2: Nullo siquidem certo fine montium aut fluminum terminabatur quem disposi-
tae limitis custodiae tuerentur…
35
Pan. Lat. 5 [9]. 18. 4: Nam quid ego alarum et cohortium castra percenseam toto Rheni et
Histri et Euphrates limite restituta?
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 223

in the discourse of 31 March 307,36 four in the one delivered at the end of July
31037 and five in the one delivered in 313.38 Most of these occurrences (15)
carry a wide meaning, describing the imperial frontier in general, however
there is explicit evidence that the term limes was also used in the context of
referring to more specific frontiers, such as of the border provinces (limes
Raetiae; Germaniae Raetiaeque limites; Gallia; the Danubian provinces of
Dalmatia and Pannonia).39 Six other occurrences are associated with a natural
barrier that delineates the area inhabited by the Romans, specifically with the
rivers Rhine, Danube, Tigris and Euphrates (Histri limes; Rheni et Histri et
Euphrates limites).40 In one case only, the term is associated with a river
(Padus) that demarcates not the external frontiers of the state, but an ‘internal’
border – Padi limes.41
One may notice that the term limes is encountered particularly in orations
belonging to the first part of the corpus, whose content revolves mostly, as
suggested some time ago by Marie-Claude L’Huillier, around the exaltation of
the political-military function of the emperor during the campaigns against his
and the state’s true adversary – the barbarian from across the border; it is not
by accident that, in the proximity of the term limes, we can repeatedly find the
words barbarus and natio.42
In Latin usage, limes was meant to refer to the fines extremi imperii Roma-
ni.43 In modern research, it is considered to be ‘un terme impropre pour un
concept ambigu’,44 a fact that led to a historiographic debate over its meaning.
As is well known, many a scholar associate the term with a frontier of the

36
Pan. Lat. 6 [7]. 14. 1: te, iuvenis, indefessum ire per limites qua Romanum barbaris gen-
tibus instat imperium…
37
Pan. Lat. 7 [6]. 7. 1: Cuius etiam suprema illa expeditio non Britannica tropaea, ut vulgo
creditum est, expetivit, sed dis iam vocantibus ad intimum terrarum limen accessit; 11. 5: Con-
tra hinc per intervalla disposita magis ornant limitem castella quam protegunt; 13. 1: cum
tamen hoc tu magis ad gloriam imperii tui et ornatum limitis facias quam ad facultatem, quotiens
velis, in hosticum transeundi…; 21. 2: Ecce enim, dum a limite paulisper abscesseras, quibus se
terroribus barbarorum perfidia iactaverat…
38
Pan. Lat. 9 [12]. 2. 6: Rhenum tu quidem tot limite dispositis exercitibus tutum reliqueras;
14. 6: Quotiens milites in contionem vocabat, se solum cum illis imperare, alios per limites pro
se militare iactabat; 15. 1: ad resistendum Padi limite…; 21. 5: sed eodem impetu quo redieras
in Gallias tuas perrexisti ad inferiorem Germaniae limitem…; 22. 5: sed inopinato consilio usus
abeundi, simulato enim nuntio maioris in superiore limite tumultus…
39
Pan. Lat. 2 [10]. 2. 4; 9. 1; 3 [11]. 3. 9; 5. 4; 13. 4; 4 [8]. 1. 4; 3. 3; 18. 2; 6 [7]. 14. 1;
7 [6]. 7. 1; 11. 5; 21. 2; 9 [12]. 14. 6: ad inferiorem Germaniae limitem; 21. 5; 22. 5.
40
Pan. Lat. 2 [10]. 2. 6; 7. 3; 4 [8]. 13. 3; 5 [9]. 18. 4; 7 [6]. 13. 1; 9 [12]. 2. 6.
41
Pan. Lat. 9 [12]. 15. 1.
42
L’Huiller 1992, 216, 222‒23.
43
TLL s.v. limes, 1415, 19–20; Forni 1987; Lassandro 1987, 296; de Trizio 2009, 68, 96;
Bucci 2015, 141‒42.
44
Carrié 1995, 31‒41.
224 N. ZUGRAVU

military type that delineates the Roman territory from the one beyond, whereas
others scholars maintain that limes was never used in reference to a militarised
structure or a defensive system against the barbarians, but to a frontier district
that was in late antiquity under the command of a dux limitis;45 thus, it would
have had mainly ‘una funzione organizzativa più che diffensiva’.46 Results
from recent research based mainly on ‘military archaeology’,47 while not con-
tradicting entirely the latter view, add nuance by employing sources from both
the time of the high empire (for example Tacitus)48 and also of the late period
(for example: Libanius,49 Themistius,50 Ammianus Marcellinus,51 Historia
Augusta,52 Codex Theodosianus,53 Zosimus54), the Panegyrici Latini included.
The Gallo-Roman orators often state that there are troops stationed on the
limes – legiones, alae, cohortes, etc.55 – that it is strengthened with fortifica-
tions (castella, castra) or that it owns other various constructions built for
military purposes, such as bridges.56 When speaking of the manner of putting
together such an assembly, some of them use the verb munio,57 which is known
to be part of the technical military vocabulary, to refer to the process of forti-
fying by means of building walls, ditches, etc.58 Moreover, we encounter the
principle instated around mid-4th century by a military technical expert –

45
Isaac 1988; 1990; Mayerson 1989; Whittaker 1994; 2004, 28‒49; Elton 1996, 126‒28;
Martin 1999; Whately 2013b, 246–54.
46
Traina 2004, 207.
47
Janniard 2006, 134.
48
Tacitus Agr. 41. 1: nec iam de limite imperii et ripa, sed de hibernis legionum et possessione
dubitatum; Germ. 29. 3: mox limite acto promotisque praesidiis; Ann. 1. 50; Graham 2006, 62.
49
Libanius Or. 59. 44; 102.
50
Themistius Or. 10. 135D–139A, especially 138B; Or. 16; Graham 2006, 103.
51
Drijvers 2011; Bocci 2011‒12; Elton 2013, 670, 671; Kalafikis 2014.
52
SHA Hadr. 12. 6, etc.; Schwartz, 1970.
53
CTh 15. 1. 13 (364); Elton 2013, 669.
54
Zosimus 2. 34. 1‒2; Elton 2013, 669.
55
Pan. Lat. 2 [10]. 2. 4: illa fortissimorum sede legionum; 4 [8]. 13. 3: … equestribus…
pedestribus copiis; 18. 2; 5 [9]. 18. 4: Nam quid ego alarum et cohortium castra percenseam
toto Rheni et Histri et Euphrates limite restituta?; 9 [12]. 2. 6: Rhenum tu quidem tot limite
dispositis exercitibus tutum reliqueras; 3. 2; 21. 3. For the frontier troops at this time, cf. Dietz
1993; Scharf 2005; Elton 2013.
56
Pan. Lat. 2 [10]. 2. 5; 3 [11]. 13. 4; 4 [8]. 18. 2; 5 [9]. 18. 4; 7 [6]. 11. 1; 11. 5; 13. 1‒5;
14. 1; 11 [3]. 4. 6; Lassandro 1987, 296; Nixon et al. 1994, 111, n. 6, 234, n. 51, 235‒36, nn.
56‒57; Elton 2013, 670‒71. For an analysis of the relatively recent bibliography on the fortifica-
tions and the entire military infrastructure at the various frontiers of the later Roman empire, see
Scharf 2005; Graham 2006, 62‒63, 103‒07; Băjenaru 2010; Sarantis and Christie 2013b; Saran-
tis 2013a; 2013b; Elton 2013, 655‒57, 669‒73; Howard-Johnston 2013; Collins et al. 2015.
57
Pan. Lat. 2 [10]. 7. 4: Ecquis umquam ante vos principes non gratulatus est Gallias illo
amne muniri?; see also 5 [9]. 19. 2: Quippe, ut initio dixi, nulli umquam antehac principes pari
cura belli munia et huiusmodi pacis ornamenta coluerunt.
58
de Trizio 2009, 97‒99; Bucci 2015, 110.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 225

the anonymous author of De rebus bellicis: it must be created at the frontiers,

he says, an uninterrupted line of castella positioned every 1000 feet and con-
nected by a wall with durable observation towers.59 The anonymous panegyric
of 310 in honour of Constantine evokes such a picture: contra hinc per inter-
valla disposita magis ornant limitem castella quam protegunt.60 Thus, the bor-
der provinces are militarised provinces: which was the patria of Maximian
and that of Diocletian,61 where was the land of their birth (stirps),62 where had
each of them been educated and trained (educatus institutusque)?63 In a fron-
tier province (in illo limite), answers Claudius Mamertinus in 289 about
Maximian,64 namely in Pannonia, known for the virtus of its inhabitants65 and
where the bravest legions are stationed (illa fortissimorum sede legionum);66
not in a land of leisure and luxury (non enim in otiosa aliqua deliciisque cor-
rupta parte terrarum nati institutique estis), he also says in the discourse of
291 in reference to both sovereigns, but in those provinces (in his provinciis)
which, on account of bordering the enemy ‒ even a defeated one (fracto… op-
positus hosti) are accustomed to being constantly in arms (armis tamen semper
instructus exercet), to withstand fatigue and hardship (patientia); here, the
entire life is military service (in quibus omnis vita militia est), and women
even are braver than the men of other nations (quarum etiam feminae cetera-
rum gentium viris fortiores sunt).67 Only in this sense limes may designate
‘une division administrative’,68 a ‘ressort géographique del territorio control-
lato dal dux’.69 Lastly, the panegyrics clearly support, as noted above, the
association of limes with a natural barrier, specifically with great rivers, which,
unsurprisingly, provide protection for the empire.70 In this sense, Mamertinus
was saying of the Rhine in 289 that Nature itself had laid it out (ipsa sic
Natura duxisse) in order to delimit (ut eo limite) the Roman provinces (Romanae
provinciae) from the barbarian ferocity (immanitas barbariae),71 and of the

59
De rebus bellicis 20. 1: Est praeterea inter commoda rei publicae utilis limitum cura
ambientium ubique latus imperii; quorum tutelae assidua melius castela prospicient, ita ut
millenis interiecta passibus stabili muro et firmissimis turribus erigantur.
60
Pan. Lat. 7 [6]. 11. 5.
61
Pan. Lat. 2 [10]. 2. 2; 3 [11]. 4. 1.
62
Pan. Lat. 3 [11]. 4. 1.
63
Pan. Lat. 2 [10]. 2. 4; 3 [11]. 4. 1: institutorum.
64
Pan. Lat. 2 [10]. 2. 4; Nixon et al. 1994, 56, n. 11; de Trizio 2009, 68.
65
Pan. Lat. 2 [10]. 2. 2; de Trizio 2009, 66‒67.
66
Pan. Lat. 2 [10]. 2. 4.
67
Pan. Lat. 3 [11]. 3. 9.
68
Glad 2017, 133.
69
Traina 2004, 215.
70
For the natural frontiers of the time, see Graham 2006, 51‒75.
71
Pan. Lat. 2 [10]. 7. 3 (de Trizio 2009, 95‒97); 10 [4]. 18. 1 (Laudani 2014, 234‒35).
226 N. ZUGRAVU

Euphrates, that it protected in an embrace the rich and fertile Syria (opimam
illam fertilemque Syriam velut amplexu suo tegebat Euphrates).72 There were
guard posts on the banks of the rivers,73 military boats patrolling on the water,74
and structures built to facilitate the crossing into barbarian territory.75 This
reality, of the militarised limes, which had to be known to them also in view
of the places where the orations were delivered, offered the panegyrists,
I believe, the justification they needed to affirm that, alongside the sovereign’s
military prowess (virtus),76 it was the limes that held under control the barbar-
ian nations (per limites qua Romanum barbaris gentibus instat imperium),77
thus enabling the inhabitants of the frontier provinces (cum summo metu nos-
tro)78 to experience security and peace instead of fear (cum securitate nostra;79
soluto animo ac libero sumus;80 pax81).
2) Terminus is encountered only five times: twice in the panegyric of 289,82
once in that of 29183 and twice in the one of 297.84 The notion has divergent
meanings. Thus, on the one hand, it connotes more clear-cut nuances than the
term limes, in that terminus expresses very clearly the idea of the territorialisa-
tion of the empire (as in the high period),85 by referring to a sharp delimitation

72
Pan. Lat. 2 [10]. 7. 5; de Trizio 2009, 99.
73
Pan. Lat. 4 [8]. 18. 2; 9 [12]. 2. 6; 3. 1‒2. For the high empire, see Flor. 2. 30: Drusus…
in Rheni quidem ripa quinquaginta amplius castella direxit.
74
Pan. Lat. 7 [6]. 13. 1: quippe cum totus armatis navibus Rhenus instructus sit; 9 [12]. 3.
2: Et quid opus erat ipsi Rheno instructis et militibus et classibus…; 22. 6: tot Rheni alveo
oppleto navibus. For the fluvial naval bases of the time, see Elton 2013, 673‒74.
75
Pan. Lat. 4 [8]. 2. 1: a ponte Rheni usque ad Danubii transitum Guntiensem; 7 [6]. 13. 1:
Agrippinensi ponte.
76
Pan. Lat. 2 [10]. 2. 6; 6. 3; 7. 6; 9. 1‒3; 3 [11]. 2. 4; 7. 5; 4 [8]. 1. 5; 9. 5; 10. 4; 5 [9].
19. 1; 19. 4; 6 [6]. 3. 3; 5. 3; 10. 4; 11. 1‒2; 12. 4; 9 [12]. 3. 1; 10. 3, etc.; L’Huiller 1992,
223‒35.
77
Pan. Lat. 6 [7]. 14. 1; see also 7 [6]. 13. 1: Insuper etiam Agrippinensi ponte faciundo reli-
quiis adflictae gentis insultas…
78
Pan. Lat. 2 [10]. 7. 4; de Trizio 2009, 98.
79
Pan. Lat. 2 [10]. 7. 4; de Trizio 2009, 98.
80
Pan. Lat. 2 [10]. 7. 6.
81
Pan. Lat. 2 [10]. 6. 3; 14. 4: profundissima pax; 4 [8]. 20. 2‒3; 7 [6]. 11. 1; 9 [12]. 2. 6;
3. 2.
82
Pan. Lat. 2 [10]. 7. 2: Quod autem maius evenire potuit illa tua in Germaniam transgres-
sione, qua tu primus omnium, imperator, probasti Romani imperii nullum esse terminum nisi qui
tuorum esset armorum? 10. 1: Vos vero, qui imperium non terrae sed caeli regionibus
terminatis…
83
Pan. Lat. 3 [11] 16. 2: Sancte Iuppiter et Hercules bone, tandem bella civilia ad gentes illa
vesania dignas transtulistis, omnemque illam rabiem extra terminos huius imperii in terras hos-
tium distulistis.
84
Pan. Lat. 4 [8]. 3. 3: qui Romanae potentiae terminos virtute protulerant, imperium filio
pietate debebant; 20. 4: Quae a vobis ita recuperata est ut illae quoque nationes terminis eius-
dem insulae cohaerentes vestris nutibus obsequantur.
85
Tacitus Ann. 1, 11: intra terminos imperii.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 227

of the imperial area (hic Imperium) from the enemy lands (terrae hostium), as
can be inferred unambiguously from the genethliacus Maximiani Augusti
delivered on 21 July(?) 291 by Mamertinus86 and from the anonymous pane-
gyric honouring Constantius Caesar in 297.87 On the other hand, the word is
employed to better underline the imperial expansionist policy: the empire,
says the above-mentioned anonymous author, has no borders except for those
outlined by the armies (probasti Romani imperii nullum esse terminum nisi qui
tuorum esset armorum?),88 which help the emperors, in addition to the latter’s
own merit, to expand the limits of the Roman power (qui Romanae potentiae
terminos virtute protulerant).89 According to Mamertinus who was similarly
perorating in 291, echoing a Ciceronian phrase, for the empire administered by
Diocletian and Maximian earth is no boundary, only heaven (non terrae sed
caeli regionibus terminatis).90
3) Finis is attested four times: twice in the 297 discourse in honour of
Constantius,91 once in the panegyric delivered in 310 for Constantine,92 once
in the gratiarum actio delivered in 362 by Mamertinus.93 The word had vari-
ous meanings. Thus, in the rhetorical geography of the anonymous author who
was eulogising Constantius Caesar in 297, finis conveys the ancient connota-
tion of ‘boundary/end/limit of the world’, since it is used, with this meaning,
next to Oceanus (fines… Oceani).94 It is a well-established fact that, ever since
the Homeric poems, the Ocean had been considered to be the limit of the
known world, the place where real and imaginary geography cross paths;95 we
can encounter the term with the same meaning in writings of the 4th century
as well, such as Avienus’ Ora maritima96 or the anonymous Expositio totius
mundi et gentium.97 In the same vein, for the unspecified orator of Augusto-
dunum who, at the end of July 310, delivered a laudatio for Constantine, finis

