You are on page 1of 444

N ETW ORKS AN D RELIGIOUS

IN N OVATION IN THE ROM AN EM P IRE

A. C. F. Collar

Submitted by Anna Caroline Folly Collar, to the University of Exeter as


a thesis for the degree of Doctor of Philosophy in Ancient History, July
2008.
This thesis is available for Library use on the understanding that it is
copyright material and that no quotation from the thesis may be
published without proper acknowledgement.
I certify that all material in this thesis which is not my own work has
been identified and that no material has previously been submitted and
approved for the award of a degree by this or any other University.

Abstract

Why do some religious movements succeed and spread, while others,


seemingly equally popular and successful at a certain time, ultimately
fail? It is from this starting point that this thesis approaches religious
success or failure in the Roman Empire: exploring a new analytical
method for understanding religious change: network theory. The thesis
forms two parts.
Part I sets out the theoretical frameworks. The focus of network
theory is on the processes by which innovation spreads: how
interconnectedness facilitates change. Although some innovations
might be superior, viewing success or failure as the result of interplay
between inherent qualities of a religious movement and the structure of
the social environment in which it is embedded means it is possible to
reduce value judgements about superiority or inferiority. The discussion
then turns to religious change. The key point is that sociologists of
religion can explain something of the processes of religious conversion
(or recruitment) and the success or failure of a religious movement
through an analysis of social interactions. Finally, I explain how I shall
use networks both as a heuristic approach and a practical modelling
technique to apply to the epigraphic data, and detail some of the
previous application of networks to archaeological test cases.
Part II applies these methods to the epigraphic data of three
religions. In Chapter Four, I examine the cult Jupiter Dolichenus,
arguing that the previous explanations for the success of the cult are
untenable, showing from the epigraphy that the cult spread through a
strong-tie network of Roman military officials. In Chapter Five, I look at
the development of Jewish identity in the Diaspora, showing that,
during the second century AD, Diaspora Jews began to actively display
their Jewish identity in their epitaphs. I argue that this re-Judaization
ii

represents the activation of an ethno-cultural network, as a response


to the destruction of the Temple in Jerusalem and the crushing of the
Bar Kokhba rebellion; the visible remains of the rabbinic reforms. In
Chapter Six, I discuss the cult of the Highest God, Theos Hypsistos,
taking Mitchells argument further to suggest that the huge increase in
the dedications during the second-third centuries is not simply a
reflection of the epigraphic habit, but rather, that the cult of Hypsistos
was swelled by the Gentile god-fearers, as a result of the changes
happening within Judaism itself at this time.

iii

Contents

Title, with declaration

Abstract

ii

Contents

iii

acknowledgements

iv

List of maps

Abbreviations

vi

Introduction

1.

Theoretical Framework I: Evolution, Networks, Innovation

2.

Theoretical Framework II: Sociology of Religion. Religious


Innovation & Conversion

30

3.

Methodology: Networks in Archaeology. Analytical Models

52

4.

The Cult of Jupiter Dolichenus.

70

Military networks on the edges of Empire


5.

The Jewish Diaspora.


Ethnic networks and the activation of Jewish identity

6.

The Cult of the Most High God.


God-fearers and the redefinition of the Jewish-Gentile
relationship

13
1
18
8

Epilogue

24
6

APPENDIX: CONCORDANCE OF EPIGRAPHIC MATERIAL

24
8

BIBLIOGRAPHY

32
7

iv

Acknowledgements

I have many people to thank here. Primary of these is my supervisor


Stephen Mitchell, whose intellectual generosity and vision has made
this thesis possible. He has consistently allowed me the freedom to
pursue new lines of thought, and offered his lucid criticism, careful
guidance and unstinting support. I cannot thank him enough. My ideas
have also been fundamentally shaped by the innovative work of Carl
Knappett, whose incisive and intelligent discussion has always
demanded more of me. Without his influence I should be a laggard
indeed. My thanks also go to Elena Isayev and Martin Millett, for such a
stimulating and enjoyable viva.
A number of institutions have generously supported my research,
both financially and in kind: The AHRC, The British Institute at Ankara,
The British School at Athens, the Kluge Center at the Library of
Congress and the Forschungsstelle Asia Minor in Mnster. My thanks
go to Elmar Schwertheim and Engelbert Winter, who welcomed me to
the Forschungsstelle, and who incorporated me into the Dlk Baba
Tepesi excavation team. In my third year I was fortunate enough to be a
British Council fellow at the Kluge Center, the staff of which I thank for
their great hospitality. I am indebted to Irad Malkin for pushing for the
inclusion of my networks paper in the first volume of the special edition
of the Mediterranean Historical Review. I am grateful to Tom Elliott at
the Ancient World Mapping Center for the use of the Barrington Atlas.
The chapter on Jupiter Dolichenus has been greatly improved through
conversations with and corrections by Michael Blmer, my thanks to
him for sharing with me his considerable knowledge. Leif Isaksen
helped develop my views on networks; my thoughts on innovation and
evolution have benefited from conversations with physicists and
philosophers: Jake Snow, who has given me dense science books, and
far more besides; Harry Platanakis, who has offered me
encouragement, and interested and demanding debate; and Jonathan
Davies, who has pushed me in the last few months to think more deeply
about evolution and emergence. I am also extremely grateful to Pete
Lockley for proof reading this for me.
During my second year at Manchester University, I was privileged
enough to be taught by Geoff Stone, who rekindled my love of the
ancient world and who has supported my efforts ever since, to whom I
offer my warmest thanks. Jessica Bogo, Stefanie Reetz and Elsa Suckle
were great friends to me during my time in America, and Amie
Brautigam, and Jrgen, Jasper and Zo Thomsen provided a muchneeded surrogate family. Tristan Carter made my time at Ankara much
more fun; and Michael Scott, Amalia Kakissis and Jill Hillditch did the
same for me in Athens, as well as back in Cambridge and Exeter.
v

Working with the Classics and Ancient History staff and postgraduates
at Exeter has been a pleasure, my thanks to all of them, especially to
my office companions of the last year, Pauline Hanesworth, Gillian
Ramsey and Anna Blurtsyan, who have all offered helpful suggestions
and encouragement. My time at Exeter would have been far less
without Cassie Hague, Susannah Cornwall, Bee Taylor, Sandra
Lohmann, Sarah Lorimer and especially, my blood, Rebecca Catto. Their
support, brilliance and humour have kept me sane. Pete Vardon has
punctuated this year with adventure and misadventure, and has put up
with me in between. Finally, my enormous love and thanks go to my
family, for all their emotional, intellectual and financial support: my
mum, Alison Carter, for her patience, humour and unfailing belief in
me; my inspirational and hilarious sisters Beth and Clio; and my
wonderful and encouraging grandparents, Eileen and Leslie Carter.
Lastly, I thank my dad, Nigel Collar, for goading, challenge, and
argument, and for always reminding me of the value of knowledge for
its own sake. This is for him.

vi

List of Maps

Chapter 4: the cult of Jupiter Dolichenus


Map 4a Distribution

12
5

Map 4b initial proximal point analysis

12
6

Map 4c ppa: First century BC-AD 150

12
7

Map 4d PPa: AD 150-211

12
8

Map 4e PPA: AD 212-253

12
9

Map 4f PPA: AD 253-300

13
0

Chapter 5: the Jewish Diaspora


Map 5a testified movements across the Diaspora

18
1

Map 5b distribution

18
2

Map 5c initial proximal point analysis

18
3

Map 5d PPA: first and second centuries AD

18
4

Map 5e PPA: third century AD

18
5

Map 5f PPA: fourth century AD

18
6

Map 5g ppa: fifth-sixth centuries AD

18
7

Chapter 6: the cult of the most high god


vii

Map 6a distribution

24
1

Map 6b initial proximal point analysis

24
2

Map 6c PPA: Hellenistic to the first century BC

24
3

Map 6d ppa: The first century AD

24
4

Map 6e PPA: The second-third centuries

24
5

viii

Abbreviations

Epigraphic Corpora
CCID
CIJ
CIL
IJO 1-3

JIGRE
JIWE
SEG
TAM

M. Hrig, E. Schwertheim, Corpus Cultus Iovi Dolicheni


J. -B. Frey, Corpus Inscriptionum Judaicarum
Corpus Inscriptionum Latinarum
D. Noy, A. Panayotov, and H. Bloedhorn; W. Ameling; D.
Noy and H. Bloedhorn; Inscriptiones Judaicae Orientis
vols. I-III
W. Norbury and D. Noy, Jewish Inscriptions of GraecoRoman Egypt
D. Noy, Jewish Inscriptions of Western Europe I
Supplementum Epigraphicum Graecum
Tituli Asia Minoris

Other abbreviations
AA
AJA
AJP
AJS
AJSLL
APA
ASR
AW
BA
BMCR
CJZC
CQ
DM

American Antiquity
American Journal of Archaeology
American Journal of Philology
American Journal of Sociology
The American Journal of Semitic Languages and
Literatures
Transactions and Proceedings of the American
Philological Association
American Sociological Review
Antike Welt
Biblical Archaeologist
Bryn Mawr Classical Review
Corpus jdischer Zeugnisse aus der Cyrenaika
The Classical Quarterly
Damaszener Mitteilungen
ix

EA
HSCP
HThR
IM
JAOS
JECS
JJS
JPE
JQR
JRA
JRS
JSJ
JSS
JSSR
MA
MHR
NRSV
PNAS
RMP
SIAM
TLS
WA
ZPE

Epigraphica Anatolica
Harvard Studies in Classical Philology
Harvard Theological Review
Istanbuller Mitteilungen
Journal of the American Oriental Society
Journal of Early Christian Studies
Journal of Jewish Studies
Journal of Political Economy
The Jewish Quarterly Review
Journal of Roman Archaeology
Journal of Roman Studies
Journal for the Study of Judaism
Jewish Social Studies
Journal for the Scientific Study of Religion
Mediterraneaneo Antico
Mediterranean Historical Review
New Revised Standard Version
Proceedings of the National Academy of Sciences of the
United States of America
Reviews of Modern Physics
Society for Industrial and Applied Mathematics
Times Literary Supplement
World Archaeology
Zeitschrift fr Papyrologie und Epigraphik

Introduction

Why do some religious movements succeed and spread, while others,


seemingly equally popular and successful at a certain time, ultimately
fail? It is from this starting point that this thesis approaches the
question of religious success or failure in the Roman Empire. It
presents and explores the application of the analytical method of
network theory for understanding religious change. Instead of focusing
on the intrinsic qualities of a religious innovation to explain its
success, this approach highlights the role that the interconnectivity of
the network plays in driving success or failure. By considering the
connectivity of the network, it is possible to explain why some
ostensibly valuable innovations are unsuccessful.
The chronological framework extends from the Hellenistic to the
early Byzantine era, but the focus of the thesis is on the evidence from
the Imperial period, the first to third centuries AD, partly because the
majority of the epigraphic evidence dates from this period. By this time,
the land encompassed by Imperial rule stretched from Spain to the
Crimea,

and

from

Scotland

to

southern

Egypt.

Under

this

administration lived a huge variety of people, speaking any number of


languages or dialects, and worshipping any number of local or global
deities. The Roman government of this plethora of social, linguistic and
religious forms provides a veneer of similarity, of a new lingua franca,
of a globalised environment, but how genuine was this? Even if the
sameness was superficial, the trappings of Roman administration and
the necessity of defending borders entailed a number of universalising
features being imposed on the landscapes and the people. Most
1

physically present were the roads and bridges that were built to
facilitate the movement of the military, the accoutrements of the
military,

and

the

long-distance

communication

that

the

Empire

required. Alongside these physical additions to the landscape came the


ideological reminders of participation in Empire the statues and
monuments to Imperial rule, bathhouses, temples, and the presence,
along the limes, of the soldiers themselves. Latin as well as Greek
became a pan-Empire language, used for documents (albeit more
rarely)

even

to

the

borders

with

Parthia,

brought

with

the

administration of the provincial governors or with the soldiers of the


legions. But for all these unifying factors within the Roman Empire, the
varieties of local identities below the surface did not disappear. There is
evidence for resistance to the Romans and to Romanisation, both
active (for example, the revolting province of Judaea) and passive (for
example, non-adoption of Roman culture). Equally there is evidence
for the eager adoption of Roman practices or names, as markers of
status or political allegiance.
Religious

movements

spread

across

this

environment

with

differing degrees of success. This thesis will investigate three kinds of


network that were operative in the Roman world and the roles they had
in the dissemination of new religious information: military, ethnic and
cultic. Each of these networks has different configurations and ways of
spreading information.

The thesis essentially forms two parts. The first sets out the theoretical
frameworks that are being drawn on, questioned, and developed. The
second consists of an exploration of three case studies from the ancient
world and examines bodies of epigraphic evidence that show the
operation of three kinds of network military, ethnic, and religious.

Chapter 1 sets out the theoretical framework for understanding


change, and the various approaches to the study of innovation and the
transmission of innovation. The framework is multidisciplinary: drawing
on theory from biology, physics, mathematics, anthropology, sociology
and archaeology. The chapter begins with an examination of the
dominant conceptual paradigm for understanding change: evolution by
natural selection. This paradigm can be usefully applied to cultural
change, and explain the emergence of novel cultural forms. The main
focus of the discussion, however, is on the process by which innovation
spreads, using network theory as understood from literature in physics,
sociology,

and

anthropology

to

explain

the

properties

of

interconnectedness that facilitate change. By assessing success or


failure as the result of interplay between the inherent qualities of a
religious movement and the structure of the social environment in
which it is embedded it is possible to reduce subjective value
judgements about the superiority or inferiority of a religious innovation.
Because innovations can both be alien introductions and arise
naturally from within an environment, considering the connectivity of
the network means it is possible to explain why some ostensibly
valuable innovations are sometimes unsuccessful. Central to these
explanations of social change is the theory of the diffusion of
innovations. This body of theory is mainly drawn from sociology, which
discusses vulnerability and status in the process and transmission of
innovations; and from physics, which takes these observations and
quantifies

them

mathematically,

drawing

conclusions

about

the

boundaries and connectedness of networks that can result in the


phenomenon of information cascade.
Chapter 2 sets out the second theoretical framework being used
in the thesis. Using theory from the sociology of religion, this chapter
moves the discussion of change specifically back into the sphere of
religion, and investigates the processes of innovation, adoption and
diffusion as they are understood in sociological terms. The sociological
3

research into the processes of conversion to modern religious


movements forms a counterpoint to the theories of the previous
chapter, and is complementary to network theory. The central point is
that sociologists of religion can go some way to explaining the
processes of religious conversion (or recruitment) and the success or
failure of a religious movement through an analysis of

social

interactions. This does not undermine the importance of the values of


the religious movement, but suggests that an analysis combining an
understanding of the ideology and the social system can present a fuller
picture of the spread, success or failure of religious innovations.
Chapter 3 sets out the network methods I will use in my approach
to ancient religious data, both as a heuristic way of approaching and as
a practical modelling technique to apply to the epigraphic data. I then
outline some previous applications of the network approach to
archaeological test cases that have informed my study. Analytical
methods that are examined are Proximal Point Analysis as a model for
settlement growth and centrality in the Early Bronze Age Cyclades by
Cyprian Broodbank; the interrelationship of dispersion-colonisation
networks during the emergence of Greek identity in the Archaic
period, as proposed by Irad Malkin; the more scientifically driven
utilisation

of complexity

theory and power-laws to explain the

emergence of power and hierarchy in the Neolithic as detailed by R.


Alexander Bentley; and the collaborative work of Carl Knappett with
physicists Tim Evans and Ray Rivers for understanding the interactions
of sites and the costs of maintenance in the Aegean Bronze Age.

I then move on to apply these methods to the ancient evidence.


My first case study in chapter 4 examines the cult of Jupiter Dolichenus.
The god was originally a local Near Eastern storm god from the region
of Doliche in Commagene, but like other Oriental religions, during the
middle Imperial period, the cult spread along the Danube and Rhine
valleys, the German limes and along Hadrians Wall. The previous
explanations for the success of the cult the participation of the Roman
legions in Eastern campaigns; the presence of Syrian recruits within
the structure of the army; and zealous Syrian traders are all argued to
be untenable. By contrast, it is shown from the epigraphy that the cult
spread through a highly connected strong-tie network of Roman
military officials. This is supported by a network analysis, and a
network explanation for the end of the cult of Dolichenus is proposed.
My second case study in chapter 5 looks at the material
pertaining to the Jewish Diaspora from the Hellenistic period to late
antiquity. During the second century AD the Jews of the Graeco-Roman
Diaspora became very concerned with the active display of their Jewish
identity in their epitaphs. It is argued that this represents the
activation of the ethnic/ethno-cultural network of the Jewish Diaspora,
as a response to the destruction of the Temple in Jerusalem in AD 70
and the crushing of the Bar Kokhba rebellion in Judaea in AD 135.
Although the epigraphic habit is partly responsible for the increase in
inscriptions at this time, the marked difference between the epigraphy
from before and after the fall of the Temple is striking, and it is
suggested that this represents a re-Judaization of the Diaspora, to be
understood as the visible remains of the rabbinic reforms enacted in
Judaism at this time. Finally, it is suggested that the network can be
described as having been vulnerable to the religious innovations of the
rabbis, an observation which may in turn shed light on the parallel
success of Christianity in this period.
My final case study in chapter 6 is related to the Jewish Diaspora,
but is concerned with the cult of the Highest God, Theos Hypsistos. As
5

the title is the translation of the Hebrew name for God, El Elyon, the
cult has been associated with Judaism. There is a huge increase in
dedications to the cult during the second-third centuries AD, but the
inscriptions relating to the cult are often extremely brief. Instead of
focusing on this rather limited information contained in the inscriptions,
we can understand the cult as a totality from the distribution pattern of
its evidence, and show that it did not have a long-range diffusive
appeal. Mitchell made the connection of the cult with the god-fearers
theosebeis associated with the Jewish synagogues. This group of
people are often referred to in the literature, but appear quite
infrequently in the epigraphy in comparison with the cult of Theos
Hypsistos. If the epigraphic habit can be held solely responsible for the
increase in dedications to Theos Hypsistos, then why do the dedications
of the god-fearers not also increase at this time? I here take Mitchells
argument further, and suggest that the huge increase in the dedications
is not simply a reflection of the epigraphic habit, but rather that the cult
of Hypsistos was swelled by the god-fearers, as a result of the changes
happening within Judaism itself.
The use of networks and network theory offers a new and
rewarding method of analysing ancient historical subjects of innovation
and the spread of ideas. Although there are certainly some issues with
the approach that need further theorising and further application, I
hope that this thesis will demonstrate the potential and value of the
methodology.

Part i
theory

Chapte r 1.
Theoretical Framework. 1
Evolution, Networks, Innovation

Introduction
This chapter sets out several approaches to the study of innovation and
the transmission of innovation, and provides the theoretical framework
that will then be used to approach the spread of religious innovations in
the Roman Empire. The theories that will be examined here are drawn
from

number

of

disciplines:

biology,

physics,

mathematics,

anthropology, sociology and archaeology. These have been used to


explain the transmission of a wide range of data: genes, cultures,
information, technologies, diseases and ideas. Although the range of
starting data and disciplines

means

there

are some important

differences between these approaches, the core ideas about the


transmission of innovations have much in common.
The

point

of

commonality

between

all

these

theoretical

approaches is the importance of the environment the connectivity of


the network in determining the success or failure of an innovation.
This perspective is not simply reliant on a progressive view of the
intrinsic superiority of the innovation itself. Although some innovations
might ostensibly be more valuable than others, subjective and
historically determined value judgements about the superiority or
inferiority of an innovation are replaced by viewing success or failure as
the result of the interplay between the inherent value of an innovation
1

A shortened version of this chapter has been published in the Mediterranean


Historical Review, vol. 22, no. 1, June 2007, pp. 149-162, entitled Network Theory
and Religious Innovation; and in a Routledge collected volume (forthcoming).

and the environmental structure to which it belongs. Innovations can


either be alien introductions or arise naturally from within an
environment. In either case it is possible for the connectivity of the
network to explain why some innovations that seem ostensibly valuable
are sometimes unsuccessful, while others make headway.
Seen from this perspective, social, cultural, and religious change,
whether gradual or sudden, can be viewed as a process resulting from
the network interconnections. This network can be actively manipulated
as well as passively receptive to the introduction of an innovation, but
the most important aspect is the nature and connectivity of the network
itself, and its ability to transmit information.
This chapter begins by locating the grounds of the argument
within archaeological research, the role of evolution as the paradigm
for understanding innovation, and change in both biological and
cultural terms. Having argued that evolution by natural selection can
go some way to explaining the emergence of novel cultural forms, the
chapter then discusses the process by which innovation spreads, using
theories

that

have

been

developed

in

physics,

sociology,

and

anthropology to explain the properties of interconnectedness that


facilitate change. Central to these explanations of social change is the
theory of the diffusion of innovations. This body of theory is mainly
drawn from sociology, which has introduced the notions of vulnerability
and status in the process and transmission of innovations, and from
physics, which quantifies them mathematically, drawing conclusions
about the boundaries and connectedness of networks that can result in
the phenomenon of information cascade.
The chapter concludes by examining some of the issues that arise
from mathematical network theory and its applicability to social
situations, including the vertical/horizontal divide in the transmission
and adoption of new religious ideas.

Cultural Change, Evolution, and Innovation

Archaeology, by its very nature, takes a long view, recording continuity


and change in the material record of human activity. Continuity or
change can be macro-scale: for example, the transition from huntergatherer existence to that of settled agriculture; meso-scale: for
example, the adoption of the potters wheel; or micro-scale: for
example, the developments in the burial practices of a particular
community. The extent of continuity or change is analysed by
constructing and comparing typologies of material culture and these
typologies can span great periods of time. However, chronological
reconstruction, as a historical approach, tends to minimize variability in
the material evidence and attempts to build seamless narratives.
James McGlade and Sander Van der Leeuw suggest that this is because
archaeology sets out to reconstruct long-term history and has been
conditioned to pursue coherence and similarity as the mainsprings of
classification: disorder, discontinuity and difference have no place in
this scheme; they are aberrant categories which must be underplayed
or judiciously edited out of the interpretive and explanatory discourse. 2
This is partly a result of a residual philosophical positivism in the
discipline. The dominance of the conception of history as a strict linear
process is as much a part of ancient history as it is part of archaeology.
McGlade and Van der Leeuw have criticised the search in archaeology
for sequential narratives, and advocate instead a diffuse, dissipative
structures approach within an understanding of evolutionary change, 3
an approach that can be usefully applied to ancient history.
Understanding why change happens requires an understanding of
the processes by which changes are wrought: from innovation to
2

Van der Leeuw, S. E., and McGlade, J., eds., Time, Process, and Structured
Transformation in Archaeology, London: Routledge, 1997, p. 5.
3
Van der Leeuw and McGlade, Structured Transformation, p. 11.

10

dominance. Whats New?, Van der Leeuw and Robin Torrences


collection of papers on innovation4 begins with a challenge to the view
current in archaeology during the 1980s, where stasis had come to be
regarded as the norm, and change as an aberration: even when
archaeologists do question why something changed, the answer is
usually sought in terms of the need for change or the benefit and
value of the new behaviour, with much less regard for the costs. 5 The
approaches published in the volume built on sociological studies that
sought to understand innovation and adoption not as a single event
which has a simple cause and effect, but rather as a complex, ongoing
process of the production of cultural variety, working from conception
to legitimisation. Intrinsic to this is the study of the innovations that did
not become dominant, or which happened only on a small scale
understanding the reasons that lie behind the refusal to implement a
new technology provides important insights into the way economic,
social, and ideological factors constrain and determine behaviour. 6

Evolution: a paradigm for change


The dominant paradigm for understanding change is that of evolution.
There are certain value-laden problems inherent in applying the
commonly understood notion of Darwinian evolution by natural
selection to cultural systems and society. The survival of the fittest
means that the ways of living and interacting that have been eradicated
by other cultural norms should somehow be considered inferior, or at
least, have failed because they were unable to adapt they were not
fit enough. A fuller discussion of the nature of the issues involved will
be beneficial here, because the concept is such a powerful one, and so
engrained in Western thought processes.
4

Van der Leeuw, S. E., and Torrence, R., eds., Whats New? A closer look at the
process of innovation, London: Unwin Hyman, 1989.
5
Van der Leeuw, Torrence, Whats New?, p. 2
6
Van der Leeuw, Torrence, Whats New?, p. 5.

11

The structure of an organism is fully contained within its genetic


code, its DNA, of which the organism inherits a hybrid of its parents at
conception, and is programmed to develop in particular ways to create
a

replica

of

its

own

particular

species.

Individual

ontogenetic

development is subject to environmental pressures, as, for example, in


the water plant, arrowroot, which changes leaf shape depending on the
depth of water it germinates in.7 The reconfiguration of parent genes in
the offspring of every generation ontogenesis is not the same as the
evolution of a species phylogenesis although new beneficial traits in
the offspring may lead to the reproduction of those traits.
Darwins account of evolution, put simply, was the separation of
causes of ontogenetic variation, as coming from internal factors, and
causes of phylogenetic variation, as being imposed from the external
environment by way of natural selection. 8 It has been generally taken
as

standard

evolutionary

thinking

that

biological

innovations

(essentially, randomly occurring genetic mutations genetic drift)


internal to the organism are only successful if they favour survival or
reproductive success in a particular environment, i.e. that the organism
adapts to that environment. Species variation therefore results from the
best fit of these random biological innovations into differing, preexistent environments.
An implication of adaptationist evolutionary thinking is that the
situations now dominant are those that are the fittest for their
particular environment.9 The theoretical physicist Peter Allen has
7

See Wilson, E. O., Consilience: The Unity of Knowledge, New York: Knopf, 1998, p.
137. For similar examples see also Maynard Smith, J., Shaping Life: Genes, Embryos
and Evolution, New Haven: Yale University Press, 1999.
8
Lewontin, R. C., Gene, Organism and Environment, in Cycles of Contingency:
Developmental Systems and Evolution, Oyama, S., Griffiths, R. D., and Gray, P. E.,
(eds.), Cambridge, Mass: MIT Press, 2001, p. 59.
9
There is not space to discuss this further here, but there is a discussion to be had
about whether natural selection is still appropriate to describe the situation of many
species teetering on the brink of extinction. Active hunting to extinction of species
over the past 300 years as a product of human global expansion; the present
anomalous dominance of our own species, and the transformation or destruction of
various environments by humans is actively obstructing the ability of other organisms
that share our environment to evolve. Can it really be argued that these animals,
birds, fish, plants, and human societies that are now extinct simply did not adapt
quickly enough to their changing environment?

12

modelled the process, which requires a level of error-making in the


evolutionary process and postulated that the system operates at a suboptimal level. He argues that because there exists apparently random
or highly eccentric behaviour, which at that time is meaningless, and on
average loss-making. However, in order to maintain adaptivity to the
environment, some stochastic, risk-taking behaviour is retained by
evolution. In short, then, evolution is both driven by, and leads to
microscopic diversity and individual variability. 10 The system, (if it can
be so defined) although less than optimally efficient, apparently runs at
the best level to maintain diversity.
Gould and Lewontin have spearheaded the critique of this
adaptationist notion of evolution. They argue that it is misleading for
evolutionary biology to attempt to explain all features of diversity as
explicitly

resulting

from

the

adaptive

process.

Analogously,

the

spandrels of San Marco, so elaborate, harmonious and purposeful, are


not claimed to be the cause of the surrounding architecture. 11 This is
because the adaptationist notion of evolution understands speciation, or
phylogenesis, through the metaphor of problem and solution the
environment poses the problem; the organisms posit solutions, of
which the best is finally chosen. The organism proposes; the
environment

disposes.12 Lewontins

counter-proposal argues that

instead of evolution being viewed as an essentially adaptive process,


whereby organisms mutate, develop and survive in response to their
environment, it is better conceptualised as a process of construction.
Genes, organisms and environments are in reciprocal interaction with
each other in such a way that each is both cause and effect. 13
Environments and ecological niches that are so perfectly suited to
10

Allen, P. M., Modelling innovation and change, in Van der Leeuw, S. E., and R.
Torrence, eds., What's New?: A Closer Look at the Process of Innovation, One World
Archaeology, London: Unwin Hyman, 1989, p. 271.
11
Gould, S. J., and Lewontin, R. C., The Spandrels of San Marco and the Panglossian
Paradigm: A Critique of the Adaptationist Programme, in Proceedings of the Royal
Society of London. Series B, Biological Sciences, Vol. 205, no. 1161, The Evolution of
Adaptation by Natural Selection, 1979, pp. 581-598.
12
Lewontin, Gene, p. 60.
13
Lewontin, Gene, p. 61.

13

every variant of biological life do not simply exist extrinsically from the
organisms that inhabit them: they are created by the animal themselves
as a consequence of their own life activities. 14 The environment of an
organism is determined by what that organism finds relevant:
organisms do not adapt to their environments; they construct them out
of the bits and pieces of the external world.15 In this conception of
evolution by natural selection, change is not simply a response to
external, uncontrollable factors. The oppressive notion of best fit is
sidelined: evolution becomes a mutually constructive dialectic between
organism and its environment.

Evolution and cultural change


These approaches and critiques in biology are relevant here, because
the paradigm of natural selection has been frequently used in
discussion and analysis of the processes of cultural change. Boyd and
Richerson have developed the view that culture is a part of biology,
defining culture as the information affecting [their] phenotype acquired
by individuals by imitation or teaching. 16 Culture is seen as an
inherited system, and cultural variation is then the result of either
errors of repetition (as in random genetic mutation), active imitation (a
process that cannot exist in genetic evolution); or systematic change,
including guided variation (where an inherited system is modified by an
individual in light of their own experience), and bias (either direct,
because the individual has a number of options to choose from; or
indirect, where a choice is made because it links to another individual
who is being used as a model for something else).
Boyd and Richerson conclude that both guided variation and
direct bias are often relatively weak factors in explaining change, and
14

Lewontin, Gene, p. 64.


Lewontin, Gene, p. 64.
16
Boyd, R., and Richerson, P. J., Culture and the Evolutionary Process, Chicago:
University of Chicago Press, 1985, p. 283.
15

14

that imitation rather than learning is responsible for much change in


cultural traits. They argue that indirect bias is largely responsible for
propagating cultural traits, in a similar way to natural selection it
seems likely that individuals characterized by some cultural variants
will be more likely to survive or attain social positions that cause them
to be imitated than individuals characterized by other variants. When
this is true, natural selection will increase the frequency of those
variants.17 Observation and imitation of successful individuals can lead
to these characteristics being locked-in and propagated, leading to
diversity and runaway traits similar to that of sexual selection in
animals. This echoes the wave-of-advance model, describing the spread
of advantageous genes through a population, where any trait that exists
alongside the advantageous one is carried along with it, regardless of
intrinsic superiority: described as cultural hitchhiking.18
Shennan points out that biologists and evolutionary theorists have
been concerned with how cultural traditions are handed on and
modified through time, in other words, with the conditioning of people
into a cultural environment through learning. In a similar way, social
theorists such as Bourdieu and Giddens have argued for the creation of
a habitus, of an environment as the daily production and reproduction
of the institutionalized practices of their society. 19 Shennan concludes
that a combination of Boyd and Richersons emphasis on cultural
transmission with Giddens aim of explaining how societies persist
across time and space provides a justification for the long-standing
archaeological assumption that it is change and not stability which
needs explaining.20
17

Boyd and Richerson, Culture, p. 285.


Ackland, G. J., Signitzer, M., Stratford, K., Cohen, M. H., Cultural hitchhiking on the
wave of advance of beneficial technologies, in PNAS, vol. 104, no. 21, 2007, pp. 87148719.
19
Shennan, S., Cultural Transmission and Cultural Change, in Van der Leeuw, S. E.,
and R. Torrence, eds., What's New?: A Closer Look at the Process of Innovation, One
World Archaeology, London: Unwin Hyman, 1989, p. 332, using Bourdieu, P., Outline
of a Theory of Practice, Cambridge: Cambridge University Press, 1977, and Giddens,
A., The Constitution of Society, Cambridge: Polity Press, 1984.
20
Shennan, Transmission, p. 339.
18

15

In archaeology in particular, material culture operates precisely


as

an

endogenously

generated

transmissible

environment

[]

channelling future decisions in particular directions and acting as a


source of cultural transmission, in addition to, and conditioning, the
imitation and teaching/learning processes.21 This view of the evolution
of culture and ideas correlates usefully with the constructionist
argument in biology: that culture and environment are locked into a
mutually creative process.22

Understanding the processes of change


Theories of cultural evolution provide a link between micro-scale action
and macro-scale social properties. Shennan, summarising the argument
of Harre,23 writes that for social change to occur, small-scale changes
must spread through populations and a Darwinist perspective provides
an appropriate way of viewing this. It is the processes by which social
change occurs the spread of the small-scale changes through
populations that is the focus of this thesis.
The theory of emergence in the natural and physical sciences is
also concerned with this link between micro and macro: but the
emphasis of the theory is on the process by which this occurs. The
mainstay of emergence is that complex phenomena result from
relatively simple inputs. Emergent behaviour basically describes the
self-organising properties of systems, and the resulting collective
action.
21

Shennan, Transmission, p. 342.


It should be noted that Richard Dawkins coined the term meme to describe the
way that ideas propagate via human consciousness, likening the process to that of
genetic propagation through organisms over generations. The theory has come under
considerable attack from various angles, and will not be discussed here because the
emphasis of the theory is on the memetic value of the idea itself, rather than on the
process by which it is transmitted. Dawkins, R., The Selfish Gene, Oxford: Oxford
University Press, 1976, p. 189-201.
23
Shennan, Transmission, p. 342, using Harre, R., Social Being, Oxford: Basil
Blackwell, 1979.
22

16

It is a given that any innovation must both be determined by and


possess an inherent value within the context in which it was developed.
The structure of the environment determines and limits both the nature
of the innovation itself and how profoundly the innovation will spread. It
is the inherited, vertical aspect of the culture that determines the
nature of an innovation, and it is the environment, or the connectivity of
the network, the horizontal aspect, that determines that innovations
propagation.

Innovation, Change and Network Theory

Understanding how the structure of the environment in which an


innovation happens can determine the profundity of its transmission is
key to understanding why some innovations are successful, and others
fail. By conceiving an environment as a network, a space characterised
by interconnected nodes, whose interconnections have the ability to
affect each other, it is possible to analyse success or failure without sole
recourse to a value judgement of the nature of the innovation itself.
There is a vast body of literature on network theory, especially in
physics and sociology. In sociology, network has been used to describe
the structure of a society through abstracting from the concrete
population and its behaviour the pattern or network (or system) of
relationships between actors in their capacity of playing roles relative
to one another.24 The term is developed to include the further linkage
of the links themselves and the important consequence that, what
happens so-to-speak between one pair of knots, must affect what
happens between other adjacent ones. 25 It is the change in perspective
24

Nadel, S. F., The Theory of Social Structure, London: Cohen & West Ltd, 1957 p. 12,
quoting Parsons, T., Essays in Sociological Theory, pure and applied, Glencoe, Ill: Free
Press, 1949, p. 34.
25
Nadel, Social Structure, p. 16.

17

afforded by studying interconnectedness that gives the theory its


weight: how and why information travels across the network is a result
of the connections, not just of the individual nodes.
Advances in computing technology and the potential of in silico
modelling have to a large extent driven the rapid development of
network theory by mathematicians and physicists, who have applied the
principles to a huge range of subjects, both social and non-social. What
follows here is an overview of the key mathematical ideas: phase
transition the sudden dramatic shift from one network state to
another; the small-world network, made up of clusters interlocking
with each other through weak and strong ties; the short path-length
property of the small-world network which allows any two individuals to
connect relatively easily; and power laws, where certain nodes are
hubs in the network. The sociological theory of the diffusion of
innovations and the relationship of individual thresholds with regard to
innovation are shown to be central to the process of information
cascade: the success (or failure) of a particular innovation.

Phase transition
Network theory in mathematics derives from random graph theory.
Mathematicians have long known that on a randomly connected
network a phase transition between disconnection and connection
occurs when the network is connected enough for most of the nodes
to have joined up into clusters, with a few interconnecting links
between the clusters through the addition of a relatively small number
of new links. The result is the joining of these isolated groups into one
interconnected cluster, known as the giant component. 26 The network
undergoes a dramatic and sudden leap from one state to the other.
26

The normal component is the term for the set of nodes to which a node belongs,
i.e. its cluster, which nodes it is linked to. See Watts, D. J., Six Degrees: The Science
of a Connected Age, New York: Norton, 2003, p. 45-6.

18

Physical examples of phase transition include the magnetisation of iron


molecules, or the freezing of water. What the giant component does is
allow communications across the whole network; i.e. when a network is
not connected by the giant component, events are only felt locally.
Absence of centrality is associated with phase transition and is vital to
understanding the spread of information across networks. No single
molecule causes a piece of iron to magnetise. It is a decentralised,
emergent process.
Small events, behavioural switches, and individual choices
percolate through the system and can lead to widespread change, and
they could come from anywhere. Far from the centre determining the
action on a network, the centre is created by that action. Watts states
that the network centrality of individuals, or any centrality for that
matter, would tell us little or nothing about the outcome, because the
centre emerges only as a consequence of the event itself.27

Small worlds, short paths and overlaps


But randomly connected graphs or networks are not representative of
the real world. Social networks are by nature neither entirely regular
nor entirely random: they are made up of close-knit (regular) clusters,
formed by geography, religion, family and so on, intersected by longdistance (random) links, for example, somebody with whom nobody else
in the cluster has any contact. This is known as a small-world
network.28 As would be expected, the occurrence

of close-knit

communities the clustering of neighbouring nodes is more frequent


than the occurrence of long distance connections. The long distance
connections are then the links that transgress the boundaries of a local
cluster, becoming shortcuts to other clusters in the network. It is a
27

Watts, Degrees, p. 53.


Watts, D., and Strogatz, S. H., Collective dynamics of small world networks,
Nature, vol. 393, 1998.
28

19

global

network

phenomenon

that

arises

from

local

network

interactions: the small-world has a short path length (i.e. direct


access) to other clusters and individuals.
A short path in a social network can be either a weak or strong
tie, where strength and weakness refer only to the fidelity of the
connection. Strength of an interpersonal tie was defined by Granovetter
as a combination of the amount of time, the emotional intensity, the
intimacy (mutual confiding), and reciprocal services. 29 Weak ties are
therefore ties without these characteristics, representative of indirect
or random contacts. They act as low fidelity bridges between different
groups, and the fewer indirect contacts one has the more encapsulated
he will be in terms of knowledge of the world beyond his own friendship
circle.30 Where strong ties breed local cohesion, which can lead to
overall fragmentation31 i.e. groups become too inward looking and
therefore separate from other groups, converse to their name,
Granovetter

observed

weak

ties

to

be

extremely

powerful

at

facilitating transmission across the network. This is because they cross


network distance, (the number of nodes one has to pass through to
reach the desired node) directly accessing totally separate clusters or
individuals. This ability to cross social distance makes weak ties
particularly important for certain types of diffusion in social networks.
This is especially relevant to diffusion that does not require frequent
contact or trust for example, disease, or passing on information about
new jobs.32 These types of link do not, however, generally exert great
influence on people where fundamental issues of change or adoption of
new ideas are concerned.33

29

Granovetter. M., The Strength of Weak Ties, AJS, vol. 78, no. 6, 1973, p. 1361.
Granovetter, Strength, p. 1371.
31
Granovetter, Strength, p. 1378.
32
See Granovetter, Strength; Watts, Degrees, p. 49.
33
The issue is more complex than presented here, as certain individuals especially in
the case of religious diffusion, missionaries who come into a community as a weak
tie may well possess a characteristic such as great charisma, knowledge, status, or
wealth which will make them more likely to be influential.
30

20

This is because, in Granovetters terms, most social ties are


strong, and are responsible for local cohesion. They are reflective of
fundamental facets of human identity, which can be defined (somewhat
simplistically) by group membership. By belonging (or choosing not to
belong) to certain groups, people acquire aspects of identity that drive
the makeup of their social network. Most of these groups are clusters,
strong tie groups trusted people, who may form closed triads the
sociological term for a situation where all three people know each other
or else have other markers of strength, such as frequency or length of
contact.34 Socially, it is these people that form the core of an
individuals environment, and so exert most influence on an individual
when it comes to complex transmission of ideas or information.
Mathematically however, the diffusion of information through
these strong ties of a local social network is problematic. The path
length of strong ties is long, i.e. the information must make lots of little
hops through the different clusters and may take a long time to travel
from one side of the network to the other. However, empirical social
network data has shown that strong ties can still have a relatively short
path length, that allow information to pass through the strong-tie social
network relatively fast: Shi et al. found that removing the weak ties in a
test case did not disconnect the network; rather, the network sheds
some nodes and shrinks modestly.

35

They conclude that a high-fidelity

strong-tie social network, made up of overlapping clusters, spreads


information at almost the same efficiency as the network linked
together with long-distance weak ties. A combination of the two is the
most accurate representation of the real world where both overlaps
and weak ties connect the network. Like the giant component in
physics, these connections make what is known as information
cascade equivalent to a phase transition possible on a social
network.
34

Shi, X., Adamic, L., Strauss, M., Networks of Strong Ties, published as arXiv:archive.condmat/0605279, at http://arxiv.org/abs/cond-mat/0605279, 2006, p. 1.
35
Shi et al, Strong, p. 2.

21


Power laws and restriction of power laws
An important development in understanding the spread of information
on networks resulted from the discovery of Reka Albert and AlbertLaszlo Barabsi that instead of following a normal distribution 36 when
plotted graphically, many real world networks are highly skewed. 37 The
majority of nodes are poorly connected, while a few have massive
connectivity (i.e. the network features people who know everybody,
particular airports, or huge cities). The resulting graph follows what is
known as a power law, which represents many small nodes coexisting
with a few very large ones. Power law networks have hubs, nodes that
are disproportionately well connected.
Albert and Barabsi discovered this network property by running
an experiment where the network could grow. They modelled their
growing network by starting with a small group of nodes, and added
new nodes one at a time. Each new node had two links with which it
could connect to the existing network, the more senior nodes. At each
moment all nodes have an equal chance to be linked to, resulting in a
clear advantage for the senior nodes. Indeed, apart from some rare
statistical fluctuations, the first nodes in Model A will be the richest,
since these nodes have had the longest time to collect links. 38 Their
developed model showed that hubs exist because of a combination of
growth and preferential attachment, i.e. the probability of a new node
joining to an existing node is proportional to the number of links that
node already has. A node attracts new links largely on the basis of

36

The line on the graph creates a curve indicative of the rule of the average i.e. very
few nodes, if any, have no connections, and very few nodes, if any, have an excessive
number of connections.
37
Albert, R., Barabsi, A.-L., Statistical mechanics of complex networks, RMP 74,
2002, pp. 4797.
38
Barabsi, A-L., Linked the new science of networks, Cambridge, MA: Perseus,
2002, p. 83.

22

already having links: part of the reason hubs exist is because they
attract more frequent connections.
Seniority within a network is one of the main reasons a node will
become a hub, but other aspects of identity are also factors, for
example, size, wealth, reputation, location, efficiency, or power, and all
of these are essentially chance variables. The chance variables
possessed by a node weight the network, causing the lock in of
fluctuations, which amplifies differences over time and can ossify a
network into a semi-static state.
Power laws have been found to occur frequently in many realworld situations, and are demonstrated, for example, by the distribution
of wealth, or the popularity of websites. However, human social
interactions are different and far subtler than any other kinds of
networks. Primarily, the limitations of peoples social contacts and
environments mean that they cannot necessarily choose to connect with
a hub. Additionally, they may not want to: the anthropologist Michael
Schnegg39 observed that reciprocity and the recollection of beneficial
past acts restrict the development of hubs and power laws in social
networks. This leads to the conclusion that in social networks,
maximum utility is not as important to most members as reciprocal
altruism and being fair as supported by biological data. 40 Blending
reciprocity and memory into transactions reproduces networks much
better correlated to the social world.41 This is not to say that hubs do
not exist in social networks. Rather, they exist, but their hub status
does not automatically mean they are universally powerful in attracting
links, as opposed to known nodes within a social setting, with which
there has been positive historical interaction. Hubs do however play an
important role in transmitting information over a network. In order to
understand how hubs function in different network situations, it is
39

Schnegg, M., Reciprocity and the Emergence of Power Laws in Social Networks,
published as arXiv:arch-ive.physics/0603005, at http://arxiv.org/abs/physics/0603005,
2006.
40
See Dawkins, Selfish Gene, pp. 183-88.
41
Schnegg, Reciprocity, p. 8.

23

necessary first to discuss the processes by which information diffuses


through a network.

Using the network transmitting information


Network distance the number of steps between two nodes is more
important

in

analysing

information

transmission

than

physical

distance.42 Distance is not measured in just one way however, there are
many types geographical, social, professional every aspect of
identity can be a measure of distance. People use knowledge of the
identities of their local network to assess the best route the shortest
path through which to spread information known as directed search.
Stanley Milgrams letter sending experiment43 in the 1950s provided
evidence of the existence of short paths, and also that people are very
good at finding them. The multidimensional nature of social identity
means that social barriers can often be crossed in the process of
transmission of messages or ideas, and this allows the six degrees of
separation phenomenon to work.
A search for another node, person, group, or target, can be
conducted in two ways. First, one can conduct a broadcast search,
which shows no discrimination and where every neighbour of every
node is involved (which is how disease and computer viruses work),
with the effect of saturating the system. In a power law network, this
type of search is made easier because the chances of coming across a
high-degree node (i.e. a hub with many links) are high, as the
neighbours of the high-degree vertices [nodes] account for a significant
fraction of all the vertices in the network. On average, therefore, we
need only go a few steps along the chain before we find a vertex with a

42

However, in the ancient world, physical distance constricts network closeness in a


different way to modern networks.
43
Milgram, S., The Small-World Problem, Psychology Today, vol. 1, pp. 61-67.

24

neighbour that has the information we are looking for. 44 Second, one
can conduct a directed search, where a node uses the information it has
regarding the most suitable route in order to get to the node it wishes
to reach (the six degrees of separation method).
Viruses perform broadcast searches across the network in order
to spread themselves as far as possible. The directed search is the
intelligent search, where choices are made based upon knowledge of
nodal identity. People use knowledge of the identities of their local
network to assess the best route, the shortest path along which to
spread information. If knowledge of nodal identity can be used to seek
out the most powerful hub through which to spread information, the
search is likely to be even more successful.
When the nodes in a network that are vulnerable to a particular
innovation can be linked together through this process of search, they
form what is known as a percolating vulnerable cluster. The
percolating vulnerable cluster allows information to cascade across it.
It can encompass anything from a massive power failure following the
sabotage of a hub, to a fashion trend or a financial bubble. This can be
rapid, as with an out-of-nowhere success story, or slow, as with the
change of societal norms.
Information cascades, like phase transition in physics, are
demonstrated by the network as a whole displaying emergent or selforganising behaviour: when individuals stop acting like individuals and
behave as though part of an organised group. These cascades are selfperpetuating picking up new adherents largely on the strength of
having attracted previous ones. Hence, an initial shock can propagate
throughout a very large system, even when the shock itself is small. 45
In social networks, small shocks happen all the time, but the pandemic,
or massive information cascade is extremely rare. The system remains,

44

Newman, M., The structure and function of complex networks, in SIAM Review 45,
2003, p. 45.
45
Watts, Degrees, p. 206.

25

on the whole, unaffected by most innovations. Yet some sweep through


the network. Why?

Vulnerability to innovation: thresholds and neighbours


Effective diffusion occurs when the nodes that are vulnerable to the
new idea, technology etc. are found. If we focus on the connectivity of
the network rather than on the inherent value of the innovation, then it
is possible to see that the shocks that trigger cascade do not
necessarily represent superior stimuli. The success of an innovation can
rather be viewed an indication of the network structure in which it
happens to land, whether the local environment is vulnerable or
stable with regard to that particular innovation. There are two ways of
identifying a node as vulnerable: either it has a low threshold (thus, a
predisposition for change); or [] it possesses only a very few
neighbours, each of which thereby exerts significant influence. 46
When people react to an innovation, they are generally making a
social decision, led by the opinions of others when they do not have
enough information to make a rational decision themselves. People have
what is known as a critical threshold where their opinion about
something can jump abruptly from one alternative to another. If in
doubt about a course of action, people will refer to the actions of those
around them. Additionally, people are often reluctant to deviate from
their local norm, as an act of self-preservation. Many people
consistently opt to minimise potential risk [] by observing and
emulating the actions of others. Even where we explicitly eschew the
majority, we are rarely behaving entirely as atomistic individuals or as
pure contrarians.47 In some situations, an individuals threshold may be
very high, when complete unanimity of solicited opinion is required to
cause the individual to switch their opinion about something, and even
46

Watts, Degrees, p. 233. This distinction may be particularly pertinent in modelling


religious change.
47
Watts, Degrees, p. 210.

26

then they may not do so. Often, however, the threshold is much lower,
when there are fewer obvious differences between alternatives. It is at
this level that neighbourly influence is vital to an individuals decision
to switch, because small shocks lead to big shifts in mass behavior only
if people happen to be very close to the borderline between
alternatives.48 If an individual observes enough endorsement of a
particular choice by people around him, he is likely to change,
especially if those people are part of socially powerful relationships,
regarded as being of higher status, or known to have made good
decisions in the past.
These reactions are part of a problem solving mechanism, which
Herbert Simon49 in the 1950s described as bounded rationality.50
Rational behaviour is generally considered all human beings goal, but
cognitive constraints and lack of access to information restrict an
individuals ability to act completely rationally in all situations. A
number of externalities affect behaviour. These include, amongst
others, coercive externalities, the peer pressure of gangs, where
beliefs alter in response to the expressed beliefs of a majority of others,
and information externalities, where a decision is constrained by access
to knowledge.
If enough people surrounding an individual endorse a particular
choice, the individual in question is likely to switch. When an individual,
having observed the actions of those ahead of him, follow[s] the
behavior of the preceding individual without regard to his own
information,

51

he/she may help trigger an information cascade. The

threshold of individuals, their susceptibility to innovation, is central to


understanding information cascades in social networks.
48

Bikhchandani, S, Hirshleifer, D, and Welch, I, A Theory of Fads, Fashion, Custom,


and Cultural Change as Informational Cascades, in JPE, vol. 100, no. 5, 1992, p. 994.
49
Simon, H. A., A Behavioral Model of Rational Choice, in Models of Man, Social and
Rational: Mathematical Essays on Rational Human Behavior in a Social Setting, New
York: Wiley, 1957.
50
Non-human network information cascades will of course not be subject to bounded
rationality. Instead, the positions of hubs in relation to the node where the innovation
(e.g. power failure) hits will be central to the model.
51
Bikhchandani et al, Theory of Fads, p. 994.

27


Innovation: theory of adoption
There are differing degrees of susceptibility to innovation within a
population. The variance in thresholds creates a normal distribution,
the graph pattern discussed above in relation to power laws. There is a
particular point on the graph that forms the core of diffusion, the
tipping point:52 from 10 percent adoption to 20 percent adoption is
the heart of the diffusion process. After that point, it is often impossible
to stop the further diffusion of a new idea, even if one wished to do
so.53 Within a population, there will be a number of different types of
person, displaying differing degrees of susceptibility to innovation, and
this will be subject to change depending on the nature of the
innovation. In the 1960s, Rogers, in his discussion of the adoption of
technological and organisational innovation, identified five groups in
the population that have became the standard definitions for degrees of
susceptibility.54
Innovators. This group forms the first 2.5 percent of the
population.55 Their main characteristic is adventurousness. Interest in
new ideas leads them out of a local circle of peer networks and into
52

See Gladwell, M., The Tipping Point, Great Britain: Abacus, 2000, which explores
the phenomenon in various social situations.
53
Rogers, E., Diffusion of Innovations, fourth edition, New York: Free Press, 1995, p.
259.
54
These categories and social characteristics are drawn from mainly modern case
studies. There are some important differences that should be highlighted if they are to
be applied effectively to past societies. Aside from obvious differences such as speed
of communication and widespread permeation of the media, it is worth questioning
some aspects of the status of innovators: for example, financial independence, or
involvement in non-geographically based friendship circles. Intercommunication
between local communities might be a more useful model to think about for antiquity,
as might patronage of innovators by elites, or at least, technological innovation as an
indicator of prestige. However, aside from these minor points, it seems reasonable to
assume that the categories are broadly applicable to past society, although
identification of early adopting individuals, for example, may prove difficult. An
exception would be the apostle Paul, whose manipulation of his own networks and
active missions make him a clear opinion leader. See introduction to Barabsi, A-L,
Linked.
55
For a discussion of the method used for the percentage divisions, see Rogers,
Innovations, p. 262.

28

more cosmopolite social relationships. Communication patterns and


friendships among a clique of innovators are common, even though the
geographical distance between the innovators may be considerable. []
Control of substantial financial resources is helpful to absorb the
possible

losses

from

an

unprofitable

innovation.

The

ability

to

understand and apply complex technical knowledge is also needed. The


innovator must be able to cope with a high degree of uncertainty about
an innovation at the time of adoption. [] While an innovator may not
be respected by the other members of a local system, the innovator
plays an important role in the diffusion process: that of launching the
new idea in the system by importing the innovation from outside of the
systems boundaries.56
Early adopters. This group forms the next 13.5 percent. Of all
categories, early adopters are the most respected by their peer groups.
They are a more integrated part of the local social system than are
innovators. [] This adopter category, more than any other, has the
greatest degree of opinion leadership in most systems. Potential
adopters look to early adopters for advice and information about an
innovation. [] This adopter category is generally sought by change
agents as a local missionary for speeding the diffusion process. Because
early adopters are not too far ahead of the average individual in
innovativeness, they serve as a role model for many other members of a
social system. Early adopters help trigger the critical mass when they
adopt an innovation.57
Early majority. This category forms the next 34 percent of the
population. They are deliberators, and adopt an innovation just before
the rest of the system. The early majority interact frequently with their
peers, but seldom hold positions of opinion leadership in a system. The
early majoritys unique position between the very early and the
relatively late to adopt makes them an important link in the diffusion
process. They provide interconnectedness in the systems interpersonal
56
57

Rogers, Innovations, p. 263-4.


Rogers, Innovations, p. 264.

29

networks. [They] may deliberate for some time before completely


adopting a new idea.58
Late majority. This group makes up the next 34 percent of the
population. They are sceptical about the new. They adopt new ideas
just after the average member of a system. [] Adoption may be both
an economic necessity for the late majority and the result of increasing
network pressures from peers. [] The weight of system norms must
definitely favor an innovation before the late majority are convinced.
The pressure of peers is necessary to motivate adoption.59
Laggards. The final group makes up the last 16 percent. They are
the

traditionalists.

They

possess

almost

no

opinion

leadership.

Laggards are the most localite in their outlook of all adopter categories:
many are near-isolates in the social networks of their system. The point
of reference for the laggard is the past. Decisions are often made in
terms of what has been done previously, and these individuals interact
primarily with others who also have relatively traditional values.
Laggards tend to be suspicious of innovations and of change agents.
Their innovation-decision process is relatively lengthy, with adoption
and use lagging far behind awareness-knowledge of a new idea.60
This schematic model needs some modification if it is to
correspond to real situations. In reality innovators and laggards form
part of a continuous scale from awareness to adoption, suggesting that
the percentages are really only rough guides, and the categories may
need modification to apply to innovations in ideologies and beliefs.
Many factors affect an individuals position on the scale, including age,
education, environment, class, gender, religion and financial and social
status. It may also be that an individual will occupy different categories
in different situations, and depending on the nature of the innovation.
People also have varying degrees of trust in different sources of
information.
58
59
60

Rogers, Innovations, p. 264-5.


Rogers, Innovations, p. 265.
Rogers, Innovations, p. 265.

30

Rogers generalised the characteristics of adopter categories


under three headings: socioeconomic status, personality values, and
communication

behaviour.

Socioeconomically,

some

of

the

characteristics of earlier adopters were better education and higher


social status measured by variables such as income, occupational
prestige, or identification with a social class. In personality, they were
less likely to have a relatively closed belief system, a greater ability to
deal with abstractions, greater rationality and intelligence, the capacity
to cope with uncertainty, a more favourable attitude towards science,
and higher aspirations. In communication terms, they were likely to
have more social participation, to be more highly interconnected
through interpersonal networks that reached beyond the local system,
to have more contact with change agents, and to have a higher degree
of opinion leadership.

Innovators and opinion leaders


Rogers set out the division between innovator and early adopter. The
innovator may be an outsider, with stronger links to an extracommunity world, perhaps something of a social isolate within his
geographical environment. The early adopter is by contrast tightly
enmeshed in the local community, and an opinion leader held in high
regard by its members. Layton61 responded by observing that: in
setting up a rather rigid set of analytic categories, Rogers takes the
(anthropological) observers position as a constant. Actors are classified
according to the extent that their responsiveness to innovation is in
advance of, or behind, that of the modal member of the population.
Rogers, however, point out that the judgement of an observer within
that population will depend on the observers own responsiveness to
61

Layton, R., Pellaport, in Van der Leeuw, S. E., and R. Torrence, eds., What's New?:
A Closer Look at the Process of Innovation, One World Archaeology, London: Unwin
Hyman, 1989, p. 48.

31

new traits. In other words, a persons position is relational, both in


their own opinion and that of others. For example, some members of
the community might class an individual (positively or negatively) as an
innovator; others might define the same person as an opinion leader.
Layton observed that it is those individuals who are the first to
adopt innovations who are of prime importance. They act both as a
bridge and a buffer between their fellow villagers and the outside world
by bringing detailed knowledge of innovations into that community. 62
The innovators must be connected to early adopters opinion leaders
if the innovation is to spread, let alone cause a cascade. Rogers
observed that opinion leaders were not people at the top of things so
much as people at the edge of things, not leaders within groups so
much as brokers between groups.63 They are the strong-tie overlaps.
They must also be socially accessible, i.e. have greater social
participation than those who follow their opinions.
Finding and using the connectivity of opinion leaders is the most
effective

way

to

spread

information

through

network.

The

interconnectedness of innovators to early adopters, and of early


adopters to the rest of the population can also be described
mathematically. This will help to illuminate the differences between
types of information cascade, and the ways people adopt innovations.

Boundaries, cascade, and cascade failure


The process of informational cascade relies on two related factors:
vulnerability, which can be either inherent to the identity of the node,
i.e. being an early adopter, or couched in terms of neighbourly
influence; and connectedness, the ability to transmit information to lots
of other nodes. Watts uses the term critical upper degree to describe a
62

Layton, Pellaport, p. 50-1.


Rogers, Innovations, (fifth edition) p. 317, quoting Burt, R. S., The Social Capital of
Opinion Leaders, in The ANNALS of the American Academy of Political and Social
Science, vol. 566, no. 1, 1999, p. 37.
63

32

nodes boundary between maintaining the status quo and adopting the
innovation. If a node has more neighbours than its critical upper
degree, then it will be stable with respect to single neighbour
influences, and if not, it will be vulnerable. Variability of degree our
observation [] that some people have more friends or simply solicit
more opinions than others is therefore central to the stability of
individuals and consequently to the dynamics of cascades.64
An innovation can only spread if the innovators are connected to
unstable nodes, equivalent to early adopters: in Watts model, these are
nodes that will switch on if they have one active neighbour. The more
early adopters there are in a population, the more likely the innovation
is to cascade. The larger the connected cluster of early adopters in
which the innovation lands, the further it will spread. If the vulnerable
cluster that is hit by an innovation (that is, the cluster containing an
innovator) happens to percolate through the whole network, then the
innovation will trigger a global cascade.65 It is this, the percolating
vulnerable cluster of unstable nodes that makes global information
cascades possible.
In mathematics, nodes are more stable the more neighbours they
have because they are less likely to be influenced into adopting by just
one of them. So a highly connected node a hub will be the most
stable in a network. By definition, stable nodes cannot be included in a
vulnerable cluster, so that in mathematics, the vulnerable cluster needs
to percolate effectively in the absence of the most-connected nodes in
the network.66 This is deeply counter-intuitive, as it would be expected
that the hubs in a system to be the most useful to the percolation
process. This paradox will be examined further below.
Mathematically, cascades fail to occur for three reasons. First, if
no nodes threshold is low enough. In such cases, this might represent
no need for the innovation, or the innovation does not persuade.
64
65
66

Watts, Degrees, p. 234


Watts, Degrees, p. 235-6.
Watts, Degrees, p. 237.

33

Second, if the network is not sufficiently connected. A disconnected


network is vulnerable to innovation, but does not percolate fully;
therefore the innovation is unable to spread beyond the initial area.
Third, if it is too well connected. In this case the majority of nodes are
stable because of neighbourly influence and so, even though the
network percolates fully, it cannot form a vulnerable cluster. This
analysis creates two boundaries within which information cascades can
occur the lower boundary, below which the nodes have too few
neighbours, and the upper boundary, above which the nodes have
become too densely connected.67
Full network cascades only happen if the innovation activates the
vulnerable cluster of highly connected nodes near the upper boundary,
where the network around the cluster is locally stable. The percolating
vulnerable cluster is tightly integrated with the rest of the network
here, and so initially stable nodes will be exposed to multiple early
adopters, causing them to activate. These prerequisites are difficult to
achieve, making the upper boundary cascade very rare, but when an
innovation

does

activate

the

vulnerable

cluster

here,

the

high

connectivity means the entire network will follow and global cascade
occurs. It is important to note that because upper boundary nodes are
so well connected, it follows also that cascades are almost as likely to
be triggered by an individual with an average number of neighbours as
someone to whom many people pay attention [] being simply wellconnected is less important than being connected to individuals who
can be influenced easily.68
This means that to spread information at the upper boundary,
network connectivity is not as important as being connected to
individuals with a low threshold. Conversely, because nodes at the
lower boundary are vulnerable to neighbourly influence, highly
connected individuals are disproportionately effective in propagating
67

See Watts, Degrees, p. 237-8.


Watts, Degrees, p. 243. This could aid understanding of how cults propagate,
through deliberate search for individuals with low thresholds or few friends to keep
them stable.
68

34

social contagion.69 This fits better with diffusion of innovation theory,


which classes early adopters opinion-leaders with both long distance
links and high social integration as the most effective promoters of
new ideas or technology. There is an important difference between how
cascades happen at the two different boundaries: lower boundary
cascades, i.e., partial cascades, are therefore much more frequent than
upper boundary cascades.

Problems and ways forward


There are a number of issues to be addressed when thinking about
cascade across social networks as opposed to mathematical network
models. First of all, why are highly connected nodes not effective at the
upper boundary too? It has been shown that there are two types of
cascade lower boundary, where hubs play a central role in diffusion,
and upper boundary, where hubs are too inherently stable to be
effective transmitters. This may make sense mathematically, and
perhaps works in some network examples, but it is socially paradoxical.
Early adopters, who by definition are respected opinion-leaders, are
likely to be well-connected nodes, possibly even a local hub, and
certainly personalities who provide overlaps between groups. In real
terms, these well-connected individuals represent the most effective
way of spreading information. It is difficult to see how an innovation
can spread on a social network without these nodes being involved,
regardless of their mathematical stability.
This problem arises partly because identity has for simplicitys
sake been left out of the mathematics. The node with many links may be
regarded as active or passive (and this could change depending on the
situation) in other words, social connections can run in different
directions: the node can be initiating the connection, or be being
69

Watts, Degrees, p. 240.

35

connected to. Sometimes this will be a mutual relationship, but active


and passive links need to be distinguished in the developed model of
social cascades. The frequency of the interaction, and its fidelity, must
be considered in the theorisation of diffusion in social networks: there
is connective directionality and asymmetry as well as variance in values
of different group participations.
How is the success or failure of an innovation judged? Is it by the
time taken for an innovation to be accepted, a general depth of
understanding, or legitimisation by a social leader; or is an innovation
only to be counted as successful if it becomes the norm? Additionally,
it must also be asked how the new information is converted into
knowledge that is trusted and applied, not simply possessed.
The missing third link in the diffusive chain, the early majority
central to the legitimisation of an innovation has been left out of the
mathematical model. Mathematically, the network does not explain the
legitimisation process it ignores it by making the stable nodes all
identical, simply switching under the collective pressure of being
connected to many early adopters. Socially, this is not the case.
Understanding

legitimisation

is

central

to

understanding

how

innovations go from being small pockets of adoption to being cascades


of information.
Perhaps this suggests that is not viable to talk about upper
boundary, full network cascades when thinking about the transmission
of certain types of innovations across social networks. Perhaps there
are no universal innovations, that apply to all human beings in all
societies: what actually happens is that full network cascades can exist
only in closed networks societies and groups where a particular norm
will be in operation. Furthermore, there is a distinction to be drawn
between technological and ideological innovations and their differing
abilities to percolate fully across a network: technology can be
empirically tested, and religious belief cannot.

36


Spreading

religious

innovation:

horizontal

and

vertical

transmission
As noted by the cultural evolutionists in the earlier part of this chapter,
information spreads along both horizontal and vertical axes. Horizontal
transmission refers to the imitative process, by which Boyd and
Richerson suggest much of cultural change occurs; whereas vertical
transmission requires a depth of knowledge and experience, for
instance through conditioned learning,70 the culture inherited from
parents and lived environment. An individuals religious environment is
generally inherited.
There are two separate cases to be considered with regard to
religious innovation: first, the generation of a new religious movement,
and second, the adoption of an established religious movement by new
worshippers. The generation of a novel religious movement might well
result from horizontal imitation, but it will naturally ossify into a
format with a vertical transmission process, in much the same way as a
genetic mutation might be inherited and passed on through natural
selection. The conversion to a religious movement, whether recently or
immemorially established, is a more complex process, and it is this
subject that the next chapter will seek to address.

Chapte r 2.
Theoretical framework II: sociology of
religion
Religious Innovation & Conversion.

70

See Knappett, C., Thinking Through Material Culture: an interdisciplinary


perspective, Philadelphia: University of Pennsylvania Press, 2005.

37

Introduction

This chapter moves the discussion of change specifically back into the
sphere of religion, and investigates the processes of innovation,
adoption and diffusion as they are understood in sociological terms.
Sociological research into the processes of conversion found in modern
religious movements forms a more grounded counterpoint to the
abstracted theories explored in the previous chapter. The core theory of
this chapter is the work of the sociologist of religion Rodney Stark and
his various co-authors, supplemented by re-assessments, critiques and
clarifications of their ideas and observations.
The central point of the chapter is that sociologists of religion can
go some way to explaining the processes of religious conversion (or
recruitment) and the success or failure of a religious movement
through an analysis of social interactions. Stark and Bainbridge 71 have
shown that although aspects of social status and background are
important to an individuals general receptiveness to new religious
movements, it is their social contacts their network which are
crucial to an individuals actual recruitment to a new religious
organisation. The basis for successful conversionist movements is
growth through social networks, through a structure of direct and
intimate interpersonal contacts. Most new religious movements fail
because they quickly become closed, or semiclosed networks. []
Successful

movements

discover

techniques

for

remaining

open

networks, able to reach out and into new adjacent social networks. 72
This does not mean that the values of the religious movement are
irrelevant, rather that an analysis that combines understanding the

71

Stark, R., and Bainbridge, W. S., Theory of Religion, New York: Peter Lang, 1987.
Stark, R., The Rise of Christianity, New Jersey: Princeton University Press, 1996, p.
20.
72

38

ideology and the social system gives a more rounded picture of the
spread, success or failure of religious movements.
This is essentially the same logic as that of the network theories
explored in the previous chapter. Rather than assuming the superior
quality of a religious stimulus that proves to be successful as a given
factor, recruitment to religious groups and their success should be
understood as a social process: the direct and intimate contacts in
peoples social networks the strong ties drive religious change.
Understanding

religious

innovation

and

conversion

from

this

perspective does not pass judgement on the truth or beneficial


qualities of the ideology, it instead engages with the social aspects of
why humans believe and why they change what they believe. This
approach attempts to give a more objective explanation for why some
religious movements succeed and some fail. The ideology of a religion
will always be part of the explanation for a movements success in
recruitment and growth. However, regardless of the persuasiveness of
an ideology, or the transcendence of a vision, conversion does not
happen in a social vacuum.
An overview of the theories pertaining to the creation and
development of religious organisations, the social environments,
character and operations of church, cult and sect movements, and the
processes of conversion (or recruitment) to these movements are
discussed first. Following this, the trajectories of religious movements
are examined, before the chapter ends by outlining some of the issues
that need to be considered in the application of modern sociological
theory to antiquity, namely, the differences between monotheist and
polytheist environments as spaces in which religious innovation and
conversion can happen.

39

The Creation of Religion: the Stark-Bainbridge model

In the late 1980s, Stark and Bainbridge formulated their controversial


scientific theory of religion based upon the rational-choice economic
system of rewards, costs, and compensators. It is a formal deductive
theory of human action, based upon axioms and the propositions that
can be logically derived from those axioms. The knowledge of the
inevitability of death is at the heart of the theory, which they claim
leads almost universally to the creation of the ultimate unobtainable
religious reward: eternal life. They argue for a progressive evolution
of religion from magic to polytheism to monotheism, deducing that over
time, economic exchange mechanisms lead to gods of larger and larger
scope, until monotheism, or more specifically a duality between good
and evil, emerges as a logical conclusion. From this, they argue, come
religions that do not provide specific compensators. These may run
their course until God vanishes completely, 73 and secularisation is the
end result.
The theory has received not a small amount of criticism. As with
any economic model, it is essentially reductionist. Some of the major
issues with this conception of the creation and development of religion
need to be outlined. First, it comes from a positivist standpoint, since it
assumes that there exists a causal, non-random relationship between
social phenomena. Specific social situations consistently account for the
appearance of social phenomena, which are assumed to be universal. 74
Second, a quintessentially American, capitalist perspective underpins
the theory: the central notion is that actors are constantly on the prowl
for rewards and never seized with the notion that there are, in
principle, insurmountable social obstacles to achieving rewards. Should
rewards not be achieved, the actor does not retreat to a dark corner of

73

Stark and Bainbridge, Theory, p. 112.


Simpson, J. H., The Stark-Bainbridge Theory of Religion, in JSSR, vol. 29, no. 3,
1990, p. 367-371.
74

40

the society and hatch a plot to overthrow the system. 75 Rewards are
distributed unevenly throughout society, and this is largely determined
by the personal characteristics of the actors. Third, the theory rests on
the presupposition inherent in rational-choice theory: that rational
beliefs and actions always display logical consistency and are explained
by reasons, whereas irrational beliefs and actions are inconsistent and
require explanation by causes.76 The observer cannot distinguish
between the rational or irrational beliefs internal to the actor that drive
an action: to the actor themselves, the action they choose to make is
rational. When an actor is called to defend an action that is radically at
variance with prevailing beliefs, the legitimisations chosen for it will
bring the action into compatibility with the environment. 77 The
reductionism of the theory also ignores the special status of religious
belief within conventional discourse and rationality. The belief in a
higher power, which is beyond rational explanation, means that
religious activity is not subject to the same rules and does not conform
to the same patterns as other forms of behaviour. With other forms of
activity, for example, economic behaviour, if a certain course of action is
unproductive, we abandon it in favour of something else, whereas this
is not the case in religious situations, for example, the increase in
prayer

in

times

of

distress,

even

though

those

prayers

are

unanswered.78 Finally, the authors freely use the concept of evolution


to describe the process of religious development. Although they do
recognize that this process can be reversed, for example, following
societal upheaval, the term is used without acknowledging the
(commonly misconceived) inherent value-claims of superiority, and the
notion of competition that evolution entails.
75

Simpson, Theory of Religion, p. 371.


Wallis, R., and Bruce, S., Sociological Theory, Religion and Collective Action,
Chapter Two, Accounting for Action: Defending the common sense heresy, Northern
Ireland: The Queens University of Belfast, 1986, p. 16.
77
Wallis and Bruce, Theory, p. 18.
78
See Phillips, D. Z., Religion Without Explanation, Oxford: Blackwell, 1976, which
offers a thorough philosophical critique of attempts to rationalise religious behaviour
by reducing it to conform with patterns of secular activity.
76

41

The

modernist-atheist

notion

at

the

core

of

Stark

and

Bainbridges theory is that religion is not sui generis and therefore


cannot be accounted for on its own terms. Although there are many
issues that can be raised about the political and philosophical
standpoints from which the theory is generated, there are nevertheless
many useful observations and conclusions that are drawn about the
social processes that influence the development and success of
religious movements. This first section gives a brief outline of the
general theory, before moving the discussion to the more relevant
specifics of social tension, recruitment to cults and sects, and the social
interactions that are fundamental to this process.

Rewards and Costs


The starting point of A Theory of Religion is that humans seek what
they perceive to be rewards and avoid what they perceive to be costs 79
and that the human mind naturally seeks explanations for how and why
rewards may be obtained and costs are incurred. Stark and Bainbridge
propose that generally, it is fairly easy to find a successful explanation
and to solve the problem of obtaining a desired reward, but that
sometimes, rewards are in limited supply, and some, e.g. eternal life, do
not exist at all.80 They argue that many rewards have a temporary
nature and are destroyed when they are used, i.e. they are consumed,
and so must be acquired again and again. This leads to the exchange of
rewards, and therefore, the assumption that high exchange ratios are
sought. Their final axiom is that individual and social attributes which
determine power are unequally distributed among persons and groups
in any society.81 Although this has an inherent capitalist perspective, it

79
80
81

Stark and Bainbridge, Theory, p. 27


Stark and Bainbridge, Theory, p. 31
Stark and Bainbridge, Theory, p. 32

42

is generally true that a series of exchanges can result in asymmetry


over time. It is used as the basis for a series of propositions.
Power is the degree of control an individual has over their exchange
ratio, based upon various aspects of their inherited, achieved or
ascribed statuses. Stark and Bainbridge observe that rewards that
exist only in limited supply are particularly susceptible to the exercise
of power and that as a consequence of higher exchange ratios, these
rewards will be likely to end up in the hands of those who are socially
powerful and not of those who are weak.82 The powerful then acquire a
monopoly over rewards whose supply is limited, and those rewards
become less available to others. People are willing to pay high costs for
rewards that they desire greatly.
A difficulty arises from the axiom that states that certain rewards
are in short supply or may not exist at all. How do people assess this
possibility? Stark and Bainbridge propose that sometimes it is
impossible to be certain that a reward does not exist; and that when a
desired reward is relatively unavailable, explanations that promise to
provide it are costly and difficult to evaluate correctly; and that the
more valued or general a reward, the more difficult will be evaluation of
explanations about how to obtain it. Explanations that set limited time
on the appearance of the reward are likely to be invalidated, which
leads to the next proposition, in the absence of a desired reward,
explanations often will be accepted which posit attainment of the
reward in the distant future or in some other non-verifiable context. 83
These propositions set out to explain why people often continue to
follow a set course of action, even when they are incorrect or unproven.
This explanation does not take account of the fact that people may
simply be operating at a level of normalcy within their cultural
framework and may involve a misapprehension of what is rational or
irrational nature in specific religious belief systems.

82
83

Stark and Bainbridge, Theory, p. 33.


Stark and Bainbridge, Theory, p. 35.

43


Compensators
Stark and Bainbridge argue that compensators are invented as a way
to substitute for the strongly desired reward: they are both a promise of
a future reward and an explanation for how to achieve it, and as such
they are treated and exchanged as if they were rewards. Stark and
Bainbridge make the assumption that humans will always prefer
rewards to compensators, and will try to exchange compensators for
rewards.84
From this, Stark and Bainbridge introduce religion. They argue
from their series of axioms and propositions that religion is a naturally
emerging phenomenon across all human societies. Their general
compensators, those that act as substitutes for clusters of rewards, or
for rewards of great scope or value, can be supported only by
supernatural explanations. Such general compensators are sought to
explain the many questions that humans have about their lives and
sufferings. Some of these difficult questions require a supernatural
answer, because with why questions, one is either condemned to
chance or fate. As Stark and Bainbridge write: to seek the purpose
of life is to demand that it have one. The word purpose is not
compatible with blind chance, but assumes the existence of intentions
or motives. These assume a consciousness. For the universe to have a
purpose, it must be directed by a conscious agent or agents.85
This idea of universal purpose distinguishes religion from magic,
and magic from science. Magic is used to explain or manipulate
specifics about the world; as opposed to the general questions to which
religion claims to offer answers. Magic can thus be empirically
evaluated and is adopted in societies lacking the means to test its
efficacy. The modern eras new god, science, answers general questions

84
85

Stark and Bainbridge, Theory, p. 37.


Stark and Bainbridge, Theory, p. 39.

44

without resort to purpose, and in most cases, fulfils the criterion of


empirical evaluation.86
Major issues have been raised with Stark and Bainbridges
understanding of the compensator and the reward, and their conception
of human desire in general as being entirely materialistic. Wallis and
Bruce argue in response that the compensator does not in fact provide
a substitute for the tangible this-worldly reward. It may make present
misery easier to endure while we wait to secure that reward in the
future, but it is only the future reward itself that can substitute for what
we cannot have here and now.87 They also criticise the description of
reward, as Stark and Bainbridge assume that rewards are only viable if
they are tangible, concrete and immediate. Anything else somehow
becomes merely symbolic, unreal and thus a substitute for some
present gratification.88 The promises of religion, then, cannot be
desired for their own sake, but must always arise as a substitute for
something else, an assumption derided by Wallis and Bruce as
substantively atheistic and repetitive of the reductionist errors of
Durkheim, Marx and Freud in construing religion as reducible to
mundane considerations.89 They observe that since much of sociology
is committed to uncovering hidden motives lying beneath public
rhetoric, it tends to reject the possibility that symbolic goals and
values could really be what actors sought in such enterprises.90

86

Stark and Bainbridge, Theory, p. 41. This is not to suggest that science answers
why questions generally it offers answers to how questions. As Stephen Hawking
asked at the end of A Brief History of Time, the how of the universe have begun to be
answered, but the why is a question for philosophy.
87
Wallis and Bruce, Theory, Chapter Three, A Critique of the Stark-Bainbridge Theory
of Religion, p. 50-52.
88
Wallis and Bruce, Theory, p. 52.
89
Wallis and Bruce, Theory, p. 53.
90
Wallis and Bruce, Theory, p. 59.

45

Religious organisations and specialists


Although these are important criticisms of the methodologies inherent
in the Theory of Religion, Stark and Bainbridge nevertheless make
some useful definitions and conclusions concerning religion within
society: religion is a system to explain fundamental questions about
human life, and to provide general compensators in place of highly
valued

rewards.

Religious

organisations

are

therefore

social

enterprises whose primary purpose is to create, maintain, and


exchange supernaturally based general compensators. 91 The emphasis
is on the social dimensions of religion: the social functions of group
membership

and

reinforcement,

and

the

fact

that

religious

organisations provide immediate rewards alongside supernaturally


based compensators. These rewards are generally social leadership
positions, status, respect, companionship but they may also be
financial, for example the giving of alms, or the accumulated wealth of
some religious leaders.92
Stark and Bainbridge go on to argue that societies will evolve
religious specialists as a process of general specialisation. Because
these specialists then enjoy a beneficial relationship with the gods, and
exert certain controls over the norms of the society, these specialists
will constitute an elite. Religious organisations and positions are
accordingly controlled by the powerful. 93 While this may well be largely
true in terms of social function, it does not do full justice to the
motivation

of

religious

specialists.

What

about

those

religious

specialists who are outside organisations, for instance, ascetics and


hermits? An issue with these statements is the mutually reinforcing
nature of the elite relationship to power. In some cases, powerful people
certainly seek power through membership or control of religious
specialisation (for example, the Medici family in Renaissance Florence
91

Stark and Bainbridge, Theory, p. 42


Stark and Bainbridge, Theory, p. 43.
93
Stark and Bainbridge, Theory, p. 97-102. A counter-example to this might be the
Quakers, who, although there are priests in some churches, operate with a
deliberately anti-hierarchical social arrangement whereby each individual can express
their faith in God.
92

46

and their relationship with the Papacy) but this is not always the case.
To describe desire for religious leadership as either the preserve of the
already powerful or as driven by a functionalist desire for power or
status misses out many of the internal factors that draw people to these
positions, for example, compassion, desire to initiate social care or
repair, or simply a profound personal spirituality. While these feelings
are likely to find their expression in a religious norm, it does not hold
that the people who are religious specialists are motivated solely or
primarily by desire for power or status. This again highlights the
problem identified by Wallis and Bruce, that the sociologists need to
find an explanation in social function is contradicted by the reasons
given for his actions by the agent.

Churches, sects, and social divisions


Religious organisations operate as part of a social environment. People
join or are born into these environments and as such, adopt the
religions of those environments. Socialisation conditions standards and
norms, and religion can be defined as a chain that links the individual
into a community of past, present and future members, into a tradition
or collective memory that is the basis of that communitys existence.94
Within society, different religious organisations occupy different
positions, characterised essentially by the level of tension between
them and their environment. Tension, in this context, is shown by the
extent to which a religious unit differs from the socio-cultural standard.
Churches are characterised by low-key religious participation, deal
mainly in direct social rewards, and operate at low tension with the
surrounding social and cultural environment. Sects, by contrast, are
characterised by fervent religious commitment and rejection of the
surrounding social and cultural environment and place their emphasis
94

Hervieu-Lger, D., Religion as a Chain of Memory, English edition, Cambridge:


Polity Press, 2000.

47

on compensators.95 The compensators offered by sects generally include


religious experiences, feelings of moral superiority, prayer and divine
aid. These contrast with the socially constituted rewards offered by
churches, which include status and community legitimacy, participation
in organisations affiliated with the church, and the socialisation of
children.96
Sects are schismatic. They form as divisions from existing
organisations in their religious tradition, and are characterised by a
move towards a high-tension relationship with their environment.
Schism is most likely to be linked to previously existing social cleavages
divisions of a network across which there are relatively few strong
attachments and are most likely in groups that are strongly socially
divided in this way.97 Strong attachments are key to schismatic
movement: these represent the most highly valued members of an
individuals immediate network, and as such are those with whom that
individuals exchange ratios are highest, to couch it in economic
terms.98 Sects largely tend to be populated by the disaffected members
of a religious environment, whose powerlessness with regard to gaining
rewards through exchange demands a series of more efficacious
compensators. Stark and Bainbridge argue that these moderately
powerful but relatively deprived disaffected members of society are
offered an opportunity to increase their own rewards and social status
by organising a sect movement.99 Although it is difficult to classify the
desire to occupy a position of spiritual leadership as a simplistic
functionalist desire for power and status, this characterisation of a sect
as a schismatic, high-tension group formed of largely disaffected and
powerless individuals seems to be reasonable.
Empirical studies suggest that distinctions between church and
sect are often linked to socio-economic status (SES), confirming Stark
95
96
97
98
99

As defined by Johnson, B, On church and sect, ASR 28.4, 1963, pp. 539-549.
Stark and Bainbridge, Theory, p. 46.
Stark and Bainbridge, Theory, p. 131.
Stark and Bainbridge, Theory, p. 133.
Stark and Bainbridge, Theory, see p. 145-6.

48

and Bainbridges theory, that SES should be positively associated with


religious

rewards

and

negatively

associated

with

religious

compensators with the exception that SES should be unrelated to


belief in life after death.100 From Glock and Starks observations101 of
modern American religious involvement, the upper and middle classes
are predominantly involved with church based religion. However, the
vast majority of the unchurched population was found to be of a
disproportionately lower SES, an issue to which we shall return. 102
There was also a clear link found between the more highly educated
sectors of society and lack of religious belief.103

Deviance: Sects and Cults


Sociologists of religion have articulated the fundamental difference
between a cult and a sect. Stark and Bainbridge define a sect as a
deviant religious organisation with traditional beliefs and practices, as
an intensification of the norm; as opposed to a cult, which represents a
deviant

religious

organisation

with

novel

beliefs

and

practices.

However, both cults and sects operate at high tension within their
surrounding social environment. Because they are deviant groups,
they represent a departure from the cultural norms in such a way as to
incur the imposition of extraordinary costs from those who maintain the
culture.104 They raise the costs of exchange across the boundaries of
that culture. As a result, deviant groups in high tension with their
cultural environment may incur punishment by the followers of the
norm, either by a conscious policy of imposing special costs upon the

100

Stark and Bainbridge, Theory, p. 47.


Glock C. Y., and Stark, R., American Piety: The Nature of Religious Commitment,
Berkeley: University of California Press, 1968.
102
Stark and Bainbridge, Theory, p. 48.
103
How true might this be for the educated elites during the Roman period?
104
Stark and Bainbridge, Theory, p. 124.
101

49

deviants, by attempts to change their behaviour, or by driving away or


destroying the deviant group.105
Cults are innovative religious movements with novel beliefs, and
as such, they form part of a two (or more) -step process: first, they
invent new religious ideas, and second, they gain social acceptance for
these ideas at least within a group large enough to sustain them. 106
Cults have been found to be most prevalent in societies that do not have
a dominant religious tradition supported by the elite: because the
deviance on which cult formation and success is predicated is less
likely to be punished in such an environment. Anthropology, psychology
and sociology propose three general approaches to the generation of
new cultic movements: the psychopathology model, the entrepreneur
model, and the subculture-evolution model.
The psychopathology model takes the creation of cult to be a
response to an individuals personal or social crisis. It postulates a
mentally unstable individual creating compensators that meet his/her
needs during a psychotic episode which, if it coincides with a time of
societal upheaval or crisis, where there are others with similar needs,
may result in a group of people whose needs are met by the founding
individuals set of compensators.107 The entrepreneur model sees the
creation of cults somewhat cynically, as businesses that provide their
customers with a product. The founders are then individuals who are
motivated largely by desire for profit, a notion generally derived from a
prior involvement with a successful cult. This leads to the use of
previously experienced formulae to create new compensators, leading
to cult lineages that share a number of features. 108 The subcultureevolution model understands the creation of cult as the expression of
novel social systems, small groups composed of a few intimately
interacting individuals, generally formed from populations involved
with the occult milieu. They are thought to be the result of failed
105
106
107
108

Stark
Stark
Stark
Stark

and
and
and
and

Bainbridge,
Bainbridge,
Bainbridge,
Bainbridge,

Theory,
Theory,
Theory,
Theory,

p. 125.
p. 156
see p. 158-68.
see p. 168-78.

50

collective attempts to gain scarce or nonexistent rewards, which


originated when a group committed itself to attainment of those
rewards. By collectively working towards this aim, they begin to
exchange other rewards, and as their attempts to gain the nonexistent
or scarce rewards fail, they begin to also exchange compensators. If
this intra-group exchange becomes sufficiently intense, the group will
become encapsulated, in extreme cases undergoing social implosion.
Once separated from external control, the group develops and
consolidates a novel culture. Successful cult evolution then ends (or
ends the beginning) with a new religious group faced with the task of
extracting resources from the surrounding environment.109
It is observed that over time, cults naturally become more
introverted and intra-connected. Exchanges become internalised and
reinforce the social bonds of the religious group. This is also described
as social implosion, as the open network gradually becomes more
closed.110 Small high-tension sects are particularly vulnerable to this
process, as they move from schism with a traditional church toward
religious innovation and eventual socially closed cult status. Stark and
Bainbridge qualify this with the observation that to the extent that a
society does not punish religious innovation, small sects tend to evolve
into cults111 i.e. the group will begin to exchange novel beliefs as a
result of their isolation.
Environments undergoing rapid cultural or social changes tend to
produce situations where cults or sects can form. Innovation in religion
often happens in response to encounters with new technology, or social
crisis war, famine, disease etc. The disruption to social networks
following rapid cultural or social change creates new social needs or
conditions.112

109
110
111
112

Stark
Stark
Stark
Stark

and
and
and
and

Bainbridge,
Bainbridge,
Bainbridge,
Bainbridge,

Theory,
Theory,
Theory,
Theory,

see p. 179-187.
see p. 185.
p. 187.
see p. 189.

