TOPOGRAPHIES OF POWER

IN THE EARLY MIDDLE AGES

This page intentionally left blank
 TOPOGRAPHIES OF POWER
IN THE EARLY MIDDLE AGES

EDITED BY

MAYKE DE JONG AND FRANS THEUWS

WITH
CARINE VAN RHIJN

BRILL
LEIDEN BOSTON KLN

2001
 This book is printed on acid-free paper.

Library of Congress Cataloging-in-Publication Data

The Library of Congress Cataloging-in-Publication Data is also available.

Die Deutsche Bibliothek - CIP-Einheitsaufnahme

Topographies of power in the early Middle Ages / edited by Mayke de Jong
and Frans Theuws. With Carine van Rhijn. Leiden ; Boston ; Kln : Brill,
2001
(The transformation of the Roman world ; Vol. 6)
ISBN 9004117342

ISSN 13864165
ISBN 90 04 11734 2

Copyright 2001 by Koninklijke Brill NV, Leiden, The Netherlands

All rights reserved. No part of this publication may be reproduced, translated, stored in
a retrieval system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without prior written
permission from the publisher.

Authorization to photocopy items for internal or personal

use is granted by Brill provided that
the appropriate fees are paid directly to The Copyright
Clearance Center, 222 Rosewood Drive, Suite 910
Danvers MA 01923, USA.
Fees are subject to change.

printed in the netherlands

 CONTENTS

Acknowledgements ...................................................................... vii

Mayke de Jong and Frans Theuws

Abbreviations .............................................................................. ix

Topographies of power: introduction ........................................ 1

Chris Wickham

Cemeteries as places of power .................................................. 9

Heinrich Hrke

Topography and the creation of public space in early

medieval Constantinople ........................................................ 31
Leslie Brubaker

Topography, celebration, and power: the making of a

papal Rome in the eighth and ninth centuries .................. 45
Thomas F.X. Noble

Monuments and memory:

repossessing ancient remains in early medieval Gaul ........ 93
Bonnie Effros

Cordoba in the Vita vel passio Argenteae ...................................... 119

Ann Christys

Topographies of holy power in sixth-century Gaul ................ 137

Ian Wood

Maastricht as a centre of power in the early Middle Ages .... 155

Frans Theuws

Aachen as a place of power ...................................................... 217

Janet L. Nelson
vi

Convents, violence, and competition for power in

seventh-century Francia .......................................................... 243
Rgine Le Jan

One site, many meanings: Saint-Maurice dAgaune as a

place of power in the early Middle Ages ............................ 271
Barbara H. Rosenwein

Monastic prisoners or opting out? Political coercion and

honour in the Frankish kingdoms ........................................ 291
Mayke de Jong

Monasteries in a peripheral area: seventh-century Gallaecia .... 329

Pablo C. Daz

Aedificatio sancti loci: the making of a ninth-century

holy place ................................................................................ 361
Julia M.H. Smith

People, places and power in Carolingian society .................... 397

Matthew Innes

The regia and the hring barbarian places of power .............. 439
Walter Pohl

Asgard reconstructed? Gudme a central place in

the North ................................................................................ 467
Lotte Hedeager

The lower Vistula area as a region of power and

its continental contacts .......................................................... 509
Przemyslaw Urba czyk

Topographies of Power: Some conclusions .............................. 533

Mayke de Jong and Frans Theuws

Primary sources .......................................................................... 547

Literature .................................................................................... 554
Index ............................................................................................ 597
 ACKNOWLEDGEMENTS

