Sei sulla pagina 1di 401

i

Commemorating the Dead


ii
iii
Commemorating
the Dead
Texts and Artifacts in Context
Studies of Roman, Jewish, and
Christian Burials
Edited by
Laurie Brink, O.P. and Deborah Green
with an Introduction by
Richard Saller
Walter de Gruyter Berlin New York
iv
Printed on acid-free paper which falls within the guidelines,
of the ANSI to ensure permanence and durability.
ISBN 978-3-11-020054-6
Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie;
detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.
Copyright 2008 by Walter de Gruyter GmbH & Co. KG, 10785 Berlin
All rights reserved, including those of translation into foreign languages. No part of this book may be
reproduced or transmitted in any form or by any means, electronic or mechanical, including photo-
copy, recording or any information storage and retrieval system, without permission in writing from
the publisher.
Printed in Germany
Cover design: Martin Zech, Bremen
Cover photos: Laurie Brink, O.P., and Margaret M. Mitchell
Typesetting: Drlemann Satz GmbH & Co. KG, Lemfrde
Printing and binding: Hubert & Co. GmbH & Co. KG, Gttingen

v
To Robert M. Grant, Professor Emeritus of New Testament and Early
Christian Literature, the University of Chicago, whose lifelong
commitment to investigating the origins of early Christianity within
the cultural, historical, and social context of ancient Rome inspired
this project.
In memorial to Estelle Shohet Brettman, l z, the founder of the
International Catacomb Society, who desired to preserve and
document the Roman catacombs in order to illustrate the common
influences on Jewish and Christian iconography. May this publication
forward her legacy.
vi
Preface vii
Laurie Brink, O.P., and Deborah Green
Preface
Our fascination with the artifacts of the ancient world began when we
worked on the archaeological dig at Caesarea Maritima, Israel. As two
text-based doctoral students of Judaism and early Christianity, our ex-
perience of uncovering and analyzing material remains provided a view
into the ancient world that had produced our texts, a view that reading
alone could not fully provide. The richness of the conversations among
Israeli and American archaeologists, ancient historians, and textual
scholars further fueled our recognition that fostering dialogues across
the divide of academic disciplines would benefit not only our own aca-
demic work but the work of all who participated, helping to lower the
wall of academic territorialism. More than thirty years ago, Church
historian, Robert M. Grant, acknowledged the chasm between the re-
search efforts of classicists and those of scholars of Christianity:
The early history of Christianity is Roman history, and I should claim that
Roman history itself needs the collaboration of those who try to relate the
Christian movement to the whole life of the Empire, not explaining everything
Christian in Roman terms or everything Roman in Christian terms but trying to
understand identities, similarities, and differences.
1
Our own attendance at conferences and participation in interdisci-
plinary courses further confirmed our recognition that research about
the ancient world would benefit from cross-disciplinary work in which
scholars of various academic fields investigated, analyzed, and inter-
preted texts and artifacts in context. Similar work by Eric Meyers and
James Strange in the areas of Judaism and Christianity in the Galilee
laid the foundation for a more expansive project that would include
comparisons of the two religions with Roman religions and cultural
1
R. M. Grant, Introduction: Christian and Roman History, in The Catacombs
and the Colosseum: The Roman Empire as the Setting of Primitive Christianity
(ed. Stephen Benko and John J. ORourke; Valley Forge: Judson Press, 1971), 24.
viii Preface
dispositions.
2
However, we also recognized the need to confine the
topics of study in order to accomplish meaningful comparisons.
One arena in which historians, archaeologists, and scholars of Ju-
daism and early Christianity may share equal footing is investigations
of Imperial period burials. The distinctions and similarities among
Roman, Jewish, and Christian burials can provide evidence of social
networks, family life, and, perhaps, religious sensibilities. Is the Roman
development from columbaria to catacombs the result of evolving
religious identities or simply a matter of a change in burial fashions?
Did Christians practice inhumation in imitation of the Jews or was
this an expression of an early Christian theology of resurrection?
What Greco-Roman funerary images were taken over and baptized
as Christian ones? Do the material remains from Jewish burials evi-
dence an adherence to ancient customs, new beliefs about an afterlife,
or adaptation of rituals from surrounding cultures? Investigating the
emergence of a Christian or Jewish material culture that may be dis-
tinct from general Roman practices requires that the material culture
be viewed, whenever possible, in situ, through multiple disciplinary
lenses and in light of ancient texts. Scholars of Roman history and clas-
sics (John Bodel, Richard Saller, Andrew Wallace-Hadrill), archaeo-
logy (Susan Stevens, Amy Hirschfeld), history of Judaism (Deborah
Green), Christian history (Robin M. Jensen) and the New Testament
(David Balch, Laurie Brink, O.P., Margaret Mitchell, Carolyn Osiek,
R.S.C.J.) were invited to participate in what we affectionately called
The Grateful Dead Project. The project comprised three stages:
x
Two weeks of field research in Rome and Tunisia (June 2004).
x
The Shohet Conference on Roman, Jewish, and Christian Burials,
held at The Divinity School of the University of Chicago (May
2005).
x
This publication of articles resulting from the project.
Commemorating the Dead: Texts and Artifacts in Context is an attempt
to build bridges across the divide of disciplines simply by creating a con-
2
Eric M. Meyers and James F. Strange, Archaeology, the Rabbis, and Early Chris-
tianity (Nashville: Abingdon, 1981). Also see the interdisciplinary work of Caro-
lyn Osiek and David Balch, Families in the New Testament World: Households
and House Churches (Louisville, Ky.: Westminster John Knox, 1997) and Daniel
Schowalter and Steven Friesen, Urban Religion in Roman Corinth: Interdisciplin-
ary Approaches (Cambridge, Mass.: Harvard University Press, 2005).
Preface ix
versation wherein we learn from each other. Though certainly not the
first of such attempts, the project did incorporate several unique fea-
tures. First, the discussion extended beyond the institutional boundaries
of the University of Chicago where we had first met as doctoral students.
It included faculty and doctoral students from the British School at
Rome, Brite Divinity School, Brown University, Catholic Theological
Union, Northwestern University, the University of Maryland, the Uni-
versity of Michigan, the University of Chicago, the University of Or-
egon, Randolph College, and Vanderbilt University.
Second, in addition to creating a community of colleagues, we
wanted to pave the way for future collaborative efforts among the next
generation of scholars. To that end, we asked doctoral students in
complementary fields to respond to the faculty papers at the Shohet
Conference. We are grateful to the respondents: Terri Bednarz,
R.S.M., Brandon Cline, Fanny Dolanksy, Joel Dries, Joan Downs,
Patricia Duncan, Annal Frenz, Annette Huizenga, Lee M. Jefferson,
Meira Kensky, Young-Ho Park, Brad Peper, Matthew Perry, Trevor
Thompson, Janet Spittler, Jennifer Stabler, Karen Stern, James
Weaver, and Ann Marie Yasin. Their comments and insights helped
shape the final version of the articles that appear in this volume.
This project also extended beyond the library, the office, and the
conference hall. We secured and sustained a three-year commitment
from respected scholars because we promised them an opportunity to
do fieldwork together. As a team, we were able to stand in front of
the impressive funerary monument of Flavius Secundus in Kasserine,
Tunisia, with its one-hundred-and-ten-line inscription and read it in
situ. Together, we explored the dark, damp tunnels of the Roman cata-
combs, noting the similarities between Roman and Jewish iconography
and pondering the emergence of Christian symbols. We visited tiny
museums in Lamta, Salakta, and El Jem, Tunisia and castle-size ones
in Rome and Baia. We waited patiently for custodians to secure access
to treasures seldom viewed by the public at large. And, when evening
came, we shared food and drink, in lively discussion, debate, and happy
conversation about the experiences of the day. We did all of this with
scholars who share an interest in the ancient world but who view that
world through different disciplinary and methodological lenses. It has
been our immense pleasure to be a part of this adventure, and we hope
it has been fun and valuable for them as well.
This project received the first grant from the Shohet Scholars Program
to be awarded by the International Catacomb Society (ICS) of Boston.
x Preface
The generous financial support of ICS combined with the encourage-
ment of its executive director, Amy Hirschfeld, are chiefly responsible for
the projects success. In addition, we want to thank Amy for the countless
digital images she took during the field research. These are available for
viewing on the ICS website (www.catacombsociety.org/archive.html).
No project of this size and duration is the inspiration of one person.
Laurie traveled to Rome and Pompeii with David Balch, J. Patout
Burns, Robin Jensen, and Carolyn Osiek in August 2000, and realized
the value of viewing the material remains in situ with a team of
scholars. Similarly, the interdisciplinary conference on the family in
early Christianity facilitated by Carolyn Osiek and David Balch at
Brite Divinity School in 2000 presented a model for our Shohet Con-
ference. The introduction Robin Jensen provided to the International
Catacomb Society initiated a relationship that has provided financial
support and encouragement. The scholars who participated each con-
tributed their expertise at various points in the project. Andrew Wal-
lace-Hadrill and the British School at Rome provided a hospitable
home during our time in Rome. Andrew, along with John Bodel, Robin
Jensen, and Susan Stevens served as guides at various sites throughout
the two-week research trip. Unofficial fellow travelers Patout Burns
and Tanya Luhrmann offered insights from their respective academic
fields that enriched the conversations throughout.
We would also like to acknowledge the support of Richard Rosen-
garten, Dean of the Divinity School at the University of Chicago, who
offered invaluable advice in the planning stages of the project as did
the former Dean of Students Winnifred Fallers Sullivan and Admin-
istrator Sandra Peppers. The Divinity School and the Marty Center
for the Advanced Study of Religion assisted in the promotion and
hosting of the Shohet Conference. The University of Oregon supplied
extra funds for Deborah to travel to Israel to view burial sites and meet
with Israeli scholars.
Patrick Alexander, formerly with Walter de Gruyter, has been a sup-
porter of the project from its initial proposal in 2003. The editors and
contributors are thankful for his editorial insights. Dr. Sabine Vogt,
Editor for the Classical and Ancient Near Eastern Studies of Walter
de Gruyter guided our project to publication, for which we are, indeed,
grateful.
Personally, we would like to thank our families, friends, and col-
leagues for their support throughout the preparation of this publi-
cation. Special thanks to Deborahs husband, Reuben Zahler, and
Preface xi
their son, Joshua Zahler, for their love and patience, and to Lauries re-
ligious community, the Dominican Sisters of Sinsinawa, WI., and, in
particular, Betsy Pawlicki, O.P., for their continual encouragement and
understanding, especially at the initial stages of the project.
Commemorating the Dead has been based on the belief that conver-
sation among scholars who share similar interests in the ancient world,
but who differ in their disciplinary approaches, has the potential to
bear much scholarly fruit. May this book be the first of many harvests.
xii Preface
Table of Contents xiii
Table of Contents
Laurie Brink, O.P. (Catholic Theological Union)
and Deborah Green (University of Oregon)
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Richard Saller (Stanford University)
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Archaeology and Artifacts
Amy K. Hirschfeld (International Catacomb Society)
Chapter 1. An Overview of the Intellectual History of
Catacomb Archaeology . . . . . . . . . . . . . . . . . . . . . . 11
Andrew Wallace-Hadrill (The British School at Rome)
Chapter 2. Housing the Dead: The Tomb as House in
Roman Italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Susan T. Stevens (Randolph College)
Chapter 3. Commemorating the Dead in the Communal
Cemeteries of Carthage . . . . . . . . . . . . . . . . . . . . . . 79
Ritual and Religious Rites
Robin M. Jensen (Vanderbilt University Divinity School)
Chapter 4. Dining with the Dead: From the Mensa to the
Altar in Christian Late Antiquity . . . . . . . . . . . . . . . . . 107
Deborah Green (University of Oregon)
Chapter 5. Sweet Spices in the Tomb: An Initial Study
on the Use of Perfume in Jewish Burials . . . . . . . . . . . . . 145
xiv Table of Contents
Patronal Relations and Changes in Burial Practices
John Bodel (Brown University)
Chapter 6. From Columbaria to Catacombs: Collective
Burial in Pagan and Christian Rome . . . . . . . . . . . . . . . 177
Carolyn Osiek (Brite Divinity School)
Chapter 7. Roman and Christian Burial Practices
and the Patronage of Women . . . . . . . . . . . . . . . . . . . 243
Envisioning Context and Meaning
David Balch (Pacific Lutheran Theological Seminary)
Chapter 8. From Endymion in Roman Domus to Jonah
in Christian Catacombs: From Houses of the Living to Houses
for the Dead. Iconography and Religion in Transition . . . . . 273
Margaret M. Mitchell (University of Chicago)
Chapter 9. Looking for Abericus: Reimagining Contexts
of Interpretation of the Earliest Christian Inscription . . . . 303
Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
Index of Terms . . . . . . . . . . . . . . . . . . . . . . . . . . 379
Introduction 1
Richard Saller
Introduction
The death of a family member, friend, or dependent was a far more
common experience in the lives of Romans than for us today.
1
Conse-
quently, it is to be expected that commemoration of the dead would
figure more prominently in our textual and material records from the
Roman world than in comparable records today. And yet, burials and
the associated artifacts from the Roman world are disproportionately
overrepresented in the material record and therefore command the at-
tention of Roman historians and archaeologists, ever hungry for more
evidence. John Bodel estimates that we know of only 1% or so of the
burials from the city of Rome from 25 B.C.E. to 325 C.E. (Chapter 6).
That is a soberingly small fraction, and yet it is a much more sub-
stantial material attestation of a facet of ordinary life in Rome than
survives for any other aspect of sociocultural life. Hence, the dispro-
portionate attention of archaeologists and historians to funerary arti-
facts is understandable and warranted.
2
This volume brings together a set of excellent chapters on disparate
topics related to burial. The unifying rationale for the collection and
the preceding conferences is that our understanding of burial practices,
and the societies that gave rise to them, will be deepened and enriched
by bringing together the material record of archaeologists and the texts
of historians from both the classical Roman and the Christian fields of
study. This involves crossing the divides among four traditional aca-
demic specialties. The chronological span covers a millennium from
the middle Republic (thirdsecond centuries B.C.E.) to the seventh
1
The mortality rate in developed countries today is under 15 per 1,000, in contrast
to the Roman mortality rate of nearly 40 per 1,000 per year.
2
As an illustration of my point, all of the efforts by demographic and family his-
torians to derive useful generalizations about the Roman life course from epi-
taphs would be unnecessary if census records for the living survived from the
whole empire.
2 Richard Saller
century C.E., and the geographical span runs from Rome to North Af-
rica to Palestine. My introduction seeks to sketch briefly a broad frame-
work for the chapters and to draw attention to some thematic and
methodological threads running through the diverse contributions.
Narratives and Themes
More than one of the following chapters has a title with the phrase
from to , implying a developmental narrative. The reader should
understand that these are not pieces of a single narrative, but multiple
narratives for different locations and classes. For the cities of Rome
and Pompeii, Andrew Wallace-Hadrill (Chapter 2) examines house
tombs from the middle Republic, the late Republic-Early Empire, and
the High Empire. Like Roman houses with their dual function of in-
ward-looking familial space and outward-looking assertion of status
to the public, house-tombs had a dual function that shifted in empha-
sis through the three periods from inward-looking in the thirdsecond
centuries B.C.E. to outward-looking at the turn of the era and back to
inward-looking in the mid-second century C.E.
Other types of burials from the same region and centuries are ana-
lyzed in an entirely different narrative in Bodels account of the rise of
columbaria at the end of the Republic and the transition to catacombs
two centuries later. The reason for the entirely different narratives lies
in the fact that Bodels study is based on larger collective burial groups
of more modest economic strata on average than Wallace-Hadrills.
Columbaria were first built as a way to accommodate the ashes of slave
familiae of aristocratic Romans in structures separate from the aristo-
crats own monumental graves. The columbarium form then spread to
serve modest Romans more broadly. Bodel points out that columbaria
came to be associated with collegia, which provided the infrastructure
for the autonomous self-regulation needed to organize and maintain
a condominium of burials over time on a scale larger than the family.
In the late second and third centuries C.E. the spatial arrangement of
extended groups of burials shifted from the columbarium to the cata-
comb. The same period also saw the spread of Christianity and the shift
from cremation to inhumation, but Bodel stresses that these were three
separate developments and not causally related. Indeed, as Deborah
Green notes in Chapter 5, Jewish burials in Rome went through a com-
parable shift from family (or familial burials) to catacombs, and did
Introduction 3
so earlier than in the Christian community. The fact that catacombs are
strongly associated in the modern mind with the early Christians to the
neglect of Jewish catacombs is a result, as Amy Hirschfeld shows in
Chapter One, of the charged implications that Christian catacombs
have had in modern religious debates.
Susan Stevenss analysis of five late antique cemeteries around Car-
thage (Chapter 3) takes us to the period after the establishment of
Christianity and beyond the collapse of Roman rule. None of these
burial grounds extended the use of a classical Roman cemetery. Stevens
contrasts the spatial arrangements of the two wealthier cemeteries with
the other three: whereas the latter graves were largely anonymous and
spatially difficult to visit, the former were laid out in a hierarchy of
privilege and in a fashion to accommodate visits from the living (for
purposes described in Robin Jensens contribution in Chapter 4).
Alongside the spatial development of Roman burials went a socio-
political development described in the chapters of Jensen and Carolyn
Osiek (Chapter 7). In the classical era, burial and the rites associated
with it were largely a private matter, based in part on Roman law,
especially the law of property. Tombs and columbaria were private
property allocated to members of the familia and other dependents by
the property owner. As Osiek points out, a proper burial space was one
of the favors (beneficia) in the arsenal of a Roman patron, and that
patron could be a female property owner, just as well as a male. The
provision of burial spaces for early church members by propertied
women, Osiek suggests, explains why some of the major catacombs in
Rome bear the names of women (Domitilla, Priscilla, and Commo-
dilla). This patronage gave women status and influence in the early
church that gradually came to be monopolized by ecclesiastical auth-
orities in the third and fourth centuries.
The theme of assertion of Episcopal authority also motivates
Jensens account of the transformation of funerary banquets in the
third and fourth centuries in North Africa. With textual and archaeo-
logical evidence, Jensen documents the practice of meals (mensae) in
the graveyards to maintain the links between the living and their de-
ceased loved ones. This pre-Christian practice was so entrenched that
it continued among Christians through the fourth century, despite the
efforts of church authorities to suppress it and the accompanying
drunken excesses. Bishops sought to change the rites to bring them
under the control of the church in the form of celebration of the Eu-
charist.
4 Richard Saller
The funerary mensa is one of several examples in this volume of the
early Christian community drawing on the classical cultural repertoire
and transforming it. As another illustration, David Balch in Chapter 8
traces the transformation of the artistic representation of Endymion
into Jonah, who became the most common figure by far in early Chris-
tian art found in funerary and other contexts. Endymion, represented
as a handsome, nude youth, might seem to be a less than obvious
choice, but Balch makes the case for the aesthetic and spiritual reasons
to settle on his figure. And the Christians were not the only ones to
adopt Endymion, Balch notes, because he was a figure who would ab-
sorb many projected meanings.
Margaret Mitchells essay (Chapter 9) on Aberciuss epitaph ex-
plores with great subtlety the possibilities of the multiplicity of mean-
ings in an era of cultural change around 200 C.E. Aberciuss monument,
now in the Vatican Museum, is one of those very rare extant inscrip-
tions to be accompanied by a later literary text explaining its origin.
One would have thought that the text, the Vita Abercii, would clarify
the meaning of the epitaph, in which Abercius speaks to the passerby/
reader in the first person, but Mitchell shows that the Vita attempts to
delimit the meaning of the inscribed lines and thus to obscure the sig-
nificant intentional ambiguities that allowed the earliest Christian
readers to understand the meaning quite differently from non-Chris-
tian passersby. Mitchells argument raises major methodological is-
sues in the interpretation of other monuments, to which I will return.
Methodological Issues
The interactions among the participants in the volume during the tour
of the sites and at the conference heightened everyones awareness of the
methodological complexities of bringing textual and material evidence
together in a historical argument. The extensive survival of funerary
texts and artifacts makes it especially tempting to use evidence from
burials to deduce something important about the society, culture, and
economy of living Romans. But one might argue that the texts and arti-
facts from burial sites are evidence only for burial self-representation,
and that it is unwarranted to draw inferences about living society.
3
Such
3
Suzanne Dixon, Reading Roman Women: Sources, Genres, and Real Life (Lon-
don: Duckworth, 2001) has argued this position forcefully.
Introduction 5
skepticism strikes me as a salutary contribution to the scholarly de-
bate, but can be overstated. Every sort of evidence from every context
requires sophisticated methodological care to avoid nave inferences.
The essays in this volume display the requisite sophistication as they
explore various aspects of the symbols and rituals that connected the
living to the dead.
I would suggest several broad methodological points about that
connection. First, both the funerary texts and the visual imagery of
burial offer idealized representations of roles and relationships in liv-
ing society (Bodel). Whether it be the epithet attached to the deceased
(e.g., most chaste wife) or the architectural motifs on the outside
and inside of the tomb (as described in Wallace-Hadrills chapter), the
idealizations had their roots in the world of the living and can provide
evidence for characteristic features and values of that world.
In addition, there is the obvious point that the dead could not bury
themselves: burials in all their ritual and material aspects were the
creations of living social entities whether they be familial (Wallace-
Hadrill) or patronal relationships (Osiek) or collegia (Bodel) or Chris-
tian communities (Osiek and Jensen). Furthermore, several of the
following chapters show vividly how the cultural practices and motifs
associated with death drew on those of the living. Most obviously,
Endymion was taken from the dining rooms of the living and trans-
formed into Jonah for the burial sites of the Christians (Balch). The
sanctity of burials was protected by the living through a legal frame-
work of private property enforced by the living state apparatus (Osiek).
Indeed, burial is one of the great collective action problems for so-
cieties, because the deceased cannot themselves perform the rites or
protect the inviolability of their tombs and monuments in perpetuity,
nor can their individual families. When state authority broke down,
the tombs were vulnerable to being plundered (Stevens, Hirschfeld). In
sum, the living were linked to the dead by systems of meaning, of rit-
uals and social practices, and of law.
A second methodological point indeed, the basis for the whole
enterprise embodied in the volume is the importance of context, em-
phasized explicitly by Mitchell. Her essay on the epitaph of Abercius
shows that the more we know about a text or artifact, the more intri-
cate the issues of context become. Even the particular form of display
of this famous inscription in the Vatican Museum turns out to depend
on modern religious ideology and has demonstrably affected scholarly
interpretations of the text. The implication of Mitchells fundamental
6 Richard Saller
point might be thought a reason for despair, since historians rarely
know the full story of how a particular text or artifact came to exist
and then to survive to the present. The despair seems pointless, since
historians and archaeologists always work on incomplete information
about context. Rather, Mitchell shows the constructive path forward
by reminding the reader that we need to reckon with the limits of our
knowledge.
A third, related methodological principle is the importance of cul-
tural context for understanding burial practices. In the absence of cul-
tural knowledge, there is a temptation to offer utilitarian explanations
for artifacts and texts based on our own modern experience. In her
contribution on perfume vials found in Jewish burials, Green argues
convincingly against using a facile utilitarian approach. Despite the
superficial plausibility of the idea that the perfume was intended to
cover up the stench of the corpse, she shows that this explanation can-
not account for all the material evidence. It is more likely that the per-
fume should be understood in the same way as the other personal ar-
tifacts found of sentimental value with the deceased.
My fourth methodological point is that the context for any given
artifact or text was not static. Mitchell notes that the readers response
to the epitaph of Abercius may have changed over time as the surround-
ing graveyard changed to become a more definitively Christian context.
Not only did the necropolis develop over time, but so also did many of
the tombs. We need to imagine the house tombs, the columbaria, and
the catacombs as works in progress, rather than the finished product
left to posterity. And if the finished house-tomb or columbarium or
catacomb was the result of a series of decisions over time, we often do
not know which member of the family or community made each deci-
sion. Osiek shows how it is possible to draw on what is generically
known about the social relations of patronage to sketch the develop-
ment from private graveyards above ground to the underground cata-
combs controlled by the church. Often we cannot be more precise than
these generic explanations.
A fifth and related point is the risk of simplifying context by sharply
separating ethnic and religious groups and reifying them with the his-
torians hindsight. The ancient realities were more fluid. Green makes
the point that it is a mistake to think in terms of a unified category of
Jewish, because it fails to take into account changes over time. As a
result, she is very careful to specify the period and place she is stu-
dying. Bodel argues that the Christian catacombs are unlikely to
Introduction 7
have been filled only by Christians, because that identification would
imply that, of the 150,000 or so Roman burials about which something
is known, an improbably high proportion would be Christian.
A final methodological point is that the relevant context includes
the history of scholarship on ancient burial leading to our current cat-
egories and understandings. Hirschfelds essay on the history of cata-
comb archaeology shows that not only does the past influence the
present but also the present influences the treatment of the past. Some
of the most famous figures associated with the excavation of the cata-
combs were driven more by the religious disputes of their era than by
their time spent in excavation of the catacombs. Bodel demonstrates
that some of the categories at the base of modern scholarship, such as
columbarium and catacomb, are not ancient categories but mod-
ern constructs. And Mitchell describes how two interpretive circles, the
Vita Abercii and the modern Vatican Museum, have framed the inter-
pretation of Aberciuss funerary inscription.
In sum, the methodological complexities require the crossing of dis-
ciplinary boundaries and chronological specialties to a degree beyond
the capacity of most individual scholars. As a result, the cross-disciplin-
ary discussions that inform the contributions to this volume are vital.
The other participants and I are most grateful to the International
Catacomb Society and Laurie Brink for making them possible.
8 Richard Saller
Archaeology and Artifacts
History of Catacomb Archaeology 11
Amy K. Hirschfeld
Chapter 1
An Overview of the Intellectual History
of Catacomb Archaeology
The past is essential and inescapable. Without it we
would lack any identity, nothing would be familiar, and
the present would make no sense. Yet the past is also a
weighty burden that cripples innovation and forecloses
the future. How do we recognize and cope with this
heritage, which at the same time sustains and constrains
us? What benefits does it provide, what costs does it
exact?
1
David Lowenthal
Introduction
Archaeology, in the broadest sense of the study of the past through
the material culture produced by peoples of the past, involves many
disciplines (such as natural sciences, anthropology, history, religious
studies, and art history) and many disciplines rely, in part, on archaeo-
logical evidence. Context is essential in archaeological research not
only the context of the past and its material remains as they are exca-
vated but also the context of each present in which research has been
1
David Lowenthal, The Past is a Foreign Country (Cambridge: Cambridge Uni-
versity Press, 1985). I am grateful to Laurie Brink for inviting me to participate
in this project and to the International Catacomb Society for supporting my
participation. I am also grateful to Carolyn Osiek for helpful comments on an
earlier version of this paper. Sections of this paper are based on Exploration of
the Catacombs in Vaults of Memory: The Roman Jewish Catacombs and Their
Context in the Ancient Mediterranean World, Estelle S. Brettman, Amy K. Hirsch-
feld, and Florence Z. Wolsky. Forthcoming.
12 Amy K. Hirschfeld
conducted. For sites such as the Roman catacombs
2
that have, for the
most part, been continuously known instead of newly excavated or dis-
covered, new information and insight is often gained from new ap-
proaches and from excavating the archives of previous study.
Traditional questions such as How does the past influence the pres-
ent? and How has the present grown out of the past? are increas-
ingly being reversed as scholars in many disciplines have become more
self-reflexive. The complimentary questions being asked are How
does the present influence the past? and How is the interpretation of
the past affected, restricted, or prescribed by present social, political,
and religious conditions? Such questions can be productively exam-
ined from the perspective of the various presents in which the Roman
catacombs have been studied.
The study of the Roman catacombs has been an important part of a
long-standing tradition of religious inquiry. The catacombs are some-
what unusual as a subject of archaeological study in their almost inex-
tricable relationship to a living religion that has primarily been in con-
trol of their study and guardianship.
The manner in which the catacombs have previously been studied and
presented to academic audiences (in scholarly publications) and gen-
eral audiences (in popular writing, tourist sites, and museum displays)
forms an essential component of the evidence many later scholars have
relied upon in their interpretations of the catacombs. The authors of
much of the past academic and popular writing about the catacombs
viewed them as sites of connection to a venerated religious past that
could be used to legitimize the religious present. Many ancient Chris-
tians wanted to be buried in the catacombs near the graves of saints
and martyrs to be close to the sacred, and this same desire likely moti-
vated many later researchers to study these sites.
Throughout the history of their study, the catacombs have often not
been studied objectively in their own right but instead been used as a
source of evidence to support established ideas (especially of religious
history). Many of the main figures in the history of catacomb explora-
2
Throughout this paper when I use the term catacombs, I am referring to the
Christian catacombs of Rome. I also mention the Roman Jewish catacombs but
will identify them explicitly as Jewish. I do not address the question of the accu-
racy of the identification of entire catacombs as Christian or Jewish, but fol-
low the traditional identifications based on the presence of iconographic and
epigraphic indicators.
History of Catacomb Archaeology 13
tion did not pursue their research for the sake of objective science but
rather for the sake of material documentation of already-known relig-
ious facts. In the following paper, I give an overview of some of the
main figures and historical trends in the study of the catacombs to pro-
vide some context for research involving the catacombs.
The Catacombs from the Sixth Century
through the Middle Ages
The catacombs ceased to be actively used for burials in the fourth
through sixth centuries C.E., having been gradually replaced by above-
ground cemeteries. Even after active burial in the catacombs ceased,
the catacombs continued to be regularly visited. Many of the cata-
combs were converted into martyrs shrines in the fourth and fifth
centuries, and aboveground sanctuaries and cemeterial basilicas, some-
times linked to the catacombs by tunnels, were built for the veneration
of the martyrs. Notably, Pope Damasus (366384) enlarged and decor-
ated sections of the catacombs that contained martyrs tombs, for
example, the Crypt of the Popes in the catacomb of S. Callisto.
The maintenance and renovation of these sacred areas by papal
authorities continued until at least the eighth century, and they at-
tracted both Roman Christians and pilgrims who visited the cata-
combs for prayer and devotion. After the last repairs to the catacombs
and martyrial shrines made by Popes Hadrian and Leo III at the end
of the eighth century and beginning of the ninth century, there was
an extensive removal of relics of the saints from their shrines and
burial places in the catacombs, which were all outside the city walls, to
churches within the city walls. This large-scale removal may have been
prompted by the presence of foreign invaders or could have resulted
from gradual changes in ecclesiastical policy to allow the transpor-
tation of relics.
3
Eventually, the catacombs were abandoned, with the
exception of a few galleries in certain catacombs that continued to be
visited, primarily those located below martyrial churches.
4
Itinerari,
3
See Irina Tassa Oryshkevich, The History of the Roman Catacombs from the
Age of Constantine to the Renaissance (Ph.D. dissertation, Columbia Univer-
sity, 2003).
4
Especially the catacomb of S. Sebastiano, which became a center for the venera-
tion of Peter and Paul by the end of the third century, as evidenced by pilgrims
14 Amy K. Hirschfeld
written as guides for pilgrims, especially during the seventh and eighth
centuries, and the Liber Pontificalis,
5
begun in the late fifthearly sixth
century, were consulted by later explorers attempting to discover and
identify the catacombs.
The common conception that the catacombs were entirely forgotten
after the translation of relics to churches in the eight and ninth cen-
turies until their supposed rediscovery in the sixteenth century is
somewhat misleading. In addition to the few catacombs located below
martyrial churches that were continuously visited, the Mirabilia urbis
Romae, a popular guidebook for Rome begun in the twelfth century
and enjoying popularity for several centuries, included several cata-
combs in most of its editions,
6
and numerous visitors in the fourteenth
and fifteenth centuries, including Renaissance humanists such as Pet-
rarch, wrote about their visits to the catacombs.
7
However, there was
not much interest in the careful exploration, study, and recording of
the catacombs until the late fifteenth and early sixteenth centuries. By
that time, the locations and even the names of some of the catacombs
likely had, in fact, been forgotten.
Latin and Greek graffiti invoking the two apostles, which can still be seen today.
Other catacombs that continued to be visited because of their connection to active
churches were S. Lorenzo, S. Pancrazio, S. Agnese, and S. Valentino. Vincenzo
Fiocchi Nicolai, Fabrizio Bisconti, and Danilo Mazzoleni, The Christian Cata-
combs of Rome: History, Decoration, Inscriptions (Regensburg: Schnell & Steiner,
1999), 9.
5
The Book of the Popes is a biographical history of the popes from St. Peter
through the fifteenth century and includes information on their burial places. As
a work written and compiled by many different individuals over a long period,
its historical reliability has always been somewhat questionable. The reliability
of the itinerari is also uncertain. Nonetheless, they were frequently used as
sources of topographical information by later scholars.
6
For example, the catacombs of Commodilla, S. Callisto, Praetestato, Priscilla,
and Domitilla, among others. Francis Morgan Nichols, ed. and trans., The Mar-
vels of Rome: Mirabilia Urbis Romae (2d ed.; New York: Italica Press, 1986).
7
Oryshkevich, History of the Roman Catacombs, 6773, 1027. See this entire
dissertation for evidence that the catacombs were never forgotten. For a dis-
cussion of the later history of the Christian catacombs, see J. Osborne, The
Roman Catacombs in the Middle Ages, Papers of the British School at Rome 53
(1985): 278328.
History of Catacomb Archaeology 15
The Catacombs in the Fifteenth and Sixteenth Centuries
The fifteenth and sixteenth centuries are characterized by systematic
exploration and study of the catacombs, the use of the evidence from
the catacombs for contemporary religious purposes, and the creation
of popular literature focused on the catacombs.
In the fifteenth and sixteenth centuries, the renewed interest in and
knowledge of the catacombs made them an important destination for
many travelers to Rome who followed guidebooks such as the Mirabilia
urbis Romae and who wrote about their impressions of the catacombs,
although some of their descriptions seem to be copied from the guide-
books. The belief in purification and remission of sins in the presence
of martyrs relics accounts for much of the appeal of the catacombs for
visitors. Stories of getting lost in the catacombs and having a spiritual
reawakening become a common literary theme in the following cen-
turies.
8
Some of the most notorious visitors to the catacombs in the late
fifteenth century did not leave a written record of their visits to the cata-
combs, except in the graffiti they marked on the walls of the catacombs.
The Accademia Romana degli antiquari, led by the flamboyant Pompo-
nio Leto (14281498), wished to broaden their knowledge of classical
antiquity. This unconventional group referred to themselves as unanimes
perscrutatores antiquitatis (investigators of antiquity). They visited the
catacombs of S. Callisto, Ss. Pietro and Marcellino, Praetestato, and
Priscilla.
9
Pope Paul II considered their actions heretical, and members
of the Accademia were prosecuted as pagans conspiring against the
pope. Leto and other members of the Accademia were imprisoned in
the Castel SantAngelo for almost a year, but evidence against them
could not be procured. Interestingly, the graffiti they left in the cata-
comb of S. Callisto, which would have provided the evidence the auth-
orities needed but was not inscribed until after their imprisonment,
8
Robert W. Gaston, British Travellers and Scholars in the Roman Catacombs,
Journal of the Warburg and Courtauld Institutes 46 (1983): 147.
9
Pasquale Testini, Le catacombe e gli antichi cimiteri cristiani in Roma (Bologna:
Cappelli Editore, 1966), 1516; see also Ludwig Hertling and Engelbert Kirsch-
baum, The Roman Catacombs and Their Martyrs (Milwaukee: Bruce Publishing
Co., 1956). Hertling and Kirschbaum disparagingly refer to the Accademia
Romana as half-pagan humanists [with] no interest in Christian antiquities
(p. 3).
16 Amy K. Hirschfeld
was not known until the nineteenth century, when it was discovered by
Giovanni Battista de Rossi.
10
Religiously oriented investigation of the catacombs during this
period began with Onofrius Panvinius (15291568), an Augustinian
monk who conducted a well-organized, methodical study of Christian
cemeteries of the ancient world.
11
Panvinius is credited with beginning
a trend of scientific Christian archaeology and was renowned as a great
church historian and archaeologist. He focused on the cemeteries of
Rome and researched the then-available historical, ecclesiastical, and
epigraphic resources. Panvinius was the first person to classify ancient
Christian inscriptions, and his work indicates that he was aware of
regions of 43 catacombs. He is not known to have explored them, how-
ever. Although Panvinius is known as an archaeologist, his work was
almost entirely based on a survey of literary rather than material evi-
dence, an approach followed by many other later scholars who are
nevertheless also known for conducting scientific archaeology.
St. Philip Neri (15151595) founded the Cenacolo Filippino, or Con-
gregation of the Oratory, in his quest to promote the Counter-Reforma-
tion movement of the Roman Catholic Church, restore early Christian
religious practices, and trace a history of Church events. He was known
for spending long hours in the catacombs meditating and frequently
preached to his followers on visits to the catacomb of S. Sebastiano.
Neris devotional focus on the martyrs of the catacombs and their mess-
age of suffering and redemption set the stage for the intense interest in
the exploration of the catacombs by Antonio Bosio and others in the
sixteenth century and also set the devotional tone that would character-
ize much catacomb study. Neri appointed his closest follower, Cesare
Baronius, to continue his work.
Cardinal Cesare Baronius (15381607), who succeeded Philip Neri
as leader of the Oratory, is credited with inaugurating the scientific
study of Church history, much as Panvinius was credited with doing
for Christian archaeology. By the second half of the sixteenth century,
renewed interest in the archaeology of early Christianity was being
supported by efforts on the part of both Catholics and Protestants
10
The graffiti included the phrases regnante pomponio pontifice maximo (when Pom-
ponio reigned as Pontifex Maximus) and romanarum puparum delitiae (delights of
Roman girls). Oryshkevich, History of the Roman Catacombs, 21923.
11
De ritu sepeliendi mortuos apud veteres christianos, et eorundem coemeteriis liber
(Cologne, 1568).
History of Catacomb Archaeology 17
to trace early Church history. Protestant interests led to the production
of the Magdeburg Centuries (15591574), a history that criticized
the Roman Catholic Church. Baroniuss monumental work Annales
ecclesiastici (15981607) was written as a response to the Magdeburg
Centuries and made extensive use of previously ignored collections of
Roman manuscripts. This publication earned Baronius the title
Father of Ecclesiastical History, and he gained a reputation for thor-
ough, penetrating research. Baroniuss quest for accuracy, as well as
the regard for the catacombs instilled in him by Neri, led him to the
catacombs as a frequent subject of study. His research was almost ex-
clusively text-based and served the Roman Catholic Church and the
Counter-Reformation.
In 1578 an intact Christian catacomb containing frescoes, sarcop-
hagi, and inscriptions was discovered on the Via Salaria Nuova. Pre-
viously known catacombs had been stripped of relics and artifacts and
did not contain as extensive or detailed frescoes as the newly discovered
catacomb contained. The discovery captured the imagination of the
public and scholars alike who visited the catacomb in great numbers.
This catacomb, strikingly painted with Old and New Testament scenes,
seemed to many a city buried beneath Rome and gave rise to the term
Roma sotterranea.
12
The discovery is heralded by later scholars as the
event that marked the rediscovery of the catacombs and the birth
of Christian archaeology.
12
The catacomb on the Via Salaria Nuova was incorrectly identified as the cata-
comb of Priscilla by Baronius and as the Ostriano cemetery by Alfonso Chacon
and later by Bosio. Soon after its discovery, the catacomb was buried by a land-
slide, precipitated by the continued extraction of pozzolana (a volcanic rock).
Antonio Bosio, who was only three years old at the time of the initial discovery,
was later frustrated by his inability to visit the catacomb and claimed that the
diggers who were trapped in the collapse received just retribution. James Steven-
son, The Catacombs: Life and Death in Early Christianity (Nashville: T. Nelson,
1985), 51. The cemetery was rediscovered in 1921 by Enrico Josi and again mis-
takenly identified, this time as the catacomb of the Giordani. The catacomb was
recently identified as a private cemetery and is now called the Anonymous
Cemetery of the Via Anapo. Philippe Pergola, Le catacombe romane: storia e
topografia (Rome: Carocci editore, 1998), 12530; Fabrizio Mancinelli, Cata-
combs and Basilicas: The Early Christians in Rome (Florence: Scala, 1981),
4546. Various identifications of this catacomb have persisted, even in recent
works, e.g., W. H. C. Frend, The Archaeology of Early Christianity (Minneapolis,
Minn.: Fortress Press, 1996); Stevenson, The Catacombs; and Gaston, British
Travellers, 14465.
18 Amy K. Hirschfeld
This rediscovery of the catacombs was presented as almost provi-
dential for the Roman Church, which now had a well-known, popular
example in the catacomb frescoes of the early Christian use of images.
The Council of Trent in 1563, just fifteen years earlier, had confirmed
the value of the visual image, and the evidence from the catacombs
often took center stage in Reformation debates about sacred imagery.
The idea of a providential rediscovery and its impact was likely
popularized by later scholars, especially de Rossi, who identified
May 31, 1578, as the birth date of Christian archaeology. The first
mention of the archaeological evidence from the catacombs in treatises
dealing with images is in Baroniuss Annales ecclesiastici (15981607),
twenty years after the discovery of the catacomb on the Via Salaria
Nuova.
13
Among those who kept alive the newly awakened interest in cata-
combs were the Spanish Dominican Alfonso Chacon (15401599),
known as Ciacconio, and the Flemish laymen Philip van Winghe
(d. 1592)
14
and Jean LHeureux, known as Macarius. This noble
triumvirate
15
documented the catacomb on the Via Salaria Nuova
as well as the catacombs of Priscilla, Ss. Pietro and Marcellino, S. Val-
entino, and S. Callisto, which was recorded at this time as Coemeterium
Zephyrini. As scholars with antiquarian interests, their work focused
on early Christian burial practices in a comparative perspective.
Although Chacons annotated interpretations were not very accu-
rate and often amusing in their misinformation, much of his work was
incorporated in Bosios later work. His copyists frequently employed
the artistic vernacular of the day, making entertaining images that
were not particularly conducive to serious study.
16
Chacons colleague
Philip van Winghe, aware of the deficiencies in his friends work, repro-
duced and revised the material more accurately. Unfortunately, van
Winghe died prematurely in 1592 without having completed his work,
which was not preserved in its entirety.
13
Gaston, British Travellers, 145, n. 4. See also Oryshkevich, History of the
Roman Catacombs.
14
See Cornelis Schuddeboom, Philips van Winghe (15601592) en het ontstaan van
de Christelijke archeologie (Haren: Geldermalsen Publications, 1996).
15
Giovanni Battista de Rossi, La Roma sotterranea cristiani descritta ed illustrate
(Rome, 1864), vol. I, 14.
16
See Anthony Grafton, ed., Rome Reborn: The Vatican Library and Renaissance
Culture (New Haven: Yale University Press, 1993).
History of Catacomb Archaeology 19
LHeureux authored an important study of the monuments but it
was not published until 1856 by the eminent Christian archaeologist
and Jesuit scholar Father Raffaele Garrucci (18121885),
17
and thus
was not an important influence on contemporary study.
Antonio Bosio
Antonio Bosio (15751629) is regarded as the Columbus of Roma
Sotterranea the first person to extensively explore the catacombs
themselves as opposed to previous scholars who explored texts for
catacomb references. He was enthusiastically devoted to the study of
the catacombs and investigated them tirelessly, although he did not ac-
tually conduct excavations or explore obstructed passages.
Bosios extensive explorations enabled him to amass a wealth of new
information and earned him the tribute of numerous later scholars.
Despite Bosios extensive investigations, he was able to identify only
a very few Christian monuments by name and trace their history be-
cause of the lack of adequate topographical evidence.
Although Bosio is considered by many to be the first to approach
the study the catacombs scientifically, his main concern was for the
spiritual value of the monuments he was investigating and this concern
is apparent in his work. He considered the catacombs to be tangible
evidence of the early Church of the martyrs and of Romes status as
successor to the early Church.
Bosio, as well as many scholars who followed him and emulated
his example, considered textual sources almost exclusively in the inter-
pretation of archaeological material, and curious omissions in his
work indicate that he recorded only material that served his devotional
purpose and affirmed the descent of the current Roman Catholic
Church from the early Church of the martyrs. He did not record all of
the paintings or sculptural fragments he surely saw. For example, he
neglected to describe or copy in detail the famed frescoes and stuccoes
of the orans figures, the Virgin with child and prophet, and the Good
Shepherd in the catacomb of Priscilla, although he and his illustrator
17
Hagioglypta sive picturae et sculpturae sacrae antiquiores praesertim quae Romae
reperiuntur explicatae a Joanne LHeureux (Macario), edited by Raffaele Gar-
rucci (Lutetiae Parisorum: J. A. Toulouse, 1856).
20 Amy K. Hirschfeld
Toccafondo clearly saw them, as evidenced by their autographs
sprawled over the frescoes in the catacomb.
Bosio had not completed his monumental work Roma sotterranea at
the time of his death in 1629. In 1634, Roma sotterranea was published,
after having been edited, emended, and cut by Giovanni Severani.
18
Because Bosios original manuscripts were preserved, the nature of the
edits to the published version can be determined. A section that Bosio
intended to include but that Severani cut clearly focused on Bosios in-
tent to document the early Church as a means to justify contemporary
practice.
19
Severani also seems to have omitted or simplified sections
that Bosio, in his zeal of completeness, had included but would likely
have provided ammunition for Protestant polemicists.
20
The Italian
edition of Roma Sotterranea edited by Severani was not widely known.
A Latin edition prepared by Paolo Aringhi and published in 1651 was
widely disseminated throughout Europe. Aringhi made numerous
changes to the work infusing it with a distinctly polemicist Counter-
Reformation tone and anti-Jewish sentiment that were not consistent
with Bosios original.
Bosios work had a great influence on later scholars, and he was par-
ticularly admired by Giovanni Battista de Rossi, who is perhaps the
most well known of all scholars who have studied the catacombs.
Giovanni Battista de Rossi
The nineteenth century was witness to a veritable explosion of interest
in the catacombs, both for the purposes of scientific study and for
current religious debates being carried out in academic and popular
literature.
Giovanni Battista de Rossi (18221894) is renowned as the father
of modern scientific Christian archaeology. De Rossi was appointed
by his friend and teacher the Jesuit scholar Father Giuseppe Marchi
18
The frontispiece is imprinted with the date 1632, the date that printing was initi-
ated.
19
Simon Ditchfield, Text Before Trowel: Antonio Bosios Roma Sotterranea
Revisited, in The Church Retrospective (ed. R. N. Swanson; Studies in Church
History 33; Woodbridge, Suffolk, U.K., Ecclesiastical History Society, 1997),
353.
20
Ibid., 356.
History of Catacomb Archaeology 21
(17951860) to create a corpus of Christian monuments. This project
entailed the systematic cataloging of Christian epigraphy, a project
that de Rossi envisioned would, for the first time, partially reconcile it
with topographical criteria.
De Rossi contributed to more than two hundred publications, many
of which have had a lasting influence on catacomb scholarship and
are still consulted today. In collaboration with his brother Michele,
a mathematician and geologist, de Rossi documented a detailed topo-
graphical study of the Roman catacombs published in three volumes
entitled La Roma sotterranea cristiana (18641877), a name he likely
selected in tribute to Bosio. De Rossi inspired and provided the proto-
type for further topographical study with the publication of his Bullet-
tino di Archeologia Cristiana, which documented the fruits of his many
years of laborious research and was published in a simultaneous French
edition, which contributed to the internationalization of catacomb
studies at this time.
De Rossi figured largely in the establishment of the Pontifical Com-
mission of Sacred Archaeology on January 6, 1852, by Pope Pius IX,
who supported and favored de Rossi. The commission instituted
strong measures to control exploration of the Christian cemeteries and
to end violations of the catacombs, and sponsored continued study
and excavation by de Rossi and others.
De Rossis extraordinary popularity in Rome and among inter-
national scholars, his extensive publications, and his labors in the ser-
vice of Christian archaeology stimulated great interest in the subject.
He befriended, mentored, and inspired the work of many, including
J. Spencer Northcote and W. P. Brownlow, who first made de Rossis
Roma sotterranea available in English.
21
De Rossis protg Orazio Marucchi popularized archaeology in the
later nineteenth century with his prolific writing on the catacombs,
22
and another distinguished follower of de Rossi, Mariano Armellini
21
J. Spencer Northcote and W. P. Brownlow, Roma Sotterranea or Some Account of
the Roman Catacombs, Especially of the Cemetery of San Callisto, Complied from
the Works of Commendatore de Rossi with the Consent of the Author (London:
Longmans, Green, Reader, and Dyer, 1869).
22
Orazio Marucchi, Le catacombe romane secondo gli ultimi studi e le piu recenti
scoperte: compendio della Roma Sotterranea con molte piante parziali dei cimiteri
e riproduzioni di monumenti. (Rome: Desclee, Lefebvre, 1903); Le catacombe
romane. Opera postuma. Preface and biography by Enrico Josi (Rome, 1933).
22 Amy K. Hirschfeld
(18521896), made notable contributions.
23
In 1903 Josef Wilpert
(18571944), also a student and successor of de Rossi, published a
comprehensive documentation of the paintings and sarcophagi of the
catacombs. In spite of some limitations, his work long served as a basic
research tool.
24
The Nineteen-Century Early Christian Novel
In addition to being the focus of intense scholarly attention in Italy and
abroad during the nineteenth century, the catacombs were also the
focus of popular interest and made appearances in much popular re-
ligious literature of the period, especially in Victorian England. Relig-
ious authors used the wealth of archaeological information obtained
from the prolific discoveries and research of the nineteenth century for
polemical purposes. They viewed the Early Church of the catacombs
as a legitimizing predecessor to their own doctrines and practices.
25
The catacombs easily captured the attention of a Victorian public
fascinated with death and pleasurable terror, and several religious
novels based on the early Christians of the catacombs and the cata-
comb martyrs became extremely popular. The authors of these novels,
almost all clergymen themselves, were intentional participants in seri-
ous current theological debates, and at the same time, they appealed to
general readers with detailed, sensationalized descriptions of death
and torture. These novels address religious and social issues of great
importance at a time when there existed a major rift in the English
Catholic church between those who favored the Roman Catholic tradi-
tion and ritual and those who supported Liberal Catholicism. The
novels are also examples of the vehemence of religious polemic and the
23
Mariano Armellini, Gli antichi cimiteri cristiani di Roma e dItalia (Rome: Tipo-
grafia Poliglotta, 1893).
24
Josef Wilpert, Die Malereien der Katakomben Roms (Freiburg im Breisgau: Her-
dersche Verlagshandlung, 1903). See Reiner Srries, De Rossi, Wilpert und die
christliche Archologie um 1888 and Die Malereien der Katakomben Roms
(18861903) in Josef Wilpert: Ein Leben im Dienste der christlichen Archologie,
18571944 (Wrzburg: Bergstadtverlag Wilhelm Gottlieb Korn, 1998), 2734,
3555.
25
Wendel W. Meyer, The Phial of Blood Controversy and the Decline of the Lib-
eral Catholic Movement, Journal of Ecclesiastical History 46, no. 1 (1995):
7677.
History of Catacomb Archaeology 23
blatant rhetorical strategies by which paper opponents were demol-
ished
26
that characterize most of the popular religious press of this
period.
The three most important and popular novels in this category are
Hypatia, or New Foes With an Old Face (18521853), by Charles
Kingsley; Fabiola, or The Church of the Catacombs (1854), by Nicholas
Wiseman; and Callista: A Tale of the Third Century, by John Henry
Newman (1855). These three novels are similar in their use of a histori-
cal setting of martyrdom and the Early Church and their distinctly
polemical, propagandist tone.
27
Charles Kingsley
28
used a different approach than most of the auth-
ors of Early Christian novels. He used the example of history in his
novel not to recall a venerable past but rather as a negative example of
the corruption of the past to compel his readers to find their own jus-
tifications for their faith in the present instead of looking to the past.
His heroine, Hypatia, is murdered after her acceptance of Christ
in fifth-century Alexandria (when Christianity was already the state
religion). Kingsley used the death of Hypatia to question the value the
Roman Catholic Church placed on the patristic past, which Kingsley
believed to be corrupt.
29
Nicholas Wiseman claimed that the nineteenth-century Roman
Church was consistent with the Church of the catacombs.
30
Fabiola
was published in the series the Catholic Popular Library, which may
26
David J. DeLaura, Review of Gains and Losses: Novels of Faith and Doubt in Vic-
torian England, by Robert Lee Wolfe, Nineteenth-Century Fiction 33, no. 2
(1978): 25155.
27
Leon B. Litvack, Callista, Martyrdom, and the Early Christian Novel in the
Victorian Age, Nineteen-Century Contexts 17, no. 2 (1993): 164.
28
When Kingsley wrote Hypatia, he was the Rector of Eversley Church and was
committed to social reform and the Christian Socialist movement. He later
became Regius Professor of Modern History at Cambridge (18591869), was
appointed chaplain to Queen Victoria in 1859, became the private tutor to the
future Edward VII in 1861, and was appointed the canon of Westminster Abbey
in 1873.
29
Litvack, Callista, 16465. It has been suggested that the murder of the histori-
cal Hypatia of Alexandria, a mathametician and Platonic philosopher, may have
been a political assassination as much as it was the work of a Christian mob act-
ing against a well-known pagan. Maria Dzielska, Hypatia of Alexandria (Cam-
bridge: Harvard University Press, 1995). Kingsleys fictionalization of Hypatias
last-minute conversion has no known historical basis.
30
At this time, it was commonly believed that the catacombs were monuments of
the first century, an idea that has since been disproven.
24 Amy K. Hirschfeld
have been originated for the purpose of refuting Kingsleys version of
the early Church.
31
Wiseman wrote Fabiola to respond to Kingsleys
negative portrayal of the early church in Hypatia and also to promote
English Roman Catholicism. Wiseman had been made the Archbishop
of Westminster in 1850 when the Roman Catholic hierarchy was re-
stored in England. In the novel, Fabiola, a pagan in Rome, converts to
Christianity after witnessing martyrdoms, which Wiseman describes in
gruesome detail. Wisemans novel enjoyed immense popularity, prob-
ably due in large part to the graphic, sensationalized descriptions of
the martyrdoms, and went through numerous editions, translations
into other languages, and even an adaptation for the stage, The Youth-
ful Martyrs of Rome by Frederick Oakeley.
32
John Henry Newman believed that the sufferings of the martyrs
were emblematic of the trials that all Christians must undergo to gain
salvation, and some of the themes in Callista probably represented his
own painful struggle in converting from Anglicanism.
33
Newman wrote
Callista (at Wisemans request and published anonymously in the
Catholic Popular Library) in what seems to be almost direct response
to Kingsleys Hypatia.
34
Newman details Callistas spiritual struggle
leading to her conversion and her eventual martyrdom. Callista also
enjoyed great popularity and also was adapted for the stage in The
Convert Martyr by Frederick Husenbeth.
35
31
Charlotte E. Crawford, Newmans Callista and the Catholic Popular Library,
Modern Language Review 45 (1950): 219, cited in Susann Dorman, Hypatia
and Callista: The Initial Skirmish between Kingsley and Newman, Nineteenth-
Century Fiction, 34. no. 2 (1979): 17393.
32
Litvack, Callista 16567.
33
Ibid., 16263. Newman was ordained as an Anglican priest in 1825, and he was
a well-known scholar at Oxford. His conversion to Roman Catholicism in 1845
was shocking because he had been the central figure the Tractarian Movement
(or Oxford Movement, 18331845), which sought to affirm the High Church of
England as the middle ground between Roman Catholicism and Protestantism.
He was ordained as a Catholic priest in 1847, and in 1879 he was made a cardinal
by Pope Leo XIII.
34
See Dorman, Hypatia and Callista, for a comprehensive comparison of Hy-
patia and Callista.
35
Litvack, Callista 16770.
History of Catacomb Archaeology 25
Nineteenth-Century Popular Religious Press
In addition to writing extraordinarily successful Early Christian novels,
Newman and Wiseman played important roles in the popular religious
press of the period. The Roman catacombs were the subject of many
popular articles published between 1848 and the mid-1860s in The
Rambler, a publication of the Liberal Catholic Movement (and of which
Newman was briefly the editor in 1859); the Edinburgh Review; and the
Dublin Review, founded by Wiseman.
36
The debate between the Ultra-
montanes, who advocated traditional Roman devotions, centralized ec-
clesiastical authority, and veneration of saintly relics, and the Liberal
Catholics, who advocated intellectual freedom and saw scholarly and
scientific inquiry as independent of religious oversight, reached a height
in the Phial of Blood Controversy,
37
which was largely carried out in
the pages of The Rambler and in which the principles of scientific in-
quiry were tested against the authority of the Church.
Graves of the martyrs in the catacombs were officially identified by
glass or ceramic containers of blood affixed to the plaster outside of
the graves per a decree of 1668 issued by a commission appointed by
Clement IX. A question emerged in the mid-nineteenth century about
whether these containers held the actual blood of the martyr, collected
at the time of the martyrdom, or instead contained eucharistic wine, as
was commonly found in such containers in ordinary graves. The sug-
gestion that the containers supposed to identify the graves of the mar-
tyrs held only eucharistic wine called into question the authenticity of
the relics of the martyrs and sparked a vicious debate.
In 1860, Richards Simpson, then editor of The Rambler, suggested
that the question could be answered by a microscopic investigation of
36
See Joseph Altholz, The Liberal Catholic Movement in England: The Rambler
and Its Contributors, 18481864 (London: Burns & Oates, 1962); id., The Reli-
gious Press in Britain, 17601900 (Westport, Conn.: Greenwood Press, 1989),
99102. Mary Heimann, Catholic Devotion in Victorian England (Oxford and
New York: Oxford University Press, 1995) suggests that the most vocal propon-
ents on each side of the debate between the Ultramontanes and Liberal Catho-
lics likely viewed the debate in a larger context than that of England alone and
were either foreign or outsiders in some other way.
37
I am grateful to The Rev. Wendel W. Meyer, Ph.D., for providing me with an off-
print of his article The Phial of Blood Controversy and the Decline of the Lib-
eral Catholic Movement, which first introduced me to this fascinating debate
and on which this discussion relies.
26 Amy K. Hirschfeld
the contents of the containers. Although this type of testing would not
be given a second thought today, at the time, it was a daring suggestion
that seemed to many people like an attempt to test the Churchs auth-
ority by chemical analysis.
38
A chemical analysis of the contents of
sixty of these containers from the catacombs was eventually published.
All of the contents were found to be neither blood nor wine but instead
iron rust deposits that had effloresced from the glass.
39
In the con-
tinued debate that followed, many of the parties, which now included
even de Rossi in Rome, were hesitant to overtly question the Churchs
authority in the name of scientific inquiry.
40
In 1863, despite the scien-
tific evidence to the contrary, the 1668 decree was reaffirmed by the
Roman Catholic Church and signaled in many ways the downfall of
Liberal Catholicism in nineteenth-century England and the inability
of scientific evidence to stand up to the power of the Church.
Replicas of the Catacombs
at the Turn of the Twentieth Century
The extraordinary popular interest in the catacombs during the nine-
teenth century makes it likely that knowledge of the catacombs was
widespread both outside of Italy and outside of the academic world.
This widespread popular interest is likely responsible in part for two
interesting cases in which replicas of the Roman catacombs were pro-
duced at the turn of the twentieth century with the cooperation of the
Pontifical Commission of Sacred Archaeology.
41
In Valkenburg, The Netherlands, exact replicas of sections of the
catacombs of Priscilla, Domitilla, and S. Sebastiano were created in
38
Ibid., 86.
39
Today, scholars commonly agree that the containers held perfume. Ibid., 89, 93.
40
J. S. Northcote, an former Anglican priest and convert to the Church of Rome
who is well known for compiling and publishing the English version of de Rossis
Roma Sotterranea, wrote a popular series of articles on the Roman catacombs
for The Rambler in 18481849 and was initially a strong supporter of the publi-
cation. He later joined the opponents of The Rambler when the editors could not
appreciate the dilemma that Northcote felt so keenly, the need to balance the
search for truth with concern for the beliefs of the faithful. Ibid., 8788.
41
These instances of replication of the Roman catacombs in Valkenburg and in Re-
imans murals are the subject of a larger research project I am currently under-
taking.
History of Catacomb Archaeology 27
abandoned limestone quarries. The creators of these replicas sought to
evoke the religious experience of the Roman catacombs and had finan-
cial purposes as well in creating a tourist attraction and providing
work for the unemployed. They went to much effort, in cooperation
with Roman researchers, to make their replicas as exact as possible.
Orazzio Marucchi, de Rossis protg, was present at the opening of
the Valkenburg catacombs in July 1910. The Valkenburg catacombs re-
main a popular tourist attraction to this day.
42
Ivan Tsvetaev (18471913),
43
founder of the Pushkin Museum, com-
missioned artist Fyodor Reiman in the last decade of the nineteenth
century to make exact copies of paintings in the Christian catacombs
for the founding collections of Pushkin State Museum of Fine Arts
and possibly for a never-published atlas of early Christian frescoes.
44
Reiman spent 12 years in the catacombs making his watercolor copies
of the paintings and nearly lost his eyesight in the process. Interest-
ingly, Reimans watercolors were never exhibited until 2000, when the
exhibit Under the Vaults of the Roman Catacombs was shown at the
Pushkin Museum in honor of the Papal Jubilee.
45
42
Joep Didden, The Catacombs of Valkenburg in Limburg, SOK-Mededelin-
gen 26 (1996). See also the website of the Valkenburg catacombs, http://www.
katakomben.nl/.
43
Tsvetaev was the son of a priest and himself studied in the seminary at Vladimir.
He was a professor of antiquities and the Latin language in Moscow University
and is considered the first Russian specialist in Latin epigraphy. Tsvetaev
founded the Pushkin Museum (which opened in 1912 as the Moscow Museum
of Fine Arts and was originally part of Moscow University) on the model of the
Cabinet of Fine Arts and Antiquities of the Moscow University, which con-
tained scientifically precise casts and copies of ancient sculpture and art for edu-
cational purposes. Museum History, The Pushkin State Museum of Fine Arts,
1998, http://www.museum.ru/gmii/defengl.htm, viewed 20 March 2005.
44
Reimans watercolors were not published, and I found mention of the fact
that they were commissioned for an atlas of Christian frescoes only in the Bol-
lettino di informazione of the Italian Embassy in Russia, 28 February 2000,
http://www.ambrusital.mid.ru/AnbRusItal/it/bull8.htm#14, viewed 18 July
2001.
45
Unfortunately, no catalog of this exhibition was published. Personal correspon-
dence, M. Axenenko and T. Vorobjeva, Pushkin State Museum of Fine Art,
March 7, 2001.
28 Amy K. Hirschfeld
The Study of the Jewish Catacombs
I discuss the Roman Jewish catacombs separately rather than incor-
porate them into the previous discussion primarily to point out some
differences in how they have been studied and because they have been
treated separately in many ways by researchers, even up to the present
day. There are only six known Jewish catacombs in Rome, and only
one of those (the Monteverde catacomb) was known prior to the mid-
nineteenth century. The Jewish catacombs are not well known outside
of academia, and certainly never enjoyed the immense popular interest
that the Christian catacombs did in the nineteenth century. Although
the fact that there are so many fewer Jewish catacombs than Christian
catacombs accounts in part for a seeming lack of attention to the Jew-
ish catacombs, the fact that their investigation did not bear as heavily
on modern Judaism as the Christian catacombs did on modern Roman
Catholicism likely accounts for many of the differences in how they
have been studied, perceived, and presented.
There is no known documentation of pilgrimages to the Jewish cata-
combs after active burials cease,
46
and they certainly did not enjoy the
continued maintenance and refurbishing that the Christian catacombs
did before the removal of the relics to city churches.
Before the first discovery of a Jewish catacomb (that of Monteverde
in 1602), Jewish epigraphy was recorded in the sixteenth century
by van Winghe, Chacon, and Claude Menestrier.
47
This epigraphy was
46
However, the Itinerary of Benjamin of Tudela seems to indicate that there was an
awareness of the Monteverde catacomb in the twelfth century. Benjamin of
Tudela was a rabbi from Spain who undertook extensive travels to southern
Europe and Palestine in 11651173. He noted the existence of a cave in a hill on
the bank of the Tiber in which the ten Jewish martyrs were buried. This obser-
vation could indicate that in 1166 Roman Jews knew of an ancient Jewish under-
ground cemetery in this location. Marcus Nathan Adler, trans., The Itinerary of
Benjamin of Tudela (London: Henry Frowde, 1907). Caution must be taken in
viewing this itinerary (or any others) as historical fact as it is replete with mir-
acle stories, legends, and embellishments. However, it is descended from a long
tradition of rabbinic writing, an important feature of which is the affirmation or
validation of the present by association with significant events of antiquity.
Leonard Victor Rutgers, The Jews in Late Ancient Rome: Evidence of Cultural
Interaction in the Roman Diaspora (Leiden: Brill, 1995), 35.
47
Cod. Vat. Lat. 10545, fol. 150, cited in H. J. Leon, The Jewish Catacombs and
Inscriptions of Rome: An Account of Their Discovery and Subsequent History,
Hebrew Union College Annual 5 (1928): 299314
History of Catacomb Archaeology 29
probably found above ground and is no longer extant. Almost a cen-
tury later, in 1685, Jacob Spon published the Miscellanea eruditae anti-
quitatis, which included three Jewish epitaphs from ancient Rome, two
of which were those originally recorded by de Winghe.
Bosio initially discovered the Monteverde catacomb in 1602, but
he did not document it extensively, likely because it did not provide
evidence consistent with his devotional purpose. In the few pages
he does devote to a description of the catacomb, he states that he
includes a description of a Jewish cemetery in a work on sacred cem-
eteries so that it will be known that our cemeteries have never been
profaned nor contaminated by the bodies of either Hebrews or Gen-
tiles.
48
After Bosios discovery of the Monteverde catacomb, few efforts
were later made to locate this first and only recorded Roman Jewish
catacomb. It was not until the first half of the eighteenth century that
interest in Monteverde was reawakened, when Giuseppe Bianchini
claimed to have entered the necropolis with the archaeologist Cardinal
Domenico Passionei.
49
Subsequently, Gaetano Migliore, probably
after the mid-eighteenth century, visited the catacomb but was forced
to withdraw because of continuous rock slides. In an amusing, candid
account, he divulges that in spite of the imminent danger, he explored
the catacomb in order to ingratiate himself with scholars.
50
For nearly the next one hundred years, interest in the Monteverde
catacomb diminished to the point that even its location was unknown.
In 1843, Father Giuseppe Marchi (one of de Rossis mentors) tried to
locate the entrance to Monteverde, but his efforts proved fruitless. In
1879, the well-known Christian archaeologist Mariano Armellini de-
clared that he had found the entrance to Monteverde but that it was
blocked with soil, which had to be removed. Nikolaus Mller, who was
later to become the principal investigator of Monteverde, failed to
locate this burial ground in 1884 and again in 1888. M. Seymour de
Ricci, a French archaeologist, also tried and failed in his efforts to lo-
cate Monteverde in 1900 and 1904.
51
48
Quoted in Rutgers, The Jews in Late Ancient Rome, 10.
49
Leon, Jewish Catacombs, 303, believes Passionei did not visit Monteverde and
instead copied Bosios description and illustrations.
50
Cod. Vat. Lat. 9143, fol. 127, quoted in H. J. Leon, Jews of Ancient Rome (Phil-
adelphia: Jewish Publication Society of America, 1960).
51
Leon, Jewish Catacombs, 306307. In his New Tales of Old Rome (Boston:
Houghton Mifflin, 1901), Rodolfo Lanciani failed to distinguish this burial spot
30 Amy K. Hirschfeld
In the last half of the nineteenth century, in addition to investigating
Christian cemeteries, the Pontifical Commission of Sacred Archaeol-
ogy also supported the exploration of Jewish catacombs. These efforts
were probably influenced by Marchi and his encouragement of the
search to rediscover Monteverde. Marchis interest in Jewish cata-
combs was based in the desire to better understand early Christianity
rather than to understand the Jewish community of ancient Rome.
52
A second Jewish cemetery, the Vigna Randanini, was discovered in a
vineyard in 1857.
53
Excavations were made by the owner, Giuseppe
Randanini, and later by his son, Ignazio Randanini. Raffaele Garrucci
(18121885) explored and published a more detailed description of
Vigna Randanini
54
than did his predecessor E. Herzog.
55
Almost three-
quarters of a century later, Father Jean Baptiste Frey explored this
catacomb.
56
from a Christian cemetery because of the absence of ritual symbols on the frag-
ments found in the earth and the presence of so many identified catacombs
honeycombing the area.
52
Rutgers, The Jews in Late Ancient Rome, 32. Rutgers suggests that Marchi be-
lieved the Christian catacombs were based on the Jewish catacombs and saw this
as an illustration of the typically Christian view that Jesus had come not to ab-
rogate the Law, but rather to perfect it. [T]he architects of the Christain cata-
combs brought to perfection a mode of burial that had remained imperfect
among the people who had invented it.
53
Orazio Marucchi, Breve guida del cimitero giudaico di Vigna Randanini (Rome,
1884) dates the discovery of the catacomb to 1857 but Leon, Jewish Cata-
combs, 309, gives the date May 1, 1859, for the discovery as does Jean Baptiste
Frey, Corpus inscriptionum judiacarum (Vatican City, 19361952), 53. Marucchi
gives 1859 as the date of the first excavations, likely referring to Garruccis work
rather than the informal explorations of the owners after discovering the cata-
comb in 1857. Visconti mentions in his 1861 article on the excavations of the
Vigna Randanini that the Jewish hypogeum had been discovered in these last
years. Scavi di Vigna Randanini, Bulletino dellinstituto de corrispondenza
archeologica (1861): 16. David Noy, Jewish Inscriptions of Western Europe. Vol-
ume 2, The City of Rome (Cambridge: Cambridge University Press, 1995), 173,
follows Marucchis dates of 1857 for the discovery and 1859 for the excavation.
54
Cimitero degli antichi Ebrei scoperto recentemente in Vigna Randanini (Rome,
1862); Dissertazioni archeologiche di vario argomento (Rome, 18641865).
55
Le catacombe degli Ebrei in Vigna Rondanini [sic], Bulletino dellinstituto de
corrispondenza archeologica 1861: 91104.
56
Nouvelles inscriptions indites de la catacombe juive de la via Appia, Rivista
di archeologia cristiana 10 (1933): 2750. Freys monumental Corpus inscriptio-
num judiacarum was long considered the authoritative reference on the Jewish