51

Conversion, or recruitment

Moving on from the theoretical definitions and explanations of novel


religious forms and the social situations that produce religious
innovation,

we

now

turn

to

the

process

of

conversion,

or

recruitment.113 Breaking with an implicit culture, a set of communal


memories encapsulated in ritual and religious practice, is a remarkable
act. Realigning oneself to a different set of beliefs within any sociocultural hegemony requires extraordinary circumstances: how does this
happen? How, why, and who join religious groups, whether they are
established churches, or new sect or cult movements? It is an external,
social process as well as an internal, psychological process. Both of
these factors must be considered.

Conversion: seven conditions?


Psychological and social reasons for conversion are not necessarily
separate. Stark and Bainbridge argue that those individuals who are
likely to be recruited to a new (to the individual, not necessarily to the
world) religious group are those members of society who are free to
deviate114 from the accepted cultural norms. This is related to the
degree to which they have stakes in social conformity, i.e. social
attachments, investments, and involvements. It is argued that these

113

Terminologically, Stark and Bainbridge discard the word conversion because of


the implication it has for actual internal transformation of the individual, preferring
instead to use recruitment or affiliation to describe the social process of joining a
religious organisation.
114
This is not to suggest that non-conformists are drawn automatically to religion.
Choosing non-conformity also allows access to a non-conformist society, with its own
set of attachments, involvements and loyalties.

52

members of society can be generally characterised as deprived in


terms of rewards, compensators, and self-esteem.115
The Lofland-Stark conversion model of 1965 identified seven
psychological and social conditions that make an individual more open
to deviation. The first three are predisposing conditions and the last
four situational factors: 1, experience of enduring and acutely-felt
tensions; 2, within a religious problem solving perspective; 3, resulting
in self-designation as a religious seeker; 4, the potential convert must
come across the religious group at a turning-point in his or her life; 5,
form an affective bond with one or more members; 6, neutralize or cut
contact with outside non-members; and 7, experience

intensive

interaction with other converts.116 These conditions are required for


conversion to a group considered deviant by the norm in any given
society i.e. a cult or a sect. This analysis, however, does not pose the
question of the conditions relevant to a decision to join a non-deviant
religious group.
Empirical observations of a Buddhist cult in America in the 1970s
caused Snow and Phillips to challenge the Lofland-Stark model. 117
Citing G. H. Meads observation that meanings of events are not
intrinsic, but rather are conditioned by an individuals current situation,
they argue that the psychological criteria for conversion are liable to
charges that these involve a retrospective re-characterisation of the
past. Although this may be the case, it is nevertheless important not to
subsume the individuals psychological reasons for conversion beneath
the wider social factors that the sociologist might argue to be the
driving factors of conversion. Wallis and Bruce argue against Marxist
and functionalist social accounts: the presumption must be that [the
actors] characterisations of their actions and their accounts of why
they are performing them are the correct ones.118
115

Stark and Bainbridge, Theory, p. 190-191.


Lofland, J and Stark, R, Becoming a world-saver: a theory of conversion to a
deviant perspective, ASR 30, 1965, pp. 862-875.
117
Snow, D., Phillips, C. L., The Lofland-Stark conversion model: a critical
reassessment, in Social Problems, vol. 27, no. 4, 1980.
118
Wallis and Bruce, Theory, p. 19-29.
116

53

There are clearly problems with interpretation post-event. These


stem from both the subjective nature of the individuals accounts of
conversion and from the attempted objective understanding of wider
social reasons that underpin conversion. Although Lofland and Starks
approach is essentially functionalist, their theoretical framework for
understanding the process of conversion does incorporate both
psychological and social factors. The issues with the model that are
raised by Snow and Phillips from their empirical work are also included
here.
1) Tension
By tension, Lofland and Stark mean a strong, unmet desire, which is
felt prior to the affiliation with the new religious group. It is
problematic to ascertain the role of tension in conversion, as it is
possible to reinvent the tensions felt at the time of conversion to align
with feelings in the new, converted present. Snow and Phillips found
that reported feelings of tension may well have formed part of the
reasons given for conversion, but they conclude that it is not
necessarily a pre-disposing condition.119 It is possible to hypothesise
about the wider social tensions that might create a general mentality of
unmet desire, on which more below.
2) Religious problem-solving perspective
Lofland and Stark mean by this that, in order for conversion to be likely,
there should be a concordance of outlook between the way an
individual approaches solving the problems he faces and the potential
methods of doing so political, spiritual, psychological, etc. However,
Snow and Phillips research did not suggest that converts were
necessarily in possession of a religious problem-solving perspective,
and so where it does exist, they argue that it constitutes a facilitative
rather than a necessary precondition for conversion.120

119
120

Snow and Phillips, Critical Reassessment, p. 435.


Snow and Phillips, Critical Reassessment, p. 438.

54

3) Being a religious seeker


Lofland

and

Stark

suggest

that

this

religious

problem-solving

perspective leads into the individual self-classifying as a religious


seeker someone who searches for religious meaning to interpret and
resolve his discontent.121 However, the subjects of Snow and Phillips
research did not show particular leaning towards religion prior to their
conversion. The authors also point out that an individuals self-definition
as a seeker is subject to retrospective reassignment. They conclude
that this aspect of Lofland and Starks analysis is not usefully applicable
when attempting to understand why people convert, and certainly does
not constitute a necessary precondition.
4) Turning Points
The issue of retrospective re-assessment is equally applicable to the
concept of a turning point. Retrospectively, one can classify almost any
event or non-event as a turning point that led to the adoption of the
new faith. However, the situations Lofland and Stark alluded to as
turning points were objective, such as bereavement, divorce, losing a
job, finishing school, moving house etc. Snow and Phillips research
supported the conclusion that many converts refer to their conversion
in terms of a turning point, but that they do not necessarily mean
objective turning points such as Lofland and Stark assumed. The
subjects of Snow and Phillips research defined their turning points
not as moments of external caesura, but rather as the internal moment
when they come to align themselves with the movement emotionally,
cognitively, and morally seeing themselves at one with the group. 122 A
turning point for any given individual is a subjective assessment of their
own situation and emotions, rather than one that can be objectively
discerned.
These four criteria are all concerned with the individuals internal
emotional landscape. As Wallis and Bruce argue these are certainly vital
to an individuals susceptibility to and motivation towards conversion,
121
122

Lofland and Stark, Becoming a world-saver, p. 868.


Snow and Phillips, Critical Reassessment, p. 439.

55

but they are also extremely difficult to ascertain after the event and are
subject to retrospective editing. Furthermore, their usefulness in an
application of theory to the ancient world is severely limited. The last
three criteria are concerned with the external, social network factors
involved in conversion.
5) Affective bonds
Building relationships with the other members of the religious
organisation is the key for recruitment of new converts: as Stark and
Bainbridge put it, the development of valued new social relations with
members the affiliation process is essential. 123 Lofland and Stark
argue that people with few social relationships are more likely to be
open to any kind of new social exchange, and so more ready to accept
those that bring religious affiliation with them. In addition, if an
individuals closest and most valued social associates shift their
religious affiliation, the individual in question is quite likely to follow
suit.124

Snow

and

Phillips

research

supported

this

analysis

of

recruitment, with 82 percent of their sample found to have been


recruited to the cult through pre-existing social bonds.125
6) Cutting outside contact
Lofland and Stark argued that full conversion was less likely without
the cutting of extra-group bonds. However, Snow and Phillips findings
provided little support for the hypothesis, linked to the fact that 82
percent of their subjects reported recruitment through pre-existing
social networks. They suggest that this observation might be to do with
the particular Buddhist cult they investigated, conversion to which may
lead to the strengthening of bonds particularly with family, co-workers
and classmates outside the cult.126 They suggest that the reason for this
may be because of the particular non-communal nature of the cult.
Other communal religious groups may be more demanding and might
require the breaking of extra-group affective ties in order to convert
123
124
125
126

Stark
Stark
Snow
Snow

and Bainbridge, Theory, p. 233.


and Bainbridge, Theory, p. 234.
and Phillips, Critical Reassessment, p. 440.
and Phillips, Critical Reassessment, p. 441.

56

fully. However, the social ties of new members provide good new
avenues along which a religious movement can spread. The dichotomy
may be characterised as the decision between the maintenance of
insularity and the necessity of expansion.
7) Intensive intra-cult interaction
Lofland and Stark suggest that intensive interaction with group
members is necessary for the transition of the initiate to becoming a
full, devoted member of the religious movement. Snow and Phillips
observations fully support this, and they go on to suggest that intensive
interaction is perhaps the most important factor in the conversion
process once the prospect has been informed about and brought into
contact with the movement.127
Therefore, Snow and Phillips research supports certain aspects
of Lofland and Starks criteria namely, those that are concerned with
the importance of social relationships. This is because attempting to
reconstruct the set of internal feelings that converts experienced prior
to their joining a religious group is both difficult and fallible, whereas
understanding something of their social relationships is generally
easier. The most relevant conclusion here is that social networks are
shown to form a major part of the explanation for conversion and for
the growth of religious groups.

Wider social factors in the conversion process


The above analyses have been concerned with the conversion of the
individual

and

the

factors

that

might

make

them

individually

susceptible. However, there are also wider social factors that impact on
the religious environment and increase the likelihood of conversion.
These are, broadly: instability, class, and deprivation.

127

Snow and Phillips, Critical Reassessment, p. 442.

57

One of the functionalist social explanations for conversion is the


elevation of the converts relative social status. We have seen that cults
and sects are considered deviant within an established religious
milieu, and are categorised by a degree of tension with it. Although the
status of the religious group within that milieu is lower, because they
are a minority, because they represent or believe something that is
alien, or because they may have certain attached social stigmas, the
status of individuals within the social structures of the group is
heightened relative to the outside world.
Potential recruits to deviant religious groups are generally drawn
from socially unstable sectors of society. Their instability and
vulnerability is connected to non-integration in a social network. This
occurs

at

times

of

decreased

attachments,

investments

and

involvements that are manifest in close social networks. 128 These can
range from seasonal employment to bereavement or relocation. In
social network terms, people with decreased attachments can clearly be
viewed as unstable or vulnerable. Unstable individuals or families
could therefore be understood as viewing an investment in the relative
higher status available within the deviant group as worthwhile.
Class and deprivation, i.e., being a victim of unfair exchange
ratios, also have an effect on the likelihood of joining social movements,
whether religious or secular.129 In 20th Century America the upper and
middle classes have formed the core of the church going population.
They are more likely to support the religious status quo, because the
dominant religious group and the political state are interconnected, and
such groups are generally supportive of the political and social situation
as it stands. These observations suggest that those attracted to
deviant religious organisations should be largely drawn from the
socially powerless and deprived. This may be true for America, where
left-wing politics remains weak;130 but Stark showed that, by contrast,
128

Stark and Bainbridge, Theory, p. 229.


See Stark, R., Class, Radicalism, and Religious Involvement in Great Britain, ASR,
vol. 29, 1964, p. 699.
130
Stark and Bainbridge, Theory, p. 49.
129

58

in Europe,131 the energy of social groups with the least socio-economic


power was quite likely to be channelled into leftist political activism, i.e.
into the fight for change. However, Starks research is based on modern
religious and political observations, and the political activism of the
lower classes cannot really be traced back into history further than the
Industrial Revolution. What would the situation have been in a
predominantly peasant culture? It is tempting to argue that, prior to the
tendency towards secularisation of leftist politics, the traditional
religious environment underwent periodic evolutionary sectarian
spasms in response to social unrest and upheaval among the lower
classes.
The upper and middle classes, being in possession of better
exchange ratios, i.e. with better education, more money, and more
social status, should therefore be less likely to be receptive to affiliating
with new religious groups, because they have no wish to change the
socio-political status quo. However, elites are not immune to the appeal
of new religious groups. Stark and Bainbridge explain this by
suggesting that, although members of the elite may be relatively
powerful at a general social level, they can still suffer individual
deprivations that their social position cannot change: for example, they
may seek health or beauty or love. And although they may be part of a
powerful class, they may not be powerful individually. 132 Stark and
Bainbridge note however that elites are more likely to be drawn to
innovative

cults

as

opposed

to

sects,

because

sects

generally

condemned or attacked the elite. Cults, on the other hand, do not


oppose the established church; and because they represent religious
innovation, those who adopt a new cult will have some characteristics
of the early adopters discussed in the previous chapter. Because elites
are more powerful, they have the resources with which to experiment,

131
132

Stark, Class, pp. 698-706.


Stark and Bainbridge, Theory, p. 235.

59

and fail, with religious innovation. Elites can also have wider ranging
social networks, making it easier for the cult to spread through them.133

The Trajectories of Religious Movements


Once they are formed, new religious movements can generally be
observed to progress in two ways. Innovative groups that maintain high
tension with their environment risk punishment or sanctions being
imposed upon them from outside. This encourages increased intragroup exchange, leading to the social implosion of the group; and the
resultant isolation limits the recruitment potential from external society.
However, groups such as these can continue to grow even when deeply
isolated,

both

through

intra-group

childbirth

and

through

the

recruitment of social isolates.134 Isolates are not considered useful


with regards to religious growth, as they do not bring an extended
social network that may be tapped for further recruits. Most religious
groups that do not reach a certain size maintain high tension and
isolation from the surrounding society, and eventually collapse and die
out. It is notable that there is an optimal size that will help to determine
the groups success or failure an observation corresponding to the
tipping point of collective action.
Alternatively, the tension of a religious movement with the
surrounding environment can begin to dissipate. This is due either to
increasing numbers of converts, whose participation help to make the
group the norm in that particular environment. Alternatively, tension
can dissipate through the process of social evaporation. This describes
133

Stark and Bainbridge, Theory, p. 236-7.


For example, the Amish population of Pennsylvania continues to grow despite the
fact that they do not admit outsiders, and have a 40% defection rate. The Mormon
Church, although isolated in many ways, has a strong proselytising element combined
with high rates of childbirth, making it still one of the fastest-growing religious
movements of modern times (see Stark, R., and Bainbridge, W. S., Networks of Faith:
Interpersonal Bonds and Recruitment to Cults and Sects, in AJS, vol. 85, no. 6, 1980,
pp. 1376-1395).
134

60

the process of defection by dissatisfied members, leaving a higher


percentage of members of the group who are satisfied. Stark and
Bainbridge note that the lower the tension of a religious group, the
more strongly will social evaporation tend to decrease the tension of
the group135 with the opposite also being true. From this they then
argue that the more socially powerful participants will bring the rest of
the group towards lower tension with the surrounding environment. It
is at this point that formerly stable members of the surrounding
religious environment would be drawn towards the previously hightension but now low-tension religious group because the group has a
greater level of social integration and it is here that social equilibrium
will be found.
Stark and Bainbridge argue that, unless it is very socially
isolated, a new religious organisation will recruit members rapidly at
formation, because there are wide availabilities of statuses within the
new group, as well as high exchange of compensators and rewards.
This proposition suggests that there is a crucial cut-off point in degree
of isolation. New organisations slightly too isolated, will become fully
encapsulated, become confirmed in their high tension, and fail to grow
or change. Organisations that are born in slightly less social isolation
will be able to grow rapidly and at least have the opportunity to evolve
in response to success.136 This is the equivalent of an innovation
starting at the lower or upper boundary in network theory terms the
connectivity of the social network drives the ultimate success or failure
of a religious movement. Successful religious movements expand their
numbers through existing social relationships, and generally move
toward lower tension with their surrounding environment. Most new
religious movements will not prosper, and will remain at high tension or
die out, once the extent of their social networks have been exhausted.

135
136

Stark and Bainbridge, Theory, p. 259


Stark and Bainbridge, Theory, p. 265.

61

Application to antiquity

Can the assessments of the sociology of modern religious movements


work across a greater time depth? In applying these theoretical
frameworks to the adoption and spread of religions in antiquity, there
are a number of caveats to take into account. Primarily, historical
research is limited by the amount of evidence that is available, and by
the impossibility of empirical study. Aspects of social status or religious
feeling may be seen or inferred from epigraphic records or literature,
but the problem always remains: ancient historians have only a) what
the individual in question wanted to present about themselves to a
wider public, and b) what archaeology has revealed so far.
In addition, there is the central methodological issue, which
cannot be fully explored here, of the dominant modern monotheistic
tradition in which the sociologists of religion make their observations.
Can the same conclusions concerning innovation and tension within a
monotheist society be drawn about polytheist society? A polytheistic
environment is fundamentally different from a largely monotheistic
environment, which only came into existence in the ancient world
during the fourth century AD. Polytheism understood the world through
a plurality of deities. Monotheism acknowledged one supreme deity and
the belief that one system should explain everything. Polytheistic
society should therefore be more accepting of novel religious forms,
and in fact, it might be impossible to think about religious conversion
as a concept to apply within a polytheist society. People could and did
dedicate to a number of deities, depending on many factors: social
background, position, status, time of year, geographical location,
occupation and so on. Sociological rules based in the observation of 20th
century religious movements within a monotheistic society should be

62

reconsidered when attempting to understand the generation and


adoption of new cult forms in the ancient polytheistic world.

Elites and the dominant religious tradition: polytheism &


monotheism
The reconsideration of sociological rules for polytheist society is far too
large a subject to tackle here, however, it will be useful to explore one
issue a little further: Stark and Bainbridge suggest that innovative cults
are less prevalent in an environment where a dominant religious
tradition is supported by the socio-political elite. This is because when
there exists a mutually supportive relationship between this elite
(essentially, the sector of society which can exert control over the
environment) and one hegemonic form of religious expression, the
opportunity for and acceptability of deviance from either of these
powerful factors in society is limited.
The dominant, polytheistic religious tradition of the Roman
Empire was supported by the elite. It is true that within this, some
conservative aristocrats and literati did complain about the new, nontraditional religions that were paraded through the streets of Rome
from Elagabal to Isis, Cybele to Mithras; and that tensions were felt at
certain times, manifested, for example, in the discrimination against
alien religious forms. Jews and Christians were persecuted, ostensibly
for their religious deviance but also, and possibly more usefully,
interpretable as being due to their political deviance and refusal to
conform to the social and political norms of the Roman Empire.
Equally, however, there are plentiful examples of elite adherence
to novel religious forms Judaism in particular (see Chapter Five); and
the fact that so many new and alien cults could flourish in Rome and
across the empire supports the suggestion that during this period, the
attachment of the elites to traditional religious forms that Stark and
63

Bainbridge postulate was beginning to decline. This may in part be due


to the nature of empire creation: as the sphere of what was considered
Roman expanded, and the social barriers to leadership lessened
(consider the emperors drawn from the ranks of the military) the
acceptance of other religious and cultural forms so increased. The
polytheist cultural and religious space shared by a multiplicity of deities
grew with the accommodation of new peoples, languages, and land
acquisitions, and it may even be possible to see religious innovation as
a response to the encounter with the Roman Empire. Irad Malkin 137 has
argued that what was seen as Greek only came into being as a result of
colonisation in the Archaic period. Likewise, Greg Woolf 138 has argued
that Roman culture was shaped as the Empire expanded: that the
quality of Roman-ness, or what was seen by Romans and their subjects
to be Roman, was only formed in response to the encounter with
difference. The social crisis that followed occupation and subjugation
could have prompted periods of intense religious development in a
similar fashion to the creation of Roman-ness. It is reasonable to
suggest that new religious forms were created by the process of
Romanisation to serve the new, universal environment to which the
empires inhabitants now belonged.
Although the pre-Christian Roman Empire ostensibly possessed
traditional religious mores supported by the elites, it can nevertheless
be viewed as an open, tolerant environment where a plethora of cult
forms could and did exist alongside each other. It was not until
Constantines conversion and the official labelling of the Roman Empire
as monotheist and specifically Christian, that this kind of supportive
relationship between the elite and a singular religious form took shape.
The often violent putting down of the heresy and schism that plagued
the early Church shows this relationship between the dominant elite
137

Malkin, I., Networks and the Emergence of Greek Identity, MHR, 18, 2003, pp.
56-74.
138
Woolf, G., Becoming Roman: The Origins of Provincial Civilization in Gaul,
Cambridge: Cambridge University Press, 1998.

64

and a hegemonic religious form quite clearly: and it is a situation


substantially different from the previous environment of polytheism.

The Pattern of Evidence and Social Networks


In an examination of whether these sociological observations of religion
in the modern period can be used for understanding religious
innovation

and

conversion

in

antiquity,

there

is

no

scope

for

ascertaining empirically the internal, psychological reasons people


might give for their adoption of a new religion. All we have are the
patterns left behind by their actions. That an actors rational decisions
lie behind the pattern is of course a given, as is the attractiveness of a
cult. However, without knowledge of what those choices and that
attractiveness were, it only remains possible to test the social network
properties

identified

by

Stark

and

Bainbridge,

and

attempt

to

reconstruct the social connections that introduced the religious


movements and influenced individuals reasons and rationality for
worshipping a particular deity or deities. We can only in exceptional
circumstances ascertain the individuals reasons for following a
particular religious movement, but we can try to understand something
of the reasons for the cults success or failure. It is, essentially a
pragmatic approach to the evidence that is available after the event.
To do this, an examination of ancient social networks is vital.
Harland distinguished five common types social networks that drove
religious or other associations. There were groups that drew primarily
on (1) household connections, (2) ethnic or geographic connections, (3)
neighbourhood connections, (4) occupational connections, and (5) cult
or temple connections. These sets of social linkages are often
interrelated with issues concerning the self-understandings or identities

65

of particular associations, and they also provide clues regarding the


economic and social standings of their members.139
My case studies will examine the power of three kinds of social
network to communicate religious ideas: occupational, in the case of
Jupiter Dolichenus; ethnic, in the case of the Jewish Diaspora; and
cultic, in the case of the cult of Theos Hypsistos.

139

Harland, P., Associations, Synagogues and Congregations, Fortress Press:


Minneapolis, 2003, p. 29.

66

Chapte r 3.
Methodology: Networks in Archaeology.
Analytical models

Introduction

Archaeologists need to be concerned with the operations of wider


systems that influence change in the material record, and they have
offered many interpretive frameworks for understanding them: from
world systems theory to the ideas of punctuated equilibrium drawn
from chaos theory. Understanding change in antiquity through the
theoretical approaches discussed in the previous two chapters is still
relatively new to the discipline. I suggest that an approach that
combines both the wider understanding of the network properties that
facilitate change, as discussed in Chapter One, and the social networks
that are fundamental to the transmission of religious innovation, as
seen in Chapter Two, can be usefully applied to religious data from the
Roman Empire. This chapter sets out the methods I will use in my
approach, and critically examines a few of the ways in which networks
have previously been used as analytical models for archaeological and
ancient historical data.
I explain how I shall use networks both as a heuristic approach
and as a modelling technique in application to my religious epigraphic
datasets. For my practical model, I use Proximal Point Analysis (PPA) as
a simple technique for examining the centrality and isolation of nodes
and areas in the network; more heuristically, I take the idea of the
social network of human interconnections as a facilitator of change, and
67

use the opportunities

for discovering personal information

that

epigraphic data presents us with to move away from physical and


network geography towards an interconnected social geography.
Following this outline of how I shall apply network methods to my
datasets, I briefly examine and summarise some of the contemporary
uses of the network model in archaeology that have influenced my
approach. A few archaeologists have begun to apply aspects of network
theory to archaeological data to focus on the interactions of the ancient
world. The examples outlined here are the use of Proximal Point
Analysis as a model for settlement growth and centrality in the Early
Bronze Age Cyclades by Cyprian Broodbank; the interrelationship of
dispersion/colonisation and connective networks and the emergence of
a universalised Greek identity in the Archaic period as proposed by Irad
Malkin; the more scientifically-driven utilisation of complexity theory
and power-laws to explain the emergence of power and hierarchy in the
Neolithic as detailed by R. Alexander Bentley and the collaborative
work of Carl Knappett with physicists Tim Evans and Ray Rivers for
understanding the interactions of sites and the costs of maintaining
those site interactions in the Aegean Bronze Age.
Finally, I will outline some problems: those inherent in epigraphic
analysis, those of network manipulation by individuals or groups, and
those arising from the particular nature of this study. The previous uses
of network models in archaeology have been concerned with longerterm social change, for example, the development of settlement
patterns, or technological change. Religious change in archaeology has
not been studied in this way before.

Networks in the present analysis

68

The decision to apply network theoretical methods to the epigraphic


data pertaining to various religious groups in the Roman Empire was
driven by the observation of Stephen Mitchell in his article on Theos
Hypsistos: The number of inscriptions for the cults of Zeus and Theos
Hypsistos is large and expanding rapidly. The geographical range which
they cover is huge, extending from Achaea and Macedonia to the
eastern parts of Asia Minor and to the edge of the Syrian desert, from
Rostov on the Don to the Nile Delta [] Hypsistos was one of the most
widely worshipped gods of the eastern Mediterranean world. 140 The
inscriptions were also noted to be extremely uniform. How should such
uniformity across such a broad and varied geographical area be
explained? My decision was to approach the problem by studying the
distribution pattern itself, rather than by looking for an intrinsic factor
in the cult that might explain its propagation. Instead of using a topdown method to focus on the inherent quality of a religious idea, I
undertake a bottom-up analysis of the physical location of epigraphic
data, the end result of the process of transmission: the social routes
that religious innovation moved across, the interactions between people
that drive religious change.
There are a number of reasons why inverting the approach in this
way is important for reappraising religious data. First, the approach is
egalitarian: each dedication, however simple, plays some kind of role in
the wider picture of a cult. Second, there is no assumption of centre.
Centres and peripheries arise from interactions on the network, not
from judgements made by archaeologists. Third, it visualises otherwise
invisible ideological connections between people and places, and allows
us to hypothesise on the broader movements of ideas and the routes
they took across the Mediterranean. By combining an understanding of
the general properties of networks that facilitate change with that of
the structure of direct and intimate interpersonal contacts 141 that
140

Mitchell, S., The Cult of Theos Hypsistos between Pagans, Jews, and Christians, in
Pagan Monotheism in Late Antiquity, Oxford: Clarendon Press, 1998, p. 99.
141
Stark, R., The Rise of Christianity, New Jersey: Princeton University Press, 1996, p.
20.

69

specifically drive the transmission of religious innovation, the changing


religious environment of the Roman Empire can be understood in a new
way.
My approach draws on previous applications of networks in
archaeology, both in theoretical and in practical terms. The practical
technique I have chosen for my network analysis is Proximal Point
Analysis (PPA), which is most suitable for the scope of the current
application. The technique and Cyprian Broodbanks application of it is
outlined below, where it is shown to be a useful tool for the visualisation
of centrality and isolation as a product of interactions on a network. My
analysis will rely on real-world data to allocate points, and although
there will always be new information to add to the distribution map and
a developed version of the model will need to be flexible, my
preliminary analysis of network connectivity will reveal centrality,
isolation and parochialism that should feed back into the conclusions
drawn from an analysis of the epigraphic data. Initially, an egalitarian
logic will be applied to all network nodes, which makes them all the
same size and shows no bias towards the data. Following this analysis
as a starting picture of the overall distribution pattern, points will then
be mapped with the aim of expressing something of the dynamic nature
of diffusion: by using appropriate developmental time periods; by
building in centres of gravity i.e. places with known large
communities, or simply by making a distinction in the power of a node
to attract new links, with places that have two or more pieces of
epigraphic data being more powerful than those with only single
epigraphic attestations.
The analyses are underpinned by the theoretical frameworks of
the previous two chapters and by some of the ideas that will be outlined
here. Irad Malkins approach to understanding the creation of identity
in an expanding colonial world is one that could have useful
implications for thinking about widely-distributed cults, and specifically,
about the Jewish Diaspora, for example by considering how much the
70

sense of Jewish identity in the Diaspora came from the process of


dispersion itself. Could this idea of a widely dispersed religious network
be applied to non-ethnically linked religious movements, such as the
cults of Jupiter Dolichenus or Theos Hypsistos? Malkin augments
historical sources with observable phenomena to identify points of
contact along the network lines. This is broadly similar to the way of
approaching cult through the epigraphic data supplementing the
primary evidence of objects and cult places with literary sources.
The more mathematically advanced analyses taken by the sociophysicists in Alex Bentley and Herbert Maschners volume and by Carl
Knappett, Tim Evans and Ray Rivers are somewhat beyond the scope of
this project and my own mathematical ability. Their analyses suggest
potentially illuminating methods for future investigations into religious
transmission using more advanced modelling techniques. However, it is
the two concepts articulated in Bentley and Maschners article on the
avalanche of ideas that shall be considered particularly relevant in this
study: that of change resulting from stochastic network growth through
emulation, and that of change as an emergent phenomenon caused by
self-organising properties of a network. However, the applicability of
these terms will be considered only after a full exploration of the
epigraphic data, and the examination of three types of network as
facilitators of the transmission of new religious information: the military
network and Jupiter Dolichenus; the ethnic network of the Jewish
Diaspora; and the religious network of the cult of Theos Hypsistos.
The reasons for my choice of these three religious groups in
particular are related to my involvement with Stephen Mitchells Pagan
Monotheism in its Intellectual Context project at Exeter. My initial task
was to further investigate the cult of Theos Hypsistos. This cult is
connected to the Jewish Diaspora, an examination of which was
therefore a natural progression. My decision to analyse the cult of
Jupiter Dolichenus as a comparative study arose as a result of my

71

involvement in the excavation of the site of Doliche in Turkey with the


Forschungsstelle Asia Minor team at the Universitt Mnster.

72

Previous network methods in archaeology

This next section outlines four uses of network models in previous


studies of archaeological subjects, the most detailed of which is the PPA
of Broodbank, the technique I have adopted because of its simplicity. It
is important to note that all these examinations are concerned with the
modelling of super-social patterns: of the interactions that led to
development of settlements in the EBA Cyclades; of the creation of
what constituted Greek identity through encounter with other; of the
emergence of hierarchy in the Neolithic; and in the expansion and
collapse of interactions between sites in the Aegean. My data are very
different. Epigraphy has the potential to be used to span and
incorporate both the macro- and the micro-scales of analysis: by
focusing on both the overall distribution patterns and on the individual
life the partial record of which is in the stone itself. This thesis aims to
synthesise the interpretations offered by these different scales.

Application One: Settlement growth in the EBA Cyclades


Broodbank142 uses Proximal Point Analysis to understand the growth of
settlement, centrality and interaction in the Early Bronze Age Cyclades,
factoring into his analysis problems such as distance, inter-island
visibility, and method of travel. He simulated interaction networks in
the Cyclades, before comparing his results to the actual material.
Proximal Point Analysis (PPA) is a simple and transparent 143
technique to predict patterns of interaction between distributed points.
It takes as a starting convention the assumption that each point
connects

with

those

three

points

142

nearest

to

it,

and

these

Broodbank, C., An Island Archaeology of the Early Cyclades, Cambridge:


Cambridge University Press, 2000.
143
Broodbank, Cyclades, p. 180.

73

interconnections then build networks. It is a gravity model, where


centres develop over time. The distribution of points in a real world
environment is generally uneven, so some points will collect more than
three links because of their geographical positioning because they
become the closest target for other points to connect to. The method
assumes that communities interact most intensely with their closest
geographical neighbours. It is important to note that PPA indicates
relative degrees of connection, rather than absolute presence or
absence. PPA has been used to explore centrality and isolation, by
theorising which points are best connected and where the most
effective communication routes appear to be. A strength of the method,
as Broodbank points out, is that the model has an inbuilt egalitarianism,
because all points can be evenly weighted and initially treated without
assumptions about hierarchy.144 It is a simple technique, but one that
offers a good unbiased starting point from which to explore ancient
networks, and one that makes no assumptions about the data.
Broodbanks field of analysis is the EBA Cyclades, where the
known data is quite unevenly spread. Some places have been
extensively surveyed, whilst others are relatively archaeologically
unknown. Broodbank suggests three ways of dealing with this
discrepancy, which is obviously not an uncommon problem in most
archaeological analyses. The first method assigns an arbitrary number
of points to an island or area, and follows fixed rules about the nature of
their interaction, dismissed by Broodbank as a technique that would
create an ahistorical network. Unless the points represent real
distributions of people, what emerges is a study not of how islanders
interact, but of how islands as geographical entities do.145 Physical
geography can only explain so much about human interaction patterns.

144

Broodbank, Cyclades, p. 181. This is certainly an extremely useful starting point for
simulative network analysis; however, in developing the technique to apply to realworld data, it may be equally important to factor in knowledge of hierarchies, gravity
and centres.
145
Broodbank, Cyclades, p. 181-2.

74

The second uses

known site information,

although unless

relatively complete this is problematic for settlement analysis because


of the discrepancy and accident involved in discovery of sites, and the
reconfiguration of the network that is necessary whenever a new site is
discovered. Another issue is that many sites may not have been
contemporaneous, and the network will therefore present a false
picture. However, Broodbank observes that if such information is
accessible, the scope for analysis is all but unlimited. 146 It is this
approach that I will be using in my PPA, the reasons for which will be
detailed below.
The third approach, and the one that Broodbank uses, is to take a
rough sketch of what is known from the data, and to simulate
distribution over the entire area, following certain rules. This bases the
model in reality, but also predicts possible areas of high archaeological
interest. If a sufficient sensitivity to contexts and changes can be
attained, PPA of island groups can transcend deterministic analysis, and
develop into a relatively subtle tool for investigating the many ways in
which culture and history may develop within, but also profoundly
rework, the configurational features of a given islandscape.147
Broodbank allocated one point to an island (or group of small
islands) per 150km2 of landmass, which worked out as about one point
per larger island. Population growth was simulated by lowering the
area in which a point was allocated first to 100km 2, then to 75km2, and
finally to 50km2. His four analyses simulated a population increase from
about 0.7 people to about 2.0 per square km, both of which fall within
the postulated population for the time.
The PPA analysis first showed that during the periods when
population was lower, the overall network was more homogeneous, and
as the population increased over time, sub clusters become more
clearly defined. Second, it showed that treating an island as an
individual entity is useless at this stage in Cycladic history, and that the
146
147

Broodbank, Cyclades, p. 182.


Broodbank, Cyclades, p. 183.

75

establishment of populations is better assessed in terms of networks of


mutually

sustaining

communities148

which

transcend

insular

boundaries. Third, as population increased, maritime links decline in


importance to the larger islands, which become more introspective and
self-supporting; by contrast, for the smaller islands it is increasingly
necessary to maintain them. This ties in with the fourth observation,
that isolation can mean two things: first, the relative difficulty in
making and maintaining contact with others, called by Broodbank
remote; and second, the self-sufficiency that comes with dense
occupation, called parochial.
Isolation was measured by physical distance from neighbouring
nodes: all points on average over 20 km (this is above the one-day, oneway travel range that Broodbank set) from their nearest neighbour
have the potential to be classed as remote, and those points whose
average distance to their nearest neighbour is 10 km or less (within a
one-day there-and-back range) have the potential to be classed as
parochial. Reciprocity is factored into understanding the network by
making a rule that identifies areas as remote as being only those that
need outside contacts more than outsiders need them, as identified by
points at least one of whose three links represents a non-reciprocal
connection, and adds that an areas remote status increases in
extremity if this applies to two or three of its links.149
Many of the smaller islands were found to be remote, but
surprisingly, so was the larger island of Syros. Broodbank suggested
inherent nodal factors that might account for the smaller islands
remoteness, including lack of mineral resources or arable land, which
would make them unattractive for larger islands interaction networks.
Syros, however, possesses fertile land, and theoretically, should be easy
to reach. What is more, the remoteness of Syros was marked and
continuing through the sequence, and highlights the value of the

148
149

Broodbank, Cyclades, p. 187.


Broodbank, Cyclades, p. 191.

76

method for overturning intuition, which may be false. 150 The PPA
illustrated areas of interest and anomalies in the ancient network,
which would not have always been intuitively predictable.
Broodbank therefore dismisses these inherent, nodal explanations
for network centrality that might be intuitively postulated: arable land,
mineral resources, or good harbours. He points out that specialisation
in response to agricultural poverty can be applied across most of the
Cyclades, that none of the major sites are close to mineral resources,
and in fact are often at some distance from these areas, and that good
harbours are largely irrelevant at this period because of the use of
canoes.151 Environmental determinism for the emergence of settlements
is

therefore

strongly

denied

as

causal

explanation.

Instead,

Broodbank concludes that it is the interactions between land and sea,


and island and settlement groups, which should be considered the
driving force of settlement centrality. Centrality can be measured in two
ways, both through the efficiency of interaction, i.e. the short path
lengths between sites, and by intensity of interaction, marked by the
number of links to other sites, i.e. the hub status of a site.
The PPA method can visualise growth, development and decay
through analysis of connectivity. Broodbank used the model to
investigate reasons why settlements emerge in the places they do, and
whether current knowledge of sites is roughly in accordance with the
simulated model. He asked why settlements that can be classed as
centres emerge in EB II, and specifically, why the Grotta-Pelos culture
shifted its deposition from Paros to other islands. He reasons that
centrality in the interaction networks of the Cyclades drives settlement
location.152
The simulation of population growth in the EBA Cyclades matched
50% of the archaeologically known centres. What this tells us is that
the changes in local interaction networks resulting from population
150
151
152

Broodbank, Cyclades, see p. 193.


Broodbank, Cyclades, see p. 237-8.
Broodbank, Cyclades, see p. 237.

77

growth have a fundamental effect on the location of communication


centres.153 Further exploration of the three sites the model missed
reveals this is more successful than it might seem at first.
A centre was predicted in the north of the island of Tenos, where
there is no known settlement. There is, however, a large settlement at
Chalandriani-Kastri, on the northeast edge of the remote island of
Syros. Broodbank suggests that the centre which ought to have
emerged

at

Tenos

was

suppressed

by

the

Chalandriani-Kastri

settlement, or alternatively that the centre for this part of the Cyclades
was consciously relocated to Syros. He reiterates that relocation would
not be for reasons internal to Syros, but rather explains it through the
visual control the position affords for the surrounding seascapes. What
Chalandriani-Kastri [] sought to control was not Syros or Keros, nor
neighbouring islands per se (they probably had little to do with the far
sides of Andros, Tenos and Mykonos, or Naxos, Amorgos and Ios), but
rather the maritime and coastal islandscapes, complete with their
traceries of peoples movements and other activities, that stretched out
in front of them.154 Psychological control as a factor in certain strategic
locations is highlighted convincingly as a result of this kind of network
analysis.
Another simulation miss was Agia Irini on Kea. However, this
site, situated close to the mainland Peloponnese, had a different
function within the interaction networks. Instead of being a locally
networked site, Agia Irini acted as a gateway into the Cyclades from the
mainland, forming a centre at an inter-regional scale. However, access
to and importance within different networks cannot explain the
analytical miss of Skarkos on Ios. This stands out as the only large
settlement that cannot be convincingly interpreted as a communication
hub. Combined with the doubts about its overall similarity to the other
major EB II sites, this strengthens the possibility that Skarkos was

153
154

Broodbank, Cyclades, p. 239.


Broodbank, Cyclades, p. 242-4.

78

indeed a different kind of community with different reasons for


existing.155
On the whole, Broodbanks PPA simulates and predicts a network
of sites and attachments which is not dissimilar to what is known from
real world data. What it misses, however, are sites with cross-network
positions, and sites that seem to have a different kind of function. At
this period in history, the analysis shows that the crucial scale of
centrality has proved to be that of the local networks, suggesting that
all these centres,

save Agia Irini,

emerged through

bottom-up

processes, rather than as the result of top-down, long-range factors.


This

need

not

imply

that

these

centres

subsequently

confined

themselves to the scales of activity to which they owed their origins. 156
As a combined predictive-explanatory model, the simple technique of
the PPA helps to fill in data gaps and highlight potential areas of
interest.
Because of the different nature and different aims of my study,
some of the predictive elements of Broodbanks PPA are not relevant. I
am not attempting to simulate where we might look for new epigraphic
finds; rather, I am attempting to ascertain something of the connections
that led to the distribution patterns of religious data, i.e. the records of
individual religious choices, and why they were made. Of the three
methods Broodbank describes, the second is the therefore most
appropriate to my data: using real-world information. Although the
problems of discrepancy, accident of discovery, and the possibility for
reconfiguration are of course to be acknowledged, the amount of
available data still present a good opportunity for this application of
this real-world approach. One of the particularly useful observations
that Broodbank makes is that of the differing types of isolation:
remoteness and parochialism. These kinds of network configurations
will appear in my PPA analyses, and, because they will refer to actual
known data, highlight areas of interest in terms of the quality of the
155
156

Broodbank, Cyclades, p. 244.


Broodbank, Cyclades, p. 245.

79

religion in those areas. Broodbank uses PPA to analyse a possible


demographic change from a modelling perspective, before comparing it
to actual data. I invert this sequence, analysing the actual epigraphic
data first, before using a PPA to model the possible interconnections to
compare with the conclusions drawn from the epigraphic analysis.

Application Two: Networks and Identity


Malkin157 has used theorised interactions to challenge assumed
hierarchies of centre and periphery in his analysis of the creation of
Greek identity. His focus is the early colonisation movements during the
Archaic period, where, he argues, what constituted Greek-ness was
still being defined. Many different aspects of identity were involved in
the process regional, colonial relations to the mother city, linguistic,
religious and ethnic but that it was the confrontation of the colonising
Greeks with new people and cultures in the places they colonised that
articulated how Greeks saw themselves and which therefore drove the
formation of what came to be understood as Greek identity.
Using the notion of difference from an alien other as a
mechanism of group self-definition is a commonly used model for
colonisation. Malkin argues the case of the co-creation of notions of
self with that of other with regard to the Archaic period: the
commonalities of the colonial experience, combined with the extended
and increasingly varied geographical horizons, had taught Greeks that
the variety among them was far less than that which they were
encountering overseas. Awareness of sameness occurs not when
people are close to each other (in fact, that is when they pay particular
attention to their differences) but when they are far apart. It is distance

157

Malkin, I., Networks and the Emergence of Greek identity, in MHR 18, 2003, pp.
56-74.

80

that creates the virtual centre. The more the connecting cables are
stretched, the stronger they become.158
Malkin

observes

that

the

experience

of

colonists

in

new

landscapes and environments, and the contact with new people and
languages led to the definition of what it was to be a colonist through
the recreation of known sacred landmarks and reproduction of certain
civic monuments. It was this cross-colonial fertilisation that began to
exemplify Greekness, and percolated back to the old so-called
centre,159 informing the creation of what and who was Greek and what
and who was non-Greek. He argues that these kinds of definitions were
needed to express the new hyper-ethno-linguistic identity that was
exemplified by the colonial experience. 160 The further use of mythology
and founding stories created an intellectual network that began to view
itself as of equivalent oldness in the colonies as in the metropoleis.161
One of the central observations Malkin makes is that the
direction of the flow on the network could and did change over time.
Initially, the interactions of mother city and colony were symbolically
cast

in

kinship

terms,

but

in

reality

the

multidirectional,

accumulating links are more significant than the true origins. 162 It is
the interactions on the network that build the identity of the nodes.

Application Three: Complexity Theory and Archaeology


Bentley and Maschner163 et al. have taken a more mathematicallyscientifically modelled approach to the application of complexity theory
to archaeological problems. They write in their introduction to the
collection of papers in Complex Systems and Archaeology that a goal of
158

Malkin, Networks, p. 59.


Malkin, Networks, p. 71.
160
Malkin, Networks, p. 63.
161
Malkin, Networks, p. 65-6.
162
Malkin, Networks, p. 67.
163
Bentley, R. A., and Maschner, H. D. G., eds., Complex Systems and Archaeology,
Salt Lake City: University of Utah Press, 2003.
159

81

complexity theory is to discover how the movements at a small scale


translate into emergent phenomena at a larger scale, or, if that is not
possible, at least what emergent properties can be expected.164 The
collected volume brings together archaeologists and anthropologists
using networks at different levels, both highly mathematically, such as
in the work of Bentley himself, and also those using complexity theory
more

as

methodological

starting

point

for

understanding

archaeological phenomena.
Bentleys research is particularly focused on the phenomenon of
scale-free networks and the creation of power laws. He uses the model
of scale-free networks to identify power laws of inequality in Neolithic
Europe, and in the lengths of long barrows in southern England. He
plots the length of barrows, and argues that what can be seen is a curve
beginning with normal distribution, but having the graphic tail of a
power law. He uses this observation to argue the case that over time,
the rich seem to become richer with these Neolithic long barrows,
which suggest that status in Neolithic Wessex accumulated within a
scale-free network.165
An issue that might be raised here is Bentleys discovery of power
laws within many different environments: whilst it is helpful to be able
to discuss increasing inequality in a society, does it really count as a
power law when, for instance in the Neolithic barrow example, the
curve starts off following a different distribution? This does not mean
that the use of the model is not illuminating, just that care must be
taken in identifications and the conclusions that are drawn from them.
When a power law can be identified, Bentley concludes, it may offer an
insight into the transition to new forms of society, especially through
contact with other groups.166

164
165
166

Bentley and Maschner, Complex Systems, p. 5.


Bentley, R. A., Scale-Free Network Growth, in Complex Systems, p. 41.
Bentley, Scale-Free, p. 42.

82

More

relevant

here

is

Bentley

and

Maschners

work

on

information cascade, avalanches of ideas in the volume. 167 They


explore two explanations for the phenomenon of large-scale and sudden
change that might well be applicable to my data. The first is that of
events occurring within a constantly growing, interconnected network,
in that one triggering event may give rise to multiple consequent
events:168

an

approach

that

can

incorporate

contingency

and

punctuated change. This model is arboreal, in that all subsequent


events can be linked to a single original event. The other explanation
they explore is that of self-organised criticality, which is less
deterministic than the network growth model,169 and has the potential
to capture more complexity.
They use the first model of stochastic network growth to explain
power laws found in citation curves and to explain the spread of ideas
on Neolithic pottery. They examine the case study modelled by Bentley
and

Shennan170

that

implies

that

pottery

styles

were

copied

preferentially from nodes that were already well-connected hubs and


these were the prestigious households. Horizontal transmission of
ideas, emulation and prestige are key factors in this explanation of
mass change. They make the important observation that in studying
cultural

evolution,

we

may

need

to

place

more

emphasis

on

transmission rather than focusing exclusively on natural selection.171


The second model, that of self-organisation, offers a different
explanation for how cultural change is enacted. This is a rhizomatic
model, where change is a result of a slowly-driven, interactiondominated

threshold

system

(Jensen

1998:126)

that

naturally

gravitates toward a critical state in which some perturbations only


trigger small changes, while others can cause an avalanche of
167

Bentley, R. A., Maschner, H. D. G., Avalanches of Ideas in Complex Systems, pp.


61-73.
168
Bentley and Maschner, Avalanches, p. 61.
169
Bentley and Maschner, Avalanches, p. 62.
170
Bentley, R. A., and Shennan, S. J., Cultural Transmission and Stochastic Network
Growth, in AA, Vol. 68, No. 3, 2003, pp. 459-485.
171
Bentley and Maschner, Avalanches, p. 67.

83

consequent events.172 They use the model as a useful one for describing
technological evolution, which tends to be a punctuated process.
Thresholds and low-level interactions are key to this kind of transition.
Both of these explanations for the phenomenon of sudden religious
change can be found in antiquity, and will be examined in my own case
studies.

Application Four: Networks and Relational Space


Knappett (in archaeology) and Evans and Rivers (both theoretical
physicists) have collaboratively developed the network model in their
use of complex mathematics as a way to understand the articulation of
the

physical

and

relational

dimensions

of

regional

interaction

networks173 in the Bronze Age Aegean. By building a computer


programme where parameters can be changed, they have run
simulations of expansion and collapse in the interactions between
Crete, the Cyclades and the mainlands of Greece and Asia Minor. This
allowed them to test the hypothesis that challenges site centrism, by
treating sites as secondary and the interactions between them as
primary,

and

to

argue

that

the

interactions

themselves

might

contribute to the size and status of the sites in question.174


One of the most important parameters that Knappett, Evans and
Rivers set is that of cost of interaction and the consequences of cost.
Long distance links are more likely to be maintained if a node is large
and wealthy, with the opposite also being true. This forces us to realise
that if a network is indeed created over a large asymmetrical grid of
this kind, then large sites are likely to feature. Furthermore, large sites
172

Bentley and Maschner, Avalanches, p. 67.


Evans, T., Knappett, C., Rivers, R., Using statistical physics to understand
relational space: A case study from Mediterranean prehistory, in Complexity
Perspectives on Innovation and Social Change, in D. Lane, D. Pumain, S. van der
Leeuw and G. West, (eds.) Berlin: Springer, 2007, p. 19 (forthcoming).
174
Evans, Knappett, Rivers, relational space, p. 2.
173

84

searching for information about resource availability are much more


likely to target other large sites in that quest. This means that
gravitational pull needs to be taken into account also when we examine
such networks the tendency of like to seek out like.175
In developing their network analysis from Broodbanks PPA
model, they further incorporate asymmetry, directionality, and the costs
involved in maintenance of interactions. This model begins to be able to
manipulate some of the social conditions of the network. I attempt to
consider some of these issues in my own analyses, but a full-scale
mathematical analysis is beyond the scope of this thesis. However, I do
factor into my network analyses (albeit un-mathematically) some of the
principles they outline, such as that of gravitational pull.

Some issues relevant to the current analysis

We have seen that network theory (in various forms) can be applied to
the spread of certain phenomena in the ancient world, but none of
these is religious, even in a broad definition. The combination of
network analytical techniques with those of the sociology of religion is
therefore an important augmentation to this subject.
The development of powerful computing technologies has made
possible analyses of datasets that were previously prohibitively large.
There are, however, some problems with the application of network
theory to archaeological data. Bentley admits that the identification of
network connections in the archaeological record is much more
challenging than with modern data, and that we have no hope of
quantifying them with the degree of detail possible in graph-theoretical
models or for modern datasets such as for co-authors. Nonetheless,
they existed, and the small world and scale-free network model still
175

Evans, Knappett, Rivers, relational space, p. 9.

85

offer qualitative insights for prehistoric networks of agents and


connections.176
A major issue is the fact that there are only limited amounts of
data and these present only limited amounts of information about
individuals and their connections. There are also wider factors to
consider that are relevant to the period in question here: the
epigraphic

habit

of

the

first-second

centuries

AD,

the

active

manipulation of the network, and the structures that formed part of the
Roman Empire, including global features such as the Imperial cult or
the trappings of universal civilisation, architectural uniformities and
civic institutions, such as bath-houses.

Literacy and the Epigraphic Habit


The term the epigraphic habit has been coined to describe the
enormous rise in inscriptions that is observable in many parts of the
Roman Empire between the first and third centuries AD, especially
between AD 50 and 250. It seems that the pattern is related to the
Romanisation of the Empire and to the articulation of social status, but
MacMullen warns against using this patterning of the data to support
grand theories, for example, that of the decline of Rome. 177 Epitaphs
form by far the largest body of epigraphic data. 178 The growth in
funerary memorial inscriptions during the Roman period is marked also
by the naming of the commemorator, as well as the deceased.
Meyer argues that one of the main reasons for the erection of
funerary monuments was the proclamation of status 179 and that interest
in affirming status varies according to the value of the status in relation
176

Bentley, R. A., Introduction to Complex Systems, in Bentley and Maschner,


Complex Systems, p. 19.
177
MacMullen, R., The Epigraphic Habit in the Roman Empire, AJP 103, 1982.
178
Meyer, E., Explaining the Epigraphic Habit in the Roman Empire, JRS, 1990, p.
75.
179
Meyer, Epigraphic, p. 83.

86

to the rest of the population, going some way to explaining the variation
in the habit over time. In the earlier periods of the Empire, status was
more coveted and more of a privilege, and as such, records of it were
part of an agenda of self-aggrandizement linked to the mode of
expression through memorialisation. Meyer argues that this trend
varied across the Empire, and the East was apparently less concerned
with expressions of Roman citizenship and status, although Mitchell
notes that the relative rarity of Roman citizenship in eastern cities
probably indicates that it was still a genuine privilege, especially in the
first century AD.180 Following the granting of universal Roman
citizenship by Caracalla in AD 212, the status value of that citizenship
diminished considerably, and the marking of it in epitaphic form
apparently declined.181 A secondary aspect of the increased volume of
epigraphic monuments is the implication that a larger proportion of the
population could read them.

Active vs. Passive network manipulation


Studying the networks that facilitated religious change is the study of a
process that is not necessarily deliberately directed; however, it quite
often can be. Although the examination of the processes of religious
transmission has great potential to illuminate passive aspects such as
centrality and isolation, what it does not fully take into account is the
potential for the active manipulation of the network by individuals or by
groups. Mission and missionaries have contributed fundamentally to the
spread of Christianity, as exemplified in the early period by the active
missions of the apostle Paul. Sophists and rhetors played an important
180

Mitchell, S., Pers. Comm.


However, it is notable that following the universal grant of the Constitutio
Antonina, the name Aurelius was extremely widely adopted and widely recorded,
especially in Asia Minor, as a marker of the individuals Roman citizenship. This too
might stand in opposition to Meyers claim that an interest in marking citizen status
declined at this point.
181

87

role in the communication of information and ideas in the ancient


world. Traders with investments in certain innovations or technologies
would also have had reasons to actively manipulate the network.
Charismatic individuals have the power to direct or attempt to direct
action on a network; in fact, they may sometimes possess hub status
as Paul is argued to have done by Barabsi in his introduction to
Linked.182 Although locating these individuals in the historical record is
virtually impossible, a future direction of network research might
attempt to simulate the effects of hub-like charismatic individuals on a
network.

State Mission
The coercive power of the Roman state is another dimension of network
manipulation that is relevant to understanding the real world situation.
The centralisation of control of taxes and laws, among other trappings
of Imperial governance, imposes a hierarchical structure onto the
environment. When the manipulation of the network by a centralised
power becomes too strong, the ability of the network to be freely
dynamic is limited. As Stark and Bainbridge pointed out (see Chapter
Two), when religion and state join forces, their combined power to
manipulate the environment is hugely increased.
The Roman Imperial cult is a good example of a widely distributed
cultic phenomenon, centrally driven by a state mission to create a
level of respect and obedience to the Roman leadership, manifest in
religious form. It also served to formulate a dialogue of patronage and
mutual benefit on the provinces, and as a tool of the Empire acted as an
imposing reminder of the might of the Emperor and Roman ideals.
Whether or not people actually believed that the Emperor was a god is
almost entirely arbitrary they may well have done. The refusal of early
182

Barabsi, Linked.

88

Christians to sacrifice to the Emperor may indicate some kind of belief


on the part of those who did offer sacrifice; but equally, the punishment
meted out upon the refusenik Christians is just as likely to have been
because of their apparent objection to Roman rule, rather than their
non-belief in the divinity of the Emperor.

These various issues of the epigraphic habit, network manipulation, and


the force of the Roman state will impact on the ways we interpret
ancient data. Keeping them in mind, we now turn to the application of
network analytical methods to my three specifically religious case
studies: the cult of Jupiter Dolichenus, as an investigation of the role of
the military network in propagating religious innovation; the network
formed by the Jewish Diaspora, to examine the role shared ethnicity
may have played in communication of new religious information; and
finally, the evidence for the cult of Theos Hypsistos, my original starting
point, and the assessment of a specifically religious network.

Part ii
application
89

Chapte r 4.
The Cult of Jupiter Dolichenus.
Military networks on the edges of empire

Introduction

In this chapter the cult of Jupiter Dolichenus is used as a test case for
the diffusion of innovative cults across the Roman Empire within the
milieu of polytheism. The aim is to illuminate the role military networks
play in cult transmission. Through an assessment of the epigraphic
evidence and the pattern it forms, combined with the diffusion over
time, something can be understood about the connections between
people who worshipped Jupiter Dolichenus, where and why they
adopted the cult, and the factors involved.
The Roman cult of Jupiter Dolichenus is distributed widely across
the western empire, especially along the northern frontiers in Germania
and Britannia, like other oriental deities such as Mithras or the Dea
Syria. This patterning of the evidence alone implies that Jupiter
Dolichenus is a different manifestation from the Bronze and Iron Age
forms of the deity from which he descends; but a brief examination of
the origins and forms of the cult and historical developments in Syria
prior

to

the

Roman

conquest

will

deepen

the

picture

of

the

transformation of the cult in the Roman period.


The explanation for the westward diffusion of the cult has always
centred on the participation of legions in eastern campaigns, a
connection with Syrians in the Roman army, and the zealous Syrian
traders who influenced its transmission. 183 This case study explores
183

Cumont, F., Oriental Religions in Roman Paganism, New York: Dover Publications,
1956 (originally published 1911); Merlat, P., Rpertoire des inscriptions et monuments

90

these suggestions and proposes instead that the cult was diffused via
the military network of the officer class. The analysis of the epigraphic
material on its own terms shows that the traditional suggestions for the
spread of the cult are inadequate, and demonstrates the mobility of the
officers in the Roman army. Following this, network analyses are used
to visualise the evidence and support this hypothesis, beginning with a
Proximal Point Analysis (PPA) and then mapping the cult development
over time through four PPA networks in chunks of fifty years.
Finally, an investigation of the role of priests and the militarycivilian interface adds depth to understanding of the spaces, both
physical and cognitive, for cults in Roman military society, and the
interaction between military and civilian on the borders of the Roman
Empire.

Origins

This section locates the origins of the cult of Jupiter Dolichenus within
an intellectual and religious framework. Jupiter Dolichenus is the name
given to the Roman manifestation of the Near Eastern storm god from
the town of Doliche, now in southeast Turkey. The plurality of sky-storm
deities of the region were known variously in different earlier periods
and places as Adad (Early Bronze Age Mesopotamian), Hadad (BronzeIron Age Semitic), Teshub (Bronze Age Hurrian/Hittite), Tarhunzas
(Bronze-Iron Age Luwian), or Baal Shamin (Bronze Age Semitic). It is
not suggested that these deities were identical, or syncretised by the
people who worshipped them, rather that there is an observable
regional theme of an all-powerful weather deity, whose individual
characteristics blend with and differentiate from each other in multiple
figures du culte de Jupiter Dolichenus, Paris: P. Geuthner, 1951; Speidel, M, The
Religion of Iuppiter Dolichenus in the Roman Army, EPRO 63: Leiden, 1978.

91

ways. As Millar notes with regard to the cult of Atargatis at Hierapolis,


it is futile to try to define exactly what the deity, or her cult, really
was. In attempting to do so modern observers have a painful tendency
to show no more logical self-awareness than ancient ones, and to forget
that ancient deities were whatever different things observers or
worshippers chose to regard them as.184
It is Hadad/Zeus Hadados that is found in the region of Doliche.
From the later Iron Age to Roman periods, the worship of Hadad/Zeus
Hadados in the Near East was focused on a number of temples aside
from Doliche: Aleppo, Hierapolis/Membij (although subordinate to his
consort, Atargatis by this period, possibly always), and Damascus. 185 In
these temples he is the Hadad of that locality, and many of these places
provide evidence of some regional continuity between the Iron Age and
Roman periods.
It seems that during the Imperial period, however, the Hadad of
Doliche underwent a transformation from a major west-Semitic deity
whose worship was essentially fairly localised, into the pan-Roman
Jupiter Dolichenus. The Dolichenus epithet is unequivocally linked to
this period. Why was it that the cult of Doliche spread so profoundly
across the western Empire?

Historical background of the region


The area of southeast Turkey and northern Syria from where the cult
originated borders the Syrian Desert and formed both the western edge
of the various Bronze and Iron Age empires that were largely focused
on the fertile crescent of Mesopotamia and Iran, as well as the eastern
fringe of the Hittite Empire focused on Anatolia and of the post-Hittite
184

Millar, F., The Roman Near East, Cambridge, Mass: Harvard University Press,
1993, p. 244.
185
Bunnens, G., The storm god in Northern Syria and Southern Anatolia from Hadad
of Aleppo to Jupiter Dolichenus, in Hutter, M., and Hutter-Braunsar, S., eds. Offizielle
Religion, locale Kulte und individuelle Religiositt, AOAT 318, 2004, p. 65.

92

Luwian-Aramaean culture labelled variously as Late Hittite or SyroHittite.186


The Bronze Age centres were mostly destroyed, and some were
never reoccupied. The region does not appear to have suffered major
cultural or ethnic disjuncture however, even with the destruction and
incorporation of the Iron Age centres into the Assyrian empire in the
late eighth century BC.187 This lack of visible break is argued to indicate
a new ethnic group, the Aramaeans. 188 The Assyrians settled extensively
in both fortified palatial structures as well as across the landscape,
testified to by the evidence for a multitude de petits tablissements
ruraux.189 The subsequent Achaemenid Empire unified a number of
different states and people into one system.190 Grainger deduces from
the lack of archaeological evidence from this period that taxation and
foreign rule had left the region with few cities, and that the population
was all but exclusively rural. This was in a country which had been
thickly sown with cities four centuries before. 191 The lack of evidence
usually means the period has been defined archaeologically as between
post-Assyrian

and

pre-Hellenistic,

and

has

been

interpreted

as

representing a time of depopulation, supported by the archaeological


absences in some sites that were important in the earlier Iron Age, such
as Aleppo, Carchemish, Tell Rifaat and Hama, which were all relatively
unused in this period.192 Foreign imports, such as numismatic material
or Attic pottery, have been used to date archaeological layers to this
time.
186

Novk, M., Arameans and Luwians processes of an acculturation, in W. H. van


Soldt, ed., Ethnicity in ancient Mesopotamia, papers read at the 48th Rencontre
Assyriologique Internationale, Leiden 1-4 July 2002, Leiden: Nederlands Instituut voor
het Nabije Oosten, 2005, pp. 252-256.
187
Akkermans, P. M. M. G., and Schwartz, G. M., The Archaeology of Syria,
Cambridge: Cambridge University Press, 2003, p. 361.
188
Novk, Arameans, p. 253.
189
B. Lyonnet, La prsence achmnide en Syrie du Nord-Est, in Briant, P. ed.,
Larchologie de lempire achmnide: Nouvelles recherches, Persika 6, Paris: De
Boccard, 2005, p. 131.
190
Kuhrt, A., Sherwin-White, S., From Samarkhand to Sardis, London: Duckworth,
1993.
191
Grainger, J. D., The Cities of Seleukid Syria, Oxford: Clarendon Press, 1990, p. 7.
192
Grainger, Syria, p. 27.

93

With the results of new archaeological research in the region this


view is changing, and as Millar argues,193 lack of evidence does not
necessarily mean that there were no people. Graves from Deve Hyuk
and Tell Ahmar testify to garrisoned Persian soldiers along the
Euphrates, and the Royal Road ran south of the Tur Abdin through
Harran to Carcemish, as a major trade and communications route.
Nomadic or semi-nomadic populations are more difficult to locate
archaeologically, but certainly existed in the area.194 Fuensanta and
Charvat also support the view that the Achaemenid presence in the
area of the Euphrates-Tigris was of a commercial nature, especially
connected with the control of the iron resources of the region.195
If

there had

been genuine

decreased inhabitation

of

the

Achaemenid landscape of Syria, the sanctuaries surely would have


undergone decline also. This does not appear to be the case, and
Bambyce/Mabbog (Hierapolis) was still minting coins during the time of
Alexanders conquest,196 providing evidence for its retention of its
status as a cultic and political centre. These coins bear, moreover,
Aramaic legends referring to various priests of the cult, including one
called servant of Hadad,197 which make explicit the indigenous
Aramaic culture of the area. The major Hadad sanctuaries of Aleppo
and Doliche also have no particular archaeological evidence for
disjuncture or discontinuity from the Iron Age to Hellenistic period, and
Mazzoni and others argue by contrast that the Persian period saw a
flourishing of these cult centres as supra-regional sites.198 It has been
suggested that after the collapse of the pre-Achaemenid Iron Age city
states there was a shift towards new cult centres. Without a picture of
193

Millar, F., The Problem of Hellenistic Syria, in Kuhrt, A., Sherwin-White, S., eds.,
Hellenism in the East, London: Duckworth, 1987, pp. 110-133.
194
Lyonnet, achmnide, p. 143.
195
Fuensanta, J. Gil, Chavet, P, Birecik achmnide et lage du fer IIIB dans le sud-est
anatolien, in P. Briant, ed., Larchologie de lempire achmnide: Nouvelles
recherches, Persika 6, Paris: De Boccard, 2005, p. 152.
196
Grainger, Syria, p. 26.
197
Millar, Syria, p. 126.
198
Mazzoni, S., Temples in the City and the Countryside: New Trends in the Iron Age
Syria, DM 13, 2002, pp. 89-99.

94

the transformation of the sacred landscape, it is difficult to retain the


picture of decline or change in the general population under the
Achaemenids.
Undoubtedly, however, Alexanders conquest and the subsequent
Macedonian rule shifted the (elite at least) focus of the region towards
the Greek-speaking world. The Seleukid Empire retained the scope of
the Achaemenid Empire before it, and with it the core administration
centres in Mesopotamia and north Syria.199 Seleucus I certainly founded
or re-founded a number of cities in north Syria c. 303 BC, including
Antioch, Apamea, Seleucia and Laodicea, which Grainger argues marks
the beginning of a re-urbanisation process of the rural population
native people who would have been hostile to the Macedonian
settlers.200 Iron Age tells were re-used as the acropoleis for Apamea and
Beroia, and the Greek or Macedonian settlers were given substantial
portions of land.201 The (re)creation of these centres in north Syria
changed the appearance of the landscape in physical and social terms,
but as Millar points out, only in this area.202 The rest of Syria was not
subject to similar mass-colonisation, and the survival of non-Greek
culture is evident across the whole region.
In terms of the cult of Doliche, what is noteworthy is that there is
no dateable Hellenistic evidence for the cult under the name
Dolichenus. The cult existed, with the god addressed as Zeus, Hadad, or
Zeus Hadados, and probably also Theos Epekoos, a popular epithet in
the area, but there is no evidence for the cult of Dolichenus per se. It
is reasonable to conclude that the name of the town of Doliche itself
became the focus for the Roman cult only, and prior to this, the deity
was addressed as the local Baal, Hadad, being given the Greek name
Zeus and so making explicit the point that the people engaged in his
worship were both native Aramaeans as well as Greek colonists. With
the arrival of the Romans, the god naturally also became known as
199
200
201
202

Kuhrt & Sherwin-White, Samarkhand.


Grainger, Syria, p. 110.
Grainger, Syria, p. 111-113.
Millar, Syria, p. 116.

95

Jupiter, and the affixing of the Dolichenus epithet makes it clear that
the site of Doliche, as opposed to Aleppo or Hierapolis, was the local
focus for the Romans in the region. What reasons were there for this?

96

Doliche
The town of Doliche, about 10km north of the modern Turkish city of
Gaziantep, could have been one of the four cities of Commagene
mentioned in an inscription on a bridge in the north of Commagene
built during the reign of Septimius Severus.203 Hellenistic-Roman
Doliche was situated between the important centres of Edessa, Harran,
Samosata and Antioch and major roads passed through it, as indicated
by the Roman milestones from the area.204 In the valleys to the north of
the town, two Roman bridges also indicate the direction of the roads
leading to Samosata, probably an important Commagenian city and site
of the later Roman legionary fortress for XVI Flavia Firma.205 Remains
of the Hellenistic-Roman town are seen in the quarries, the double cave
Mithraia, and various Latin and Greek inscriptions and architectural
fragments. There are two surviving rock-cut churches, and the town
was a bishopric until the 11th century AD.
The sanctuary to the god who became known as Jupiter
Dolichenus206 was located on the summit of Dlk Baba Tepesi, a
prominent hill c. 1211m high, to the south of the town of Doliche. The
hill is visible for at least 20km in every direction, and from some places,
for example, the Roman temple at Ksk in the hills to the northeast, this
distance is closer to 50km. It is possible to see the Tepe from the citadel
of Aleppo,207 and, on a clear day, one can see from the hill itself the
distinctive sanctuary-tumulus of Nemrud Dag, c. 150km away to the
northeast.

203

Butcher, Kevin, Roman Syria and the Near East, London: The British Museum
Press, 2003, p. 114.
204
See also the Tabula Peutingeriana.
205
Knox MElderry, R., The Legions of the Euphrates Frontier, CQ, vol. 3, no. 1, Jan
1909, pp. 44-53.
206
The site has been under excavation since 2001 by a German team from the
Forschungsstelle Asia Minor at the Universitt Mnster, under the direction of Prof.
Dr. Engelbert Winter. My thanks are due to them for allowing me to participate in the
excavation. See excavation reports: Blmer, M., and Winter, E., Der Dlk Baba
Tepesi bei Doliche und das Heiligtum des Iupiter Dolichenus (Vorbericht 1, 2), IM 55,
56, 2005, 2006.
207
See photograph on p. 74 of Gonella et al, Die Zitadelle von Aleppo und der Tempel
des Wettergottes: neue Forschungen und Entdeckungen, Mnster: Rhema, 2005.

97

Knowledge about the exact Iron Age use of the site is limited,
however. It is clear that it was significant during this period, as a large
number of Iron Age seals have been found, and an Achaemenid bulls
head capital reused within a later context confirms monumental preHellenistic architecture. An Iron Age ash deposit c. 1.5m thick,
containing enormous quantities of animal bones, 208 could be interpreted
as the remains of a festival similar to the spring fire-festival at
Hierapolis described in passage 49 in Lucians De Dea Syria:
They cut down tall trees and set them up in the court, and after
that they bring goats and sheep and other live animals and hang
them from the trees; together with them are birds and clothes
and gold and silver objects. Everything once complete, they carry
the offerings round the trees and set fire to them: the whole lot
immediately go up in flames. Many people come to this festival
from Syria and all the surrounding countries, and they all bring
their own offerings and have standards fashioned in similitude.209
It is known from inscriptions that an annual holocaust took place
in April for Bel at Palmyra, and at Harran in the medieval period, 210 so it
is not unreasonable to extrapolate that a similarly large sacrificial
festival took place on the hill at Doliche. This testifies to the supposedly
diminished rural population being committed to their local religious
cults during this period. There is no evidence for Bronze Age use of the
site,211 and if this really is the case, as it seems, then there is a question
as to when and why the hill was chosen as a suitable site for this cult.
Had there been a location that was abandoned after the Bronze Age
destructions, or was this an entirely new cult centre? Was it simply a
local temple, or did it form part of a cultic network?
The existence of the Hellenistic-Roman temple is confirmed from
the quantities of monumental architecture sculptural fragments of
208

See the 2007 excavation report (forthcoming).


Lightfoot, J. L., On the Syrian Goddess, Oxford, New York: Oxford University Press,
2003.
210
See Lightfoot, Syrian, p. 503-4.
211
Although much earlier seal-stones have been found, presumably heirlooms,
deposited at a much later date.
209

98

architrave, columns and reused high-quality ashlar masonry the


excavation of an extensive basalt paved area, and roof tiles stamped
IERATIKH. The paving formed a (presumably public) temenos-piazza. The
temple may have been destroyed during the c. AD 256 expeditions of
the Sassanian king Shapur I, who names the town of Doliche as one of
the places he annihilated,212 although there is no clear destruction level
to prove this. At any rate, the temple is not mentioned after this, and
there is further late antique use of the site as well as evidence for
Christian and Islamic occupation.
The position of the hill, with sight lines into the northern
mountains as well as across the southern plains, may have afforded it a
role in the surveillance of the area; and the temple at least would have
had a symbolic domination of the surrounding countryside.

Early iconography
There is great diversity and complexity in the Bronze and Iron Age
depictions of the various manifestations of the storm god, as Hadad,
Teshub or Tarhunzas, shifting across regions and times. Some general
observations that can be made across all these is that he is most often
bearded with long hair, with a crown, conical headdress or cap, and
wears an Egyptianising kilt or a so-called oriental robe, i.e. non-Greek
style tunic and trousers. Distinctions in type have been argued for
through the presence or absence of horns on the conical headdress. 213
Generally, his attributes are an axe, and either a thunderbolt or a long
staff/spear with foliate end. Representations from Doliche and the
surrounding area show the god on the back of a bull, 214 holding a
212

Sprengling, M., Shahpuhr I, the Great on the Kaabah of Zoroaster, AJSLL, vol. 57,
no. 4, October 1940, pp. 341-429.
213
Bunnens, G., The storm god in Northern Syria and Southern Anatolia from Hadad
of Aleppo to Jupiter Dolichenus, in Hutter, M., and Hutter-Braunsar, S., eds. Offizielle
Religion, locale Kulte und individuelle Religiositt, AOAT 318, 2004.
214
The few extant representations of the storm god Baal from the Canaanite Late
Bronze and Iron Age I periods (c. 1500-1000 BC) do not show him with a bull, he

99

thunderbolt and axe, but the style is archaic, almost indistinguishable


from representations of the god a thousand years earlier. In addition, he
is often associated with an eagle, possibly related to the winged disc, a
common feature of the early iconography.215
In the region, the storm deity was naturally associated with the
destruction wreaked by flash flooding. Deighton216 argues that the
indigenous Anatolian weather god controlled rivers and ground
waters; and is associated with rain only by proxy finding resonance
with the literary presentation of the Babylonian Adad Let him block
below, and not raise flood-water from the springs. (SBV, Atrahasis II/iv.)
Drijvers217 observes that Lucian (DDS 12, 13) attributes the founding of
the temple of Atargatis at Hierapolis to Deucalion, after the great flood,
and suggests that the oldest tradition linked the flood and its
destructive and life-giving power to Hadad alone and his semeion, and
perhaps

that

Atargatis

gradually

became

more

important

as

representative of the life-giving power of the water. By the time of the


Roman adoption of the cult, any associations the god had with either
the destructive or life-giving qualities of water in the earlier periods
had vanished or were not transmitted.

either stands alone, often in a smiting stance, or on seals and amulets he is sometimes
depicted as standing on a lion or horse see Cornelius, I., The Iconography of the
Canaanite Gods Reshef and Baal, Orbis Biblicus et Orientalis 140, Gttingen:
University Press Fribourg Switzerland, 1994, and Novk, Arameans, p. 256. By
contrast, the Hittite Teshub, and the Iron Age Hadad of Aleppo are represented on the
back of a bull and/or two mountain deities, or ascending a bull- drawn chariot. The
connection of the storm deity with the bull has been argued persuasively by Novk to
relate to the syncretism of the moon god with the storm god during the post-Hittite
Aramaean period see Novk, Arameans, p. 256.
215
This has usually been interpreted as a solar disc, but Novk argues, it could also be
taken as a lunar symbol, showing the crescent of the new moon and the disc of the
full moon. Novk, Arameans, p. 256.
216
Deighton, H. J., The Weather-God in Hittite Anatolia, Oxford: BAR International
Series 143, 1982.
217
Drijvers, H. J. W., Cults and Beliefs at Edessa, Leiden: Brill, 1980, p. 95-6.

100

Iconography of the Roman cult


When Jupiter Dolichenus is depicted in the western Roman Diaspora, he
is almost always dressed in a standard Roman military outfit: leather
panelled kilt, cloak, sword, greaves and embossed breastplate. This is
an oft-noted style in the representation of deities from the east, and so
cannot be taken as a distinctive feature of Jupiter Dolichenus. Various
other aspects, however, are characteristic of the cult in the west: his
usual attributes are a thunderbolt and an axe; and the most distinctive
feature is that he is very often depicted standing on the back of a bull.
There is a high level of uniformity in the Roman depictions of the
deity, with only a few anomalies, some examples naked, or in Oriental
robe and trousers. Butcher suggests that the images of Dolichenus
were faithfully replicated in various media across the empire, but in
each case the image referred to an original statue, associated with a
particular place.218 There is no archaeological evidence to indicate
what kind of cult statue might have existed in the sanctuary at Doliche,
or even if there was one. A newly discovered basalt stele from the
site,219 dated to the Roman period, represents the god and goddess on
the back of a bull and hind respectively. It is stylistically almost
identical to Bronze and Iron Age representations of the deities. It has
also been argued to replicate the cult statue, 220 contra Butcher, which if
it is the case, indicates a profound difference between how the cult was
conceived in its Syrian homeland and the form in which it was
transmitted across the Western empire. Even if it does not represent
the cult statue, it nevertheless highlights this dramatic opposition of
styles. Jupiter Dolichenus in Doliche was deeply stylistically connected
to his Bronze and Iron Age past, connecting the people who worshipped
him into a long artistic, religious and presumably literary tradition. The
cult in the west, by contrast, is linked to the Hadad of Doliche only
through the gods stance on the bull and his attributes, his epithet, and
218
219
220

Butcher, Syria, p. 336


See excavation reports, 2007 (forthcoming).
M. Blmer, pers comm.

101

the infrequently occurring phrase, ubi ferrum nascitur, where iron is


born. This phrase has been taken to refer to the iron mines of the
Taurus Mountains,221 and if this interpretation is correct, represents an
example of precise knowledge of the gods region of origin. No doubt
the worship of Jupiter Dolichenus was accompanied or fostered by
ancient stories or songs, perhaps describing the rolling thunder over
the iron-rich mountains, but without this supplementary evidence to
link the region of Doliche to the Roman cult Diaspora, what is apparent
is that the cult underwent a major transformation.

Speidels argument: The Dolichenian Pantheon


Sanders noted the popularity of the cult of Dolichenus with the army in
1902, suggesting that Commagenian cohorts would have propagated
their ancestral worship. However he suggests that its success more
generally was due to Syrian traders, slaves from the region, and the
zealousness of the priests.222 Cumont likewise claimed that its
popularity with the Roman army was due to the presence of
Commagenian traders and soldiers within the ranks. In his 1978 book,
The Religion of Jupiter Dolichenus in the Roman Army, Speidel agrees
that these people must have worshipped their native deity, although he
notes that there are very few Commagenians actually known from the
evidence. He acknowledges that while other units recruited from the
Orient, or campaigning there, took an active part in spreading the cult,
the Commagenian units proper cannot be shown to have made a
difference. His argument, by contrast, is that there was something
innate in the cult that caused its success: Obviously, it took more than
native zeal to win the world for a little-known local god, however old:
221

There is no iron in Doliche itself. On the history and importance of iron production
in Commagene from the Hittite period onwards, see Roesch, K., Kommagene das
Land ubi ferrum nascitur, in AW, 6, 1975, pp. 15-17.
222
Sanders, C. S., Jupiter Dolichenus, JAOS, vol. 23, 1902, pp. 84-92.

102

the cult must have possessed an intrinsic appeal, at least for other
Orientals.223
He goes on to argue that this intrinsic appeal was the theology of
the cult. He then attempts to reconstruct something of what this
theology might have entailed, and his main point is that the presence of
other deities both iconographically and in dedications proves the
existence of a Dolichenian Pantheon.
It is true that Jupiter Dolichenus was sometimes paired with Juno
Dolichena, and this makes a superficial theological link to the wellknown partnership of Hadad and Atargatis in Syria. The cult of
Atargatis as the Dea Syria spread in a similarly profound fashion across
the western Empire, her consort Hadad subjugated. Unlike the
Atargatis/Hadad partnership at Hierapolis, and in a similar way to the
spread of the cult of the Dea Syria alone, in the Roman Dolichenian
Diaspora, Jupiter is dominant, if Juno is shown at all. When she is, she
wears a long dress and cloak or veil, holding both a mirror or poppy
seed

head,

and

staff,

and

occasionally

peacock.

She

is

iconographically different from Atargatis, as she stands on the back of a


deer, while Atargatis animal is the lion. Atargatis was a regional
goddess in her own right, with her own particular cult and iconography.
A goddess on a hind, possibly a local version of Atargatis, seems to have
been worshipped at Doliche alongside Jupiter Dolichenus/Hadad. 224
Juno Dolichena in the Roman Diaspora is not as central, and the cult
does not seem to revolve around their divine partnership. This is
another marker of the transformation of the cult by the worshippers in
the process of transmitting it from the region of Doliche to the rest of
the Roman world.
Speidels proposed pantheon consists of Juno and the pairings of
Apollo Kitharoidos with Diana Lucifera, Sol Invictus with Luna, and the
Dioskouroi. He sees them as deities in their own right within the
pantheon, with meaningful roles in Dolichenian doctrine, although the
223
224

Speidel, Dolichenus, p. 8.
See especially the newly discovered stele of the god and goddess, mentioned above.

103

exact nature of their roles or the theology of the cult he leaves


unanswered. He attempts to answer, however, speculating that the
pairing of Apollo and Diana is perhaps equivalent to the IranianCommagenian Mithras and Anahita [] Or is Apollo here a healer god
[] and does Diana stand for Nemesis; or are the two meant to
symbolize emperor and empress?225 He takes Lucians report about the
collection of statues at Hierapolis as support for the interpretation of a
pantheon of Dolichenus.
There is, however, no real archaeological evidence for a wider
pantheon such as this at Doliche itself, although there has been found
the aforementioned stele of the god and goddess, an altar with a
depiction of a deer,226 and an altar with an enthroned goddess with two
lions, in the style of Atargatis. 227 If a pantheon associated with
Dolichenus really can be discerned in the western evidence, then it was
apparently a response to the needs of the western worshippers, rather
than a wholesale exportation of an archetypal Commagenian cult form
from the sanctuary in Doliche. Although some aspects of a pantheon of
the Atargatis cult are detectable in Hierapolis, there is very little
evidence in the corpus for the worship of Jupiter Dolichenus with any
other deity aside from Juno, or very occasionally Sol. The examples
Speidel gives of the inclusion of Apollo and Diana are few, and he
admits that Apollo and Diana were worshipped by the equites
singulares Augusti since the early second century, which may also have
contributed to their appearance on this plaque.228 He gives further
examples of Apollo and Diana as local gods of soldiers in the Balkans,
making it likely that these deities were included in dedications because

225

Speidel, Dolichenus, p. 23
See Winter, E., Die Grabung auf dem Dlk Baba Tepesi, in PATRIS PANTROFOS
KOMMAGHNH, Neue Funde und Forschungen zwischen Taurus und Euphrat, Asia
Minor Studien, Band 60, Bonn: Habelt, 2008.
227
See Blmer, Winter, Dlk, p. 197. The discovery at the site of images of the
goddess with both lions and deer may be relevant to the role of Juno Dolichena, but
this is not the place for a detailed discussion of this issue.
228
Speidel, Dolichenus, p. 24.
226

104

they were sacred to the worshippers themselves, rather than as a


formal part of the cult of Dolichenus.
It seems that the so-called pantheon reveals more about the
needs and backgrounds of the people dedicating than about the
theology of the cult itself. The pantheon Speidel finds does not
necessarily have a greater significance other than that the worshippers
of Jupiter Dolichenus were from a polytheistic background, accustomed
to dedicating to a number of different deities for different reasons, at
different times or in different places. This would suggest that the
majority of the cult followers worshipped Jupiter Dolichenus as he came
to them, supplementing their dedications as they were accustomed, and
as is exemplified in the dedication from Hadrians Wall, where Juno is
also addressed as the local deity, Caelesti Brigantia (CCID 565).229
My approach here, in opposition to Speidels, is to look at the
development and trajectory of the cult from a network perspective.
Instead of taking the intrinsic appeal of the cult as the point of central
importance, the focus here will be the connectivity of the networks of
people that were involved in the cult transmission.

Hypothesis: Activation of a Roman Military Network

The

cult

of

Jupiter

Dolichenus

in

the

Roman

world

is

found

concentrated in the northern and eastern borders: in Dacia, both


Pannonias, both Moesias, Raetia, Noricum, both Germanias, and in
Britannia. The earliest evidence for the cult in the West dates to the
beginning of the second century AD, but as Schwertheim notes, 230 some
229

All references to the body of material pertaining to the cult of Jupiter Dolichenus
come from Hrig, M., and Schwertheim, E., Corpus Cultus Iovi Dolicheni (CCID),
tudes prliminaires aux religions orientales dans lEmpire romain, Leiden: Brill,
1987.
230
Schwertheim, E., Iupiter Dolichenus, der Zeus von Doliche und der
kommagenische Knigskult, in Studien zum antiken Kleinasien: Friedrich Karl Drner

105

knowledge of the cult can be assumed prior to this, shown through


inscriptions recording the rebuilding of temples to the god for
example at Voreda in Britannia (CCID 577). Although certainly present
earlier, it is particularly during the second century AD that the cult
increased dramatically in popularity across the military zones of the
northern and western provinces of the Roman Empire.231
The western transmission of the cult of Dolichenus has been
associated with the general Roman trend of increased popularity of
cults originating in the east including for example, Mithras, Isis, and
the Dea Syria. Scholars have explained this phenomenon by appealing
to the quality of the eastern cults, judged to have an intrinsic pull on
human passions, in contrast to the cold and prosaic Roman religion. 232
The way the worship of the Syrian deities in particular diffused so
profoundly has been suggested, again by Cumont, to be initially through
Syrian slave populations during the second century BC, and then
because of the veritable colonization of the Latin provinces by Syrian
merchants.233
Likewise, the popularity of the cult with soldiers has been
suggested to result from the presence of Syrian recruits or regiments
within the wider structure of the Roman army: Syrian troops and, to a
lesser extent, Syrian merchants, slaves and freedmen carried the cult
of this obscure divinity far and wide through the Roman world.234
In addition, Cumont and Merlat235 suggest that Jupiter Dolichenus
became a semi-official tutelary deity of the army, through a particular
appeal to itinerant soldiers in his simple function as a protective battlezum 80. Geburtstag gewidmet, Forschungsstelle Asia Minor im Seminar fr Alte
Geschichte der Westflischen-Wilhelms-Universitt Mnster, A. Schtte, D. Pohl, J.
Teichmann (Eds.) Asia Minor Studien, Band 3, Bonn: Habelt, 1991, p. 37.
231
The epigraphic habit of the second and third centuries AD must of course be taken
into account, but cannot entirely explain the enormous increase in popularity of the
cult in the early second century.
232
Cumont, Oriental, p. 28.
233
Cumont, Oriental, p. 105-108.
234
Cook, A. B., Zeus, vol. 1, Cambridge: Cambridge University Press, 1914, p. 607.
235
Cumont, Dolichenus, p. 1278, 1903, Merlat, P., Jupiter Dolichenus, essai
dinterpretation et de synthese, Paris: Presses Universitaires de France, 1960, p. 100
ff.

106

god. If this is the case, or unless there is something missing from the
interpretation of the earlier storm god type, then it highlights the
Roman transformation from his earlier manifestations, as the Bronze
and Iron Age storm god is not explicitly martial.236
Speidel responded by arguing that the military aspect of the cult
had been overstated, and that less than two in five of the cults
inscriptions mention soldiers.237 My updated total is c. 121/430 using
military terms of any sort, and a further 48 probably connected to the
military via immediate context or inference, amounting to c. 169/430.
Military dedications therefore still account for roughly two in five
pieces of evidence; Speidel concluded that hardly more than half the
dedications come from the military.238 His wording is negatively
phrased to support his point: because much of the hinterland of
Noricum, the Pannonias, the Moesias and Dacia, as well as Italy,
Dalmatia and Thrace were largely free from garrisons; 239 the civilian
aspect of the cult had been overlooked. By rectifying the dichotomy that
previous scholars had seen between military and civilian worshippers,
Speidel argued that defining the cult of Dolichenus as a military cult is
in opposition to the evidence. The cults success with the military was
therefore not because it was simple, but precisely because it offered
more to its military followers than just protection and victory, who
followed as complex a creed as the civilians.240
Aspects of both these arguments are true. There was a definite
transformation of the deity for the Roman audience; and there is no
division between what soldiers and civilians believed. With these
statements of earlier scholars in mind, I here re-approach the evidence
with the aim of understanding what the transformation of the deity
236

Although the stele from Tell Ahmar (c. 8th century BC) is dedicated to Tarhunzas as
god of the army (see Bunnens, G., A New Luwian Stele and the cult of the Storm God
at Til Barsib-Masuwari, Peeters: Louvain, 2006). My thanks to Michael Blmer for
bringing this to my attention.
237
Speidel, Dolichenus, p. 11, n. 35. Merlat counted 80/264, Speidel 97/254.
238
Speidel, Dolichenus, p. 39.
239
Speidel, Dolichenus, p. 38.
240
Speidel, Dolichenus, p. 11.