This book is part of the series produced by the European Science

Foundation programme The Transformation of the Roman World. Our
working-group went by the name of Power and Society. We first
published Rituals of power from late Antiquity to the early Middle Ages
(2000). The present volume is its successor, and owes much to the
discussions and debates that raged in 19951997, when TRW was
still running, and all of us were able to meet on a regular basis. We
remain very grateful to the European Science Foundation and to
the co-ordinators of the TRW programme for having made our co-
operation possible.
Once the programme came to an end, we looked for an oppor-
tunity to have one more meeting in order to prepare this publica-
tion. To our delight and gratitude, the Rockefeller Foundation invited
us to organise a workshop in its splendid Bellagio Study and Confer-
ence Center on the shores of the Lago di Como. On 711 September
1998 we profited from the excellent care of the Centers staff and
the ideal surroundings for a stay filled with both hard work and
pleasant relaxation. The Rockefeller Foundation also enabled us to
invite extra participants from outside our ESF-group to contribute
to this venture on Topographies of Power. Hence, we were joined by
Leslie Brubaker, Ann Christys, Albrecht Diem, Bonnie Effros, Matthew
Innes, Tom Noble, Walter Pohl, Barbara Rosenwein and Chris
Wickham.
During preparations for the workshop and in Bellagio itself, we
had the benefit of Carine van Rhijns formidable powers of organ-
isation. Subsequently, she helped us to edit this volume; without all
her efficient and cheerful work, this book would have reached the
press much later. We received a lot of help from many quarters.
Barbara van Rhijn-Dekker kindly checked the English of one chapter;
Walter Pohl lent a hand with editing; Anna Adamska solved some
problems concerning Polish typography; Helmut Reimitz helped with
the bibliography; Janneke Raaijmakers and Rob Meens gave us edi-
torial (and moral) support at the final stage. But above all, the edi-
tors want to thank the contributors for sending us chapters that were
stimulating to read and edit, and for promptly and patiently answering
viii

our countless e-mails. The ESF-programme The Transformation of the

Roman World remains a source of inspiration to all the members of
the group Power and Society. The workshop in Bellagio was a wor-
thy finale to a co-operation that has now formally ended but in
all other possible ways, it continues.

M J
F T
 ABBREVIATIONS

AA SS Acta Sanctorum
AF Annales Fuldenses
Annales ESC Annales. conomies, socits, cultures
BAR British Archaeological Reports
BHL Bibliotheca Hagiographica Latina
Bulletin KNOB Bulletin van de Koninklijke Nederlandse Oudheidkundige
Bond
CBA Council for British Archaeology
CCM Corpus Consuetudinem Monasticarum
CCSL Corpus Christianorum Series Latina
CDF Codex Diplomaticus Fuldensis
ChLA Chartae Latinae Antiquiores
CL Codex Laureshamensis
CSEL Corpus Scriptorum Ecclesiasticorum Latinorum
CSM Corpus Scriptorum Muzarabicorum
DHGE Dictionnaire dHistoire et de Gographie Ecclsiastique
EME Early Medieval Europe
IMR Series Institute for Medieval Research Series
LP Liber Pontificalis
MGH Monumenta Germaniae Historica
AA Auctores Antiquissimi
Capit. Capitularia
Conc. Concilia
DD Diplomata
Epp. Epistolae
LL Leges
Poet.Lat. Poetae Latini
SRG Scriptores Rerum Germanicarum
SRL Scriptores Rerum Longobardicarum et italicarum
SRM Scriptores Rerum Merovingicarum
SS Scriptores
Migne PL Patrologia Latina, ed. J.-P. Migne
Migne PG Patrologia Graeca, ed. J.-P. Migne
MIG Mitteilungen des stereichisches Institut fr Geschichts-
forschung
PBSR Papers of the British School at Rome
Settimane Settimane di studio del centro italiano di studi sullalto
medioevo
SC Sources Chrtiennes
TRAC Theoretical Roman Archaeology Conference
x

TRW Transformation of the Roman World

UBF Urkundenbuch des Klosters Fulda
ZRG Zeitschrift fr Rechtsgeschichte
ZRG kan.Abt. Zeitschrift fr Rechtsgeschichte, kanonistische Abteilung
 TOPOGRAPHIES OF POWER: INTRODUCTION