catacomb inscriptions. Harry Leon updated and corrected the Jewish catacomb
History of Catacomb Archaeology 31
The next unexpected discovery of a Jewish catacomb, that of the Via
Labicana, now called Via Casilina, was made in late 1882 by Orazio
Marucchi, who was just commencing his notable career. Aware that he
had come upon an ancient Jewish cemetery because of the presence of
a menorah, Marucchi apprized his mentor, de Rossi, of his discovery.
After viewing the site, de Rossi realized the importance of the dis-
covery and urged Marucchi to publish a report on this catacomb.
57
The investigation of the Jewish cemetery of Via Labicana/Casilina
was supported by the Pontifical Commission, as were the investi-
gations of the small catacombs of the Vigna Cimarra and the Via
Appia Pignatelli (which at that time was mistakenly considered to be
Jewish), discovered by quarriers on the property of the Prince of Tor-
lonia near the Cafarella
58
and explored by Nikolaus Mller. Mller
went on to become the major explorer of Jewish catacombs in Italy,
and at the behest of the Commission investigated the Monteverde
catacomb, which had been exposed by a landslide in the early years of
the twentieth century.
Because Mller considered the exploration of Monteverde to be of
the utmost urgency, as did the Pontifical Commission, he conducted
a series of excavations with the permission of the proprietors of the
estate in 19041905 and in the spring and autumn of 1906. The 1906
excavations were funded by the Berlin Society for the Advancement of
Knowledge of Judaism. Mller had to abandon excavations in 1909,
and his death in 1912 prevented him from completing his publication
of this catacomb and compiling a complete description of the artifacts
retrieved during his explorations.
Mller had given Marucchi his incomplete Italian manuscript in
1912 before his death so that Marucchi could record it in the Acts of
the Papal Academy. He had left his other records and German manu-
script to his brothers, his legal heirs, who turned the material over to
a scholarly committee at the New Testament Seminars, at the Royal
inscriptions from Freys corpus (and followed Freys numbering of the inscrip-
tions) in the Jews of Ancient Rome, and more recently, David Noys Jewish
Inscriptions of Western Europe has become a standard reference for Jewish cata-
comb inscriptions, updating and correcting previous work as well as supplying
detailed bibliography for each inscription.
57
Orazio Marucchi, Di un nuova cimitero giudaico scoperto sulla via Labicana,
Dissertazioni della Pontificia Accademia Romana di Archeologia, ser. 2, t. 2
(1884): 497532.
58
Frey, CIJ, 50.
32 Amy K. Hirschfeld
University of Berlin. The Berlin Society for the Advancement of Knowl-
edge of Judaism, at the committees request, commissioned Nikos
Bees, Mllers assistant, to complete and publish Mllers German
manuscript.
59
Marucchi published Mllers Italian manuscript in 1915,
adhering exactly to the original.
60
Toward the end of 1913, a new section of Monteverde was revealed by
chance. Baron Rodolfo Kanzler, Secretary of the Commission of Sacred
Archeology, later wrote the official report
61
on the investigation of this
region by inspectors Enrico Josi and Giorgio Schneider Graziosi.
Josi and Schneider Graziosi were requested by the Pontifical Commis-
sion of Sacred Archaeology, in agreement with the Rome Superintend-
ency of Monuments, to remove all of the artifacts because this new
section of the catacomb was in such imminent danger of collapse.
One of the final important dates in the history of Monteverde was
1919, significant for the investigation of the last remaining section by
Roberto Paribeni.
62
The 24 inscriptions retrieved from this exploration
raised the total number actually discovered in the Monteverde cata-
comb over the years to over 200.
63
This is the largest number of
definitely Jewish inscriptions retrieved from any Roman catacomb. On
October 14, 1928, a devastating collapse occurred, destroying the cata-
comb completely for all practical purposes. The area is now covered
with apartment buildings.
In 1918 a Jewish catacomb was accidentally disclosed by laborers
strengthening the foundations of the stables on the grounds of the Villa
Torlonia on the Via Nomentana. De Rossi had predicted in 1865
such a discovery when he commented on the fact that if the agger
with synagogue nearby mentioned in the Publius Corfidius Signinus
59
Nikolaus Mller and Nikos Bees, Die Inschriften der jdischen Katakombe am
Monteverde (Leipzig: O. Harrasowitz, 1919).
60
Il cimitero degli antichi Ebrei posto sulla Via Portuense, Dissertazioni della
Pontificia Accademia Romana di Archeologia, ser. II, 12 (1915): 205318. Maruc-
chi added to the end of Mllers publication a note and an appendix of photos of
inscriptions (which had been sent by Mller without instructions about where
they should be included).
61
Scoperta di una nouva regione del cimitero giudaico della Via Portuense.
Nuovo bullettino di archeologia cristiana 21 (1915): 152157.
62
Via Portuense. Inscrizioni del cimitero giudaico di Monteverde, Notizie degli
scavi di Antichit 46 (1919): 6070.
63
Noy, Jewish Inscriptions, 2: 1172, includes 202 inscriptions and 56 fragments
with letters or symbols. Leon, Jews, 74, recorded 206 inscriptions.
History of Catacomb Archaeology 33
inscription
64
was the well-known Esquiline agger (as generally agreed
upon by scholars), a search should be made for Jewish cemeteries on
the Via Tiburtina or Via Nomentana. True to de Rossis prescience,
more than fifty years later Jewish catacombs were discovered under the
Villa Torlonia on the Via Nomentana. In 1919 the Prince of Torlonia,
65
who was also a senator, financed excavations under the technical direc-
tion of engineer Agostino Valente, assisted by the Roman Soprinten-
denza of excavations, represented by Roberto Paribeni. Several years
later, Hermann W. Beyer and Hans Lietzmann explored these burial
grounds more fully and recorded in greater detail their findings.
66
Many of the artifacts from the excavations of the Jewish catacombs
sponsored by the Pontifical Commission of Sacred Archeology were
removed and placed in the Sala Giudaica, a specially designed room in
the Museo Cristiano Lateranense.
67
The documentation of an ex-
hibition of the material in the Sala Giudaica concluded with the obser-
vation that all the scholars would be grateful to the Prefecture of the
Apostolic Palaces and to the administration of the museums for having
added this very important epigraphic group to the notable Christian
Lateran collection, an appropriate location because of the close rela-
tionships between the Jewish monuments and those of primitive Chris-
tianity.
68
This statement could be said to characterize much of the
way that the Jewish catacombs have historically been studied they
have been primarily studied from the perspective of their relationship
to early Christianity and not in their own right. The exploration of the
Roman Jewish catacombs has often been closely tied to the efforts of
64
Frey, CIJ, inscription 531, 391; De Rossi, Bullettino di archeologia cristiana 3,
no. 12 (1865): 95. This non-Jewish Latin epitaph is dedicated to Publius Cor-
fidius Signinus, who sold fruits in a stall near the proseucha (synagogue) next to
the agger. This epitaph attests to the fact that a synagogue existed at that lo-
cation and could have been that of the Siburesians (residents of the Subura), at
least four members of which were buried in the Torlonia catacombs, the closest
burial grounds for this congregation.
65
The history of nineteenth- and twentieth-century Roman archaeology appears
to be closely interwoven with the Princes of Torlonia, particularly the exca-
vations involving the burial finds of the Jews of ancient Rome.
66
Hermann W. Beyer and Hans Lietzmann, Die jdische Katacombe der Villa Tor-
lonia in Rom, Jdische Denkmler, vol. 1 (Berlin: W. de Gruyter and Co., 1930).
67
The Christian Museum was founded in 1854 by Pius IX in the Lateran Palace
and was moved in 1963 to the Vatican Museum as the Museo Pio Cristiano.
68
Giorgio Schneider-Graziosi, La nuova sala giudaica nel Museo Lateranense,
Nuovo bullettino di archeologia christiana 21 (1915): 56.
34 Amy K. Hirschfeld
the individuals who owned the land under which the catacombs were
found (e.g., the Randaninis
69
and the Prince of Torlonia), a situation
quite different from the Christian catacombs, which are often associ-
ated with martyrial churches and owned by the Church rather than by
any individual.
The Catacombs in the Twentieth Century
In 1929, the excavation and preservation of the catacombs of Rome
and Italy were officially entrusted to the Vatican in accordance with
Article 33 of the Concordat, one of the three sections of the Lateran
Pacts of 1929 dealing with the Roman Catholic Churchs ecclesiastical
relations with the Italian state. According to the Concordat, all of the
catacombs of Rome and Italy were directly administered by the Pon-
tifical Commission of Sacred Archeology.
In addition to extensive work on the Christian catacombs during the
twentieth century, the Pontifical Commission, restored and system-
atized the Jewish catacomb of Randanini and cleared and thoroughly
explored the catacombs under the Villa Torlonia.
The discovery in 19551956 of a small pagan-Christian catacomb on
the Via Latina was a complete surprise because its existence had never
been documented, probably because it was privately owned and did
not contain venerated tombs.
70
The fact that the discovery of the Via
Latina catacomb was so unexpected because it had not been docu-
mented is an indication of how heavily researchers relied on texts, to
the virtual exclusion of all other methods, for the exploration of the
catacombs.
On February 18, 1984, a revision of the Concordat was signed that
called for the administration of the extant Jewish catacombs to be
handed over to the Italian government. This marks the first official,
formal separation of the study of the Christian and Jewish catacombs
and has had significant effects on the continued maintenance and
study of the Jewish catacombs, in particular.
69
Even today, the maintenance and limited accessibility to the Randanini cata-
comb is made possible by the dedication and generosity of the Del Gallo family,
current owners of the land over the Randanini catacomb.
70
Antonio Ferrua, La pitture della nuova catacomba di via Latina (Vatican City:
Pontifical Institute of Christian Archaeology, 1960).
History of Catacomb Archaeology 35
The Present Study and Maintenance of the Catacombs
The Christian catacombs continue to be studied and maintained by the
Pontifical Commission of Sacred Archaeology and are an important
destination for tourists and pilgrims visiting Rome. In a speech to
the Pontifical Commission on 7 June 1996, Pope John Paul II indicated
the significance of the Christian catacombs as destinations for modern
pilgrims, in particular those visiting Rome for the Papal Jubilee in 2000:
By visiting these monuments, one comes into contact with the evocative
traces of early Christianity and one can, so to speak, tangibly sense the
faith that motivated those ancient Christian communities. Visitors
will be able to feel the atmosphere of the first conversions to the Gos-
pel. the catacombs should be a necessary destination for Holy Year
pilgrims. Thank you [members of the Pontifical Commission of
Sacred Archaeology] for your efforts and for the professional contribu-
tion you are making to evangelization with your activities.
71
In another
speech at a plenary assembly of the Pontifical Commission on 16 January
1998, the pope stated: your attention is appropriately focused on the
pastoral benefits of these famous monuments of Christian antiquity [the
catacombs]. In the silence of the catacombs, the pilgrim of the Year
2000 can rediscover or revive his religious identity on a sort of spiritual
journey that, by starting from the first testimonies of the faith, brings
him to the reasons for the new evangelization and to its demands.
72
Could this implicit mandate of the Pontifical Commission, the
organization that grants permission for access to direct study of the
Christian catacombs and is responsible for their presentation to the
public, be a potential obstruction to certain types of research that
might be perceived as contrary to the aims of the Church? A modern
scholar can productively consider, especially in light of the history of
catacomb studies, whether any tension might exist between pastoral
appreciation of the catacombs and scholarly research.
The Jewish catacombs remain under the administration of the Italian
state, which has recently undertaken a highly controversial program of
71
LOsservatore Romano, weekly edition in English, 19 June 1996, p. 7 quoted in
The Popes Speeches Concerning the Catacombs, The Christian Catacombs of
Rome website, http://www.catacombe.roma.it/en/discorso.html.
72
From LOsservatore Romano, quoted in The Popes Speeches Concerning the
Catacombs, The Christian Catacombs of Rome website, http://www.catacombe.
roma.it/en/discorso.html.
36 Amy K. Hirschfeld
privatization of Italian cultural heritage, in which culturally significant
sites have been sold to international investment firms and private
investors to generate funds to reduce Italys budget deficit or finance
public works.
73
What effect might privatization have on the Jewish
catacombs, which have received limited attention since they have been
in the hands of the Italian state since the revision of the Concordat
in 1984? A recent restoration project for the Villa Torlonia, which in-
cluded the construction of an underground parking lot near the villa,
did not include any consideration of the Jewish catacombs located
under the villa. Only after the Jewish Community in Rome publicly
denounced the government in a newspaper article and negotiated on
behalf of the catacombs was money allocated to the study of the cata-
combs and the plans for the underground parking lot abandoned.
74
The International Council on Museums and Sites recently reported
on the risk to cultural heritage when some sites are not given the same
priority as other examples of archaeological heritage, because they are
manifestations of particular historical periods or cultures. This
arises as a potential threat when one cultural group does not recognize
a segment of the archaeological heritage as relating to their current so-
cietys cultural tradition. As a result, alternative periods are given
greater priority for research and conservation as they are deemed to be
important to the dominant societys cultural identity.
75
Italy is cited in
this report as one of the countries where this risk is present.
The recent reporting of radiocarbon dates from the Torlonia cata-
comb
76
indicates that there is still much new data to be obtained from
the study of the catacombs and is an interesting modern example of the
73
Privatization has sparked heated political and social debate among politicians,
scholars, and the general public. See Salvatore Settis, Italia S.P.A. (Turin: Einaudi,
2002) and Roland Benedikter, Privatisation of Italian Cultural Heritage, In-
ternational Journal of Heritage Studies 10, no. 2 (2004): 36989.
74
Jessica Dello Russo, 500 Million Italian Lire to Finance Jewish Catacomb
Study, 3 February 2001, http://www.catacombsociety.org/nfr_2-3-2000.html.
75
Heritage at Risk, ICOMOS World Report 2001/2002 on Monuments and Sites in
Danger (Munich: K. G. Saur Verlag, 2001). Online. http://www.international.
icomos.org/risk/2001/icahm2001.htm, viewed 4 May 2005.
76
First reported in L. V. Rutgers, A. F. M. de Jong, and K. van der Borg, Radio-
carbon Dates from the Jewish Catacombs of Rome, Radiocarbon 44, no. 2
(2002): 541547 and brought to a wide audience after being published in Nature
(L. V. Rutgers, A. F. M. de Jong, K. van der Borg, and Imogen Poole, Jewish
Inspiration of Christian Catacombs, Nature 436 [21 July 2005]: 339) and sub-
sequently being reported by international media outlets.
History of Catacomb Archaeology 37
intersection between scholarly study and the popular press. The radio-
carbon dates for the Torlonia catacomb indicate that it was in use in
the second century C.E., at least a hundred years before the earliest
Christian catacombs. The authors of the study suggest that this could
indicate that Jewish catacombs influenced the development of the
Christian catacombs, contrary to the common belief that burial in
catacombs was begun as a Christian practice. They caution that no
final determination can be made without radiocarbon dating from
the Christian catacombs. When this story was picked up by the inter-
national media after being reported in Nature, the headlines and
related stories could give readers a range of different impressions. For
example, in the article Catacombs Had Jewish Origin, Not Chris-
tian,
77
under the heading Bursting Bubbles? the author reports that
the Italian media were disconcerted by the study, and that Romes
daily Il Messaggero wrote that the last myth on the catacombs has
fallen. the Christians did not even invent them.
In Catacomb Find Boosts Early Christian-Jewish Ties, Study
Says,
78
the author presents the findings of the study in terms of the
possibility of Jewish influence on Christian catacombs and gives some
detail on the historical lack of attention to the Jewish catacombs in
scholarship. He quotes a classical archaeologist as saying that the Jews
were treated extremely badly in seventeen-century Catholic Rome
when catacomb studies were first intensifying and since the 17
th
cen-
tury, its been traditional that catacomb archaeology is done by
members of the Catholic Church and nobody else.
In the article, Did Christains Copy Jewish Catacombs?
79
the
author clearly mentions the studys recommendation that radiocarbon
tests must be conducted for the Christian catacombs before any defi-
nite conclusions can be drawn and concludes with the statement Re-
gardless of whether another catacomb is found to be older the larger
point is that Jews and Christians co-existed peacefully for centuries
and clearly influenced each others culture.
77
Rossella Lorenzi, Discovery News, 26 July 2005, http://www.abc.net.au/science/
news/ancient/AncientRepublish_1422611.htm, viewed 31 March 2006.
78
James Owen, National Geographic News, 20 July 2005, http://news.nationalgeo
graphic.com/news/2005/07/0720_050720_christianity.html, viewed 31 March
2006.
79
Michael Schirber, LiveScience, 20 July 2005, http://www.msnbc.msn.com/
id/8644832, viewed 31 March 2006.
38 Amy K. Hirschfeld
Conclusion
To return to the opening quote from David Lowenthal, How do we
recognize and cope with this heritage, which at the same time sustains
and constrains us? What benefits does it provide, what costs does it
exact? The study of the Christian catacombs has been for the most
part a text-based, devotional enterprise, from which the Jewish cata-
combs have been excluded or to which they have been subordinated.
Periods of intense interest in the catacombs can often be attributed
more to religious polemics than to objective scholarship. How does
this intellectual history continue to influence the study of the cata-
combs today? Although scholars often cannot comprehensively review
the historical and intellectual context of every piece of past scholarship
that they consult, reflection on the history of a field such as catacomb
studies and the place of current research in that historical trajectory
can only serve to deepen and broaden scholarly inquiry.
Housing the Dead 39
Andrew Wallace-Hadrill
Chapter 2
Housing the Dead:
The Tomb as House in Roman Italy
Domus ista, domus!
That the tomb was a house for the dead was a topos, a commonplace
of imperial Latin literature. So Statius, poet of Domitians court,
composed a suitable lament for the death of Priscilla, wife of the
mighty imperial freedmen Abascantus (Silvae 5.1).
1
By leaving the
composition for a suitable interval, he was able to comment on her
tomb. Its massive marble construction would defy the erosion of time.
The statues of goddesses that graced it would not shame the divinities
themselves. The household servants (famuli ) and the usual crowd of at-
tendants are present for the obsequies, and for the regular rituals of
celebratory meals.
domus ista, domus! quis triste sepulcrum/dixerit?
It is a house, a true house. Who could call it a sad sepulchre? (237f.)
Statius does not comment on the form; in fact, a classic circular mau-
soleum, if the traditional identification is right with a monument on
the Via Appia by Domine Quo Vadis.
2
It is not the shape of the tomb,
so much as its activity which provokes his outburst: the crowd of atten-
1
I am grateful to all my fellow participants for stimulation and discussion, and
in particular to my discussants, Brandon Cline and Young-Ho Park. I also owe
an especial debt to Regina Gee and Robert Coates-Stephens for sharing with
me their own knowledge of Roman burials. For help with sourcing illustrations,
I am indebted to Adam Gutteridge, Dr. Greta Stefani (Soprintendenza Arche-
ologica di Pompei), Prof. Henner von Hesberg, Dr. Sylvia Diebner (German Ar-
chaeological Institute, Rome), and Prof. Filippo Coarelli.
2
See A. La Regina, ed., Lexicon topographicum urbis Romae: suburbium, vol. 1
(Rome: Quasar, 2001), s.v. Appia Via, 101 with fig.83.
40 Andrew Wallace-Hadrill
dants, servants, and others, not to speak of the dignified company of
the goddesses, mean that Priscilla is not left sadly on her own, but con-
tinues in death as in life to be in good company.
Another freedman, this time fictional, Trimalchio, is famous for his
reflections on how to house the dead in proper style.
Are you going to build my tomb as I instructed? I do want you to be sure to put
my puppy at the feet of my statue, and wreaths, and unguents and all Petraites
best fights. Your kind act will give me life after death. The dimensions now: a
hundred feet of street front, two hundred of depth for I want every kind of fruit
around my ashes, and a generous vineyard. Because it really is nonsense for a
person to have a nice house when hes alive and not to worry about the one in
which hes got to live for rather longer. So the most important thing is a notice
this monument does not descend to the heir (Petronius Satyricon 71,68,
trans. Purcell)
3
Petroniuss masterly parody scores off so many familiar features of
early imperial burials, and above all, those of the freedman.
4
Again, we
cannot be quite sure of the form of the tomb. The standard formula for
the dimensions (so many feet in fronte, so many in agro) refers of course
to the entire plot, with its ample provision for vines and flowers as in
many cepotaphia. But it sounds less like the form archaeologists call a
house-tomb, and more like the freemen burials of the period outside
the Herculaneum gate of Pompeii, with a larger enclosure around a
monumental altar. Specifically, the tomb of Naevoleia Tyche and Mu-
natius Faustus, Augustalis and paganus, honoured by the Council with
a bisellium (Fig. 2.1), seems to fit Trimalchios prescription,
be sure to have ships in full sail on the of my monument, and me sitting on a
platform in full official dress with five gold rings dishing out cash to the people
from a bag
5
3
Nicholas Purcell, Tomb and Suburb in Rmische Grberstraen. Selbstdarstel-
lung Status Standard. (eds. Henner von Hesberg and Paul Zanker; Munich:
Verlag der Bayerischen Akademie der Wissenschaften: in Kommission bei der
C. H. Beckschen Verlagsbuchhandlung, 1987), 25.
4
The passage is discussed in detail by J. Whitehead, The Cena Trimalchionis
and Biographical Narration in Roman Middle-Class Art, in Narrative and
Event in Ancient Art (ed. P. J. Holliday; Cambridge: Cambridge University Press,
1993), 299325. On freedmen burials, P. Zanker, Grabreliefs rmischer Frei-
gelassener, Jahrbuch d.deutschen archologischen Instituts 90 (1975): 267315.
On Trimalchio, G. Rowe, Trimalchios World, Scripta classica israelica 20
(2001): 22545.
5
For the parallel, see V. Kockel, Die Grabbauten vor dem Herkulaner Tor in Pom-
peji (Mainz: P. v. Zabern, 1983), 105f.
Housing the Dead 41
2.1. Tomb of Naevoleia Tyche
and Munatius Faustus
at Herculaneum gate of Pompeii:
(a) general view of side, with
ship in sail; (b) inscription
on faade, with scene of
liberation distribution (Soprin-
tendenza Archeologica di
Pompeii. Used with permission.)
a
b
42 Andrew Wallace-Hadrill
Again, what makes the tomb a house is not so much shape as the ex-
tension of the activities of lifetime, the commercial success, the popu-
lar benefactions, and the garden which ensured that the family could
have regular festivals to celebrate around the tomb, the Parentalia,
the Rosalia, and the Violaria that are specified in so many inscrip-
tions.
6
People, not walls, make a house as well as a city.
The tomb-as-house metaphor continued to flourish throughout the
empire, as the numerous passages cited by the Thesaurus Linguae Lati-
nae show (TLL IV, 1979 s.v. domus 1B2c), and into late antiquity. The
Codex Theodosianus shows the deep concerns about the destruction
of tombs of the successors of Constantine (who of course destroyed
tombs to build his basilica for St. Peter), and the language of houses
strangely interweaves their protests. So Constantius II, from Milan in
356 or 357:
Those who violate the habitations of the shades, the homes, so to speak, of the
dead, appear to perpetrate a two-fold crime. For they both despoil the buried
dead by the destruction of their tombs, and they contaminate the living by the
use of this material in living (Codex Theodosianus 9.17.4).
So not only are tombs like homes, they specifically risk contamination
by confusion with the homes of the living.
We learn that some men too eager for gain destroy tombs, and transfer the build-
ing material to their own homes (Codex Theodosianus 9.17.3, Constantius,
356 C.E.).
The trouble of course is that tombs are so close to houses that the
elements are in part interchangeable, a point reinforced by Julian:
Some men even take away from the tombs ornaments for their dining rooms and
porticoes (Codex Theodosianus 9.17.5).
The very fact that tombs were places for dining rendered them the more
suitable for despoliation for the benefit of the houses of the living.
Funerary epigraphy itself bears out the persistence in Latin epitaphs
of the house/tomb analogy. Richmond Lattimore gathered a selection
of the passages, noting the frequency of the expression aeterna domus.
7
6
See in general J. Toynbee, Death and Burial in the Roman World (Ithaca, N.Y.:
Cornell University Press, 1971; repr., Baltimore: John Hopkins University Press,
1996), 614; K. Hopkins, Death and Renewal. Sociological Studies in Roman His-
tory 2 (Cambridge: Cambridge University Press, 1983), 233.
7
R. Lattimore, Themes in Greek and Roman Epitaphs (Urbana, IL: University of
Illinois Press, 1962), 165ff.
Housing the Dead 43
The expression is ambiguous since sometimes it refers to the Greek
concept of Hades as the eternal house of the dead, but often too the
reference to the tomb is explicit:
haec domus aeterna est, hic sum situs, hic ero semper
Here is my eternal home, here I lie, here shall I be for ever.
8
And again, the conscious interplay of the houses of the living and the
dead is to the fore:
aedes aedificat dives, sapiens monumentum;
hospitium est illud corporis, hic domus est.
The rich man builds a house, the wise man a monument;
the first is a lodging for the body, the second a home.
9
By the familiar paradox, the domus is degraded to the status of a tem-
porary lodging house, while the funerary monument becomes the true
home, the domus.
As Lattimore interestingly comments, the tomb/house figure seems
to be a great deal more common in Latin epitaphs than Greek; in-
deed, the Greek passages are late, meaning from Greek areas under
Roman rule, and often look very much like translations.
10
Several
scholars have recently pursued the house/tomb analogy, including
Keith Hopkins, Nicholas Purcell, Richard Saller, and Valerie Hope.
11
In particular, John Pattersons interesting chapter on Living and
Dying in the City of Rome looks in parallel at the houses and the
tombs of rich and poor, and ends by concluding with the observation
that the link goes back to the Villanovan hut-urns of the beginning of
the first millennium.
12
A visit to a modern Italian cemetery like the
Campo Verano in Rome, with its characteristic house-like family
8
Lattimore, Themes, 168 = Carmina Latina Epigraphica (ed. F. Buecheler; Leipzig:
Teubner, 189597), 434.15, Pisaurum.
9
Lattimore, Themes, 168 = Carmina Latina Epigraphica (ed. F. Buecheler; Leipzig:
Teubner, 189597), 1488.12, Rome.
10
Lattimore, Themes, 165.
11
K. Hopkins, Death and Renewal, 20156; Purcell, Tomb and Suburb; R. P.
Saller, Patriarchy, Property and Death in the Roman Family (Cambridge: Cam-
bridge University Press, 1994), 95101; V. M. Hope, A Roof Over the Dead:
Communal Tombs and Family Structure in Domestic Space in the Roman World:
Pompeii and Beyond (eds. R. Laurence and A. Wallace-Hadrill (Portsmouth,
R.I.: JRA, 1997), 6988.
12
J. R. Patterson, Living and Dying in the City of Rome: Houses and Tombs in
Ancient Rome: The Archaeology of the Eternal City (eds. Hazel Dodge and Jon
Coulson; Oxford: Oxford University School of Archaeology, 2000), 25989.
44 Andrew Wallace-Hadrill
tombs, in stark contrast to the separate gravestones of the Acattolico
cemetery reserved for non-Catholic foreigners, but above all northern
European and American Protestants, might lead one to think that here
is one of those great cultural continuities. Is it somehow specifically
Italian to link burial to house and family?
But to leap from a Villanovan hut-urn to the Campo Verano seems
to me altogether too risky a project. If there is some degree of persis-
tence of the tomb/house analogy even through the ancient Roman
period, the apparent continuity masks some fairly fundamental shifts.
Where does the analogy actually get us, or where did it get them? Meta-
phors are slippery, shifting things, which refuse to be pinned down at
the moment you most need to push them. The Romans evidently en-
joyed playing with the analogy, and so do modern scholars talking
about the Romans, but it is one thing to use the comparison as a rhe-
torical trope or figure, another as an argument or hypothesis. Signifi-
cantly, the majority of scholars mentioned above draw attention to the
house-tomb link almost as an aside. Only Richard Saller, who has some
investment in the potential of extracting information about the struc-
ture of the Roman family from tombs and their inscriptions, comes
close to incorporating it in his argument (and even he is admirably cau-
tious); as Valerie Hope suggested, tombs seem to tell us more about the
role of the freemen and servile household in the family than about the
nuclear family.
To begin to assess the significance of the tomb/house link, we need
also to understand its limits. Scholars can be curiously uncritical about
this. In particular, the brick-built house-tombs that characterise the
Vatican necropolis and the Isola Sacra have led to enthusiastic appro-
priation of the analogy. Saller well quotes Toynbee and Ward-Perkinss
evocative commentary on the Vatican necropolis:
The Vatican house-tombs, and their counterparts elsewhere, so simple without,
so richly decked and colourful within, were surely regarded as places in which
the dead, in some sense or at some times, resided. Hidden away behind stout
doors and seen only by members of the owners families on anniversaries and
feast-days, when sacrifices, ceremonial meals, and ritual washings took place, all
this luxuriant internal ornament and art must have been designed as much to de-
light the dead as to gratify and instruct the survivors.
13
13
J. Toynbee, and J. B. Ward-Perkins, The Shrine of St. Peter and the Vatican Ex-
cavations (New York: Pantheon Books, 1957) 113f., cited by Saller, Patriarchy, 97.
Housing the Dead 45
Here the ancient trope of the grave as the eternal house of the dead
is put to work to explain an apparent paradox, the disproportionate
investment in the artistic decoration of the invisible inside of the tomb
(behind stout doors), rather than the visible outside. And yet the
explanation contains its own visible contradiction. The authors are well
aware that the family of the dead regularly penetrate behind those stout
doors for festival celebrations, and yet it is assumed that the decoration
is for the benefit of the dead not the living. Or at least the survivors are
building as much to delight the dead as to gratify themselves. But the
most spectacularly decorated of these house-tombs, that of Valerius
Herma, as the inscriptions tell us, was built by Valerius in his lifetime.
Are the survivors building for the dead Herma, or is the living Herma
building for the living, and to ensure a continued presence of the sur-
vivors at his own tomb?
The tomb/house analogy is partial. Even the house-like appearance
of the faade is misleading.
14
We are so used to suburban houses with
pitched roofs, that we instantly recognise a house in the formula of a
rectangular front with door and windows and a pitched roof. But did
not Roman domus roofs pitch inwards to the impluvium? Then, what
sort of a house opens inwards to a single chamber? Sometimes there is
provision for the sloping couches of a triclinium, either inside the tomb
as in the very interesting examples of the tombs outside the gates of
Ostia studied by Boschung,
15
or immediately outside as in the tomb, as
in the case of the tomb of Verria Zosime at the Isola Sacra. But while
the triclinium is an evident derivative of, and allusion to, domestic
arrangements, there is no attempt to evoke the internal architectural
arrangements of a Roman house. All of which is simply to say that the
analogy is a partial one, and raises the question of its limits and effec-
tiveness.
Another set of questions about limits is raised by the very frequency
of house-tombs in these locations. If the house form was effective, why
is it only one among many?
16
Look down the streets of tombs of Pom-
peii, with their carefree mixture of styles, circular mausolea, altars,
14
I owe this point, and stimulating discussion of the issue, to Regina Gee.
15
D. Boschung, Die republikanischen und frhkauserzeitlichen Nekropolen vor
den Toren Ostias in Rmische Grberstraen, 111124.
16
The typology of tombs is surveyed by H. von Hesberg, Monumenta. I sepolcri
romani e la loro architettura (Milan: Longanesi, 1994) 71230, esp. on house
form 8992, one of the briefest sections.
46 Andrew Wallace-Hadrill
miniature temples, exhedrae, columns, enclosures with the semi-carved
heads of columellae, all such splendid diversity, alongside a few
examples that can reasonably be classified as house-tombs, and one is
bound to ask what symbolic or other function was better performed by
the house-tomb that was not also performed by the others? We have
already seen that Trimalchios tomb, for all his anxiety to make it a
house for ever, seems not to have been imagined as a house tomb.
Finally, we may ask whether we are not simply carried away by the
pleasure of the rhetorical trope into pushing the analogy further than
it can bear. Take the columbarium, the remarkable pigeon-loft form of
the late republic and early empire, which in its most dramatic examples
provided capacity for several hundred urns. Keith Hopkins, whose in-
terest is in a crowded city and its forgotten masses, sees in the colum-
barium the analogue to the metropolitan insula, with its many floors
and packed tenements.
17
Nicholas Purcell by contrast, who observes
that the most important examples were built for the servile households
and dependants of the high aristocracy, says the analogy is rather with
the domus with its endless attics and tabernae and ramifications for
the long and short-term stay of the dependents, not the insula.
18
But
close though Purcells image of the domus is to my own, I cannot ad-
judicate here between Hopkins and Purcell, for the real architecture of
domestic space (internal divisions and floors) is simply absent, and
each scholars metaphor makes its point, just so far as it can be pushed.
Such questions lead me to suppose that there is room for a more
thorough and critical survey of the linkage of house and tomb. Part of
what I have to say is that a higher degree of critical distance is in place.
But above all I wish to argue that the analogy cannot get us very far
until it can be incorporated into a hypothesis, and this is what I would
like, in however provisional a form, to offer. In doing so I shall lean
heavily on what I have already written about the Roman house. I then
suggested that one way in which to understand the underlying dynamic
of the Roman house was to see it as a tension between two different di-
mensions or axes. The house is Janus-like, looking in two directions,
outwards and inwards. It looks outwards to the world beyond its
doors, foris, and to those visitors from outside who penetrate its doors.
In looking to the outsiders (strangers, clients, guests, friends, outsiders
in varying degrees) it seeks to impress, and makes statements about the
17
Hopkins, Death and Renewal, 21417.
18
Purcell, Tomb and Suburb,39.
Housing the Dead 47
status and identity of the insiders. Simultaneously it looks inwards,
domi, for it is a space also articulated for the insiders, who have their
own crucial social distinctions: slave and free, men and women, adults
and children. The complexity of reading Roman domestic space, I ar-
gued, derives the imperative for the same set of spaces to communicate
in both directions at once, inside and outside.
19
Similar considerations, I now suggest, are equally valid for the Roman
tomb. It looks outward, to the passing visitor, the hospes often invoked
by the epitaph, the unknown stranger who stands for everybody, since
the tomb is deliberately placed (or at least in many cases is placed) close
to the major thoroughfares leading into the city. Many tombs, not least
that of Trimalchio, had their eye primarily to the passerby. Tombs
are consequently major public declarations of identity and status, the
assumption implicit in the subtitle of Rmische Grberstraen Selbst-
darstellung-Status-Standard. But they also look inward, to a closed
circle of the family, those who gather with their wine and roses and
violets on the festal days, and gradually, one by one, take their resting
places within. One of the fundamental functions of funerary rites is the
reintegration of the family group, shattered by the brutality of loss.
20
The family is not ruptured, but continues: funeral masks, portrait
statues, inscriptions work to maintain the continuity. If Toynbee and
Ward-Perkins are surprised that the art is inside behind closed doors,
their surprise is that this function has taken precedence over the func-
tion of declarations of identity and status to the world outside.
This might suggest a simple dichotomy: the exterior aspects of the
tomb serve an outwards-looking function, the interior aspects serve
an inward-looking function. But just as with the house, it is vital to
appreciate that the outwards/inwards divide is more complex than
that. In the case of the house, simplistic distinctions of public versus
private areas are not helpful: the public penetrates the most private
recesses of the house (the masters bedroom, or the latrine by the
kitchen), and the private penetrates the public (women and children
are not kept away from the public area of the atrium but share it, and
19
A. Wallace-Hadrill, Houses and Society in Pompeii and Herculaneum (Princeton:
Princeton University Press, 1994), 812.
20
R. W. Chapman, I. Kinnes, K. Randesborg, eds, The Archaeology of Death
(Cambridge: Cambridge University Press, 1981); I. Morris, Death-Ritual and
Social Structure in Classical Antiquity (Cambridge: Cambridge University Press,
1992), 10.
48 Andrew Wallace-Hadrill
slaves are present at every point). In the case of the tomb, the gaze of
the passerby may rest on the family united in its festivities, and to
some extent (varying greatly from case to case) the tomb serves pre-
cisely as a public representation of the intimate unit of the family. Who
is displayed is a critical decision. Trimalchio in a tantrum threatens
Fortunata to exclude her statue from his tomb; he is confident he
wants to display his own superabundant importance, and will commit
himself on a puppy to be carved at his feet, but is less certain whether
he wants to let his wife in on the display (Sat. 74.17). Herein, of course,
his gross vulgarity. It is because so many Roman tombs at least to some
extent put the family unit on public display that Saller and Shaw could
make such telling use of their inscriptions.
21
My hypothesis, then, proposes that both functions external and
internal are simultaneously present in all Roman burials, but that
the balance and relationship between them can vary substantially, and
that we can detect changing patterns over time. To illustrate the con-
cept, I offer one Pompeian example of what might be termed the Trim-
alchio syndrome, where the balance seems to be tipped strongly in
favour of the external function, and yet the internal function is indeed
present, despite a certain level of uncertainty. This is the tomb of
Vesonius Phileros (Fig. 2.2) in the necropolis outside the Porta Nocera.
22
The form of the tomb is of an aedicula, a miniature temple with a
simple opening and pediment on a high podium, looking down on the
street. Within the portico are three statues, two males flanking one fe-
male. The titulus identifies the characters: P(ublius) Vesonius Phileros,
G(aiae) l(ibertus), Augustalis, his patrona, Vesonia P(ubli) f(ilia), and
M(arcus) Ofellius Faustus M(arci) L(ibertus) amicus. Here is indeed a
strange family grouping, the freedman and his female patron, and an
unrelated man tied only by friendship. The monument belongs to the
familiar type of one erected in the life of the commemorator/com-
memorated, vivos monument(um) fecit sibi et suis. Here is conspicuous
self-display; the element of status display is only enhanced by the fact
that AVGVSTALIS is added at a subsequent point in a second hand.
Vesonius may seem anything but a family man: no wife, no children,
not even freedmen generalised as libertis libertabusque posterisque suis.
21
R. P. Saller and B. Shaw, Tombstones and Roman Family Relations in the Prin-
cipate: Civilians, Soldiers and Slaves. JRS 74 (1984): 12456.
22
A. DAmbrosio and S. De Caro, Un impegno per Pompei. Fotopiano e documen-
tazione della necropoli di Porta Nocera (Milan, 1983), tomb number 23 OS.
Housing the Dead 49
2.2. Tomb of Vesonius Phileros at
Porta Nocera necropolis, Pompeii:
(a) statues of Vesonius, Vesonia,
and Orfellius; (b) inscription to three
commemorated persons, with later
inscription added below cursing
Orfellius (Soprintendenza Archeologica
di Pompeii. Used with permission.)
a
b
50 Andrew Wallace-Hadrill
And yet his monument constitutes a pseudo-family, displays it, and
then goes on in a second and longer inscription to display the despair
at the breakdown of the group so constituted and displayed.
23
Hospes paullisper morare si non est molestum (Stranger, stay a while
if it is not a nuisance, and hear my sad tale). It is a cautionary tale
(quid evitas cognosce). The syntax is faltering, the sentence breathless,
the cri-de-coeur rings loud and clear. This man whom I had hoped to
be my friend! By him accusers were instigated against me and proceed-
ings were started. I thank the gods and my own innocence, I am freed
of all nuisance. The one of us who lies, him may neither the gods of the
house nor the gods below receive.
24
The curse upon the false friend is
eloquent of the expectations of the tomb. For the very act of including
his friend in the memorial display constitutes him as a family member.
The curse excludes him simultaneously from the gods of the house,
di penates, and the gods below, di inferi. The function of the tomb then
is to facilitate that link. The display of those united around the di
penates, the family gods of the living house, projects their unity into
the underworld, the house of the dead. In the end, Vesonius stands
stripped to eternity of his pseudo-family, uncertain what to display to
the outside world, to the passing hospes.
But we should not shed a tear for him too hastily. One of the most
important features of Pompeian tombs is the survival within them of
individual headstones, called columellae, which often bear the name of
the buried (Fig. 2.3). There were no less than 18 such headstones within
Vesoniuss tomb. Apart from himself and his patrona, Vesonia, we find
a Vesonius Proculus, who died at age 13, a Vesonia Urbana, who lived
to 20, and a (H)eliodorus, who lived to 18. At this point we can only
guess the story. The patrona sounds to have been his partner as well as
former owner. Presumably they are the parents of Vesonius Proculus
and Vesonia Urbana. Heliodorus should be one of their slaves, as in all
likelihood are the 13 other unnamed columellae, unless any of them
were freedmen. At least we can be confident Marcus Ofellius was not
23
See now for a much improved reading of the inscription E. Rodrguez Almeida,
Topografia e vita romana: da Augusto a Costantino (Rome: Elenco, 2001), 91103.
24
Rodrguezs text is: Hospes paullisper morare/ si non est molestum et quid evites/
cognosce. Amicum hunc quem speraveram mi esse! Ab eo mihi accusato/res sub-
iecti et iudicia instaurata. Deis/ gratias ago et meae innocentiae: omni molestia
liberatus sum. Qui nostrum mentitur,/ eum nec di penates nec inferi recipiant.
Improved readings are ab eo for [h]abeo, and accusatores for the senseless
accusato res.
Housing the Dead 51
a
2.3. Columellae within tomb of
Vesonius Phileros at Porta
Nocera necropolis, Pompeii:
(a) general view of columellae in
interior of tomb; (b) detail of
columella of (H)eliodorus.
(Upper photograph by William
van Adringa. Soprintendenza
Archeologica di Pompei. Used
with permission.)
b
52 Andrew Wallace-Hadrill
among them. The interior of the tomb thus reconstitutes the family so
partially glimpsed on the exterior.
25
To explore in detail this two-faced internal/external aspect of the
tomb throughout Roman Italy would strain the limits of space and
time of this paper. Instead, I wish to illustrate what seem to be the sub-
stantial changes over time of Roman practice by exemplifying three
moments: the first, a mid-Republican moment when Roman material
culture still has palpable links to Etruscan practice, through the tomb
of the Scipios; the second, a moment of late Republican/early imperial
transition seen through the necropolis of the Porta Nocera at Pompeii;
the third, the high imperial tradition of the house-tombs of the me-
tropolis seen through the tomb of Valerius Herma beneath St. Peters.
In each case, I shall risk overgeneralization through exemplification;
my interest is underlining the substantial contrasts across time as
much as the continuities.
The Mid-Republic and the Tomb of the Scipios
The dearth of evidence from Rome of either houses or tombs predating
the late Republic, coupled with the sharp imbalance in evidence from
Etruria in the same period between abundant tombs and scarce houses,
has long meant that Etruscan tombs have had to work very heavily to
supply the gaps in our knowledge of housing in both areas, and of
tombs in Rome.
26
Carandinis excavations at the foot of the Palatine,
coupled with the chance discovery of a suburban villa site beneath
Renzo Pianos new Auditorium, now mean we are on better ground in
talking about Roman houses of the period between the sixth and sec-
ond centuries. Even so, it is striking to observe how in the publication
of the Palatine houses by Carandini and Carafa, Etruscan tombs are
still put under heavy contribution to establish the development of the
atrium houses.
27
Just how plausible are their reconstructions of house
plans of sixth-century Roman atriate houses with central impluvia is
not a theme I wish to pursue here, though it must be said it takes cour-
25
For the inscriptions, DAmbrosio and De Caro, Un impegno per Pompei.
26
Hesberg, Monumenta, 2932 for an overview of this early period.
27
A. Carandini and P. Carafa, Palatium e sacra via I. Prima delle mura, let delle
mura e let case arcaiche. Bollettino di archeologia, vols. 3133 (Rome: Istituto
poligrafico dello Stato 1995), esp. 237ff., 266ff.
Housing the Dead 53
age to reconstruct an entire atrium-house when the evidence is the odd
disjointed fragment of walling.
That Etruscan tombs have long seemed to offer a reflection of hous-
ing is scarcely surprising. Consider the striking transformation that
takes places in the Banditaccia necropolis of Cerveteri in the late archaic
period. In the seventh century the cemetery is characterised by circular
tumuli, some very large, some quite modest. The burial chambers be-
neath them appear in plan rather like bacilli: long corridors with short
side elements towards the end, and a culminating chamber. Then in the
early sixth century the form of the burial chambers beneath the tumuli
changes significantly. They become square in plan, with a characteristic
three-fold division: short entrance flanked by two chambers lead to a
wide central chamber, and at its back, a group of three equal cells. The
formal links with the atrium house seem irresistible: fauces flanked by
two rooms lead into atrium, leading to tablinum flanked by two rooms.
Then in the late sixth century, the tumulus shell is dropped, to be re-
placed by neatly aligned streets of cube tombs (tombe a dado). We
seem to be witnessing an urban revolution, a move from villages of
huts to colonial cities laid out on a grid pattern. At the same period we
find these neat streets of tombs at a number of other sites, including
the necropolis at Crocifisso del Tufo at Orvieto, and the Monterozzi
cemetery at Tarquinia.
It seems quite reasonable to interpret these tombs as deliberately
evoking a domestic parallel. This seems to be underlined by the
evocation of domestic features like windows, doors, and pilasters, and
particularly by the ceilings which often evoke a pitched roof, with cen-
tral beam and downwards-sloping rafters. Yet these pitched roofs
are as ambivalent as those of the Vatican house tombs. Colonna,
followed by Carandini, argues strongly for the introduction of the
compluviate roof and central impluvium from as early as the sixth
century.
28
That is the rereading of the fifth-century houses of Marza-
botto, long supposed to have been covered by displuviate, outward-
sloping, pitched roofs, but now argued on the basis of their internal
drainage arrangements to have been compluviate. If it is the case that
as early as the sixth and fifth centuries, the characteristic image of the
atrium was the inwards sloping roof and impluvium, were the tombs
with their outwards-sloping roofs really evoking atria?
28
G. Colonna, Urbanistica e architettura in Rasenna. Storia e civilt degli Etru-
schi (ed. Massimo Pallotino; Milan: Libri Scheiwiller, 1986) 371530.
54 Andrew Wallace-Hadrill
Short of substantial new evidence about early housing in both Rome
and Etruria, this debate is liable to loop in circles. But for present pur-
poses, it is enough to observe that Etruscan tombs persistently had fea-
tures that evoke a domestic context, like chairs, couches, and pillows,
taken to a height in the fourth-century Tomba dei Rilievi at Cerveteri,
in which the plastered walls are decorated by a splendid range of fur-
nishings which are at least partly domestic (though also partly ritual,
pointing to sacrifice). But where the domestic imagery gains its rele-
vance is in the function of such tombs in reconstituting and represent-
ing the family. The architecture in itself, by creating a series of beds,
arranged with a strong sense of hierarchy, the preferred position being
the central niche of the tablinum, and possibilities of subgroupings in
the lateral chambers, points to the desire to represent the occupants as
a structured group. The abundant epigraphic material confirms that
the typical group was the multigenerational family.
A classic example, from a period of close interaction with the middle
Republican Roman aristocracy, is the tomb of the Volumnii at Peru-
gia.
29
Dating from the late third century, the tomb is located outside
the town at the bottom of the hill. Externally it is unremarkable: steps
lead down to an underground chamber hewn from the soft tufo. At the
bottom is a large rectangular hall (atrium) with a pitched roof and
rafters, with a main chamber (tablinum) on the central axis, and two
lateral chambers (Fig. 2.4). The main chamber contains the remains
of seven named members of the Volumnius/Velimna family. Arnth
Velimnas, the founder of the tomb, dominates from his high couch
with pillows and drapes, held aloft by two winged daemons. To the left,
his daughter Veilia Velimnei is the one female of the group, the unmar-
ried daughter of Arnth. Male descendants (Thefri, Avle, Larth, and
Vel) stretch down to the last, early imperial, member of the group,
Publius Volumnius Violens, Roman enough to Latinise his script and
name, but still Etruscan enough to give his matronym. His elegantly
carved ash urn is in the form of a rectangular building with a pitched
and tiled roof, double doors, and Corinthian pilasters; a house, it is
normally said, though the form is a great deal closer to a temple than a
house.
I linger over this Etruscan background in order to bring out a point
relevant to the issue of the internal/external functions of the tomb.
29
S. Haynes, Etruscan Civilization: A Cultural History (Los Angeles: J. Paul Getty
Trust, 2000), 379ff.
Housing the Dead 55
Despite the appearance of streets of tombs from the sixth century, the
tombs of the classical Etruscan tradition have a relatively minor en-
gagement with the external display of status. Indeed, the great tumuli
of the seventh century make notable statements in the landscape, as
indeed do some of the rock-cut tombs as at Sovana. But the streets of
Cerveteri-Banditaccia, or Tarquinia-Monterozzi, or Orvieto-Croce-
fisso are not streets in the sense of Roman Grberstraen, major thor-
oughfares where the public pass, but internal paths within a cemetery.
Again, the Volumnii tomb at Perugia may be on a main road, but ex-
ternally nothing survives to mark it as conspicuous; only once you
2.4. Tomb of Volumnii, Perugia, view of atrium and tablinum
(German Archaeological Institute in Rome. Used with permission.)
56 Andrew Wallace-Hadrill
have descended the stairs does it make its impression, as is true of the
famous frescoed tombs of Tarquinia.
30
The internal function of all these tombs, on the other hand, is very
strong. They go to considerable lengths to construct the family group
as a living continuity. Architectural evocations of domestic structures,
hierarchical disposition of multiple burial couches, decorative details,
frequently evoking the theme of banqueting, figured representations
of the dead, and extensive inscriptions underlining their relationships,
all work together to ensure that the living visitor to the chamber will be
impressed, and presumably identify strongly with the family group to
which they by definition belong. These tombs, unlike those of the via
Appia, were not designed for tourists.
The tomb of the Scipios off the via Appia finds itself in an ambiva-
lent role (Figs. 2.57). As one of the rare surviving examples of a burial
place of a noble family from the middle Republic, it has to serve as the
illustration of everything Polybius has to say about the vital import-
ance of display of family continuity in the ritual of a noble Roman
funeral.
31
Certainly other families made impressive tombs, though it
is from the literature that we hear how the tomb of the Marcelli at the
Porta Capena had three statues with the notable inscription, three
Marcelli nine times consul;
32
and Cicero famously attests the impres-
sion made on one leaving the Porta Capena by the tombs of Calati-
nus, the Scipiones, the Servilii and the Metelli (Tusculan Disputations
1.7.13). Yet we know, and Cicero remarks on it, that the Cornelii were
in some sense exceptions in their funerary practice: they continued to
follow the old Roman practice of inhumation when cremation had be-
come the standard Roman practice, the mos Romanus as Tacitus calls
30
For overviews, S. Haynes, Etruscan Civilization, 14271; F. Prayon, Architec-
ture in Etruscan Life and Afterlife (ed. L. Bonfante; Detroit: Wayne State Uni-
versity Press, 1986), 17493.
31
The third-century tomb from the Esquiline with its historical paintings of the
Fabii and Fannii, suggestive of Fabius Pictor, is another possible example:
F. Coarelli, Roma medio repubblicana: aspetti culturali di Roma e del Lazio nei
secoli IV e III a.C. (Rome, 1973), 200208; E. La Rocca, Linguaggio artistico
e ideologia politica a Roma in et repubblicana, in Roma e lItalia: Radices
Imperii (ed. G. Pugliese Carratelli; Milan: Scheiwiller, 1990), 3547, figs 15669;
F. Oriolo, Sepulcrum: Fabii/Fannii, Lexicon topographicum urbis Romae
(ed. M. Steinby, vol. 4; Rome: Quasar, 1999), 288.
32
Ascanius Commentary on Cicero in Pisonem 44; F. Coarelli, Sepulcrum: M.
Claudius Marcellus, Lexicon topographicum urbis Romae vol. 4, 27980.
Housing the Dead 57
it.
33
Their sepulcrum may have been in some ways a deliberate display
of a consciously maintained difference.
As analysed by Coarelli, the tomb goes back in its earliest form to
the early third century (fairly close in time to that of the Volumnii).
34
Cn. Cornelius Scipio Barbatus, consul of 298 B.C.E., is taken to be the
founder, and like Arnth Velimnas, he dominates his family from the
axial position at the center at the back of the tomb (Fig. 2.5). Unlike the
tomb of the Volumnii, this is rather crudely hewn from the tufo. The
main chamber is broadly square in plan, with four tufo pillars left
33
Cicero de legibus 2.56; cf. Pliny Natural Histories 7.187; Tacitus Annals 16.6;
on cremation and inhumation, see Morris, Death-Ritual, 3169.
34
F. Coarelli, Il sepolcro degli Scipioni (1972), reprinted in Revixit Ars. Arte e ideo-
logia a Roma. Dai modelli ellenistici alla tradizione repubblicana (Rome: Quasar,
1996), 179238; cf. with recent bibliography, F. Zevi, Sepulcrum (Corneliorum)
Scipionum, Lexicon topographicum urbis Romae vol. 4, 28185.
2.5. Tomb of the Scipiones, Rome, tomb of Scipio Barbatus
(Filippo Coarelli. Used with permission.)
58 Andrew Wallace-Hadrill
more of less symmetrically disposed (Fig. 2.6). The eventual capacity
of the tomb was 30 couches, which, Coarelli points out, corresponds
with the likely total of members of the family between the early third
and mid-second centuries. This is to say that the interior of the tomb
had already reached capacity in the mid-second century when its fa-
ade was rebuilt, as Coarelli hypothesizes, by Scipio Aemilianus him-
self (Fig. 2.7). Alternatively, one might argue that the interior was re-
modelled at this period to fit the existing burials.
Only nine of the sarcophagi survive, each with an inscription, six in
verse. All the verses are in Saturnians, except the last in the series, that
for the praetor of 139 B.C.E., which is in elegiac couplets. Since Satur-
nians were standard in early Latin poetry (such as Livius Andronicus
and Naevius), and the shift to Greek verse forms (elegiacs and
hexameters) is linked with Ennius, it is particularly intriguing to know
what the role of Ennius was in the reshaping of this tomb. For Cicero
and others attest Enniuss statue there (pro Archia 22), while Livy
(38.56.4) reports that its faade carried statues of three men, Scipio
Africanus, Scipio Asiaticus, and Ennius. The use by the Scipios of
verse epitaphs evidently correlates with their persistent patronage of
poets.
35
We cannot draw comparisons or make contrasts between the Sci-
pionic tomb and its Roman mid-republican rivals, for lack of evidence,
but at least we can make some comparisons with the tomb of the Vol-
umnii. Both are multigenerational family tombs, making much of
the agnatic male descent line. As with the Volumnii, the Scipios have
only one surviving female sarcophagus, that of Paulla Cornelia, wife
of Hispallus, which is slipped behind that of the founder Barbatus.
They stretch over at least five generations, though the total duration of
usage of the tomb was extended by the fact that the Cornelii Lentuli
used it in the early empire for the incineration burials. In so far as it
underlines the importance of the agnatic descent group, it reflects per-
fectly Polybiuss account of the aristocratic funeral and its parade of im-
personated ancestors. Like the Volumnii tomb, it originally contained
portraits in peperino; one was stolen, another has been attributed
to Ennius, but is unlikely to be so; but there are other possible candi-
35
Cf. Ovid Ars Am. 3.409; Valerius Maximus 8.14.1; Pliny Natural Histories 7.114.
On the inscriptions, H. I. Flower, Ancestor Masks and Aristocratic Power in
Roman Culture (Oxford: Clarendon, 1996), 15980.
Housing the Dead 59
2.6. Tomb of the Scipios, Rome, plan
(Filippo Coarelli. Used with permission.)
2.7. Tomb of the Scipios, Rome, reconstruction of facade
(Filippo Coarelli. Used with permission.)
60 Andrew Wallace-Hadrill
dates.
36
Like the Volumnii tomb, care is taken by the inscriptions to
identify the name and relationships of the commemorated, but the
verse inscriptions also allow the res gestae to be celebrated.
It is not easy to imagine that the envisaged audience of the interior of
the Scipio tomb was anyone other than the Scipios themselves. There
were certainly parts of the noble funerary ritual that were designed
to impress the public and consolidate the reputation and political clout
of the family, particularly the parade and public speeches described by
Polybius. But the sarcophagi and their verse inscriptions did not serve,
even if they reflected, this external function. Rather we are in the world
described by Sallust in the preface to the Jugurthine War:
I have often heard that Q. (Fabius) Maximus and P. Scipio, among other leading
figures in our city, used to say that when they looked at the images of their an-
cestors, they felt strongly inspired to virtue (Bellum Iugurthinum 4.5).
The visit to the tomb, like the viewing of ancestral portraits, serves to
admonish and inspire new generations, consolidating the family, con-
structing it as a continuity over time.
37
There was of course an external aspect to the tomb, in the faade
which partly survives, but has to be reconstructed as it is by Coarelli as
a more magnificent example of mid-second-century hellenistic archi-
tecture in order to accommodate the statues described by Cicero and
Livy. Two points may be made here. The first is that the location of
the tomb, set back from the main road and at an angle to it, seems ill-
designed to catch the attention of passersby on the via Appia. Assuming
it is right that Scipio Barbatus established its location, the implication
is that in the early third century this external function of the tomb was
not regarded as primary. Only in a second moment does it acquire an
imposing faade, and by then it is too late to remedy the location. Pos-
sibly the monumental faade raised its visibility sufficiently to be seen
from the junction where the via Appia and the via Latina part, though
that would require the absence of competing structures on the inter-
vening triangle of land. At least Ciceros words suggest that it made
36
Note however that the supposed portrait of Ennius must be one of the earlier
Scipiones,: Coarelli, Il sepolcro degli Scipioni, 23237; L. Giuliani, Bildnis und
Botschaft: hermeutische Untersuchungen zur Bildniskunst der rmischen Republik
(Frankfort: Suhrkamp,1986), 17275.
37
On noble funerary rituals, see J. Bodel, Death on Display: Looking at Roman
Funerals, in The Art of Ancient Spectacle (eds. B. Bergman and C. Kondoleon,
New Haven: Yale University Press, 1999) 25981; Flower, Ancestor Masks, 15980.
Housing the Dead 61
an impression as the traveller left the Porta Capena. The second point
is to observe the misfit between the figures celebrated, as least as
recounted to us by literary sources, and the inhabitants of the tomb
itself. Scipio Africanus was buried at Liternum, and there is some
doubt whether the poet Ennius was buried here or at his native Rudiae.
The faade paraded a rather different view of the Scipiones from the
multigenerational lesson contained within.
It is, as we have already remarked, dangerous to generalise from
a single instance. But it might be reasonable to hypothesize that the
Roman tombs of the early to middle Republic were closer to those of
contemporary Etruria than of the late Republic. The prominent dis-
play of eye-catching funerary monuments along the Appian and other
ways presents itself to us as a feature of the late Republic, from the mid-
second century on. These monuments may appear very traditional,
and in line with the Polybian account of eye-catching noble funerals,
but there is a good chance they are innovative, a new appropriation of
old traditions, a monumentalization of the popular impact of the fu-
nerary ritual.
38
As the emphasis shifts from projecting the continuity of
the household beyond death to display of magnificence, the architec-
tural language of the domestic becomes less important.
Pompeii, Porta Nocera Necropolis:
A Late Republican and Early Imperial Transition?
What characterizes the classic streets of tombs of the last century
B.C.E. and the first C.E., as we meet them in Rome, Pompeii, Sarsina,
Aquileia, and the locations assembled in Rmische Grberstraen, is a
formal diversity. Beyond doubt that diversity reflects a strong impulse
to competitive display. But does it say more than that? The authors
note the failure of their conference to establish a semantics of the di-
verse topologies.
39
But did the variety have a semantic significance at
all? If you chose an altar or a mini-temple, were you showing yourself
more pious? If a tumulus-shaped mausoleum, were you more heroic?
If a triumphal arch, more military? If a palace-faade, more regal? If a
38
Cf. von Hesberg and Zanker, Rmische Grberstraen, 9; von Hesberg, Monu-
menta 3238.
39
Cf. von Hesberg and Zanker, Rmische Grberstraen, 11. The typology is sup-
plied by von Hesberg, Monumenta ch.4.
62 Andrew Wallace-Hadrill
house-faade, more domestic? What is surely most striking about this
formal diversity is its indifference. The inscriptions follow the same
formulae on all types of structure. There is no meaningful distinction
of the burials of magistrates versus freedmen, or of Augustales versus
ordinary freedmen, of men against women, of people of different eth-
nic origin, or even of period. As we move down the extraordinary
clutter of the via Nocera necropolis, we can see the instinct to keep
ringing the changes. We can see too the vast differences between rich
and poor, from Eumachias gigantic semicircular exhedra construction
at one extreme (11 OS), to the fragile little niche tomb of Castricia
Prisca with its perished plaster decoration of garlands, cupids, and
birds (25 OS). But can we say that they are making different statements
about their identity, status, or family affiliations?
For all the paraded difference, these tombs have in fact certain fun-
damental factors in common. The variety of the external aspects con-
ceals a consistent relationship between the external and internal func-
tions of the tomb. Architectural variety in the outward appearance of
the monumentum reflects the common desire to catch the eye of the
passersby and inform them about the identity of those who lie buried.
Hospes, paullisper morare, si non tibi molestum est There are so many
others buried along the road, and the traveller may be in a hurry, but,
please, stranger, tarry a while and hear my tale. Vesonius spells it out
more explicitly than others, but they all have a tale to tell. In contrast
to the homogeneous cube tombs of Cerveteri or Orvieto, which
neither stand on the main road, nor seek to stand out architecturally,
nor contain more than minimal inscriptional evidence about the occu-
pants (maybe a family name incised above the lintel), inscriptions and
portraits reward the stranger who tarries outside the Porta Nocera.
We may be struck by how strong was the instinct to portray on the
exterior of the tomb. Portraits are common in Etruscan cemeteries too,
but they belong, together with the inscriptions, on the inside, within
the family chamber. The typical Etruscan portrait is a full figure reclin-
ing, either at full length on the lid of the sarcophagus, or in abbreviated
form above an ash-urn. Roman funerary portraiture shows as much
variety as the architecture of the tombs: full-length standing figures,
like Vesoniuss group, or Marcus Octavius and Vertia Philumina a few
tombs down at 13 OS, or seated figures, like the couple on the far side
of Eumachias tomb (9 OS), reduced to anonymity by the removal of
their inscription, or simple portrait busts, most strikingly in the tomb
of the Flavii, with its symmetrically arranged niches in two rows, six
Housing the Dead 63
below and eight above, that await the arrival of the Flavii to come
(their death cycle cut off by Vesuvius), but in notably asymmetrical
positions to the right display the chunky busts of Flavius Philoxenus and
Flavia Agathe (Fig. 2.8). The tomb within has two separate chambers;
the western one contained the ollae, identified by labels in carbon, as
Flavius Philoxenus and Flavia Agathe. The external portraits therefore
correspond to the chamber within, in relation to which they are in fact
symmetrically placed, rather than to the monument as a whole.
The Flavii help to underline the vital point: that the external por-
traiture is a mirror of the internal function of the tomb, which consists
in burial chamber, pots containing ashes, and further identifying la-
bels. Thanks to the superb publication of the old excavations at Porta
Nocera by Stefano De Caro and Antonio DAmbrosio, and above all
to their extension of the line of graves in a new excavation,
40
we can re-
store the fragile traces of the internal aspects of these tombs which are
concealed to the modern visitor as to the ancient passerby. The use of
carbon to record the names of the Flavii within reminds us of how the
numerous apparently anonymous burials in columbaria and chamber
tombs must in fact have had labels in evanescent materials, carbon or
red pigment on terracotta, plaster, and surely frequently wood. We ac-
cept far too easily the idea that naming was a privilege for the master
of the house and his close family; it is the use of incised marble that is
the privilege.
What makes this point most forcefully at Pompeii is the use of col-
umellae within the chamber to mark the burial place of the individual.
These headstones, as we have seen with Vesoniuss tomb, evoke the shape
of a head without attempting a portrait. The rear part is rounded, and
in the case of females often sketches out a head of hair. Their front sur-
face is always flat, and serves as a support for an inscription. It is these
columellae, rather than the external inscriptions or portraits, which
provide the evidence for the location of the buried.
The most remarkable example is that of Munatius Faustus and Nae-
voleia Tyche (Fig. 2.9). We have already met this couple on the splen-
didly carved marble altar outside the Herculaneum gate, with their
ship, like Trimalchios, in full sail. But that altar, it would appear, was a
cenotaph, for outside the Porta Nocera, they have another tomb (9 ES).
This is less ostentatious, taking the form of a rectangular enclosure with
40
A. DAmbrosio and S. De Caro, La necropoli di Porta Nocera. Campagna di
scavo 1983, Rmische Grberstraen, 199228.
64 Andrew Wallace-Hadrill
a gabled roof. The fact that it is one of a pair adds to the vague impres-
sion of a row of houses. Externally, a marble inscription on the gable
identifies C. Munatius Faustus Augustal(is) et pagan(us) d(ecreto)
d(ecurionum) sibi et Naevoleiae Tyche coniugi. Internally, there are
eight columellae, recording Munatius Faustus himself (misspelt as
Fausus), one L. Naevoleius Eutrape(lus), taken to be the freedmen of
Naevoleia, but perhaps more plausibly her father or patron, the freed-
woman Munatia Euche, and five slaves, Helpis, Primigenia, Arsinoe,
and Psyche, all of whom died at the age of 3 or less, and Atimetus who
died at 26. The interesting absentee is Naevoleia Tyche; and though
it has been assumed that the new tomb at the Herculaneum gate was a
cenotaph, there is surely a chance that she is buried there, having out-
lived her husband; if indeed she was not still alive at the moment of the
eruption, and still planning to transfer her husbands remains to their
fancy new tomb.
41
I take one further illustration of the inward/outward rhythm of the
Porta Nocera burials from the new excavations, where because of their
freshness, the significance of the columellae is particularly visible.
Tomb F north is formally similar to the tomb of Munatius Faustus: a
rectangular enclosure with a gabled roof. In the gable, the inscription
announces C. Veranius Rufus Q.f. IIvir. It is worth taking note that
this city magistrate, a duumvir, has exactly the same tomb type as an
Augustalis, and that in neither case is the tomb particularly eye-catch-
ing. The dedication is made by his fathers freedwoman and one may
assume his partner: Verania Q.l. Clara optimo patrono sibi et suis. In-
side the low, arched doorway is visible a line of half a dozen columellae.
The central couple, neatly framed by the doorway, are Verania Q.l.
Clara and Q Veranius Q.f. Rufus, though this time his office is given
not as duumvir but aedile, a usage which is paralleled in Pompeii (that
is to say, the aediles described themselves, being a pair, as duumvirs, so
creating a constructive confusion of their precise rank). The other four
columellae are without inscriptions, or as I prefer to put it, without
surviving legible labels.
We have seen in the cases of Veranius and Verania, Munatius and
Naevoleia, Flavius Philoxeus and Flavia Agathe, and Vesonius Phil-
eros, that there is a close relationship between the external and internal
aspects of the tomb. The essential feature of the tomb is in fact the en-
closure or chamber: it is here that the family members are assembled,
41
Kockel, Die Grabbauten vor dem Herkulanertor, 107.
Housing the Dead 65
2.8. Tomb of Flavius Philoxenus and Flavia Agathe, Porta Nocera, Pompeii
(German Archaeological Institute. Used with permission.)
2.9. Tomb of Munatius Faustus and Naevoleia Tyche, Porta Nocera, Pompeii
(Soprintendenza Archeologica di Pompei. Used with permission.)
66 Andrew Wallace-Hadrill
and here too presumably that the ritual activities of the surviving
family focused on the Parentalia and the days of roses and violets. But
in comparison to the Etruscan tombs, they have been turned inside
out. The columellae represent minimalist markers of the presence of
the deceased; the portraits and detailed inscriptions are displayed for
public consumption on the outside. The phenomenon is so marked
that we run the risk of noticing only the external aspects and therefore
the external function: we think of the Roman tomb as a monumentum
to boast the status of the dead to the outside world. In truth, this is
only the outwards face of a structure that still has a critical internal
function in reconstituting the family.
By my argument, then, these tombs are indeed parallel to the houses
of the living. One notable feature of the Pompeian domus is that the
external function (the desire to impress the visitor from outside) is so
strong that it almost overwhelms the internal functions of a family
structure. Women and children prove relatively elusive within the
house. But of course the internal function is still there. The link be-
tween tombs and houses lies not in their typology (if the tomb of Mu-
natius Faustus at the Porta Nocera is more house-like, his altar-tomb
at the Porta Ercolano is less so), but in their management of the rela-
tionship of the external and internal function. Where the tombs of this
late republican/early imperial period seem to be historically distinctive
is in the extraordinary degree of importance attributed to the external
function; and that, by no coincidence, is also true of the treatment of
domestic space.
Valerius Herma and the High Imperial Necropolis
My third and last moment is the mid-second century C.E. floruit of the
street of the tombs beneath the Vatican (Fig. 2.10). That they were in
some sense representative is suggested by the close typological par-
allels with the Isola Sacra necropolis with its tight chronological range
from Trajanic to Severan.
42
True, we are looking at burials overwhel-
mingly of freedmen, and not of the urban elite. Equally, the poor
are under-represented. That the vision is partial is brought out by the
42
I. Baldassare, La necropoli della Isola Sacra (Porto), Rmische Grberstraen,
125138.
Housing the Dead 67
equally important necropolis of the Vatican autopark. Thanks to Mar-
gareta Steinbys careful publication, we can recover the full clutter of
an ancient graveyard, where the neat rectangular structures of chamber
tombs are surrounded by a dense spread of simpler burials, in urns, or
capuchin tents of tiles, or simple wooden boxes that have rotted away.
43
That is a vital reminder that brick-built chamber tombs were no uni-
versal norm, but a specific effort to group the dead together.
The brick facades of the St Peters necropolis, or the Isola Sacra,
have often put visitors in mind of rows of houses.
44
The analogy, as
we have seen, has its limits; but coming to this material from the
Pompeian Porta Nocera, what must surely strike us is a sense of uni-
formity. It is because modern houses often come in rows of uniform
brick structures that the image seems irrepressible. Although, as Eck
has shown, there is a considerable range in size of recorded plots that
follow the formula, so many feet in fronte, so many in agro, there is a
notable cluster around a uniform size of around 1012 feet wide and
as many deep.
45
To call these frontages homogeneous is not to say
they are without individuality: the occasional scenes at the Isola
Sacra representing the trade activities of the deceased are especially
43
E. M. Steinby, C. Coletti, M. B. Carre, and M. T. Cipriano, La necropoli della via
Triumphalis: il tratto sotto lautoparco Vaticano (Atti della Pontificia Accademia
Romana di Archeologia, ser. 3. Memorie, vol. 17; Rome: Quasar, 2003).
44
Toynbee and Ward-Perkins, Shrine of St. Peter, 132ff.; H. von Hesberg, Planung
und Ausgestaltung der Nekropolen Roms im 2. Jh. n. Chr, Rmische Grber-
straen, 4360.
45
W. Eck, Rmische Grabinschriften. Aussageabsicht und Aussagefhigkeit im
funerren Kontext, Rmische Grberstraen, 6184.
2.10. Vatican (St. Peters necropolis), plan and section.
(German Archaeological Institute. Used with permission.)
68 Andrew Wallace-Hadrill
effective in this sense. Nevertheless, compared to the competitive di-
versity of the Pompeian streets, there seems little attempt to catch the
eye of the passerby. The one pyramid at the Isola Sacra is not, like
that of Cestius in Rome, a conspicuous monumentum aere perennius,
but a modest miniature.
In a word, there seems to have been another flip-round in the
relative importance of the external and internal function. Valerius
Hermas tomb is a powerful illustration because of the sheer richness
of its internal decoration (Figs. 2.1115).
46
The magnificent stucco
work enriches an internal architecture elegantly articulated with
niches, and ranges statues of the gods and philosophers, and of Val-
erius Herma, his wife, daughter, son, and perhaps patron. As in Pom-
peii, there is a close relationship between the presentation outside, in
the form of an inscription, and that inside. From outside, we meet the
family:
C Valerius Herma fecit et
Flaviae T.f. Olympiadi coiugi et
Valeriae Maximae filiae et C Valerio
Olympiano filio et suis libertis
libertabusque posterisque eorum.
The classic nuclear family grouping is extended, just as at Pompeii and
across Italy, by the generic grouping of freedmen and freedwomen and
their descendants.
But it is only inside that we can get a grip on the family. Flavius Her-
ma presents himself repeatedly, almost obsessively.
47
The marble panel
of his sarcophagus reintroduces himself and his wife:
C Valerius Herma dum
vivo mihi feci et
Flaviae T.f. coiugi.
46
The definitive publication is H. Mielsch and H. von Hesberg, Die heidnische
Nekropole unter St. Peter in Rom: die Mausoleen AD (Atti della Pontificia Ac-
cademia Romana di Archeologia, ser. 3, Memorie, vol. 16, 2. Roma: Lerma di
Bretschneider). See also Toynbee and Ward-Perkins, Shrine of St. Peter, 8287;
Eck, Rmische Grabinschriften, 7173. I am indebted to the forthcoming
paper by Regina Gee, Being Greek in Rome: Identity, Memory, and Status in
the Tomb of Gaius Valerius Herma.
47
Eck, Rmische Grabinschriften, 7173 for the inscriptions.
Housing the Dead 69
2.11. Tomb of Valerius Herma, necropolis of the Vatican, St. Peters, plan.
(German Archaeological Institute. Used with permission.)
70 Andrew Wallace-Hadrill
2
.
1
2
.
T
o
m
b