107

entailed, and how the cult diffused so widely and swiftly. The ideology
and nature of the god has always been identified as the central factor
for understanding the reasons for the cults popularity. Drawing on the
conclusions made by sociologists of religion detailed in Chapter 2, and
given that nothing is known about the cultic theology or ideology for
Jupiter Dolichenus, I here shift the explanation given for the adoption of
or conversion to the cult away from the intrinsic value of the god
towards an approach that understands this as a social process, relating
to status and established communication networks. To make sense of
the diffusion pattern, the focus here is not the nature of the god himself,
but the networks that facilitated the spread of his worship.
I show here that the cult diffused across a military network that
was already in position. The generally homogeneous depictions of the
god as well as the relatively short time period in which the diffusion
took place can be described as the activation of a network the cult
travelled in a coherent and unified form through established social
networks. The people who adopted the worship of the deity were
already in place, forming an open system of communication who were
they, and what did they have in common? When, and why did they
choose Jupiter Dolichenus? Can the cult patterning inform the
interpreter about the centralisation of the cult form, or is the pattern
reflective of peoples social networks?
The role of ethnic groups, i.e. Syrians, and the role of
occupational groups, i.e. traders, will both also be further investigated.
I expect to find that the role of traders in the diffusion of the cult was
negligible, as in social terms, traders are naturally on the outside of the
strong-tie social networks that need to be infiltrated in order to spread
a new religious belief. Syrians, who were themselves traders as well as
soldiers and priests, may prove to be more powerful in the dynamics of
the cult transmission. Where possible from the epigraphy, observations
about the different kinds of roles played by Syrians, and the varying
degrees of influence these had will be explored.
108


Military Status: the army and civilians
The military cult diffusion can be thought of in terms of relative status.
It has been supposed that the cult travelled to the west via Syrian
recruits and regiments who continued to worship their traditional local
god within their new social context of the Roman army, or was adopted
by legions serving in the region of Doliche who took it back with them
to the west.
In Syria itself, the Syrian soldiers serving there presumably
continued their worship of Hadad/Dolichenus as before. As local men
who knew the land, the people, and the language, they may have had an
element of status within the ordinary ranks of the army. It is likely that
local men were also appointed to the officer class for the same reasons.
Non-Syrian soldiers and officers on campaign in the region or stationed
near Doliche would have been exposed to the cult of a Jupiter-type god,
both through their Syrian colleagues and because the temple exerted a
level

of

dominance

in

the

landscape.

The

name

of

Zeus

Hadados/Dolichaios would swiftly have been translated by them into


Jupiter Dolichenus.
Beyond the Syrian homeland of Jupiter Dolichenus, I argue that
the cult was transmitted across military networks already in position.
Innovative

religious

movements

pass

through

strong-tie

social

relationships families, close friends, and neighbours but in a military


environment, these roles are played by comrades-in-arms. The strongtie relationships would exist among close colleagues of similar rank.
Among the infantry, these are comrades who live and fight together,
closely tied by geography and friendship. In the officer class, these
relationships often involve geographical distance, but with regular and
repeated communication links. The social networks linking officers
provides a clear explanation for the spread of the cult through the army
109

at such speed, and across the Roman Empire to such profundity. As men
of higher status to start with have considerable influence over their
subordinates, it might be expected that the cult would filter down to the
lower status soldiers also. If this hypothesis were correct, then it would
be expected to find large numbers of officer class military men, for
example, centurions, represented in the epigraphic evidence, as well as
some dedications by lower status military men. The presence of Syrian
recruits or legions would be expected to play an important role earlier
in the diffusion, but tail off as the movement gathered adherents. The
cult had priests, but their function and offices are largely unknown. It
may be that if a continued Syrian presence is found, it will be in the
administering of the cult.
Dedications with no explicit indication of a military connection by
my count total 257/430, well over half. However, some of these simply
have no inscription at all, military or otherwise, and many of them are
found in close proximity to the military dedications and military zones.
It may be that many of the civilians represented in the evidence are
actually invisibly connected with the military i.e. as partners (only
after Septimius Severus as legitimate wives) and families of the
soldiers.241
This transmission to the non-immediate family civilian populations
near the army occupied territory can also be explained through relative
social status. The hypothesis for a military-to-civilian diffused cult
involves understanding the status of the soldiers and officers as a
special and clearly delineated sub-group within a frontier setting. The
relationship of locals to the occupying army was undoubtedly complex,
with the obvious likelihood of hostility as well as compliance. The
members of the army would, however, have been in possession of
higher status, simply through being the arm of the state: in control of
the land, the taxes, and requiring huge quantities of physical support.
241

The ban on marriage for serving soldiers was created by Augustus and dissolved by
Septimius Severus in AD 197. See Phang, S. E., The Marriage of Roman Soldiers (13
BC AD 235). Law and Family in the Imperial Army, Leiden: Brill, 2001, p. 3.

110

Connections with the local population would have been multifaceted,


and again, if close ties were made, religious beliefs might have passed
across them. This might imply that non-Roman civilians who were
attracted to the soldiers cults viewed them as markers of a relative
elite identity. If this hypothesis were reasonable, it would be expected
that the evidence from non-military contexts might be found to be
Romanising in some way, through experiment with nomenclature,
adoption of Roman titles, or other expansion of the spheres of identity.
The epigraphic material will be able to show some of these things if
they exist. Some civilian dedications, however, clearly represent
independent cult followers. These will be examined for unifying
features, such as ethnicity, immigrant status or ties to a particular
place.

Issues
Some important caveats to add here are concerned with the nature of
epigraphic evidence.

The epigraphic habit, especially of the Romans

within the context of the army, must be recalled. Most of the


inscriptions from Roman Britain, for example, are found in highly
militarized zones on Hadrians Wall, and around the veteran cities of
the south and east. The fashion of inscribing stones had not penetrated
into the blank areas of the country, and so the skewing of the evidence
should not then be taken to mean there were no Romans or other
people worshipping Jupiter Dolichenus in those areas where the cult is
unattested. This in fact is quite likely to have been the case, but it is
important to note that there were areas of widespread general
epigraphic silence.
There is also the intrinsic problem that those who can afford
inscribed dedications are those who already possess a higher-thanaverage income and social status. Lower status offerings of comestible
111

or degradable materials will not be represented, and many small metal


votives are not likely to have survived.242 However, this problem is
equally applicable across all studies that rely on epigraphy, and
although it is important to remember the bias in the evidence, it should
not necessitate its dismissal. In this case, there are enough inscribed
dedications from persons of humbler status (slaves, freedmen, ordinary
soldiers) to suggest that the epigraphic picture is not impossibly
skewed towards higher status individuals.

242

It is worth considering also the performative nature of setting up altars or


dedications, and the physical acts of prayer and sacrifice that are generally lost to the
modern interpreter. Inscriptions sometimes mention more than one dedicant, families,
or whole troops of comrades-in-arms. These people may not have been able to afford
an inscribed dedication individually, but could record their belief as a community. It
may be imagined that they were set up in the presence of all of them and so the
dedication becomes a communal event.

112

Epigraphic Analysis

This section examines the epigraphic material to create a sociological


profile of the cult worshippers. By building up a picture of the
characteristics and iconography of the cult, and the ethnicities and
occupations of the worshippers and priests from the evidence of their
dedications and buildings, it will be possible to examine the hypotheses
outlined above.
The Nature of the Evidence
The evidence for the cult of Jupiter Dolichenus can basically be divided
into four types. There are buildings that are archaeologically attested,
Dolichenums, that are identified as such through inscriptions or clear
iconography, these number nineteen, of which three are slightly
uncertain. There are twenty-two other buildings or sacred places to the
god that are archaeologically unknown but attested from the epigraphy,
of which four are uncertain. By far the largest of the types of evidence
are votive inscriptions, generally on stone altars, columns, or statue
bases that are explicitly to Jupiter Dolichenus, this group numbers 308,
of which some are from the archaeological context of a known
Dolichenum. Finally, there is evidence not explicitly connected to
Dolichenus worship, but that is associated with the cult for various
reasons. Some of these reasons are good seven non-explicit
dedications are from clear contexts, fifty bear clear iconography of the
cult and six are dedicated to Juno and are connected by context, but
some are more uncertain or subjective, for example, the bronze hands
and triangles.
Bronze hands are a cult offering also found associated with other
cults for example, Theos Hypsistos (CCID 44).243 Four hands are
243

This dedication has been included in the corpus for Dolichenus on the strength of
its form only. I suggest that it does not belong here, and should be considered
alongside the cult evidence for Theos Hypsistos, see chapter 6.

113

explicitly dedicated to Jupiter Dolichenus, from Moesia Inferior (CCID


70), Dacia (or the Ukraine) (CCID 177), Nida (Heddernheim) in
Germania Superior (CCID 520) and Cappadocia (CCID 43).244
The bronze triangles245 by contrast relate more specifically to the
cult of Dolichenus and are sometimes found in clusters. They are often
divided into registers, with, alongside Jupiter on the back of a bull,
representations of various other deities246 including: Juno, an eagle in
the top register, busts of Sol and Luna, the Dioskouroi, Hercules,
Sarapis, Minerva, and Nike. Two were found in the hoard from Maueran-der-Url (CCID 294; 295) and two in the Dolichenum at Nida in
Germania Superior (CCID 511; 512); two from Trisisamum in Noricum
(CCID 327, 328); and two from Lussonium in Pannonia Inferior (CCID
201; 202). Two are alleged to be from Syria but were both bought in
the Munich art market (CCID 6; 7). If the iconography of the triangles
can be used to understand the nature of the cult, then the two
supposedly originating from Syria are remarkable in their difference to
the western triangles: they show only the god, presumably Dolichenus,
the busts of the Sun and Moon, and on one, another god, or a priest
(CCID 6). The triangles from the west, by contrast, which largely
feature depictions of other deities, suggest that the pantheon is a
western feature.
The silver votive palms and sheets are all either from the hoard at
Mauer-an-der-Url in Noricum (fourteen) or from Nida (Heddernheim) in
Germania Superior (five). The survival of these clusters of delicate
items is surely accidental, and they were a common type of dedication
244

Aside from these definite dedications, three other bronze hands have been found,
uninscribed and without clear iconography, that may relate to the cult of Jupiter
Dolichenus. They come from Dacia (CCID 171), Thrace (CCID 53; 55), and Pannonia
Inferior (CCID 189).
245
It has been suggested by my colleague, M. Blmer (Mnster) that the triangle
might represent the top of the semeion of Atargatis, linking the Dolichenus cult more
strongly with that of Hadad/Atargatis at Hierapolis. They are generally interpreted as
functioning as cultic standard-tops, to be carried on a pole in procession or stood in a
particular place of worship, perhaps the sign of Dolichenus mentioned in an
inscription from Dura Europus (CCID 39).
246
Much of Speidels argument for the so-called Dolichenian pantheon rests on the
iconography on the triangles.

114

but generally do not survive.247 Men with clear military links dedicated
all the votives from Nida. By contrast, the votives from Mauer-an-derUrl

are

notable

for

the

extremely

high

proportion

of

women

represented; the total hoard is twenty-six items, of which individual


women gave ten. There are in general very few female dedicants in the
cult. When women do appear, it is most often within a family dedication,
with a male family member, or with children. The exceptions are these
ten fronds, and the only known inscription from Tungrorum in Belgia,
(LAnne Epigraphique 2002, 1011).

Military and non-military characteristics


There are a number of criteria for defining a dedication as military.
Primarily, there is the explicit mention of military terms generally
either a particular rank within the army, or the name of a legion. Of the
430 dedications, 59 are without inscription, leaving 371 inscribed
monuments.248 121 of these dedications can be explicitly connected to
the Roman army. In addition, there are forty-eight further dedications
(some inscribed, some not) that are assigned to the military through
physical context or close geographical association. Proportionally, these
amount to almost half the total number of inscribed dedications. Within
the group classed as military dedications, there are individuals, groups
of soldiers, officers, and officers or individuals on behalf of their
comrades.
Criteria for classification of dedications as explicitly non-military
are harder to define, as many of the inscriptions do not give any details
about the worshippers, but the main other occupations that are
247

See the hoard of silver leaves from Vichy dedicated to Jupiter Sabasius, in Moore,
C. H., The Distribution of Oriental Cults in the Gauls and the Germanies, APA, vol.
38, 1907, p. 117.
248
Uninscribed dedications without a useful context are unhelpful in the following
analysis. They will however be included in the assessment the pattern of the evidence,
in the section on network analysis.

115

mentioned are traders and priests. Dedications by women might usually


be classed as civilian, but here, depending on their context, they are
likely to be connected to the military. There are also some dedications
without any indication of profession that are found in the hinterlands,
away from the military borders.
The traditional hypotheses will be examined first: the exposure to
the cult of legions that had been used in the eastern campaigns; the
influence of Syrian troops within the army; and finally, the role of trade
and traders, especially Syrians. After showing these hypotheses to be
unsound, I then investigate the role of the communications network of
higher status army officers. In the final section, the non-military
evidence will be examined to ascertain the part that Syrian priests
played in the diffusion; the military-civilian interface; and dedications
from the hinterlands.

Chronological framework and initial dispersion


Over 200 of the dedications are undated, which accounts for roughly
half the total of inscribed and uninscribed monuments. Dating for the
remaining evidence is either precise, through record of the exact date
of inscription, or approximate, through reference to the ruling Emperor
or other officials. Probable dating applies to the rest, using style of
letter cutting or other typology, context or independent knowledge, or
scholarly opinion. The range of those that can be reasonably accurately
dated is the first century BC-first century AD in the region of Doliche
(CCID 22; 23) to c. AD 300 in Colonia Ulpia Ratiaria in Moesia
Superior (CCID 111). The epigraphic habit has been mentioned
previously, but it is useful to recall again the huge general increase in
epigraphic material between the second and third centuries AD. Even
so, the geographical breadth of the cult by the early second century is
particularly remarkable. It is here also worth noting that the lack of
116

Dolichenus dedications post AD 300 confirms the decline in public


display of pagan cult in the fourth century, and the particular decline of
this cult, the reasons for which will be examined later.
It is unsurprising that the earliest evidence comes from the region
of Doliche, but there is little epigraphic material and it is difficult to
assert exact dates. The majority of the evidence comprises stelai and
reliefs of the deity. The style of these representations is often
archaizing, making them additionally difficult to date accurately. In the
inscriptions from Syria, the deity, where he is addressed at all, is called
the great-sighted god (CCID 28); the listening god (CCID 6); the
holy god (CCID 20); Theos Dolichenus (CCID 2); Zeus Megistos of
Doliche (CCID 33; 34); or simply the god (CCID 9). The name Jupiter
Dolichenus in the region seems to appear in the second-third century, at
Hierapolis (LAnne Epigraphique 1998, 1430) and Dura Europus
(CCID 32; 39). These dedications have explicit military connections.
The western diffusion is first explicitly testified to at the end of
the first century AD. The cult is found in Rome in a dedication by a
praefectus vigilum in AD 92 (CCID 434), although this is a disputed
date. However, the next dated piece of evidence is only 33 years later,
in a dedication discovered in a building identified as a Dolichenum at
Lambaesis in north Africa from AD 125 (CCID 620). The cult here
apparently already had a reasonable following, as it records the
dedication of the temple. Temples are additionally found during the
reign of Antoninus Pius in Carnuntum (CCID 217), Balaklawa in
Chersonesus Taurica (LAnne Epigraphique 1998, 1156), and at
some point between 120-160 in Voreda (Plumpton Wall) in Britannia
(CCID 577), these last two clearly fairly well established, as both were
already undergoing restoration at this time. In 138, the cult is found at
Praetorium Latobicorum in Pannonia Superior (CCID 275), in Dacia at
Pojejena de Sus at some point before AD 132 (CCID 172), and in
Apulum, also in Dacia, and on Hadrians Wall in Britannia during the
reign of Antoninus Pius (CCID 151; 564). It is not unreasonable to
117

assume that these dedications may have been housed in a sacred


building or space of some kind, and that the cult enjoyed a level of
popularity beyond these few surviving inscriptions.
These early pieces of evidence are notable for the geographical
breadth they span from North Africa, Britannia, Dacia and Pannonia,
to the north shore of the Black Sea. It is clear that this is no organic
diffusion through areas of geographical proximity to Doliche. Of these
nine dedications, six have an explicit military connection: the legatus
pro praetore who dedicated the temple at Lambaesis; the centurion of II
Augusta on Hadrians Wall; the centurion of a vexillatio of I Italica at
Balaklawa who undertook the restoration of the temple; the prefect of
the cohors II Gallorum in Britannia who also restored the temple there;
the prefect of the cohors V Gallorum, who dedicated in Dacia; and the
beneficiarius consularis of XIIII Gemina at Praetorium Latobicorum.
This supports the proposition that wealthy and powerful men within the
army influenced the spread of the cult westward. Although the
inscriptions from Carnuntum and Apulum are not explicitly military,
they must be closely associated because these places were major
Roman legionary camps at the time. There are, however, no active
Syrian connections made in the inscriptions, and the Syrian link has
been central for understanding the diffusion of the cult in the west. Had
these men served in Syria and returned, bringing the cult? Or had the
connection to Syria been made prior to their devotional offerings as in
at least four of these places there was an existing cult building of
Dolichenus?
What does seem clear, however, is that the initial dispersion of the
cult to the corners of the empire was through military avenues.
Following this, the number of dedications made between c. AD 160 and
the end of the Severan dynasty increases dramatically. This suggests
that, if it is true that legions that served in the east brought the cult
back with them, then the campaigns against the Parthians by Lucius
Verus between AD 162-66, and those of Septimius Severus c. AD 198
118

should be of central importance in explaining this exponential rise in


popularity at this time. Does the evidence support this interpretation?
Or does the cult diffuse of its own accord, through other lines of
communication?

The role of the eastern campaigns in the cult diffusion


One of the main explanations offered for the success of the cult in the
west assumes that legions drafted to the east were exposed to the cult
in its homeland, and took it back with them to the west when the
campaign was over. To ascertain if there are correlations to be found
between participation in eastern campaigns and the presence of the
cult in those legions, dedications by members of legions that had served
in Syria are discussed here.
The eastern campaigns that will be useful to examine are:
Trajans Parthian wars between 114-117; the quashing of the Bar
Kokhba revolt under Hadrian between 132-135; Lucius Verus Parthian
expedition between 161-167; Septimius Severus campaigns against
Pescennius Niger in 195 and against the Parthians in 197-8 and again in
208; and possibly also the third century campaigns Caracallas in 215,
Severus Alexanders between 230-232, and Gordian IIIs in 238.
Trajans Parthian wars 114-117
Five legions found in the dedications to Dolichenus are known to have
taken part in the campaigns of Trajan against Parthia: I Adiutrix, III
Augusta, X Fretensis, XVI Flavia, and XXX Ulpia Victrix. I Adiutrix
moved from Pannonia to participate in the Dacian wars, and remained
in Dacia. After Trajans wars it returned to Pannonia and was based in
Brigetio. Vexillationes of III Augusta, based usually in Lambaesis,
fought in Trajans campaign, and it is known that at least 43 Syrian

119

soldiers were recruited to the legion at this time. 249 X Fretensis was
based in Judaea and Syria and had been involved in other campaigns in
the area. It was this legion that destroyed Qumran, and besieged
Jerusalem, Machaerus and Masada, before being stationed in Jerusalem
following the Jewish war. XVI Flavia was formed from the disgraced XVI
Gallica, and sent to the eastern provinces as punishment by Vespasian.
After being involved in Trajans campaigns, Hadrian sent it to
Samosata. XXX Ulpia Victrix was raised by Trajan c. AD 100 to fight in
the Dacian wars. It was based in Dacia, although it is likely that part of
the legion participated in the Parthian campaigns.
Most of the inscriptions dedicated by members of these legions
are, however, dated much later than the period of Trajans wars: all the
dedications by III Augusta date to the third century (CCID 615; 616;
621-624; 626; 627; 630); a dedication by X Fretensis is found between
167-180 (CCID 138); XVI Flavia in 211 (CCID 32); and two inscriptions
of XXX Ulpia Victrix, dating from 211, (CCID 547); and 228 (CCID
541). It is implausible that the participation of these legions in Trajans
campaign can have had any influence on these cult dedications, some a
century later.
There are four inscriptions of I Adiutrix, three from the
Dolichenum

in

Brigetio,

the

legions

camp,

and

another

from

Aschaffenburg in Germania Superior, (CCID 538). The Aschaffenburg


altar is dated to 191, which means it can be discounted here, but the
dedications from Brigetio are either undated or broadly placed in the
second half of the second century. It may be that they are earlier rather
than later, and it is possible the legion was exposed to the cult in Syria
and brought it back to Pannonia; however, even if this were so, it still
leaves a considerable gap between the time of military engagement and
the evidence. The legion also took part in Septimius Severus
campaigns in the closing years of the second century, which will be
examined below. It seems fairly apparent that the dedicants from these
249

Pollard, N., Soldiers, Cities and Civilians in Roman Syria, Ann Arbor: The University
of Michigan Press, 2000, p. 117.

120

legions did not adopt the cult of Jupiter Dolichenus while on Trajans
campaigns, and alternative campaigns or explanations must be sought.
The one inscription that might plausibly have a connection with
this campaign is mentioned above the dedication from Lambaesis
dating to AD 125 of the temple to Dolichenus by the legatus pro
praetore Sextus Iulius Maior, a senator from Nysa on the river
Maeander in Asia. Although there is no explicit mention of his legion, it
must be III Augusta. It is not necessarily improbable to hypothesise that
he encountered the cult of Doliche while leading a vexillatio of another
legion in the east when he was more junior, dedicating the temple a
decade later.
The Bar Kokhba revolt 132-135
Four legions that dedicate to Dolichenus are known to have taken part
in suppressing the Bar Kokhba revolt in Judaea: III Cyrenaica, X
Fretensis, X Gemina and XI Claudia. III Cyrenaica was in Alexandria,
used in campaigns in the region against the Parthians, and the Jewish
wars of 66-70 before its involvement in Bar Kokhba. X Fretensis had
been heavily involved in the fighting in Judaea, and was stationed in
Jerusalem. X Gemina was in Germania Inferior until AD 103 before
moving to Aquincum and then Vindobona in Pannonia, where it
remained until the fifth century. Vexillationes of the legion were
involved in Bar Kokhba. XI Claudia was in Brigetio until AD 104, before
moving to Durostorum in Moesia Inferior, where it remained until the
fifth century. Some vexillationes took part in quelling the Bar Kokhba
revolt.
Both the inscriptions for III Cyrenaica are from the Esquiline
Dolichenum and date from 191, too late to reasonably make a case for
connection with Bar Kokhba. Likewise, the dedication by the member of
X Fretensis in Dacia dates from between 167-180 (CCID 138), also too
late. Similarly, the five inscriptions of X Gemina, from Pannonia
Superior and the Esquiline Dolichenum in Rome: two are un- or broadly
121

dated to the second century (CCID 222; 277), but the other three date
between 180-218 (CCID 223; 270; 415), also too late to reconstruct a
plausible connection with involvement in Bar Kokhba. The only
dedication of XI Claudia is undated (LAnne Epigraphique 2001,
1733).
As well as the fact that the dates of the dedications are too far
removed from the legions involvement in the campaign, it is also
unlikely that quashing the Bar Kokhba rebellion played any part in the
diffusion of the cult of Dolichenus, simply because Judaea is some way
from the north Syrian homeland of the cult. It is possible that legions
may have marched through the region of Doliche en route, but it seems
implausible to speculate that the cult would have had time to transmit
during a relatively short stay.250
Lucius Verus Parthian expedition 161-167
The campaigns of Lucius Verus against Parthia are a more likely
candidate for how the military might have been exposed to the cult in
its homeland, as it is precisely at this period that the dedications begin
to increase dramatically. Supporting this, seven legions who make
Dolichenus dedications are known to have been involved: II Adiutrix, III
Cyrenaica, IIII Flavia, V Macedonica, X Gemina, XIIII Gemina, and XVI
Flavia. II Adiutrix was based in Aquincum after Trajans Dacian wars,
and vexillationes were involved in the campaigns of Lucius Verus. III
Cyrenaica was based in Egypt, and had served in the region on previous
occasions. IIII Flavia was in Moesia Superior and had fought in Trajans
Dacian wars, before being involved in Lucius Verus campaigns. V
Macedonica returned to Moesia in AD 71 after successes against the
Parthians and in the Jewish wars. Trajan used it in his Dacian wars,
before Lucius Verus took it on his Parthian campaigns. The legion
afterwards returned to Porolissum. X Gemina was in Pannonia until the
fifth century, with vexillationes taking part in Verus wars. XIIII Gemina
250

Although further archaeological investigation is necessary in order to ascertain the


scale of military involvement in the area of Doliche.

122

was involved in Trajans wars in Dacia between AD 101-106, and then


posted to Carnuntum, c. 114, where it stayed for the next three
centuries. XVI Flavia had been camped at Samosata since the period of
Hadrian.
One undated inscription by a veteran of II Adiutrix was found in
the legionary camp at Aquincum (CCID 183). The deity is addressed as
Jupiter Dulcenus Heliopolitanus, making a connection to Syria through
the Heliopolis locative, but also syncretising Dolichenus with the
popular Jupiter Heliopolitanus. The misspelling of Dolichenus as
Dulcenus represents a mishearing or second-hand appropriation of the
deitys name, as it is clearly Latinised and has implications of
sweetness. Without an accurate date, it is impossible to draw further
conclusions from this inscription.
As noted above, both dedications by III Cyrenaica are from the
Esquiline, dating from 191. The thirty year lag between the legions
involvement in the wars of Lucius Verus suggest that these dedicants
did not adopt the cult during their time on campaign, but via another
route. Likewise, the time gap is a problem for interpreting the one
inscription mentioning IIII Flavia in this way, as it dates from between
185-192 (CCID 449). One of the dedications accounting for V
Macedonica in the evidence is undated (CCID 404); the other is an
altar found in Samum in Dacia that dates to AD 243 (CCID 131), which
is far too late to associate with this campaign. The inscriptions
pertaining to X Gemina date between 180-218 (CCID 223; 270; 415),
or are unspecific (CCID 222; 277); the minimum twenty-five year lag is
also too long to reconstruct picking the cult up with Verus campaigns.
Both inscriptions that mention XIIII Gemina are unconnected with
deployment in Syria: as one was dedicated between AD 235-238 (CCID
232), and the other in 138 (or, according to CIL, in 159) but either
dating is too early (CCID 275). The one dedication by XVI Flavia is too
late, 211 (CCID 32).

123

Intuitively, it is reasonable to think that the participation by the


military in the campaigns of Lucius Verus near the homeland of the cult
of Jupiter Dolichenus might be in some way responsible for bringing the
cult back to the west. However, further examination of the evidence for
this claim has shown it to be untenable, with all the dedications by
members of the legions involved occurring after too great a time lag.
Septimius Severus campaigns 195, 197-8 and 208
Like the campaigns of Lucius Verus, the eastern engagements of the
African emperor Septimius Severus have also been hypothesised as a
time when the cult of Jupiter Dolichenus might have been further
adopted and disseminated back to the west. The nine legions who
dedicate to Dolichenus who were involved are: I Adiutrix, I Italica, I
Parthica, II Italica, IIII Scythica, VIII Augusta, XI Claudia, XIIII Gemina,
and XVI Flavia. I Adiutrix, whose base camp was at Brigetio, marched
on Rome in support of Septimius Severus, and was involved in all of his
campaigns in the east. I Italica, based at Novae in Moesia, also
supported Septimius, besieging Pescennius Niger in Byzantium with XI
Claudia, fighting in the battle of Issus in 194. It may have been further
involved in the Parthian campaigns. I Parthica was raised c. 195
specifically for these campaigns, and the legion remained in the east, at
Singara (Sinjar, Iraq) until 360. II Italica was raised by Marcus Aurelius
c. 165 and stationed in Lauriacum (Lorch) in Noricum. It was used
against Pescennius Niger and in the Parthian campaigns. IIII Scythica
were based at Zeugma, fifty kilometres from Doliche itself, and was
involved in these campaigns of Septimius. VIII Augusta was based
somewhere on the Rhine, later at Strasbourg. XI Claudia, based in
Durostorum in Moesia Inferior, fought alongside the other Moesian
legion I Italica against Pescennius Niger at Issus. Septimius Severus
himself had commanded XIIII Gemina, based at Carnuntum. XVI Flavia
was at Samosata.

124

The four dedications of I Adiutrix are all too early to have been
affected by these wars, being either undated, broadly dated to the
second half of the second century (CCID 253; 241; 242), or precisely
to 191 (CCID 538). One inscription mentions II Italica, from Virunum
(Zollfeld) and dates to the second half of the third century (CCID 342).
This is too long a time gap to propose a direct link between the eastern
campaigns and the cult presence, moreover, the altar was found in a
building tentatively identified as a Dolichenum, testifying to the cults
presence at an earlier date. Two inscriptions by members of VIII
Augusta are known, both dated prior to these campaigns (CCID 538;
539). The inscription of XI Claudia is undated (LAnne Epigraphique
2001, 1733), but is from Novae, the headquarters of I Italica, the
legion they fought alongside against Pescennius Niger. However, this
dedication may not relate to this period at all, and reflects instead the
communications that existed between geographically close camps. The
two inscriptions of XIIII Gemina have been discussed above, neither
dated to a period that could associate the cult with the eastern
campaigns, and moreover, a Dolichenum is attested in Carnuntum by
the period of Antoninus Pius.
Two of the four inscriptions of I Italica are too early, being dated
to 139-161 (LAnne Epigraphique 1998, 1156; 1158); but the other
two may not be. One from Novae, the legions headquarters, is dated
broadly to the second-third century (CCID 74), and one from
Dionysopolis in Moesia, is dated precisely to 214 (CCID 71), and may
result from engagement in Septimius Severus campaigns. Likewise, the
one inscription of I Parthica dates from 206, found in Obernburg, was
given by a vexillatio of XXII Primigenia that were led by a centurion of I
Parthica (CCID 537). As the dedication is by the soldiers of XXII
Primigenia themselves, it has little to do with I Parthicas involvement
in Septimius campaigns, although if the centurion were Syrian, which
is not unlikely given the date of the inscription and his legions genesis,
then he may have been an adherent of the cult in its place of origin who
125

brought it with him to XXII Primigenia. However, what seems more


likely is that members of XXII Primigenia already worshipped the god,
and the centurion of I Parthica had very little to do with its
transmission. What it certainly does show is the vast movement of
officers across the empire, which will be discussed further below.
Not unreasonably, it might be expected that the legions based in
the region, IIII Scythica (Zeugma) and XVI Flavia (Samosata), would
have adopted the cult of Jupiter Dolichenus early and fairly extensively.
However, only four dedications to Dolichenus by these legions are
known: three by IIII Scythica, one from the Aventine Dolichenum (CCID
404), and two from Dura Europus (CCID 32; 35); and one by XVI
Flavia, also from Dura (CCID 32). The Aventine inscription and one of
the Dura inscriptions are undated, and the other two are precisely
dated to 211. Vexillationes of both legions were present in Dura at this
time, and the Dolichenum there seems to have been a purely military
installation.251 This is good evidence for the campaigns of Septimius
Severus having an effect of the dispersion of the cult, but not in the way
conventionally suggested. The men of these legions, more than any
others, would have had the opportunity to discover the cult and temple
of Doliche, as they were stationed very close by. However, they did not
apparently adopt the cult before the end of the second century, after
they had been based at Zeugma and Samosata for a century or so. This
seems to indicate that interaction with other legions during Septimius
campaigns may have been the driving factor in the adoption of the cult
here, rather than geographical proximity to its homeland, as it may be
with the inscriptions of I Italica and I Parthica discussed above.
The third century expeditions
Caracallas expedition between 215-217, that of Severus Alexander
between 230-232 and Gordian III from 238 are all somewhat late to
251

See Pollard, N., The Roman army as total institution in the Near East?, The
Roman Army in the East, ed. D. Kennedy, JRA supplementary series no. 18, Ann Arbor:
JRA, 1996, p. 222.

126

account for the adoption and spread of the cult of Jupiter Dolichenus in
the west, as a large proportion of the dedications are dated to the end
of the second century and beginning of the third. Caracalla and Gordion
used I Adiutrix, whose base was in Brigetio in Pannonia. The
Dolichenum there was established by the end of the second century,
before these campaigns; and the other dedication by this legion from
Aschaffenburg dates from 191 (CCID 538). Caracalla also used III
Cyrenaica, but both the dedications associated with this legion are also
from 191 (CCID 408; 409). The only legion that dedicates to
Dolichenus and that took part in the campaigns of Alexander Severus is
XXX Ulpia Victrix. However, that members of this unit picked the cult
up in Syria can be immediately discounted, as both dedications date
from before the campaign (CCID 547; 541). Gordian III also used II
Adiutrix and possibly XXII Primigenia. II Adiutrix was based in
Aquincum with vexillationes used against the Sassanids, however, the
one dedication to Dolichenus, found in Aquincum, is undated (see
above). XXII Primigenia was based in Mogontiacum in Germania
Superior, remaining there until around the third century. Vexillationes
also helped build the Antonine Wall. All five of the dedications by this
legion are dated prior to their involvement with the campaigns in the
east (CCID 508; 524; 531; 536; 537).
It is clear from this analysis that there is very little in the
evidence to support the notion that the cult of Jupiter Dolichenus was
picked up by soldiers engaged in any of the eastern campaigns. In this
case, there must be other reasons to account for its diffusion through
these various legions. The other central explanation given has been the
presence of Syrian recruits bringing the cult with them from their
homeland, which will be examined next.

127

The role of Syrian recruits


Analysts from Cumont onwards have also attributed the success the
cult of Jupiter Dolichenus in the army in part to the presence of Syrian
recruits, who zealously continued their worship of the deity of their
homeland in the army, influencing and converting their comrades.
Moore writes that, for the most part the cult came into the Rhine valley
from the valley of the Danube and the provinces contiguous to it, as did
the worship of Mithras. We know that the legions in Moesia were filled
from

Asia,

that

in

Dacia

there

were

two

or

three

cohortes

Commagenorum, as well as many Asiatics settled there by Trajan. Some


of these would naturally have brought this worship with them.252
Does the evidence support this reasonable explanation? Evidence
taken here to represent Syrians in the army is direct: explicit mention
of Syrian ethnicity through name, hometown or province. The use of
Greek is not enough to postulate an explicitly Syrian connection,
although it could certainly represent easterners in general. The
discussion looks at both whole units that were raised in Syria or the
region, as well as the presence of individual recruits or groups of
recruits within the wider structure of the army.
A few military units were raised in the east and were probably
composed largely of Syrians, at least at their inception. Legio I Parthica
was raised by Septimius Severus for his Parthian campaigns. However,
as noted above, the one dedication (CCID 537) to Jupiter Dolichenus
mentioning this legion makes no explicit link to Syria, although the
centurion in command was almost certainly Syrian and had been
involved in the eastern campaigns. It is dated to 206, so it is too late
within the corpus to extrapolate any central involvement in the diffusion
of the cult; moreover there is no evidence to suggest that his
connection with Jupiter Dolichenus did not in fact arise from his time in
Germany with XXII Primigenia. An altar from the legions base camp at
Mogontiacum supports this. The centurion of XXII Primigenia, Domitius
252

Moore, Oriental Cults, p. 122-123.

128

Aesclepiades, from Arethusa in Syria, (CCID 524) dedicated it between


211-217, after the cult is first found in the legion. Moore suggests that,
in his transfer from legion to legion the centurion remained faithful to
the god of his native country. 253 This may be so, but his is one of the
only pieces of evidence for this being the case. The relatively late date
moreover suggests that his role as a Syrian in spreading the cult of his
native country was minimal there were already worshippers of
Dolichenus in his legion five years before his dedication.
That some cohorts were originally from Syria is suggested by
their names, and it is possible that they were levied from the region
during or following Trajans Parthian campaigns. Veterans of I
Canathenorum (from Canatha, in the heart of Syria Trachontis)
dedicated an altar in Serviodorum in Raetia in 163 (CCID 485), a
relatively early date in the evidence, making it plausible that these
veterans had brought the deity with them from Syria. Similarly, the
cohort I Damascenorum was based in Friedberg in Germania Superior
between 120-160, and at some point during the second century Tiberius
Claudius, a centurion of the cohort, dedicated a silver votive sheet in
Nida (Heddernheim) (CCID 518). The cohort II Flavia Commagenorum
was in Dacia following Trajans Dacian Wars. If these soldiers were in
Dacia this early, then they should be heavily implicated in the popularity
of the cult in that province, supporting Moores thesis. Unfortunately,
the only piece of evidence explicitly pertaining to the unit is undated:
an altar from Micia given by the prefect of the cohort (CCID 159). It is
however notable that many of the dedications from Dacia explicitly
name Jupiter Dolichenus as being of Commagene (for example, CCID
146;

147;

152;

160;

162),

plausibly

connected

to

strong

Commagenian presence in the region, II Flavia Commagenorum being


part.
The cohort XX Palmyrenorum was the principal garrison at Dura
Europus.254 A papyrus text from Dura records Aelius Avitus as centurion
253
254

Moore, Oriental Cults, p. 122.


Butcher, Roman Syria, p. 55.

129

of an unspecified legion and praepositus of the cohort in 239 (CCID


39), but the cult at Dura seems to be late, and unconnected with the
temple at Doliche or Syrian recruits in particular.
It is notable that there are in fact very few individual recruits with
Syrian connections represented in the evidence. The earliest individual
with a potentially Syrian cognomen is Caius Julius Marinus, a miles of
the Classis Praetoria Misenensis. He dedicated a stone deers head in
the Esquiline Dolichenum at some point prior to 212 (CCID 411). A
man with the same name is found later in the second century dedicating
a bronze hand at Nida (CCID 520); he is probably the same man, but
he had moved up the ranks and was by this point a centurion of the
Numeri Brittonum Gurvedensium. Dating from the same period and
also found in Rome, Caius Julius Dionysius dedicated to the paterno deo
Comageno (CCID 433). He was a miles of the same Classis, and names
himself of the natione Surus. It seems that the cult in this fleet has
strong connections to the Commagenian homeland, perhaps indicating
that it arrived in it early on.
A few Syrians may well be present in the legion III Augusta,
which was based in Lambaesis: an altar from Lepcis Magna was
dedicated between AD 208-210 by a centurion of the legion, Titus
Flavius Marinus, whose cognomen may indicate a link to Syria (CCID
615). An altar from Thanadassa (Ain Wif), dating from AD 210-211, was
dedicated by Marcus Caninius Adiutor Faustinianus, who was both the
prefect of cohors II Hamiorum and the praepositus of a vexillatio of III
Augusta. Hamiorum may have been from the Syrian city of Hama, and
was on the Numidian front with Septimius Severus (CCID 616). Finally,
an altar dating from 222-238 (CCID 627) was given by a centurion,
Valerius Rufus. He may be from Syrian Antioch,255 as a tombstone is
known that reads: D. M. L. Valeri L. f. Rufi domo Antiochia (CIL VIII
2997). If this is the case, then he may have been part of the
replenishment of the legion by men from III Gallica by Elagabalus c.
255

See commentary in the CCID by Hrig and Schwertheim, no. 637.

130

219. III Gallica was on duty in Syria, but is not mentioned in any
Dolichenus dedication. There was a temple to Dolichenus in Lambaesis
from 125, and so these potentially Syrian members of the legion a
century later can be discounted as genuine forces in the adoption of the
cult here.
Another

man

probably

from

Syrian

Antioch

is

Flavius

Antiochianus, the prefect of the cohort I civium Romanorum equitatae,


and also leader of the cohort IIII Vindelicorum (CCID 527). He
dedicated an altar in Germania Superior in 211.
Aurelius Theoteknos, whose cognomen is a translation of the
Syrian Barlaha, dedicated a stele from the Dolichenum area of Dura
Europus (CCID 35). He was general of the legion IIII Scythica, and it
seems quite likely that he was ethnically Syrian. The rest of the
inscriptions from Dura all date to the third century, his pseudopraenomen Aurelius makes it reasonable to place this dedication with
them, which is again too late to be useful in thinking about the routes of
the cult diffusion.
The individual soldiers found in the corpus that potentially have a
Syrian connection number but a handful, and the units with Syrian
connections are also very few. The rest of the evidence is all too late
and too sparse to support Cumont and others argument for a zealous
Syrian contingent in the military.
The only really plausible connection with Syrian recruits carrying
the cult with them is found in the evidence for the cohorts I
Canathenorum, I Damascenorum, and II Flavia Commagenorum. These
units seem to be the most likely candidates for the initial introduction of
the cult into the western provinces. If this were the case, then the
earliest

evidence

for

the

cult

should

be

in

Dacia

(II

Flavia

Commagenorum), Germania Superior (I Damascenorum) and Raetia (I


Canathenorum). However, the earliest evidence, as seen above, is from
Africa, Pannonia Superior, Britannia, and the Chersonesus Taurica, as
well as the altar from Apulum in Dacia. The earliest evidence from
131

Raetia is the dedication from 163 by the veterans of I Canathenorum


discussed above. The earliest from Germania Superior dates to 191, and
the dedicant is from Savaria in Pannonia Superior (CCID 538), so I
Damascenorum can be discounted as having influenced the adoption of
the cult.
It seems from this that Moores suggestion is correct, contra
Speidels assertion that, because there is such meagre evidence, the
influence of II Flavia Commagenorum should be discounted (see above).
The Commagenian connection explicitly expressed in the dedications
from Dacia has already been discussed, and one of the earliest pieces of
evidence comes from Apulum. It can legitimately be argued that the
cult was introduced into Dacia by the Commagenian cohort and the
settled Syrians in the population. The only other locations where
Commagene is explicitly mentioned are Pannonia Superior and Rome
both places where the cult is earliest attested. If the introduction of the
cult into the west can be aligned with the presence of II Flavia
Commagenorum, then what is to be understood from the striking lack
of Syrian recruits in the rest of the evidence? This can be explained if it
is accepted that the cult was not diffused by Syrian recruits, but instead
by powerful military men who had adopted the cult in Dacia. The cult
originally had a strong connection with Commagene, but as it was
transmitted beyond natives of Syria and Commagene this connection
became less important. The influence of easterners on the cult diffusion
has been overstated, as it is also suggested that Syrian traders played a
part in the transmission, which will be discussed next.

Supplying the army: the influence of traders


The final explanation given for the diffusion of the cult of Jupiter
Dolichenus has been the presence of Syrian traders. 256 However, there
256

Cumont, Oriental, p. 25.

132

are an extremely small number of traders in the evidence, and from


that number, only two make themselves known clearly as Syrians.
Moreover, although the status of traders as long-distance weak ties
makes them very important in, for example, the diffusion of disease,
technological innovation or information requiring little social change,
their social ability to influence people concerning religion is very
limited because they are outside the tight social networks that religious
information and change travels across.
However, there clearly are some traders that dedicated to Jupiter
Dolichenus. It is my suggestion here that traders were introduced to
the cult through their role in the supplying the army. The find locations
where traders are attested, which all have connections with the
military, supports this argument.
The dedications by traders are found all across the empire, mainly
dating to the early third century. One with an explicit military
connection has been found in Rome: Publius Aelius Miron dedicated
two items to Dolichenus in 218, probably in the Esquiline Dolichenum
(CCID 415; 416). He is connected to the legion X Gemina in an
unknown capacity. That legion was based at this time in Vindobona in
Pannonia.
A limestone relief dedicated in Brigetio by Domitius Titus dates to
the second-third century AD. He had been decurion at Seleucia Zeugma
(CCID 239), and is argued by Hrig and Schwertheim to be a former
decurion now occupied in trade like many Orientals before and after
him.257 Hrig and Schwertheim class him as a civic decurion of Syrian
origin, but it is unclear whether his role was civic or military as it is
not stated, and associations of veteran soldiers had decurions. Either
way, he has connections with the military both through the find spot,
the base camp of I Adiutrix, and his link with Zeugma, IIII Scythicas
base.

257

See commentary in the CCID by Hrig and Schwertheim, no. 239.

133

In 221, Lucius Viducius dedicated a slab in Eboracum (York); 258


and Claudius Fronto, town councillor and trader from Aquincum in
Dacia dedicated a column in Augusta Traiana in Thrace (CCID 52). The
cohors I Athiotorum had built a Dolichenum during the reign of
Septimius Severus (LAnne Epigraphique 1999, 1374), it is likely
that this column was dedicated in this building, placing it most probably
in the third century. In 217 at Mogontiacum, base camp of XXII
Primigenia, an altar was given by Caius Julius Maternus (CCID 525);
two silver votives from the hoard at Mauer-an-der-Url in Noricum that
possibly date to the reign of Commodus were given by the trader
Augustus Aurelius, one to Hercules (CCID 306); and one to Jupiter
(CCID 311). An unnamed person, known simply as Surus, offered
another votive here. There is nothing to indicate that Augustus Aurelius
was a Syrian, but there is the possibility that he was this unnamed
person.
In Dacia, two inscriptions by explicitly named Syrian traders have
been found a column from Apulum was dedicated at the turn of the
second-third century by Aurelius Alexander and Flavius Suri (CCID
153); and an undated altar from Ulpia Traiana Sarmizegetusa was
given by Gaius Gaianus and Proculus Apollophanes Suri (CCID 169).
The same Gaianus is also found on another Dolichenus inscription in his
role as part of the builders guild (CCID 165).
Some dedications are undated, but those that are clearly dated
come from the beginning of the third century, making it even more
unlikely that these men had any influence on the success of the cult.
From this, it is apparent that there are fewer Syrian traders
represented in the evidence than has been previously suggested, and
that the impact of traders in general and Syrian traders in particular in
the spread of the cult of Dolichenus was minimal. It seems likely, in
fact, that these traders picked up the cult from their dealings with the
military, and not the other way round.
258

See Irby-Massie, G. L., Military Religion in Roman Britain, Leiden: Brill, 1999,
Catalogue no. 395, p. 278.

134

The communications network of Roman military officials

With all of the traditional explanations for the success of the cult of
Jupiter Dolichenus in the west proving to be untenable or unlikely, what
then can account for its popularity and the profundity of its diffusion?
What does the exponential rise in dedications from c. 160 onwards
represent? The epigraphic habit of course must be remembered, but I
suggest here that the enormous success of the cult at this time can be
understood as an information cascade the activation of a military
network that was already established.

Administration: structure & communication


The generals of the army were also provincial governors, the legates of
the Emperor, titled legati Augusti pro praetore or, in certain regions,
proconsul. A generals staff would have consisted of two groups his
bodyguards, the singulares, and his administrative support staff, the
men engaged in the issue of orders and information. Goldsworthy
writes that,
We can do little more than guess at the composition of the
commanders staff proper. Titus seems to have had centurions
seconded from their units in his immediate following and Arrian
had both centurions and decurions close to him (BJ 6. 262,
Ectaxis 22). We know that many officers did serve away from their
units in the retinues of magistrates and commanders, but it is
impossible to do more than guess at their precise function. For
instance, would a general have sent an order to a subordinate via
an officer such as a centurion or merely have used one of his
bodyguard as a messenger? It might have been useful for an
officer, with a better appreciation of the situation than a ranker, to
135

carry orders, since the recipient could have questioned him about
the generals intentions.259
It is reasonable to assume that the communications network of
the army consisted of middle-ranking officers, and that this class of
soldier were regularly moving between legions and across the Empire.
As early as 1907, Clifford Moore noted that the role of officers in the
diffusion of cults was important referring to a dedication to Jupiter
Sabasius, he observes that the place of the dedicators origin, the
colony Emona in Pannonia, shows how a centurion, transferred from
one province to another, might be an important agent in the diffusion of
foreign cults.260
An overview of the epigraphic material illuminates this argument
further. There are twenty-four known centurions that dedicate to Jupiter
Dolichenus, two of which seem to have had a connection to the
Praetorian Guard. There are also fourteen prefects, five signiferi, five
tribunes, five beneficarii, three legati propraetores and one general.
There are also two primipili, five miles, eight optiones, five veterans of
unspecified rank, and two custodians of the armoury. It is clear that the
overwhelming majority of the dedicants from the military inscriptions
are men of middling-high status. Although there are foot soldiers and
veterans of unspecified rank, by far the largest social group is that of
centurions, prefects, and tribunes. There are in addition a number of
men with special status: signiferi, aquiliferi, primipili and a dux ripae,
the commander of the river Euphrates border. Although as previously
mentioned there is an issue concerning the disposable income
necessary

for

inscribed

dedications,

the

evidence

provides

an

unequivocal picture of the officer class of the Roman military being


closely involved with the cult of Jupiter Dolichenus.

259

Goldsworthy, A. K., The Roman Army at War 100 BC-AD 200, Oxford: Clarendon
Press, 1996, p. 123-4.
260
Moore, Oriental Cults, p. 116.

136

The mobility of the officer class


The Commagenian cohorts in Dacia, officers among them, had brought
the cult into the sphere of the military. The proposal here is that
because officers were highly mobile and communicating frequently, this
fact and their shared status created strong-tie bonds between them,
despite, and perhaps because of, geographical distance. This resulted in
an active physical and emotional network across which innovation and
information was able to flow. The cult of Jupiter Dolichenus became
established through and spread across this network of interactions. The
mobility of the officer class can be clearly seen in the epigraphic
material.
The Esquiline Dolichenum, suggested by Speidel to be primarily a
military temple, is shown to be particularly cosmopolitan, with officers
and soldiers from across the empire donating there. Centurions of III
Cyrenaica, based in Alexandria, augmented the sacrarium, added
columns to the assembly and donated a nymphaeum as well as other
cultic installations (CCID 408; 409) in 191, representing the expansion
of the building. Two men have been encountered before: the miles
Caius Julius Marinus, of the Classis Praetoria Misenensis based in
Misenum in Campania gave a stone deers head prior to 212 (CCID
411), he is probably the same man as the centurion of the same name
in Germania Superior (CCID 520); and in 218, Publius Aelius Miron,
who was associated in some way with X Gemina, who were based in
Vindobona in Pannonia, dedicated there twice, (CCID 415; 416). The
Aventine Dolichenum was less cosmopolitan, but still there is an
inscription from here that was dedicated by an eques who was also a
priest (CCID 404), marked with the eagles of V Macedonica and IIII
Scutica (Scythica). This represents a considerable journey from either
IIII Scythicas base camp in Zeugma, or V Macedonicas at Potaissa in
Dacia, showing the long-distance travel undertaken by more senior
soldiers; this is particularly interesting as he is also a priest. The travel
137

and role of priests, especially Syrian priests, will be examined further


below.
The scenario in the provinces shows similar movement of officers.
That men moved across the British frontiers is demonstrated by an
undated altar from Aesica (Great Chesters) on Hadrians Wall, that was
dedicated by Lucius Maximus Gaetulicus, a centurion of XX Valeria
Victrix, stationed at Chester (CCID 561), also found dedicating to
Apollo at Newsteads (RIB 2120). Centurions of II Augusta, which was
based at Glevum (Gloucester) and Isca Augusta (Caerleon) are found
dedicating on Hadrians Wall: between 138-161 Marcus Liburnius
Fronto at Condercum (CCID 564), he may be the same Liburnius
Fronto from XX Valeria Victrix known from RIB 2077; and in the first
half of the third century a dedication from Magis was given by an
officer who was also in command of a vexillatio of VI Victrix, based at
Hadrians Wall (CCID 575). Another altar from Magis shows the
international links of the frontiers, as it was given by Julius Valentinus,
ordinatus of Germania Superior (CCID 576) A later dedication from
between 235-238 at Camboglanna was given by Flavius Maximianus, a
tribune of cohort I Aelia Dacorum, and ex evocato of cohors I Praetoria
Maximinia (CCID 572).
In Germania, the Aschaffenburg altar was dedicated in 191 by the
former aquilifer of I Adiutrix, whose base was in Brigetio in Pannonia.
He was a local recruit to that legion, originally from Savaria, the
regional capital of Pannonia. He had by this point worked his way up to
centurion status, part of the command of VIII Augusta (CCID 538).
Longer international movements are shown by the dedication in GrossKrotzenburg in 211 by the prefect of the cohort I civium Romanorum
equitatae, Flavius Antiochianus, (CCID 527); he was also leader of the
cohort IIII Vindelicorum, based in Caesaraea in Mauretania. The altar
from Obernburg (CCID 537) has already been discussed in relation to
the campaigns of Septimius Severus, but it shows also the transfer of
officers from legion to legion. Dating from 206, it records the dedication
138

by a centurion of I Parthica while he was leading a vexillatio of XXII


Primigenia,

based

in

the

provincial

capital

at

Mogontiacum.

Unsurprisingly, vexillatio members of XXII Primigenia are found in


dedications from around the German frontiers: a horned altar was given
in Saalburg in 205 (CCID 508); another altar from Obernburg was
given in 207 (CCID 536); and an altar from the Dolichenum in
Stockstadt was dedicated in 214 (CCID 531). Another altar from this
Dolichenum was given by the prefect of the cohort I Aquitanorum, Titus
Fabius Liberalis (CCID 530), a man with the same name commanded a
vexillatio of legio VI Victrix, usually based on Hadrians Wall. Lucius
Caecilius Caecilianus, a prefect of I Aquitanorum gave a base and an
altar in this temple (CCID 532; 533): he is also known from Aelia
Augusta Mercurialis Thaenitanorum in Africa.
Pannonia was home to the legions of X Gemina at Vindobona and
XIIII Gemina at Carnuntum. Carnuntum was seemingly a cult centre for
Jupiter Dolichenus, with a Dolichenum in the town from the mid second
century at the latest.261 It received dedications by visitors from other
military units aside from the home legion of XIIII Gemina: Gaius
Spurius Silvanus, a centurion of X Gemina, dedicated a base between
180-183 (CCID 223); Atilius Primus, a second century centurion of XIII
Gemina but ex evocatus of X Gemina dedicated a limestone relief of
Dolichenus on a bull (CCID 222). His legion was based in Apulum in
Dacia at this point; but he clearly retained his connection with both his
earlier legion and the centre of worship.
The officers of the two legions are found across the region:
Marcus Aurelius Valentinus, beneficiarius consularis of XIIII Gemina,
dedicated an altar in Praetorium Latobicorum, on the route into
Dalmatia and Italy, on the 1st November 138262 (CCID 275). In Savaria,
between the Tenths base in Vindobona and that of the Fourteenths in

261

Fourteen other inscriptions were also found in Carnuntum; the town as a cult
centre will be discussed in greater detail below.
262
Or, according to CIL, in 159.

139

Carnuntum, two consular beneficarii of X Gemina dedicated an altar in


208 (CCID 270).
Officers of other military units are also well represented in
Pannonia. Valerius Aelianus, a signifer of XIII Gemina, based at Apulum,
gave an altar at Emona (CCID 273); and much further south in Aquae
Balissae, two altars were dedicated by centurions, one of an unnamed
legion, Secundius Restutus (CCID 277), and one dated between 198208, by a centurion of VII Gemina, whose usual base was in Tarraco in
Hispania (CCID 276). A further international link is perhaps made
through his wifes name, Atticillia, which may make a link with Athens.
The province of Dacia, as might be expected given that it was
here the cult was first adopted by the military, has a few early examples
of very long-distance movement of officers. The prefect of the cohort V
Gallorum, Quintus Petronius Novatus, gave an altar in Pojejena de Sus
that must date to pre-132, as the cohort was back in Moesia Superior
by roughly this date (CCID 172). His wide-ranging career is known
from another dedication from Mauretania. Likewise, Publius Caius
Valerianus, centurion of X Fretensis, based in Judaea, gave an altar in
Domnestri in Dacia between 167-180 (CCID 138). Marcus Aurelius
took detachments to fight the Marcomanni in Dacia. This seems to
indicate clearly that although the legion was based in the Near East,
the cult of Dolichenus was only encountered on the limes of the
Danube.
Proving that the cult was popular across Dacia and until the end
of the date range of the evidence (and just before the evacuation of the
province), an altar from Samum dating from 243 was dedicated by a
beneficiarius of V Macedonica, Publius Aelius Proculinus. The base
camp of the legion was at Potaissa, nearly 75km south of the find spot
(CCID 131).
The connections Moesia Superior had with the rest of the empire
is shown in a dedication of a bronze badge in Iulia Concordia in Italy by
a centurion of IIII Flavia, based in Moesia, given between 185-192
140

(CCID 449). Iulia Concordia is on the route between the frontier and
Rome, and this inscription supports the suggestion that it was the
officers taking information between the generals and Rome. The more
local interactions of the troops in Moesia are shown by an undated
relief dedicated by a member of XI Claudia that was found in Novae,
headquarters of I Italica (LAnne Epigraphique 2001, 1733).
Dura Europus as a Euphrates frontier town received dedications
by legions from elsewhere: a member of XVI Flavia, based in Samosata
(CCID 32), in 211, and by Aurelius Theoteknos of IIII Scythica (CCID
35), who were at Zeugma.
North Africa also provides examples of the movement of highstatus officials. A man who was both prefect of cohors II Hamiorum and
praepositus of a

vexillatio

of III

Augusta, whose base was a

considerable distance away in Lambaesis gave an altar dating from 210211 in Thanadassa (CCID 616). In Lambaesis itself, the legatus Sextus
Iulius Maior, who dedicated the temple there in 125, was a native of
Asia Minor (CCID 620); there is also later evidence of international
links, in the dedication from post-253 by Tiberius Memmius, son of
Tiberius Palatina Ulpianus Roma, (i.e. of the Palatini tribe in Rome),
who was prefect of the cohort II Civium Romanorum and also tribune of
III Augusta. In addition, his wife has a Greek name, Veratia Athenais
(CCID 622).
The summation of this evidence is that the officer class of the
Roman army dedicating to Jupiter Dolichenus were highly physically
mobile, supporting the suggestion that it was these men who were
instrumental in carrying the cult of Jupiter Dolichenus to the far
reaches of the Roman Empire. Because their interaction network is
almost completely invisible, these inscriptions provide important proof
that

communication

and

travel

by

this

group

of

people

was

commonplace and expected. It will be very helpful to visualise the


hypothetical networks that might have existed, to ascertain whether the

141

same conclusions can be drawn from a network analysis as can be from


an epigraphic analysis.

142

Visualising the Network

This analytical method assesses the pattern of the evidence as a totality.


A distribution map (Map 4A) shows the places where the dedications
have been found, clearly illustrating that the cult follows the Rhine and
Danube valleys, with some outlying clusters in Britannia and the region
of Doliche: limited information that offers no further insight into the
routes of transmission. By linking the nodes into a network, it is
possible to visualise the communication routes suggested here to be
central to understanding the diffusion of the cult.
An initial Proximal Point Analysis (PPA) captures the end picture
of the data and links every known node to its three closest neighbours.
This model shows the centrality and isolation of places on the network.
It is a preliminary analysis and clearly not a reflection of the actual
connections that existed between these sites, and moreover covers the
whole date range. What it does allow is an initial observation of the
geographically determined network: the empty spaces, long-distance
overland or maritime links, and some of the constraints the terrain had
on communications.
A directed PPA responds to what is known of the date range of the
evidence, by mapping the find spots in blocks of fifty years. Each find
location is once again linked to its three closest neighbours, two of
which are required to be already established. This analysis helps to
visualise the spread of the cult in actual terms, highlighting centres of
diffusion and, because of the requirement of prior establishment,
suggests a more centralised diffusive method. The spread of a cult is
likely to have been a more organic process of contact and adoption, but
this analysis serves as a hyper-real version of the picture.

143

Proximal Point Analysis (Map 4B)


The initial PPA of the cult find spots allows some clear observations to
be drawn. The network and connectivity in the region of Doliche in
Syria are almost entirely separate from the rest of the western network.
The tight, integrated network created by the sites round Doliche is
unsurprising, and reflective of the local nature of the cult in this region.
The analysis also highlights the relative isolation of the cult in Dura
Europus, Caesarea and Cappadocia, here due to geographical distance.
The separation of this part of the network from the rest, however, may
reflect a different quality of the cult in these more distant places, i.e.
that they are not necessarily part of the local cult network round
Doliche. In fact, the cult finds support this interpretation: the cult at
Dura has been demonstrated to be a late, military arrival, not
necessarily connected with Doliche itself; the one find from Caesarea is
dated to the second-third century and was dedicated by a Latin-named
man in Greek (CCID 30), probably representative of a man out of his
usual context; and the one find from Cappadocia is a bronze hand, a
portable offering (CCID 43). What is also missing is the western link to
Rome that important military cities in this area certainly had. Dorylaion
provides the only link across into the west, and its isolation is clear, as
is the greater likelihood of it being connected with the cult in Thrace,
rather than Doliche.
By contrast, the western network is highly connected, with few
areas being very isolated. Places that do appear to be somewhat
isolated and have long-distance links are the North African sites at
Lepcis Magna, Thanadassa and Lambaesis; Saldanha in Hispania; the
sites of Massalia and Antibes in Gallia, Ager Morinorum in the north;
and Brancaster and Isca Silurum in Britannia. Although it might be
viewed as problematic that the distances between these sites is so great
and that the network could not reflect a plausible link between them, it
is profoundly useful to visualise these long-distance links. What the
analysis does highlight is the remoteness of these cult find locations
144

and so suggests potential routes that might have brought the cult to
them. The find at Isca Silurum, for example, is quite likely to have a
connection

with

the

military

cult

on

Hadrians

Wall;

likewise

Brancaster. Equally, the other isolated sites in the Mediterranean are


mainly port sites, and as such are quite likely to have been connected.
The finds from these places again might support the slightly different
network quality in this area: the marble statuette from Massilia was
found in the harbour, suggesting it resulted from a shipwreck (CCID
602); the inscription from Antibes is fragmentary and uncertain (CCID
601); the one find from the Iberian peninsula was given by a veteran of
VII Gemina, whose base was in modern Len (CCID 609) and the find
from Turris Libisonis (CCID 468) may make reference to Ravenna,
possibly to be interpreted as meaning the Classis of the city, known
from two other Dolichenus inscriptions. These dedications round the
north shore of the Mediterranean seem likely to be connected with the
sea, and are linked with Italy via North Africa to Misenum, base of the
Classis Praetoria Misenensis. The finds from North Africa all have
military connections, making it extremely likely that the dedications
from the Mediterranean have some link with the early temple and cult
known from Lambaesis.
What is also striking in the western Mediterranean is the
apparent separation in the cult networks of the north and south of Italy.
Although the PPA has not identified Rome as a major hub, this divide is
not simply reflective of a problem with the analysis. Northern Italy was
close to the frontiers, and the finds fall along the path of the overland
route into and out of Rome. Southern Italy was not, and the connectivity
of the southern network centres on Misenum and the bay of Naples.
The finds from the southern Italian sites also suggest a different quality
of the cult, and indeed some of the finds are not explicitly to
Dolichenus. It is the only place in the west that Greek is represented
(CCID 467), although this is unsurprising in the ancient Greek colony
of Naples; there are two magical alphabets with no explicit link to
145

Jupiter Dolichenus (CCID 465; 466); and there is a Syrian priest at


Tarracina (CCID 462). Two finds from Ostia make explicitly military (or
naval) links with the south through their mention of the Classis
Praetoria

Misenensis

(CCID

440;

441).

The

complete

lack

of

connectivity between the two areas prompts the suggestion that there
were two separate routes for the cult diffusion and adoption in Italy
from the military frontiers in the north, and via naval connections with
the Classis Praetoria Misenensis in the south, perhaps bolstered by
eastern immigrants to Rome and the region.
The main body of the rest of the network is tightly integrated, and
follows, with very few deviations, the geographical features of the
northern frontier, concentrated in particular on the Danube and Rhine
valleys. The mountainous interiors of the Alps, Dalmatia and Thrace
present themselves as a clear obstacle to the cult transmission, and it is
noteworthy that although this analysis has no geographical costs or
directionality included, the network resulting from the positions of the
find spots implies geographical barriers. More importantly however,
these barriers were also social plenty of people lived in these regions,
but they were not exposed to the cult because they did not interact with
the people who transmitted it, showing these mountainous regions to
be socially as well as geographically isolated. Religious innovation
moved through receptive social space: highlighted by the cults success
in the mountains of Dacia. This represents, superficially, as equally
inaccessible geography as Dalmatia or the Alps, but the cult was
instead profoundly popular, thanks to the presence of the Roman army.
The network of the evidence follows the military frontiers of the
Rhine, Danube and northern Britannia, with particularly tight clusters
round the legionary base camp at Mogontiacum and Hadrians Wall.
These network formations represent a level of internal isolation, where
interaction is very introspective, and perhaps implies a very rapid
adoption. In the case of Germania Superior, the evidence supports this
claim of rapid, faddish adoption, with the greatest number of
146

dedications from between 190-217. In Britannia, of those that can be


dated, the largest number of dedications do fall into the first half of the
third century, but the date range is much greater. Elsewhere the cult
pattern is more evenly distributed, with some important corridors, for
example along the lower Danube from Colonia Ulpia Ratiaria to Novae
in Moesia, from where the cult spreads more freely; or the upper
Danube through Raetia. This is likely correlative to the spaces that the
military inhabited.
Some parts of the network seem to be offshoots from the main
body running along these frontiers: the pocket in Moesia Superior; the
path into Thrace via Cabyle; and the chain of sites along the Dalmatian
coast. These places are some distance from the main military
communications arteries, and the local networks they are part of reflect
this, with the networks in Thrace and Moesia Superior being quite
clustered, and that along the Dalmatian coast quite extended. It is
notable that all these sites, bar Augusta Traiana, are single dedication
find spots, suggesting immediately that these areas were cultic
backwaters, anomalous in some way. In fact, the evidence supports the
identification of these places as unusual: the dedications from the
Dalmatian coast either reveal nothing about their dedicants or were
given by priests of the cult, one of whom is Syrian (CCID 123). In
Thrace, there is also a dedication by Greek priests at Cillae (CCID 54);
but there are also military connections. The cohort I Athiotorum built
the temple to Dolichenus in Cabyle (LAnne Epigraphique 1999,
1374), and two of the dedications from Augusta Traiana make explicit
links to Porolissum in Dacia (CCID 50) or Aquincum in Pannonia (CCID
52). Again, in the Moesian cluster, there are priests, possibly oriental
(CCID 115); a Greek servus vilicus (CCID 116); and a Syrian with
connections to the army possibly a veteran (CCID 126). It is a high
percentage of both easterners and cultic officials to find in such a small
dataset. The interface between military and civilian, and the role that
priests and Syrians played in this interface will be further discussed
147

below. The network highlighted these nodes because of their geography,


but their geographical configuration also highlights them as of a
different social makeup.

Developing Networks over time (Maps 4C-4F)


Following from these initial observations of clustering and isolation in
the pattern, the following Proximal Point Analysis will build in the date
range of the evidence. By plotting the distributions in chunks of fifty
years, the analysis reveals a clearer picture of the diffusion of the cult
over time and places that can be identified as centres. Once again,
nodes connect to their three closest neighbours, at least two of which
must be already established.
First century BC-AD 150: Map 4C
The first map is fairly simple, with a cluster of links only in the region of
Doliche. The fact that only a few sites in the west have early dedications
means that the links here are very long, reflecting both the breadth of
the Roman Empire at this point and the great distances the people
involved in the cult were spread across. As has been seen above,
members of the military did indeed travel all across the empire, so what
initially looks unnaturally distant is an important visualisation of the
huge range of this communications network.
It is striking that because of the distances involved, Doliche has
only one long-range link, to Balaklawa in the north of the Black Sea. I
suggest that this reflects a real division between the cult in Doliche and
the west: but the link ought to run through Dacia, where the
Commagenian cohort was based. In fact, the Balaklawa dedication
(LAnne Epigraphique 1998, 1156) was given by a member of I
Italica, based at Novae in Moesia, shifting the real network closer to
Dacia. The single link to Doliche and the focus of the network in the
148

west on the central sites of Rome, Praetorium Latobicorum, Carnuntum


and Apulum would appear to confirm my argument from the epigraphy
that the cult was adopted independently of any involvement with
eastern campaigns.
AD 150-211: Map 4D
The end date of 211 was chosen because many inscriptions can be
dated to before or after the death of Septimius Severus, and the map
shows

the

extraordinary

increase

in

cult

dedications.

What

is

immediately clear is that the triumvirate of Carnuntum, Apulum and


Praetorium Latobicorum are centres of diffusion, with many links
radiating into their local area as well as over much greater distances.
Carnuntum has most connections with the Rhine sites; Apulum linking
with Moesia Inferior and Superior as well as Dacia; and Praetorium
Latobicorum acting as a bridge between both of these areas and Italy
and Dalmatia. Balaklawa acts as a centre only for the new sites in
Moesia, underlining the maritime link that was made between these
two areas on the previous map.
Rome as a centre is on a smaller scale than might be expected,
linking with Africa and internal Italy rather than the provinces. This
suggests that the cult was not diffusing from the city but rather that
people were coming to Rome, and dedicating while there. The number
of military inscriptions supports this suggestion.
The network in Britannia appears quite introspective, although
the links that are made between the cult cluster in Germania Inferior
and that on Hadrians Wall provides a deepening of the corridor of cult
diffusion along the Rhine between Britannia and the centre at
Carnuntum.
The network around Doliche grows a small amount with the
introduction

of

Dura

connections

with

the

Europus,
western

but

does

network,

not

gain

unsurprising

any

further

given

the

geographical distances involved. The cult at Dura is known to have


149

been a preserve of the military, suggesting that although it connects


with the much earlier sites in the region, this gives something of a false
impression of its nature.
AD 212-253: Map 4E
The third map in the series demonstrates the increasing popularity of
the cult of Jupiter Dolichenus in the first half of the third century, with
the network thickening but essentially retaining the same shape. The
major change is seen in the clustering of the network at the edges. In
Britannia, it is particularly introspective, but the crop of new
dedications in Germania Inferior and the coast of Moesia Inferior
suggests that this was a period of increasing local interactions.
By contrast, in the centre of the network around Carnuntum,
Apulum and Praetorium Latobicorum, there are markedly fewer new
cult locations, with a few in northern Dacia and some along the course
of the Danube.
Some new coastal sites in Italy and Sardinia increase the
centrality of Rome in this period, but the general picture remains the
same: Rome was a place of cult reception rather than diffusion. The
network around Doliche expands with the dedications from Caesarea,
Perrhe and Hierapolis. These new sites are probably connected with the
military networks seen in the west, as the Dura dedications above, and
indeed, a soldier gave the Perrhe dedication.263
AD 253-300: Map 4F
Although there are a few dedications from this period, only two occur in
a new site, Gerulata, near Carnuntum (CCID 235) and Stobi (CCID
48), which hardly changes the network at all. Other places where late
dedications are found include Rome (CCID 384); Colonia Ulpia Ratiaria
(CCID 111); Dura Europus (CCID 33); and Lambaesis (CCID 621263

Blmer, M., Facella, M., Ein Weihrelief fr Iupiter Dolichenus aus der Nekropole
von Perrhe, PATRIS PANTROFOS KOMMAGHNH, Neue Funde und Forschungen
zwischen Taurus und Euphrat, Asia Minor Studien, Band 60, Bonn: Habelt, 2008, pp.
189-200.

150

623), but in these places simply represents a continuation of the cult.


However, there are only a few late dedications: the cults popularity
was waning. Reasons that might explain this decline include the
devastation of the area round Doliche, and possibly the temple itself, by
Shapur I, increased pressure from other religious movements (in this
case the likely candidate would be Christianity) or because the network
itself had begun to dissolve. These suggestions will be examined in
further detail in the final section of this chapter.
The above network analyses have supported the conclusions
drawn from the epigraphy that there was a cult network separation
between the region of Doliche and the west. The initial Proximal Point
Analysis has highlighted the anomalies in the network, and the
geographical and social spaces and barriers that enabled or hindered
cult diffusion. Through the four temporally developing networks, the
analysis has demonstrated the probable cult centres, and also the
increasing localisation of interactions towards the middle of the third
century.

Beyond the military: networks and receptive social space

The discussion above has focused on the transmission of the cult of


Jupiter Dolichenus through military avenues, diffusing across the
Roman Empire via the in-place network of officials in the army.
However, as noted above, only half of the dedications can be connected
with the military. This final section places the rest of the dedications in
context: examining the centres of diffusion highlighted by the network
analysis, the role of Syrians and priests, and the military-civilian
interface.

151


Centres of diffusion
Three places were highlighted in the network analysis as centres of
diffusion, because they appear early in the evidence: Apulum in Dacia,
and Carnuntum and Praetorium Latobicorum in Pannonia Superior. The
evidence supports Apulum and Carnuntum as correctly identified, but
Praetorium Latobicorum is a network miss, and an alternative regional
centre at Virunum is suggested.
Apulum was base camp for XIII Gemina when Dacia was under
Roman control, and appears in the evidence in the period of Antoninus
Pius (CCID 151), where Jupiter Dolichenus is addressed with the
phrase, nato ubi ferrum exoritur, born where iron springs forth,
thought to refer to the iron-rich Taurus mountains. A total of five
inscriptions are known. XIII Gemina is implied in CCID 155 and 156,
and explicitly mentioned in 154, given by the Syrian priest Flavius
Barhadados on behalf of the legion. The god of Commagene and a priest
of Antioch (perhaps the same Flavius Barhadados?) are mentioned in
CCID 152; and Syrian traders dedicated CCID 153. These pieces of
evidence make both strong Syrian and military connections, and as the
cohort II Flavia Commagenorum are known from Dacia, it makes it
reasonable to support the network identification of Apulum as a place
for cross-fertilisation of religious ideas and a regional centre of the cult
of Jupiter Dolichenus.
Carnuntum was base camp for XIIII Gemina and a Dolichenum is
known from the evidence by the period of Hadrian (CCID 217). The
town was a major redistribution point, situated on the Danube and as
the end point of the overland Aquileia-Carnuntum route. 264 Sixteen
dedications come from Carnuntum, eleven from the Dolichenum in the
civilian quarter of the city. However, explicitly military inscriptions
come from both the civilian city (CCID 222; 223) and the legionary
264

Fulford, M., Territorial Expansion and the Roman Empire, in WA, vol. 23, no. 3,
1992.

152

(CCID 232), suggesting that the division of the two areas is somewhat
arbitrary, and that soldiers and civilians intermingled freely in
Carnuntum.
Two dedications were given by military men not of XIIII Gemina:
CCID 223, given by a centurion of X Gemina and CCID 222, given by a
centurion of XIII Gemina, previously of X Gemina. XIII Gemina was at
this time based in Apulum, a considerable distance away, implying that
this man retained strong ties both to his old legion and old temple.
These dedications support the notion implied by the network analysis
that the Dolichenum in Carnuntum was an important inter-regional cult
centre, and this assertion is backed up by the Oriental priests
Antiochus and Marinus known from CCID 229. Many other priests
are also known here, but do not have any obvious Syrian connections
(CCID 219; 220; 221), and CCID 221 also mentions a Greek scribe,
Zosimus, which suggests the presence of native Greek speakers in
Carnuntum. It could be hypothesised that the town, as an established
place of interaction between civilian and military, was also a place
where the more esoteric aspects of the cult were discussed and
expounded by priests of the cult from both east and west. The role of
priests of Dolichenus will be discussed in further detail below.
Three dedications have been found in Praetorium Latobicorum,
and although the worship of Dolichenus was established early here a
beneficarius of XIIII Gemina donated in AD 138 (CCID 275) it does
not appear to have been a major centre of diffusion. The other two
inscriptions make no explicit military connections, but three men who
donated

to

Jupiter

Dolichenus

and

Jupiter

Heliopolitanus

call

themselves fratres (CCID 274), which may refer to either military or


religious brotherhood. A temple certainly stood here at least by end of
the second century, as an inscription mentions the building and
furnishing of the temple (CCID 640).
The network analysis therefore picked out two of three major
centres of cult diffusion, and they are noteworthy because of the
153

presence of Syrians and priests of the cult. The evidence from


Praetorium Latobicorum does not support the suggestion that it was a
centre. It may be that there is further evidence from the site that has
not yet been found, but also, because the analysis is simply weighted by
date, Praetorium became a centre in network terms, but the real
centre should instead be looked for nearby, perhaps at Poetovio, Aquae
Balissae or Virunum. Jupiter Dolichenus was worshipped at Virunum as
early as c. the mid-second century, and eight inscriptions have been
found in the Dolichenum there. Some Syrians are known at Virunum too
(CCID 330; 335), lending support to the suggestion that it might be
the centre missed by the network analysis.

Syrians and priests


It would appear from the above analysis that Syrians or Orientals found
in the Dolichenian corpus are connected with the centres of cult
diffusion in some way: not, as has been shown above, as members of the
military diffusion, but in another function, perhaps as officials of the
cult. Where a Syrian states an occupation, it is almost invariably nonmilitary, and moreover, a large number of Syrians or Orientals are also
identified as priests.
That the priesthood was a specific occupation is clearly visible
from the epigraphic evidence. Speidel suggests that perhaps oriental
religions needed holy men of special birth or initiation [] to direct the
rites of worship and guard the doctrines of the faith 265 and that this is
particularly evident in the cult of Dolichenus.
There are eighty-six priests known from the evidence, 266 doing the
dedicating themselves in thirty-five instances, or being mentioned in a
dedication in twenty-four instances. A single priest either dedicating or
being mentioned occurs thirty-six times; two or three priests together
265
266

Speidel, Dolichenus, p. 46.


Not, however, from eighty-six separate inscriptions.

154

twenty-four times. Just under half (forty-one) of all the priests known
are of oriental background, twenty-four with explicitly Syrian or Semitic
names, and seventeen with Greek names. Only six of this total also
makes a connection to the military. A further forty-five known priests
reveal no information about their ethnicity, and only seven of these
make an explicit connection with the army.
It is clear that there was a high number of Orientals or Syrians in
the priesthood, suggesting that within the cult of Dolichenus, cult
officials were religious specialists with a connection to the region of
Doliche or to Syria, imported to administer the cult on the frontiers.
However, the presence of an equal number of men without explicitly
Oriental names or links suggests also that non-Syrians were accepted
into the priesthood, and trained on the job or in situ by the established
priests. The relatively high number of triads or dyads of priests support
this interpretation: in a number of cases, a non-Oriental name will
appear alongside Syrians or Greeks (CCID 60; 61; 77; 207; 371; 408).
In all cases, the connections with the army are very infrequent,
implying

that

priests

were

separate

class.

However,

cult

administration and military participation were not mutually exclusive,


and the example of Ulpius Chresimus (a Parthian priest in Rome, who
also seems to have been part of the army CCID 419; 420) leads
Speidel to conclude that there was no Roman policy to preclude a
soldier from being involved in the priesthood. This suggestion may be
supported by the markedly uneven geographical distribution of priests.
Nineteen can be identified in Rome; twelve are known from Moesia
Inferior; seven from Dacia; and six from Pannonia Superior, while there
are no known priests in Britannia or Germania Inferior, and only two
from Germania Superior. It would be reasonable to suggest that more
religious administrators are found in places where the cult was most
popular, as implied by the evidence from Dacia, Moesia and Pannonia.
However, it fails as an explanation when it is recalled that there are
enormous numbers of worshippers in Britannia and the Germanies,
155

suggesting that in these places, the cult operated differently. The


evidence may simply be unfortunately lacking, but this is very unlikely.
In the northern provinces, it may be that priests of Dolichenus either
did not name themselves as such, or people who were not specifically
trained as priests administered the cult.
The conclusions that can be drawn from this evidence is that
priests

of

the

cult

were

regularly

transported

or

transporting

themselves to the west from Syria and the eastern provinces for the
purpose of administration of the cult. They presumably formed a
regularly communicating network of religious brethren of Jupiter
Dolichenus, but that this was not an entirely Oriental preserve. The
numbers of non-Oriental names, coupled with the gaps in the evidence
for priests and Syrians in the northern provinces, combine to suggest
firstly, that non-Orientals were trained in the priesthood of Dolichenus
and secondly, that in some areas, the distinction between religious
official and ordinary worshipper may have been considerably more
blurred.

The military-civilian interface


Roughly half the inscribed dedications to Jupiter Dolichenus can be
identified as military, a further thirty-six given by priests, and the ten or
so traders known from the evidence have been dealt with above. This
still leaves a considerable number of dedications that either give no
indication of their military or non-military status, or were dedicated by
civilians. Who were they, and what can be understood from their
inscriptions about the cults arrival and adoption?
Although there are so many inscriptions without a clear military
marker, it is nevertheless difficult to find purely civilian contexts along
the Danube and Rhine limes. The presence of the army in these border
zones was fairly ubiquitous, and the epigraphic evidence for civilians in
156

these areas show that most are closely connected to military contexts,
for example, the close interactions between military and civilian in
Carnuntum, discussed above.
Dedications by women are an obvious example of the inclusion of
civilians within the cult. However, although the cult of Dolichenus was
apparently more inclusive than the cult of Mithras, female dedicants
are still infrequent: the votives from Mauer-an-der-Url; the lone cult
dedication from Belgia (LAnne Epigraphique 2002, 1011); the two
women whose dedications constitute the finds from Trieste (CCID 445,
446); and Magunna, who gave an altar in Blatobulgium in Britannia
(CCID 555). It is generally the case that where women are found in the
dedications, it is within a familial structure either as wives, mothers or
daughters.
Serving soldiers were not allowed to marry until the time of
Septimius Severus, and the paucity of native named women in the
evidence Claudius Rufinuss wife, Octavia Comsilla in Virunum (CCID
331); Magunna in Blatobulgium (CCID 555); and Matugena in Maueran-der-Url (CCID 314) confirms that, although plenty of relationships
between native women and military men must have existed, these
relationships were not formalised. The women from the rest of the
family dedications generally have ordinary Roman names; although
occasionally Greek or Oriental names are attested, Veratia Athenais in
Lambaesis, (CCID 622); Marina in Mauer-an-der-Url (CCID 305); and
Apollonia in Sucidava in Dacia (CCID 176).
The local civilian populations would have been engaged in
providing the necessities for the conquering armies food, building
materials, clothing and equipment. There is some evidence of the
supply chain that existed to provide the peripheries of the empire with
this support from the core, for example, grain from Egypt, or oil from
Baetica.267 Much of the armys supplies must have been sourced locally,
as tax in kind for the empire. Civic decurions that may have been
267

Fulford, Territorial Expansion, p. 298-99.

157

involved in this interaction are found a number of times, some of whom


were also explicitly connected to the military: a municipal decurion of
Murselensium is found in Acumincum (CCID 208); a town councillor
from Aquincum found in Thrace, see also below (CCID 52); the
decurion of the colony of Apulum (CCID 157); three civic leaders, one
of whom was a veteran and decurion (so probably an example of a
military decurion) in Porolissum (LAnne Epigraphique 2001, 1707);
another municipal decurion who was also connected with the military in
Brigetio (CCID 238); a town councillor of Dionysopolis in Moesia who
was also a beneficiarius of I Italica, (CCID 71); and the decurion from
Seleucia Zeugma found in Brigetio in Pannonia who may or may not
have been civic (CCID 239). It is clear that many of the men involved in
running the towns were ex-military officials, and even when there is no
military connection, the location and context of the town is in the
military zone.
People who were definitely civilians are more readily found in the
non-border zones, Italia, Dalmatia and Thrace. Some fairly clear
conclusions can be drawn from a brief survey of these data.
In Dalmatia, four out of seven inscriptions mention Greek or
Syrian names: Euphemus at Arupium (CCID 121); the priests Flavius
Faladus and Domitius Apollinaris at Narona (CCID 124); the priest
Aurelius Germanus Barhala and his Syrian wife at Salona (CCID 123);
and a man who explicitly calls himself Syrian from Prizren, in the high
mountains of Illyria (CCID 126). His dedication also makes a military
connection and also features Asclepius, Hygeia and Telephoros,
suggesting that this man associated Jupiter Dolichenus with healing
deities. Another priest is known (CCID 125), suggesting that the cults
occurrence in the province was largely to do with the religious officials
who administered it. This suggestion that that cult in Dalmatia was the
preserve of easterners is supported by the fact that each of the seven
dedications are single finds, there is no evidence for wider communities
of Jupiter Dolichenus worshippers, and the find spots are mainly located
158

in settlements on the coast. Dalmatia is very mountainous internally,


and the cult does not appear to have spread to the native civilians.
In Thrace, the earliest evidence is dated to the period of
Septimius Severus, suggesting that the cult in this province was a fairly
late arrival. This first dedication is military recording the building of
the temple to Dolichenus in Cabyle (LAnne Epigraphique 1999,
1374). The rest of the dedications from the province either have
connections with Oriental or Greek priests (CCID 50; 51; 54) or link to
the northern frontiers (CCID 52). Even though the dedication is by a
civilian trader, it has an implicit link with the military because the
dedicant was originally from Aquincum.
Italy provides a number of inscriptions to Jupiter Dolichenus that
have no clear military link, but often, these are single finds. A lone
dedication suggests that there was not necessarily a wider community
of worshippers. An organisation of veterans in Ateste accounts for the
only group of worshippers known outside of Rome (CCID 451). This
may help to indicate the nature of the dedicants in the lone find spots: a
candidatus dedicated at Brixia (CCID 453), and other men dedicate
singly at Padua (CCID 450), Aecae (CCID 461), and Bononia (CCID
454). Perhaps they had encountered the cult during military service.
Speculations aside, there are explicitly military dedications all dating
from the 180s from Ravenna (CCID 456), Ostia (CCID 440; 441) and
Iulia Concordia (CCID 449), places where there is more than one piece
of evidence, supporting the notion that the introduction of the cult into
northern Italy was via the military. Two freedmen are found in Italy, one
in Ariminum who has an Oriental connection (CCID 458); and one in
Tusculum (CCID 444). Priests account for the dedications from
Caesena (CCID 457) and Tarracina (CCID 462)
From this analysis of the evidence for civilian worshippers, it is
clear that there are still strong ties to the military. Where there are
dedications without clear links to the army, they have Syrian
connections, and are most often dedications by priests. The cult did not
159

transmit profoundly into the civilian populations where there were no


military networks to facilitate that transmission.

Conclusions: Networks and places for new cults

This chapter has shown that the traditional explanations for the success
of the cult of Jupiter Dolichenus within the military and in the west are
not sufficient. Understanding the adoption of the cult as a social
process driven by social networks, by contrast, allows the evidence to
be viewed from a different angle, and supports the interpretation that
this particular innovative religious movement was transmitted swiftly
across a network of military officials. The evidence suggests also that
the cult came to a sudden end. Can this same network explain the
reasons for this?