86
Pan. Lat. 3 [11]. 16. 2.
87
Pan. Lat. 4 [8]. 20. 4: nationes terminis eiusdem insulae /scl. Britannia/.
88
Pan. Lat. 2 [10]. 7. 2.
89
Pan. Lat. 4 [8]. 3. 3.
90
Pan. Lat. 2 [10]. 10. 1. Cf. Cicero Cat. 3. 26: [Pompeius] fines vestri non terrae, sed caeli
regionibus terminaret; de Trizio 2009, 28, 107.
91
Pan. Lat. 4 [8]. 18. 2: Nullo siquidem certo fine montium aut fluminum terminabatur quem
dispositae limitis custodiae tuerentur; 20. 5: fines… Oceani.
92
Pan. Lat. 7 [6]. 9. 4: Di boni, quid hoc est quod semper ex aliquo supremo fine mundi nova
deum nomina universo orbi colenda descendunt? 9. 5: Sacratiora sunt profecto Mediterraneis
loca vicina caelo, et inde propius a dis mittitur imperator ubi terra finitur.
93
Pan. Lat. 11 [3]. 14. 1: Cum igitur inter egregia negotia itinere confecto usque ad Thra-
ciae fines perventum foret…
94
Pan. Lat. 4 [8]. 20. 5.
95
Garbarino 2005, 33.
96
Balbao Salgado 1995‒96.
97
Expositio 59: Inde Oceanum… est ibi finis mundi.
228 N. ZUGRAVU

designates the extreme limit of the world (supremus finis mundi), of all known
universe (universus orbis), bordered by the Nile, by India and by Britannia; it
is at these boundaries that the world ends (ubi terra finitur).98 On the second
hand, the finis is contrasted with the limes, as can be observed by reading the
oration delivered in honour of Constantius in 297: the first term designates the
natural border, while the second – the artificial one: nullo siquidem certi fine
montium aut fluminum terminabatur quem dispositae limitis custodiae tueren-
tur…99 Lastly, the finis designates the border of a Roman administrative unit
(Thraciae fines), as suggested in Mamertinus’ gratiarum actio of 362.100
4) Ripa has five occurrences – once in each of the panegyrics of 289,101
297102 and 362,103 and twice in that of 362104 – and terms the natural, fluvial
border of the empire, different from the artificial, political, military, terrestrial
one (limes);105 the distinction had already been applied by Tacitus.106 Ripa
separates naturally the area under Roman rule (noster) from the one under the
rule of other nations, which is foreign (barbarus), if we are to note the termi-
nology used in the anonymous discourse of 310.107 This role is played by ripa
Rheni.108 Beyond it (ultra Rhenum,109 trans Rhenum110), one is on barbarian
soil (in hosticum transeundi),111 from where, across the frozen waters of the
river (gelu Rhenus), hosts of Germanic peoples, including the Franks, pass into
Roman territory (transire Rhenum)112 – whereas on this side (cis Rhenum,113
contra hinc114), one is in Roman territory (illa ripa), where the farmer culti-
vates the bank without the need of taking up arms to defend himself and the
flocks can immerse themselves all along the two-horned river (arat illam…

98
Pan. Lat. 7 [6]. 9. 4.
99
Pan. Lat. 4 [8]. 18. 2.
100
Pan. Lat. 11 [3]. 14. 1.
101
Pan. Lat. 2 [10]. 2. 4: et ripas peragrabo Rheni…
102
Pan. Lat. 4 [8]. 13. 3: quantoslibet valebat exercitus Maximianus in ripa!
103
Pan. Lat. 11 [3]. 7. 2: cum dexteriorem incliti fluminis ripam utriusque sexus, omnium
ordinum, armatorum atque inermium perpetuus ordo praetexeret.
104
Pan. Lat. 7 [6]. 11. 5: Arat illam terribilem aliquando ripam inermus agricola, et toto
nostri greges bicorni <amne> mersantur; 13. 1: quippe cum totus armatis navibus Rhenus
instructus sit et ripis omnibus usque ad Oceanum dispositus miles immineat.
105
Carrié 1995, 36.
106
Tacitus Agr. 41. 1: nec iam de limite imperii et ripa, sed de hibernis legionum et posses-
sione dubitatum; Bucci 2015, 141‒42.
107
Pan. Lat. 7 [6]. 13. 2.
108
Pan. Lat. 2 [10]. 2. 6.
109
Pan. Lat. 2 [10]. 7. 7.
110
Pan. Lat. 6 [7]. 8. 4; Bucci 2015, 108.
111
Pan. Lat. 7 [6]. 13. 1.
112
Pan. Lat. 7 [6]. 6. 4; 11. 1‒3.
113
Pan. Lat. 6 [7]. 4. 2.
114
Pan. Lat. 7 [6]. 11. 5.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 229

ripam inermus agricola, et tot nostri greges bicorni <amne> mersantur);115

military patrol boats navigate the Rhine while intimidating soldiers guard its
banks right down to the Ocean (quippe cum totus armatis navibus Rhenus
instructus sit et ripis omnibus usque ad Oceanum dispositus miles immine-
at).116 Maximian’s presence here (Maximianus in ripa), the author of the pan-
egyric of 297 remarks, is more powerful than any army, inculcating fear in the
enemy.117 The same delimiting role is played by the Danube: in the gratiarum
actio de consulatu suo delivered by Mamertinus on 1 January 362 at Constan-
tinople, we can read that on the right bank of this illustrious river (cum dexte-
riorem incliti fluminis ripam) is imperial territory, while on the left bank (ad
laevam), on the other side of the great river (ex parte altera), is barbarian ter-
ritory (barbaricum solum).118 Finally, while the Panegyrici does not refer spe-
cifically to the ripa of the Euphrates and Tigris, mentions of these two rivers
allow the conclusion that their banks were delimiting the lands of the Romans
and the Parthians (Persians), respectively.119

The Empire’s Frontiers and Space in the Panegyrici Latini

Like many other Graeco-Latin authors from the Imperial age, the Gallo-Roman
orators had a double perception about the frontiers of the empire. The first has
a purely rhetorical, idealising, maximalist connotation – the Roman empire is
a world hegemon that owns both the political sphere and the physical domain.120
The state governed by Diocletian and Maximian knows no boundaries on
earth; the sky is its only limit, as Mamertinus proclaimed in 289 (non terrae
sed caeli regionibus terminatis).121 Each and every region on earth and in hea-
ven (ex omni terrarum caelique regione) belongs to the empire, according to
a peroration by an anonymous orator in 297.122 One year later, Eumenius
maintained as well that the Roman state displayed total dominance (potentia)

115
Pan. Lat. 7 [6]. 11. 5.
116
Pan. Lat. 7 [6]. 13. 1; see also 4 [8]. 14. 3‒4: equestribus… pedestribus copiis… arma-
tis… diversis classibus; 9 [12]. 3. 2: Et quid opus erat ipsi Rheno instructis et militibus et
classibus…
117
Pan. Lat. 4 [8]. 13. 3; Bucci 2015, 108‒09.
118
Pan. Lat. 11 [3]. 7. 1‒3; 8. 3‒4. For the high period, see Seneca QN 9: Danuvius Sar-
matica ac Romana disterminet. On Danube as limes and a Roman-barbarian point of interaction
in the later period, see Guzmán Armario 2012; Glad 2017.
119
Pan. Lat. 2 [10]. 2. 6: Ibo scilicet virtutis tuae vestigiis colligendis… perque omnem qua
tendit Euphraten…; 4 [8]. 3. 3: Partho quippe ultra Tigrim redacto.
120
Hidalgo de la Vega 2003; Plácido Suárez 2008.
121
Pan. Lat. 4 [8]. 3. 3.
122
Pan. Lat. 4 [8]. 20. 3.
230 N. ZUGRAVU

over land and sea (Romana res plurimum terra et mari valuit).123 This hegem-
ony is continuously growing, for, as one anonymous orator was claiming in
297, by defeating each and every enemy (omni hoste), the emperors, by virtue
of their bravery extended the frontiers of the Roman power (Romanae poten-
tiae terminos virtute protulerant), while the state (res publica), by becoming
more secure (certa securitas), has grown and is destined to grow further still
(aucta atque augenda res publica).124 We are dealing here, thus, with the belief
in the endless expansion of Rome, exalted by Virgil in the time of Augustus
(imperium sine fine)125 and by Aelius Aristides in the time of Antoninus
Pius.126 This maximalist geography, which can be encountered until late in the
Latin literature127 is, ultimately, a geography of a sovereign’s universal power
of sacred origin. According to our panegyrists, just as the principes are epipha-
nies of the tutelary divinities of the empire (conspicuus et praesens Iuppiter
cominus; sanctus Iuppiter; Hercules bonus; imperator Hercules; dei/numina
praesentes, etc.),128 having an immortalis/divina origo,129 a majesty akin (cog-
nata maiestas) to the respective deities,130 a will (voluntas) of the same author-
ity as that of the supreme father (summi patris sequatur auctoritas),131 and
their power delegated by Iuppiter,132 there exists a similitude between the
earthly universe and the celestial world (totius mundi caelestiumque rerum
similitudinem), as advocated by the anonymous orator who delivered the pan-
egyric for Constantius Caesar in 297.133 Hence, the empire is the terrestrial
projection of the celestial, and thus indivisible empire (imperium singularis,
imperium commune, patrimonium indivisum, res publica una), and as such the

123
Pan. Lat. 5 [9]. 19. 4.
124
Pan. Lat. 4 [8]. 3. 2‒3.
125
Virgil Aen. 1. 278: His ego nec metas nec tempora pono/ Imperium sine fine dedi; Hardie
1986.
126
Aelius Aristides Romae Encomium 97‒99; Desideri and Fontanella 2013.
127
Rutilius Namatianus De reditu suo 1. 55‒58; Claudian Cons. Stil. 2. 60; 3. 132‒133,
136‒137, 139, 159‒160: nec terminus unquam/ Romanae dicioni erit; III cons. Hon. 57; IV cons.
Hon. 42; Laud. Seren. 44; Prudentius Contra Symm. 1. 543; Fontaine 1983.
128
Pan. Lat. 2 [10]. 1. 1‒4; 2. 1: qui te praesentem intuemur deum (cf. Cicero Marcell. 10:
quem praesentem intuemur) (de Trizio 2009, 27, 64); 2. 3; 4. 2; 7. 5‒6; 10. 2; 13. 3‒4; 3 [11].
2. 3‒4; 3. 3; 3. 6‒7; 10. 5; 11. 1; 16. 2; 5 [9]. 8. 1; 6 [7]. 8. 2; 8. 4; 11. 3; 7 [6]. 2. 5; 11. 3;
22. 1; 9 [12]. 2. 4; 3. 1; 4. 4; 19. 6; 10 [4]. 7. 3, etc.; Nixon et al. 1994, 34, 44‒45, 49‒51,
79‒80; Rees 2002, 39; Lagioia 2004; de Trizio 2009, 55‒56, 63‒64, 67, 100; Bucci 2015,
104‒05, 125.
129
Pan. Lat. 2 [10]. 2. 3; 3 [11]. 3. 7; Nixon et al. 1994, 79‒80; de Trizio 2009, 67.
130
Pan. Lat. 4 [8]. 4. 1.
131
Pan. Lat. 5 [9]. 15. 3; Bucci 2015, 134‒35.
132
Pan. Lat. 6 [7]. 12. 6: Recipe, Iuppiter, quod commodasti; Bucci 2015, 133.
133
Pan. Lat. 4 [8]. 20. 3; L’Huiller 1992, 373‒74.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 231

imperial sovereignty is unrestricted, ubiquitous and unshared by associates ‒

an idea one can encounter in more than one oration.134
The second perspective about the imperial frontiers featured in the Panegy-
rici Latini is the rhetorical-realist one, wherein amplificatio rhetorica is inter-
woven with elements of political-administrative geography specific to the
times of their delivery. As notes by several scholars, from the 3rd century
onwards, the concept of imperium sine fine did not simply vanish from the
Roman political thought, but was replaced by a different view, based on
a more acute sense of reality,135 according to which the empire has a well-
defined territory, is territorialised, is separated from others by certain border
lines, qui barbarous Romanosque diuideret, as one reads in the Historia
Augusta.136 For the Gallo-Roman orators too, the Roman state is confined to
a space, however the description of this spatiality is not completely devoid of
rhetorical nuances, as can be expected in the case of this sort of text. Thus, the
imperial space is equivalent with the immense area delimited by the natural
barriers, i.e. by the great rivers – the Rhine to the north-west, Danube/Hister to
the north, the Euphrates and the Tigris to the east, the Ocean to the west and
the Nile to the south:137 Rhenus et Hister et Nilus et cum gemino Tigris
Euphrate et uterque qua <so>lem accipit ac reddit Oceanus et quidquid est
inter ista terrarum et fluminum et litorum, said Mamertinus in 291, referring
to the territory governed by virtue of the potentia exercised equally by the
diarchs Diocletian and Maximian (tam facili sunt aequanimitate communia).138
An ancient ideological and historiographical theme is echoed here: the rivers

134
Pan. Lat. 2 [10]. 9. 4‒5; 10. 1: tantum vim tantum potestatem mutuo vobis impartire; 11.
1‒3; 3 [11]. 6. 3; 6. 5; 6. 6‒7: Vobis Rhenus et Hister et Nilus et cum gemino Tigris Euphrate
et uterque qua <so>lem accipit ac reddit Oceanus et quidquid est inter ista terrarum et fluminum
et litorum, tam facili sunt aequanimitate communia quam sibi gaudent esse communem oculi
diem. Ita duplices vobis divinae potentiae fructus pietas vestra largitur: et suo uterque fruitur et
consortis imperio; 7. 1‒7; 13. 5: neque enim pars ulla terrarum maiestatis vestrae praesentia
caret, etiam cum ipsi abesse videamini; 14. 2‒3: /Iuppiter/ numen tamen eius ac mentem tot
infusam esse mundo… Divinitatem vestram ubique versari, omnes terras omniaque maria plena
esse vestri; 4 [8]. 15. 6: ubique vim vestrae divinitatis esse, ubi vultus vestri, ubi signa coler-
rentur; 20. 5; 6 [7]. 9. 2: summarum… rerum socium semper habuisses; 7 [6]. 1. 4: socia maies-
tas; L’Huiller 1992, 373‒74; Nixon et al. 1994, 58, n. 16, 59, n. 18, 141, n. 73, 215, 295, nn. 7;
Leadbetter 1998; de Trizio 2009, 73, 110‒11; Bucci 2015, 116.
135
Arce 2001.
136
SHA Hadr. 11. 2.
137
Pan. Lat. 2 [10]. 2. 6; see also 2 [10]. 7. 3‒5 (Rhine, Euphrates); 11. 7; 12. 2‒3; 12. 6
(Oceanus); 3 [11]. 6. 6 (Rhine, Ister, Nile, Tigris, Euphrates, Oceanus); 4 [8]. 2. 1 (Rhine, Danu-
bius); 3. 2 (Tigris); 6. 1; 6. 4 (Oceanus); 8. 1 (Rhine); 17. 4 (Oceanus); 18. 3 (Oceanus); 5 [9].
18. 1 (Rheni cornua); 18. 4 (Ister, Euphrates, Rhine); 20. 3 (Oceanus); 21. 1 (gemina Persidos
flumina, Nile); 6 [7]. 4. 2; 8. 3 (Rhine); 11 [3]. 7. 1‒3; 8. 4; 9. 1 (Ister, Danubius). The same
idea in Libanius Or. LIX; Malosse 2003, 40‒41.
138
Pan. Lat. 3 [11]. 6. 6.
232 N. ZUGRAVU

demarcate the ‘geography of imperial power’,139 as stated explicitly by the Res

gestae divi Augusti.140 When referring to larger geographical units, the empire
was said to contain the traditional areas of the oikoumene – Libya/Africa,
Europe and Asia, which were, according to Mamertinus in 362, under the rule
of Julian (Libyae Europae Asiaeque regnatur est).141 There is little doubt that
both views reflect perfectly what Pierre-Louis Malosse described as ‘le topos
encomiastique des extrémité de l’Empire et du monde’.142 This rhetorical
geography emphasises the idea that, from a territorial point of view, the empire
is a fluid reality, and there may always be circumstances favourable to the
expansion; this is what emerges from a discourse of 297 by an anonymous
orator: the Roman state (Romana res publica), he says, is a solidum imperium,
which holds in one embrace all that once was not Roman; there are, of course,
other territories that can be conquered if the will of the emperors and reasons
of state so require (et ex aliis quidem partibus aliqua restant, quae, si voluntas
vel ratio rem desiderent, possitis adquirere), but there is no longer any motive
to move forward unless the sovereigns would seek the limits of the Ocean (fi-
nes ipsius quaerantur Oceani), an act that nature has forbidden so far; all that
which is worthy of undefeated emperors, he concludes, has been already
attained (omnia, inquam, invictissimi principes, vestra sunt quae digna sunt
vobis).143 The same idea can be grasped from the oration of another anony-
mous, delivered in 310 in front of Constantine: having liberated Britannia, his
father, Constantius, had not considered further conquests to be a duty worthy
of pursuit – he had not set out for the woodlands and swamplands of the Cal-
edonians and Picts (nec Caledonum aliorumque Pictorum silvas et paludes),
for neighbouring Ireland (nec Hiberniam proximam), or for the very remote
Thule/Iceland? (nec Thylen ultimam), and not even for the legendary Isles of
the Blessed (nec ipsas, si quae sunt, Fortunatorum insulas).144
This world, this geographical and political reality is described ‒ of course,
according to the conventions of rhetorical geography, by using an amplificatio
rhetorica, a hyperbolising tone, thus associating geography and politics ‒ by
the orator Eumenius in a discourse of 298 in which he was thanking the sover-
eigns for the restoration of the so-called Maenian schools (scholae Maenianae)
of Augustodunum. He mentions the map displayed beneath their porticoes (in