o
f

V
a
l
e
r
i
u
s

H
e
r
m
a
,

n
e
c
r
o
p
o
l
i
s

o
f

t
h
e

V
a
t
i
c
a
n
,

S
t
.

P
e
t
e
r

s
,

c
e
n
t
r
a
l

s
e
c
t
i
o
n

w
i
t
h

s
a
r
c
o
p
h
a
g
u
s

o
f

H
e
r
m
a
.
(
G
e
r
m
a
n

A
r
c
h
a
e
o
l
o
g
i
c
a
l

I
n
s
t
i
t
u
t
e
.

U
s
e
d

w
i
t
h

p
e
r
m
i
s
s
i
o
n
.
)
Housing the Dead 71
b
2.13. Tomb of Valerius Herma,
necropolis of the Vatican,
St. Peters, marble portrait of
Herma (a), death mask (b)
(German Archaeological
Institute in Rome. Used with
permission.)
a
72 Andrew Wallace-Hadrill
2.14. Tomb of Valerius Herma, necropolis of the Vatican, St. Peters, gilded
stucco portrait of son Valerius Olympianus (a) and death mask (b). (German
Archaeological Institute in Rome. Used with permission.)
2.15. Tomb of Valerius
Herma, necropolis of the
Vatican, St. Peters, death
mask of infant. (German
Archaeological Institute in
Rome. Used with permission.)
a b
Housing the Dead 73
The lettering is elegant, the grammar a touch uncertain (dum vivo
combines the dative of vivo mihi feci with the fragmentary clause
dum vivus eram). He introduces himself again on his sons sar-
cophagus:
C Valerio Olympiano qui vixit
annis IIII menses V dies XIII
C Valerius Herma pater.
The early loss of his four-year-old son could well be the occasion of
his building of a tomb dum vivo. But it could equally have been on
the loss of his 12-year-old daughter, Valeria Maxima, whose titulus can
be reconstructed on the model of her brothers:
[Valeriae] C.f. M[aximae
quae vixit an]nis XII [mens.?
dieb. ? C Valerius Herma pater.]
Since both dead children figure on the titulus at the entrance, we can
assume both children died before the tombs construction.
Valerius also marked the loss of a foster-child of 3 years old, Val-
erius Asiaticus:
C Valerio Asiatico
alumno C Valerius Herma
qui vixit an. III m. XI d. III.
Asiaticus must have become alumnus of Herma on the death of his
mother, Asia, who is commemorated by her husband, Valerius Prin-
ceps, presumably a freedman or fellow freedman of Herma:
C Valerius Princeps [Va]
leriae Asiae libertae i[ncom]
parabili quae vix[it ann ]
mecum [ann ]
Valeriuss nuclear family thus extends through the links of manu-
mission and fostership. But it also extends to another alumnus, an
8 year old from a different family:
C Appaieni Ca-
ti qui vix. ann. VIII
m. X d. XXVIII alumno
dulc.cui locum optu-
lit C Val. Herma in
frontf (sic) ped. V sarcofag-
go terra deposito.
74 Andrew Wallace-Hadrill
Finally, the family is extended to the wife of a freedman, Valerius
Eutychas, though she was apparently a slave:
Dynateni C Valerius Eutychas
coiugi benemerenti fecit
permissu C Valeri
Hermaes patroni optimi.
The presence of Herma himself is felt massively in this epigraphic
ensemble. So it is in the stucco decoration of the tomb. The figure of
the god which occupies the central niche opposite the entrance is too
damaged for certain identification, but Mielsch feels confident in
seeing in him Hermes (rather than Guarduccis Apollo/Harpocrates).
Even more striking, columns are replaced in the decoration by the
squared pilasters of Herms; of the original 23 Herms, 10 heads survive.
Since these are a non-standard decorative form for a tomb, one infers
that Herma is playing a deliberate game with his name.
His self-representation goes far beyond punning. In the bearded fig-
ure in sacrificial pose on the west wall, Mielsch identifies the portrait
of Valerius Herma (Guarducci had seen this implausibly as the
emperor Marcus), in the female figures that flank him, his daughter
Valeria Maxima, his wife Flavia Olympias. An older male figure is
identified as his patron, C. Valerius. But these stucco representations
(assuming that they do indeed consist in the family group) are backed
up by two well-carved marble portraits, of a bearded man and an idea-
lized woman wearing the turban-like hairnet that is typical of the sec-
ond century. They are identified as Valerius Herma (Fig. 2.13a) and
Flavia Olympias.
Herma might be thought to have done well in terms of self-repro-
duction. But he is not finished. His wife seems to be subject of a further
portrait, this time in stucco, looking older and more tight-lipped, but
wearing the same turban-like hairnet. It is strange that these remark-
able portraits have not attracted more attention from those concerned
with realism and idealism in Roman portraiture. The stucco por-
trait series continues. A young woman with wavy hair, and a young boy
with short-cut hair with a quiff at the back, are taken to be portraits of
the prematurely dead Valeria Maxima and Valerius Olympianus. In
the latter case, the portrait is gilded, indicating an especial importance.
Even so, the portrait gallery is not complete. Three startling death
masks complete the collection. One (surviving as a mould) shows a
bearded figure so close to the bearded portrait as to make the choice of
Housing the Dead 75
Valerius Herma seem inescapable (Fig. 2.13b). The final two, even
more powerful, show a young boy with long eyelashes, hard not to take
as the 4-year-old Valerius Olympianus (Fig. 2.14), and an even younger
child, not identified by Mielsch, though the fosterchild Valerius Asiati-
cus might be a candidate (Fig. 2.15).
After this extraordinary catalogue of self-representation, let us take
stock. Without pushing any of the individual identifications too hard,
it is fairly evident that Valerius Herma projects himself inside his tomb
with an insistence that puts even a Trimalchio to shame. From outside
we see him in the titulus; inside we see him in his own sarcophagus
inscription, and in those of his many dependents. His face looks down
on us from the stucco decoration, from a marble bust, and even from
a death mask. He ensures that his wife is also represented multiply
within, along with his children and possibly alumni. His theophoric
name seems to play even into the decorative scheme of herms.
But while we cannot mistake the urgency of his projection of himself
and his family and dependents, it is only from within the tomb that we
can pick up the message. Unlike Trimalchio, he is not interested in
making an ostentatious statement about himself to the world: his trim
brick faade speaks respectability but not vanity. It is for the benefit of
himself and his close circle that he makes his not inconsiderable finan-
cial investment. We have seen Toynbee and Ward-Perkins comment
with surprise on the richness of the internal decoration indeed, the
quality is quite remarkable. But we may find difficulty with their con-
cept that this was done for the benefit of the dead, to make them feel
at home in their eternal abode. It is surely done with an eye to the
living, that is to say Valerius himself, who as he lost members of his
family spent perhaps even more of his life than he would have wel-
comed in the tomb, amid the cycle of regular annual rituals; for the
benefit of the survivors in his circle, who wished to remember their
loved ones; and for the benefit of the future generations which Herma
surely hoped would continue to remember and revere him. That is to
say, the functions of the tomb are predominantly internal; the external
function is present but subsidiary.
This is perhaps the place to add a comment about portraiture in
Roman funerary art. There is an uncomfortable misfit between archaeo-
logical reality on the one hand, and on the other the well-known
accounts by Polybius of masks and impersonation in the Roman funer-
ary ritual, and by Pliny of the atrium as a location for ancestral images
with tituli linked by red thread. No archaeological evidence instanti-
76 Andrew Wallace-Hadrill
ates these descriptions: there is no known example of any sort of mask
that might be used for impersonation at a funeral, nor of an atrium
with a collection of portraits such as might be linked by red thread.
48
These passages are so much cited because they are assumed to provide
the key to what we actually do find: numerous portraits in connection
with tombs. If the Scipio tomb originally included portraits, as we have
suggested, it might make the tomb an evocation of the Plinian atrium.
The best example of a collection of ancestral portraits are those from
the tomb of the Licinii in Rome of the early first century C.E., which
have made their way, including the famous portrait of Pompey, to the
Ny Carlsberg Glyptotek always supposing the finds are genuine, not
a nineteenth-century fake assembled to order for the market by those
who well knew their Pliny.
49
The overwhelming assumption, apparently supported by Polybius
and Pliny, is that Roman portraiture was all about status, that is to say
about its external function, or advertising the image of the portrayed
to the outside world. That is certainly borne out by the portraits on the
tombs of Pompeii, not to speak of those on the via Appia and other
streets of Rome, including the serried ranks of freemen solemnly
framed in the windows from which they look out on the world.
50
But
portraits, like houses and tombs, have an internal as well as an external
function. They are a mechanism whereby a family represents itself to
itself and constructs its identity. That is also what Polybius and Pliny
are saying. A tomb like that of Valerius Herma shows this function at
work.
Conclusion
In sum, my argument is that the analogy between tomb and house in
Roman Italy is perhaps stronger than we suspected. I remain sceptical
about the importance of the formal architectural evocations. These are
present, but always partial. It seems to me risky to reconstruct the
image of the Etruscan house on the basis of the Etruscan tomb, how-
48
Flower, Ancestor Masks, 3640.
49
P. Kragelund and M. Moltesen, The Licinian Tomb: Fact or Fiction? (Copen-
hagen: Ny Carlsberg Glyptotek, 2003).
50
D. Kleiner, Roman Group Portraiture: The Funerary Reliefs of the Late Republic
and Early Empire (New York: Garland, 1977).
Housing the Dead 77
ever plausible the links. But it is a game of allusion where one needs to
know both sides of the equation to see how it works. On the other
hand, I have argued that the tomb is a conscious extension of the two-
fold function of the house, internally to articulate the household, ex-
ternally to present the household to the world. The tomb, like the
house, is where two worlds intersect, the world of the family or house-
hold, and the world beyond. In providing a home in which the dead
are reassembled with the family of the living, the tomb also invites the
world beyond, the passing stranger, to take note, shed a tear, or gasp
with astonishment.
But the most interesting conclusion, I feel, is one about which we
must be tentative without a more extensive examination of the evi-
dence. It is that over time, there are significant shifts in the balance
between this internal and external function.
51
In the mid Republic and
the high Empire, I have argued, the internal function is dominant.
Tombs are about representing the family or household to itself. The
main difference is that the mid-Republican family, like the Etruscan, is
one with significant duration over time, across several generations,
while that of the Empire is short-lived, and recruits the servile house-
hold to bolster its numbers. The high imperial model is by no means
a return to the mid-republican one, but a new one suitable to the
changed society of the empire. Hence I have deliberately characterised
the late Republic/early Empire as a transitional period, to counter the
impression it always creates as the classic and timeless expression of
the true Roman way. The novelty lies in the vigorous and competitive
interest in self-representation to the world outside; though I have tried
to underline that the internal function persists and should never be
overlooked. The tomb, like the house, enables this constant dialogue.
51
Cf. the similar conclusions of von Hesberg, Monumenta, 27779.
78 Andrew Wallace-Hadrill
Commemorating the Dead 79
Susan T. Stevens
Chapter 3
Commemorating the Dead
in the Communal Cemeteries of Carthage
1
Since Lantiers 1922 landmark article Notes de topographie carthagi-
noise, burials at Carthage have been discussed mostly in the context
of the citys growth and transformation, reflections, in part, of changes
in attitude toward the dead in Late Antiquity wrought by the beliefs
and practices of Christianity.
2
As a result, much emphasis has been
given to the location of burials during the period and the typology of
tombs, topics that tend to mask the distinctiveness of individual sites.
3
Thus, fifth- through seventh-century burials at Carthage tend to be
treated in one of two mutually exclusive categories: Christian, that is
burials in and around Christian basilicas and other cult buildings, or
urban, individual tombs or small groups of tombs, not specifically
Christian, in and around buildings of the city or in cemeteries on its
1
I am grateful to Laurie Brink and my other colleagues in this volume for dis-
cussions of commemoration of the dead that have added a dimension to my
study of cemeteries in a time and place where commemoration is often difficult
to recognize. An early version of this paper was delivered as part of a panel, Ur-
banism in North Africa: Beyond the Forum, at the Archaeological Institute of
America Annual Meeting, Montreal, January 2006.
2
Raymond Lantier, Notes de topographie carthaginoise. Cimitires romains
et chrtiens de Carthage, CRAI (1922): 2228; Henry R. Hurst, The Late
Roman-Byzantine Defences of Carthage, in Excavations at Carthage: The Brit-
ish Mission Vol.1.1: The Avenue du Prsident Habib Bourguiba, Salammbo: The
Site and Finds other than Pottery (eds. Henry R. Hurst and Simon P. Roskams;
Sheffield: The British Academy, 1984), 3141; Liliane Ennabli, Carthage:
Une mtropole chrtienne du 4e la fin du 7e sicle. (tudes dantiquits afri-
caines; Paris: CNRS ditions, 1997).
3
A recent welcome exception is Naomi J. Norman, Death and Burial of Roman
Children: The Case of the Yasmina Cemetery at Carthage: Part 1: Setting the
Stage, Mortality 7.3 (2002): 30223.
80 Susan T. Stevens
periphery.
4
These models leave out of discussion burial sites that do not
fit comfortably. One example is the cemetery that includes mass graves
at Falbe Point 90 on the north coastal periphery of Carthage. The di-
versity of practice at this site and others warns that traditional models
of urban and Christian burial, developed from coarse-grained evi-
dence when little specific archaeological data was available, may have
been too broadly applied. The dichotomy implied by these models tends
to obscure a shared ideology of communal burial in the cemeteries of
this period.
5
The overall aim of this paper is to explore the internal logic of re-
cently excavated and published cemeteries of the fifth through seventh
centuries by focusing, initially, on five cemeteries: two in and around
the basilicas of Bir el Knissia and Bir Ftouha, two associated with the
city wall, the Theodosian Wall and Circus cemeteries, and the seem-
ingly anomalous burial site at Falbe Point 90.
6
Each of these five sites
4
For the first category see Nol Duval, Linhumation privilgie en Tunisie et en
Tripolitaine, in Linhumation privilgie du 4e au 7e sicle en Occident, (eds. Yvette
Duval and Jean-Charles Picard: Paris, De Boccard, 1986), 1342 and Les ncro-
poles chrtiennes dAfrique du Nord, in Monuments funraires, institutions
autochtones en Afrique du Nord antique et mdivale, VIe colloque international sur
lhistoire et larchologie de lAfrique du Nord, (ed. Pol Trousset: ditions CTHS:
Paris, 1995), 187205 which set Carthage in the context of other, better known
North African Christian sites. For the second category see Susan T. Stevens, Sp-
ultures tardives intra-muros Carthage, in Monuments funraires, 20717, id.,
Transitional Neighborhoods and Suburban Frontiers in Late- and Post-Roman
Carthage, in Shifting Frontiers in Late Antiquity, (eds. Ralph W. Mathisen and
Hagith Sivan; Aldershot: Ashgate, 1996), 187200, and more recently, Anna
Leone,Linumazione in spazio urbano Cartagine tra 5 e 7 secolo D.C., An-
tiquit Tardive 10 (2002): 23348 which is especially useful for fig. 5, an updated
plan of the later city, id., Changing Urban Landscapes: Burials in North African
Cities from the Late Antique to Byzantine periods, in Mortuary Landscapes in
North Africa (eds. David. Stone and Lea M. Stirling; Toronto: University of To-
ronto Press, 2007), 164203. The same author generously shared the manuscript of
a book in progress, Transition Revisited: Decline and Ordered Evolution in North
African Towns (Zeugitania, Byzacena, Tripolitania) from Late Antiquity to the
Arab conquest that sets Carthage in the context of other North African cities.
5
G. Cantino Wataghin, The Ideology of Urban Burials,in The Idea and Ideal of
the Town between Late Antiquity and the Early Middle Ages, (eds. Gian P. Brogiolo
and Bryan Ward-Perkins; Leiden: Brill, 1999), 14763.
6
Following Lantier, I use the word cemetery in its modern sense of collective burial.
For the debate over the words ancient origins see ric Rebillard, KOIMHTH-
PION et COEMETERIUM: tombe, tombe sainte, ncropole, MEFRA 105.2
(1993): 9751001.
Commemorating the Dead 81
has fixed limits in space and time, clear principles of spatial arrange-
ment, and a burial koine in the consistency of tomb types, markers, and
gifts. These features reflect a conformity to tradition at each site that
can be taken as evidence of a distinctive collective identity, even if the
specific nature of the community cannot be ascertained.
7
The details of
each cemetery reveal the importance of individual and group identity
and landmarks of social status within each community, as well as be-
tween communities. Together, the sites, with a handful of other re-
cently excavated cemeteries of the period, suggest a more nuanced pic-
ture of the fifth- through seventh-century cemeteries at Carthage.
Rather than being a continuous zone of cemeteries,
8
the urban pe-
riphery of Carthage may have been a fluid landscape in which distinct
cemeteries within specific enclosures or clustered around landmarks
came and went like alluvial islands.
The basilica at Bir el Knissia was built just outside a gate in the late
Roman city wall on Kardo 5 east in the late fifth-early sixth century. It
is known from excavations in 192223 by Delattre and in 199092 by a
team with access to Delattres field notes and excavation plan.
9
Indi-
cations that the basilica was a cemetery church by design are its original
beaten earth floor, an apsed structure attached to its west wall, and a
presumed atrium at the north, all used for burials.
10
The burial function
of the complex was expanded in the later sixth and seventh centuries:
porticoed courtyards were added to the east and west long sides of the
basilica in the Justinianic period (540566) and an east courtyard was
attached to the east wall of the basilica in the post-Justinianic period
(575+), a change of plan that may have linked the original basilica to a
large symmetrical building to its northeast, perhaps a second basilica.
11
7
Ann Marie Yasin, Funerary Monuments and Collective Identity: From Roman
Family to Christian Community, Art Bulletin 87.3 (September 2005): 43357,
esp. 44245.
8
Ennabli, Carthage, 56.
9
Alfred-Louis Delattre, Fouilles sur lemplacement dune basilique prs de
Douar-ech-Chott Carthage, CRAI (1922): 30207, and La basilique de
Bir-el-Knissia Carthage, CRAI (1923): 44951; Susan T. Stevens, Bir el Knis-
sia at Carthage: A Rediscovered Cemetery Church: Report no. 1 (JRASup. 7;
Ann Arbor, 1993).
10
As at Demna, Stif, and Uppenna, see Duval Ncropoles chrtiennes, 203.
11
Stevens, Bir el Knissia, 3038. In his review, Nel Duval suggests that the sym-
metrical building might be a second basilica, La basilique de Bir el Knissia
Carthage: une fouille du Pre Delattre redcouverte et rtudie, Antiquit Tar-
dive 3 (1995): 295.
82 Susan T. Stevens
How far east the basilicas burial area extended is unknown. Delattres
excavation plan suggests that its structures extended to the north to
within 10 meters of the rural extension of Kardo 5 east and to the west
perhaps as far as a perpendicular rural road that may have confined
the southwest annex. At the southern end of the complex, structures
and burials appear to have extended as far as 15 meters southeast of
the southwest corner of the basilica, an area that bears a remarkable
similarity to the burial area that lay immediately outside the apse and
southeast corner of the basilica. At the southern extent of the burial
complex were one robbed burial, fragments of three Christian tomb in-
scriptions, and disarticulated human bone representing a minimum
number of three individuals. The burial lay perpendicular to two phases
of a SW-NE-aligned wall, and probably lay inside a structure, the align-
ment of which, while unlike that of the basilica proper, was similar to
that of the southwest annex of the basilica. The burials associated with
the basilica did not extend as far south as sondages 1 and 6 where typi-
cal domestic contexts roughly contemporary in date with the basilica
marked the edge of the low plateau on which the complex was built.
12
A Roman necropolis just outside the gate of the city has been con-
sidered a kind of predecessor of the cemetery in and around the Chris-
tian basilica, although some 50 meters separate the Roman tombs
from Bir el Knissias symmetrical building.
13
Indeed, the paucity of
Roman epitaphs at Bir el Knissia seems to argue against any real con-
tinuity: only nine of 89 small fragments of funerary inscription found
in the complex are arguably Roman, most of which appear to have
been reused as paving or cut for frames for other tomb markers.
Only one retained evidence of attachment to the masonry of a Roman
monument.
14
Seventy-four tombs are known from inside the basilica and its an-
nexes. Forty-seven were shown on the 19221923 excavation plan or in
detailed sketches in the diary, though Delattres passing mentions
of spultures indicate that he encountered many more burials than he
investigated or recorded. The 1990 excavations encountered the traces
of another 27 tombs, though the minimum number of 55 individuals
represented among the disarticulated bone suggests many more within
12
Stevens, Bir el Knissia, 4, fig. 3 and 6771.
13
Lantier,Notes de topographie, fig. 1, no. 10; Duval, Ncropoles chrtiennes,
193.
14
L. Ennabli, Inscriptions de Bir el Knissia, in Stevens, Bir el Knissia, 25788.
Commemorating the Dead 83
the limited area excavation.
15
Although the excavated tombs represent
only a sampling of the cemetery, they appeared in every area of the
complex, including some burials originally placed outside the basilica
that were later incorporated into its annexes. The distribution of funer-
ary inscriptions from 1990 suggests that tombs were most concen-
trated in the symmetrical building and in the basilica proper.
16
The tombs probably belong to all phases of the basilica, from the
late fifth to the late seventh century, though very few can be more spe-
cifically dated. In a pattern long recognized for churches, the tombs
were arranged parallel to the walls of the complex.
17
The vast majority
of the tombs were aligned either with the NNW-SSE long walls or
SSW-NNE short walls of the basilica, with five tombs in the southwest
annex area aligned roughly WSW-ENE conforming to a later cross
wall that was not perpendicular to the basilicas long axis. Among the
tombs where an orientation of the body was clear, no preference was
obvious. Limited evidence from the east aisle suggests that the burials
inside the basilica may have belonged to two sequences, arranged ver-
tically. The relatively deep-shafted NNW-SSE graves cut from the
beaten earth floor, including Delattres best-preserved burials, seem to
date to the late fifth to mid-sixth century, while the comparatively shal-
low graves aligned SSW-NNE and cut from the level of the mosaic
pavement belonged to the lates sixth and seventh centuries.
18
The vast majority of graves were formae, shafts cut into the pave-
ments of the complex at the bottom of which the deceased were placed
simply in the earth or contained in cists of stone, masonry, or terra-
cotta.
19
The tops of many shafts were marked with an inscription in
mosaic or stone, most frequently commemorating a single individual.
However, four inscriptions from this site commemorated two and four
individuals.
20
15
Stevens, Bir el Knissia, after 24, 38, 47, 1024, 12127, 14450, 18386; Cherie
K. Walth and Laura J. Miller, Burials and Disarticulated Human Bone (1990),
in Stevens, Bir el Knissia, 19192.
16
Ennabli, Inscriptions, 25788 catalogues 22 inscriptions from the NE annex
and 14 from inside the basilica.
17
Duval, Ncropoles chrtiennes, 206.
18
Stevens, Bir el Knissia, 3046.
19
Three amphora burials were also recorded: outside the apse aligned N-S with the
basilica, oriented S-N just outside the chancel entrance and oriented S-N against
the east wall of the basilica.
20
Stevens, Bir el Knissia, 35; Ennabli,Inscriptions, 272.
84 Susan T. Stevens
Four group tombs stand out in the Bir el Knissia cemetery. These
consisted of three to nine prefabricated tombs built side by side all
at one time in a single masonry structure. Two group tombs lay inside
the basilica, a structure with nine graves in the apse, and another with
four tombs associated with a small cistern in the floor of the chancel.
The other two group tombs were located in the western annexes of the
basilica: eight tombs in a hypogeum under the west portico floor, three
in the southwest annex. How or if these groups or their individual
members were marked at floor level is not attested. While the group
tombs conform to the larger community in that they are individual
formae, the fact that they were built together all at once instead of being
dug ad hoc defines them as a group apart from other tombs, asserting a
smaller group identity within the larger cemetery community. However,
because group tombs located inside the basilica were in areas ordinarily
restricted to the clergy, clergy may have been the small community ex-
pressed here. Similarly, clergy were distinguished from laymen in the
cemetery churches at Demna, Stif, and Hadra 1 by epitaphs that in-
cluded the name and titles of clergymen, but only the names of
laymen.
21
The other two group tombs at Bir el Knissia may have been
family monuments, and therefore like the numerous small masonry
tomb groups (for two to five individuals) and even tomb monuments
found in the large burial enclosure southeast of the basilica of Midfa,
and outside that basilicas north corner.
22
Their location in com-
paratively informal and less privileged spaces of the basilica may have
enabled families to assert their smaller group identity.
23
The closest par-
allel to the hypogeum at Bir el Knissia may be the late-fourth- or early-
fifth-century hypogeum of Flavius Valens found by Delattre against the
southwest corner of the basilica at Damous el Karita.
24
21
Yasin,Funerary Monuments, 447.
22
Liliane Ennabli, Les inscriptions funraires chrtiennes II. La basilique de Midfa
(Tunis and Rome: Institut national darcheologie et dart and cole franaise de
Rome, 1982), 11, fig. 5.
23
The southwest annex at Bir el Knissia also included a highly unusual funerary
monument for a single individual, a caisson shown in F4 of Delattres plan and
p. 4 of his carnet, Stevens, Bir el Knissia, 40. Duval, La basilique de Bir el Knis-
sia, 291, interprets this feature as the basin of an earlier Roman house.
24
Heimo Dolenz, Damous el Karita: Die sterreichisch-tunesischen Ausgrabungen
der Jahre 1996 und 1997 im Saalbau und der Memoria des Pilgerheiligtums Da-
mous el Karita in Karthago (sterreichisches Archologisches Institut Sonder-
schriften 35; Vienna, 2001), Abb. 54, Beil.1/10.
Commemorating the Dead 85
Delattre made no mention of any grave goods, and none were found
among the surviving burials of the 1990 excavation, but the profound
disturbance of the basilica by grave robbers suggests that enough
tombs were furnished to make the practice profitable. Indeed, uninten-
tional disturbance of tombs would have been unlikely because many
formae were marked at floor level, a practice attested at Bir el Knissia
by 89 recorded epitaphs. In fact, the elevation of the new mosaic floor
some 0.5 meters above the original beaten earth surface sealed the
layer of early tombs. This strategy to accommodate a new layer of
burials cut from the new floor, without disturbing earlier ones, sug-
gests the value placed on the integrity of tombs.
The inscriptions were all in Latin except for two in Greek; most were
on stone (marble, limestone, kadel, in descending order) with at least
eight in mosaic. The epitaphs have a standard format: a single name
followed by some or all of the fomula fidelis in pace vixit annis
depositus with a date and a limited range of familiar iconography. In
this respect, the Bir el Knissia community conforms to the epigraphic
traditions for this period in Carthage known from other basilica cem-
eteries.
25
The cognomina, with the exception of Siricia (cat. no. 6), are
well-attested, though no names of Germanic origin are recorded as
they were at nearby Bir el Knissia 2 and other basilica cemeteries. This
is surprising given the origin of the basilica in the late Vandal period,
and it may be a clue to the ethnic identity and religious persuasion
of the basilica community: Roman and Catholic as opposed to Vandal
and Arian.
26
Little other information can be gleaned about the
cemetery population. The few preserved burials and disarticulated
bone from the 1990 excavations suggested that immature individuals
(16 of 55) were under-represented, perhaps because they were largely
excluded from the area excavated inside the basilica.
27
This pattern is
discernable on the 19221923 excavation plan: the standardized adult-
sized shafts predominate, with only three very small tombs shown out-
side the basilica proper, probably representing the graves of children,
to which may be added at least one of Delattres two amphora burials,
25
Ennabli, Inscriptions, 28788 and id., Les inscriptions funraires chrtiennes de
la basilique dite de Sainte-Monique Carthage (Tunis and Rome: Institut
national darchologie et dart and cole franaise de Rome, 1975), 5969,
7782, 87.
26
Stevens, Bir el Knissia, 16.
27
Walth and Miller, Burials,19293.
86 Susan T. Stevens
the small globular amphora shown outside the east wall of the basilica.
A similar imbalance is better documented at the so-called basilica of
Sainte Monique where only 28 of 86 individuals identified by age on
their epitaphs are classified as immature.
28
Clearly, the defining feature of the Bir el Knissia community as a
whole was its association with the structures of the basilica complex,
a living monument that commemorated members of the community.
As a communal monument the Bir el Knissia basilica was small and
rather simple both in plan and adornment by comparison with other
burial churches at Carthage. On the other hand, the basilica is unusual
in being a new foundation in the late fifth or early sixth century with
architectural ornament specially made for this structure rather than
being composed of spolia.
29
The probable density of burials at Bir el Knissia, in and around a ba-
silica that expanded both horizontally and vertically to accommodate
more burials, suggests a large and not particularly exclusive community.
Within this community the abundance of inscriptions and the concern
for the integrity of tombs attest to the value placed on the individual.
The prevalence of forma-type tombs and the standard repertory of
formulae, iconography, and nomenclature reflected on their markers
suggest a deliberate conformity to tradition.
30
Hierarchy within the Bir
el Knissia cemetery is also expressed by the privilege attached to indi-
vidual graves and tomb groups in the chancel and apse as opposed to
burials in the rest of the basilica, to graves inside the basilica as com-
pared with those in its annexes, to those in pre-fabricated group tombs
compared to individual formae placed ad hoc.
Beyond the notion of communal commemoration, what attracted
the community to Bir el Knissia at the outset is not clear. While the
answer usually given for suburban churches is the prospect of burial
ad sanctos, no specific evidence at Bir el Knissia supports its origin as
a martyrial church. In fact, of the suburban basilicas at Carthage only
Basilica Maiorum (Midfa) produced circumstantial archaeological
evidence, an inscription with the names of Perpetua and Felicitas and
28
Ennabli, Sainte-Monique, 9192: five sub adults (1317 years), 12 children
(312 years), and one infant (fetal-2 years).
29
Nad Ferchiou, Les lments architecturaux (1990), in Stevens, Bir el Knissia,
25455.
30
Yasin, Funerary Monuments, 44245.
Commemorating the Dead 87
a crypt at the center of the basilica, suggesting it was martyrial in
origin.
31
The elaborate basilica complex at Bir Ftouha on the northwestern
outskirts of Carthage was built approximately 1 kilometer outside the
city wall, the most distant of the known surburban basilicas. Dis-
covered in 1895 by A.-L. Delattre, the complex was partially exposed
in 1897 by P. Gauckler in the process of lifting mosaics, but not exten-
sively excavated until the 19941999 excavations.
32
The origins of the
basilica complex may have been four tombs around a column foun-
dation that were cut through a beaten earth floor dating to the early
sixth century, though the identity of these individuals cannot now be re-
covered. The four individual tombs of pre-basilica date were incorpor-
ated into the foundations of the Byzantine north building. One tomb
was one of adult dimensions, aligned SE-NW. The other three were of
dimensions appropriate for children. While the reason for their irregu-
lar alignments is unclear for the pre-basilica phase, such alignments are
characteristic of tombs in centralized structures in the Byzantine com-
plex. Two of the tomb shafts were sealed by a coarse gray mortar and
cobble layer, presumably a practical measure to prevent disturbance.
One of the these (tomb 30), the only one of the four that was bottomed,
contained coffin nails in situ on the carefully cut hard clay bottom of
the shaft that otherwise contained no tomb structure. A concentration
of loose mosaic tesserae suggested that at least one of these tombs
(tomb 31) may originally have been marked by a tomb mosaic.
Thus, the first documented use of this part of the Bir Ftouha field
for burial appears to be the early sixth century. The only traces of pre-
Christian cemetery were three fragments of epitaph (cat. nos. 34, 7)
and the only other pre-basilica structure was a single wall of anom-
alous alignment and unknown date beneath the south portico. An ex-
tensive site that Delattre explored in 19281929 some 50 meters to the
northwest of the basilica complex, included a small bath complex, tre-
foil funerary chapel with sarcophagi, and a few Christian funerary in-
31
Ennabli, Midfa, 517 reproduces early excavation drawings, plans, and photo-
graphs.
32
Alfred-Louis Delattre, Inscriptions chrtiennes, Cosmos 542 (June 1895):
33739; Paul Gauckler, Fouilles, Marche du service des antiquits 1898, 7; id.,
Inventaire des mosaques de la Gaule et de lAfrique 2: Afrique proconsulaire (Tuni-
sie) (Paris: Ernest Leroux, 1910), 26364; Susan T. Stevens, Angela V. Kali-
nowski and Hans vanderLeest, Bir Ftouha: A Pilgrimage Church Complex at
Carthage, (JRASup 59; Portsmouth, R.I., 2005).
88 Susan T. Stevens
scriptions. While the full extent of the Bir Ftouha basilica complex and
its cemetery is not known, the results of a magnetometer survey in
2000 and the marking of tombeaux between the two Bir Ftouha sites
on an early-twentieth-century map of the field suggest that it extended
some distance west of the recent excavations, perhaps as far as Del-
attres 19281929 site.
33
The whole Byzantine basilica complex was built all at one time in the
late 540s, its plan modified slightly in a second phase dating from the
last third of the sixth through the last half of the seventh century. At
Bir Ftouha 95 tombs belong to the basilica cemetery. Original to the
construction of the basilica was a roughly cruciform masonry burial
structure aligned E-W with the long axis of the complex. This was
oriented to the cardinal compass points rather than with the rural cad-
astration. The tombs of this structure, described by Delattre as finely
plastered auges,
34
lay under the floor of the chancel and apse and ac-
commodated 16 individual tombs in one layer or as many as 32 in two.
The best-preserved tomb structure (tomb 1) was a rather narrow lower
tomb revetted in marble about 0.5 meters deep with a ledge for a cover
slab, above which was place for another deeper but wider tomb under
the presumed marble-tiled floor of the basilica.
35
The tombs were all of
a standard adult size and were probably marked at floor level by tomb
mosaics or inscriptions in stone, of which Delattre found fragments,
one for a presbyter and one apparently for a monk. After the construc-
tion of the basilica floor two additional tombs (or four in two levels)
appear to have been added, extending the burial structure by the length
of two tombs end to end down the middle of the nave, perhaps under a
prostoon in the fifth and sixth bays.
36
Disarticulated human bone in the
heavily robbed chancel amounted to a minimum number of three
adults, with an additional two adults attested in the disturbed fill of the
best preserved later tomb in the nave (tomb 4).
37
33
P. J. Bordy, Carte archologique et topographique des ruines de Carthage (1897);
Alfred-Louis Delattre, Sance du 27 juillet, CRAI (1928): 25254 and Les
fouilles de Bir Ftouha, CRAI (1929): 2329; Stevens, Kalinowski, and vander-
Leest, Bir Ftouha, 2126, 58082.
34
Ibid., 1534, esp. 2021. The recent excavation found another three fragments of
stone inscription in the chancel area.
35
Ibid., 35114, esp. 44, fig. 2.7 and 45, fig. 2.8.
36
Ibid., 53, fig 2.14.
37
Cherie K. Walth,Human Bone, in Stevens, Kalinowski, and vanderLeest, Bir
Ftouha, 48283.
Commemorating the Dead 89
Beginning in the last third of the sixth century 26 individual tombs
were cut through the mosaic floor of the west building and an attached
southwest external room, though only one burial of two adults in a
single shaft escaped tomb robbing. In the west building the burials were
concentrated in the eastern part of the ambulatory nearest the basilicas
narthex, in a variety of alignments, generally parallel or perpendicular
to the polygonal outer walls of the building. In the buildings central
space the tombs seem to have been aligned with column foundations.
The 16 tombs with some surviving structure were of a size appropriate
for adults (1.842.2 meters 0.50.8 meter and 1.061.53 meters deep)
buried in stone cists of various kinds and levels of elaboration. Two in
situ fragments of different mosaic inscriptions were found (tombs
1617), as well as several examples of cobble bedding for tomb markers
near floor level in other tombs. One of the tomb mosaics was a small
rectangular marker (0.50 0.15 meter) for Be or Verudus;
38
the
other was probably a tomb-sized marker incorporating a multiple-line
inscription panel. Four fragments of inscription suggest that other
tombs may have been marked with marble slabs. The absence of tomb
intercutting and the pattern of targeted tomb robbing suggest that all
the tombs were clearly marked.
The centralized north building yielded 18 individual tombs inserted
ad hoc into its flagged floor in phase 2 of the basilica complex, though
there are likely to have been many more tombs in the unexcavated
two thirds of the structure. The principle behind the placement of the
tombs, while less clear than in the west building, appears similar: five
tombs were aligned parallel to the south wall of the building, the
reused column foundation, and each other; the rest apparently were
aligned with paving stones that seem to have been laid in a pattern
radiating from the center of the building. As in the west building the
tombs are of a size and shape appropriate for adults. Only tomb
26 yielded any indication of tomb structure: it was unusually large
(2.4 0.75 meters), deep (1.8 meters), and nicely plastered on the in-
terior. The north end of the tomb was marked at floor level by a frame
enclosing a firm cobble and mortar bedding, in which a new or reused
paving stone once lay.
39
The pattern of targeted tomb robbing and the
38
Henry Maguire, Mosaics, in Stevens, Kalinowski, and vanderLeest, Bir
Ftouha, 33233, fig. 6.25.
39
Stevens, Kalinowski, and vanderLeest, Bir Ftouha, 109112, fig. 2.55 and more
generally 57476.
90 Susan T. Stevens
absence of tomb intercutting suggest that the other tombs in the north
building were also clearly marked, like those in the west building. In
fact, disturbance of the pre-basilica-phase tombs was likely to have
been prevented by markers, albeit anonymous ones, in the Byzantine
floor: the paving stones over the early tombs were aligned with the
grave beneath instead of conforming to the radiating pattern of the rest
of the floor.
An additional 15 tombs (alone or in pairs) were inserted ad hoc into
the floors outside the monumental center of the complex or outside the
complex altogether, varying in alignment according to nearby walls
or features. One sub-adult tomb consisted of a shaft without any struc-
ture 1 meter below the floor of the north hall of the ambulatory (tomb 5,
burial 1). It was centrally located and oriented N-S parallel to the long
walls of the room. Another, evidently for a child, was also aligned N-S
and lay inside an adjacent structure to the west of the north hall of
the ambulatory. Two side-by-side graves, marked by fragmentary but
in situ tomb-sized mosaic panels, were aligned roughly parallel to the
curving northeast wall of the north peristyle. Of these two tombs, the
more northerly was definitely child-sized. The better preserved of the
two, for Gaudiosa, was oriented W-E, its child-sized panel enclosed in
an unusually wide border.
40
Another tomb was probably located under
the floor of northeast room 2, aligned with its walls. The nineteenth-
century excavators recovered two other tomb mosaics for the children
Adeodatus and Redibibus or Redibiba, laid end to end, probably in the
south peristyle. Recent excavations found two other child-sized simple
shaft tombs: one in the ambulatory of the baptistery, aligned locally
roughly NE-SW with the outside wall of the building and centralized
at its entrance, and another single tomb outside the southwest curving
wall of the south peristyle, aligned roughly E-W, not quite either par-
allel or perpendicular to the walls of the area. One tomb was cut into
the courtyard floor north of the west building, and two in the south
yard: one adult-sized tomb against the polygonal wall of the west
building and one child-sized, locally centralized and aligned N-S, par-
allel to the west wall of the narthex.
41
Finally a cluster of three individ-
ual tombs, aligned E-W, lay against the outside south wall of southeast
room 2, one of these, tomb 7, consisted of a stone cist. Although no
identifiable tombs were found in the south courtyard or portico, the
40
Maguire, Mosaics, 325, fig. 6.18.
41
Stevens, Kalinowski, and vanderLeest, Bir Ftouha, 105109, fig. 2.50.
Commemorating the Dead 91
five inscription fragments found there may have been redeposited from
tombs outside the south portico wall.
Ninety-five excavated tombs is a surprisingly low number for this
large and extensively excavated a complex with well-preserved floors.
The pattern of distribution of these tombs is striking. Just over a third
of the tombs are in the masonry structure at the core of the basilica, a
group tomb that differs significantly from the group tombs in the Bir el
Knissia basilica because of its size, its central location, and the fact
that it was original to the construction of the monument. This group
burial appears to have been designed as a kind of internal martyrium
around which the rest of the basilica was built. The Bir Ftouha basil-
ica, rather than housing the cemetery, was a basilica ad corpus, in the
vicinity of which some privileged individuals were buried. Even privi-
leged tombs were excluded from the basilica proper, its narthex, apse
ambulatory, and south courtyard area. The real communal burial monu-
ments at Bir Ftouha were the west building, southwest external room,
and the north building which housed nearly half of the attested tombs
in the basilica cemetery, arranged according to the logic of each space.
While the west and north buildings were probably entrance buildings
by design, the former had higher status as an elaborately decorated main
entrance on the long axis of the basilica than the latter, a subsidiary en-
trance on the minor axis of the complex. The cemetery population of
both buildings appears to have consisted of adults who enjoyed the
status of a smaller group identity associated with a functioning basilica
complex. The least privileged members of the Bir Ftouha community,
perhaps those not yet full members, appear to have been buried singly
or in pairs in the less important, outlying buildings, outdoor spaces of
the complex, or adjacent to but outside the complex. Of these 15 tombs,
as many as nine were for children, an indication that children in the
Bir Ftouha community were differentially treated, sometimes deferen-
tially, as in the case of the single tomb in the baptistery ambulatory. It
is also worth noting, in reflecting on the possible status of this commu-
nity, that only formae were found at Bir Ftouha, by contrast with Bir el
Knissia which also had amphora burials.
Both the Bir el Knissia and Bir Ftouha cemeteries were separate
from other cemeteries of the city. Their communal monuments, the ba-
silicas, were visible and easily accessible by suburban roads, though
not located on the main roads from the city. The organization of the
cemeteries was local, conforming to the buildings with which they were
associated that were themselves not aligned with the Roman rural cad-
92 Susan T. Stevens
astration. Both cemeteries, with their individual graves indicated by
tomb markers and lack of intercutting, appear to have preserved the
identity of individuals within the community. The cemeteries also re-
flect a clear hierarchy with the greatest privilege granted to group
tombs closest to the core of the basilica and to group and individual
tombs inside the basilica and its annexes. Single tombs in outside
spaces probably represented the least privileged members of the basilica
community. Both cemetery populations appear to reflect a differential
treatment of children and adults, perhaps reflecting a prohibition
against the unbaptised being buried inside the church.
42
If architec-
tural elaboration, obvious hierarchy, and strict control of the type and
arrangement of tombs reflect privilege, then Bir Ftouhas community
appears the more privileged of the two.
The Theodosian Wall cemetery was located in the northwest quad-
rant of the city, approximately 200 meters west of the main north
gate (on the Kardo maximus).
43
It occupied a long narrow area about
5 30 meters between the city wall and a ditch outside it, an area that
had once been part of the urban grid of streets at the intersection of
Decumanus 5 north with Kardines 5 and 6 west. The 213 primary in-
humation burials of the cemetery are dated to the period between the
construction of the city wall ca. 430 and the mid-sixth century. This
cemetery appears not to have had a Roman phase.
The cemetery was organized with individual graves aligned NW-SE,
roughly parallel to the Theodosian city wall, with the greatest density
of graves closest to the wall, and thinning out further from it. While the
plan of the complete cemetery gives the impression of being organized
in serried rows, its chronological development suggests rather that
42
Augustine, De sepultura catechumenorum 7 (F. Dolbeau, Augustin dHippone:
vingt-six sermons au peuple dAfrique 1, Collection des tudes augustiniennes,
srie antiquit: Paris, 1996) mentioned by Yasin, Funerary Monuments, 451.
43
Andrea Carandini, Lucilla Anselmino, Clementina Panella, et al., Gli scavi ita-
liani a Cartagine: Rapporto preliminare delle campagne 19731977, Quaderni
di Archeologia della Libia 13 (1983): 761; Lucilla Anselmino, Le secteur nord-
ouest de la ville, in Pour Sauver Carthage (ed. Abdelmajid Ennabli; Paris and
Tunis: Institut national de larchologie et lart and UNESCO, 1992), 125130;
Mark B. Garrison, A Late Roman/Early Byzantine Cemetery at Carthage:
The University of Michigan Excavations at Carthage, Archaeological News 15
(1990): 2329; Mark B. Garrison and Susan T. Stevens, Le cimitire du Mur de
Thodose, in Pour Sauver Carthage, 13134; Susan T. Stevens, A Late Roman
Urban Population in a Cemetery of Vandalic Date at Carthage, JRA 8 (1995):
26370 with a final report in preparation.
Commemorating the Dead 93
tombs of roughly the same date may have been clustered. Much
clearer is the vertical arrangement of graves with burials over one
another in irregular stacks or overlapping end to end, a system that
was consistent over time and was probably intentional. This vertical
arrangement of tombs, the precedent for which may have been set
by two early cist burials, one over the other sealed under the same
marker, may have been devised in the absence of grave markers to
avoid disturbing recent graves. It appears likely that some vertical
stacks are family plots.
Except for a handful of double burials of siblings or mothers and
infants buried at the same time in the same grave, the burials were for
single individuals. The vast majority of the graves (142) were simple pit
inhumations in which the bodies were covered only with earth, though
some graves were also lined with stones or covered by large pieces
of amphora. In the 71 most elaborate grave structures, bodies were
contained in amphoras, stone, or mud-brick cists. The only discern-
ible grave goods were a coin or coins in 36 of the graves, and a few
personal items (earrings, a pair of tweezers). The differential treat-
ment of infants is attested in this cemetery both in a rather isolated
infant section, and by the fact that elsewhere only infants were buried
in amphorae. The cemetery population was made up of 63 percent
adults, 37 percent infants-sub adults, an expected proportion for a
pre-industrial population.
The cemetery contained one possible monument, an enclosure wall
that may have set off two tombs, albeit temporarily. Eleven tombs had
inscribed tomb mosaics, of which two were iconographically identifi-
able as Christian. Nine of these markers were laid against each other,
a small group distinguished from the larger community by their orien-
tation. The mosaic-marked burials in the eastern part of the cemetery
were unusually deep, in sturdy cists, suggesting intent to protect
them from disturbance. The intensity of later burials around and cut
through this mosaic group suggests their perceived importance. The
mosaics originally enhanced the sense of smaller-group identity and
later may have served as a kind of communal monument for this part
of the cemetery. Indeed, the other two mosaic-covered tombs in the
cemetery lay isolated in other parts of the cemetery and did not attract
later tombs. Another homogeneous group of burials set somewhat
apart from the larger community included six young infants. All the
tombs of the cemetery are for individuals except where members of
same nuclear family were buried at the same time, or in two cases,
94 Susan T. Stevens
apparently in sequence (one over the other in the bisomum of Codbul-
deus, or side by side inside the enclosure wall). While some tomb
clusters appeared to be family plots, other individuals were grouped
by age (infant section) or commemorative status (graves marked with
mosaics).
The off-the-beaten-path location of the cemetery and absence of
functional buildings or streets in the vicinity would not have promoted
visitation. Furthermore, there were no recognizable paths between
graves, few monuments and markers, and no evidence for ritual activity,
the usual indicators of a continuing association of the living with the
dead. While a few cases of bone reburials suggest some management
of the cemetery and a certain respect for earlier burials, the abundance
of bone dumps and disarticulated bones indicate that disturbance of
earlier graves by later ones was routine. The cemeterys users could not
have ascribed much importance either to the physical commemoration
of the individual or the integrity of the grave. The cemetery nevertheless
expresses a communal identity in the limited range of simple tomb
types and the prevalence of coins as grave goods. The chronology of the
cemetery, that is, its beginning with the building of the Theodosian city
wall and its end with the destruction of the wall, the arrangement of
tombs parallel to the city wall and the concentration of tombs against it
suggests that the city wall acted as the collective marker of the cem-
etery. Within the cemetery a smaller community may have been formed
by the interlocking of mosaic markers, subsequently destroyed by later
burials, making them their collective marker.
The Circus cemetery is located in the southwest quadrant of the city
close to the circuss northwest end, near the carceres, between the long
southwest wall of the circus and the Theodosian city wall.
44
The use
of the area before the construction of the city wall is unknown. After
its construction the slope up to the north was gradually filled in with
debris and flattened out. Access to the cemetery is unknown, though
it may have been from the area of the circus, the bays of which were
occupied for domestic habitation after the circus no longer func-
44
Simon P. Ellis, John H. Humphrey, and Judy P. Marshall, The Theodosian Wall
and the Cemetery, The Circus and a Byzantine Cemetery at Carthage, (ed. John
H. Humphrey; Ann Arbor: University of Michigan Press, 1988), 179256 and
Simon P. Ellis and John H. Humphrey, Interpretation and Analysis of the Cem-
etery, in The Circus and a Byzantine Cemetery, 32536.
Commemorating the Dead 95
tioned.
45
The earliest human burial in the area was of three adults in a
shallow pit probably dating to the late sixth century, a burial strati-
graphically linked to a large dump of partially articulated equids, in-
cluding at least 14 individuals.
46
The formal burial area that followed, dating to the first half of the
seventh century, included 50 inhumations (27 primary and 23 dis-
turbed burials), although the cemetery was more extensive than the ex-
cavated area.
47
The excavated portion of the cemetery was approxi-
mately 20 meters wide and 30 meters long, with all but two graves
oriented NW-SE, aligned roughly parallel to the city wall, with the
greatest density of graves close to the wall, and thinning out to the
northeast. The Circus cemetery was newly founded, probably in the
late sixth century, in quite a remote and probably poor area of the
city. It included no enclosures or paths, only one obvious mud-brick
marker (though other markers may have been removed in the leveling
of the area), and no inscriptions either in situ or in fragments. The ma-
jority of the graves were in well-defined rectangular pits, most had
some kind of solid tomb structure, a stone or mud-brick cist or an am-
phora, and at least three graves (possibly as many as six) contained evi-
dence for a coffin. Grave goods were found only in three burials: a
complete cooking pot placed in the west end of the shaft (above the
head) over the cover slabs of two cist burials and coins scattered in the
shaft fill of one grave. The extensive intercutting of tombs suggests that
the location of graves was rarely known, and a concomitant opportun-
istic robbing and reuse of stone from earlier graves is well attested.
Although the numbers are small the cemetery population consisted of
approximately 60 percent adults and 40 percent infants-sub adults and
the ratio of sexable skeletons was 1:1. The Circus cemetery contrasts
with the Theodosian Wall cemetery in its lack of small groups or other
suggestions of differential treatment of some individuals in the com-
munity.
45
The circus seems to have functioned into the sixth century but not into the
seventh, Naomi J. Norman, Le cirque romain, in Pour Sauver Carthage, 162.
Five additional burials were found within the confines of the ruined circus.
46
Kevin Rielly, A Collection of Equid Skeletons from the Cemetery, in The Cir-
cus and a Byzantine Cemetery, 297300.
47
A minimum of 30 individuals were represented among the redeposited bone and
the excavators remark that many rectangular pits from which no bones were
recovered may have been robbed graves. Ellis and Humphrey, Interpretation,
330.
96 Susan T. Stevens
The cemetery at Falbe Point 90, now dated to the mid-fifth through
sixth century, was excavated by a Danish team in 1975, 1977, and 1981.
48
The cemetery includes at least 44 individuals buried in three parts of the
vaulted substructures of a fourth-century Roman villa which were re-
turned to domestic use after burial activity ceased in the sixth century.
Although the cemetery includes mass burials, one of only two sites with
mass graves known at Carthage in this period,
49
the general interpre-
tation of the cemetery as a burial site for victims of epidemic or famine
masks other characteristics highlighted in the latest excavation report.
50
In AO, the southernmost room, the excavators report 30 skeletons
laid in three shallow horizons. The earliest layer of burials included 14
individuals: seven (mostly infants under 1 year) that lay both under
and on the feet of seven W-E-oriented adults aligned with the walls of
the structure. Ten centimeters above these in the same sandy fill were
six S-N-oriented adults, again aligned with the walls of the vault. A
third group burial, including three adults and two infants less than two
years old, was in the SE corner of the room, though both the original
orientation of these individuals and their chronological relationship to
the other two groups was uncertain. The plan suggests, however, that
this last group contained the jumbled remains of five individuals cut
into a seventh adult burial in the horizon of six N-S-oriented adults for
a total of 26 individuals.
51
The room is interpreted as a mass grave of
individuals who died and were buried at the same time, identified by
48
Sren Dietz and S. Trolle, Premier rapport prliminaire sur les fouilles danoises
Carthage: les campagnes de 1975 et 1977, National Museum of Denmark
Working Papers 10, Copenhagen, 1979; Lucinda Neuru, Late Roman Pottery:
A Fifth-Century Deposit from Carthage, Antiquits africaines 16 (1980):
195212, discussed by John H. Humphrey, Vandal and Byzantine Carthage:
Some New Archaeological Evidence, in New Light on Ancient Carthage (ed.
John Pedley, Ann Arbor: University of Michigan Press, 1980), 109113. Sren
Dietz, Fouilles danoises Carthage 19751984, Cahiers des tudes anciennes
16 (1984): 10718; Erik Poulsen, Tombs of the 4th5th centuries A.D. in the
Danish Sector at Carthage (Falbe Site no. 90), Cahiers des tudes anciennes 18
(1986): 141159, although the probable later date suggested in the article is not
reflected in its title. The latter was most recently discussed by Norman, Roman
Children, 4244.
49
The other possible mass burial dates to the first half of the sixth century and was
found in the Italian Taglio 3A of the Theodosian Wall cemetery, Carandini et al.,
Gli scavi italiani a Cartagine, 46.
50
Poulsen, Tombs, 141159.
51
Poulsen, Tombs, 152, fig. 1a.
Commemorating the Dead 97
pottery and coins in the sandy fill of the room to be after the second
half of the fifth century. The assemblage of at least one complete vessel
and many broken ones concentrated in the northwest corner of the
room near the head and torso of skeleton AO19, the two complete jugs
at the right shoulder of skeleton AO 20 (or perhaps at the feet of skel-
eton AO 28) near the center of the west wall as well as the three com-
plete jugs at the feet of skeleton AO 22 against the east wall of the
room, are interpreted as a collective offering.
52
However, with two such clearly delineated strata of perpendicularly
oriented burials, it is tempting to refine the excavators interpretation
of a single mass grave, into two (or perhaps three) different group
tombs for individuals who died at the same time. Indeed, it is not un-
usual to find only ten centimeters of fill between individuals buried se-
quentially, as in the uneven stacks of burials at the Theodosian Wall
cemetery. There individuals buried together in the same tomb had no
dirt between them and layers of cemetery earth laid down at what must
have been different times were virtually impossible to distinguish at the
time of excavation.
In the central room (AG), two groups of adults were buried in the
same horizon and interpreted by the anthropologists as parts of a mass
grave, collectively commemorated by the two whole jugs against the
west wall of the room. The stratigraphy indicates, however, that there
were at least two and perhaps three phases of burial in the room. Two
well-preserved adults oriented roughly N-S, and aligned with the walls,
were buried together (AG 8 supine and AG 9 on its right side facing
AG 8) in the west end of the room, accompanied by a series of vessels
that lay along the right side of AG 8, from shoulder to knee, and per-
haps two whole jugs.
53
Pottery suggests a mid-fifth-century terminus
post quem for the burial of these two individuals. The second and
slightly later group (with a terminus post quem from pottery of the
mid- to late fifth century) consisted of three adults (AG 57), oriented
roughly west-east. A close examination of the plan invites a further
complication to the sequence of burial in the room: skeleton AG 5
appears isolated so much further west and south of the pair AG 67
(buried in very close proximity to each other) as to raise a question
about whether it was a single burial or part of a group of three. Poulsen
himself suggests that a very disturbed but partially articulated skeleton
52
Ibid., 15556, figs. 23.
53
Ibid., 157, fig. 4.
98 Susan T. Stevens
between the two groups may represent the remains of a single individ-
ual buried on a N-S alignment, presumably earlier than the two ident-
ified group tombs in the room.
54
Finally, the remains of numerous infants or small children were
found in three groups in the northern room (CL):
55
CL 5 consisted of
three burials in the northwest corner of the room; CL 6 included at
least eight individuals, two amphora burials, a skeleton in the north-
east corner of the room and the disturbed remains of at least five other
individuals against the north wall. CL 7 had at least two burials that
occupied the southeast quadrant of the room. The burials within these
groups, as well as the groups themselves, cut one another and were
therefore not contemporaneous. Coin evidence suggests that the burials
began in the early fifth century and continued at least into the early
sixth. The burials, in amphoras or covered with a combination of am-
phora sherds and stones, appear to have been disturbed because they
were placed ad hoc, in a variety of alignments that did not follow the
walls of the room.
The cemetery had a distinctive and consistently applied burial prac-
tice. Adults, buried singly or in groups side by side, were buried supine
with the right arm under the right side, aligned with the walls of the
structure. The pattern was flexible enough to be adapted by AG 9,
an individual buried on the right side to face AG 8, the other half of a
pair buried together. The same flexibility characterizes the treatment
of children who, perhaps in the circumstances of famine or epidemic,
were buried with adults in AO who died at the same time. In less stress-
ful circumstances they were buried sequentially with other children in
CL, in varied alignments that seem to have little to do with the vault
that contained them. The children in CL had some of the same goods
(coins, whole jars) as the adults in AO and AG, but other goods in CL
(animal bone and wooden plaques) appear to distinguish the children
from the adult population of the cemetery. This cemetery demon-
strates not only that groups, pairs, and individuals were differentially
treated, but also that the grouping and placement of individuals in dis-
tinct parts of the cemetery in rooms AO, AG, and CL, may have been
determined by the circumstances of their deaths.
On the strength of this survey of fifth- through seventh-century cem-
eteries at Carthage some general observations are warranted, although
54
Ibid., 146.
55
Ibid., 148, fig. 1b.
Commemorating the Dead 99
they are hardly conclusive in the absence of final site reports. Each of
the five cemeteries was identified by a shared burial practice and a col-
lective marker, a building or structure, in use or reuse. Within each
cemetery individuals or small groups of individuals with a distinctive
identity within the community were commemorated to a greater or
lesser degree by tomb enclosures or markers. The feature that most
clearly distinguishes the Bir el Knissia and Bir Ftouha cemeteries from
the Theodosian Wall, Circus, and Falbe Point 90 cemeteries is the
character and degree of commemoration of the dead. At Bir el Knissia
and Bir Ftouha the dead were commemorated not only by their basil-
ica as the overarching collective monument of their community, but
also by individual markers and group monuments for small groups of
individuals within the larger community. The endurance of individ-
uals, groups, and the whole community of the dead was ensured by the
forma-type tomb which was principally designed for a single individ-
ual, and also was usually marked. The dead, in a collective sense, were
also incorporated into the community of the living.
In comparison, the dead in the Theodosian Wall, Circus, and Falbe
Point 90 cemeteries seem to have been separated from the living, in
largely anonymous and ephemeral tombs, in cemeteries in peripheral
or uninhabited areas of the city. In two cases, however, they were
closely associated with the functioning city wall. The general absence
of individual markers in these cemeteries and the extensive intercutting
of graves suggest a tacit acknowledgement of the impermanence of
burial and a recognition of the temporal limits of the community.
Finally, the social stratification implied by the presence of small
groups of graves in some cemeteries suggests higher social status than
in cemeteries where distinctions between members are not apparent. In
other words, commemoration of the dead in this period at Carthage, as
in earlier periods and elsewhere, appears more pronounced among the
privileged, and this elite expressed itself by affiliation to a church and a
living community. The less privileged may have been imitating the elite
by developing a distinctive burial koine of their own and adapting ear-
lier structures as collective monuments.
None of the five cemeteries featured here appears to extend the life
of a Roman necropolis, either chronologically or spatially. A similar
discontinuity characterizes two other recently excavated sites. At Yas-
mina on the southwestern outskirts of the Roman city, an elite necrop-
olis that included a second-century three-story stucco monument for
M.Vibius Tertullus, possibly of consular family, and the mausoleum of
100 Susan T. Stevens
a circus sparsor and his wife, dating to the first quarter of the third cen-
tury. This necropolis was abandoned by its original users by the early
fourth century when burial activity ceased.
56
A new inhumation cem-
etery, apparently for individuals of lower social status, developed in
the fifth and sixth centuries and may have continued later. The late
cemetery included 60 tombs, approximately a third of which were for
children under the age of seven, the only tombs which had associated
grave goods. Burials in the cemetery were in a variety of structures:
simple pits, stone-lined trenches, occasional sandstone cists, covered
by broken amphoras and cobbles, or pitched stone or tiles. The un-
marked tombs, some of which were stacked on top of each other, were
clustered around and were cut into the still-standing early tomb monu-
ments that may have acted as collective markers for the later cemetery.
A Roman necropolis along Kardo 2 east included high-status monu-
ments of the third century, a hypogeum of the late fourth or early fifth
century, and six individual late-fifth- or early-sixth-century inhumation
graves aligned with the hypogeum entrance. All were leveled in the sec-
ond third of the sixth century to make room for a Christian memoria,
associated with the basilica of Damous el Karita. The side chambers of
the new memoria may have been designed for high-status burials.
57
At
the still-poorly-understood Christian basilica, the pattern of burial
may also have changed. Formae (excavated by Delattre in the nine-
teenth century) cut through the floor of the basilicas nave, aisles, and
great hemicycle atrium in the later fourth and fifth centuries (phase 1)
may have become less frequent in the early sixth century. In their place
more exclusive burial rooms were incorporated into the basilicas
phase 2 annexes and perhaps into its later sixth-century, phase 3, funer-
ary hall.
58
The establishment of new cemeteries in this period and the concomi-
tant loss of older ones should be read, at the very least, as part of the
same social fragmentation of the citys population evinced by the small
56
Naomi Norman and Anne Haeckl The Jasmina Necropolis at Carthage, JRA
6 (1993): 23850. I am grateful to Norman for sharing the text of her paper,
Death and Burial in Roman Carthage, delivered at the Classical Association
of the Midwest and South, Southern Section, Birmingham, Ala., Nov. 2002, ac-
cess to a website devoted to the Yasmina necropolis, and a list of the late burials,
which were discussed in part in Norman, Roman children, 30508.
57
Dolenz, Damous el Karita, 4351.
58
Ibid., 2139.
Commemorating the Dead 101
groups of burials in the buildings of the former urban center.
59
Per-
haps, in addition to enclosing and protecting the dead, the reuse of
Classical buildings for burial provided a communal identity for the
graves and means of collective commemoration similar to that found
in cemeteries on the periphery.
Another recently excavated cemetery site, Bir el Jebanna, suggests
first that the two phenomena, the insertion of burials in Classical
buildings and the establishment of new communal cemeteries, may be
contemporaneous. Second, the two phenomena suggest a similar com-
munal ideology. A second-century public bathhouse on the periphery
of the city fell into disuse in the fourth century. The use of Bir el Jeb-
anna for burials began in the fourth century, before it could have been
transformed by being excluded from the urban fabric by the construc-
tion of the Theodosian city wall in the early fifth.
By the mid-fourth century a series of NE-SW-oriented graves, aligned
with the walls of the bathhouse, were first cut into the floor of the origi-
nal north and later south rooms of the baths. By the late fourth century,
a more numerous series of NW-SE-oriented burials were cut into the
fill representing the collapse of the bathhouse, though probably also
aligned with its ruined walls. A small group of shallow masonry tombs
oriented NE-SW, outside the bath complex to the northeast, and a
stack of burials in an adjacent room also dated to the fourth century.
Burial in and around the ruined bathhouse appears to have continued
into the mid- to late sixth century, the date of the latest burial, a neon-
ate in a shallow amphora tomb.
60
By the sixth century a new cemetery had established itself behind an
enclosure wall that may have been built as early as the late fourth cen-
tury at the eastern extent of the ruined bathhouse. Among the more
than 30 tombs recently excavated in this cemetery were some marked
59
Leone, Linumazione, 23348.
60
J. J. Rossiter, A Roman Bathhouse at Bir el Jebanna: Preliminary Report on the
Excavations (19941997), in Carthage Papers: The Early Colonys Economy,
Water Supply, a Public Bath and the Mobilization of State Olive Oil (JRASup 28,
Portsmouth, R.I., 1998), 11213; 103115; id., Excavations at Bir el Jebanna,
Carthage (1994): A Roman Bathhouse Rediscovered, Actes du 8e colloque inter-
national sur lhistoire et larchologie de lAfrique du nord, Tabarka (Tunisie) 813
Mai 2000 Tunis (2003), 491501. Rossiter generously shared the text of his paper
From Bath-house to Cemetery: The Transformation of Suburban Space at Bir el
Jebbana, Carthage, delivered to the Socit dtude du Maghreb Prhistorique,
Antique et Mdivale.
102 Susan T. Stevens
with architectural fragments from the bathhouse, fragmentary inscrip-
tions, and tomb mosaics. Limited finds from the tombs fall within the
standard repertory of grave goods of the period, coins and a few items
of personal adornment (worked bone objects, bronze rings, and brace-
lets) dated to the fourth through the sixth century. As in the Theodosian
Wall cemetery, Christian tomb mosaics were found at Bir el Jebanna
without an attested Christian cult building. This cemetery included
the mosaic-covered cupola tomb for the child Theodora excavated
by Delattre in the nineteenth century. Despite their similarity in date,
the burials in and around the bathhouse and the cemetery behind the
enclosure wall were not specifically connected to each other. Fur-
thermore, neither group of burials can be considered a continuation of
the extensive first through third century necropolis known as the Cem-
etery des officiales, that lay on the line of the Decumanus maximus
adjacent to the west gate of the city.
61
Together the featured cemeteries demonstrate a shared burial ideo-
logy expressed in (mostly) individual tombs placed ad hoc in horizon-
tal and vertical relation to each other that is strikingly different from
that lying behind the familial monuments of Roman necropoleis of
firstthird-century Carthage. The chronology of these sites suggests a
recognizable development of that communal ideology. Bir el Jebanna is
a reminder that this development was already underway in the fourth
century, at least in the periphery of Carthage.
The Bir el Knissia, Theodosian Wall, and Falbe Point 90 cemeteries
were all founded in the fifth century and, despite the differences in so-
cial status of their communities, demonstrate a communal burial ideo-
logy that is quite inclusive. At Bir el Knissia, the social hierarchy of the
community is expressed in two ways, familial, with group tombs out-
side the core of the basilica and clerical, with group tombs inside it. At
the Theodosian Wall cemetery some groups of tombs stood out, such
as the possible family enclosure, and the cluster of tombs marked by
mosaics. The latter, rather than being honored as grave markers for in-
dividuals, were destroyed in the process of being transformed into a
collective marker for later graves. The flexibility of burial practice at
the Falbe Point 90 cemetery is most obviously demonstrated by the
61
Lantier, Notes de topographie, fig. 1, no. 14; Robert tienne and Georges
Fabre, Dmographie et classe sociale: lexample du cimitire des officiales de
Carthage, in Recherches sur les structures sociales dans lantiquit classique,
Caen 2526 Avril 1969 (Paris: ditions CNRS, 1970), 8197.
Commemorating the Dead 103
presence of mass graves together with burials in pairs or as individuals.
The variety of orientation of graves, abundance of grave gifts of a li-
mited variety, and even possible collective markers in some rooms, can
be interpreted as other indicators. Some of the gifts, unusual by the
standard of the other cemeteries, suggests a tightly knit community,
though its status is unknown.
The origins of the Circus and Bir Ftouha cemeteries fall squarely in
the Byzantine sixth century and later. The Circus cemetery, as the pres-
ence of amphora burials and absence of inscriptions probably indi-
cate, is lower in social status than the Bir Ftouha cemetery. Never-
theless, these two cemeteries look similar in that individuals with few
gifts were buried in a limited number of tomb types strictly arranged
within the cemetery. This austere pattern is confirmed by the sixth- and
seventh-century cemetery at Le Kram that lay approximately 200
meters outside a southern stretch of the Theodosian city wall, and was
arranged and governed by principles similar to those at the Circus
cemetery.
62
The Le Kram cemetery consisted of 50 unmarked tombs of
two simple types, invariably oriented NW-SE. Seventy percent of the
graves were for adults in stone cists that lay deep (between 1.5 and
3 meters) and thirty percent of the graves were burials for infants and
children (except for one anomalous adult) either in large African cylin-
drical or small eastern globular amphoras. Both the Circus and Le
Kram cemeteries have less variation in tomb type and orientation and
fewer grave goods than the Theodosian Wall cemetery.
Thus, the relatively expansive and fluid burial practice seen in cem-
eteries originating in the Vandal fifth century seems to have given way to
the rather austere and tightly controlled system in cemeteries originating
in the Byzantine sixth century and later that are more homogenous
and egalitarian in appearance. However, the differential treatment of
children in cemeteries,
63
though manifested in various ways, continued
unabated from the fifth through the seventh century at Carthage, per-
haps an indication of the strength of the African and Roman roots of its
inhabitants.
62
Mohammed K. Annabi, Deux ncropoles au sud de la ville, in Ennabli, Pour
Sauver Carthage, 18687.
63
Norman, Roman Children, 3745, explores the pattern in Africa proconsu-
laris.
104 Susan T. Stevens
Ritual and Religious Rites
Dining with the Dead 107
Robin M. Jensen
Chapter 4
Dining with the Dead:
From the Mensa to the Altar in
Christian Late Antiquity
Roman tombs were gathering places for the living as well as for the
dead. Family members and friends came to graves at regular intervals
in order to honor the departed by sharing a meal with them. Since
tombs, then as now, displayed the wealth or social status of the de-
ceased and their heirs, more elaborate private family enclosures in-
cluded furnishings and facilities for pouring libations, and preparing
and sharing simple food offerings with the shades (manes) of the dead.
Cemeteries featured communal banqueting tables that might be used
by visitors to collective grave areas. Grave goods included drinking
cups, bowls, and other dishware. Consolation, convivium, and nourish-
ment were thus offered to mourners as well as to their departed loved
ones, and social contact was established at least briefly, between the
upper and lower worlds.
Visual and epigraphic artifacts as well as textual evidence offer both
concrete and verbal testimony to this funerary practice and demon-
strate that it was continued by converts to the Christian religion, who
also adapted it for the feasts of their martyrs and saints despite the
often vehement disapproval of church officials. Gradually, the tradi-
tion of eating a meal with the dead was also transformed into the prac-
tice of celebrating a eucharist at an ordinary funeral. First at the
tomb, then at the altar, the church family gathered to hear the tales of
heroism and to eat a meal celebrating the lives of their spiritual as
well as blood ancestors. Funerals and food, then as now, are a natural
combination.
Pictorial representations of the deceased reclining on a couch (kline)
and enjoying a banquet are nearly ubiquitous in Greek and Roman fu-
108 Robin M. Jensen
nerary sculpture from the fifth century B.C.E to the fourth century
C.E.
1
Art historians traditionally refer to this iconographic motif as
Totenmahl or meal of the dead. Most of these were crafted for
non-Christian clients and show a reclining figure (usually male) hold-
ing a drinking cup. Spouses, children, and servants often appear in the
composition, wives usually seated next to their reclining husbands.
Other details, such as pets, flowers, birds, and small tables or trays of
food may be included (Fig. 4.1). The images are sometimes carved in
the round, but also carved in relief on sarcophagus fronts or on free-
standing monuments.
Funerary meal scenes appear also on a few polychrome mosaic
tomb coverings from Roman Africa.
2
Although most of the surviving
examples of these tomb covers were produced for Christian clients
(and do not portray meals), a small group of pagan tomb covers dem-
onstrates that non-Christians occasionally ordered such decorative
sepulchral embellishments. A few, from the area of Thina and now in
the Muse Archologique de Sfax, specifically represent the deceased
on a dining couch, holding a drinking cup as if toasting the viewer.
A small tripod table stands in front of the couch, laid with delicacies
appropriate for an underworld repast (Fig. 4.2). In a matched set
made for a husband and wife, eroti (naked and winged children)
bring baskets of red roses, and small winged musicians play on
stringed instruments (panduria). Along with bird and garlands, the
1
For evidence dating to a period before the common era, see Jean-Marie Dentzer,
Le motif du banquet couch dans le proche-orient et le monde grec du VIIe au
IVe sicle avant J.-C. (Rome: cole franaise de Rome, 1982). Among the many
volumes on Roman sarcophagi, the following (in English) are recommended:
Michael Koortbojian, Myth, Meaning, and Memory on Roman Sarcophagi
(Berkeley, Calif.: University of California, 1995); Guntram Koch, Roman Funer-
ary Sculpture: Catalogue of the Collections (Malibu, Calif.: J. Paul Getty Mu-
seum, 1988); and Susan Walker, Memorials to the Roman Dead (London:
Trustees of the British Museum, 1985).
2
Many catalogues of Roman African mosaics contain examples of these tomb
pavements. The most comprehensive works on Christian funerary mosaics, how-
ever, are the as-yet-unpublished dissertations of Margaret Alexander Early
Christian Tomb Mosaics of North Africa (Ph.D. diss., New York University,
1958), and James Terry, Christian Tomb Mosaics of Late Roman, Vandalic
and Byzantine Byzacenal Mosaics, Tunisia (Ph.D. diss., University of Missouri
Columbia, 1998), as well as Nol Duval, La mosaque funraire dans lart pa-
lochretien (Ravenna: Longo, 1976). See also Paul-Albert Fvrier, Mosaques
funraires Chrtiennes dates dAfrique du nord, ACIAC 6 (1965): 43356.
Dining with the Dead 109
4
.
1
.
F
r
o
n
t

p
a
n
e
l

f
r
o
m

a

s
a
r
c
o
p
h
a
g
u
s

i
n

t
h
e

J
.

P
a
u
l

G
e
t
t
y

M
u
s
e
u
m
.
(
T
h
e

J
.

P
a
u
l

G
e
t
t
y

M
u
s
e
u
m
,

V
i
l
l
a

C
o
l
l
e
c
t
i
o
n
,

M
a
l
i
b
u
,

C
a
l
i
f
o
r
n
i
a
.

U
s
e
d

w
i
t
h

P
e
r
m
i
s
s
i
o
n
.
)
110 Robin M. Jensen
4
.
2
.
M
o
s
a
i
c

t
o
m
b

c
o
v
e
r

f
r
o
m

T
h
i
n
a
.

M
u
s

e

A
r
c
h

o
l
o
g
i
q
u
e

d
e

S
f
a
x
.

(
a
u
t
h
o
r

s

p
h
o
t
o
.
)
Dining with the Dead 111
iconography portrays the pleasures of a luxurious earthly life, or per-
haps a hoped-for blissful afterlife (Fig. 4.3).
3
A different type of image one that portrays a group of diners (rather
than a couple or small family gathering), reclining at a semicircular
table (stibadium) and sharing a convivial banquet is frequently seen
on both Christian and pagan monuments from third- and fourth-cen-
tury Rome. In these, like the others, a small tripod table usually stands
in front of the couch. Wine cups are visible on the table or in raised
hands.
4
This banquet motif appears on both pagan and Christian sar-
cophagi, as well as painted on walls of pagan and Christian hypogea
in the Roman catacombs (Figs. 4.45).
5
In both pagan and Christian
examples the assembly consists of seven (but sometimes five or twelve)
diners reclining around a table set with wine, bread, and fish. In one
well-known fresco, from the (pagan) Hypogeum of Vibia, the Good
Angel (Angelus Bonus) guides the deceased (Vibia) through the gate of
Paradise by the Good Angel. In the same image, Vibia appears again,
seated at a table with five others who, like her, were judged by the
good ones (bonorum iudicio iudacati, Fig. 4.6).
6
Scholars have interpreted these scenes in various ways. Whether
the iconography alludes to some aspect of funerary practice, represents
the deceaseds past life, or offers an optimistic view of the afterlife is
a matter of debate. Katharine Dunbabin has recently argued that all
are possibilities that the scene probably denoted different things in
3
A fourth-century mosaic from Antioch gives a different, notable example from
another region. See the depiction of a funeral banquet now in the Worcester Art
Museum, showing women reclining on couches or serving the meal. Illustration
in Christine Kondoleon, ed., Antioch: The Lost Ancient City (Princeton: Prince-
ton University Press, 2000), 1212.
4
See Katherine M. D. Dunbabin, The Roman Banquet: Images of Conviviality,
esp. chap. 4, Drinking in the Tomb, (Cambridge: Cambridge University Press,
2003), 10340; as well as her earlier article, Triclinium and Stibadium, in
Dining in a Classical Context (ed. William J. Slater; Ann Arbor: University of
Michigan Press), 1991.
5
For discussion of these images, typological categorization, and the iconographi-
cal distinction between pagan and Christian examples see Friedrich Gerke, Die
christlichen Sarkophage der vorkonstantinischen Zeit (Berlin: Walter de Gruyter,
1940), 12342.
6
On the Christian images see Josef Engemann, Der Ehrenplatz beim antiken
Sigmamahl in JAC 9 (1982): 23949; Elisabeth Jastrz ebowska, Iconographie
des banquets aux IIIeIVe sicles, RecAug, 19 (1979): 390; and Robin Jensen,
Dining in Heaven, BRev 14 (1998): 3239, 48.
112 Robin M. Jensen
4
.
3
.
M
o
s
a
i
c

t
o
m
b

c
o
v
e
r

f
r
o
m

T
h
i
n
a
.

M
u
s

e

A
r
c
h

o
l
o
g
i
q
u
e

d
e

S
f
a
x
.

(
a
u
t
h
o
r

s

p
h
o
t
o
.
)
Dining with the Dead 113
4
.
4
.
P
a
i
n
t
i
n
g

f
r
o
m

c
a
t
a
c
o
m
b

o
f

S
t
.