The end of the cult: a network explanation


Various suggestions have been made for the sudden end to the cult:
that the temple in Doliche may have been destroyed by the Sassanian
invasions of Shapur I in the mid-third century and that this caused the
centre of the cult to fall apart; or that it could not compete with other
new religious movements, specifically Christianity. Both of these
explanations rely on an intrinsic quality of the cult itself: either that
when the gods temple at Doliche was shown to be fallible, people lost
faith in the god himself, or that the ideology espoused by Christianity
was superior. I have shown in this chapter that the cult of Jupiter
Dolichenus spread across the social network of official military men

160

that communicated frequently. The end of the cult can be explained in


the same way.
The fifty years following the death of Severus Alexander in 235
was a time of considerable internal discord, described as a period of
military anarchy268 marked by the habit of treason.269 The mounting
pressures on the Rhine and Danube frontiers as well as losses in Syria,
led to rivalries in the military, civil war and a succession of pretenders
to the Imperial throne. Decius supplanted Philip the Arab in 249, and
left the Danube border open to the Goths and various other invading
tribes, who had been pushed into the Balkans by the incursions of
nomadic tribes from central Asia. Decius was killed fighting in the
Balkans in 251 and the governor of the Moesias was proclaimed
emperor by his troops, only to be usurped by his successor in 253, who
survived three months.
Romes frontiers were in disarray and under attack from all sides,
and, moreover, plague had decimated the army and the empire when
Valerian was proclaimed in 253, and restored some measure of
discipline in the military forces.270 The frontiers of the north and the
east required two supreme military commanders, and Valerians son
Gallienus attempted to take control of the Rhine-Danube borders while
Valerian confronted Shapur in the east, was captured by him and died
in 260. In the years that followed, Spain and Gaul elected a civilian
empire and set up their capital in Bordeaux, Zenobia expanded
Palmyras empire to Antioch and lower Egypt, and the emperor
Claudius Gothicus defeated and settled the Goths on the Danube. His
successor Aurelian evacuated the army and civilians from Dacia in 270,
and moved on to destroy Palmyra and the Gallic empire, earning the
title

Restitutor

Orbis

in

273.271

Diocletians

appointment

of

Maximianus as Caesar was a different approach to government,


268

Boardman, J., Griffin, J., Murray, O., eds., The Roman World, Oxford: Oxford
University Press, 1986, p. 443.
269
Cary, M., Scullard, H. H., A History of Rome, 2nd Edition, London: Macmillan Press,
1975, p. 508.
270
Cary, Rome, p. 509.
271
See Cary, Rome, p. 509-516.

161

acknowledging that, with unstable frontiers on all sides, a unified


empire was impossible under only one man.
The military and political turbulence of this period coupled with
the plague meant that the established military communications of
central Imperial governance crumbled. This network and the military
officials were the backbone of the cult of Jupiter Dolichenus, supported
by the Syrian priests. The area of Doliche having been overrun by
Shapur meant it was unlikely that cult officials from Syria were able to
interact with their compatriots in the west, and, moreover, the cultic
heartlands of Dacia in the west were evacuated. With these various
fractures all occurring at the same time, the cult of Jupiter Dolichenus
could not sustain beyond the end of the third century, although the few
dedications between 253-300 testify to its limping on in certain places:
Rome, Gerulata in Pannonia, Colonia Ulpia Ratiaria, Virunum and
Lambaesis.

Networks and cognitive space


Speidel observed that the cult followers cover a wide range of social
classes, but they are found mainly on the frontiers and in cosmopolitan
cities, places inhabited by floating populations. Senatorial elites,
soldiers, traders, freedmen, slaves, and aristocrats in the imperial
service272 are all represented, and are all to some extent removed from
their usual social environment. It has been shown that the cult
disseminated across a network of military men that was already in
place, men outside of their traditional context, and so, in Starks terms,
free to deviate from their traditional deities, open to new religious
forms, especially those espoused by other members of their social class.
Transient people have reconfigured social networks; they are likely to
be vulnerable to the social benefits of cult membership. The uncertain
272

Speidel, Dolichenus, p. 76.

162

collective cognitive space at the edges of an empire means it is entirely


logical that new cults spread quickly in these places.
They did not, however, transport the cult back to the places from
which they had floated: and as has been seen, the internal landscapes
of Spain, France, Italy, Greece, Thrace and Asia Minor are almost
entirely untouched by the cult. Where the cult is found in these places,
it is most often as a single dedication, suggesting that, without the
social network of fellow worshippers, they did not continue to revere
Jupiter Dolichenus when they returned home to their traditional gods. It
seems that worshippers engaged with the cult in a temporary way
during their time on the frontiers, where the cult was fashionable.
The cult of Jupiter Dolichenus jumped certain areas: Asia Minor
and Greece in the east and Hispania and Gallia in the west. Speidel
suggests that these lacunae can be explained because the cult had no
appeal for the conservative people of the countryside or in cities with
stable populations.273 This classification of the populations of these
places as conservative makes value judgements about both the nature
of the cult of Jupiter Dolichenus and about the people who worshipped
or did not worship him. The inhabitants of these areas were simply not
part of the military networks that transmitted the cult: it was not the
case that there was no cognitive space for religious innovation in these
areas, or no social need for it, as will be seen in the case of Theos
Hypsistos (chapter 6).

273

Speidel, Dolichenus, p. 76.

163

Chapte r 5.
The Jewish Diaspora.
Ethnic networks and the activation of Jewish
identity

Introduction

This chapter uses the Jewish communities in the Roman Empire as a


second case study for the investigation of the role networks play in the
transmission of religious information. The aim here is to examine the
communicative power of the network of the Jewish Diaspora and to
explore how, when and why it was used to diffuse new ideas about
Jewish identity. Assessing the pattern formed by the epigraphic material
and how it changed over time draws a bottom-up picture of the
developments in the Diaspora, and allows an understanding of how
ordinary people marked their ethno-religious identity, and the reasons
why the way they did this might have changed.
Numerous studies discuss the accommodation of Jews within
Gentile society; the different levels of Jewish acculturation; and the
religious and cultural engagement between Gentiles and the Jewish
minorities. By contrast however, this case study examines how Jewish
identity was internally re-formed: it is argued that, if the rabbinic
reforms were necessitated by the destruction wrought in Judaea
between AD 66-135, this cataclysm also activated the ethnicity of the
Diaspora Jews, and that a dramatic reiteration of Jewish identity can
clearly be seen in the epigraphic data from across the Diaspora.
The analysis begins with a brief survey of the central historical
events and scholarly opinion regarding Diaspora Jewry and sources of
131

Jewish difference and disconnect in their Graeco-Roman environment.


The focus is then on the epigraphic data (methodologically slightly
suspect as it may be, pace Rutgers274), although literary sources are
used

as

an

occasional

supplement.275

The

epigraphic

evidence

illuminates the differences in Jewish self-identification in the Diaspora


before and after the sixty years or so between the destruction of the
Temple and the quashing of the Bar Kokhba revolt. It is argued that,
before the destruction of the Temple, Jews in the Diaspora did not need
to define themselves as Jewish, because there was an inherent centre to
their religious life. Post-destruction, the real and cognitive centre of
Judaism was gone, changing Jewish existence forever. Judaism itself of
necessity transformed as a result. Likewise the Jews in the Diaspora,
and the religious authorities in Palestine utilised the newly activated
ethnic network of the Diaspora to transmit the tenets of rabbinic
halakhah, shown epigraphically as a renewed knowledge of the wider
Jewish network. This is visible in terms of self-classification as Jewish,
references to the Old Testament and Jewish Law, the use of Hebrew,
and of explicitly Jewish symbols and names.
Network models are used to support this hypothesis by visualising
how information might have been transmitted across the Diaspora. This
begins with a Proximal Point Analysis (PPA) of the entire HellenisticRoman

Diaspora,

highlighting

centrality

and

isolation

from

the

geographical patterning alone; and then moves into a developed PPA


network, mapping the Hebraization process between the second and
fifth centuries AD. These analyses highlight the dialogue that took place
intra Diaspora and reconstruct the potential routes used to spread new
religious information across the network.
Finally, it is suggested that the spread of rabbinic halakhah can
be understood an information cascade across the network of the Jewish
274

Rutgers, L. V., The Hidden Heritage of Diaspora Judaism, Leuven: Peeters, 1998, p.
31-41.
275
The literature has been noted to be widely varying in reliability as evidence:
ranging from Josephus exaggerated claims for the legal status of Alexandrian Jews to
the conceivably Jewish nomenclature in summary papyri fragments.

132

Diaspora, and if this is the case, then it is suggested that this same
phenomenon of information cascade may be instructive when thinking
about the rise of Christianity.

Background: Jews in a Gentile world

It is first necessary to situate the Diaspora within an historical


framework. At various points, the Jews were dispersed, whether
through enforced resettlement as prisoners of war or as refugees from
conflict in Judaea, or voluntarily due to economic incentives. This
section comprises a brief look at accounts of early dispersions and
perceptions of Jewish identity in the early Diaspora. The discussion then
moves on to assess scholarly opinion on the various issues that being
Jewish in the Hellenistic-Roman environment entailed, and how it
impacted on both Jews themselves and their Gentile neighbours.
In different periods and places, and for different individuals, the
integration or separation of Jews within the Graeco-Roman environment
ranged from total assimilation to rigorous separatism. The Jewish
experience differed widely across time and space, and as such, it is
extremely difficult to make claims about the Jewish Diaspora as a
totality. However, regardless of the variations inherent in Jewish
participation or non-participation with their Diaspora environments, the
presence of any culturally and ethnically different or separate group in
a dominant276 population causes a level of dissonance.277 At certain
times, but by no means at all times, this was surely the case for both

276

Although the varying populations across the Roman Empire may not have had a
dominant ethnicity, there were dominant cultural aspects.
277
See Collins, J. J., Between Athens and Jerusalem, New York: The Crossroad
Publishing Company, 1986, p. 8, following L. Festinger, A Theory of Cognitive
Dissonance, Evanston, Ill., and White Plains, N.Y.: Row, Peterson & Co., 1957.

133

the Jewish settlers and for their various neighbours in the Roman
Empire.
The aspects of Jewish identity most likely to cause dissonance
would have been those most difficult to reconcile with Roman life. This
included both practical social issues such as food prohibitions,
circumcision, Sabbath observance, and family values, and theological
issues such as belief in one abstract, aniconic God. These aspects of
Jewish religious or cultural practice are then taken as markers
representing Jewishness. While it is true that the real picture of the
Jewish Diaspora is far more subtle than this simplistic labelling allows,
it is these socio-religious aspects of Judaism that became its indicators,
and so it is these that allow a modern observer to attempt to
reconstruct the situations of Jews in the Graeco-Roman world.

The Graeco-Roman Diaspora


The actual history and the perceived notions about the history of the
Diaspora are difficult to disentangle. The testimonies we have to the
early dispersions are almost entirely Biblical, although they are
supplemented by later Jewish authors such as Josephus, who estimates,
for example, the numbers of Babylonian Jews who returned to Israel
during the Achaemenid period. It is fair to say that the Assyrian and
Babylonian deportations happened when the religion of Israel was still
much more fluid, less consolidated and less differentiated from
paganism than it became later. 278 Judaism needed to flex and respond
to the new situation of the Jewish people, and a considerable part of the
O.T. is given to theological explanations for the various destructions and
deportations wreaked upon the Jewish nation during the Assyrian and
278

Schrer, E., The history of the Jewish People in the Age of Jesus Christ, vol. III.i,
revised version: Vermes, E., Millar, F., Goodman, M., Edinburgh: T & T Clark, 1986, p.
5. All references to Schrer hereafter are to the revised edition of the text, marked as
Schrer2.

134

Persian periods. Dispersion is seen as a punishment 279 by God for the


sins of the Jewish people: at Gen. 15.13, God says to Abraham, Know of
a surety that your descendents shall be immigrants in a land not their
own, where they shall be slaves, and be oppressed for four hundred
years.280
The articulation of Jewish identity through expulsion and exile is a
major theme of the O.T. Exodus records the enslavement and
oppression of the Israelites and their expulsion281 from Egypt. The
subsequent return to the Promised Land under the guidance of Moses
is marked by the detailed exposition of Jewish separation and
difference: through the details and laws of garment and sacrifice given
to Moses on Sinai, affirmations of the superiority of the Israelites, and
commands to show no tolerance for other religions in Judaea and to
destroy their idols and temples. Exodus and Leviticus mark out Gods
requirements of the Jews, detailing the practical Laws that mark the
Jewish peoples covenant with God, and it is here that the status of the
Hebrews as distinct is really explicitly stated, linking with the themes of
expulsion and rejection. The Jews minority experiences and the
hostility shown them by their environments create an initial separatist
understanding of Judaism. It is notable that this also entails selfreflexive, tolerant morals: at Lev. 19:33, Gods commandment to the
Jewish people is not to oppress the alien in their midst, because they
were once aliens in Egypt.
It is apparent that the early determined and strengthened Jewish
self-identification, and Judaism only took a normative shape as a
response to the destruction of the first Temple and the initial dispersion
279

It is instructive to recall the present-day ultra-Orthodox Iranian Jews whose


agreement with the political authorities in Iran that Israel should not exist stems from
an understanding of dispersion as the will of God: divine punishment of the tribes of
Israel, as demonstrated at Lev. 26:33, You I will scatter among the nations, and I will
unsheathe the sword against you.
280
All Biblical quotations are taken from the NRSV, Oxford: Oxford University Press,
1995.
281
Peter Schfer notes the dichotomy between the version of the story in Exodus,
which tells it as a voluntary departure of the Jews, and the Egyptian and GraecoRoman traditions of the story where the Jews were expelled for ethnic reasons.
Judeophobia, Cambridge: Harvard University Press, 1997, p. 15.

135

into Babylon: defining more strongly what constituted being Jewish


became more necessary as an ethnic minority. The physical distances
between communities combined with the heightened perception of
difference between themselves and their Gentile neighbours can be
usefully

understood

as an illustration

of Irad Malkins

elegant

suggestion that the distance creates the virtual centre.282


Following these earlier exiles into the east, according to Josephus
the dispersion of the Jews into Asia Minor began under Antiochos III, c.
210-201 BC.283 Alexander enforced resettlement of some Jews in Egypt,
and the Letter of Aristeas records the transportation of one hundred
thousand Jews to Egypt following Ptolemys invasion of Judaea, and the
settlement of thirty thousand of these in military colonies, adding to the
already

established

communities.284

Although

he

allows

for

the

possibility of coercion, Collins views the early western Diaspora as not


due to any external compulsion,285 and the massive growth of the
population of Jews in Egypt and the Hellenistic world as an economic
response to commercial intercourse, a view shared by Lightstone. 286
Aside from the Hellenistic resettlements, Pompey is known to have
captured and transported Jewish prisoners of war to Rome when he
took Judaea in 61 BC, and many Jewish slaves were sold after the
crushing of the Bar Kokhba rebellion. 287 It seems therefore that much of
the origin of the western Diaspora can be linked with enforced
resettlement, but it must also be assumed that some Jews in the
Graeco-Roman period moved and added to the communities outside
282

Malkin, I., Networks and the Emergence of Greek Identity, in MHR 18, 2003, pp.
56-74.
283
Josephus, Ant. 14. His quotation of a letter at 12.147-53 records the transportation
of 2000 Jewish families to fortresses and strategic places in Phrygia and Lydia, who
were given land and permitted to live by their own laws. The authenticity of the
documents Josephus uses has been questioned, and it may be that this letter was an
apologetic document penned by Jews themselves. Barclay, J. M. G., Jews in the
Mediterranean Diaspora, Berkeley & Los Angeles: University of California Press,
1996, pp. 260-262.
284
Letter of Aristeas, 12-14.
285
Collins, Jerusalem, p. 3.
286
Lightstone, J. N., Migration (Forced and Voluntary), Communication and the
Transformation of Judaism in the Greco-Roman Period: Prolegomena, paper given at
the CSBS meeting, 2006, p. 2.
287
Schrer2, History, vol. I, p. 553.

136

Judaea of their own free will. Voluntary as opposed to enforced exile


entails a profoundly different cognitive reaction to and interaction with
the Gentile environment. For example, to what extent was Judaea
perceived as the homeland by voluntary exiles, did it come to be
viewed as simply the religious centre, rather than political or cultural?
How did Jews define themselves within a chosen alien society and how
did these societies themselves respond to Jewish belief and practice?

Aspects of Diaspora
This section briefly examines aspects of Jewish culture that were
potential

sources

of

disconnect

or

interest

between

Diaspora

communities and their neighbours, and some of the scholarly opinion


about the integration or rejection of Jews in the Diaspora. Marking of
Jewish identity through adherence to food laws, the practice of
circumcision, keeping of the Sabbath and worship of a single deity
comprise the most recognisable and stark social and theological
divisions between Jews and Gentiles. In particular the food laws and
Sabbath observance, (and to a lesser extent and within certain
environments such as the gymnasium only, circumcision), would have
been the most obvious to non-Jews, and which can be taken as standard
public markers of Jewishness.288 It is notable therefore, that it is these
aspects that are often played down in Hellenistic Jewish literature, in
order to find social favour and acceptance.289 Theoretically, adherence
to the Laws entails deeply ghettoised living arrangements, but for most
of the time and in most places Jews seem to have lived side by side with
Gentiles in the cities and countryside. There are, however, some
examples of Jewish districts in Hellenistic Egypt and according to
Josephus, also in Sardis, where some resident Jews are likely to have
had decreased contact with the Gentile world. Barclay makes the
288
289

Collins, Jerusalem, p. 7.
Collins, Jerusalem, p. 143.

137

important point that levels of assimilation were likely to have been


different between genders: the least assimilated Egyptian Jews were
Jewish women who lived in wholly or largely Jewish districts. 290
However, what is apparent from Barclays analysis of the sources is that
adherence to these supposedly canonical regulations differed widely
both within different communities and more generally across the
Diaspora in its entirety. Moreover, levels of assimilation or strictness
differed across time.
The advent of Roman control of the Mediterranean led to issues
about who had access to Roman citizenship and its privileges. A major
point of conflict between Jewish and Roman norms occurred in the
context of military service, both because Jewish-Roman soldiers could
not observe the Sabbath and because they were obliged to show loyalty
through religious acts, which were incompatible with strict Judaic
observance. Otherwise, Jewish monotheistic belief in the face of a
dominant polytheist culture seems to have been problematic only when
brought into relief against the social aspects of traditional pagan
religious involvement: i.e. the partaking of sacrificial meat at communal
civic festivals. The political situation in Alexandria at the turn of the
millennium, where social demarcations had been historically drawn
between Greeks and non-Greeks, highlights the problems involved in
deciding citizenship. The Alexandrian Greeks were automatically given
Roman citizen status, whereas the rest of the population of Egypt,
including the Jews, were subject to the laographia, or poll tax,
introduced by Augustus in 24/23 BC. 291 The apostasy that Roman
citizenship required was violently rejected by the author of 3
Maccabees, but the exemption from the tax and the benefits of
citizenship was still an aspiration for some Jews, and some did
apostatise.292 Tiberius Julius Alexander, nephew of Philo, is the most
famous example of an Alexandrian Jew giving up his heritage in order
290
291
292

Barclay, Jews, p. 118.


Barclay, Jews, p. 49.
See Collins, Jerusalem, pp. 102-111.

138

to participate fully within the administrative structures of the Roman


world, having a public career, and eventually becoming prefect of
Egypt.293
However, although there were certainly problems at times,
aspects of Judaism were also clearly attractive to Gentiles. Jewish
monotheism was understood by some within the Hellenistic-Roman
world as part of an abstract philosophical framework, and was clearly
attractive to certain sectors of the community, which will be explored
further in the analysis of the spread of Jewish beliefs beyond the
boundaries of ethnicity, forming part of the discussion of the cult of
Theos Hypsistos in the following chapter. It is noteworthy that Judaism
was particularly appealing to the upper classes in Rome, with famous
examples of Judaizers being Poppaea, mistress and then wife of Nero,
or Domitians cousin and his wife, who were respectively executed and
exiled on charges of atheism.294 The appeal to elites (although not elites
only) is an important social aspect of Judaism, and shows that Jewish
beliefs

and

culture

cannot

have

been

considered

as

entirely

contemptible to Romans as Juvenals Satire 14 might imply.

The Temple and the Synagogue


As a direct result of exile in Babylon and the destruction of the first
Temple, the essential centre of what had originally constituted Judaism
was removed, and a new type of community-based centre was of
necessity instigated: the replacement of the Temple with a protosynagogue arrangement, and prayers and the reading and exposition of
scripture for animal sacrifice. Pilchiks observation that Ezekiel already
knew of the assemblies of the exiled in Babylonia, those informal
293

Barclay, Jews, p. 105-106.


Josephus, Ant. 20.195; C.A.2.255, 280; Dio, Roman History, 67.14.1-3. See also
Mitchell, S., Herods People: Roman-Jewish Sympathies in the rise of Christianity, in
TLS, March 6, 1998; Liebeschuetz, W., The Influence of Judaism among Non-Jews in
the Imperial Period, in JJS, Vol. LII, No. 2, 2001, pp.235-252.
294

139

gatherings for prayer and study where the community and the faith was
kept alive,295 indicates that this was a natural result of dispersion.
Irad Malkin has argued that distance creates the virtual
centre296 in his analysis of the beginnings of a universal notion of
Greekness and a recognisable Greek identity during the colonisations
of the Archaic period. It has been forcefully argued by Niehr 297 that this
post-exilic period during which the second Temple was rebuilt was
marked by the fundamental development of the central notion of
Judaism as an aniconic faith. He claims that the destruction of the first
Temple also involved the destruction of the cult statue of YHWH, and
that various O.T. books, Psalms in particular, make (somewhat
obfuscated) reference to this statue. For example, at Dt. 4:12, 15, the
visibility of the face of God is denied, you saw no form. This, Niehr
argues, reflects the fact that the actual statue, the physical face of
God, had been destroyed. If this supposedly central tenet of Judaism
took shape only during this later period, and as a response to the
destruction of an earlier cult form, then it is reasonable to assume that
other aspects of Jewish identity were also created as a response to
destruction and exile, supporting Malkins theory of the importance of
distance in the creation of unity.
A couple of examples from sixth-second century Egypt support
the notion that the unity of Judaism was still not fully formed. During
the post-exilic period when Deuteronomy was written, the command
that the Jewish God shall only have one Temple (Dt. 12:2-27) was either
unknown or disregarded: the Jewish military colony at Elephantine is
recorded in a papyrus document from 410 BC, and their temple to
Yaho had existed from before 525 BC. 298 It is not certain what exactly
happened in this temple, but the passage demonstrates that in the
295

Pilchik, E. E., Judaism outside the Holy Land, New York: Bloch Publishing Company,
1964, p. 10.
296
Malkin, Networks, p. 59.
297
Niehr, H., In Search of YHWHs Cult Statue in the First Temple, in van der Toorn,
K., (ed.), The Image and the Book, Iconic Cults, Aniconism, and the Rise of Book
Religion in Israel and the Ancient Near East, Leuven: Peeters, 1997, pp. 73-95.
298
Schrer2, History, vol. III.i, p. 39.

140

earlier stages of the Diaspora, Jewish identity was apparently not bound
to the singular Temple in Jerusalem. This group of mercenary soldiers
outside Palestine at least felt it to be no contravention of Jewish law to
establish their own temple, and may even have perceived this to be a
necessary way of showing their piety to the God of Israel.299
Later still it seems that notions of the Laws were flexible enough
for the exiled High Priest of Jerusalem, Onias IV, to build a temple to
the Jewish God in the military colony of Leontopolis in the mid-second
century BC. Jewish cult was enacted there by priests and survived until
c. AD 70. However, it was never regarded as legitimate in Palestine, and
the Egyptian Jews apparently also continued to revere Jerusalem as the
focus of their religion,300 suggesting that by this stage the centrality of
the Temple was universal.
The proto-synagogue arose originally as a response to the
Babylonian exile, but became an intrinsic part of the way Judaism was
practiced. The synagogue of the Graeco-Roman Diaspora, as a place for
gathering, prayer and instruction, had developed further under the
influence of Hellenistic civic institutions. It acted as an architectural,
cultural, and religious focus for Diaspora communities, a role that
further increased after the destruction of the second Temple in
Jerusalem in AD 70.

Gentile reactions to Diaspora: Anti-Semitism


At various moments, certain Diaspora communities suffered extreme
persecution by their neighbours the pogrom in Alexandria in AD 38
documented by Josephus and Philo being the most well known and most
violent but also the destruction of the temple at Elephantine in 410
299

It also shows that in this period Jews were involved in mercenary warfare outside
Judaea in sufficient numbers to merit building another temple in the first place,
supporting Collins observation that in some cases, Jews joined the Diaspora of their
own accord.
300
Schrer2, History, vol. III.i, pp. 145-147.

141

BC, and the various expulsions from Rome. Peter Schfer argues that
these pogroms and the Graeco-Roman literature concerning Jews
reveals a level of anti-Semitism in Graeco-Roman society: To what
degree the Jews were separate is not important [] The only crucial
question is what the Greco-Egyptian and Greek authors made out of it.
They turned Jewish separateness into a monstrous conspiracy against
humankind and the values shared by all civilized human beings, and it
is therefore their attitude which determines anti-Semitism.301
Although there is certainly hostility towards the Jews in some
literature,302 as well as evidence of certain civic authorities hindering
the free practice of Judaism,303 the term anti-Semitism has extremely
strong historical connotations, and its use is inflammatory when
regarding events and attitudes of two thousand years ago. The pogroms
generally represent brief episodes driven by external political unrest;
and the literature is also written with certain political aims in mind.
Tacitus abusive digression, for example, is related to the Roman
destruction of the Temple and the capture of Jerusalem. These violent
episodes and the examples of hostile literature in no way comprise the
whole story of Jewish existence in the Diaspora, nor of the GraecoRoman responses to Judaism. The elite Roman Jewish sympathisers
have already been briefly mentioned above, and Josephus and the Acts
of the Apostles testify to the Jewish communitys open door policy
towards Gentiles, in other words, to Gentile interest in Jewish society
and religion and the acceptance Jews showed to them. Attitudes
towards proselytes and god-fearers will be further discussed below.
These examples stand against the notion that there was in general a
strong level of Graeco-Roman anti-Semitism in antiquity, although the
chronology is important. There is more evidence of toleration and
sympathy in the first century BC/AD up to the war of 66-70 and the fall
301

Schfer, Judeophobia, p. 209-10.


For example, Juvenals Satires; Tacitus, Histories 5.4-5; the Graeco-Egyptian Apion,
who disseminated ideas such as those propagated by Tacitus and against whom
Josephus wrote his counter-apologetic.
303
Schrer2, History, vol. III.i, p. 150.
302

142

of the Temple. Following this, there is more evidence for Roman antiJewishness, in particular under Domitian, and especially under Hadrian,
whose war of 132-5 seems to have been driven by active antiJewishness.
As Drijvers points out, our literary sources [] stress the
differences, so that Gentiles, Jews and Christians appear as almost
totally different groups in society. [] Religious texts stress ideological
differences; religious practice is often a shared experience of a basically
social character.304 Literary elites represent one aspect of society, one
viewpoint, and are subject to propagandising or political machinations.
The evidence for the daily lives and interactions of Jews and their
neighbours presents a different kind of story. Drijvers envisages the city
he studies, late antique Edessa, as a place where Jews, Christians and
Gentiles interacted frequently: [They] did their shopping at the
common market-place, suffered from the same diseases, epidemics and
wars, and therefore shared a lot of ideas and concepts about which they
talked with each other. They were buried in the same cemeteries, caves
around the city, and got the same education if they could afford it. They
lived on each others doorsteps, shared common experiences and
usually spoke the same language.305
Generalisations about Gentile attitudes towards Jews that are
drawn from a few literary sources present a one-sided version of a
much more complex and much more interesting story. Although it is
important to know what the Graeco-Roman literary elites were writing
and thinking, it is equally important to know what divided or brought
together the average Jew and the average non-Jew in their everyday
lives. The epigraphic material goes some way to recovering these
interactions, and suggests that the general attitude towards Diaspora
communities living within the general population of the Roman Empire
was not permanently hostile. Rather, much of the evidence points to a
304

Drijvers, H., Syrian Christianity and Judaism in The Jews among the Pagans and
Christians, Lieu, J., North, J., & Rajak T., (eds.), London: Routledge, 1992, p. 128-9.
305
Drijvers, Christianity, p. 128.

143

level of integration and acceptance of the Jews as part of a plethora of


different ethnicities and religious views that were found across the
Empire. As Rutgers argues, the situation was subject to continuous ups
and downs.306

Expansion of ethnicity I: Jewish-Gentile intermarriage


It has been assumed that Diaspora communities existed because of the
active or passive dispersion of ethnic Jews. This belief is based on the
assumption that Mosaic Law was rigidly observed, and does not take
into account the potential for populations to expand exogamously,
through

marriage

with

non-Jews,

or

through

attracting

Gentile

proselytes.
Marriage to non-Jews is banned at various places in the
Pentateuch, and the book of Ezra condemns the wives and children of
intermarriages to be cast off at 10:11-44. This indicates that when
Ezra was composed in the fifth-fourth century BC, Mosaic Law was not
unequivocally adhered to. To what extent did later Diaspora populations
observe these prohibitions? There is scant epigraphic data referring to
marriage outside Judaism. Only JIGRE, 40 from Leontopolis, or IJO 1,
Ach6 from Larissa, provide possible evidence for intermarriage.
Likewise, there is sparse literary evidence: Acts 16:1-3 records Timothy,
the son of a Jewish mother and a Greek father, and Josephus mentions
the marriage also recorded in the Acts (24:24) between the high-status
Jewess Drusilla, descendent of Herod, and Felix, the Roman procurator
of Judaea.307 This paucity of evidence might suggest that the Law was
generally upheld and that Jew-Gentile unions were rare. However, those
Jews who married non-converted Gentiles probably assimilated with the
Gentile community more fully, perhaps discarding markers of their

306
307

Rutgers, Heritage, p. 22.


Josephus, Ant. 20.141-144.

144

Jewish identity and are so invisible in the archaeological record. 308


Equally, Gentiles who converted and married into Judaism may have
taken on Jewish names or identified themselves by Jewish attributes,
and so are likewise indistinguishable. 309 It is possible that attitudes to
intermarriage varied, but within the Jewish populations that continued
to attest themselves as Jewish, it can be assumed that Mosaic Law was
largely adhered to, but was flexible enough to incorporate converted
Gentiles.
However, there are practical considerations concerning JewGentile marriage, or indeed, marriage outside the home community.
One of these is how large a population has to be to sustain nonincestuous endogamy. Broodbank,310 in his study of settlement patterns
in the early Bronze Age Cyclades, uses the guideline figure of 400
individuals (i.e. between 40-80 families) as the minimum number of
people required to sustain non-incestuous endogamy. Certainly many
Jewish communities, including the metropoleis of Rome, Alexandria,
and Antioch, were large enough for exogamy to be unnecessary, but this
may have not been the case in other, more peripheral Diaspora
communities.
An archaeologically or epigraphically known synagogue should
therefore be understood to indicate that the community was large
enough to sustain itself. Although there is the inherent methodological
problem of chance in archaeological and epigraphic survival, where
there is no synagogue attested, it might be plausible to view the
community as auxiliary, and propose that these places had a need for
more communication with the Gentile environment, or to suppose that
308

Support for the decrease in assimilation following intermarriage is provided by


recent sociological studies of exogamy in modern Jewish communities; see Barclay,
Jews, p. 411, using Ellman, Y., Intermarriage in the United States, JSS 49, 1987, 126.
309
There is also an important methodological issue, pointed out by Daniel Stoekl Ben
Ezra in his review of IJO 3, that records of Jewishness on epitaphs probably represents
only
the
Jewish
hard
core,
BMCR,
2005.02.23,
http://ccat.sas.upenn.edu/bmcr/2005/2005-02-23.html.
310
Broodbank, C., An Island Archaeology of the early Cyclades, Cambridge:
Cambridge University Press, 2000.

145

connections by marriage were made through longer-distance contacts


to Jewish groups in other locations. Depending on the size and growth
rate of communities too small to sustain endogamously, it might be
hypothesised that they died out or disappeared, either because the
Jewish population married Gentile partners and were gradually
absorbed into the Gentile world, or because they tried to retain their
endogamous practice but simply ran out of reproductive steam. It could
be argued, therefore, that dedicated synagogue buildings are only
represented when the community was large enough to sustain one, and
that prior to this, people met in private houses, open air or other
spaces. Indeed, has been argued that the synagogue grew out of the
private house as meeting place.311

Expansion of ethnicity II: Judaizers, proselytes, god-fearers &


mission
If there is scant primary evidence for Jewish-Gentile intermarriage,
then the evidence for inclusive Jewish attitudes towards proselytes and
god-fearers suggests that, by proxy, attitudes towards mixed marriages
may have had an inherent level of flexibility that has been obscured. It
is important to recognise that communities had the potential to grow
through incorporation of Gentiles. Judaism was especially attractive to
women,312 and there is no evidence positively to refute the hypothesis,
which has been widely canvassed, that converts made up a great
proportion of the Jewish population. 313 For Goodman, this shows that
the Jewish nation had accepted the principle that it was open to anyone
to integrate him or herself into its political and social community simply
311

Harland, P. A., Associations, Synagogues and Congregations: Claiming a place in


Mediterranean society, Minneapolis: Fortress Press, 2003, p. 31.
312
Collins, p. 163, among others. Reasons suggested are family values, i.e. lifelong
monogamy and sacredness of all children, social values that were both adopted by
Christianity.
313
Goodman, M., Mission and Conversion: Proselytizing in the Religious History of the
Roman Empire, Oxford: Clarendon Press, 1994, p. 63.

146

by acceptance of Jewish religious customs and that the potential


flexing of communal boundaries entailed by such a notion is quite
astounding.314
Accepting non-Jews into Judaism, providing they conformed to the
social and religious strictures of the faith, is not, however, the same as
the conduction of active mission to gain converts. It has been the
standard position of various scholars since Schrer that, because
Christianity was a missionary religion, active proselytising must have
been a practice shared with contemporary Judaism. Literary sources
are cited as exposing this habit, for example, the comments of Valerius
Maximus and Cassius Dio that the expulsion of the Jews from Rome in
139 BC and in AD 19 was a reaction to active Jewish proselytism,315
snippets of literature that suggest Jews were proud of winning
proselytes,316 or Jesus admonition of the Pharisees and scribes at
Matthew 23:15, who would cross sea and land to make a single
convert. These passages have all been used to argue that active Jewish
proselytism must have existed before 100 AD.
Goodman finds the evidence used to present the case for active
Jewish mission in this period substantially lacking, and in fact, equally
interpretable in the opposing way. He argues, for example, that the
comments of Valerius Maximus and Dio could be taken as indicative of a
heightened Roman awareness both of Jews and of proselytism, rather
than proof of active mission. Likewise, he shows that, before AD 100,
the term proselyte was very rare, and could apply to both Jews and
Gentiles. He argues that the text of Matthew, so often used to support
the claim for Jewish mission, was written at the end of the first century
and in fact reflects the change in meaning that the term proshvluto" was
undergoing at that time, coming eventually to mean solely a Gentile
who has become Jewish. The passage in Matthew is argued by
Goodman to be an admonition of the Pharisees for attempting to
314
315
316

Goodman, Mission, p. 61.


See Goodman, Mission, p. 68.
For example, Josephus on the royal converts of Adiabene Ant. 20.34-48.

147

persuade other Jews to follow Pharisaic halakhah, norms of behaviour,


rather than an example of a general Jewish trend to gain Gentile
proselytes.317
Goodman argues that Jews in the first century AD did not see any
particular value in active proselytising of the Gentiles, and that the
conversion of potential partners before marriage, and the circumcision
of male domestic servants do not signify missionary zeal, but rather, a
desire to reinforce the groups boundary and solidarity, not to open it
up to the outside.318 The fact that there were proselytes should not be
understood as indicating a Jewish desire to win them, and the
acceptance of proselytes into the Jewish religious and social group was
at this stage essentially passive. Although Goodman argues that Jews
before AD 100 may have engaged in apologetic mission to win
sympathisers, he does not see evidence for a proselytising mission to
seek full converts. The range and fluidity of terms found in the
evidence, from proselytes to god-fearers, Judaizers, and sympathisers,
supports this interpretation.

Jewish

reactions

to

Christianity:

changing

attitudes

to

proselytism?
However, Goodman argues that the situation began to change during
the Talmudic period, between c. AD 100-500. In the latter part of this
period, converting to Judaism (especially if the convert had previously
been a Christian) began to be punished by the Roman-Christian state.
Laws against circumcision, endorsing capital punishment for doctors
performing the operation, and the confiscation of the property of
converts to Judaism are found in the Codex Theodosianus. In addition,
the advent of Christian mission may have changed the Jewish attitude
towards proselytism.
317
318

See Goodman, Mission, pp. 69-90.


Goodman, Mission, p. 78.

148

All this might imply that in the middle and later Roman Empire,
Jews were engaged in active mission to both pagans and Christians.
However, it can also equally be interpreted to mean that the impetus
was with the convert and the Jews themselves simply accepted converts
as they always had done.319 Christian Judaizing, i.e. fraternising with
Jews or attending synagogue meetings, was almost endemic 320 for
example, John Chrysostoms congregation in Antioch but Chrysostom
is castigating his Christian flock for going to the synagogue, rather than
accusing the Jews of tempting away his congregation. Some evidence
that might indicate an active Jewish proselytism can be found in the
rabbinic texts, although it is far from unanimous. Goodman suggests
that, after the Temple destruction and certainly by the third century,
and despite the Roman strictures on proselytising, some rabbis were
enthusiastic about Jewish mission and saw it as a duty. 321 The most
plausible reason Goodman sees for this change in attitude is the need to
compete with Christianity, which was proving successful at winning
converts away from both paganism and from Judaism.
This may be the case. However, Judaism also underwent a
massive change independent of Christianity, and the rabbinic period
was marked by the composition of the Mishnah c. AD 200, as a
summary of practice aimed at universalising norms of behaviour,
halakhah. Tessa Rajak states that the process reached its climax,
though in no sense its conclusion, with the massive and heterogeneous
compilation of the sixth century, known to us as the Babylonian
Talmud.322 Rabbinic halakhah encouraged the sharpening of Jewish
identity and the renewed adherence to reiterated Jewish Laws.
My approach here then is to examine the epigraphic evidence for
Diaspora reflections of these developments in Judaea. I argue that in
this period, Jewish identity underwent a fundamental reappraisal, and
319

Albeit with some Rabbinic illiberalism at certain times see Goodman, Mission,
chapters 4-6.
320
Goodman, Mission, p. 143.
321
See Goodman, Mission, pp. 145-153.
322
Rajak, T., The Jewish Community and its Boundaries, in The Jews among Pagans
and Christians, London & New York: Routledge, 1992, p. 12.

149

becomes

deliberately

and

explicitly

visible

to

us

and

to

their

contemporary Gentile neighbours through the increasing use of Jewish


names and the use of Jewish symbols in the monumental evidence from
the Diaspora. We can thus at this stage begin to see how widespread
the communities of the Graeco-Roman Diaspora really were, and can
usefully apply network analysis to them to understand the connectivity
of the ethnic citizenship323 of Judaism; the communication routes along
which the rabbinic ideas might have spread, and how the dispersed
Jewish community might have interrelated between its scattered parts.
Although this is a picture of Judaism in the later Empire, it may have
some usefulness as a model for the connectivity of the earlier Diaspora.
The universally renewed notion of Judaism and of the Jewish nation as a
whole set down in the Mishnah was a response to the destruction of the
Temple and of Judaea. The transmission of the newly reiterated tenets
of halakhah it advocated to the Jews of the Diaspora can be understood
as the activation of a network built on an understanding of shared
ethnicity.

Hypothesis: Utilisation of a Network of Shared Ethnicity

The Jewish Diaspora was spread throughout the Roman world, reaching
Hispania, southern Gallia and Pannonia by the fourth century AD, as
well as the ancient communities beyond the eastern borders of the
Roman Empire in Babylon. However, communities were most densely
located closer to Judaea in the major cities and the eastern provinces of
Egypt, Asia Minor, Greece and Cyprus. The breadth and depth of the
dispersion was due also to the creation of commercial opportunities
resulting from the Roman Empire and the preceding Hellenistic
323

Liebeschuetz, Judaism, p. 249.

150

kingdoms, as well as being a consequence of forced settlements, and


periodical replenishments by refugees and prisoners of war.
The Diaspora was a network of ethnically linked groups of people,
with a universal notion of Jewish Law and the distinctiveness of the
Jewish people and faith, as set out in the Torah. Some of the Laws,
especially related to food restrictions, had been allegorised by the
Hellenistic Diaspora literary elites, and taken as symbolic of the social
and moral separation that maintained Jewish righteousness. 324 Since the
concepts of religion were transmitted in book form, which increased
the potential for standardisation, transportation and access, there was
an intrinsic unity to the Jewish religious community. Because shared
Jewish identity involved most aspects of daily life, ethnic links can be
understood as strong-ties, and could therefore be extremely powerful
for the transmission of social and cultural information and innovation.
The Jewish Diaspora offers then a unique network with which to
examine

the

communicative

power

of

shared

ethnicity

in

the

Hellenistic-Roman world. Defining ethnicity can sometimes be a


complex issue, as ethnicity itself is subject to reconstruction and
redefinition

by

different

people

in

different

environments.

The

epigraphic data generally masks subtleties of this kind however, and so


can only be taken at face value: those who chose to define themselves
as Jewish are understood to be ethnically Jewish. My hypothesis here,
then, is to test the communicative power that ought to be inherent in
such an ethnic network.
I show that the epigraphic evidence, as the record of the lives of
ordinary Jews, the largely static minority populations making up the
Diaspora, undergoes a stark change, beginning in the second-third
centuries AD. In the early Hellenistic-Roman Diaspora, Jews integrated
with Gentile communities, adopted Hellenised names and practices, and
engaged with certain aspects of Graeco-Roman culture to a degree.
Even though the Diaspora was considerable by this period, there is very
324

For example, by the author of the Letter of Aristeas, see Barclay, Jews, pp. 138-150.

151

little evidence for Jewish self-identification, limited to particular sociopolitical contexts: emancipation of slaves and the collective dedications
of prayer-houses in Egypt.
The destruction of the Temple and the fifty years that followed
changed the lives of Jews in the Diaspora dramatically. Following the
cataclysm

in

Judaea,

the

epigraphy

reveals

the

widespread

dissemination and adoption of explicitly Jewish names, symbols and


language. This could be argued to be a uniform response to the rising
tide of intolerant Christianity, but this explanation presupposes a
universal Christian attitude to Jews, which is implausible at this early
date; and, moreover, the process began before Christianity had been
adopted as the state religion of the Roman Empire. Rather, the
Hebraization of the Diaspora should instead be related to the internal
change in Judaism, which occurred after the destruction of the Temple
and of Judaea. By recognising and interpreting the visible symbols of
the new universalised halakhah of the rabbinic reforms, we may analyse
these as a demonstration of how, at this time, the ethnic community of
the Jews created a vulnerable, dynamic network based on strong-tie
familial connections.

Epigraphic Analysis

There are some features of the epigraphic record that can be used to
identify Jewish communities in the Diaspora, although there are certain
caveats that must be borne in mind: the epigraphic habit of the second
century, the epitaphic nature of most of the available epigraphy, and the
fact that Jews who did not feel themselves to be so strongly aligned to
the core community might be less likely to present their Jewish identity
epigraphically. Essentially, these people are lost to the modern observer.
152

Only individuals whose Jewish identity is explicitly revealed by


inscriptions, or who are securely located within an established Jewish
context, are included in the body of material studied here.
The inscriptions are predominantly in Greek, even in Italy, which
suggests that Italian Jews were strongly linked to the Greek-speaking
world, as Greek was the contemporary lingua franca for most of the
Diaspora. The identifying features of Jewish inscriptions include the
mention

of

the

synagogue,

proseuche

the

prayer-house,

or

sometimes the gerousia as a place of gathering and worship. It is


sometimes difficult to be entirely sure about the Jewishness of these
terms, as non-Jews use all three to describe places of meeting.
Likewise, offices within these domains archisynagogos, gerousiarch,
and presbyter are all found in non-Jewish contexts, and so texts that
mention only of one of these offices without further Jewish indicators
are not included. In later examples, Rabbi is quite clearly Jewish, but it
should be noted that the term does not necessarily denote someone
who had formal priestly rank. Other clear indicators are the
occurrence of Hebrew in the inscriptions, the use of Semitic personal
names, and the depiction of Jewish symbols, (the menorah, lulab, etrog
and shofar being the most common). We also sometimes encounter
specific reference to the Laws or the Sabbath.
This analysis relies on the various corpuses of Jewish inscriptions,
Freys Corpus Inscriptionum Judaicum, (CIJ), Amelings, Noy and
Bloedhorns,

and

Noy,

Panayotov

and

Bloedhorns

Inscriptiones

Judaicae Orientis vols. I-III (IJO 1-3), Horbury and Noys Jewish
Inscriptions

of Graeco-Roman Egypt

(JIGRE), and Noys

Jewish

Inscriptions of Western Europe (JIWE), supplemented by the revised


Schrer. Unless otherwise stated, the inscriptions are in Greek, and
these make up more than two thirds of the 900+ items of evidence used
here.

153

The Diaspora before AD 70


The synagogue (or other form of proseuche, prayer/meeting place) had
become the local focus of Diaspora Judaism, and the exposition of Law
and the reading of the sacred texts happened in these spaces. Goodman
argues that it was here, by separating or being separated from their
traditional land, that Jews were required to articulate anew the tenets
of their faith and Law, craft the constituents and boundaries of Judaism,
negotiate flexibilities and define themselves as had not been necessary
in the Judaean homeland.325 This process involved both the stricter
imposition of certain prohibitions and the suppression of divisive
aspects in order to define the place of the Jews within the GraecoRoman environment.
However, the epigraphic evidence from pre-AD 70 shows very
little sign either of relaxation of Jewish regulations or the tightening of
Jewish boundaries. Jewish personal names and the use of the term
Ioudaios or Ebraios as a marker of Jewish ethnicity were the most
important epigraphic indicators, suggesting a commonly understood
and unified notion of being Jewish, and a need to mark it. On the other
hand, the use of non-Jewish personal names indicates a level of
integration within Graeco-Roman society.
A survey of the earlier Diaspora evidence shows that the
occasions where Jews explicitly name themselves as such were very
limited. The use of Ioudaios (and it is Ioudaios rather than Ebraios at
this stage, a point to which we shall return) is rare, and is confined to
inscriptions that refer to slaves or ex-slaves, who indicated themselves
to be Jewish seemingly as an indicator of heritage and place of origin. It
is otherwise only found in Egypt. The statement of their Jewish identity
by Egyptian Jews in the Hellenistic period seems likely to be an
indicator of their status under Greek rule and established claim to
certain social and political privileges.
325

Goodman, M., Jews, Greeks, and Romans in Jews in a Graeco-Roman World, M.


Goodman, (ed.), Oxford: Oxford University Press, 1998.

154

Inscriptions recording the manumission of Jewish slaves are found


quite widely dispersed across the early Diaspora. One of the earliest,
ca. 300-250 BC, was found in Oropus in Boeotia. It was set up in the
Amphiareion, to record the emancipation of Moschos son of Moschion
the Jew. Moschos appears to have received instruction from the gods
Amphiaraus and Hygeia in a dream that he should be set free (IJO 1,
Ach45, also CIJ I, p. 82.) Because Moschos refers to the pagan Greek
deities, and makes his dedication within the sacred precinct of the
Amphiareion following incubation there, this inscription is taken as the
earliest evidence on the Greek mainland for a Hellenized Jew. Although
this is clearly the case, it is also true that Moschos understands himself
still as Jewish.
More explicitly Jewish slaves were being emancipated in Greece
in the mid-second century BC. Inscriptions from Delphi record the
manumission of Ioudaios, and a decade later Antigona, with her
daughters Dorothea and Theodora (IJO 1, Ach42; 43). They are called
to; gevno" jIoudaivon, of the race of the Jews. They may have been

prisoners of war from the Maccabaean period. 326 Ergasion the


Samaritan (IJO 1, Ach41) was a member of a pagan group in Athens in
the fourth or third century BC; it is unclear whether he was stating his
faith and/or ethnicity.
Inscriptions in the west that explicitly define the person as being
Jewish are only found in Aquileia on the Adriatic in the first century BC,
where a freedman, Lucius Aiacius Dama, possibly from Damascus, calls
himself Iudaeus (CIJ I, 643); and at Villamesias in the Spanish
hinterland where a Latin epitaph records Alucius Roscius, 327 a freedman
defining himself as a Jew (JIWE, 188). The inscription is however only
loosely dated between the first and third centuries AD, and he may

326

See Schrer2 Vol. III.i, p. 65, or Williams, M. H., The Jews among the Greeks and
the Romans A Diasporan Sourcebook, Baltimore: The Johns Hopkins University
Press, 1998, p. 5.
327
His name is unusual, and Alucius may be connected with the Celtiberian name,
Allucius.

155

therefore have been enslaved after any of the revolts in Judaea or the
Diaspora.
In Gorgippia on the north shore of the Black Sea Jews by race
are mentioned in a manumission text dated to AD 59-60 (IJO 1, BS23).
A manumission text from Phanagoria dating to AD 52 consecrates three
slaves328 to the proseuche, with the synagogue of the Jews providing
guardianship (IJO 1, BS18). Four further manumission inscriptions
from Panticipaeum of the first-second century mention the permission
or guardianship of the synagogue of the Jews/Jews and god-fearers (IJO
1, BS5; BS6; BS7; BS9). The Jewish synagogue was apparently a locus
for the legal transaction of slaves.
The Jews of Egypt often state their Jewishness during the
Hellenistic period apparently for a distinct political purpose, to be
understood in the context as a marker of separation, probably from the
Egyptians rather than the Greeks. Plentiful papyrus fragments from the
third-second centuries BC use explicitly Jewish names, mention the
proseuche and the term jIoudaivon.329 These are found across lower and
middle Egypt and supported by evidence from inscriptions, for example,
the third century BC proseuche at Schedia was dedicated by the Jews
(JIGRE, 22) like that at Crocodilopolis (JIGRE, 117); the second
century BC Jews of Xenephyris dedicated the gate to their proseuche
(JIGRE, 24); there are also similar dedications by the Jews of Nitriai
(JIGRE, 25) and Athribis (JIGRE, 27). Men with Greek names
(Theodotus son of Dorion, Ptolemy son of Dionysios) who call
themselves the Jew dedicate at the temple of Pan at El-Kanais from the
second-first century BC (JIGRE, 121; 122).330 There is no later
evidence from Egypt for this kind of self-definition, perhaps related to

328

Two of the slaves have unusual, possibly Iranian names, Karsandanos and Karagos,
which might indicate some kind of interaction between Persia and the Black Sea
region.
329
See Schrer2, History, Vol. III.i, pp. 46-57.
330
The editors of Schrer had dated these inscriptions to the second century AD, see
History, Vol. III.i, p. 58, but I here adopt the revised date suggested by Noy in JIGRE.

156

the hostility towards the Jews at certain points during the Roman
period.
Few individuals stated their Jewish heritage or ethnicity in the
period before the fall of the Temple, and when they did, they did so for
different reasons. In Egypt the Jews had the distinct politico-legislative
purpose of distinguishing themselves from the Egyptian populace to
secure privileges from the Hellenistic rulers. In the Black Sea, the
Aegean and the West where singular rather than collective marking of
Jewish identity is found, there is a clear connection with freedom from
enslavement. The general absence of explicit statements about Jewish
ethnicity in the rest of the epigraphic evidence for the Diaspora in the
period pre-AD 70 is therefore all the more striking. The literary
evidence records that Rome had a Diaspora community by the second
century BC, and Josephus records the dispersion of the Jews into Asia
Minor and the west under the Seleukids.331 The only epigraphic
evidence from Asia Minor at this time is a late Hellenistic inscription
from Caunos in Caria that records members of a Samaritan family, of
whom at least the father, Simon, was originally from Shechem, the
Samaritan capital under Mount Gerizim. The rest of the family have
Greek names, including Dionysia and Cleopatra (IJO 2, 24). There is
epigraphic evidence for Jews or Samaritans in Athens and Delos in the
second century BC: Simon, son of Ananias in Athens (IJO 1, Ach33),
and the murdered Jewish women Heraclea and Marthina from Delos
(IJO 1, Ach70; 71). A Jew named Ioudaios manumitted his slave
Amyntas in Delphi, which is a rare example of a Jew manumitting a
slave in a pagan temple (IJO 1, Ach44). It is possible, because
Ioudaios is a personal name also used as an ethnic, that Ioudaios was a
freedman himself. There were Jews in Cyrenaica and on Cyprus during

331

Josephus, Ant. 14. His quotation of a letter at 12.147-53 records the transportation
of 2000 Jewish families to fortresses and strategic places in Phrygia and Lydia, who
were given land and permitted to live by their own laws. The authenticity of the
documents Josephus uses has been questioned, and it may be that this letter was an
apologetic document penned by Jews themselves. Barclay, Jews, pp. 260-262.

157

the Ptolemaic period according to Josephus, 332 and the Cypriot


community is known also from three Phoenician inscriptions from the
fourth century BC. The names are Semitic: Haggai, son of Abdi (IJO 3,
Cyp6); Muttun-Astart, son of Azariah, son of Muttun, son of Shalom, the
chief of the scribes (IJO 3, Cyp7); and Shalom, son of Asaphyahu (IJO
3, Cyp8).
From this assessment of the earlier epigraphic evidence, and the
absence in most places of explicit statements of Jewish identity, it
should be deduced that Jews in the pre-AD 70 Mediterranean Diaspora
responded to the Graeco-Roman world by assimilating to a degree that
meant it was either unnecessary or even perhaps undesirable to
identify oneself explicitly as Jewish in inscriptions. This is supported by
the fact that although Jews sometimes used Biblical names, theophoric
names such as Heraclea at Delos, Dionysia at Caunos and MuttunAstart, gift of Astarte on Cyprus hardly indicate piety towards the
Jewish god. The evidence of Moschos incubation in the Amphiareion or
the dedications in the temple of Pan suggests that using pagan temples
was not necessarily inappropriate or incompatible with possessing
Jewish ethnicity in the Hellenistic period.
The lack of explicit statements of Jewish identity in inscriptions
may be due in part to a strong centralised religious and financial
relationship within the Diaspora to the Jerusalem Temple, implying an
inherent Jewish identity that did not need external expression. The
probable reference to the Day of Atonement on Delos (IJO 1, Ach70)
shows knowledge of and participation in the Jewish festival year. These
connections between Jerusalem and the Diaspora communities are also
visible in the actions of Judaeas rulers. Inscriptions from Delos and
Syros honouring King Herod the Great (IJO 1, Ach38-9; 74) show that
the Diaspora was intimately connected to political events in Judaea; and
an inscription dating from 4-39 AD from Delos (IJO 1, Ach69), found in
the propylon of the temple of Apollo, is an honorific by the Athenians for
332

Josephus, Ant. 13.284 (Cyprus); C. Ap. ii 4 (44) (Cyrenaica).

158

Herod Antipas. It is suggested that he improved the propylon, and


supports the notion that the community on Delos in particular was very
much engaged with the political structures in both Athens and Judaea,
and also that the leaders in Judaea were involved with Diaspora
communities even to the extent of making donations to pagan buildings.
Delos, with its status as a free port and major commercial centre,
emerges as an important Diaspora focal point where Jews clearly
interacted frequently with Greeks and people from across the Greek
world. In the first century BC, four people with Greek names dedicated
to Theos Hypsistos, the Greek translation of the Hebrew El Elyon,
including Zosas from Paros (IJO 1, Ach60), and some of these
inscriptions were found in GD80, the building that has been identified
by some as the synagogue (IJO 1, Ach61; 62; 63). These may either be
Greeks dedicating to the Jewish God, to Theos Hypsistos as a distinct
cult333 or Jews adopting Greek names, but either way, they show the
high level of interaction between Jews and Gentiles during the late
Hellenistic period. The dedication by the man from Paros perhaps even
implies an intra-archipelagian community with Delos at its spiritual and
physical centre. The abundant epigraphy of Delos contrasts with the
poverty of epigraphic evidence in Rome and Asia Minor, whose Jewish
communities are only attested by literary sources. This may simply be a
pocket of epigraphic fashion at Delos, but may also indicate the
relative wealth and self-confidence of its Jewish population.
It seems clear from the lack of ordinary Jews in the Diaspora
explicitly named as such, and the special situations of those who are,
that before the destruction of the Temple, the notion of Jewish culture
and ethnicity in the Diaspora was somehow inherent

and not

prominently advertised. Jews did not need to state their Jewishness,


and

Goodmans

argument

that

simply

being

in

the

Diaspora

necessitated, for ordinary Jews, the renewed articulation of the tenets


of Jewish Law and the reinforcement of boundaries is certainly not
333

This will be discussed further in Chapter 6.

159

epigraphically clear. Jews were given pagan names, Jewish rulers


donated to pagan buildings, and Gentiles were interested in Jewish cult.
Because Jerusalem was the centre in certain absolute and specific
religious and fiscal terms, Judaism before AD 70, with a book and a
singular Temple at its heart, was understood by Jews to be fully formed.
Jews engaged with and responded to the circumstances of their life in
the Diaspora without losing this sense of attachment to the Jerusalem
temple, which was at the centre of their religion. However, when the
emotional and religious heart of Judaism, enshrined in the Temple, was
destroyed, how did the Diaspora react?

The destruction of the Temple


The Jerusalem Temple in the Hellenistic-Roman period was a solid
financial and religious focus for the Diaspora: both through the annual
didrachma Temple tax levied on all men over 20, in accordance with the
command of Dt. 12:26 that the sacred donations that are due from you,
and your votive gifts, you shall bring to the place that the LORD will
choose, and the substantial numbers of pilgrims.334 The generation of
normative and accessible Jewish writings in Greek335 may have aided
this notion of centre. Paying the Temple tax would certainly have been
an important part of every ordinary male Jews life, who was thereby
required to undertake long-distance travel to the central administrative
and religious point. In Jerusalem, there would have been opportunity
for discussion of Law and scripture as well as for interaction with other
members of Diasporan and Palestinian Jewish communities. One piece
of epigraphic material provides proof of the Diaspora relationship with
the Temple, the donation of flooring by a man from Rhodes (IJO 2, 10)
334

As recorded by Philo, De Spec. Leg. i 12 (69); and Josephus reckoning of the


numbers attending the festivals in Jerusalem at 2,700,000, B. J. vi 9, 3 (425). See
Schrer2, History, vol. III.i, pp. 148-149.
335
Including the LXX, Sibylline Oracles, 2 Maccabees, and the letter of Aristeas, see
Collins, Jerusalem, p. 61-86.

160

from the rubble of a Herodian palace south of the Temple, also


destroyed in AD 70.
Goodman claims that for the Romans, the destruction of the
second Temple in 70 AD was a act of political machismo and posturing
by Titus following the Jewish revolt, essentially an insignificant event in
a comparatively minor provincial backwater. 336 For the Jews in Judaea,
however, the destruction of the Temple marked a major change in the
way Judaism was practised and how it thought about itself. The
combination of the decimation, subjugation and poverty of the people of
Judaea, the renaming of the province Syria Palaestina, and the
destruction of the Temple itself had enormous psychological, spiritual
and financial consequences, kick-starting the period of the Bar Kokhba
revolt and the rabbinic reforms, that led to the composition of the
Mishnah c. AD 200. What about the situation in the Diaspora? The
editors of Schrer argue that given the Temple tax and the pilgrimages
recorded by Philo and Josephus, the destruction must have been of
profound import. The subsequent revolts in Cyprus, Egypt and
Cyrenaica are manifest reactions to the destruction. The destruction is
lamented in Diaspora works of literature, such as the Fourth and Fifth
Sibylline Oracles.337 Josephus, writing Contra Apionem probably in the
period following the assassination of Domitian, located the essence of
Judaism in the rites of the Temple. 338 What then for Judaism when the
rites of the Temple were no more? The only centre left to the people
was the Torah.339
Aside from the psychological import, the most immediate changes
for the majority of Diaspora Jews would have been the transformation of
the Temple tax into the fiscus Judaicus, now payable to the Romans,
and, in certain places, the influx of Judaean refugees. Prisoners of war
swelled numbers in the Diaspora communities in Italy, with refugees
from the conflict probably fleeing to places geographically proximate to
336
337
338
339

Goodman, Mission, p. 43.


Collins, Jerusalem, p. 152.
See Goodman, Mission, p. 45.
Schrer2, History, Vol. I, p. 513.

161

Judaea. One such prisoner of war in Italy was Claudia Aster, an Imperial
freedmans prisoner-slave from Jerusalem recorded in a Latin epitaph
from between 70-95 AD in Naples (CIJ I, 556). In Athens, a few first
century AD epitaphs may represent refugees from Judaea: Matthaia
from Aradus, married to a man from Sidon (CIJ I, 715f); Ammia of
Jerusalem (IJO 1, Ach26); Ammia, daughter of Philo, a Samaritan, who
married a man from Antioch (IJO 1, Ach35); and Theodora, daughter of
Themison, a Samaritan (IJO 1, Ach36). Likewise, a dedication from the
Serapeion on Delos by Praulus of Samaria, dating from 100 AD (IJO 1,
Ach68); the first century epitaph of Justus from Tiberias in Taenarum
on the southern tip of the Peloponnese (IJO 1, Ach55); and a first
century AD list from Rhodes that includes a man from Jerusalem (IJO 2,
9), may all represent first or second generation refugees.
However, the longer-term effects of the destruction of the Temple
on the Diaspora communities were cognitive, seen in the revolts that
took place in the following fifty years. Judaea as a centralising force
must have lost considerable power, and with the most important centre
of gravity of the Diaspora network removed, the opportunity for travel
to and the exchange of information and ideas with the perceived
homeland was gone. How did the Jews both in Judaea and the Diaspora
deal with the challenges to their faith and their status as the chosen
people? It has been argued that, like the Bar Kokhba revolt, the various
violent revolts in the Diaspora over the next fifty years, in Cyrenaica,
Cyprus and Egypt, marked the powerful messianic expectations 340 of
the dispersed Jewish nation following the destruction of the Temple.
However, the revolts were all quashed, and instead, the most important
reaction to the destruction of the centre is to be seen in the rise of
rabbinic Judaism, which promoted a greater focus on the texts of the
Torah and stricter adherence to the Laws governing norms of behaviour
340

Schrer2, History, p. 149, quoting the argument of M. Hengel, Messianische


Hoffnung und politischer Radikalismus in der jdisch-hellenistischen Diaspora.
Zur Frage der Voraussetzungen des jdischen Aufstandes unter Trajan 115-117
n.Chr., D. Hellholm (ed.), Apocalypticism in the Mediterranean world and the Near
East, 1979, pp. 655-86.

162

halakhah: it was precisely the annihilation of Israels political


existence which led to the triumph of rabbinic Judaism.341

The transmission of rabbinic Judaism in the Diaspora

The epigraphic evidence for the Diaspora shows a massive increase in


explicit statements of Jewish identity from the second century onwards.
I argue that this should be interpreted as evidence that the new
religious authorities in Palestine used the tightening strong-tie familial
connections of the ethnic network of the anxious, vulnerable Diaspora
to transmit the religious and social discipline of rabbinic Judaism. The
reforms of rabbinic Judaism arose in Judaea, beginning in the years
following the destruction of the Temple and the Bar Kokhba revolt,
emphasising reading and interpreting the Torah and standardising
norms of behaviour.342 This re-construction of Judaism following the
cataclysms and the new universalised halakhah are clearly manifest in
the records of the ordinary people of the Diaspora. The indicators found
on Jewish monuments that reflect an increased awareness of a common
Jewish practice, history and behaviour include specifically Jewish
symbols, i.e. the menorah, lulab, etrog, and shofar, as referents to a
universalised ritual and the religious calendar,343 and the use of Hebrew
as a marker of education and a revived knowledge of the sacred texts,
the Torah, Jewish Law, and Jewish history. Additionally the increasing
use of specifically Jewish name forms also provides a subtle indication
of the universal engendering of a more strongly defined Jewishness,
341

Schrer2, History, Vol. I, p. 555.


Rajak, The Jewish Community, p. 11-12.
343
The menorah is the seven-branched lamp that had stood in the Temple before its
destruction by Titus and the transportation and parade of Jewish cult items in Rome.
The lulab (palm branch) and etrog (citrus) are associated with the Feast of
Tabernacles (Sukkot); the shofar is the trumpet used to sound the new moon and the
beginnings of certain feasts.
342

163

matched by the trend during the third-fourth centuries AD for


individuals to defines themselves as Jews or, more specifically,
Hebrews.344
It has been argued by Williams that the Hebraization of names in
the Diaspora during the fourth-fifth centuries was a reaction to
Christianitys appropriation of Biblical names and the impositions of
increasingly intolerant attitudes of the Christian emperors towards
people of other religions.345 It might be argued therefore that increase
in these other indicators of Jewishness might also be a reaction to
Christianity. However, because these changes begin to be enacted
across the Diaspora before Christianity was the state religion, I instead
suggest that this represents the internal transformation of Judaism,
which was a reaction to the devastation of Judaea and the destruction
of the centre of the religion. That the Jews of the Diaspora felt a strong
emotional connection with Judaea is evident from two fourth century or
later inscriptions from Acmonia in Phrygia IJO 2, 169, a prayer for all
the fatherland, and IJO 2, 170, a prayer for peace in Israel and
Jerusalem.

Marking ethnicity and education through piety toward the Laws


References to events in Jewish history, festivals, and the patriarchs
indicate knowledge of and participation in a Jewish world beyond the
local environment. However, there are only a very few instances in the
epigraphy: the second century BC inscription from Rheneia (IJO 1,
Ach70), that may make reference to the Day of Atonement, Yom
Kippur; the fourth century AD epitaph from Catania which mentions the
344

This change in terminology has been linked to the destruction of Judaea as a


province, but may be better explained by the renewed emphasis on Jewish history.
345
Williams, M. H., Semitic Name-Use by Jews in Roman Asia Minor and the Dating of
the Aphrodisias Stele Inscriptions, in E. Matthews, (ed.), Old and New Worlds in
Greek Onomastics, Proceedings of the British Academy, 148, Oxford: Oxford
University Press, 2007, p. 192.

164

patriarchs (CIJ I, 650, see below); or the text from Argos that
specifically mentions the patriarchs, ethnarchs and the honour of the
Sages (CIJ I, 719). It must be assumed that although these were well
known Jewish institutions, whose history was explained in the
synagogue or proseuche, they were inappropriate for the grave.
The Jewish Law346 is mentioned more often, most frequently in the
west: Rome, Italy and Sicily. Across the rest of the Diaspora, the Law is
mentioned on one inscription each from Macedonia, Greece, Asia Minor
and Egypt. Ten inscriptions from Rome mention the Laws. From the Via
Appia cemetery (dated between the second and fourth centuries) there
are the epitaphs of Eukarpos, philonomos, which is decorated with a
menorah (CIJ I, 111); Eusebius, nomomathes, (CIJ I, 113); Krispina, a
philentolos (CIJ I, 132); an unnamed nomomathes, whose grave is
decorated with a menorah and a lion (CIJ I, 193); an unnamed
nomodidaskolos, (CIJ I, 201) which also uses the formula from
Proverbs 10:7, the memory of the righteous is a blessing, discussed
below; and a philolaos, philentolos and philopenes (CIJ I, 203). The use
of laos in this context generally refers to the Jewish people. An
inscription that may refer to the Law is a Latin epitaph in Greek letters
for a woman who is described as being a good disciple in (CIJ I, 215),
an indication that the Jewish population was speaking Latin but that
Greek was still the written language. From the Via Portuensis come the
epitaphs of Eusebis, a didaskalos nomomathes (CIJ I, 333); and a
plaque in Latin dated to the beginning of the second century for Regina,
obedient to the Law (CIJ I, 476). A Latin and Greek epitaph records
Victorina who died in 330, dikaia, osia, philentolos; her epitaph is
marked with a menorah, etrog and shofar (CIJ I, 482). A sarcophagus
from the Via Nomentana is inscribed in Latin to Julia Irene Arista, who
was filled with the virtue of God and the faith of the chosen people,
who observed the Laws exactly (CIJ I, 72). Quite a number of epitaphs
from the Roman cemeteries end with the formula sleeps in peace with
346

Mosaic Law, meaning food restrictions, circumcision, Sabbath observance and


prohibition of intermarriage.

165

the righteous, and righteous God is also found occasionally across the
rest of the Diaspora. Righteousness in this context means pious
adherence to the Laws, and also suggests that the Jewish belief that
those who had lived a righteous life became angels was fairly
widespread.
The Law is mentioned elsewhere in Italy. A Greek metrical
epitaph from Lorium on the coast near Rome remembers Rufinus, godfearer, learned in the holy Laws and wisdom (JIWE, 12).347 The
mention of the Laws in Jewish inscriptions is generally a phenomenon
of late antiquity and therefore it is my suggestion that this inscription
should be similarly dated. A third century synagogue donation from
Ostia records the building of the synagogue and ark for the holy law
by Mindius Faustus (JIWE, 13). A Hebrew and Latin epitaph from
Catania for Aurelius Samohil and his wife Lasiferina (JIWE, 145)
requests respect for the patriarchs and for the Jewish Laws. It dates
from AD 383, and Fergus Millar 348 uses it as a clear example of an
unambiguously Jewish person combining the secular and Jewish dating
systems. He questions who is meant by the patriarcha: the biblical
patriarchs, local Jewish officials, or contemporary Jewish patriarchs in
Palestine. The last interpretation is settled on, and would imply a
relatively integrated and homogenous Jewish world, in which all or
most of the communities in the provinces of the Roman Empire
observed a Judaism closely resembling that rabbinic Judaism of
Palestine, which was just giving birth to the Palestinian or Jerusalem
347

This inscription has provoked considerable controversy, with arguments proposed


for its Jewish, Christian or pagan character. Rufinus is explicitly called qewsebhv", i.e.
a Gentile devoted to the Jewish God, but the reference to the ajgivwn te novmwn has
been argued by many to be an indication that he was a full Jew, for example, Palmer
Bonz, M., The Jewish Donor Inscriptions from Aphrodisias: are they both third
century, and who are the Theosebeis?, in HSCP, 96., p. 291-292. Levinskaya also
thinks Rufinus is likely to be Jewish: Levinskaya, I., The Book of Acts in its Diaspora
Setting, vol. 5, Grand Rapids, Mich: W.B. Eerdmans, 1996, p. 67. The god-fearers and
the case of Rufinus will be discussed in Chapter 6; here, it is simply notable that
Gentiles could still be considered as pious as Jews. The reforms of the rabbis did not
exclude Gentiles or deny their ability to know God and Jewish Law. They simply
clarified their status. This inscription shows that Judaism was still open to Gentiles,
and marks Rufinus participation in and knowledge of a wider Jewish world.
348
Millar, F., Jews of the Graeco-Roman Diaspora, in The Jews among the Pagans and
Christians, Lieu, J., North, J., & Rajak T., (eds.), London: Routledge, 1992.

166

Talmud or which the Talmud presents to us as the Judaism of late


Roman Palestine.349 The use of Latin suggests that in Sicily by this
point some (possibly more highly educated) Jews were using Latin as
their first language; however, another two epitaphs from Catania dating
to the fourth-fifth century are in Greek, for presbyters, Irenaeus and
Jason, who did not offend the Law (JIWE, 148; 149). Across the water
in Rabato on Malta, a graffito from the fourth-fifth century catacombs
records Eulogia, a presbyter and gerousiarch, as a lover of the Law
(JIWE, 163).
Further east, allusions to the Laws are much less common. In
Stobi, a column dating from the second-third century documents the
building of various parts of the synagogue by the Roman citizen
Tiberius Polycharmos, the father of the synagogue, and follower of all
the Jewish prescriptions (CIJ I, 694). A third-fourth century epitaph
from Argos records Aurelius Ioses as invoking the power of the Law
(IJO 1, Ach51). This inscription also mentions the patriarchs (here
thought, as in Catania, to refer to the contemporary office in Palestine),
ethnarchs and the honour of the sages, making a strong link back to
Judaea and the Palestinian Judaism heritage, supporting Millars
suggestion about the Sicilian inscription. The only inscription that
refers to Jewish Law in Asia Minor is the setting up, at some point after
AD 212, of a heroon in Apamea in Phrygia (IJO 2, 179) by Aurelius
Rufus and his wife Aurelia Tatiane. It invokes knowledge of the Laws of
the Jews in a curse formula to prevent the grave being reused by
anyone else. Finally, a late Roman Greek and Hebrew epitaph from the
Roman tower in Alexandria, records Lady Roua (possibly Ruth),
daughter of Borouch, as being Law-abiding (JIGRE, 15).
The references to the Law largely date from the fourth-fifth
century, which supports Millars suggestion that in this period the
communities of the Diaspora were following a common Judaism
directed by Palestine. It is also quite likely that these people represent
349

Millar, Jews, p. 98.

167

the better-educated end of the spectrum of Jews in the Diaspora who


may have been involved in the teaching or exposition of Laws. This is
supported by the use of the formula from Proverbs and the terms
nomomathes and nomodidaskalos on texts from Rome. It is also evident
that reference to the Law is much more common in the Latin West than
in the East. A reason for this could be that, as the centre of the Empire
and because of the large and well-established Jewish community there,
the dialogue between the Jewish leaders in Rome and the authorities in
Judaea was more regular. This perhaps then resulted in a more involved
exegesis and deeper knowledge of Torah in the western communities
than in those in the east, and as a consequence, the communications
between Rome and the other western Diaspora communities involved a
greater level of prescription, manifest in reference to the Laws. It may
also be that the eastern Diaspora communities were older than those in
the West, mostly dating in their origins before the fall of the Temple,
and preserved their own traditions more effectively. It is notable that
Hebrew makes more progress as a sacred script in the West than in the
East, where Greek always remained heavily predominant.

The

activation

of

ethnicity

through

symbol,

language

&

onomastics
However, inscriptions that mention Jewish history or contemporary
Palestinian patriarchs, or that imply knowledge of Jewish Law represent
only a small percentage of the Jewish epigraphic corpus. During the
third-fourth centuries, what becomes widespread across the Empire is
the use of Jewish symbols, names and the formulaic use of Hebrew. This
is partly to do with the epitaphic context of most Jewish inscriptions,
where it was easier to use a symbolic shorthand to denote religious or
ethnic affiliation. But such symbols had markedly not been used in the
earlier period. The sudden and dramatic increase in the use of these
168

explicit identifiers should be connected with the universalising reforms


of the Palestinian rabbis, and therefore with an increased knowledge of
standardised Torah and the ritual calendar, festivals, and the use of the
Hebrew

language.

Assessing

the

distribution

patterns

of

these

epigraphic innovations, which were markers of pan-Judaic culture, will


therefore reveal places in the Diaspora that had early contact with the
Palestinian reforms, and illuminate something about the routes and
speed of information transmission across the ethnic network.
During the first century AD, Hebrew was only very occasionally
used in the Diaspora. It is used for a graffito from Pompeii (CIJ I,
562),350 and possibly on an inscription from Tyre (IJO 3, Syr9). The real
revival of the language only begins in the second-third centuries,
coinciding with the introduction of the menorah as a universally
recognised symbol. During the second-third centuries the menorah
occurs only in Greece and Egypt, whereas Hebrew is found only in
Egypt and places proximate to Judaea. In Egypt the language and the
symbols occasionally appear together; for example, during the second
century, a Hebrew inscription for the son of Lazarus from the Christian
cemetery at Antinoopolis is decorated with a menorah and lulab
(JIGRE, 119). There is however little other evidence for Jews here. It
seems that this mans tomb was specifically differentiated as Jewish,
although it was acceptable or necessary for him to be buried in a
Christian cemetery. Likewise at Al-Minya, the Hebrew inscription for
Judan is marked with three menorahs, two shofars, and an etrog and
lulab (JIGRE, 118). It is particularly noteworthy that these Hebrew
Egyptian inscriptions that use the menorah also use specifically Jewish
names, suggesting that these aspects were closely linked. During the
second-third centuries all three occurrences of the menorah on Greek
inscriptions in Greece also feature Hebrew or Semitic names. In Athens
the menorah appears with a lulab and etrog on the grave of Beniames
350

Another graffito from Pompeii in Latin reads Sodom and Gomorrah, which
necessitates knowledge of Genesis and also suggests a fire-and-brimstone style of
preaching, as it is thought to relate to the destruction of the city by earthquake before
the eruption (CIJ I, 567).

169

(Benjamin), son of Lachares, probably an Attic name (IJO 1, Ach27).


He was a provscolo", which may indicate that he belonged to a
specifically Jewish school in Athens. The menorah also appears on the
epitaph for Lady Panto, daughter of Maronius (a Semitic, if not biblical
form) from Arcadia (IJO 1, Ach52), and on an epitaph from Plataea for
Issachar, son of Heraclides (IJO 1, Ach46). Benjamin and Issachar are
the sons of Greek-named fathers, which supports the notion that the
use of the menorah and the use of Hebrew names were part of the same
scheme of re-Judaization.
However the use of Hebrew is mainly found at this stage in places
close to Judaea or well-established Jewish centres. Cohen from Berytus
and Joseph from Phaine in Syria were buried in Beth Shearim, Josephs
epitaph using the Hebrew shalom (IJO 3, Syr39; 25). The longest
Biblical Hebrew inscription from antiquity is the text of Dt. 6:4-9,
carved on lintels in Palmyra, and probably dates from the third century
(IJO 3, Syr44-47). Further afield, at Panticipaeum, part of the Hebrew
inscription wishes peace on the deceased young man, who is named in
Greek as Isaac (IJO 1, BS13). Hebrew is also used in a synagogue
dedication at Alexandria (JIGRE 17). Again it is clear that the use of
Hebrew is connected with the use of Hebraizing names.
The use of the menorah increases massively during the third
century. It is found at Panticipaeum (IJO 1, BS10-12; 15), where some
of these inscriptions also use Hebraizing names, Seimon (15) and
Samouelos (12); the rest use Greek names or there is no name attested.
In Asia Minor, the menorah, lulab, shofar, and etrog are found on the
epitaph of Aurelius Alexandros at Claudiopolis (IJO 2, 152) and on the
grave of Aurelius Ethelasios and Aurelia Thamar, the son of a
Macedonian, who was an ajnagnwvsto", a reader at Nicomedia (IJO 2,
156). The latter text makes an explicit link between the reading of
scripture and the use of Jewish symbols and names, as well as pointing
to movement between northwest Anatolia and Macedonia. At Dorylaion
the menorah appears on the grave of Esauos, Menothemis, and
170

Aponymios(?) (IJO 2, 185), again the symbol appears alongside a


Hebrew name, and the unusual Greek name Menothemis may be an
indirect allusion to Jewish Law. At Hierapolis in Phrygia a sarcophagus
simply marked Ioudeon uses the menorah, it is thought it was perhaps
intended to mark the collective property of the Jews (IJO 2, 187); it is
also found on a sarcophagus for Marcus Aurelius Gaius Theodorianos
and his wife Aurelia Zenonis and their children, Sanbathios and Zenon.
Sanbathios is named again in the last section as Aurelius Sanbathios,
Ioudaion, presumably marking his Roman citizenship granted after AD
212 (IJO 2, 200). At Corycus the menorah is used on the grave of
Aurelius Eusanbatios, who was a citizen and a councillor of Corycus,
and his wife Matrona (IJO 2, 236). It goes on to read, do not despair,
no-one is immortal, save he who is one, who brought us here into the
sphere of the planets. The first part of this phrase is commonly found
in Jewish epitaphs, and the second reflects the belief that the souls of
the righteous dead would become stars, the angels of God. There is
clearly here a concordance between the Jewish symbol, the name
invoking the Sabbath, and the knowledge of a Jewish religious text.
Elsewhere in the east during the third century, the menorah is found on
Cos on the grave of Eutychos (IJO 2, 7); on a stamp or seal from
Morphou on Cyprus, with a lulab and etrog, which reads I receive fair
hopes (IJO 3, Cyp2); at Dura Europus in an Aramaic graffito from the
later synagogue (IJO 3, Syr89); and at Beth Shearim, in a Greek and
Hebrew epitaph recording Daniel son of Adda from Tyre (IJO 3, Syr7).
On the Balkan Peninsula, the menorah is found on three third
century inscriptions from Phthiotic Thebes (IJO 1, Ach18; 20; 21). One
is a memorial for Peristeria, archegissa; she may be the wife of an
archegos or an archegoness in her own right; another is for Theodotus
and Leontia, (Theodotus was a translation of the Hebrew Jonathan); and
the last for Paregorius and Eutychia. Paregorius was a name widely
used by Jews and corresponds to the Hebrew Menachem, consoler. A
further epitaph from Thebes for Saoul and Anna is possibly a little later,
171

dating from the third-fourth century (IJO 1, Ach17). At Philippopolis,


mosaic flooring in the synagogue was donated by Cosmanius, also
called Joseph, and decorated with a menorah, lulab, and etrog (IJO 1,
Thr1); at Solva in Pannonia a Latin inscription in Greek letters records
Judah and Cassia, and is decorated with a menorah (CIJ I, 676); the
menorah is found in a tomb in an otherwise pagan necropolis in Doclea
in Dalmatia (Duklju, near Podgorica-Titograd, IJO 1, p. 20, Dalmatia
introduction, fn 2); and at Hydrantum on a Greek and Hebrew epitaph
(CIJ I, 632), where the Hebrew reads sleep in peace with the
righteous. From this evidence, it is once again clear that the use of
Jewish symbols and names were linked. However, the use of Hebrew in
the third century was not adopted so readily, perhaps due to the strong
local tradition of Hellenized Jewish communities, which continued to
use Greek as the lingua franca.
The Hebraization of the Roman Jewish community is testified to
by the 115 or so examples of the use of Jewish symbols from Rome,
most dated to the second-fourth centuries. The menorah is the
commonest, and most often appears alone, but also variously with the
lulab, shofar, etrog and aron (shrine for the Torah scrolls). These are
also sometimes found alone. Notably only 26 of these 115 inscriptions
also feature Hebrew, mention Jews or Hebrews, or use explicitly Jewish
names. The absence of Hebrew suggests that the majority of the Roman
community was fairly poorly educated, but perhaps the absence of
names or explicit mention of Jews or Hebrews suggests that the
community was more closely integrated into their Latin environment,
that in the context of a Jewish cemetery it was unnecessary to state
ones Jewishness, or even that it was imprudent, in this context at
period, to draw attention to ones Jewishness through the use of
Hebrew names. The formula from Proverbs 10:7, the memory of the
righteous is a blessing, or versions of it, are found in the Via Portuensis
on the grave of Makedonis, son of Alexander, from Caesarea in
Palestine (CIJ I, 370); on the Via Appia on the grave of a gerousiarch,
172

Theophilos (CIJ I, 119); on the grave of the nomodidaskalos mentioned


above (CIJ I, 201); and on that of Hilaros, the archon of the synagogue
of the Volumnenses (CIJ I, 343). These men are all connected with the
leadership of the synagogue or else are explicitly foreign, suggesting
that knowledge of the text was perhaps confined to these more
educated people, or that the use of these formulas was particularly
appropriate for community leaders.
In the fourth century, the Hebraization process becomes really
apparent across the Empire. In the west beyond Rome, the communities
of the Diaspora eagerly take up the trend; a reason for this might be
that, aside from the inland centre at Venosa, they are all coastal
communities that may have been exposed to direct influence from
Judaea. This is particularly the situation represented by a Greek and
Hebrew epitaph from Brusciano with a menorah, lulab and shofar, for
Rabbi Abba Maris (JIWE, 22), whose Aramaic name and title suggests
that he was perhaps an imported religious expert. The menorah is
found with the lulab and shofar at Naples in a Latin and Hebrew
epitaph for Flaes the Hebrew (JIWE, 37). On Sicily the menorah
appears in the catacombs at Syracuse (CIJ I, 652); at Acrilla on the
tomb of a child, Jason (JIWE, 155); and on three epitaphs from Catania,
once with an etrog, and once on a Latin and Hebrew epitaph for Samuel
and Lasiferina, discussed by Millar, above (respectively JIWE, 146;
150; CIJ I, 650). The menorah, lulab and shofar are found on five
Latin/Latin and Hebrew inscriptions from Sulcis and Porto Torres in
Sardinia (CIJ I, 657; 658; JIWE, 173; 175; 176), the name Juda is
used twice. A menorah is also found on an epitaph from Rabato on
Malta for Dionysia, also called Irene (JIWE, 166), and in North Africa
at Carthage, where the menorah is frequently represented in the
extensive Jewish necropolis, and also at Sullecthum and Thagura, Oea
in Tripolitania where the epitaph of a female presbyter is decorated

173

with a menorah, on a Hebrew inscription from Thaenae and in Numidia


at Henchir Fouara, near Tebessa (CIL VIII, 16701).351
Across the Adriatic in the fourth century, the menorah is found as
decoration or as a Jewish marker of portable objects, on a lamp from
Narona (IJO 1, p. 20, Dalmatia introduction, fn 1) and a bronze seal of
Eustathios from Stobi (IJO 1, Mac2). Two menorahs are found on the
Latin-in-Greek epitaph of Anastasios and Decusana and their son
Benjamin at Aquincum (CIJ I, 675), and aside from the use of the
Biblical name, it is also attached to a Heis Theos inscription, of which
there are very few in the Jewish corpus: found otherwise only in Syria
and in one example of Samaritan Hebrew from Thessalonica dating
from the fourth-sixth century (IJO 1, Mac17), which uses the blessing
text of Numbers 6:22-27.352 This feature and the use of Greek to express
Latin make this inscription unusual, as is the find spot: it is the only
known Jewish inscription from the important Roman military camp of
Aquincum. That both inscriptions decorated with a menorah found in
Pannonia use Greek letters to express Latin (see above epitaph from
nearby Solva) perhaps reveals something about the transmission of the
symbol and the people involved: it seems that Latin was their preferred
language, but Greek was deemed to be the proper language of Jewish
epitaphs.
In the east, Hebraization spreads profoundly during the fourth
century, manifested predominantly in the use of the menorah. In Asia
Minor the menorah is found at Chalcedon (IJO 2, 150; 151); both are
epitaphs for Jewish-named sons of Greek-named presbyters, Jacob, son
of Leontios, and Sanbatis, son of Gerontios. Sanbatis tablet also
features a lulab, etrog and shofar. The menorah appears in a synagogue
at Priene;353 and as a graffito in the theatre at Aphrodisias (IJO 2, 17);
but remarkably, only twice in the great synagogue at Sardis (IJO 2, 55;
351

Schrer2, History, III.i, pp. 62-64.