139
Purcell 2012; Campbell 2012.
140
RG 26. 2. 4; Roman 1983; Vanotti 1987.
141
Pan. Lat. 11 [3]. 27. 2; Nixon et al. 1994, 430, n. 165.
142
Malosse 2006.
143
Pan. Lat. 4 [8]. 20. 2‒3; 2. 5.
144
Pan. Lat. 7 [6]. 7. 1‒2; Nixon et al. 1994, 227, n. 28; Lassandro and Micunco 2000, 228,
nn. 23‒24.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 233

illis porticibus), on which ‒ he said ‒ young people can see (videat) and con-
template every day (cotidie) all the lands (omnes terras), all the seas (cuncta
maria), the cities, the peoples, the nations (urbium gentium nationum) that the
undefeated principes have restored because of their piety, have defeated
because of their courage, or have made tremble in horror (quidquid invictissimi
principes… aut pietate restituunt aut virtute devincunt aut terrore
devinciunt).
Siquidem illic… omnium cum nominibus suis locorum situs spatia intervalla des-
cripta sunt, quidquid ubique fluminum oritur et conditur, quacumque se litorum
sinus flectunt, qua vel ambitu cingit orbem vel impetu irrumpit Oceanus. Ibi
fortissimorum imperatorum pulcherrimae res gestae per diversa regionum argu-
menta recolantur, dum calentibus semperque venientibus victoriarum nuntiis revi-
suntur gemina Persidos flumina et Libyae arva sitientia et convexa Rheni cornua
et Nili ora multifida; dumque sibi ad haec singula intuentium animus adfingit aut
sub tua, Diocletiane Auguste, clementia Aegyptum furore posito quiescentem aut
te, Maximiane invicte, perculsa Maurorum agmina fulminantem aut sub dextera
tua, domine Constanti, Bataviam Britanniamque squalidum caput silvis et flucti-
bus exserentem aut te, Maximiane Caesar, Persicos arcus pharetrasque calcantem.
Nunc enim, nunc demum iuvat orbem spectare depictum, cum in illo nihil vide-
mus alienum.145
On this map, which reminds us of that of the porticus Vipsania, on which,
as Pliny the Elder informs us, was depicted the orbis terrarum,146 we have,
therefore, the ‘inventory of the world’, to use the title of a book by Claude
Nicolet, whose pretext was the Augustan Res gestae and Vipsanius Agrippa’s
map ‒ so ‘a geographic repertoire and a catalogue of conquests and military
actions’.147 From this point of view, the panegyrists are part of the authentic
Roman ideological tradition that one may encounter, for example, in the title
of Augustus’ epigraphic memoirs ‒ Rerum gestarum divi Augusti, quibus
orbem terrarum imperio populi Romani subiecit; Bella terra et mari… externa
toto in orbe terrarum.148

The Meanings of the Frontier in the Panegyrici Latini

The discourses of the Gallo-Roman orators of the 3rd-4th centuries highlight

the multiple meanings of the frontier.

145
Pan. Lat. 5 [9]. 20. 1‒3; 21. 1‒3; L’Huiller 1992, 284; Nixon et al. 1994, 171‒77,
nn. 76‒86; Graham 2006, 51‒52.
146
Pliny NH 3. 17.
147
Nicolet 1988; 1990.
148
RG 3. 1. Cf. Pan. Lat. 3 [11]. 19. 4: itemque /scl. bona sidera et amica/ praeter victorias
toto orbe terrarum partas etiam navalia trophaea promittunt; 4 [8]. 17. 2‒3.
234 N. ZUGRAVU

The Legitimacy of Universal Power

The express indication of the boundaries of imperial space is meant to empha-

sise the extent and legitimacy of sovereign power. Its universal dimension is
best reflected in the mention by Mamertinus, in the genethliacus of 291, of the
two Oceans (uterque qua <so>lem accipit ac reddit Oceanus).149 According to
the myth of Heracles, the Ocean is at the end of the world, which the hero has
integrated with civilisation by cleaning it of monsters.150 The same was done
by Maximianus Herculius and Constantius Herculius. The former, after being
triumphant around the world (toto quidem orbem victorem), prepares himself
to fight at the River Ebro and the Western Ocean (a flumine Hibero et conscio
occidui solis Oceano), where Hercules did,151 and on the island of Britannia,
whose border (termini)152 lies beyond the Ocean (ultra Oceanum),153 at the
threshold of the earth (ad intimum terrarum limen).154 The one he is about to
face there is a true misfortune for the Occident (Occidentis plaga),155 a pira-
ta,156 the author of a nefarium, indignissimum latrocinium,157 i.e. of another
monster (prodigium),158 but much more horrifying than that misshapen three-
headed shepherd (non pastorem trino capite deformem sed prodigium multo
taetrius opprimentem) – the mythological Geryon; it is the usurper Carausius
(Augustus between autumn 286 and early summer 293), whom the orator how-
ever does not name.159 Thus, inserted into the divine world as a descendant of
Hercules (Herculea gens),160 Maximian, like Hercules paccator terrarum,161
like Hercules Victor, who had relieved the world from a range of perils,162 car-
ries the fight right down to the Ocean,163 so to the ends of the world, in pursuit

149
Pan. Lat. 3 [11]. 6. 6; L’Huiller 1992, 374.
150
Amiotti 1987, 13‒20; Díez del Corral Corredoira 2004.
151
Pan. Lat. 2 [10]. 2. 1; de Trizio 2009, 63; 2005.
152
Pan. Lat. 4 [8]. 20. 4.
153
Pan. Lat. 4 [8]. 20. 3.
154
Pan. Lat. 7 [6]. 7. 1; see also 6 [6]. 14. 3: dum vicinos ortus repetit occasu (Lassandro
2000a, 214, n. 38; Bucci 2015, 143).
155
Pan. Lat. 2 [10]. 2. 1.
156
Pan. Lat. 2 [10]. 12. 1; 4 [8]. 12. 1.
157
Pan. Lat. 4 [8]. 12. 1.
158
Pan. Lat. 2 [10]. 2. 1; de Trizio 2009, 65.
159
Pan. Lat. 2 [10]. 2. 1; Nixon et al. 1994, 55, n. 9, 72, n. 42; de Trizio 2009, 64‒66.
160
Pan. Lat. 2 [10]. 1. 3; 7. 6; 10. 2; 3 [11]. 3. 6; 5 [9]. 8. 1; 7 [6]. 2. 5; see also above.
161
Pan. Lat. 2 [10]. 11. 6; de Trizio 2009, 113.
162
Pan. Lat. 3 [11]. 3. 6‒7: terras omnes et nemora pacavit, urbes dominis crudelibus libe-
ravit, etiam caelo dirarum alitum volucra tela detraxit, etiam terrores inferum abducto custode
compescuit… multa faciant… similia Victoris. See also Lassandro 2006; Rees 2002, 42‒44; de
Trizio 2009, 60, 122; Iglesias García 2013, 126‒27, 132.
163
Pan. Lat. 2 [10]. 11. 7.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 235

of the restoration of legitimate authority. However, only Caesar Herculius –

that is to say, Constantius – whose avus is Hercules and whose pater is Her-
culius (Maximian),164 managed to free the island in the Ocean from evil,165 to
reintegrate it into the sphere of civilisation (ad humanos cultus), and restore it
to the beneficial light of Rome (ad conspectum Romanae lucis emersit);166 he
is, therefore, vindex et liberator.167 This kind of confrontation, which is cen-
tred on Hercules like characters fighting for legitimacy, is a recurrent motif in
ancient literature.168
On the other hand, as the anonymous author singing Constantine’s praise at
Augusta Treverorum at the end of July 310 stated, it was from Britannia, situ-
ated as it was at the end of the world (ad intimum terrarum limen),169 like
other lands at the ends of the earth giving the nova deum numina, that a new
divinity, destined to be venerated by the whole universe (universus orbis),
revealed itself.170 For just as Mercurius, coming from the Nile, and Liber,
arrived from India, had become manifest to men as gods,171 so the fortunata
Britannia gave the world, in accordance with the will of the celestial deities,
an emperor with a twofold divine origin – Constantine,172 descending from
divus Claudius173 and Constantius Pius, welcomed among the gods (receptus
est consenssu caelitum)174 on the shore of the Ocean – genitor deorum;175 after
going to Gallia, to Apollo ‒ Apollo tuus,176 says the orator – after recognising
himself in it and thus becoming a praesentissimus deus,177 he is destined to

164
Pan. Lat. 5 [9]. 8. 1.
165
Pan. Lat. 7 [6]. 5. 1; 5. 3‒4; 6. 1.
166
Pan. Lat. 5 [9]. 18. 2‒3; see also 4 [8]. 9. 5‒6; 6 [7]. 4. 3.
167
Pan. Lat. 4 [8]. 19. 1.
168
Moretti 1999; de Trizio 2009, 58; Bernard 2012.
169
Pan. Lat. 7 [6]. 7. 1.
170
Pan. Lat. 7 [6]. 9. 4: Di boni, quid hoc est quod semper ex aliquo supremo fine mundi
nova deum nomina universo orbi colenda descendunt?
171
Pan. Lat. 7 [6]. 9. 4: Sic Mercurius a Nilo, cuius fluminis origo nescitur, sic Liber a Indis
prope consciis solis orientis deos se gentibus ostendere praesentes; Nixon et al. 1994, 232,
n. 43; Lassandro and Micunco 2000, 232, nn. 34‒35.
172
Pan. Lat. 7 [6]. 4. 1; 7. 3; 9. 1; 9. 5: Sacratiora sunt profecto Mediterraneis loca vicina
caelo, et inde propius a dis mittitur imperator ubi terra finitur.
173
Pan. Lat. 7 [6]. 2. 1‒5; see also 8 [5]. 4. 2: divum Claudium parentem tuum.
174
Pan. Lat. 7 [6]. 7. 2; see also 6 [7]. 3. 3: divi… Constantii filium; 14. 3: dive Constanti,
quem curru [et] paene conspicuo… Sol ipse invecturus caelo excepit; 8 [5]. 4. 4: divus pater
tuus; 9 [12]. 25. 3: divus… Constantii; Bucci 2015, 78‒79, 142‒43.
175
Pan. Lat. 7 [6]. 7. 2; cf. Il. 14. 201 and 206; Lassandro and Micunco 2000, 229, n. 26;
Bucci 2015, 102.
176
Pan. Lat. 7 [6]. 21. 4; Müller-Rettig 1990; Nixon et al. 1994, 248‒50, nn. 91‒92.
177
Pan. Lat. 7 [6]. 22. 1; 17. 4: deus.
236 N. ZUGRAVU

have, as the vatum carmina divina had predicted for Apollo, the reign of the
whole world (totius mundi regna).178

The Political, Ideological and Cultural Divide

The frontiers delimit two opposing political, ideological and cultural spaces ‒
the civilised, pacific Roman order, and the conflict-generating barbarian socie-
ties. The panegyrists echo the political philosophy and ideology inherited from
past Graeco-Latin literature: the world is dichotomous, i.e. divided between
two entities that define themselves on the basis of divergent standards – civili-
sation and barbary. The former is composed of sedentary, agrarian human
communities, and is shaped in accordance with a technical (urban), moral and
juridical model; in the words of the orator Mamertinus, it is a space of amoena
loca and nobiles urbes.179 The latter is governed by nature, is wild, fluid,
decentred180 – the Latin literature speaks of a locus immensus, ignotus, infor-
mis, deformis, horridus, adversus, barbarus, ferox.181 As with other late
authors, for the Gallo-Roman orators too finis demarcates two opposing
spaces – Imperium and barbaria.182 Within the area bounded by borders there
is an orderly universe, a harmoniously enclosed space – orbis (orbis Romanus,
noster orbis).183 This is Romanum imperium184 – a human community, a civi-
lised world, the inhabited world (quidquid homines colunt),185 composed of
nationes186 governed by laws and norms (mores, leges, mandata, instituta,

178
Pan. Lat. 7 [6]. 21. 5: Vidisti teque in illius specie recognovisti, cui totius mundi regna
deberi vatum carmina divina cecinerunt; Müller-Rettig 1990; Nixon et al. 1994, 215, 250‒51,
n. 93.
179
Pan. Lat. 3 [11]. 4. 1.
180
Mund-Dopchie 2004.
181
Borca 2004.
182
Elton 1996, 126, n. 4; Guzmán Armario 2003.
183
In the panegyrics surveyed here I have identified 23 occurrences of the term orbis, which
the panegyrists used to designate the Roman state; of course, the connotation is aesthetical, poeti-
cal: Pan. Lat. 2 [10]. 2. 1 (totius orbis); 3. 3 (totius orbis); 9. 1 (orbis); 14. 4 (totius orbis); 3
[11]. 2. 4 (orbis); 6. 3 (orbis Romanus); 13. 4 (orbis); 4 [8]. 3. 2 (orbis); 4. 3 (orbis terrarum);
10. 4 (totius orbis terrarum); 17. 3 (orbis terrarum); 5 [9]. 17. 5 (orbis); 21. 3 (orbis); 6 [7]. 1.
3 (totius orbis); 14. 1 (orbis); 8 [5]. 9. 3 (totius orbis); 10 [4]. 3. 3 (orbis terrarum); 35. 1
(orbis); 35. 2 (totius orbis); 36. 1 (orbis); 9 [12]. 3. 5 (totius orbis); 26. 5 (orbis); 11 [2]. 13. 3
(orbis terrarum); see also 12 [2]. 14. 2 (noster orbis). In two instances, the empire is called
mundus ‒ Pan. Lat. 8 [5]. 10. 2; 10 [4]. 12. 3.
184
Pan. Lat. 2 [10]. 1. 5; 7. 2; 9. 4; 10. 1; 11. 2; 13. 2; 2 [11]. 6. 5; 12. 2; 13. 4; 14. 1; 16.
1‒2; 22. 5; 4 [8]. 10. 4; 16. 3; 17. 3; 20. 2; 5 [9]. 5. 1; 6 [7]. 14. 1; 7 [6]. 1. 2; 2. 2; 6. 2; 10.
5; 15. 5; 8 [5]. 2. 1; 3. 1; 9 [12]. 1. 3; 10 [4]. 30. 5; 31. 2; 11 [2]. 3. 1; 11. 1; 15. 2.
185
Pan. Lat. 2 [10]. 13. 2.
186
Pan. Lat. 2 [10]. 4. 4; 8 [5]. 15. 5.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 237

iura),187 which, together, constitute the Romanus populus.188 Hence the empire
is not only a geographical reality, but also a juridical one (Romana res
publica).189 It is organised in praefecturae190 and provinciae191 administered
by public servants in authority (imperii apparatus, praefecti, iudices, duces,
magistratus, etc.).192 It has cities (urbes, oppida, civitates) adorned with pri-
vate (privatae aedes, domus) and public buildings (publica tecta, loca publica,
opera) (palatia, moenia, muri, templa, fora, basilicae, porticus, circus, tribu-
nalia, deambulacra, gymnasia, aqua, caldaria, lavacra, scholae, sedes iusti-
tiae, horrea, cella),193 villages (vici, pagi), ports (portus) harbouring fleets
(classes), cultivated arable land (agri consiti), rich flocks, mines (metalla),
roads (viae).194 At the same time, the empire and the emperors are insepara-
ble.195 The former belongs fully to the latter (cum universa teneatis),196 who
govern it,197 defend it (rem publicam vindicare)198 and permanently care for it
(cura),199 for everyone’s sake (per omnium salute);200 they exercise their