Chris Wickham

Mayke de Jong, in her proposal for the Bellagio conference, set out
some of the parameters she wanted the participants to work with,
in their analyses of the locations in which power was exercised and
represented. She argued that one of the central transformations in
the post-Roman West (as opposed to the post-Roman East) was that
the city-based Roman world, orientated towards the Mediterranean,
was replaced by societies which resorted to a plurality of locations
of power: cities still, but also palaces, fortifications, monasteries and
other ecclesiastical foundations, where royalty and aristocrats could
publicly express their status and might; in non-Roman parts of Europe
one could add the central places and trading settlements that have
recently been the subject of much analysis in, in particular, Scandinavia.
She went on to say: Such geographical dimensions and divergences
should all be taken into account. At the heart of our interest, how-
ever, is the intricate connection between the physical topography of
power and its mental counterpart. The physical topography of the
past has mostly been the province of archaeologists, while historians
have concentrated on the domain of mentalities, primarily derived
from texts. By linking the debates in the two disciplines, we hope
to get new insights into the exercise of power and its spatial dimen-
sions. Obvious places for consideration are the cult places of the
early middle ages, especially monasteries. In fact, the cult of relics,
and its control by aristocratic and royal power, is a precious indi-
cator of the development of a public arena very different from that
of the Roman world. But there are also contemporary and past land-
scapes and townscapes which need to be taken into consideration
with regard to the definition of new power positions. A Roman ruin
was not a neutral site for quarrying stones; it evoked a glorious past
and invited re-use of a site loaded with memory. According to hagio-
graphical conventions, new monastic foundations were situated in a
wilderness, but quite a few were in fact located near Roman baths
and cult sites.
2

These were our principal starting-points when we considered our

contributions for this conference. I was asked to begin the conference
with a set of reactions to Maykes proposals, as set out above, reac-
tions that were to be as off-the-cuff as possible. The same applies
to this introduction, which essentially does nothing more than set
out the six sets of distinctions that I presented then. These distinc-
tions must be understood as highly generic and provisional; some of
them were extensively developed in debate at Bellagio, and in many
cases the papers that follow go well beyond them. But they were
our starting-points for discussion, as were Maykes proposals for the
texts of the papers, and so are perhaps worth recording in the printed
version of the conference as well.

W-?

Power relations always need to be analysed through an understanding

of (to paraphrase Lenin) who is doing what to whom. In the con-
text of this conference, which is focussed on public representations
of power, this means the question of audience: who is receiving the
message? how are they persuaded to receive it in the right way?
and so on. Even if we just consider the context of secular power
(and this is not the only context we will be looking at), we could
distinguish between several different sorts of power relationship, all
of them with topographical implications: kings representing power to
aristocrats; aristocrats representing power to other aristocrats; aris-
tocrats representing power to kings (as with Masona of Mrida to
Leovigild, or Adalbert II of Tuscany to Louis III);1 kings and aris-
tocrats representing power to non-aristocrats (as in the placitum, or
in the March/Mayfield, or, more generally, on campaign, or in
estate-centres at, say, harvest); non-aristocrats representing power to
aristocrats and kings (this being the least likely box to be filled, but
there are some examples of it, such as the Swedish placita in the Vita
Anskarii, ceremonial occasions in which Swedes could argue with and
defy kings, sometimes successfully).2 Each of these have different para-

1
Vitas sanctorum patrum Emeritensium, ed. A. Maya Snchez (Turnhout, 1992), 5.38;
Liutprand of Cremona, Antapodosis, ed. J. Becker, MGH SRG (Hannover, 1915), 2.38.
2
Rimbert, Vita Anskarii, ed. G. Waitz, MGH SRG 55 (Hannover, 1884), c. 19,
pp. 267.
 : 3

meters, different locations, different sets of symbolic geographies. To

be brief about it: it is a bad idea to mix them up.