Amulets also often use Biblical texts, but they are not included here because
although they represent the wider dissemination of Jewish culture and religious ideas,
they are portable and not specifically Jewish.
353
Rutgers, Heritage, p. 102.
352

174

135), where one is a fragment of an actual marble menorah, the other


is on a stamp-amulet. It might seem extraordinary that such a wealthy
and well-known building during this period should provide only these
two examples of the symbol; however, this is easily explained because it
is the synagogue building itself that is known from Sardis, rather than
the epitaphs that generally comprise the majority of the rest of the
Jewish corpus. The menorah was a simple way of marking Jewish
identity in death. As an actual central feature of community worship,
the menorah itself was present as a feature of the synagogue and part
of the ritual itself and did not need to be depicted. The Hebraization of
the community at Sardis is instead seen in the fact that Hebrew is used
a number of times here, (IJO 2, 56; 105-109), one records Shemarayah
son of Elijah (56), another Yohanan (107), as indeclinable Hebrew
names. Further Hebraizing names are recorded in the synagogue
dedications, as discussed by Williams.354
Elsewhere in Asia Minor the menorah appears at Ephesus on a
piece of marble barrier, associated by inscription with the altar (IJO 2,
31); and at Philadelphia on the epitaph of Joseph, and on part of a
funerary table decorated also with ivy, lulab, and etrog, for Hesychios
and Judas (IJO 2, 50; 51). The name Hesychios in this context is
thought to translate Noah. In Hierapolis in Phrygia the menorah
continues to be used and is found alongside a lulab and shofar on the
grave of Tryphon, Ioudeos (IJO 2, 194). At Tavium the menorah is used
on the grave of Sarah (IJO 2, 166), and in Sebastopolis, all three Jewish
inscriptions have the menorah (IJO 2, 159-161). One is in Greek and
Hebrew, for Lampetis, archon (160), one for Sarah, presbyter (161),
and the last is for a child, Despoina, and is also decorated with lulab
and shofar. At Seleucia ad Calycadnus two menorahs appear on the
grave of Theodorus (IJO 2, 245), and at Corycus the motif of paired
menorahs features on three epitaphs (IJO 2, 232; 237; 239). One is for
Abas, son of Symon, most pious under the priests (232); there is also
354

Williams, Semitic, pp. 173-197.

175

Eusambatios, Ioudeos, a perfumer (237) and Julius, another perfumer


(239). This coincidence of Jewish names, Jewish self-definition and the
use of the menorah is echoed by other contemporary epitaphs from
Corycus Samoes (IJO 2, 241), Samuel (IJO 2, 242), and Moses the
Hebrew (IJO 2, 240). Rabbi Samuel, archisynagogos of Phrygia, is
known in a Greek and Hebrew epitaph from Docimeion (IJO 2, 184).
Curses from Deuteronomy are used on epitaphs to deter thieves in
Acmonia, dating from 248-9 (IJO 2, 173; 174), and in Laodicea ad
Lycus in the second-third century (IJO 2, 213). An epitaph from
Laodicea Catececaumene in Lycaonia ends by mentioning the wrath of
the undying God, which, although is not necessarily directly Biblical, is
clearly related to the descriptions of the Jewish God in the Pentateuch
(IJO 2, 227).355
Elsewhere in the east during the fourth century, the menorah is
found again at Panticipaeum on epitaphs for men with Biblical names
Samuel and Symeon (IJO 1, BS14; 16). The only known inscription
from Chersonesos is a Hebrew graffito by Hananiah, dated to the
fourth-fifth century, probably from the synagogue there (IJO 1, BS2). In
AD 385-6 Sambathion the archon was remembered in Hebrew at Byblos
(IJO 3, Syr30). Hebrew occurs on the epitaph of Anna at Corinth (IJO
1, Ach49), where the synagogue of the Hebrews is attested by the third
century (CIJ 1, 718) and on Syros in the Aegean the typical Jewish
symbols were carved onto the rocks on the shore with inscriptions
invoking the Lords help for a safe voyage (IJO 1, Ach72; 73):
Eunomios, whose name recalls the Jewish Law, asks the Lords help for
himself and his Naxian crew (72), and Heortylis the Jew dedicates in
the name of the living God for his safe return (73). These distinguished
the Jewish prayers from sixteen similar texts put up by Christian crews.
Rabbi Atticus is known from a column from Lapethos on Cyprus, (IJO 3,
355

For the influence of Jews on Christianity at Laodicea Catacecaumene, see Mitchell,


S., The Cult of Theos Hypsistos between Pagans, Jews, and Christians, in P.
Athanassiadi and M. Frede, (eds.) Pagan Monotheism in Late Antiquity, Oxford:
Clarendon Press, 1999, p. 123.

176

Cyp1). In Macedonia and Thessaly the menorah, lulab, etrog and shofar
are again found at Philippopolis on the mosaic flooring of the
synagogue (IJO 1, Thr2), this time dedicated by a man called Isaac; at
Larissa a menorah appears on an epitaph for Alexander, prostates and
scholasticus (IJO 1, Ach5). He is clearly a local patron of some kind,
and probably also involved in the exegesis of the texts. It also occurs
twice at Beroea (IJO 1, Mac8; 11), once on the grave of the three-yearold Theodosius, described as Ebreos. The epitaph of Sophia of Gortyn
reads mnhvmh dikeva" ij" ew`na, which Noy and Bloedhorn suggest is
derived from the phrase in Proverbs 10:7, the memory of the righteous
is a blessing (IJO 1, Cre3). Sophia was also an archisynagogissa,
suggesting that she was educated and of some standing within the
synagogue community. It is notable that in the fourth century, the use
of Hebrew is still quite rare.
During the fourth-fifth centuries, the use of Jewish names,
symbols and Hebrew proliferate more widely. At Bizye in Thrace, the
epitaph of the presbyter Rebecca is decorated with menorah and etrog
(IJO 1, Thr3); at Almyros, that of Juda and Asteria (Esther?) has a
menorah (IJO 1, Ach24); and menorahs are found on two epitaphs
from Athens (CIJ I, 712; 713). One of these features only Greek names;
the other is dated a little later and has a lulab and shofar and is for
Theodoula and Moses. In Asia Minor, a menorah is depicted on an
inscription reading a prayer for all the fatherland at Acmonia, Hebrew
is attested a little earlier in a text calling for peace to Israel and
Jerusalem and this place (IJO 2, 169; 170). The only Jewish
inscription from Heraclea-Perinthos is an epitaph decorated with a
menorah, lulab, etrog, shofar and fire shovel (IJO 1, Thr4); and a
menorah, lulab, shofar and etrog feature on the epitaph of Samuel son
of Jacob at Docimeion (IJO 2, 183). Psalm 136:25, who gives food to all
flesh, for his grace is everlasting is inscribed on a block decorated with
a menorah from Nicaea (IJO 2, 153); and at Tavium, three inscriptions
have menorahs, two with the name Matheios (IJO 2, 163-165). A
177

menorah is used at Corycus on the grave of Anastasios and Jacob (IJO


2, 234); and painted onto a pillar from the Hippodrome at Tyre,
presumably marking a stall (IJO 3, Syr10). Hebrew is found in Smyrna
on the grave of a presbyter and son of Jacob (IJO 2, 41).
The fourth, fifth and sixth centuries sees the process of
Hebraization spreading much more profoundly in the west. Hebrew,
Hebraizing names and the menorah occur in Sicily during the fourthfifth centuries. The only two epitaphs from Sofiana respectively use the
name Judas Sabatias and the menorah (JIWE, 157; 158), likewise at
Agrigentum, where of the two inscriptions testifying to the community,
one bears a menorah and the other mentions a poor Jew (CIJ I, 654;
JIWE, 160). On mainland Italy, Venosa emerges clearly as a major
Jewish centre, with almost all of the inscriptions using Hebrew, Jewish
names or the menorah (CIJ I, 569-570; 575; 578; 579; 581; 584-586;
593-597; 599-600; 606-614; 616; JIWE, 43; 107). One of these
epitaphs records an elegy made by two apostuli and two Rabbis (611)
which is probably evidence of a direct link to Palestine. Jewish
communities are either expanding across the rest of southern Italy in
this period or else are motivated to take up these markers of their
Jewish identity, as several inscriptions from Naples use Hebrew or are
decorated with the menorah and other Jewish symbols (JIWE, 31-37;
CIJ I, 558-559), one of which mentions a Rabbi (JIWE, 36). Likewise
the use of Hebrew, the menorah and the formula from Proverbs 10:7,
the memory of the righteous is a blessing, are found at Oria, and Anna
is recorded as one who knew the Laws of her faith and saw the face of
God (CIJ I, 634-5); and a menorah is part of the synagogue mosaic
decoration at Bova Marina (JIWE, 140).
In the north of Italy and Hispania, a similar scenario of packaged
Hebrew, Jewish symbols and Hebraizing names occurs. In Milan, two
fifth century epitaphs use Hebrew and the menorah and other symbols,
one is for an Alexandrian man called Jose (CIJ I, 644; 646). In Tortosa,
an epitaph mentioning Rabbi Juda (JIWE, 183) and at Tarraco, an
178

epitaph for a Rabbi and archisynagogos from Cyzicus both use Hebrew
(JIWE, 186, IJO 2, 148); and an early medieval epitaph from Auch in
France uses Hebrew, the menorah, and was carved by Ionas (CIJ I,
671). More fleeting evidence for the percolation of Hebrew in this
period is found in the examples of a poorly preserved Hebrew
inscription from Bari (CIJ I, 633); a fragment of amphora from
Ravenna (JIWE, 10); and a lamp inscribed with the text of Proverbs
6:23, Because the commandment is a lamp, and the law a light from
Nola, dated to the seventh century (CIJ I, 554).
That the Diaspora Jews had an in-depth knowledge of the text of
the O.T. in Hebrew or Greek is not clear from the epigraphic record.
Selected passages of the Hebrew Bible were singled out, the use of
Proverbs 10:7 and the Deuteronomic curses being the most frequent.
These clearly were particularly appropriate to their epitaphic context,
and although the rest of the Hebrew O.T. or Septuagint were
presumably known to a degree, they rarely appear in epitaphs. This
suggests that knowledge of the literature was essentially oral, captured
perhaps in the strong representation of names linked to Jewish history:
in particular Jacob (twenty-one); Isaac (ten); Sara/h (eleven) and
Abraham and Solomon (three each). The name Juda/h/s is found forty
times, by far the most popular Hebrew name, which was presumably
adopted for its unmistakeable ethnic connotations.
However, what is apparent from this survey is that there are some
areas that are almost entirely empty of Hebrew namely, the interior of
Asia Minor, most of Greece, and Dalmatia and the northern provinces.
Hebrew is, by contrast, well diffused across the communities of
southern Italy and the western Mediterranean. What causes this divide
in language use between east and west? Hebrew was brought to the
Diaspora, but only adopted in some places. Why? A possible reason for
this patterning might be the coastal location of many of the sites in the
west Mediterranean, i.e. that maritime trade and immigration routes
brought people with new religious information directly from other
179

places, crossing, as it were, network distance. However, in addition the


use of Hebrew may indicate highly educated status, representing not
only a literate person, with the ability to read Torah in the original but
also the presence of somebody with the ability to carve it. On a more
secondary level, as the majority of the Hebrew inscriptions are funerary
formulas, most often reading peace or blessings, it may represent
social emulation or superficial adoption of the language, spreading
almost like a fashion.
The interior of Asia Minor, which is so full of Jewish communities,
is however the most interesting lacuna in the use of Hebrew. It might
be suggested that, as these communities are fairly close together
geographically and also in some senses quite isolated, this would have
entailed a level of social introspection and conservatism. It is clear
however, that the Jews of Asia Minor readily adopted use of the symbol
of the menorah, but that the use of Hebrew did not take root. It is
notable that when Hebrew is found in Asia Minor, it occurs in places
with large Jewish communities: Sardis, Acmonia, and Smyrna. The most
likely explanation of this phenomenon is that Greek was embedded as
the language both of everyday life and of worship in the Jewish
communities of Asia Minor, and this simply prevented the wider
adoption of Hebrew, although it did not prevent these communities
from adopting the other cultural markers of their religion as part of the
wider development of Jewish identity in later antiquity.

Movements of people and ideas across the Diaspora


It is abundantly clear that the epigraphic evidence undergoes a major
change from the third century onwards, with an enormous increase in
Jewish self-identification, manifest in the use of Jewish names, terms,
symbols and Hebrew. This observable phenomenon can only be
adequately explained as the epigraphic reflection of the transmission of
180

rabbinic reforms to the ordinary people of the Diaspora, as visual


evidence for the spread of universalising halakhah. The infrequent but
notable occurrence of rabbis in the evidence, some of who clearly
originated elsewhere, suggests that there may have been a clear
pattern of sending out religious authorities to aid the process of
Hebraization in the Diaspora. The rabbis are found more widely in
Spain and Italy, than in the Greek-speaking East: Rabbi Atticus on
Cyprus (IJO 3, Cyp1), Rabbi Arbiades at Naveh in Syria (IJO 3, Syr36);
and Rabbi Samuel, archisynagogos of Phrygia in Docimeion (IJO 2,
184). This supports the above suggestion that the eastern Diaspora was
perhaps more inward looking and perpetuated its earlier traditions
even after the transformation of Judaism in the mid second century.
However, aside from the possibly centralised movements promoted by
the religious authorities, other types of migrations and interactions
across the Roman Empire will have also contributed to the introduction
and acceptance of new religious information and norms of behaviour.
Particularly, people who travelled long distances across the
Diaspora from the region of Judaea might be suspected as being
harbingers of religious innovation. Examples from the epigraphy
include: in 100 AD, Praulus of Samaria, who was included on a list in
the Serapeion on Delos (IJO 1, Ach68); in the second century, a man
from Jerusalem found in Iasos, supporting a pagan festival, who may
have been granted metoikos status by the city (IJO 2, 21); in the midthird century, a decurion in the cities of Ascalon and Damascus who was
of the Terentine tribe of Scythopolis, found in Ostia (JIWE, 15);
between the second-fourth centuries, Jews of Israel in the Via
Nomentana cemetery in Rome (CIJ I, 21); Ionios (perhaps a Hellenised
form of Jonah) from Sepphoris (CIJ I, 362); and the wife of a man from
Caesarea, presumably in Palestine (possibly Mauretania) (CIJ I, 25),
both buried in the Via Portuensis cemetery in Rome; a man from
Tiberias found in Senia on the Dalmatian coast, during the fourth-fifth

181

century (IJO I, Dal2); and in the fifth century a child of Syrianos in


Venosa (CIJ I, 579).
The strong-ties of marital links will have aided the diffusion and
acceptance of ideas. Early evidence for long-distance marriages is
provided by Debbora from Antioch (in Pisidia) who was married to a
man from Sillyon in Pamphylia and recorded in Apollonia in Phrygia in
the first-second century (IJO 2, 180). A much later inscription from the
region of the Via Portuensis records Sigismund (perhaps a Burgundian),
whose wife Sarra appears to be from Hydrantum (CIJ I, 499); the wife
of a soldier in a Jewish troop of Emesenes who was buried in Iulia
Concordia in the fourth-fifth century, suggesting either that the soldier
had married her as a local, or that she had accompanied him from
Emesa on deployment (CIJ I, 640); and later still, a seventh-eighth
century inscription in Hebrew from Tarentum is for the wife of Leon,
son of David of Melos, making a link with the Aegean islands (CIJ I,
621).
As well as longer-distance travel, there is also evidence for more
localised movements, probably reflective of the more commonplace
shorter-distance interactions in the Diaspora. These people act as lowlevel connectors of the network. For example, in the second-third
century, epitaphs from Corycus record a man from Anemourion, further
south along the coast of Cilicia (IJO 2, 233), and a couple who were
citizens of Seleukia, fifteen miles away (IJO 2, 236). In the mid-third
century, Amathbel, a Palmyrene, was remembered at Dura Europus
(IJO 3, Syr90); and the Persian language inscriptions from the
synagogue at Dura testify to the cross-border connections with Persia
(IJO 3, Syr111-126). Post-AD 212, Aurelius Heortasius, son of Iulianus,
died in Hierapolis in Phrygia but was a citizen of the nearby Lydian city
of Tripolis

(IJO

2,

191). In the third-fourth

century, Aurelius

Alexandros, who was from nearby Gordiene, died on the territory of


Claudiopolis (IJO 2, 152); in the fourth century, a man from Hypaipa
gave a donation to the synagogue in Sardis (IJO 2, 95); between 200
182

and 352, an Antiochene gerousiarch was buried in Apamea-on-theOrontes (IJO 3, Syr74); and in the same place at the end of the fourth
century, an archisynagogos of Antioch was mentioned in the synagogue
(IJO 3, Syr53). In the west, in the sixth century, a family from Saranda,
in modern southern Albania, moved across the Adriatic to Venosa
(JIWE, 107).
The movement was mainly east to west, reflecting the continual
migratory flows from the Near East to Rome, which are particularly
evident in evidence from late antiquity.356 In the first century, a Roman
citizen named Gaius Seius Ptolemy, a Samaritan, died in Sicily at
Termini Imerese. His name and the use of chaire on his tombstone
suggest an Egyptian link (JIWE, 161). In the second century, the son of
Menippos from Samaria died in Kamiros on Rhodes (IJO 2, 11); during
the second-fourth centuries, a man from Laodicea (either Laodicea on
the Lycus or Laodicea Catececaumene in Asia Minor) was buried in the
Via Portuensis (CIJ I, 296); in the third century, Antiochus the
Samaritan was buried in Hipponion (JIWE, 138); in the fourth-fifth
century, Benjamin, prostates of Caesarea, was buried in Naples (JIWE,
30). This could be the Palestinian but could also be the Mauretanian
city, because JIWE, 31, also from Naples, is the fifth century epitaph of
a Gaudiosus, explicitly from Mauretania. In the fifth century in Beroea
in Macedonia, the community was being bolstered by Egyptian Jews, for
example, Joses the Alexandrian (IJO 1, Mac6); and in AD 539 a
Samaritan woman died at Salonae in Dalmatia (IJO I, Dal4).
However, it was not entirely one-way flow. That the Jews of the
Diaspora were interacting with the Judaean homeland is testified to by
two epitaphs from Jaffa, one for an elder of the synagogue of the
Cappadocians who was from Tarsus in Cilicia (IJO 2, 249), the other
mentioning Symmachos from Chios (IJO 2, 4). It is well known that
Beth Shearim was a resting place for many better-off Jews (even of
senatorial rank IJO 3, Syr26) from the eastern Diaspora communities,
356

See Noy, D., Foreigners at Rome, London: Duckworth with The Classical Press of
Wales, 2000, pp. 54-84.

183

with Jews from Tyre (IJO 3, Syr6; 7; 8), Sidon (IJO 3, Syr17; 18),
Byblos (IJO 3, Syr32), Iamour (IJO 3, Syr21), Antioch (IJO 3, Syr74),
Palmyra (IJO 3, Syr51), and Berytus (IJO 3, Syr26) all found there.
Plotting these movements (see Map 5A) reveals quite clearly that
the inter-community movements and interactions have a strong east to
west bias; however, it is notable that there are no really clear centres
in the western Diaspora. It might be expected that Rome would exert
more of a gravitational pull, but this is not the case, with people also
moving (or being moved) to the coast and interior of Asia Minor,
Greece, and southern Italy. What is worth noting too is that many of the
inscriptions that mention another place of origin come from places on
the coasts, implying that harbour towns were particularly attractive
destinations for immigrants, and that these often also stated their place
of origin on their epitaphs.
From this assessment, it is apparent that Jews in the Diaspora
were regularly moving across it from their own volition as well as from
external compunction. The tangible movement documented by these
inscriptions adds an important real-world dimension to the argument
for the less visible movement of ideas, suggested by the progressive
Hebraization of the rest of the Diaspora. It will therefore also be helpful
to visualise the hypothetical networks that might have existed, to
ascertain what the network analysis adds to the epigraphic analysis.

Visualising the Network

This section takes the epigraphic data and uses network principles to
hypothesise on the now largely invisible communication routes and
potential avenues of information transmission across the Jewish
Diaspora. A distribution map (Map 5B) shows the places where Jewish
184

inscriptions

have

been

found,

clearly

demonstrating

that

the

communities were concentrated in Greece and Asia Minor, with some in


Sicily, Egypt, and the Near East: limited information that simply takes a
snapshot of all the evidence, and offers no insight into the interactions
between them. By linking the nodes into a network, it is possible to
visualise the communication links that are suggested here to be central
to understanding the diffusion of rabbinic halakhah.
An initial Proximal Point Analysis (PPA) of every piece of
epigraphic data captures the end picture, linking every known node,
regardless of date, to its three closest neighbours. This first model
highlights

clusters,

isolated

communities,

and

centres.

It

is

preliminary analytical model and clearly not a reflection of the actual


connections that existed between these sites. What it does allow is an
initial observation of the geographically determined Diaspora network:
the empty spaces, long-distance overland or maritime links, and some
of the constraints that terrain imposed on communications.
The second series of network maps are a developed version of the
PPA. These networks visualise the Hebraization process between the
second and sixth centuries AD, mapping therefore only those finds that
use Hebrew, Jewish symbols, or make explicit reference to Israel. This
series responds to what is known of the date range of the evidence, by
mapping the find spots in hundred year blocks, and it builds in Judaea
as the place from which the reforms were disseminated. Rome is also
treated as an auxiliary centre of gravity. Each find location is once again
linked to three other locations. However, instead of these being the
three closest neighbours: one is always Judaea, and one must be a preexisting link, except in later cases when there are three established
nearby neighbours. This analysis helps to visualise in actual terms the
spread of information across the Diaspora, highlighting the growth of
localised centres, clusters and routes of information transmission. The
spread of halakhah was driven by the rabbinical centres of Palestine,
which is clear from the hyper-real picture the maps present, but these
185

also highlight the potential routes of a more organic process of contact


and adoption. On the maps, a site with only one epigraphic find is
written in lowercase lettering; a site with more than one find or an
inscription that reveals a community or synagogue is in uppercase.

Proximal Point Analysis: Map 5C


Even though the initial PPA is a static picture with inherent limitations,
it has a useful function as a preliminary snapshot of the geographical
pattern of data, nearest neighbours and interactions of the Diaspora
communities.

It

highlights

places

of

geographical

isolation

and

connectivity on the grounds of physical location alone.


Immediately clear are the long distance links across the western
and northern Diaspora, as opposed to the tightly geographically
integrated eastern networks of Asia Minor, Egypt and Syria. In the
west, the island sites on Ibiza, Malta, and Sardinia clearly provide
important links between Hispania, North Africa and Rome; Ibiza in
actuality emerging as an inter-regional centre. Romes gravitational
pull is however probably underrepresented, although it does emerge as
a centre. There is a clear disconnect between southern Italy, which was
linked via Hydrantum to Greece (Hydrantum was a major port for
immigration from Greece and Epirus) and the northern sites, which
concentrate around the Po valley and link with the northern provinces
and the Dalmatian coast. This may quite reasonably reflect a divide
between the two areas and highlight patterns of local interaction and
the different origins of their Jewish communities. The sparser network
in the northern Adriatic is composed mainly of single finds, suggesting
that the Jewish presence here was more superficial and involved longer
links.

This is borne out by the epigraphy: for example, the Latin

epitaph from Aquileia is for a slave, possibly captured during Pompeys


wars in Judaea (CIJ I, 643). The only evidence from Ravenna is an
186

amphora fragment inscribed in Hebrew (JIWE, 10), and the find from
Concordia is a Latin epitaph of a woman whose husband was a soldier
from Emesa (CIJ I, 640). The larger sites with more established
communities are later: the synagogue in Brescia dates from the fourth
century. Likewise, the inscriptions from Mediolanum do not date to
before the fifth century AD.
The most striking thing about the western Diaspora is the
separation of Magna Graecia from the rest of Italy, and the clear
importance of Sicily as a local network, connecting with Rabato on
Malta, Carthage and Naro in Africa Proconsularis and the tip of
Calabria. Tauromenium is a local centre at the Italian end of the island;
Rabato serves as an offshore communications hub between Sicily and
Tripolitana.

Another

interesting

area

is

the

little

pocket

of

introspection in the sites along the Danube limes in Pannonia, which


must be connected either with participation in or supply of the military
camps indeed, the praepositus stationis of a camp at Spondilla made a
dedication in the synagogue at Intercisa (CIJ I, 677). The long-distance
links into Raetia and Germania Superior are somewhat misleading, as
both of the finds at Regensburg and Badenweiler (CIJ I, 673; 674) are
portable amuletic texts that may not actually have been used by Jews
per se; although they do show that Jewish religious ideas were
attractive and taken far beyond places where there were established
communities.
Moving into the eastern Empire, although northern Greece,
Moesia Inferior and Thrace also have diffused long-distance links,
westward into Dalmatia via Stobi and Doclea, and eastwards into
Bithynia through Chalcedon, their communities were bigger with
attested synagogues. The missionary journeys of Paul into northern
Greece also testify to the importance of these places in the Jewish
Diaspora, and also to the intercommunication between them, regardless
of their geographical distribution. This area of the network clearly also
forms an important corridor into Dalmatia, as well as connecting down
187

into Athens and southern Greece. The network of southern Greece is


tightly integrated, unsurprisingly with a centre at Athens. The area
links to southern Italy via Patras, and to Crete via Taenarum on the
southern tip of the Peloponnese. Eastwards the connection across the
Aegean to Asia Minor is via Syros, Delos and Ikaria into Teos, Ephesus
and Chios. Crete also connects northwards to the rest of Greece via
Melos.
The network in Asia Minor is tightly integrated and evenly
spaced, presenting a picture of regular short-distance interactions, and
is most even along the coast of Ionia and Caria, extending into the
coastal islands. Nysa and Teos emerge as centres, and although there
were probably synagogue buildings in both places, there is no
indication that these were local centres of diffusion, and perhaps
instead should be better characterised as communication corridors
between the surrounding communities. Although this analysis takes no
account of geographical costs or directionality, the network reflects the
geographical features of the landscape. The sites in Lycia, Pisidia and
Pamphylia mainly connect amongst themselves and along the coast of
Cilicia, rather than across the high mountains of internal Lycia. This
creates a pocket of introspection, supported by the fact that most of
these are single inscription find spots. However, this pocket does link
up into Phrygia to Apollonia, noteworthy in particular because the
inscription here makes explicit the connection: it is the epitaph of
Debbora, of Pisidian Antioch, who married a man from Sillyon in
Pamphylia (IJO 2, 180). Apart from this link to the south, the sites in
Phrygia are also somewhat separate, although linked also to Caria and
Ionia

through

the

Eumeneia-Hierapolis-Aphrodisias

corridor.

The

Phrygian network extends north into the sites of Bithynia across a


slightly more diffuse network via Cotiaeon and Germa. Amastris
provides the only link to the communities of the north shore of the
Black Sea, which is otherwise completely isolated. The centrality of
Panticipaeum and the isolation of the Black Sea sites is quite reflective
188

of reality. Although the Jewish Diaspora extended eastwards into Persia,


the

border

between

Rome

and

Parthia

must

have

restricted

communications in this direction. In the high Anatolian plateau there is


another network of somewhat isolated communities in Cappadocia,
which links through Laodicea Catececaumene into Phrygia, and via
Tyana across the Taurus Mountains south into Tarsus.
On first impressions, the network on Cyprus also appears
introspective, with only a couple of links out of the north to the coast of
Cilicia. The high mountain ridge between north and south seems not to
divide the network, until it is recalled that in the early period, Jews are
epigraphically attested only on the eastern side of the island at Kition in
the fourth century BC (IJO 3, Cyp7-9), and in the first century BC at
Kourion (IJO 3, Cyp5). The link to the Near East at that time is explicit
the inscriptions at Kition are in Phoenician. There were many Jews in
Salamis, as attested at Acts 13:4. The strong communications between
Cyprus and Judaea continued, as the Jews revolted under Trajan,
demonstrating interaction with Palestine and the rest of the Diaspora.
However, the network connectivity has been skewed by the postbanishment357 return of Jews to the island in the third-fourth centuries,
when they are found at Salamis, Golgoi, Lapethos and Morphou. This
has reconfigured the network structure on the island by masking the
links Kition and Kourion had with Phoenicia, creating a stronger
internal network, and highlighting instead the new links to Cilicia.
The network in Syria shows two clusters, the coastal cities and
the pocket of communities inland in Trachontis. The network in
northern Syria is more diffuse, with a hub at Antioch, which connects to
Dura Europus and Edessa. These two sites look isolated, but this may
be misleading as the Jews in Dura and Edessa probably communicated
with Persia, especially in the former case. Egypt and Cyrenaica are the
only parts of the network that are entirely isolated. In Egypt, the
communities form a long line of interconnected sites along the Nile.
357

Cassius Dio, Roman History, LXVIII, 32, 1-3, see Schrer2, History, Vol. III.i, p. 68.

189

This prediction is quite accurate, as the Egyptian Diaspora was well


established and, as a consequence, fairly introspective. The PPA misses
Alexandria as a major Mediterranean hub, but the lack of an emergent
centre in Egypt suggests that the population was evenly diffused and
integrated. Cyrenaica is also entirely introspective. The rest of North
Africa is, by contrast, fairly well integrated into the western networks
via Sicily, Sardinia, and Ibiza.
This initial analysis captures pockets of isolation, and this reflects
the real situation that may be supposed in the hinterlands of Asia
Minor, Egypt, Cyrenaica and the Black Sea. It is notable that two of
these more introspective areas, Egypt and Cyrenaica, were regions
where Jewish rebellions occurred. Corridors of communication exist
between North Africa and Sicily, between Greece and Asia Minor via
Delos, and between Thrace and Bithynia via Chalcedon. The divide
between north and south Italy is probably indicative of reality, although
the gravitational pull of major cities is absent, and as a result the major
Jewish communities in Rome, as well as Alexandria and Jerusalem, do
not look very important. This is partly to do with the fact that this model
does not have different costs built in for land and sea connections a
more

complex

version

of

this

analysis

would

factor

in

these

differentials. This analysis is also only built on epigraphic data, and a


more developed version should also include the literary evidence for
Diaspora communities, which would add a more realistic gravity to
these three major cities.

Networks over time: the process of Hebraization (Maps 5D-5G)


Following from these initial observations of clustering and isolation, the
following Proximal Point Analyses build in the date range of the
evidence for the Hebraization of the Diaspora. Only evidence using
Hebrew, Jewish symbols or making an explicit mention of Israel or a
190

Rabbi is used here. Once again, new nodes connect not strictly to their
three closest neighbours, but, as explained above, to Judaea and to one
established connection, except in later cases when there are three
neighbours close by. This builds into the network the role of Judaea as
the place from which the reforms were disseminated as well as
attempting

to

simulate

more

localised

contact

and

exposure

Hebraization. Rome is also treated as a centre of gravity.


Although this skews the maps towards having very long-distance
links and makes Judaea a heavy centre, it acts as a counterpoint to the
initial, un-weighted PPA and creates a picture of more realistic
interaction patterns, the idea of rabbinic mission to the Diaspora, in
which

places were first exposed to the halakhic reforms, and

subsequent localised spread of information. The analyses cannot reflect


how communities declined, as the evidence is too unclear to ascertain
this, so once a piece of epigraphic evidence is recorded; it is assumed
that Jews continued to be present. The exception to this rule is Pompeii,
which, although it is not removed from the network itself, because
doing so would reconfigure the earlier network, receives no new links
after AD 79.

Map 5D: first and second centuries AD


This map comprises the sparse evidence for the early stages of
Hebraization after the destruction of the Temple in AD 70 and the Bar
Kokhba revolt of the 130s. The links are naturally very long: to Ibiza,
Rome, Pompeii, Athens and southern Greece, and middle Egypt. Some
are not connected with the rabbinic reforms, for example, the fragment
of amphora, inscribed with Hebrew (or Samaritan Hebrew) from Ibiza
simply implies long-distance trade with Judaea, and the evidence from
Pompeii is connected with slaves presumably taken to Italy after the
Jewish Wars. However, the targeting of the established communities in
191

Rome and Athens, and to a lesser extent Egypt, is fairly clear. It is


however more interesting at this stage to look at the gaps in the
network, which in Cyrenaica and Egypt can be connected with the
quashing of revolts, but in Asia Minor is quite striking. Presumably,
attempts to bring the reforms into these well-established communities
(as highlighted by the missionary journeys of Paul through the
synagogues of Asia Minor) were less successful, because these were so
well established and perhaps too tightly integrated to be socially
receptive to the innovation.

Map 5E: third century AD


The jump in the network connectivity in the third century is impressive.
What is immediately clear is that Hebraization was not an organic
process, spread through geographically proximate places, but rather a
pan-Empire phenomenon: from Caesarea in Mauretania to Pannonia,
from Sicily to the Black Sea. Many of the places where Hebraization is
found at this stage are coastal Carthage, Catania, Hydrantum, Kos
and Corycus, which highlights the importance of geographical position
within the network for exposure to and acceptance of new religious
ideas. However, the sites in the hinterland of Asia Minor also begin to
open up at this stage, notably in Phrygia and Bithynia. A reason for this
might be found in the contemporaneous flourishing of Christianity in
internal Asia Minor, which although it did not drive the process of
Hebraization, provided a localised change in the dynamics of the Jewish
communities. Corycus on the Cilician coast is disconnected with the
rest of Asia Minor and connects instead with Cyprus, where a vow of
Rabbi Atticus and a limestone seal with Jewish symbols were found in
this period (IJO 3, Cyp1-2). Hebraization is also found spreading into
Syria, although only to a superficial degree at this stage. The process
does not continue in Egypt.
192


Map 5F: fourth century AD
The process of Hebraization in this period does become more organic in
Asia Minor, Syria and Sicily. The communities on the coast of Ionia are
drawn inland towards the tightly interconnected sites in Phrygia.
Phrygia, via Docimeion and Dorylaion, is the connective corridor to the
sites in Bithynia, with Nicomedia emerging as a centre between the
Black Sea sites and the rest of Asia Minor. This highlights an increase in
cross-Euxine connectivity centred near the Bosphorus that may in part
reflect the transferral of the capital of the Roman Empire to
Constantinople. A similarly tight network emerges in the sites round the
southeast corner of Sicily, also including Malta; and likewise in coastal
Syria and inland in Trachontis. The network between the communities
on the heel of Italy probably represents a similarly introspective
process. The evenly distributed pattern of Hebraization seen in the
well-established communities of Thessaly and Macedonia may result
from exposure to the phenomenon, Hebraization being seen already in
the third century in Phthiotic Thebes and Philippopolis. The network in
the western Mediterranean appears to have been far more centralised
than in the east, reflecting the continuing importance of Rome. The
apparent lack of an eastern centre outside of Judaea is therefore
particularly noteworthy in contrast.

Map 5G: fifth-sixth centuries AD


The final map shows a deepening of the process of Hebraization,
especially at a localised level. The notable growth in the east is in the
region of Constantinople, creating an area of introspection, doubtless
reflecting the new capitals gravitational pull. In the west, Hebraization
193

is spread through the locally integrated networks of communities in


Sicily and Calabria. The geographical distances involved in the western
network are much longer than in the east, suggesting that the network
was of a different quality. It seems to indicate that either the west was
more centralised around the Roman hub, or alternatively that the
places on the coast of Spain were more cosmopolitan and had longdistance links elsewhere. This latter suggestion is supported by the find
from Tarragona, recording a Rabbi Latous from Cyzicus (JIWE, 186).
The sites along the coast of Spain and into southern France actually
make up an almost separately interacting network, which may reflect
the disintegration of the centralising Roman force: it was during this
period that the western provinces came under barbarian control and
the Empire split.

What this has shown is how Hebraization as the visible remains of the
pan-Judaic universalising reforms might have travelled across the
Diaspora. The analyses have shown that it was a centralised process,
occurring throughout the Roman Empire during and after the third
century. It may be that certain large Jewish communities were targeted,
for example, Sardis, Athens, and Rome. It has shown the difference in
the network structure in east and west, in particular the more regular
eastern network, especially in Asia Minor, implying a more organic
adoption, based on localised interactions. Further, the models have
highlighted areas of introspection and more gradual diffusion (Sicily
and Syria), and shown that some places were more receptive to new
information than others. It has also illuminated some interesting
lacunae namely, Egypt, and to a lesser extent, northern Syria. The
revolt in Egypt resulted in the destruction of a large part of the Jewish
population there, which offers some kind of explanation for the lacuna.
Also missing is the large Jewish community of Antioch, who are known
194

from literary sources and from inscriptions from elsewhere in northern


Syria. The civil unrest between the Jews and Gentiles may account in
part for the absence of epigraphic evidence.

Conclusions: Activating familial networks of ethnicity


This chapter has shown that the Jewish communities in the Roman
Empire underwent a Diaspora-wide process of Hebraization. The
destruction of the Temple in Jerusalem and the subsequent quashing of
the Bar Kokhba and other Diasporan revolts changed the situation of
Jewish communities both in Judaea and in the Diaspora. Without an
actual centre to the religion, Judaism ultimately turned inwards. A
newly heightened sense of persecution activated the familial ethnic
bonds already in existence, making a strong-tie network that promoted
religious innovation. The rabbinic reforms were spread in this way.

Information

cascade

in

the

Diaspora:

Hebraization

and

Christianity
The

rapid

and

universal

process

of

Hebraization,

manifest

epigraphically in the use of Hebrew, Jewish symbols and the panEmpire rise in popularity of explicitly Jewish names, can be understood
as the result of

an information cascade. The strong-tie ethnic

connectivity of the Jewish Diaspora that was activated in the years


following the destruction of the Temple and the Bar Kokhba revolt made
the network increasingly susceptible to religious innovation. The Jewish
Diaspora became a percolating vulnerable cluster, and the religious
authorities in Palestine used it to transmit the rabbinic reforms.

195

Christianity made a similar use of the same the strong-tie ethnic


network and the sense of anxiety and persecution that was latent in the
Jewish communities at the end of the first century, except that it took
Christians in the opposite direction from the Jews. Pauline Christianity
was marked by the application of linguistic terms for close family to all
those who lived in the community of Christ. The creation of pseudofamily is a commonly used persuasive rhetorical device to highlight the
ethnic bond between Paul and his audiences Paul addresses the mob
in Jerusalem in Aramaic at Acts 22, calling them brothers and fathers,
and again when he addresses the council of chief priests at 23:6 and it
has the effect of creating a cognitive strong-tie network between them.
The Acts of the Apostles always refers to the Christian believers as
brothers (for example, Acts 1:15; 3:17) and as such, reinforced the
social ties that bound them. Christianity was able to persuade and
convert effectively both by utilising the actual strong-tie familial
network of the Jewish Diaspora and by simultaneously manufacturing a
new one for believers in Christ.
Because the Jews in the Diaspora were ready for change, the
network they formed was utterly vulnerable to the religious innovations
brought by people who preached change. Martin Hengel has argued 358
that the Jews in the Diaspora had strong messianic hopes in the
decades following the destruction of the Temple, which were manifest
in the various revolts and also in the person of Simon Bar Kokhba. At
the same time, the sect of Christianity believed that it offered a
messiah, and also an explanation for the cataclysm punishment of the
Jewish people by God for failing to recognise the divinity of His Son.
Rabbinic halakhah and Christianity are different manifestations of the
same cognitive response to the disasters that befell the Jewish people,
and both required an internal change in the religion: promoting better
adherence to Laws and moral codes. Both used the same type of strong-

358

See footnote 340.

196

tie network to transmit their message, both of which swept across the
Roman Empire.

197

Chapte r 6.
The Cult of the Most High God.
God-fearers and the redefi nition of the
Jewish-Gentile relationship

Introduction

This chapter examines the evidence for the cult of Theos Hypsistos
the highest God as a third investigation of the role networks play in
the spread of new religious ideas. Theos Hypsistos is the Septuagint
translation of the Hebrew El Elyon, and the cult has long been
associated with Judaism. The evidence is found mostly in Greece, Asia
Minor and the eastern Roman Empire, and spans from the second
century BC in Macedonia until the fifth century in Phoenicia, reaching a
zenith of popularity in the second to third centuries AD. 359 The
interpretation of the cultic evidence has sparked no small amount of
controversy, and some scholars have proposed that the material should
be divided between the Jewish and pagan usages of the title. 360
However, Mitchell has shown that this process of attempting to divide
the evidence only serves to highlight how difficult it is to do so, and I
agree with his argument that the evidence should be understood to
359

This has been argued to be a result of the epigraphic habit rather than an indicator
of a rise in popularity of the cult see Mitchell, S., The Cult of Theos Hypsistos
between Pagans, Jews, and Christians, in Pagan Monotheism in Late Antiquity,
Oxford: Clarendon Press, 1998, pp. 108-110. However, there may be associated
reasons for an increase in the particular use of the name Theos Hypsistos, which will
be examined here.
360
See in particular Trebilco, P., Jewish Communities in Asia Minor, Cambridge:
Cambridge University Press, 1991; Bowersock, G., The Highest God with particular
reference to North Pontus, Hyperboreus 8, 2002, pp. 353-363.

188

represent one cult with a level of internal unity, broadly to be termed


aniconic and monotheistic.
The arguments are as follows. Mitchell presented the case for a
unified cult of Theos Hypsistos, associated more or less with the Jewish
Diaspora, a group of Hypsistarians or Hypsistari, who should also be
identified in the theosebeis or god-fearer inscriptions associated with
the Jewish community. Other scholars have argued that the body of
material represents a common term for the highest god, which was
used to describe a variety of different deities. Interpreting the same
body of material, Stein argues that the inscriptions testifying to these
cults (Hypsistos alone, Zeus Hypsistos and Theos Hypsistos) cannot be
regarded as one and the same cult of pagan-Jewish character. One
should rather distinguish between different cults of pagan origin, of
Jewish origin and those which comprise elements of both.

He also

disagrees with Mitchells analysis and argues that the theosebeis are
not to be identified with the adherents of Theos Hypsistos. 361 Similarly
Bowersock argues that Theos Hypsistos is not always one and the same
god, linked with Judaism, rather, that the cult term was applicable to
many different gods, the Jewish god among them. Wallraff also shares
the view that there was nothing like a coherent cult of Theos Hypsistos,
and argues that there was no specific ritual or specific priests of the
cult. He sees no evidence clearly suggesting monotheistic tendencies,
but prefers to describe the cult with the term henolatry or
henotheism.362
Mitchell, by contrast, suggests that the difficulty in dividing the
Jewish and pagan usage of the term reveals much about the nature of
the cult and the religious climate in the Roman Empire, Asia Minor in
particular, arguing that the judgement is entirely arbitrary and that
the cult had room for pagans and for Jews. More than that it shows
that the principal categories into which we divide the religious
361

Stein, M., Die Verehrung des Theos Hypsistos: ein allumfassender pagan-jdischer
Synkretismus?, in EA 33, 2001, 119-126.
362
Wallraff, M., Pagan Monotheism in Late Antiquity. Remarks on a Recent
Publication, in MA 6, 2003, pp. 534-5.

189

groupings of late antiquity are simply inappropriate or misleading when


applied to the beliefs and practices of a significant proportion of the
population of the eastern Roman empire. 363 Mitchell argues that the
Hypsistos inscriptions form a unified corpus of material, representative
of one aniconic cult, basically monotheistic or henotheistic in nature,
closely linked to the monotheism of the Jews.364
That the inscriptions form a unified corpus is the point from
which this analysis starts. Although the individual items of epigraphic
evidence are often very brief and somewhat broadly dated, some useful
conclusions may still be drawn from them. The data is analysed in three
sections. The first looks at the material from the Hellenistic period-first
century BC. The tentative suggestion is made that interactions with
Samaritans/Jews on Delos may have been instrumental in the early
propagation of the cult. The next section examines the material until
the end of the first century AD, and the final section analyses the
epigraphy from the second-third centuries

AD. The number of

dedications increases enormously during this final period, and suggests


three general conclusions: that the cult transgressed social divides; that
worship was conducted in a variety of places; and that the spaces in the
Roman Empire where the cult is absent are revealing as to its nature
and relationship with Judaism. Especially given the scant information
that can be gleaned from the epigraphy, understanding of the cult is
increased through a visualisation of a network analysis, beginning with
a Proximal Point Analysis (PPA) and then mapping the cult development
over time through three PPA networks that follow the temporal
divisions made in the epigraphic analysis.
The hypothesis proposed here is that the profound spread of the
cult of Theos Hypsistos in the second-third centuries is not simply a
result of the epigraphic habit, but also represents the Gentile response
to the tightening of boundaries within Judaism. Judaism in this period
363

Mitchell, Hypsistos, p. 115.


Caveats to add to this are the dedications in some areas, especially Thrace, which
represent an anthropomorphic Zeus figure; and the oracle from Oinoanda in Lycia,
which defines a cohort of angels or subordinate deities. This shall be examined below.
364

190

was undergoing a major renewal following the destruction of the


Temple in AD 70 and the putting down of the Bar Kokhba revolt in 135.
It was argued in the previous chapter that, beginning around the
second-third centuries, the Jewish communities of the Diaspora were
re-Judaized, and that this process is visible epigraphically in the
exponential rise in explicitly Jewish names, symbols and the use of
Hebrew. It is suggested that this process must also have impacted on
the Gentile god-fearing community that had long been attached to the
synagogues and to worship of the Jewish God. Although the rabbinic
reforms

did not explicitly

separate

Gentiles from Judaism, the

reiteration of the Laws made the boundaries between them starker, and
this may have had the effect of distancing them from Jewish society. The
religious network of erstwhile god-fearers continued to worship the God
they believed in, just as a separate cult, that of Theos Hypsistos. This is
not to suggest that the separation between Jews and Gentiles lasted, as
the Aphrodisias inscription of late antiquity, possibly dating from the 5 th
century, lists god-fearers alongside Jews (and proselytes) and shows
that, by this stage, Jews and Gentiles were once again closely
associated. But the gaps in the evidence for god-fearers and proselytes
in the second and third centuries need to be explained. The suggestion
here is that the popularity of the cult of Theos Hypsistos at this time
can go some way towards offering an explanation.

191

Origins: Translating the Name of God & Gentile god-fearers

In the third century BC, the Greek translators of the LXX used the term
Theos Hypsistos to describe the Jewish God in his mystical function as
El Elyon, as opposed to YHWH or Adonai. However, the occurrences in
the Hebrew of the combined form El Elyon are extremely limited. At
Gen. 16:18-19365 and 22, the Tetragrammaton and El Elyon are cognate:
And King Melchizedek of Salem brought out bread and wine; he was
priest of God Most High. / He blessed him and said, Blessed be Abram
by God Most High, maker of heaven and earth; and blessed be God
Most High, who has delivered your enemies into your hand!, and But
Abram said to the king of Sodom, I have sworn to the LORD, God Most
High, maker of heaven and earth. The combination of El and Elyon is
found again in the Psalms, Ps. 78:35 and 56, They remembered that
God was their rock, the Most High God their redeemer. / Yet they
tested the Most High God, and rebelled against him.
However, otherwise, the references are simply to Elyon alone, the
Most High. This occurs much more frequently, especially in the poetic
verses, for example, in Ps. 9:2 I will sing praise in your name, O Most
High; 18:13 The LORD also thundered in the heavens, and the Most
High uttered his voice; 46:4 There is a river whose streams make glad
the city of God, the holy habitation of the Most High; 47:2 For the
LORD, the Most High, is awesome, a great king over all the earth; 91:1
You who live in the shelter of the Most High. It is also used throughout
2 Esdras especially, for example: 4:2 do you think you can comprehend
the way of the Most High?; 7:33 The Most High shall be revealed on
the seat of judgement; 7:42 only the splendour of the glory of the Most
High; and in a particularly notable section, 7:79 If it is one of those
who have shown scorn and have not kept the way of the Most High,
who have despised his law and hated those who fear God. This refers
365

All Biblical quotations are taken from the NRSV, Oxford: Oxford University Press,
1995.

192

to the Jews as those who fear God, but came to be used for Gentile
worshippers of the Jewish deity.
This combined name of God is also found a few times in the N.T.,
where Jesus is called jIhsou` uiJe; tou` qeou` tou` uJyivstou at Mark 5:7
and Luke 8:28. Other places that mention Jesus in connection with the
name Hypsistos are found at Luke 1:32 (uiJo" uJyivstou); 1:35 (duvnami"
uJyivstou); and 1:76 (profhvth" uJyivstou). In other N.T contexts the

expression does not necessarily connote any link to Jesus, for instance
in Acts 16:17, where the slave girl cries out that Paul and his
companions were slaves of the Most High God, and in Heb. 7:1 in
reference to Melchizedek as the priest of the Most High God. It is
apparent that use of the term Theos Hypsistos within either the Old or
New Testament texts is rare. However, Hypsistos alone is rather more
common.
Mitchell notes that, because the term is originally Jewish, there
has been a strong academic bias to divide those Hypsistos inscriptions
that are Jewish from those that are pagan. His opinion, to the
contrary, is that: most pagan or Jewish examples of the term Theos
Hypsistos are formally indistinguishable from one another and that the
arguments for assigning them to either category are rarely decisive. 366
Instead he argues that it is more fruitful to view this as shared
terminology between pagans and Jews, and should as such prompt
questioning at a deeper level about the nature of the common ground
between worshippers of Theos Hypsistos and the Jews of the Diaspora.

Common ground in Hellenistic literature


Jewish and Gentile literary examples of this shared ground are detailed
by Collins,367 who examines early Greek commentators on Judaism, for
366

Mitchell, Hypsistos, p. 112.


Collins, J., Between Athens and Jerusalem, Jewish Identity in the Hellenistic
Diaspora, The Crossroad Publishing Company: New York, 1986.
367

193

example, Hecataeus, Strabo and Varro, and finds them in general


respectful of Judaism and interested particularly in its philosophical
aspects: it appears then that there was a dimension of Judaism which
was quite attractive to the Hellenistic world. This was its philosophical
dimension, its ethical code, and aniconic God.368
Although this attitude was not representative of the entire
Hellenistic environment, similar philosophical standpoints existed
among the educated elites of the Hellenistic world. Collins notes that
the fragments of Jewish sibylline writings preserved in Theophilus and
Lactantius attest God as one in both unity and uniqueness, eternal,
invisible and self-begotten, reflecting a philosophical Judaism that is
indistinguishable from philosophical Hellenistic religion. They mention
no specifics associated with Jewish religious identity, and so Collins
concludes that there is no clear distinction between the ethics of
Judaism and those of a God-fearer, or even of a philosophically minded
pagan.369
Similarly, the gnomic writings of Pseudo-Phocylides are soft on
polytheism

and

idolatry,

implying

instead

practical

Jewish

monotheism.370 Collins argues that this is in order to minimize the


disjunction from Hellenistic civil and social life required of the convert
to Judaism, but also of the Jew who wished to participate fully in
Hellenistic life, and the God-fearing pagan, who may not have broken
with polytheism at all.371 It is apparent that the aspects of Judaism
more difficult to reconcile with pagan civic life and values were played
down by the Jewish writers at the time.372

368
369
370
371
372

Collins, Jerusalem, p. 8.
Collins, Jerusalem, p. 153.
Collins, Jerusalem, p. 145.
Collins, Jerusalem, p. 146.
See also Schrer2, History, vol. III.i. pp. 153-155.

194

After the destruction of the Temple: A change in attitude


This is in marked contrast to some of the later Jewish writings. The
Jewish trend towards hostility seemingly begins only after AD 70.
Regardless of individual political attitudes, it is, after all, difficult to
imagine the general Jewish population being particularly tolerant of or
welcoming to Gentiles after

various Emperors and the

Roman

administration had destroyed their Temple, imposed harsh religious and


financial restrictions upon them as a distinct group, and massacred
Diasporan and Judaean Jews. It is unsurprising to find this attitude
reflected in the literature. Probably composed in the period leading up
to the Egyptian revolt, the fifth sibylline oracle in particular perpetrates
vehement anti-Roman rhetoric and a general negative attitude to
Gentiles.373 Although this bitterness is also levelled at the Egyptians, it
is Rome that is denounced as immoral, and because it was Rome that
destroyed Jerusalem, is anathematised as the new Babylon. A saviour is
fantasised, who, at one point, will restore Jerusalem and the Temple (vs.
422). However, as Collins writes, not only will the restored Judea enjoy
peace. It will be freed from the unclean foot of the Greeks (264) and
sexual immorality (430). More clearly than in any previous document of
the Egyptian Diaspora, the exaltation of Jerusalem is accompanied by
the destruction of other nations.374
Pharisaic Judaism in its strictest form was not interested in
Gentiles, as Yahweh was regarded as the God specifically of Israel; but
this view was fundamentally surmounted by the prophetic idea of God.
[] If God is one, he is not only God of the Jews but also of the
gentiles.375 Although not actively hostile, the rabbinic reforms of the
Mishnah made clear distinctions between the Laws for Jews and the
Laws for Gentile proselytes, as we shall see below. The open hostility
towards Gentiles as expressed by the sibylline literature and the
Diasporan revolts at the turn of the first-second century, and by the
373
374
375

Collins, Jerusalem, pp. 122-128.


Collins, Jerusalem, p. 127.
Schrer2, History, vol. III.i. p. 159.

195

renewed differentiation between Jews and Gentiles set out in the


rabbinic reforms in the following decades, must have had an impact on
everyday Jewish-pagan interactions.
Therefore, in an investigation of the cult of Theos Hypsistos as a
cult with common ground between Jew and pagan, of prime importance
is

the

understanding

of

something

about

those

Gentiles

who

worshipped the Jewish God. Gentile proselytes and god-fearers have


been briefly mentioned in the previous chapter, but here their status
and position with regard to Judaism will be examined in more detail.

Proselytes in Judaism
The writings of the rabbis draw clear distinctions as to the duties and
rights of proselytes, including marriage rights (for example, female
proselytes were forbidden from marrying priests, and their daughters
could only do so if one of the parents was a Jew by birth). 376 The editors
of Schrer conclude that, the very care with which these distinctions
are drawn shows that in essentials proselytes were regarded by the
rabbis as of equal status with born Israelites in regard to duties and
rights.377 The fact that the rabbis set out the status of proselytes so
explicitly, and the testimony from Acts 2:10, referring to Jews and
proselytes from every nation gathered in Jerusalem, are evidence for
the many Gentile converts to Judaism. Yet the epigraphic material that
makes explicit reference to full proselytes is extremely limited, there
are only fifteen explicit records of full proselytes in the Jewish corpus. 378
There may be many more, not identified as such but sometimes hinted
at epigraphically by double names (a person also called a second

376

Schrer2, History, vol. III.i. p. 175, referring to mYeb. 6:5, mKid. 4:7, mBik. 1:5.
Schrer2, History, vol. III.i. p. 175-76.
378
Rutgers, L. V., Archaeological Evidence for the Interaction of Jews and Non-Jews in
Late Antiquity, AJA 96, no. 1 (Jan., 1992), p. 115: the evidence for proselytes to
Judaism in Rome itself is unfortunately scarce in the extreme.
377

196

name, often more Jewish),379 but this lack of actively described


proselytes is striking. This could either be taken to mean that, once
entered into fully, the Jewish community did not find it necessary for
proselytes to mark their status on their epitaphs; or by contrast that the
Jewish community was largely exclusive, with clear boundaries and
definitions of who was and who was not actually Jewish, and to what
extent. It may be most useful to take both of these statements as true,
and suggest therefore that when the term proselyte was used in
inscriptions, it referred specifically to a person who died while still in
the middle ground between Gentile and Jew, still occupying the liminal
position of a convert as they were gaining full Jewish status. Another
possible explanation for the rarity of proselytes in the epigraphic
evidence has been suggested: that the term may have been comparable
to the Roman status of freedman, libertus, which only lasted for the
lifetime of the freed slave; his descendants became full citizens, liberi.
Similarly the status of proselyte would therefore have been liminal in
the sense that it marked the transition of one familys status from
gentile to Jew, but lasted only for the first generation.380
Who were these liminal people, and where were they? Seven of
the explicitly recorded proselytes are from the Roman cemeteries,
dated to between the second and fourth centuries AD. The inscription
CIJ I, 37 refers to the archon of the Jews of the Subura region as a
protector of proselytes and father of widows and orphans. This implies
that proselytes within the synagogue were of a comparable social
vulnerability to widows and orphans, that all these groups required
special protection, and also that proselytes were quite numerous. The
various people across Rome who are recorded in death as proselytes
include Eirene, a three-year-old adoptive ward (qrepth) of Jews of
Israel, (CIJ I, 21); Crescens Sincerius, called by his mother Iudaeus
proselytus, in Latin (CIJ I, 68); a proselyte sister, Crysis, remembered
by her brother Mannacius (CIJ I, 222 which is also in Latin); an
379
380

Schrer2, History, vol. III.i, p. 174-75.


Mitchell, S., Pers. Comm.

197

unnamed Ioudea proselytos who is also called theosebes (CIJ I, 202); a


man probably a slave with a Greek name, Nicetas, memorialised by
his patron, Dionysias, in Latin (CIJ I, 256); Felicitas probably also a
slave, her social vulnerability particularly marked by the word
peregrina had been a proselyte for six years and was also
remembered by her patron in Latin (CIJ I, 462); and Veturia Paulla, an
86 year old woman who had been a proselyte for sixteen years but who
achieved the position of mother of the synagogues of Campus and
Volumnius.381
The epitaphs are found across the city, in the Via Portuensis, Via
Nomentana and Via Appia cemeteries. There are notably more in Latin
than in Greek, suggesting that these were Romans recorded in their
mother tongue. The two cases of family members recorded as
proselytes pose an interesting question namely, why would bloodrelated siblings or children be defined as proselytes rather than full
Jews? The epitaphs are found in the Jewish cemeteries. Do they reveal,
as Eirene does explicitly, a Jewish culture of adoption of unwanted
Gentile children, or Jewish-Gentile intermarriage, the children of which
were not defined as fully Jewish, as, for example, Timothy in Acts 16,
which would make Crysis either a half-sister or adoptive sister, and
Crescens the adopted son of his mother; or perhaps, that the families
from which these proselytes emerge were already Jewish sympathisers
i.e. god-fearers, an interpretation supported by CIJ I, 202 where the
deceased is called both theosebes and proselyte. The individuals who
became

proselytes

had

opted

for

full

conversion

to

Judaism,

presumably including circumcision for males. This interpretation


suggests the label was a badge of distinction, and the other, non
proselyte, family members responsible for the burials were happy to see
it made clear on the gravestones.382

381

See Murray, M., Playing a Jewish Game: Gentile Christian Judaizing in the First and
Second Centuries, CE, Waterloo, ON: Wilfrid Laurier University Press, 2004, p. 13.
382
Mitchell, S., Pers. Comm.

198

Aside from the much later example from the fifth century of a
proselyte in Venosa (CIJ I, 576), there are only a few other proselytes
attested in the entire Jewish epigraphic corpus of over 900 inscriptions.
One is a man from Tyre who, in the first century, was buried in an
ossuary in Jerusalem, remembered in both Greek and Hebrew (IJO 3,
Syr9). Also in Jerusalem in this period are found Ioudas (CIJ II, 1385),
Diogenes (Bagatti-Milik, 21383), and Salome (Bagatti-Milik, 31). Another
is Ariston from Apamea-on-the-Orontes, who was also memorialised on
an inscription on an ossuary from Jerusalem in the first century BC/AD.
The Greek and Aramaic inscription reads Ariston. Ariston of Apamea,
Judah the proselyte (IJO 3, Syr72). Judah was the commonest name
for proselytes,384 but it is unclear whether Ariston and Judah are the
same man, with Judah the name taken upon full conversion. It is also
possible that the inscription refers to two people, where it should be
assumed that these were not blood relations, but may again represent
adoptive family. These examples provide explicit support for Acts 2:510, that records devout Jews and proselytes from every nation under
heaven living in Jerusalem. Other proselytes are Sarra, in Cyrene
(CJZC,385 12); and an unnamed proselyte in an Aramaic inscription
from Dura Europus, dating from AD 244-5 (IJO 3, Syr84).
Eirene provides evidence for a culture of adoptive family within
Judaism. Other proselytes may have been part of sympathetic godfearing Gentile families who were thoroughly familiar with Judaism, no
doubt as a result of synagogue attendance (cf. Josephus in Contra
Apionem). This explanation implies that members of sympathetic
Gentile families that underwent the full conversion to become
proselytes did so with the full and supportive approval of their families.
This suggests that the contiguity of god-fearer families to the full Jewish
community was a crucial network tie.
383

Bagatti, B., and Milik, J. T., Gli scavi del Dominus Flevit Parte I: La Necropoli del
Periodo Romano, Tipografia dei PP. Francescani, Jerusalem, 1958.
384
See commentary in IJO 3, Syr72, p. 115.
385
Lderitz, G., Corpus jdischer Zeugnisse aus der Cyrenaika, Weisbaden: Dr. Ludwig
Reichert, 1983.

199

What is most apparent from this handful of inscriptions, however,


is just how few proselytes there are in the evidence. Given what is
known about the numbers of proselytes from the various literary
sources, it is extremely surprising to find such a small quantity attested
in the epigraphy. An explanation that might account for this is that,
once proselytes were fully incorporated into Judaism and the Jewish
social network, they were considered as Jews and as such, it was not
necessary to mark their status on their epitaph, or, as suggested above,
the status did not continue to be marked into the second generation.

God-fearers in Judaism
However, the more epigraphically common people at the edge of
Judaism are the god-fearers. They are well known from Acts, where at
13:16 Paul says to the synagogue in Antioch in Pisidia, Israelite men
and those who fear God, oi phoboumenoi ton theon, and implied at 14,
when Paul and Barnabas speak in the synagogue of Iconium and
convert a great number of both Jews and Greeks. They are markedly
not called proselytes, and are now generally understood as those people
with varying degrees of interest in Judaism or participation in the
activities of the synagogue, but who had not undergone full conversion.
They were Gentiles who attended synagogues, a formal group attached
to a Jewish community, and distinguished both from Jews and from full
proselytes.386 The terms for god-fearer are metuens in Latin387 and
theosebes, sebomenos or phoboumenos in Greek. However, again, there
are not a particularly great number attested epigraphically, considering
the evidence from the literary sources. The inscriptions seem to be in
the individuals mother tongue, as when found in Italy, they are in Latin
and when in the east, in Greek. A few of the eastern god-fearers appear
386

Schrer2, History, vol. III.i. p. 166.


The heretical Judaizing group of Caelicolae mentioned by St Augustine in Africa
may be the same as god-fearers, as this is the Latin term used to translate sebomenoi
in the Acts. See Mitchell, Hypsistos, p. 125.
387

200

to be of high-status they donate, for example, buildings, mosaic


decorations, a menorah, but many are ordinary people. That they were
integrated into the wider community is indicated by the fact they are
known to have had seats in the theatre.
In the west, there are relatively few attested god-fearers, and
they are generally from the later imperial period. Three are known from
Rome, all memorialised in Latin. Amelios Valens, from the Via Salaria
cemetery was a fifteen-year old Roman of the equestrian class (CIJ I,
5), so of a high status background. The other two are both from the Via
Appia cemetery, and so date from the second-fourth centuries AD. A
woman, Eparchia, is designated a god-fearer by the Greek term,
theosebes, (CIJ I, 228) she also has a Greek name, yet the inscription is
in Latin. She may have been an immigrant from the provinces, as
perhaps suggested by her name. Larcia Quadratillas epitaph was cut
into a previous inscribed stone; she is named as a Romanae metuens
(CIJ I, 285).
A Greek metrical epitaph from Lorium on the coast near Rome
remembers a young man, Rufinus, aged 21, a god-fearer, learned in the
holy Laws and wisdom (JIWE, 12). I have suggested in the previous
chapter that, because references to Jewish Law seem generally to be a
phenomenon of late antiquity, this inscription should be placed at the
later end of the dating scale. Rufinus is explicitly called qewsebhv";
however, the ajgivwn te novmwn have been argued to mean that he was
an exceptionally pious full Jew.388 There is no mention of any other
Jewish edifice, or any symbol or name to support this identification, as
might be expected of such an apparently pious Jew, considering the
process of Hebraization that Diaspora Jewry was undergoing at the
time. It seems from this conspicuous absence that Rufinus should be
thought of as a particularly devout young Gentile man, qewsebhv", as
described, perhaps unable to become a full convert due to the laws
against proselytising and circumcision at the time. An inscription for
388

See the brief discussion in the previous chapter, p. 157, note 347.

201

Marcus, a god-fearer in Venosa, is in Latin and dates from the fourthfifth century (JIWE, 113). The only other mention of god-fearers in the
west is a Latin epitaph dating from the third-fifth centuries 389 from Pola,
now in Croatia. It is for Aurelia Soteria, religioni(s) iud(a)eicae
metuenti, of the Jewish religion, god-fearer (CIJ I, 642). Her sons,
Aurelius Soter and Aurelius Stephanus, dedicated the stone for her, and
moreover, Aurelia Soteria herself had set up an epitaph for her fosterdaughter, Aurelia Rufina (CIJ I, 641). This was apparently an extended
family, suggesting that the Jewish familial ideology of adoption was
transferred to Gentile god-fearers.390
There

are

two

god-fearer

inscriptions

that

pose

more

interpretative difficulties. In the Black Sea region, there was a large


community of Jews at Panticipaeum by the first century AD, and a
manumission text there refers to the sunagogh`" tw`n jIoudaivwn kai; qeo;n
sebw`n, which could either be interpreted as the synagogue of the Jews

and god-fearers or as the synagogue of the Jews who are also godfearers (IJO 1, BS7). Because there is no other explicit mention of
god-fearers in the Black Sea region, it has been argued that the second
interpretation is correct,391 however, there are many inscriptions in the
region to Theos Hypsistos, and IJO 1, BS4 refers to Theos Hypsistos
and the proseuche. Should this inscription then be understood as
Jewish, or as belonging to the cult of Hypsistos as a separate entity?
The common ground is abundantly clear. Likewise, inscriptions from the
theatre in Miletus from the second-third century name the places where
Jews and god-fearers sat (IJO 2, 37; 38), but again the wording is
unclear and both the interpretations of meaning Jews who are also godfearers, or Jews and the associated group of theosebeis have been
389

The fact these people all had Aurelii names suggests that they were proud of their
Roman citizenship and indicates that these inscriptions should be dated earlier in the
third century, closer to the Constitutio Antoniniana in AD 212, as opposed to the fifth
century.
390
Supported by the inscription from the Bosporan kingdom that refers to the
adopted brothers who worship Theos Hypsistos, (96; 98; 100; 101).
391
Simon, M., Le Christianisme antique et son contexte religieux. Scripta Varia II,
Wissenschaftl. Unters. z. N.T. 23, Tbingen, 1981, Siegert, F., Gottesfrchtige und
Sympathisanten JSJ 4, 1973.

202

suggested, although it is now generally accepted that the term


theosebes refers to a Gentile god-fearer.
There are other examples of god-fearers in Asia Minor. The term
is found a few times, and all the inscriptions are in Greek. An Imperial
period epitaph from Cos is for Eirene the god-fearer (IJO 2, 6), and at
Tralles in the middle of the third century, Claudia Capitolina made a
donation of foundations and the marble cladding of the steps
presumably of the synagogue. She is described as ajxiologwtavte, which,
like lamprotatos, indicates senatorial or equestrian class. She is also
called qewsebh;" (IJO 2, 27), and as her husband is known to have been
a Roman senator and priest of Zeus Larasios, 392 she is a classic example
of a high-status woman with an interest in Judaism. An earlier but
similar case is found at Acmonia, where Iulia Severa restored the
synagogue (IJO 2, 168). She was a local high-ranking lady from the
time of Nero, and high priestess of the Imperial cult. It is apparent that
this position was entirely compatible with being involved with the
synagogue, and although not specified, it might be supposed that her
patronage of the synagogue indicates god-fearer status.
In the all-holy synagogue of the Hebrews in Philadelphia in the
third-fourth century, Eustathios, a god-fearer (theosebes), made a
donation of a basin, from himself and his sister-in-law Athanasia, for the
memory of his brother Hermophilos (IJO 2, 49). Athanasias name may
indicate a connection with Christianity.
There was a large Jewish community at Sardis, attested in the
literature from the first century BC 393 but probably established some
time

before.

The

synagogue

provides

evidence

of

god-fearers

(theosebeis) dedicating there during the fourth-sixth centuries until the


destruction of the building (IJO 2, 67; 68; 83; 84; 123; 125; 132). Two
pieces of mosaic flooring were certainly given before AD 378, by
Aurelius Eulogios (68) and Aurelius Polyippos (67). Other fourth
century dedicators include a man who simply calls himself Eutychianos
392
393

See commentary in IJO 2, 27, p. 141.


Josephus, Ant. 14.235.

203

(123) and there is a group of god-fearers who donated together (125).


Between the fourth and sixth centuries, Aurelius Hermogenes citizen of
Sardis records his donation of a menorah (132), and Leontios is
recorded twice in the fifth-sixth centuries (83; 84). These inscriptions
reveal that god-fearers were still a recognised sub-section of the Jewish
community, even as late as the destruction of the city by Persia of the
city in AD 616. They were wealthy enough to donate substantial pieces
of the synagogue building, they were citizens, and the lower-status
god-fearers clubbed together in order to display their piety.
The stele recording Jews and god-fearers in Aphrodisias published
by Reynolds and Tannenbaum in 1987 provides the conclusive evidence
for the reality of a defined category of Gentile God-fearers attached to
a Jewish community394 and was originally dated to the third century AD
(IJO 2, 14). It has recently been reappraised, divided into two separate
inscriptions, and dated much later with Chaniotis and Ameling
suggesting the fourth century for the earlier inscription and the fifth
century for the later; whereas Williams goes further and suggests the
fifth-sixth century, in particular, sometime after the plague of the
540s.395 Williams reason for dating the stele so much later than
originally suggested is because of the considerably higher proportion of
indeclinable Semitic names represented, which she argues reflects a
process of deliberate Hebraization as a response both to Christian
appropriation of Biblical names during the early Byzantine period and
the oppression of the Jews by the increasingly intolerant Christian
state.
There is therefore good epigraphic evidence for Gentiles devoted
to the synagogue. The Acts of the Apostles and the epigraphic evidence
from Panticipaeum, discussed above, show that, like proselytes, Gentile
god-fearers were part of synagogue life as early as the first century AD.
394

Schrer2, History, vol. III.i, p. 26.


See Williams, M. H., Semitic Name-Use by Jews in Roman Asia Minor and the
Dating of the Aphrodisias Stele Inscriptions, in E. Matthews, (ed.), Old and New
Worlds in Greek Onomastics, Proceedings of the British Academy, 148, Oxford: Oxford
University Press, 2007, p. 191.
395

204

However, it is notable that, although there are a couple of examples of


god-fearers that date from the first century, most of the epigraphic
evidence is considerably later, dating from the third century onwards.
Why are no god-fearers attested in the intervening period?

God-fearers and Theos Hypsistos


The link between the Gentile god-fearers and the cult of Theos
Hypsistos is made explicitly by Cyril of Alexandria, who, at the end of
the fifth century, said that worshippers of Theos Hypsistos called
themselves theosebeis, who also worshipped the Earth and Heaven,
the Sun and the Moon, and the brightest stars. Gregory of Nazianzus
made a connection between the worshippers of Hypsistos and the Jews.
Mitchell argues from these explicit literary testimonies, from the very
similar

geographical

distribution

of

Hypsistos

and

god-fearer

inscriptions, and from the inscriptions from Tanais, that record the
god-fearing

(sebomenoi)

adopted

brothers

who

worship

Theos

Hypsistos (96, 98, 100, 101), that theosebes was therefore a specific,
technical term used to describe themselves by the worshippers of Theos
Hypsistos.396
However, aside from these inscriptions from Tanais, almost all of
the god-fearer inscriptions detailed above are explicitly connected with
the synagogue or a Jewish community, and as noted, most are later than
the third century. The cult dedications to Hypsistos, by contrast, almost
never make an explicit connection with a Jewish place of worship;
indeed, there are independently known sanctuaries to Hypsistos, for
example, at the Pnyx in Athens, Oinoanda in Lycia, and at Iasos in
Caria. This suggests that the worshippers of Theos Hypsistos and those
who called themselves god-fearers, although connected, are to be
differentiated.
396

Mitchell, Hypsistos, p. 119.

205

It is here proposed that, although the cult existed beforehand, the


large body of Hypsistos inscriptions from the second-third centuries in
particular can be interpreted as representative of a sect-like Judaism
formed by Gentile god-fearers. When and why might these people have
formed or joined the cult, and moved away from the religious outskirts
of the synagogue to form a movement in its own right? Some internal
change must have facilitated this process.

Hypothesis: Necessary Activation of a Network of Believers

The epigraphic evidence for the cult of Theos Hypsistos is found across
Greece, Asia Minor and the eastern Roman Empire, with the major body
of inscriptions dating from the second to third centuries AD (see Map
6A). The worshippers range in status and occupation, there are Roman
colonists and slaves represented, and the cult was popular with women.
The appeal of the god, as noted by Mitchell, was therefore apparently
exceptionally wide ranging from peasant farmers to city dwellers,
slaves to slave-owners, men and women, and covered all sorts of
professions. This strongly suggests that the cult did not transmit across
professional groupings or guilds, i.e. that the routes of diffusion ought
to be looked for elsewhere family, friends, neighbourhood, etc. This is
in marked contrast with, for example, the transmission routes of the
cult of Jupiter Dolichenus. Since nothing really links these people in
terms of status or of occupation, then how did the cult come to be
adopted by all of them?
What these people do have in common is the name they used to
address their deity. The general lack of associated iconography
suggests that the cult of Hypsistos was aniconic; and the absence of any
real syncretism with other deities supports the notion that the cult was
206

essentially monotheistic or henotheistic. If it were the case that all the


name Hypsistos represented was a superlative given to any local god
considered to be the Most High, then it would surely be reasonable to
expect to find those deities named. With the one obvious exception of
Zeus, this is very rarely the case. Therefore, this name is better
understood to refer to one deity, a cult in common between these
divergent people in divergent places. Unlike the Jupiter Dolichenus
worshippers, there is no apparent network of occupation and status,
and unlike the Jewish Diaspora, there is no common bond of ethnicity to
link these people. The hypothesis tested here therefore takes the name
they use in common as an indication of a religious network.
The cult must have originated somewhere. The translation of the
Torah into Greek with the use of the term Theos Hypsistos to describe
the Jewish God as El Elyon, and the common ground between
Hellenistic philosophy and the philosophical aspects of Judaism coincide
with the first attestations of the term Hypsistos in Greek contexts. It
might be expected that this would have taken place in Egypt, but this
was not apparently the case, possibly because Egypt was too riven with
social tensions. Instead, it is proposed that the origins of the cult of
Hypsistos among the Greeks should be sought in Delos, as a long
established and sacred point of juncture between the Hellenistic
kingdoms, a free port and a place where cultures, ethnicities and
religions could cross-fertilise.
However, the main emphasis of the following section is on the cult
in its own right, and its development in chronological terms. It is
proposed that, as a result of the profound impact of the destruction of
the Temple in Jerusalem in AD 70 and the subsequent revolts and
decimation of the Jewish populations in Judaea, Cyrene, Egypt and
Cyprus, the situation for the Gentile god-fearers in Judaism changed. It
is hard to see how relations between Jews and Gentiles could have
continued as before during this time of great upheaval. The temporary
separation of Jew and Gentile in the second to third centuries was part
207

of

the

earliest

reactions

to

the

destruction

and

the

nascent

dissemination of the rabbinic reforms. These Gentile god-fearers, then,


without their Jewish community structure to gather in, found a place
within the existing, somewhat Judaizing cult of Theos/Zeus Hypsistos.

Epigraphic analysis

As Mitchell has already collected most of the evidence for the cult of
Hypsistos, references here are to his catalogue.397 Material published
since 1999 is included with full references. The inscriptions are largely
simple, often indicating the name of the dedicant and the name of the
deity, and most have no iconographic features. Almost all the
dedications are in Greek. The aim here is to understand the
development

of

the

cult,

so

the

approach

to

the

material

is

chronological. It is unfortunate that many of the inscriptions are only


roughly dated, but the concentration of evidence in the second and
third centuries, and its virtual disappearance in the fourth century,
serve to tell us something about the cults trajectory. The material is
organised into two main temporal stages: the earliest attestations and
the first century AD; and the period following, the second and third
centuries AD.

The earliest attestations


The term Hypsistos is first found epigraphically in the Hellenistic
period, when it is generally used to refer to Zeus, or, much more
occasionally, to an unnamed Theos. The inscriptions discussed here are
397

Mitchell, Hypsistos, pp. 128-147.

208

dated either to the Hellenistic period, understood traditionally as


meaning between Alexanders death in 323 BC and the annexation by
Rome of Greece in 146 BC, or roughly to the second-first century BC.
The inscriptions have been found in Macedonian Edessa, where
Charis Alexandros and Demetrios Charitos dedicate to Zeus Hypsistos
(41); in Odessus on the Black Sea, where the one also called Papias
built this house and the doors and dedicated them in prayer in the place
of Zeus Hypsistos (81); at Delos, where an inscription is for Zeus
Hypsistos and the unnamed gods of the dedicated altar (110a) on
Mount

Cythus,

where

there

is

an

attested

sanctuary

of

Zeus

Hypsistos;398 on the islands of Imbros, where a dedication to Zeus


Hypsistos was given by Athenaios (113); and Skiathos, where the
inscription to Zeus Hypsistos mentions the city and makes reference to
a building, perhaps a sanctuary (118); likewise at Iasos in Caria, where
two boundary stones to Zeus Hypsistos (129; 130) may have marked a
sanctuary, as a dedication simply to Hypsistos was also found (131); in
nearby Miletus there is another dedication to Zeus Hypsistos (134); and
in Mylasa in Caria, a lease document was made by the crown-wearer,
Aristeos son of Melanos son of Apollonios, priest of Zeus Hypsistos and
Agathe Tyche (137). Caria was briefly ruled by Macedonia under Philip
V c. 200 BC, which makes an important link in this period between
these two areas with attested Zeus Hypsistos worship.
Inscriptions that use Theos instead of Zeus are more rare in this
period, and when they do occur, are directly related to Jewish
communities. A straightforward use of Theos Hypsistos by explicitly
named Jews comes from Athribis in the second-first century BC: Jews
of Athribis dedicated the proseuche to Theos Hypsistos (285). An
inscription from Alexandria that, although it does not explicitly name
Jews, mentions the proseuche and so has been argued to be Jewish,
reads to Theos Hypsistos, the temple precincts and the proseuche and
the places next to it (283). Additionally, Theos Hypsistos is invoked in
398

Noy, D., Panayotov, A., Bloedhorn, H., IJO, vol. I, Eastern Europe, Tbingen: Moer
Siebeck, 2004, p. 218.

209

two virtually identical epitaphs from Rheneia, the burial island of Delos,
dated between the second and first centuries BC (IJO 1, Ach70; 71;
Mitchell, 110). The editors of IJO 1 state that the reference to Theos
Hypsistos here is clearly to the Jewish God, as indicated by the
following phrase, ton Kyrion ton pneumaton kai pasis sarkos, which is
almost an exact citation of the LXX text of Num. 16.22 and 27.16. They
find further Septuagint parallels, including a reference to the Lord who
sees everything, and the angels of god which do not presuppose a
special angelic cult. They are invoked to carry out Gods vengeance. 399
Thus all the Hellenistic attestations of Theos Hypsistos have explicit
Jewish connotations.
The fact that most of these early inscriptions identify the deity as
Zeus rather than Theos suggests primarily that the people dedicating
were Greek and Macedonian, supported by the personal names that
occur in the inscriptions, whereas the dedications to Theos Hypsistos in
this period are closely connected, sometimes explicitly, with Jewish
communities. It is expedient to recall the Letter of Aristeas, probably
composed in the mid-second century BC, where it is stated that the
Jews worship the same god that the Greeks know by the name Zeus.400
The use of the term Zeus Hypsistos began in the Greek world in
precisely the same period as the Jews were translating the Torah into
Greek and using Theos Hypsistos to describe the Jewish God. There
are also a number of dedicated sanctuaries to the god in this period;
some built structures and others apparently open-air spaces. This
variance within the space for veneration of the god reveals that the
worshippers were not following a set paradigm, but, rather, adapted to
the situations they found themselves in. Certainly at this time there was
no definitive understanding of how the space Diaspora Jews should
gather and worship in should be construed.
399

IJO 1, p. 238-239.
See Bartlett, J. R., Jews in the Hellenistic World. Josephus, Aristeas, the Sibylline
Oracles, Eupolemus, Cambridge: Cambridge University Press, 1985, p. 14; Collins,
Jerusalem, p. 179.
400

210

However, during the first century BC, the distinction that might
be drawn between Jewish or Greek use of the term Hypsistos
becomes less apparent, and Zeus and Theos become more contiguous.
This supports Mitchells point that, instead of assuming that the
inscriptions need to be sorted into Jewish and pagan groups we should
try to see if they make sense as a single body of material, treated on its
own terms.401 Two inscriptions from the Fayoum in Egypt demonstrate
this quite clearly: the first, dating from 69-57 BC, mentions a religious
guild of Zeus Hypsistos worshippers (287), and the second, dated to 29
BC, is to the Theos Megas Megas Hypsistos on behalf of Epitychia also
called Dionysia, and her husband Harpochras (286). Epitychia and
Harpochras might well have been Jews, and Jews were certainly
present in the Fayoum during the second century BC, as Eleazar and his
wife Eirene set up a stele during this period (JIGRE, 115). However,
this shows the difficulty in making a distinction between the Jewish
and the Greek cults, and that often, distinguishing them in this way is
based on arbitrary interpretative decisions.
Elsewhere in the Mediterranean, Zeus and Theos are both used.
In Macedonia, the dedications continue to be to Zeus. At Antigoneia,
Quintus Markios Noumerios, presumably a Roman colonist, dedicated a
statue to Zeus Hypsistos (SEG XLVI, 726), and at Edessa, Zoilos
Alexandros dedicated to Zeus Hypsistos for his children (38). At Azoros
Elassonas in Thessaly, by contrast, Hermokles gave a stele decorated
with an eagle simply to Hypsistos (SEG XLVI, 640). In Spitali on
Cyprus, Aristokles Koukomes offered up his prayer to Theos Hypsistos
(265), and on Delos, Laodice and Lysimachus gave two thanksgiving
inscriptions

to

Theos

Hypsistos

(respectively

107;

108).

The

dedications of Laodice and Lysimachus were found in a building on


Delos, GD80, tentatively, but by no means certainly, identified as a
Jewish synagogue building by the contemporary offering from there by
Lysimachus and Agathocles (IJO
401

Mitchell, Hypsistos, p. 100.

211

1,

Ach65) that mentions the

proseuche. It has been argued from this that Laodice and Lysimachus
must therefore have been ethnically Jewish, but Mitchell makes the
point that although GD80 may have been a synagogue of the Jews, it is
also true that the sanctuary is also a Greek one, containing dedications
set up by persons with Greek names to Theos Hypsistos. 402 What is
apparent from the debate over the synagogue and the Hypsistos
dedications on Delos is that by this stage, the use of Theos Hypsistos in
a place where there were also known to be Jews does not automatically
mean that the dedication should be classified as Jewish.
The cultic divide is, however, quite apparent between inscriptions
that, although not necessarily made by Jews, can be closely associated
with

Jewish

populations,

i.e.

the

Egyptian,

Cypriot

and

Delian

inscriptions, and those that cannot. The former group almost all use
Theos. The dedications from Macedonia, Caria, and the Aegean islands,
aside from Delos, all use Zeus. There are no explicitly known Jewish
communities in these places at this time, although there are established
synagogues in Macedonia by the time of Paul, and there were Jews in
various cities in Caria by the first-second century AD. This apparent
lack of connection with Jewish communities has been taken as proof
that the Hypsistos dedications do not represent a single cult, and
instead that it was a term applied to any god considered to be highest.
However, because the dedications are very simple, and at this stage
there are still very few of them, it is difficult to make generalisations.
Moreover, the features that connect them the lack of any other named
deities and the use of the term also applied to the Jewish God are
notable. It might be more useful to consider that what this divide
represents is that those worshippers of Hypsistos who were regularly
exposed to Jewish practice and communities termed the deity Theos,
perhaps that they considered themselves to be god-fearers; whereas
those worshippers whose local Jewish community was not large enough
to sustain a synagogue, or those who had only briefly encountered the
402

Mitchell, Hypsistos, p. 98.

212

god perhaps in a foreign location or second-hand referred to


Hypsistos by the name of a god they recognised: Zeus.

Digression: Gerizim, Delos and the early Hypsistos dedications


Two places where the earliest Hypsistos inscriptions are found are also
marked by the explicit presence of Samaritans: Caria, and most notably,
Delos. Evidence concerning the ancient Samaritans is extremely
limited, with a few epigraphic attestations and some mentions in Jewish
literature. The Samaritans were a schismatic Jewish sect that claimed it
was Mount Gerizim in Samaria, rather than Mount Zion in Jerusalem,
that God had intended as the site for his one Temple. They built their
alternative temple there, revealed

by archaeological excavations

conducted during the 1960s and dated to between 300-124 BC. 403
During the later Hellenistic period, this temple was destroyed:
according to Josephus, in one of two expeditions by John Hyrcanus
against Samaria, either c. 128 BC or 108 BC. 404 It is difficult to
ascertain the relationship between Jews and Samaritans: it has been
argued variously that the divide and animosity between them stemmed
from the post-exilic period, or from the Hellenistic conquest, when
Alexander sanctioned the building of the Samaritan temple, or, as
Crown argues, that Samaritans were basically considered as Jews until
the period following Bar Kokhba.405 There was, however, certainly
hostility between them prior to this period, and the Samaritans
apparently rejoiced at the destruction of the Jerusalem Temple in AD
70.406

403

Bull, R. J., Wright, G. E., Newly Discovered Temples on Mt. Gerizim in Jordan,
HThR 58, 1965, pp. 234-237.
404
Josephus, Ant. XIII, 10.
405
Crown, A. D., Redating the Schism between the Judaeans and the Samaritans,
JQR, New Series, vol. 82, no. 1-2, pp. 17-50, 1991.
406
Gaster, M., The Samaritans: Their History, Doctrines and Literature, Oxford: Oxford
University Press, 1923, p. 37.

213

What is important here is that, following the Bar Kokhba revolt


and the destruction of Judaea, Hadrian built a temple to Zeus Hypsistos
on Mount Gerizim, the remains of which were discovered in the
excavations of Bull and Wright. It seems that Hadrian, who also enacted
anti-Jewish legislation, rebuilt the temple of their rivals with the
express purpose of further humiliating the Jews of Judaea, so
contributing to the division between Jew and Samaritan. The epithet
used for Zeus is the particularly noteworthy aspect of this temple. 2
Maccabees records the suppression of Judaism in the Hellenistic
period; 6:1-2 recalls that the king sent an Athenian senator to compel
the Jews to forsake the laws of their ancestors and no longer to live by
the laws of God; also to pollute the temple in Jerusalem and to call it the
temple of Olympian Zeus, and to call the one in Gerizim the temple of
Zeus-the-Friend-of-Strangers, as did the people who lived in that place
(my

emphasis).

Montgomery407

noted

that

Eusebius408

presents

extensive extracts of Eupolemus, who wrote at the time of Alexander. In


sections 17 and 18, in a Midrash on the life of Abraham, Gerizim is
referred to as ojvro" uJyivstou, the Mount of the Most High. Josephus
records that the Samaritans petitioned Antiochus Epiphanes with the
request that they should be treated differently from the Jews, and that
their temple, which at present has no name at all, should be called the
Temple of Jupiter Hellenios.409 Of course these accounts are shot
through with anti-Samaritan bias, but what they go some way to
demonstrating is that, in the Hellenistic period, it is not implausible to
imagine that the name Zeus and the epithet Hypsistos were associated
with the Samaritan God, and perhaps even that some Samaritans
themselves addressed him by this name. The temple that Hadrian built
to Zeus Hypsistos should not, therefore, be thought an entirely new and
baseless foundation. Moreover, it was not necessarily an act that would
407

Montgomery, J. A., The Samaritans: the earliest Jewish sect, their history, theology
and literature, New York: KTAV, 1968, p. 284.
408
Eusebius, Praep. Evang. IX, 17, 18, 26, 30-34, 39.
409
Josephus, Ant. xii 5.5.