187
Pan. Lat. 2 [10]. 6. 4; 4 [8]. 21. 1; 10 [4]. 38. 4; 11 [3]. 4. 3‒7; 10. 3.
188
Pan. Lat. 4 [8]. 19. 2; 7 [6]. 1. 2; 11 [3]. 14. 5.
189
Pan. Lat. 4 [8]. 20. 2; 6 [7]. 13. 5.
190
Pan. Lat. 11 [3]. 15. 5; 21. 5; 22. 2; 23. 3; 31. 5.
191
Pan. Lat. 2 [10]. 7. 3; 4 [8]. 4. 3; 9. 1; 10. 1; 17. 4; 6 [7]. 12. 5; 7 [6]. 23. 2; 8 [5]. 7. 7;
11 [3]. 1. 5; 7. 1; 9. 2, etc.; for example: Italia (4 [8]. 10. 3; 18. 5), Britannia (3 [11]. 3. 3; 4 [8].
7. 4; 9. 5; 11. 1; 15. 2; 17. 2; 18. 4; 18. 6‒7; 20. 3; 21. 2; 5 [9]. 18. 3; 21. 2; 6 [7]. 4. 2‒3),
Gallia (2 [10]. 5. 1; 6. 1; 7. 4; 4 [8]. 4. 2; 6. 1; 12. 1; 6 [7]. 8. 3; 12. 5; 7 [6]. 6. 2; 9 [12]. 21.
5; 10 [4]. 38. 3; 11 [3]. 3. 1; 4. 3; 4. 5, etc.), Batavia (3 [11]. 3. 3; 5 [9]. 21. 2), Germania (4
[8]. 3. 3; 9 [12]. 21. 5), Raetia (2 [10]. 9. 1; 3 [11]. 5. 4; 16. 1; 4 [8]. 3. 3; 10. 2), Noricum (4
[8]. 10. 2), Pannonia (3 [11]. 4. 2; 4 [8]. 10. 2), Aegyptus (4 [8]. 10. 2; 5 [9]. 21. 2), Syria opima
et fertilis (3 [11]. 4. 2; 4 [8]. 10. 2), Hispania (4 [8]. 18. 5), Africa (4 [8]. 18. 5; 11 [3]. 14. 5),
Libya (5 [9]. 21. 1), Mauretania (6 [7]. 8. 6), Illyricum (11 [3]. 6. 2; 9. 4), Macedonia (11 [3]. 9.
4), Thracia (4 [8]. 21. 1; 13. 3; 14. 1), etc.
192
Pan. Lat. 2 [10]. 2. 1; 3. 3; 7 [6]. 23. 2; 8 [5]. 2. 1; 11 [3]. 1. 5; 4. 2; 5. 2; 14. 4; 15. 1-2;
16. 1; 17. 1; 20. 5; 21. 1; 22. 2‒3; 24. 5; 30. 4; Nixon et al. 1994, 58‒59, n. 17, 394, n. 8, 398,
n. 25; de Trizio 2009, 74‒75.
193
Pan. Lat. 2 [10]. 3. 4; 3 [11]. 11. 1; 11. 3; 4 [8]. 6. 1; 9. 1; 17. 1; 21. 2; 5 [9]; 6 [7]. 6.
2; 11. 3; 7 [6]. 1. 5; 15. 6; 18. 1‒7; 19; 21. 3; 21. 7; 22. 1‒6; 23. 1‒2; 8 [5]. 1. 1; 1. 3; 2. 1;
4. 4; 8. 4; 13. 5‒6; 14. 3; 15. 1‒5; 10 [4]. 17. 3; 38. 4; 11 [3]. 4. 3‒4; 4. 6; 5. 2; 7. 3; 8. 4; 9.
2‒4; 10. 1; 11. 1‒2; 14. 1; 29. 4.
194
Pan. Lat. 3 [11]. 10. 5; 15. 2‒4; 4 [8]. 6. 1‒2; 7. 3; 11. 1; 13. 4; 15. 5; 7 [6]. 18. 3; 19.
1; 19. 4; 8 [5]. 6. 4; 6. 7‒8; 7. 2; 8. 1; 10 [4]. 35. 4‒5; 11 [3]. 10. 1, etc.
195
Pan. Lat. 6 [7]. 2. 2.
196
Pan. Lat. 4 [8]. 20. 5.
197
Pan. Lat. 2 [10]. 4. 2: salutarem manum gubernaculis addidisti; 14. 2: institutio impera-
toria; 3 [11]. 13. 4: rei publicae administratione; 4 [8]. 3. 3; 14. 1: in administranda re publica;
14. 2; 6 [7]. 2. 5: communis salutis gubernacula; 12. 5: imperare; 11. 4; 12. 7; 14. 1; 9 [12].
26. 5: gubernaculis orbis; 10 [4]. 27. 2.
198
Pan. Lat. 7 [6]. 10. 1.
199
Pan. Lat. 2 [10]. 3. 3‒4; 4 [8]. 4. 1; 20. 5; 6 [7]. 2. 1; 9. 3; 11. 7; 11 [3]. 4. 5‒6; 10. 1‒3;
de Trizio 2009, 71‒73; Bucci 2015, 70‒71, 116‒17, 129.
200
Pan. Lat. 2 [10]. 3. 4.
238 N. ZUGRAVU

power (imperium,201 imperii potestas,202 principatus203) and bestow their ben-

efactions upon it (beneficia, munera),204 especially pax,205 pietas and felicitas,
but also other bona and divitiae.206 For, as an orator said in 291, although they
are governing such a large state (vos tantae rei publicae administratione sus-
cepta), although they are solicited everywhere by so many cities (tot urbes), so
many fortifications (tot castra), so many frontiers (tot limites), so many rivers,
mountains and shores scattered in the Roman empire (tot circumiecta Romano
imperio flumina montes litora vocant), the principes have so much strength of
the soul and are so successful that the world is not afraid at all for its safety
(nihilominus orbe securo);207 no part of the earth is deprived of the presence
of your majesty, even when you are far away (neque enim pars ulla terrarum
maiestatis vestrae praesentia caret, etiam cum ipsi abesse videamini), he con-
cludes.208 The same conviction was shared in 297 by the anonymous orator of
the panegyric delivered in honour of Constantius Caesar: the goods distributed
by the divine power of the emperors (numen vestri),209 more numerous than
even those of the gods, were destined for the whole world (vestra in orbem
terrarum distributa propre plura sunt quam deorum).210 Under these condi-
tions, was it not natural for the authors of these encomia to allow themselves
to think in terms that did not seem utopian? They thought of themselves to be
witnesses to a rejuvenated Saturnian golden age, to a time of happiness, to
a state which had been given the chance to flourish anew (renascens res
publica).211

201
Pan. Lat. 2 [10]. 1. 1; 1. 5; 3. 1; 3. 3; 14. 4; 3 [11]. 2. 2; 3. 3‒4; 6. 7; 7. 6; 16. 6; 4 [8].
3. 3; 12. 2; 19. 2; 6 [7]. 1. 1; 1. 2; 2. 2; 3. 2; 5. 2‒3; 7. 2: summum imperium; 7. 4; 7. 6; 9. 6;
10. 1; 11. 1; 11. 4; 12. 2; 12. 4; 13. 5; 14. 1; 14. 3‒4; 7 [6]. 2. 3; 2. 5: honor imperii; 3. 1; 4.
1; 7. 3; 8. 5‒6; 10. 1; 13. 1; 15. 4; 16. 1; 8 [5]. 13. 1; 9 [12]. 2. 3; 24. 4; 10 [4]. 2. 2; 2. 6; 16.
4; 16. 6; 11 [3]. 13. 1: imperium universae rei publicae; 13. 3; 14. 4; 32. 3.
202
Pan. Lat. 2 [10]. 10. 1; 3 [11]. 6. 3: summa potestas; 13. 3; 5 [9]. 15. 3: imperandi potes-
tas; 6 [7]. 14. 5: par imperii potestate; 7 [6]. 15. 1: potestas… imperii; 10 [4]. 8. 2; 11 [3]. 17.
3.
203
Pan. Lat. 10 [4]. 16. 4; 11 [3]. 13. 3: iustus principatus.
204
Pan. Lat. 6 [7]. 1. 2; 7 [6]. 22. 5; 8 [5]. 9. 2; 10. 3; 11. 1; 11. 3; 10 [4]. 35. 1; 11 [3]. 9.
1, etc.; Bucci 2015, 63.
205
Pan. Lat. 2 [10]. 14. 4; 11 [3]. 7. 1.
206
Pan. Lat. 2 [10]. 11. 7; 3 [11]. 13. 1; 18. 1‒5; 19. 2; 4 [8]. 20. 1; 3 [5]. 14. 1‒3; 7 [6].
22. 1; 10 [4]. 35. 1; 11 [3]. 4. 1‒7; 8. 4; 9. 1‒4; 10. 2; 11. 1, etc.; de Trizio 2009, 114.
207
Pan. Lat. 3 [11]. 13. 4.
208
Pan. Lat. 3 [11]. 13. 5.
209
Pan. Lat. 4 [8]. 4. 2.
210
Pan. Lat. 4 [8]. 4. 3.
211
289: Pan. Lat. 2 [10]. 13. 1: Felix igitur talibus, Roma, principibus…; felix, inquam, et
multo nunc felicior quam sub Remo et Romulo tuis; tuque potissimum… has provincias tuas
frequenter inlustres, et profundissima licet pace florentes adventu numinis tui reddas feliciores
(de Trizio 2009, 120‒21); 14. 4 (de Trizio 2009, 166‒67); 291: Pan. Lat. 3 [11]. 1. 3; 16. 1:
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 239

Beyond the limites, there are hostile nations (hostes212); beyond the Euphrates
and ultra Tigrim – the Parthians (Persians)213 and the nearby Palmyrene214 and
Saraceni;215 ultra Oceanum, not far from Britannia,216 trans Rhenum217 and on
the left bank (ad laevam) of the Danube218 ‒ barbarae nationes/gentes,219
populi 220 which are identified only by names (Bructeri, Chamavi, Cherusci,
Lanciones, Alamanni, Tubantes) that spread terror (horror), as Nazarius says in
the panegyric delivered in 321 in honour of Constantine;221 they flood in all over
the place, in an immeasurable number (innumerae gentes222), in an indefinable,
far-reaching space (terrae hostium,223 barbaricum solum,224 ab ultimis barbariae
litoribus225), with regions, rivers and mountains with unknown names (inaudita
regionum fluviorum montium nomina),226 presenting almost no sign of civilisa-
tion – domus, terrae (cultivated lands?), vici 227 – impossible to identify
more precisely even when denominations such as Alamannia,228 Germania,229

imperii vestri felicitas; 18. 5: felicitatem istam, optimi imperatores, pietate meruistis! 297: Pan.
Lat. 4 [8]. 3. 1: aurea illa saecula, quae non diu quondam Saturno rege viguerunt, nunc aeternis
auspiciis Iovis et Herculis renascuntur; 298: Pan. Lat. 5 [9]. 18. 1: felicitas saeculi; 307: Pan.
Lat. 6 [7]. 1. 3; 10. 1 (Bucci 2015, 64, 119); 312: Pan. Lat. 8 [5]. 13. 1: Nobis ergo praecipue
te principem di immortales creaverunt, quibus singulis haec est nata felicitas, ex quo tu imperare
coepisti; 321: Pan. Lat. 10 [4]. 2. 2: salutaris imperii; 2. 6: felicis cursus imperii; 38. 5: Hic
denique status rem est, ut obtinendae potius felicitatis votum geramus qua, augendae cupiditatem
(Laudani 2014, 79‒80, 84, 444); 362: Pan. Lat. 11 [3]. 15. 1: Cis pauculos dies in novum ac
florentem statum re publica restituta; Seager 1984, 152.
212
Pan. Lat. 2 [10]. 8. 2; 10 [4]. 3. 5; 7. 1; 11 [3]. 4. 7; 5. 2, etc.
213
Pan. Lat. 2 [10]. 7. 5; 3 [11]. 9. 2; 4 [8]. 3. 3: Partho quippe ultra Tigrim redacto; 10. 2:
tunc se nimium et Parthus extulerat; 10. 4; 5 [9]. 21. 1‒2. For the confusion between Parthi and
Persae in the 4th-century sources, cf. Chauvot 2016, 157‒66.
214
Pan. Lat. 4 [8]. 10. 2.
215
Pan. Lat. 3 [11]. 5. 4.
216
Pan. Lat. 4 [8]. 20. 3‒4.
217
Pan. Lat. 7 [6]. 8. 4.
218
Pan. Lat. 11 [3]. 7. 2.
219
Pan. Lat. 2 [10]. 5. 1; 4 [8]. 20. 4; 6 [7]. 8. 4; 7 [6]. 6. 2; 10 [4]. 17. 2; 18. 5; de Trizio
2009, 83; Laudani 2014, 230, 240; Bucci 2015, 109.
220
Pan. Lat. 10 [4]. 18. 5.
221
Pan. Lat. 10 [4]. 18. 1: Quid memorem Bructeros, quid Chamavos, quid Cheruscos Lan-
cioneas Alamannos Tubantes? Bellicum strepunt nomina, et immanitas barbariae in ipsis voca-
bulis adhibet horrorem; Nixon et al. 1994, 363, nn. 79‒80; Laudani 2014, 233‒36.
222
Pan. Lat. 10 [4]. 18. 6; Laudani 2014, 242.
223
Pan. Lat. 3 [11]. 16. 2.
224
Pan. Lat. 11 [3]. 8. 4.
225
Pan. Lat. 7 [6]. 6. 2.
226
Pan. Lat. 11 [3]. 6. 2.
227
Pan. Lat. 7 [6]. 12. 3 (vici); 9 [12]. 22. 6 (domus, terrae).
228
Pan. Lat. 4 [8]. 2. 1; 10. 4; 11 [3]. 4. 5; 6. 2; 13. 3.
229
Pan. Lat. 2 [10]. 7. 2; 9. 1‒2; 6 [7]. 8. 5; 11 [3]. 4. 5.
240 N. ZUGRAVU

Sarmatia,230 Francia231 exist, or when allusions are made to the existence of

regna ‒ type territorial structures;232 from there, sometimes reunified (congre-
gata, coactae) in temporary alliances (conspirationes),233 they rush into the
empire spreading horror, viciousness, ferocity (immanitas),234 and bringing the
provinces to ruin.235
In the gratiarum actio de consulatu suo, which was delivered on 1 January
362 at Constantinople, Mamertinus illustrates, if in a triumphant, encomiastic
tone, a most compelling image of the dichotomy between the Roman space
and the space beyond: one of civilisation, of cities, of social and moral order
on this side, and of barbarity, of the instability generated by the endless crowd
of barbarae nationes, on the other.236 Thus, describing Julian’s journey along
the course of the Ister (longissimo cursu Histrum; per proximi fluminis), the
orator stated that on the right bank of that famous river (cum dexteriorem in-
cliti fluminis ripam), that is, in the empire, the emperor was greeted by both
sexes (utriusque sexus), of all social orders (omnium ordinum), armed or not
(armatorum atque inermium); here there were cities (urbes; Romana oppida)
to which the princeps noster distributed, not wheat (allusions to the Greek
hero Triptolemos), but immunitates, privilegia, pecunia, bona spes, libertas,
divitiae. Instead, on the left bank (ad laevam), on the other side of the river (ex
parte altera), on barbarian soil (in barbaricum solum), barbarian peoples could
be seen crying and praying on their knees (despiceretur… in miserabiles pre-
ces genu nixa barbaria); to many, the emperor gave forgiveness and peace
(innumerabilibus barbaris data venia et munus pacis indultum), whilst in oth-
ers he instilled the terror of war, confusion, flight and fear (terrorem bellicum
trepidationes fugas formidines obserentem).237 ‘Aucune réalisme dans ce pas-
sage’, wrote a commentator.238 It is, of course, true, but, beyond the rhetorical
note, the authors of the panegyrics, like many of their contemporaries, knew
perfectly well that there was a foreign territory beyond the Danube/Ister. The
fact that in the middle of the 4th century the river was recognised as a border
that irremediably separated the empire from the barbarian world is demon-
strated by the conclusion of the treaty of 369 (after some, of 370) between