The early middle ages saw three basic types of political system: strong
states, that is to say Byzantium and the Caliphate (and the Roman
empire before them); weak states, above all the Romano-Germanic
kingdoms of the centuries after 550, the Franks, Visigoths and Lom-
bards ( joined after c. 750 by the Anglo-Saxons); and very weak states
or (according to ones definition of states), stateless societies, such
as early Saxon England, Scandinavia, the western Slav lands, or the
Celtic lands. It must be expected that the articulation of power would
work differently in each, and, if we want to compare power usefully,
we need to recognise these differences.
In Byzantium, say, public power was a given: if one got an office
one could of course play politics with it, but one already had author-
ity simply by right of office, and also direct power in the framework
of the hierarchy of office. Furthermore, emperors and other politi-
cal leaders could control that hierarchy very directly, for official posi-
tions were funded by taxation; gaining office meant gaining access
to that funding, and losing office meant the loss of it. This is the
major feature that separated strong states from weak ones, which
were not dependent on taxation and which associated office-holding
above all with grants of land, as in the Frankish kingdoms: it was
far harder to take land back from subordinates who were disloyal
or out of favour (or simply inept) than it was to dismiss officials.
This distinction is well-known, of course; but it has relevance when
considering topographies of political power, at least if that power
is connected to the state. Rulers everywhere needed to construct
awe, and thus consent, through the geography of buildings, cities,
landscapes, or through different forms of ritual, but their aims were
different in different state systems: so, speaking very schematically,
Western rulers were often most concerned to ward off regional sep-
aratism, by rituals of association; so did Byzantine emperors, but
they needed to ward off coups as well, because real power was always
most associated in Byzantium with control of central government.
The way Byzantine emperors represented power was differently con-
structed as a result. In the third group, furthermore, systems where
4

there was little public political apparatus at all, one might expect
representations of power to be more direct, and to include more
people as well as to be differently constructed because polities in
this group usually had little of the Roman past to draw on, and
often did not yet have access to the rituals of Christianity either.
These are basic distinctions that need to be addressed before we
compare, if we want to be sure of comparing like with like. I would
add, finally, a further distinction, at a different level: the topographical
representation of the power of great landowners when confronting
neighbours or dependants showed more continuities across Europe,
even though its material forms could change, from the villa rustica to
the mead-hall or the hunting-lodge. But it must be noted that not
all aristocrats actually were great landowners: some had dependants
or clients who were not subject to them by the rules of land tenure
that had survived from the Roman Empire, and who had to be per-
suaded into obedience rather than simply coerced. Here, too, the
representation of power would have to be more inclusive, and maybe
also more materially generous in the provision of feasts, for exam-
ple to its potential audience.

The above characterisations focus on secular power; religious power,

by contrast, often works differently. Gregory of Tours books of the
virtues of St Martin show the saints relics exercising power, with
Gregory systematically recording it, but not necessarily always with
political results.3 Did Gregorys effective control of this miracle-work-
ing make him more powerful in Tours, or with respect to kings? He
thought the second (kings are regularly depicted as afraid of the
miraculous power of St Martin); he does not really say whether he
also thought the first, perhaps because bishops did not strictly need
a major saint to dominate as small a city as Tours. At the very least,
one would have to say that Gregorys power had multiple roots,
drawing both from the St Martin tradition and from the political-
religious traditions of local episcopal/aristocratic power in Gaul that

3
De virtutibus S. Martini, ed. B. Krusch, MGH SRM 1, 2 (Hannover, 1885); the
key reference-points for this text are P. Brown, The cult of the saints (London, 1981);
L. Pietri, La ville de Tours du IV e au VI e sicle (Rome, 1983); R. Van Dam, Saints and
their miracles in late antique Gaul (Princeton, 1993).
 : 5