214

have displeased the Samaritans themselves, 410 rather, that they may
well have viewed it as a pious act of rebuilding. Hadrian was making a
political statement as much as a religious one: by rebuilding the temple
on Gerizim under the name Zeus Hypsistos, he managed both to further
humiliate the Jews in Judaea, and at the same time, acknowledge the
differences between the Jews and the Samaritans and sanction
Samaritan worship.
In the Hellenistic period, inscriptions testify to Samaritan
populations in the Aegean. In Athens, Ergasion the Samaritan is
mentioned as part of a thiasos inscription, dating to the fourth-third
century (IJO 1, Ach41), he was clearly engaged with Gentile activities.
From the acropolis at Caunos, an immigrant family originally from
Shechem (Neapolis, modern Nablus, the Samaritan city in the foothills
of Mount Gerizim) is attested in the late Hellenistic period (IJO 2, 24).
Crown argues that, after the destruction of Shechem in the second
century BC, the Samaritan Diaspora grew rapidly and should be
considered as having the same range as that of the Jews: through
Egypt, Greece, North Africa, Italy, Sicily and south into the lands
adjacent to the Red Sea.411 This may be the case, but there is little
supporting

epigraphic

evidence,

partly

because

in

general,

no

distinction can be made between Jewish and Samaritan inscriptions of


this period.
Delos is the exception however, where, as we have seen above,
there was also a mountain sanctuary to Zeus Hypsistos. The earliest
inscription dates to between 250-175 BC, and reads: The Israelites [on
Delos?] who make offerings to the temple (on the) holy Garizim
[Argarizin] honoured Menippus (son) of Artemidorus, from Heraclion,
himself and his descendants, for constructing and dedicating from his
own funds in a prayer [=vow] of God and crowned with a golden
wreath and (IJO 1, Ach66).
410

However, it is also true that the Samaritans had been adversely affected by
Hadrians anti-Jewish law forbidding circumcision.
411
Crown, Samaritans, p. 25.

215

A very similar inscription was found with Ach66. This records


that the Israelites of Delos who make offerings to the temple (on)
Garizim crown with a golden wreath Serapion (son) of Jason, from
Knossos, for his beneficence to them (IJO 1, Ach67). This inscription
has been dated on palaeographic grounds to 150-50 BC, but may be
closer in period to Ach66.
It is clear then, that, like the Jews, Samaritans in the Diaspora
paid an annual tax to their temple on Gerizim, and that a Samaritan
building of some sort existed on Delos. Although the word proseuche is
used here, it is apparent from the grammar that it means a vow to God
rather than a prayer-hall. The highly distinguished donors, Menippus
and Serapion, were not necessarily Samaritans; the names are
otherwise unattested in Samaritan or Jewish inscriptions. Because of
the reference to Knossos, the editors of IJO think that Heraclion is most
likely a reference to Heraclion/Heraclea in Crete, but also consider the
possibility that this refers to another Heraclea, perhaps Heraclea under
Latmus, which would make another link to Caria.
The editors of IJO suggest that, because this is the first
attestation of the term Israelite to denote a Samaritan, this was
specifically chosen to distinguish the Samaritans from the Ioudaioi on
the island, although alternatively it has been suggested to be a
reference to Samaritans as originating from the north part of Israel, as
opposed to Judaea.412 It is perhaps relevant that Ioudaioi are never
explicitly mentioned in inscriptions from Delos, although they are
known to have lived there from various other sources.413
What these two inscriptions reveal is that non-Samaritan men
from the wider Aegean context were sufficiently connected with the
Samaritan community on Delos to make donations substantial enough
to merit their crowning with the most expensive and highly regarded
golden wreaths, and the recording of these acts on stone for posterity;
412

See the argument presented in Kraabel, A. T., New evidence of the Samaritan
Diaspora has been found on Delos, in BA, vol. 47, no. 1, 1984, pp. 44-46.
413
For example, 1. Macc. 15.23 lists Delos as a place where the Romans sent a letter
about the Jews in 140 BC.

216

and

that

the

Samaritan

community

on

Delos

had

pan-Aegean

connections. More importantly, these dedications highlight the unique


position of Delos as a place where cultures and ethnicities could
intermingle and exchange.414 The islands status during the Hellenistic
period as a free port facilitated trade, especially the trade in slaves. 415
People from across the Mediterranean world interacted and grew
wealthy here. It might be hypothesised that the Samaritan community
on Delos introduced the cult of Hypsistos into the Aegean, perhaps
worshipping alongside Greeks in the sanctuary to Zeus Hypsistos on
Mount Cythus. It may have been an international sanctuary. This
Jewish sect had already acknowledged that their God, friendly to
strangers, could be equated with the Greek Zeus, and moreover, their
later-rebuilt temple in Samaria was dedicated to Zeus Hypsistos. They,
like the Jews, honoured non-Samaritan donors and welcomed Gentile
worshippers. Perhaps it could be surmised that people who had
encountered the cult in their dealings with the Samaritans on Delos
transmitted the worship of Zeus Hypsistos back to the important and
predominantly coastal places from whence they came.

The Cult of Hypsistos in the Imperial period: The First Century


AD
In the Hellenistic period and the first century BC, dedications to either
Zeus or Theos Hypsistos are relatively rare. During the first-third
centuries AD, the instances of the term Hypsistos increases massively.
However, a survey of the material dated to the first century reveals it to
still be relatively small, and still quite varied. In some areas, the cultic
414

This had long been the case Susan and Andrew Sherratt remark on the growth, in
the eighth century BC, of international sanctuaries at nodal points on the maritime
routes, of which Delos was one. Sherratt, S., and Sherratt, A., The Mediterranean
economy in the early first millennium BC, in WA, vol. 24, no. 3, Ancient Trade: New
Perspectives, 1993, p. 367.
415
Noy et al, IJO 1, pp. 210-219.

217

character seems relatively pagan, but elsewhere it could easily be


described as Jewish or at least, Judaizing. The dedications are found
across the Mediterranean from Gorgippia on the Black Sea to
Chersonesos on Crete. Although there are still only a few inscriptions
from this period, some notable conclusions can be drawn from them.
Many of the dedications are by Roman citizens. Two of these are
from Mytilene: first, a dedication decorated with an eagle and an olive
wreath to Theos Hypsistos by Marcus Pompios Lukaos with his wife
Phoebe (115); and second, by a Roman family: Gaius Cornelius
Christios, Cornelia Phallousa, and Gaius Cornelius Secundus, wintering
in the islands, in thanks to Theos Hypsistos (116). This family was
apparently en route, perhaps to Asia Minor. Although no Jewish
community is mentioned in Mytilene, Paul crossed over to Asia Minor
from Macedonia via the city in this period (Acts 20:14). Across the
water in Miletopolis in Mysia, Tiberius Claudius Syntrophos set up a
dedication at the command of Zeus Hypsistos Brontaios (185), the stele
shows a figure of Zeus holding a thunderbolt, depicted beside an altar,
a herm, and a female figure lying on the ground.
Two dedications from Thrace provide evidence for the contiguity
of the names Zeus and Theos: the first, dating from AD 25, is a
dedication from Selymbria by Gaius Julius Proklos in thanksgiving to
the holy Theos Hypsistos, on behalf of Rhoemetalkes and Pythodoros for
saving them from the dangers of the war of Koilalitikon (68); the
second is from Kavalla, dated to 36-48 AD, and was dedicated to Zeus
Hypsistos by the superintendent and workmen at the quarries, on
behalf of the King of Thrace, Rhoemetalkes, son of Cotys and his
children (60). Jews are attested epigraphically in Thrace in the Imperial
period.
An inscription dated to AD 51 from Edessa is to Zeus Hypsistos
for the salvation of Marcus Vibius Ambova, which then lists a group of
men

and

one

woman,

who

are

variously

Romans

and

Greeks/Macedonians: C. Pontius Torquatus, P. Vettius Narcissus, L.


218

Liburnius Chrysippos, G. Flavius Alypos, Secundus son of Adymos,


Melete daughter of Apollodoros, Apollonides son of Theudas, M. Vibius
Hermeros, Epaphras son of Damothares, M. Antonios Moustios Crispus,
set up when M. Attius Longus... was priest (SEG XLVI, 744). These
men may represent a group of Zeus Hypsistos worshippers; they are
notably composed of Roman citizens and presumably, members of the
local elite.
In

Thessalonica,

Titus

Flavius

Euktimenos,

trikleinarchos,

president of a dining-group, dedicated to Theos Hypsistos (56);


suggesting that here at least, the cult followed traditional Greek
parameters. In AD 74-5, Gaius Julius Orios was instructed in a dream to
make his dedication to Theos Hypsistos the great saviour, as a
thanksgiving for making him money, and saving him from the great
danger of the sea (55). He may not have been a merchant, but was
certainly involved with maritime activity in some way. The inscription
also records the priest Marcus Vetius Proclus, suggesting that the cult
had a body of religious officials. It might be interpreted to mean that
Marcus Vetius the priest was a Jewish official in the synagogue, as there
was a synagogue in Thessalonica with attached sebomenon Hellenon by
the time of Paul, however, the attestation of a priest of the cult of Zeus
Hypsistos at Edessa could also support the identification of this priest
as an official of the cult of Hypsistos as a separate entity.
These

higher

status

inscriptions

are

concentrated

in

the

northwest Aegean, where it is notable that the cult is mainly that of


Zeus as opposed to Theos. Lower status dedications from the first
century

are

by

contrast

rare,

probably

partly

because

these

worshippers could not afford dedications that survive. However, since


there are some lower-status inscriptions in the corpus, this may to some
extent also reflect the demographic profile of the cult followers. They
are found more across Asia Minor and the southern Aegean, and
notably, most of them use Theos rather than Zeus. A dedication was
given by Tation, to Helios Theos Hypsistos in Pergamon (186), which
219

connects the god with the sun; and in Prusa in Bithynia, a simple
collective prayer was offered to Zeus Hypsistos by the entire village
because of a gift of ten measures of corn given to them by Paterion (?)
(189). In Chersonesos in Crete, Tertula offered her prayer to Theos
Hypsistos, decorated with an eagle (121); almost identical is the
dedication of Sozousa on Cyprus (262); and that of Cosmos in Athens,
which is probably first century, and which is decorated with a picture of
a torso (22).416 It is notable that most of these places that use the term
Theos have attested Jewish communities at this time.
Finally, there are the manumission texts from the Black Sea.
Three inscriptions from Gorgippia are dated to the second half of the
first century, and all use Theos Hypsistos in the manumission formula
(IJO 1, BS20; 21; 22). Another inscription that follows a similar
formula and so has been restored to read Theos Hypsistos is a
thanksgiving plaque for Metrotimus for salvation from megalon
kindunon (IJO 1, BS27). In two of the three manumission texts, Theos
Hypsistos appears alongside a call for the protection of the freed slave
by Zeus, Earth and Sun (IJO 1, BS20; 22). This final call has caused
some commentators to claim that therefore they cannot be Jewish; yet
conversely, some have defined them so because all three also use the
term blessed, eulogetos, regarded as specifically Jewish at this date. In
addition, BS20 mentions the proseuche, used to support some
commentators argument that it is a fully Jewish inscription. The editors
of the IJO explain the situation by arguing that the manumittor was
Jewish and the pagan deities were included for the benefit of the freed
slave.417 This may be the case, as these inscriptions do not mention the
protection of the Jewish community, as found in examples from
Panticipaeum (BS5; 6; 9) or Phanagoria (BS18).418 However, IJO 1,
416

There is a strong association with healing in the cult in Athens, as seen in the
dedications of the next century, which often display body parts.
417
IJO 1, p. 307.
418
These three deities do have an abstract quality however, and that Helios has been
found elsewhere in the corpus suggests that perhaps this is representative of a
traditional formula being used alongside Hypsistos that is not at odds with the cult.

220

BS21, which follows exactly the same formula, contains no mention of


the proseuche to argue for its being Jewish, and no mention of pagan
deities to argue against it. Although, as the editors note, Kittel pointed
out the illogicality of accepting one inscription as Jewish and not the
other,419 what it shows is the tenuousness of some of the identifications
and the common terminology existing between these supposedly
separate definitions of pagan or Jewish.
These

inscriptions

therefore

particularly

highlight

the

methodological difficulties in dividing Jewish from pagan inscriptions


to Theos Hypsistos, as there is considerable evidence in the Black Sea
region for Jewish communities and the pervasiveness of Jewish religious
ideas and terminology. Instead of attempting to divide the Hypsistos
dedications

into

different

inscriptions

should

be

ethno-religious

regarded

as

boxes,

revealing

the
the

Gorgippia
pan-Empire

phenomenon of Gentile involvement in Judaism. Another example from


the area that makes the common ground very clear is the manumission
of a slave woman, Elpis, in Panticipaeum (IJO 1, BS7). She is released,
except from a bond of service to the proseuche, and this is done with
the guardianship of the community of Jews and god-fearers.420 As we
have seen, it is now the standard view among scholars that the godfearers were non-Jewish sympathizers with Jewish beliefs, Gentile
attendees of Jewish synagogues.421 Why would the language godfearers employed in epigraphic dedications be any different to that of
their ethnically Jewish co-worshippers? In other words, it is more and
more difficult to separate people in the epigraphic evidence.
It is these inscriptions from Gorgippia that truly mark the
profound connection of the cult term with Judaism at this time, and that
Theos Hypsistos should be closely associated with Jewish communities
419

IJO 1, p. 310.
There is a slight grammatical issue with the text; leading to the suggestion that
sebwn is a participle referring back to the slave. According to the editors of the IJO,
this seems a very forced explanation (p. 282), and in the light of the term being
found elsewhere in connection with Jewish communities, it is more reasonable to
assume that it refers to god-fearers as an integrated part of the Jewish community.
421
Mitchell, Hypsistos, p. 120-121.
420

221

and Gentile involvement with the Jewish synagogue. However, although


this is the case in the Black Sea region, discerning this trend explicitly
elsewhere is not easy. As we have seen, the Hypsistos cult is found
particularly in Macedonia, Thrace and northern Asia Minor, and seems
to be particularly popular with the Roman colonist elite there. The Acts
of the Apostles testifies to the many large Jewish communities with
attached Gentile god-fearers in Thessalonica, Beroia, and Philippi
during the first century, and it is well-known that Judaism had attracted
high-status individuals in Rome itself at this time in particular, Neros
wife Poppaea and Domitians cousin.422 Although many of these high
status inscriptions are to Zeus, some are to Theos. The pocket of cult in
Macedonia might be taken as indicative of the elite being attracted to
abstract monotheism, whether explicitly related to Judaism or not. It is,
however, possibly more useful to regard the phenomenon as the cult
spreading through this particular social network of Roman colonists
people who travelled, were outside their usual social networks, who
were far from home and so also in the associated local elites. With
regard to the lower status dedications that predominantly use Theos,
nearby Jewish communities can often be identified. There were Jews in
Pergamon at the time of Cicero, and Josephus refers to JewishPergamene friendship;423 there were Jews long-attested in Athens, Crete
and Cyprus;424 and although there are no Jews specifically in Prusa,
there were Jews in other cities in Bithynia at this time.425
What is also apparent is that even though the use of the term
Hypsistos can often be closely associated with Judaism, there was space
for individual understanding of the deity. By this time, Theos appears as
regularly as Zeus, often contiguously, and there is as such little
distinction that can be drawn between them to claim they represent
different cults. What the choice of name instead should be taken to
422

See Chapter 5, p. 140.


Josephus, Ant. xiv 10, 22 (247-55).
424
Epigraphic evidence for Jews in Athens and Cyprus dates from the fourth century
BC; there were certainly also Jews on Crete, testified to by I Mac. 15:23 and Josephus
in numerous places. See Schrer2, History, vol. III.i p. 69.
425
See Schrer2, History, vol. III.i, p. 36.
423

222

reflect is the background of the worshippers. They used the familiar


name of the highest God in the Greek pantheon to describe a less
familiar deity, an interpretation supported by the association of both
names with the image of the eagle. As Mitchell writes, the worshippers
chose a designation that would have seemed self-evidently correct and
appropriate, and which tells us something about their personal religious
convictions.426 This reflection of personal religious belief is also seen in
association of the god with Helios in Pergamon, or even perhaps the
inclusion of Zeus, Earth and Sun in Gorgippia.

The second-third centuries AD


The number of inscriptions to Hypsistos, and Theos Hypsistos in
particular, increases dramatically during the second to third centuries.
Mitchell has argued that this is a result of the epigraphic habit, and
should not be thought of as reflecting a rise in popularity: the fact that
the great majority of the inscriptions for Theos Hypsistos belong to the
later imperial period cannot be taken as an indication that the cult was
first introduced, or even became more popular then. 427 It is very likely
that this explanation does largely account for the sudden boom in
inscriptions

to the deity in this

period.

However, it

is worth

investigating the material further to discern whether the epigraphic


habit also disguises a deeper change. There were, by the first century
AD, Jewish Diaspora communities with attached god-fearers throughout
most of Greece, Macedonia and Asia Minor, testified to by inscriptions,
commentators at the time, and the Acts of the Apostles. There is,
however, no sudden increase in dedications made by god-fearers at this
time, which would be expected if the epigraphic habit were simply the
explanation for the increase in Hypsistos dedications.

426
427

Mitchell, Hypsistos, p. 102.


Mitchell, Hypsistos, p. 108.

223

Gentile god-fearers, the destruction of the Temple, & Bar


Kokhba

Why then, are the god-fearers, so present in the Acts of the Apostles
and the literary evidence, so absent in the epigraphy, given the
epigraphic habit? As shown above, there are relatively few god-fearers
known from the epigraphic corpus. Yet the god-fearers have been
argued to have the same beliefs as the worshippers of Theos Hypsistos,
and indeed, to be the same group of people. The inscriptions from the
Black Sea region in the first century AD clearly show the connection
between Judaism, god-fearers and the worship of Theos Hypsistos. It is
suggested here then, that alongside the epigraphic habit, an additional
reason for the massive increase in dedications to Theos Hypsistos at
this time was because the Gentile god-fearers who had worshipped him
as attendees of the Jewish synagogue had been activated, and were
not able or did not wish to call themselves god-fearers in this period.
This problem of self-definition for the Gentile worshippers of the
Jewish God stemmed from the profound changes enacted in Judaism at
this time. The psychological trauma for Jews following the destruction
of the Temple in AD 70 and the events of the following sixty years
cannot be underestimated. The imposition of the increased fiscus
Judaicus post-AD 70 further fuelled the pre-existing social tensions that
led to the uprisings in Egypt, Cyrenaica, Cyprus and Mesopotamia, and
the subsequent destruction of the Jewish communities in those places.
The added insult of the foundation of Aelia Capitolina on the ruins of
Jerusalem led to the revolt of Simon Bar Kokhba and his followers and
their final defeat in 135. Judaea was practically a desert; and to ensure
Jerusalems pagan character, Hadrian drove out those Jews that
remained and forbade all Jews from entering Jerusalem on pain of
224

death, a law that remained in place until at least the fourth century,
when Constantine allowed them to mourn the destruction of Jerusalem
once a year.428 Without the physical Temple at its core, and with no hope
of Jewish political independence, the strictures and Laws of Judaism
became a spiritual Temple, reconstituted and reiterated by the rabbis in
Judaea and Babylon. The composition of the Mishnah c. AD 200 and the
renewed emphases of rabbinic Judaism on adherence to the Law
resulted in the dissemination of these reforms, and, as argued in the
previous chapter, the period between the third and fifth centuries was
marked by the Hebraization of the communities in the Diaspora.
Prior to this, the destruction in Judaea produced a period of
Jewish anxiety and anger, manifest in the Diasporan revolts. Equally,
the Jewish uprisings and revolts across the Roman Empire led to
increased Roman stigmatisation of the Jewish communities. Both Jewish
unrest and Roman suspicion led to a level of mutual distrust and
separation, with the effect of the tightening of the boundaries of
Judaism. That the iteration of explicit Jewish identity is so clearly visible
epigraphically in the process of Hebraization leads to the question: how
did the Gentiles attached to the synagogues, the non-Jews who
worshipped the Jewish God, react? It is here suggested that this period
between the second and fourth centuries, when the hostilities and
tension between Jewish communities and the Roman government was
at its height, was marked by the reiteration of Jewish Law. The social
divisions

already

in

existence

between

Jew

and

Gentile

were

heightened, and consequentially, Gentile god-fearers would have been


excluded to a degree from participation in Jewish worship and life, and,
moreover, may have harboured misgivings themselves. With access to
the God they believed in either denied or disapproved of, these large
communities of Gentiles could react in one of two ways: by either
finding a place in the swelling ranks of the Christians, or alternatively,
by joining the partially formed and quite widespread Judaizing cult and
428

Schrer2, History, vol. I, p. 553-557.

225

continue their worship under the Jewish name for God Theos
Hypsistos.
The epigraphy is, however, notoriously uninformative. In this
sense, the evidence that supports this hypothesis is, to an extent,
negative. Nevertheless, certain conclusions can be drawn from an
examination of the epigraphy. The range of people, occupations and
statuses suggest that the cult was not centralised in any way, but
rather, was a ground-level, grass-roots kind of movement that did not
transmit via occupational, ethnic or elite networks, which supports the
notion that this was a religious network, perhaps already fully or
partially formed. The outdoor/extra-mural or reused places where
people gathered indicates that there may have been no physical space
for the worshippers in the cities, suggesting a less formalised sacral
arrangement without specific hierarchies a consequence, perhaps, of
cult followers rejecting the traditional space for worship, or indicating
that they had themselves been rejected. Finally, the geographical
restrictions on the spread of the cult suggest that it lacked long-range,
independent diffusive appeal, and had arisen at a more local level as a
response to localised actions, interactions or exposure.
This hypothesis is not, of course, intended to apply to the entirety
of the inscriptions that refer to Hypsistos it has been amply
demonstrated that, even if some explicit connections can be made with
Judaism, this is by no means applicable to all of the evidence all of the
time. Instead, it is suggested here that the massive increase in
dedications, in particular those to Theos Hypsistos in the second-third
century, may represent some kind of epigraphic trace of those Gentile
god-fearers that had previously been part of the Jewish communities.

People, status & occupation

226

The snippets of information in the epigraphic record about peoples


occupation or status reveal the cults extraordinarily wide social range.
That the cult had long been popular with the Roman colonists and local
elites in Macedonia has been seen above, and further evidence for high
status individuals is found in the later epigraphy: P. Aelios Arrianus
Alexandros, a Roman councillor of the Dacian colony of Sarmizegetusa
dedicated to Theos Hypsistos in Mytilene (117), making an explicit link
between

these

places;

Ulpius

Carpos,

town

councillor,

prophet

(profhvth") of the most holy Theos Hypsistos who was honoured by the
statio of the civic gardeners is attested in Miletus (135; 136); and
Statius Rufinus, a centurion (eJkatovnarcho") dedicated to Theos
Hypsistos Epekoos in Nicomedia (192). Servants and slaves are also
found in the corpus: in Beroia, the servants of Eros son of Eubiotos
dedicate to Zeus Hypsistos (34); in Sparta the slave of Claudius
Pratolaos offered to Zeus Hypsistos (25); and Koronos, a publicus, civic
slave, dedicated to Theos Hypsistos in Knossos (120).
The occasional mentions of occupation in the later epigraphy
testify to people of both lower and middling status. In Kavalla in Thrace,
Tarsas the blacksmith dedicated simply to Hypsistos (61); in Acmonia,
Aurelius Tatis Onesimos the bronze-smith dedicated with his wife from
his own means to Theos Hypsistos (205); Theon the builder dedicated
to Theos Hypsistos in Kition (245); in Nacolea, Gaius son of Manes
dedicated to Hypsistos for saving his cattle (219); in Apamea, Aurelius
Paulus, also called Epithymitos, a doctor, dedicated to Theos Hypsistos
(214); in Miletus, there are shell-fishers who dedicate to the aJgiotavto"
Theos Hypsistos (136); and in Gortyn, Euphranor the flute-player and
Zosimos the goldsmith both dedicate to Theos Hypsistos (122; 123). A
man from Kozani in Macedonia, Chryseros Philippos, dedicated vines
and trees to kyrios Zeus Hypsistos (46) which suggests he may have
been a farmer, but also that he was probably a land-owner. The

227

dedication of vines may indicate knowledge of the golden vine in the


Jerusalem Temple, as mentioned by Josephus.429
The cult was popular with women, with dedications made by
women accounting for roughly 1/6 of the 336 known inscriptions. In
some places, for example, the Pnyx, most of the dedications are by
women. Female dedicants are also found across the Greek islands and
mainland, Macedonia, Dacia, Crete and Cyprus, Rome, Asia Minor, and
Phoenicia.430 They sometimes dedicate alone, sometimes for or with a
family member. They are often concerned with healing or at least, in
certain places the god was associated with healing, 431 and this power
seems to have attracted more than the average numbers of female
dedicants. The adoption of the cult by women presents an interesting
case how did the cult reach them? Their husbands or family may have
exposed them to it, but equally, the role of cultic transmission within
female contexts should not be overlooked. The high number of women
involved in the cult also links it with the demographic profile of Judaism
and Christianity, both known to have been particularly attractive to
women.432 The family values espoused by both Judaism and Christianity
favoured women and, moreover, women were able to hold positions of
authority within Jewish and Christian communities. Although there are
no immediate indicators within the present corpus of a similar situation
in the Hypsistos cult, there are only very few indications of any kind as
to the cults internal religious structure.
429

Josephus, Ant. XV, xi, 3.


Dedications made by women include: Athens: (2; 3; 6-10; 12; 14; 15; 21); Delos
(107; 109); Sparta (28); Beroia (37); Perinthus (63); Philippopolis (67); Serdica (71);
Sarmizegetusa (78); Chersonesos, Crete (121); Rome (125); Stratonicaea in Caria
(144; 156); Ephesus (159); Saittai, Lydia (172); Thyateira (177; SEG XLIX, 1709);
Pergamum (187); Sebastopolis (193); Synaus (223); Seluecia ad Calycadnus (239);
Sinope (SEG LII, 1240); Pessinus, Galatia (SEG XLVI, 1703); Smyrna (162); Golgi
(257; 259); Mathikoloni, Cyprus (260); Paphos (262), and Byblos (268).
431
That the Jewish God was associated with the power to heal was widely believed in
antiquity, and Jewish holy names are found frequently in amulets and other healing or
protective devices. See Liebeschuetz, W., The influence of Judaism among non-Jews in
the Imperial Period, JJS vol. LII, no. 2, 2001, p. 248.
432
Josephus records the people of Damascus who wished to eliminate the local Jews,
but feared their wives, all of whom, with a few exceptions, had gone over to the
Jewish religion. Jewish War 2.560-1.
430

228

For example, there is the leader of a dining club in Thessalonica,


mentioned above (56), and a few dedications by priests. Although there
is no reason to suppose that the cult did not have priests in the usual
way, some have argued that the priests from the corpus should be
viewed as priests of different deities. The only instance where this
seems to be the case is the dedication to Theos Hypsistos at Andeda in
Pisidia by a priest of Mn Ouranios, a native Anatolian moon-god (228).
It has been argued that this represents either a dedication to Theos
Hypsistos by a priest of Mn (Mitchell), or alternatively that this is
evidence of the name Theos Hypsistos being used to describe Mn
Ouranios himself (Belayche). The rest of the inscriptions that mention
priests give no indication that they are not priests of Hypsistos himself:
in AD 100, Tryphon to Theos Hypsistos in Paphos (261); by an unnamed
man to Theos Hypsistos in Serdica (72); Hermogenous to Theos
Hypsistos in Pirot, Serbia (75); Diophantos Akiamos to Theos Hypsistos
the great god in Philadelphia (171); in AD 172, Neikiphoros son of
Hermokrates to Theos Hypsistos in Thyaera in Lydia (175); Gaius
Olympios Paulos to Zeus Hypsistos in Dion (SEG LIII, 598); in Beroia,
two Roman citizens, P. Cornelius Rufus and Sextus Popillius Phil-,
donated to Zeus Hypsistos in their official function as diakonoi
(deacons) of the cult (36). An epitaphic inscription for Gourdos from
Iconium (237), which is clarified by another inscription for the same
man that states he was in fact a Christian. 433 Further terminology
similar to that of the Christians is found in a few inscriptions: in
Serdica, one man uses his full three names Ponponius Theodoulos
(69) the second name could be considered Christianizing. The only
Hypsistos inscription known at Delphi was given by a doulos of the god
(31). The man is named as Titus Flavius Megalinos, making it likely that
he was a Roman citizen. The inscription is damaged, making it
impossible to ascertain whether the god was addressed here as Zeus or
Theos, but it is decorated with a crescent moon. However it should also
433

See Mitchell, Hypsistos, p. 123.

229

be mentioned that observance of the moon is of course very important


for determining the Jewish ritual calendar.
Priests are found in the corpora of Judaism as much as in
pagan cults. Conversely, Jewish sacral terminology is rarely found in
corpora of pagan cults, so the inscription from Pydna in Macedonia
dating

from

AD

250

that

clearly

demonstrates

the

crossover

terminology between Judaism and the cult of Hypsistos is something of


an exception (51). The stele records a group who meet together in
religious worship of Theos Zeus Hypsistos. It is the religious offices
that follow that make the inscription extraordinary: the archon Aurelius
Nigerionos, the archisynagogos Aurelius Kypiona, and the prostates
Aurelius Severos.434 All of these terms were regularly used in the Jewish
synagogue, and in fact rarely occur outside a Jewish or Judaizing
context.435 This fact suggests that, the cult of Hypsistos, even when
addressed as Zeus, should at this stage be cognitively connected in
some way with Judaism. The list of following names includes Roman
citizens and many Aurelii; both women and men; and a potentially godfearing or Christianizing name, Aurelius Theodoulos.
It is apparent that people from all different backgrounds and
professions worshipped Hypsistos: peasant farmers and city dwellers,
slaves and slave-owners, men and women. Aside from the pocket of
elites in Macedonia, there are no apparent links that can be made
between these people in terms of status or of occupation, implying that
the cult did not transmit across the networks formed by professional
groupings or guilds. How, then, did the cult come to be adopted by all of
them? The routes of diffusion ought to be looked for elsewhere
through a pre-formed network that already incorporated this plethora
of professions and social statuses.

434

See also Cormack, J. M. R., Zeus Hypsistos at Pydna, in Mlanges Hlleniques


offerts a Georges Daux, Paris: ditions E. de Boccard, 1974, pp. 51-55.
435
See Mitchell, Hypsistos, p. 101, n. 41, Rajak, T., and Noy, D., Archisynagogoi:
Office, Title and Social Status in the Graeco-Roman Synagogue, JRS 83, 1993, pp. 7593.

230


Places of Hypsistos worship
As seen above, there are a number of known or implied sanctuaries in
the earliest evidence for the cult of Hypsistos: the boundary stones that
may mark a sanctuary space for Zeus Hypsistos in Iasos (129; 130); the
dedicated doors in the tovpon of Zeus Hypsistos in Odessus (81); and
the Zeus Hypsistos sanctuary attested on Mount Cythus on Delos.
Although there is some variation in the types of sanctuary space
associated with Hypsistos, it is clear that a number of these early
sanctuaries were in the open air. In the Tosefta, R. Judah mentions the
open air of the synagogue, which implies that Jewish worship or
gathering could take place outside or in a place without a roofed
structure,436 in warmer climes or, perhaps, in places before there was a
large enough community to merit a building. Because the Torah was
holy, it consecrated the place it was in the presence of the Torah
made the building a sanctuary; study of the Torah thus became a cultic
act.437
In the later period of the cult, between the first-third centuries
AD, there are some other indications of the ritual space afforded the
Highest God. A third century inscription from Cotiaeum in Phrygia
reads: Aurelius Alexandros Timotheos and his wife Aurelia Ammia,
offered their vow to Theos Hypsistos in prayer, together with their
children Attikos, Artemon, Timotheos, Alexandros and Platon, set up the
columns and the propylon (215), which reveals a sanctuary space with
an entry area, but nothing more. Timotheos is clearly wealthy. The
pious name Timotheos, he who honours God, was certainly also used
by god-fearers/Christians (cf. Acts 16).
436

Zahavy, T., Studies in Jewish Prayer, Lanham: University Press of America, 1990, p.
72-73.
437
Van der Horst, P. W., Was the synagogue a place of Sabbath worship before 70
CE?, in Jews, Christians and Polytheists in the Ancient Synagogue: cultural
interaction during the Graeco-Roman period, Fine, S., (ed.), London: Routledge, 1999,
p. 36.

231

One of the largest sanctuaries to Hypsistos was in Athens, where


worshippers reused the Pnyx, carving fifty-eight niches and inscriptions
into the back walls. It was probably founded at the end of the first
century.438 Dedications have been found elsewhere in the city, and are
thought to have originated in the Pnyx. The contiguity of Zeus and
Theos here reinforces the interchangeable nature of the names,
however, the majority of the dedications are simply given to Hypsistos.
Twelve of the inscriptions have a connection with healing, either
through representations of the body part or in the text itself, which may
make a connection with the Jewish God by proxy (see note 416). The
Jewish community in Athens was well established and is attested both
in the epigraphic material (which also represents Samaritans) and in
Acts 17:16, which records also that there were sebomenoi in the
synagogue.
The niches in the back walls of the Pnyx sanctuary are
particularly noteworthy, as it is likely that these were places for the
dedication of lamps. Many lamps, as well as terracotta figurines, glass
unguentaria

and terra

sigillata

vessels were discovered in the

excavations of the sanctuary.439 Lamps form a major part of Jewish


worship, being lit at the beginning of the Sabbath, and Josephus 440
records the spread of the practice of lamp lighting amongst the
Gentiles. It is quite likely that the presence of niches and the known
dedications of lamps to Hypsistos demonstrate Josephus claim. The
third century dedication of a lamp to Theos Hypsistos by Chromatis at
Oinoanda illustrates this (234); and the contemporaneous oracle set
into the city wall just by her dedication and niche indicates that this
was the place where the worship of the deity occurred. The area is
outside and extra-mural, and carved precisely at the point of the old
438

Forsn, B., The Sanctuary of Zeus Hypsistos and the date and construction of Pnyx
III, in B. Forsn, G. Stanton, eds., The Pnyx in the History of Athens, proceedings of
an International Colloquium organised by the Finnish Institute at Athens, 7-9 October
1994, Helsinki: Vammalan Kirjapaino Oy, 1996, p. 49.
439
See Forsn, Hypsistos.
440
Josephus, C. Ap, 2.282, 39.

232

city wall which was first struck by the rays of the rising sun. 441 The text
reads:
Born of itself, untaught, without a mother, unshakeable, not
contained in a name, known by many names, dwelling in fire, this
is god. We, his angels, are a small part of god. To you who ask this
question about god, what his essential nature is, he has
pronounced that Aether is god who sees all, on whom you should
gaze and pray at dawn, looking towards the sunrise.442
It is apparent from this that the worship of Hypsistos was here
associated with the reappearance of the sun, so therefore an outdoor
sanctuary would be particularly important. This suggests that the
followers had quite specific times for gathering lamps are only really
visible, necessary, or impressive in the dark perhaps in keeping with
the Jewish Sabbath. Mitchell suggests that lamps represent the earthly
version of the heavenly fire, and indicates that individuals could partake
in a universal god through their humble gift.
Further explicit dedications of lamps are known in the Hypsistos
corpus. In first-second century AD Hierocaesarea: Teimotheos Diagoros
Labrantidis and Moschion Teimotheos his wife pray to Theos Hypsistos
at this altar. Diagoras, Teimotheos, Pytheos, the sons of Timotheos son
of Diagoros Labrantidai set up these lamps to Hypsistos (169). The use
of the pious name Timotheos is again notable. At Pella, the dedication of
an actual lamp simply to Hypsistos was discovered, decorated with an
eagle on a bucranium, dated to the second century (SEG XLVI, 785).
Another dedication that can be restored as pertaining to lamp
dedication is to Theos Hypsistos Epekoos from Tiberiopolis in Phrygia,
dating to the second-third century (225). The fact that all of these
examples of lamp-dedication are from the second century, and that they
all use Theos, or simply Hypsistos may suggest that this habit had been
adopted from close contact with Jewish practice.
441

Mitchell, Hypsistos, pp. 87-90, referring to the argument of Hall, A. S., The
Klarian Oracle at Oenoanda, ZPE, vol. 32, 1978, pp. 263-268.
442
Translation Mitchell, Hypsistos, p. 86; epigraphic reference 233.

233

In Sibidunda in Pisidia, Artimas and Marcia made a dedication of


incense and columns which might indicate a covered area to Theos
Hypsistos, whom they also called the abstract Hagia Kataphyge Holy
Refuge (230). This is the translation in the LXX of Divine Refuge in
Psalm 31,443 and has meant that this dedication is thought to be
particularly Judaizing. In the mid-third century in Tanais there was an
association of worshippers of Theos Hypsistos who call themselves
brothers (92-98; 100; 101), and although there are no explicitly
known Jews from Tanais, this was also a particularly Judaizing cultic
group: in prayer to Theos Hypsistos Epekoos, the association for the
sacred things, Papas Chrestos and the synagogue (102). This last
inscription is dated precisely to AD 244. With the mention of the
synagogue, this could of course be understood as a group of Jews in
Tanais. However, the same issues apply as to the Gorgippia inscriptions
discussed above. These could equally be Gentile worshippers of
Hypsistos defining their cult building and their God with the same name
as that of their Jewish co-worshippers, in the same way as the cultic
officials at Pydna in Macedonia; and perhaps also indicate the close
links that may have been retained between the Jews and the erstwhile
god-fearers. The identical situation is seen in the territory of Ancyra in
the third century in an inscription to the great Theos Hypsistos and the
heavens and the holy messengers of him and with the same
worshippers of the proseuche (202).
Writing in the fourth century, Pausanias444 supplements the
epigraphy. He mentions that in front of the entrance to the Erechtheion
in Athens there was an altar to Zeus Hypsistos, where they sacrifice
nothing that breathes, but they put sweet-cakes there and the rite
allows not even the use of wine (1.26.6). There were two altars to Zeus
Hypsistos at Olympia (5.15.5), in the south east of the sanctuary on the
443

Another inscription that refers to Theos Hypsistos as refuge was made by


Dimitrios on Cyprus (256); this dedication uses asuloi rather than kataphyge but may
be an attempt to render the same concept.
444
Pausanias, Guide to Greece, vol. 1: Central Greece, trans. Levi, P., London: Penguin
Books, 1971.

234

way to the hippodrome and in the vicinity of an altar of the Fates and
one of Hermes. There was cult for Hypsistos at Corinth (2.2.8), where
there were three statues, one with no title, one called Zeus of the
Underworld, and one Zeus the All-highest, all in the open air. It is
unclear what these statues looked like; the untitled one at least may
have been aniconic in some way. At Thebes, Pausanias discussion is
about the gates of the city, and by the High gates is a sanctuary of Zeus
the all-Highest (9.8.5). This last location sounds very like the sanctuary
at Oinoanda, discussed above, that was located at the highest point in
the city near the city wall; the statue of the god at Corinth was also in
the open air. At Olympia and Athens, the cult was able to be included in
the sanctuary space of other deities.
In sum, this seems a somewhat disparate collection of Judaizing
practice, buildings, and reused or extramural public spaces, with no
particular links to be discerned between them. However, the emphasis
is on altars in the open air and outside spaces. It also appears that, in
the second and third centuries, there are fewer attested sanctuaries to
Hypsistos than in the earlier period of the cult. However, a recently
published inscription from Thessalonica provides evidence that in the
Imperial period the worshippers of Hypsistos were sometimes able to
build their own temples from scratch: Zoilos son of Menon son of
Menandros son of Dionysios, and Kratisto daughter of Menon son of
Sosibios of the tribe of Antiochis, his wife, both from the tribe Antiochis
(?) furnished the foundations of the temple of Zeus Hypsistos from their
own means (SEG LII, 650). What is particularly noteworthy about this
inscription is that these wealthy devotees of Zeus Hypsistos were
founding a temple for his worship in the second-third century AD. The
cult of Zeus Hypsistos had been popular among the elites of Macedonia
since the Hellenistic period, and in Thessalonica itself since the first
century AD. What could account for this necessity to build a new temple
to a god that had been worshipped in the city for at least a century?
Most of the other inscriptions in Thessalonica are to Theos Hypsistos,
235

and the Jewish synagogue there is well attested. Could this inscription
be understood as the foundation of a separate Hypsistos temple as a
result of the separation of Gentile god-fearers from the synagogue?

Geographical Absences
There are some important places where the cult is not attested at this
time. As Mitchell suggests, it may be that the god was worshipped
under

different

name,

or

the

worshippers

called

themselves

something different, for example, the Caelicolae in North Africa


mentioned by St Augustine (see note 327). This may be the case, but
the simple fact that the name Hypsistos is not used marks a
fundamental difference between them. It is my suggestion here that the
geographical spaces where the cult is not attested show that this was
not a cult with long-range, independent diffusive appeal. By contrast, it
is argued that this shows that the cult had arisen at a local level, as a
response to localised interactions or exposure. In particular, in the
same way that the patterns for the cult of Hypsistos and the Jewish
Diaspora can be fairly closely matched, 445 the absences in the evidence
for the cult can be related quite closely to places where there were not
large Jewish populations, or where those populations had been
destroyed.
The notable exception is Rome, where the Jewish population was
very large, yet there is only one inscription, in Greek, to Theos
Hypsistos, given by Claudia Pistis, the faithful (125). As seen above,
there are also very few god-fearers attested from Rome. Certainly in
the first century AD Judaism is known to have attracted Romans, and
Poppaea Sabina is called theosebes by Josephus. It seems likely that the
absence of epigraphic examples may be accidental; or perhaps that in
Rome, the Gentile adherents of Judaism were known by another name;
445

See Mitchell, Hypsistos.

236

or that, as the cult of Hypsistos did not have an established precedent


in the city, god-fearing Gentiles adopted Christianity instead.
In Rome and across the Latin west, oriental cults and gods with
Greek names and heritage were widely diffused, but the cult of
Hypsistos is barely attested. An inscription to Theos Hypsistos Epekoos
was found in a cultic cave on the coast north of Otranto in Calabria
(SEG XLV, 1482), alongside other dedications in the cave to Hypsistos,
Epekoos, Aniketos, and Optimus Maximus. The inscription to Theos
Hypsistos was dedicated by a freedman of the Anicii invoking protection
on behalf of his patrons and wishing a safe journey for a ship, but aside
from this cave and Claudia Pistis in Rome, the cult in Italy is entirely
absent. The situation is the same in the northern and western provinces
of the Roman Empire, where the only other known dedication is an
amulet from Valentia on the coast of Hispania (126). A few Jewish
inscriptions are known from along the coast of Spain, in Sardinia, and
around the northern shores of the Adriatic, but as with the considerable
Jewish community attested in Sicily, the inscriptions are mostly dated to
the third century or later.
This was also clearly not a cult that made any impression in the
Roman army, which was such a fertile ground for the proliferation of
new cult movements, as has been seen in Chapter 4. The most obvious
explanation for this is that worshippers were never part of the military
networks that spread religious information but why not? The only
Hypsistos dedications from the militarised northern provinces are from
Dacia, where the names Zeus and Theos also appear alongside each
other in Apulum and Sarmizegetusa. Iulius Ateimetos dedicated a votive
depicting ears to Theos Hypsistos (79); Aelia Cassia set up a
thanksgiving to Theos Hypsistos Epekoos (78); and Aelius Apolinarius,
a guardian, and Maxima addressed the deity as Zeus Hypsistos
Epekoos (77). The appearance of the name Aelius may indicate that
these people were quite high status, perhaps having been given Roman
citizenship by Hadrian, or descended from people who had been. The
237

only inscription to Hypsistos solely in Latin is from Apulum, reading to


the highest Jupiter, most superior (Iovi summo exsuperantissimo), ruler
of divine and human affairs, arbitrator of destiny (76). It seems that
the dedicants are all relatively high status Romans, although some have
a Greek nomen, and use Greek as the language for dedications. Very
few Jews are known from the northern provinces, and none in Dacia: a
fourth century Jewish archisynagogos is found in Oescus in Moesia
Inferior. Jews had fundamental issues with Roman military service,
although there are two known Jewish synagogues on the limes in
Pannonia. Because of the commercial and military connections of the
northern provinces, the populations of these places are quite likely to
have been highly mobile, and these Hypsistos worshippers in Dacia
probably introduced the cult from elsewhere.
Egypt is almost completely devoid of evidence for the worship of
Hypsistos in the imperial period. Equally, there are no known godfearers in Egypt, suggesting that the social divide between Greek and
Egyptian Gentiles and Jews in Egypt was too wide to bridge. Both Zeus
and Theos appear in the few pieces of Hypsistos material dating from
the Hellenistic period. However, there is only one dedication dated to
the first-second century AD, which is, notably, an inscription from the
Jewish colony of Leontopolis, mentioning Theos Hypsistos and the
proseuche (288). Egypt was an area where, in this period in particular,
the interactions between Jew and Gentile had been rife with social
tension, leading to the pogrom in Alexandria and finally, under Trajan,
the revolt and the decimation of the Jewish population. Furthermore,
there was apparently no established body of Gentile worshippers of the
Jewish God. The social tensions in Egypt, and the lack of god-fearers go
some way towards explaining the later absence of the cult of Hypsistos
in the region. The revolt and destruction of the Jews in Cyrenaica might
also be recalled; however, the differing situation on Cyprus is
noteworthy. The Jews of Cyprus revolted under Trajan also, massacring
many non-Jewish inhabitants and destroying Salamis, for which they
238

were banned from the island. They seem to have begun to re-establish
themselves in the third-fourth centuries. In the intervening period, the
cult of Hypsistos is very well attested, with most of the inscriptions
dating from the second-third centuries. It is notable that every single
dedication is to Theos Hypsistos and there is a fairly strong healing
connection. It is plausible to imagine that these might represent Gentile
god-fearers who developed the cult of Theos Hypsistos after the Jewish
uprising and banishment. There is no indication of what kinds of spaces
they worshipped in.
This survey of the areas where Hypsistos is not attested
highlights the nature of the religious network across which the cult
diffused. The cult appears to have been highly localised, spreading
across a network that was already formed. It did not transmit over long
distances like other eastern cults the extremely few epigraphic finds
from Spain and Italy are generally maritime or transportable items. The
uniformity of the second and third century dedications, and the fact
that by far the majority of the inscriptions use Theos instead of Zeus to
address the god supports the notion that the dedicants at this time had
something in common: a prior affiliation with Judaism.

The Cult of Hypsistos: divergent trends


However, there is at this time also the indication of divergence within
the cult of Hypsistos. Some worshippers seem to want to incorporate
aspects of traditional polytheism, visible in the few dedications that
make an association with other deities. For example, a thanksgiving
dedication by two men from Nysa in Lycia is for Theos Hypsistos with
the Metri Oreia, at the calling of all the gods and goddesses (232); and
a dedication from Cos mentions a pantheon associated with Zeus
Hypsistos (105a): Hera Ourania, Poseidon Asphaleios (an often-used
epithet for Poseidon, meaning he who secures from earthquakes),
239

Apollo and all the gods. It appears that this individual certainly
understood the cult of Hypsistos in a more traditional polytheist sense,
finding it necessary or desirable to include the traditional Hellenic
pantheon alongside Hypsistos. However, what is apparent is that these
few examples cannot be used to argue that the cult simply fits the
normal pattern of polytheism: that polytheist tendencies are so
occasional supports the notion that in general, the cult of Hypsistos was
monotheistic. The exceptions underline the rule.
A particular pocket of local tradition is seen in the survival of
seventeen dedications from Stratonicaea in Caria, most of which are to
Zeus rather than Theos (140-156). The earliest dedication is dated to
the period of Antoninus Pius, with the rest being roughly second-third
century. Almost all of the dedications are to Dii; uJyivstw/ kai; qeivw/
either Zeus the highest God, or Zeus Hypsistos with an abstract divine
power, to theion.446 Angeloi also occur frequently, reminiscent of the
common Jewish idea that the righteous dead become angels or mingle
with the angels. Pseudo-Phocylides refers to the heavenly bodies as
blessed ones and says that the dead become theoi neither of which,
in Collins opinion, should be taken as deviation from monotheism. 447
These facets of the dedications both suggest a strong Jewish aspect to
the cult, as Pseudo-Phocylides refers to the righteous Jewish dead as
theoi that mix with the angels. Angel worship was part of the
indigenous religious culture from other places in Asia Minor, and this
also links with the description of God and his angels in the oracle from
Oinoanda. The earliest known inscription, dated between 138-161, is
particularly notable, as it names a pantheon with Zeus Hypsistos
Hekate Soteria, Zeus Capetolios and the Tyche of the emperor
Antoninus Pius (140). Mitchell suggests that this shows Zeus Hypsistos
was one of the main civic deities of Stratonicaea, alongside Hekate,
who was known for her famous extra-mural sanctuary at nearby Lagina,
where an inscription following the formula of Stratonicaea has been
446
447

Mitchell, Hypsistos, p. 102.


Collins, Jerusalem, p. 145.

240

found: Dii; uJyivstw/ kai; qeivw/ tw`/ basilikw`/ (157). Both these prominent
local gods were then coupled in a familiar fashion with the emblematic
Roman cults of Jupiter Capitolinus and the Fortuna of the ruling
emperor. This suggests that at Stratonicaea, to a greater degree than
elsewhere in the evidence, the worship of Zeus Hypsistos was part of
mainstream civic paganism. However, the association in the majority of
the dedications with to theion suggests that it had much in common
with the Theos Hypsistos of other contexts.448
While some dedications from this period incorporate other deities,
equally, some worshippers are more Judaizing. Some of these, such as
the dedication that mentions the Hagia Kataphyge, the synagogue
building in Tanais, or Judaizing official terminology, such as that at
Pydna, have been discussed above. Some references may be more
covert. In Thessalonica, Quintus Urbanus set up a dedication to Zeus
Hypsistos with Nemesis, qea;n dikaivan, goddess of righteousness (54),
showing the association between the Hypsistos cult and concepts of
justice and vengeance within Greek thought and religious ideas. An
inscription from Nacolea in Phrygia (220) is to Theos Hypsistos with
Hosio kai Di-, which Mitchell is tempted to read as Dikaios, the abstract
god of justice found often in Phrygia. 449 Although there is a lack of
space on the stone for the necessary letters, the frequency of the terms
in the region means that the use of Hosio kai Di- is almost enough to
conclude that this is what the stonecutter meant. Further support for
this argument is found in an example of Hypsistos in conjunction with
Hosios kai Dikaios from north-east Lydia (SEG XLVIII, 1427 now TAM
V.3, 1637) where the stone was first cut to read Theos Hypsistos, with
the upsilon, psi and sigma overcut to read Hosios. It is, however,
unclear as to whether this dedication is to the Theos Hosios kai Dikaios
the pious and righteous God, or to Theos and the Holy and the Just
as separate entities. The editors of SEG do conclude however, that, in
Phrygia, the cult of Hypsistos should be considered to be in some way
448
449

Mitchell, S., Pers. Comm.


Ricl, M., Hosios kai Dikaios, EA, 18, 1991, pp. 1-70; 19, 1992, pp. 71-103.

241

associated with the Holy and the Just. Although the dedications are
largely uniform, there was clearly room within the cult for the
expression of individuals personal religious beliefs or strong local cults,
especially in the later Imperial period.

It is clear that the cult of Hypsistos existed as an entity separate from


Judaism to a greater or lesser degree from the Hellenistic period
onwards. We have seen also, however, that many aspects of the
epigraphic record link the cult of Hypsistos strongly to Judaism, and
although the material is often simple and not particularly revealing, the
above examination of the epigraphy has shown that a number of
conclusions can be drawn from the corpus taken as a whole. However,
these conclusions are, to an extent, negative. First, that the cult was
not limited to any gender, sector of society (except perhaps in the case
of the Macedonian elites) or profession the range of the people in the
evidence are connected solely through the common term used to
address their deity they form, in other words, a religious network.
Second, that the spaces where Hypsistos worship occurred were
somewhat heterogeneous; an implication being that there was not a
fixed space for worship, although there were certainly priests of the
cult and a tendency towards open air sanctuary space. This might
further suggest that devotees did not necessarily have much choice of
the place in which they worshipped, but more important, it shows that a
built physical location was not a pre-requisite for the cult to thrive.
Finally, that the places where the cult is not attested, or is not attested
during the later Roman Empire, are glaringly apparent. These areas, in
particular Egypt and the militarised northern provinces where oriental
cults were so profoundly popular, were areas under the Empire without
Jewish populations and the associated god-fearers. The cult of
Hypsistos could only move through receptive social space.
242

The cult could not reach into social space where there were no
existing networks to facilitate transmission. One explanation that might
help to incorporate these three aspects of the epigraphic evidence is
that, in the second century, the numbers of the existing cult of
Hypsistos, which was fairly evenly distributed across the eastern
Mediterranean and had been influenced to a degree by Judaism, was
swelled by erstwhile god-fearers, who brought with them a partially
formed network of believers. These were people from all social
backgrounds who had been associated with Judaism men, women,
poor, wealthy, from all kinds of professions. At precisely the time that
Judaism was undergoing radical change and upheaval, and when Jew
and Gentile relations were becoming in some places extremely hostile,
the cult of Theos Hypsistos boomed.
It will be helpful to supplement these hypothetical explanations
by visualising the connections that might have existed between the
worshippers of Hypsistos themselves and comparing them with the
network for the Jewish Diaspora. The following network analysis will
add depth to the conclusions from the epigraphy.

Visualising the Network

Here the pattern of the evidence is assessed as a totality. Using


networks allows the hypothetical reconstruction of largely invisible
communication routes and the avenues of information transmission that
framed and shaped the spread of the cult of Hypsistos. The simple
distribution map (Map

6A) shows the places where Hypsistos

inscriptions have been found, concentrated in Macedonia, Greece and


Asia Minor: purely geographical data that can offer no further
information as to the development of the cult or its transmission.
243

Linking the nodes together into a network visualises the potential


communication routes between these places and reveals patterns and
routes in the distribution as a whole.
An initial Proximal Point Analysis (PPA) of every piece of
epigraphic data captures the end picture of the cult, linking every
known node, regardless of date, to its three closest neighbours. This
type of model highlights clusters, isolation and centres. It is a
preliminary analysis of data that spans roughly six centuries and as
such is not a reflection of the actual connections that existed between
these sites. However what it does do is allow an initial observation of
the geographically determined network: the empty spaces, longdistance overland or maritime links, and some of the constraints the
terrain had on communications.
The three subsequent directed Proximal Point Analyses respond
to what is known of the date range of the evidence, by mapping the find
spots in three temporal chunks to match those of the epigraphic
analysis. Each find location is once again linked to its three closest
neighbours, but because of the loose dating of these cult finds, only one
of which is required to be already established. This analysis helps to
visualise the spread of the cult in actual terms, highlighting centres of
diffusion and building localised network pockets of interaction. A
comparison with the PPA of the Jewish Diaspora allows the relationship
between the two to be seen more clearly. On the maps, a site with only
one epigraphic find is written in lowercase lettering; a site with more
than one find or an inscription that reveals a building is in uppercase.

Proximal Point Analysis: Map 6B


The initial proximal point analysis of the Hypsistos evidence shows the
concentration, interconnectedness and evenness of the cult finds in
Macedonia and western Asia Minor, implying a localised and somewhat
244

uniform diffusion. However, the networks of the more distant areas are
also noteworthy namely, the separation of the networks in Bosporan
Cimmeria and Paphlagonia and Pontus and that of Egypt, the Negev
and Syria. Crete and Cyprus also represent introverted and somewhat
disconnected clusters. These network formations might indicate that
the cult was of a different quality: and Cyprus, Egypt, Syria, Bosporan
Cimmeria, and to a lesser extent, Crete, are notable for the strong
Jewish connections in those places. In Crete, Cyprus and the Cimmerian
Bosporus only dedications to Theos Hypsistos are found; comparison
with the network for the Jewish Diaspora will help to further illuminate
these connections. The interpretation of the network of Paphlagonia
and Pontus will also benefit from this comparison.
In the more tightly integrated network of Greece, Macedonia and
Asia Minor, some other features are immediately noticeable. The divide
between Macedonia, Thrace and Thessaly, and the southern Greek
networks is quite apparent, and may indicate that the cult was of a
different form there is certainly a divide in terms of address between
these two areas, with the southern locations, with the exception of
Sparta, mostly choosing Theos rather than Zeus.
The

pathway

across

the

south

Aegean

is

through

Delos,

highlighting the island as a bridge between southern Greece and Asia


Minor. Through the north, it is between Thrace and Mysia via Lemnos
and Imbros, implying the connectivity of these areas.
Across Asia Minor itself, it is interesting that the sites in Lycia
and Pisidia connect up the river valleys into Phrygia, rather than across
the mountains into Caria, reflecting the geographical barriers and
indicating that the cult in Lycia might be more strongly linked with
Phrygia. Likewise, the network around Byzantium, the Sea of Marmaris
and coastal Moesia connects into central Phrygia rather than across
Thrace to Macedonia. These features combine to highlight Phrygia as
an area of central importance to the cult, and create a hub with six
links at Aizanoi. Other hubs are at Athens and Corinth, both places
245

where there are large known Jewish communities; Antigoneia in


Macedonia, which seems to have been highlighted for its between-ness
linking Dacia and Moesia Superior into Macedonia and Thrace; and
Emesa in Syria, a major cultic centre and certainly by the fourth
century AD, with a Jewish community large enough to support an entire
regiment, known from an inscription from Iulia Concordia (CIJ I, 640).
It is notable that the network is very evenly weighted and as such that
there are very few nodes that have hub status. This suggests that,
rather than being a cult movement that was diffused centrally, it was a
more grass roots, localised emergent process. This network pattern
also supports the notion that it does not really represent a diffusion of
ideas, rather the activation of a pre-existing group.
The more distant northern and western cult find spots, in Spain,
Rome, and Dacia, highlight their own isolation through the long links
across huge stretches of water or land. This offers a clue as to the
routes by which the cult came to be found in these places i.e., that
these dedications were by people out of their ordinary context. It might
be supposed that the worshippers in Apulum and Sarmizegetusa were
connected with the Roman military and commercial operations in the
area, and that these places had long-distance connections with the rest
of the Roman world is indicated by the inscription of a Roman citizen
and councillor of Sarmizegetusa attested in Mytilene in the second
century (117). The single find in Rome must presumably be connected
with the Jewish community there; and the two other western locations,
Valencia and Torre dellOrso in Calabria, also both single finds, are
surely connected with maritime activity. Indeed, the eu[ploia inscription
from Calabria makes this explicit (SEG XLV, 1482); and the find from
Valencia is really in a different category, as it is a portable lead amulet
(126).

246

Directed Network Analysis (Maps 6C-6E)


Following from these initial observations of clustering and isolation in
the pattern, the following Proximal Point Analysis will build in the date
range of the evidence. By plotting the distributions in chunks, the
analysis reveals a clearer picture of the diffusion of the cult over time
and places that can be identified as centres. Once again, nodes connect
to their three closest neighbours, one of which must be already
established. What this sequence is intended to show is the potential
diffusion routes based solely on the date and pattern of the evidence.
This series of networks will then be compared with the networks for the
Jewish Diaspora.

The Hellenistic period to the end of the first century BC: Map 6C
The early cult network is simple: divided between that in Egypt and
Cyprus and that in the Aegean. In the network centred on the Aegean,
it is notable that the locations are mainly coastal Delos, Imbros,
Skiathos, Miletus, Iasos, Prusa and Odessus. These are at some
distance from each other, making the network distances fairly long, but
their coastal locations imply maritime links. The separation of the
Aegean network is supported by the fact that almost all of these are
dedications to Zeus Hypsistos.
Delos,

with

five

links,

is

shown

to

be

crucial

to

the

interconnectivity of the cult in the Aegean, likewise Imbros provides a


stepping-stone between the Aegean and Odessus and Prusa. Although
both of these are single finds, they reveal that the cult was quite well
established in both these places at this early stage, the inscription from
Odessus recording the donation of parts of a building to Zeus Hypsistos
(81); and the dedication from Prusa being on behalf of the whole village
(189). The cult was not therefore simply superficially adopted. The
pockets of introspection in Caria and Macedonia represent localised
247

cult movements, focused respectively around Miletus and Iasos and


Edessa.
The separation of the Cypro-Egyptian network is quite striking,
especially in comparison with the initial PPA, which posed no links
between them. The cult is surely related to the considerable Jewish
populations in both of these places, supported by the fact that the
dedications in both these areas use Theos, aside from the first century
guild of Zeus Hypsistos worshippers in the Fayoum (287). As noted
above, two of the Egyptian inscriptions also mention the proseuche.
Egypt and Cyprus were closely linked; indeed Cyprus was for a long
time a major part of the Ptolemaic empire. Egypt also at various times
possessed Caria, Lycia, some Aegean islands and parts of Thrace and
the Chersonese, which serves to remind us of the pan-Aegean
communications and taxation that took place in the Hellenistic
period.450

The first century AD: Map 6D


The following hundred years show a substantial expansion of the
network, and also that it does so quite evenly. It is at this period that
the cult is first attested in the Black Sea region, in internal Anatolia, in
southern Greece, and stretches superficially into Rome to the west.
The network is very even, with most links fairly short and many
nodes that have six or seven links. Delos is slightly larger as a hub, with
new links to Crete and Athens, but its influence is essentially limited to
the southern Aegean. At this stage, the islands fortune was in decline,
and it is Imbros that acts as a major juncture between the cult in
Macedonia and the Sea of Marmaris. There is only one dedication from
Imbros (113), however, so what the network really highlights is the
level of connectivity between Thrace and Macedonia and northwest
450

Rostovtzeff, M., Social and Economic History of the Hellenistic World, vol. 1,
Clarendon Press: Oxford, 1941, p. 332-339.

248

Asia Minor. It is pertinent to remember that it was at this time that Paul
made his crossing over into Macedonia from Troas, arriving in Philippi
(Acts, 16:8-12).
The pocket of cult in Caria barely changes in this period, whereas
the network in Macedonia expands outwards and becomes less
introspective, joining south into Greece, north to Serdica and eastwards
along the coast into Thrace. Again, Pauls missionary journeys
throughout this area are expedient to recall, highlighting the actual
connectivity and transmission of religious ideas that was happening at
the time. A smaller pocket also appears clustering around Prusa in the
region of the Sea of Propontis.
It is quite illuminating that the network between Cyprus and
Egypt hardly expands and remains entirely separate. The lack of cult
dedications in this period suggests that there was no real social space
for the cult. That this network does not join up with the Aegean network
suggests that the interactions between Cyprus and Egypt remained
markedly introspective, and just a little later than the temporal
boundaries of this map, under Trajan, the Jews of Cyprus revolted along
with Cyrenaica and Egypt, proving that close communications between
the Jewish communities in both places were still in existence.

The second-third centuries: Map 6E


This last map in the series illustrates the extraordinary increase in the
number of dedications to Hypsistos in this period. The expansion of the
network connectivity is quite extreme, but it is immediately obvious
that the shape of the network essentially retains the same shape.
It is only in this period that the network between Egypt and
Cyprus connects with the rest of the cultic network in the Aegean, via
Cyprus and the sites on the coast of Cilicia. The cult on Cyprus itself
becomes profoundly introspective, which may in part be reflective of
249

the broken link between Cyprus and Egypt following the Jewish revolts.
The coast of Phoenicia and Palmyra in the hinterland appear in this
period, and connect into Cyprus. Egypt gains no new nodes, clearly
paralleling the destruction of the Jewish population there, but links are
made into Egypt from the new sites in the Negev in Judaea and Petra in
Nabataea. The find from the Negev is decorated with a menorah (281),
indicating that this was a fully Jewish dedication.
In Asia Minor, the network connects the coastal cities of Caria
and Ionia, and spreads into the cities of central Anatolia. Hierocaesarea
and Pessinus emerge as local centres, due to the presence of the cult in
the previous century, however, even though these sites have more than
the average number of links, the resulting network is still extremely
decentralised. This makes apparent the speed and the profundity of the
adoption of the cult in these areas, supporting the interpretation of the
epigraphy that it was not a centralised cult, but rather, represents in
some way the activation of a network already in place. It additionally
indicates the epigraphic habit being adopted as a sudden, emergent
phenomenon in these areas. This analysis also shows the same
geographical barriers as the initial PPA. Lycia and Pisidia connect up
into Phrygia, while the knot of cult finds Caria makes a tightly
introspective network, implying the cult here has a traditional
character; although it does connect via Aphrodisias into Phrygia.
Similarly, the cult network in Thrace and Macedonia in particular
becomes very tightly integrated, implying that this area the cult was
localised and standardised.
The expansion of the Roman Empire under Trajan into the
province of Dacia early in this period is made clear with the find from
Apulum, with the network suggesting that the person who dedicated
came into this area from Moesia Superior. It is notable that the cult
found no followers further west along the Danube, when other eastern
cults were proving so popular among the soldiers in the Roman army; in
fact, the only indication that there might have been military interest is
250

found in the location of the Hypsistos finds at Sarmizegetusa. The cult


is otherwise completely absent from the Danube valley, except at the
mouth of the river and a little along the coast. The people dedicating to
this deity up in the mountains of Dacia must then be connected with
cult followers further south, made clear by the network connections.

Network Comparison: the cult of Hypsistos and the Jewish


Diaspora
A comparison of the two PPA analyses for both the Hypsistos material
and the material pertaining to the Jewish Diaspora reveals the striking
similarity in their distribution. This section refers to Map 5C in Chapter
5, and Map 6B here.
However, an immediately identifiable difference is that the Jewish
Diaspora permeates through the western Mediterranean, where the
cult of Hypsistos is absent. This is explained simply through the fact
that much of this epigraphic evidence for the Jewish Diaspora in the
west is either superficial or dated to the third century or later. The cult
of Hypsistos is also clearly missing from North Africa. The Jewish
network is heavily introspective in Egypt, whereas the cult of Hypsistos
is only found in the Nile delta region and only superficially.
However, the similarities between the two networks far outweigh
the differences. The distributions for both in the eastern Mediterranean
are virtually identical. In particular, the position of Delos as a bridge
between Athens and southern Greece to the Dodecanese and the coast
of Caria and Ionia is very similar. This is partly geographically
determined but also makes clear the unique position that Delos
occupied in antiquity. A node with such high network betweenness like
Delos would have provided fertile ground for the exchange of
information and religious ideas. Lycia and Pisidia are linked to Phrygia
in both the Jewish and Hypsistos networks, while the networks in Caria
251

and Ionia are entirely separate. Also similar are the networks round the
Sea of Propontis.
The cats cradle of both networks in central Asia Minor is
particularly notable, and shows that interactions in this area would
have been frequent and fairly intense. It is important that in this area,
no centre emerges for either network, suggesting that the diffusion of
ideas in Phrygia, Galatia and the hinterlands of Caria, Ionia and Mysia
might have been a particularly heterarchical process, driven by lowlevel localised interactions.
However,

the

network

configurations

in

the

eastern

Mediterranean also show some interesting differences. Although both


groups share three of the sites in northern Asia Minor (Amastris,
Sebastopolis, Tavium), the networks they form are quite different. The
Jewish Diaspora links into the north shore of the Black Sea, into
Bithynia and down into Lycaonia, while Hypsistos forms an entirely
separate network. However, a comparison of these configurations and
noting the points they share suggest that they probably shared
connective routes: cross-Euxine for Hypsistos, and between the eastern
steppe and the coast of Paphlagonia for the Jewish Diaspora. The Jewish
communities of Asia Minor are, aside from via the Sea of Propontis and
Delos, apparently unconnected with Macedonia and Greece, whereas
the

Hypsistos

cult

makes

strong

link

between

Mysia

and

Macedonia/Thrace: the recollection of Pauls routes between the Jewish


communities of Asia Minor and the Aegean basin counters this
somewhat false picture and remind us that the networks were in
actuality more similar. Likewise, the disconnection of Crete in the
Hypsistos PPA network is countered by the links into Melos and the
southern Peloponnese seen in the Jewish network, as well as in the
developed Hypsistos PPAs.

252

These network analyses have supported the conclusions drawn from the
epigraphy that the cult of Hypsistos emerged in a decentralised fashion.
Network hubs potentially had an early effect in diffusing the cult, Delos
being an obvious candidate, but also the sites of Miletus-Iasos had an
important effect on their local environment, and Hierocaesarea and
Pessinus during the first-second centuries. However, the network for
the cult of Hypsistos is in general extremely decentralised, and it is
noticeable that in Macedonia, no site emerged as having particular
network centrality, supporting the suggestion that diffusion was
decentralised or heterarchical. The initial Proximal Point Analysis
highlighted the areas in the network that might be considered
anomalous or of a different quality. Through the three temporally
developing networks, the analysis has further demonstrated the lack of
cultic centres, and the localised, emergent interactions that seem to
have driven the cult of Hypsistos. The comparison with the PPA network
for the Jewish Diaspora reveal the striking similarities between the
diffusion patterns, and highlight some of the different configurations of
the networks, but also the potential points of contact between them.

Conclusions: Religious networks and competition

The epigraphic evidence for the cult of Hypsistos is not as informative


as that for the cult of Jupiter Dolichenus or that of the Jewish Diaspora,
and moreover, the dating is often very loose, so it is not easy to make
substantiated claims regarding the nature of the cult or its relationship
to Judaism. Given these barriers, this chapter has gone some way to
making some broad proposals about the cult.
It has been suggested that Delos played a pivotal role in the
introduction of the cult of Hypsistos into the Greek world. Delos
253

occupies a central position in both the epigraphic and network


analyses, and as an international cult centre and free port in the
Hellenistic period, people of considerable status from across the
Mediterranean were frequent visitors. The opportunity for the exchange
of religious ideas is clear in such a context, and the high status
individuals who visited the island would have been in a good position to
bring new ideas back to their homelands, and have the social power to
influence others.
It has been shown that the cult was diffused relatively evenly
during the first century AD, across the Mediterranean world, some
areas more Judaizing, some more pagan. At this period Jewish
communities are also found in the same locations where the cult is
known. It has been argued that the cult underwent a major change
during the second-third centuries, when the attestations increase
enormously and almost all use Theos rather than Zeus. Rather than this
simply being a product of the epigraphic habit, although this certainly
played a role, the facts that the cult crosses social divides and is not
confined to any professional or other group; has no fixed paradigm of
space for worship; and is markedly absent from certain areas
especially in the Western Mediterranean where it might have been
expected to occur, together suggest the activation of a previously
established religious network. The increase in dedications at this time
represents a phenomenon with an emergent or self-organised quality,
arising across particular spaces in the eastern Mediterranean, Anatolia,
and the Black Sea world, all at the same time, due to a series of
localised interactions.
These are markedly regions where there were known Jewish
communities. Furthermore most of these environments, especially in
the interior of Asia Minor, were largely agrarian and populated on the
whole by peasant farmers. New influxes of immigrant settlers would
have been quite exceptional. The low-level and regular interactions
between Jew and Gentile in these areas Hellenised the Jews and
254

Judaized the Gentiles and this resulted in considerable Gentile


attendance at Jewish religious gatherings. The Acts of the Apostles
records god-fearers in all the synagogues that Paul visits yet the godfearers are almost entirely absent from the epigraphic record in Asia
Minor, with a few examples from either the first or the third centuries.
If the epigraphic habit as a fashion spread over central Asia Minor in
the second-third century, then why did the god-fearers not also adopt
this habit?
This analysis goes beyond Mitchells argument that these people
who had been associated with Judaism should also be found in the
evidence for the cult of Theos Hypsistos, and proposes that the reason
that they did not call themselves god-fearers in the second-third
centuries was because it was unattractive or even dangerous for
gentiles to associate themselves overtly with Judaism following the
destruction of the Temple and the revolt led by Simon Bar Kokhba. In
the previous chapter, it was argued that the strong-tie ethnic
connectivity of the Jewish Diaspora was activated in the years following
these disasters, and that this made the network increasingly susceptible
to religious innovation, utilised by the religious authorities in Palestine
to transmit the rabbinic reforms. These new religious strictures must
have had some effect on the Gentile god-fearers, and would have led to
a re-articulation of the relationship between Jew and Gentile. The godfearers already formed their own network as a result of their
participation in the synagogue, and could either join the burgeoning
messianic sect of Christianity, or the already established cult of Theos
Hypsistos. It is suggested here that many of them did accounting, in
part, for the huge increase in dedications at this time.

255

The end of the cult


The dedications to Hypsistos drop off almost completely by the fourth
century, with five known inscriptions dated to this period. One from the
Negev bears the explicit Jewish symbol of the menorah (281); one from
Hadriani in Mysia is explicitly Christian (184). Might the type of
religious network formed by Hypsistos worshippers account for the end
of the cult? It is apparent that during this period, god-fearers again
begin to be found in the epigraphic evidence, suggesting that the
Jewish communities were opening up to the outside once again.
Christianity had boomed in popularity and was adopted as the state
religion of the Roman Empire at the beginning of the fourth century.
The end of the cult of Hypsistos was inevitable: with no particular bond
of ethnicity, class or profession to link the worshippers, the religious
network was only as strong as the cult itself. Although Christianity too
crossed these kinds of social divides, it had a missionary core and
impetus. The example of Gregory of Nazianzus father, who died in AD
374, is a pertinent illustration of the strong competition from
Christianity that the cult of Hypsistos faced. He had originally been a
worshipper of Hypsistos, but had been converted some fifty years
previously, by a travelling group of Orthodox Christian bishops on their
way to the council of Nicaea.451 In the face of the powerful missionary
ethic of Christianity, and the re-opening of the doors of the Jewish
synagogue, the cult of Theos Hypsistos was unable to survive.

451

Gregory of Nazianzus, Or. 18.5 (PG 35. 990), in Mitchell, Hypsistos, p. 94-95.

256

Epilogue

Networks of interconnections between nodes are used in the modern


world to explain a variety of types of change across a variety of
disciplines, from disease epidemics to systems failure. The aim of this
thesis was to begin to explore how insights from network models might
be applied to the ancient world.
Understanding how the structure of the environment can
determine the diffusion of an innovation is key to understanding why
some innovations are successful and others fail. By conceiving of an
environment as a space characterised by interconnected nodes, a
network, it is possible to analyse the success or failure of an innovation
without recourse to a value judgement about the nature of the
innovation itself. The emphasis of network theory is on the processes by
which innovation and social change occur: the combination of the
inherited, vertical aspect of the culture that determines the nature of
an innovation, and the environment, or the connectivity of the network,
the

horizontal

aspect,

that

determines

the

profundity

of

that

innovations propagation.
I have shown that network theory can be used both as a heuristic
device and as a practical modelling tool for re-approaching the subject
of religious change in the ancient world, allowing us to re-conceptualise
the reasons for the success or failure of religious movements. Instead of
focusing simply on what the epigraphic data can tell the modern
observer, approaching the evidence from a network perspective has
brought

other

aspects

to

the

forefront

of

interpretation.

The

distribution patterns as well as the social patterns of epigraphic


material can be analysed, showing the social networks that were in
246

place that facilitated the transmission and adoption of new religious


information.
The three case studies examined here have illuminated different
ways in which religious movements operated in the ancient world. The
analysis of the cult of Jupiter Dolichenus has shown that the actual
innovation of an entire cultic form can spread across a strong-tie social
network, and that the particular strong-tie network of the officer class
of the Roman military was conducive to the swift propagation and
dissolution of novel religious forms. The success (and failure) of Jupiter
Dolichenus might be described as an information cascade, a stochastic
network growth triggered by a single change or adoption, but that gave
rise to multiple consequent events.
By contrast, the cult network of Theos Hypsistos has been shown
to have properties which suggest a state of self-organised criticality: in
that a series of small interactions between Jew and Gentile combined to
create a fragile system, manifest in the god-fearers, that was subject to
mass change given the right environmental factors. The political
situation created by the destruction of the Temple in Jerusalem and the
beginning of the rabbinic reforms was just such an environment for
god-fearing Gentiles. The emergence in the second-third centuries of
the cult of Theos Hypsistos marks the epigraphic trace of a selforganised reaction to the tensions between Jew and Gentile.
The discussion of the spread of information across the ethnic
network formed by the Jewish Diaspora might be described as
combining elements of an information cascade driven by a single event,
with a network that displays a level of self-organised criticality. The reJudaization of the Diaspora following the destruction of the Temple in
AD 70 and the Bar Kokhba revolt was manifest in the epigraphy
through the clear statement of Jewish identity. This fashion spread
rapidly through the established religious format across the activated
network of the Jewish Diaspora, catalysed by external political factors

247

and events and driven by the rabbinic reforms, but also showing that
the network was fragile and ready for change.
Understanding change in ancient history can be aided by the
incorporation and further development of ideas in network theory.
Although this thesis has presented only a preliminary examination of
the ways in which network thinking can be applied to ancient historical
subjects, I hope it has shown something of the potential in this method
of analysis. My analysis of these three religious movements from the
broad viewpoint of network theory has allowed the building of an
alternative picture about the interactions that created them. I hope to
have shown that religious conversion and change cannot simply be
attributed

to

the

superiority

of

the

religious

innovation:

it

is

fundamentally driven by the social networks formed by the believers,


and these networks are embodied by the patterns in the epigraphic
data.

248

AP P EN DIX: C ON C ORDAN C E OF
EP IGRAP HIC MATERIAL
1.

2.

3.

Jupiter Dolichenus:
by catalogue number

249

by region

262

The jewish diaspora:


By catalogue number

275

By region

293

Theos Hypsistos:
By catalogue number

311

By region

319

249

Jupi ter Dol i ch en u s

by catalogue number
Publication/s
A. Pellegrino, Les cultes
de Jupiter Dolichnien
et de Jupiter
Hliopolitain Ostie, in:
G. M. Bellelli and U.
Bianchi (eds), Orientalia
Sacra Urbis Romae.
Dolichena et
Heliopolitana, Rom
1996, S. 563-583, S.
564, Fig. 3.
E. Sanzi, Sur une
inscription romaine en
rapport avec le culte
dolichnien, in
Orientalia Sacra Urbis
Romae. Dolichena et
Heliopolitana, S. 257 ff.
G. L. Irby-Massie,
Military Religion in
Roman Britain,
Leiden/Boston/Kln
1999, S. 278.
G. L. Irby-Massie,
Military Religion, S.
280.
G.L. Irby-Massie,
Military Religion, S.
279.
M. Hrig, E.
Schwertheim, CCID,
1987, 2
Hrig, Schwertheim
1987, 3
Hrig, Schwertheim
1987, 4
Hrig, Schwertheim
1987, 6
Hrig, Schwertheim
1987, 7
Hrig, Schwertheim
1987, 8
Hrig, Schwertheim
1987, 9

Location

Region

Type

Date

Ostia

Italia

stone

Rome

Italia

sarcophag
us

Eboracum (York)

Britannia

inscription 221 AD

Habitancum Risingham

Britannia

inscription

3rd century
AD

Maglona (old
Carlisle)

Britannia

altar

197 AD

Doliche - Dlk

Syria

altar

57/58 AD

Doliche - Dlk

Syria

slab

Pre-256 AD

Doliche - Dlk

Syria

slab

Pre-256 AD

Doliche - Dlk

Syria

Mid-1st
century AD

Doliche - Dlk

Syria

votive
triangle
votive
triangle

Zafer, nr. Antep

Syria

stele

Kekliktepe, nr.
Antep

Syria

stele

End 2nd, beg.


3rd century

250

AD
Hrig, Schwertheim
1987, 12
Hrig, Schwertheim
1987, 16
Hrig, Schwertheim
1987, 17
Hrig, Schwertheim
1987, 18
Hrig, Schwertheim
1987, 19
Hrig, Schwertheim
1987, 20
Hrig, Schwertheim
1987, 21

Antep (Doliche)

Syria

sealstone

21 AD

Tilhalit

Syria

stele

Syria

stele

1st century
AD

Syria

sculptural
group

Imperial

Syria

stele

First half 1st


century AD

Syria

column
drum

Syria

stele

Maras Germanikeia
between Maras &
Birecik
Kurdini Tepe, nr.
Alacakilise
Kurdini Tepe, nr.
Alacakilise
Gonca Dagi, nr.
Asagi Kalecik

First half 1st


century AD
1st century
BC/1st
century AD
1st century
BC/1st
century AD

Hrig, Schwertheim
1987, 22

Zeytintepe, nr.
Baspinar

Syria

stele

Hrig, Schwertheim
1987, 23

Zeytintepe, nr.
Baspinar

Syria

stele

Kurcuoglu

Syria

stele

Kurcuoglu

Syria

stele

Syria

stele

Syria

stele

Mastala

Syria

stele

Hierapolis Membidj

Syria

stele

Caesarea

Judea

block

2nd/3rd
century AD

altar

211 AD

altar

June-Oct 251
AD

altar

251-53 AD

stele

text

239 AD,
27/28 May

Hrig, Schwertheim
1987, 24
Hrig, Schwertheim
1987, 25
Hrig, Schwertheim
1987, 26
Hrig, Schwertheim
1987, 27
Hrig, Schwertheim
1987, 28
Hrig, Schwertheim
1987, 29
Hrig, Schwertheim
1987, 30
Hrig, Schwertheim
1987, 32
Hrig, Schwertheim
1987, 33
Hrig, Schwertheim
1987, 34
Hrig, Schwertheim
1987, 35
Hrig, Schwertheim
1987, 39
Hrig, Schwertheim
1987, 40
Hrig, Schwertheim
1987, 41
Hrig, Schwertheim
1987, 43
Hrig, Schwertheim
1987, 44
Hrig, Schwertheim
1987, 45
Hrig, Schwertheim
1987, 48

Khaltan, in Tal des


Afrin
Khaltan, in Tal des
Afrin

Mesopota
mia
Mesopota
mia
Mesopota
mia
Mesopota
mia
Mesopota
mia

Dura Europus
Dura Europus
Dura Europus
Dura Europus
Dura Europus

1st/2nd
century AD
1st/2nd
century AD
2nd century
AD

Beka'a

Syria

hand

Lebanon

Syria

hand

Comana
Cappadociae - Sar

Cappadoci
a

hand

1st century
AD

Asia Minor

Asia Minor hand

Asia Minor

Asia Minor hand

sculptural
group

Stobi

251

Mid-3rd
century AD

Hrig, Schwertheim
1987, 50
Hrig, Schwertheim
1987, 51
Hrig, Schwertheim
1987, 52
Hrig, Schwertheim
1987, 53
Hrig, Schwertheim
1987, 54
Hrig, Schwertheim
1987, 55
Hrig, Schwertheim
1987, 60
Hrig, Schwertheim
1987, 61
Hrig, Schwertheim
1987, 62
Hrig, Schwertheim
1987, 63
Hrig, Schwertheim
1987, 64
Hrig, Schwertheim
1987, 65
Hrig, Schwertheim
1987, 66
Hrig, Schwertheim
1987, 67
Hrig, Schwertheim
1987, 68
Hrig, Schwertheim
1987, 69
Hrig, Schwertheim
1987, 70
Hrig, Schwertheim
1987, 71
Hrig, Schwertheim
1987, 73
Hrig, Schwertheim
1987, 74
Hrig, Schwertheim
1987, 75
Hrig, Schwertheim
1987, 76
Hrig, Schwertheim
1987, 77
Hrig, Schwertheim
1987, 85
Hrig, Schwertheim
1987, 86
Hrig, Schwertheim
1987, 87
Hrig, Schwertheim

Augusta Traiana Stamovo


Augusta Traiana Stara Zagora
Augusta Traiana Stara Zagora

Thracia

slab

222-235 AD

Thracia

altar

212-217
(214) AD

Thracia

column

Thracia

hand

Cillae - Cerna Gora Thracia

base

202-211 AD

Haskovo

Thracia

hand

Moesia
Inferior

column

212, 27 Feb8th Apr. 217

tablet

218-222 AD

sculptural
group

222-235 AD

statuette

altar

198-209 AD

altar

212-222 AD

statuette

End 2nd, beg.


3rd century
AD

slab

altar

statuette

First half 3rd


century AD

hand

235-239 AD

altar

214 AD

votive
table

Beg. 3rd
century AD
2nd/3rd
century AD

Gorni Voden

btwn Noviodunum
and Troesmis Nikulitel
Troesmis Meidanchioi
Troesmis - Cerna
Histria
Histria or
Durostorum
Vicus Quintionis
(nr. Histria)
Tomis - Constanta

Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior

Moesia
Inferior
Moesia
Tropaeum Traiani
Inferior
Bezmer, prob. from Moesia
Durostorum
Inferior
Moesia
Bizone - Varna
Inferior
Moesia
Dionysopolis Inferior
Balcik
Muratu

Nikolaevka, nr.
Varna
Novae - Steklen
Novae - Steklen
Emporium
Piretensium Gorsko Kosovo
Dragoevo, near
Preslav
Vinimacium Kostolac
Pincum - Veliko
Gradiste
Karatas, near
Kladovo
Egeta - Brza

Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia

252

altar
altar

altar

altar

238-244 AD

sculptural
group

statuette

3rd century
AD

altar

post 211 AD

sculptural

1987, 89

Palanka

Superior

Hrig, Schwertheim
1987, 90

Egeta - Brza
Palanka

Moesia
Superior

Hrig, Schwertheim
1987, 91
Hrig, Schwertheim
1987, 92
Hrig, Schwertheim
1987, 93
Hrig, Schwertheim
1987, 94
Hrig, Schwertheim
1987, 95
Hrig, Schwertheim
1987, 97
Hrig, Schwertheim
1987, 103
Hrig, Schwertheim
1987, 104
Hrig, Schwertheim
1987, 105
Hrig, Schwertheim
1987, 106
Hrig, Schwertheim
1987, 107

Egeta - Brza
Palanka
Egeta - Brza
Palanka
Egeta - Brza
Palanka
Egeta - Brza
Palanka
Egeta - Brza
Palanka
Egeta - Brza
Palanka
Romulianum? Jasen
Romulianum? Jasen
Romulianum? Jasen
Romulianum? Jasen
Bononia/Jasen Junija Alba

Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior

Hrig, Schwertheim
1987, 108

Vidin

Moesia
Superior

Hrig, Schwertheim
1987, 109
Hrig, Schwertheim
1987, 110
Hrig, Schwertheim
1987, 111
Hrig, Schwertheim
1987, 112
Hrig, Schwertheim
1987, 113
Hrig, Schwertheim
1987, 114
Hrig, Schwertheim
1987, 115
Hrig, Schwertheim
1987, 116
Hrig, Schwertheim
1987, 120
Hrig, Schwertheim
1987, 121
Hrig, Schwertheim
1987, 122
Hrig, Schwertheim
1987, 123

Colonia Ulpia
Ratiaria - Arcar
Colonia Ulpia
Ratiaria - Arcar
Colonia Ulpia
Ratiaria - Arcar
Colonia Ulpia
Ratiaria - Arcar
Colonia Ulpia
Ratiaria - Kosava
nr. Arcar

Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior

Timacum Minus Ravna


Gracanica
Kumanovo

group
statue

1-1st half 2nd


century AD
or 2nd/3rd
century AD

statuette

statuette

statuette

2nd/3rd
century AD

statuette

tablet

Mid-3rd
century AD

relief

votive
triangle

3rd century
AD
3rd century
AD

tablet
tablet

statuette

statuette
statuette

2nd/3rd
century AD
End 2nd, beg.
3rd century
AD

altar

altar

altar

c. 300 AD

altar

Beg. 3rd
century AD

Moesia
Superior

altar

198-208 AD

Moesia
Superior

sculptural
group

Moesia
Superior
Moesia
Superior

votive
Beg. 3rd
inscription century AD
216 AD, 1st
altar
Nov

Metulum Josephsthal/Munjav Dalmatia


a
Arupium - Vitalj
Dalmatia
near Otocac

Beg. 3rd
century AD

altar

Beg. 3rd
century AD

column

Vrlika

Dalmatia

Salonae - Salone

Dalmatia

grave
inscription

253

Hrig, Schwertheim
1987, 124
Hrig, Schwertheim
1987, 125
Hrig, Schwertheim
1987, 126
Hrig, Schwertheim
1987, 130
Hrig, Schwertheim
1987, 131
Hrig, Schwertheim
1987, 132
Hrig, Schwertheim
1987, 133
Hrig, Schwertheim
1987, 134
Hrig, Schwertheim
1987, 135/6
Hrig, Schwertheim
1987, 137
Hrig, Schwertheim
1987, 138
Hrig, Schwertheim
1987, 139
Hrig, Schwertheim
1987, 140
Hrig, Schwertheim
1987, 141
Hrig, Schwertheim
1987, 142
Hrig, Schwertheim
1987, 143
Hrig, Schwertheim
1987, 144
Hrig, Schwertheim
1987, 145
Hrig, Schwertheim
1987, 146
Hrig, Schwertheim
1987, 147
Hrig, Schwertheim
1987, 148
Hrig, Schwertheim
1987, 149
Hrig, Schwertheim
1987, 150
Hrig, Schwertheim
1987, 151
Hrig, Schwertheim
1987, 152
Hrig, Schwertheim
1987, 153

Narona - Hama

Dalmatia

altar

193 AD

Japra - Majdaniste

Dalmatia

base

Prizren

Dalmatia

altar

Samum - Casei

Dacia

fragment

Samum - Casei

Dacia

altar

243 AD

Samum - Casei

Dacia

altar

224 AD

Ilisua, district
Bistrita-Nasaud

Dacia

statuette

End 2nd, beg.