S
e
b
a
s
t
i
a
n
o
,

R
o
m
e
.

T
o
m
b

o
f

M
.

C
l
o
d
i
u
s

H
e
r
m
e
s
.
(
P
h
o
t
o
:

E
s
t
e
l
l
e

S
.

B
r
e
t
t
m
a
n
,

I
n
t
e
r
n
a
t
i
o
n
a
l

C
a
t
a
c
o
m
b

S
o
c
i
e
t
y
.

U
s
e
d

w
i
t
h

p
e
r
m
i
s
s
i
o
n
.
)
114 Robin M. Jensen
4
.
5
.
P
a
i
n
t
i
n
g

f
r
o
m

c
a
t
a
c
o
m
b

o
f

C
a
l
l
i
x
t
u
s
,

R
o
m
e
.

(
P
h
o
t
o
:

E
s
t
e
l
l
e

S
.

B
r
e
t
t
m
a
n
,

I
n
t
e
r
n
a
t
i
o
n
a
l

C
a
t
a
c
o
m
b

S
o
c
i
e
t
y
.
U
s
e
d

w
i
t
h

p
e
r
m
i
s
s
i
o
n
.
)
Dining with the Dead 115
4
.
6
.
P
a
i
n
t
i
n
g

f
r
o
m

H
y
p
o
g
e
u
m

o
f

V
i
b
i
a
,

R
o
m
e
.

(
P
h
o
t
o
:

E
s
t
e
l
l
e

S
.

B
r
e
t
t
m
a
n
,

I
n
t
e
r
n
a
t
i
o
n
a
l

C
a
t
a
c
o
m
b

S
o
c
i
e
t
y
.
U
s
e
d

w
i
t
h

p
e
r
m
i
s
s
i
o
n
.
)
116 Robin M. Jensen
different places and times, and also that it may have been intentionally
ambivalent and lacked any clear or definable content, allowing differ-
ent viewers to read whatever meaning they chose into the image a
flexibility that could have added to the themes popularity.
7
Accompanying inscriptions sometimes provide explanations, many
of them being of the eat, drink, and be merry for tomorrow you may
be dead type, written in the first-person voice of the deceased, who
advises the passerby to enjoy the transitory pleasures of life, or asks
the visitor to pour him a drink. One monument, that of Flavius Agri-
cola, today in the Indianapolis Museum of Art, puts words into the de-
ceaseds mouth, advising his friends to mix up wine and drink and keep
pretty girls handy since death comes all too quickly.
8
Another, the sar-
cophagus of Titus Aelius Euangelus and his wife Gaudenia Nicene,
now in Malibus Getty Museum, includes an inscription that asks the
reader to pour unmixed wine for him a patient man.
9
Others, like
an inscription from the city of Rome, expressed the expectation that the
deceased couple would attend the funeral feast and enjoy themselves
along with everyone there.
10
Descriptions of funeral banquets also appear in literature. Petro-
niuss fictional comic freedman, Trimalchio, solicits a promise that his
friends would erect a lavish monument to decorate his tomb, which
would include a sculpted representation of dining couches (triclinia)
with a gathering of people enjoying themselves at a banquet. He re-
quests the image of his pet dog to be placed at the feet of his portrait
statue, along with some banqueting wreaths, a sun dial (a kind of mem-
ento mori to remind the visitor of lifes brevity), and the fights of a
champion gladiator. He further asks that large jars filled with wine be
placed at his right hand so that the libations poured over his bones
might do him as much good dead as when he was alive.
11
Taken at face value, these pleas suggest that many ancient Romans
believed that even after death, disembodied spirits could partake in
some kind of nourishment. But the nourishment likely was intended as
much for the surviving as for the dead. According to documentary evi-
7
Dunbabin, Roman Banquet, 109, 126, 140.
8
This monument described at length by Dunbabin, Roman Banquet, 1034.
9
See Koch, Roman Funerary Sculpture: Catalogue of the Collections, 2427.
10
CIL 6.26554. See the discussion of this and other texts in Keith Hopkins, Death
and Renewal (Cambridge: Cambridge University Press, 1983), 23334.
11
Petronius, Satyr. 71. See also Apuleius, Metam. 8.9. Note the sundial on the
Christian sarcophagus fragment (Fig. 4.11).
Dining with the Dead 117
dence, entertaining at the tomb was a fairly common practice and
gatherings, although intended to be solemn, could sometimes get a bit
rowdy, especially when mourners ate and drank to excess. Leftovers,
once the party had broken up, might be gathered later by the desti-
tute.
12
Thus the living honored the dead by dining with them, a ritual
that obliged them also to confront (and perhaps boldly laugh in the
face of) the transitory nature of life and its ephemeral pleasures.
Traditional Romans celebrated funeral banquets at the graves of
family and friends, first on the day of burial (silicernium) and then
again on the ninth day after the funeral (cena novendialis) which indi-
cated the end of the official mourning period.
13
Very little data exists to
indicate the actual kinds of food consumed at the grave. Cups, loaves,
and fish appear in the imagery, as well as the heads of a pig and fowl, and
a joint of beef. Both archaeological and textual evidence indicate that
grain, wine, oil, incense and flowers were brought as offerings for the
dead, either scattered or poured on and into the tomb itself.
14
Ovid rec-
ommends that the gifts left for the dead be modest (a little scattered
grain, some salt, bread soaked in wine, loose violets, or flower garlands)
but added that neglecting the festival would court disaster. Another
12
See Dennis Smith, From Symposium to Eucharist: The Banquet in the Early
Christian World (Minneapolis: Fortress Press, 2003), 401; Jocelyn M. C. Toyn-
bee, Death and Burial in the Roman World (Ithaca, N.Y.: Cornell University
Press, 1971; repr., Baltimore: Johns Hopkins University Press, 1996) 51; and
Jon Davies, Death, Burial and Rebirth in the Religions of Antiquity (London:
Routledge, 1999), 13954.
13
In the Greek-speaking part of the ancient world, a common practice was to cel-
ebrate the meal on the third day after death. See discussion of the Roman festi-
vals in Toynbee, Death and Burial, 5051, 614.
14
In regard to the types of food consumed see Cyrille Vogel, Le poisson, aliment
du repas funraire chrtien? in Paganisme, Judasme, Christianisme: influences et
affrontements dans le monde antique, mlanges offers Marcel Simon (Paris: di-
tions E. de Boccard, 1978), 23343. Vogel argues that fish is the special food of
the funeral banquet, especially for Christians as it holds both eschatological and
eucharistic signficiance. See also Andrew McGowan, Ascetic Eucharists: Food
and Drink in Early Christian Ritual Meals (New York: Oxford University Press,
1999), 12735. Note here the funerary inscriptions of Abercius (second century;
see Margaret Mitchells chapter in this volume.) and Pectorius (fourth century) in
which fish, along with wine and bread, are part of a ritual meal. Cf. Johannes
Quasten, Patrology (Utrecht: Spectrum Publishers, 1966) I.17175 for trans-
lation. Platters of fowl (chickens?) appear on a late-third-century sarcophagus in
the Museo Gregoriano Profano in the Vatican, and the leg of a calf on a fragment
in the Museo Pio Cristiano, inv. # 31662 (Lat. no. 117).
118 Robin M. Jensen
meal could be held on the fortieth day after death, and then in sub-
sequent years on the deceaseds birthday (dies natalis), and on the an-
nual festival set aside for commemoration of ancestors (the parentalia),
from February 13 to 21. Furthermore, during those days, marriages
were prohibited, the family hearth kept unkindled, business suspended,
and the temples closed with altars incense-free.
15
Another festival known
as dies rosationis or rosalia in May or June, when family members
brought roses to the graves of their kinfolk, may be the reason for these
flowers to appear in fresco or mosaic on tomb walls or coverings.
16
As Ovid warns, neglecting these rituals could be perilous.
17
At least
some ancients tried to observe these festivals with decorum and rever-
ence making them occasions for paying pious respect to the dead.
Ausonius of Bordeaux, tutor to the Emperor Gratian, prefect of Gaul,
consul in 379 C.E., and a Christian convert, wrote a series of memorial
verses to honor deceased ancestors, the Parentalia. The respectful but
grief-filled sentiments expressed in this document suggest that for him,
at least, this was a bittersweet time of recollection and filial devotion,
rather than a time for a drunken revel.
18
One of Ausoniuss epitaphs (written for the tomb of a happy man)
commands the passing stranger to sprinkle my ashes well with un-
mixed wine and sweet-scented oil of nard and to bring balsam and
roses to his tearless urn.
19
Such libations were accommodated by
providing tombs with holes or pipes for pouring liquids (Fig. 4.7). These
feeding tubes projected above graves, often made from the necks of
broken or even buried amphorae which held the remains (both cre-
mated and inhumed) of the deceased. Grave markers (cippi ) often had
attached platforms for food offerings sometimes with indentations in
the shapes of the foods themselves (bread or fish, especially), which
suggest a representative as well as actual offering (Fig. 4.8). Archaeol-
ogists also have found permanent tables, either semicircular (stiba-
dia), biclinia, or triclinia forms, attached to a family mausoleum or in a
common area of the cemetery, to facilitate the meal shared by the liv-
15
Ovid, Fasti 2.53370.
16
On the rosalia see Rosalia, Pauly-Wissowa, ser. 2, vol. 1 (1920), cols. 1111
1115; On the practice of bringing roses to the tomb see inscriptions in Hermann
Dessau, ILS, 7213, 7258, 8369, 8370, 8371, 8372, 8373, and 8374 (et rosas suo
termpore deducerent).
17
This is also clear in Porphyry (on Horace) Ep. 2.2.209 and in Apuleius, Metam. 8.9.
18
Ausonius, Parent. passim.
19
Ausonius, Epit. 31; cited also in Toynbee, Death and Burial, 63.
Dining with the Dead 119
4.7. Mensa from the Capitoline Museum,
Rome. (authors photo)
4.8. Mensa from Algeria, from area of
Timgad. (photo: Michael Flecky)
120 Robin M. Jensen
ing (Fig. 4.9).
20
Elaborate vessels, dishware, and other furniture could
be provided both as grave goods and as utensils to be used by visiting
relatives, or even painted onto the walls of very important tombs
(Fig. 4.10). Some burial sites included hearths for cooking, and cem-
eteries even provided water fountains and channeling systems for
purification and post-meal washing up.
21
The diversity of funerary banquet facilities or equipment is paral-
leled by the variety of iconography depicting or pertaining to the prac-
tice. The kind and quality of the monuments or furnishings differs ac-
cording to date and geography, but also reflects the social status and
wealth of the patron. Similarly, the Roman sarcophagi bearing the most
elaborate banquet scenes have been shown to come from the middle
classes or like Trimalchio freedmen or their descendents, a case well
demonstrated by the graves discovered in the necropolis under the Vati-
can car park.
22
Dunbabin astutely argues that wealthy freedmen, un-
able to exploit the ancestral images that played so great a part in the
Roman upper class funeral chose to portray themselves as enjoying a
luxurious banquet as a part of their compensatory funeral display.
23
Despite their rejection of many other aspects of pagan culture,
Christians continued these traditional funerary practices probably
because they did not view giving honors to their dead relations as hav-
ing anything to do with the pagan god, religion, or idols. Meanwhile,
church leaders were attempting to transfer these customary practices
from cemetery to the church by encouraging mourners to observe the
anniversary of a loved ones death with alms and eucharistic offerings
rather than food shared at a tomb.
In his treatise On Monogamy, Tertullian remarks on the duties of
a wife to her dead husband and makes passing reference to the tradi-
20
The semicircular shaped table was especially associated with dining al fresco ac-
cording to Dunbabin, Triclinium and Stibadium, 1325.
21
On water in cemeteries see also the discussion of tombs at Isola Sacra in Toyn-
bee, Death and Burial, 136; also in Tipasa, Paul-Albert Fvrier, A propos du
repas funraire: culte et sociabilit, CahArch 26 (1977): 2945. An inscription
from Rome records the transfer of rights in a tomb that included 24 urns, the use
of a kitchen, and a well for drawing water, CIL 6.14614.
22
On the issue of social class and representations of actual feasts, see Andrew
Wallace-Hadrill, Imaginary Feasts: Pictures of Success on the Bay of Naples,
in Ostia, Cicero, Gamala, Feasts and the Economy: Papers in Memory of John H.
DArms (ed. Anna Gallina Zevi and John H. Humphrey, JRASup. 57; Port-
smouth, R.I., 2004), 10926.
23
Dunbabin, Roman Banquet, 11213.
Dining with the Dead 121
4.9. Biclinium from Isola Sacra. (authors photo)
4.10. Tableware on tomb of Vestorius Priscus, Pompeii. (authors photo)
122 Robin M. Jensen
tional funeral customs that he assumes even Christian women would
observe: Indeed, she prays for his soul, and requests refreshment in
the waiting period (refrigerium interim) for him, and companionship
(with him) in the first resurrection; and she makes offerings on the
anniversaries of his falling asleep. For, unless she does these things, it is
as if she has truly divorced him.
24
In another place, Tertullian admon-
ishes a widower against remarrying since doing so would require the
husband to offer the annual oblations for the first, in the presence of
the second an awkward situation.
25
The anniversary offerings that
Tertullian mentions might refer to food or liquids left at the grave as
nourishment for the body as well as the soul awaiting the general res-
urrection (and reaffirmation of the marriage bond in heaven), but it
might also refer to gifts brought to the church.
26
In his treatise On the
Soldiers Crown, he describes such rituals as baptism, eucharist, fast-
ing, and prayer, and includes occasions when we make offerings for
the dead.
27
In his Treatise on the Soul, Tertullian mocks the offerings
brought by pagans to the tombs, as being more for the enjoyment of
the living than for the benefit of the departed, and notes that while
reclining at a sumptuous funeral banquet, no one would dare to speak
24
Tertullian, Mon. 10.4, authors translation. The verb here (offert) does not make
it clear, however, what or where she offers. See also Uxor. 2.8.
25
Tertullian, Exh. cast. 11
26
The Latin word refrigerium actually means refreshment or a cooling off but
in the context of a tomb inscription, it referred to a state of blessed rest or re-
pose. In early Christian texts it probably referred to the time of waiting before
the general resurrection, e.g., the term refrigerium interim which seems to have
been coined by Tertullian. See also Tertullian, Test. 4 where he describes the
pagan customs of wishing the bones and ashes a bene refrigeria and bene requi-
escat. See Tertullian, Marc. 4.32.13, where interim refrigerium refers to the rest
of Lazarus in the bosom of Abraham (Luke 16.1931). See also Passio. Perp. 8;
Cyprian Epp. 6.3.1, 30.7.2; and Augustine Gen. litt. 8.5. It appears on many early
Christian epitaphs. See Enrico Josi, Refrigerium, EC 10, 62731; Andr Par-
rot, Le refrigerium dans lau del (Paris: Librairie E. Leroux, 1937); and Chris-
tine Mohrmann, tudes sur le latin des chrtiens, II (Rome: Edizioni i storia
e letteratura, 1961), 8192. The idea that the refrigerium may refer specifically
to the funeral meal (actual refreshments) or a heavenly banquet is suggested
by some inscriptions in the catacombs including at the triclia at S. Sebastiano,
see discussion below (and fn. 29). On the early Christian belief in the time after
death and its reflection in the iconography of Christian burial places see
Alfred Stuiber, Refrigerium Interim: Die Vorstellungen vom Zwischenzustand und
die frhchristliche Grabeskunst (Bonn: Peter Hanstein Verlag, 1957).
27
Tertullian, Cor. 3 (oblations prodefunctis pro nataliciis annua die facimus).
Dining with the Dead 123
ill of the dead, since they are thought to be, in some manner, present
at the party.
28
On the other hand, Tertullian also condemns those who
do these things, since they are associated with idolatry and the feasts of
the pagan gods. Citing the text from 1 Cor 10.21, he asserts that offer-
ing funeral oblations or partaking in what is offered at the banquet is
akin to sitting down at the table of the demons.
29
Paintings of banquet scenes, found on walls in the Christian cata-
combs of Rome or carved on early Christian sarcophagi demonstrate
that Christians continued to share the traditional meals with the dead
(Fig. 4.11). Although historians of the past have attempted to find a spe-
cifically Christian significance either a liturgical reference or a repre-
sentation of a biblical scene in these images their obvious similarities to
parallel pagan paintings counter their being interpreted as portraying a
Christian agape or eucharist, Christ and his disciples at the Last Supper,
or referring to the gospel story of the multiplication of the loaves and
fish. The more straightforward conclusion, that these are scenes of ac-
tual funeral banquets, or evocations of the future paradisiacal banquet
like that of Vibia (Fig. 4.6), better explains the compositions in any case,
which either display or lack key details necessary for those identifica-
tions (e.g., the fish on the table as well as the seating arrangement argues
against these as representations of a late-third- or early-fourth-century
28
Tertullian, Test. 4. Here he comments on the inconsistency of believing that the
dead are beyond feeling, but at the same time making them offerings and worry-
ing about their opinion. See also Res. 1.
29
Tertullian, Spec. 13. See also Apol. 13.
4.11. Sarcophagus from Museo Pio Cristano, Rome. (authors photo)
124 Robin M. Jensen
eucharist; the number of diners makes no sense for a scene of the Last
Supper).
30
Some particularly well-known images, from the catacomb of
Peter and Marcellinus even give captions to some of the diners, calling
evocatively named women servants (Irene and Agape) to bring them
more warm, mixed wine (Fig. 4.12). The group in this painting looks
anything but solemn or mournful, as they raise their glasses and order
refills.
31
Certain graffiti from the triclia under S. Sebastiano at the Mem-
oria Apostolorum (site of the translation of relics of both Peter and Paul)
specifically use the term refrigeria to mean either a funerary meal or
heavenly banquet, in either case perhaps hoped to have been shared with
the saints (Peter and Paul) as well as with deceased family members.
32
In addition to evidence from the paintings in the Christian cata-
combs in Rome are funerary inscriptions. For example, an epitaph of a
Christian woman named Aelia Secundula, dated to 299 C.E. from the
African province of Mauretania Sitifensis, gives insight into how these
funeral meals might have been observed, in this region; it describes the
placing of a stone table, laying out of food and drink, reciting of eu-
logies and telling of stories about the deceased, long into the night:
Memoria Aeliae Secundulae.
Funeri mu[l]ta quid[e]m condigna iam misimus omneS,
Insuper ar[a]equ[e] deposte Secundulae matrI
Lapideam placuit nobis atponere mensaM,
In qua magna eius memorantes plurima factA,
Dum cibi ponuntur calicesq[ue]. E[i] copertaE,
Vulnus ut sanetur nos rod[ens] pectore saeuuM
Libentur fabul[as] dum sera redimus horA
Castae matri bonae laudesq[ue], vetula dormiT
Ipas, q[uae] nutri[i]t, iaces et sobriae semper.
V[ixit] a[nnis] LXXV a[nno] p[rovinciae] CCLX,
Statulenia Iulia fecit.
30
Identifications of these images as portraying eucharists or agape meals or
alternately of the Last Supper or the multiplication of loaves are common-
place. For instance see Graydon Snyder, Ante Pacem: Archaeological Evidence of
Church Life before Constantine (Rev. ed.; Macon, Ga.: Mercer University Press,
2003), 12426; Such identifications have been used to argue that certain Chris-
tians were still celebrating agape meals into the early fourth century, or that
women were celebrating the eucharist in certain instances. For example see Do-
rothy Irvin, The Ministry of Women in the Early Church, Duke Divinity
School Review 45.2 (1980): 7686. See also the articles noted above, fn. 6.
31
See Fvrier, A propos du repas funraire: culte et sociabilit, which considers
the message of conviviality imparted by these particular images.
32
See examples, discussion, and bibliography in Snyder, Ante Pacem, 25158.
Dining with the Dead 125
4
.
1
2
.
C
a
t
a
c
o
m
b

o
f

P
e
t
e
r

a
n
d

M
a
r
c
e
l
l
i
n
u
s
,

b
a
n
q
u
e
t

w
i
t
h

I
r
e
n
e

a
n
d

A
g
a
p
e
.
(
P
h
o
t
o
:

E
s
t
e
l
l
e

S
.

B
r
e
t
t
m
a
n
,

I
n
t
e
r
n
a
t
i
o
n
a
l

C
a
t
a
c
o
m
b

S
o
c
i
e
t
y
.

U
s
e
d

w
i
t
h

p
e
r
m
i
s
s
i
o
n
.
)
126 Robin M. Jensen
To the memory of Aelia Secundula
We all sent many worthy things for her funeral.
Further near the altar dedicated to Mother Secundula,
It pleases us to place a stone table
On which, we placing food and covered cups,
Remember her many great deeds.
In order to heal the savage wound gnawing at our breast,
We freely recount stories at a late hour,
And give praises to the good and chaste mother, who sleeps in her old age.
She, who nourished us, lies soberly forever.
She lived to be seventy-five years of age, and died in the 260th year of the
province.
Made by Statulenia Julia.
33
Mensae in cemeteries are found throughout the Roman world. In ad-
dition to Italy and Africa, examples can be seen Spain, Dalmatia, Ger-
many, and in the catacombs of Malta (Fig. 4.13).
34
Somewhat to the west
of Aelias tomb is the ancient site of Tipasa in Mauretania Caesarien-
sis, which includes two huge cemetery areas outside the eastern and
western city walls. These areas were equipped with tables for memorial
feasts, many of them in very good condition and covered with mosaic
decoration, along with cisterns and systems for drawing water and
channeling it onto the tombs. In addition to open-air burials, these
areas included a martyrs shrine (Sta. Salsa), and a basilica built pri-
marily to house funeral banquets and private memorial services. One
of these, built around 400 C.E. by the bishop Alexander to provide a
place for his own tomb, also contained the burials of his nine prede-
cessors as well as a number of other, probably more ordinary, burials.
A structure roughly 23 by 14 meters in size, this basilicas nave and
aisles are filled with graves and feature several semicircular stone
couches for the celebration of funeral meals (Fig. 4.14).
35
A stone mosaic mensa cover from the late fourth century, recovered
from an nearby area known as the necropolis of Matares, contains a
33
Diehl, ILCV 1. 1570; CIL 8.20.277. This inscription cited also by Johannes
Quasten in his very helpful article, Vetus Superstitio et Nova Religio: The Prob-
lem of Refrigerium in the Ancient Church of North Africa, HTR 33 (1940):
25366.
34
See the article by X. Barral i Altet, Mensae et repas funraires dans la pninsule
ibrique, ACIAC 9.2 (1978): 4969, for discussion of funerary menase in Spain.
35
On the site of Tipasa and these cemeteries see Paul-Albert Fvrier, Le culte des
martyrs en Afrique et ses plus anciens monuments, Corso di cultura sullarte Ra-
vennate e Bizantina 17 (1970): 191215.
Dining with the Dead 127
4.13. Mensa from St. Pauls catacomb in Rabat, Malta. (authors photo)
4.14. Mensa from Tipasa, basilica of Alexander. (authors photo)
128 Robin M. Jensen
legend embodying the optimistic spirit of such banquets. In addition
to images of fish, typical of North African mosaics, is a legend that
reads: IN DEO, PAX ET CONCORDIA SIT CONVIVIO NOSTRO.
In God (Christ), may peace and concord be on our banquet (Fig. 4.15).
As these cases show, although Christians continued to honor their
ordinary dead family members, they also extended these funerary
practices to honor deceased clergy (bishops in particular) and incor-
porated them into the cult of the saints individuals who were members
of the extended church family, but who also functioned as patrons and
intercessors.
36
Oral reading of the martyrs heroic deeds (acta), singing
songs of praise, and sharing food on a saints birthday (natalacius)
into heaven was a kind of communion with that holy person. Cyprian
of Carthage urges his congregants to record the days of martyrs
deaths, so that they might celebrate them afterwards with offerings and
sacrifices as well as festive meals.
37
Thus the martyrs celebrations were
noted on the church calendar and commemorated with sacrifices (pre-
sumably a eucharist), offerings, and a banquet. Such adaptation of the
funerary banquet allowed a special kind of communion with holy men
and women. Saints shrines came to be augmented with banqueting
facilities that could accommodate pilgrims bringing food offerings
to the tombs of their spiritual, rather than their biological, ancestors.
The inscription on a large stone slab, discovered in northwest Altava
(ancient Mauretania) identifies it as a mensa dedicated to St. Janua-
rius. The text, which is difficult to interpret, indicates that it might have
been placed in a church (basilica dominica) and used for the euchar-
istic liturgy.
38
The same evolution of ritual action, space, and table characterized
those other places in the Roman milieu where the practice of holding
banquets at the tomb was ancient and entrenched. As noted above, the
triclia in the Memoria Apostolorum on Romes Via Appia Antica dis-
plays graffiti produced by ancient pilgrims which invoke the blessings of
Peter and Paul on the funerary banquets (here refrigeria) celebrated at
36
Tertullian, Cor. 3, mentions making annual offering for the dead as birthday
honors.
37
Cyprian, Epp. 12.2.1; 39.3.1; see also Tertullian, Scorp. 7.2 on the singing of
songs in honor of the martyrs.
38
See Jean Marcillet-Jaubert, Les inscriptions dAltava (Aix-en-Provence: Gap,
ditions Orphrys, 1968), 3234; and Fvrier, Le cult des martyrs en Afrique,
191215.
Dining with the Dead 129
4
.
1
5
.
M
e
n
s
a

f
r
o
m

r
e
g
i
o
n

o
f

M
a
t
a
r
e
s

(
n
e
a
r

T
i
p
a
s
a
)
.

(
a
u
t
h
o
r

s

p
h
o
t
o
)
130 Robin M. Jensen
that place.
39
Another well-known example, excavated in an ancient
pagan and Christian cemetery beneath the Cathedral of Bonn, gives a
clear example of a matryrium equipped with a mensa. An enclosure had
been built around or near the sarcophagi of what were presumed to be
four saints, and funerary mensae and benches were discovered, probably
those used by late-third- or early-fourth-century pilgrims.
40
The trans-
formation of burial places into pilgrimage sites required architectural
modifications, gathering spaces, and mensae. These eventually became
chapels with eucharistic altars.
41
Such arrangements still exist today as
Christian visitors continue to hold services in the catacombs of Rome.
The area around these shrines became desirable for burials ad sanc-
tos. Thus burial continued in the proximity to the saints tombs well
after they were set apart as holy places. In the late fourth century Pope
Damasus began to identify the tombs and promote the cult of the mar-
tyrs in the catacombs of Rome, even composing epigrams which he had
inscribed on marble plaques and placed near the saints remains.
42
Pre-
sumably many of these places also included mensae that later would
have been used as small eucharistic altars for occasional celebrations.
For instance, the catacomb of Domitillas crypt of Veneranda features
a sarcophagus that appears to have been appropriated as a mensa de-
dicated to the cult of St. Petronilla. The fresco in the arcosolium over
this sarcophagus shows Petronilla escorting Veneranda into paradise
(Fig. 4.16). Presumably, family members commemorating their ordi-
39
The Memoria Apostolorum (part of the catacomb of St. Sebastian) probably dates
to the mid-third century. For general discussion and bibliography see Antonio Fer-
rua, La basilica e la catacombe di S. Sebastiano (Vatican City: Pontificia Commis-
sione di Archeologia Sacra, 1990); and Elisabeth Jastrz ebowska, Untersuchengen
zum christlichen Totenmahl auf Grund der Monumente des 3. und 4. Jarhhunderts
unter der Basilika des hl. Sebastian in Rom (Frankfurt am Main: P.D. Lang, 1981).
40
See bibliography in G. Snyder, Ante Pacem, 164. See also discussion of Salona in
Snyder, and bibliography.
41
For a (dated but traditional and illuminatingly pious) study of this phenomenon
see Ludwig Hertling and Englebert Kirshbaum, Die rmische Katakomben und
ihre Martyrer (Vienna: Verlag Herter, 1950); trans.: The Roman Catacombs and
Their Martyrs (Milwaukee, Wis.: Bruce Publishing, 1956), esp. chap. 3, The
Tombs of the Martyrs, 4986.
42
Epigrammata Damasiana. On Damasuss activities see Jean Guyon, Damase
et lillustration des martyrs: les accents de la devotion et lenjeu dun pastorale,
in Martyrium in Multidisciplinary Perspective (eds. Mathijs Lamberigts and
Peter van Deun; Louvaine: Peters, 1995), 15777; and Victor Saxr, Damase
et le calendrier des ftes de martyrs de lglise romaine, Saecularia Damasiana
(Rome: Pontificial Institute of Christian Archaeology, 1986), 5988.
Dining with the Dead 131
4.16. Veneranda with St. Petronilla, catacomb of Domitilla, Rome. (Photo:
Estelle S. Brettman, International Catacomb Society. Used with permission.)
132 Robin M. Jensen
nary dead sometimes had to share tight spaces with pilgrims arriving
to honor a saint.
Monumental examples this complex of tomb, dining room, and
martyrs shrine, include the extra-urban Roman cemetery basilicas of
St. Agnes, Ss. Peter and Marcellinus, and St. Lawrence. Large funerary
banquet halls erected at these sites, near to (but not on top of) the mar-
tyrium, housed those who came to honor family members buried
within (or nearby) and provided a venue for the saints commemor-
ation a time when crowds would fill the hall, celebrating with songs
and drink, sometimes to bawdy and inebriated excess.
43
This adap-
tation of the funerary banquet created certain new problems that
church officials needed to resolve.
The structures built to serve these festivities were not regular parish
churches like those built inside the walls of the city for regular Sunday
eucharistic celebrations. Although many were equipped with altars and
baptismal fonts along with mensae and couches for the purpose of
serving both private funeral banquets and public saints feasts, they
neither had resident clergy, nor were the seat of a bishop. The presence
of an altar in the large hall as well as one inside the small shrine enclo-
sure indicates that a eucharist could be held either place, perhaps
moved to the hall when the size of the gathering required a larger
space. At the same time, this larger and more removed space could ac-
commodate the (sometimes rowdy) activities associated with the vigil.
Jerome complained to one of his correspondents that night-watches in
the basilicas of the martyrs were spoiled by young men and women of
bad reputations who behaved scandalously.
44
Before long, church offi-
cials attempted to put a stop to these practices, which they saw as dis-
respectful, disorderly, and fundamentally profane.
St. Peters, undoubtedly the most popular pilgrimage church in
Rome, was likely built originally to serve as a funeral hall (rather than
as the papal basilica that it appears today).
45
Augustine decries the
regular spectacle of inebriated pilgrims coming to visit the saint. Al-
43
See Richard Krautheimer, Mensa-Coemeterium-Martyrium, CahArch 11
(1960): 1540 reprinted in Studies in Early Christian, Medieval, and Renaissance
Art (New York: New York University Press, 1969), 3558.
44
Vig. 9.
45
On the reconsideration of the original structure of St. Peters as a funerary
church with saints shrine, see Alberto C. Carpiceci and Richard Krautheimer,
Nuovi dati sullantica basilica di San Pietro in Vaticano, Bolletino darte del
ministero per i beni culturali e ambientali 81 (1996): 184.
Dining with the Dead 133
though such behavior was forbidden, he explained, the prohibition
was hard to enforce because the place was distant from the residence
of the bishop and in so large a city of people living according to the
flesh.
46
Simultaneously, however, St. Peters shrine served as a venue
for banquets honoring dead members of prominent Christian families.
Paulinus of Nola, in a condolence letter to the Roman senator Pamma-
chius upon the death of his wife, mentions the funerary feast in the
basilica of the apostle and commends his friend for using the occa-
sion for a charitable act. For, instead of only inviting members of his
own social class, Pammachius had opened the doors to a large crowd
of the hungry poor. This, Paulinus approvingly comments, recalled the
story of the multiplication of the loaves and fish. By feeding bodies in
need, Pammachius has garnered Gods good will and also refreshed
the soul of his dead wife.
47
An earlier effort to turn rowdy or exclusive funeral feasts either
into more sober occasions, or into opportunities to show charity, are
recorded in the Apostolic Constitutions, which instructs Christians on
how to conduct decorous and charitable funeral banquets: Let the
third day of the departed be celebrated with psalms, and lessons, and
prayers, on account of him who rose within the space of three days;
and let the ninth day be celebrated in remembrance of the living, and
of the departed, and the fortieth day according to the ancient pattern:
for so did the people lament Moses, and the anniversary day in memo-
ry of him. And let alms be given to the poor out of his goods for a
memorial of him. The document further urges those who attend their
memorials to feast with good order and refrain from drinking to ex-
cess.
48
Dated to the early fourth century, Constantines Oration to the
Assembly of the Saints contrasts the commemoration of a Christian
martyr with pagan funerary festivities, pointing out that the Christian
ceremonies include a sacrifice of thanksgiving in honor of the saint,
a bloodless, harmless sacrifice with neither frankincense nor fire but
only enough pure light (i.e., torches and candles at the tomb) to sat-
isfy the assembled worshipers. The oration further praises Christians
46
Augustine, Ep. 29.1011.
47
Paulinus, Ep. 13. 1114. Pammachiuss wife, Paulina, was the daughter of Paula
and sister to Eustochium and Blesilla. After her death, Pammachius went into a
monastery and dedicated his life to aiding the poor.
48
Ap. Const. 8.4244, trans. ANF 7, 498.
134 Robin M. Jensen
for transforming funeral feasts into occasions for a charitable display,
while subtly acknowledging the tenaciousness of the old, more self-in-
dulgent behaviors: Many too there are whose charitable spirit leads
them to prepare a temperate banquet for the comfort of the needy,
and the relief of those who had been driven from their homes; a custom
which can only be deemed burdensome by those whose thoughts
are not accordant with the divine and sacred doctrine.
49
This kind of
practice may have grown out of those earlier agape meals Tertullian
refers to as those feasts we give for the relief of the poor.
50
Gradually the memorial banquet function of the halls was repressed
and abandoned, while the adjacent buildings that included the actual
saints shrine were enlarged and transformed into basilicas dedicated
to martyrs whose relics were placed under their main altars. In this way
funeral mensae in cemeteries were adapted and moved inside of churches
and funeral rites began to include a eucharistic celebration at the
table of the saint, rather than a meal at the table of the dead. The
practice of celebrating a eucharist at an actual funeral (rather than the
traditional memorial feast at the grave) is not mentioned in the literary
sources before the end of the fourth century and then only indirectly.
A council of African bishops forbade the celebration of a eucharist in
the presence of a corpse, specifically prohibiting the practice of putting
the consecrated bread into its mouth.
51
Other decrees allow the euchar-
ist to be offered as part of the burial only if participants had fasted
(prior to the first meal of the day).
52
Augustine, describing Monicas
funeral, notes that the practice in Italy was different than that of
Africa, in that they celebrated the sacrifice of our redemption at the
tomb, in the presence of the corpse.
53
Meanwhile, as the saints festival was moved inside of the church,
the termmensa gradually came to refer to any eucharistic table or altar,
but especially denoted an altar found at a major saints shrine.
54
That
49
Orat. Const. 12; trans. NPNF 1, 57071. This oration, appended to Eusebiuss
Vita Const. is not certainly an original work of the Emperor himself.
50
Tertullian, Apol. 39 (siquidem inopes quosque refrigerio isto iuvamus). See Fv-
rier, A propos du repas funraire, 405, in which he argues that this transition
from private banquet to occasion for almsgiving can be seen clearly at Tipasa.
51
Con. Hippo A 393 c. 4, CCSL 149.21.
52
Ibid.
53
Augustine, Conf. 9.12.32 (cum offerretur pro ea sacrificium pretii nostri iam iuxta
sepulchrum posito cadauere).
54
See Krautheimer, Mensa-Coemeterium-Martyrium, passim.
Dining with the Dead 135
the early mensae were mostly made of stone may explain why stone
altars began to replace wooden ones in the late fourth and early
fifth centuries.
55
A well-known effort to move a saints celebration inside a church
where the memorial ritual would be conducted with decency and de-
corum is illustrated by the way the fore-mentioned Paulinus of Nola,
managed his own pilgrimage site. Paulinus tries to discourage the
rowdy revelers at Felixs shrine in Nola by luring them inside the basil-
ica proper by the novel addition of paintings on the walls. Address-
ing a visitor to the shrine, he outlines his problem:
Now the greater number among the crowds here are country-folk, not without
belief but unskilled in reading. For years they have been used to following pro-
fane cults in which their god was their belly, and at last they have turned as con-
verts to Christ out of admiration for the undisputed achievements of the saints
performed in Christs name. Notice in what numbers they assemble from all the
country districts, and how they roam around, their unsophisticated minds be-
guiled in devotion See how they now in great numbers keep vigil and prolong
their joy throughout the night, dispelling sleep with joy and darkness with
torchlight. I only wish they would channel this joy in sober prayer and not in-
troduce their wine cups within the holy thresholds Their naivety is uncon-
scious of the extent of their guilt, and their sins arise from devotion, for they
wrongly believe that the saints are delighted to have their tombs doused with
reeking wine.
Paulinus then offers his solution pictures on the walls as a competing
attraction:
This was why we thought it useful to enliven all the houses of Felix with paint-
ings on sacred themes, in the hope that they would excite the interests of the
rustics by their attractive appearance, for the sketches are painted in various co-
55
For documentary evidence that early altars were made of wood see Augustine,
Cresc. 3.47, Ep. 185.27; Optatus, Donat. 6.1; Athanasius, H. Ar. 56. Sigma-
shaped stone tables, found in churches (especially in Provence and Africa) and
often referred to as agape tables may have been used as offering tables rather
than as eucharistic altars. See W. Eugene Kleinbauer, Table Top with Lobed
Border, in Age of Spirituality: Late Antique and Early Christian Art, Third to
Seventh Century (ed. Kurt Weitzmann; New York: Metropolitan Museum of Art
1979), 63738. On the transformation of altars see Catherine Metzger, Le
mobilier liturgique, in Naissance des arts chrtiens (eds. Nol Duval et al.; Paris:
Ministre de la culture et de la communication, 1991), 25667; X. Barral i Altet,
Mensae et repas funraire, 678; H. Leclercq, Autel, DACL 2 (1924):
315589; and Johann P. Kirsch and Theodor Klauser, Altar, RAC 1 (1950):
33454.
136 Robin M. Jensen
lors. Over them are explanatory inscriptions, the written word revealing the
theme outlined by the painters hand. So when all the country folk point out and
read over to each other the subjects painted, they turn more slowly to thoughts
of food, since the feast of fasting is so pleasant to the eye. In this way, as the
paintings beguile their hunger, their astonishment may allow better behavior to
develop in them as they gape, their drink is sobriety, and they forget the long-
ing for excessive wine.
56
In his Confessions, Augustine recalls another campaign to bring the
celebration of the saint inside the church proper (and the proper
church) and to squelch the practice of Christians banqueting in
cemeteries. Monica, carrying food to a martyrs shrine in Milan (as
was her former custom in Numidia), was stopped by the doorkeeper
with orders from Ambrose to bar anyone from bringing in food and
drink. Augustine comments on how hard it was for her to discontinue
a custom which she had long practiced, one motivated by devotion
rather than personal pleasure, and even seemed a little surprised at
Ambroses influence over his mother (one her son apparently did not
have). The bishop of Milan successfully admonished her to turn her
pagan custom into a pious celebration inside the church where she
would attend a eucharistic banquet rather than leave food at a grave.
She happily abstained, offering her gifts instead to the needy and
bringing a heart full of purer vows to the memorials of the martyrs.
57
As in Nola, this incident illustrates a process that would eventually
lead to saints relics being placed under the main sacramental altar
thereby joining all three feasts (memorial, martyrs feast, and euchar-
istic sacrifice).
Official attempts to transform the ancient practice of dining with the
dead by converting boisterous and unruly celebrations at tombs into
sober and respectful liturgies in churches may have underestimated the
difficulty of such an undertaking. Such efforts met resistance, some-
times overt and riotous, and sometimes passive and merely stubborn.
Augustine himself, attempting to curb the wild parties that character-
ized the feast of Cyprian in Carthage, or the feast of Leontius, the mar-
tyr saint of Hippo, realized that he might as well compromise and allow
family memorial meals to continue, at least. Nevertheless, in his treat-
56
Paulinus, Carm. 27.542, trans. P. G. Walsh, The Poems of Paulinus of Nola
(Ancient Christian Writers Series 40; New York: Newman Press, 1975), 290291.
57
Augustine, Conf. 6.2.
Dining with the Dead 137
ise, The City of God, Augustine praises those better families who had
abandoned the practice.
58
Documents recounting the progress of a more complicated case, that
of the Mensa of Cyprian in Carthage, vividly show how difficult it was
for church leaders (in this case Augustine of Hippo and Aurelius of
Carthage) to gain control of the festival of a beloved local saint. Cy-
prian, the bishop of Carthage, martyred in 258, was the perhaps the
most important and revered martyr of African Christians holy to Do-
natists and Catholics alike. His shrine was the destination of pilgrims
from all parts of the region, and his feast day, September 14, was the
occasion for a momentous and sometimes riotous celebration.
In one of more than a dozen extant sermons he preached on the vigil
or feast of Cyprian between 394 and 419,
59
Augustine refers to the
saints shrine in Carthage by its traditional name, the Mensa Cypriani.
Although its location is no longer certain, it would have been well
known to anyone who visited Carthage in Augustines time. This par-
ticular shrine, one of at least three dedicated to Cyprian, was con-
structed at the place known as the Ager Sexti, the site of Cyprians
execution by the Roman governor, on grounds just behind his own
residence. Cyprians beheading was witnessed by a throng of his fol-
lowers, who rushed to dip cloths in his blood and who bore his body
away in triumph for burial in the cemetery of Macrobius Candidianus
on the Mappalian Way.
60
Concerned to disabuse his audience of certain misunderstandings,
Augustine explains that the shrines name The Table of Cyprian
was given because Cyprian had been martyred at that place, not be-
cause he had dined there: And because by this very sacrifice of himself
he prepared this table; not as one on which to feed or be fed, but as one
on which sacrifice might be offered to God, to whom he offered his
very self. Therefore while it was called Cyprians table, this mensa
58
Augustine, Civ. 7.26; much of the following discussion was covered more briefly
in Quasten, Vetus Superstitio. Note, however, that some of Augustines ser-
mons hint at a Christianized version of the Parentalia, oriented more toward a
churchly celebration of days for remembering the departed on which sermons
might be preached on Christian beliefs about death and resurrection. See Serm
172, 173.1 and possibly 361.
59
Augustine, Serm. 308A, 309, 310, 311, 312, 313, 313A, 313B, 313C, 313D, 313E
and 313F; Enarrat. Ps. 32.2, 323.
60
Acta Proconsularis (Sancti Cypriani) 5 (CSEL 3.3), 11314; trans. H. Musurillo,
The Acts of the Christian Martyrs (Oxford: Clarendon Press, 1972), 1734.
138 Robin M. Jensen
was actually Gods table. The very place where Cyprian was once sur-
rounded by his persecutors is now a table surrounded by worshipers.
61
Such an explanation indicates that Augustine meant a eucharistic
altar when he spoke of the mensa, a equation strengthened in two other
(possibly earlier) sermons given at this same shrine where he identified
an altar (altare) honoring Cyprian but raised up for God.
62
He also
instructed his listeners in the difference between the honor paid to
martyr and that paid to God, and admonished them to be respectful by
celebrating the saint in a holy way as Christians. After all, he says,
we have not erected an altar to Cyprian as though he were God, but
we have made an altar to the true God out of Cyprian.
63
Here Augustine tries to rein in certain aspects of the martyr cult
by carefully redefining the purpose of this popular pilgrimage site as a
basilica dedicated to the worship of God, not to disrespectful feasts
in honor of a saint. The name of this particular shrine, however, must
have given him some qualms since the word mensa never applied to
place of sacrifice or the shedding of blood, but rather only used for an
ordinary dining table, a funerary table, or some other ritual or cultic
table.
64
Augustines equation of the words mensa and altare suggests
that he was eager to transform the cult through its ceremonial cel-
ebration in (Gods) eucharist rather than a festival banquet. Further-
more, when in other places Augustine applies the term mensa Domini
(or mensa Dominica) to the eucharist, he contrasted it with mensa
diaboli, a translation of trapezes kuriou found in 1 Cor 10.21.
65
Augustine may have attempted to rename the shrine itself, referring to
it elsewhere as the Domus Sancti Cypriani.
66
The shrine known as the Mensa Cypriani had existed for more than a
century by the time Augustine preached his sermons there, and must
have been called by this name from the beginning. As noted above, the
61
Augustine, Serm. 310.2.
62
Ibid., Serm. 313.5; 313A.5. Hill argues for a date of 419, although 406 is possible
for the former and 401 for the latter.
63
Ibid., Serm. 313A.5.
64
See Cicero, Leg. 2.26.66; Virgil, Aen. 2.995 mensaeque deorum; Cic. Har. resp.
57; Pliny, Nat Hist. 25.59, Iovis mensa.
65
See Augustine, Peccat. Merit. 1.24.32; Ep. 149.16, Ser. 31.1.2 for example.
66
Author has adopted this title from Enarrat. Ps. 32.3, where Augustine may
in fact be using it to refer to the Mensa Cypriani in order to make a point. On
the matter of terminology, mensa = altare, see R. Krautheimer, Mensa-Coeme-
terium-Martyrium, 4950.
Dining with the Dead 139
terminology is puzzling, since the shrine commemorated the site where
Cyprian was executed, not the place where he was buried. Located in
the Ager Sexti (the Estate of Sextus), where the proconsul Galerius
Maximus was temporarily residing for reasons of his health, it must
have been built in an open field outside the city walls.
67
Cyprians tomb
was in another pilgrimage site, the basilica known as the Mappalia,
also outside the city walls in the suburban cemetery of the Procurator
Macrobius Candidianius, on the Mappalian way (near the fish-
ponds).
68
Based on traditional funerary practices, one would expect
this shrine to have a funerary mensa, and perhaps it did, even though its
name does not reflect that possibility. In any case, by Augustines time,
the shrine presumably also held an altar for eucharistic celebrations.
Sources indicate that Augustine preached in both places on the feast
of Cyprian. Based on textual evidence it seems likely that he preached
more than a dozen sermons in Carthage, five at the Mappalia and at
least eight at the Mensa Cypriani. The sermons at the Mappalia were
likely preached on the evening of the feast (vigil), while the Mensa was
his venue on the next day thereby distinguishing the two shrines by
the type of commemorative celebration held in each.
69
The evening cel-
67
Acta Proconsularia (Sancti Cypriani), 2 and 5, as noted above. The Mensa also is
mentioned in Augustine, Enarrat. Ps. 80 (4, 23), which seem to have been
preached there on Cyprians feast day.
68
About the two basilicas, see Victor of Vita, Hist. Pers. Af. Prov. 1.16. Their pres-
ent-day locations (ruins) are still debated; see Othmar Perler, Les Voyages de
Saint Augustine (Paris: tudes Augustiniennes, 1969), 42021; Liliane Ennabli,
Les inscriptions funraires chrtiennes de la basilique dite de Sainte-Monique
Carthage (Rome: cole Franaise de Rome 1975), 1216; and Yvette Duval,
Loca sanctorum: le culte des martyres en Afrique du IVe au VII sicle, vol. 2
(Rome: cole Franaise de Rome, 1982), 67577.
69
This corresponds to the identification of the place and time of his preaching of
Serm. 308A (clearly preached on the vigil) and Enarrat. Ps. 32.2, and might be
surmised of Serm. 311, 312, and 313C (if the definition of the dies natalis can be
extended to include the evening vigil), and Serm. 313F (the latter identified as
being preached in the evening). This pattern corresponds to the view of Maria
Boulding, The Works of Saint Augustine: Expositions of the Psalms III/15 (Hyde
Park, N.Y.: New City Press, 2000), that Enarrat Ps. 32.2 was preached in the vigil
at the Mappalia and then the Enarrat. Ps. 32.3 the next day at the Mensa. See
dating of the sermons by Edmund Hill, The Works of Saint Augustine, Sermons
III/9 (Hyde Park, N.Y.: New City Press, 1994), 12122 who suggests that Serm.
308A, 313C and 313F were all preached at the same time, thus arguing against
this pattern all three sermons being in the Mappalia: vigil, next morning, and
next evening.
140 Robin M. Jensen
ebration associated primarily with the Mappalia was particularly dis-
orderly and included dancing, singing impious songs, and excessive
eating and drinking.
70
In any case, Augustine and his fellow bishop of Carthage, Aurelius,
made efforts to at least control the celebratory excesses if not eradicate
the practice of festival banquets on saints days or altogether. This was,
no doubt, part of their larger program generally to contain and control
the popular martyr cult in North Africa, and to distinguish their con-
gregation from that of the Donatists with regard to the decorum by
which they honored their saints.
Augustines sermons allude to his desire to control the disorderly
partying at the Mappalia, which in his mind more dishonored than
honored the saint. He claims to have had some success. In a sermon
probably preached in the year 405, he says this place (Mappalia) was
once invaded by the pestilential rowdiness of dancers and resounded
with the singing of impious songs. But now, he says, based on the
initiative of Aurelius (our brother bishop), the abuse had stopped.
These things dont go on here any longer.
71
The change may have started around 392, when Augustine wrote to
Aurelius of Carthage, urging him to join him in reforming the martyrs
feasts.
72
Citing Pauls letter to the Romans, not in feasting and
drunkenness, not in fornication and impurity, not in strife and jeal-
ousy; rather, put on the Lord Jesus Christ and do not provide for the
flesh with its desires (Rom. 13.1314), Augustine laments the sacri-
legious feasting, drinking, and general foulness at the tombs of
saints and the places for the sacraments.
73
Grudgingly tolerating
such misbehavior in private funerals, Augustine concentrates on elimi-
nating such practices in public, religious spaces and official saints
commemorations. He adds that the churches of Africa were lagging
behind churches elsewhere, and is thus shamed by the relative laxity in
correcting these moral failures.
74
70
Augustine, Serm. 311.5.
71
Augustine, Serm. 311.56, as cited above. Serm. 313A urges people to behave as
Christians and in a holy way (cf. Enarrat. Ps. 32.5).
72
Augustine, Ep. 22.
73
This is the same text (Rom 13.1314) that Augustine cites in Conf. 8.12. It was
the passage he opened his bible to in the garden, the one that that finally con-
verted him.
74
He may be referring, in particular, to Ambroses suppression of the funerary cult
at Milan, see above.
Dining with the Dead 141
Although he believed that the remedy required the authority of a
council, he advises Aurelius to begin at home, since, he argues, other
churches will be embarrassed to retain what the church at Carthage
had corrected. Reminding Aurelius that as a deacon he had con-
demned such practices, he encourages him now to take firm, but not
harsh steps to follow up. Once these depraved celebrations in the mar-
tyrs shrines were eliminated, Augustine optimistically expects that the
dissolute funeral banquets held for the ordinary dead would also grad-
ually come to an end, and that mourners would voluntarily replace
the feasts with almsgiving to the poor and commemorate their dear
ones inside the church, rather than at the tomb.
75
Augustine thus re-
veals his larger program to transform ordinary funerals into chari-
table and pious celebrations within churches, and to get out of the
cemeteries.
Augustine was satisfied in his desire for this matter to come before a
council of African bishops. Canon 29 of the council held at Hippo in
393 probably represents a follow-up to Augustines letter to Aurelius:
Neither bishops nor clergy shall dine in the church, except when
necessary for the hospitality shown to travelers, but then the people
shall be prohibited from this kind of banquet as much as possible.
76
Although this canon offers no prohibition of banquets at family
tombs, it nearly caused that above-mentioned riot when it was imposed
on the church at Hippo during the feast of the martyr Leontius in 394.
Augustine describes his success in suppressing the protest in a letter to
Alypius of Thagaste, and summarizes the arguments of the opposition.
Having preached a few days earlier on the Gospel passage do not give
what is holy to dogs, and do not cast your pearls before swine, Augus-
tine believes he had made his point but unfortunately to only the few
who came out to hear the sermon.
Trying again, he preached the next day on the story of Jesus driving
the money changers from the Temple, and paralleled the den of
thieves with drunken revelers. On the next day, the feast day itself,
people still complained about the suppression of their celebration, ask-
ing why now? when others before Augustine had allowed the parties
to go on. Augustine responds that in earlier times, church officials
had to tolerate certain pagan practices because of the weakness of the
75
Augustine, Ep. 22.26.
76
Brev. Hipp. 29. CCL, CSLIX, 41
142 Robin M. Jensen
newly converted, who were used to their long-standing pleasures.
77
But
by now, he points out, Christians ought to be ready to live as true
Christians, and moreover no less pious than members of those other
churches across the sea. Apparently his words had good effect, and the
crisis was averted and the celebration was marked with both modesty
and piety and, as Augustine asserts kept quite differently from the
heretics (Donatists) in their basilica, where the customary carnal feast-
ing and drinking took place.
78
This exchange reveals Augustines mo-
tives for this reform, which includes his desire to show up the Donatists
as immoral in comparison with the pious Catholics.
79
Augustines position against the feasts of the martyrs was a compro-
mise, however. As he explains in his letter to Aurelius, the practice of
ordinary funeral feasts was much more entrenched and almost impos-
sible to stamp out. Toleration for certain kinds of private celebrations
was probably politically wise and although he disapproved, Augustine
probably turned a blind eye to those long-cherished traditions, es-
pecially as they offered some consolation to mourning family members.
Nevertheless, as noted above, Augustine recognizes and praises those
better families who had abandoned the practice.
80
At the same time, the legislation of the African church prohibited
the giving of the eucharist to a corpse, and even the celebration of the
eucharist in the presence of a corpse. Legislation further forbade the
eucharist as part of a funeral ritual that took place after midday (since
mourners could not be assumed to have fasted before receiving the sac-
rament).
81
The need for such rules show that by the late fourth century,
the eucharist taken in the church and at its altar rather than at the
77
In roughly contemporary exchanges with the Manichees, Augustine had to ac-
knowledge that many ignorant or still-superstitious Catholic Christians had
forgotten their vows to abstain from pagan practices and still drink to excess
over the dead and bury themselves over the buried in gluttonous funeral feasts
in the name of religion. See Mor. ecc. 34.75 (ca. 388); also Faust. 20.21 (ca. 397),
where he says that some things must be borne for a while since intemperance is
even worse than impiety. Here Augustine also insists that Christians distinguish
between sacrificing to the martyrs and sacrificing to God in memory of the mar-
tyrs. Worship, he proclaims, is due to God alone.
78
Augustine, Ep. 29
79
See also Optatus of Milevis, Donat. 3.4, where he mentions the multiplication of
martyrs altars and tables.
80
Augustine, Civ. 7.26
81
Council of Hippo 393, canon 4 (CCSL 149, 21); Brev. Hipp. 160.28 (CCSL 149,
41).
Dining with the Dead 143
grave must have become part of the ritual surrounding most Chris-
tian funerals, and that certain practices like the feeding of the conse-
crated elements of the sacrament to a corpse (viaticum) have a long
tradition.
82
And, finally, the eucharist itself never really ceased to be a
certain kind of funeral meal a meal at which a once-dead host is now
living and present.
Archaeological evidence also shows how entrenched the funerary
cult was. Given its antiquity it could not be eradicated. In fact, church
officials even today try to bring the saints festival into the church from
off the street, and admonish their parishioners to respect the dead with
proper funeral etiquette. Nevertheless ordinary people still eat and
drink at wakes, and they still get a little boisterous at festivals in honor
of their saints.
82
Prayers for the dead at the eucharistic feast, however, are attested much earlier.
See, for example Cyprian, Ep. 1.2.1, which specifies the naming of the dead at the
altar during the prayer of the bishop.
144 Robin M. Jensen
Sweet Spices in the Tomb 145
Deborah Green
Chapter 5
Sweet Spices in the Tomb:
An Initial Study on the Use of Perfume
in Jewish Burials
How might we account for the numerous perfume bottles (unguen-
taria) both glass and ceramic found at Jewish burial sites in Pales-
tine in the early centuries C.E.?
1
A few years ago, scholars thought
these might be lacrimaria (tear bottles), brought to burial sites by
mourners,
2
or balsamaria (balsam bottles), so named for the scented
oil the bottles were thought to have contained. Others considered the
bottles to be part of food and other offerings either buried with the
dead for use in the afterlife or brought to the dead for ancestor wor-
ship. Both of these theories rely upon the persistence of customs that
may be traced back to the First Temple and earlier periods.
3
Today,
1
I would like to thank Laurie Brink for including me in this exciting project. I
would also like to extend my appreciation to Annal Frenz and Karen Stern for
their responses to my paper at the May 2005 conference.
2
See Erwin R. Goodenough, Jewish Symbols in the Greco-Roman World (13 vols.;
New York: Pantheon Books, 19531968), 1:165 and Amos Kloner and Boaz Zis-
sou, The Necropolis of Jerusalem in the Second Temple Period (in Hebrew) (Jeru-
salem: Yad Izhak Ben-Zvi; The Israel Exploration Society, 2003), 59.
3
Kloner and Zissou, The Necropolis of Jerusalem, 59. For a discussion of Iron Age
ancestor worship and needs and nourishment in the afterlife, see Elizabeth Bloch-
Smith, Judahite Burial Practices and Beliefs about the Dead (Sheffield: Sheffield
Academic Press, 1992). For the Roman period in the land of Israel, see Byron
McCane, Roll Back the Stone: Death and Burial in the World of Jesus (Harrisburg,
Pa.: Trinity Press Internatonal, 2003), 1415. McCane gives the example of the
Kidron Valley tomb as reported by N. Avigad, A Depository of Inscribed Ossu-
aries in the Kidron Valley, IEJ 12 (1962): 112. McCane also discusses in detail
the possibility of an early Roman period cult of the dead, 4952. Related to the
cult of the dead is the mourning practice of cutting the body; this ritual is
forbidden in the Hebrew Bible (Lev 19:28) and is also referred to in m. Mak. 3:5.
146 Deborah Green
several archaeologists and historians believe that the perfume in the
bottles was employed to mask the scent of the decomposing corpse.
4
This conjecture, seemingly sound on its surface, is often derived either
from contemporary western cultural attitudes toward decomposing
corpses or from a literal reading of a single line from the Babylonian
Talmud. However, when analyzed critically and comprehensively, the
archaeological, textual, and biological evidence point to reasons other
than the utilitarian for burying the dead with spices and perfume. This
study examines relevant rabbinic texts and compares them to the ma-
terial remains of perfume bottles in order to elucidate the three phases
of Jewish burial in which spices, in the form of either perfume or in-
cense, may have been used; the phase of burial in which scent may have
been employed to mask the stench of a decomposing corpse; and the
reason, if not to cover odors, perfume bottles may have been interred
with the dead.
5
The Problem of Evidence
Before launching into this task, an overview of Jewish burial practices
would seem to be in order. However, such an endeavor can be an all-
consuming, perhaps even futile, enterprise. As Ian Morris has pointed
out, the material remains of ancient burials are simply the remnants of
funerals ritual acts steeped in meaning, tradition, and emotion. As
such, the miscellany of physical evidence cannot adequately illuminate
the process of the rituals or the significance attached to burials. In ad-
dition, archaeological remains, along with studies of epigraphic and
other literary evidence, provide only singular or fractionalized snap-
shots of burials and their attendant rituals located in a specific place
and period, making the inexact science of extrapolation a necessary
means for surmising anything about ancient burials.
6
These problems are aptly demonstrated in the study of ancient Jew-
ish burials in the land of Israel, where we can describe the evolution of
4
For example, McCane, Roll Back the Stone, 15, 48.
5
This initial study is also part of a larger project on the metaphors and interpre-
tations concerning scent, spices, perfume, and incense in the Hebrew Bible and
rabbinic midrash and their connection to realia.
6
Ian Morris, Death-Ritual and Social Structure in Classical Antiquity (Cam-
bridge: Cambridge University Press, 1992), 115.
Sweet Spices in the Tomb 147
particular practices but are left with many unresolved questions. For
example, in the First Temple period, rock-cut tombs appear to reflect
family burials with benches or beds cut into the rock in order to ac-
commodate decomposing corpses. A compartment under one of the
benches or a separate repository would be located in the tomb to house
the collected bones of ancestors.
7
In such cases, the bones were piled
together.
8
During the late Second Temple period, although most
tombs continued to be family structures, bodies were either buried di-
rectly in rock-cut niches or placed in coffins which were then placed
into niches. When the bones were collected, they were placed individ-
ually or with one or two other skeletal remains (often women with
children) into separate boxes called ossuaries.
9
Although we can see
the change over time in these burial practices, the surviving evidence
represents only a small portion of the buried populace most likely the
richest members of the community. We have no method to determine
how or in what manner the majority of the population were buried,
nor do we know the procedures or rituals leading up to the burial for
those entombed in the structures or for the rest of the population. Fin-
ally, although scholars have conjectured as to the meaning and signifi-
cance of the collection of bones into ossuaries, no primary text exists
that explains plainly why Jews performed such a rite.
10
By the third century C.E., most tombs are no longer family struc-
tures; instead, several families and those without familial neighbors
are buried together. Some tomb caves, such as those at Beth Shearim,
7
For example, the St. tienne tombs in Jerusalem (located at the cole Biblique).
Gabriel Barkay and Amos Kloner, Jerusalem Tombs from the Days of the First
Temple, BAR 12 (1986): 2239.
8
Bloch-Smith, Judahite Burial Practices, 146149.
9
Rachel Hachlili, Ancient Jewish Art and Archaeology in the Land of Israel
(Leiden: Brill, 1988), 97; L. Y. Rahmani, A Catalogue of Jewish Ossuaries in the
Collections of the State of Israel (Jerusalem: Israel Antiquities Authority, Israel
Academy of Sciences and Humanities, 1994), 16, 21; Gideon Avni and Zvi
Greenhut, The Akeldama Tombs: Three Burial Caves in the Kidron Valley, Jeru-
salem (Jerusalem: Israel Antiquities Authority, 1996), 118120; Rachel Hachlili
and Ann E. Killebrew, Jericho: The Jewish Cemetery of the Second Temple
Period, (Jerusalem: Israel Antiquities Authority, 1999), 19295; Jodi Magness,
The Burials of Jesus and James, JBL 124/1 (2005): 132; Rachel Hachlili, Jew-
ish Funerary Customs, Practices and Rites in the Second Temple Period (Leiden:
Brill, 2005), 94, 32224, 483.
10
This fact has not deterred scholars from trying. For a recent discussion and re-
view of opinions, see Magness, Burials, 129.
148 Deborah Green
are quite large, were in use for several centuries (making the dating of
individual burials difficult), and include burials of both local and
Diasporic Jews.
11
Similar to the earlier burials, it is likely that only
wealthy or high status individuals are buried in these tombs.
12
How-
ever, either because of changes in style or variations in tradition pos-
sibly related to the different cities of origin of the mourners, a wide
variety of burial practices is attested in the caves. In Beth Shearim we
see the employment of two distinct types of burial niches (Figs. 5.13):
the kokh (a burial niche usually dug perpendicularly to the tomb wall)
and the arcosolium (a rectangular niche dug horizontally or parallel to
the wall with an arched top).
13
Different types of sarcophagi (Fig. 5.4)
and some ossuaries are also present; the former becoming more preva-
lent during the third century C.E. (the period of greatest expansion of
the tombs) and the latter becoming less common. The decorations and
epigraphy on the tomb walls, markers, and sarcophagi are quite varied.
11
For the dating of Beth Shearim, see Benjamin Mazar, Beth Shearim: Report on
the Excavations during 19361940 (vol. 1; New Brunswick: Rutgers University
Press, 1973), 22; Nahman Avigad, Beth Shearim: Report on the Excavations dur-
ing 19531958 (vol. 3; New Brunswick: Rutgers University Press, 1971), 260;
Fanny Vitto, Byzantine Mosaics at Bet Shearim: New Evidence for the History
of the Site, Atiqot 28 (1996): 13741. Although five periods of building are
found at the necropolis (Herodian through Arab), Avigad and Mazar viewed the
most significant as Periods I through III. Avigad outlines these as Period I (from
the Herodian through the first half of the second century C.E.), Period II (from
the second half of the second century to the beginning of the third century C.E.),
and Period III (Phase A, from the middle of the third to the fourth century C.E.
and Phase B, from the first half of the fourth century to the destruction of Beth
Shearim in 352 C.E.). However, Vitto demonstrates that the city continued
to be inhabited and the necropolis to be used after the mid-fourth century. In the
Byzantine period, Bet Shearim apparently enjoyed a second period of prosper-
ity , Vitto, Byzantine Mosaics, p. 138. For information on the local and
Diasporic Jews buried there, see Avigad, Beth Shearim, 25961.
12
Moshe Schwabe and Baruch Lifshitz, Beth Shearim: The Greek Inscriptions
(vol. 2. New Brunswick: Rutgers University Press, 1974), 219221.
13
Of note, the arcosolia in Beth Shearim often hold more than one body. In sev-
eral cases these receptacles are large enough to accommodate the burial of three
bodies placed perpendicularly to the wall and another body placed horizontally
at the back of the niche. For more precise information on the dating of these
niche types at Beth Shearim, see Avigad, Beth Shearim, 259. In several sites
throughout Palestine, the kokh is the most common type of burial niche. The
kokhim are often dug at ground level deep into the wall of the cave, and the body
is placed perpendicular to the surface of the wall so the feet are facing toward
the room.
Sweet Spices in the Tomb 149
As mentioned, some of these differences may be the result of changing
practices, but others may simply reflect differences in custom (local vs.
foreign, familial, or fashion). Therefore, even though we may describe
general changes in burial customs over time, the account remains pro-
hibitively sparse, as it concerns only the wealthiest or highest status of
the population and gives no insights into the processes or motivations
behind such practices.
Further problems are reflected in previous scholarship. The study of
ancient Jewish burials reflects biases similar to those apparent in other
areas of Jewish studies. To a large degree, most of these notions have
been corrected over time, but their fallout may nevertheless uncon-
sciously influence or consciously stymie the researcher. The first is the
preconception that Diasporic Jews who lived across the expanse of the
Roman Empire were isolated, either by choice or by force, from their
host cultures. Leonard Rutgers, among others, has gone a long way
to disprove this misconception.
14
In his thorough study of the Late
Roman period Jewish catacombs in Rome, Rutgers assesses the ma-
terial remains and epigraphy in relation to their Christian and wider-
Roman counterparts. In addition, he touches on burials throughout
the Diaspora and Palestine in order to draw inner-Jewish comparisons.
Although there is great similarity between the Christian and Jewish
catacombs in Rome, there is a wide divergence in burial customs found
throughout Jewish communities of this time period. In the cases of
other Diasporic communities, Rutgers finds that local Jewish practices
have much in common with their host cultures.
15
He further finds
that in several instances in Rome, the decorative art found on sar-
cophagi and tomb walls appears to have been crafted in Roman
shops, most likely by Roman craftsman.
16
These findings indicate that
the Jews of Rome had regular commercial contact with their Roman
neighbors.
Therefore, it should not be a surprise that the type of Jewish burial
most common in Rome in the Late Roman period is the catacomb, an
14
Leonard Victor Rutgers, The Jews in Late Ancient Rome: Evidence of Cultural
Interaction in the Roman Diaspora, (Leiden: E. J. Brill, 1995). Particularly for the
case of Rome, see also Harry J. Leon, The Jews of Ancient Rome (Philadelphia:
Jewish Publication Society of America, 1960; repr., Peabody, Mass.: Hendrick-
son Publishers, 1995). For Rome and other areas of the Diaspora, see the early
groundbreaking work of Goodenough, Jewish Symbols, vol 2.
15
Rutgers, The Jews in Late Ancient Rome, 6567.
16
Ibid., 6881.
150 Deborah Green
5.1. Kokh, Beth Shearim, Israel. (authors photo)
5.2. A variety of burial types: arcosolium, loculus, pit, sarcophagus,
Beth Shearim, Israel. (authors photo)
Sweet Spices in the Tomb 151
5.4. Sarcophagus, Beth Shearim, Israel. (authors photo)
5.3. Two kokhim, Beth Shearim, Israel. (authors photo)
152 Deborah Green
underground tunnel complex built and in use for approximately
300 years,
17
and which has several features in common with evolving
Roman tomb types. The catacombs contain columns of horizontal
niches (loculi ) cut into the tunnel walls. Corpses were laid into these
niches which were then sealed with a mixture of rubble and brick and
marked with marble or other types of inscribed markers. The catacomb
complexes are exceedingly large and appear to have contained a large
portion of the Jewish community.
18
These underground cemeteries re-
flect a similar development in tomb structure in the wider Roman com-
munity; that is, the move from hypogea (small underground rooms in
which the remains of families are found) to larger burial complexes sup-
ported by membership in particular subgroups of the population.
Only in a few cases in the Jewish catacombs of Rome do we find the
kokh type of niche so popular in Palestine.
19
Although not in abun-
dance, there are also several examples of distinctly Jewish (and in some
cases, non-Jewish) hypogea in the catacombs which contain only a
few loculi and arcosolia indicating that these may have been family
tombs.
20
Unfortunately, very few grave goods survive from the Jewish
catacombs and virtually none of these goods have been recorded in situ,
making comparisons with those of Palestinian or other Diasporic
burials almost impossible.
21
17
Very recent scholarship suggests dating the Villa Torlonia catacomb to the sec-
ond century C.E. The entrance to the lower catacomb level may have been dug as
early as 50 B.C.E., while the upper level of the catacomb dates to 400 C.E. These
estimates derive from radiocarbon dating recently completed and published
by Leonard Rutgers, Klaas Van der Borg, Arie F. M. de Jong, and Imogen Poole,
Jewish Inspiration of Christian Catacombs, Nature 436 (21 July 2005): 339.
18
This suggestion is derived from, among other points, the poor quality of many of
the inscriptions and artistic details; that is, the mix of wealthy and poor commu-
nity members suggests the desire to bury as much of the community population
as possible. Rugters, The Jews in Late Ancient Rome, 56; Leon, The Jews of
Ancient Rome, 25759.
19
Except for the Vigna Randanini catacomb, which contains several kokhim;
Rutgers, The Jews in Late Ancient Rome, 6264.
20
For a discussion on Painted Rooms I and II in the Vigna Randanini catacomb,
see Rutgers, The Jews in Late Ancient Rome, 5455. Rutgers surmises that these
rooms were separate hypogea, possibly pagan, that were later connected as the
distinctly Jewish catacomb was expanded.
21
However, Rutgers does discuss parallels in artistic production between Jewish
burials in Rome and in Palestine, 8892. For evidence of the grave goods (par-
ticularly glass), see Goodenough, Jewish Symbols, vol. 3.
Sweet Spices in the Tomb 153
According to Rutgers, the surviving evidence of the catacombs them-
selves, the inscriptions (which are predominantly in Greek and Latin),
the onomastic data from the inscriptions, and the decorative art found
within the complexes does shed some light on the Late Roman Jewish
community. On the basis of this evidence, Rutgers concludes that this
community not only had extensive contact with the wider Roman com-
munity but felt perfectly at ease in adopting or transforming some
Roman styles (e.g., highly decorated sarcophagi with images of victo-
rae, tomb paintings of vines, etc.) just as they rejected other Roman
burial customs or rituals (e.g., cremation and pictorial representations
of the deceased).
22
The Jews of Rome, therefore, appear to be free to
imitate the Roman iconography and other practices they admire as well
as to develop their own styles in burial without complete assimilation
into the general culture. At the same time, although the Jews of Rome
do not seem to be tied to any strict legal code with respect to burial,
they do appear to observe traditions and practices similar to Jews in
other parts of the empire (e.g., inhumation, employment of the kokh,
and widespread use of the menorah as a decorative element).
A second assumption long-held in the study of Jewish history was
that from the period of 70 C.E. (the destruction of the Second Temple)
onward, the rabbis maintained a hegemony over the Jewish commu-
nity writ large. This notion derives in part from the uncritical reliance
scholars placed on meticulous study of rabbinic texts for historical
information and in part from a lack of material evidence.
23
How-
ever, as Jewish studies became increasingly integrated into Humanities
programs, scholars began to adopt methods and practices from other
disciplines (e.g., history, anthropology, and literary studies) and to
reassess the textual data. Likewise, as numerous archaeological exca-
vations were completed and published, this new brand of scholars was
also required to account for the seeming disparities between text and
artifact. Several scholars attempted to address these issues via a syn-
thesis of the data, thereby indicating that the rabbis did not have the
power they asserted in their texts or that other Jews either maintained
their own customs and traditions or adopted Hellenistic mors contrary
22
Rutgers, The Jews in Late Ancient Rome, 9299.
23
For a more complete discussion, see Jacob Neusner, The Demise of Normative
Judaism, A Review Essay, Judaism 15 (1966): 23040, in which he addresses the
problems of writing Jewish history that the studies of such scholars as George
Foote Moore and Ephraim Urbach highlight.
154 Deborah Green
to the desires of the rabbis.
24
Often attendant to this view was the
opinion that the rabbis shunned any and all Hellenistic (Roman or
pagan) customs and culture.
However, as investigations continued, the facts, at least in Palestine,
proved more complex. How much power or authority any group of
rabbis wielded at any specific time or place is still being hypothesized
but remains unknown and is relatively unimportant for this study.
25
Of
great importance is the recent work of several scholars which demon-
strates that the textual and material evidence often reveal the rabbis
26
to be just as Hellenized and influenced by Roman culture as other local
and Diasporic Jews. In addition, more than a few scholars, through
careful textual and philological study, have determined that the rabbis
lived in a society that was both Roman and Jewish just as their counter-
parts did across the Roman empire.
27
These Palestinian rabbis seem to
have been comfortable with, and adopted as their own, those customs
that did not run contrary to other rabbinic norms or observances.
For example, it was Hellenistic custom to light incense after meals, and
evidence of this practice in rabbinic households is described in the
24
Goodenough, Jewish Symbols, vols. 13. See also Jacob Neusner, Notes on
Goodenoughs Jewish Symbols, IVIII, Conservative Judaism 17, nos. 34
(1963): 7792.
25
One exception to the notion of the limited influence of any group of rabbis may
be the role of Judah Ha-nasi (the Prince) as Patriarch in the late second century
and early third century C.E.
26
There is great danger in using the term the rabbis, as different groups or
schools of rabbis seem to be not only at odds with those people who may dis-
agree or appear to threaten them (e.g., other Jews, women, and pagans), but also
at odds with each other. In addition, the literature represents redaction and editing
of several hundred years of interpretation, and therefore may not reflect accurately
the views of the various generations. To avoid confusion, scholars often use the
term class of rabbis; however, this term presents other obvious problems.
27
For example, see Steven Fine, Art and Judaism in the Greco-Roman World: Toward
a New Jewish Archaeology. (Cambridge: Cambridge University Press, 2005); Seth
Schwartz, Gamliel in Aphrodites Bath: Palestinian Judaism and Urban Culture
in the Third and Fourth Centuries, Talmud Yerushalmi and Graeco-Roman Cul-
ture (ed. Peter Schfer; Texte und Studien zum antiken Judentum; Tbingen:
Mohr Siebeck, 1998), 20317. Saul Lieberman, Greek in Jewish Palestine (1942;
repr., New York: The Jewish Theological Seminary of America, 1994); Morton
Smith, The Image of God: Notes on the Hellenization of Judaism, with Especial
Reference to Goodenoughs Work on Jewish Symbols, Bulletin of the John Ry-
lands Library Manchester 40, no. 2 (1958): 473512.
Sweet Spices in the Tomb 155
Talmud.
28
Surprisingly, room fumigation after meals was so popular
that the description includes methods for fumigating during the Sab-
bath when one is not allowed to kindle fire. Similarly, at the necropolis
of Beth Shearim, where both Diasporic Jews and rabbis are buried,
ornate Hellenistic decorations appear on sarcophagi, grave goods, and
the walls.
Of special importance for the study is the fact that the rabbinic
literature provides some of the only textual evidence extant on Jewish
burial practices and customs in Palestine. In many instances, it pro-
vides the only descriptions of corpse preparation, funeral rites, and
secondary bone collection. Because careful review of the material indi-
cates that the rabbis are not necessarily legislating new or previously
unknown rituals and certainly are not legislating against then current
rituals, the textual evidence can be a valuable tool in the study of Jew-
ish burials, particularly if we compare it where possible to the archae-
ological evidence.
The methodology employed in this paper may be considered to be
overly parochial and no doubt would run aground of comparativists.
However, whenever possible, the paper attempts to consider only Pal-
estinian textual evidence that can be compared with archaeological
evidence from the same location and approximate time period in an at-
tempt to uncover the rationale that either already existed or developed
during the early rabbinic period for the employment of perfume in Jew-
ish burials in the early centuries C.E. in Palestine. The goal is to de-
cipher the texts and compare them to the material remains in order to
understand how the rabbis justified these practices and to begin to
consider why perfume bottles were buried with the dead.
This study acknowledges as its starting point that the practice of
burying the dead with perfume bottles was already ancient and wide-
spread throughout Palestine and the Mediterranean at the time of the
early rabbinic period (Fig. 5.58). Unfortunately, most of the material
evidence from Jerusalem the area containing the largest known
quantity of Jewish burials all but ceases in 70 C.E. For this study, the
most useful evidence would be the burial remains from Sepphoris and
Tiberias. These cities were among the largest Jewish urban centers in
Palestine at the time, served at various times as the seat of the Patri-
archate, and are represented as paradigms for Jewish city life in much
28
m. Besah 2:7, b. Ber. 43b and b. Besah 22b.
156 Deborah Green
5.5. Unguentaria, Maktar Museum, Tunisia. (Photo: Amy Hirschfeld,
International Catacomb Society. Used with permission.)
5.6. Unguentaria and assorted glass artifacts, Carthage Museum, Tunisia.
(Photo: Amy Hirschfeld, International Catacomb Society. Used with permission.)
Sweet Spices in the Tomb 157
5.7. Assorted glass artifacts including candlestick bottles
(unguentaria), Carthage Museum, Tunisia. (Photo: Amy Hirsch-
feld, International Catacomb Society. Used with permission.)
5.8. Bone fragments, unguentaria and other grave goods, Via Latina catacomb,
Cubiculum A. (Photo: Estelle S. Brettman, International Catacomb Society. Used
with permission.)
158 Deborah Green
of rabbinic literature. However, no complete excavation of the cem-
eteries has yet been accomplished. Finally, the most abundant avail-
able evidence is at Beth Shearim, the difficulty in dating of which has
already been discussed.
An Overview of the Evidence of Rabbinic Texts
The early rabbinic legal corpora are classified into tractates or units
based on the various aspects of organized communal Jewish life as
perceived by the early rabbis. In most cases, these are intimately linked
to the laws and regulations of the Torah through interpretation. There-
fore, the division of Zeraim (Seeds), contains such subunits as Bera-
khot (Blessings), Peah (Gleanings), and Terumot (Heave Offerings).
Although almost every conceivable situation may be addressed for
certain legal issues, surprisingly, no division exists dedicated to the
subjects of death, burial customs, or mourning. Only the Babylonian
Talmuds minor tractate of Semahot,
29
which is generally dated to the
eighth or ninth century C.E., refers to these customs.
30
While the trac-
tate includes issues and decisions regarding death, burial customs, and
other post-burial events and questions, it is difficult to know whether
the customs described therein reflect accurately the much earlier
stratum of rabbinic practice in Palestine or whether they reflect the
late period in Babylonia from which they derive. These issues apply to
all tractates of the Babylonian Talmud which is parsimonious in its
methodology and often so terse in its description as to require lengthy
interpretation, conjecture, and outright speculation by the reader.
Therefore, to obtain relevant data on burial practices for the early cen-
turies C.E., we must first scour the early Palestinian Jewish texts and
resort to the Babylonian Talmuds comments on these earlier verses
only where necessary. Two texts serve as the primary sources for such
information: the Mishnah and the Tosefta. Until recently, the Mish-
nah, which serves as the basis for both the Jerusalem and Babylonian
29
Ironically, semahot means happy occasions.
30
Strack and Stemberger date the tractate, also known as Ebel Rabbati, to the
eighth century, noting that Zlotnick and Meyers date it to the third century. H. L.
Strack and G. Stemberger, Introduction to the Talmud and Midrash (trans. Mar-
kus Bockmuehl; Edinburgh, T&T Clark, 1991), 249. David Kraemer dates Se-
mahot to the Geonic period (ninth century). David Kraemer, The Meanings of
Death in Rabbinic Judaism, (London: Routledge, 2000), 9.
Sweet Spices in the Tomb 159
Talmuds, was considered to have been redacted in 200 C.E. and to rep-
resent the first rabbinic legal code. The Tosefta was thought to date ap-
proximately 100 years later than the Mishnah. Organized similarly to
the Mishnah, the Tosefta includes much of the same material but also
evidences several differences. Scholars accounted for these differences
as changes, debating whether observance of the laws actually changed
or whether the rabbis changed these laws in order to align them with
popular practice or new customs. However, because recent scholarship
has reopened the case for the primacy of the texts,
31
these changes in
law are also suspect.
Regardless of their status as relevant codes of conduct, the Mishnah
and Tosefta present elements of Jewish burial practices in the Roman
period that may be pieced together by means of close textual study.
Upon death, the corpse would be anointed with oil, rinsed, and
wrapped in a linen shroud or other type of linen garment.
32
As part
of this dressing process, the chin would be tied, the eyes closed, and
the orifices stopped up.
33
The body was placed into a bier or coffin
or onto some type of wooden structure, often referred to as a bed
(hum). The bed was then carried in a funeral procession from the home
of the deceased through the community to the outskirts of town and
then to the burial site.
34
During the procession, stops might be made
for the hired wailers, often women, to sing or lament and clap their
hands loudly.
35
Frequently, the burial sites were man-made caves into
which were carved deep recesses for the burial (kokhim).
36
In the late
second Temple and early rabbinic periods, a second form of burial,
known as the collection of bones,
37
often took place approximately
31
Among others, see Judith Hauptman, Rereading the Mishnah: A New Approach
to Ancient Jewish Texts (Tbingen: Mohr Siebeck, 2005). Hauptman dates the
Mishnah to the early third century C.E., but views it as an amalgam of the two
older texts, the ur-Mishnah and the Tosefta, and other materials , 21.
32
m. Kil. 9:4; m. S