went back to the late empire, and indeed earlier. But these multi-
ple elements did not necessarily all act in the same way. For exam-
ple, when Gregory faced down kings, which he did on more than
one occasion, it is not clear that he only did so because he had St
Martin to help him; bishops with Gregorys personal charisma and
family connections could act pretty well without saintly authority.
Gregory certainly sought to use St Martin; so did the kings, for that
matter. But there is also a respect in which St Martin himself just
was powerful, and this power was not strictly commensurate with
the power of kings or even bishops. The same was true of living
saints: that they could have secular political power ascribed to them
cannot be doubted (think of Daniel the Stylite during the usurpa-
tion of Basiliscus), but it would be unwise to read their authority
only through political events Daniels aim was to save souls first,
to intervene in politics a distant second.4
This power that St Martin and Daniel just had was as Daniel
thought, at least given them by God, but was at least as much
given them by their audience, which could be very large on occa-
sion. Saints, like kings in fact, were to an extent prisoners of the
expectations of their audiences, in the sort of uncontrollable power
network that has best been characterised by Michel Foucault.5 A
major defining element of this power was that it was not exclusively
used for secular ends; we may indeed even wonder whether any spir-
itually-based power could risk being too exclusively used for secular
political purposes, which might in the end undermine its spiritual
identity. When monasteries, in particular, are referred to as centres
of power, we need to ask what type of power, and how it worked:
how, for example, did St Denis actually help Dagobert and his heirs?
What did people do there? For our purposes here, however, it must
also be noted that religious power could have a different topography
from secular power; furthermore, access to it might be wider (Daniel
on his column, just outside Constantinople), or more controlled (access
to relics was often carefully limited), or entirely restricted (as with
the more enclosed monasteries). These restrictions might have different
valencies from the broadly analogous restrictions on access to kings,
precisely because religious and secular power were not the same.

4
Vita Danielis, ed. H. Delehaye, Les saints stylites (Paris, 1923), cc. 6984; see
P. Brown, The rise and function of the holy man in late Antiquity, Journal of
Roman studies 61 (1971), pp. 80101, at pp. 923.
5
M. Foucault, History of sexuality I (London, 1979), pp. 927.
6

Some of the most obvious problems in political topography hang on

architecture. Architecture or its absence is an essential theatrical
prop when considering how power is represented. The open space
holds one sort of message: it is particularly often found in the case
of representations to wide groups, or non-aristocratic groupsthe
religious faithful, the army, free men as a political community. Build-
ings can be analysed in detail for how their spatial structure pro-
motes a particular form of political message. One example of this is
the throne-room of Mad nat al-Zahr just outside Crdoba, with its
precise placings for officials of different types. Another, closely par-
allel, is the Great Palace of Constantinople, with its mechanical
devices designed explicitly to impress.6 On a less elaborate level, how
a Germanic hall worked can equally clearly be analysed.7 We could
ask where western royal palaces fitted on this spectrum. We could
also ask how much architecture created restrictions, and how much
it reflected restrictions in political participation that were already in
existence: whether the political society of Merovingian Francia or
(still more) Byzantium was already restricted to a small group of
Knigsnhe who were the only people allowed inside a palaces walls,
or whether the existence of the walls of the palace, and the tradi-
tions of operating politically inside them, themselves created such
narrow political communities.

Power was inevitably constructed differently in cities. It was easier

to use architecture as a prop, and, where cities were also demo-
graphic centres, there was a larger ready-made audience, including
a non-aristocratic one. A Byzantine emperor might be more hidden
behind his palace walls than a Merovingian king was, but he was
also closer to the people, because, when he did emerge, the people
were living right there, and could interact with him, in Hagia Sophia,
or before the palace gate, or in the Hippodrome. Cities also allowed

6
M. Barcel, El califa patente, in: R. Pastor e.a. eds., Estructuras y formas del
poder en la historia (Salamanca, 1991), pp. 5171; Liutprand, Antapodosis, 6.5.
7
See for example F. Herschend, The idea of the good in late Iron Age society (Uppsala,
1998).
 : 7

for one particular aspect of the topography of power, processions,

whose paths, timing, and personnel were capable of minute regula-
tion, with changes in them extremely visible to very large audiences
indeed. Processions were indeed sufficiently potent an expression of
power that they could themselves serve to delimit cities, or to link
cities with extramural cult sites, such as Sta Eulalia in Mrida or St
Martin in Tours, whose spatial attraction was otherwise capable of
pulling cities apart, into islands of microurbanism, as at Tours, or
Trier, or Rome itself. All these political representations would work
differently in the countryside, where one would have to work harder
to get an audience together (in fact just assembling the audience
would be an attribute of power), or else where the audience would
be just that much less immediate, receiving the secular or religious
message second hand, or by rumour, that is to say in less control-
lable ways.