230
Pan. Lat. 3 [11]. 5. 4; 16. 1; 4 [8]. 10. 4.
231
Pan. Lat. 7 [6]. 6. 2; 10. 2.
232
Pan. Lat. 10 [4]. 18. 5: tot regna (Laudani 2014, 240); 11 [3]. 6. 2: ultima regna.
233
Pan. Lat. 10 [4]. 18. 5‒6; Nixon et al. 1994, 364, n. 81; Laudani 2014, 240–42.
234
Pan. Lat. 2 [10]. 7. 3; 3 [11]. 17. 4; 7 [6]. 2. 2; 6. 4; 10 [4]. 18. 1; Lassandro 1980, 197;
de Trizio 2009, 96‒97; Laudani 2014, 234‒35.
235
Pan. Lat. 2 [10]. 5. 1; 3 [11]. 7. 2, etc.
236
Lassandro 1986; 1998; L’Huiller 1992, 288.
237
Pan. Lat. 11 [3]. 7. 1‒3; 8. 3‒4.
238
L’Huiller 1992, 289.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 241

Valens and Athanaricus on the river, as Ammianus Marcellinus and Themis-

tius inform us.239

Frontiers as Areas of Confrontation and Integration240

The empire, as one learns from Libanius’ Oratio LIX and the anonymous
author of the De rebus bellicis, is surrounded from all sides by uncountable
barbarians.241 The panegyrists too state that the Roman frontiers are constantly
attacked by hostes, however the emperors’ justifiable intervention (bellum
necessarium)242 constitute the starting point of the imperial victory ‒ ‘la
marque spécifique de la domination’.243 That is because, as the anonymous
author of a discourse delivered in 307 in honour of Maximian and Constantine
on the occasion of the latter’s marriage to Fausta remarked, it is from the lim-
ites that the empire keeps the barbarian nations under control (per limites qua
Romanum barbaris gentibus instat imperium).244 Along them (per limites),245
on the Ister (per totum Histri limitem), throughout the course of the Euphrates
(per omnem qua tendit Euphratem), on the banks of the Rhine (ripae Rheni),
on the shores of the Ocean (litus Oceani),246 on the Tigris,247 in the enemy’s
territories (in propriis sedibus),248 beyond the Rhine (ultra Rhenum,249 trans
Rhenum250), in Germania251 or Alamannia,252 or beyond the Danube, in Sarma-
tia,253 expeditiones take place, always leading to victories and the forced sub-
mission of the empire’s enemies to the banners of Rome (Romanis armis sig-
nisque cesserunt).254 Also, extra terminos imperii, the sovereigns espouse

239
Themistius Or. 10. 132C–D; Ammianus Marcellinus 27. 5. 9.
240
See also Lassandro 1986; 2000b, 36‒38.
241
Libanius Or. 29. 135; De rebus bellicis 6. 1. For the enemies of the empire in the 3rd‒4th
century, see Elton 2013, 674‒75.
242
Pan. Lat. 4 [8]. 13. 1.
243
L’Huiller 1992, 290.
244
Pan. Lat. 6 [7]. 14. 1.
245
Pan. Lat. 9 [12]. 14. 6.
246
Pan. Lat. 2 [10]. 2. 6.
247
Pan. Lat. 4 [8]. 3. 3.
248
Pan. Lat. 2 [10]. 8. 2.
249
Pan. Lat. 2 [10]. 7. 7.
250
Pan. Lat. 6 [7]. 8. 4.
251
Pan. Lat. 2 [10]. 7. 2; 9. 1‒2 (de Trizio 2009, 103‒05); VI [7], 8, 5.
252
Pan. Lat. 4 [8]. 2. 1; 10. 4.
253
Pan. Lat. 3 [11]. 5. 4; 7. 1; 16. 1; 4 [8]. 10. 4.
254
Pan. Lat. 4 [8]. 11. 4. For example: Pan. Lat. 2 [10]. 5. 1‒4; 7. 1‒7; 3 [11]. 5. 1‒5; 7.
1‒7; 16. 1; 4 [8]. 1. 4; 2. 1; 3. 2‒3; 5. 1‒2; 10. 4; 5 [9]. 18. 5; 6 [7]. 4. 2; 7 [6]. 12. 1‒3; 9 [12].
4. 6; 14. 6; 10 [4]. 3. 3; 6. 2; 17. 1‒2; 18. 1‒6, etc.; L’Huiller 1992, 221‒22; de Trizio 2009,
82‒88; Laudani 2014, 97; Bucci 2015, 108‒09.
242 N. ZUGRAVU

a subtle politics of creating enmity among the barbarae nationes, contributing

thus to their self-annihilation:255 ‘Barbari ad arma, sed invicem dimicaturi!
Vicerunt barbari, sed consanguineos suos!’, perorated Mamertinus in 291.256
Following such or different methods (by imposing client kings,257 capturing or
killing leaders,258 taking hostages,259 creating secure areas,260 constructing
bridges for observation and control,261 transferring on imperial soil262), one
obtains the friendship (amicitia) and submission of (obsequium), the loyalty
(pietas), and the peace (pax, otium) with the enemies.263 All these reinforce not
only the conviction that the state has regained control over its devastated prov-
inces (tot provinciae restitutae),264 but also that its frontiers have expanded
(<tot> prolati limites).265
On the other hand, the space in the proximity of the frontiers is also a space
of coexistence and integration. In 307, recalling the military successes of
Constantius I, the anonymous orator showed that he had appeased in victory
many barbarian nations (plurimas ille barbaras nationes victoriis domuit) and
had pacified them with kindness (venia mitigavit).266 In the last part of the
sentence, the panegyrist voiced another more pragmatic, more sophisticated
mode of conduct toward the barbarians – which the Romans had embraced
mainly since the reign of Marcus Aurelius – a conduct guided by venia, as our
orator says (however the concept can be encountered in other orations as well,

255
Pan. Lat. 3 [11]. 16‒18.
256
Pan. Lat. 3 [11]. 18. 3.
257
Pan. Lat. 2 [10]. 10. 3: cum tam multi reges, imperator, vestri clientes sint, cum per te
regnum receperit Gennoboudes, a te vero munus acceperit; de Trizio 2009, 107‒08.
258
Pan. Lat. 4 [8]. 2. 1; 7 [6]. 10. 2; 10. 5; 11. 5; 11. 6: compendium est devincendorum
hostium duces sustulisse; 12. 1; 10 [4]. 16. 5‒6; 11 [3]. 6. 2; Nixon et al. 1994, 233, n. 45, 361,
n. 73; Laudani 2014, 221‒24.
259
Pan. Lat. 7 [6]. 13. 5.
260
Pan. Lat. 7 [6]. 11. 4: Iam ne procul quidem Rhenum audetis accolere.
261
Pan. Lat. 7 [6]. 13. 1: Insuper etiam Agrippinensi ponte faciundo reliquiis adflictae gentis
insultas.
262
Pan. Lat. 4 [8]. 21. 1; 6 [7]. 8. 7; 7 [6]. 5. 3; L’Huiller 1992, 281, 283.
263
Pan. Lat. 2 [10]. 10. 4‒7; 4 [8]. 20. 3 (quies; armis domitum; pietas); 6 [7]. 8. 5 (amici-
tia); 13. 5 (obsequium; pax); 9 [12]. 23. 2 (imperator noster amicorum regum admittat obse-
quia); 10 [4]. 35. 3‒4 (otium; pax); 38. 3-4 (amicitia; omnia foris placida); 11 [3]. 7. 1; 7. 3
(pax); L’Huiller 1992, 280.
264
Pan. Lat. 4 [8]. 1. 4; 3. 3; 17. 2; 20. 4, etc.
265
Pan. Lat. 4 [8]. 1. 4; see also 2 [10]. 7. 7: quidquid ultra Rhenum prospicio, Romanum est
(de Trizio 2009, 101); 9. 1: Romanum limitem victoria protulit; 3 [11]. 5. 4: transeo limitem
Raetiae repentina hostium clade promotum; 16. 3; 4 [8]. 3. 3: porrectis usque ad Danubii caput
Germaniae Raetiaeque limitibus… qui Romanae potentiae terminos virtute protulerant; 9 [12].
21. 5: post annuam expeditionem statim bellum auspicatus a Tiberi ad Rhenum, immo… a Tusco
Albula ad Germanicum Albam prolaturus imperium.
266
Pan. Lat. 6 [7]. 4. 4.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 243

next to clementia267), or, as the poet Rutilius Namatianus wrote at the begin-
ning of the 5th century, by victrix clementia,268 i.e. a policy of acceptance that
allowed some contingents of adjacent populations to settle on imperial soil.
It was a new strategy, designed to provide the empire with agricultural labour
and new recruits, which is mentioned in 4th-century sources as well (Ammi-
anus Marcellinus, Themistius, Claudius Claudianus, Rutilius Namatianus,
etc.).269 There are many references to this in the Panegyrici Latini too. The
nationes allowed by Constantius I to settle in desertis Galliae regionibus, as
one of the orators said in 310, had proved to be of use to the Roman empire
(Romani imperii… iuvarent), for they were working the land in times of peace
(pacem… cultu) and provide soldiers in times of war (et arma dilectu).270
In the frontier provinces, devastated and depopulated by the barbarians them-
selves, as another orator declared in 297, the former vagrants and predators
repopulate and cultivate as peasants (cultores barbari) or coloni (incolae) the
abandoned lands (loca deserta, arva iacentia) of the Gallo-Roman landowners
which come back to life under their ploughs, rebuild the old houses and restore
the public buildings and temples (extructione veterum domum et refectione
operum publicorum et templorum) as craftsmen (artifices) ‒ finally, they sub-
mit to military discipline and serve as soldiers (obsequiis teritur et tergo coer-
cetur et servire se militiae nomine gratulatur);271 so the barbarians, tamed by
weapon (armis domiti), then chained by kindness (pietate devincti),272 are now
subject to Roman laws (in leges… excoluit).273 The orationes from which we
extracted the fragments just quoted echoes what some specialists have called
‘the cultural model of Roman hegemony’,274 the stereotype of peaceful subser-
vience (pacificum iugum, as the same poet Rutilius Namatianus called it275)

267
Pan. Lat. 7 [6]. 5. 3 (clementia); 10. 1 (clementia) 10. 4 (venia); 10. 4 (Cautior licet sit
qui devinctos habet venia perduelles); 11. 3 (venia); 11 [3]. 7. 3 (venia); Heim 1992, 283; Trizio
2009, 93; Bucci 2015, 86.
268
Rutilius Namatianus De reditu suo 1. 69: Mitigat armatas victrix clementia vires; Heim
1992, 284, n. 25.
269
Daly 1972; Heim 1992, 282‒89; Raimondi 2000; Danvoye 2007; Sánchez Medina; Castello
2010.
270
Pan. Lat. 7 [6]. 6. 2; 12. 3; Nixon et al. 1994, 235, n. 55.
271
Pan. Lat. 4 [8]. 1. 4; 8. 4; 9. 1‒5; 21. 1‒2; Heim 1992, 284, 285; L’Huiller 1992, 281‒83;
Nixon et al. 1994, 142‒44, nn. 76‒78.
272
Pan. Lat. 4 [8]. 20. 3; Heim 1992, 284.
273
Pan. Lat. 4 [8]. 21. 1; Heim 1992, 284; L’Huiller 1992, 281‒82.
274
L’Huiller 1992, 283.
275
Rutilius Namatianus De reditu suo 1. 80.
244 N. ZUGRAVU

through ‘civilisation’,276 acculturation or Romanisation,277 which Tacitus had

formulated with genius in Vita Agricolae: referring to the material or spiritual
benefactions that his father-in-law bestowed upon the Brittani, Tacitus con-
cluded: Idque apus imperitos humanitas vocabatur, cum pars servitutis
esset.278
Therefore, the references to the frontiers of the Roman empire in the Pane-
gyrici Latini provide important elements of imperial ideology and propaganda
and, at the same time, necessary data for historical reconstruction; this con-
firms the observation that they are set at the interference between history and
rhetoric, between the ‘informative and the conventional registers of
stereotypy’.279

BIBLIOGRAPHY

Abbatepaolo, M. 2012: ‘Panegyrici Latini e laudationes rinascimentali: un esempio di

continuità letteraria’. Classica et Christiana 7.2, 337‒51.
Amiotti, G. 1987: ‘Le Collonne d’Ercole e i limiti dell’ecumene’. In Sordi 1987,
13‒20.
Arce, J. 2001: ‘Frontier of the Late Roman Empire: Perceptions and Realities’. In
Pohl, W., Wood, I. and Reimitz, H. (eds.), The Transformation of Frontiers from
Late Antiquity to Carolingians (The Transformation of the Roman World 10)
(Leiden/Boston), 5‒13.
Balbao Salgado, A. 1995‒96: ‘Más allá de las columnas de Hércules: el discurso espa-
cial en la Ora Maritima de Avieno’. Studia Historica. Historia Antigua 13‒14,
195‒204.
Balbo, A. 2017a: ‘Roman oratory and power under the Flavians: some case studies
from Pliny the Younger’. In Bessone, F. and Fucecchi, M. (eds.), The Literary
Genres in the Flavian Age: Canons, Transformations, Reception (Trends in
Classics Suppl. 51) (Berlin/Boston), 83‒99.
—. 2017b: ‘Some Thoughts on the Image of Rome in Late Antique Authors: Ausonius,
Symmachus, the Panegyrists’. Horizons 8.1, 27‒45.
Băjenaru, C. 2010: Minor Fortifications in the Balkan-Danubian Area from Diocletian
to Justinian (Cluj-Napoca).
Barbero, A. 2016: Costantino il vincitore (Rome).