abb. 23:4, 5; m. Maa s. S

. 5:12.
33
m. S

abb. 23:5 and t. S

abb. 17:18.
34
m. Meg. 4:3; m. Ber. 3:1; m. B. Bat. 6:7; m. Sanh. 2:1 (2:3).
35
m. B. Mesia 6:1; m. Meg. 3:3, 4:3; m. Moed Qat. 3:8; m. Ket. 4:4 (also discusses
the playing of flutes); t. S

abb. 17 (also discusses the playing of instruments);


m. Menah. 10:9.
36
m. Moed Qat. 1:6; m. B. Bat. 6:8. The term (kokh) arises from the rabbinic tex-
tual material. Archaeological evidence in the burial caves indicates other types
of recesses in addition to kokhim (i.e., arcosolia, see note above), although it is
also possible that kokh was a general term for burial niche.
37
m. Sanh. 6:6; m. Moed Qat. 1:5.
160 Deborah Green
one year after the initial burial. From the archaeological evidence, we
find that these bones were placed into small stone boxes known as os-
suaries (see above). For the most part, ossuaries were phased out after
the destruction of the Second Temple in 70 C.E., although evidence of
ossuary use exists through the third century C.E.
38
Rabbinic References to Perfume in Burial
Spices were likely used in each of the three phases of burial: corpse
preparation, funeral procession, and interment.
39
Sometimes the texts
are clear about which phase is being discussed, but often this may
be quite difficult to discern. For example, the Mishnah stipulates the
following:
No blessing may be said over the lamp or the spices of idolaters, or over the lamp
or spices of the dead, or over the lamp or spices [placed] before the idols of idola-
ters. No blessing may be said over a lamp until one can enjoy its light.
40
This law contains a four-part structure. The first three injunctions
focus on lamps and spices: those belonging to idolaters, those for the
dead, and those for idolatry. The last section seems to be an addition to
the foregoing; namely, that as for lights and occasions on which one
does say a blessing over a lamp, one does not do so until the light is lit.
Of note, the law regarding lamps and spices for the dead is wedged
between the lamps and spices of the idolaters and idolatry. It would
seem that the rabbinic voices want to stipulate a difference between
those lamps and spices owned by idolaters and those actually used in
idolatrous practices. One does not say a blessing for either, but never-
theless a distinction is made. The rabbis may wish to imply that other
people, in addition to those who perform idolatry, are considered
idolaters.
41
The three injunctions together also imply that before one
intones a blessing for a pleasant lamp-light or fragrance, one must first
38
See Rahmani, A Catalogue, 21.
39
These phases should not be confused with the three-stage pattern of funerals
first outlined by Robert Hertz, Death and the Right Hand (1907; repr., Glencoe,
Ill.: Free Press, 1960) and Arnold van Gennep, The Rites of Passage (trans.
Monika B. Vizedom and Gabrielle L. Caffe; 1909; repr., Chicago: University of
Chicago Press, 1960).
40
m. Ber. 8:6.
41
This point is addressed below in the discussion on b. Ber. 53a.
Sweet Spices in the Tomb 161
check the source and significance of such a light or scent; that is, who is
the kindler of the light or incense (or person wearing the perfume) or
for what purpose is the light kindled or the scent produced. These
possibilities raise several questions about rabbinic attitudes toward the
dead. Do the dead own the spices and lamps, or are they employed in
honor of the dead, or are the practices associated with the dead re-
lated, in the perception of these anonymous rabbis, with the practices
of idolatry? All these possibilities lay pregnant in the passage without
further elucidation. If we focus on the portion of the passage that con-
cerns only the lamps and spices of the dead, several other questions
arise: What are the spices used for the dead? In which phase of burial
are they used? How are they used?
A close reading of other early rabbinic statements may help answer
the questions this Mishnah raises. The first possibility concerns the
anointing of the dead body. The washing, anointing, and wrapping of
the body is discussed in m. S

abb. 23:5 in reference to the types of work


one is allowed to perform during the Sabbath:
They may make ready [on the Sabbath] all the needs of the dead, anoint and
rinse it,
42
only [provided that] they do not move any of its limbs.
43
They may
draw the mattress away from beneath it and let it lie on sand that it may be the
longer preserved; they may bind up the chin they may not close a corpses eyes
on the Sabbath
The first part of the passage refers directly to the first step of the
burial: washing and anointing of the body. Death is a dirty business,
and a corpse needs to be washed or rinsed soon after death as the ori-
fices may leak before, during, or just after the event.
44
Because typical
bathing practices at this time involved anointment with scented oils,
45
42
This passage belies a fascinating problem; that is, anointing is mentioned be-
fore rinsing (or washing). However, with reference to bathing, usually wash-
ing occurs prior to anointing. See m. Taan. 1:6, where one is not allowed to
wash or anoint (hkycbv hjyxrb) during a fast.
43
Although limb could also refer to the genitals or penis, it is unlikely in this
passage.
44
For the stopping up of the orifices, see t. S

abb. 17:18.
45
See m. Taan. 1:6. Perfumed oils were regularly used from the Greek through
Roman periods by both men and women as part of the bathing process. Both
sexes would apply aromatic oil to the head and hair. See Michal Dayagi-Men-
dels, Perfumes and Cosmetics in the Ancient World (Jerusalem: Keter Publishing
House, 1991), 1634; Andrew Dalby, Empire of Pleasures: Luxury and Indul-
gence in the Roman World (London: Routledge, 2002), 24547 and passim; Flo-
162 Deborah Green
it is likely that this anointing of the corpse was performed with scented
oils as well. It is worth noting that it is unlikely that scented oil was
used to mask the stench of decomposition. It is possible that particular
diseases may have caused early on-set decomposition but, for the most
part, in temperate climes, the odor of decomposition is not detectible
by the human nose until the second or third day.
The burial of Jesus, as recounted in New Testament sources such as
Mark and John, also provides evidence for anointment after death
with scented oils. Mark 16:1, describes, When the Sabbath was over,
Mary Magdalene, and Mary, the mother of James, and Salome bought
spices so that they might go and anoint him.
46
Similarly, in John
19:3940 we learn, Nicodemus, who had at first come to Jesus by
night, also came, bringing a mixture of myrrh and aloes, weighing
about a hundred pounds. They took the body of Jesus and wrapped it
with the spices in linen cloths, according to the burial custom of the
Jews. The books of Matthew and Mark also present a figurative ac-
count of this custom, as the anointing occurs before the death of Jesus.
In these accounts, an unidentified woman comes to the house of Simon
the leper, where Jesus and his disciples are eating dinner, and anoints
Jesus with a costly ointment from an alabaster jar (Matt 26:67,
Mark 14:38). The disciples become angry over the extravagance, but
Jesus reminds them, She has performed a good service for me. You
will always have the poor with you, but you will not always have me.
By pouring this ointment on my body, she has prepared me for burial
(Matt 26:1012).
47
rence Dupont, Daily Life in Ancient Rome (trans. Christopher Woddall; Oxford:
Basil Blackwell Ltd., 1992), 264, 266. Men involved in wrestling or other exercise
would also apply oil before bodily exertion and scrape off the oil using a metallic
device (strigil) before bathing. Several of these have been recovered in areas of
Roman period Jewish communities (Dayagi-Mendels, Perfumes and Cosmetics,
1634. Of interest, as depicted in Dayagi-Mendels volume, The Israel Museums
collection of alabaster bottles and other unguentaria dates back to pre-Israelite
periods of the second millennium B.C.E). Cf. Saul Liebermans discussion on
m. S

habb. 22:6, which states, They may oil and massage their stomach but
not exercise (the body) and not scrape. They may not go down to the hmydrvq
and may not use artificial emetics. Lieberman, Greek in Jewish Palestine,
93.
46
See also Luke 23:5524:1.
47
Cf. Mark 14:68 in which we are informed of the specific ingredient: nard or
spikenard. In both accounts, the alabaster jar adds to the image of costliness.
Sweet Spices in the Tomb 163
Although the employment of scented oil appears to have been a
regular practice of burial preparation,
48
other than this slim textual
evidence, there is no direct proof about where this event took place. In
the New Testament accounts it would seem that the spices are brought
after the body is interred. However, this may be out of necessity be-
cause of the manner of Jesus death, the lateness of the day at the time
he was buried, or the impending Sabbath. According to the Mishnaic
passage, it seems that washing and anointing regularly occurred in the
home before burial, and home preparation is the most likely arena for
these activities.
49
At home, proper attention, care, and respect for the
loved one could be given before the public ceremony of burial. Private
corpse preparation also accords well with the rabbinic sense of propri-
ety, modesty, and public decorum stipulated throughout the rabbinic
texts.
50
Finally, because the body appears to be already in a bier, coffin,
or upon some other type of bedding during its procession to the burial
cave, it seems likely that preparation of the body has already taken
place before arrival at the grave.
As to the question of whether the Mishnaic passage concerning the
lamp and spices for the dead (m. Ber. 8:6) is referring to the oil used to
anoint the dead before burial, it is unlikely. The lamp and spices pas-
sage specifies lamps and proscribes a blessing. The anointing passage
(m. S

abb. 23:5), however, discusses neither lamps nor blessings. There-


fore, it is unlikely that the first passage is referring to anointing oil
when it intones that no blessing should be said over the lamp or spices
of the dead.
The second use of spices in the burial of the dead may be either in-
cense that is lit during the procession to the burial cave or some kind of
scented oil that was sprinkled on the bier during such a procession.
Evidence of the funeral procession is found in several places through-
48
It appears that in addition to regular bathing and burial practices, aromatics
were also used after dining. Evidence in the Babylonian Talmud suggests that
scented oil was employed to rid the hands of unwanted residue from eating
(b. Ber. 53a) and that incense was lit after meals (see above).
49
The fact that Jesus does not have a home in Jerusalem may also explain why his
body is prepared in the burial cave.
50
While this is generally the case for all men, the laws concerning modesty and
public propriety are quite stringent for the students of rabbis (see b. Ber. 43b).
Even stricter are the laws for women (e.g., Gen. Rab. 8:12 and m. Moed Qat. 3:8
with direct reference to burial, as the bier of woman is not to be put down in pub-
lic out of respect).
164 Deborah Green
out the Mishnah,
51
but for the use of incense or perfume during the
procession it is the Tosefta which contains two interesting passages.
The first, t. S

eqalim 1.12, concerns funds that are collected for express


purposes but are found to be surpluses:
the surplus [of money collected for] the poor, [must be used] for the poor. The
surplus [of money collected for] the redemption of captives, [must be used] for
the captives The surplus [of money collected for] the dead, [must be used] for
the dead. The surplus [of money collected for] a [particular] dead person, [must
be given to] his heirs. Rabbi Meir says, The surplus for a [particular] dead per-
son will be left until Elijah will come. Rabbi Nathan says, [With] the surplus for
a [particular] dead person, they build a structure over his tomb, or he may
sprinkle perfume
52
for him before
53
his bier
54
The last two lines of this series of instructions are of particular inter-
est. Rabbi Meir insists that the surplus monies from burials should be
kept until the prophet Elijah comes to decide what shall be done with
them. Rabbi Nathan disagrees and mentions two appropriate uses:
erection of a marker or some other structure over the gravesite or ex-
penditure of the money on perfumed oil to sprinkle before the bier
during the funeral procession. Although this passage indicates
clearly that the spices were used in the procession (i.e., before the
bier), it does not explain the rationale or motivation underlying
such practice.
The next passage, from t. Nid. 9:16, explains the use of spices in the
funeral procession:
51
See m. Ber. 3:1; m. Meg. 3:3, 4:3; m. Moed Qat. 3:8; m. B. Bat. 6:7, 8; m. Sanh.
2:3, and m. Menah
.
10:9.
52
xlz; that is, sprinkled fluid, or perfume. However, this is not the regular
word for perfume or spices (,ymsb).
53
Zuckermandel has ynpb (in front of ), M. S. Zuckermandel, Tosephta: Based on
the Erfurt and Vienna Codices (Jerusalem: Wahrmann Books, 1970), 651. Lieber-
man has ynpl (before or in front of ), Saul Lieberman, Tosefta Ki-Fshutah
(New York: Jewish Theological Seminary 1992), 204.
54
t. S

eqal. 1:12. Lieberman also cites the y. S

eqal. 10:2 and b. Sem. 12:9, Tosefta


Ki-Fshutah, 673. However, the issue in b. Sem. 12:9 is whether one should
sprinkle wine and oil onto the bodies at the time of burial (or on the bones at the
time of secondary burial, as Kloner reads see below), as instructed by Rabbi
Akiba. R. Simeon ben Nanos disagrees and states that oil but not wine should
be sprinkled, and the later sages affirm that neither wine nor oil should be
sprinkled.
Sweet Spices in the Tomb 165
At first they would immerse on the basis of women who died while menstruating.
Subsequently, they immersed every one of them on account of the honor of
women. At first they brought out incense
55
before those [who died] of sickness of
the bowels.
56
Subsequently, they brought out [incense] before every one of them
on account of the honor of the dead.
57
Because Niddah is the tractate concerning the condition of unclean-
ness, particularly with respect to menstruating women, the beginning
of this passage focuses on the immersion of objects that come in con-
tact with menstruating women who die. However, as we move through
the passage, we can see that the focus is on the change in burial cus-
toms over time. The word subsequently (vrzx) is literally translated
as they returned. This may imply either that the custom changed or
that the rabbis returned to the issue and changed their earlier ruling.
One might reasonably infer that the rabbis changed their earlier ruling
to align the law with widespread practice.
More importantly, the passage imparts valuable information with
respect to spices namely that their use was widespread as part of the
burial process in the early centuries C.E. At one point, only those who
died from intestinal problems had incense lit before their biers,
58
but by
the time of the Tosefta, incense was lit before everyones bier. Further,
it appears from this passage that the burial phase during which the
burning of incense took place was the funeral procession, since the
term before every one is similar to the term before his bier as seen
in the S

heqalim passage. In addition, we do not find references to in-


cense in any of the passages concerning corpse preparation nor is there
any significant archaeological evidence of incense burners at burial
55
rmgvm. These are spices or perfume placed on hot coals (i.e., incense).
56
Or, intestines.
57
Of note, this passage continues to describe other burial customs that change over
time: At first they would bring out the rich in a high bed (>grd) and the poor in
a box (Xbylk, i.e., coffin or bier). Subsequently, they would bring out some in a
bed and some in a box on account of the honor of the poor.
(17)
At first they
would bring [food] to the house of mourning of the poor in a colored glass vessel
but to the rich in a white glass vessel. Subsequently, they brought out some in
colored [vessels] and some in white [vessels] on account of the honor of the poor.
At first, anyone who had someone who died, his expenses (or, his departure)
were more difficult for him than his death. Everyone began to set down their
dead and flee. Rabban Gamliel set the example of disregard for the custom him-
self, [and then] everyone acted according to Rabban Gamliel.
58
Or, as might be the case, rabbinic law allowed incense to be lit before the biers
only of those who died of intestinal illness.
166 Deborah Green
sites from which we may deduce that incense was not employed as
part of either the preparation or the interment phases.
59
This passage raises additional questions: Why was incense lit for
those who died of intestinal disease and, if this passage represents more
than a mere change in rabbinic legislation, for what reasons was the
practice changed? Is it possible that in the beginning, incense was
needed to mask the odor of those who died from intestinal illness?
60
If
so, then it is possible that incense was later employed to mask the odors
of everyone. But why not cover the odors of everyone from the start?
The answer: It is unlikely that incense was needed to cover the odors of
everyone, as the stench of decomposition would not be so bad so soon
after death. As already mentioned, decomposition does not usually
have a scent for the first few days, but these effects can obtain more
rapidly due to either increased heat or humidity or decomposition that
begins before death. Incense might be necessary, though, if the funeral
was delayed for a day or two. We have already seen that one may not
bury a corpse on the Sabbath. Further, even though Jewish law requires
burial as soon as possible, the Mishnah cites a qualification under
which it is permissible to delay a funeral for a day in order to honor
the deceased by bringing a coffin and burial clothing (m. Sanh. 6:5).
Although it is possible that incense would be used to cover the odor of
decomposition, it is just as possible that the Tosefta has explained
exactly why incense was lit for everyone: in order to honor the dead.
As obvious as the concept of honoring the dead may be, the ref-
erence to honor (dvbk) might be the key to unlocking the mystery
of the first Mishnaic passage regarding the lights and spices of the
dead. The Talmudic discussion of this Mishnaic passage also includes
the term honor, but instead of coupling honor with spices, this
text pairs honor with the lights:
A blessing may not be said over the lights or spices of the dead. What is the reason?
The light is kindled only in honor [of the dead], and the work of the spices is to
remove the odor. Rab Judah said in the name of Rab, Wherever it is found that
[a lamp would be carried] before him (the dead person) during the day and dur-
ing the night, no blessing is said over it. But wherever it is found that [a lamp
would be carried] before him only during the night, then a blessing is said. Rab
59
Only Goodenough mentions incense burners with reference to the city of Gezer,
Jewish Symbols, 1:165.
60
This might explain why the incense passage follows the immersion passage, as
both concern the removal of something either contamination or contaminat-
ing scent.
Sweet Spices in the Tomb 167
Huna said, Spices of the privy and oil used to remove filth, we do not say bless-
ings over. [Is this] to say that wherever [a spice] is not used for scent no blessing
is said over it? An objection: [One] who enters a spice store and smells the scent
even if he sits there all day long he only says a blessing once.
61
The discussion, part of a larger discourse on blessing pleasant experi-
ences (,ynhn tvkrb), assumes that a blessing is regularly said both when
one smells a pleasant fragrance and when lights are kindled. For the
lamps and spices of the dead, however, one does not say such blessings.
The explanation is that the light is in honor of the person who has died,
while the spices are employed to cover up the stench of the decompos-
ing body neither use honors God. These are new pieces of infor-
mation not seen in the earliest sources. To review, the only occurrence
of lights in those texts is the m. Ber. passage on the lights and spices
for the dead, and the purpose of the lights does not appear in that
passage. As for spices, the only information we see in the early sources
is from the Toseftan passage that describes their purpose as honoring
the dead. Therefore, the Babylonian rationale given here (to mask the
odor) clearly contradicts the Tosefta.
62
However, because the Tosefta
passage and this Talmudic passage both cite honoring the dead dur-
ing the funeral procession as the reason and time for these rituals, one
might surmise that our first Mishnaic passage which discusses the
lamps and spices for the dead is referring to the incense and lights
used in the funeral procession. This accords with y. Ber. 8.6.
As for b. Talmuds rationale that the spices mask the odor of the
decomposing corpse, three possibilities corresponding to the three
phases of burial must be addressed. While it is possible that the corpse
is malodorous during preparation for burial or during the funeral pro-
cession, it is unlikely that the odor was so bad as to require camou-
flage.
63
The odor from putrefaction is strongest 10 to 20 days after
61
b. Ber. 53a. For comparison, see also y. Ber. 8.7 and honor of the living.
62
See y. Ber. 8.6. Lights and spices placed on top of the bier are not blessed. Those
placed before it are blessed, as they are for the honor of the living.
63
In addition, we have no way of knowing whether corpse decomposition was
considered foul-smelling at all by the people in the early centuries C.E. The evalu-
ation of whether an odor is fragrant or foul is entirely culture-specific. See
William Ian Miller, The Anatomy of Disgust (Cambridge, Mass.: Harvard Uni-
versity Press, 1997), 1518. For example, in the West today certain body odors
connected to perspiration or vaginal discharge are considered by society to be
offensive, and therefore the market is flooded with perfumes, antiperspirants, de-
odorants, and douches. Similarly, in cultures where decomposition represents
168 Deborah Green
death;
64
therefore, masking an unpleasant odor would be most neces-
sary during interment.
65
The spices in the Talmudic passage could
refer either to perfume or to incense left at the burial cave for the use of
those who must enter the cave during corpse decomposition. However,
perfume sitting in a bottle is not an effective means of masking the
odor of an area, as the scent does not spread efficiently. It is much
more likely that mourners or others who needed to come to the burial
caves during this period would light incense to mask the stench rather
than employ perfume.
66
The flaw in this theory is that incense burners
are not found at the burial sites.
67
rebirth or sacrifice to the deity, it is entirely possible that these scents would not
only be tolerable but pleasant. The Berawan of north-central Borneo store the de-
composing corpse either inside the longhouse, where the entire community lives,
or on a raised platform in the jungle, precisely so that they may be near the body
in order to ensure it is not reanimated by evil spirits. They may also collect the de-
composition liquids in sacred vessels. The neighbors of the Berawan, who have
similar customs, may consume the liquids with rice. See Peter Metcalf, Death
Be Not Strange, in Magic, Witchcraft, and Religion: An Anthropological Study of
the Supernatural (ed. Arthur I. Lehmann and James E. Myers; Mountain View,
Calif.: Mayfield Publishing Company, 1993), 325. Beyond the issue of cultural
construct, we find in the faunal realm that animal decomposition actually at-
tracts certain animals. See for example Steven A. Smith and Richard A. Paselk,
Olfactory Sensitivity of the Turkey Vulture (Cathartes Aura) to Three Carrion-
Associated Odorants, The Auk 103 (1986): 58692. Although these researchers
dispute the generally accepted scholarship on response to particular odorants,
they allow that response to several other odorants relative to decomposition may
trigger responses in turkey vultures.
64
This could be later for those corpses in cool caves. In addition, it is questionable
how strong the scent of the decomposing corpses was considering that they were
sealed and inaccessible to several types of carnivorous organisms whose feasting
and other activities (e.g., laying eggs and attracting other organisms) at the
corpse site often increase the rate and odor of decomposition by increasing the
body temperature.
65
Of note, Arpad A. Vass, Senior Staff Scientist at Oak Ridge National Labora-
tory and Adjunct Associate Professor at the University of Tennessee in Forensic
Anthropology, has indicated (via email communication) that he does not believe
even unsealed corpses would have smelled so bad that people would not have
been able to eat nearby an interesting point considering that some ancient
peoples included feasting or commemorative meals at the graveside.
66
As already mentioned, incense was regularly used after meals either to cover the
scent of the meal or for pure enjoyment, so the Jews of the early Roman period in
Israel would be familiar with its other potential uses.
67
While it is possible that extensive looting of the tombs might account for the lack
of incense burners, one might still expect to see remnants of at least a few broken
Sweet Spices in the Tomb 169
It would seem then that the Talmud presents its own interpretation
of the spices for the dead perhaps evaluating them anachronistically
as a custom that has gone out of favor or as a rite with which the later
Babylonian rabbis are unfamiliar. It is also quite likely that the Talmud
employs this comment on the spices of the dead simply to introduce the
predominant themes of the passage; that is, the subject of the incense of
idolatry, its connection to Jewish women who regularly fumigate their
garments with incense, and rabbinic ambivalence toward these femi-
nine customs.
68
I have written extensively on this passage elsewhere.
69
As such, it seems inherently dangerous to base the analysis of perfume
bottles at burial sites on this one particular line from the Talmud.
Unguentaria at Burial Sites
How then are we to understand the preponderance of perfume bottles
evidenced in the archaeological data? According to Amos Kloner, be-
cause far fewer lamps are found at Second Temple burial sites than
burners recorded in excavation reports. It is also possible that incense or other
fragrant materials were burned in bowls or other receptacles (as depicted on the
walls of the Hellenistic tomb at Marisa, which is not a Jewish site), but recorded
evidence does not so indicate.
68
The Talmudic passage continues as follows: [If] he goes in and out and in and
out, he says a blessing each time. And this is a case in which it is not used for
scent, and yet he makes a blessing. Indeed [it is] too [used] for smell! [Its use is
that] it will be smelled by people, and [they] will come and make purchases of it.
Our rabbis taught: If one is walking outside the town and smells a [pleasant]
scent, if the majority of the inhabitants are idolaters, he does not say a blessing.
But if the majority are Israelites, he does say a blessing. Rabbi Yossi says, Even
if the majority are Israelites, he does not say a blessing, because the daughters of
Israel light incense for witchcraft. Do all of them light incense for witchcraft?
A minority was used for witchcraft and so too a minority for scenting garments.
Consequently, the majority is not for making scent, and wherever the greater
part is not used for making scent, a blessing is not said over it. Rab Hiya bar
Abba said, Rabbi Yohannan said, If one is walking on the Sabbath evenings in
Tiberias or at the end of Sabbath in Sepphoris and smells a [pleasant] scent, he
should not say a blessing because the presumption is that it is only the scenting
of garments. Our rabbis taught: If one is walking in a market of idolaters and
enjoys smelling [the pleasant scent of spices], this is a sin. b. Ber. 53a.
69
See Deborah A. Green, Soothing Odors: The Transformation of Scent in
Ancient Israelite and Ancient Jewish Literature (Ph.D. diss., University of Chi-
cago, 2003), 27491.
170 Deborah Green
perfume bottles, the lamps must have been used to light up the caves.
70
This would mean that the lamps found in the caves are not coincident
with those discussed by the rabbis since we determined that the refer-
ences to lamps describe those used in the funeral procession. As for the
perfume bottles, the occasional evidence of these bottles in ossuaries
suggests to Kloner that perfume and other precious liquids, such as
wine, were used to sprinkle on the bones during collection.
71
Since
bone-gathering is described in the Mishnah, without reference to the
sprinkling of oil, this argument does not seem persuasive.
As mentioned earlier, many scholars assume from the New Testa-
ment sources that corpse preparation occurred in the caves, and that
the bottles were left behind because of contamination from the dead.
72
While corpse contamination is a persuasive argument for both per-
fume bottles and lamps being left in the caves, we have already seen
that corpse preparation probably occurred in the home (m. S