Finally, it is worth asking how one can build up sets of places of

power, wider spiritual geographies, landscapes of power (a rural image
above all, although it is fair to add that urban micro-landscapes of
power are also possible, linking in many cases sources of power and
architectures of power of different types). One example of this is the
Paris region, where, in the seventh century, one can identify a net-
work of palaces stretching around the city and up the Oise to the
North that is sufficiently clear-cut that one could refer to it as a
symbolic royal geography, inside which Neustrian kings operated
politically, and which was sufficiently clearly articulated that it could
on occasion generate substantial audiences for major rituals (such as,
to take but one example, the huge procession, with an ever-chang-
ing composition, that took St Audoins body from Clichy back to
Rouen in the 680s.)8 This geography depended on Paris, which was
a real urban centre in this period, and which was the meeting-point
of all the routes of the region; another important element in it was
the two great suburban monasteries of the Paris region, St Denis

8
Vita Audoini, ed. B. Krusch, MGH SRM 5 (Hannover, 1910), cc. 1518. For
the palaces, see J. Barbier, Le systme palatial Franc: gense et fonctionnement
dans le nord-ouest du regnum, Bibliothque de lEcole des Chartes 145 (1990), pp. 24599,
at pp. 25579.
8

and St Germain. In a sense, one could see the monastic and pala-
tial penumbra of Paris as an extreme example of the tendency of
cities to fragment in our period into islands of power, as with the
opposition between the Lateran, the Vatican, and the Palatine in
Rome. (Whether the Parisis was linked together by processions in
the same way as Rome was is less clear, however, and, lets face it,
unlikely.) We might ask how clearly this spatial geography was vis-
ible to the eyes of contemporaries: in particular, whether this Parisian
royal landscape could be clearly counterposed at the time to what
we could characterise as the much more aristocratic landscape of
the Brie to its east, whose power-points were the villas and aristo-
cratic monasteries dotted around Meaux.9 It seems probable to me
that this particular distinction was already clear in the seventh cen-
tury. But it seems certain to me that these sorts of symbolic land-
scapes were in general legible, as a totality, to their chosen audience(s).
The (Neustrian) Frankish political community understood their mean-
ings, just as the Byzantine political community understood the Kre-
mlinology of who was present or absent at processions or in the
Hippodrome, or the Visigothic political community understood the
rituals of public humiliation in Toledo that stud our scarce political
histories for Spain.
We need, first, to reconstruct these political topographies. We then
need, however, to undertake the far harder task of figuring out how
they worked and why, and what for.10

9
See for example A. Bergengrn, Adel und Grundherrschaft im Merovingerreich
(Wiesbaden, 1958), pp. 6584.
10
The discussion that followed this presentation was sufficiently varied that it
would have allowed a completely different introduction to be written one, how-
ever, that would have borrowed too blatantly the ideas of others. Some of the points
raised, however, would have altered my introduction sufficiently that they deserve
record. Walter Pohl proposed a seventh opposition, between military and non-mil-
itary power. Ian Wood pointed out that audiences are constantly changing, between
a standard church service and a festival for example. Opinion was sharply divided
over whether the religious vs. secular distinction was useful or not. Mayke de Jong
stressed the importance of rumour and curiosity when assessing isolated monaster-
ies as places of power. Przemek Urbanczyk stressed that in stateless societies, many
of which specifically avoided concentrations of power, the whole way these ques-
tions are posed would have to be different; he and several of the archaeologists
present, in particular Frans Theuws and Lotte Hedeager, also stressed that they
needed to start with the material evidence, with how power was conceptualised,
my major stress in this introduction, coming second. To these commentators, and
to all the others who contributed to that discussion, my thanks. Thanks also to
Leslie Brubaker and Mayke de Jong for critiquing this text. The conclusion to this
book will pick up their points, and develop them more fully.
 CEMETERIES AS PLACES OF POWER*