276
For example: Pan. Lat. 7 [6]. 5. 3: nec contentus vicisse ipsas in Romanas transtulit
nationes, ut non solum arma sed etiam feritatem ponere cogerentur; L’Huiller 1992, 283; Liba-
nius Or. 59. 93, 131‒132; Libanius displays a general attitude that is anti-barbarian – cf. Malosse
2002, 171‒72; 2009.
277
Janniard 2006; Graham 2006, 120‒21; Celeste Berardi 2007; Ziche 2011, 204‒07; Ellis
2011, 245‒47; Malaspina 2011; Chauvot 2016, 197‒226, 259‒99.
278
Tacitus Agr. 21. 1‒3.
279
L’Huiller 1992.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 245

Bernard, G. 2012: ‘Monstres et héros des confins occidentaux de l’oikumène’. In

Prados, F., García, I. and Bernard, G. (eds.), Confines: el extremo del mundo
durante la Antigüedad (Alicante), 73‒94.
Béranger, J. 1970: ‘L’expression de la divinité dans les Panégyriques Latins’. Museum
Helveticum 27.4, 241‒54.
Bianchi, E.A. 2016: ‘I Bagaudi alla presa di Autun (270 AD). Vecchie questioni, nuove
conferme’. Classica et Christiana 11, 9‒41.
Blockley, R. 2012: ‘The Panegyric of Claudius Mamertinus on the Emperor Julian’. In
Rees 2012, 349–59.
Bocci, S. 2011–12: ‘La frontiera danubiana in Ammiano Marcellino’. Geographia
Antiqua 20–21, 105–66.
Borca, F. 2004: Confrontarsi con l’Altro: I Romani e la Germania (Milan).
Borgognoni, R. 2010: ‘Parlare alle istituzioni, parlare delle istituzioni: retorica, verità
e persuasione nell’Oriente tardoantico’. In Bonamente, G. and Lizzi Testa, R.
(eds.), Istituzioni, carismi ed esercizio del potere (IV–VI secolo d.C.) (Bari),
77‒90.
Bucci, T. 2010: ‘Massimino da imperator invictus a perfidus hostis’. Invigilata
Lucernis 32, 35‒42.
—. 2012: ‘Formule di ‘modestia’ negli exordia dei Panegyrici Latini’. Classica et
Christiana 7.1, 43‒56.
—. (ed.) 2015: Anonimo, Il panegirico del 307 per Massimiano e Costantino (Bari).
Campbell, B. 2012: Rivers and the Power of Ancient Rome (Chapel Hill).
Cañizar Palacios, J.L. 2009: ‘Algunas consideraciones sobre la ambiguitas religiosa en
el s. IV d. C. los discursos de Claudio Mamertino y Pacato’. Klio 91.1, 443‒57.
—. 2013: ‘Tortura y castigo en los panegíricos latinos’. In Crifò, G. and Giglio, S.
(eds.), Atti dell’Accademia Romanistica Costantiniana. XIX Convegno Inter-
nazionale: In memoria di Franca de Marini Avonzo (Organizzare Sorvegliare
Ponire. Il controllo dei corpi e delle menti nel diritto della Tarda Antiquità,
Spello-Perugia 25–27 giugno 2009) (Perugia), 203‒27.
—. 2017: ‘Reflexiones sobre la perceptión e idea de la frontiera en la Hispania del s.
V. d. C.’. Studia Historica. Historia Antigua 35, 53‒74.
Carrié, J.-M. 1995: ‘1993: ouverture des frontières de l’Empire romain?’. In
Rousselle, A. (ed.), Frontières terrestres, frontières célestes dans l’Antiquité
(Presses universitaires de Perpignan, Études 20) (Paris), 31‒53.
Casella, M. 2017: Galerio. Il tetrarca infine tollerante (Monografie del Centro
Ricerche e Documentazione sull’Antichita Classica 43) (Roma).
Castello, M.G. 2010: ‘La crisi dell’Impero e la frantumazione dell’illusione di rinascita.
La Gratiarum Actio di Decimio Magno Ausonio’. Historia 59.2, 189‒205.
Celeste Berardi, C. 2007: ‘Barbari exsecrati gladios suos ad aratra conversi sunt
(Oros., adv. pag. 7, 41, 7). Il tema della pacificazione dei barbari tra IV e V sec.
d.C.’. Auctores Nostri 5, 17‒37.
Chauvot, A. 2016: Les «barbares» des Romains. Répresentantions et confrontations
(Metz).
Collins, R., Symonds, M.F.A. and Weber, M. 2015: Roman Military Architecture on
the Frontiers: Armies and Their Architecture in Late Antiquity (Oxford/
Philadelphia).
Daly, L.J. 1972: ‘The Mandarins and the Barbarians: The Response of Themistius to
the Gothic Challenge’. Historia 21.2, 351‒79.
246 N. ZUGRAVU

Danvoye, S. 2007: ‘Contacts pacifiques et violents entre les Barbares et les Romains
dans les panégyriques de Claudien’. Latomus 66.1, 132‒49.
de Trizio, M.S. 2005: ‘Un’eco virgiliana nel Panegirico di Mamertino del 289 d.C.’.
Invigilata Lucernis 27, 155‒68.
—. 2006: ‘Echi ciceroniani nel panegyrico di Mamertino per l’imperatore Massimiano
(289 d.C.)’. Invigilata Lucernis 28, 61‒74.
—. 2007: ‘La propaganda della Concordia Augustorum nei Panegyrici di Mamertino’.
Classica et Christiana 2, 65‒78.
—. (ed.) 2009: Panegirico di Mamertino per Massimiano e Diocleziano (Panegyrici
Latini 2 [10]) (Biblioteca tardoantica 2) (Bari).
Desideri, P. and Fontanella, F. (ed.) 2013: Elio Aristide e la legittimazione greca
dell’impero di Roma (Bologna).
Díez del Corral Corredoira, P. 2004: ‘El mar en el fin del mundo: Océano en la musi-
varia de Gallaecia’. Gallaecia 23, 35‒56.
Dietz, K. 1993: ‘Cohortes, ripae, pedaturae. Zur Entwicklung der Grenzlegionen in
der Spätantike’. In Dietz, K., Hennig, D. and Kaletsch, H. (eds.), Spätantike und
frühes Christentum. Adolf Lippold zum 65. Geburtstag gewidmet (Würzburg),
279‒329.
Donciu, R. 2012: L’empereur Maxence (Munera 34) (Bari).
Drijvers, J.W. 2011: ‘The Limits of the Empire in the Res Gestae of Ammianus
Marcellinus’. In Hekster, O. and Kaizer, T. (eds.), Frontiers in the Roman World
(Proceedings of the Ninth Workshop of the International Network Impact of
Empire, Durham, 16–19 April 2009) (Impact of Empire 13) (Leiden/Boston),
13‒29.
Ellis, L. 2011: ‘Elusive Places: A Chorological Approach to Identity and Territory in
Scythia Minor (Second–Seventh Centuries)’. In Mathisen, R.W. and Shanzer, D.
(eds.), Romans, Barbarians, and the Transformation of the Roman World: Cul-
tural Interaction and the Creation of Identity in Late Antiquity (Farnham),
241‒51.
Elton, H. 1996: ‘Defining Romans, Barbarians, and the Roman Frontier’. In Mathisen
and Sivan 1996, 126‒35.
—. 2013: ‘Imperial Campaigns between Diocletian and Honorius, A.D. 284–423: The
Rhine Frontier and the Western Provinces’. In Sarantis and Christie 2013a,
657‒68.
Enenkel, K. 2000: ‘Panegyrische Geschichtsmythologisierung und Propaganda: Zur
Interpretation des Panegyricus Latinus VI’. Hermes 128, 91‒126.
Fontaine, J. 1983: ‘De l’universalisme antique aux particularismes médiévaux: la con-
science du temps et de l’espace dans l’Antiquité tardive’. In Popoli e paesi nella
cultura altomedievale: 23‒29 aprile 1981, vol. 1 (Settimane di studio del Centro
italiano di studi sull’alto Medioevo 29) (Spoleto), 13‒45.
Forni, G. 1987: ‘Limes: nozioni e nomenclature’. In Sordi 1987, 282‒94.
Galletier, É. (ed.) 1949: Panégyriques Latins, vol. 1 (Paris).
—. 1952: Panégyriques Latins, vol. 2 (Paris).
—. 1955: Panégyriques Latins, vol. 3 (Paris).
Garbarino, G. 2005: ‘Viaggi in capo al mondo da Catulli a Seneca’. In Gargano, A.
and Squillante, M. (eds.), Il viaggio nella letteratura occidentale tra mito e sim-
bolo (Naples), 23‒44.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 247

García Ruiz, M.P. (ed.) 2006: Claudio Mamertino, Panegírico (Gratiarum actio) al
Emperador Juliano (Pamplona).
—. 2008a: ‘La evolución de la imagen política del emperador Juliano a través de los
discursos consulares. Mamertino, “Pan.” III (11) y Libanio, “Or.” XII’. Minerva
21, 137‒53.
—. 2008b: ‘Una lectura de la gratiarum actio de Claudio Mamertino a la luz de los
primeros escritos de Juliano’. Emerita 76.2, 231‒52.
—. 2013: ‘Rethinking the political role of Pliny’s Panegyricus in the Panegyrici
Latini’. Arethusa 46, 195‒216.
Giuliese, C. 2005: ‘Costantino stratega nei Panegirici del 313 e del 321’. Invigilata
Lucernis 27, 169‒75.
—. 2007: ‘La prosopopea di Roma nei Panegyrici del 313 e del 321’. Classica et
Christiana 2, 97‒105.
Glad, D. 2017: ‘Le Danube à l’époque romaine tardive et protobyzantine (284-614/5).
Limes ou espace d’échanges?’. In Baron, N., Boissellier, S., Clément, F., and
Sabaté, F. (eds.), Ériger et borner diocèses et principautés au Moyen Âge: Limites
et frontières II (Villeneuve d’Ascq), 131‒43.
Graham, M.W. 2006: News and Frontier Consiousness in Late Roman Empire (Ann
Arbor).
Grünewald, T. 1990: Constantinus Maximus Augustus. Herrschaftspropaganda in der
zeitgenössischen Überlieferung (Historia Einzelschriften 64) (Stuttgart).
Guzmán Armario, F.J. 2003: ‘El «relevo de la barbarie». La evolución histórica de un
fecundo aequetipo clásico’. Veleia 20, 331‒40.
—. 2010: ‘Teoría y praxis del poder imperial en el siglo IV: lo que nos enseñan los
panegíricos’. In Bravo, G. and González Salinero, R. (eds.), Toga y daga: Teoría
y praxis de la política en Roma (Actas del VII Coloquio de la Asociación Inter-
disciplinar de Estudios Romanos) (Madrid), 265‒74.
—. 2012: ‘La frontera del Danubio durante el Imperio Romano: ¿Barrera o espacio
para el inercambio cultural?’. In Prados, F., García, I. and Bernard, G. (eds.),
Confines: el extremo del mundo durante la Antigüdad (Alicante), 385‒96.
Hardie, P.R. 1986: Virgil’s Aeneid: Cosmos and Imperium (Oxford).
Hartmann, U. 2008: ‘Quellen’. In Johne, K.-P. (ed.), Die Zeit der Soldaten-Kaiser.
Krise und Transformation des Römischen Reiches im 3. Jahrundert n.Chr. (235–
284), vol. 1 (Berlin), 39‒40.
Heim, F. 1992: ‘Clémence ou extermination: le pouvoir impérial et les barbares au
IVe siècle’. Ktema 18, 281‒95.
Hidalgo de la Vega, M.J. 2003: ‘Algunas reflexiones sobre los límites del oikumene en
el Imperio Romano’. Gerión 23.1, 271‒85.
Hostein, A. 2012: La Cité et l’Empereur: Les Eduens dans l’Empire romain d’après
les Panégyriques Latins (Paris).
—. 2016: ‘Le Panégyrique latin VII (6) ou l’éloge de la pietas de Constantin’. In
Guichard, L., Gutsfeld, A. and Richard, F. (eds.), Constantin et la Gaule: Autour
de la vision de Grand (Actes du colloque international Constantin et la Gaule,
Grand et Domrémy-la-Pucelle, 6 et 7 octobre 2010) (Nancy/Paris), 35‒48.
Howard-Johnston, J. 2013: ‘Military Infrastructure in the Roman Provinces North and
South of the Armenian Taurus in Late Antiquity’. In Sarantis and Christie 2013a,
853‒91.
248 N. ZUGRAVU

Iglesias García, S. 2013: ‘El epíteto invictus durante los siglos III d.C. y IV d.C. en el
ámbito imperial’. Antesteria 2, 121‒41.
Isaac, B. 1988: ‘The Meaning of the terms Limes and Limitanei’. Journal of Roman
Studies 78, 125‒47.
—. 1990: The Limits of Empire: The Roman Army in the East (Oxford).
Janniard, S. 2001: ‘L’armée romaine tardive dans quelques travaux récents. 2e partie:
stratégies et techniques militaires’. L’Antiquité Tardive 9, 351‒61.
—. 2006: ‘Armée et «acculturation» dans l’Orient romain tardif. L’exemple des con-
fins syro-mésopotamiens (Ve–VIe s. ap. J.-C.)’. Mélanges de l’École française de
Rome. Antiquité 118.1, 133‒49.
Kalafikis, G. 2014: ‘Ammianus Marcellinus on the Military Strategy of the Emperor
Valentinianus (364–375 AD): General Principles and Implementation’. Byzantiaka
31, 26‒30.
Kasprzyk, M. 2016: ‘Les mesures de Constantin en faveur de la cité des Éduens (Pan.
Lat. VIII (5)): réponse à une crise locale ou mesure politique?’. In Guichard, L.,
Gutsfeld, A. and Richard, F. (eds.), Constantin et la Gaule: Autour de la vision
de Grand (Actes du colloque international Constantin et la Gaule, Grand et
Domrémy-la-Pucelle, 6 et 7 octobre 2010) (Nancy/Paris), 199‒216.
L’Huiller, M.-C. 1992: L’Empire des mots: orateurs gaulois et empereurs romains,
3e et 4e siècles (Annales littéraires de l’Université de Besançon 464) (Besançon/
Paris).
La Bua, G. 2009: ‘Laus deorum e strutture inniche nei Panegirici latini di etá imperi-
ale’. Rhetorica 27.2, 142‒58.
Lagioia, A. 2004: ‘L’‘epifania’ di Giuliano in Illiria e il Palladio (Paneg. 11,6,4)’.
Invigilata Lucernis 26, 123‒37.
Lassandro, D. 1986: ‘L’integrazzione romanobarbarica nei Panegyrici Latini’. In Sordi,
M. (ed.), L’Europa nel mondo antico (Contributi dell’Istituto di storia antica 12)
(Milan), 153‒59.
—. 1987: ‘Il limes renano nei Panegyrici Latini’. In Sordi 1987, 295‒300.
—. (ed.) 1992: XII Panegyrici Latini (Turin).
—. 1993: ‘Oriente e Occidente nei Panegirici Latini’. In Conca, F., Gualandri, I. and
Lozza, G. (eds.), Politica, cultura e religione nell’Impero Romano (secoli IV–VI)
tra Oriente e Occidente (Atti del Secondo Convegno dell’Associazione di Studi
Tardoantichi) (Naples), 219‒25.
—. 1998: ‘La riva sinistra del Danubio e la Gratiarum actio di Claudio Mamertino
all’imperatore Giuliano (362 d.C.)’. Studia Antiqua et Archaeologica 5, 175‒88.
—. 2000a: Sacratissimus Imperator. L’immagine del princeps nell’oratoria tardoantica
(Quaderni di Invigilata Lucernis 8) (Bari).
—. 2000b: ‘Introduzione’. In Lassandro and Micunco 2000, 9–58.
—. 2001: ‘Reno e Danubio nei Panegyrici Latini’. Studia Antiqua et Archaeologica 8,
205‒09.
—. 2006: ‘L’imperator invictus in alcuni testi tardoantichi’. Classica et Christiana 1,
99‒110.
—. 2013: ‘Bella movere docent, melius qui rura moverent (Sigeberto): rivolte conta-
dine di età tardoantica e tradizione medievale’. Classica et Christiana 8.1, 175‒85.
—. 2015: ‘La legittimazione religiosa dell’auctoritas imperiale di Costantino nei pane-
girici’. Classica et Christiana 10, 213‒24.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 249

Lassandro, D. and Micunco, G. (ed.) 2000: Panegirici Latini (Turin).