3rd century
AD

Certiae - Romita

Dacia

altar

238-244 AD

Porolissum

Dacia

Buciumi

Dacia

altar

211-212 AD

Dacia

altar

167-180 AD

Dacia

statuette

Napoca - Cluj

Dacia

altar

198-208 AD

Napoca - Cluj

Dacia

altar

Potaissa - Turda

Dacia

votive
triangle

Potaissa - Turda

Dacia

altar

Domnestri,
municipality
Mariselu, area
Bistrita
Gherla, in area of
Cluj

Potaissa - Turda

Dacia

altar

End 2nd, beg.


3rd century
AD

Potaissa - Turda

Dacia

altar

Ampelum - Zlatna

Dacia

altar

Ampelum - Zlatna

Dacia

column

Ampelum - Zlatna

Dacia

column

End 2nd, beg.


3rd century
AD

Ampelum - Zlatna

Dacia

column

Ampelum - Zlatna

Dacia

altar

238-244 AD

Apulum - Alba Iulia Dacia

fragment

138-161 AD

Apulum - Alba Iulia Dacia

fragment

Apulum - Alba Iulia Dacia

column

End 2nd, beg.


3rd century

254

AD
Hrig, Schwertheim
1987, 154
Hrig, Schwertheim
1987, 155

Apulum - Alba Iulia Dacia

altar

Apulum - Alba Iulia Dacia

altar

198-211 AD

Hrig, Schwertheim
1987, 156

Kozlard,
Coslar/Apulum

Dacia

votive
table

End 2nd, beg.


3rd century
AD

Hrig, Schwertheim
1987, 157

Blaj/Apulum

Dacia

votive
relief

Hrig, Schwertheim
1987, 158

Sibiu
Apulum/Sarmizeget Dacia
usa

Micia - Vetel

Dacia

altar

Micia - Vetel

Dacia

altar

Sincrai,
deployment front
for Aquae

Dacia

altar

209-214. Feb.
211 AD

Sibiu - Sacadate

Dacia

altar

Dacia

altar

Dacia

altar

Dacia

altar

Dacia

relief

Catunele de Sus

Dacia

hand

Pojejena de Sus

Dacia

altar

Pre-132?

Racari, region of
Dolj

Dacia

votive
triangle

Amarastii de Jos

Dacia

statuette

First half 3rd


century AD
End 2nd
century AD

Desa, near Ratiaria Dacia

statuette

Sucidava

Dacia

altar

Dacia

hand

Hrig, Schwertheim
1987, 159
Hrig, Schwertheim
1987, 160
Hrig, Schwertheim
1987, 161
Hrig, Schwertheim
1987, 162
Hrig, Schwertheim
1987, 163
Hrig, Schwertheim
1987, 165
Hrig, Schwertheim
1987, 169
Hrig, Schwertheim
1987, 170
Hrig, Schwertheim
1987, 171
Hrig, Schwertheim
1987, 172
Hrig, Schwertheim
1987, 173
Hrig, Schwertheim
1987, 174
Hrig, Schwertheim
1987, 175
Hrig, Schwertheim
1987, 176
Hrig, Schwertheim
1987, 177
Hrig, Schwertheim
1987, 182
Hrig, Schwertheim
1987, 183
Hrig, Schwertheim
1987, 184
Hrig, Schwertheim
1987, 185
Hrig, Schwertheim
1987, 188
Hrig, Schwertheim
1987, 189

Tibiscum Caransebes
Ulpia Traiana Sarmizegetusa
Ulpia Traiana Sarmizegetusa
Ulpia Traiana Sarmizegetusa

Mychkovo (or else


Myszkow in
Ukraine!)
Tokod, near
Esztergom
Aquincum - Obuda
Aquincum?
Aquincum
Vetus Salina Adony
Vetus Salina Adony

255

Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior

End 2nd
century AD
Beg. 3rd
century AD

Pre-mid-3rd
century AD
Second half
2nd century
AD

altar

222-235 AD

base

Post-100 AD

sculptural
group

stele

228 AD

bust

hand

Hrig, Schwertheim
1987, 200

Gorsium Sarpentele

Pannonia
Inferior

Hrig, Schwertheim
1987, 201
Hrig, Schwertheim
1987, 202
Hrig, Schwertheim
1987, 204
Hrig, Schwertheim
1987, 205
Hrig, Schwertheim
1987, 206

Lussonium Komlod
Lussonium Komlod
Lussonium Komlod
Mursa - Klisa, near
Osijek
Acumincum - Stari
Slankamen

Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior

Hrig, Schwertheim
1987, 207

Acumincum - Stari
Slankamen

Pannonia
Inferior

Hrig, Schwertheim
1987, 208
Hrig, Schwertheim
1987, 209

Pannonia
Inferior
Pannonia
Inferior

Hrig, Schwertheim
1987, 217
Hrig, Schwertheim
1987, 218

Acumincum Surduk
Burgenae - Novi
Banovci
Sirmium Srijemska
Mitrovica
Sirmium Srijemska
Mitrovica
Carnuntum Pfaffenberg
Carnuntum Petronell

Hrig, Schwertheim
1987, 219

Carnuntum Petronell

Hrig, Schwertheim
1987, 220
Hrig, Schwertheim
1987, 221
Hrig, Schwertheim
1987, 222
Hrig, Schwertheim
1987, 223
Hrig, Schwertheim
1987, 224
Hrig, Schwertheim
1987, 226
Hrig, Schwertheim
1987, 228
Hrig, Schwertheim
1987, 229
Hrig, Schwertheim
1987, 230
Hrig, Schwertheim
1987, 231
Hrig, Schwertheim
1987, 232
Hrig, Schwertheim
1987, 233

Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum - Bad
Deutsch-Altenburg
Carnuntum - Bad
Deutsch-Altenburg
Carnuntum - Bad
Deutsch-Altenburg
Carnuntum - Bad
Deutsch-Altenburg

Hrig, Schwertheim
1987, 210
Hrig, Schwertheim
1987, 211

256

altar
votive
triangle
votive
triangle

Beg. 3rd
century AD,
or 202 AD
3rd century
AD
-

altar

altar

statuette

statuette

End 2nd, beg.


3rd century
AD

postament
s

base

Pannonia
Inferior

fragment

193-211 AD

Pannonia
Inferior

altar

block

128-138 AD

votive
table

altar

base

altar

183 AD

relief

2nd century
AD

base in
altar form

180-183 AD

statue

arm

votive
table

altar

Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior

sculptural
group
votive
table
votive
table
statuette

235-238 AD
3rd century
AD

Hrig, Schwertheim
1987, 234
Hrig, Schwertheim
1987, 235
Hrig, Schwertheim
1987, 237
Hrig, Schwertheim
1987, 238
Hrig, Schwertheim
1987, 239
Hrig, Schwertheim
1987, 240
Hrig, Schwertheim
1987, 241
Hrig, Schwertheim
1987, 242
Hrig, Schwertheim
1987, 253
Hrig, Schwertheim
1987, 263
Hrig, Schwertheim
1987, 265
Hrig, Schwertheim
1987, 266
Hrig, Schwertheim
1987, 267
Hrig, Schwertheim
1987, 268
Hrig, Schwertheim
1987, 269
Hrig, Schwertheim
1987, 270
Hrig, Schwertheim
1987, 271
Hrig, Schwertheim
1987, 272
Hrig, Schwertheim
1987, 273
Hrig, Schwertheim
1987, 274
Hrig, Schwertheim
1987, 275
Hrig, Schwertheim
1987, 276
Hrig, Schwertheim
1987, 277
Hrig, Schwertheim
1987, 282
Hrig, Schwertheim
1987, 283
Hrig, Schwertheim
1987, 284
Hrig, Schwertheim

Gerulata - Rusovce
Gerulata - Rusovce
Brigetio - Oszony
Brigetio - Oszony
Brigetio - Oszony
Brigetio - Oszony
Brigetio - Oszony
Brigetio - Oszony
Brigetio - Oszony
Mullendorf, near
Eisenstadt
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Emona
Praetorium
Latobicorum Trebnje
Praetorium
Latobicorum Trebnje
Aquae Balissae Daruvar
Aquae Balissae Daruvar
Poetovio - Ptuj
Poetovio - Ptuj
Poetovio - Ptuj
Lauriacum - Enns

257

Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior

altar

block

4th century
AD

slab

First half 3rd


century AD

fragment

relief

2nd/3rd
century AD

relief

statue base sculptural


group

base

Second half
2nd century
AD

altar

altar
altar

2nd/3rd
century AD
2nd/3rd
century AD

altar

bowl

fragment

altar

208 AD

sculptural
group

2nd/3rd
century AD

fragment

altar

Pannonia
Superior

altar

Pannonia
Superior

votive
138 AD, 1st
inscription Nov

Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Noricum

altar

198-208 AD

altar

altar

189 AD

stone

207 AD

altar
bowl

3rd century
AD
-

1987, 285
Hrig, Schwertheim
1987, 286
Hrig, Schwertheim
1987, 287
Hrig, Schwertheim
1987, 289
Hrig, Schwertheim
1987, 291
Hrig, Schwertheim
1987, 292
Hrig, Schwertheim
1987, 293
Hrig, Schwertheim
1987, 294
Hrig, Schwertheim
1987, 295
Hrig, Schwertheim
1987, 298
Hrig, Schwertheim
1987, 299
Hrig, Schwertheim
1987, 300
Hrig, Schwertheim
1987, 301
Hrig, Schwertheim
1987, 302
Hrig, Schwertheim
1987, 303
Hrig, Schwertheim
1987, 304
Hrig, Schwertheim
1987, 305
Hrig, Schwertheim
1987, 306
Hrig, Schwertheim
1987, 307
Hrig, Schwertheim
1987, 308
Hrig, Schwertheim
1987, 309
Hrig, Schwertheim
1987, 310
Hrig, Schwertheim
1987, 311
Hrig, Schwertheim
1987, 312
Hrig, Schwertheim
1987, 313
Hrig, Schwertheim
1987, 314
Hrig, Schwertheim
1987, 315
Hrig, Schwertheim
1987, 316

Lauriacum - Enns

Noricum

vase

Ovilava - Wels

Noricum

votive slab

beg. 3rd
century AD

Bedaium

Noricum

text

Noricum

statuette

Noricum

sculptural
group

Noricum

statuette

Locus Felicis (?) Mauer an der Url


Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url

258

Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum

votive
triangle
votive
triangle
votive
palm
votive
palm
votive
palm
votive
palm
votive
palm
votive
votive
sheet
votive
sheet

Noricum

votive

Period of
Commodus?

Noricum

votive

Noricum

votive
sheet

Noricum

votive

Noricum

votive

Noricum

votive

Noricum

votive

Noricum
Noricum
Noricum
Noricum

votive
sheet
votive
palm
votive
sheet
votive
sheet

Hrig, Schwertheim
1987, 317
Hrig, Schwertheim
1987, 318
Hrig, Schwertheim
1987, 327

Locus Felicis (?) Mauer an der Url


Locus Felicis (?) Mauer an der Url
Trigisamum Traismauer

Hrig, Schwertheim
1987, 328

Trigisamum Traismauer

Noricum

votive
triangle

260/70 AD/
end 2nd
century AD

Virunum - Zollfeld

Noricum

slab

189 AD

Virunum - Zollfeld

Noricum

round altar -

Virunum - Zollfeld

Noricum

round altar -

Virunum - Zollfeld

Noricum

fragment

Virunum - Zollfeld

Noricum

altar

Hrig, Schwertheim
1987, 342

Virunum - Zollfeld

Noricum

base

Hrig, Schwertheim
1987, 344

Virunum - Zollfeld

Noricum

altar

Virunum - Zollfeld

Noricum

altar

Noricum

votive
inscription

Noricum

relief

Feldkirchen

Noricum

fragment

Colatio Windischgraz

Noricum

slab

Rome

Italia

altar

150 AD

Rome

Italia

altar

150 AD

Hrig, Schwertheim
1987, 362

Rome

Italia

altar

Hrig, Schwertheim
1987, 363

Rome

Italia

relief

Hrig, Schwertheim
1987, 364

Rome

Italia

relief

Hrig, Schwertheim
1987, 365

Rome

Italia

relief

Rome

Italia

relief

Rome

Italia

relief

Rome

Italia

slab

Hrig, Schwertheim
1987, 330
Hrig, Schwertheim
1987, 331
Hrig, Schwertheim
1987, 332
Hrig, Schwertheim
1987, 333
Hrig, Schwertheim
1987, 335

Hrig, Schwertheim
1987, 345
Hrig, Schwertheim
1987, 346
Hrig, Schwertheim
1987, 347
Hrig, Schwertheim
1987, 349
Hrig, Schwertheim
1987, 350
Hrig, Schwertheim
1987, 356
Hrig, Schwertheim
1987, 357

Hrig, Schwertheim
1987, 367
Hrig, Schwertheim
1987, 368
Hrig, Schwertheim

Horzendorf, near
St. Veit
Lamprechtskogel,
near Waisenberg

259

Noricum
Noricum
Noricum

votive
sheet
votive
sheet
votive
triangle

Second half
3rd century
AD
Piccottinni
places
between 197207 AD
Mid-2nd
century AD

Second half
2nd century
AD
183 AD, 1
March
Second half
2nd century
AD
Second half
2nd century
AD

1987, 370
Hrig, Schwertheim
1987, 371
Hrig, Schwertheim
1987, 372

Rome

Italia

relief

Late 2nd
century AD

Rome

Italia

altar

198-209 AD
First half 3rd
century AD

Hrig, Schwertheim
1987, 373

Rome

Italia

slab

Hrig, Schwertheim
1987, 375

Rome

Italia

slab

Rome

Italia

slab

Rome

Italia

slab

Rome

Italia

base

Rome

Italia

round base -

Rome

Italia

statue

Rome

Italia

slab

First half 3rd


century AD

Rome

Italia

fragment

222-235 AD

Rome

Italia

altar

244 AD, 10
Oct.

Rome

Italia

base

244 AD

Rome

Italia

slab

Rome

Italia

relief

Rome

Italia

fragments

Rome

Italia

inscription -

Rome

Italia

inscription -

Rome

Italia

column

Rome

Italia

relief

Rome

Italia

slab

191 AD, 31
July

Rome

Italia

slab

191/2 AD

Rome

Italia

sculptural
group

End 2nd
century AD

Rome

Italia

head

Pre-212 AD

Rome

Italia

slab

198-209 AD

Rome

Italia

slab

218 AD

Hrig, Schwertheim
1987, 376
Hrig, Schwertheim
1987, 377
Hrig, Schwertheim
1987, 378
Hrig, Schwertheim
1987, 379
Hrig, Schwertheim
1987, 380
Hrig, Schwertheim
1987, 381
Hrig, Schwertheim
1987, 382
Hrig, Schwertheim
1987, 383
Hrig, Schwertheim
1987, 384
Hrig, Schwertheim
1987, 385
Hrig, Schwertheim
1987, 386
Hrig, Schwertheim
1987, 401
Hrig, Schwertheim
1987, 402
Hrig, Schwertheim
1987, 403
Hrig, Schwertheim
1987, 404
Hrig, Schwertheim
1987, 405
Hrig, Schwertheim
1987, 408
Hrig, Schwertheim
1987, 409
Hrig, Schwertheim
1987, 410
Hrig, Schwertheim
1987, 411
Hrig, Schwertheim
1987, 414
Hrig, Schwertheim
1987, 415

260

Second half
3rd century
AD
First half 3rd
century AD

Second half
3rd century
AD
3rd century
AD
-

Hrig, Schwertheim
1987, 416
Hrig, Schwertheim
1987, 418
Hrig, Schwertheim
1987, 419
Hrig, Schwertheim
1987, 420
Hrig, Schwertheim
1987, 421
Hrig, Schwertheim
1987, 422
Hrig, Schwertheim
1987, 423
Hrig, Schwertheim
1987, 424
Hrig, Schwertheim
1987, 425
Hrig, Schwertheim
1987, 426
Hrig, Schwertheim
1987, 427
Hrig, Schwertheim
1987, 428
Hrig, Schwertheim
1987, 429
Hrig, Schwertheim
1987, 430
Hrig, Schwertheim
1987, 431
Hrig, Schwertheim
1987, 432
Hrig, Schwertheim
1987, 433
Hrig, Schwertheim
1987, 434
Hrig, Schwertheim
1987, 440
Hrig, Schwertheim
1987, 441
Hrig, Schwertheim
1987, 442
Hrig, Schwertheim
1987, 443
Hrig, Schwertheim
1987, 444
Hrig, Schwertheim
1987, 445
Hrig, Schwertheim
1987, 446
Hrig, Schwertheim
1987, 447
Hrig, Schwertheim
1987, 448
Hrig, Schwertheim
1987, 449

218 AD, 14
March,
3rd century
AD
2nd/3rd
century AD

Rome

Italia

altar

Rome

Italia

altar

Rome

Italia

relief

Rome

Italia

relief

Rome

Italia

slab

201 AD, 11
Nov

Rome

Italia

inscription 200-208 AD

Rome

Italia

base

Rome

Italia

inscription -

Rome

Italia

inscription 270-275 AD?

Rome

Italia

slab

Rome

Italia

Rome

Italia

altar

Rome

Italia

altar

Rome

Italia

tablet

Rome - Tivoli

Italia

slab

Rome

Italia

block

196 AD, 29
June

Rome

Italia

round base Pre-212 AD

Rome

Italia

inscription 92 AD?

Ostia

Italia

slab

186 AD

Ostia

Italia

slab

191/2 AD

Ostia

Italia

slab

Mid-2nd
century AD

Ostia

Italia

tablet

Tusculum - Frascati Italia

altar

Tergeste - Trieste

Italia

altar

Tergeste - Trieste

Italia

altar

Aquileia

Italia

altar

Italia

statuette

3rd century
AD

Italia

badge

185-192 AD

near Iulia
Concordia (Lison)
near Iulia
Concordia

261

Hrig, Schwertheim
1987, 450
Hrig, Schwertheim
1987, 451
Hrig, Schwertheim
1987, 452
Hrig, Schwertheim
1987, 453
Hrig, Schwertheim
1987, 454
Hrig, Schwertheim
1987, 455
Hrig, Schwertheim
1987, 456
Hrig, Schwertheim
1987, 457
Hrig, Schwertheim
1987, 458
Hrig, Schwertheim
1987, 459
Hrig, Schwertheim
1987, 460
Hrig, Schwertheim
1987, 461
Hrig, Schwertheim
1987, 462
Hrig, Schwertheim
1987, 463
Hrig, Schwertheim
1987, 465
Hrig, Schwertheim
1987, 466
Hrig, Schwertheim
1987, 467
Hrig, Schwertheim
1987, 468
Hrig, Schwertheim
1987, 475
Hrig, Schwertheim
1987, 476
Hrig, Schwertheim
1987, 477
Hrig, Schwertheim
1987, 480
Hrig, Schwertheim
1987, 481
Hrig, Schwertheim
1987, 485
Hrig, Schwertheim
1987, 491
Hrig, Schwertheim
1987, 493
Hrig, Schwertheim
1987, 497

Padua

Italia

altar

Ateste - Este

Italia

tablet

Severan

Atria - Adria

Italia

tablet

222-235 AD

Brixia - Brescia

Italia

altar

Bononia - Bologna

Italia

slab

Ravenna

Italia

altar

193-217 AD

Ravenna

Italia

inscription -

Caesena - Cesena

Italia

relief

Ariminum - Rimini

Italia

altar

Ariminum - Rimini

Italia

altar

2nd/3rd
century AD
2nd/3rd
century AD

Histonium - Il vasto
Italia
d'Aimone

column

Aecae - Troja

Italia

Second half
2nd century
AD

Tarracina Terracine

Italia

statue

Misenum

Italia

fragment

Puteoli - Pozzuoli

Italia

slab

Naples

Italia

slab

Naples

Italia

relief

Severan

Turris Libisonis Porto Torres,


Sardinia

Italia

inscription 211/212 AD

Aquileia - Aalen

Raetia

votive
triangle

Aquileia - Aalen

Raetia

block

Faimingen

Raetia

relief

Raetia

tablet

Severan

Raetia

tablet

Severan

Raetia

altar

163 AD, 11
April

altar

bull

222-235 AD

altar

Statio Vetonianis Pfunz


Statio Vetonianis Pfunz
Serviodurum Straubing
Zugmantel
Zugmantel
Kastell Alteburg

262

Germania
Superior
Germania
Superior
Germania
Superior

Beg. 3rd
century AD
End 2nd
century AD

Hrig, Schwertheim
1987, 499
Hrig, Schwertheim
1987, 500
Hrig, Schwertheim
1987, 501
Hrig, Schwertheim
1987, 508
Hrig, Schwertheim
1987, 509
Hrig, Schwertheim
1987, 511
Hrig, Schwertheim
1987, 512

Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior

Saalburg
Saalburg
Saalburg
Saalburg
Saalburg
Nida Heddernheim
Nida Heddernheim

altar

End 2nd
century AD

altar

altar

altar

205 AD

altar

votive
triangle
votive
triangle

Hrig, Schwertheim
1987, 514

Nida Heddernheim

Hrig, Schwertheim
1987, 515
Hrig, Schwertheim
1987, 516
Hrig, Schwertheim
1987, 517
Hrig, Schwertheim
1987, 518

Nida Heddernheim
Nida Heddernheim
Nida Heddernheim
Nida Heddernheim

Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior

votive
sheet
votive
sheet
votive
sheet
votive
sheet

Hrig, Schwertheim
1987, 519

Nida Heddernheim

Germania
Superior

altar

Hrig, Schwertheim
1987, 520

Nida Heddernheim

Germania
Superior

hand

Hrig, Schwertheim
1987, 521
Hrig, Schwertheim
1987, 522
Hrig, Schwertheim
1987, 523

Nida Heddernheim
Nida Heddernheim

Germania
Superior
Germania
Superior

Mogontiacum Mainz

Germania
Superior

Hrig, Schwertheim
1987, 524
Hrig, Schwertheim
1987, 525
Hrig, Schwertheim
1987, 526
Hrig, Schwertheim
1987, 527
Hrig, Schwertheim
1987, 529
Hrig, Schwertheim
1987, 530
Hrig, Schwertheim
1987, 531
Hrig, Schwertheim
1987, 532
Hrig, Schwertheim
1987, 533
Hrig, Schwertheim

Mogontiacum Mainz
Mogontiacum Mainz
Aquae Mattiacorum
- Wiesbaden

Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania

Gross-Krotzenburg
Stockstadt
Stockstadt
Stockstadt
Stockstadt
Stockstadt
Stockstadt

263

votive
sheet

2nd century
AD
2nd century
AD
Second half
2nd century
AD
Second half
2nd century
AD

tablet

altar

altar

Beg. 3rd
century AD

altar

211-217 AD

altar

217 AD, 23
May

votive
table

194 AD

altar

211 AD

inscription altar

altar

214 AD

base

altar

altar

211-217 AD

1987, 534
Hrig, Schwertheim
1987, 536
Hrig, Schwertheim
1987, 537
Hrig, Schwertheim
1987, 538
Hrig, Schwertheim
1987, 539
Hrig, Schwertheim
1987, 540
Hrig, Schwertheim
1987, 541
Hrig, Schwertheim
1987, 547
Hrig, Schwertheim
1987, 549
Hrig, Schwertheim
1987, 550
Hrig, Schwertheim
1987, 554
Hrig, Schwertheim
1987, 555
Hrig, Schwertheim
1987, 556
Hrig, Schwertheim
1987, 557
Hrig, Schwertheim
1987, 559
Hrig, Schwertheim
1987, 560
Hrig, Schwertheim
1987, 561
Hrig, Schwertheim
1987, 562
Hrig, Schwertheim
1987, 564
Hrig, Schwertheim
1987, 565
Hrig, Schwertheim
1987, 567
Hrig, Schwertheim
1987, 572
Hrig, Schwertheim
1987, 573
Hrig, Schwertheim
1987, 574
Hrig, Schwertheim
1987, 575
Hrig, Schwertheim

Superior
Germania
Obernburg
Superior
Germania
Obernburg
Superior
Aschaffenburg
Germania
(Obernburg)
Superior
Germania
Portus? - Pforzheim
Superior
Germania
Grinario - Kongen
Superior
Vetera Germania
Furstenberg, near
Inferior
Xanten
Colonia
Germania
Agrippinensis Inferior
Koln
Germania
Bonn
Inferior
Rigomagus Germania
Remagen
Inferior
Antonine Wall Britannia
Croy Hill
Blatobulgium Britannia
Birrens
Castra
Exploratorum Britannia
Netherby
Habitancum Britannia
Risingham

264

207 AD

altar

206 AD

altar

191 AD

inscription Pre-185 AD
inscription altar

228 AD

tablet

211 AD

statuette

altar

250 AD

relief

Beg. reign of
Commodus

altar

altar

altar

altar

Britannia

altar

Britannia

altar

Britannia

inscription

Early 3rd
century AD

Britannia

altar

138-161 AD

Britannia

altar

Beg. 3rd
century AD

Britannia

relief

Britannia

altar

235-238 AD

Britannia

altar

Britannia

inscription -

Britannia

base

Britannia

altar

Banna - Bew Castle Britannia


Aesica (Hadrian's
Wall) - Great
Chesters
Aesica (Hadrian's
Wall) - Great
Chesters
Cilurnum
(Hadrian's Wall) Chesters
Condercum
(Hadrian's Wall) Benwell
Corstopitum Corbridge on Tyne
Corstopitum Corbridge on Tyne
Camboglanna Birdoswald
Magnis (Hadrian's
Wall) - Carvoran
Magis Piercebridge
Magis Piercebridge
Magis - Gainford

altar

First half 3rd


century AD
Merlat puts it

1987, 576
Hrig, Schwertheim
1987, 577
Hrig, Schwertheim
1987, 582
Hrig, Schwertheim
1987, 586
Hrig, Schwertheim
1987, 596
Hrig, Schwertheim
1987, 601
Hrig, Schwertheim
1987, 602
Hrig, Schwertheim
1987, 609
Hrig, Schwertheim
1987, 615
Hrig, Schwertheim
1987, 616
Hrig, Schwertheim
1987, 620
Hrig, Schwertheim
1987, 621
Hrig, Schwertheim
1987, 622
Hrig, Schwertheim
1987, 623
Hrig, Schwertheim
1987, 624
Hrig, Schwertheim
1987, 625
Hrig, Schwertheim
1987, 626
Hrig, Schwertheim
1987, 627
Hrig, Schwertheim
1987, 628
Hrig, Schwertheim
1987, 629

Voreda - Plumpton
Wall (Old Penrith)
Branodunum Brancaster
Isca Silurum Caerleon
Ager Morinorum Halinghem
Between Antibes
and Vallauris
Massilia Marseilles
Tarraconensis Saldanha
Leptis Magna Lepcis
Thanadassa - Ain
Wif
Lambaesis Lambese
Lambaesis Lambese
Lambaesis Lambese
Lambaesis Lambese
Lambaesis Lambese
Lambaesis Lambese
Lambaesis Lambese
Lambaesis Lambese
Lambaesis Lambese
Lambaesis Lambese

Hrig, Schwertheim
1987, 630
Hrig, Schwertheim
1987, 637
Hrig, Schwertheim
1987, 638
Hrig, Schwertheim
1987, 639
Hrig, Schwertheim
1987, 640
LAnne Epigraphique
1988, 962
LAnne Epigraphique
1990, 785
LAnne Epigraphique
1991, 500

at 217 AD
Britannia

inscription 120-160 AD

Britannia

rim of jug

Britannia

altar

161-169 AD

Gallia

base

Gallia

inscription -

Gallia

statuette

Hispania

altar

Africa

altar

Africa

altar

Africa

slab

Africa

inscription Post-253 AD

Africa

altar

Post-253 AD

Africa

altar

Post-253 AD

Africa

altar

222-235 AD

Africa

base

Africa

altar

222-238 AD

Africa

altar

222-238 AD

Africa

altar

Africa

inscription -

Lambaesis Lambese

Africa

altar

Second half
3rd century
AD

Minta Sarmizegetusa

Dacia

column

Gezmisaza

Dacia

altar

Carnuntum Petronell
Praetorium
Latobicorum Trebnje

Pannonia
Superior

altar

Pannonia
Superior

slab

196-197 AD

Sibiu - Sacadate

Dacia

altar

Wallsee

Noricum

altar

Campania
(Minturnes?)

Italia

tablet

265

Between 11th
April 208-210
Between 210211 AD
125-126 AD

LAnne Epigraphique
1997, 1244
LAnne Epigraphique
1998, 183
LAnne Epigraphique
1998, 273
LAnne Epigraphique
1998, 1143
LAnne Epigraphique
1998, 1144
LAnne Epigraphique
1998, 1156
LAnne Epigraphique
1998, 1157
LAnne Epigraphique
1998, 1158
LAnne Epigraphique
1998, 1159
LAnne Epigraphique
1998, 1160
LAnne Epigraphique
1998, 1430
LAnne Epigraphique
1999, 1374
LAnne Epigraphique
1999, 1784
LAnne Epigraphique
2001, 1706
LAnne Epigraphique
2001, 1707
LAnne Epigraphique
2001, 1733
LAnne Epigraphique
2002, 1011
SEG 1989, 1586
SEG 1998, 1871
V. Najdenova, Jupiter
Dolichenus in lower
Moesia and Thrace
ANRW II, 18, 2, Berlin
1989, S. 1362-1396, S.
1386f.

Neviodunum Krsko

Pannonia
Superior

altar

Rome

Italia

vase

mid-2nd/mid3rd century
AD

Ostia

Italia

slab

Sacidava, near
Rasova & Dunareni
Sacidava, near
Rasova & Dunareni
Balaklawa, 10km s
of Sebastopol
Balaklawa, 10km s
of Sebastopol
Balaklawa, 10km s
of Sebastopol
Balaklawa, 10km s
of Sebastopol
Balaklawa, 10km s
of Sebastopol
Khirbet Khalid,
HierapolisBambyke

Moesia
Inferior
Moesia
Inferior
Chersonisu
s Taurica
Chersonisu
s Taurica
Chersonisu
s Taurica
Chersonisu
s Taurica
Chersonisu
s Taurica

altar

222-235 AD

slab

2nd/3rd
century AD

slab

139-161 AD

altar

altar

table

bowl

Syria

base

2nd/3rd
century AD

Cabyle

Thracia

slab

Reign of
Septimius
Severus

Melleus?

Africa

slab

Porolissum

Dacia

statuette

reign of
Gordian

Porolissum

Dacia

altar

238-244 AD

Novae - Svistov

Moesia
Inferior

slab

Belgia

block

2nd/3rd
century AD

Syria
Syria

seals
altar

200-250 AD

Moesia
Inferior

gravestone -

Atuatuca
Tungrorum Tongres
Nikopolis
Doliche?

Histria

266

Jupi ter Dol i ch en u s

by REGION
Region
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Asia Minor
Asia Minor
Belgia
Britannia
Britannia
Britannia

Location
Lambaesis
Lambese
Lambaesis
Lambese
Lambaesis
Lambese
Lambaesis
Lambese
Lambaesis
Lambese
Lambaesis
Lambese
Lambaesis
Lambese
Lambaesis
Lambese
Lambaesis
Lambese
Lambaesis
Lambese

Type

Publication/s
Hrig, Schwertheim
slab
125-126 AD
1987, 620
Hrig, Schwertheim
inscription Post-253 AD
1987, 621
Hrig, Schwertheim
altar
Post-253 AD
1987, 622
Hrig, Schwertheim
altar
Post-253 AD
1987, 623
Hrig, Schwertheim
altar
222-235 AD
1987, 624
Hrig, Schwertheim
base
1987, 625
Hrig, Schwertheim
altar
222-238 AD
1987, 626
Hrig, Schwertheim
altar
222-238 AD
1987, 627
Hrig, Schwertheim
altar
1987, 628
Hrig, Schwertheim
inscription 1987, 629
Second half
Lambaesis Hrig, Schwertheim
altar
3rd century
Lambese
1987, 630
AD
Between
Leptis Magna Hrig, Schwertheim
altar
11th April
Lepcis
1987, 615
208-210
LAnne Epigraphique
Melleus?
slab
1999, 1784
Between
Hrig, Schwertheim
Thanadassa - Ain Wif altar
210-211 AD 1987, 616
Hrig, Schwertheim
Asia Minor
hand
1987, 44
Hrig, Schwertheim
Asia Minor
hand
1987, 45
Atuatuca Tungrorum
2nd/3rd
LAnne Epigraphique
block
- Tongres
century AD 2002, 1011
Aesica (Hadrian's
Hrig, Schwertheim
Wall) - Great
altar
1987, 560
Chesters
Aesica (Hadrian's
Hrig, Schwertheim
Wall) - Great
altar
1987, 561
Chesters
Antonine Wall - Croy
Beg. reign
Hrig, Schwertheim
relief
Hill
of
1987, 554
-

267

Date

Commodus
Britannia
Britannia
Britannia
Britannia
Britannia
Britannia
Britannia
Britannia
Britannia

Banna - Bew Castle


Blatobulgium Birrens
Branodunum Brancaster
Camboglanna Birdoswald
Castra Exploratorum
- Netherby
Cilurnum (Hadrian's
Wall) - Chesters
Condercum
(Hadrian's Wall) Benwell
Corstopitum Corbridge on Tyne
Corstopitum Corbridge on Tyne

altar

altar

rim of jug

altar

235-238 AD

altar

inscription

Early 3rd
century AD

altar

138-161 AD

altar

Beg. 3rd
century AD

relief

Britannia

Eboracum (York)

inscription 221 AD

Britannia

Habitancum Risingham

inscription 3rd C AD

Britannia
Britannia

Habitancum Risingham
Isca Silurum Caerleon

altar

altar

161-169 AD

altar

Merlat puts
it in 217 AD

Britannia

Magis - Gainford

Britannia

Magis - Piercebridge inscription -

Britannia

Magis - Piercebridge base

First half 3rd


century AD

Britannia

Maglona (old
Carlisle)

altar

197 AD

altar

Britannia
Britannia
Cappadocia
Chersonisus
Taurica
Chersonisus
Taurica
Chersonisus
Taurica
Chersonisus
Taurica
Chersonisus
Taurica
Dacia

Magnis (Hadrian's
Wall) - Carvoran
Voreda - Plumpton
Wall (Old Penrith)
Comana
Cappadociae - Sar
Balaklawa, 10km s
of Sebastopol
Balaklawa, 10km s
of Sebastopol
Balaklawa, 10km s
of Sebastopol
Balaklawa, 10km s
of Sebastopol
Balaklawa, 10km s
of Sebastopol
Amarastii de Jos

inscription 120-160
hand

1st century
AD

slab

139-161 AD

altar

altar

table

bowl

statuette

End 2nd

268

Hrig, Schwertheim
1987, 559
Hrig, Schwertheim
1987, 555
Hrig, Schwertheim
1987, 582
Hrig, Schwertheim
1987, 572
Hrig, Schwertheim
1987, 556
Hrig, Schwertheim
1987, 562
Hrig, Schwertheim
1987, 564
Hrig, Schwertheim
1987, 565
Hrig, Schwertheim
1987, 567
G. L. Irby-Massie,
Military Religion in
Roman Britain, 1999, S.
278.
G. L. Irby-Massie,
Military Religion, S.
280.
Hrig, Schwertheim
1987, 557
Hrig, Schwertheim
1987, 586
Hrig, Schwertheim
1987, 576
Hrig, Schwertheim
1987, 574
Hrig, Schwertheim
1987, 575
G.L. Irby-Massie,
Military Religion, S.
279.
Hrig, Schwertheim
1987, 573
Hrig, Schwertheim
1987, 577
Hrig, Schwertheim
1987, 43
LAnne Epigraphique
1998, 1156
LAnne Epigraphique
1998, 1157
LAnne Epigraphique
1998, 1158
LAnne Epigraphique
1998, 1159
LAnne Epigraphique
1998, 1160
Hrig, Schwertheim

century AD
Dacia

Ampelum - Zlatna

altar

Dacia

Ampelum - Zlatna

column

End 2nd, beg.


Hrig, Schwertheim
3rd century
1987, 148
AD
Hrig, Schwertheim
1987, 149
Hrig, Schwertheim
238-244 AD
1987, 150
Hrig, Schwertheim
138-161 AD
1987, 151
Hrig, Schwertheim
1987, 152
End 2nd, beg.
Hrig, Schwertheim
3rd century
1987, 153
AD
Hrig, Schwertheim
1987, 154
Hrig, Schwertheim
198-211 AD
1987, 155
Hrig, Schwertheim
1987, 157
Hrig, Schwertheim
211-212 AD
1987, 137
Hrig, Schwertheim
1987, 171
Hrig, Schwertheim
238-244 AD
1987, 134
Hrig, Schwertheim
1987, 175

Dacia

Ampelum - Zlatna

column

Dacia

Ampelum - Zlatna

column

Dacia

Ampelum - Zlatna

altar

Dacia

Apulum - Alba Iulia

fragment

Dacia

Apulum - Alba Iulia

fragment

Dacia

Apulum - Alba Iulia

column

Dacia

Apulum - Alba Iulia

altar

Dacia

Apulum - Alba Iulia

altar

Dacia

Blaj/Apulum

votive
relief

Dacia

Buciumi

altar

Dacia

Catunele de Sus

hand

Dacia

Certiae - Romita

altar

Dacia

Desa, near Ratiaria

statuette

Dacia

Domnestri,
municipality
Mariselu, area
Bistrita

altar

167-180 AD

Dacia

Gezmisaza

altar

Dacia

Gherla, in area of
Cluj

statuette

Dacia

Ilisua, district
Bistrita-Nasaud

statuette

Dacia

Kozlard,
Coslar/Apulum

votive
table

Dacia

Micia - Vetel

altar

Dacia

Micia - Vetel

altar

column

Dacia
Dacia
Dacia

Minta Sarmizegetusa
Mychkovo (or else
Myszkow in
Ukraine!)
Napoca - Cluj

hand
altar

269

1987, 174
Hrig, Schwertheim
1987, 146
Hrig, Schwertheim
1987, 147

End 2nd, beg.


3rd century
AD
End 2nd, beg.
3rd century
AD

second half
2nd century
AD
198-208 AD

Hrig, Schwertheim
1987, 138
Hrig, Schwertheim
1987, 638
Hrig, Schwertheim
1987, 139
Hrig, Schwertheim
1987, 133
Hrig, Schwertheim
1987, 156
Hrig, Schwertheim
1987, 159
Hrig, Schwertheim
1987, 160
Hrig, Schwertheim
1987, 637
Hrig, Schwertheim
1987, 177
Hrig, Schwertheim

Dacia

Napoca - Cluj

altar

Dacia

Pojejena de Sus

altar

Pre-132?

Dacia

Porolissum

Dacia

Porolissum

statuette

Reign of
Gordian

Dacia

Porolissum

altar

238-244 AD

Dacia

Potaissa - Turda

votive
triangle

Dacia

Potaissa - Turda

altar

Dacia

Potaissa - Turda

altar

Dacia

Potaissa - Turda

altar

Dacia

Racari, region of
Dolj

votive
triangle

Dacia

Samum - Casei

fragment

Dacia

Samum - Casei

altar

Dacia

Samum - Casei

altar

Dacia

Sibiu - Sacadate

altar

Dacia

Sibiu - Sacadate

altar

Dacia
Dacia
Dacia
Dacia
Dacia
Dacia
Dacia
Dalmatia

Sibiu
Apulum/Sarmizegetu ?
sa
Sincrai, deployment
altar
front for Aquae
Sucidava

altar

Tibiscum altar
Caransebes
Ulpia Traiana altar
Sarmizegetusa
Ulpia Traiana altar
Sarmizegetusa
Ulpia Traiana relief
Sarmizegetusa
Arupium - Vitalj near
column
Otocac
base

End 2nd, beg.


Hrig, Schwertheim
3rd century
1987, 144
AD
Hrig, Schwertheim
1987, 145
First half 3rd Hrig, Schwertheim
century AD 1987, 173
Hrig, Schwertheim
1987, 130
Hrig, Schwertheim
243 AD
1987, 131
Hrig, Schwertheim
224 AD
1987, 132
Hrig, Schwertheim
1987, 162
LAnne Epigraphique
1988, 962
209-214.
Feb. 211 AD
Before mid3rd century
AD
End 2nd
century AD
Beg. 3rd
century AD
-

Dalmatia

Japra - Majdaniste

Dalmatia

Metulum altar
Josephsthal/Munjava

Beg. 3rd
century AD

Dalmatia

Narona - Hama

altar

193 AD

Dalmatia

Prizren

altar

270

1987, 140
Hrig, Schwertheim
1987, 141
Hrig, Schwertheim
1987, 172
Hrig, Schwertheim
1987, 135/6
LAnne Epigraphique
2001, 1706
LAnne Epigraphique
2001, 1707
Hrig, Schwertheim
1987, 142
Hrig, Schwertheim
1987, 143

Hrig, Schwertheim
1987, 158
Hrig, Schwertheim
1987, 161
Hrig, Schwertheim
1987, 176
Hrig, Schwertheim
1987, 163
Hrig, Schwertheim
1987, 165
Hrig, Schwertheim
1987, 169
Hrig, Schwertheim
1987, 170
Hrig, Schwertheim
1987, 121
Hrig, Schwertheim
1987, 125
Hrig, Schwertheim
1987, 120
Hrig, Schwertheim
1987, 124
Hrig, Schwertheim

Dalmatia

Salonae - Salone

grave
inscription

Dalmatia

Vrlika

Gallia
Gallia
Gallia
Germania
Inferior
Germania
Inferior
Germania
Inferior
Germania
Inferior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior

Ager Morinorum base


Halinghem
Between Antibes and
inscription Vallauris
Massilia - Marseilles statuette

Bonn

statuette

tablet

211 AD

altar

250 AD

altar

228 AD

votive
table

194 AD

altar

191 AD

Colonia
Agrippinensis - Koln
Rigomagus Remagen
Vetera Furstenberg, near
Xanten
Aquae Mattiacorum Wiesbaden
Aschaffenburg
(Obernburg)
Grinario - Kongen

inscription -

Gross-Krotzenburg

altar

211 AD

Kastell Alteburg

altar

altar

Beg. 3rd
century AD

altar

211-217 AD

altar

217 AD, 23
May

Mogontiacum Mainz
Mogontiacum Mainz
Mogontiacum Mainz
Nida - Heddernheim
Nida - Heddernheim
Nida - Heddernheim
Nida - Heddernheim
Nida - Heddernheim
Nida - Heddernheim
Nida - Heddernheim

votive
triangle
votive
triangle
votive
sheet
votive
sheet
votive
sheet
votive
sheet
votive
sheet

Germania
Superior

Nida - Heddernheim altar

Germania
Superior

Nida - Heddernheim hand

Germania

Nida - Heddernheim tablet

271

2nd century
AD
2nd century
AD
Second half
2nd century
AD
Second half
2nd century
AD
-

1987, 126
Hrig, Schwertheim
1987, 123
Hrig, Schwertheim
1987, 122
Hrig, Schwertheim
1987, 596
Hrig, Schwertheim
1987, 601
Hrig, Schwertheim
1987, 602
Hrig, Schwertheim
1987, 549
Hrig, Schwertheim
1987, 547
Hrig, Schwertheim
1987, 550
Hrig, Schwertheim
1987, 541
Hrig, Schwertheim
1987, 526
Hrig, Schwertheim
1987, 538
Hrig, Schwertheim
1987, 540
Hrig, Schwertheim
1987, 527
Hrig, Schwertheim
1987, 497
Hrig, Schwertheim
1987, 523
Hrig, Schwertheim
1987, 524
Hrig, Schwertheim
1987, 525
Hrig, Schwertheim
1987, 511
Hrig, Schwertheim
1987, 512
Hrig, Schwertheim
1987, 514
Hrig, Schwertheim
1987, 515
Hrig, Schwertheim
1987, 516
Hrig, Schwertheim
1987, 517
Hrig, Schwertheim
1987, 518
Hrig, Schwertheim
1987, 519
Hrig, Schwertheim
1987, 520
Hrig, Schwertheim

Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior
Germania
Superior

Nida - Heddernheim altar

Obernburg

altar

207 AD

Obernburg

altar

206 AD

Portus? - Pforzheim

inscription Pre-185 AD

Saalburg

altar

End 2nd
century AD

Saalburg

altar

Saalburg

altar

Saalburg

altar

205 AD

Saalburg

altar

Stockstadt

inscription -

Stockstadt

altar

Stockstadt

altar

214 AD

Stockstadt

base

Stockstadt

altar

Stockstadt

altar

211-217 AD

Zugmantel

altar

Zugmantel

bull

222-235 AD

Hispania

Tarraconensis Saldanha

altar

Italia

Aecae - Troja

Second half
2nd century
AD

Italia

Aquileia

altar

Italia

Ariminum - Rimini

altar

Italia

Ariminum - Rimini

altar

Italia

Ateste - Este

tablet

Severan

Italia

Atria - Adria

tablet

222-235 AD

Italia

Bononia - Bologna

slab

Italia

Brixia - Brescia

altar

Italia

Caesena - Cesena

relief

272

2nd/3rd
century AD
2nd/3rd
century AD

1987, 521
Hrig, Schwertheim
1987, 522
Hrig, Schwertheim
1987, 536
Hrig, Schwertheim
1987, 537
Hrig, Schwertheim
1987, 539
Hrig, Schwertheim
1987, 499
Hrig, Schwertheim
1987, 500
Hrig, Schwertheim
1987, 501
Hrig, Schwertheim
1987, 508
Hrig, Schwertheim
1987, 509
Hrig, Schwertheim
1987, 529
Hrig, Schwertheim
1987, 530
Hrig, Schwertheim
1987, 531
Hrig, Schwertheim
1987, 532
Hrig, Schwertheim
1987, 533
Hrig, Schwertheim
1987, 534
Hrig, Schwertheim
1987, 491
Hrig, Schwertheim
1987, 493
Hrig, Schwertheim
1987, 609
Hrig, Schwertheim
1987, 461
Hrig, Schwertheim
1987, 447
Hrig, Schwertheim
1987, 458
Hrig, Schwertheim
1987, 459
Hrig, Schwertheim
1987, 451
Hrig, Schwertheim
1987, 452
Hrig, Schwertheim
1987, 454
Hrig, Schwertheim
1987, 453
Hrig, Schwertheim
1987, 457

Italia
Italia

Campania
(Minturnes?)
Histonium - Il vasto
d'Aimone

tablet

column

Italia

Misenum

fragment

Italia

Naples

slab

Italia

Naples

relief

Severan

Italia

near Iulia Concordia badge?!

185-192 AD

Italia

near Iulia Concordia


statuette
(Lison)

3rd century
AD

Italia

Ostia

stone

Italia

Ostia

slab

186 AD

Italia

Ostia

slab

191/2 AD

Italia

Ostia

slab

Mid-2nd
century AD

Italia

Ostia

tablet

Italia

Ostia

slab

Italia

Padua

altar

Italia

Puteoli - Pozzuoli

slab

Italia

Ravenna

altar

193-217 AD

Italia

Ravenna

inscription -

Italia

Rome

sarcophag
us

Italia

Rome

altar

150 AD

Italia

Rome

altar

150 AD

Italia

Rome

altar

Second half

273

LAnne Epigraphique
1991, 500
Hrig, Schwertheim
1987, 460
Hrig, Schwertheim
1987, 463
Hrig, Schwertheim
1987, 466
Hrig, Schwertheim
1987, 467
Hrig, Schwertheim
1987, 449
Hrig, Schwertheim
1987, 448
A. Pellegrino, Les cultes
de Jupiter Dolichnien
et de Jupiter
Hliopolitain Ostie,
in: G. M. Bellelli, U.
Bianchi, Orientalia
Sacra Urbis Romae.
Dolichena et
Heliopolitana, Rom
1996, S. 563-583, 564,
Fig. 3.
Hrig, Schwertheim
1987, 440
Hrig, Schwertheim
1987, 441
Hrig, Schwertheim
1987, 442
Hrig, Schwertheim
1987, 443
LAnne Epigraphique
1998, 273
Hrig, Schwertheim
1987, 450
Hrig, Schwertheim
1987, 465
Hrig, Schwertheim
1987, 455
Hrig, Schwertheim
1987, 456
E. Sanzi, Sur une
inscription romaine en
rapport avec le culte
dolichnien, in Bellelli,
Bianchi (Eds.),
Orientalia Sacra Urbis
Romae. Dolichena et
Heliopolitana, Rom
1996, S. 257
Hrig, Schwertheim
1987, 356
Hrig, Schwertheim
1987, 357
Hrig, Schwertheim

2nd century
AD
183 AD, 1
March
Second half
2nd century
AD
Second half
2nd century
AD

Italia

Rome

relief

Italia

Rome

relief

Italia

Rome

relief

Italia

Rome

relief

Italia

Rome

relief

Italia

Rome

slab

Italia

Rome

relief

Late 2nd
century AD

Italia

Rome

altar

198-209 AD

Italia

Rome

slab

Italia

Rome

slab

Italia

Rome

slab

Italia

Rome

slab

Italia

Rome

base

Italia

Rome

round base -

Italia

Rome

statue

Italia

Rome

slab

First half 3rd


century AD

Italia

Rome

fragment

222-235 AD

Italia

Rome

altar

244 AD, 10
Oct.

Italia

Rome

base

244 AD

Italia

Rome

slab

Italia

Rome

relief

Italia

Rome

fragments

Italia

Rome

inscription -

Italia

Rome

inscription -

Italia

Rome

column

1987, 362
Hrig, Schwertheim
1987, 363
Hrig, Schwertheim
1987, 364
Hrig, Schwertheim
1987, 365
Hrig, Schwertheim
1987, 367
Hrig, Schwertheim
1987, 368
Hrig, Schwertheim
1987, 370
Hrig, Schwertheim
1987, 371
Hrig, Schwertheim
1987, 372
Hrig, Schwertheim
1987, 373

First half 3rd


century AD
Second half
Hrig, Schwertheim
3rd century
1987, 375
AD
Hrig, Schwertheim
First half 3rd
1987, 376
century AD

274

Second half
3rd century
AD
3rd century
AD
-

Hrig, Schwertheim
1987, 377
Hrig, Schwertheim
1987, 378
Hrig, Schwertheim
1987, 379
Hrig, Schwertheim
1987, 380
Hrig, Schwertheim
1987, 381
Hrig, Schwertheim
1987, 382
Hrig, Schwertheim
1987, 383
Hrig, Schwertheim
1987, 384
Hrig, Schwertheim
1987, 385
Hrig, Schwertheim
1987, 386
Hrig, Schwertheim
1987, 401
Hrig, Schwertheim
1987, 402
Hrig, Schwertheim
1987, 403
Hrig, Schwertheim
1987, 404

Italia

Rome

relief

Italia

Rome

slab

Italia

Rome

slab

Italia

Rome

sculptural
group

Italia

Rome

head

Italia

Rome

slab

Italia

Rome

slab

Italia

Rome

altar

Italia

Rome

altar

Italia

Rome

relief

Italia

Rome

relief

Italia

Rome

slab

Italia

Rome

inscription

Italia

Rome

base

Italia

Rome

inscription

Italia

Rome

inscription

Italia

Rome

slab

Italia

Rome

Italia

Rome

altar

Italia

Rome

altar

Italia

Rome

tablet

Italia

Rome

block

Italia

Rome

round base

Italia

Rome

inscription

Italia

Rome

vase

Italia

Rome - Tivoli

slab

Italia

Tarracina - Terracine statue

Italia

Tergeste - Trieste

altar

275

Hrig, Schwertheim
1987, 405
191 AD, 31 Hrig, Schwertheim
July
1987, 408
Hrig, Schwertheim
191/2 AD
1987, 409
End 2nd
Hrig, Schwertheim
century AD 1987, 410
Hrig, Schwertheim
Pre-212 AD
1987, 411
Hrig, Schwertheim
198-209 AD
1987, 414
Hrig, Schwertheim
218 AD
1987, 415
218 AD, 14 Hrig, Schwertheim
March,
1987, 416
3rd century
Hrig, Schwertheim
AD
1987, 418
2nd/3rd
Hrig, Schwertheim
century AD 1987, 419
Hrig, Schwertheim
1987, 420
201 AD, 11 Hrig, Schwertheim
Nov
1987, 421
Hrig, Schwertheim
200-208 AD
1987, 422
Hrig, Schwertheim
1987, 423
Hrig, Schwertheim
1987, 424
Hrig, Schwertheim
270-275 AD?
1987, 425
Hrig, Schwertheim
1987, 426
Hrig, Schwertheim
1987, 427
Hrig, Schwertheim
1987, 428
Hrig, Schwertheim
1987, 429
Hrig, Schwertheim
1987, 430
196 AD, 29 Hrig, Schwertheim
June
1987, 432
Hrig, Schwertheim
Pre-212 AD
1987, 433
Hrig, Schwertheim
92 AD?
1987, 434
Mid-2nd/midLAnne Epigraphique
3rd century
1998, 183
AD
Hrig, Schwertheim
1987, 431
Hrig, Schwertheim
1987, 462
Hrig, Schwertheim
1987, 445
-

Italia

Tergeste - Trieste

altar

Italia

Turris Libisonis Porto Torres,


Sardinia

inscription 211/212 AD

Italia

Tusculum - Frascati

altar

Judea

Caesarea

block

2nd/3rd
century AD

Dura Europus

altar

211 AD

Dura Europus

altar

June-Oct
251 AD

Dura Europus

altar

251-53 AD

Dura Europus

stele

Dura Europus

text

Bezmer, prob. from


Durostorum

statuette

Bizone - Varna

hand

235-239 AD

btwn Noviodunum
and Troesmis Nikulitel

column

212, 27th
Feb-8th Apr.
217

Mesopotami
a
Mesopotami
a
Mesopotami
a
Mesopotami
a
Mesopotami
a
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior

Dionysopolis - Balcik altar


Dragoevo, near
altar
Preslav
Emporium
Piretensium - Gorsko altar
Kosovo

239 AD,
27/28 May
First half 3rd
century AD

214 AD
238-244 AD
-

Moesia
Inferior

Histria

statuette

Moesia
Inferior

Histria

gravestone -

Histria or
Durostorum

altar

198-209 AD

Muratu

slab

Nikolaevka, nr.
Varna

votive
table

Novae - Steklen

altar

Beg. 3rd
century AD
2nd/3rd
century AD

Novae - Steklen

altar

Novae - Svistov

slab

altar

222-235 AD

slab

2nd/3rd
century AD

Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior

Sacidava,
Rasova &
Sacidava,
Rasova &

near
Dunareni
near
Dunareni

276

Hrig, Schwertheim
1987, 446
Hrig, Schwertheim
1987, 468
Hrig, Schwertheim
1987, 444
Hrig, Schwertheim
1987, 30
Hrig, Schwertheim
1987, 32
Hrig, Schwertheim
1987, 33
Hrig, Schwertheim
1987, 34
Hrig, Schwertheim
1987, 35
Hrig, Schwertheim
1987, 39
Hrig, Schwertheim
1987, 69
Hrig, Schwertheim
1987, 70
Hrig, Schwertheim
1987, 60
Hrig, Schwertheim
1987, 71
Hrig, Schwertheim
1987, 77
Hrig, Schwertheim
1987, 76
Hrig, Schwertheim
1987, 63
V. Najdenova, Jupiter
Dolichenus in lower
Moesia and Thrace, in:
ANRW II, 18, 2, Berlin
1989, 1362-1396, S.
1386f.
Hrig, Schwertheim
1987, 64
Hrig, Schwertheim
1987, 67
Hrig, Schwertheim
1987, 73
Hrig, Schwertheim
1987, 74
Hrig, Schwertheim
1987, 75
LAnne Epigraphique
2001, 1733
LAnne Epigraphique
1998, 1143
LAnne Epigraphique
1998, 1144

Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Inferior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia
Superior
Moesia

Tomis - Constanta

statuette

Troesmis - Cerna

sculptural
group

Troesmis Meidanchioi

tablet

Tropaeum Traiani

altar

Vicus Quintionis (nr.


Histria)
Bononia/Jasen Junija Alba
Colonia Ulpia
Ratiaria - Arcar
Colonia Ulpia
Ratiaria - Arcar
Colonia Ulpia
Ratiaria - Arcar
Colonia Ulpia
Ratiaria - Arcar
Colonia Ulpia
Ratiaria - Kosava nr.
Arcar
Egeta - Brza Palanka

altar
statuette
altar
altar
altar
altar

End 2nd, beg.


Hrig, Schwertheim
3rd century
1987, 66
AD
Hrig, Schwertheim
222-235 AD
1987, 62
Hrig, Schwertheim
218-222 AD
1987, 61
Hrig, Schwertheim
1987, 68
Hrig, Schwertheim
212-222 AD
1987, 65
2nd/3rd
Hrig, Schwertheim
century AD 1987, 107
Hrig, Schwertheim
1987, 109
Hrig, Schwertheim
1987, 110
Hrig, Schwertheim
c. 300 AD
1987, 111
Beg. 3rd
Hrig, Schwertheim
century AD 1987, 112

altar

198-208 AD

Hrig, Schwertheim
1987, 113

sculptural
group

Hrig, Schwertheim
1987, 89

First half 2nd


century AD Hrig, Schwertheim
Egeta - Brza Palanka statue
or 2nd/3rd
1987, 90
century AD
Hrig, Schwertheim
Egeta - Brza Palanka statuette
1987, 91
Hrig, Schwertheim
Egeta - Brza Palanka statuette
1987, 92
2nd/3rd
Hrig, Schwertheim
Egeta - Brza Palanka statuette
century AD 1987, 93
Hrig, Schwertheim
Egeta - Brza Palanka statuette
1987, 94
Mid-3rd
Hrig, Schwertheim
Egeta - Brza Palanka tablet
century AD 1987, 95
Hrig, Schwertheim
Egeta - Brza Palanka relief
1987, 97
votive
Beg. 3rd
Hrig, Schwertheim
Gracanica
inscription century AD 1987, 115
Karatas, near
Hrig, Schwertheim
altar
Post-211 AD
Kladovo
1987, 87
216 AD, 1st
Hrig, Schwertheim
Kumanovo
altar
Nov
1987, 116
Pincum - Veliko
3rd century
Hrig, Schwertheim
statuette
Gradiste
AD
1987, 86
votive
3rd century
Hrig, Schwertheim
Romulianum? - Jasen
triangle
AD
1987, 103
3rd century
Hrig, Schwertheim
Romulianum? - Jasen tablet
AD
1987, 104
Hrig, Schwertheim
Romulianum? - Jasen tablet
1987, 105
Romulianum? - Jasen statuette
Hrig, Schwertheim

277

Superior
Moesia
Superior

Timacum Minus Ravna

sculptural
group

Moesia
Superior

Vidin

statuette

Moesia
Superior

Vinimacium Kostolac

sculptural
group

Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum

Beg. 3rd
century AD
End 2nd, beg.
3rd century
AD

1987, 106
Hrig, Schwertheim
1987, 114
Hrig, Schwertheim
1987, 108

Hrig, Schwertheim
1987, 85
Hrig, Schwertheim
Bedaium
text
1987, 289
Colatio Hrig, Schwertheim
slab
Windischgraz
1987, 350
Hrig, Schwertheim
Feldkirchen
fragment
1987, 349
Horzendorf, near St. votive
Hrig, Schwertheim
Veit
inscription
1987, 346
Lamprechtskogel,
Hrig, Schwertheim
relief
near Waisenberg
1987, 347
Hrig, Schwertheim
Lauriacum - Enns
bowl
1987, 285
Hrig, Schwertheim
Lauriacum - Enns
vase
1987, 286
Locus Felicis (?) Hrig, Schwertheim
statuette
Mauer an der Url
1987, 291
Locus Felicis (?) sculptural
Hrig, Schwertheim
Mauer an der Url
group
1987, 292
Locus Felicis (?) Hrig, Schwertheim
statuette
Mauer an der Url
1987, 293
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
triangle
1987, 294
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
triangle
1987, 295
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
palm
1987, 298
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
palm
1987, 299
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
palm
1987, 300
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
palm
1987, 301
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
palm
1987, 302
Locus Felicis (?) Hrig, Schwertheim
votive
Mauer an der Url
1987, 303
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
sheet
1987, 304
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
sheet
1987, 305
Locus Felicis (?) Period of
Hrig, Schwertheim
votive
Mauer an der Url
Commodus? 1987, 306
Locus Felicis (?) Hrig, Schwertheim
votive
Mauer an der Url
1987, 307
Locus Felicis (?) votive
Hrig, Schwertheim
Mauer an der Url
sheet
1987, 308
Locus Felicis (?) Hrig, Schwertheim
votive
Mauer an der Url
1987, 309

278

Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum
Noricum

Locus Felicis (?) Mauer an der Url


Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url
Locus Felicis (?) Mauer an der Url

votive

votive

votive

votive
sheet
votive
palm
votive
sheet
votive
sheet
votive
sheet
votive
sheet

Noricum

Ovilava - Wels

votive slab

Beg. 3rd
century AD

Noricum

Trigisamum Traismauer

votive
triangle

Noricum

Trigisamum Traismauer

votive
triangle

260-70/ end
2nd century
AD

Noricum

Virunum - Zollfeld

slab

189 AD

Noricum

Virunum - Zollfeld

round altar -

Noricum

Virunum - Zollfeld

round altar -

Noricum

Virunum - Zollfeld

fragment

Noricum

Virunum - Zollfeld

altar

Noricum

Virunum - Zollfeld

base

Noricum

Virunum - Zollfeld

altar

Noricum

Virunum - Zollfeld

altar

Noricum

Wallsee

altar

Pannonia
Inferior

Acumincum - Stari
Slankamen

statuette

Pannonia
Inferior

Acumincum - Stari
Slankamen

statuette

Acumincum - Surduk

postament
s

Aquincum

stele

Aquincum - Obuda

base

Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior

279

Second half
3rd century
AD
Piccottinni
puts it
between
197-207 AD
Mid-2nd
century AD

Hrig, Schwertheim
1987, 310
Hrig, Schwertheim
1987, 311
Hrig, Schwertheim
1987, 312
Hrig, Schwertheim
1987, 313
Hrig, Schwertheim
1987, 314
Hrig, Schwertheim
1987, 315
Hrig, Schwertheim
1987, 316
Hrig, Schwertheim
1987, 317
Hrig, Schwertheim
1987, 318
Hrig, Schwertheim
1987, 287
Hrig, Schwertheim
1987, 327
Hrig, Schwertheim
1987, 328
Hrig, Schwertheim
1987, 330
Hrig, Schwertheim
1987, 331
Hrig, Schwertheim
1987, 332
Hrig, Schwertheim
1987, 333
Hrig, Schwertheim
1987, 335
Hrig, Schwertheim
1987, 342
Hrig, Schwertheim
1987, 344
Hrig, Schwertheim
1987, 345
LAnne Epigraphique
1990, 785
Hrig, Schwertheim
1987, 206

End 2nd, beg.


Hrig, Schwertheim
3rd century
1987, 207
AD
Hrig, Schwertheim
1987, 208
Hrig, Schwertheim
228 AD
1987, 185
Hrig, Schwertheim
Post-100 AD
1987, 183

Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Inferior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior

Aquincum?

sculptural
group

Burgenae - Novi
Banovci

base

Gorsium Sarpentele
Lussonium - Komlod
Lussonium - Komlod
Lussonium - Komlod
Mursa - Klisa, near
Osijek
Sirmium - Srijemska
Mitrovica
Sirmium - Srijemska
Mitrovica
Tokod, near
Esztergom

altar
votive
triangle
votive
triangle

Beg. 3rd
century AD,
or 202 AD
3rd century
AD
-

altar

altar

fragment

193-211 AD

altar

altar

222-235 AD

Vetus Salina - Adony bust

Vetus Salina - Adony hand

Aquae Balissae Daruvar


Aquae Balissae Daruvar

altar

198-208 AD

altar

Brigetio - Oszony

slab

First half 3rd


century AD

Brigetio - Oszony

fragment

Brigetio - Oszony

relief

2nd/3rd
century AD

Brigetio - Oszony

relief

Brigetio - Oszony

statue base -

Brigetio - Oszony

sculptural
group

Pannonia
Superior

Brigetio - Oszony

base

Second half
2nd century
AD

Pannonia
Superior
Pannonia
Superior
Pannonia
Superior

Carnuntum - Bad
Deutsch-Altenburg
Carnuntum - Bad
Deutsch-Altenburg
Carnuntum - Bad
Deutsch-Altenburg

sculptural
group
votive
table
votive
table

Pannonia
Superior

Carnuntum - Bad
Deutsch-Altenburg

statuette

3rd century
AD

Pannonia
Superior
Pannonia
Superior

Carnuntum Petronell
Carnuntum Petronell

votive
table

altar

280

235-238 AD

Hrig, Schwertheim
1987, 184
Hrig, Schwertheim
1987, 209
Hrig, Schwertheim
1987, 200
Hrig, Schwertheim
1987, 201
Hrig, Schwertheim
1987, 202
Hrig, Schwertheim
1987, 204
Hrig, Schwertheim
1987, 205
Hrig, Schwertheim
1987, 210
Hrig, Schwertheim
1987, 211
Hrig, Schwertheim
1987, 182
Hrig, Schwertheim
1987, 188
Hrig, Schwertheim
1987, 189
Hrig, Schwertheim
1987, 276
Hrig, Schwertheim
1987, 277
Hrig, Schwertheim
1987, 237
Hrig, Schwertheim
1987, 238
Hrig, Schwertheim
1987, 239
Hrig, Schwertheim
1987, 240
Hrig, Schwertheim
1987, 241
Hrig, Schwertheim
1987, 242
Hrig, Schwertheim
1987, 253
Hrig, Schwertheim
1987, 230
Hrig, Schwertheim
1987, 231
Hrig, Schwertheim
1987, 232
Hrig, Schwertheim
1987, 233
Hrig, Schwertheim
1987, 218
Hrig, Schwertheim
1987, 219

Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior
Pannonia
Superior

Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Petronell
Carnuntum Pfaffenberg

base

altar

183 AD

relief

2nd century
AD

base in
altar form

180-183 AD

statue

arm

votive
table

altar

altar

block

128-138 AD

Emona

altar

Gerulata - Rusovce

altar

Gerulata - Rusovce

block

4th century
AD

Mullendorf, near
Eisenstadt

altar

Neviodunum - Krsko altar

Poetovio - Ptuj

altar

189 AD

Poetovio - Ptuj

stone

207 AD

Poetovio - Ptuj

altar

3rd century
AD

altar

Praetorium
Latobicorum Trebnje
Praetorium
Latobicorum Trebnje
Praetorium
Latobicorum Trebnje
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely
Savaria Szombathely

Hrig, Schwertheim
1987, 220
Hrig, Schwertheim
1987, 221
Hrig, Schwertheim
1987, 222
Hrig, Schwertheim
1987, 223
Hrig, Schwertheim
1987, 224
Hrig, Schwertheim
1987, 226
Hrig, Schwertheim
1987, 228
Hrig, Schwertheim
1987, 229
Hrig, Schwertheim
1987, 639
Hrig, Schwertheim
1987, 217
Hrig, Schwertheim
1987, 273
Hrig, Schwertheim
1987, 234
Hrig, Schwertheim
1987, 235
Hrig, Schwertheim
1987, 263
LAnne Epigraphique
1997, 1244
Hrig, Schwertheim
1987, 282
Hrig, Schwertheim
1987, 283
Hrig, Schwertheim
1987, 284
Hrig, Schwertheim
1987, 274

votive
138 AD, 1st
inscription Nov

Hrig, Schwertheim
1987, 275

slab

Hrig, Schwertheim
1987, 640

196-197 AD

altar
altar

2nd/3rd
century AD
2nd/3rd
century AD

altar

bowl

fragment

altar

208 AD

281

Hrig, Schwertheim
1987, 265
Hrig, Schwertheim
1987, 266
Hrig, Schwertheim
1987, 267
Hrig, Schwertheim
1987, 268
Hrig, Schwertheim
1987, 269
Hrig, Schwertheim
1987, 270

Pannonia
Superior
Pannonia
Superior

Savaria Szombathely
Savaria Szombathely

Raetia

Aquileia - Aalen

votive
triangle

Raetia

Aquileia - Aalen

block

Raetia

Faimingen

relief

Raetia
Raetia
Raetia

Serviodurum Straubing
Statio Vetonianis Pfunz
Statio Vetonianis Pfunz

sculptural
group
fragment

altar
tablet
tablet

Syria

Antep (Doliche)

sealstone

Syria

Beka'a

hand

Syria

between Maras &


Birecik

sculptural
group

Syria

Doliche - Dlk

altar

Syria

Doliche - Dlk

slab

Syria

Doliche - Dlk

slab

Syria

Doliche - Dlk

Syria

Doliche - Dlk

Syria

Doliche?
Gonca Dagi, nr.
Asagi Kalecik

Syria

votive
triangle
votive
triangle
altar
stele

Syria

Hierapolis - Membidj stele

Syria

Kekliktepe, nr. Antep stele

Syria
Syria
Syria

Khaltan, in Tal des


Afrin
Khaltan, in Tal des
Afrin
Khirbet Khalid,
Hierapolis-Bambyke

stele
stele
base

Syria

Kurcuoglu

stele

Syria

Kurcuoglu

stele

Syria
Syria
Syria

Kurdini Tepe, nr.


Alacakilise
Kurdini Tepe, nr.
Alacakilise
Lebanon

stele
column
drum
hand

282

2nd/3rd
century AD

Hrig, Schwertheim
1987, 271
Hrig, Schwertheim
1987, 272
Hrig, Schwertheim
1987, 475
Beg. 3rd
Hrig, Schwertheim
century AD 1987, 476
End 2nd
Hrig, Schwertheim
century AD 1987, 477
163 AD, 11 Hrig, Schwertheim
April
1987, 485
Hrig, Schwertheim
Severan
1987, 480
Hrig, Schwertheim
Severan
1987, 481
Hrig, Schwertheim
21 AD
1987, 12
Hrig, Schwertheim
1987, 40
Hrig, Schwertheim
Imperial
1987, 18
Hrig, Schwertheim
57/58 AD
1987, 2
Hrig, Schwertheim
Pre-256 AD
1987, 3
Hrig, Schwertheim
Pre-256 AD
1987, 4
Hrig, Schwertheim
Mid-1st C AD
1987, 6
Hrig, Schwertheim
1987, 7
200-250 AD SEG 1998, 1871
First half 1st Hrig, Schwertheim
century AD 1987, 21
Hrig, Schwertheim
1987, 29
End 2nd, beg.
Hrig, Schwertheim
3rd century
1987, 9
AD
nd
1/2
Hrig, Schwertheim
century AD 1987, 26
1/2nd
Hrig, Schwertheim
century AD 1987, 27
2nd/3rd
LAnne Epigraphique
century AD 1998, 1430
Hrig, Schwertheim
1987, 24
Hrig, Schwertheim
1987, 25
first half 1st Hrig, Schwertheim
century AD 1987, 19
Hrig, Schwertheim
1987, 20
Hrig, Schwertheim
1987, 41

Syria

Maras - Germanikeia stele

Syria

Mastala

stele

Syria

Nikopolis

seals

1st century
AD
2nd century
AD
-

Syria

Tilhalit

stele

Syria

Zafer, nr. Antep

stele

Syria

Zeytintepe, nr.
Baspinar

Syria

Zeytintepe, nr.
Baspinar

Thracia
Thracia
Thracia

Augusta Traiana Stamovo


Augusta Traiana Stara Zagora
Augusta Traiana Stara Zagora

1st century
BC/1st
century AD

stele

stele

1st century
BC/1st
century AD

slab

222-235 AD

column

212-217
(214) AD
Reign of
Septimius
Severus

altar

Thracia

Cabyle

slab

Thracia

Cillae - Cerna Gora

base

202-211 AD

Thracia

Gorni Voden

hand

Thracia

Haskovo

hand

Macedonia

Stobi

sculptural
group

Mid-3rd
century AD

283

Hrig, Schwertheim
1987, 17
Hrig, Schwertheim
1987, 28
SEG 1989, 1586
Hrig, Schwertheim
1987, 16
Hrig, Schwertheim
1987, 8
Hrig, Schwertheim
1987, 22
Hrig, Schwertheim
1987, 23
Hrig, Schwertheim
1987, 50
Hrig, Schwertheim
1987, 52
Hrig, Schwertheim
1987,51
LAnne Epigraphique
1999, 1374
Hrig, Schwertheim
1987, 54
Hrig, Schwertheim
1987, 53
Hrig, Schwertheim
1987, 55
Hrig, Schwertheim
1987, 48

THE JEW ISH DIASP ORA

by catalogue number
Publication/
Region
s
Frey, J. -B.,
Corpus
Inscriptionu
m
Judaicarum
I, Rome:
Italia
Pontificio
istituto di
archeologia
cristiana,
1936, 1

Location

Italia

Via Flaminia

CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ

Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia

Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via

3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
30
31

Date

2nd-4th Century
AD

Via Flaminia

CIJ I, 2
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,

Type

Salaria
Salaria
Salaria
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana

284

plaque
plaque
plaque
painted epitaph
painted epitaph
painted epitaph
tablet
painted epitaph
painted epitaph
painted epitaph
painted epitaph
fragment
painted epitaph
painted epitaph
painted epitaph
painted epitaph
graffiti
painted epitaph
plaque
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph

2nd-4th Century
AD
-

CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ

I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,

32
33
34
35
36
37
38
39
41
42
43
44
45
46
47
48
50
51
52
53
55
56
67
68
69
70
72

Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia

Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via
Via

Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana
Nomentana

CIJ I, 73

Italia

Via Labicana

painted epitaph

CIJ I, 75

Italia

Via Labicana

painted epitaph

CIJ I, 77

Italia

Via Labicana

painted epitaph

CIJ I, 87

Italia

Via Appia

graffiti

CIJ I, 88

Italia

Via Appia

plaque

CIJ I, 89

Italia

Via Appia

fragment

CIJ I, 90

Italia

Via Appia

plaque

CIJ I, 95

Italia

Via Appia

plaque

CIJ I, 97

Italia

Via Appia

plaque

CIJ I, 99

Italia

Via Appia

fragment

CIJ I, 100

Italia

Via Appia

inscription

CIJ I, 102

Italia

Via Appia

plaque

CIJ I, 103

Italia

Via Appia

CIJ I, 105

Italia

Via Appia

inscription

CIJ I, 106

Italia

Via Appia

fragment

285

painted epitaph
painted epitaph
plaque
painted epitaph
fragment
fragment
painted epitaph
painted epitaph
painted epitaph
painted epitaph
stele
painted epitaph
plaque
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
plaque
plaque
plaque
sarcophagus

2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 108

Italia

Via Appia

plaque

CIJ I, 109

Italia

Via Appia

sarcophagus

CIJ I, 110

Italia

Via Appia

fragment

CIJ I, 111

Italia

Via Appia

plaque

CIJ I, 113

Italia

Via Appia

painted plaque

CIJ I, 118

Italia

Via Appia

plaque

CIJ I, 119

Italia

Via Appia

fragment

CIJ I, 120

Italia

Via Appia

inscription

CIJ I, 121

Italia

Via Appia

plaque

CIJ I, 122

Italia

Via Appia

plaque

CIJ I, 125

Italia

Via Appia

plaque

CIJ I, 126

Italia

Via Appia

plaque

CIJ I, 130

Italia

Via Appia

fragment

CIJ I, 132

Italia

Via Appia

fragment

CIJ I, 136

Italia

Via Appia

plaque

CIJ I, 138

Italia

Via Appia

plaque

CIJ I, 139

Italia

Via Appia

fragment

CIJ I, 140

Italia

Via Appia

inscription

CIJ I, 141

Italia

Via Appia

plaque

CIJ I, 142

Italia

Via Appia

plaque

CIJ I, 143

Italia

Via Appia

plaque

CIJ I, 145

Italia

Via Appia

plaque

CIJ I, 146

Italia

Via Appia

plaque

CIJ I, 147

Italia

Via Appia

painted plaque

CIJ I, 148

Italia

Via Appia

plaque

CIJ I, 149

Italia

Via Appia

plaque

CIJ I, 150

Italia

Via Appia

plaque

CIJ I, 151

Italia

Via Appia

plaque

286

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 152

Italia

Via Appia

plaque

CIJ I, 153

Italia

Via Appia

plaque

CIJ I, 155

Italia

Via Appia

plaque

CIJ I, 156

Italia

Via Appia

fragment

CIJ I, 157

Italia

Via Appia

plaque

CIJ I, 161

Italia

Via Appia

plaque

CIJ I, 163

Italia

Via Appia

graffiti

CIJ I, 165

Italia

Via Appia

graffiti

CIJ I, 166

Italia

Via Appia

fragment

CIJ I, 167

Italia

Via Appia

painted epitaph

CIJ I, 173

Italia

Via Appia

fragment

CIJ I, 176

Italia

Via Appia

fragment

CIJ I, 180

Italia

Via Appia

fragment

CIJ I, 193

Italia

Via Appia

fragment

CIJ I, 200

Italia

Via Appia

painted plaque

CIJ I, 201

Italia

Via Appia

plaque

CIJ I, 202

Italia

Via Appia

fragment

CIJ I, 203

Italia

Via Appia

fragment

CIJ I, 206

Italia

Via Appia

plaque

CIJ I, 209

Italia

Via Appia

plaque

CIJ I, 210

Italia

Via Appia

inscription

CIJ I, 212

Italia

Via Appia

inscription

CIJ I, 213

Italia

Via Appia

plaque

CIJ I, 215

Italia

Via Appia

plaque

CIJ I, 217

Italia

Via Appia

plaque

CIJ I, 221

Italia

Via Appia

plaque

CIJ I, 222

Italia

Via Appia

plaque

CIJ I, 225

Italia

Via Appia

plaque

287

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 228

Italia

Via Appia

inscription

CIJ I, 229

Italia

Via Appia

plaque

CIJ I, 230

Italia

Via Appia

plaque

CIJ I, 231

Italia

Via Appia

plaque

CIJ I, 233

Italia

Via Appia

plaque

CIJ I, 246

Italia

Via Appia

painted epitaph

CIJ I, 248

Italia

Via Appia

plaque

CIJ I, 249

Italia

Via Appia

fragment

CIJ I, 250

Italia

Via Appia

plaque

CIJ I, 252

Italia

Via Appia

plaque

CIJ I, 254

Italia

Via Appia

plaque

CIJ I, 255

Italia

Via Appia

graffiti

CIJ I, 256

Italia

Via Appia

plaque

CIJ I, 257

Italia

Via Appia

plaque

CIJ I, 260

Italia

Via Appia

plaque

CIJ I, 261

Italia

Via Appia

graffiti

CIJ I, 263

Italia

Via Appia

plaque

CIJ I, 265

Italia

Via Appia

inscription

CIJ I, 268

Italia

Via Appia

plaque

CIJ I, 269

Italia

Via Appia

plaque

CIJ I, 281

Italia

Via Appia

inscription

CIJ I, 282

Italia

Via Appia

fragment

CIJ I, 283

Italia

Via Appia

sarcophagus

CIJ I, 285

Italia

Via Appia

plaque

CIJ I, 287

Italia

Via Ostiensis

sarcophagus

CIJ I, 290

Italia

Via Portuensis

plaque

CIJ I, 291

Italia

Via Portuensis

fragment

CIJ I, 293

Italia

Via Portuensis

painted brick

288

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 296

Italia

Via Portuensis

plaque

CIJ I, 298

Italia

Via Portuensis

inscription

CIJ I, 299

Italia

Via Portuensis

plaque

CIJ I, 300

Italia

Via Portuensis

plaque

CIJ I, 301

Italia

Via Portuensis

plaque

CIJ I, 304

Italia

Via Portuensis

plaque

CIJ I, 306

Italia

Via Portuensis

plaque

CIJ I, 307

Italia

Via Portuensis

inscription

CIJ I, 309

Italia

Via Portuensis

plaque

CIJ I, 310

Italia

Via Portuensis

plaque

CIJ I, 312

Italia

Via Portuensis

plaque

CIJ I, 315

Italia

Via Portuensis

plaque

CIJ I, 316

Italia

Via Portuensis

plaque

CIJ I, 317

Italia

Via Portuensis

plaque

CIJ I, 318

Italia

Via Portuensis

plaque

CIJ I, 319

Italia

Via Portuensis

plaque

CIJ I, 321

Italia

Via Portuensis

inscription

CIJ I, 323

Italia

Via Portuensis

plaque

CIJ I, 324

Italia

Via Portuensis

plaque

CIJ I, 325

Italia

Via Portuensis

plaque

CIJ I, 327

Italia

Via Portuensis

plaque

CIJ I, 328

Italia

Via Portuensis

plaque

CIJ I, 329

Italia

Via Portuensis

plaque

CIJ I, 331

Italia

Via Portuensis

painted epitaph

CIJ I, 332

Italia

Via Portuensis

fragment

CIJ I, 333

Italia

Via Portuensis

plaque

CIJ I, 334

Italia

Via Portuensis

plaque

CIJ I, 335

Italia

Via Portuensis

fragment

289

2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 336

Italia

Via Portuensis

plaque

CIJ I, 337

Italia

Via Portuensis

inscription

CIJ I, 340

Italia

Via Portuensis

plaque

CIJ I, 343

Italia

Via Portuensis

inscription

CIJ I, 345

Italia

Via Portuensis

fragment

CIJ I, 346

Italia

Via Portuensis

plaque

CIJ I, 347

Italia

Via Portuensis

plaque

CIJ I, 348

Italia

Via Portuensis

plaque

CIJ I, 349

Italia

Via Portuensis

plaque

CIJ I, 351

Italia

Via Portuensis

plaque

CIJ I, 353

Italia

Via Portuensis

plaque

CIJ I, 354

Italia

Via Portuensis

plaque

CIJ I, 355

Italia

Via Portuensis

fragment

CIJ I, 358

Italia

Via Portuensis

plaque

CIJ I, 361

Italia

Via Portuensis

plaque

CIJ I, 362

Italia

Via Portuensis

plaque

CIJ I, 364

Italia

Via Portuensis

plaque

CIJ I, 367

Italia

Via Portuensis

plaque

CIJ I, 368

Italia

Via Portuensis

inscription

CIJ I, 369

Italia

Via Portuensis

plaque

CIJ I, 370

Italia

Via Portuensis

plaque

CIJ I, 371

Italia

Via Portuensis

plaque

CIJ I, 372

Italia

Via Portuensis

plaque

CIJ I, 373

Italia

Via Portuensis

painted brick

CIJ I, 374

Italia

Via Portuensis

plaque

CIJ I, 375

Italia

Via Portuensis

plaque

CIJ I, 376

Italia

Via Portuensis

plaque

CIJ I, 378

Italia

Via Portuensis

fragment

290

2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 379

Italia

Via Portuensis

plaque

CIJ I, 380

Italia

Via Portuensis

plaque

CIJ I, 382

Italia

Via Portuensis

plaque

CIJ I, 383

Italia

Via Portuensis

fragment

CIJ I, 384

Italia

Via Portuensis

inscription

CIJ I, 385

Italia

Via Portuensis

plaque

CIJ I, 390

Italia

Via Portuensis

inscription

CIJ I, 391

Italia

Via Portuensis

painted plaque

CIJ I, 392

Italia

Via Portuensis

plaque

CIJ I, 394

Italia

Via Portuensis

plaque

CIJ I, 395

Italia

Via Portuensis

painted epitaph

CIJ I, 396

Italia

Via Portuensis

plaque

CIJ I, 397

Italia

Via Portuensis

plaque

CIJ I, 398

Italia

Via Portuensis

plaque

CIJ I, 399

Italia

Via Portuensis

inscription

CIJ I, 400

Italia

Via Portuensis

plaque

CIJ I, 401

Italia

Via Portuensis

plaque

CIJ I, 403

Italia

Via Portuensis

painted epitaph

CIJ I, 405

Italia

Via Portuensis

inscription

CIJ I, 408

Italia

Via Portuensis

plaque

CIJ I, 409

Italia

Via Portuensis

plaque

CIJ I, 411

Italia

Via Portuensis

plaque

CIJ I, 413

Italia

Via Portuensis

fragment

CIJ I, 416

Italia

Via Portuensis

plaque

CIJ I, 417

Italia

Via Portuensis

inscription

CIJ I, 418

Italia

Via Portuensis

plaque

CIJ I, 419

Italia

Via Portuensis

painted epitaph

CIJ I, 425

Italia

Via Portuensis

plaque

291

2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 433

Italia

Via Portuensis

plaque

CIJ I, 456

Italia

Via Portuensis

plaque

CIJ I, 457

Italia

Via Portuensis

inscription

CIJ I, 458

Italia

Via Portuensis

inscription

CIJ I, 460

Italia

Via Portuensis

plaque

CIJ I, 462

Italia

Via Portuensis

plaque

CIJ I, 463

Italia

Via Portuensis

fragment

CIJ I, 464

Italia

Via Portuensis

fragment

CIJ I, 465

Italia

Via Portuensis

plaque

CIJ I, 466

Italia

Via Portuensis

plaque

CIJ I, 467

Italia

Via Portuensis

plaque

CIJ I, 470

Italia

Via Portuensis

plaque

CIJ I, 472

Italia

Via Portuensis

plaque

CIJ I, 476

Italia

Via Portuensis

plaque

CIJ I, 477

Italia

Via Portuensis

plaque

CIJ I, 478

Italia

Via Portuensis

plaque

CIJ I, 479

Italia

Via Portuensis

plaque

CIJ I, 480

Italia

Via Portuensis

plaque

CIJ I, 481

Italia

Via Portuensis

fragment

CIJ I, 482

Italia

Via Portuensis

inscription

CIJ I, 483

Italia

Via Portuensis

plaque

CIJ I, 494

Italia

CIJ I, 496

Italia

CIJ I, 497

Italia

CIJ I, 499

Italia

CIJ I, 533

Italia

Castel Porziano

fragment

CIJ I, 534

Italia

Ostia

disc

CIJ I, 535
CIJ I, 536
CIJ I, 537

Italia
Italia
Italia

Porto
Porto
Porto

plaque
plaque
plaque

region of Via
Portuensis
region of Via
Portuensis
region of Via
Portuensis
region of Via
Portuensis

292

2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
beg. 2nd Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
AD 330, 14 May
2nd-4th Century
AD

fragment

sarcophagus

plaque

fragment

First half 2nd


Century AD
2nd-4th Century
AD
-

CIJ
CIJ
CIJ
CIJ
CIJ
CIJ

I,
I,
I,
I,
I,
I,

538
539
540
543
548
552

Italia
Italia
Italia
Italia
Italia
Italia

Porto
Porto
Porto
Porto
Porto
Fondi

plaque
fragment
plaque
plaque
fragment
epitaph

CIJ I, 553

Italia

Capua

epitaph

CIJ I, 554

Italia

Nola

lamp

CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ

555
556
558
559
561
562
563
564
566
567
568
569
570
575
576
578
579
581
584
585
586
587
590
593
594
595
596
597
599
600
606
607

Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia

Naples
Naples
Naples
Naples
Puzzeoli
Pompeii
Pompeii
Pompeii
Pompeii
Pompeii
Salerno
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa
Venosa

seal
epitaph
epitaph
epitaph
epitaph?
graffiti
amphora
graffiti
graffiti
graffiti
inscription
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
epitaph
painted epitaph
epitaph
epitaph
epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph
painted epitaph

CIJ I, 608

Italia

Venosa

painted epitaph

CIJ I, 609
CIJ I, 610

Italia
Italia

Venosa
Venosa

CIJ I, 611

Italia

Venosa

CIJ I, 612

Italia

Venosa

painted epitaph
painted epitaph
painted
inscription
painted epitaph

CIJ I, 613

Italia

Venosa

painted epitaph

CIJ I, 614

Italia

Venosa

painted epitaph

CIJ I, 616

Italia

Venosa

painted epitaph

CIJ I, 620

Italia

Tarentum

inscription

I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,

293

2nd Century AD
uncertain
2nd-4th Century
AD
7th Century AD or
later?
uncertain
70-95 AD
5th-6th Century AD
5th-6th Century AD
1st Century AD
pre AD 79
pre AD 79
pre AD 79
pre AD 79
pre AD 79
5th-6th Century AD
5th-6th Century AD
5th Century AD
5th Century AD
5th Century AD
5th Century AD
5th Century AD
5th Century AD
4-5th Century AD
5th Century AD
5th Century AD
5th Century AD
4-5th Century AD
5th Century AD
5th Century AD
5th-6th Century AD
5th Century AD
5th Century AD
5th Century AD
5th-6th Century AD
5th-6th Century AD
early 6th Century
AD
5th-6th Century AD
5th-6th Century AD
early 6th Century
AD
5th-6th Century AD
early 6th Century
AD
mid 6th Century
AD
early 6th Century
AD
-

CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ

I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,

Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia

Tarentum
Tarentum
Tarentum
Tarentum
Tarentum
Tarentum
Tarentum
Tarentum
Tarentum
Otranto
Bari
Oria
Oria

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
inscription
stele
epitaph

CIJ I, 636

Etruria

Civitavecchia

epitaph

CIJ I, 637
CIJ I, 638

Italia
Italia

Ferrara
Brescia

inscription
epitaph?