abb.
23:5).
73
In addition, the disparity between lamps and perfume bottles is
not accounted for in this theory, as many more lamps would have been
needed in order to see what one was doing in these very dark caves.
There are also those scholars who believe the perfume bottles were
brought by mourners simply to cover the odor of the decomposing
corpses.
74
Although it is possible that some of the bottles may have
been used for this purpose, as already mentioned, incense would have
been the preferred method of odor masking. Further, how might one
explain the many perfume bottles that are sealed within the kokhim?
75
70
See Kloner and Zissou, The Necropolis, 5960.
71
Ibid., 60, wherein Kloner and Zissou cite b. Sem. 12:9.
72
Ibid., 60. On the problems of ethnographic interpretation of funerary remains
by archaeologists and anthropologists, see Peter J. Ucko, Ethnography and Ar-
chaeological Interpretation of Funerary Remains, World Archaeology 1 (1969):
26280. On the surprising rites and customs that seem counterintuitive to eth-
nographers, anthropologists, and archaeologists whose frame of reference is
Western culture, see Metcalf, Death Be Not Strange, in Magic, Witchcraft, and
Religion.
73
See also McCane, Roll Back the Stone, 48.
74
Ibid., 15, 48.
75
See Elena Kogan-Zehavi, Settlement Remains and Tombs at Khirbet Tabaliya
(in Hebrew), Atiqot 40 (2000): 5379; Fanny Vito, Burial Caves from the Second
Temple Period in Jerusalem (Mount Scopus, Givat Hamivtar, Neveh Yaaqov),
Atiqot 40 (2000): 65121; Hachlili and Killebrew, Jericho, 17691; Mazar, Beth
Shearim, 173; Avigad, Beth Shearim, 68, 201 (in arcosolium). However, these
bottles may date to the Late Roman period. See also the Akeldama tombs for
Sweet Spices in the Tomb 171
These cannot be explained as being for use by mourners. Rather, two
suggestions present themselves as most reasonable. The first is that the
bottles are the personal effects of the deceased similar to cosmetic
bottles,
76
spindle whorls, rings, jewelry and other precious personal
items also found in sealed kokhim.
77
Even the late tractate of b. Sema-
hot describes bridegrooms being buried with their pens, inkwells, and
marriage documents. Samuel, the Small, a scribe, is buried with his writ-
ing tablet.
78
These bottles may have been important to the deceased,
79
and so mourners wished their loved ones to be buried with them. To ful-
fill this desire, the perfume bottles would have been buried in the sealed
kokh or arcosolium with the deceased. Occasionally, the bottles may have
been transported to the ossuary with the bones. Because thieves were
unlikely to find much use in the bottles, robbers would have thrown them
out into the chamber of the cave when hunting for jewelry and other
precious effects. This would explain the broken bottles found in open
chambers. Other suggestions are that the vessels were broken on purpose
to reduce the risk of tomb robbery or as a part of the funeral rites,
80
but neither of these claims can be substantiated.
A second possibility is that the bottles are related to other grave
goods, including cooking pots, bowls, jugs, lamps, and storage
jars. Similar to much of this household pottery, the glass perfume
bottles of the Roman period are not well-made, which may indicate
that they were not personal items but were specifically produced for
burials.
81
Further, almost all of the unguentaria from the Second
examples of perfume bottles used in non-Jewish burials in the Late Roman
period (first through third centries C. E.); Gideon Avni and Zvi Greenhut, The
Akeldama Tombs.
76
For example, glass containers for kohl (used for eye make-up), see Dayagi-Men-
dels, Perfumes and Cosmetics, 3658.
77
Ibid. and Uza Zevulun and Yael Olienik, Function and Design in the Talmudic
Period (Tel Aviv: Haaretz Museum, 1978), 96105 (in Hebrew), 5157 (in Eng-
lish). For evidence from the First Temple period see Bloch-Smith, Judahite
Burial Practices, 90.
78
b. Sem. 8:7. Samuel, the Small, is a second generation Tanna (c. 90130 C.E.),
Strack and Stemberger, Introduction to the Talmud, 78.
79
This might explain why bottles are found more often buried with the bones of
women.
80
For a more expansive list of possibilities, see Hachlili, Jewish Funerary Customs,
390.
81
Virginia Anderson-Stojanovi c, The Chronology and Function of Ceramic Un-
guentaria, AJA 91, no. 1 (1987): 120. Although Anderson-Stojanovi cs main
172 Deborah Green
Temple period were made from pottery similar to these other house-
hold items. Rachel Hachlili has identified storage jars as being located
most often at the entrance to tombs, whereas cooking pots may
be found inside tombs on shelves, in pits, or even in kokhim.
82
This
is in contrast to Kloner who argues that cooking pots are not usually
found in kokhim.
83
A definite connection between the placement of
perfume bottles and cooking pots has yet to be determined.
84
And,
although several scholars have speculated on the meaning of cook-
ing pots as a vestige of the rite of meal offerings or as a symbol of
the commemorative meals executed in the Greco-Roman world they
have not tied the significance of the unguentaria to these food-oriented
theories.
In essence, the perfume bottles, cooking pots, and other grave goods
are symbolic of habits and activities that no longer occur in death
perhaps a striking reminder to the mourners of the death, separation,
and loss of the loved one. They may signify consolation either for the
mourners or for the dead that the loved one is not completely alone or
uncared for. Or, as Saul Lieberman indicates, these Jews may under-
concern is the Hellenistic and early Roman funerary unguentaria from Stobi in
Yugoslavian Macedonia, the presence of these bottles is so widespread that
Hachlili cites this article in her discussions on Second Temple burials in Israel;
Hachlili, Jewish Funerary Customs, 38384. Alabastra bottles are also found at
these sites, but in far fewer quantities. In the later Roman period, after 70 C.E.,
pottery unguentaria are replaced by glass bottles. Ibid., 38385.
82
Ibid., 386; Ann E. Killebrew, The Pottery, Jericho: The Jewish Cemetery in the
Second Temple Period (eds. Rachel Hachlili and Ann E. Killebrew; Jerusalem:
Israel Antiquities Authority, 1999), 123; Rachel Hachlili and Ann E. Killebrew,
Burial Customs and Conclusions, Jericho: The Jewish Cemetery in the Second
Temple Period (eds. Rachel Hachlili and Ann E. Killebrew; Jerusalem: Israel An-
tiquities Authority, 1999), 168.
83
Kloner and Zissou, The Necropolis, 6062, in which cooking pots are located in
the outer rooms (or tomb areas) and perfume bottles may be sealed in the ko-
khim or deposited in ossuaries. In the Akeldama tombs, for example, the per-
fume bottles are all found near and around the coffins and bones particularly
those burials of the later Roman period while the cooking pots and other jars
are found in other areas of the larger chambers. Avni and Greenhut, The Akel-
dama Tombs, 12329.
84
Unfortunately, many archaeologists in the past did not record the precise place-
ment of such items, and there is ample evidence of tomb robbing and secondary
use of tombs. As a result, we may never be able to determine whether there is a
connection between the placement of unguentaria and cooking pots.
Sweet Spices in the Tomb 173
stand their dead as sensate beings who are still able to hear and feel.
85
As such, these artifacts, whether personally owned or not, might orient
and comfort the dead or be necessary items for resurrection.
86
Another
possibility is that these items mark a liminal or ambiguous stage in the
lifecycle of both participant and family members. The personal effects
that travel with the individual may be symbolic of this transitory
stage in which the individual is still a member of the family or commu-
nity but not in the same manner as before.
87
Any of these reasons could
either be vestiges of ancient customs that then became regular rites
performed in honor and memory of the dead or the continuation of
an emotionally and ritually significant embodiment of then-current
belief.
In sum, if we consider that each stage in the burial process reflects
love, honor, and respect for the dead, as demonstrated by washing and
care for the body, the lighting of incense and lights before the bier, and
interment with personal effects or grave goods, then it seems reasonable
that the spices at each stage are a symbol in some manner of the esteem
felt for the dead by those who grieved for them. And, while it is possible
that perfume bottles found in graves or at burial sites had other pur-
poses as well, the simplest reason for their presence seems to be that
they belonged to the dead and, as such, were valued by the living.
85
Saul Lieberman, Some Aspects of After Life in Early Rabbinic Literature,
Harry Austryn Wolfson Jubilee Volume (vol. 2; Jerusalem: American Academy
for Jewish Research, 1965). David Kraemer, The Meanings of Death, also ad-
dresses this issue at some length.
86
The Babylonian and some of the Palestinian rabbis maintained that at the time
of resurrection, the dead would arise in the same clothes which they wore when
they were buried, according to Lieberman, Some Aspects of After Life in Early
Rabbinic Literature, 51011.
87
I thank Annal Frenz and her keen insights on the similarities between marriage
and death for this point.
174 Deborah Green
Patronal Relations
and Changes in Burial Practices
177
John Bodel
Chapter 6
From Columbaria to Catacombs:
Collective Burial in Pagan and Christian Rome
1
Mommsen fait ma desolation.
Il entre dans lerudition ecclsiastique
comme un rhinoceros dans un champ de vigne,
crasant droite et gauche, sans smouvoir du dgat.*
Students of classical Roman institutions and scholars of early Chris-
tianity have not always seen eye to eye, nor do historians and archaeo-
logists invariably agree. More than a century and a half ago, two of the
greatest, Theodor Mommsen and Giovanni Battisa de Rossi, found
mutual inspiration and took equal pleasure in debating the role of
Roman associations (collegia) in burial at Rome during the first three
centuries of the common era; together they set a high standard for pro-
ductive and amicable disagreement on a subject central to our con-
1
Louis Duchesne, Director of the cole franaise de Rome, in a letter to Giovanni
Battista De Rossi from Paris, 13 November 1892, quoted by Jonathan S. Perry,
The Roman Collegia: The Modern Evolution of an Ancient Concept (Leiden and
Boston: Brill, 2006), 56. My thanks are due to many: first, to the editors and es-
pecially to Laurie Brink, for the inspiration and dedication needed not only to
produce this volume but to arrange the splendid study tour and conference that
preceded it; to Patricia Duncan and Bradley Peper, my learned and tactful re-
spondents in Chicago; to my fellow participant Carolyn Osiek, who provided
detailed criticism on a subsequent written draft; to Simonetta Serra, whose pro-
bing skepticism and generously shared knowledge helped to make the argument
less vulnerable; to Vincenzo Fiocchi Nicolai, for valuable conversation and ex-
pert opinion at a late stage; and to responsive audiences in Chicago, New York,
New Haven, Cologne, Bonn, Munich, and Rome, who improved individual points
in more ways than can be mentioned. For all the help given, the essay ought to
be better than it is; remaining errors of judgment and fact are the authors re-
sponsibility alone.
**
178 John Bodel
cerns.
1
This chapter (to compare small things to great), the product of
an equally amicable and stimulating collaboration and debate, hopes
to cultivate the more useful elements of that example without wreaking
unnecessary havoc in the vineyard.
Burial at Rome:
Problems of Evidence and Interpretation
Burial space in ancient Rome was always limited and frequently con-
tested. This was true from the beginning of the Iron Age in Italy,
around 900 B.C.E., when the few cremation graves in what later be-
came the Roman Forum began to be intermingled with inhumation
burials of the sort found among the indigenous peoples of the Apen-
nine hills, to the fourth century C.E., when Constantine destroyed an
early imperial necropolis along the Via Triumphalis in order to build
an imposing funerary basilica above the spot on the Vatican hill be-
lieved to mark the tomb of St. Peter.
2
During the three and a half cen-
1
On this rivalry, see the illuminating discussion of Perry, Roman Collegia,
2360.
2
For the early Iron Age culture in Latium, see Timothy J. Cornell, The Beginnings
of Rome (London: Routledge, 1995), 4853. The discovery in early 2006 of a few
late Bronze Age pit inhumation burials in the Forum of Caesar on the lower
slopes of the Campidoglio has pushed back by about a century the earliest
known graves and thus the earliest evidence of human habitation in the area.
Constantines basilica rose on the site of an earlier (mid-second century) shrine
to St. Peter: see Jocelyn Toynbee and John Ward Perkins, The Shrine of St. Peter
and the Vatican Excavations (London: Longmans, Green and Co., 1956), 1213
for the destruction of existing tombs; further on the Via Triumphalis necropolis,
Eva Margareta Steinby, Caterina Coletti, M.-B. Carre, and Maria Teresa Ci-
priano, La necropoli della via Triumphalis il tratto sotto lautoparco Vaticano
(Atti della Pontificia Accademia Romana di Archeologia, ser. 3. Memorie, vol. 17;
Rome: Quasar, 2003), esp. 2239. For the history of the site, see briefly Mary
Beard, John North, and Simon Price, Religions of Rome, vol. 1 (Cambridge:
Cambridge University Press, 1998), 26869 (shrine of the 170s), 36869 (Con-
stantines basilica), 37677 (subsequent building), with further bibliography. It
was not until 563 C.E., when the First Council of Braga reversed the longstand-
ing Roman prohibition against intramural graves by allowing burials around the
walls of churches, that the competition for burial space in the suburban zone
began to ease: see R. Naz, Dictionnaire de Droit Canonique, vol. 3 (Paris: Letou-
zey et An, 1942), col. 730; cf. Orma Robinson, The Roman Law on Burials and
Burial Grounds, The Irish Jurist 10 (1975): 186.
179
turies of pre-Christian imperial Rome between the reigns of Augustus
and Constantine, when the population of the city numbered between
750,000 and 1,000,000 inhabitants, the suburbs of the city must have
accommodated between 10,500,000 and 14,000,000 burials.
3
Of these
we have traces of perhaps 150,000 or less than 1.5 percent of the total.
4
In generalizing about broad trends, even during the best documented
periods, we should not forget how little we know. Nor does the surviv-
ing evidence provide a representative selection of all Roman burials;
our sample is biased by the chances of survival and recovery and in-
herently tends to favor commemorative monuments and epitaphs over
unmarked and anonymous graves and thus tells us mainly about com-
paratively privileged segments of the population.
Columbaria and Collegia
If the state of our evidence raises one set of problems, our explanations
of it raise another. Major changes in Roman funerary behavior have
traditionally been linked to changes in the social or cultural order
mass migrations to the city in one scenario, the arrival of a new
religion in another but the conventional hermeneutical strategy of
interpreting historical outcomes in light of their presumed historical
causes in this case meets an impasse in the evidence. The invention of
a new form of burial monument (the columbarium) around 25 B.C.E.
and the emergence in the management of funerary rites at about the
3
Estimates of the population of early imperial Rome and Italy continue to spark
controversy, in part because the confines of the city are variously defined
(or presumed), but scholarly consensus seems to have settled on a figure between
750,000 and 1,000,000 for Rome and its surrounding suburbium during the first
three centuries C.E.: see recently, Geofrey Kron, The Augustan Census Figures
and the Population of Italy, Athenaeum 93 (2005): 487 and n. 251 for Rome;
Rob Witcher, The Extended Metropolis: Urbs, Suburbium, and Population,
JRA 18 (2005): 126 and n. 44, 129. Elio Lo Cascio, Le procedure di recensus
dalla tarda repubblica al tardo antico e il calcolo dela popolazione di Rome, in
La Rome impriale: dmographie et logistique (Collection de lcole Franaise de
Rome 230) (ed. Catherine Virlouvet; Rome: cole franaise de Rome, 1997),
376. For ancient mortality rates, see John Bodel, Dealing with the Dead, in
Death and Disease in the Ancient City (eds. Valerie M. Hope and Eireann Mar-
shall; London: Routledge, 2000), 12829.
4
See the Appendix.
From Columbaria to Catacombs
180 John Bodel
same time of a social institution previously unconnected with them
(the collegium) have usually been linked together and explained as the
result of the demographic pressures created by a large influx of new
residents to the capital following the Augustan peace.
5
According to
this view, as the city expanded beyond the capacity of the traditional
social support network of families and private patrons to meet the
burial needs of an increasingly heterogeneous and rootless urban poor,
new mechanisms sprang up to fill the void. Social upheaval exposed
gaps in the system, which the organism then evolved in order to fill.
There is much of value in such an analysis, but columbaria were ex-
pensive structures and served no-less-privileged groups indeed, in
many respects, no different groups than traditional tomb monuments
of the same period.
6
Nor, as several recent studies have shown, did the
collegia replace the family in caring for the burial of the dead. What
evidence we have in fact shows families operating in their customary
roles within the framework (both administrative and architectural)
of the collegia and columbaria.
7
What these institutions represented in
5
So, e.g., Keith Hopkins, Death and Renewal (Cambridge: Cambridge University
Press, 1983), 21517; Nicholas Purcell, Tomb and Suburb, in Rmische
Grberstraen: Selbstdarstellung, Status, Standard: Kolloquium in Mnchen vom
28 bis 30. Oktober 1985 (eds. Henner von Hesberg and Paul Zanker; Munich:
Verlag der Bayerischen Akademie der Wissenschaften: in Kommission bei der
C. H. Beckschen Verlagsbuchhandlung, 1987), 3839; John Patterson, Patron-
age, Collegia and Burial in Imperial Rome, in Death in Towns: Urban Responses
to the Dying and the Dead, 1001600 (ed. Steven Bassett; Leicester: Leicester
University Press, 1992), 2223.
6
Columbaria represented a substantial financial outlay: Ian Morris, Death-Ritual
and Social Structure in Classical Antiquity (Cambridge: Cambridge University
Press, 1992), 4247. For the commercial trade and speculation in tomb monu-
ments, see also Stefan Schrumpf, Bestattung und Bestattungswesen im rmischen
Reich: Ablauf, soziale Dimension und konomische Bedeutung der Totenfrsorge im
lateinischen Western (Gtingen: V+R unipress/Bonn University Press, 2006),
16272, discussing (inter alia) CIL 6.9189 and the dossier of texts relating to col-
umbaria and attesting transactions involving large numbers of burial niches, often
in multiples of ten: CIL 6.7803 (10 columbaria, 40 ollae), 13557 (100 ollae), 15551
(65 ollae), 17780 (30 ollae), 28126 (20 columbaria, 43 ollae); further below, n. 81.
7
See, e.g., Patterson, Patronage, Collegia and Burial, 23; Hopkins, Death and
Renewal, 21314; Kinuko Hasegawa, The collegia domestica in the Elite Roman
Households: The Evidence of Domestic Funeral Clubs for Slaves and Freed-
men, in Studies in Latin Literature and Roman History XII (ed. Carl Deroux;
Collection Latomus 287; Brussels: Latomus, 2005), 260; Onno Van Nijf, The
Civic World of Professional Associations in the Roman East (Amsterdam: J. C.
Gieben, 1997), 33; further Jean Pierre Waltzing, tude historique sur les corpor-
181
most cases were not alternatives to the traditional mechanisms of sup-
port but overarching structures that enabled the family and individual
patrons to perform their conventional roles within newly defined so-
cial and physical contexts. Demographic change necessitated new so-
lutions to traditional problems but did not fundamentally alter the
social channels through which they were addressed.
Christian Catacombs?
The situation is similar with the advent of Christianity, the early im-
perial shift from cremation to inhumation, and the invention of cata-
combs. Despite repeated attempts to prove otherwise, what has rightly
been called the biggest single event in ancient burial the change in
practice of tens of millions of people across the western empire from
burning to interring their dead, which transformed the suburban land-
scape of Rome has persistently defied both theological and sociohis-
torical explanation.
8
That the main period of transition, from the late
first century C.E. to the late second century, roughly coincides with the
beginning of the spread of Christianity across the western empire no
doubt helps to explain the appeal of linking these two major cultural
shifts, one involving practice, the other belief, but no plausible causal
relationship between the two has ever been found.
9
As for catacombs, a
ations professionelles chez les Romains, vol. 4 (Brussels and Louvain: C. Peeters,
18951900), 50910, 51820, listing many examples. The continuity of the nu-
clear family as the primary social unit represented in epitaphs is demonstrated
by Richard Saller and Brent Shaw, Tombstones and Roman Family Relations
in the Principate: Civilians, Soldiers, and Slaves, JRS 74 (1984): 12456.
8
Quote from Morris, Death-Ritual and Social Structure, 31, who ultimately favors
a modified version of the demographic explanation: it was not people carrying
the idea but the idea itself that came from the east and percolated from the
top down, through the diffusion of Hellenistic culture among the upper classes,
rather than from the bottom up, via proselytic immigrants from Palestine and
Judaea.
9
The rate of growth of the early Christian community must have varied widely
over time and place, but it clearly began small, with fewer than 1,000 members
(mainly, no doubt, in the eastern Mediterranean) around the middle of the first
century, and grew by the end of the third century, on one widely accepted esti-
mate, to perhaps 10 percent of the population of the empire, or roughly six mil-
lion: see Keith Hopkins, Christian Number and Its Implications, JECS 6, no. 2
(1998): 185226, esp. 19295. The popular transition from cremation to inhu-
mation around Rome began around the middle of the first century and is evident
From Columbaria to Catacombs
182 John Bodel
handful of tendentious texts written during the third century by even
fewer Christian bishops and apologists about particular circumstances
at Carthage, Alexandria, and Rome has been taken to show that Chris-
tians of that era systematically segregated themselves in death in
special Christian cemeteries under ground, but the evidentiary value of
the testimony from such proselytizing polemicists is questionable, and
the archaeological record, though regularly pressed into argument, re-
mains equivocal.
10
In fact, early Christian bishops seem to have taken
little interest in the funerary behavior of contemporary Christians and
evidently exercised only minimal control over cemeteries before the
time of Constantine.
11
Of the three texts regularly cited to demonstrate
that Christians avoided burial with pagans, the only one dating to be-
already then among the lower classes (slaves and freedmen): see Franca Taglietti,
Ancora su incinerazione e inumazione: la necropolis dellIsola Sacra, in R-
mischer Bestattungsbrauch und Beigabensiten in Rom, Norditalian, und den Nord-
westprovinzen von der spten Republik bis in die Kaiserzeit (eds. Michael
Heinzelmann, Jacopo Ortalli, Peter Fasold, and Marion Witteyer; Palila 8;
Wiesbaden: Dr. Ludwig Reichert Verlag, 2001), 14958; cf. Steinby, La necrop-
olis della Via Triumphalis, 2930.
10
For the most frequently cited texts Tertullian, Scap. 3.1 (ca. 203 C.E.), Apol.
39.56 (ca. 197 C.E.); Cyprian, Epist. 67.6.2 (251 C.E.) (North Africa); Origen,
Hom. Jer. (ca. 200230) (Alexandria); and [Hippolytus], Philosoph. 9.12.14
(ca. 198217) and Traditio Apostolica 40 (Rome) see below, p. 205 and n. 54.
The difference between perception and reality must be weighed carefully in as-
sessing the value of such testimony: see Hopkins, Christian Number, 18687,
expressing skepticism. For a good recent statement of the accepted view, see Vin-
cenzo Fiocchi Nicolai, The Origin and Development of Roman Catacombs, in
The Christian Catacombs of Rome, History, Decoration, Inscriptions (eds. Vin-
cenzo Fiocchi Nicolai, Fabrizio Bisconti, and Danilo Mazzoleni; Regensburg:
Schnell and Steiner, 1999), 1317.
11
The last point is controversial, but see ric Rebillard, Religion et sepulture: lg-
lise, les vivants, et les morts dans lantiquit tardive (Paris: ditions de lcole des
Hautes tudes en Sciences Sociales, 2003); Mark J. Johnson, Pagan-Christian
Burial Practices of the Fourth Century: Shared Tombs? JECS 5, no. 1 (1997):
4049; and Jill Harries, Death and the Dead in the Late Roman West, in Death
in Towns: Urban Responses to the Dying and the Dead, 1001600 (ed. Steven
Bassett: Leicester: Leicester University Press, 1992), 61, all emphasizing the dis-
tinction between official concern for martyrs tombs and relics and the general
lack of interest in the burial of ordinary Christians; see also Osiek, below in this
volume; and below, pp. 2024. It is not until the Council of Paderborn in 785 C.E.
that a general rule was promulgated that Christians be buried in church cem-
eteries rather than in pagan tombs: see Charles Hefele, Histoire des Counciles
daprs les documents originaux, vol. 3 (trans. Henri Leclerq; Paris: Letouzey et
An, 1910), 994.
183
fore the time of Constantine shows exactly the opposite. In a letter of
251 C.E. to the clergy and Christians of Spain, Cyprian accuses two
Spanish bishops, Basilides and Martial, of having accepted letters of
idolatry during the persecutions of Decius and complains that Martial
had long been frequenting the disgraceful and filthy banquets of a
pagan collegium and had placed his sons in the same collegium and, in
the manner of foreign peoples, had buried them with strangers among
profane graves. At the time, Cyprian was quarrelling with Stephen,
bishop at Rome, who had previously reinstated both Basilides and
Martial to their sees, so his charges against the two Spaniards were
hardly disinterested and must be seen as part of a larger polemic with
his Roman rival. More importantly, Stephens actions show that, re-
gardless of Cyprians opinion, when Martial as bishop had buried his
sons according to the practices of a pagan funerary collegium, he did
not violate any ecumenical principle serious enough to prevent his sub-
sequent rehabilitation and moreover had done so openly and without
fear of retribution. Whatever the currency of Cyprians views among
the Christian community at Carthage, the attitude they represented
was parochial and had no authority over or bearing on Christian
burial practices in Spain and at Rome.
12
This is not the place to review the remaining literary and archeologi-
cal evidence in detail, but a simple demographic consideration may
perhaps illustrate one problematic aspect of the current orthodoxy. If
we accept the consensus opinion that the population of imperial Rome
comprised between 750,000 and 1,000,000 residents during the third
century, and if we further accept a reasonable estimate of the size of the
Christian community at Rome of possibly as many as 7,000 at the be-
ginning of that century and follow a plausible model of its projected
growth across the empire of 40 percent per decade, then we must sup-
pose that by the middle of the century the community counted some
37,000 members, and about the time that Diocletian began persecuting
Christians systematically, during the first years of the fourth century,
they numbered around 200,000 at Rome and constituted between 20 and
12
Cyprian, Epist. 67.6.2, Martialis quoque praeter gentilium turpia et lutulenta
convivia in collegio diu frequentata et filios in eodem collegio exterarum gentium
more apud profana sepulcra depositos et alienigenis consepultos The other two
passages Hilary of Poitiers, Mat. 7.11 and Theodoret, Graec. affect. curatio
8.29 belong to the middle of the fourth century and to the first half of the fifth
century respectively. On all three see Johnson, Pagan-Christian Burial Prac-
tices, 4546.
From Columbaria to Catacombs
184 John Bodel
27 percent of the urban population.
13
Those figures compare favorably
with the most careful existing attempt to estimate the number of Chris-
tians at Rome from the literary sources, made originally more than a
century ago, which arrived at a figure for the middle of the century of
30,000.
14
(At the same rate of growth, the community of Roman Chris-
tians would have surpassed 750,000 by the year 340 and 1,000,000 by
the year 350. Once Christians constituted nearly 100 percent of the
urban population, then virtually all Roman burials, whether in cata-
combs or elsewhere, would naturally in some sense have been Chris-
tian. At this point, of course, the notion of purely Christian cata-
combs becomes unproblematic.
15
)
13
Calculating the size of the early Christian community is fraught with difficulties,
but one recent estimate for the entire empire places their number around
210,000 at the beginning of the third century and close to six million at its close,
with a disproportionate concentration in urban centers, especially in the eastern
Mediterranean: see Hopkins, Christian Number, 192 and Fig. 1. For the size
of the Christian community at Rome around 200 C.E., see Robert M. Grant,
Early Christianity and Society: Seven Studies (San Francisco: Harper and Row,
1977), 6. For the growth rate of 40 percent per decade and other relevant figures,
see Rodney Stark, The Rise of Christianity: A Sociologist Reconsiders History
(Princeton, N. J.: Princeton University Press, 1996), 413, 12945 (on cities across
the empire). The actual (projected) number of Christians at Rome in 250 C.E.
would be 37,647; in 300 C.E. 202,474.
14
Adolf von Harnack, Die Mission und Ausbreitung des Christentums in den ersten
drei Jahrhunderten, vol. 2 (4th ed.; Leipzig: J. C. Hinrichs, 1924), 806, relying pri-
marily on a famous letter written by Cornelius, bishop at Rome in 251253, to
Fabius, bishop of Antioch, boasting of the diversity and number of episcopal
personnel at the capital 46 priests, 7 deacons, 7 subdeacons, 42 acolytes, 52
exorcists, and various lectors and claiming that 1,500 widows and poor persons
were supported by communal charity: Eusebius, Hist. eccl. 6.43.11. All such fig-
ures must be regarded with caution: see Hopkins, Christian Number, 18792;
cf. Luce Pietri, Les resistances: de la polemique paenne la persecution de
Diocltien, in Histoire du Christianisme des origines nos jours II. Naisance
dune Chrtient (250430) (ed. Luce Pietri: Paris: Descle-Fayard, 1995), 134.
15
The actual (projected) figures: in 340 C.E., 777,826 Christians at Rome; in
350 C.E., 1,088,956. Across the empire a rate of growth of 40 percent per decade
would have resulted in some 33,880,000 Christians by 350 C.E., or 56.5 percent
of a population of 60 million: see Stark, The Rise of Christianity, 7 and Table 1.1.
Of course as the absolute numbers, and thus the percentage of a more or less
static (if not actually declining) urban population, increased, the rate of growth
would have slowed, but the contours of an exponential curve make it clear that
by the middle of the fourth century virtually all those buried in catacombs were
probably in some sense Christian. Who counted as Christian, of course, is
problematic and largely a matter of perception: among early Christian writers,
185
If we then compare this hypothetical number of urban Christians
around the year 300 with the estimated number of individual burial
spaces provided by the Roman catacombs before the time of Constan-
tine some 41,800 (see the Appendix) we confront a notable para-
dox. If virtually all the spaces in these purpose-built catacombs were
occupied by Christians, as is regularly maintained, and if Christians
died at the same presumed rate as other Romans (roughly 40 per 1,000
per year), then, supposing that the rate of growth of the urban Chris-
tian community over the course of the century was constant (it almost
certainly was not, but we are here interested only in a hypothetical
model), some 234,500 Christians would have died at Rome over the
course of the century. Bearing in mind the general figures proposed
earlier for the percentage of burials of all types at Rome during the
three and a half centuries before Constantine for which we have any
evidence at all (less than 1.5 percent), we must then conclude that re-
markably we have evidence for nearly 18 percent of all the Christian
burials of the third century but only a minuscule portion less than
one hundredth of one percent for the disposition of others who died
at Rome during the same period. All of these figures are crude esti-
mates and any could be disputed, but even if each were off by 50 per-
cent we would still be left with a disproportionately high represen-
tation of Christian graves. That is possible, of course, but perhaps
unlikely. Even if we grant the pious efforts of later generations of
Christians to cultivate the tombs of their early brethren (to say nothing
of the cult of the martyrs), and even if we note the general lack of
Christian concern for the graves of non-believers (to say nothing of the
willful destruction of them during the middle ages), a more plausible
hypothesis might suppose that the evidence we have for Roman burial
during the third century is more or less equally unrepresentative of all
groups and that Christians and non-Christians filled the burial spaces
of the catacombs, as well as the other sorts of graves in use during the
period, in numbers more or less reflective of their numbers within the
general population.
If that is so, then not only did Christian dogma about the fate of the
soul after death have little to do with the broad change in preferred
the term was a persuasive, hopeful and often porous category, used optimisti-
cally: Hopkins, Christian Number, 18687, quote from 187. Among lapsed
believers who abrogated their faith during the persecutions, on the other hand,
many must subsequently have regarded themselves as (again) Christian.
From Columbaria to Catacombs
186 John Bodel
method of disposal that accompanied the rise of Christianity in the west,
but the impulse of urban Christians to be inhumed in collective cem-
eteries underground may not have particularly distinguished them from
their pagan and Jewish contemporaries.
16
Two points, related but dis-
tinct, are relevant in this regard. First, and most easily demonstrated,
the creation of private cemeteries by groups identified by a common re-
ligious bond was not peculiar to the Christians at Rome. In addition to
half a dozen well-recognized Jewish catacombs , we may note, for in-
stance, during the same periods and in the same regions in which the so-
called Christian catacombs were being created and developed, collective
cemeteries established by collegia of Aesculapius and Hygia and of Sil-
vanus beside the Via Appia between the first and the third mile.
17
Indeed,
if a recent reassessment of the organization of Jewish burials at Rome is
correct, Roman Christians will have learned the use of catacombs from
the local Jewish community, who had developed their own subterranean
burial grounds at Rome beginning already in the first century B.C.E.,
16
The idea that Christian ideology influenced the change in preferred method of
burial goes back to a misapprehension of Minucius Felix, Octavius 11.4, on the
Christian condemnation of cremation. See the famous refutation of Arthur D.
Nock, Cremation and Burial in the Roman Empire, HTR 25 (1932): 32159;
repr. in Arthur Darby Nock: Essays on Religion and the Ancient World (ed. Zeph
Stewart; Cambridge, Mass.: Harvard University Press, 1972), 277307 and, on
Christian imagery in funerary art of the same period, Sarcophagi and Sym-
bolism, AJA 50 (1946): 14070; repr. in Arthur Darby Nock: Essays, 60641. For
pagan catacombs, see Philippe Pergola, Il praedium Domitillae sulla via Ar-
deatina: analisi storicotopografica delle testimonianze pagane fino alla met del
III sec. d.C., RACrist 55 (1979): 31335, hedging the argument, and the dis-
cussion among Pergola, Umberto Fasola, and Vincenzo Fiocchi Nicolai, in Actes
du XIe Congrs International dArchologie Chrtienne (Rome: cole franaise de
Rome and Pontificio Istituto di Archeologia Cristiana, 1989), 2:120710.
17
For the Jewish catacombs of Rome, see Cinzia Vismara, I cimiteri ebraici di
Roma, in Societ romana e impero tardoantico II. Roma. Politica, economia,
paesaggio urbano (ed. Andrea Giardina: Rome: Laterza, 1986), 35192, 490503
and (less reliably) Leonard V. Rutgers, The Jews in Late Ancient Rome: Evidence
of Cultural Interaction in the Roman Diaspora (Leiden: E. J. Brill, 1995). For col-
lective cemeteries of other religious groups, see Lucrezia Spera, Il paesaggio sub-
urbano di Roma dallantichit al medioevo. Il comprensorio tra le vie Latina e
Ardeatina dalle Mura Aureliane al III miglio (Rome: LErma di Bretschneider,
1999), 53 (Aesculapius and Hygia: a schola and burial facilities on the left side of
the Via Appia between the first and second mile, ca. 150 C.E.; cf. CIL 6.10234;
ILS 7213; AE 1937, 161); 13839 (Silvanus: between the second and the third
mile, second century; cf. CIL 6.10231 = ILS 7313); further 35861, 46364; and,
in general, Waltzing, tude Historique, 1:19798.
187
and who, like Diaspora Jews elsewhere throughout the Mediterranean,
relied on family for burial arrangements and followed the funerary cus-
toms and organization of the local population.
18
Second, and more importantly, not only is it obvious that not all fol-
lowers of a particular religion were buried in such collective monu-
ments, it is equally clear that burial in such places was not normally re-
stricted to devotees of a particular religion. Occasionally one finds
testators or the owners of tombs attempting to prescribe burial within
them to followers of a specific sect an epitaph of the third century
from the catacombs of Domitilla, for example, declares that a certain
M. Antonius Res(ti)tutus made the hypogeum for himself and his
household trusting in the Lord; another (which may or may not be
Christian) of late-second-century date from beside the Via Nomentana
identifies a monument owned by a Valerius Mercurius and two other
persons and intended for freedmen and descendants who belong to
my religion but these declarations, which (it may be noted) do not
explicitly exclude burial to anyone but merely designate certain cat-
egories of persons welcomed, are the exceptions that prove the rule.
19
More often with collective monuments established by groups linked
with a specific religious identity we find envisaged the possibility if not
the actual fact of the burial of others who do not belong to the same
sect. So, for example, the foundation document of the collegium of
Aesculapius and Hygia explicitly allows for a member to bequeath his
place to a son, or brother, or freedman, without specifying any other
qualification than payment of a funerary fee.
20
18
Margaret H. Williams, The Organisation of Jewish Burials in Ancient Rome
in the Light of Evidence from Palestine and the Diaspora, ZPE 101 (1994):
16582, especially 17582, arguing convincingly for the absence of control by
synagogues over the burial arrangements of Jews at Rome.
19
ILCV 1597 = ICUR 6555, M. Antonius Restutus fecit ypogeu(m) sibi et suis
fidentibus in Domino; CIL 6.10412 = ILCV 3824 = ILS 8337, Monumentum
Valeri Mercuri et Iulittes Iuliani et Quintilies Verecundes libertis libertabusque
posterisque eorum ad religionem pertinentes meam ; cf. also ILCV 3681 = AE
1923, 66, Faltoniae Hilaritati dominae filiae carissimae quae hoc coemeterium a
solo sua pecunia fecit et hu{h}ic religioni donavit with ric Rebillard, KOIMH-
THRION et COEMETERIUM: tombe, tombe sainte, ncropole, MEFRA 105
(1993): 979, Osiek, below, p. 247; CIL 6.10411 = ILCV 3826.
20
CIL 6.10234: si quis locum suum legare volet filio vel fratri vel liberto, dumtaxat
ut inferat arkae n(ostrae) partem dimidiam funeratici. Exclusivity was not un-
known in Roman collegia but was more characteristic of the associations organ-
ized by trades than of those defined by religions: see below p. 192.
From Columbaria to Catacombs
188 John Bodel
In fact, it is now generally recognized that throughout the third and
fourth centuries many Christians continued to be buried in familial
monuments of the traditional sort, long after catacombs came into
widespread use; that the Christian ideology of egalitarianism, which the
networked galleries supposedly promoted, is belied by their accom-
modation from the outset of privileged areas for monumental tombs set
off from the other burial spaces both architecturally and decoratively;
and that many of the so-called Christian catacombs originated from
and regularly incorporated subterranean pagan burial areas.
21
To these
now generally conceded points, especially the last, we may add that un-
equivocal evidence exists not only for the incorporation within Chris-
tian catacombs of earlier pagan hypogea but for the contemporary
burial side by side, throughout the third and fourth centuries, of Chris-
tians and pagans, not only within a single monument but in adjacent
subterranean spaces connected by tunnels and galleries. Among the
more striking examples of the latter are the catacombs of Agnese beside
the Via Nomentana, where pagan hypogea were left intact and access-
ible after a Christian cemetery was installed on the site, and the so-
called catacombs of Vibia next to the cemetery of Praetextatus at the
second mile of the Via Appia, where the frescoes decorating certain
arched inhumation niches (arcosolia) identify the burial spaces of three
priests of Mithras, a devotee of Sabazius, and his pagan wife, while
other inscriptions clearly mark the graves of Christians all, it seems,
dating from the second half of the fourth century.
22
21
See recently, e.g., Vincenzo Fiocchi Nicolai and Jean Guyon, Introduzione, in
Origine delle Catacombe Romane. Atti della giornata tematica dei Seminari di
Archeologia Cristiana (Roma 21 marzo 2005) (eds. Vincenzo Fiocchi Nicolai
and Jean Guyon; Vatican City: Pontificio Istituto di Archeologia Cristiana, 2006),
which Prof. Fiocchi Nicolai kindly shared with me in advance of publication;
for non-egalitarian design, see further, below p. 224. Notable examples at Rome
of Christian catacombs originating from pagan hypogea include the so-called
hypogeum of the Flavii in the cemetery of Domitilla and the hypogaeum of the
Acilii in the catacombs of Priscilla: see, e.g., Philippe Pergola, La region dite des
Flavii Aurelii dans la catacombe de Domitille, MEFRA 95 (1983): 183248 and
Antonio Ferrua, Iscrizioni pagane della Catacomba di Priscilla, Archivio della
societ romana di storia patria 110 (1987): 519, both with further references.
22
See Johnson, Pagan-Christian Burial Practices, 5059, with examples from
Rome and elsewhere. For S. Agnese, see also Umberto Fasola, La Regio IV del
cimitero di S. Agnese, RAC 50 (1974): 175205, with the remarks of Johnson,
Pagan-Christian Burial Practices, 50; for Vibia, see CIL 6.142 =ILS 3961 =CLE
1317 and Spera, Il paesaggio suburbano, 17475, 4001, with further references.
189
When all the exceptions are taken into account and all the qualifi-
cations have been duly noted, little remains at the foundation of the
conventional view but the conviction that purpose-built catacombs
must have been Christian from the outset because after Constantine
the cult of the martyrs caused Christians to vie for burial in close prox-
imity to the tombs of their saints and thus to expand the already exist-
ing networks of galleries into vast subterranean complexes, which
Christians (numerically predominant now, in the urban population)
naturally controlled and eventually monopolized. Whether the origi-
nal underground cemeteries developed during the third century were
exclusively Christian, on the other hand, is considerably less certain.
Such a situation is demographically improbable and archaeologically
dubious and, on the basis of the ambivalent literary sources available
to us, must remain decidedly open.
Collective Burial
The conventional explanations of these two major innovations in burial
practices during the first three centuries C.E. the invention of colum-
baria and the development of catacombs fail to convince because the
new funerary forms did not in fact respond to the social pressures that
are thought to have given rise to them: columbaria and collegia did not
replace families and patrons, and catacombs were not invented and de-
signed to accommodate the particular ideological beliefs and religious
behavior of early Christians. What columbaria and catacombs have in
common, and what distinguishes them from other Roman monument
types, is their capacity to accommodate burials in groupings larger than
and different from the traditional units of the nuclear family (normally
buried in so-called sepulchra hereditaria) and the extended household
provided for in sepulchra familiaria, which included, in the well-known
epitaphic phrase, freedmen and freedwomen and their descendants
(libertis libertabusque posterisque eorum).
23
If we start from this obser-
23
For the basic definition, see Dig. 11.7.5 (Gaius), Familiaria sepulchra dicuntur
quae quis sibi familiaeque suae constituit, hereditaria autem quae quis sibi heredi-
busque suis constituit. The distinction was purely a matter of law: see Sergio Laz-
zarini, Sepulcra familiaria: unindagine epigrafico-giuridica (Milan: CEDAM,
1991), 711 (on Dig. 11.7.6.pr, below n. 73) with Fernand de Visscher, Le droit
des tombeaux romains (Milan: Giuffr Editore, 1963), 93102. Architecturally
there was no difference in form between the two types, and the legal distinction
From Columbaria to Catacombs
190 John Bodel
vation and ask first what possibilities for collective burial the new
architectural forms encouraged and only then proceed to consider
what social purposes they may have served, we may perhaps avoid
some of the pitfalls that have impeded progress from the opposite di-
rection. First, however, it will be necessary to establish one basic me-
thodological point about the analysis of ancient burial practices and to
dispel two common misconceptions about Roman funerary behavior
that have led some otherwise valuable investigations astray.
Wild Geese and Red Herrings
Method first. In considering the disposition of collective cemeteries we
must resist the tendency to assume that burial arrangements for the
dead corresponded directly to the social organization of the living, and
in particular we must avoid succumbing to the housing fallacy ac-
cording to which the internal configuration of columbaria and cata-
combs somehow reflected the distribution of space in contemporary
domestic architecture.
24
The social arrangements articulated within
a columbarium or catacombs are unlike those ever lived out in a house
or apartment block. Romans did not, in fact, live as they died, nor did
they arrange themselves in death as they did in life: the easy analogies
break down almost as soon as they begin to be examined, whether we
between them could become blurred in practice: see e.g., Valerie Hope, A Roof
over the Dead: Communal Tombs and Family Structure, in Domestic Space in
the Roman World: Pompeii and Beyond (eds. Ray Laurence and Andrew Wallace-
Hadrill; Portsmouth, R.I.: Journal of Roman Archaeology, 1997), 6988; Mario
Amelotti, Una Visita a Pietro e a Popilio Eracla, in Collatio iuris romani:
tudes ddies Hans Ankum loccasion de son 65e anniversaire (eds. R. Feen-
stra, A. S. Hartkamp, J. E. Spruit, P. J. Sijpestein, L. C. Winkel; Amsterdam:
J. C. Gieben, 1995), 45.
24
Purcell, Tomb and Suburb, 39, for example, corrects Hopkinss comparison of
columbaria to insulae (Death and Renewal, 214) by arguing that the housing
which parallels it [the columbarium] is the domus, Each view has something to
recommend it, but both are fundamentally incorrect. More cautious (and more
accurate) is the view of Hope, A Roof over the Dead, who likens familial and
household structure to tomb configuration only to the extent that both were
fluid and changed unpredictably. For the organization of space within the
Roman house, see Andrew Wallace-Hadrill, The Social Structure of the Roman
House, PBSR 56 (1988): 4397 and, briefly, id., Houses and Society in Pompeii
and Herculaneum (Princeton, N. J.: Princeton University Press, 1994), 1416.
191
consider the positioning of sarcophagi within the tomb of the Scipios
or the placement of grave markers in the familial tombs of imperial
Romans of moderate means. A recent analysis of a selection of early
imperial tombs from Isola Sacra and from the section of the Via
Triumphalis necropolis under the Vatican parking lot, for example,
shows that in many cases no particular burial place within a monu-
ment was marked out more than any other and that where a hierarchi-
cal arrangement privileging a central location is found, the principal
focus was less often centered on the proprietor of the tomb or the
paterfamilias than on the first person buried in the monument or on
one whose principal claim seems to have been the affection in which he
or she was held by the owner.
25
It hardly requires observing that Romans of the empire grouped them-
selves socially according to various criteria, depending upon the context.
Within the home, familial ties and hierarchies dominated, but outside the
household, various criteria might dictate not only membership but rank
within a group. Domestic collegia regularly and naturally subsumed fam-
ilial structures in organizing the burial behavior of their members within
collective monuments, but other formal or semiformal voluntary associa-
tions for instance, those that channeled political activity or access to
social services bore more complex and variable relationships with the
funerary activities of their constituents. The vici of Rome, for example,
provided administrative structure and corporate organization for vari-
ous political, social, and religious activities, but residency in a neighbor-
hood played little, if any, role in determining funerary behavior.
26
Simi-
larly, at Pompeii neighborhood groups exhibited corporate organization
and acted collectively at times in endorsing local politicians as did pro-
fessional collegia of dyers (infectores) and fruit-sellers (pomarii), reli-
gious associations such as the devotees of Isis (Isiaci universi ), the inhab-
25
Francisca Feraudi-Grunais, Inschriften und Selbstdarstellung in stadtrmi-
schen Grabbauten (Libitina 3; Rome: Quasar, 2003), 2542, esp. 4142. Even
when an original decorative scheme or arrangement of burial places seems de-
signed to focus attention on a particular location, the practicalities of continued
use of the monument frequently subvert the program, notably when subsequent
graves intrude into the decorative framework: in brief, functionality trumped
aesthetics in determining how burial spaces were used (4254, esp. 43).
26
See J. Bert Lott, The Neighborhoods of Augustan Rome (Cambridge: Cambridge
University Press, 2004), 4560, 10617. For the mistaken idea that the region in
which Jews in Rome were buried depended upon the location of their syna-
gogues, see Williams, Organisation of Jewish Burials, 16575.
From Columbaria to Catacombs
192 John Bodel
itants of a single residential block (the insula Satriorum), the patrons (it
seems) of the baths of Venus (Veneri or Veneriosi), and informal groups
waggishly identifying themselves as petty-thieves (furunculi), gladiator-
ial fans (spectaculi spectantes), sleepers (dormientes), and late-drinkers
(seri bibi) but none of these groups appears anywhere in the organiza-
tion of Pompeian cemeteries or tombs.
27
In certain cases, however, corporate identity in life might be reframed
(rather than replicated) in death. Groups of clients collectively endors-
ing a patron for political office might expect in return to have their burial
arrangements provided for, and worshippers of a particular cult, as
noted earlier, might band together in death in collective cemeteries. Pro-
fessional associations sometimes projected their exclusivity into their fu-
nerary facilities: at Rome a collegium of cooks in the imperial household
had a special burial complex between the second and third mile of the
Via Appia, and an association of ivory and citron-wood workers, which
probably provided burials as well as banquets, restricted membership
to practitioners of those trades.
28
Soldiers lived a highly regimented
life and found in their military units a social institution that provided
much of the structure traditionally associated with the family. Special
units of them at Rome, such as the equites singulares and the detachment
of sailors from the Misene fleet assigned to rigging the awnings above
the Colosseum, segregated themselves in separate burial grounds, but
others of no less specialized and even more lite status, such as the prae-
torian guards, preferred individual burial among civilians in the general
necropoleis outside the city gates.
29
Even with corporate institutions, the
27
For the topographical organization of Pompeii, see Paavo Castrn, Ordo Popu-
lusque Pompeianus: Polity and Society in Roman Pompeii (2nd ed.; Rome: Bardi
Editori, 1983), 7982. For the groups of electoral supporters, see James L.
Franklin, Pompeii: The Electoral Programmata, Campaigns and Politics, A.D.
7179 (Rome: American Academy in Rome, 1980), 2124, with references. For
the humorously named pseudo-collegia, Waltzing, tude historique, vol. 1, 51 n. 2
aptly compares the jocular references in Horace (Serm. 1.2.1) and Apuleius
(Metam. 7.7.4) to associations of Syrian flute-girls (ambubaiarum collegia) and
bandits (latronis collegium) respectively.
28
For the imperial cooks, see Spera, Paesaggio suburbano, 18788 with CIL
6.7458, 8750, 8751, the last reused to cover a loculus in the nearby catacombs of
Callistus, a section of which, to judge from a pair of graffiti (ICUR 14815ab),
was evidently known as the regio cocorum. For the collegium of ivory and citron-
wood workers, see CIL 6.33885 = ILS 7214.
29
For the legion as a social institution, see Ramsay MacMullen, The Legion as a
Society, Historia 33 (1984): 44056. For the equites singulares, see Jean Guyon,
193
correspondence between membership in life and community in death
was variable and unpredictable. It stands to reason, then, that principles
of organization and hierarchy recognizable in one are seldom found
reflected in the other. Burial architecture in antiquity did not intend to
replicate the circumstances of the living but instead enabled abstract ex-
pressions of ideal social orders that seldom, if ever, corresponded di-
rectly to the way human relationships were enacted in life.
30
Finally, before we turn to the problematic question of definition,
it will be necessary to dispel two common misconceptions about the
so-called sepulchra familiaria, those which provided for freedmen and
freedwomen to be buried along with their patrons. First, it is not the
case, as is sometimes maintained, that this type of monument, which
was most characteristic of the second century C.E. and thus represents
the principal alternative method of group burial to columbaria at the
beginning of the century and to catacombs toward the end, first came
into use during the latter half of the first century.
31
Familial tombs
existed already during the final decades of the Republic, when (it
seems) it was ex-slaves themselves (often the innovators in commemo-
rative funerary behavior) who favored the form.
32
Whatever role fam-
ilial tombs played in the changing face of collective burial during the
Le cimitire aux deux lauriers: recherches sur les catacombes romaines (Rome:
cole franaise de Rome, 1987), 3033; for the Misene sailors, see Spera, Paes-
aggio suburbano, 158; for the praetorian guard, see Marcel Durry, Les Cohortes
Prtoriennes (Paris: De Boccard, 1968), 6063.
30
For the basic methodological point, see, e.g., Bruno DAgostino, Societ dei
vivi, communit dei morti: Un rapporto difficile, Dialoghi di archeologia 3,1
(1985): 4758 and, more polemically, Rick Jones, Rules for the Living and the
Dead: Funerary Practices and Social Organization, in Rmerzeitliche Grber
als Quellen zu Religion, Bevlkerungsstruktur und Sozialgeschichte (ed. Manuela
Struck: Mainz: Institut fr Vor- und Frhgeschichte der Johannes Gutenberg-
Universitt Mainz, 1993), 24754.
31
So, e.g., Francisca Feraudi-Grunais, Ubi diutius nobis habitandum est. Die
Innendekoration der kaiserzeitlichen Grber Roms (Palilia 9; Wiesbaden: Dr. Lud-
wig Reichert Verlag, 2001), 15253, and ead. Inschriften und Selbstdarstel-
lung, 36, apparently confusing the advent of the architectural form of the
Kammergrber with the juridical capacity for familial burial, which clearly
existed earlier.
32
The clearest indication is provided by epitaphs that include the standard phrase
libertis libertabusque (in various orthographies), e.g., CIL 1
2
1226, 1236, 1277,
1278, 1330, 1401, 1638; sometimes libertis alone, e.g., CIL 1
2
1286, 1308, 1313,
1355, 1398, 1568; occasionally in the complete formula with posteris also, e.g.,
CIL 1
2
1319, 1334, 2213; AE 1990, 345.
From Columbaria to Catacombs
194 John Bodel
second century, they were not first invented to respond to new needs
first arising then. Second, we should not imagine that familial tombs,
by including freedmen as well as family members within the monu-
ment, somehow reflect a growing magnanimity of patrons toward
their households. Roman landowners establishing testamentary foun-
dations and trusts to allow generations of ex-slaves to inherit their
monuments were less interested in preserving landed property within
their families than they were in perpetuating their own names by pas-
sing the properties on to their freedmen.
33
In this respect, what familial
tombs illustrate is not generosity but ego. Nor, on the other hand,
should we imagine that during the Republic Roman slaves were norm-
ally deprived of formal burial and that columbaria first made this
opportunity available to them. Since slaves did not have juridical
personae, testamentary dispositions and the epitaphic formulae that
reflect them naturally took no formal account of their ultimate fate in
death. Although definitive physical evidence of the gravesites of slaves
is difficult to identify, psychological plausibility and the apparently
commonplace presence of unmarked or anonymous graves within
tomb enclosures suggest that household slaves often found burial in
familial and even hereditary monuments.
34
Let us assume that the primary reason for the introduction of both
the columbarium and catacombs was the simple demographic need cre-
ated by a limited amount of land on the outskirts of Rome and the ever-
accumulating demands placed on it by successive generations of Ro-
mans united (whatever their religious beliefs or social circumstances)
by a persistent desire to bury their dead in the suburban soil and a re-
33
David Johnston, Trusts and Tombs, ZPE 72 (1988): 8187 and id. The Roman
Law of Trusts (Oxford: Oxford University Press, 1988), 7797.
34
Werner Eck, Rmische Grabinschriften. Aussageabsicht und Aussagefhigkeit im
funerren Kontext, in Rmische Grberstraen: Selbstdarstellung, Status, Stan-
dard (eds. Henner von Hesberg and Paul Zanker; Munich: Verlag der Bayerischen
Akademie der Wissenschaften: in Kommission bei der C. H. Beckschen Verlags-
buchhandlung, 1987), 7374 argued the point persuasively for the second and
third centuries C.E., and the same can be said for the period of the late Republic as
well. Eck believes that during the Republic dead slaves were unceremoniously
dumped in the Esquiline puticuli; I have argued elsewhere that disposal in these
loca publica was more often the fate of the indigent free: Dealing with the Dead,
12835. A contract of Augustan date for the undertaking concession at Puteoli
provides for regular, albeit less formal, burial for slaves: see AE 1971, 88 II.2223,
with Franois Hinard and Jean C. Dumont, Libitina: pompes funbres et supplices
en Campanie lpoque dAuguste (Paris: De Boccard, 2003), 13132 ad loc.
195
luctance to give up the commemorative habit of regular pilgrimages to
local grave sites. It remains to investigate how these new burial forms
accommodated those desires and what further possibilities for social
expression they may have supported or imposed. Here we are ham-
pered by some fundamental problems of definition, which we must
now address, for it is clear that those who write about columbaria, cata-
combs, and cemeteries do not always mean the same things by the terms.
Problems of Definition
Our problems of interpretation begin with terminology. Unfortunately
they are acute, since each of the three principal terms used to describe
collective burial spaces during the first three centuries C.E. colum-
barium, catacombs, and cemetery (otjtjtov or coemeterium) is,
in standard usage, a modern invention that remains elusively ambiva-
lent or imprecise.
Columbarium, in antiquity, meant dovecote, a nesting-box for a
pair of pigeons, and then, in a transferred sense, niche for an ash urn
or more precisely, since pigeons kept for breeding were (and are) norm-
ally kept in pairs, bipartite niche for a pair of urns, the most charac-
teristic form.
35
The word is never used in literary sources in this latter
sense but is found in some forty inscriptions, all but two from Rome or
Ostia.
36
This extended usage, like the form itself, evidently originated
at the capital and was virtually restricted to its environs.
In modern usage, however, the term columbarium does not normally
refer to the individual niches or cavities for urns but to the architectural
35
For the Roman dovecote (columbarium or, in fashionable Greek parlance,
rtotrotortov), see the description by Varro, Rust. 3.7.4: Singulis paribus
(sc. columbarum) columbaria fiunt rutunda in ordinem crebra, ordines quam plu-
rimi possunt a terra usque ad cameram; Round nesting places are made for each
pair (sc. of pigeons), close to each other in a row, and as many rows as possible
are built from the ground up to the vaulted ceiling.
36
Cf. Thesaurus Linguae Latinae 3:1734, s.v. columbarium. Of the two texts
found elsewhere, one is from Spain (CIL 2.2002); the other, from Antium, 30
miles southwest of Rome (CIL 10.8299), evidently describes a bipartite niche:
columbaria II ollarum IIII. See further Diz. Epigr. 2:46465, s.v. columbarium
(Ettore De Ruggiero); RE IV.1:593, s.v. columbarium (E. Samter). Other
transferred uses applied to the niches in walls to hold beams and, in ships, to
oarlocks (TLL). For the primary sense of dovecote, see also Pliny, Historia Natu-
ralis 17.51 and Palladius 1.24.
From Columbaria to Catacombs
196 John Bodel
structures that housed them large or small tomb buildings, built above
or below ground (or sometimes both) and distinguishable from other
monumental tombs mainly by their interior configuration, which is
marked by plastered walls and pillars systematically lined, from floor to
ceiling, with rows of niches accessible via wooden ladders or stairways
and scaffolding (Fig. 6.1). Regimented order and symmetry are charac-
teristic of the form. Each niche contained one or more cavities ollae
normally a pair, sometimes as many as four or six, usually incorporating
a terracotta funerary urn within the wall but occasionally allowing an
urn to be inserted and removed independently. Individual niches some-
times received personal attention small shelves built out of the wall to
hold offerings, a painted border or marble frame surrounding the niche,
personalized decoration, a terracotta or stone tablet cut to cover over
the opening to the niche and inscribed with the name of the owner or of
the person buried there but the interior decoration of the chamber was
normally uniform and sometimes divided the rows of niches by horizon-
tal bands of thematically related motifs (Fig. 6.2).
37
The architectural
form flourished from the time of Augustus to that of Hadrian, little
more than 150 years; as a funerary fashion, in other words, it was short-
lived. The termcolumbarium first appears with this modern sense in the
eighteenth century in reference to the very largest such structures, the
ones originally intended for the massive households of the great families
of Rome, and only became common in the nineteenth century.
38
Today it
37
For the decorative program of the largest of the columbaria in the Doria
Pamphilj necropolis (for which see below. n. 57), which exhibits themes similar
to (but less systematicaly disposed than) those in the nearby columbarium of
C. Scribonius Menophilus (see Fig 6.2), see Roger Ling, The Paintings of the
Columbarium of Villa Doria Pamphili in Rome, in Functional and Spatial
Analysis of Wall Painting (Proceedings of the Fifth International Congress on
Ancient Wall Painting) (ed. Eric M. Moormann; Leiden: Stichting Bulletin
Antieke Beschaving, 1993), 12735, emphasizing the generic (non funerary) na-
ture of the scenes represented, and esp. 129 on the uniformity of the design.
38
Maria Letizia Caldelli and Cecilia Ricci, Monumentum familiae Statiliorum: Un
Riesame (Rome: Edizioni Quasar, 1999), 76 n. 93 cite an unpublished tesi di
laurea by Simona Crea that traces the modern sense of the term to Antonio
Francesco Goris publication of the largest known columbarium, that of the
household of the empress Livia, under the title Monumentum sive columbarium
libertorum et servorum Liviae Augustae et Caesarum Romae detectum in Via
Appia anno MDCCXXVI (Florence, 1727). For this work and for Goris un-
stated rivalry with Francesco Bianchini, who published first, under the title
Camera ed inscrizioni sepolcrali de liberti, servi, ed ufficiali della casa di Augusto
scoperte nella Via Appia (Rome 1727), the same excavations of 1726 that un-
197
has grown in application to encompass almost any tomb monument
with niches for cremation burials in the walls.
In contrast to the origin of the term columbarium, catacombs has
no basis whatsoever in ancient terminology (the singular noun is a lexi-
cal anomaly) but derives instead from the Greek phrase oto upo
covered the monument, see Maria Raina Fehl, Archaeologists at Work in 1726:
The Columbarium of the Household of Livia Augusta, in Ultra Terminum Va-
gari. Scritti in onore di Carl Nylander (eds. Brje Magnusson, Stefania Renzetti,
Paolo Vian, and Sever J. Voicu; Rome: Edizioni Quasar, 1997), 8992.
From Columbaria to Catacombs
6.1. Drawing by Antonio Buonamici of the long wall of the first room of the
columbarium of the household of Livia beside the Via Appia (reproduced from
Francesco Bianchini, Camera ed inscrizioni sepulcrali deliberti, servi, ed ufficiali
della casa di Augusto scoperte nella Via Appia, ed illustrate con le annotazioni di
Monsignor Francesco Bianchini Veronese, lanno MDCCXXVI (Rome: Giovanni
Maria Salvioni, nellarchiginnasio della Sapienza, 1727), tav. IV [BAV Cicognara
VIII 3617A], after Maria Raina Fehl, Archaeologists at Work in 1726: The Col-
umbarium of the Household of Livia Augusta, in Ultra Terminum Vagari. Scritti
in onore di Carl Nylander (eds. Brje Magnusson, Stefania Renzetti, Paolo Vian,
and Sever J. Voicu: Rome: Edizioni Quasar, 1997), fig. 2.
198 John Bodel
(near the hollows) in reference to the abandoned quarries, sandpits,
and cisterns from which, later in the fourth and fifth centuries, profes-
sional gravediggers (fossores) often started digging the networks
of burial tunnels to which the term nowadays regularly applies.
39
In an-
39
So, convincingly, Bruno Luiselli, Il toponimo Catacumbas e Odilone di San
Medardo, MEFRA 98 (1986): 85254, finding a Latin parallel for the usage in
6.2. Columbarium of C. Scribonius C.l. Menophilus on the Janiculum hill beside
the Via Aurelia, main room, long wall, showing rows of niches with painted tabulae
ansatae beneath each niche. The rows are divided systematically by painted bands
of (from bottom to top): garlands; flowers, fruits, birds, and Dionysiac motifs;
sacro-idyllic landscapes; and narrative scenes with human figures. Several of the
niches are sealed with mortar or with stone or terracotta plaques; two, one of which
had been enlarged and enhanced with a marble tablet bearing an epitaph (in situ),
have had marble frames removed. (authors photo)
199
tiquity the phrase occurs specifically, and for a long time only, in the
topographical designation cimiterium ad catacumbas, in reference to
the subterranean Christian cemetery excavated out from the pozzolana
quarries near the third mile of the Via Appia in the vicinity of S. Seb-
astiano, where already by the middle of the third century there was an
important memorial cult of the apostles Peter and Paul.
40
Colloquial and vague, the original topographical designation leaves
open the question of a typological distinction between the purpose-
built networks of tunnels designed to accommodate hundreds of inhu-
mation burials, and other linked hypogaea and underground cemeteries,
which are occasionally found beside the Via Appia and elsewhere
the Passion of Saint Sebastian in the phrase ad arenas used to describe the place
where the martyrs Marcus and Marcellianus were buried (Act. Sanct., Ian.
II:642). The neologism catacumbas is first attested in the Chronica Urbis
Romae inserted in the register of the Chronographer of 354 and edited some
twenty years previously in reference to building activity in the region by the em-
peror Maxentius (fecit et circum in catecumbas). Otherwise in antiquity it occurs
only and always in reference to the Christian cemetery on the site (see the next
note). Giuseppe Marchi, Monumenti delle arti cristiane primitive nella metropoli
del cristianesimo (Rome: Tip. di C. Puccinelli, 1844), 209 and De Rossi, Roma
sotterranea 3:427, interpreted kymbas as deriving from Latin cubare, to sleep.
More recently, Antonio Ferrua, La basilica e la catacomba di S. Sebastiano (2nd
ed.; Vatican City: Pontificia Commisione di Archeologia Sacra, 1990), 11, has
associated it with cumba, ship, in reference to a hypothetical sign or relief in
the area advertising an inn and depicting two or more ships. Kata in the phrase
means in the vicinity of (a late usage), so that, strictly, the construction ad
catacumbas in the phrase cimiterium ad catacumbas is redundant. Such bilingual
tautologies are not uncommon in late Latin topographical designations. Petro-
logical explorations during the early 1940s showed that understanding of the
geophysical properties of the tufa quarries outside Rome was greater among
gravediggers during the first and second centuries than later during the third and
fourth, which explains why later diggers often built new complexes off of the
older networks: see Gioacchino De Angelis DOssat, La geologia delle cata-
combe romane (3 vols; Vatican City: Pontificio Istituto di Archeologia Cristiana,
19391943). For the role of the fossores, who not only dug but sold burials spaces
in the catacombs, see Jean Guyon, La vente des tombes travers lpigraphie de
la Rome chrtienne (IIIe VIIe sicles): le role des fossores, mansionarii, praepo-
siti, et prtres, MEFRA 86 (1974): 54996.
40
For the site and its designation, see Anna Maria Nieddu, Catacumbas ad, in
Lexicon Topographicum Urbis Romae. Suburbium (ed. Adriano La Regina;
vol. 2; Rome: Edizioni Quasar, 2004), 7982, on the name and Catacumbas
coemeterium. Cimitero sopratterra, ibid., 8286; Rafaella Giuliani, Cimitero
sotterraneo, ibid., 8693.
From Columbaria to Catacombs
200 John Bodel
outside Rome already during the second century.
41
Whereas the la-
bel columbarium aims to characterize an architectural form, the term
catacombs merely identifies a place. What is more, the burial mode
primarily associated with the phrase can only be inferred from develop-
ments of a later period than that when the expression was originally
employed. In reference to the earliest underground cemetery on the
site a concentrated grouping of columbaria and familial mausolea,
three of them disposed around a sunken piazzola with niches for
inhumations excavated out of the sides the term is not only anachron-
istic but misapplied.
42
The only distinctive architectural features
universally recognized are columns and rows of large niches for
inhumation burials (loculi ) lining the walls (Fig. 6.3). Since many cata-
combs incorporated or originated from earlier hypogea, and since the
developed versions regularly included quadrilateral burial chambers
(cubicula) similar in form to the earlier and independent subterranean
monuments, the question naturally arises when a series of linked hypo-
gea becomes a catacomb. The solution most commonly proposed is to
identify as proper catacombs only those that could be entered directly
from the ground and were designed from the outset to offer as many
spaces for inhumation burials as possible, but typological distinctions
are difficult to draw when one considers closely the various architec-
tural configurations of underground burial complexes outside Rome
during the second and third centuries, and in practice the term cata-
combs has come to be reserved for those that are presumed to have been
controlled by official religious authorities and therefore to be charac-
terized by exclusive religious affiliations (Christian, Jewish),
whereas those of indeterminate or private status are labeled hypogea. A
spurious semantic precision in this case does not conceal the circularity
of the reasoning, nor does it resolve the basic terminological problem.
43
41
For an overview of various types of subterranean burial complexes dating from
the second century, see Spera, Paesaggio suburbano, 35562.
42
See Spera, Paesaggio suburbano, 20925; Nieddu, Cimitero sopratterra, 8283.
43
For the standard distinction between Christian catacombs and pagan hypo-
gea, see, e.g., Fiocchi Nicolai, Origin and Development, 1617 and Leonard
V. Rutgers, Subterranean Rome: In Search of the Roots of Christianity in the
Catacombs of the Eternal City (Leuven: Peeters, 2000), 64. For a typically prob-
lematic case, compare the discussions of the so-called Catacombs of Vibia
by Spera, Paesaggio suburbano, 17276 and Palmira Maria Barbini, in Philippe
Pergola, Le catacombe romane: storia e topografia (Rome: La Nuova Italia
Scientifica, 1997), 17780.
201 From Columbaria to Catacombs
6.3. Catacombs of Domitilla, gallery with loculi, after Vincenzo Fiocchi Nicolai,
Fabrizio Bisconti, and Danilo Mazzoleni, The Christian Catacombs of Rome:
History, Decoration, Inscriptions (Regensburg: Schnell and Steiner, 1999), 76.
202 John Bodel
Coemeterium, the transliterated form of the Greek otjtjtov,
came into Latin usage in Christian contexts toward the beginning of
the third century, when it referred to the final resting place of a dead
person.
44
The classical Greek term, based on a verb used metaphor-
ically of the sleep of death already in the Iliad and attested in its de-
nominative form, both in a literal sense (sleeping room) and meta-
phorically of a burial chamber, already by the third century B.C.E.,
had an obvious appeal to those who imagined death as a transitional
sleeping period (the refrigerium interim) between life and the resur-
rection.
45
By the end of antiquity this metaphorical usage had been
extended and transferred, by a process of pars pro toto similar to
that which created the modern concept of the columbarium, to our
concept of cemetery, that is a collection of individual tombs or rest-
ing places; but cemeterium and otjtjtov continued to be regu-
larly used also in the literal sense in reference to sleeping places (indi-
vidual and collective, that is, dormitories) down into medieval and
Byzantine times.
46
The philological crux lies in determining at what point this later, ex-
tended usage first came into currency, but the more important histori-
cal question is what, precisely, our earliest literary sources mean when
they refer to coemeteria or rather what one source meant in using
the Greek term otjtjtov, since all the earliest Latin attestations
of coemeterium, both pagan and Christian, clearly refer to individual
tombs. In a poorly transmitted passage of an anonymous pamphlet
written against the election of Callistus as bishop at Rome in 217 and
attributed to Hippolytus, the author claims that Zephyrinus, bishop at
Rome in 198, in the following year recalled Callistus from exile and as-
44
Tert. Anima 5.17 (of 197 C.E.) appears to be the earliest attestation of the term in
a Christian context, but note also CIL 8.7543; 11.1700 = ILCV 2171; ILCV
3681. It first appears in Greek in [Hippolytus], Philosoph. 9.12.14 (below n. 48).
45
E.g., Homer, Il. 11.241; sleeping room: IG 7.235.43;burial chamber: IG
3.3545. For the Christian sleep of death, see Marbury B. Ogle, The Sleep of
Death, MAAR 11 (1933): 81117, esp. 8586 and Alfred C. Rush, Death and
Burial in Christian Antiquity (Washington, D.C.: Catholic University of America
Press, 1941), 1222. Both rightly trace the Christian (and Jewish) usage to the
long classical tradition of figuring death as sleep in funerary epigraphy and art.
46
Rebillard, Religion et sepulture, 14 n. 11 cites Philippe Gauthier, Revue des
tudes Byzantines 43 (1985): 5165, for a typicon of Theotokos Kecharitomene
(twelfth century) in which otjtjtov is applied both to a dormitory and to a
cemetery.
203
signed himri t otjtjtov.
47
In attempting to deduce the meaning
of this appointment, de Rossi, through a series of possible but by no
means inevitable suppositions, arrived at the conclusion that Zephyri-
nus had placed Callistus as deacon in charge of a Christian funerary
society and had given him management over the first official Christian
cemetery in Rome.
48
Subsequent scholarship has long since discarded
important elements of de Rossis formulation there were no official
Christian funerary societies or indeed specifically funerary collegia
of any sort, and the notion of a central Church at this date, let alone
of an official cemetery owned by a church, is doubtful but only
recently has a careful study of the Greek and Latin terms by ric
Rebillard shown there to be no firm evidence for the use of either to
mean cemetery in a general sense before the beginning of the sixth
century; in earlier Christian texts the words seem always to refer to
martyrs tombs or to the shrines surrounding them (martyria).
49
That
47
[Hippolytus], Philosoph. 9.12.14, Mr0 ou oijotv Zrutvo <u>
o(u)voorv(ov) o0tv <0>ruv <rrtv> tv otootootv to0 jou,
rtijor<v r> t u ioi u o u, o toutou <otv> rtoyoyv 0 to0
`Av0riou ri t otjtjtov otrotjorv. After the death of [Victor],
Zephyrinus, wishing to have him (Callistus) as a colleague in the institution of
the clergy, honored him to his own detriment and, transferring him for his sake
from Antium (?), appointed him to the koimeterion.
48
Giovanni Battista de Rossi, Esame archeologico e critico della storia di s. Cal-
listo narrata nel libro nono dei Filosofumene, Bulletino di Archeologia Cris-
tiana 4 (1866): 114, 1733; and La Roma sotterranea cristiana (Rome:
Cromo-litografia Pontificia, 1864), 1:197204. The syllogistic argument is car-
ried mainly by assertions of the must have variety, e.g., Egli impossibile,
che la chiesa romana tanto numerosa e potente non abbia avuto a quei d alcun
grande cimitero commune (197).
49
Rebillard, KOIMHTHPION et COEMETERIUM, summarized in Religion et
sepulture, 1123. Hippolytus himself uses otjtjtov to mean tomb in his com-
mentary on Daniel 11, 3646 (4.51). For the long-standing legal restriction on cor-
porate ownership of property, especially as regards Christian cemeteries, see, e.g.,
Peter Lampe, From Paul to Valentinus: Christians at Rome in the First Two Cen-
turies (ed. Marshall D. Johnson; trans. Michael Steinhauser; Minneapolis, Minn.:
Fortress Press, 2003), 36972, assembling also the supposed evidence none of it
unequivocal for a change in legal situation around the time of Fabians pontifi-
cate (ca. 236250 C.E.). For the fiction of a specific class of funerary collegia,
see Frank Ausbttel, Untersuchungen zu den Verein im Westen des rmischen Rei-
ches (Kallmnz: M. Lassleb, 1982); cf. John S. Kloppenborg, Collegia and Thia-
soi: Issues in Function, Taxonomy, and Membership, in Voluntary Associations in
the Graeco-Roman World (eds. John S. Kloppenborg and Stephen G. Wilson; Lon-
From Columbaria to Catacombs
204 John Bodel
Callistus himself, according to the Liber Pontificalis, was ultimately
buried in the catacombs of Calepodius on the Via Aurelia proves noth-
ing but lends little support to de Rossis view that he was appointed by
the Church as official overseer of the first or most important Christian
cemetery of Rome.
50
The position held by Callistus was in any case not
a regular post but an ad hoc assignment, probably coincident with the
end of the practice of burying high-ranking clergy near the tomb of
St. Peter on the Vatican, specifically, perhaps, in order to oversee the
arrangement of a new collective tomb for Romes bishops on Zephyri-
nuss private plot.
51
The custom of appointing overseers to manage pri-
vate tomb properties was common in Roman society, and there is no
reason to suspect that Callistuss role departed in any fundamental
way from that tradition.
52
The same charge of overreading can be levelled against de Rossis
interpretation of the burial enclosures used by Christians in North
Africa. Tertullians accusation of local hostility toward Christians and
the areae of their burials (areae sepulturarum nostrarum) at Carthage,
like several similar references to Christian burial enclosures elsewhere
in his writings, says nothing about official ownership or adminis-
tration of these cemeteries by the church.
53
Individual Christians could
don and New York: Routledge, 1996), 2023; and Rebillard, Religion et sepulture,
5171.
50
Lib. Pontif. I:62, cymiterio Calepodi, via Aurelia. Extraordinary efforts to explain
the burial of Callistus elsewhere than in the catacombs that bear his name
recently, for example, by invoking the tumultuous circumstances of his death
in Trastevere (Acta Sanctorum octobris, V [Paris 1868], 430): see Roberto Gior-
diani, Et sepultus est iuxta corpus beati Petri in Vaticano: qualche consider-
azione sul problema delle sepolture dei papi nellantichit, Vetera Christiano-
rum 40 (2003): 3045 betray discomfort with the inconvenience of the
circumstantial evidence for the conventional view.
51
Rebillard, KOIMHTHRION et COEMETERIUM, 98891. For the likeli-
hood that Zephyrinus personally owned the property, see Lampe, From Paul to
Valentinus, 2627.
52
For custodes (often freedmen of the proprietor) assigned to private tombs, see,
e.g., Petr. 71.8; CIL 6.9832; EDR 5184; cf. Liber Pontificalis 51.8 (314 C.E.), cus-
tos martyrum. More often in inscriptions the revenue-producing properties
attached to tomb monuments custodiae causa are explicitly recorded, whereas
the managers of the properties themselves go unmentioned: see Diz. Epigr.
2:1426 s.v. Custodia (Ettore De Ruggiero).
53
Tert. Ad Scapulam 3.1, addressed to the proconsul of Africa in 212 but invoking
an episode of a decade earlier: Tamen, sicut supra diximus, doleamus necesse est,
quod nulla civitas impune latura sit sanguinis nostri effusionem; sicut et sub Hil-
205
and no doubt often did congregate together in death (especially within
Christian families) and may well have been inclined toward a form of
euergetism that favored burial of the poor over more traditional dis-
tributions of public largesse, but the areae to which Tertullian refers fit
comfortably into a long pagan tradition of private donations by indi-
viduals of cemetery plots for public use (sometimes with restrictions
imposed on those who could be buried in them) and have no precedent
or likelihood of precedent in official sponsorship by a corporate entity
such as a church.
54
ariano praeside, cum de areis sepulturarum nostrarum acclamassent: Areae non
sint! Areae ipsorum non fuerunt: messes enim suas non egerunt. Rebillard, Re-
ligion et sepulture, 1722, notes the double entendre in Tertullians phrase (areae
= threshing floors and burial enclosures) and rightly dismisses the notion of
a technical, specifically Christian, application of the term in the present context.
For areae as burial enclosures in non-Christian texts, see below, n. 55.
54
Christian concern for the nourishment and burial of the poor (egeni) is touted by
Tertullian elsewhere (Apol. 39.56); cf. also Aristides of Athens, Apol. 15.6;
Lactantius Inst. 6.12; and [Hippolytus], Trad. Ap. 40, a work of composite and
probably later date falsely attributed to Hippolytus: see Alistair Stewart-Sykes,
On the Apostolic Tradition (Crestwood, N.Y.: St. Vladimirs Seminary Press,
2002), 2022. For private donations of land for Christian catacombs at Rome,
see Vincenzo Fiocchi Nicolai, Euergetismo eccleisastico e laico nelle iscrizioni
paleocristiane del Lazio, in Historiam pictura refert: miscellanea in onore di
Padre Alejandro Recio Veganzones (Vatican City: Pontifico Istituto di Archeolo-
gia Cristiana, 1994), 24345 (on ILCV 3681A) and Carolyn Osiek, below in this
volume. For traditional Roman precedent, note the late Republican donation of
a private cemetery at Sarsina by a certain Horatius Balbus for individual burials
of his fellow townsmen and residents, except those who pledge themselves as
gladiators, or commit suicide by hanging, or practice a dirty profession (muni-
cipibus [su]eis incoleisque extra au[ct]orateis et quei sibei [la]queo manu(m)
attulissent et quei quaestum spurcum professi essent) (CIL 1
2
.2123 = 11.6528 =
ILS 7846 = ILLRP 662) with Giancarlo Susini, Fundus Fangonianus, Studi
Romagnoli 20 (1969): 33339; id. Sarsina. Studi di anitichit (S. Giovanni in Per-
siceto: F.A.R.A.P, 1982), 26369; a similar donation at Tolentinum in the first
century C.E. to the townspeople of Tolentinum (municipibu[s] Tolentinati-
bu[s]) (CIL 9.5570 = ILS 7847) with Gianfranco Paci, Tolentinum, in Supple-
menta Italica, nuova serie 11 (Rome: Edizioni di storia e letteratura, 1993),
6567; and the early imperial funeral subsidy of Marius Lupercianus at Bergo-
mum (CIL 5.5128 = ILS 6726), with John Bodel, Graveyards and Groves: A
Study of the Lex Lucerina (American Journal of Ancient History 11) (Cambridge
Mass.: E. Badian, 1994), 1819 and 34 n. 137. For the long-standing legal pro-
hibition against corporate ownership of property, especially in reference to
burial sites, see above, n. 49.
From Columbaria to Catacombs
206 John Bodel
The terms otjtjtov and coemeterium, which say nothing about
any particular form of burial arrangement, raise a semantic problem of
a different sort from the challenges posed by the ambiguities inherent
in columbaria and catacombs. Areae, surface enclosures for
burial, present no particular interpretive difficulties of any sort, since
neither the name nor the significance of the term is in dispute, and
both are found already in earlier non-Christian sources.
55
What re-
mains in doubt is merely the question of the supervision of individual
Christian areae, which can only be determined, if at all, in individual
cases. The key to understanding the significance of the major inno-
vations in collective burial arrangements during the first three cen-
turies of the common era the rise of the columbarium during the first
century and then, accompanying a widespread change in method of
disposal during the second, the proliferation of catacombs in the
third lies in interpreting the form and function of these two osten-
sibly similar and yet fundamentally different modes of burial. It is to
that question that we now turn.
Form and Function
Two related tendencies have characterized, and impeded, recent study
of the historical significance of columbaria and catacombs. The first is
a failure to distinguish clearly between formal and functional criteria
in classifying and analyzing the archaeological evidence; the second is
an assumption that, because the two modes of collective burial share
superficial similarities of form, their functions must also be similar.
Each may be addressed briefly.
We noted above that the term columbarium is nowadays used as a
purely formal designation to describe any monumental tomb charac-
terized by rows of cremation niches lining the walls. Recently, however,
the usefulness of such a broad application of the term has begun to be
questioned: as currently employed, the word is appropriately applied
to most of the tombs of first-century date found in the outskirts of
Rome, which housed the remains of nuclear families and their immedi-
ate households as well as larger, more or less differentiated groups. The
55
See, e.g., ILS 7296, 7899, 8217, 8325, 8326, 8334, 8339, 8347 with Diz. Epigr.
1:654 s.v. Area pura 5, Area di sepolcri (Ettore De Ruggiero).
207
authors of the most recent detailed study of one of the large columba-
ria of Rome, that of the familia of the Statilii near the Porta Maggiore,
advocate a return to the more restricted usage of the eighteenth cen-
tury, when it applied only to the largest such monuments, those of the
households of the most prominent senatorial families of Rome, and
suggest that the identification of a niche tomb as a columbarium not be
based merely on size.
56
Although they do not specify what other cri-
teria might be relevant to distinguish columbaria from other monuments
designed for cremation burials, a helpful distinction emerges from
their discussion between those intended to house the remains of family
members and household staff together (traditional familial tombs),
which tend to be smaller, and those exclusively devoted to the burial
of members of a slave household or of a mixed group not defined by
ties of kinship. According to these criteria only nine monuments from
the city of Rome and none from elsewhere qualify as columbaria in
the proper sense, the smallest of which accommodated more than
200 burials.
57
According to this typology, the total number of niches is
a characteristic but incidental feature of the classification. Whether or
not such a restrictive definition is of use to archaeologists interested
in categorizing different types of tomb monuments, it is helpful for our
56
Caldelli and Ricci, Monumentum familiae Statiliorum, 60.
57
Of more than 40 columbaria excavated between 1700 and 1920 and published
in CIL, the only ones that fit the description are the monumentum familiae Li-
viae, which housed more than 1100 burials over little more than seventy years
(ca. 1080 C.E.?); the monument of the Statilii, with three rooms: one compris-
ing some 700 loculi, of which only 381 were used, from the Augustan or early
Tiberian era until 53 C.E., the other two first opened in 66 C.E. and in use until
end of the first century; two columbaria discovered beneath the Villa Doria Pam-
philj, one, in 1838, with some 500 niches for nearly 1,000 burials (ca. 10 C.E.),
the other in 1984 (the monument of C. Scribonius C.l. Menophilus), with more
than 500 burials of the Julio-Claudian era; three from the Vigna Codini between
the Via Appia and Via Latina: one, possibly of late Tiberian date, containing
500 niches, another (late Augustan?) with some 150 niches, each comprising two
cavities for urns for some 300 burials, and the third of unknown capacity but
yielding some 180 inscriptions (of Tiberian date but in use, perhaps, until the
second century); the monumentum familiae Volusiorum Saturninorum, with some
200 burials (and over 190 inscriptions) (ca. 2097 C.E., most ca. 50 C.E.); and a
monument of Augustan date unearthed outside the porta Praenestina in the
same region as the monument of the familia of the Statilii, with 118 niches for
double burials: cf. Caldelli and Ricci, Monumentum familiae Statiliorum, 6064,
with further references.
From Columbaria to Catacombs
208 John Bodel
purposes in shifting the focus from an architectural form and an arbi-
trary number to a social use, which usefully highlights the area in
which the new funerary facilities (an outgrowth, precisely, of the fam-
ilial monuments favored by the more prominent families of the late
Republic) marked a change in the organization of burial at Rome: for
the first time, it seems, collective tombs housed together, on the one
hand, slaves and freedmen of particular households apart from their
freeborn owners and patrons and, on the other, miscellaneous groups
of persons (sometimes in familial groupings, but not exclusively so) not
related to each other by blood or ownership.
The separation of form and function is not so clear in current dis-
cussions of subterranean cemeteries and graves. As noted above, the
primary criteria used to identify proper catacombs are purely for-
mal: independent entrances from ground level, intensive use of avail-
able space for inhumation burials, and open design intended to
allow expansion through the extension of existing galleries or the cre-
ation of new ones, normally according to a regular plan. In an effort
further to distinguish purpose-built communal catacombs from pri-
vate underground burial complexes, Hugo Brandenburg draws a func-
tional distinction between smaller hypogaea comprising up to a dozen
or so clearly mixed, pagan and Christian burials and larger catacombs
accommodating hundreds and thousands, in which religious distinc-
tions are less apparent; but he then compromises the categorization by
extrapolating from it a formal rule that correlates size with religious
exclusivity.
58
As it happens, and as he notes, intermediate numbers are
rare, but they do exist, and where they do they tend to be problematic,
since it is not always easy to tell when smaller hypogea were expanded
and made more uniform haphazardly or when they were extended by
original design, nor indeed can we presume to know in most cases what
considerations may have motivated the development of private hypo-
58
Hugo Brandenburg, berlegungen zu Ursprung und Entstehung der Kata-
komben Roms, in Vivarium. Festschrift Theodor Klauser zum 90. Geburtstag
(eds. Ernst Dassmann and Klaus Thraede; Mnster: Aschendorff, 1984), 39 and
4445. The idea that pagans disdained the use of larger catacombs because they
provided little opportunity for those of greater wealth and status to display it
(45) is disproved by the abundant evidence of funerary self-representation in
cubicula already at an early date: see Fiocchi Nicolai, Origin and Develop-
ment, 2223 and above, n. 21.
209
gea into larger funerary complexes.
59
Recent analysis of the architec-
tural and decorative program of the controversial Via Latina cemetery,
for example, a supposedly private hypogeum complex of the third
quarter of the fourth century that provided burial for up to 400 per-
sons (and thus approaches the size of proper catacombs), where
indisputably pagan figured scenes are intermingled with Chris-
tian iconography throughout a series of luxurious cubicula joined
by galleries with loculi, shows that all the spaces were designed, con-
structed, and (variously) decorated according to a single homogeneous
plan.
60
Philippe Pergola proposes an ostensibly more clear formal distinc-
tion between closed and open underground cemeteries, the former
being those intended to house a predetermined and fixed number of
burials, the latter those capable of expansion through existing galleries
and along the principal axes, but then complicates it by drawing a sec-
ondary distinction within the open type between those centered on
and systematized according to a monumental presentation of the
family of the owner and those which, though not entirely unreflective
of distinctions of status, nonetheless present a more uniform and
homogeneous aspect.
61
He would have done better to stop with the
categories open and closed a useful polarity for typological
analysis or to begin with familial and communal (or perhaps
better collective), concepts useful for evaluating purpose and use,
than to have mixed together the two types of criteria, so that form and
function become confused. It is clear that in Roman funerary architec-
ture the two categories are indeed related, but we will not be able to
59
The dating of catacombs is fraught with difficulties and often hangs precariously
on the stylistic dating of frescoes: see, e.g., J. G. Deckers, Wie genau ist eine Ka-
takombe zu datieren? Das Bespiel SS. Marcellino e Pietro, in Memoriam Sanc-
torum Venerantes. Miscellanea in onore di Monsignor Victor Saxer (Studi di
Antichit Cristiana 48; Vatican City: Pontificio Istituto di Archeologia Cris-
tiana, 1992), 21738, and Norbert Zimmermann, Werkstattgruppen rmischer
Katakombenmalerei (Jahrbuch fr Antike und Christentum Ergnzungsband 35;
Mnster Westfalen: Aschendorff Verlag, 2002), who advocates a more scientific
approach through analysis by workshops and iconography together.
60
See Zimmermann, Werkstattgruppen, 61125; cf. Johnson, Pagan-Christian
Burial Practices, 5658.
61
Pergola, Le catacombe romane, 6062. The division into classes (60) inevitably
allows for a variety of indeterminate intermediate types.
From Columbaria to Catacombs
210 John Bodel
recognize either for what it is if we fail to distinguish each clearly from
the other in our initial analysis.
62
Apart from this categorical confusion of design and purpose, noth-
ing has clouded the picture more thoroughly than a repeated insistence
on the obvious but superficial points of formal similarity between col-
umbaria and catacombs and a failure to recognize their less striking
but more fundamental formal differences. To observe that catacombs,
like columbaria, feature multiple rows of niches and were designed to
accommodate numerous burials within a minimum amount of space
says no more than that both offered an economical response to the
(demographically predictable and socially inevitable) ever-increasing
demand for burial space in the neighborhood of Rome. But to claim
that [columbaria] were not the only solution: the catacomb works in
the same way is, I think, to misrepresent their essential differences in
orientation.
63
The question of how these two modes of collective burial
channeled that demand must now be addressed.
Columbaria: Members Only
With columbaria, the architectural form, though capable of housing
large numbers of burials, remains closed (if we may borrow Pergolas
formulation for the classification of catacombs). In this respect the
largest of the collective monuments is essentially no different from the
smallest familial tombs restricting entry to named family members
and, in the formulaic phrase, freedmen and freedwomen and their de-
scendants. The possibilities for membership within the community
are finite, even if, in principle, they extend (with the inclusion of all
possible descendants) indefinitely into the future. The largest of the
columbaria, ironically those devoted to the slave and freed staff of the
great houses of Rome provide the clearest indications of the limi-
tations of the form, and the very largest of them, that of the household
of Livia, offers the most unequivocal evidence of all (see Fig. 6.1,
62
In fact, one can point to individual examples of the closed type of hypogeum
which present a largely uniform appearance and do not privilege any one burial
space or chamber which is simply to say that Roman hypogea are equally sus-
ceptible to formal and functional analysis: see, e.g., Feraudi-Grunais, Inschriften
und Selbstdarstellung, for examples from the Vatican necropolis: 2930 (mau-
soleum F), 3334 (Tomb of the Octavii), 3941 (Tomb XXX).
63
Purcell, Tomb and Suburb, 39.
211
p. 197).
64
Among some 140 surviving epitaphs from the first room of
the monument, which was in use from the end of the Augustan era
down to the time of Claudius, more than 50 were inscribed more than
once and another 40 record two independent names in the nomi-
native.
65
The implication is that individual ollae were used for more
than one burial and, despite a no doubt already-existing legal prohib-
ition against defacing or erasing epitaphs, were being reused and this
within the space of a few decades. Evidently the nearly 1,100 spaces set
aside for Livias staff and their dependents were insufficient to house
the remains of those eligible for burial in the monument, and the desire
to be included among that group apparently outweighed fear of the
consequences of violating the burial of another.
66
Nothing could illus-
trate more clearly the comparative pull on funerary behavior of the
competing social forces of solidarity with fellow members of a corpor-
ate group (the staff of Livia) and social ambition for individual repre-
sentation within a privileged community.
At the same time, the monument, which was evidently administered
by a collegium of Livias slaves and freedmen, housed burials also of
the servants of Livias husband, son, daughter-in-law Antonia, and
grandchildren, as well as persons with no obvious connection to the
imperial house.
67
Furthermore, other of Livias slaves and freedmen
were buried in columbaria primarily devoted to different households
(such as those of the children of Drusus or the younger Marcella) or in
64
For the monument, its discovery, and interpretation, see Helke Kammerer Grot-
haus, Camere sepolcrali de liberti e liberte di Livia Augusta ed altri Caesari,
MEFRA 91 (1979): 31542; Fehl, Archaeologists at Work.
65
Double burials in a single olla: e.g., CIL 6.3945, 3946, 3992, 8944.
66
See Caldelli and Ricci, Monumentum familiae Statiliorum, 6061, aptly citing
Paulus, Sententiae 5.1.21.8, qui monumento inscriptos titulos eraserit sepul-
chrum violasse videtur.
67
See, e.g., CIL 6.3951 (a slave of Tiberius); 6.3959 (a freedman of Augustus);
6.3971 (a slave of Nero Caesar); 4.4049 (P. Caetennius Heraclis); 6.4051 (Corne-
lius Chius); 6.4057 (Fuscus, slave of a freedwoman of Antonia, the mother of
Claudius); etc. Jukka Korpela, Die Grabinschriften des Kolumbariiums Liber-
torum Liviae Augustae: eine Quellenkritische Untersuchung, Arctos 15 (1981):
5366 analyzes onomastic aspects of the 670 names recorded on the 376 surviv-
ing inscriptions from the monument: 137 are certain slaves, 184 freedmen (55);
two thirds are men; of the 114 identifiable persons who erected epitaphs, only 30
are women (57). No system seems to have controlled the elements of nomencla-
ture used, and the names of those who erected the monuments were evidently
less important than those of the ones commemorated. For the collegium itself,
see CIL 6.4305 and below, n. 69.
From Columbaria to Catacombs
212 John Bodel
their own individual monuments.
68
Though closed, membership in the
group of those admitted to Livias monument was not strictly defined
according to membership in the group for which it was intended. Per-
sonal associations, marital relationships in some cases but clearly not
in all, evidently determined who sought access to this limited resource;
deciding who was to be included and where apparently fell to the ad-
ministrators of the collegium, probably with the necessary approval of
the aristocratic patron.
69
Normally the problem of overcrowding did not arise. Three of the
largest columbaria in Rome were evidently never used to capacity, but
even in more modest familial tombs, the invitations to freedmen and
freedwomen and their descendants to be buried within the family
monument may never have been taken up by more than a few.
70
Many
ex-slaves and, a fortiori, their descendants preferred to advertise their
names on their own monuments. Indeed, if we accept the implications
of Lily Ross Taylors classic discussion of the number of urban resi-
dents of the first and second centuries carrying some mixture of servile
blood in their veins, most people in Rome would have had their burial
assured several times over by the pervasive opportunities held out by
these largely underactivated formulae.
71
Accordingly, many of those
advertising such magnanimity on their tombstones must have counted
on the limits of their generosity not being put to the test.
72
The promise
68
For slaves buried elsewhere, cf., e.g., CIL 6.4448 (monumentum Marcellae)
6.6213; 6. 8727; 6.8903, with Susan Treggiari, Jobs in the Household of Livia,
PBSR 43 n.s. 30 (1975): 4849 and the lists on 7277. With the familia of the
Statilii, at least 39 slaves and freedmen were buried outside the household col-
umbarium: Caldelli and Ricci, Monumentum familiae Statiliorum, 13540.
69
See Kammerer Grothaus, Camere sepolcrali, 326, on the collegium libertorum
et servorum Liviae and the collegium magnum tribunorum divae Liviae; in general,
Hasegawa Collegia domestica, 25256, for the administrative organization of
the collegia domestica, and 26165 for the role of patrons in granting permission
for burial within the columbarium; see also below, n. 98.
70
Caldelli and Ricci, Monumentum familiae Statiliorum, 57.
71
Lily Ross Taylor, Freedmen and Freeborn in the Epitaphs of Imperial Rome,
AJP 82 (1961): 11332, esp. 128.
72
A striking illustration of this confidence is found in an epitaph of perhaps Fla-
vian date from the section of the Via Triumphalis necropolis under the Vatican
parking lot dedicated by a husband to himself, his wife, and their descendants
and inscribed on a single stele marking (it seems) an individual burial! See
Veikko Vnnen et al., Le iscrizioni della necropoli dellautoparco Vaticano
(Rome: Institutum Romanum Finlandiae, 1973), 4041 n. 27, with Steinby, La
necropolis della Via Triumphalis, 57.
213
to provide burial could never be considered completely idle, however,
when the judgment of the pontiffs, who controlled Roman tomb law,
could be expected generally to follow the written wishes of the testator,
as indicated, even, by an inscribed epitaphic formula. So we find, at
the beginning of the third century, the jurist Papinian expressing
the opinion that freedmen can neither be buried nor bury others [in
a familial monument] unless they are heirs to their patron, although
some people have inscribed on their tomb that they have built it for
themselves and their freedmen.
73
For the patron, it was the idea of
representing oneself publicly as a beneficent dominus or domina that
made the gesture worth making, despite the potential cost. In the same
fashion, for slaves and ex-slaves of the empress Livia (who were in a
better position than most to control independent financial resources in
a peculium), obtaining even ephemeral recognition in a prestigious
burial location was evidently preferable to securing a more permanent
memorial elsewhere. What mattered was to be on the inside, no matter
how futile the hope for lasting commemoration.
This emphasis on the right to be included finds its counterpart in ex-
clusionary expressions prohibiting one or more persons by name from
burial in a familial tomb or monument.
74
More stridently, if vainly,
since such privately (and posthumously) imposed penalties had no
legal force, impressively large fines were threatened against any who in-
troduced alien burials into the tomb.
75
With columbaria and the later
73
Dig. 11.7.6.pr. (Ulpian 25 ad ed.), liberti autem nec sepeliri nec alios inferre
potuerunt, nisi heredes extiterunt patrono, quamvis quidam inscripserint monu-
mentum sibi libertisque suis fecisse. After citing Papinian, Ulpian goes on to say
that there has very often been a ruling to this effect, et ita Papinianus respondit
et saepissime idem constitutum est.
74
E.g., ILS 7602, 7660, 82838286, and many examples of more specific formulae
of inclusion and exclusion, often used in combination, in ILS 82598282. See
now also Silvia Orlandi, Heredes, alieni, ingrati, ceteri. Ammisione ed esclu-
sioni, in Libitina e dintorni (Libitina 3; ed. Silvio Panciera; Rome: Edizioni
Quasar, 2004), 35984. For the legal aspects, see also de Visscher, Droit des tom-
beaux, 1036.
75
On the increasingly extravagant (if idle) threats to exact monetary fines, see Gian
Luca Gregori, Si quis contra legem sepulcri fecerit: violazioni e pene pecu-
niare, in Libitina e dintorni (Libitina 3; ed. Silvio Panciera; Rome: Edizioni
Quasar, 2004), 391404, esp. 4023: fines ranged from HS 1,000 to HS 350,000
during the second and early third centuries, with HS 50,000 evidently represent-
ing a standard amount; beginning in the third century amounts of HS 100,000
and higher became common; cf. de Visscher, Droit des tombeaux, 11223.
From Columbaria to Catacombs
214 John Bodel
familial monuments of this sort, what mattered was to be in or out. De-
fining the space physically, with a perimeter barrier (maceria) en-
closing the plot and an imposing monumental structure designed to
impress those viewing it from the outside, and verbally, with a titulus
declaring the size of the plot (pedes tot in fronte, tot in agro) and speci-
fying persons and categories of persons eligible to be included within
the monument, served to segregate the members of that circumscribed
community from the rest of society.
76
So deeply did this impulse pen-
etrate into the mentality of those who chose this form of burial that we
sometimes find miniature family tombs erected within columbaria and
epitaphs marking individual ollae that include the standard formulae
promising burial also libertis libertabusque posterisque eorum.
77
Once inside, with the exception of a distinctive emphasis sometimes,
but not always, or even normally, focused on a central burial spot, little
attention is devoted to distinguishing individuals: small uniform head-
stones line the rear and side walls of familial tombs at Pompeii, and
columbaria, as noted earlier (above, p. 196) generally exhibit uniform
decoration across rows of niches if not entire walls.
78
In the colum-
barium of C. Scribonius Menophilus in the Villa Doria Pamphilj, the
olla identified by Menophiluss epitaph is unobtrusively located in the
second row from the bottom next to a door into a secondary room. In
the monument of Livias household, the draftsmanship and carving of
some of the small placards found in front of individual ollae are so
poor that it is difficult to imagine the broken and reused stones as even
temporary markers of actual graves rather than mere place holders,
76
For the practice of staking out the dimensions of tomb plots with declarative
inscriptions, see Giovannella Cresci Marrone and Margherita Tirelli, eds., Ter-
minavit sepulcrum. I recinti funerari nelle necropoli di Altino (Rome: Edizioni
Quasar, 2005), especially the article by Gian Luca Gregori, Definizione e mis-
urazione dello spazio funerario nellepigrafia repubblicana e protoimperiale di
Roma. Unindagine campione, 77126.
77
For miniature family tombs, see Hanne Sigismund Nielsen, The Physical Con-
text of Roman Epitaphs and the Structure of the Roman Family, ARID 23
(1996): 41, 56, 58 n. 27; cf. also above, n. 72. For columbaria inscriptions dedi-
cated libertis libertabusque posterisque eorum, see, e.g., CIL 6.5678, 7729, 8100,
and especially 6823, for a plot measuring 2 3.5 Roman feet.
78
For the rarity of central focus in a sample of familial monuments of the first and
second centuries, see Feraudi-Grunais, Inschriften und Selbstdarstelung,
4142. For the Pompeian stelae (columellae), see Hope, A Roof over the Dead,
8284.
215
but several of the ollae they identified contained ashes.
79
Nor, evi-
dently, was much care taken in columbaria to preserve groupings of
close kin, perhaps of necessity. A detailed study of familial relation-
ships represented in the largest of the Vigna Codini catacombs ex-
cavated by Pietro Campana in 1840, with some 198 epitaphs in situ
and another hundred found loose inside, failed to reveal any strong
evidence of nuclear families buried together but did find numerous in-
stances of families being split up among individual ollae located in dif-
ferent parts of the monument.
80
Indeed, the grouping of ollae in pairs
within a single niche, the configuration which seems to have suggested
to the Romans the designation columbarium, although it would have
served well for couples, is singularly ill-suited to the unified comme-
moration of nuclear families. Inscriptions sometimes mention multiple
ollae acquired by a single person in different rows of a columbarium
allotted during different rounds of a lottery or selection process, or the
resale of ollae acquired on speculation by entrepreneurs.
81
Consider-
79
See Caldelli and Ricci, Monumentum familiae Statiliorum, 6061. Conversely, in
some columbaria not all unmarked ollae were unused. In the first of the two col-
umbaria discovered on the property of the Villa Doria Pamphilj, for example, all
of the ollae contained ashes, but not all were marked with inscribed or (appar-
ently) painted inscriptions. In the nearby columbarium of Scribonius Meno-
philus, where both inscribed and painted inscriptions are found (the latter
normally in the painted tabulae ansatae provided beneath each niche in the uni-
form decorative scheme) and both occasionally bear the same name, it seems
that the painted tituli may have identified the owners of the niches, whereas the
inscribed placards recorded the epitaphs of those buried in them.
80
Nielsen, Physical Context, 4344 noting that both of the (only) two certain
close-kin groupings found in the columbarium also reflected ties to patrons out-
side the family: cf. CIL 6.4923, 5035, and 5074 (family of M. Valerius Futianus)
and 6.50465047 (Veturia Helena). This columbarium was evidently built by an
entrepreneur who sold spaces in it to any who wanted. For other columbaria
found in the Vigna Codini constructed for professional collegia or slave house-
holds, see Lucia Parri, Iscrizioni funerarie, colombari, e liberti: il terzo ipogeo
di Vigna Codini ed alcuni dei suoi epitaffi, Atene e Roma 43 (1998): 5460,
esp. 5557.
81
Cf. e.g., ILS 7892 (five individual ollae acquired in five consecutive rounds of
a lottery); cf. 7893. For ollae chosen extra sortem or (rarely) contiguous units,
cf. ILS 7889 sine sorte ab socis quas vellet ollae sexs datae sunt; 7900a, ollas
habet continentes VI; further Maria Laetizia Caldelli, Simona Crea, and Claudia
Ricci, Donare, emere, vendere, ius habere, possidere, concedere similia: don-
azione e compravendita, propriet, possesso, diritto sul sepolcro e diritti di se-
poltura, in Libitina e dintorni (Libitina 3; ed. Silvio Panciera; Rome: Edizioni
Quasar, 2004), 311 (Ricci), noting that some of the large columbaria evidently
From Columbaria to Catacombs
216 John Bodel
ation of the bipartite columbarium slabs from Rome those which
record two epitaphs side by side on a single marble tablet although it
reveals a sizeable number of associations that might reflect relation-
ships of intimacy in life, equally clearly preserves a number of pairings
that do not. The overwhelming impression is of fragmentation at an in-
itial distribution of burial places within the monument and frequent
redistribution of individual ollae or niches by subsequent gift or sale.
In this respect, Purcells analysis is fully on target in characterizing the
nature of such monuments as expressing neither individuality nor
membership of a mass society but incorporation in a group a few
hundred or a few thousand strong.
82
But if we wish to pursue the housing analogy, we should observe that
the most appropriate model for the columbarium is neither the apart-
ment block (insula), as Hopkins would have it, nor the domus, as Pur-
cell maintains, but rather the divided households with separate lararia
for kin and slave familiae, such as in the House of the Vettii at Pom-
peii.
83
For what is truly novel about the largest of the columbaria is
neither their sheer size nor even their essential presentation, which,
like the familial tombs, is exclusionary and extroverted, but rather the
segregation of the slave household from the kinship group a rupture
that shattered the fiction of the paterfamilias treating his household
slaves in loco filiorum (as if they were his children) and opened the
way for the broader bipartite division of society that would emerge
more formally a century later with the segregation of the privileged
did not allow this: already Brizio observed that none of the more than 400 in-
scriptions from the columbarium of the Statilii mentions sale, donation, or
acquisition. Note also Giuseppe Gatti, Singolari iscrizioni dellaedificium
XXXVI sociorum sulla Via Latina, BCAR 10 (1882): 328, esp. 38; Schrumpf,
Bestattung und Bestattungswesen, above n. 6.
82
Purcell, Tomb and Suburb, 39. The claim that the internal configuration of
columbaria expresses hierarchies within the group and that variations in the
size and dcor of individual niches and ollae indicate minute gradations of
status (38), on the other hand, is difficult to sustain (no examples are cited):
certain niches were indeed more lavishly decorated or more advantageously lo-
cated than others, but we have no idea what criteria determined who occupied
them. If, as several indications suggest, distribution by lottery was the norm,
hierarchies of status would have been difficult to support, even if desired.
83
For household lararia as markers of separate households of servile familiae
and freeborn proprietors, see John Bodel, Ciceros Minerva, forthcoming in
Household and Family Religion in Antiquity: Contextual and Comparative Per-
spectives (eds. John Bodel and Saul Olyan; Oxford: Blackwell, 2008).
217
elite from the more humble free population, or, in Roman terms, with
the discrimination of honestiores and humiliores.
84
At another level, the
most suitable analogy to the columbarium within the sphere of do-
mestic housing is the modern condominium complex, a cooperative
organization created and designed less to provide a minimum neces-
sary service than to cater to those with sufficient resources to expend
on amenity and willing to forego individual preference on a small
scale in favor of more lavish accommodation and guaranteed care
within a collectivity.
It remains to consider briefly the administration of these monu-
ments. The earliest of the large columbaria were those constructed for
the household staffs of the senatorial families of Rome. In creating sep-
arate tombs for a group that, by definition, had no legally recognized
kinship relations and thus no familial hierarchy to govern the distribu-
tion of burial spaces within a collective monument, the aristocratic
slave owners who provided (or at least allowed) these structures made
possible the creation of a new system of tomb management based
upon other principles than those that governed the administration of
familial monuments.
85
In theory slave-owner patrons might have con-
trolled admission into these structures and determined the internal
arrangement of burials within them, but in practice, it seems, they exer-
cised their prerogative only seldom and with discretion. Nor, it seems,
did the informal (but nonetheless real) families of slave partners (con-
tubernales) and their offspring exert their familial identity sufficiently
to maintain kinship groupings within the collective spaces. Instead, a
84
For the concept of honestiores and humiliores and its implications for the divi-
sion of Roman society, see Peter Garnsey, Social Status and Legal Privilege in the
Roman Empire (Oxford: Clarendon Press, 1970), 22133.
85
For slaves as legally kinless, see, e.g., Dig. 38.8.12 (Ulpian) and 38.10.10.5
(Paul) with William Warwick Buckland, The Roman Law of Slavery (Cam-
bridge: The University Press, 1908), 7679. In addition to the few surviving col-
umbaria devoted to the households of senatorial families (above, n. 57),
inscriptions dedicated to the freedmen and familia of various prominent per-
sons (two from Aquileia, the rest from Rome) attest another sixteen examples of
burial places or tombs reserved for particular households: cf. ILS 78487860,
7862; cf. also ILS 7861 (C.E. 136). We do not know what sort of monuments the
inscriptions originally adorned; some of the plots were fairly small: cf. e.g., ILS
7850 (15 16 Roman feet), 7857 (10.5 12 feet), 7862 (12 24 feet); but others
were clearly large enough to have accommodated large columbaria of the
sort that occasionally survive: e.g., ILS 7585 (13 45 Roman feet), 7854 (35 35
feet), 7860 (32 32 feet, at Aquileia).
From Columbaria to Catacombs
218 John Bodel
loose corporate structure seems generally to have furnished the organ-
izing principle of tomb administration. The flexibility of this model of
self-government lent itself readily to other groups who wished to con-
solidate resources in the interest of providing more lavish communal
facilities for the regular commemorative rituals that Roman funerary
custom required than individual members could finance independ-
ently, whether the groups were held together by a common interest,
such as a shared religion or occupation, or were defined merely by a
mutual desire to share in the services and amenities that collective
membership offered.
An inscription set up in 16 C.E. by two freedmen administrators of a
familial columbarium beside the Via Labicana outside Rome illustrates
well the sorts of communal amenity that membership in a columbarium
might provide, as well as the subtle blend of autonomous corporate
self-government and collective dependence upon a patron that the
management of such properties seems often to have entailed.
Titus Cocceius Gaa and Titus Cocceius Patiens, quaestors for the third time (of
the domestic funerary collegium), according to the will of the decurions (of the
collegium) set up the square dining table in the pavilion, the sideboard and base,
the sundial, the fountain basin with supports, the marble well, the stucco-work
above the wall of the middle path with the tiled roof, the little travertine column
beneath the sundial, the projecting roof in front of the portico, the scales and
weights. And, thanks to the kindness and generosity of their patron Titus, they
undertook the clearing of a place behind the further perimeter wall and the
transferring of the crematories from the furthest fence to there and the construc-
tion of a path there and a doorway. And the same men with public money dec-
orated those places which their patron Titus had granted to his decurions with
the seeds of vines and fruits and flowers and all sorts of greenery, in the consul-
ship of Sisenna Taurus and L. Scribonius Libo.
The inscription goes on in hexameters to urge readers to recognize in
the expense incurred the just observance of piety and, for peace of
mind, to follow the example of those who created and tended the fu-
nerary garden during their lifetimes, so as to be remembered and cared
for by others after their deaths.
86
The message it conveys, indirectly but
86
CIL 6.10237 = ILS 7870; for the poem, CLE 371: T. T. Coccei Gaa et Patiens
quaest(ores) III mensam quadratam in trichil(a), abacum cum basi, horologium,
labrum cum fulmentis, marmor putiale, crustas supra parietem itineris medi cum te-
gulis, columellam sub horologio Tiburtina(m?) 7 (sic) protectum ante porticum,
trutinam et pondera d(e) d(ecurionum) s(ententia) posuerunt; et locum post
maceriam ulteriorem emendum ustrinasque de consaepto ultimo in eum locum trai-
219
no less clearly, is one of exclusivity: the garden appurtenances cata-
logued in detail, the reference to worthy expenditure (impensae cau-
sam et iustam curam), and the appeal to successors to reciprocate
the commemorative care bestowed, are designed not only to encourage
imitation among future members of the society but to call attention to
the privileges from which others were excluded.
Catacombs: World without End?
Catacombs, by contrast, were open, ill-defined spaces, infinitely ex-
pandable (or at least creating the impression of being so) and offering
little or no external public aspect. Upon entering a columbarium, one
recognized the traditional boundaries of place in the Roman world
border stones, a circuit fence or wall marking the perimeter of the plot,
the four walls of an enclosed rectilinear space. To be sure, catacombs
too, or rather their central areas, were often marked out at the surface
by perimeter walls and boundary stones and later, after the time of
Constantine, by large funerary basilicas that provided monumental
focus on the tombs of martyrs and formed the locus of concentrated
burials both above and below ground.
87
At certain entrances some had
outposts for caretakers, who monitored and perhaps restricted visits to
the subterranean galleries. But most eventually could be entered from
any of several different stairwells, many of which seem to have allowed
free access, and the circumscribed surface burials sub divo gave little
ciendas et iter ad eum locum ianuamque beneficio et liberalitate T. patroni facienda
curaverunt; idemque vitium pomorumq(ue) et florum viridiumque omnium generum
seminibus ea loca quae T. p(atronus) decurionibus suis adtribuerat ex pecunia pub-
lica adornaverunt, Sisenna Tauro L. Scribonio Libone co(n)s(ulibus). See further
Bodel Roman Tomb Gardens, forthcoming in Gardens of the Roman Empire
(ed. Wilhelmina Jashemski; Cambridge: Cambridge University Press, 2008).
87
For an overview, see Umberto M. Fasola and Vincenzo Fiocchi Nicolai, Le
necropoli durante la formazione della citt cristiana, in Actes du XIe Congrs
International dArcheologie Chrtienne. Volume II. La Topographie Chrtienne
des Grandes Capitales (Rome: cole franaise de Rome, 1989), 117079. For ear-
lier surface cemeteries overyling the site of catacombs, see also Osiek, below,
24850. For funerary basilicas, which often took the shape of a racecourse, see
Vincenzo Fiocchi Nicolai, La nuova basilica circiforme della Via Ardeatina,
RendPontAcc 68 (19951996) [1999]: 69233, esp., on burials, 14575 (by Maria
Paola Del Moro), with further bibliography.
From Columbaria to Catacombs
220 John Bodel
hint of the extensive galleries and networked cubicula that opened
out below.
88
Once inside, moreover, the visitor to a Roman catacomb experienced
a sensation of having entered an alien, somewhat disorienting world, in
which each tunnel and gallery connected with other tunnels or gal-
leries, so that there was no sense of center and periphery and no clear
demarcation of finite limits. In some early catacombs, such as the first
floor of the catacombs of Priscilla, datable to the first decades of the
third century, the galleries meandered unpredictably along the more
easily excavated seams of the tufa beds, so that anyone attempting
to follow an orderly route would easily loose a sense of direction
(Fig. 6.4).
89
In others, such as the catacombs of Callistus in their ear-
liest phase (around 200 C.E.), the main galleries of a regular plan laid
out in correspondence with a clearly demarcated surface plot, origi-
nally dug only as far as the boundaries of the area, were designed
to allow subsequent extension underground beyond the confines of the
surface cemetery (Fig. 6.5).
90
Even where systematic planning pro-
duced a regular grid of networked tunnels, as in a second, lower level
of the catacombs of Priscilla dug out beneath the first in a character-
istic fishbone pattern several decades later, near the beginning of
the fourth century, one gets the impression of construction by module,
88
At Rome caretakers quarters have been identified securely only at the cata-
combs of Praetextatus: see Antonio Ferrua, Un vestibolo della catacomba di
Prestestato, RACrist 40 (1964): 14665. Other possible examples may have
existed at the so-called Villa piccola of S. Sebastiano and perhaps at the upper
level of the entrance to the hypogeum of the Flavii at the catacombs of Domitilla
(see Fasola and Fiocchi Nicolai, Le necropolis, 1179), but most stairwell en-
trances have left no evidence of being guarded.
89
See Francesco Tolotti, Il Cimitero di Priscilla: Studio di topografia e architettura
(Vatican City: Societ amici delle catacombe presso Pontificio Istituto di Arch-
eologia Cristiana, 1970), 63106, 17189.
90
For the large surface plot (100 250 Roman feet, ca. 30 75 meters,) circum-
scribed by a fence and developed already as a burial area during the first and
second centuries, see Spera, Paesaggio suburbano, 10923. For the main sub-
terranean galleries of the so-called Area I (Regio A), which were entered by
separate staircases at the corners of the surface plot, and the transverse ortho-
gonal tunnels planned and subsequenlty dug between them, see Paul Styger,
Lorigine del cimitero di S. Callisto sullAppia, RendPontAcc 4 (19251926):
11219; Brandenburg, Ursprung und Entstehung, 9192; and Donatella Nuzzo,
Tipologia sepolcrale delle catacombe romane: i cimiteri ipogei delle vie Ostiense,
Ardeatina, e Appia (British Archaeological Reports, International Series 905)
(Oxford: Archeopress, 2000), 9095.
221 From Columbaria to Catacombs
6
.
4
.
P
l
a
n

o
f

t
h
e

c
e
n
t
r
a
l

s
e
c
t
i
o
n

o
f

t
h
e

f
i
r
s
t

f
l
o
o
r

o
f

t
h
e

c
a
t
a
c
o
m
b
s

o
f

P
r
i
s
c
i
l
l
a

b
e
s
i
d
e

t
h
e

V
i
a

S
a
l
a
r
i
a

N
o
v
a
,

a
f
t
e
r

T
o
l
o
t
t
i
,

C
i
m
i
t
e
r
o

d
i
P
r
i
s
c
i
l
l
a
,

p
l
.