Heinrich Hrke

Archaeologists have studied the relationship between burial places

and power for some time, explicitly and within a theoretical frame-
work since the 1980s at least, implicitly for a much longer time.
General considerations as well as archaeological and historical evi-
dence suggest that cemeteries have their own symbolic power, and
are an arena for the display and negotiation of social and political
power. Thus, cemeteries may be part of the plurality of locations of
power which emerged in the early Middle Ages, and which appears
to be one of the hallmarks of the transformation of the Roman
world.
Given the problems of inferring power from archaeological evi-
dence, this paper will approach the subject from a theoretical per-
spective. Rather than presenting evidence and asking how it may
relate to power, we will first consider the nature of cemeteries and
what goes on in them. This will be done from a quasi-contemporary
perspective, not from the perspective of much later archaeologists.
We will then ask where power comes into this, and which evidence
from early medieval cemeteries may relate to the various types of
power present at burial sites.
The emphasis of this paper will be on cemeteries rather than bur-
ial sites in general. The reason for this emphasis is the Roman and

* I am grateful to the Rockefeller Foundation for the opportunity to spend a

week at Bellagio, together with the other members of our European Science
Foundation workshop, and to the European Science Foundation and the Research
Board of Reading University for travel expenses. The first draft of this paper was
written whilst I was holding a Senior Research Fellowship of the British Academy
and the Leverhulme Trust in 1997/98; I am indebted to both institutions. I would
also like to express my thanks to the members of our ESF workshop for the criti-
cal and stimulating discussion of my paper at Bellagio. My Reading University col-
leagues Tony Walter, Richard Bradley and Robert Chapman kindly read and
commented on the final draft of the paper. Additional bibliographical references
were provided by Roberts Coates-Stevens (Reading), Joanna Sofaer-Derevenski
(Cambridge), Frans Theuws (Amsterdam), Chris Wickham (Birmingham) and Ian
Wood (Leeds). I am grateful to Janet DeLaine (Reading) for providing one of the
illustrations, and to Jrn Staecker (Lund) for helping to track down the other.
10

post-Roman preference for burial of inhumations or cremations in

cemeteries. Isolated burial of individuals did occur during this period:
the remote roadside grave of the grandfather of Sidonius Apollinaris
is a case in point,1 as is the use of isolated barrows for the burial
of high-status individuals in seventh-century England.2 However, it
seems that such isolated burial was a comparatively rare phenome-
non or at least, it is rare in the archaeological record. A third
type of grave location was burial inside churches, which emerged as
a royal and aristocratic rite in western Europe by the later sixth cen-
tury.3 Church burial was an integral part of the gradual shift from
Roman extramural to medieval intramural burial, the beginnings of
which are discernible by the sixth century.4
There is likely to have been a conceptual distinction between bur-
ial in a designated, communal or collective burial area, and isolated
burial, be that in a prominent or undistinguished location. Even so,
the three types of burial sites share a number of the features dis-
cussed below. There may have been some overlap or transition
between them, anyway, be that in the form of small clusters of graves
or barrows, or in the development of burial crypts in churches, or
graveyards around churches with burials inside.
1
Sidonius Apollinaris, Epistolae, III.12, transl. O.M. Dalton, The letters of Sidonius,
2 vols. (Oxford, 1915); cf. J. Harries, Sidonius Apollinaris and the fall of Rome, A.D.
407 485 (Oxford, 1994), pp. 2629; I. Wood, Spultures ecclsiastiques et sna-
toriales dans la valle du Rhne (400600), Mdivales 31 (1996), pp. 1327, part.
p. 13. For overviews of Roman burial rites, tombs and cemeteries, cf. J.M.C.
Toynbee, Death and burial in the Roman world, Aspects of Greek and Roman life
(London, 1971); R. Reece ed., Burial in the Roman world, Council for British Archaeology
Research Report 22 (London, 1977). For a brief introduction to post-Roman bur-
ial rites and places, cf. G. Halsall, Early medieval cemeteries, New Light on the Dark
Ages 1 (Glasgow, 1995).
2
J. Shephard, The social identity of the individual in isolated barrows and bar-
row cemeteries in Anglo-Saxon England, in B.C. Burnham and J. Kingsbury eds.,
Space, hierarchy and society, BAR International Series 59 (Oxford, 1979), pp. 4779;
H. Williams, Placing the dead: investigating the location of wealthy barrow buri-
als in seventh century England, in M. Rundkvist ed., Grave matters: Eight studies of
first millennium A.D. burials in Crimea, England and southern Scandinavia, BAR International
Series 781 (Oxford, 1999), pp. 5786.
3
Cf. e.g. J. Werner, Frankish royal tombs in the cathedrals of Cologne and Saint-
Denis, Antiquity 38 (1964), pp. 20116; O. Doppelfeld and R. Pirling, Frnkische Frsten
im Rheinland, Schriften des Rheinischen Landesmuseums Bonn 2 (Dsseldorf, 1966).
4
For Rome, cf. J. Osborne, The Roman catacombs in the Middle Ages, Papers
of the British School at Rome 53 (1985), pp. 278328, part. pp. 279284; R. Meneghini
and R. Santangeli Valenzani, Sepolture intramuranee e pesaggio urbano a Roma
tra V e VII secolo, in L. Paroli and P. Delogu eds., La storia economica di Roma
nellalto Medioevo alla luce dei recenti scavi archaeologici (Florence, 1993), pp. 89111; cf.
the contributions in S. Bassett ed., Death in towns: Urban responses to the dying and the
dead, 1001600 (Leicester, 1992).
 11