Laudani, C. (ed.) 2014: Nazario, Panegirico in onore di Costantino (Bari).
Leadbetter, B. 1998: ‘Patrimonium indivisum?’. Chiron 28, 213‒28.
Lee, A.D. 1993: Information and Frontiers: Roman Foreign Relations in Late
Antiquity (Cambridge).
—. 2007: War in Late Antiquity: A Social History (Oxford).
Lippold, A. 2012: ‘The Ideal of the Ruler and Attachment to Tradition in Pacatus’
Panegyric’. In Rees 2012, 360‒86.
Long, J. 2009: ‘How to Read a Halo: Three (or More) Versions of Constantine’s
Vision’. In Cain, A. and Lenski, N. (eds.), The Power of Religion in Late Antiquity
(Farnham), 227‒35.
Lopetegui, G. 2015: ‘Retórica e imagen imperial en la obra de los epitomadores y
panegyristas del siglo IV’. Veleia 36, 63‒82.
MacCormack, S.G. 1981: Art and Ceremony in Late Antiquity (Transformation of the
Classical Heritage 1) (Berkeley/London).
—. 2012: ‘Imagery in Panegyrics’. In Rees 2012, 240‒50.
Malaspina, E. 2011: ‘Incontri di popoli e prove di convivenza (III/V secolo)’. Classica
et Christiana 6.2, 473‒512.
Malosse, P.-L. 1998–99: ‘Fausser avec du vrai: l’art du masque dans les éloges royaux,
d’après l’exemple de Libanios’. Plekos 1 (http://www.plekos.uni-muenchen.
de/98,99/amalosse.html) (consulted 25 April 2018).
—. 2002: ‘Le modèle du mauvais empereur chez Libanios’. Pallas 60, 165‒74.
—. 2003: ‘Introduction’. In Malosse, P.-L. (ed.), Libanios, Discours, IV, Discours IX,
(Paris).
—. 2006: ‘Rhétorique et géographie: le topos encomiastique des extrémités de
l’Empire et du monde’. Pallas 72, 205‒18.
—. 2009: ‘Il retore e il potere: Libanio e l’imperatore Valente’. In Laudizi, G. (ed.),
Satura rudina: Studi in onore di Pietro Luigi Leone (Lecce), 84‒89.
Marotta, V. 2010: ‘Gli dèi governano il mondo. La trasmissione del potere imperiale
in età tetrarchica’. In Costabile, F. (ed.), Polis: Studi interdisciplinari sul mondo
antico, vol. 3 (Reggio Calabria), 171‒88.
Martin, S.P. 1999: Rome and the Enemy: Imperial Strategy in the Principate (Ber-
keley/London).
Mathisen, R.W. and Sivan, H.S. (eds.) 1996: Shifting Frontiers in Late Antiquity
(Papers from the First Interdisciplinary Conferences on Late Antiquity, the
University of Kansas, March, 1995) (Aldershot).
Mayerson, P. 1989: ‘The Meaning of the World Limes (Límiton) in the Papyri’. ZPE
77, 287‒91.
Mazza, M. 2005: Cultura guerra e diplomazia nella Tarda Antichità: tre studi
(Catania).
Moretti, G. 1999: ‘Via della virtù, lotta coi mostri e viaggio ai confini del mondo: il
modello di Eracle nel IX libro del «Bellum civile»’. In Esposito, P. and Nicastri,
L. (eds.), Interpretare Lucano. Miscellanea di studi (Quaderni del Dipartimento
di scienze dell’antichità 22) (Naples), 237‒52.
Morgan, M.M. and Lieu, S.N.C. 1989: ‘A speech of thanks to the Emperor Julian
given by Claudius Mamertinus (on the occasion of his elevation to the Consulate)
(1 Jan., 362). Latin Panegyric XI (3) (Claudii Mamertini gratiarum actio de
250 N. ZUGRAVU

consulato suo Iuliano Imp.)’. In Lieu, S.N.C. (ed.), The Emperor Julian:
Panegyric and Polemic. Claudius Mamertinus, John Chrysostom, Ephrem the
Syrian (Liverpool), 1‒38.
Mouchová, B. 2002: ‘Das Bild des Usurpators in der Preisrede des Pacatus’. Graeco-
latina Pragensia 19, 63‒71.
Müller-Rettig, B. (ed.) 1990: Der Panegyricus des Jahres 310. Auf Konstantin der
Grossen: Übersetzung und historisch-philologischer Kommentar (Stuttgart).
—. (ed.) 2008: Panegyrici Latini. Lobreden auf römische Kaiser 1: Von Diokletian bis
Konstantin (Darmstadt).
Mund-Dopchie, M. 2004: ‘La frontière entre le civilisé et le sauvage dans l’imaginaire
de l’Occident latin: usages et mésusages des critères antiques’. Folia Electronica
Classica (Louvain-la-Neuve) 7 (http://bcs.fltr.ucl.ac.be/FE/o7/civilise.html) (con-
sulted 25 April 2018).
Nicolet, C. 1988: L’inventaire du monde: Géographie et politique aux origines de
l’Empire romain (Paris).
—. 1990: Space, Geography, and Politics in the Early Roman Empire (Ann Arbor).
Nixon, C.E.V., Rodgers, B.S. and Mynors, R.A. Baskerville- (eds.) 1994: In Praise of
Later Roman Emperors: The Panegyrici Latini (Berkeley/Oxford).
Orihuela Sancho, R. 1996: ‘Consideraciones sobre el Panegírico XI(3) del Corpus
Panegyricorum Latinorum’. Myrtia 11, 47‒60.
Pérez Sánchez, D. 2003: ‘Panegírico y ciudad: tradición y control ideológico en la
antigüedad tardía’. Studia Historica. Historia Antigua 21, 223‒45.
Pernot, L. 1993: La rhétorique de l’éloge dans le monde gréco-romain, 2 vols.
(Collection des études augustiniennes. Série Antiquité 137–138) (Paris).
Piacente, L. 2004: ‘Il Danubio negli autori latini tardoantichi’. In Santelia, S. (ed.),
Italia e Romania. Storia, Cultura e Civiltà a confronto (Atti del IV Convegno di
Studi italo-romeno, Bari, 21–23 ottobre 2002) (Quaderni di Invigilata Lucernis
21) (Bari), 173‒80.
Plácido Suárez, D. 2008: ‘La ecúmene romana: espacios de integración y exclusión.
Presentación’. Studia Historica. Historia Antigua 26, 15‒20.
Pohl, W., Wood, I. and Reimitz, H. (eds.) 2001: The Transformation of Frontiers from
Late Antiquity to the Carolingians (Transformations of the Roman World 10)
(Leiden/Boston).
Purcell, N. 2012: ‘Rivers and the geography of power’. Pallas 90, 373‒87.
Raimondi, M. 2000: ‘Temistio e la prima guerra gotica di Valente’. Mediterraneo
Antico 3.2, 633‒83.
Rees, R. 2002: Layers of Loyalty in Latin Panegyric AD 289–307 (Oxford).
—. (ed.) 2012: Latin Panegyric (Oxford).
Roberto, U. 2014: Diocleziano (Rome).
—. 2017: ‘Il Panegirico di Eumenio. Educazione, scuola e politica nell’Impero di
Diocleziano’. Annali di storia dell’educazione, 24, 106‒18.
Rocco, M. 2017: ‘Ottaviano Augusto nella panegirista tardoantica. Dal deus praesens
latino-ellenistico alla teologia politica cristiana’. Athenaeum 105.1, 153‒86.
Roche, P. (ed.) 2011: Pliny’s Praise: The Panegyricus in the Roman World
(Cambridge).
Rodgers, B. 1986: ‘Divine Insinuation in the Panegyrici Latini’. Historia 35, 69‒104
—. 2012: ‘Divine Insinuation in the Panegyrici Latini’. In Rees 2012, 284‒334.
 THE FRONTIERS OF THE EMPIRE IN THE PANEGYRICI LATINI 251

Rodríguez Gervás, M.J. 1991: Propaganda política y opinión pública en los panegíri-
cos latinos del bajo imperio (Acta Salmanticensia. Estudios históricos y geográfi-
cos 79) (Salamanca).
Roman, Y. 1983: ‘Auguste, l’Océan Atlantique et l’impérialisme romain’. Ktema 8, 261‒68.
Ronning, C. 2007: Herrscherpanegyrik unter Trajan und Konstantin. Studien zur sym-
bolischen Kommunication in der römischen Kaiserzeit (Studien und Texte zu
Antike und Christentum 42) (Tübingen).
Rousselle, A. 1976: ‘La chronologie de Maximien Hercule et le mythe de la Tétrarchie’.
Dialogues d’Histoire Ancienne 2, 445‒46.
Ryan, F.X. 2009: ‘Pacatus on the menmonic capabilities of republican political fig-
ures’. Leeds International Classical Studies 8.1 (www.leeds.ac.uk/classics/lics/)
(consulted 25 April 2018).
Sabbah, G. 1984: ‘De la rhétorique à la communication politique: les Panégyriques
Latins’. Bulletin de l’Association Guillaume Budé 43, 363‒88.
Sánchez Medina, E. 2010: ‘Entre filantropía y pragmatismo: los Discursos Públicos de
Temistio al emperador Teodosio y el problema bárbaro’. In Bravo, G. and González
Salinero, R. (eds.), Toga y daga: Teoría y praxis de la política en Roma (Actas del
VII Coloquio de la Asociación Interdisciplinar de Estudios Romanos) (Madrid),
291‒305.
Sancho Gómez, M.P. 2011: Guerra y política en el Imperio Romano de Occidente
(337–361). Usurpadores, bárbaros, y querellas dinásticas: los problemas del
Imperio en Occidente después de Constantino el Grande [Editorial Académica
Española].
Sarantis, A. 2013a: ‘Fortifications in Africa: A Bibliographic Essay’. In Sarantis and
Christie 2013a, 297‒315.
—. 2013b: ‘Fortifications in the East: A Bibliographic Essay’. In Sarantis and Christie
2013a, 317–70.
Sarantis, A. and Christie, N. (eds.) 2013a: War and Warfare in Late Antiquity: Current
Perspectives, 2 vols. (Late Antique Archaeology 8.1–2) (Leiden/Boston).
—. 2013b: ‘Fortifications in the West: A Bibliographic Essay’. In Sarantis and Christie
2013a, 255‒96.
Saylor Rodgers, B. 1986: ‘Divine Insinuation in the Panegyrici Latini’. Historia 35,
69‒104.
Scharf, R. 2005: Der Dux Mogontiacensis und die Notitia Dignitatum: Eine Studie zur
spätantiken Grenzverteidigung (Reallexikon der germanischen Altertumskunde
Ergänzungsbände 50) (Berlin/New York).
Schwartz, J. 1970: ‘Le limes selon l’Histoire Auguste’. In Bonner Historia-Augusta-
Colloquium 7: 1968/1969 (Bonn) 233‒38.
Seager, R. 1984: ‘Some Imperial Virtues in the Latin Prose Panegyrics. The Demands
of Propaganda and the Dynamics of Litterary Composition’. In Cairns, F. (ed.),
Papers of the Liverpool Latin Seminar, vol. 4 (ARCA 11) (Liverpool), 129‒75.
Sordi, M. (ed.) 1987: Il confine nel mondo classico (Contributi dell’Istituto di storia
antica 13) (Milan).
Tommasi Moreschini, C.O. 2016: ‘Coping with the Ancient Gods, Celebrating
Christian Emperors, Proclaiming Roman Eternity: Rhetoric and Religion in Late
Antique Latin Panegyric’. In Kahlos, M. (ed.), Emperor and the Divine: Rome
and its Influence (Helsinki), 177‒209.
252 N. ZUGRAVU

Traina, G. 2004: ‘La frontiera armena dell’Impero romano. I due punti di vista’. In
Moatti, C. (ed.), La mobilité des personnes en Méditerranée de l’Antiquité à
l’Époque moderne. Procédures de contrôle et documents d’identification (Collec-
tion de l’École française de Rome 341) (Rome), 205‒23.
Turcan-Verkerk, A.-M. 2003: Un poète latin chrétien redécouvert: Latinus Pacatus
Drepanius, panégyriste de Théodose (Collection Latomus 276) (Brussels).
Vanotti, G. 1987: ‘Prospettive ecumeniche e limiti reali nella definizione dei confini
augustei’. In Sordi 1987, 234‒49.
Ware, C. 2017: ‘The Lies the Poets Tell. Poetry in Prose Panegyrics’. In Elsner, J. and
Hérnandez Lobato, J. (eds.), The Poetics of Late Latin Literature (New York),
345‒69.
Warmington, B. 2012: ‘Aspects of Constantinian Propaganda in the Panegyrici
Latini’. In Rees 2012, 335‒48.
Whately, C. 2013a: ‘War in Late Antiquity: Secondary Works, Literary Sources and
Material Evidence’. In Sarantis and Christie 2013a, 101‒51.
—. 2013b: ‘Strategy, Diplomacy and Frontiers: A Bibliographic Essay’. In Sarantis
and Christie 2013a, 239‒54.
Whitby, M. (ed.) 1998: The Propaganda of Power: The Role of Panegyric in Late
Antiquity (Mnemosyne Suppl. 183) (Leiden/Boston/Cologne).
Whittaker, C.R. 1994: Frontiers of the Roman Empire: A Social and Economic Study
(Baltimore).
—. 2004: Rome and its Frontiers: The Dynamics of Empire (London/New York).
Ziche, H. 2011: ‘Barbarian Raiders and Barbarian Peasants: Model of Ideological and
Economic Integration’. In Mathisen, R.W. and Shanzer, D. (eds.), Romans,
Barbarians, and the Transformation of the Roman World: Cultural Interaction
and the Creation of Identity in Late Antiquity (Farnham), 199‒219.
 INDEX

Abritus 99–101 Baetica 21, 169

Adrianople 206 Bagaudae 221
Aegyssus 99, 101 Barbaricum 46, 181, 194, 196–99, 203,
Aelius Aristides 230 208, 229, 239, 240
Africa 32, 169, 232, 237 Barboşi (Şendreni) 72, 73, 98, 99, 101,
Agrippensis 46 109, 110
Aizanoi 44 Basilia 47
ala I Tungrorum 65, 97, 103 Bassiana 96, 105, 200
ala III Asturum 68 Batavia 233, 237
Alexandria 33, 45, 46, 70 Batavos 96
Alsóhetény 200, 201 Beroe 107, 118
Amastris 47 Bétique see Baetica
Ammianus Marcellinus 203–06, 224, Black Sea region 36, 44, 47, 107, 108,
243 111, 112, 116–19
Ampelum 159–60 Blandiana 105
Ancyra 15, 43 Bohemia 197
Antoninus Pius 61, 104, 164, 179, 230 Brigetio 103
Apamea 96 Britannia 39, 40, 45, 59, 64, 71, 223,
Apulum 69, 70, 154, 160–62, 174, 175, 227, 232–35, 237, 239
178, 180–82, 185, 186 Brixia 148
Aquileia 46, 10, 111, 114 Brohltal 37
Aquincum 62, 68, 99, 187 Buciumi 65, 179, 182, 186
Ara Agrippinensium 40, 41 Budalia 26
Arcobadara/Arcobara 97, 105, 174, 179, Burnum 73
180, 186 Byzantium 37, 45
Arrisu Menander 58
Arucci 169 Cabyle 100
Asclepius 59, 61, 62, 73 Calarasi 109
Asia Minor 36, 41, 108, 111, 117, 118 Callatis 22, 33, 36, 45, 48, 51
Augusta Raurica/Rauricorum 116, 178, Capidava 99, 101, 107, 111, 112, 114–
179, 181, 187 16
Augusta Traiana 46, 104 Cappadocia 64
Augusta Treverorum 37, 42, 45, 178, Caracalla 40, 96, 103
219, 220, 235 Carausius 234
Augustodunum 219, 220, 227, 232 Carnuntum 178, 185, 187
Augustus 57, 61, 230, 233 Carsium 72, 73, 107, 111, 112, 114,
Aurelian 181, 184, 185 116, 118
Ausdecensium 25 Carthage 45
Avienus 227 Cassius Dio 23
Axiopolis 33, 46, 47 Certiae 182, 186
Axios 33 Cetate-Turnul Mare 110, 111
254 INDEX

Chilia 160 98, 102, 104, 107–11, 114, 116–19,

Cibalae 26 142, 154, 163, 173, 178–80, 182,
Cicero 66, 227 184–86, 196–98, 200–02, 206, 208–
Cilicia 26 10, 212, 222–24, 229, 231, 239–42
Cioroiu Nou 145, 146, 154 Dardania 26
Cius (Gârliciu) 22, 102 Darnithithi 96
Classis Flavia Moesica 67, 72–74, 76 Dasminium 51
Claudiopolis 26 Diana (Karataš) 74
Claudius 23, 27, 28, 62 Dimum (Belene) 98, 102, 103
Claudius II 184 Diocletian 194, 201, 209, 212, 220,
Claudius Claudianus (Claudian) 230, 225, 227, 229, 231, 233
243 Dinogetia 101
Claudius Mamertinus 220, 225, 227– Dobrudja 102, 107, 108, 210
29, 231, 232, 234, 236, 240, 242 Doliche 95, 105
client kings/states (Roman) 193, 195, Dolna Bešovica 14
203, 206, 212 Domitian 62
Cluj etc. see Napoca Drobeta 68, 69, 97, 162–63, 175, 181,
cohortes Aurelia Daradanorum 60, 63, 182, 185, 186
67, 68 Durostorum 72, 73, 99, 107, 111, 112,
coins 111, 173–88 114–18
Cologne 37, 40
Commodus 48, 64 Ebro 234
Constantine/Constantinian(s) 194, 200– Eburacum 39–41
03, 209, 213, 220, 221, 225, 227, 228, Egypt 117, 118, 213, 233, 237
230, 232, 235 Ennion of Sidon (glass) 113
Constantinople 212, 220, 229, 240 Euphrates 222–24, 226, 229, 231, 239
Constantius I 220, 227, 228, 230, 232,
234, 235, 238, 242 Flavia Solva 178, 185, 187
Costoboces 28 Forum Hadriani 47
Cotis (king) 7
Cotul Celicului 109 Galen 61, 62, 66, 71
Crito of Heraclea 61, 62 Gallia 47, 203, 220, 224, 235, 237
Cybele 163 Gallia Belgica 176, 178, 183, 188
Cyprus 111 Gallia Lugdunensis 39
Ganventa 37, 39, 40
Dacia(ns) 1, 27, 31, 33, 35, 42, 46, 47, Gârbou 69, 70
50, 52, 57–76 passim, 83–94 passim, Germania/Germany 35, 37, 45, 193–97,
95–105 passim, 110, 139, 143, 148, 222, 223, 226, 237, 239, 241
149, 154, 155, 159–69 passim, 173– Germania Inferior 40, 46, 64, 73,
87 passim, 197 176, 178, 183, 188
Dacia Malvensis 174 Germania Superior 33, 47, 50, 64,
Dacia Porolissensis 96 68, 176–78, 183–85, 187–88
Dacia Superior/Apulensis 96, 97, 149 and Rome 193–213 passim
Dacian War 60–63, 65, 75 Geryon 234
Dalmatia 73, 111, 114, 118, 223 Gherla 182
Danube/Donau/Hister/Ister 1–8, 26–28, glassware 107–37 passim
33, 47, 60, 62, 63, 65, 66, 73, 74, 97, gods see religion
 INDEX 255