I
I
.
222 John Bodel
as if additional units could be added on at will as indeed they were,
according to plan (Fig. 6.6).
91
In each of these cases the lack of tradi-
tional topographical points of reference and a plan designed from the
start to accommodate expansion created a sense of openness and
inclusiveness, in the sense that membership in the collectivity of those
sharing the cemetery was never finite but always potentially available.
At the same time, the subterranean setting provided an ambience not
only appropriate to the world of dead, who according to a deep-seated
tradition of Greco-Roman culture were to be returned at death to
mother earth, but conducive to the sort of oblique expression of an
ideal social order divorced from the compromising realities of life that
funerary architecture in antiquity normally aimed to represent.
92
Re-
moved from the natural light and freed from the contours of the sur-
face topography, the interior space of the catacombs was a world of its
own, without normal parameters. Accordingly, the iconography of the
scenes from daily life frequently found on tombstones and in painted
91
For the date, see Tolotti, Cimitero di Priscilla, 32240.
92
For the concept of mother earth as the proper recipient of the dead in Roman
culture, cf., e.g., Cicero, Leg. 2.56 with Xenophon, Cyr. 8.7.25; CIL 12.1932 =
CLE 1476; CIL 6.15493 = CLE 1129; cf. Livy 1.56.1012.
6.5. Plan of the so-called Area I
(Regio A) of the catacombs of Callistus
beside the Via Appia as developed
by the middle of the third century,
after Styger, Cimitero di S. Callisto,
103 fig. 7.
223 From Columbaria to Catacombs
6.6. Plan of the lower level of the catacombs of Priscilla, after
Fiocchi Nicolai, Origin and Development, 25 figure 20.
224 John Bodel
frescoes decorating the walls, although it superficially resembles the
naturalistic depictions of occupations and leisure activities familiar
from traditional Roman funerary art, seems always to have been to a
certain extent symbolic and over time grew more detached from reality
and increasingly ideological and abstract.
93
Within the catacombs different regions of the underground cem-
eteries were characterized by different configurations of space: in ad-
dition to the networked galleries uniformly lined with loculi, certain
areas were topographically distinguished by individual rooms and cu-
bicula carved out of the tufa and opening at irregular intervals off of
the tunnels or more systematically arranged in symmetrical groupings.
Within the rooms were not only loculi but graves of different forms
arcuated niches for individual burials (arcosolia); niche tombs in-
tended to accommodate multiple burials in the floors and walls;
window tombs, which gave access to small groups of loculi via rec-
tangular windows in the walls of the galleries; a mensa tombs, in
which trench graves running parallel to the walls were sunk into the
floors of niches and covered with slabs; a cappuccina tombs, simple
trench graves covered by gabled roof tiles; and so on.
94
Contrary to
the once popular view that the uniform simplicity of catacomb burials
reflected and promoted an egalitarian ideology within the early Chris-
tian community, the variety of architectural spaces and the multiplic-
ity of grave types, even in the early phases of development of some of
the first large catacombs, suggest rather a heterogeneous mixture of
persons of different wealth and status with no distinctively unifying
beliefs about the representation of privilege in burial. In the earliest
phases of development of the catacombs of Praetextatus, Domitilla,
and Callistus, for example, one can recognize two distinct modes of
use, which correspond to topographically distinct types of regions
within the cemeteries: in certain sections one finds series of prefabri-
cated graves, with uniform columns and rows of loculi systematically
carved out for undifferentiated use; other zones, marked by less inten-
sive exploitation, are characterized by individual cubicula and graves
93
See Fabrizio Bisconti, Mestieri nelle catacombe Romane: Appunti sul declino
delliconografia del reale nei cimiteri cristiani di Roma (Vatican City: Pontificia
Commissione di Archeologia Sacra, 2000).
94
See Nuzzo, Tipologia sepolcrale, 199204, for topographically distinct zones
within the catacombs, and 16376 for a typology of the graves found within
them.
225
of different types apparently made to order and exhibiting greater
eclecticism and elaboration in their decoration.
95
The first correspond
in certain respects (uniformity, systematization, economy of space) to
the prefabricated niches for ash urns, many with built-in ollae, found in
the columbaria, where doctrinal unanimity and religious separatism
have never been suspected; the latter, in many others, recall the free-
standing hereditary and familial tombs of the visible suburban land-
scape. The novelty in the catacombs is that the two forms of burial are
integrated with each other and housed within the same undefined
space: not only were the galleries lined with loculi able to be extended,
but the cubicula set aside for more prestigious burials, even if they
resembled the traditional familial tombs of the surface topography,
opened intermittently off of spaces that were themselves the site of
burials and were evidently accessible to any who passed them.
We have few intact catacombs like the monument of the household
of Livia or the Vigna Codini columbarium excavated by Campana,
with hundreds of grave markers preserved in place, and even where we
do, the inscriptions provide little hope of identifying familial group-
ings among the undifferentiated loculi or, indeed, in the era before
Constantine, of the religious affiliation of those buried within them.
That is partly because the catacombs have been stripped of most of
their original grave goods and portable appointments, but also because
most loculi were not marked with epitaphs, and the epitaphs that are
found tend to identify a single individual with a single name, normally
a cognomen, the least distinctive element of the nomenclature then in
use; very few provide any hint of religious belief.
96
As with the colum-
95
Nuzzo, Tipologia sepolcrale, 203. The development of specific regions devoted to
ever more elaborate architectural cubicula intensified during the reign of Con-
stantine and the pontifcates of Julius (337352 C.E.) and Laberius (352366
C.E.), when lite members of Roman society (notably senators: ICUR 5.14016,
14132, 14155, 14445) began to install expensively carved marble sarcophagi
within their familial cubicula: see Fiocchi Nicolai, Origin and Development,
3743.
96
The most serious plundering of the catacombs, by specialists known as corpis-
antari during the seventeenth and eighteenth centuries, was systematic and was
virtually sanctioned by the Catholic ecclesiastical leaders: see Pasquale Testini,
Le catacombe e gli antichi cimiteri cristiani in Roma (Bologna: Licinio Cappelli,
1966), 2126. For early catacomb epigraphy, see Carlo Carletti, Nascita e svi-
luppo del formulario epigraphico cristiano: Prassi e ideologia, in Inscriptiones
Sanctae Sedis 2. Le iscrizioni dei cristiani in Vaticano (ed. Ivan Di Stefano Man-
zella; Vatican City: Monumenti, Musei, e Gallerie Pontificie, 1997), 14546: in
From Columbaria to Catacombs
226 John Bodel
baria, little effort seems to have been made in the undifferentiated
areas to accommodate family units in groups. Here it is not the epi-
graphy that points to this conclusion but the architecture, in the dis-
tribution of the loculi of varying size throughout the galleries, where
narrow columns of small niches for infant or child burials intermit-
tently interrupt the regular series of columns and layers of adult-sized
loculi all in the interest of maximizing the use of burial space
(Fig. 6.7).
97
With such schematically imposed imbalances in the
configuration of niches, few families will have been able to bury young
children next to, or even near, their parents in their own spaces. As in
the columbaria, the regimentation of niches in rows imparted uniform-
ity, but, unlike in the columbaria, the openness of the architectural
form suggested the possibility of infinite expansion and growth.
the catacombs of Priscilla, which preserves the most coherent and complete col-
lection of catacomb inscriptions before Constantine and where some 1500 loculi
were in use before the middle of the third century (see below, pp. 23839), de Rossi
found only 303 epitaphs in place: 206 Latin, 93 Greek, 4 anepigraphic. At the
catacombs of of Saints Marcellinus and Peter beside the Via Labicana, the lar-
gest of the pre-Constantinian era, fewer than 10 percent of the (ultimately)
22,500 burial spaces seem to have had inscriptions: Jean Guyon, Dal praedium
imperiale al santuario dei martiri. Il territorio ad duas lauros, in Societ ro-
mana e impero tardoantico II. Roma. Politica, economia, paesaggio urbano (ed.
Andrea Giardina; Rome: Laterza, 1986), 479 n. 63. The Latin single-name sys-
tem became common among all Romans after Caracallas extension of Roman
citizienship to all the free: see Iiro Kajanto, The Emergence of the Late Single
Name System, in Lonomastique latine. Paris 1315 octobre 1975 (ed. Nol
Duval; Paris: Centre National de la Recherche Scientifique, 1977), 42128. In
the catacombs of Priscilla in the period before Constantine, 83 percent of the
epitaphs provide no hint of religious orientation; of the some 100 epitaphs pre-
served in the so-called Area A of the catacombs of Callistus, 75 present a single
name and 76 give no indication of religious belief; similar figures obtain for the
earliest sections of the catacombs of Praetextatus and Domitilla: see Carletti,
Formulario epigraphico cristiano.
97
Cf., e.g., rooms 56, 57, 58, and 64 in region Y of the catacombs of Saints Mar-
cellinus and Peter (ca. 295320 C.E.), where the architectural innovation was ac-
companied by the development of a new decorative design similar to that found
in the columbarium of C. Scribonius Menophilus (above, Fig. 6.2), in which
a ribbon of floral and geometric motifs uniformly divides the rows of loculi:
see Zimmermann, Werkstattgruppen, 188, 197, 23334 and pls. XXXXXXI.
Some children were no doubt placed in individual loculi with their parents: see
below n. 119, on bisoma and trisoma.
227
Collegia: A Flexible Tool
The grease that oiled the funerary machine throughout the first three
centuries of the common era and enabled the major shifting of gears in
collective burial from columbaria to catacombs was the collegium.
Once the sheer size of elite familiae outgrew the capacity of traditional
funerary architecture to reconcile the principles of providing for all
members of the household and suitably distinguishing the proprietor
and his close kin, some other mechanism than familial hierarchy,
which began with the paterfamilias and was directionless without his
authoritative presence at the top of the pyramid, was needed to con-
trol and regulate the distribution of burial space. As noted above
(pp. 21718), a loose corporate structure emerged in the earliest house-
hold columbaria, occasionally beside acknowledgment of the permission
of an aristocatic patron, to control access to the monument and the
distribution of burial places within it, as can be seen from references in
inscriptions to decuriones, curatores, magistri, quaestores, and, occa-
sionaly, to a collegium itself.
98
Even where a formally incorporated col-
legium did not exist, however, as must have been the case with those
burial associations comprised entirely of slaves, the administrative ap-
paratus of the professional collegia that first surfaced formally during
98
See Hasegawa, Collegia Domestica, 25256, 26263, with references; cf.
Waltzing, tude historique, 1:38083. For permission of a patron, the monu-
ment of the Volusii provides the most abundant evidence: cf. CIL 6.7368, 7375,
7380, 22811; for reference to a collegium, cf. CIL 6.6215, 6216, 6218, 6219 (col-
umbarium of the Statilii), 6.7282 (columbarium of the Volusii).
From Columbaria to Catacombs
6.7. Cross-section showing the distribution of long (adult-sized) and short (child
or infant) loculi in gallery D of the so-called Area I (Regio A) of the catacombs of
Callistus (see above, Fig. 6.5), after Styger, Cimitero di S. Callisto, 118 fig. 17.
228 John Bodel
the Ciceronian age provided a model for autonomous self-regulation
that filled a gap left by the relinquishing of control by a paterfamilias.
99
Collegia thus filled an administrative need and, once implicated in fu-
nerary responsibilities, quickly evolved into administrative organisms
capable, owing to the virtually ubiquitous human desire to secure a
respectable burial, of infiltrating numerous walks of Roman life. This
is the main reason why the funerary responsibilities of the collegia
were misunderstood for so long as being the distinctive purpose of one
particular type: when one looks for a common denominator that unifies
the various disparate organizations grouped together under the
general rubric of voluntary associations, provision for burial of the
members is often the most conspicuous feature and, when further
grounds for characterizing a particular association more precisely are
not apparent, that function can seem to be a defining characteristic. The
vexed question of the date and scope of a so-called senatus consultum de
collegiis tenuiorum, a measure of the Augustan or Julio-Claudian period
relaxing the restrictions imposed by a Caesarian lex Iulia de collegiis
(which applied only to Roman citizens) by permitting voluntary associ-
ations of humbler persons that served the public interest (propter utili-
tatem publicam) and intending specifically, it seems, under that rubric to
allow associations that ensured the burial of their members, need not
concern us here. It is clear that the proper burial of dead members of the
community (whether or not Roman citizens) was regarded by the jurists
as a public good; that voluntary associations of various sorts flourished
during the empire; and that providing funerals for their members,
whether or not their raison dtre, was one of the principal functions
they served.
100
Securing a proper burial a goal common to humanity
99
Waltzing, tude historique, 1:4256 (followed by many) described as private col-
legia avant la lettre a number of types of voluntary associations well attested
already during the Republic, such as religious cells (of Bacchus, e.g.), political
factions (called sodalitates, sodalicia or factiones never collegia: 49), and social
clubs (see above, n. 27). By these standards professional collegia defined by par-
ticular trades (cf. Dig. 50.6.5.12) had existed since the regal period, when, ac-
cording to legend, Numa divided the people into groups on the basis of their
occupations: cf. Plut. Numa 17.12 with Waltzing, tude historique, 1:6169.
100
For the (meager) legal evidence for the senatorial decree, see Dig. 47.22.1 (Mar-
cianus). For the (copious) modern discussion, see recently Wendy Cotter, The
Collegia and Roman Law: State Restrictions on Voluntary Associations, in Vol-
untary Associations in the Graeco-Roman World (eds. John S. Kloppenborg and
Stephen G. Wilson; New York and London: Routledge, 1996), 7489; Luuk de
Ligt, Governmental Attitudes towards Markets and Collegia, in Mercati per-
229
and socially beyond reproach (and thus effectively immune to interfer-
ence from the imperial authority) enabled the collegium to adapt and
survive, even after the official conversion of the empire to Christianity
eliminated or subverted other Roman social institutions of much longer
standing.
101
The funerary role of the collegium, ironically, was born
to meet one social need the proper burial of groups too large or
too amorphous and heterogeneous to be accommodated directly by the
traditional familial and patronal mechanisms of support but grew up
to address another, the desire for self-defining communities to express
solidarity and corporate unity within a recognized and acceptable (if al-
ways to a certain extent controversial) institutional framework.
102
The flexibility of the form has also enabled scholars to shape their
conception of the purpose and nature of the institution to suit their
own predilections and circumstances. Three of the greatest, whose
pioneering studies during the nineteenth century have formed the basis
(sometimes unquestioned) of most modern discussions, poured into
the empty container of the collegium very different mixtures of the so-
cial and political thought that percolated through their times. For
Mommsen collegia were secular organizations devoid of religious
orientation that served mainly social funtions. For de Rossi they fur-
nished the mechanism by which the early Christian community organ-
ized itself and exerted its property rights over communal cemeteries.
For Waltzing they were beneficial labor organizations, the prototypes
of the Christian Democratic professional associations that formed the
backbone of a well-run imperial society.
103
For our purposes it is only
de Rossis thesis that requires attention.
manenti e mercati periodici nel mondo romano (ed. Elio Lo Cascio; Bari: Edipug-
lia, 2000), 24252; id. D. 47,22,1,pr.-1 and the Formation of Semi-Public Col-
legia, Latomus 60 (2001): 34558. For the basic principle that the burial of
corpses was in the public interest, see Dig. 11.7.43.2a (Papinian) and 11.7.12.3
(Ulpian), with Bodel, Graveyards and Groves, 3334.
101
See the remarks of Carolyn Osiek, below p. 269, on the eventual usurpation of
the private patronage of collegia by Christian bishops during the fourth and
fifth centuries.
102
See Francesco Maria De Robertis, Causa funeris causa religionis: le commu-
nit cristiane tra normativa statale e messaggio evangelico (a proposito di D.
47,22,1), SDHI 54 (1988): 23949.
103
For the ideological currents of late-nineteenth-century European social thought
that informed the divergent theories of Mommsen, de Rossi, Waltzing, and
lesser scholars writing on the subject of funerary collegia during the same
period, see the interesting discussion of Perry, The Roman Collegia, 2388.
From Columbaria to Catacombs
230 John Bodel
According to de Rossi, collegia provided not only a protected and
formally recognized medium for Christians to congregate legally but
also a legitimate means for them to own communal property, particu-
larly burial grounds, corporately.
104
As Carolyn Osiek observes, how-
ever (below, p. 266), there is a significant difference between regarding
early Christian congregations as adapting the administrative appa-
ratus of communal voluntary associations to ensure the burial of their
members, as they surely did, and seeing the Church as a formally
constituted legal collegium which in that capacity corporately owned
collective cemeteries reserved for the burial of Christians. In fact, as is
well known, Roman law did not recognize corporate ownership of
property but regarded private communal funds, real estate, and com-
modities as belonging collectively but individually to the persons who
came together for the purpose of owning them. Thus, in the case of
collegia, a person illegally enrolled in two associations and therefore
required to withdraw from one of them would receive from the col-
legium he left the share of the common fund (ratio communis) due him,
and those who belonged to collegia judged illegal and therefore dis-
solved were permitted to divide among themselves the common funds
(pecunias communes) of the association upon its dissolution.
105
The
analogy sometimes drawn by those who argue that corporate owner-
ship of property (notably cemeteries) by communities of Christians
was recognized in practice even before Constantine in 321 C.E. for-
mally established Christian churches as juridical entities with property
rights between collegia and collective entities such as cities and col-
onies fails to recognize the difference between communal property of
this sort, which belonged collectively to the several individual owners,
and public property owned by public bodies such as cities and colo-
104
In a letter of 19 June 1882 to his friend Louis Duchesne, Director of the cole
franaise de Rome, de Rossi referred explicitly to le droit du corpus christiano-
rum, come possesseur de cimetires: see Patrick Saint-Roch, ed. Correspond-
ance de Giovanni Battitsta de Rossi et Louis Duchesne, 18731894 (Rome: cole
Franaise de Rome, 1995), 221 (letter 174) with the analysis of Perry, The
Roman Collegia, 4958.
105
Dig. 47.22.1.2 (Marcianus), Non licet autem amplius quam unum collegium lici-
tum habere et si quis in duobus fuerit, rescriptum est eligere eum oportere, in
quo magis essse velit, accepturum ex eo collegio a quo recedit id quod ei competit
ex ratione quae communis fuit. Dig. 47.22.3. pr (Marcianus), Collegia si qua fuer-
int illicita disolvuntur: sed permittitur eis, cum dissolvuntur, pecunias communes
si quas habent dividere pecuniamque inter se partiri.
231
nies, which belonged to no one individual (res nullius) but was set aside
for the common use of all (res publica).
106
A second objection concerns the common assimilation by advocates
of de Rossis view of the tenuiores identified in the senatus consultum
with the Christian poor (egeni ), whose proper burial the Apostolic
Tradition and Tertullian claim was the general responsibility of the
community.
107
The supposed equivalency of the two categories rests on
a misunderstanding of the Latin terms. Egenus, in classical and ecclesi-
astical Latin, means needy, indigent, destitute; as a substantive,
it is vox propria for pauper, one without means. The adjective tenuis,
by contrast, when applied to persons, particularly in legal contexts,
refers primarily to social standing rather than to wealth; as used sub-
stantively by jurists in its comparative form it acquires almost the
status of a technical term and in the plural defines a category equiva-
lent to that of the humiliores; specificallly it describes those who do
not belong to one of the legally recognized higher ordines (senators,
knights, and in some contexts municipal magistrates), many of whom
certainly possessed sufficient financial means to pay for their own
burials and those of their families.
108
It is evident that those whom the
Christian writers refer to as the poor in such contexts Christian
poor, it may be noted: there is no hint that Christians shared the
broader Roman conception of a public interest in the burial of all who
died in Roman territory were indeed tenuiores, but not all tenuiores
were poor, let alone indigent. Indeed, those who belonged to collegia
tended to be more prosperous than the average urban and municipal
resident and regularly received portions of higher value than common
106
For the legal concept of public property, especialy real estate, cf. e.g., Dig.
43.79, especially 43.7.1 (Pomponius), 43.8.2.3, and 43.8.2.5 (Ulpian). For the
analogy, see, e.g., de Visscher, Droit des tombeaux, 26571, cited by Osiek,
below, p. 264 n. 44.
107
See above, n. 54, especially Tert. Apol. 3839.
108
Egenus: cf., e.g., Plaut. Capt. 2.3.46; Verg. Aen. 6.91; Vulg. Deut. 15.11, Psa.
34.10. For tenuis, as applied to a segment of society, see, e.g., Cicero, Leg.
3.10.24, Fin. 2.20.66, Mur. 70, etc. with Guy Achard, Pratique rhtorique et
idologie politique dans les discours optimates de Cicron (Mnemosyne
Supplement 68; Leiden: E. J. Brill, 1981), 376; Livy 2.3.2. For legal usage, cf.
also Dig. 38.28.2, 48.19.28.2, where the term is synonymous with humiliores (see
above, n. 84); 50.6.6.12, where the category is explicity contrasted with that of
those capable of undertaking the financial obligations of municipal office
(munera civitatium) (Callistratus); note also Garnsey, Social Status, 22223.
From Columbaria to Catacombs
232 John Bodel
members of society in the hierarchically scaled distributions of public
and private largesse.
109
Collegia were indeed important in the history of the early Christian
community at Rome, but not in the way that de Rossi imagined.
Rather than forming the administrative apparatus by which a unified
Christian church exercised its legitimate property rights, they provided
a flexible model for urban Christians to organize themselves in groups
(sometimes, but by no mean always, congregations) to express a com-
mon interest in collective burial. In this they differed in particulars but
not in kind from other collegia that found convenience and solidarity
in uniting for a common purpose. So, for example, even after churches
became recognized juridical entities capable of owning (and thus of
controlling) collective cemeteries, we find particular groups of Chris-
tians, such as workmen involved in the public distribution of grain
(mensores frumentarii ), for whom Christianity may or may not have
been an important element of identity, organizing themselves into col-
legia (in this case a professional association) and providing separate
and independent burial accommodation for their members within
the large collective catacombs.
110
Behind the collegia stand individual
proprietors of funerary properties purpose-built columbaria and
catacombs private patrons in some cases but also entrepreneurs and
developers. Enterprising businessmen in the death trade, it seems, in-
spired the major developments in Roman burial architecture over the
first three centuries of empire. They operated, often, behind the scenes,
but their role in the process was fundamental. Investigating their in-
volvement in the funerary industry, however, is beyond the scope of
this essay.
109
See, e.g., Patterson, Patronage, Collegia, and Burial, 21; id. The Collegia and
the Transformation of the Towns of Italy, in LItalie dAuguste Diocltien
(Rome: Ecole Franaise de Rome, 1994), 234; Van Nijf, Civic World, 1823.
110
For the distinctively decorated cubiculum established by a collegium of mensores
frumentarii around the middle of the fourth century in a region of specially de-
signed architectural cubicula within the catacombs of Domitilla, see Phillipe
Pergola, Mensores frumentarii christiani et annone la fin de lantiquit (relec-
ture dun cycle de peintures), RACrist 66 (1990): 16784.
233
Conclusion
There is little evidence to suggest that catacombs were invented by
Christians or were originally exploited by Christians to serve their pe-
culiar socio-religious ends. The form grew naturally out of develop-
ments in the design, and probably the economics, of funerary space
introduced with the columbarium during the last decades B.C.E. and
seems to have been inspired mainly by the irresistible press of a popu-
lation that continued growing and dying beyond the capacity of the
suburban landscape to house the bodies. An independently inspired
(and, for the thesis advanced here, irrelevant) change in preferred
method of disposal from cremation to inhumation beginning in the
latter first century exacerbated but did not itself create a demand that
had by then been growing for nearly 200 years, when the introduction
of columbaria first signaled a problem. Once discovered, however, and
put into use by the mixed population of Rome, the catacomb form
quickly recommended itself to the Christian community for its open
design and otherworldly ambience, which made possible a radical re-
formulation of the theological order as expressed through the relation-
ship between the divinity and the dead.
111
The model of the underground cemetery was moreover well suited to
enabling the early Christian community to express its conception of an
ideal society through its burial customs, in the same oblique but
culturally specific way that earlier Romans had expressed theirs. By
honoring their dead communally as brothers and sisters in Christ,
Christians expanded their family to a size that soon dwarfed even
the largest of the imperial households.
112
The catacombs enabled them
to maintain the familial model of traditional Roman funerary com-
memoration without incurring the risk of running out of space, as even
the familia of the empress Livia had done. In that subterranean world
without horizons and poles, new centers of gravity naturally formed
around those with the greatest weight in the new world order, which,
increasingly, meant those most closely connected with the church and
its origins. If the archaeological dating of the early development of the
catacombs is correct, however, the cult of relics and the competitive
111
The analysis of this fundamental change by Peter Brown, The Cult of the Saints:
Its Rise and Function in Latin Christianity (Chicago: University of Chicago
Press, 1981) is too well known to require further explication.
112
See the concise but incisive remarks of Harries, Death and the Dead, 6061.
From Columbaria to Catacombs
234 John Bodel
desire among Christians for burial near their saints (tumulatio ad sanc-
tos) merely intensified but did not cause the expansion of the subter-
ranean complexes of linked hypogea into large collective cemeteries,
which was already well advanced by the middle of the third century.
Rather it provided a focus to the accumulation of graves near the new
holy centers, which often were located, not near the center of the
vast complexes, but at their periphery.
113
Thus grew up the cult of martyrs, saints, and bishops that first burst to
the surface with Constantines basilica over the grave of Peter and then
erupted repeatedly above ground later in the fourth and fifth centuries in
a series of funerary basilicas, chapels, and burial areae that ringed the
city in a constellation of mega-tombs (the great centers of Christian pil-
grimage) all communally shared by the ever-growing Christian familia.
The graves visited now were no longer those of biological relatives but of
the new Christian saints, kin to all in the ecumenical family. The familial
model of Roman funerary commemoration thus endured, even as the
concept of the family grew to encompass all who shared the Christian
faith. The traditional family endured also, of course, and continued to
assert its cohesiveness in burial through the device of the cubiculum. The
difference now was that the family unit (whatever its precise composi-
tion) no longer isolated itself in independent structures designed to
segregate the chosen few from outsiders but rather established itself
within the broader community, in the communal subterranean spaces
shared by all. Tombs in this new age were no longer final destinations
but mere way-stations, places for resting refrigerium, in the contempo-
rary Christian parlance on the way to salvation and resurrection.
The significant changes in this two-stage process, I have tried to
argue, are not in fact found where they have traditionally been located
in the growth in monument-size, from small familial tombs to large
columbaria, and in the switch from cremation to inhumation but
rather in the separation of the slave household from the biological
family, which opened the way for new, extra-familial expressions of al-
legiance and social order, now increasingly articulated through the in-
finitely adaptable instrument of the collegium, and in the movement
from above ground to beneath the surface, which enabled the develop-
ment of a burial mode ideally suited to the new theology all inclusive,
113
In this I disagree with my friend Carolyn Osiek, below p. 256, who is certainly
correct that the cult of the saints provided new focus and impetus to the expan-
sion of the vast complexes.
235
otherworldly, and capable (seemingly) of universal expansion and
growth. By the time the Christian cemeteries around Rome returned to
the surface of the land during the later fourth and fifth centuries, and
areae replaced catacombs as the preferred loca of commemoration, the
ideological foundation of the imperium Romanum had fundamentally
changed, and the traditional Roman tendency toward assertions of
privilege and rank projected itself against a new backdrop of the Chris-
tian faith. If the arguments presented in the preceding pages have any
merit, in the momentous shift that this new orientation ultimately ef-
fected in the history of European civilization, the developments in col-
lective burial practices that took place during the first three centuries
of the new Christian era played a significant part. The transition in
burial architecture from columbaria to catacombs, which replaced a
closed, isolating system of commemorative expression with an open,
inclusive form suggesting commonality and community, was exploited
to excellent effect by Christians during the century and a half after
Constantine. Whether the extensive underground cemeteries devel-
oped already during the third century belonged originally and exclus-
ively to that separatist community is considerably less certain.
Appendix
Known Burials at Rome, 25 B.C.E.325 C.E.
No one knows how many burials from the three and a half centuries of
pre-Christian imperial Rome have been reported, let alone discovered,
but by combining the figures derived from some obvious and well-rep-
resented sources with plausible estimates of the uncalculated numbers
from some recognized categories of evidence, one can arrive at an ap-
proximate total not likely, perhaps, to be off by more than 20 to 30 per-
cent a margin of error unacceptable for many purposes but useful
enough for ours, as long as the uncertainties on which it is based are
kept firmly in mind and the arguments to which it is harnessed remain
candidly tentative and exploratory. The suggested total of 150,000
known burials between the time of Augustus and that of Constantine,
then, is no more than an educated guess, but not a useless one, perhaps,
for suggesting the tiny percentage, by any reckoning, of those for which
we have any evidence at all.
Calculations that overrepresent the actual numbers known and es-
timates that err on the high side will present the case in the strongest
From Columbaria to Catacombs
236 John Bodel
possible light, since the aim is to suggest that our information con-
cerning the likely burials in the vicinity of Rome is exceedingly
meager and not necessarily representative. The estimate of 150,000,
which represents just such a figure, is rounded up from a calculated
total (149,700) derived from adding to the number of surviving epi-
taphs registered in volume six of the Corpus Inscriptionum Latinarum
devoted to the city of Rome (some 26,000); the number of funerary
insciptions destined for the new supplement to be edited by Silvio
Panciera (around 10,000); an estimate of the number of unmarked
cremation graves (mainly in columbaria) found in the suburban re-
gions of the city (some 11,000); a guess as to the number of simple,
often (but not always) unmarked, inhumation burials in shallow in-
dividual graves in suburban necropoleis (perhaps 20,000); a very
rough estimate of the number of bodies interred in loculi or inhumed
in associated surface cemeteries of the catacombs in use prior to the
time of Constantine (as many as 62,700); and acknowledgment of the
existence in what quantities we cannot know of mass graves, some
in catacombs and not only for the indigent, that sometimes com-
prised as many as 1,000 corpses (possibly 15,00020,000?). It will be
useful to summarize briefly what little evidence we have for each of
these categories in turn.
Inscribed epitaphs: the folly of relying on published inscriptions for
useful biometrical information about ancient populations is well
known, and the sources of bias in our sample need not be rehearsed.
114
The numbers that are known, however, can be counted and provide a
minimum baseline for individual graves. Many epitaphs, of course, are
dedicated to more than one person, often to three or four persons by
name (to say nothing of the collective formulae sibi et suis, etc.). But
since we cannot be certain that those included in an inscribed dedi-
cation were in fact buried where the epitaph was posted (in certain
cases they demonstrably were not), it seems safer, in order to avoid
double counting, to reckon their numbers among the anonymous
graves and to count a single epitaph as attesting a single burial. In
many instances, of course, the epitaph is detached from the grave itself
and provides all we know of the burial it commemorates; but since our
114
See, e.g., the contributions of Jean Marie Lassre, Pierre Salmon, and Keith
Hopkins, in Franois Hinard, ed., La mort, les morts, et lau del dans le monde
romain (Caen: Universit de Caen, 1987); Timothy Parkin, Demography and
Roman Society (Baltimore, Md.: Johns Hopkins University Press, 1992), 519.
237
aim is to cast the net as widely as possible, counting even an intended
burial (whether or not realized in the manner attested) seems justified.
The 26,000 epitaphs recorded in CIL 6 furnish numerous illustrations
of the situations mentioned; the estimate of 10,000 unpublished epi-
taphs from the environs of Rome derives from Professor Silvio Pan-
ciera and is based upon the archive of schede compiled and preserved
at the Department of Latin Epigraphy at the University of Rome, La
Sapienza, from which he is preparing the new supplement to CIL 6.
Unmarked cremation graves: perhaps as many as three quarters of
the some 5,500 burials accommodated in the ten largest known colum-
baria are anonymous: see above, n. 57 [N = 4,125]. If we allow the same
percentage of unmarked burials in another 45 smaller columbaria un-
covered between 1700 and 1920 and registered in CIL 6, each compris-
ing no more than 100 ollae (for a total, in other words, of no more than
3,375), and also in as many as have been uncovered between 1920 and
today, then the number of known but unmarked cremation graves in
the environs of the city amounts to, at most, slightly fewer than 11,000
(N = 10,875).
Individual surface inhumations: this is the type of burial perhaps least
likely to have survived the ravages of time, since the suburban topsoil
around Rome over the last two millennia has been so frequently tilled,
excavated, and built over that most of the burials originally consigned
to it have certainly vanished, but in recent years closer attention to
chance discoveries made in the course of construction work in the en-
virons of the city has revealed concentrations of simple surface burials
sufficient to suggest the scale of their original numbers. Hydraulic
works at Isola Sacra during the late 1980s, for example, uncovered
among the monumental tombs of the well-known necropolis outside
Portus some 650 a cappuccina graves and simple formae dating to the
second and third centuries.
115
And in 2004 construction work on a
high-speed rail line beside the Via Collatina outside Rome revealed
some 2,000 unmarked simple inhumation graves in a vast surface cem-
etery tentatively dated to the second century.
116
There is no telling how
115
See Sergio Angelucci, Ida Baldassarre, Irene Bragantini, Maria Giuseppina
Lauro, Vanni Mannucci, Alberto Mazzoleni, Chiara Morselli, and Franca Tag-
lietti, Sepolture e riti nella necropolis dellIsola Sacra, Bollettino di Archeolo-
gia 56 (1990): 49113.
116
The Via Collatina graves have not yet been published but will form the subject
of a forthcoming article by Stefano Musco and Anna Buccellato.
From Columbaria to Catacombs
238 John Bodel
many similar finds may yet be unearthed, but it is certain that many
similar necropoleis, if they existed in large numbers outside Rome,
have disappeared without a trace. The figure of some 20,000 such
simple inhumations therefore represents a generous estimate of the
number that might one day be discovered, rather than the number that
once existed, which cannot even be guessed.
Catacombs: The total number of loculi in the excavated catacombs is
not known, let alone the total number of those that might have been in
use before the time of Constantine, but approximate figures for the
larger cemeteries can be hazarded. The most serious attempt to esti-
mate the number of burials accommodated in a single large pre-Con-
stantinian complex, that of Marcellinus and Peter at the imperial prop-
erty ad duas lauros beside the Via Labicana, arrived at a total of 11,000
loculi distributed throughout two kilometers of tunnels during the first
fifty years of the use of the site following its opening around 260 C.E.,
with perhaps as many as 6,000 surface burials (sub divo) in the plot
overlying the subterranean tunnels.
117
Three other large catacombs in
use during the third century that of Priscilla beside the Via Salaria
(c. 200230 C.E., the largest of this period), Area A of the catacombs
of Callistus beside the Via Appia (c. 230240), and the catacombs of
Novatianus beside the Via Tiburtina (c. 260270) each housed be-
tween 1,200 and 1,500 loculi. Three others in use during the first half
of the third century those of Domitilla beside the Via Ardeatina
(the Area of the Flavii, c. 200230?), of Praetextatus beside the Via
Appia (c. 200230), and of Calepodius beside the Via Aurelia Vetus
(c. 230250) may each have included between 500 and 1,000 subter-
ranean burial spaces. During the second half of the third century the
existing cemeteries were expanded and new catacombs were opened: in
addition to that of Marcellinus and Peter, these included the so-called
Coemeterium Maius on the Via Nomentana (c. 250), the nucleus of
Agnese (Regio 1, c. 250), the lower levels of the catacombs of Pam-
philus (c. 260300) and Priscilla (c. 300310), and those of the Villa
Doria Pamphilj on the Via Aurelia Vetus. These are said to have con-
tained thousands of burials, but the actual figures are unknown. If
we allow a generous 2,000 to each, add another 2,000 for the total
number of loculi in a half dozen much smaller complexes dated to the
second half of the third century, and imagine every catacomb to have
117
Guyon, Dal praedium imperiale, 315; cf. 478 n. 52 probabilmente stime es-
sagerate; id. Le cimitire aux deux lauriers, 101.
239
included burials sub divo in the surface soil overlying the tunnels in the
same (generous) ratio as that estimated for the catacombs of Marcel-
linus and Peter (that is, approximately 1:2, or a range of 6001000
for the larger complexes and 250500 for the smaller ones), the total
number of burial spaces in the catacombs in use before Constantine
would amount to some 41,800.
118
Multiple burials and ossuaries: That figure (41,800) is impressive
(and, one suspects, somewhat exaggerated), but there are reasons to
mistrust any such calculation of numbers of loculi as a basis for esti-
mating the number of Romans buried in the catacombs during the
third century. Many loculi could, and some in fact did, house more
than a single burial. Not only were infants sometimes interred along
with (one presumes) a parent or parents in a single niche, but some
loculi, when found, contained two or even three adult skeletons,
lying side by side next to each other on the tufa shelves. Some of these
double and triple burials were explicitly identified in accompanying
epitaphs, but others, apparently, were not, and since no systematic rec-
ords of such multi-person loculi seem to have been kept, there is no tell-
ing how common the practice was.
119
Nor can we guess how often a
single loculus might have been cleaned out and reused for new burials
altogether, as was certainly the case with the original burial sites of
some 800(?) corpses stacked in an old pozzolana quarry converted into
an ossuary beneath Area A of the catacombs of Callistus.
120
Medi-
118
For figures, see Vincenzo Fiocchi Nicolai, Lorganizzazione dello spazio
funerario, in Christiana Loca: lo spazio cristiano nella Roma del primo millen-
nio (ed. Letizia Pani Ermini; Rome: Ministero per i Beni e le Attivit Culturali,
2000), 45 and Fiocchi Niccolai, Origin and Development, 1736, esp. 30.
119
For double burials indicated in epitaphs by the word bisomum, cf., e.g., from
the catacombs of Commodilla, ICUR 2.6030 (370 C.E.), 6110, 6128, 6183,
8680; from the Callistus, ICUR 3.3235, 9076, 9143, 9876, 10146; from the
Domitilla, ICUR 3.7354b, 7574, 7709, etc. For triple burials (trisomum), cf.,
e.g., from the Commodilla, ICUR 2.6310; from the Callistus, ICUR 3.9029,
9152; from the Domitilla, ICUR 3.8485, etc. All these examples belong to the
fourth and fifth centuries, when the practice flourished, but the lack of explicit
epigraphic testimony for double and triple burials during the third century can-
not be taken as proof that the practice did not occur.
120
Josef Wilpert, La cripta dei papi e la capella di Santa Cecilia nel cimitero di Callisto
(Rome: Descle & C. Editori Pontefici, 1910), 7580. The bodies, which had evi-
dently been removed from loculi and cubicula near the so-called Crypt of the
Popes, were laid out in rows and stacks, four meters high, with a thin layer of dirt
between each layer. After consulting a local physician, Wilpert reported that the
From Columbaria to Catacombs
240 John Bodel
eval itineraries report similar discoveries of ossuaries in the catacombs
filled with martyrs, but provide little useful information about
numbers or precise locations.
121
Mass graves: Equally problematic is the recognition that the open
spaces of certain catacombs were used for mass burials in ways that
defy precise calculation of the numbers of bodies buried there. In the
catacombs of S. Thecla beside the Via Ostiense, for example, some of
the large rooms (cameroni) opening off the galleries, each of which
provided some 70 to 100 burial spaces in narrow loculi lining the walls
from floor to ceiling and in formae sunk into the floors, were filled to
capacity with layers of corpses stacked one on top of another, each
layer separated by roof tiles or large bricks covering a corpse below.
122
Similarly, in Regio A of the catacombs of Commodilla, near the
martyrs tomb, some 45 funerary wells, 1 by 1.71.9 meters in area and
67 meters deep, were sunk into the floor, each of which accommo-
dated 1015 loculi cut in two facing columns into the walls; the centers
bodies had been deposited as skeletons, but a recent reconsideration of the evi-
dence (reported by Rafaella Giuliani in an oral communication: see below, n. 125)
suggests that the corpses were arranged in stacks before decomposing. Intact
skeletons are seldom moved without becoming disarticulated. The number 800
seems to be derived from an itinerary compiled from a report by William of
Malmesbury (twelfth century) (Notitia portarum viarum eclesiarum circa urbem
Romam e Willelmo Malmesburgensi): DCCC martyres ibidem requiescunt. Four
hundred years earlier the Itinerarium Salisburgense recorded that eighty martyrs
rest there down below (sc. the tomb of S. Cecilia) (LXXX martyres ibi requiescunt
deorsum). One suspects that perhaps a sribal error or lapsus memoriae accounts
for the expansion tenfold of the number of skeletons reported. The Epitome libri
De locis sanctorum marturum e codicibus Salisburgeni puro, Wirgeburgensi puro, et
Salisburgeni interpolato speaks vaguely of a countless number (innumerabilis
multitudo martyrum): for all these texts, see De Rossi, Roma Sotterranea, 1:180.
121
At the catacombs of Pontianus (fourthfifth century) beside the Via Portuense ad
ursum pilleatum, for example, according to the Epitome libri De locis sanctorum
marturum e codicibus Salisburgensi puro, Wirgeburgensi puro, et Salisburgensi in-
terpolato, you will find the church of S. Candida, a virgin and martyr, whose
body rests there. You descend into a cavern and you will find there a countless
number of martyrs and that whole cavern is filled with the bones of martyrs
(invenies ecclesiam s. Candidae virginis et martyris, cuius corpus ibi quiescit. De-
scendis in antrum et invenies ibi innumerabiliem multitudinem martyrum et
omnis illa spelunca impleta est ossibus martyrum): De Rossi, Roma Sotterranea,
1.182 no. II.
122
Umberto M. Fasola, La basilica sotterranea di S. Tecla e le regioni cimiteriali
vicine, RACrist 46 (1970): 23852.
241
of the wells were filled with a cappuccina tombs laid one on top of an-
other, as in the catacombs of S. Thecla; collectively they housed more
than 1,700 inhumations.
123
At the so-called Coemeterium Maius on the
Via Nomentana nearly a hundred corpses of adults and children were
found in 1956 laid in two layers in the bare soil.
124
These mass burials
belong to the fourth century after Constantine, but in 2003 a series of
rooms and galleries in a network of subterranean burial chambers at
the catacombs of Marcellinus and Peter were found to be filled with
some 1,200 corpses, said to be well dressed and dating (on the basis of
coins, jewelry, and fabric found with the remains) to between 150 and
250 C.E.
125
How many similar such finds may yet be made, or were
once made but not reported, or were reported only vaguely (as in the
medieval itineraries) is difficult to say, but there is little reason to think
that these discoveries are unique. There is therefore no point in pre-
tending that we can estimate with any confidence the numbers of Ro-
mans buried in the catacombs before the time of Constantine, but if we
take our generous calculation of the total number of loculi (41,800),
guess that no more than half of them could have been used for double
burials (+ 20,900) and allow for perhaps as many as 15,00020,000
burials in mass graves and loculi used for more than two adults, we may
not seem to underestimate the total.
If we think that we may have a rough idea of how some 150,000 Ro-
mans were buried during the three and a half centuries between
25 B.C.E. and 325 C.E., we can only guess by what means and where
the other 98.5 percent of the presumed numbers who died during that
period were buried, but it is unlikely that wholly different and unrec-
ognized means of disposal could account for any significant number of
them. Of the methods surveyed above, simple surface burials (whether
of cremations or inhumations) are perhaps the most likely to have left
123
Nuzzo, Tipologia sepolcrale, 2526.
124
Umberto M. Fasola, Le recenti scoperte agiografiche nel Coemeterium
Majus, Rend. Pont. Acc. 28 (195556): 8586 (Fig. 7).
125
This spectacular find has been published in preliminary fashion by Philippe Blan-
chard and Dominique Castex, with Michal Coquerelle, Raffaela Giuliani, and
Monica Ricciardi, A mass grave from the catacomb of Saints Peter and Marcel-
linus in Rome, second-third century AD, Antiquity 81 (2007): 98998. Initial
excavations focused on a series of rooms in regio X of the complex, of which
numbers 16, 78, and 80 were reportedly filled with stacks of bodies up to twelve
layers high.
From Columbaria to Catacombs
242 John Bodel
no trace in the archaeological record, but we must admit that our
ignorance is profound and that our best calculations barely scratch the
surface of a significant problem for any study of mortuary practices
during the early Empire. The question remains: where were the bodies
buried?
126
126
According to one recent estimate, some 30,000 tombs are known from a com-
parable period of early Chinese history, during the Qin (221206 B.C.E.) and
Han (206 B.C.E.220 C.E.) dynasties (Michael Loewe, in lecture, Brown Uni-
versity, October 2005).
Roman and Christian Burial 243
Carolyn Osiek
Chapter 7
Roman and Christian Burial Practices
and the Patronage of Women
1
We have much to learn from interdisciplinary cooperation. One of the
academic divides has been between Roman historians and Christian
historians; this project happily spans the gap between them.
2
Another
divide has been between Christian archaeologists who study material
remains and Christian historians who study texts. This chapter aims to
help bridge that gap as well.
Early in the third century, several Christian texts seem to indicate
that Christians are burying their dead in common areas. In North
Africa, Tertullian refers to animosity on the part of others towards
Christians and the areae supulturarum nostrarum, our burial fields
(Ad Scapulam 3.1). He also refers to a Christian practice of taking up a
collection once a month for a variety of charitable practices, among
them the feeding and burial of the poor (Apol. 39.56). Elsewhere in the
Apology he refers to an incident of nocturnal mob violation of Christian
burials, but this does not necessarily indicate common burial of an en-
tire Christian community together (Apol. 37). The Apostolic Tradition of
Hippolytus refers to assuring the burial of the poor in the otjtjto;
the Sahidic translation uses the Greek loanword (Trad. AP. 40).
3
Pos-
1
Thanks to my two respondents at the May 2005 conference, Annette Huizenga
and Matthew Perry, some of whose suggestions have been incorporated here.
Basic terminology and setting for this chapter are contained in John Bodels pre-
vious chapter, which should be read before this one.
2
An earlier conference in 1999 with the same aim culminated in Early Christian
Families in Context: An Interdisciplinary Dialogue, ed. David L. Balch and Ca-
rolyn Osiek (Grand Rapids, Mich.: Eerdmans, 2003).
3
Paul Bradshaw, Maxwell E. Johnson, and L. Edward Phillips, The Apostolic
Tradition (Hermeneia; Minneapolis, Minn.: Augsburg Fortress, 2002), 191.
244 Carolyn Osiek
sibly a few years later, Origen homilizes in Caesarea about the good
old days of the great martyrs, when the whole church gathered to-
gether in assembly coming 0 tuv otjtjiuv, from the cemetery
(ies) (Hom. Jer. 4.3.16). Still a few years later, at the time of the Augusti
Valerian and Gallienus (253259 C.E.), Dionysius of Alexandria
relates, as preserved by Eusebius, that Christians there were forbidden
by the sub-prefect Aemilianus to hold assemblies or to go to the
so-called cemeteries (ri to oourvo otjtjto) (Eusebius, Hist.
eccl. 7.11.10).
4
All of these references could be to familial tombs owned
by Christians and perhaps extended to needy members of the same
congregation, rather than common or community-owned cemeteries.
The most famous of these passages, and possibly the earliest, comes
from the beginning of the third century, from Hippolytuss denunci-
ation of his rival, Callistus, who, he relates, was placed by Zephyrinus,
bishop of Rome 198217, over the otjtjtov, commonly under-
stood to be the catacomb that now bears his name (Refutatio 9.12.14).
This burial complex, identified and explored in 1850 by the great nine-
teenth-century explorer of the catacombs Giovanni Battista de Rossi,
would later contain the tombs of nine bishops of Rome from 230 to
274 in the so-called Crypt of the Popes. The generic reference to the
cemetery without a specific name, at a time when there were surely
multiple burial complexes around the city of Rome being used by
Christians, led de Rossi to the conclusion that this was the first cem-
etery to be administered, if not owned outright, directly by church
authorities. Little further evidence has come to light since de Rossi to
change that judgment. But oddly, Callistus himself, according to tradi-
tion later bishop of Rome 217222, was buried elsewhere, in the cata-
comb of Calepodius on the Via Aurelia, which relativizes the claim
that the catacomb of Callistus was, at the time of the eponymous fig-
ures death, any kind of official burial place of the church of Rome.
A century later, most of the Christian burial complexes around
Rome were probably coming rapidly under administration by church
4
Vincenzo Fiocchi Nicolai, Fabrizio Bisconti, and Danilo Mazzoleni, The Chris-
tian Catacombs of Rome: History, Decoration, Inscriptions (trans. Cristina Carlo
Stella and Lori-Ann Touchette; 2d ed.; Regensburg: Schnell & Steiner, 2002),
1315; ric Rebillard, Religion et spulture: lglise, les vivants et les morts dans
lAntiquit tardive (Paris: ditions de lcole des Hautes tudes en Sciences So-
ciales, 2003), 1520. Rebillard argues that the word otjtjtov did not refer in
general to burials, but to the tombs of martyrs, but his evidence is mostly fourth
century, by which time every Christian cemetery had its martyrs.
Roman and Christian Burial 245
officials, if not by a central administration, then at least by that of local
assemblies. Yet at the same time, there is another noticeable trend:
among the names that emerge, those of some of the most prominent of
these complexes are women. By the middle of the fourth century, the
cult of the martyrs was in full swing, every major catacomb had its
memoria of its own martyrs, and pilgrimage to their tombs was becom-
ing a major enterprise, leading to the creation of underground cem-
eterial basilicas over or near the tombs of the martyrs, accessible by
staircases from ground level, roofs sometimes protruding above the
ground, as is the case, for example, in the late-fourth-century basilica of
Sts. Nereus and Achilleus in the catacomb of Domitilla. The original
use intended for these basilicas was family and eventually community
funerary banquets, leading to the all-night excesses discussed by Robin
Jensen elsewhere in this volume. The faithful continued for about an-
other century to want to be buried in the great underground complexes
ad sanctos and also above ground in the same area, as close as possible
to the holy places. The fossores (diggers) of the fourth and fifth cen-
turies had full-time jobs not only doing the actual digging but serving
as agents in the sale of burial space, the clergy having little to do with
the whole business transaction.
5
Yet there is today common agreement that all of the burial areas that
were to become Christian catacombs began as private property and
private burial areas, in most cases at a time before any Christian iden-
tity can be documented. Even most of the references given above need
not refer to common church ownership of burial property. Tertullians
allusions could be not to cemeteries reserved to Christian use, but to
private burial plots known to belong to Christian owners.
6
The same
could be true of situations referred to by Origen and the Apostolic
Tradition. Tertullians burials of the poor from the common fund could
5
Fernand de Visscher, Le droit des tombeaux romains (Milan: Giuffr, 1963),
3950; Jean Guyon, La Vente des tombes, MEFRA 86 (1974): 54996. In other
parts of the world, however, the clergy may have been more involved in the sale
of burials: witness the role of Flavia Vitalia, presbytera, in the sale of a property
in Solin, Croatia, in 425 (CIL 3:14900; Ute Eisen, Women Officeholders in Early
Christianity: Epigraphical and Literary Studies [Collegeville, Minn.: Liturgical
Press, 2000], 13132; Kevin Madigan and Carolyn Osiek, Ordained Women in the
Early Church [Baltimore, Md.: Johns Hopkins University Press, 2005], 196). De
Rossi thought that the Roman fossores were in some way members of the clergy,
but there is no convincing evidence to that effect.
6
Rebillard, Religion et spulture, 20.
246 Carolyn Osiek
still be in ground owned by generous Christian benefactors. Even the
involvement of Callistus in Christian burial practices during the epis-
copate of Zephyrinus could have been in land owned by Zephyrinuss
family and given over to community use.
7
Two centuries earlier, Cicero
had spoken of sepulchra communia when he must have meant burials of
those who held common blood ties and family relationships, not com-
mon ownership of burial ground (De Off. 1.17.5455).
But sometime by the early to mid-fourth century, the transition
from private to some kind of centralized church administration had
largely taken place, and the common assumption is that centralized
church ownership followed. How did these burial complexes evolve
from private property to massive common cemeteries, and under what
legal auspices? How did ownership shift from private to commu-
nal? What does communal mean here central administration by a
single bishop and his staff, or administration of each burial complex
by a specific titulus church center? What was the role of fami-
ly members and patrons in this process? Is there any special connec-
tion that can be traced between burial patronage and the patronage
and euergetism of women? These questions are the focus of the pres-
ent study.
Christian Burial Areas as Private Property
From early on, legislation forbid burials within the city walls of Rome.
The first law, in 451 B.C.E., was renewed by Augustus in the Lex Julia.
There were a few exceptions in extraordinary circumstances, e.g., Ju-
lius Caesars body was brought in and burned at the base of the Temple
of Castor and Pollux in the Forum in 44 B.C.E., and Trajans ashes
were buried at the base of the column in his forum in 117 C.E. (but
Hadrian had to get a senatus consultum to do it). In Roman religion,
a corpse must be hidden from the light of day, with dire consequences for
the one who violates this principle: whoever reveals to the sun a buried
body, si honestior sit, in insulam, si humilior, in metallum dari solet
(Sent. Paul. 1.21.4). A funerary monument, once a body had been
placed in it, became a locus religiosus protected by law, not the body
7
Suggested by Bradshaw et al., Apostolic Tradition, 191; Peter Lampe, From Paul
to Valentinus: Christians at Rome in the First Two Centuries (Minneapolis,
Minn.: Fortress, 2003), 2528, 36972.
Roman and Christian Burial 247
itself, but the place because of the presence of the corpse and the wider
associations of contact with divinity. Because of their religious nature,
the sale, construction, and repair of tombs were regulated by the pon-
tiffs, even well into the Constantinian era, as late as 385.
8
Roman law recognized two kinds of tombs, those for ones familia
sepulchra familiaria and those for ones heirs and other agnate kin
sepulchra hereditaria. The first kind was built by a householder for him/
herself and the members of the familia, which included blood family,
freedmen/women, slaves, and others attached to the household. The
second type excluded non-related household members and was exclus-
ively for the use of agnates and potential heirs, whether descendants or
otherwise. At the end of the line of inheritance, the last successors could
continue to designate others. In either case, the founder of a tomb had
a right to specify both inclusively and exclusively exceptions to the nor-
mal pattern, e.g., CIL 6.11027 and 14672, which exclude a specific
freedman (One wonders what stories are behind them!), or a third-
century hypogeum in the catacomb of Domitilla in which the founder
of a cubiculum says that he set up the tomb sibi et suis fidentibus
in Domino, probably thereby restricting burials to family members
who were Christian.
9
Such specific exclusions had no legal force, but
the one who set up the restrictions would have hoped that his/her
wishes were followed.
The overwhelming number of surviving tombs in Italy are of the
familial kind, intended for members of a household, not a vertical
family, with the familiar phrase sibi et suis, libertis libertabusque pos-
terisque eorum. Often the inscription actually forbids the burial of
heirs outside the familia with the abbreviation HMHNS: Hoc monu-
mentum heredem non sequetur. Many tomb inscriptions forbid the
burial of anyone not specified, or alienation of the tomb from the fam-
ilia, with threats of legal sanction and fines. In the Christian era, this
fear of tomb misuse sometimes takes on eschatological fervor: the
deacon Tetradia in Byzantine Thessaly threatens anyone who opens
her tomb with the punishment of eternal fire, while the deaconess
Athanasia in fifth-century Delphi threatens anyone who dares open
8
Digest 11.7.2.5; Cod. Theod. 9.17.2; Symmachus Epistle 2.36; see Mark J.
Johnson, Pagan-Christian Burial Practices of the Fourth Century: Shared
Tombs? JECS 5, no. 1 (1997): 3759, at p. 39.
9
ILCV 1.307, no. 1597; De Visscher, Droit des tombeaux, 9697; Johnson,
Pagan-Christian Burial Practices, 40.
248 Carolyn Osiek
her tomb, may he share the lot of Judas the [betrayer] of our Lord
Jesus Christ.
10
The familial tomb that gathered together members of a household in
several generations was a normal exercise of patronage for heads of
households. Besides members of the familia, others could be specifi-
cally included; thus the exercise of patronage could extend beyond the
household. A tomb could also be owned by more than one private per-
son; several owners could together hold the property and the right of
burial. In this case, all the owners had to agree on the acceptance of
anyone for burial in their mutually owned tomb.
Even in the case of burial areas later associated with the catacombs,
it can be assumed that the earliest burials were above ground, perhaps
consisting of a walled-in piece of property in which both incinerations
and inhumations could be placed directly in the ground. When more
burial space was needed, underground areas were excavated under the
surface property. The simplest inhumations were of bodies placed into
the ground with tiles covering the grave in an inverted V pattern, or of
incinerated ashes in urns that were buried up to their necks and filled
in with earth or sand. Such burials were still to be seen at the cemetery
of Isola Sacra in the 1970s, but have since disappeared (Fig. 7.1).
For the more well to do, a mausoleum surrounded by open prop-
erty enclosed by a low wall allowed for burials both inside the mau-
soleum and of dependents in the open ground. In some cases, the
mausoleum itself covered the entire burial space. Such dedicated
spaces can still be seen, for example, in the cemeteries lining the ap-
proaches to Pompeii (Fig. 7.2).
When more space was required, digging went underground within
the private property and then extended beyond it. This kind of
underground extension can be seen, for example, in the three private
hypogea that are today within the catacomb of S. Sebastiano, or
the Hypogeum of the Aurelii on Viale Manzoni in Rome. Laws re-
garding ownership of space below ground are not clear, but the prin-
ciple of superficies solo cedit may have applied: whoever owned an
10
Tetradia: N. I. Giannapoulos, Palaiochristianike epigraphe, Epeterias etaireias
Byzantinon spoudon 12 (1935): 26; Athanasia: J. Laurent, Delphes chrtien,
BCH 23 (1899): 20679, at pp. 27278; G. H. R. Horsley, ed., New Documents
Illustrating Early Christianity (9 vols.; North Ryde, NSW: Ancient History
Documentary Centre, Macquarie University/Grand Rapids: Eerdmans, 1981),
4.122.3, p. 240; Madigan and Osiek, Ordained Women, 7273, 9192.
Roman and Christian Burial 249
upper story of a building owned the whole building and the land on
which it stood.
11
Surface burials continued where possible, even while underground
burial areas were expanding, and even after martyrs tombs had
been established below ground, which, in the fourth century, encouraged
further burials as nearby as possible. Surface burial was less expensive
if one already owned the property, and more easily accessible. Unfor-
tunately, these aboveground burials did not stand as well the test of
time. In the land surrounding Rome, the surface has been so disturbed
over the centuries that little of the aboveground burial area is left intact
except in the case of stone mausolea that are still standing, and they, of
course, no longer hold intact contents. During excavation of the cata-
11
Suggested during conference discussion by Andrew Wallace-Hadrill, based on
information given in Felix Pirson, Mietwohnungen in Pompeji und Herkulaneum:
Untersuchungen zur Architektur, zum Wohnen und zur Sozial- und Wirtschafts-
geschichte der Vesuvstdte (Munich: F. Pfeil, 1999), 6869, citing relevant passage
in Labeo, Dig. 43.17.3.7. See also discussion of 43.17.3.4 on the roots of vines,
discussed by John Bodel in this volume.
7.1. Isola Sacra Cemetery, surface burials in 1973. (authors photo)
250 Carolyn Osiek
combs, these surface areas were neglected in favor of below ground,
where the state of preservation was greater.
One aboveground plot that did survive until excavation in 1960 was
an original area of the catacomb of Domitilla, the so-called Praedium
Domitillae, an aboveground area of sixty-two by seventeen meters sur-
rounded by a wall of opus reticulatum, where burial activity began in
the Julian or Augustan period. Ownership of the field by one or a suc-
cession of the Flaviae Domitillae was attested by inscriptions found
there: CIL 6.948, 949, 8942, and 16246 (the last a grant of funerary
land to P. Calvisius from Flavia Domitilla, thus indication of extended
relationships), discovered below ground, having fallen through from
the surface. A large mausoleum was placed there in the second century,
and the surface cemetery continued in heavy use through the fourth
century. When subterranean burials began, the staircase built to ac-
commodate them disturbed some of the surface burials, mausolea, and
7.2. Familial compound, cemetery outside
Pompeii. (Photo: L. Brink, O.P.)
Roman and Christian Burial 251
columbaria.
12
Presumably by that time, the disturbed surface burials
were from past generations and no longer maintained.
The burial complex eventually grew so large that it was originally
thought to be part of the catacomb of Callistus. Its independent exist-
ence was clarified by de Rossi in 1852. Other original areas later ab-
sorbed into the catacomb of Domitilla include two independent hypo-
gea built before the end of the second century, the so-called Hypogeum
of the Flavii Aurelii A to the south and of the Flavii Aurelii B
forty meters north of it, discovered by de Rossi in 1864. The preserved
underground cubicula were excavated below aboveground burial com-
plexes that are no longer extant. The evidence for Christian origins of
the two complexes is scant: only one inscription, and that not until the
third century.
13
Once horizontal underground extension began, the two
were quickly joined, and by the fourth century they were linked to the
larger complex and the underground basilica of Saints Nereus and
Achilleus. The third-century non-Christian Hypogeum of Ampliatus
had similar beginnings and shared the same eventual fate. Expansion
of the complex by the end of the third century made access to these
hypogea difficult.
Below the floor surface of the subterranean basilica of Saints Ne-
reus and Achilleus are several pagan sarcophagi in reuse. The usual ar-
gument is that they were robbed from older mausolea and moved
there, but in fact they may be reused in situ, where they were originally
placed in an underground extension of an aboveground mausoleum,
a single burial unit consisting of three sarcophagi and six formae
(spaces for one body dug directly in the floor of the chamber). If this is
the case, it indicates the presence of another private burial complex
above ground, destroyed when the basilica was built through the same
space.
14
12
The area was excavated by Antonio Ferrua in 1960, then covered up again: Phil-
ippe Pergola, Il praedium Domitillae sulla via Ardeatina: analisi storico-to-
pografica delle testimonianze pagane fino alla met del III sec. d. C., RACrist
55 (1979): 31335, at 31824.
13
Philippe Pergola, La region dite des Flavii Aurelii dans la catacombe de
Domitille, MEFRA 95 (1983): 183248.
14
Argued by Philippe Pergola, Les sarcophages paens remploys dans la basil-
ique des Sts. Nre et Achilles dans la catacombe de Domitille Rome: R-
flexions autour dune pratique, Historiam pictura refert: Miscellanea in onore
di Padre Alejandro Recio Veganzones, O.F.M. (Vatican City: Pontificio Istituto di
Archeologia Cristiana, 1994), 43950.
252 Carolyn Osiek
There were many women named Flavia Domitilla in the Flavian
family. Cassius Dio recounts the ill fate of one, the first-century wife of
Flavius Clemens, exiled to the island of Pandateria after the execution
of her husband for judaizing (67.14). By the fourth century, her ju-
daizing was understood as conversion to Christianity, which could
well have been confused with Judaism by a first- or second-century
Roman. For Eusebius, however, she had become the niece, not wife, of
Flavius Clemens, exiled as a Christian to the island of Ponza (Hist.
eccl. 3.18.3), where the late-fourth-century Paula of Rome went on
pilgrimage to visit her cell, as to the sanctuary of a saint (Jerome,
Ep. 108.7). Whether these were one or two women, whether attracted
to Judaism or to Christianity, has never been clarified, nor whether
the inscriptions associated with their name and found in the catacomb
of Domitilla referred to either of them or to other women of the same
name.
15
The catacomb of Priscilla on the Via Salaria poses no fewer chal-
lenges. As with the Praedium Domitillae, here too there were many
burials on the surface, neglected in excavations in favor of the cata-
comb complex. Several originally independent components formed
part of it: a nymphaeum, a cryptoporticus, the Greek Chapel (Ca-
pella Greca) with its unique artwork, and the so-called Hypogeum of
the Acilii. By the fourth century, a martyr cult of Saints Felix and
Philip made the catacomb a popular place for burials. The Hypogeum
of the Acilii, so named by de Rossi, was originally a cistern on the east
side of the complex. Only two inscriptions related to the Acilius family
were found there, and they may have fallen through from the surface.
One of them connects the names M. Acilius and Priscilla. Some other
inscriptions in Greek (Art-, Artto, Artto) were found else-
where nearby in the complex, and could be freedpersons of the same
family. The hypogeum was not originally connected to other galleries,
15
When Dio Cassiuss Flavia Domitilla was exiled, her land was presumably con-
fiscated. Therefore, any continuity of the Praedium Domitillae may be in ques-
tion. Most modern historians have assumed Eusebius was confused and turned
the wife into the niece. Among the few who argue otherwise is Paul Kerestzes,
The Jews, the Christians and Emperor Domitian, Vigiliae Christianae 27
(1973): 128. The whole episode under Domitian may have been for political
rather than religious reasons. For the entire history and a review of scholarship,
also see Philippe Pergola, La condamnation des Flaviens chrtiens sous Domi-
tien: perscution religieuse ou repression charactre politique? MEFRA 90
(1978): 40723.
Roman and Christian Burial 253
though, but was a simple space designed to receive a large sarcopha-
gus, connected to ground level by an independent staircase.
Members of the Acilius Glabrio family were consuls in 91, 152, and
186, and an Acilius Rufus in 106. The wife of the consul of 152 was
thought to be Arria L f Plaria Vera Priscilla (PIR
2
A 1120; CIL
11.6333; ILCV 1073).
16
Thus a mid-second-century origin of part of
the area is possible. De Rossi went to great pains to connect the sena-
torial family Acilius with the origins of the complex, but as usual, pro-
vable connections are tenuous. Yet the name Priscilla persevered in
connection with the catacomb.
To the south of the Hypogeum of the Acilii lay the Cryptoporticus
and Capella Greca, probably neither originally intended as burial areas,
but part of a large villa overhead that was never excavated. Toward the
end of the second century or early in the third, the Capella Greca be-
came a private burial place, probably for residents of the villa or their
familia. Burials adjacent to residential areas seem never to have been a
problem outside the city walls. North of the Capella Greca and west
of the Acilii area later lay large burial galleries, called arenaria by de
Rossi. None of these areas was originally joined to any other, but at a
late date expansion of the burial areas brought them all together into
one large complex.
17
Another smaller and later burial complex on Via Ostiense is less well
known, the catacomb of Commodilla. Discovered in 1595, it was
thought by Bosio to be the Crypt of Lucina, mentioned in early texts
and later discovered in the Callistus complex, about a mile further
east. There is no evidence of use before the fourth century. Again, ap-
parently no attention was paid to the terrain above ground during ex-
cavation. In the mid-fourth century, Pope Damasus discovered there
the burial of two martyrs, Saints Felix and Adauctus, probably killed
under Diocletian and buried in a privately owned area. The Depositio
martyrum of 354 does not mention the name, but the fifth-century
Martyrologium Hieronymianum commemorates Roma via Ostiensi
(in cimiterio) Commodellae Felicis et Adaucti. At first objects of only a
private cult, they had caused the area to become another favorite
16
See also PIR
2
P 950; CIL 6.31681; ILCV 127; ICUR 24837.
17
P.-A. Fvrier, tudes sur les catacombes romaines, CahArch 10 (1959): 126;
11 (1960): 114. For de Rossis account of discovery and discussion of the Acilius
Glabrio-Priscilla connection, Lipogeo degli Acilii Glabrioni nel cimitero di
Priscilla, BACrist ser. 4.6 (18881889): 1566, 10333.
254 Carolyn Osiek
burial zone by the mid-fourth century. The later account of two female
martyrs, the Passion of Saints Digna and Merita, places their burial in
coemeterio Commodillae eadem via (Ostiense). The two cults in the
same cemetery therefore would seem to be independent.
An unusual feature of this catacomb is its large collective pit or shaft
tombs dug beneath galleries, lined with loculi along the sides, and when
they were full, stacked with other burials separated with tiles. This is
thought to indicate a larger concentration of poorer people buried
there who could not afford anything more individual. The relative lack
of painting and the many uninscribed tombs also support the impres-
sion of large numbers of burials from modest circumstances. Burials
continued into the late fourth and early fifth centuries.
18
This is prob-
ably an example of a catacomb with a single origin that was expanded
as the need grew. The identity of the eponymous founder is unknown,
but she is likely to have been a mid- to late-fourth-century benefactor
who developed an extensive burial area for the poor, beginning from a
private complex, by acquiring the bodies of martyrs and then expan-
ding the cemetery as needed. This would not be unusual even in a pre-
Christian context: the custom of wealthy urban dwellers exercising
benefaction by donating land outside the city for burial of the general
population dates to Republican times.
19
Similar origins can be traced for other catacombs. For example,
three original centers formed the nucleus of the catacomb of Callistus.
Two are known by womens names: the crypts of Lucina and of
Balbina. The third center is probably that witnessed by Hippolytus, a
complex that, even if it was originally in private ownership, perhaps
18
B. Bagatti, Il cimitero di Commodilla o dei martiri Felice ed Adautto presso la via
Ostiense (Vatican City, 1936) 36; Carlo Carletti, Storia e topografia della ca-
tacomba di Commodilla, in Die Katakombe Commodilla. Repertorium der
Malereien (eds. J. G. Deckers, G. Mietke, A. Weiland; Vatican City: Pontificio
Istituto di Archeologia Cristiana 1994), 327; V. Fiocchi Nicolai et al., The
Christian Catacombs of Rome, 5456; photo of a shaft burial on p. 56, fig. 63.
19
Nicholas Purcell, Tomb and Suburb, in Rmische Grberstraen: Selbstdar-
stellung, Status, Standard (eds. Henner von Hesberg and Paul Zanker; Munich:
Bayerischen Akademie der Wissenschaften, 1987), 2541, cites the example
(pp. 3637) of Horatius Balbus of Sarsina who gave burial plots for the towns-
people at his own expense, excluding suicides by hanging and immoral earners
(ILS 7846), and a freedman who gave Tolentinum a plot of 282 by 200 feet for
the same purpose (ILS 7847; see also ILS 6726). Exactly what kind of people
were buried in such arrangements is not clear. Presumably, those who could
afford their own burials chose to go elsewhere.
Roman and Christian Burial 255
that of the family of Zephyrinus, was quickly turned into a common
burial area under the supervision of Callistus. It cannot have been
Callistuss own property: both the text of Hippolytus and the fact that
Callistus was not buried there contradict such a hypothesis.
Womens Patronage and Burial Euergetism
In the previous discussion of the origins of some of the major Roman
Christian catacombs, the names of a number of women have surfaced:
Domitilla, Priscilla, Commodilla, Lucina, and Balbina. There are
many other womens names associated with catacombs: Thecla, Bas-
illa, Agnes, Felicitas, and others. Of course there are also male names
associated with a variety of Roman catacombs: Callistus, Sebastian,
Calepodius, Trasone, Novatian, Hippolytus, Pamphilus, and others.
The remarkable thing about the names of the catacombs in early lists
of martyr burials is that they endured as the way to designate a par-
ticular catacomb, even when the catacomb was most famous as a place
of pilgrimage for its martyr tombs. The name of the catacomb of
Domitilla did not change to Saints Nereus and Achilleus, the cata-
comb of Priscilla did not change to Saints Felix and Philip, the cata-
comb of Commodilla was occasionally called that of Saints Felix and
Adauctus, but the name that stuck was that of Commodilla. Nor did
the original names associated with the catacombs acquire the title of
saint. Other catacombs did acquire the names of saints, the most fa-
mous being those of Saints Agnes and Lawrence, the most beloved
Roman martyrs of the early fourth century. Much later, of course,
some of the names connected with the catacombs did assume the title
of saint, when their urban churches became centers of pilgrimage, e.g.,
Santa Prisca on the Aventine (but was this supposed to be Prisca the
wife of Aquila or Priscilla of the catacomb, or another?).
The names connected to some of the principal Christian burial
places are early and persistent, and a good number of these names
are of women. We have seen that the most likely scenario for the begin-
nings of most of these burial complexes is a group of unrelated
walled-in surface burial areas containing inhumation and/or inciner-
ation graves. If set up by a wealthy benefactor, they would also prob-
ably contain mausolea. For want of space and because the popularity
of inhumation over incineration by the end of the second century cre-
ated the need for more ground and thus higher prices, the owners and
256 Carolyn Osiek
administrators of these burial spaces began to go underground below
their own property and eventually expand the underground area.
When the cult of martyrs tombs developed in the fourth century, the
desire of Christians to be buried as close as possible to the martyrs
created the need for extensive burial development, both above and
especially below ground. During this expansion, many previously in-
dependent private burial complexes became joined underground,
culminating by the late fourth and early fifth centuries in the vast laby-
rinths that the major Christian catacombs are today. This development
is not limited to Rome, but is found, among other places, in Syracuse,
Malta, and North Africa. John Bodel is correct to say in his chapter in
this book that Christians did not invent catacombs, a method of
burial also used by Jews and others, but the cult of the martyrs in the
fourth century caused the formation of the vast underground networks
that are the major Roman Christian catacombs, and these enormous
complexes are not equaled in use by any other group. In addition, be-
cause of the concentration of devotion in Rome and well-kept records
of martyrs tombs and their feast days, we have more information
about this phenomenon among Christians particularly in Rome than
anywhere else.
The historical and social link between the small private burial
groupings and the large communal complexes of the fourth century
is private patronage. This foundation of social relations in ancient
Roman society has been well studied, including the major cross-cul-
tural studies of Gellner and Waterbury, Eisenstadt and Roniger, and
Elliott,
20
and of its function in ancient Rome by Saller, Wallace-Ha-
drill, and others.
21
While most of the examples that have been preserved
are from elite classes, this is not true of all. The same social structures
20
Ernest Gellner and John Waterbury, eds., Patrons and Clients in Mediterranean
Societies (London: Duckworth, 1977); S. N. Eisenstadt and L. Roniger, Patrons,
Clients, and Friends: Interpersonal Relations and the Structure of Trust in Society
(Cambridge: Cambridge University Press, 1984); John H. Elliott, Patronage
and Clientage, in The Social Sciences and New Testament Interpretation (ed.
Richard Rohrbaugh; Peabody, Mass.: Hendrickson, 1996), 14456, with exten-
sive further bibliography.
21
Richard P. Saller, Personal Patronage under the Early Empire (Cambridge: Cam-
bridge University Press, 1982); Andrew Wallace-Hadrill, ed., Patronage in Ancient
Society (London: Routledge, 1989); Jens-Uwe Krause, Sptantike Patronats-
formen im Westen des Rmischen Reiches (Munich: C. H. Beck, 1987).
Roman and Christian Burial 257
can be expected to have been replicated as closely as possible at non-
elite levels.
Some of the actual persons involved in the creation of the Christian
burial sites may have been elites, such as a Flavia Domitilla or a Pris-
cilla married into the Acilii Glabriones. But whether elite or less than
elite, those Christians who functioned as patrons and clients presum-
ably replicated the same social structures in their own use of power to
achieve their ends. Patronage in early Christianity is now beginning to
be studied.
22
Women participated heavily in the patronage system on both sides,
as patrons and clients, and were deeply involved in both private and
public patronage. They could attend the morning salutatio of client
to patron (Juvenal, Sat. 1.12016). They participated fully in business
activities. Women who were sui iuris could conduct their own transac-
tions, though there were some legal limitations imposed. The earlier
institution of tutela, male guardianship requiring permission to alien-
ate property, was mostly inactive by the Augustan age, though former
owners could still exercise considerable control over the property of
a liberta. Other legislation prevented women from taking on liability
for the debts of others, which may have been primarily aimed to
protect women from unscrupulous husbands.
23
As is often the case
with Roman law, what is on the books is not necessarily what is done,
22
Beginning early with E. A. Judge, The Social Pattern of Christian Groups in the
First Century (London: Tyndale Press, 1960); The Early Christians as a Schol-
astic Community, Journal of Religious History (1960): 415; (1961): 12537;
Paul as a Radical Critic of Society, Interchange 16 (1974): 191203; Cultural
Conformity and Innovation in Paul: Some Clues from Contemporary Docu-
ments, Tyndale Bulletin 35 (1984): 324; continuing with Frederick W. Danker,
Benefactor: Epigraphic Study of a Graeco-Roman and New Testament Semantic
Field (St. Louis, Mo.: Clayton, 1982); more recently, John K. Chow, Patronage
and Power: A Study of Social Networks in Corinth (JSNTSup 75; Sheffield: Shef-
field Academic Press, 1992); David A. deSilva, Honor, Patronage, Kinship and
Purity: Unlocking New Testament Culture (Downers Grove, Ill.: InterVarsity
Press, 2000); Stephan J. Joubert, One Form of Social Exchange or Two? Euer-
getism, Patronage, and Testament Studies BTB 31 (2001): 1725; James R.
Harrison, Pauls Language of Grace in its Graeco-Roman Context (WUNT 2.172;
Tbingen: Mohr Siebeck, 2003); Zeba A. Crook, Reconceptualising Conversion:
Patronage, Loyalty, and Conversion in the Religions of the Ancient Mediterranean
(BZNW 130; Berlin: de Gruyter, 2004).
23
Jane F. Gardner, Women in Roman Law and Society (Bloomington: Indiana Uni-
versity Press, 1986), 23336.
258 Carolyn Osiek
and there were perhaps more exceptions than strict applications of the
law.
Women often served as patrons for other women. Cratia, the wife
of M. Cornelius Fronto, tutor of Marcus Aurelius, is called in one of
his letters to the emperor a clienta of Domitia Lucilla, the emperors
mother. As such, she visited the imperial family, staying with them
in Naples without her husband to celebrate her patrons birthday.
24
An otherwise unknown woman named Valatta on the British frontier
writes to the commanding officer of the Vindolanda outpost, Flavius
Cerialis, about a favor mediated by his wife, Sulpicia Lepidina.
25
The
epitaph of Epiphania, a second- or third-century benefactor, the well-
traveled daughter and wife of ship owners, reports that she was gener-
ous with her wealth, motivated by eusebeia, especially to abandoned
friends u yuv yuvoti, woman to women.
26
But womens patronage was not limited to women. Women could
not vote or hold elective office, yet the influence of powerful women
in the palace and the law court through their exercise of patronage,
amicitiae muliebres, was always present.
27
Roscius of Ameria, who
was later defended by Cicero in a parricide case involving political
machinations against Sulla, fled for protection in Rome to the aris-
tocrat Caecilia Metella, not to any of her abundant male relatives or
her husband, because of her amicitia with his deceased father. It
was recognized that she was his patron, not one of the male members
of her family.
28
Augustuss wife Livia had her own following and client loyalties; as a
widow, she even received the Senate in her house. Josephus recounts
her benefactions to the Herodian family, including marriage advice to
24
Marcus Cornelius Fronto (trans. C. R. Haines; vol. 1; LCL; Cambridge, Mass.:
Harvard University Press, 1962), 14551; Edward Champlin, Fronto and Anto-
nine Rome (Cambridge, Mass.: Harvard University Press, 1980), 25.
25
Alan K. Bowman and J. David Thomas, eds., Per Lepidinam: The Vindolanda
Writing-Tablets (Tabulae Vindolandenses II) (London: British Museum Press,
1994), no. 257 (inv. 85.117) 23031. The tablet is dated to period 3 of the fort,
97102/3 C.E..
26
NewDocs 2.16, pp. 5556.
27
Champlin, Fronto and Antonine Rome, 109, 171 n. 87; Richard A. Bauman,
Women and Politics in Ancient Rome (London: Routledge, 1992); Suzanne
Dixon, A Family Business: Womens Role in Patronage and Politics at
Rome 8044 B.C., Classica et Mediaevalia 34 (1983): 91124, at p. 91.
28
Ibid., 94, with other examples.
Roman and Christian Burial 259
Salome (Ant. 17.10).
29
Upon her death, the grateful Senate voted the
erection of an arch in her honor. This had never before been done for a
woman, and Tiberius would never allow it to be built. The senators
gratitude was abundant because she had saved the lives of some Senate
members, sustained their orphaned children, and helped many by pay-
ing their daughters dowries. She was so loved that she was titled infor-
mally, in parallel to Augustuss title, mater patriae. The title was never
officially granted to her, however, even after her death (Dio Cas-
sius 58.2.3).
30
Her activities on behalf of her clients are illustrative of
the kinds of benefactions expected from a powerful patron. Many
similar stories can be told about other elite women.
Antonia Caenis, freedwoman of Claudiuss mother Antonia, became
mistress of Vespasian until her death. Dio Cassius (65.14.15) gives
a vivid description of her patronal power and wealth: in exchange
for money, she granted various public offices and priesthoods, and
obtained imperial decisions and secured imperial pardons in favor of
her clients. Hers is an example of the power of women derived from
their association with male power, but it could also work the other
way. Juvenal complains of women who attend mixed dinner parties
and even host them, holding sway on politics and literature (Sat.
6.434456).
31
He also hints that the best way to social advancement
is through the patronage of some aging wealthy woman (Sat. 1.39).
Womens patronage was not limited to elites: a freedwoman named
Manlia T. l. Gnome boasts on her epitaph that she had many clients
(clientes habui multos; CIL 6.21975).
Womens patronage of groups and even cities is also well docu-
mented. For example, Euxenia, priestess of Aphrodite in Megalopolis
29
Josephus writes of other benefactions to the Herodians from imperial women,
Antonia and Agrippina the Younger. Poppaea Sabina, wife of Nero, was also said
to be mediator for Jewish causes (Ant. 18.143, 164; 20.13536; 20.18996; Life
1316). In a typical patronage maneuver, Josephus records that at Puteoli he
met an actor named Aliturus, and through him, was introduced to Poppaea. Domi-
tia, wife of Domitian, was also a personal benefactor and defender of
Josephus, toward whom she was euergetousa, benefactor (Life 429). See Shelly
Matthews, First Converts: Rich Pagan Women and the Rhetoric of Mission in Early
Judaism and Christianity (Stanford, Calif.: Stanford University Press, 2001), 3036.
30
Bauman, Women and Politics, 12429. Livias power was derived from that of
Augustus, but, like many queens and empresses, while she had it, she exercised it
quite independently.
31
See Suzanne Dixon, Reading Roman Women: Sources, Genres, and Real Life
(London: Duckworth, 2001), 101.
260 Carolyn Osiek
in the Peloponnesus in the second century B.C.E. donated a guest-
house and a wall around the temple (IG 5.2.461).
32
Plankia Magna of Perge, who held titles of ojtouyo and yuvo-
otoo in the second century C.E. both erected and had erected to her
several monuments in her city commemorating her benefactions.
33
Tation, daughter of Straton son of Empedon, from Kyme either built or
remodeled at her own expense the building and the surrounding precinct
of a synagogue, for which the Jews bestowed on her two traditional hon-
ors for a patron: a gold crown and a place of honor (oroto). Both the
wording of the inscription (the Jews honor her) and the family names
involved suggest that she was an outside benefactor, not a member of the
Jewish community (CIJ 2.738).
34
Julia Severa of Acmonia in Phrygia
held a number of distinguished priesthoods and city offices and was of a
family that was sufficiently prominent that her son entered the Senate,
yet she donated property to the local synagogue, perhaps because two of
its archons were her freedmen or clients (CIJ 2.766; MAMA 6.264).
35
Other civic benefactions in which women were involved included ali-
mentary programs for poor children. Wealthy women found this an ap-
propriate outlet for their money and a suitable way to be immortalized.
And immortalized they were. In Herculaneum, where the hardened
mud that covered the city made immediate retrieval of precious items
much more difficult than at Pompeii, more statuary was thus preserved
than at Pompeii. At Herculaneum, 40 percent of the dedicatory statues
are of women, mostly large and in bronze and metal. They were set up
alongside those of men in the theater and the forum area, without any
perceptible gender pattern.
36
32
Riet van Bremen, Women and Wealth, in Images of Women in Antiquity (eds.
Averil Cameron and Amlie Kuhrt; rev. ed.; Detroit, Mich.: Wayne State Univer-
sity Press, 1993), 223.
33
Cf. Mary Taliaferro Boatwright, Plancia Magna of Perge: Womens Roles and
Status in Roman Asia Minor, in Womens History and Ancient History (ed. Sarah
B. Pomeroy; Chapel Hill: University of North Carolina Press, 1991), 24972.
Other female gymnasiarchs are known: L. Casarico, Donne ginnasiarco, ZPE
48 (1982): 11822. There is even one in Egypt, and a female tax collector: New-
Docs 8.4, p. 49.
34
NewDocs 1.69, p. 111.
35
L. Michael White, ed., Social Networks in the Early Christian Environment: Issues
and Methods for Social History (Semeia 56; Atlanta, Ga.: Scholars Press, 1992),
1819.
36
Giuseppina Cerulli Irelli, Archaeological Research in the Area of Vesuvius:
Portraits from Herculaneum, in Pompeii and the Vesuvian Landscape (Washing-
Roman and Christian Burial 261
Women participated in and led collegia or 0iooot, professional and
social organizations of the nonelite, often patronized by an elite figure.
They held such offices as magistra, quinquennalis, sacerdos, curator,
honorata, quaestor, decurio, and immunis.
37
A well-known Pompeiian
benefactor who illustrates precisely the kind of patronage central to
this study was Eumachia, public priestess, patroness of the fullers
guild, who, in her own name and that of her son, Numistrius Fronto,
erected at her own expense a gallery, cryptoporticus, and portico for
the fullers building, which was centrally located in the forum, dedicat-
ing them herself to concordia and pietas augusta. In gratitude, the guild
erected a dedicatory statue of her with inscription, a copy of which still
stands behind their building at the forum in Pompeii (Fig. 7.3). She also
built a tomb for herself and her familia outside the city. Eumachias
tomb outside the Nucerian Gate is one of the largest funerary monu-
ments in the area, stating simply her name and filiation on one side, EV-
MACHIA L F, and on the other side, SIBI ET SVIS, for herself and
those who belong to her familia (CIL 10.810, 811, 813 and Fig. 7.4).
38
Similarly, the Jewish Rufina in second-century Smyrna, a woman
who bore the title 0touvoyuyo (synagogue ruler), recorded on a
marble plaque on her tomb that she built it for her freedmen/women
and the slaves raised in her house (tot 0rru0rot o 0rootv)
(CIJ 741). It was familial burial complexes like theirs that formed the
nuclei of most of the later Christian catacombs, as discussed above.
Many more examples could be added. This kind of evidence is im-
portant for seeing the wide range of possibilities for womens personal
patronage. Any woman who had accumulated even a modest amount
of wealth and connections could be as active in patronage relation-
ships as a man of her social status. For elite women, direct intervention
ton, D.C.: Archaeological Institute of America and the Smithsonian Institution,
1979), 1624; Caroline Dexter, The Epigraphic Evidence of Pompeiian
Women (unpublished paper) n. 18, p. 23.
37
List in J. P. Waltzing, tude historique sur les corporations professionnelles chez
les Romains depuis les origines jusqu la chute de lEmpire dOccident, vol. 4
(4 vols.; Brussels: Hayez, 1895), 25457.
38
Cf. Roy Bowen Ward, The Public Priestesses of Pompeii, in The Early Church
in Its Context: Essays in Honor of Everett Ferguson (eds. Abraham J. Malherbe,
Frederick W. Norris, and James W. Thompson; Leiden: Brill, 1998), 32327.
There is some doubt whether the building fronting on the forum was actually the
fullers meeting place, but the inscription on the frieze and the statue behind
commemorate Eumachia as stated.
262 Carolyn Osiek
7.3. Copy of statue of Eumachia
(original in Naples Museum)
behind the fullers building on
the Forum in Pompeii. (authors
photo)
7.4. Eumachias burial complex for her familia outside the Nucerian Gate, Pompeii.
(Photo: David Balch)
Roman and Christian Burial 263
in political appointments was lacking, but they exercised no less in-
fluence. One of the expected actions of such patrons was to set up a
burial place to accommodate not only themselves and immediate
family, but also members of the network of their familia, as Eumachia
and Rufina did.
39
Apparently, this first step in Christian burial practice continued into
a second phase: while the land was still in private ownership, the exer-
cise of private patronage by Christians broadened to include provision
of burial space on ones own cemetery plot to members of the church
not related by familia and without suitable alternative burial arrange-
ments. Some second- and third-century texts tell us of the developing
understanding of patronage among Christians, and their sometime re-
sistance. For example, Hermas criticizes the wealthy of his second-cen-
tury Roman community for shirking patronal duties: they get so tied
up in their business interests that they avoid lesser persons because
they do not wish to be asked for favors (Herm. Sim. 9.20.24). Such
people would incur the disdain not only of Hermas but of the Chris-
tian poor as well, to say nothing of their peers. Their repentance will
consist in doing some good, namely, generosity with their riches and
the establishment of patronage relationships. One of those good ac-
tions was granting place in their burial properties for the poor.
Other later writers under a growing church centralization are not so
encouraging. The Apostolic Tradition of Hippolytus discourages indi-
viduals from holding charity meals for the needy without clerical
supervision (Apostolic Tradition 29).
40
If this text does represent early-
third-century Rome as tradition would have it,
41
it stands at just the
time when central control is being exerted in the church in many areas.
In Carthage a generation later, Cyprian, probably like most bishops of
39
An interesting documented parallel to this development is the aggregation of
ownership of brick factories in the hands of daughters through inheritance
over the first two centuries C.E. before passing into imperial control in the third
century: Pivi Setl, Private Domini in Roman Brick Stamps of the Empire:
A Historical and Prosopographical Study of Landowners in the District of Rome
(Annales Academicae Scientiarum Fennicae. Dissertationes Humanarum Litte-
rarum; Helsinki: Suomalainen Tiedeakatemia, 1977). My gratitude to John
Bodel for this reference as for many other insights in this paper.
40
Charles A. Bobertz, The Role of Patron in the Cena Dominica of Hippolytus
Apostolic Tradition, JTS 44 (1993): 17084.
41
The origins and nature of the text are much disputed: see Bradshaw et al., Apos-
tolic Tradition, 16.
264 Carolyn Osiek
his time, wanted to consolidate patronal power in his own office by
weakening the power of wealthy members of the church, encouraging
centralized charity, and rejecting the charismatic claims of martyrs to
forgive sins. The consolidation of collection and dissemination for
charity, already evidenced by Justin and Tertullian, gradually becomes
the normal way for Christians to exercise their generosity. By then, the
patronage system has been vastly overhauled. Eventually, there is only
one major patron left for Christians: the bishop.
42
During this phase,
assumption of ownership and administration of cemeteries by the
churches under the direction of the bishop was a natural extension of
their assertion of centralized control.
From Private to Communal Ownership
At some time, in most cases during the third and early fourth century,
while still under private ownership, the underground areas of these
burial plots were extended, at first by the owners of the plots in order
to make room for burials of the poor within their own spaces. This was
the prelude to the assumption of ownership and administration by
church authorities. But how did this transition happen legally, and
what was the legal basis for church ownership?
It is well known that Roman law did not recognize corporate owner-
ship by a legal body, though it did allow multiple individual ownership
of everything from land to slaves.
43
In practice, there seems to have
been tacit recognition of corporate ownership. For example, there are
documented cases in which familial burial property is ceded as a foun-
dation to a city, colony, or collegium, apparently only when the entire
familia no longer had any survivors.
44
This implicit recognition of corporate ownership is examined in
more detail in J. P. Waltzings monumental study of professional cor-
42
William L. Countryman, The Rich Christian in the Church of the Early Empire:
Contradictions and Accommodations (New York: Edwin Mellen, 1980).
43
De Visscher, Le droit des tombeaux romains, 12327 discusses this and cites
Cicero De officiis 1.17.5455 as if Cicero were talking about corporate property,
but that need not be the case (see p. 246 above).
44
De Visscher, Le rgime juridique des plus anciens cimetires chrtiens Rome,
Analecta Bollandiana 69 (1951): 3954 (same text in Le droit des tombeaux ro-
mains, 26176), at 4449, citing the example of Junia Libertas. Some examples
given: ILS 7846, 7847.
Roman and Christian Burial 265
porations, or collegia. He cites four different ways in which Roman
legal practice implicitly recognized corporate ownership: (1) prop-
erty bequeathed to the state for accomplishment of a civil responsi-
bility (rare); (2) consecratio et dedicatio, goods belonging to a god;
(3) property held by a societas, legally understood as personal prop-
erty of many individuals; and (4) civil personification in jus privatum,
a more limited form of ownership than that of a real person.
45
The
Lex Julia of 7 C.E. required authorization of every legal collegium by
emperor or Senate, therefore presumably with this fourth form of
civil personification.
By the time of the jurists Paul and Ulpian (early third century), most
of the collegia were de facto, even if not legally, recognized as owners
of property, the legal basis being the logical extension from cities as
corporate persons to recognized organizations, which then were under
strict rules of registration and financial accountability.
46
Individuals
could donate property to collegia (e.g., CIL 6.10231; 10.444, 1579,
2112) and collegia could sell property and could reclaim the inherit-
ance of one of their freedmen (Ulpian Dig. 40.3.12), but could not
otherwise inherit by testament. Freedmen were obligated to bequeath
half their property to their patron. By the third quarter of the second
century, this probably applied to former slaves of collegia as well.
47
The evidence is not clear on the boundaries and particular charac-
teristics of collegia in the Roman world. Some certainly existed for
the purpose of mutual support in a common trade, like the fullers cor-
poration of Pompeii with their patron, Eumachia. Most probably
had some kind of religious practice associated with their meetings.
This kind of professional guild is known to have existed in most of the
Roman world, sometimes assembling in their own building, as did the
fullers of Pompeii, sometimes in a rented hall, sometimes in a private
house. In addition, it has been thought that another kind of associ-
ation of non-elites, the collegia tenuiorum composed largely of the free-
born poor, existed for the major purpose of burial societies, to assure
a decent burial of deceased members through monthly meals where
a collection was taken up for the common chest. But it is difficult to
imagine a social association with so limited a scope. Probably this is
simply another face of the same kind of social organization or club
45
Waltzing, tude historique, 2.3, pp. 43244.
46
Ibid., 43244, 47475.
47
Ibid., 45567.
266 Carolyn Osiek
that accomplished many purposes at once: social, professional, reli-
gious, and funerary.
48
The acquisition of one or more patrons to under-
write expenses and provide social status was always desirable. It has
been suggested that one of the advantages of such societies was to pool
resources and so attract patronage that individuals or smaller groups
would not be able to acquire.
49
Since de Rossi at the end of the nineteenth century, there has been
discussion of the possibility that early Christian communities may
have been associated in some way with this model of the non-elite
social club. Tertullian (Apol. 39) speaks of the monthly collection for
works of charity, and regular social meals. It is one thing to argue that
Christians deliberately used the legal form of the collegium and were
recognized as such by imperial authority; it is another to see Christians
using the collegium as a familiar form of social organization. The latter
is much more likely than the former: the idea of Christian groups as
officially recognized collegia has not been widely accepted, while they
certainly seem to share some characteristics with collegia, especially
regular common meals, a common social fund, and provision of burial
to those unable to afford it.
50
A parallel to Christian activity has sometimes been drawn with the
collegium domesticum that met in the house of Sergia Paulina in early-
second-century Rome, in which the surviving inscriptional evidence
speaks of the collegium quod est in domo Sergiae Paulinae, reminiscent of
Pauls greetings to tv ot` oiov o0tuv rjoiov (e.g., Rom16:5).
51
Besides meeting regularly in a house for an official meal, they also
seem to have had common burial space.
There is some evidence that Christians in various places continued
to belong to other collegia and were buried in their common burial
grounds. Cyprian in mid-third-century Carthage objects to the action
of bishop Martialis who allowed his son to be buried in the cemetery of
another (non-Christian) collegium (Ep. 67), presumably because he is
48
Rebillard, Religion et Spulture, 5253, 5759.
49
Purcell, Tomb and Suburb, 39.
50
For a summary of recent scholarship on this question, see Philip A. Harland,
Associations, Synagogues, and Congregations: Claiming a Place in Ancient Medi-
terranean Society (Minneapolis, Minn.: Fortress, 2003), 17882.
51
The proposition of Marta Sordi in 1971 that the group in the house of Sergia
Paulina was actually Christian has not been accepted; bibliography in Rebillard,
Religion et Spulture, 56 n14.
Roman and Christian Burial 267
a member. About the same time, Commodian condemns the decisions
of those who would be buried in such a way (Instr. 2,29,1213). Both
condemnations witness to the custom practiced by some, including a
bishop.
52
When finally Constantine in 321 established Christian churches as
full juridic persons for the purpose of inheritance, they became neces-
sarily full juridic persons for the purpose of ownership of property,
and from then on, there is no ambiguity in their legal status. If the shift
from private ownership of Christian burial properties to common
church ownership began already in the early third century, as appar-
ently witnessed by Hippolytus, then for nearly a century churches must
have shared the ambiguous position of the collegia, operating openly
but without proper juridic status as property owners. As is so often the
case in Roman law, magistrates seem to have looked the other way ex-
cept during times of actual persecution.
Even then, however, sources seem to indicate that church ownership
of real property was not questioned. The evidence is puzzling but
rather clear, at least as witnessed by fourth-century Christian sources.
For example, in the episcopal dispute with Paul of Samosata in the
270s, the deposed Paul refused to give up the oio tj rjoio,
which the emperor Aurelian decreed should belong (vrtot) to those
who were in communication with the bishops of Italy and Rome (Euse-
bius Hist. eccl. 7.30.1819). Again, Lactantius relates that during the
persecution of Diocletian and Galerius, officials came to the ekklesia
in Nicomedia, burned the books, and destroyed the interior of the
building. From their palace, the two emperors could see the building,
which they decided not to burn for fear other nearby buildings would
also catch fire. So the Pretorian Guards with axes and iron instruments
leveled the building in a few hours. (Mort. 12). In the account, no ques-
tion about legal ownership of the building is raised.
The collegia continued to exist well into the fourth century, even
though the church gradually took over the function of burial of the
poor. One section of the catacomb of Domitilla apparently belonged
to a collegium of mensores, an association responsible for procurement
and distribution of the regular dole of grain to the populace of Rome.
The corporations sometimes found new patrons in bishops, and thus
participated in the movement that was happening with regard to pri-
52
Rebillard, Religion et Spulture, 6467.
268 Carolyn Osiek
vate patronage generally: the assumption of patronal privilege into the
hands of church authorities.
53
This does not mean, however, that the practice of private patronage
in regard to cemeteries ceased entirely and immediately. Lactantius
notes that the last great work of piety is burial of strangers and the
poor (Inst. 6.12). Again, there is evidence of wealthy women patrons
who continue to function in this way. The small underground basilica
of Saints Felix and Adauctus in the catacomb of Commodilla contains
the grave of the widow Turtura, who was important enough to be im-
mortalized on a late large wall painting of the same basilica in the com-
pany of the two martyrs flanking the Virgin and Child.
54
Outside of Rome, an early-fourth-century inscription from a Chris-
tian cemetery at Velletri in Latium records a commemoration to Falto-
nia Hilaritas, dearest lady and daughter, who from the ground up made
the cemetery from her own funds and gave it to this religion (Faltoniae
Hilaritati dominae filiae carissimae quae hoc coemeterium a solo sua pe-
cunia fecit et huhic religioni donavit ILCV 3681A).
55
Seemingly, at the
time she died, the church was already in possession of what began as a
private patronal enterprise, exactly the same pattern that we have seen
began to be operative during the third century in Rome. At that earlier
time, private burial complexes that had probably been administered for
several Christian generations by patrons for the care of poor Christians
without private means of burial were eventually turned over to commu-
nal church ownership and administration by at least local leaders of par-
ticular congregations. By the fourth century, especially after Constan-
tines recognition of the legal status of the church, it became normative
for the church to assume ownership and administration of cemeteries,
increasingly under centralized administration of bishop and deacons. In
the case of Faltonia Hilaritas, the transition happened within one
generation: the woman who founded the cemetery as a private endeavor
was herself buried in the same cemetery, now owned by the church.
53
Ibid., 6870; Waltzing, tude historique, 46174.
54
Fiocchi Nicolai et al., Christian Catacombs of Rome, 63.
55
Religio originally meant an obligation having to do with the sacred, as for example
the tomb as locus religiosus. Freqently referred to is Ciceros distinction between
religio and superstitio and etymology of religio from relegere, to reread or review
(De natura deorum 2.28.7172). By the second and third centuries C.E. the term
had acquired connotations of faith practice. For other examples of Christian in-
scriptions using the word to refer to Christian faith and practice, see ILCV 1.3824,
3826 (CIL 6.10412; 10411). Cf. PW new ed., 21. 56575; DACL 14B. 229194.
Roman and Christian Burial 269
Once various church entities had taken over administration of the
cemeteries and social services in the fourth century, private patronage
did not cease. It did, however, begin to take on a different form. The
same social classes that were previously both public and private pa-
trons continued to be involved in benefactions, but now the bishop is
presented as principal initiator and patron of social projects. The re-
sources still come from lay patrons, but now they do not flaunt their
status on public monuments, as they did in the past. Rather, they as-
sume the position of humble servants of the church, relying on a future
reward.
56
Relief efforts for the poor and needy continue, but the re-
sources, supplied by lay patrons, are funneled more and more through
the hands of the bishop and his deacon assistants. Private benefaction
also now frequently takes the form of contributing to the embellish-
ment of churches or monuments, as witnessed, for example, on the in-
scriptions of an early-fifth-century altar and ciborium in the basilica of
St. Alexander bearing the names of Delicatus and the aristocratic
Junia Sabina c f (ICUR 8. 2295822959).
57
Examples like this abound.
The bishop assumes the position of major patron. Material help now
comes from the social aid programs administered by the bishop and dea-
cons. Moral and spiritual help, an actual spiritual patronage, resides in
the power of martyrs bodies in their memoria in the cemeteries, and in
the sacraments. The bishop and deacons are in control of both. These
factors help to explain the swift rise of the power of bishops beginning in
the middle of the third century and accelerated in the fourth.
58
56
Much as both ancient and modern writers, however, want to contrast the pre-
Christian patron, concerned only with the honor that would flow from his or her
generosity, with the Christian donor who selflessly gives without expecting any
earthly reward, the difference is not so neat. See, for example, Pliny the Younger,
who worries lest making known his generosity to an alimentary program might
be seen as self-aggrandizing. A noble spirit, he notes, seeks the reward of virtue,
not popular praise. Fame should be the result of good deeds, not the motive for
doing them, and there is no less merit if it does not follow (Ep. 1,8.13).
57
Vincenzio Fiocchi Nicolai, Euergetismo ecclesiastico e laico nelle iscrizioni
paleocristiane del Lazio, Historiam pictura refert: Miscellanea in onore di Padre
Alejandro Recio Veganzones O.F.M. (Vatican City: Pontificio Istituto de Arch-
eolgia Cristiana, 1994), 23752, at p. 240.
58
Y. Duval and Ch. Pietri, Euergtisme et pigraphie dans loccident chrtien
(IVeVie s.), Xe Congrs International dpigraphie Grecque et Latine, Nmes,
du 4 au 10 octobre 1992 (eds. Michel Christol and Olivier Masson; Paris: Publi-
cations de la Sorbonne, 1997), 37196. The amalgamation of patronal power in
the hands of a bishop is an interesting parallel to the role of the emperor, holder
270 Carolyn Osiek
Conclusion
The Greco-Roman system of patronage and benefaction, the back-
bone of social relations in the earlier Empire, did not cease among
Christians when they were a minority in the non-elite population, nor
when they gained social and political ascendance. One of the areas in
which they have left some evidence of their activity is the provision of
burial for their dependents, and developing from there, for other needy
persons in the community. The prominence of women in the Greco-
Roman system of patronage and benefaction is clear to those who look
at the evidence. Women were prominent, too, in Christian patronage.
One of the major ways in which we see this is the development of burial
complexes in Rome. The prominence of Christian women in this par-
ticular exercise of patronage is indicative of the significant numbers of
women who owned land and were in the position of head of household
with responsibility to provide burials for the familia, which then ex-
tended to others, especially the needy members of the church.
Gradually, however, church leadership was expanding its adminis-
tration and ownership of these properties, initially under questionable
legal arrangements, but from the time of Constantine, with full legal
power. During this development, church leaders, especially bishops,
emerge as the prominent element in the patronage system, while
private patrons become the background suppliers of materials. Both
women and men are among the private patrons who retreat to the
shadows in the new Christian order in which the right hand should not
know what the left hand is doing (Matt 6:14), the honor due to the
patron goes to the bishop, and for all others, should be reserved for the
life to come.
of supreme and universal patronal power, as argued by Fergus Millar, The Em-
peror in the Roman World (31 BCAD 337) (Ithaca, N.Y.: Cornell University
Press, 1977); cf. Richard P. Saller, Personal Patronage, 23; Andrew Wallace-Ha-
drill, Patronage in Ancient Society, 7981.
Envisioning Context and Meaning
273
David L. Balch
Chapter 8
From Endymion in
Roman Domus to Jonah in Christian Catacombs:
From Houses of the Living to Houses for the Dead.
Iconography and Religion in Transition
Pre-Constantinian Christians valued the three-stage Jonah cycle. Gray-
don Snyder lists how often thirty-one different Biblical stories are vis-
ually represented this early: Jonah-cast-into-the-sea appears 38 times,
Jonah-and-the-ketos 28 times, and Jonah-at-rest 42 times in various
media, including mosaics, wall paintings, and sarcophagi. Of the other
twenty-eight visual images found, the closest is the Sacrifice of Isaac,
which appears 8 times.
1
Christian artists emphasized the third Jonah
image, Jonah-at-rest, and represented him in the visual tradition of
Endymion, a handsome young man loved by the goddess Selene (the
moon). Endymion appears 17 times in wall frescoes in Pompeii,
2
in
Roman houses dating from the first century C.E. This essay concerns
how visual representations of Endymion made the journey from mosaics
and wall frescoes in Roman houses and tombs of the first and second
centuries C.E. to represent Jonah in Christian catacombs of the third to
fifth centuries C.E., then again moved above ground to one of the earliest
church mosaics that remains in Aquileia, in the early fourth century
C.E. Although many authors assume that Jonah is represented in the
style of Endymion, two important writers deny that Jonah images are de-
pendent on him,
3
so this essay will also inquire whether they are correct.
1
Graydon F. Snyder, Ante-Pacem. Archaeological Evidence of Church Life before
Constantine. (Rev. ed.; Macon, Ga.: Mercer University, 2003), 87.
2
PPM 10:560 (Index, s.v Endymion).
3
Hans Gabelmann, Endymion, LIMC 3.1: 72642, at 742, and Antonio Fer-
rua, Paralipomeni di Giona, in RACrist 38 (1962): 769.
274 David L. Balch
The question of method is unavoidable. One problem is that only
about 40 percent of the catacomb paintings have recently been ad-
equately published, a contrast to research on sarcophagi and inscrip-
tions.
4
Further, some earlier interpreters proceeded from the paintings
directly to Christian dogma,
5
emphasizing either their sacramental or
their eschatological meaning. The German reaction has been to study
the catacomb paintings simply as late Roman folk art and to deny the
appropriateness of interpreting the visual in light of the textual, that is,
by Patristic authors.
6
One advantage of this debate is that Early
Christian art seems to have freed itself from unambiguous interpre-
tations that united the various figurative models using a connecting
thread that related them all to a single fundamental reference.
7
This
methodological fissure means that scholars must decide whether to be
Italian or German.
4
Norbert Zimmermann, Werkstattgruppen rmischer Katakombenmalerei (JAC,
Ergnzungsband 35; Mnster: Aschendorf, 2002), 27, 29.
5
Andr Grabar, Christian Iconography: A Study of Its Origins (Bollingen
Series 35.10; Princeton N. J.: Princeton University, 1968), Parts 35: Dogma.
Josef Wilpert, Le pitture delle catacombe romane (Roma sotterranea; Rome:
Descle Lefebvre, 1903), German title: Roma sotteranea: die Malerein der Kata-
komben Roms.
6
Zimmermann, Katakombenmalerei, 29, citing Peter Dckers, Agape und Irene:
die Frauengestalten der Sigmamahlszenen mit antiken Inschriften in der Kata-
kombe der Heiligen Marcellinus und Petrus in Rom, JAC 35 (1992): 14767 and
Carlo Carletti, Origine, committenze e fruizione delle scene bibliche nella pro-
duzione romana dell III secolo, Vetera Christianorum 26 (1989): 20719. Also
Josef Engemann, Biblische Themen im Bereich der frhchristlichen Kunst,
Stimuli: Exegese und ihre Hermeneutik in Antike und Christentum: Festschrift fr
Ernst Dassmann (eds. Georg Schllgen and Clemens Scholten; JAC, Ergn-
zungsband 23; Mnster: Aschendorff, 1996), 54356. Carlettis observation
(21215) that the Biblical scenes occur in family burial chambers reserved for the
elite renders problematic Snyders hypothesis (see n. 1) that these images reflect
values of the non-elite.
7
Fabrizio Bisconti, The Decoration of Roman Catacombs, in The Christian
Catacombs of Rome: History, Decoration, Inscriptions (eds. Vincenzo Fiocchi
Nicolai, Fabrizio Bisconti, Danilo Mazzoleni; 2nd ed.; Regensburg: Schnell &
Steiner, 2002), 133.
275
Selected Second-Century-C.E. Roman Sepulchral
and Domestic Mosaics of Endymion and Third-Century
Sarcophagi of Jonah
First, the Hellenistic and Roman biography of Endymion: he is said
to have led the Aetolians from Thessaly in northern Greece to Elis, a
plain in the Peloponnese.
8
Early sources do not narrate Selenes (the
moon goddess) love of him, an addition to the saga made in Asia
Minor. When describing the Ionians and Carians near Miletus, Strabo
(XIV1.8; C636) mentions mount Latmus and a river nearby, where one
may see the sepulchre of Endymion in a cave, which Selene (=Trivia)
left her cosmic course to visit (Catullus 66.56, alluding to Callimac-
hus;
9
also Lucian
10
). Hellenistic authors expand the bucolic imagery:
Endymion is a shepherd whom Selene comes down from Olympus to
kiss, which leads to the wish, O would I were Endymion That sleeps
the unchanging slumber on (Theocritus, 3.4950 [Edmonds,
LCL]).
11
Propertius emphasizes Endymions nudity (II.15.1516).
Some interpretations are eschatological: wise souls among the
Stoics, named Endymiones, live around the moon (Tertullian, An.
55).
12
Rational souls are resolved into the moon, according to Plutarch
(The Face of the Moon, 945AB), but those enamored of the body sleep
with memories of their lives as dreams as did the soul of Endymion.
Pliny the Elder rationalizes: the first human being to observe all these
facts about her [transformations of the moon during eclipses] was
Endymion which accounts for the traditional story of his love for
her (Nat. II.43, [Rackham, LCL]). And Lucian satirizes, telling of
Endymion, king of those living on the moon, warring with the people
of the sun and their king, Phaethon (Ver. Hist. I.1112). The eschato-
logical overtones must be one aspect of the saga that led to Roman
Christians fascination, but we should also include the early Roman
Christians appreciation of aesthetic beauty.
8
Gabelmann, Endymion, LIMC 3.1: 72728 gives references to the classical
authors cited in the following two paragraphs.
9
Callimachus, Aetia (ed. Trypanis, LCL), 8085.
10
Lucian, Dial. d. (ed. Macleod, LCL), 32831.
11
Also Cicero, Tusc. 1.38.92; Amic. 1.13.4344; but see Cicero, Fin. V.20.55.
12
A fragment of Varro, Saturarum Menippearum (ed. Raymond Astbury; Leipzig:
Teubner, 1985), 1820. Lines 100108 have the title Endymiones, referring to this
Stoic conception. Werner A. Krenkel, ed., Marcus Terentius Varro Saturae Menip-
peae (Subsidia Classica 6; St. Katharinen: Scripta Mercaturae, 2002), 1.17284.
From Endymion in Roman Domus to Jonah in Christian Catacombs
276 David L. Balch
Second, selected visual representations: the Roman tombs at Isola
Sacra (between Ostia and Porta at the mouth of the Tiber river)
include an Endymion mosaic in tomb 87 dated to the Antonine era
(c. 140 C.E.), just earlier than the Christian catacomb paintings.
13
The
tomb is richly decorated by frescoes, stucco, and mosaic. The epigraph
reports that the dedicators, P. Varius Ampelus and Varia Ennuchis,
constructed the tomb for themselves, for their patron (Varia Servanda),
their children, freed persons, and their descendants, and it also prohibits
the rite of inhumation unique in Isola Sacra.
14
There are two klinai
(dining couches) in front; inside there is an edicola (niche) and an oven.
To the right of the niche inside, a male wearing a toga is represented,
and to the left a person seated, an emperor administering justice, per-
haps Trajan.
15
Below the niche one sees Thisbe who has discovered her
lover, Pyramus, killed by a lion.
16
Left of the niche is Aiace and Cas-
sandra;
17
one version of this Trojan story is that he drags her away from
the statue of Athena to rape her.
18
Above the niche Diana, spied by the
hunter Actaeon while she was bathing, turns him into a deer, and he is
then attacked and killed by his own dogs.
19
Diverse artists, whom Bald-
assarre judges had little skill, painted these frescoes. The vault was also
decorated, and finally, there is a black-and-white mosaic pavement of
Selene and Endymion who are placed in the center of a geometric de-
13
Ida Baldassarre, Irene Bragantini, Chiara Morselli, and Franca Taglietti, Nicro-
poli di Porto. Isola Sacra (Itinerari dei musei, gallerie, scavi e monumenti dIta-
lia, n.s. 38; Rome: Istituto poligrafico e zecca dello stato, 1996), 7174, fig. 28.
Hans Gabelmann, Endymion, LIMC 3.2 (1986): 55161, at 553, fig. 29.
14
Baldassarre, Nicropoli di Porto, 71, fig. 26. During the second century C.E.
Roman values changed from cremation toward inhumation, a change not gener-
ated by Christians.
15
Ostia Museum inventory #10037.
16
Ostia Museum inventory #10115. Ida Baldassarre, Piramo e Thisbe: dal mito
allimagine, in Lart dcoratif Rome la fin de la rpublique et au dbut du prin-
cipat (Collection de FR 55; Rome: Lcole Fran caise de Rome, 1981), 33751
with 11 figs., including figs. 56 of Tomb 87 at Isola Sacra.
17
Ostia Museum inventory #10114.
18
A version of this painting also occurs in the House of Menander (I 10,4) at Pom-
peii. See David L. Balch, Pauls Portrait of Christ Crucified (Gal.3:1) in Light
of Paintings and Sculptures of Suffering and Death in Pompeian and Roman
Houses, in Early Christian Families in Context: an interdisciplinary dialogue
(eds. David L. Balch and Carolyn Osiek; Grand Rapids, Mich.: Eerdmans,
2003), 84108, color plate 11.
19
Eleanor Winsor Leach, Metamorphoses of the Acteon Myth in Campanian
Painting, MDAI 88 (1981): 30732 with plates 13141.
277
sign.
20
Baldassare sees the theme of violent death in the four visual rep-
resentations of male-female couples, images placed here by a society
undergoing ideological change, one aspect of which was the internali-
zation of the experience of death.
21
Endymion also occurs in domestic settings, as in El Jem, Tunisia,
ancient Roman Thysdrus. Four couples (as in Tomb 87 above) are repre-
sented in small panels (see Figs. 8.12). The museum label reads as fol-
lows:
Endymion, a good looking shepherd, sleeping near a rock, Selene, the Moon,
admires him. Polypheme playing the lyre to charm Galate, nymph that he loves.
Dionysus, drunk, is leaning on a Satyr who reveals Ariadne. Alpheus, god of
river, attacking the nymph Arethusa. End of 2
nd
c. A.D. Maison A du terrain
Jilani Guirot.
The Museo Pio Cristiano in the Vatican exhibits a Christian sarcopha-
gus with the three scenes of Jonah dated perhaps from the final third of
the third century. The third scene exhibits the prophet in a pose like
that of Endymion.
22
Another early sarcophagus from the Basilica di S.
Maria Antiqua in Rome also has the three Jonah scenes.
23
However,
this paper focuses on comparing Pompeian visual representation with
those of the catacomb; to some extent frescoes and sarcophagi belong to
different traditions, so I will notice the latter primarily in footnotes.
24
20
Baldassarre et al., Nicropoli di Porto, 74, fig. 28.
21
Ibid., also 3334.
22
E. Jastrz ebowska, Sol und Luna auf frhchristlichen Sarkophagen: ein tradi-
tionelles Motif der offiziellen kaiserzeitlichen rmischen Kunst in christlicher
Verwendung, in Sarcofagi tardoantichi, paleocristiani e altomedievali: atti della
giornata tematica dei seminari di archeologia Cristiana (FR 8 maggio 2002)
(eds. Fabrizio Bisconti and Hugo Brandenburg; Monumenti di Antichit Cris-
tiana II Serie, 18; Vatican City: PIAC, 2004), 15563, at 159, Fig. 1. Robin Mar-
garet Jensen, Understanding Early Christian Art (New York: Routledge, 2000),
48 with figs. 13ac dates it to the late third century.
23
F. Bisconti, I sarcophagi del paradiso, in Sarcofagi, 57, 72, fig. 32. Marion Law-
rence, Three Pagan Themes in Christian Art, in Essays in Honor of Erwin Pan-
ofsky (vol. 1; ed. Millard Meiss; De Artibus Opuscula 40; New York: New York
University, 1961), 32334, at 325 dates it c. 280, later than some other scholars.
24
See Guntram Koch and Helmut Sichtermann, Rmische Sarkophage (Munich:
C. H. Beck, 1982), 14446 and Helmut Sichtermann, Spte Endymion-Sarkop-
hage: Methodisches zur Interpretation (Deutsche Beitrge zur Altertumswissen-
schaft 19; Baden-Baden: Bruno Grimm, 1966), reviewed by Josef Engemann,
JAC 10 (1967): 24750. Josef Engemann, Untersuchungen zur Sepulkralsymbolik
der spteren rmischen Kaiserzeit (JAC, Ergnzungsband 2; Munster: Aschen-
dorf, 1973), 2830, 7085. Helmut Sichtermann and Guntram Koch, Griechische
From Endymion in Roman Domus to Jonah in Christian Catacombs
278 David L. Balch
These are examples of Endymion in a Roman tomb, an African-
Roman house, and on two Christian sarcophagi from the mid- to late
second and/or third century C.E. Seeking a more complete aesthetic
and cultural context, I turn to Endymion represented in houses of the
living in Pompeii, all prior to Vesuviuss eruption in 79 C.E. I will exam-
ine selected frescoes in Pompeii before locating Jonah in some Christian
catacomb visual representations, having the question in mind whether
and how this figure made the transition from the former to the latter.
Endymion Visually Represented in Wall Frescoes
and in a Stucco Chapel in Pompeian Domus
Fabrizio Bisconti argues that the decoration of domestic, funerary
and civil buildings above ground was imitated in these underground
settings [Christian catacombs].
25
His supporting observations include
Mythen auf rmischen Sarkophagen (Bilderhefte des Deutschen Archologischen
Instituts Rom 56; Tbingen: Ernst Wasmuth, 1975), 2730. Robert Turcan,
Les sarcophages romains et le problme du symbolisme funraire, ANRW 2.6.2
(1978): 170035, esp. 170408, 171213 on Endymion; his pl. II.2 exhibits an
Endymion sarcophagus from Ostia: see Joan R. Mertens, The Metropolitan Mu-
seum of Art: Greece and Rome (New York: Dai Nippon, 1987), fig. 114, a sar-
cophagus dated 210225, the date of the earliest catacomb paintings. Fabrizio
Bisconti, I sarcophagi: officine e produzioni, in Christiana loca: lo spazio cris-
tiano nella Roma del primo millennio (ed. Letizia Pani Ermini; 2 vols.; Ministero
per i beni e le attivit culturali; Compleso di S. Michele, 5 settembre15 novembre
2000; Rome: Fratelli Palombi, 2000), vol. 1, 25763. Tobi Levenberg Kaplan, ed.,
The J. Paul Getty Museum: Handbook of the Antiquities Collection (Los Angeles:
Getty, 2002), 169: front of sarcophagus, dated c. 210, with the myth of Endymion;
Selene arrives in her chariot, then in a second scene to the right, departs.
25
Bisconti, Roman Catacombs, 85. See his pp. 89, 94 on the transition in the late
Antonine and the mature Severan periods from the fourth, architectural Pom-
peian style to the red and green linear, illusionistic style that involved demateri-
alisation and simplification. This is based on Fritz Wirth, Rmische Wandmale-
rei vom Untergang Pompeijs bis ans Ende des dritten Jahrhunderts (Berlin: 1934;
reprint: Darmstadt: Wissenschaftliche Buchgesellschaft, 1968). See Wladimiro
Dorigo, Late Roman Painting (Milan: Feltrinelli, 1966 (in Italian); New York:
Praeger, 1971), chap. 5, and Johannes Kollwitz, Die Malerei der konstanti-
nischen Zeit (Taf. ILXVII), in Akten des VII. internazionalen Kongresses fr
christliche Archologie, Trier, 511 September 1965 (SAC 28; Vatican City/Ber-
lin: PIAC, 1969), at 9398, who dates catacomb visual representations fifty years
later than the Italian De Bruyne (see n. 86 below).
279 From Endymion in Roman Domus to Jonah in Christian Catacombs
8.2. Selene and Endymion, close-up of mosaic in 8.1.
(authors photo)
8.1. Mosaic of four couples, El Jem, Tunisia. (authors photo)
280 David L. Balch
the triumph of marble in all its forms and the tripartite division of
spaces, i.e., three principle registers on walls, which crescendos as im-
ages move up the wall. He concludes that this betrays, very generi-
cally a dependence (of catacomb visual representations) on Pom-
peian wall painting, with extreme simplification of the architectural
imitation.
26
Agreeing with his argument, I will compare wall frescoes
in Pompeii of Endymion with later paintings of Jonah in Christian
catacombs in the larger aesthetic and mythical contexts of domus and
catacomb decoration.
The Casa del Sacello Iliaco (I 6,4)
Two houses in Pompeii, the Casa del Criptoportico (I 6,2)
27
and the
Casa del Sacello Iliaco (I 6,4) were united, and the decoration is of un-
usual quality.
28
In the late Republican period (4030 B.C.E) the cripto-
porticus (19) in house I 6,2 was painted in the late second style, fres-
coes whose quality is comparable with those of the House of Augustus
on the Palatine in Rome.
29
Five windows high in the arch of the vault
provide significant light for the dark cryptoporticus. In the other part
of the combined house (I 6,4), a complete renovation in the fourth style
of the atrium area was undertaken, which is to be dated a century later,
either just before Vesuviuss eruption in 79 C.E. or, according to Stroka,
26
Bisconti, Roman Catacombs, 8889.
27
I 6,4 refers to the first of nine regions into which archaeologists have divided
Pompeii, then to the sixth insula, a block of buildings typically surrounded by
streets, and third, to the entrance door number.
28
Irene Bragantini, Casa del Criptoportico e Casa del Sacello Iliaco (I 6,2.4),
PPM 1 (1990): 193277; 280329. Vittorio Spinazzola, Pompei alla luce degli
scavi nuovi di Via dellAbbondanza (anni 19101923) (Rome: La libreria dello
stato, 1953), vol. 2, 869901 on the Sacrario, and 903970 on the criptoporti-
cus. On the innovative quality of the stucco-work in the criptoporticus (19), see
Roger Ling, Stucco Decoration in Pre-Augustan Italy, PBSR 40 (1972):
1157, esp. 2455. For the sacellum see Richard Brilliant, Visual Narratives:
Storytelling in Etruscan and Roman Art (Ithaca, N.Y. : Cornell University, 1984),
6365, fig. 2.5 and Nicole Blanc, Lenigmatique Sacello Iliaco (I 6,4 E): con-
tribution ltude des cultes domestiques, in I temi figurativi nella pittura parie-
tale antica (IV sec. a.C.IV sec. d.C.). (Atti del VI convegno internazionale sulla
pittura parietale antica; ed. Daniela Scagliarini Corlita; Bologna: University
Press, 1997), 3741.
29
Bragantini, Casa del Criptoportico,194.
281
just before the earthquake in 62 C.E.
30
Triclinium (c) was virtually fin-
ished, but cubiculo (d) lacked the socle (the lowest of the three hori-
zontal bands of decoration on its walls), the sacello (e) received only
the stucco decoration in the vault and in the frieze just below the vault,
cubiculo (h) received only the upper of its three zones of horizontal
decoration, triclinium (i) lacked central wall pictures in its middle
zone, and cubiculo (l) lacked the central picture on its south wall.
An Endymion stucco is in the sacellum (e; see Fig. 8.3). From the
fauces (entrance hall) one sees the sacellum on the far, right side of the
atrium (typically the first room one enters from the street, the roof
of which has an opening to the sky) just to the right of the tablinum
(office). It is a small, walled-off space 1.9 m deep with a raised plat-
form 28 cm high; the platform covers the full width of the space, 1.5 m,
but extends only 1.3 m from the back wall towards the door. There are
thus 60 cm remaining between the platform and the door to the at-
rium. The base of the Homeric frieze is 1.9 m above the top of the plat-
form; the frieze itself, varying from 15 to 17 cm high, wraps horizon-
tally around all three walls and extends on both side walls past the
platform to the door. The semicircular lunette on the back wall is just
above the frieze and contains the Endymion stucco measuring 55 cm
high and 130 cm wide; it is under a barrel vault that extends from the
back wall only as far as the end of the platform below. Both the Endy-
mion lunette and the Homeric frieze are outlined with stucco borders.
Above the floor between the platform and the door there is an open
space (also 60 cm deep), the flat ceiling of which is 1.6 m above the
frieze.
The word sacellum designates a room set apart for the service of the
domestic cult and especially equipped for that purpose
31
with the
statue of the god(s) in a niche. Cicero (Against Verres IV24) describes
such a domestic chapel in which one could see a marble statue of cupid
by Praxiteles and a bronze Hercules by Myron, before which there
were altars, as well as bronzes of Canephoroe by Polycletus and a
wooden statue of Bona Fortuna. Boyce notes that they are rare in
30
Volker Michael Strocka, Ein missverstandener Terminus des vierten Stils: die
Casa del Sacello Iliaco in Pompeji (I 6,4) (Taf. 5061), MDAI 91 (1984):12540.
31
George K. Boyce, Corpus of the Lararia of Pompeii, MAAR 14 (1937): 5112,
with 41 plates, at 18. Also David G. Orr, Roman Domestic Religion: The Evi-
dence of the Household Shrines, ANRW 2.2 (1978):155991 with 10 plates,
at 1578.
From Endymion in Roman Domus to Jonah in Christian Catacombs
282 David L. Balch
8
.
3
.
E
n
d
y
m
i
o
n
,