CIJ I, 639

Italia

Brescia

epitaph?

CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ

Concordia
Pola
Pola
Aquileia
Milan
Milan
Milan
Catania
Syracuse
Syracuse
Agrigentum
Rabato
Macomer
Sulcis
Sulcis
Abdera
Avignon
Arles
Auch
Regensburg

epitaph
inscription
epitaph
epitaph?
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
catacombs
ring
ring
epitaph
epitaph
seal
plaque
donation
magical text

Badenweiler

magical text

CIJ I, 675
CIJ I, 676

Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Malta
Italia
Italia
Italia
Baetica
Gallia
Gallia
Gallia
Raetia
Germania
Superior
Pannonia
Pannonia

7th-8th Century AD
7th-8th Century AD
7th-8th Century AD
7th-8th Century AD
3rd-6th Century AD
4th Century AD
7th-8th Century AD
7th-8th Century AD
7th-8th Century AD
3rd Century AD
6th-8th Century AD
7th-8th Century AD
5th-6th Century AD
2nd-4th Century
AD
4th Century Ad?
4th Century or
earlier
4-5th Century AD
3-5th Century AD
1st Century BC
5th Century AD
5th Century AD
383 AD
4-5th Century AD
4-5th Century AD
5th Century AD
Republican period?
4-5th Century AD?
4-5th Century AD
3rd Century AD
4th Century AD?
7th-8th Century AD
3rd Century AD

Pannonia

epitaph
epitaph
synagogue
inscription

4th Century AD
3rd Century AD

CIJ I, 677

Aquincum
Solva (near Gran)
Intercisa (DunaPentele)

CIJ I, 678

Pannonia

Siklos

epitaph

CIJ I, 679

Sarajevo

ring

Oescus

epitaph

4th Century AD?

CIJ I, 693

Pannonia
Moesia
Inferior
Macedonia

2nd-3rd Century
AD
-

Thessaloniki

epitaph

CIJ I, 694

Macedonia

Stobi

donation

5th-6th Century AD
2nd-3rd Century
AD

CIJ I, 709

Greece

Delphi

CIJ I, 710

Greece

Delphi

I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,

621
622
623
625
627
628
629
630
631
632
633
634
635

640
641
642
643
644
645
646
650
651
652
654
655
656
657
658
665
667
669
671
673

CIJ I, 674

CIJ I, 681

manumission
text
manumission

294

233-5 AD

BC 158-7
BC 162

CIJ I, 711

Greece

Delphi

CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ
CIJ

Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece

Athens
Athens
Athens
Athens
Athens
Athens
Piraeus
Patras
Patras

CIJ I, 718

Greece

Corinth

CIJ
CIJ
CIJ
CIJ

720
722
723
724

Greece
Greece
Greece
Greece

Mantinea
Aegina
Aegina
Aegina

text
manumission
text
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
inscription
magical text
synagogue
inscription
dedication
mosaic floor
mosaic floor
magical text

CIJ I, 725

Greece

Rheneia

epitaph

Narona

stamp

4th Century AD

Larisa
Larisa

epitaph
epitaph

1st-4th Century AD
4th-6th Century AD
2nd-3rd Century
AD
uncertain
uncertain
uncertain
3rd-7th Century AD
3rd-4th Century AD
3rd Century AD
3rd Century AD
3rd Century AD
3rd-4th Century AD
5th-7th Century AD
1st Century AD?
2nd-3rd Century
AD
Roman
2nd Century AD
BC?
1st Century AD?

I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,
I,

712
713
715
715e
715f
715g
715i
716
717

Noy, D.,
Panayotov,
A.,
Bloedhorn,
H., (eds.),
Inscriptiones
Judaicae
Orientis vol. Dalmatia
I, Eastern
Europe,
Tbingen:
Mohr
Siebeck,
2004, p. 20,
fn 1.
IJO 1, Ach1
Greece
IJO 1, Ach5
Greece
IJO 1, Ach6

Greece

Larisa

epitaph

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece

Larisa
Kalyvia
Phthiotic
Phthiotic
Phthiotic
Phthiotic
Phthiotic
Phthiotic
Phthiotic
Almyra
Athens

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,

Ach7
Ach9
Ach15
Ach16
Ach17
Ach18
Ach20
Ach21
Ach23
Ach24
Ach26

Thebes
Thebes
Thebes
Thebes
Thebes
Thebes
Thebes

IJO 1, Ach27 Greece

Athens

epitaph

IJO 1, Ach31 Greece

Athens

epitaph

IJO 1, Ach33 Greece

Athens

epitaph

IJO 1, Ach35 Greece

Athens

epitaph

295

2nd-1st Century BC
4-5th Century AD
5th-6th Century AD
6 Century AD?
uncertain
1st Century AD
1st Century AD
uncertain
3rd Century AD
4th Century AD
300-50 AD
300-50 AD
end 2nd Century
BC

IJO 1, Ach36 Greece


IJO 1, Ach38 Greece
IJO 1, Ach39 Greece

Athens
Athens
Athens

IJO 1, Ach41 Greece

Athens

IJO 1, Ach45 Greece

Oropus

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece
Greece

Plataea
Corinth
Corinth
Argos
Arcadia
Coronea
Taenarum
Taenarum
Delos
Delos
Delos
Delos
Delos
Delos
Delos
Delos
Delos
Syros
Syros
Syros

IJO 1, BS1

Scythia

Olbia

IJO 1, BS2
IJO 1, BS4

Black Sea
Black Sea

Chersonesos
Panticipaeum

IJO 1, BS5

Black Sea

Panticipaeum

IJO 1, BS6

Black Sea

Panticipaeum

IJO 1, BS7

Black Sea

Panticipaeum

IJO 1, BS9

Black Sea

Panticipaeum

IJO
IJO
IJO
IJO
IJO

Black
Black
Black
Black
Black

Sea
Sea
Sea
Sea
Sea

Panticipaeum
Panticipaeum
Panticipaeum
Panticipaeum
Panticipaeum

epitaph
honorific
honorific
thiasos
inscription
manumission
text
epitaph
epitaph
epitaph
epitaph
epitaph
ephebic list
epitaph
epitaph
vow
vow
vow
vow
vow
honorific
honorific
list
statue
prayer
thanksgiving
dedication
building
inscription
graffito
dedication
manumission
text
manumission
text
manumission
text
manumission
text
epitaph
epitaph
epitaph
epitaph
epitaph

IJO 1, BS15

Black Sea

Panticipaeum

epitaph

IJO 1, BS16

Black Sea

Panticipaeum

IJO 1, BS17

Black Sea

Phanagoria

IJO 1, BS18

Black Sea

Phanagoria

IJO 1, BS19

Black Sea

Hermonassa

IJO 1, BS20

Black Sea

Gorgippia

IJO 1, BS21

Black Sea

Gorgippia

epitaph
manumission
text
manumission
text
epitaph
manumission
text
manumission
text

1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,

1,
1,
1,
1,
1,

Ach46
Ach48
Ach49
Ach51
Ach52
Ach53
Ach55
Ach56
Ach60
Ach61
Ach62
Ach63
Ach65
Ach66
Ach67
Ach68
Ach69
Ach72
Ach73
Ach74

BS10
BS11
BS12
BS13
BS14

296

1st Century AD?


37-27 BC
27-4 BC
4th-3rd Century BC
300-250 BC
2nd Century AD
3rd Century AD
4th Century AD
3rd-4th Century AD
2nd Century AD
246 AD
1st Century AD?
3rd Century AD
1-2nd Century AD
1-2nd Century AD
1st Century BC
1st Century BC
1st Century BC
250-175 BC
BC 150-50
100 AD
4-39 AD
4th Century AD?
4th Century AD
37-4 BC
2nd-3rd Century
AD
4-5th Century AD
306 AD
80 AD
1-2nd Century AD
1-2nd Century AD
1-2nd Century AD
3rd-4th Century AD
3rd-4th Century AD
3rd-4th Century AD
3rd-4th Century AD
4th Century AD?
2nd-4th Century
AD
4-5th Century AD?
17 AD
52 AD
1st Century AD
41 AD
93-123 AD

IJO 1, BS22

Black Sea

Gorgippia

IJO 1, BS23

Black Sea

Gorgippia

IJO 1, BS24

Black Sea

Gorgippia

IJO
IJO
IJO
IJO
IJO

Black Sea
Black Sea
Greece
Greece
Greece

Gorgippia
Tanais
Arcades
Arcades
Kastelli Kissamou

manumission
text
manumission
text
manumission
text
thanksgiving
amphora
epitaph
epitaph
epitaph

IJO 1, Dal1

Dalmatia

Peratovci

epitaph

IJO 1, Dal2
IJO 1, Dal3
IJO 1, Dal4
IJO 1,
Dalmatia
(tomb 281)
IJO 1, Mac2

Dalmatia
Dalmatia
Dalmatia

Senia
Salonae
Salonae

epitaph
epitaph
epitaph

45-63 AD
3rd Century AD
3rd-4th Century AD
Imperial
4-5th Century AD
2nd-3rd Century
AD
3rd-4th Century AD
4th Century AD
539 AD

Dalmatia

Doclea

tomb

3rd Century AD

Macedonia

Stobi

seal

IJO 1, Mac3

Macedonia

Stobi

vow

IJO 1, Mac4

Macedonia

Stobi

vow

IJO 1, Mac5

Macedonia

Stobi

votive

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

Macedonia
Macedonia
Macedonia
Macedonia
Macedonia
Macedonia
Macedonia
Macedonia
Macedonia

Beroea
Beroea
Beroea
Beroea
Beroea
Philippi
Thessaloniki
Thessaloniki
Thessaloniki

4th-6th Century AD
2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
5th Century AD
4-5th Century AD?
4th Century AD
4th Century AD?
4th Century AD?
3rd-4th Century AD
4th Century AD?
3rd Century AD
4th-6th Century AD

1,
1,
1,
1,
1,

1,
1,
1,
1,
1,
1,
1,
1,
1,

BS27
BS28
Cre1
Cre2
Cre3

Mac6
Mac7
Mac8
Mac9
Mac11
Mac12
Mac14
Mac15
Mac17

68 AD
59-60 AD
1-2nd Century AD

IJO 1, Pan5

Pannonia

IJO 1, Thr1
IJO 1, Thr2
IJO 1, Thr3
IJO 1, Thr4
IJO 1, Thr5
Ameling, W.,
Inscriptiones
Judaicae
Orientis,
Band II,
Kleinasien,
Tbingen:
Mohr
Siebeck,
2004, 4
IJO 2, 5
IJO 2, 5a
IJO 2, 6
IJO 2, 7

Thracia
Thracia
Thracia
Thracia
Thracia

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
dedication
synagogue
Mursa
inscription
Philippopolis
mosaic floor
Philippopolis
mosaic floor
Bizye
epitaph
Perinthus-Heraclea epitaph
Assenovgrad
votive

Asia Minor

Jaffa

epitaph

2nd-4th Century
AD

Asia
Asia
Asia
Asia

Samos
Ikaria
Kermeti
Kos town

decree
inscription
epitaph
epitaph

Imperial
5th-6th Century AD
Imperial
Imperial

Minor
Minor
Minor
Minor

297

198-210 AD
3rd Century AD
4th Century AD
4-5th Century AD
5th-6th Century AD
2nd Century AD

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,
2,
2,

8
9
10
11
14
15
16
17
18

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Palaispili
Rhodes
Rhodes
Kamiros
Aphrodisias
Aphrodisias
Aphrodisias
Aphrodisias
Aphrodisias

IJO 2, 20

Asia Minor

Hyllarima

IJO
IJO
IJO
IJO

Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor

Iasos
Iasos
Iasos
Kaunos

IJO 2, 25

Asia Minor

Myndos

IJO 2, 26

Asia Minor

Nysa

IJO 2, 27

Asia Minor

Tralles

IJO 2, 28

Asia Minor

Tralles

IJO
IJO
IJO
IJO
IJO
IJO

Asia
Asia
Asia
Asia
Asia
Asia

Ephesus
Ephesus
Ephesus
Ephesus
Ephesus
Ephesus

2,
2,
2,
2,

2,
2,
2,
2,
2,
2,

21
22
23
24

30
31
32
33
34
35

Minor
Minor
Minor
Minor
Minor
Minor

epitaph
list
donation
epitaph
donation
inscription
inscription
graffito
list
synagogue
inscription
list
list
epitaph
epitaph?
synagogue
inscription
building
inscription
donation
building
inscription
acclamation
barrier
epitaph
epitaph
epitaph
pastoral letter
building
inscription

3rd Century AD
1st Century AD
18 AD
2nd Century AD
3rd Century AD
5th-6th Century AD
5th-6th Century AD
4th Century AD
4th Century AD
3rd Century AD
2nd Century AD
80-174 AD
2nd Century AD
late Hellenistic
4th-6th Century AD
3rd-4th Century AD
3rd Century AD
1st Century AD
Imperial
4th Century AD
2nd Century AD
200 AD
212 AD
531-37 AD

IJO 2, 36

Asia Minor

Kyme

IJO 2, 37

Asia Minor

Miletus

inscription

IJO 2, 38

Asia Minor

Miletus

inscription

IJO 2, 39

Asia Minor

Miletus

inscription

IJO
IJO
IJO
IJO
IJO
IJO

Asia
Asia
Asia
Asia
Asia
Asia

Smyrna
Smyrna
Smyrna
Smyrna
Teos
Hypaipa
Magnesia ad
Sipylon
Philadelphia
Philadelphia
Philadelphia
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis

inscription
donation
donation
epitaph
donation
inscription

2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
124 AD
4-5th Century AD
2nd Century AD
Imperial
3rd Century AD
3rd Century AD

epitaph

late Imperial

donation
epitaph
epitaph
inscription
epitaph?
amulet
epitaph
amphora
amphora

3rd-4th Century AD
late Imperial
4th Century AD
200 AD
4th Century AD
3-5th Century AD
3rd Century AD
3rd-4th Century AD
3rd-4th Century AD
321-5th Century
AD
post 268 AD

2,
2,
2,
2,
2,
2,

40
41
43
45
46
47

Minor
Minor
Minor
Minor
Minor
Minor

IJO 2, 48

Asia Minor

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

2,
2,
2,
2,
2,
2,
2,
2,
2,

49
50
51
53
54
55
56
57
59

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

IJO 2, 60

Asia Minor

Sardis

mosaic

IJO 2, 62

Asia Minor

Sardis

donation

298

3rd Century AD

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,

63
64
66
67
68
69
72
73
76
77
78
79
83
84
86
87
88
90
91
92
93
94
95
97
98
100
105
106
107
108
109
114
118
119
121
122
123
124
125
129
132
133
134
135
136
146
147
148
149
150
151
152
153
154
155

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Thyateira
Cyzicus
Cyzicus
Amastris
Chalcedon
Chalcedon
Claudiopolis
Nicaea
Nicomedia
Nicomedia

prayer
prayer
prayer
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
blessing
inscription
inscription
inscription
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
epitaph
inscription
inscription
thanksgiving
epitaph
epitaph
epitaph
psalm
epitaph
epitaph

299

4-5th Century AD
341-438 AD
4-5th Century AD
351-378 AD
324-378 AD
341-383 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
5th-6th Century AD
5th-6th Century AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
5th Century AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
3rd-4th Century AD
3rd-4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th-6th Century AD
4th-6th Century AD
4th-6th Century AD
4th-6th Century AD
4th-6th Century AD
220-239 AD
5th Century AD
5th-6th Century AD
post 212 AD
4th Century AD
4th Century AD
3rd-4th Century AD
4th-6th Century AD
3rd Century AD
3rd Century AD

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,

IJO 2, 168

Asia Minor

Acmonia

IJO
IJO
IJO
IJO
IJO

Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor

Acmonia
Acmonia
Acmonia
Acmonia
Acmonia

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph?
epitaph
epitaph
epitaph
epitaph
epitaph
synagogue
inscription
prayer
prayer
donation
epitaph
epitaph

IJO 2, 174

Asia Minor

Acmonia

epitaph

IJO 2, 175
IJO 2, 176
IJO 2, 177

Asia Minor
Asia Minor
Asia Minor

Acmonia
Acmonia
Acmonia

epitaph
epitaph
epitaph

IJO 2, 178

Asia Minor

Acmonia

epitaph

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Apamea
Apollonia
Appia
Diocleia
Docimeion
Docimeion
Dorylaion
Eumeneia
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis

epitaph
epitaph
donation
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

IJO 2, 201

Asia Minor

Hierapolis

epitaph

IJO 2, 202

Asia Minor

Hierapolis

epitaph

IJO 2, 203

Asia Minor

Hierapolis

epitaph

IJO 2, 204

Asia Minor

Hierapolis

epitaph

2,
2,
2,
2,
2,

2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,

156
157
158
159
160
161
162
163
164
165
166
167
169
170
171
172
173

179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Nicomedia
Nicomedia
Nicomedia
Sebastopolis
Sebastopolis
Sebastopolis
Germa
Tavium
Tavium
Tavium
Tavium
Aizanoi

300

3rd Century AD
uncertain
3rd Century AD
4th Century AD
4th Century AD
4th Century AD
Byzantine?
5th-6th Century AD
5th-6th Century AD
5th-6th Century AD
4th Century AD
c. 175 AD
1st Century AD
Late Antique
4th Century AD
post 212 AD
243-4 AD
248-9 AD
2nd-3rd Century
AD
post 212 AD
3rd Century AD
255-6 AD
2nd-3rd Century
AD AD
post 212 AD
1st-2nd Century AD
Byzantine?
257-8 AD
5th-6th Century AD
3rd-4th Century AD
3rd Century AD
post 212 AD
3rd Century AD
3rd-4th Century AD
2nd Century AD
2nd Century AD
post 212 AD
2nd Century AD
post 212 AD
4th Century AD
2nd Century AD
post 212 AD
2nd Century AD
2nd Century AD
2nd Century AD
post 212 AD
2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD

IJO
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,

205
206
207
208
209
210
211

Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor

Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Cotiaeion
Cotiaeion

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

IJO 2, 213

Asia Minor

Laodicea ad Lycus

epitaph

IJO 2, 214

Asia Minor

Synnada

epitaph

IJO 2, 215

Asia Minor

Sibidunda

votive

IJO 2, 216
IJO 2, 218

Asia Minor
Asia Minor

Termessos
Aspendos

IJO 2, 219

Asia Minor

Side

IJO 2, 220

Asia Minor

Side

IJO
IJO
IJO
IJO
IJO
IJO

Asia
Asia
Asia
Asia
Asia
Asia

epitaph
prayer
synagogue
inscription
synagogue
inscription
epitaph
inscription
epitaph
epitaph
epitaph
epitaph

2,
2,
2,
2,
2,
2,

IJO 2, 227

Asia Minor

IJO 2, 228

Asia Minor

Limyra
Oinoanda
Tlos
Gdanmaa
Gdanmaa
Iconium
Laodicea
Katakekaumene
Sadahattin Hani

IJO 2, 229

Asia Minor

Aigai

epitaph

IJO 2, 230
IJO 2, 231
IJO 2, 232

Asia Minor
Asia Minor
Asia Minor

Diocaesarea
Diocaesarea
Korykos

prayer
epitaph
epitaph

IJO 2, 233

Asia Minor

Korykos

epitaph

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Korykos
Korykos
Korykos
Korykos
Korykos
Korykos
Korykos
Korykos
Korykos
Korykos
Seleuceia ad
Calycadnos
Seleuceia ad
Calycadnos
Seleuceia ad
Calycadnos
Selinus/Traianopoli
s

2,
2,
2,
2,
2,
2,
2,
2,
2,
2,

221
222
223
224
225
226

234
235
236
237
238
239
240
241
242
243

Minor
Minor
Minor
Minor
Minor
Minor

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

2nd Century AD
2nd Century AD
post 212 AD
post 212 AD
post 212 AD
3rd Century AD
Imperial
2nd-3rd Century
AD
1-2nd Century AD
2nd-3rd Century
AD
3rd Century AD
1-2nd Century AD
220 AD
363-390 AD
Imperial
2nd Century AD
1st Century AD
4th Century AD
4th Century AD
4th Century AD

epitaph

post 212 AD

epitaph

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

4th Century AD
2nd-4th Century
AD
4-5th Century AD
3rd Century AD
4th Century AD
2nd-3rd Century
AD
5th-6th Century AD
uncertain
3rd Century AD
4th Century AD
3rd Century AD
4th Century AD
late Roman
4th Century AD
4th Century AD
4th-6th Century AD

epitaph

4th Century AD

epitaph

4th Century AD

epitaph

Late Antique

epitaph

2nd Century AD

IJO 2, 244

Asia Minor

IJO 2, 245

Asia Minor

IJO 2, 246

Asia Minor

IJO 2, 247

Asia Minor

IJO 2, 248

Asia Minor

Tarsos

synagogue
inscription

6th Century AD

IJO 2, 249

Asia Minor

Tarsos

epitaph

2nd-3rd Century
AD

301

IJO 2, 250
IJO 2, 252
IJO 2, 253
IJO 2, 254
IJO 2, 255
IJO 2, 256
IJO 2, 258
Noy, D., and
Bloedhorn,
H., (eds.),
Inscriptiones
Judaicae
Orientis, vol.
III, Syria and
Cyprus,
Tbingen:
Mohr
Siebeck,
2004, App1
IJO 3, App2
IJO 3, App3
IJO 3, App4

Asia
Asia
Asia
Asia
Asia
Asia
Asia

Tarsos
Nevsehir
Nevsehir
Nevsehir
Nevsehir
Nevsehir
Tyana

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

4th Century AD
uncertain
uncertain
uncertain
uncertain
uncertain
uncertain

Judea

Beth She'arim

epitaph

uncertain

Judea
Judea
Judea

Beth She'arim
Beth She'arim
Beth She'arim

epitaph
epitaph
epitaph

IJO 3, App5

Judea

Beth She'arim

epitaph

IJO
IJO
IJO
IJO
IJO
IJO

App6
App7
App8
App9
Cyp1
Cyp2

Judea
Judea
Judea
Judea
Cyprus
Cyprus

Beth She'arim
Beth She'arim
Jerusalem
Jerusalem
Lapethos
Morfou

IJO 3, Cyp3

Cyprus

Golgoi

IJO
IJO
IJO
IJO
IJO
IJO
IJO

Cyprus
Cyprus
Cyprus
Cyprus
Cyprus
Syria
Syria

Salamis
Kourion
Kition
Kition
Kition
Tyre
Tyre

epitaph
epitaph
ossuary
ossuary
vow
seal/stamp
building
inscription
inscription
inscription
epitaph
epitaph
epitaph
epitaph
epitaph

3rd Century AD
uncertain
3rd Century AD
2nd-3rd Century
AD
3rd Century AD
uncertain
uncertain
uncertain
3rd-4th Century AD
3rd-4th Century AD

IJO 3, Syr3

Syria

Tyre

epitaph

IJO 3, Syr4
IJO 3, Syr5
IJO 3, Syr6
IJO 3, Syr7
IJO 3, Syr8
IJO 3, Syr9
IJO 3, Syr10
IJO 3, Syr11
IJO 3,
Syr111-126

Syria
Syria
Syria
Syria
Syria
Syria
Syria
Syria

Tyre
Tyre
Tyre
Tyre
Tyre
Tyre
Tyre
Tyre

amulet
dedication
epitaph
epitaph
epitaph
ossuary
pillar
column

3rd-4th Century AD
2nd-1st Century BC
4th Century BC
4th Century BC
4th Century BC
3rd-4th Century AD
uncertain
2nd-7th Century
AD
3rd-4th Century AD
4-5th Century AD
3rd Century AD
250-352 AD
2nd Century AD
1st Century AD?
5th-6th Century AD
5th-6th Century AD

Syria

Dura Europus

graffiti

253-4 AD

IJO 3, Syr12

Syria

Ornithopolis

IJO 3, Syr14
IJO 3, Syr16
IJO 3, Syr17

Syria
Syria
Syria

Sidon
Sidon
Sidon

3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,

Cyp4
Cyp5
Cyp6
Cyp7
Cyp8
Syr1
Syr2

Minor
Minor
Minor
Minor
Minor
Minor
Minor

synagogue
inscription
stamp
epitaph
epitaph

302

4th Century AD

uncertain
uncertain
uncertain
3rd Century AD

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,
3,

Syr18
Syr21
Syr22
Syr23
Syr24
Syr25
Syr26
Syr27
Syr28
Syr29
Syr30
Syr31
Syr32

Syria
Syria
Syria
Syria
Syria
Syria
Syria
Syria
Syria
Syria
Syria
Syria
Syria

IJO 3, Syr33

Syria

IJO 3, Syr34

Syria

Tafas

IJO 3, Syr35

Syria

Naveh

IJO 3, Syr36
IJO 3, Syr37
IJO 3, Syr38
IJO 3, Syr39
IJO 3, Syr40
IJO 3, Syr41
IJO 3, Syr42
IJO 3, Syr44,
45, 46, 47
IJO 3, Syr48
IJO 3, Syr49
IJO 3, Syr50
IJO 3, Syr51
IJO 3, Syr52
IJO 3, Syr53,
54

Syria
Syria
Syria
Syria
Syria
Syria
Syria

Naveh
Philippopolis
Damatha
Phaine
Qatana
Admedera
Damascus

Syria

Palmyra

Syria
Syria
Syria
Syria
Syria

IJO 3, Syr55

Syria

IJO 3, Syr56

Syria

IJO 3, Syr57

Syria

IJO 3, Syr58

Syria

IJO 3, Syr59

Syria

IJO 3, Syr60

Syria

IJO 3, Syr61

Syria

IJO 3, Syr62

Syria

IJO 3, Syr63

Syria

IJO 3, Syr64

Syria

IJO 3, Syr65

Syria

Palmyra
Palmyra
Palmyra
Palmyra
Palmyra
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on
Orontes
Apamea on

Syria

Sidon
Iamour
Chalcis
Mutatio Heldua
Berytus
Berytus
Berytus
Berytus
Byblos
Byblos
Byblos
Byblos
Byblos
Caesarea ad
Libanum

the
the
the
the
the
the
the
the
the
the
the
the

303

epitaph
epitaph
ossuary
mosaic
epitaph
epitaph
epitaph
amulet
epitaph
epitaph
epitaph
epitaph
epitaph

250-352 AD
3rd Century AD
1st Century AD
605-6 AD
5th-6th Century AD
3rd-4th Century AD
3rd Century AD
uncertain
4th Century AD
4th Century AD
385-6 AD
188-9 AD
3rd-4th Century AD

epitaph

3rd-4th Century AD

synagogue
inscription
synagogue
inscription
epitaph
epitaph
inscription
epitaph
officer list
inscription
amulet
biblical
inscription
graffito
epitaph
sale of tomb
epitaph
epitaph
synagogue
inscription
synagogue
inscription

4th Century AD
3rd-4th Century AD
4th Century AD
3rd-4th Century AD
uncertain
250-320 AD
3rd Century AD
5th-6th Century AD
3rd-4th Century AD
probably 3rd
Century AD
7th Century AD
212 AD
3rd Century AD
3rd Century AD
3rd Century AD
392 AD
392 AD

donation

392 AD

donation

392 AD

donation

392 AD

donation

392 AD

donation

392 AD

donation

392 AD

donation

392 AD

donation

392 AD

donation

392 AD

donation

392 AD

IJO 3, Syr66

Syria

IJO 3, Syr67

Syria

IJO 3, Syr68

Syria

IJO 3, Syr70

Syria

IJO 3, Syr71

Syria

IJO 3, Syr72

Syria

IJO 3, Syr73

Syria

IJO 3, Syr74

Syria

IJO
IJO
IJO
IJO
IJO
IJO

Syr75
Syr76
Syr78
Syr79
Syr80
Syr83

Syria
Syria
Syria
Syria
Syria
Syria

Orontes
Apamea on the
Orontes
Apamea on the
Orontes
Apamea on the
Orontes
Apamea on the
Orontes
Apamea on the
Orontes
Apamea on the
Orontes
Antioch on the
Orontes
Antioch on the
Orontes
Djebel Sim'an
Aleppo
Edessa
Edessa
Edessa
Dura Europus

IJO 3, Syr84

Syria

Dura Europus

IJO 3, Syr85

Syria

Dura Europus

IJO 3, Syr86
IJO 3, Syr87

Syria
Syria

Dura Europus
Dura Europus

IJO 3, Syr88

Syria

Dura Europus

IJO 3, Syr89
IJO 3, Syr90
IJO 3, Syr91
IJO 3, Syr92
IJO 3, Syr93
IJO 3, Syr94
IJO 3, Syr95
IJO 3, Syr96110
IJO 3,
Syr127
IJO 3,
Syr128
IJO 3,
Syr129
IJO 3,
Syr131
Noy, D., and
Horbury, W.,
Jewish
Inscriptions
of GraecoRoman
Egypt,
Cambridge:

Syria
Syria
Syria
Syria
Syria
Syria
Syria

Dura
Dura
Dura
Dura
Dura
Dura
Dura

Syria

3,
3,
3,
3,
3,
3,

donation

392 AD

donation

392 AD

donation

392 AD

blessing

392 AD

donation

392 AD

ossuary

1st Century BC/AD


uncertain

epitaph

200-352 AD

inscription
amulet
epitaph
epitaph
epitaph
graffito
building
inscription
building
inscription
donation
donation
building
inscription
graffito
graffito
graffito
graffito
graffito
graffito
graffito

272/3 AD
4-5th Century AD
1st-4th Century AD
1st-4th Century AD
1st-4th Century AD
pre 244 AD

Dura Europus

captions

244-5 AD

Syria

Dura Europus

graffito

pre 244 AD

Syria

Dura Europus

graffito

244-5 AD

Syria

Dura Europus

graffito

247-8 AD

Syria

Dura Europus

graffito

pre 254 AD

Egypt

Alexandria

epitaph

late Ptolemaic

Europus
Europus
Europus
Europus
Europus
Europus
Europus

304

244-5 AD
244-5 AD
244-5 AD
244-5 AD
244-5 AD
244-5 AD
pre 254 AD
pre 254 AD
240-1 AD
pre 254 AD
pre 254 AD
pre 254 AD

CUP, 1992, 1
JIGRE, 3
Egypt
JIGRE, 4
Egypt
JIGRE, 6
Egypt
JIGRE, 9
Egypt

Alexandria
Alexandria
Alexandria
Alexandria

epitaph
epitaph?
epitaph
dedication

JIGRE, 10

Egypt

Alexandria

epitaph

JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,

11
12
13
15
17
18
19

Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt

Alexandria
Alexandria
Alexandria
Alexandria
Alexandria
Alexandria
Alexandria

epitaph
epitaph
dedication
votive
votive
honorific
votive

JIGRE, 20

Egypt

Alexandria

dedication

JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,

22
24
25
26
27
28
29
34
36
38
39
40

Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt

Schedia
Xenephyris
Nitriai
Naucratis
Athribis
Athribis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis

dedication
dedication
dedication
statuette
dedication
dedication
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

JIGRE, 41

Egypt

Leontopolis

epitaph

JIGRE, 42

Egypt

Leontopolis

epitaph

JIGRE, 43

Egypt

Leontopolis

epitaph

JIGRE, 54

Egypt

Leontopolis

epitaph

JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,

55
56
57
58
59
60
62
63
65
74
76
81
84
86
88
93

Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt

Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

JIGRE, 95

Egypt

Leontopolis

epitaph

JIGRE, 96

Egypt

Leontopolis

epitaph

JIGRE, 98

Egypt

Leontopolis

epitaph

305

early Ptolemaic
early Ptolemaic
early Ptolemaic
2nd Century BC?
3rd-2nd Century
AD BC
late Ptolemaic
late Ptolemaic
37 BC
late Roman
Roman
3 AD
Roman
late PtolemaicRoman
246-221 BC
BC 140-116
BC 140-116
30 BC-AD 14
2nd-1st Century BC
2nd-1st Century BC
1st Century BC/AD
2nd Century BC/AD
2nd Century BC/AD
2nd Century BC/AD
2nd Century BC/AD
AD 8
2nd Century BC1st Century AD
2nd Century BC1st Century AD
AD 5
2nd Century BC1st Century AD
28 BC
2nd Century BC/AD
2nd Century BC/AD
23 BC?
1st Century BC/AD
55 or 4 BC
2nd Century BC/AD
2nd Century BC/AD
AD 1
5 BC
AD 4
1st Century BC/AD
27 BC
AD 5
2nd Century BC/AD
52 or 1 BC
27 BC or 1st
Century AD
58/7 BC
2nd Century BC-1st
Century AD

JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,

100
101
102
103
105
106
107
108
109
110
115
116

Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt

epitaph
epitaph
epitaph
epitaph
dedication
epitaph
epitaph
epitaph
epitaph
epitaph
donation
votive

26 BC
2nd Century BC/AD
2nd Century BC/AD
2nd Century BC/AD
2nd Century BC/AD
47 BC?
1st Century BC/AD
1st Century BC/AD
1st Century BC/AD
1st Century BC/AD
2nd Century BC?
29 BC

dedication

246-221 BC

Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt

Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
near Heliopolis
near Heliopolis
near Heliopolis
near Heliopolis
near Heliopolis
Fayum
Fayum
ArsinoeCrocodilopolis
near Al-Minya
Antinoopolis
Antinoopolis
El-Kanais
El-Kanais
El-Kanais
El-Kanais

JIGRE, 117

Egypt

JIGRE, 118
JIGRE, 119
JIGRE, 120
JIGRE, 121
JIGRE, 122
JIGRE, 123
JIGRE, 124
Noy, D.,
Jewish
Inscriptions
of Western
Europe,
Cambridge:
CUP, 1993, 3
JIWE, 10

epitaph?
epitaph
epitaph
thanksgiving
thanksgiving
graffito
graffito

2nd Century AD
2nd Century AD
2nd Century AD
2nd-1st Century
2nd-1st Century
2nd-1st Century
2nd-1st Century

Italia

Milan

epitaph

Italia

Ravenna

amphora

JIWE, 12

Italia

Lorium

JIWE, 13

Italia

Ostia

JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,

14
15
17
22
30
31
32
33
34
35
36
37
41
43
107
113
138

Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia

Ostia
Ostia
Porto
Brusciano
Naples
Naples
Naples
Naples
Naples
Naples
Naples
Naples
Herculaneum
Venosa
Venosa
Venosa
Hipponion

JIWE, 139

Italia

Rhegium

JIWE,
JIWE,
JIWE,
JIWE,

Italia
Italia
Italia
Italia

Bova Marina
Taormina
Catania
Catania

140
143
146
148

306

BC
BC
BC
BC

5th-6th Century AD
2nd-4th Century
metrical epitaph
AD
donation to
3rd Century AD
synagogue
epitaph
1st-2nd Century AD
honorific
253-60 AD
dedication
4th Century AD
epitaph
4-5th Century AD
epitaph
4-5th Century AD
epitaph
5th Century AD
epitaph
5th Century AD
epitaph
5th Century AD
epitaph
5th Century AD
epitaph
5th Century AD
epitaph
5th-6th Century AD
epitaph
4-5th Century AD
graffito
pre 79 AD
epitaph
4-5th Century AD
epitaph
521 AD
epitaph
4-5th Century AD
epitaph
3rd Century AD
building
4th Century AD?
inscription
mosaic
4-5th Century AD
graffito
4-5th Century AD
epitaph
4-5th Century AD
acquistion of
4-5th Century AD

JIWE, 149

Italia

Catania

JIWE, 150

Italia

Catania

JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,

Italia
Italia
Italia
Italia
Italia
Italia
Italia
Malta
Malta
Italia
Italia
Italia

Acrilla
Comiso
Sofiana
Sofiana
Sofiana
Agrigentum
Termini Imerese
Rabato
Rabato
Sulcis
Porto Torres
Porto Torres
Santa Maria del
Cami
Ibiza
Elche
Elche
Tortosa
Tarragona
Tarragona
near Tarragona
Villamesias

155
156
157
158
159
160
161
163
166
173
175
176

JIWE, 177

Majorca

JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,

Ibiza
Hispania
Hispania
Hispania
Hispania
Hispania
Hispania
Hispania

178
180
181
183
185
186
187
188

307

tomb
acquistion of
tomb
acquistion of
tomb
epitaph
amulet
epitaph
epitaph
amulet
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

4-5th Century AD
4-5th Century AD
4th Century AD
3-5th Century AD
5th Century AD
4th Century AD?
3-5th Century AD
4-5th Century AD
1st Century AD?
4-5th Century AD
4-5th Century AD
4-5th Century AD
4th Century AD?
4th Century AD?

epitaph

4-5th Century AD

amphora
mosaic floor
mosaic floor
epitaph
basin
epitaph
epitaph
epitaph

1st Century AD?


4th Century AD
4th Century AD?
5th-6th Century AD
5th-6th Century AD
5th-6th Century AD
4th-6th Century AD
1st-3rd Century AD

THE JEW ISH Di ASP ORA

By REGION
Region

Location

Type

Date

Publication
/s

Asia Minor

Acmonia

synagogue
inscription

1st Century AD

IJO 2, 168

Asia Minor

Acmonia

epitaph

Asia Minor

Acmonia

epitaph

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Acmonia
Acmonia
Acmonia
Acmonia
Acmonia
Acmonia
Acmonia
Acmonia

epitaph
epitaph
epitaph
epitaph
prayer
prayer
donation
epitaph

Asia Minor

Aigai

epitaph

Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor

Aizanoi
Amastris
Apamea
Aphrodisias
Aphrodisias
Aphrodisias

epitaph
thanksgiving
epitaph
donation
graffito
list

Asia Minor

Aphrodisias

inscription

Asia Minor

Aphrodisias

inscription

Asia Minor

Apollonia

epitaph

Asia Minor

Appia

donation

Asia Minor

Aspendos

prayer

Asia Minor
Asia Minor

Chalcedon
Chalcedon

epitaph
epitaph

Asia Minor

Claudiopolis

epitaph

Asia Minor
Asia Minor

Cotiaeion
Cotiaeion

epitaph
epitaph

Asia Minor

Cyzicus

inscription

Asia Minor
Asia Minor

Cyzicus
Diocaesarea

inscription
epitaph

Asia Minor

Diocaesarea

prayer

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

308

2nd-3rd Century
AD
2nd-3rd Century
AD
243-4 AD
248-9 AD
255-6 AD
3rd Century AD
4th Century AD
Late Antique
post 212 AD
post 212 AD
2nd-4th Century
AD
c. 175 AD
post 212 AD
post 212 AD
3rd Century AD
4th Century AD
4th Century AD
5th-6th Century
AD
5th-6th Century
AD
1st-2nd Century
AD
Byzantine?
1st-2nd Century
AD
4th Century AD
4th Century AD
3rd-4th Century
AD
3rd Century AD
Imperial
5th-6th Century
AD
5th Century AD
3rd Century AD
4th-5th Century
AD

IJO 2, 174
IJO 2, 178
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,
2,

172
173
177
176
170
169
171
175

IJO 2, 229
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,

167
149
179
14
17
18

IJO 2, 15
IJO 2, 16
IJO 2, 180
IJO 2, 181
IJO 2, 218
IJO 2, 150
IJO 2, 151
IJO 2, 152
IJO 2, 210
IJO 2, 211
IJO 2, 148
IJO 2, 147
IJO 2, 231
IJO 2, 230

Asia Minor

Diocleia

epitaph

Asia Minor

Docimeion

epitaph

Asia Minor

Docimeion

epitaph

Asia
Asia
Asia
Asia
Asia

Dorylaion
Ephesus
Ephesus
Ephesus
Ephesus

Minor
Minor
Minor
Minor
Minor

Asia Minor

Ephesus

Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor

Ephesus
Eumeneia
Gdanmaa
Gdanmaa
Germa

epitaph
epitaph
epitaph
epitaph
barrier
pastoral
letter
acclamation
epitaph
epitaph
epitaph
epitaph?

Asia Minor

Hierapolis

epitaph

Asia Minor

Hierapolis

epitaph

Asia Minor

Hierapolis

epitaph

Asia Minor

Hierapolis

epitaph

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

Asia Minor

Hierapolis

epitaph

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis
Hierapolis

Asia Minor

Hyllarima

Asia
Asia
Asia
Asia
Asia

Hypaipa
Iasos
Iasos
Iasos
Iconium

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
synagogue
inscription
inscription
list
epitaph
list
epitaph

Asia Minor

Ikaria

inscription

Asia Minor

Jaffa

epitaph

Minor
Minor
Minor
Minor
Minor

309

257-8 AD
3rd-4th Century
AD
5th-6th Century
AD
3rd Century AD
200 AD
212 AD
2nd Century AD
4th Century AD

IJO 2, 182

531-37 AD

IJO 2, 35

Imperial
post 212 AD
4th Century AD
4th Century AD
Byzantine?
2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
2nd Century AD
2nd Century AD
2nd Century AD
2nd Century AD
2nd Century AD
2nd Century AD
2nd Century AD
2nd Century AD
2nd Century AD
3rd-4th Century
AD
3rd Century AD
4th Century AD
post 212 AD
post 212 AD
post 212 AD
post 212 AD
post 212 AD
post 212 AD
post 212 AD

IJO
IJO
IJO
IJO
IJO

3rd Century AD

IJO 2, 20

3rd Century AD
2nd Century AD
2nd Century AD
80-174 AD
4th Century AD
5th-6th Century
AD
2nd-4th Century
AD

IJO
IJO
IJO
IJO
IJO

IJO 2, 184
IJO 2, 183
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,
2,
2,
2,

185
33
34
32
31
30
186
224
225
162

IJO 2, 201
IJO 2, 202
IJO 2, 203
IJO 2, 204
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,
2,
2,

189
190
192
195
197
198
199
205
206

IJO 2, 188
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,

187
194
191
193
196
200
207
208
209
47
21
23
22
226

IJO 2, 5a
IJO 2, 4

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Kamiros
Kaunos
Kermeti
Korykos
Korykos
Korykos
Korykos
Korykos
Korykos

epitaph
epitaph?
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

Asia Minor

Korykos

epitaph

Asia Minor
Asia Minor

Korykos
Korykos

epitaph
epitaph

Asia Minor

Korykos

epitaph

Asia Minor
Asia Minor

Korykos
Korykos

epitaph
epitaph

Asia Minor

Kos town

epitaph

Asia Minor

Kyme

Asia Minor

Laodicea ad Lycus

Asia Minor
Asia Minor
Asia Minor

Laodicea
Katakekaumene
Limyra
Magnesia ad
Sipylon

2nd Century AD
late Hellenistic
Imperial
2-3 C AD
3rd Century AD
3rd Century AD
4th Century AD
4th Century AD
4th Century AD
4th-6th Century
AD
4th Century AD
4th Century AD
5th-6th Century
AD
late Roman
uncertain
Imperial

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

building
inscription

3rd Century AD

IJO 2, 36

epitaph

2nd-3rd Century
AD

IJO 2, 213

epitaph

post 212 AD

IJO 2, 227

epitaph

Imperial

IJO 2, 221

epitaph

late Imperial

IJO 2, 48

Asia Minor

Miletus

inscription

Asia Minor

Miletus

inscription

Asia Minor

Miletus

inscription

Asia Minor

Myndos

Asia
Asia
Asia
Asia
Asia

Nevsehir
Nevsehir
Nevsehir
Nevsehir
Nevsehir

synagogue
inscription
epitaph
epitaph
epitaph
epitaph
epitaph

Asia Minor

Nicaea

psalm

Asia
Asia
Asia
Asia
Asia

Nicomedia
Nicomedia
Nicomedia
Nicomedia
Nicomedia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Asia Minor

Nysa

Asia Minor
Asia Minor

Oinoanda
Palaispili

epitaph
epitaph
epitaph
epitaph
epitaph
building
inscription
inscription
epitaph

Asia Minor

Philadelphia

donation

Asia Minor
Asia Minor

Philadelphia
Philadelphia

epitaph
epitaph

310

2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
4th-6th Century
AD
uncertain
uncertain
uncertain
uncertain
uncertain
4th-6th Century
AD
3rd Century AD
3rd Century AD
3rd Century AD
3rd Century AD
uncertain
3rd-4th Century
AD
2nd Century AD
3rd Century AD
3rd-4th Century
AD
4th Century AD
late Imperial

2,
2,
2,
2,
2,
2,
2,
2,
2,

11
24
6
233
236
238
232
237
239

IJO 2, 243
IJO 2, 241
IJO 2, 242
IJO 2, 234
IJO 2, 240
IJO 2, 235
IJO 2, 7

IJO 2, 37
IJO 2, 38
IJO 2, 39
IJO 2, 25
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,

252
253
254
255
256

IJO 2, 153
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,

154
155
156
158
157

IJO 2, 26
IJO 2, 222
IJO 2, 8
IJO 2, 49
IJO 2, 51
IJO 2, 50

Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor

Rhodes
Rhodes
Sadahattin Hani
Samos
Sardis

donation
list
epitaph
decree
inscription

Asia Minor

Sardis

mosaic

Asia Minor

Sardis

donation

Asia Minor

Sardis

donation

Asia Minor

Sardis

blessing

Asia Minor

Sardis

amphora

Asia Minor

Sardis

amphora

Asia Minor
Asia Minor

Sardis
Sardis

donation
prayer

Asia Minor

Sardis

amulet

Asia Minor
Asia Minor

Sardis
Sardis

donation
epitaph

Asia Minor

Sardis

prayer

Asia Minor

Sardis

prayer

Asia Minor

Sardis

donation

Asia Minor

Sardis

donation

Asia Minor

Sardis

donation

Asia Minor

Sardis

donation

Asia Minor

Sardis

donation

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis

inscription
inscription
inscription
donation
donation
donation
donation
donation
donation
donation
donation
donation
epitaph?

Asia Minor

Sardis

donation

Asia Minor

Sardis

donation

Asia Minor
Asia Minor
Asia Minor

Sardis
Sardis
Sardis

donation
donation
donation

311

18 AD
1st Century AD
4th Century AD
Imperial
200 AD
321-5th Century
AD
324-378 AD
3rd-4th Century
AD
3rd-4th Century
AD
3rd-4th Century
AD
3rd-4th Century
AD
341-383 AD
341-438 AD
3rd-5th Century
AD
351-378 AD
3rd Century AD
4th-5th Century
AD
4th-5th Century
AD
4th-6th Century
AD
4th-6th Century
AD
4th-6th Century
AD
4th-6th Century
AD
4th-6th Century
AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
4th Century AD
5th-6th Century
AD
5th-6th Century
AD
5th Century AD
post 268 AD
post 380 AD

IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,

10
9
228
5
53

IJO 2, 60
IJO 2, 68
IJO 2, 105
IJO 2, 106
IJO 2, 57
IJO 2, 59
IJO 2, 69
IJO 2, 64
IJO 2, 55
IJO 2, 67
IJO 2, 56
IJO 2, 63
IJO 2, 66
IJO 2, 132
IJO 2, 133
IJO 2, 134
IJO 2, 135
IJO 2, 136
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,

107
108
109
114
118
119
121
122
123
124
125
129
54

IJO 2, 83
IJO 2, 84
IJO 2, 94
IJO 2, 62
IJO 2, 100

Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia
Asia

Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor
Minor

Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sardis
Sebastopolis
Sebastopolis
Sebastopolis
Seleuceia ad
Calycadnos
Seleuceia ad
Calycadnos
Seleuceia ad
Calycadnos
Selinus/
Traianopolis

donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
donation
epitaph
epitaph
epitaph

post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
post 380 AD
4th Century AD
4th Century AD
4th Century AD

IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

epitaph

4th Century AD

IJO 2, 244

epitaph

4th Century AD

IJO 2, 245

epitaph

Late Antique

IJO 2, 246

epitaph

2nd Century AD

IJO 2, 247

Asia Minor

Sibidunda

votive

Asia Minor

Side

Asia Minor

Side

Asia Minor
Asia Minor

Smyrna
Smyrna

synagogue
inscription
synagogue
inscription
inscription
donation

Asia Minor

Smyrna

donation

Asia Minor

Smyrna

epitaph

Asia Minor

Synnada

epitaph

Asia Minor

Tarsos

epitaph

Asia Minor

Tarsos

epitaph

Asia Minor

Tarsos

Asia Minor

Tavium

synagogue
inscription
epitaph

Asia Minor

Tavium

epitaph

Asia Minor

Tavium

epitaph

Asia Minor

Tavium

epitaph

Asia Minor
Asia Minor
Asia Minor

Teos
Termessos
Thyateira

donation
epitaph
epitaph

Asia Minor
Asia Minor
Asia Minor
Asia Minor

312

2nd-3rd Century
AD
220 AD

2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,
2,

72
73
76
77
78
79
86
87
88
90
91
92
93
95
97
98
159
160
161

IJO 2, 215
IJO 2, 219

363-390 AD

IJO 2, 220

124 AD
2nd Century AD
4th-5th Century
AD
Imperial
1st-2nd Century
AD
2nd-3rd Century
AD
4th Century AD

IJO 2, 40
IJO 2, 43

6th Century AD

IJO 2, 248

4th Century AD
5th-6th Century
AD
5th-6th Century
AD
5th-6th Century
AD
3rd Century AD
3rd Century AD
220-239 AD

IJO 2, 166

IJO 2, 41
IJO 2, 45
IJO 2, 214
IJO 2, 249
IJO 2, 250

IJO 2, 163
IJO 2, 164
IJO 2, 165
IJO 2, 46
IJO 2, 216
IJO 2, 146

Asia Minor

Tlos

Asia Minor

Tralles

Asia Minor
Asia Minor
Baetica

Tralles
Tyana
Abdera

epitaph
building
inscription
donation
epitaph
epitaph

Black Sea

Chersonesos

graffito

Black Sea

Gorgippia

manumission

Black
Black
Black
Black
Black
Black

Gorgippia
Gorgippia
Gorgippia
Gorgippia
Gorgippia
Hermonassa

manumission
thanksgiving
manumission
manumission
manumission
epitaph

Black Sea

Panticipaeum

manumission

Black Sea

Panticipaeum

manumission

Black Sea

Panticipaeum

manumission

Black Sea

Panticipaeum

epitaph

Black Sea

Panticipaeum

dedication

Black Sea

Panticipaeum

epitaph

Black Sea

Panticipaeum

epitaph

Black Sea

Panticipaeum

epitaph

Black Sea

Panticipaeum

epitaph

Black Sea

Panticipaeum

epitaph

Black
Black
Black
Black
Black

Panticipaeum
Panticipaeum
Phanagoria
Phanagoria
Tanais

Sea
Sea
Sea
Sea
Sea
Sea

Sea
Sea
Sea
Sea
Sea

Cyprus

Golgoi

Cyprus
Cyprus
Cyprus

Kition
Kition
Kition

epitaph
manumission
manumission
manumission
amphora
building
inscription
epitaph
epitaph
epitaph

Cyprus

Kourion

inscription

Cyprus

Lapethos

vow

Cyprus

Morfou

seal/stamp

Cyprus

Salamis

inscription

Dalmatia

Doclea

tomb

313

1st Century AD

IJO 2, 223

1st Century AD

IJO 2, 28

3rd Century AD
uncertain
3rd Century AD
4th-5th Century
AD
1-2nd Century
AD
41 AD
45-63 AD
59-60 AD
68 AD
93-123 AD
1st Century AD
1-2nd Century
AD
1-2nd Century
AD
1-2nd Century
AD
2nd-3rd Century
AD
306 AD
3rd-4th Century
AD
3rd-4th Century
AD
3rd-4th Century
AD
3rd-4th Century
AD
4th-5th Century
AD?
4th Century AD?
80 AD
17 AD
52 AD
3rd Century AD

IJO 2, 27
IJO 2, 258
CIJ I, 665

4th Century AD

IJO 3, Cyp3

4th Century BC
4th Century BC
4th Century BC
2nd-1st Century
BC
3rd-4th Century
AD
3rd-4th Century
AD
3rd-4th Century
AD

IJO 3, Cyp6
IJO 3, Cyp7
IJO 3, Cyp8

3rd Century AD

IJO 1, BS2
IJO 1, BS24
IJO
IJO
IJO
IJO
IJO
IJO

1,
1,
1,
1,
1,
1,

BS20
BS27
BS23
BS22
BS21
BS19

IJO 1, BS6
IJO 1, BS7
IJO 1, BS9
IJO 1, BS15
IJO 1, BS4
IJO 1, BS10
IJO 1, BS11
IJO 1, BS12
IJO 1, BS13
IJO 1, BS16
IJO
IJO
IJO
IJO
IJO

1,
1,
1,
1,
1,

BS14
BS5
BS17
BS18
BS28

IJO 3, Cyp5
IJO 3, Cyp1
IJO 3, Cyp2
IJO 3, Cyp4
IJO 1,
Dalmatia
(tomb 281)

Dalmatia

Narona

stamp

Dalmatia

Peratovci

epitaph

Dalmatia
Dalmatia

Salonae
Salonae

epitaph
epitaph

Dalmatia

Senia

epitaph

Egypt
Egypt
Egypt

Alexandria
Alexandria
Alexandria

dedication
honorific
dedication

Egypt

Alexandria

epitaph

Egypt
Egypt
Egypt
Egypt
Egypt
Egypt

Alexandria
Alexandria
Alexandria
Alexandria
Alexandria
Alexandria

epitaph
epitaph?
epitaph
epitaph
epitaph
epitaph

Egypt

Alexandria

dedication

Egypt
Egypt
Egypt
Egypt
Egypt

Alexandria
Alexandria
Alexandria
Antinoopolis
Antinoopolis
ArsinoeCrocodilopolis

Egypt

Athribis

dedication

Egypt

Athribis

dedication

Egypt

El-Kanais

thanksgiving

Egypt

El-Kanais

thanksgiving

Egypt

El-Kanais

graffito

Egypt

El-Kanais

graffito

Egypt
Egypt

Fayum
Fayum

votive
donation

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

IJO 1

votive
votive
votive
epitaph
epitaph

4th Century AD
2nd-3rd Century
AD
4th Century AD
539 AD
3rd-4th Century
AD
2nd Century BC?
3 AD
37 BC
3rd-2nd Century
BC
early Ptolemaic
early Ptolemaic
early Ptolemaic
late Ptolemaic
late Ptolemaic
late Ptolemaic
late PtolemaicRoman
late Roman
Roman
Roman
2nd Century AD
2nd Century AD

dedication

246-221 BC

JIGRE, 117

314

2nd-1st Century
BC
2nd-1st Century
BC
2nd-1st Century
BC
2nd-1st Century
BC
2nd-1st Century
BC
2nd-1st Century
BC
29 BC
2nd Century BC?
1st Century
BC/AD
1st Century
BC/AD
1st Century
BC/AD
2nd Century BC1st Century AD
2nd Century BC1st Century AD
2nd Century BC1st Century AD
2nd Century BC1st Century AD
2nd Century

IJO 1, Dal1
IJO 1, Dal3
IJO 1, Dal4
IJO 1, Dal2
JIGRE, 9
JIGRE, 18
JIGRE, 13
JIGRE, 10
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,

3
4
6
1
11
12

JIGRE, 20
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,

15
17
19
119
120

JIGRE, 27
JIGRE, 28
JIGRE, 121
JIGRE, 122
JIGRE, 123
JIGRE, 124
JIGRE, 116
JIGRE, 115
JIGRE, 29
JIGRE, 59
JIGRE, 81
JIGRE, 41
JIGRE, 42
JIGRE, 54
JIGRE, 98
JIGRE, 101

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

dedication

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

Epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt

Leontopolis

epitaph

Egypt
Egypt
Egypt

Leontopolis
Leontopolis
Leontopolis

epitaph
epitaph
epitaph

Egypt

Leontopolis

epitaph

Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt
Egypt

Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Leontopolis
Naucratis
near Al-Minya

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
statuette
epitaph?

Egypt

near Heliopolis

epitaph

Egypt

near Heliopolis

epitaph

Egypt

near Heliopolis

epitaph

Egypt

near Heliopolis

epitaph

Egypt
Egypt
Egypt
Egypt

near Heliopolis
Nitriai
Schedia
Xenephyris

epitaph
dedication
dedication
dedication

Etruria

Civitavecchia

epitaph

315

BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
2nd Century
BC/AD
23 BC?
26 BC
27 BC
27 BC or 1st
Century AD
28 BC
5 BC
52 or 1 BC
55 or 4 BC
58/7 BC
AD 1
AD 4
AD 5
AD 5
AD 8
30 BC-AD 14
2nd Century AD
1st Century
BC/AD
1st Century
BC/AD
1st Century
BC/AD
1st Century
BC/AD
47 BC?
140-116 BC
246-221 BC
140-116 BC
2nd-4th Century
AD

JIGRE, 102
JIGRE, 103
JIGRE, 105
JIGRE, 34
JIGRE, 36
JIGRE, 38
JIGRE, 39
JIGRE, 56
JIGRE, 57
JIGRE, 62
JIGRE, 63
JIGRE, 88
JIGRE, 58
JIGRE, 100
JIGRE, 84
JIGRE, 95
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,
JIGRE,

55
74
93
60
96
65
76
43
86
40
26
118

JIGRE, 107
JIGRE, 108
JIGRE, 109
JIGRE, 110
JIGRE,
JIGRE,
JIGRE,
JIGRE,

106
25
22
24

CIJ I, 636

Gallia

Arles

plaque

CIJ I, 669

seal

7th-8th Century
AD
4th Century AD?

Gallia

Auch

donation

Gallia
Germania
Superior
Greece
Greece
Greece

Avignon
Badenweiler

magical text

CIJ I, 674

Aegina
Aegina
Aegina

magical text
mosaic floor
mosaic floor

CIJ I, 724
CIJ I, 722
CIJ I, 723

Greece

Almyra

epitaph

Greece

Arcades

epitaph

Greece
Greece

Arcades
Arcadia

epitaph
epitaph

Greece

Argos

epitaph

Greece
Greece
Greece
Greece
Greece

Athens
Athens
Athens
Athens
Athens

epitaph
epitaph
epitaph
epitaph
epitaph

Greece

Athens

epitaph

Greece
Greece
Greece

Athens
Athens
Athens

Greece

Athens

honorific
epitaph
honorific
thiasos
inscription

Greece

Athens

epitaph

Greece

Athens

epitaph

Greece
Greece
Greece

Athens
Athens
Athens

Greece

Corinth

3rd Century AD

CIJ I, 718

Greece
Greece
Greece
Greece
Greece
Greece
Greece

Corinth
Corinth
Coronea
Delos
Delos
Delos
Delos

epitaph
epitaph
epitaph
synagogue
inscription
epitaph
epitaph
ephebic list
vow
vow
vow
list

300-50 AD
300-50 AD
5th-7th Century
AD
3rd-4th Century
AD
Imperial
2nd Century AD
3rd-4th Century
AD
1st Century AD
1st Century AD
1st Century AD?
1st Century AD?
1st Century AD?
2nd-3rd Century
AD
27-4 BC
2nd Century BC?
37-27 BC
4th-3rd Century
BC
4th-5th Century
AD
5th-6th Century
AD
6th Century AD?
Roman
uncertain

IJO
IJO
IJO
IJO
IJO
IJO
IJO

Greece

Delos

vow

Greece

Delos

vow

Greece
Greece
Greece

Delos
Delos
Delos

honorific
statue
honorific

Greece

Delphi

manumission

Greece
Greece

Delphi
Delphi

manumission
manumission

3rd Century AD
4th Century AD
246 AD
1st Century BC
1st Century BC
1st Century BC
100 AD
1-2nd Century
AD
1-2nd Century
AD
250-175 BC
4-39 AD
150-50 BC
2nd-1st Century
BC
158-7 BC
162 BC

316

CIJ I, 671
CIJ I, 667

IJO 1, Ach24
IJO 1, Cre1
IJO 1, Cre2
IJO 1, Ach52
IJO 1, Ach51
CIJ I, 715f
CIJ I, 715g
IJO 1, Ach26
IJO 1, Ach35
IJO 1, Ach36
IJO 1, Ach27
IJO 1, Ach39
IJO 1, Ach33
IJO 1, Ach38
IJO 1, Ach41
CIJ I, 712
CIJ I, 713
CIJ I, 715
IJO 1, Ach31
CIJ I, 715e
1,
1,
1,
1,
1,
1,
1,

Ach48
Ach49
Ach53
Ach62
Ach63
Ach65
Ach68

IJO 1, Ach60
IJO 1, Ach61
IJO 1, Ach66
IJO 1, Ach69
IJO 1, Ach67
CIJ I, 711
CIJ I, 709
CIJ I, 710

Greece

Kalyvia

epitaph

Greece

Kastelli Kissamou

epitaph

Greece

Larisa

epitaph

Greece

Larisa

epitaph

Greece

Larisa

epitaph

Greece
Greece
Greece
Greece
Greece

Larisa
Mantinea
Oropus
Patras
Patras

epitaph
dedication
manumission
inscription
magical text

Greece

Phthiotic Thebes

epitaph

Greece

Phthiotic Thebes

epitaph

Greece

Phthiotic Thebes

epitaph

Greece
Greece
Greece
Greece
Greece
Greece

Phthiotic
Phthiotic
Phthiotic
Phthiotic
Piraeus
Plataea

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph

Greece

Rheneia

epitaph

Greece
Greece
Greece
Greece
Greece
Hispania
Hispania

Syros
Syros
Syros
Taenarum
Taenarum
Elche
Elche

dedication
thanksgiving
prayer
epitaph
epitaph
mosaic floor
mosaic floor

Hispania

near Tarragona

epitaph

Hispania

Tarragona

basin

Hispania

Tarragona

epitaph

Hispania

Tortosa

epitaph

Hispania

Villamesias

epitaph

Ibiza
Italia

Ibiza
Acrilla

amphora
epitaph

Italia

Agrigentum

epitaph

Italia
Italia

Agrigentum
Aquileia

epitaph
epitaph?

Italia

Bari

inscription

Italia

Bova Marina

mosaic

Italia

Brescia

epitaph?

Thebes
Thebes
Thebes
Thebes

317

uncertain
4th-5th Century
AD
1st-4th Century
AD
2nd-3rd Century
AD
4th-6th Century
AD
uncertain
4th Century AD
300-250 BC
3rd-4th Century
AD
3rd-4th Century
AD
3rd-7th Century
AD
3rd Century AD
3rd Century AD
3rd Century AD
uncertain
uncertain
2nd Century AD
end 2nd Century
BC
37-4 BC
4th Century AD
4th Century AD?
1st Century AD?
3rd Century AD
4th Century AD
4th Century AD?
4th-6th Century
AD
5th-6th Century
AD
5th-6th Century
AD
5th-6th Century
AD
1st-3rd Century
AD
1st Century AD?
4th Century AD
4th-5th Century
AD
5th Century AD
1st Century BC
6th-8th Century
AD
4th-5th Century
AD
4th Century AD

IJO 1, Ach9
IJO 1, Cre3
IJO 1, Ach1
IJO 1, Ach6
IJO 1, Ach5
IJO 1, Ach7
CIJ I, 720
IJO 1, Ach45
CIJ I, 716
CIJ I, 717
IJO 1, Ach17
IJO 1, Ach23
IJO 1, Ach 16
IJO 1, Ach18
IJO 1, Ach20
IJO 1, Ach21
IJO 1, Ach15
CIJ I, 715i
IJO 1, Ach46
CIJ I, 725
IJO 1,
IJO 1,
IJO 1,
IJO 1,
IJO 1,
JIWE,
JIWE,

Ach74
Ach73
Ach72
Ach55
Ach56
180
181

JIWE, 187
JIWE, 185
JIWE, 186
JIWE, 183
JIWE, 188
JIWE, 178
JIWE, 155
JIWE, 160
CIJ I, 654
CIJ I, 643
CIJ I, 633
JIWE, 140
CIJ I, 639

Italia

Brescia

epitaph?

Italia

Brusciano

epitaph

Italia

Capua

epitaph

Italia

Castel Porziano

fragment

Italia

Catania

epitaph

Italia

Catania

epitaph

Italia

Catania

Italia

Catania

Italia

Catania

Italia

Comiso

amulet

Italia

Concordia

epitaph

Italia
Italia
Italia
Italia

Ferrara
Fondi
Herculaneum
Hipponion

Italia

Lorium

Italia
Italia
Italia
Italia
Italia

Macomer
Milan
Milan
Milan
Milan

inscription
epitaph
graffito
epitaph
metrical
epitaph
ring
epitaph
epitaph
epitaph
epitaph

Italia

Naples

epitaph

Italia

Naples

epitaph

Italia
Italia
Italia
Italia
Italia

Naples
Naples
Naples
Naples
Naples

epitaph
epitaph
epitaph
epitaph
epitaph

Italia

Naples

epitaph

Italia

Naples

epitaph

Italia

Naples

epitaph

Italia
Italia

Naples
Naples

epitaph
seal

Italia

Nola

lamp

Italia

Oria

epitaph

Italia

Oria

stele

Italia

Ostia

epitaph

acquistion of
tomb
acquistion of
tomb
acquistion of
tomb

318

or earlier
4th Century AD?
4th-5th Century
AD
2nd-4th Century
AD
First half 2nd
Century AD
383 AD
4th-5th Century
AD
4th-5th Century
AD
4th-5th Century
AD
4th-5th Century
AD
3rd-5th Century
AD
4th-5th Century
AD
uncertain
pre 79 AD
3rd Century AD
2nd-4th Century
AD
5th Century AD
5th Century AD
4th-5th Century
AD
4th-5th Century
AD
5th Century AD
5th Century AD
5th Century AD
5th Century AD
5th Century AD
5th-6th Century
AD
5th-6th Century
AD
5th-6th Century
AD
70-95 AD
uncertain
7th Centuy AD
or later?
5th-6th Century
AD
7th-8th Century
AD
1-2nd Century

CIJ I, 638
JIWE, 22
CIJ I, 553
CIJ I, 533
CIJ I, 650
JIWE, 146
JIWE, 148
JIWE, 149
JIWE, 150
JIWE, 156
CIJ I, 640
CIJ I, 637
CIJ I, 552
JIWE, 41
JIWE, 138
JIWE, 12
CIJ I, 656
CIJ I, 645
JIWE, 3
CIJ I, 644
CIJ I, 646
JIWE, 30
JIWE, 37
JIWE,
JIWE,
JIWE,
JIWE,
JIWE,

31
32
33
34
35

CIJ I, 558
CIJ I, 559
JIWE, 36
CIJ I, 556
CIJ I, 555
CIJ I, 554
CIJ I, 635
CIJ I, 634
JIWE, 14

Italia

Ostia

disc

Italia

Ostia

Italia

Ostia

Italia
Italia

Otranto
Pola

honorific
donation to
synagogue
epitaph
inscription

Italia

Pola

epitaph

Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia
Italia

Pompeii
Pompeii
Pompeii
Pompeii
Pompeii
Porto
Porto
Porto
Porto
Porto
Porto
Porto
Porto
Porto
Porto Torres
Porto Torres
Puzzeoli

graffiti
amphora
graffiti
graffiti
graffiti
plaque
plaque
plaque
fragment
fragment
dedication
plaque
plaque
plaque
epitaph
epitaph
epitaph?

Italia

Ravenna

amphora

Italia
Italia
Italia
Italia

region of Via
Portuensis
region of Via
Portuensis
region of Via
Portuensis
region of Via
Portuensis

AD
2nd-4th Century
AD
253-60 AD

CIJ I, 534

3rd Century AD

JIWE, 13

3rd Century AD
3rd-5th Century
AD
pre 79 AD
pre 79 AD
pre 79 AD
pre 79 AD
pre 79 AD
2nd Century AD
4th Century AD

CIJ I, 632
CIJ I, 641

4th Century AD?


4th Century AD?
1st Century AD
5th-6th Century
AD

JIWE, 15

CIJ I, 642
CIJ I, 562
CIJ I, 563
CIJ I, 564
CIJ I, 566
CIJ I, 567
CIJ I, 537
CIJ I, 538
CIJ I, 543
CIJ I, 548
CIJ I, 539
JIWE, 17
CIJ I, 535
CIJ I, 536
CIJ I, 540
JIWE, 175
JIWE, 176
CIJ I, 561
JIWE, 10

fragment

CIJ I, 494

sarcophagus

CIJ I, 496

plaque

CIJ I, 497

fragment

CIJ I, 499

4th Century AD?

JIWE, 139

3rd-5th Century
AD
4th Century AD?
5th Century AD
4th-5th Century
AD
4th-5th Century
AD
4th-5th Century
AD?
4th-5th Century
AD
4th-5th Century
AD
4th-5th Century
AD

CIJ I, 568

Italia

Rhegium

Italia

Salerno

building
inscription
inscription

Italia

Sofiana

amulet

Italia
Italia

Sofiana
Sofiana

epitaph
epitaph

Italia

Sulcis

epitaph

Italia

Sulcis

epitaph

Italia

Sulcis

ring

Italia

Syracuse

epitaph

Italia

Syracuse

epitaph

Italia

Taormina

graffito

319

JIWE, 159
JIWE, 158
JIWE, 157
CIJ I, 658
JIWE, 173
CIJ I, 657
CIJ I, 651
CIJ I, 652
JIWE, 143

Italia

Tarentum

inscription

Italia

Tarentum

epitaph

Italia

Tarentum

epitaph

Italia

Tarentum

epitaph

Italia

Tarentum

epitaph

Italia

Tarentum

epitaph

Italia

Tarentum

epitaph

Italia

Tarentum

epitaph

Italia

Tarentum

epitaph

Italia

Tarentum

epitaph

Italia

Termini Imerese

epitaph

Italia

Venosa

epitaph

Italia

Venosa

painted
epitaph

Italia

Venosa

epitaph

Italia

Venosa

epitaph

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia
Italia

Venosa
Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
epitaph
painted
epitaph
epitaph
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
epitaph
painted
epitaph
painted

320

3rd-6th Century
AD
4th Century AD
7th-8th Century
AD
7th-8th Century
AD
7th-8th Century
AD
7th-8th Century
AD
7th-8th Century
AD
7th-8th Century
AD
7th-8th Century
AD
1st Century AD?
4th-5th Century
AD
4th-5th Century
AD
4th-5th Century
AD
4th-5th Century
AD

CIJ I, 620

5th Century AD

CIJ I, 575

5th Century AD

CIJ I, 576

5th Century AD

CIJ I, 578

5th Century AD

CIJ I, 579

5th Century AD

CIJ I, 581

5th Century AD

CIJ I, 584

5th Century AD
5th Century AD

CIJ I, 586
CIJ I, 587

5th Century AD

CIJ I, 590

5th Century AD

CIJ I, 594

5th Century AD

CIJ I, 595

5th Century AD

CIJ I, 597

5th Century AD

CIJ I, 599

5th Century AD

CIJ I, 600

521 AD
5th-6th Century
AD
5th-6th Century

JIWE, 107

CIJ I, 627
CIJ I, 628
CIJ I, 621
CIJ I, 622
CIJ I, 623
CIJ I, 625
CIJ I, 629
CIJ I, 630
CIJ I, 631
JIWE, 161
CIJ I, 585
CIJ I, 593
JIWE, 113
JIWE, 43

CIJ I, 569
CIJ I, 570

epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
inscription
painted
epitaph
painted
epitaph
painted
epitaph

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Venosa

Italia

Via Appia

inscription

Italia

Via Appia

plaque

Italia

Via Appia

Italia

Via Appia

inscription

Italia

Via Appia

fragment

Italia

Via Appia

plaque

Italia

Via Appia

sarcophagus

Italia

Via Appia

fragment

Italia

Via Appia

plaque

Italia

Via Appia

painted
plaque

Italia

Via Appia

plaque

Italia

Via Appia

fragment

Italia

Via Appia

inscription

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

321

AD
5th-6th Century
AD
5th-6th Century
AD
5th-6th Century
AD
5th-6th Century
AD
5th-6th Century
AD
5th-6th Century
AD
early 6th
Century AD
early 6th
Century AD
early 6th
Century AD
early 6th
Century AD
mid 6th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century

CIJ I, 596
CIJ I, 606
CIJ I, 607
CIJ I, 609
CIJ I, 610
CIJ I, 612
CIJ I, 608
CIJ I, 611
CIJ I, 613
CIJ I, 616
CIJ I, 614
CIJ I, 100
CIJ I, 102
CIJ I, 103
CIJ I, 105
CIJ I, 106
CIJ I, 108
CIJ I, 109
CIJ I, 110
CIJ I, 111
CIJ I, 113
CIJ I, 118
CIJ I, 119
CIJ I, 120
CIJ I, 121
CIJ I, 122
CIJ I, 125
CIJ I, 126

Italia

Via Appia

fragment

Italia

Via Appia

fragment

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

fragment

Italia

Via Appia

inscription

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

painted
plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

Plaque

Italia

Via Appia

plaque

Italia

Via Appia

fragment

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

graffiti

Italia

Via Appia

graffiti

Italia

Via Appia

fragment

Italia

Via Appia

painted
epitaph

Italia

Via Appia

fragment

Italia

Via Appia

fragment

322

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 130
CIJ I, 132
CIJ I, 136
CIJ I, 138
CIJ I, 139
CIJ I, 140
CIJ I, 141
CIJ I, 142
CIJ I, 143
CIJ I, 145
CIJ I, 146
CIJ I, 147
CIJ I, 148
CIJ I, 149
CIJ I, 150
CIJ I, 151
CIJ I, 152
CIJ I, 153
CIJ I, 155
CIJ I, 156
CIJ I, 157
CIJ I, 161
CIJ I, 163
CIJ I, 165
CIJ I, 166
CIJ I, 167
CIJ I, 173
CIJ I, 176

Italia

Via Appia

fragment

Italia

Via Appia

fragment

Italia

Via Appia

painted
plaque

Italia

Via Appia

plaque

Italia

Via Appia

fragment

Italia

Via Appia

fragment

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

inscription

Italia

Via Appia

inscription

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

Plaque

Italia

Via Appia

plaque

Italia

Via Appia

inscription

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

painted
epitaph

Italia

Via Appia

plaque

Italia

Via Appia

fragment

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

graffiti

323

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 180
CIJ I, 193
CIJ I, 200
CIJ I, 201
CIJ I, 202
CIJ I, 203
CIJ I, 206
CIJ I, 209
CIJ I, 210
CIJ I, 212
CIJ I, 213
CIJ I, 215
CIJ I, 217
CIJ I, 221
CIJ I, 222
CIJ I, 225
CIJ I, 228
CIJ I, 229
CIJ I, 230
CIJ I, 231
CIJ I, 233
CIJ I, 246
CIJ I, 248
CIJ I, 249
CIJ I, 250
CIJ I, 252
CIJ I, 254
CIJ I, 255

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

graffiti

Italia

Via Appia

plaque

Italia

Via Appia

inscription

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

inscription

Italia

Via Appia

fragment

Italia

Via Appia

sarcophagus

Italia

Via Appia

Plaque

Italia

Via Appia

graffiti

Italia

Via Appia

plaque

Italia

Via Appia

fragment

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

plaque

Italia

Via Appia

fragment

Italia

Via Flaminia

Italia

Via Flaminia

Italia

Via Labicana

Italia

Via Labicana

Italia

Via Labicana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted

324

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 256
CIJ I, 257
CIJ I, 260
CIJ I, 261
CIJ I, 263
CIJ I, 265
CIJ I, 268
CIJ I, 269
CIJ I, 281
CIJ I, 282
CIJ I, 283
CIJ I, 285
CIJ I, 87
CIJ I, 88
CIJ I, 89
CIJ I, 90
CIJ I, 95
CIJ I, 97
CIJ I, 99
CIJ I, 1
CIJ I, 2
CIJ I, 73
CIJ I, 75
CIJ I, 77

CIJ I, 10

CIJ I, 11

CIJ I, 12

CIJ I, 13

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia
Italia

Via Nomentana
Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

epitaph
fragment
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
graffiti
painted
epitaph
plaque
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
plaque
painted
epitaph
fragment
fragment
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
stele
painted
epitaph
plaque
painted
epitaph
painted

325

CIJ I, 14

CIJ I, 15

CIJ I, 16

CIJ I, 17

CIJ I, 18

CIJ I, 19

CIJ I, 20

CIJ I, 21

CIJ I, 22

CIJ I, 23

CIJ I, 24

CIJ I, 25

CIJ I, 26

CIJ I, 27

CIJ I, 28

CIJ I, 30

CIJ I, 31

CIJ I, 32

CIJ I, 33

CIJ I, 34

CIJ I, 35

CIJ I, 36
CIJ I, 37

CIJ I, 38

CIJ I, 39

CIJ I, 41

CIJ I, 42

CIJ I, 43

CIJ I, 44

CIJ I, 45

CIJ I, 46

CIJ I, 47

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia

Via Nomentana

Italia
Italia

Via Nomentana
Via Nomentana

Italia

Via Nomentana

Italia
Italia

Via Nomentana
Via Nomentana

Italia

Via Nomentana

Italia
Italia

Via Nomentana
Via Ostiensis

epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
painted
epitaph
plaque
plaque
painted
epitaph
plaque
sarcophagus
painted
epitaph
tablet
sarcophagus

Italia

Via Portuensis

plaque

Italia

Via Portuensis

fragment

Italia

Via Portuensis

painted
brick

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

Inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

326

CIJ I, 48

CIJ I, 50

CIJ I, 51

CIJ I, 52

CIJ I, 53

CIJ I, 55

CIJ I, 56

CIJ I, 6

CIJ I, 67

CIJ I, 68
CIJ I, 69

CIJ I, 7

CIJ I, 70
CIJ I, 72

CIJ I, 8

2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

CIJ I, 9
CIJ I, 287
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 290
CIJ I, 291
CIJ I, 293
CIJ I, 296
CIJ I, 298
CIJ I, 299
CIJ I, 300
CIJ I, 301
CIJ I, 304
CIJ I, 306
CIJ I, 307
CIJ I, 309
CIJ I, 310
CIJ I, 312

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

painted
epitaph

Italia

Via Portuensis

fragment

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

fragment

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

Fragment

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

327

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 315
CIJ I, 316
CIJ I, 317
CIJ I, 318
CIJ I, 319
CIJ I, 321
CIJ I, 323
CIJ I, 324
CIJ I, 325
CIJ I, 327
CIJ I, 328
CIJ I, 329
CIJ I, 331
CIJ I, 332
CIJ I, 333
CIJ I, 334
CIJ I, 335
CIJ I, 336
CIJ I, 337
CIJ I, 340
CIJ I, 343
CIJ I, 345
CIJ I, 346
CIJ I, 347
CIJ I, 348
CIJ I, 349
CIJ I, 351
CIJ I, 353

Italia

Via Portuensis

plaque

Italia

Via Portuensis

fragment

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

painted
brick

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

fragment

Italia

Via Portuensis

plaque

Italia

Via Portuensis

Plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

fragment

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

painted
plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

painted

328

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 354
CIJ I, 355
CIJ I, 358
CIJ I, 361
CIJ I, 362
CIJ I, 364
CIJ I, 367
CIJ I, 368
CIJ I, 369
CIJ I, 370
CIJ I, 371
CIJ I, 372
CIJ I, 373
CIJ I, 374
CIJ I, 375
CIJ I, 376
CIJ I, 378
CIJ I, 379
CIJ I, 380
CIJ I, 382
CIJ I, 383
CIJ I, 384
CIJ I, 385
CIJ I, 390
CIJ I, 391
CIJ I, 392
CIJ I, 394
CIJ I, 395

epitaph
Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

painted
epitaph

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

fragment

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

painted
epitaph

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

fragment

Italia

Via Portuensis

fragment

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

329

AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th
AD
2nd-4th

Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century
Century

CIJ I, 396
CIJ I, 397
CIJ I, 398
CIJ I, 399
CIJ I, 400
CIJ I, 401
CIJ I, 403
CIJ I, 405
CIJ I, 408
CIJ I, 409
CIJ I, 411
CIJ I, 413
CIJ I, 416
CIJ I, 417
CIJ I, 418
CIJ I, 419
CIJ I, 425
CIJ I, 433
CIJ I, 456
CIJ I, 457
CIJ I, 458
CIJ I, 460
CIJ I, 462
CIJ I, 463
CIJ I, 464
CIJ I, 465
CIJ I, 466
CIJ I, 467

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

plaque

Italia

Via Portuensis

fragment

Italia

Via Portuensis

plaque

Italia

Via Portuensis

inscription

Italia

Via Portuensis

plaque

Italia
Italia
Italia

Via Salaria
Via Salaria
Via Salaria

plaque
plaque
plaque

Judea

Beth She'arim

epitaph

Judea
Judea
Judea
Judea
Judea
Judea
Judea
Judea

Beth She'arim
Beth She'arim
Beth She'arim
Beth She'arim
Beth She'arim
Beth She'arim
Jerusalem
Jerusalem

epitaph
epitaph
epitaph
epitaph
epitaph
epitaph
ossuary
ossuary

Macedonia

Beroea

epitaph

Macedonia
Macedonia
Macedonia
Macedonia

Beroea
Beroea
Beroea
Beroea

epitaph
epitaph
epitaph
epitaph

Macedonia

Philippi

epitaph

Macedonia

Stobi

donation

Macedonia

Stobi

vow

Macedonia

Stobi

vow

Macedonia

Stobi

votive

Macedonia

Stobi

seal

Macedonia

Thessaloniki

epitaph

Macedonia

Thessaloniki

dedication

Macedonia
Macedonia

Thessaloniki
Thessaloniki

epitaph
epitaph

330

AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
2nd-4th Century
AD
330 AD, 14 May
beg. 2nd Century
AD
2nd-3rd Century
AD
3rd Century AD
3rd Century AD
3rd Century AD
uncertain
uncertain
uncertain
uncertain
uncertain
4th-5th Century
AD
4th Century AD
4th Century AD?
4th Century AD?
5th Century AD
3rd-4th Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
2nd-3rd Century
AD
4th-6th Century
AD
3rd Century AD
4th-6th Century
AD
4th Century AD?
5th-6th Century

CIJ I, 470
CIJ I, 472
CIJ I, 477
CIJ I, 478
CIJ I, 479
CIJ I, 480
CIJ I, 481
CIJ I, 483
CIJ I, 482
CIJ I, 476
CIJ I, 3
CIJ I, 4
CIJ I, 5
IJO 3, App5
IJO
IJO
IJO
IJO
IJO
IJO
IJO
IJO

3,
3,
3,
3,
3,
3,
3,
3,

App2
App4
App6
App1
App3
App7
App8
App9

IJO 1, Mac7
IJO
IJO
IJO
IJO

1,
1,
1,
1,

Mac8
Mac11
Mac9
Mac6

IJO 1, Mac12
CIJ I, 694
IJO 1, Mac3
IJO 1, Mac4
IJO 1, Mac5
IJO 1, Mac2
IJO 1, Mac15
IJO 1, Mac17
IJO 1, Mac14
CIJ I, 693

AD
4th-5th Century
AD
4th-5th Century
AD
4th-5th Century
AD
Republican
period?

Majorca

Santa Maria del


Cami

epitaph

Malta

Rabato

epitaph

Malta

Rabato

epitaph

Malta

Rabato

catacombs

Oescus

epitaph

4th Century AD?

CIJ I, 681

Aquincum
Intercisa (DunaPentele)

4th Century AD

CIJ I, 675

233-5 AD

CIJ I, 677

198-210 AD

IJO 1, Pan5

2nd-3rd Century
AD
3rd Century AD
3rd Century AD
2nd-3rd Century
AD
5th-6th Century
AD
4th-5th Century
AD

CIJ