Gordian III 61, 96, 103, 177, 179, 182, XXII Primigenia 48
183 Levant 108, 113, 118
Gorsium 187, 200 Libanius 224, 231, 241, 244
Grǎdişte see Sarmizegetusa Libya 232, 233, 237
Grammeni 96
Macedonia 96, 237
Hadrian 58, 67, 99, 165, 179 Macrianus 203–05
Halmyris 7, 101 Magnentius (usurper) 203
Hebros 33 Mainz see Mogontiacum
Hecate 139–55 passim Marbach am Nektar 48
Hercules 230, 234, 235 Marcianopolis 25, 104
Hermione (Achaia) 43, 45 Marcomannic Wars 64, 194, 197, 198
Herodian 59, 197 Marcus Aurelius 24, 25, 27, 29, 61, 67,
Hispania 237 242
Historia Augusta 197, 224 Marga 163
Histria/Istros 13, 33, 44, 97, 102, 103 Matrica 187
hospitals 57–76 passim Mauretania 237
Hygieia 59, 73 Mauretania Tingitana 68
Hyginus Gromaticus 59, 65 Maximian 220, 222, 225, 227–29, 231,
233–35, 241
Iaşi 110 medical care 57–76 passim
Ibida-Slava Rusa 114 Mesopotamia 117
Ilişua 65, 97 Micia 101, 163
Illyricum 237 Mihai Bravu 27
Intercisa 187 military forces/matters/veterans 15, 23,
Italy/Italia 46, 108, 109, 111, 114, 118, 27, 28, 38, 48, 57–76 passim, 83–94
237 passim, 95, 99, 103, 105, 108, 109,
194, 197, 198, and see legions
Judaea 27, 203 Misenum (fleet) 22
Jutland 199 Modestinus 66
Moesia 31–33, 35, 148
Kosmaj 63, 74 Moesia Inferior 1, 11–17 passim,
19–29 passim, 32, 33, 36, 37, 42,
Largiana 105 45, 46, 48–50, 57–76 passim,
Latinus Pacatus Drepanius 222 83–94 passim, 95, 98, 102, 107,
legions 109, 114, 117, 118, 148, 154, 155
I Italica 60, 67, 72, 73, 76, 108 Moesia Superior 33, 36, 42, 47, 50,
I Minervia 60, 73 51, 57–76 passim, 93, 155
I/II Adiutrix 23, 63 Mogontiacum/Mainz 41, 178, 187, 188,
IV Flavia 62, 63, 68, 76 196, 204, 205
V Macedonica 15, 27, 28, 38, 67, Montana 72, 73, 99–101
73, 86–88, 99, 108, 109
VII Claudia 63, 68, 76, 87, 103 Naissus 68
IX Claudia 67, 108 Napoca 139, 140, 174, 181, 186
X Gemina 63 Narona 103
XI Claudia 60, 72, 73, 76 Nehalennia (sanctuary) 37, 39, 40
XIII Gemina 70, 88, 95, 105 Neptun/Mangalia North 114
256 INDEX

Nero 23, 27, 61, 66 Raetia 64, 176, 178, 183, 187, 222, 237,
Nervii (territory of) 38 242
Nicomedia 33, 44, 45, 47 Ranisstorum 96
Nicopolis ad Istrum 19–22, 25, 45, 103 Ratiaria 26, 103, 104, 145
Nile 231, 233, 235 Ravenna (fleet) 20, 23, 101
Noricum 176, 178, 183, 185, 187, 237 Rǎzboieni 96
Novae 19, 21, 33, 51, 60, 61, 65, 69, religion/dedications etc. 14, 47–49, 59,
72, 75, 107, 110–12, 114, 116–18, 61, 62, 73, 110, 111, 115, 116, 139–
146, 147 55 passim, 159–70 passim, 230, 235,
Noviodunum 24, 74, 100, 101, 107, 238
110, 111, 115–18 Resculum 175, 186
Rhine/Rhineland 48, 112, 173, 194,
Ocean/Oceanus 227–29, 231–35, 237, 196–98, 203–05, 208, 222, 224–26,
239, 241 228, 229, 231, 233, 237, 241, 242
Odessos 15, 33, 49, 69, 116 Roman hegemony 193–213 passim,
Oescus 33, 107, 109–12, 114–16, 118 219–43 passim
Olbia 44 Rome, life in 45
Optatiana 105 Romula 145, 186
Orǎştie/Broos/Szászváros 139, 140 Rutilius Namatianus 230, 243
Ostrov 109, 110
Ovid 1–8 Salona 42
Samum (Cǎşei) 98, 163, 174, 182, 186
Padus (river) 223 Sarmatia 240, 241
Palmyrene 239 Sarmizegetusa/Grǎdişte/Várhely 46, 103,
Pannonia 196, 200, 201, 223, 225, 237 106, 139–55 passim, 163–66
Pannonia Inferior 26, 62, 64, 67, 68, Saxony-Anhalt 197
96, 176–78, 183–85, 187 Scandinavia (southern) 195, 197–99
Pannonia Superior 96, 176–78, 183– Schleswig-Holstein 199
85, 187 Scythia Minor 7, 200–02, 206
Pasewalk 198 Septimius Severus 28, 96, 101, 177
Persephone 139, 144, 148–50 Serdica 26, 104
Philip the Arab 97, 99, 100, 183 Severans 177, 178, 180
Philippopolis 25, 26, 104 Severus Alexander 101, 177
Pliny the Elder 233 Sexaginta Prista 100, 101
Poiana 110 Sidon 112, 113
Poland 194, 197 Singidunum 62, 63, 65, 67–70, 74
Pomerania (western) 198 Slovakai 194
Pompeii 110 Sopianae 187, 200
Pontus 45 Stara Zagora region 46
Porolissum(-Pomet) 46, 65, 164, 174, Stojnik 63, 64
175, 180, 182, 183, 185, 186 Strabo 108
Porticus Vipsania 233 Strassburg 203
Poşta 109 Sucidava 105, 145, 166–70, 185, 186
Potaissa 42, 146, 174, 180–83, 186 Suetonius 26, 57, 58
Priscus 206 Switzerland 111
Prudentius 230 Syria(-Palestine) 25, 64, 96, 113, 118,
Prusias ad Hypium 45, 50 226
 INDEX 257

Tacitus 23, 26, 61–63, 224, 228, 244 vici

Tarruntenus Paternus 58 Agatapara 25
Telerig 111, 114 Amlaidina 22
temples see religion Araba 25
Themistius 224, 243 Bres… 20
Theodosius 220 C[.]nisco 26
Thrace/Thracia 19, 20, 22, 23, 26, 27, Callosus 25
33, 46, 100, 103, 154, 201, 237 Cuetro 26
Thule 232 Dizerpera 20
Thuringia 197 Magaris 26
Tiber 242 Noviodunum 101
Tibiscum 101, 174, 186 Perdica 26
Tigris 223, 229, 231, 239, 241 Pereprus 25
Timacus Minus 68, 74 Rami… 22
Tium 41 Saprisara 45, 103
Tomis 1, 4, 8, 15, 33, 43–47, 49, 50, Statius 26
111, 112, 114, 116, 210 Thiuri… 22
trade routes/patterns 31–52 passim Titis 26
Trajan 58, 60, 61, 63, 65, 66, 75, 96, Trullensium 14, 16
100, 101, 165, 167, 179 Ulmetum 102
Transdierna 74 Vindenia 26
Troesmis 3, 72, 98, 101, 107, 110, 114– Vorouum Minor 22
18 Zinesdina Maior 21
Tropaeum Traiani 202 Viminacium 25, 33, 42, 47, 48, 51, 62,
Tulcea 109 63, 65, 67–71, 74, 103, 104
Tyras 44, 72, 73 Vindobona 58, 187
Vindolanda 58, 64, 70
Ulmetum 99, 102 Vipsanius Agrippa 233
Ulpia Traiana Sarmizegetusa see Sarmi- Virgil 230
zegetusa
Whipple’s Index 83, 84, 90–93
Valentinian 193, 203–06 Wiesbaden 188, 204
Valerian 181
Várhely see Sarmizegetusa Zosimus 224
Vegetius 60, 62 Zourobara 97
Vespasian 28, 58, 111, 179
 LIST OF CONTRIBUTORS

Marius Alexianu Lucretiu Mihailescu-Bîrliba

‘Alexandru Ioan Cuza’ University ‘Alexandru Ioan Cuza’ University
Faculty of History Faculty of History
Bd. Carol I, 11 Bd. Carol I 1
700506 Iași 700506 Iași
[email protected] [email protected]

Dan Aparaschivei Lucian Munteanu

Archaeological Institute Archaeological Institute
Romanian Academy, Iași branch Romanian Academy, Iași branch
Codrescu Str., no. 6, Pavilion H Codrescu Str., no. 6, Pavilion H
700479 Iași 700479 Iași
[email protected] [email protected]

Sever-Petru Boţan Ana Odochiciuc

Archaeological Institute ‘Alexandru Ioan Cuza’ University
Romanian Academy, Iași branch Faculty of History
Codrescu Str., no. 6, Pavilion H Bd. Carol I 1
700479 Iași 700506 Iași
[email protected] [email protected]

Roxana-Gabriela Curcă Annamária-Izabella Pázsint

‘Alexandru Ioan Cuza’ University Babeș-Bolyai University
Faculty of History Centre for Roman Studies
Bd. Carol I, 11 1 Mihail Kogălniceanu Street
700506 Iași Cluj-Napoca
[email protected] [email protected]

Ștefan Honcu Valentin Piftor

Archaeological Institute ‘Alexandru Ioan Cuza’ University
Romanian Academy, Iași branch Faculty of History
Codrescu Str., no. 6, Pavilion H Bd. Carol I 1
700479 Iași 700506 Iași
[email protected] [email protected]

Florin Matei-Popescu Ioan Piso

Institutul de Arheologie Babeș-Bolyai University
‘Vasile Pârvan’ Centre for Roman Studies
Str. Henri Coandă 11 1 Mihail Kogălniceanu Street
București Cluj-Napoca
[email protected] [email protected]
260 LIST OF CONTRIBUTORS

Alexander Rubel Rada Varga

Archaeological Institute Babeș-Bolyai University
Romanian Academy, Iași branch Centre for Roman Studies
Codrescu Str., no. 6, Pavilion H 1 Mihail Kogălniceanu Street
700479 Iași Cluj-Napoca
[email protected] [email protected]

Csaba Szabó Nelu Zugravu

Babeș-Bolyai University ‘Alexandru Ioan Cuza’ University
Centre for Roman Studies Faculty of History
1 Mihail Kogălniceanu Street Bd. Carol I 11
Cluj-Napoca 700506 Iași
[email protected] [email protected]
 COLLOQUIA ANTIQUA

1. G.R. TSETSKHLADZE (ed.), The Black Sea, Greece, Anatolia and Europe in
the First Millennium BC.
2. H. GENZ and D.P. MIELKE (eds.), Insights into Hittite History and Archae-
ology.
3. S.A. KOVALENKO, Sylloge Nummorum Graecorum. State Pushkin Museum
of Fine Arts. Coins of the Black Sea Region. Part I: Ancient Coins from
the Northern Black Sea Littoral.
4. A. HERMARY and G.R. TSETSKHLADZE (eds.), From the Pillars of Hercules
to the Footsteps of the Argonauts.
5. L. MIHAILESCU-BÎRLIBA, Ex Toto Orbe Romano: Immigration into Roman
Dacia. With Prosopographical Observations on the Population of Dacia.
6. P.-A. KREUZ, Die Grabreliefs aus dem Bosporanischen Reich.
7. F. DE ANGELIS (ed.), Regionalism and Globalism in Antiquity: Exploring
Their Limits.
8. A. AVRAM, Prosopographia Ponti Euxini Externa.
9. Y.N. YOUSSEF and S. MOAWAD (eds.), From Old Cairo to the New World.
Coptic Studies Presented to Gawdat Gabra on the Occasion of his Sixty-
Fifth Birthday.
10. R. ROLLINGER and K. SCHNEGG (eds.), Kulturkontakte in antiken Welten:
vom Denkmodell zum Fallbeispiel.
11. S.A. KOVALENKO, Sylloge Nummorum Graecorum. State Pushkin Museum
of Fine Arts. Coins of the Black Sea Region. Part II: Ancient Coins of the
Black Sea Littoral.
12. A.V. PODOSSINOV (ed.), The Periphery of the Classical World in Ancient
Geography and Cartography.
13. A.M. MADDEN, Corpus of Byzantine Church Mosaic Pavements from Israel
and the Palestinian Territories.
14. A. PETROVA, Funerary Reliefs from the West Pontic Area (6th–1st Centu-
ries BC).
15. A. FANTALKIN and O. TAL, Tell Qudadi: An Iron Age IIB Fortress on the
Central Mediterranean Coast of Israel (with References to Earlier and
Later Periods).
16. C.M. DRAYCOTT and M. STAMATOPOULOU (eds.), Dining and Death: Inter-
disciplinary Perspectives on the ‘Funerary Banquet’ in Ancient Art, Bur-
ial and Belief.
17. M.-P. DE HOZ, J.P. SÁNCHEZ HERNÁNDEZ and C. MOLINA VALERO (eds.),
Between Tarhuntas and Zeus Polieus: Cultural Crossroads in the Temples
and Cults of Graeco-Roman Anatolia.
262 AUTEURSNAAM

18. M. MANOLEDAKIS, G.R. TSETSKHLADZE and I. XYDOPOULOS (eds.), Essays

on the Archaeology and Ancient History of the Black Sea Littoral.
19. R.G. GÜRTEKIN DEMIR, H. CEVIZOĞLU, Y. POLAT and G. POLAT (eds.),
Archaic and Classical Western Anatolia: New Perspectives in Ceramic
Studies.
20. C. KÖRNER, Die zyprischen Königtümer im Schatten der Großreiche des
Vorderen Orients. Studien zu den zyprischen Monarchien vom 8. bis zum
4. Jh. v. Chr.
21. G.R. TSETSKHLADZE (ed.), Pessinus and Its Regional Setting. Volume 1.
22. G.R. TSETSKHLADZE (ed.), Pessinus and Its Regional Setting. Volume 2:
Work in 2009–2013.
23. I. MOGA, Religious Excitement in Ancient Anatolia. Cult and Devotional
Forms for Solar and Lunar Gods.
24. G.R. TSETSKHLADZE (ed.), Phrygia in Antiquity: From the Bronze Age to
the Byzantine Period.
25. L. MIHAILESCU-BÎRLIBA (ed.), Limes, Economy and Society in the Lower
Danubian Roman Provinces.
26. M. COSTANZI and M. DANA (eds.), Une autre façon d’être grec: inter-
actions et productions des Grecs en milieu colonial/Another Way of Being
Greek: Interactions and Cultural Innovations of the Greeks in a Colonial
Milieu.
27. G.R. TSETSKHLADZE (ed.), Ionians in the West and East.
28. G.R. TSETSKHLADZE et al. (eds.), Archaeology and History of Urartu
(Biainili).
29. M.-P. DE HOZ, J.L. GARCÍA ALONSO and L.A. GUICHARD ROMERO (eds.),
Greek Paideia and Local Traditions in the Graeco-Roman East.