s
t
u
c
c
o

s
a
c
e
l
l
u
m
,

C
a
s
a

d
e
l

S
a
c
e
l
l
o

I
l
i
a
c
o

(
I

6
,
4
)
,

P
o
m
p
e
i
i
.

(
a
u
t
h
o
r

s

p
h
o
t
o
)
283
Pompeii with only six certain examples (VI 1,1
32
; VI 15,18
33
; VII 2,20
(v)
34
; IX 8,3.6 (?)
35
; IX 9,6 (q)
36
; Villa of the Mosaic Columns
37
), each
with benches for worshipers, a niche with or without paintings, and a
permanent altar for sacrifice; our room is not among Boyces six. The
sacellum in VII 2,20 is a small, walled space comparable in size and
shape to the sacellum in the Casa del Sacello Iliaco (I 6,4). It has two
niches and an altar, but pace Boyce, no benches. The lunette exhibits
a peacock fresco instead of an Endymion stucco.
38
The sacellum in the
Villa of the Mosaic Columns also has an altar, but no benches. Nor are
there benches in Boyces plate of IX 8,3.6. To Boyces six examples
I add sacellum (d) honoring Egyptian deities in peristyle (F) of house
VI 16,7.38, one with a visual representation of a circular altar in the
lower register of the fresco as well as of Anubis, Harpocrates, Isis, and
Serapis in the upper register.
39
Boyce describing the room in I 6,4, doubts that the room was the la-
rarium (the household shrine of the familys tutelary divinities).
40
There
is a lararium near the north door on the west wall of the peristyle in the
32
Boyce, Lararia of Pompeii, plate 40, 34. Hans Eschebach, Probleme der
Wasserversorgung Pompejis, Cronache Pompeiane 5 (1979): 2460, at 59, but
without any description.
33
A. Mau, Ausgrabungen von Pompeji, MDAI 16 (1901): 283365, at 284, fig. 1
(sacellum h), and 28788: eine kleine Larenkapelle.
34
Boyce, Lararia of Pompeii, plate 41, 2. Valeria Sampaolo, Casa di N. Popidius
Priscus (VII 2,20.40), sacello (v), in PPM 6 (1996): 61558, at 65258, figs. 7493.
35
Boyce, Lararia of Pompeii,plate 40, 2, but I do not locate it in PPM 9: 9031104.
36
Boyce, Lararia of Pompeii, plate 41, 1. Valeria Sampaolo, Casa del Vinaio
(IX 9,6), PPM 10 (2003): 13142, sacello (q) at 14041, figs. 1416, a small,
walled-off space off the viridarium (p).
37
Xavier Lafon, Villa Maritima: recherches sur les villas littorales de lItalie romaine
(Bibliothque des cole Franaises dAthnes et de Rome 307; Rome: FR, 2001),
416 with fig. 150, citing Valentin Kockel and Bertold F. Weber, Die Villa delle
Colonne a Mosaico in Pompeji, MDAI 90 (1983): 5189, sacellulm d, at 8283
with figs. 1314 and pl. 35,2, another small, walled-off, decorated space.
38
Sampaolo, Casa di N. Popidius Priscus, 655, fig. 83.
39
Florian Seiler, Casa degli Amorini dorati (VI 16,7.38), PPM 5 (1994):
714845, at 76467, figs. 9399. Some authors refer to sacellum f in the House
of Octavius Quartio (II 2,4); see Shelley Hales, The Roman House and
Social Identity (Cambridge: Cambridge University, 2003), 155 with fig. 42. Ma-
riette de Vos, Casa di D. Octavius Quartio (II 2,2), in PPM 3 (1991): 7179,
figs. 4754, at 71, fig. 47, notes that there is a niche outlined in wood in the east
wall that may have exhibited the statue of a divinity, but she refers to the room as
ambiente f, not sacellum f.
40
Boyce, Lararia of Pompeii, 25, fig. 37.
From Endymion in Roman Domus to Jonah in Christian Catacombs
284 David L. Balch
other part of the house (I 6,2 [12])
41
; but nowhere in I 6,4. Boyce gives a
footnote observing that a bronze statuette of Hercules, standing,
bearded, nude except of the lion skin over his left shoulder,
42
was found
in the large room east of the fauces (c), indicating, perhaps, that it might
have been intended for the platform in the sacellum after it was finished.
Spinazzola records that in the sacellum itself, five episodes from the
Iliad are represented in stucco in the frieze: (1) Hector exiting from the
gate of Troy, (2) the combat of Hector and Achilles, (3) Achilles drag-
ging Hectors body (all in the Iliad, book 22), (4) the ransoming of Hec-
tor, and (5) Priam, guided by Hermes, returning to Troy with Hectors
body (both in Iliad 24).
43
Spinazzola thinks there is no doubt that these
episodes in the sacellum of I 6,4 refer to the decoration of nearly a cen-
tury earlier in the cryptoporticus of I 6,2, reminding viewers of the most
prominent episodes.
44
Discussing this frieze, Simon notes that the ear-
lier third style had more figures, but that fourth-style visual represen-
tations like this one in the time of Nero and Vespasian concentrate on
the protagonists, great mythical personalities like Hector and Achilles.
45
The stucco of Selene and Endymion is in the semicircular lunette
just under the barrel vault on the south wall of the sacellum (e), re-
stored today from the many fragments left by Pompeian earthquakes.
46
The base is blue, the stucco figures ivory. Selene / Diana descends in
her carriage accompanied by perhaps two Eros figures toward a sleep-
ing Endymion. Selenes carriage itself never occurs in contemporary
wall paintings, but is common later on sarcophagi; the carriage is
somehow compatible with the media of stucco and marble, but fresco
painting has a different tradition.
47
As Zimmerman investigates different workshops in a single cata-
comb, so it is also legitimate to investigate the artistic program in an en-
tire house (see n. 4344, 62, and 66) in order to give the complete aes-
41
Bragantini, Casa del Criptoportico, 197, fig. 3.
42
Boyce, Lararia of Pompeii, 18.
43
Spinazzola, Pompeii alla luce, 2:871901.
44
Ibid., 1: 544.
45
Erica Simon, Rappresentazioni mitologiche nella pittura parietale pom-
peiana, in La Pittura di Pompei (eds. Giuseppina Cerulli Irelli, et al.; Milan:
Jaca, 1990, 1991), 239; also in French: La Peinture de Pompi (Tokyo and Paris:
Hazan, 1991, 1993), vol. 1, 26776, and German: Pompejanische Wandmalerei
(Stuttgart: Belser, 1990), 23947.
46
Bragantini, Casa del Criptoportico, 303, fig. 39; 304, figs. 4142.
47
Gabelmann, Endymion, 3.1, 732 (fig. 44) and 740.
285
thetic context, as Spinazzola does with this house (I 6,4). I follow his
lead, down hallway (g) and through the second atrium (m) to room (p),
which he names the salon of philosophers and elephants.
48
Its decora-
tion belongs to the second phase of style two, so is contemporary with
the criptoporticus in I 6,2, nearly a century earlier than the sacellum in
I 6,4, but in the same house. The north wall has a representation of a gi-
gantic philosopher (megalographia) meditating before the globe of the
universe, and to his left is the Muse of astronomy, Urania.
49
Clio, Muse
of history, is portrayed on the west wall.
50
On the east wall is another
megalografia of two elephants facing each other,
51
each of them guided
by reins of myrtle in the hands of small cupids who also hold glass gob-
lets. Elsewhere in Pompei, Venus stands on the raised head and trunk of
one of the elephants; Spinazzola suggests her presence above the elep-
hants here too, although that portion of the fresco has deteriorated.
52
Spinazzola compares the statue of a meditating philosopher in Palazzo
Spada, a fresco of Tragedy meditating,
53
as well as a fresco of a poet giv-
ing friends an audience.
54
The aesthetic program in this house, including
a sacellum featuring Endymion, constitutes a portion of the mythical
and political tradition in houses of the living on which artists drew who
were selecting images to paint for burial chambers in Christian cata-
combs. The sacellum, a room set apart for the service of the domestic
cult, should be added to the architectural sources discussed for Chris-
tian house churches.
55
48
H. G. Beyen, Die Pompejanische Wanddekoration vom Zweiten bis zum Vierten
Stil, Zweiter Band, Erster Teil: Tafeln (Haag: Martinus Nijhoff, 1960), 2425,
plates 5055.
49
Bragantini, Casa del Criptoportico, 324, figs. 7677.
50
Ibid., 325, figs. 7879.
51
Ibid., 324, fig. 77.
52
Spinazzola, Pompei alla luce, 1: 56465, figs. 62425. Compare Valeria Sam-
paolo, Officina coactiliaria de Verecundus (IX 7,7), in PPM 9 (1999): 77478,
at 77677, figs. 24: Venus Pompeiana in a quadriga above four elephants.
53
Spinazzola, Pompei alla luce, 1: 56869 and 573, figs. 62829, 632. Compare de
Vos, Casa del Citarista, 11777, at 143, fig. 44a.
54
Valaria Sampaolo, VI 16,36.37, in PPM 5 (1994): 98195, at 98891, figs. 1216.
55
See L. Michael White, The Social Origins of Christian Architecture (vol. 1 of
Building Gods House in the Roman World: Architectural Adaptation among Pag-
ans, Jews and Christians; Harvard Theological Studies 42; Valley Forge, Pa. Trin-
ity Press International, 1990). Compare Dirk Steuernagle, Kult und Commu-
nity: Sacella in den Insulae von Ostia, MDAI 108 (2001): 4156, who gives
examples of second-century C.E. sacella originally in open spaces that were later
closed off for cult communities.
From Endymion in Roman Domus to Jonah in Christian Catacombs
286 David L. Balch
The Temple of Isis (VIII 7,28)
and the House of Octavius Quartio (II 2,2) in Pompeii
Another domestic setting for Endymion is the residence of the priest(s)
in the Temple of Isis (VIII 7,28). The temple was rebuilt after the earth-
quake of 62 C.E. On the grounds of the temple is a building for the use
of the priests, which includes a cubiculum (7), triclinium (8), and a
kitchen (9). The decoration is related to Egyptian motifs visually rep-
resented by the Roman fourth style. The triclinium (8) includes pic-
tures of a candelabrum on which an eagle is perched, the wise centaur
(combined horse and man) Chiron, and a fresco that has been inter-
preted as either Endymion or Narcissus.
56
Stemmer notes that the two
(or three) basic traditions of representing Endymion can both reduce
the image to Endymion alone.
57
Both Endymion and Narcissus are
hunters (and both are sometimes shepherds). The beautiful body of
both young men is emphasized by contemporary authors (for Endy-
mion see Propertius 1.15.1516 and Cicero, Tus., 1.38.92
58
; for Nar-
cissus, Ovid, Met. 3.339510
59
).
Narcissus is sometimes, but not always, distinguished by wearing
a wreath of narcissi, the flower that sprang up beside the spring
where he died. Actually in another house where a priest of Isis is vis-
ually represented, the House of Octavius Quartio (II 2,2, ambiente f,
south wall),
60
Narcissus is painted above an outdoor biclinium
61
(see Fig. 8.4). Just as in the Casa del Sacello Iliaco described above,
56
Valeria Sampaolo, Tempio di Iside (VIII 7,28), PPM 8 (1998): 732849,
at 847, figs. 22022. See David L. Balch, The Suffering of Isis/Io and Pauls
Portrait of Christ Crucified (Gal. 3:1): Frescoes in Pompeian and Roman
Houses and in the Temple of Isis in Pompeii, Journal of Religion 83, no. 1
(2003): 2455.
57
Klaus Stemmer, Casa dellAra Massima (VI 16,1517) (Huser in Pompeji 6;
Munich: Hirmer, 1992), 5155 on the iconography of the house, 5253 on Endy-
mion.
58
Gabelmann, Endymion, 3.1, 727, 737, with plates in 3.2, 55161.
59
Birgitte Rafn, Narkissos, LIMC 6.1 (1992): 70311, at 703, 709. Another vis-
ual representation of Endymion occurs with a fresco of Io=Isis, Argos and
Hermes in triclinium (37) of the Casa del Citarista (I 4,5.25), described by Ma-
rietta de Vos, PPM 1 (1990): 11777, at 12930, figs. 19, 21.
60
De Vos, Casa di D. Octavius Quartio, 42108, at 7477, figs. 5053.
61
Ibid., 10304, figs. 9193.
287
so also in this house episodes from the story of Troy are painted
(oecus h).
62
Besides Narcissuss lethal self-infatuation, death is presented in an-
other fresco above the same biclinium: Thisbe finds her lover, Pyramus,
whom a lion has killed, and she prepares to take her own life (Ovid,
Metam. 4.55ff;
63
see Fig. 8.5). Further, a scene of lions chasing deer (see
Fig. 8.6) is painted on the left wall leading to the biclinium (not repro-
duced by De Vos), and from this place one can see the amphitheater that
is virtually outside the back door. Further, on the garden side of the
door from room (f) into the garden/portico/biclinium (i) there are visual
representations of Diana spied while she is bathing nude by the hunter
Actaeon; in anger Diana metamorphosizes Actaeon into a deer, who is
then lethally attacked by his own dogs (Ovid, Metam. 3.138252;).
64
Fin-
ally, there is a marble statuette of the infant Hercules strangling two
snakes, also a popular subject in Pompeian wall paintings.
65
Preliminary Conclusions
I draw some conclusions from these observations. First, we already see
the movement from aesthetic programs in houses of the living to the
decoration of tombs for the dead: the artistic program in the House of
62
Ibid., 8498, figs. 6886. Spinazzola compares the Trojan cycles in the crypto-
porticus in I 6,2 (II, 90370) and the sacellum (e) in the Casa del Sacello Iliaco
(1, 54448; II, 869902) with room (f) in the Casa di D. Octavius Quartio
(1, 57493; II, 9711008). In the latter house the upper of the three zones on the
wall is a megalograph of Heracles/Hercules deeds at Troy. The central zone has
only a third the height of the upper zone. Brilliant, Visual Narratives, 6265, figs.
2.42.5. Antonella Coralini, Una stanza di Ercole a Pompei: la sala del dop-
pio fregio nella Casa di D. Octavius Quartio (II 2,2), in Iconografia 2001: Studi
sullImmagine (eds. Isabella Colpo, Irene Favaretto, and Francesca Ghedini;
Universit degli Studi di Padova; Rome: Quasar, 2002), 33143.
63
De Vos, Casa di D. Octavius Quartio,103, 105, figs. 9192, 94.
64
Ibid., 10001, figs. 8788. See n. 19 above. A second Narcissus is represented in
ambient (b; De Vos 55, fig. 21).
65
The Hercules statuette in Wilhelmina Feemster Jashemski and Frederick G.
Meyer, eds., The Natural History of Pompeii (Cambridge: Cambridge University,
2002), 344, fig. 286. For the fresco, see David L. Balch, Zeus, Vengeful Protector
of the Political and Domestic Order: Frescoes in Dining Rooms N and P of the
House of the Vettii in Pompeii, Mark 13:1213, and 1 Clement 6.2, in Picturing
the New Testament (eds. Annette Weisssenrieder, Frederike Wendt, and Petra von
Gemnden; WUNT 2, series 193; Tbingen: Mohr Siebeck, 2005), 6795, plate 6.
From Endymion in Roman Domus to Jonah in Christian Catacombs
288 David L. Balch
8.4. Narcissus, outdoor biclinium, House
of Octavius Quartio (II, 2,2), Pompeii.
(authors photo)
8.5. Thisbe and Pyramus, outdoor biclinium, House of
Octavius Quartio (II 2,2), Pompeii. (authors photo)
289 From Endymion in Roman Domus to Jonah in Christian Catacombs
8
.
6
.
F
r
e
s
c
o

o
f

l
i
o
n
s

c
h
a
s
i
n
g

d
e
e
r
,

w
a
l
l

f
a
c
i
n
g

o
u
t
d
o
o
r

b
i
c
l
i
n
i
u
m
,

H
o
u
s
e

o
f

O
c
t
a
v
i
u
s

Q
u
a
r
t
i
o

(
I
I
2
,
2
)
,

P
o
m
p
e
i
i
.
(
a
u
t
h
o
r

s

p
h
o
t
o
)
290 David L. Balch
Octavius Quartio is related to the decoration of Tomb 87 at Isola
Sacra. The former is not the direct source of the latter, but the aesthetic
programs are strikingly similar.
66
Two of the couples visually repre-
sented in the House of Octavius Quartio, that is, Pyramus and Thisbe,
and Diana and Actaeon, reappear in Tomb 87, as well as the figure
that is Narcissus/Endymion. The artists painting Tomb 87 added
Aiace and Cassandra, another violent Homeric episode that was also
painted in Roman domus.
67
Death was visually present on the walls of Roman houses. We rather
watch bloody deaths resulting from American and British imperialism
in Afghanistan and Iraq on TV, a medium that gives some distance,
but the residents of Pompeii trooped to the amphitheater to thrill while
animals and/or people were killed in their presence.
68
Investigating an-
other image, Kathryn Dunbabin collects a significant amount of evi-
dence for Romans displaying and playing with skeletons in Roman
triclinia, evidence that I will not repeat here, but she shows that the fa-
mous Trimalch