W ?

1. Disposal areas

The practical function of cemeteries is first and foremost that of for-

mal disposal areas of the dead. It is worth stressing this at the outset
because formal disposal as such has not been a universal phenom-
enon. There have been societies, and there were entire periods of
early human prehistory, apparently without burial and cemeteries.5
In other words: cemeteries are the result of a deliberate decision to
designate a piece of land an area for the disposal of the dead. This
decision must have been based on non-functional considerations (such
as the desire to create a focus for ritual and remembrance; cf. below)
because dead bodies can be disposed of in other ways, as is demon-
strated by ethnographic and archaeological evidence,6 and by the
absence of clear evidence for formal burial for much of early pre-
history.

2. Places of memory

In cemeteries, the dead themselves are not seen most of the time
unless the respective society practises exposure of their dead, or some
form of secondary burial (two-stage or multiple-stage burial). Exposure
was not infrequent among Plains Indian tribes who called the rite
sky burial, and is still practised in Tibet. Secondary burial has been
practised by a large number of societies, including many European
Christian groups; often this involves exhumation after the decay of
soft tissue, and the removal of bones to an ossuary.7 In such cases,

5
R. Chapman, The emergence of formal disposal areas and the problem of
megaliths in prehistoric Europe, in R. Chapman, I. Kinnes and K. Randsborg
eds., The archaeology of death (Cambridge, 1981), pp. 7181, part. pp. 746; H. Ullrich,
Totenriten und Bestattung im Palolithikum, in F. Horst and H. Keiling eds.,
Bestattungswesen und Totenkult in ur- und frhgeschichtlicher Zeit (Berlin, 1991), pp. 2334.
6
Cf. e.g. R. Huntingdon and P. Metcalf, Celebrations of death: The anthropology of
mortuary ritual (Cambridge, 1979); U. Veit, Studien zum Problem der Siedlungsbestattung im
europischen Neolithikum, Tbinger Schriften zur ur- und frhgeschichtlichen Archologie
1 (Mnster and New York, 1996).
7
For exhumation, cf. L. Danforth, The death rituals of rural Greece (Princeton, N.J.,
1982); for the ossuary at Hallstatt, cf. S. Berg, R. Rolle and H. Seemann, Der
Archologe und der Tod, Bucher Report Series (Mnchen and Luzerne, 1981), p. 83
fig. 84.
12

ossuaries would take the form or the place of monuments and become
the focus of remembrance. In other societies, what we see in ceme-
teries is just the graves of the dead and/or their monuments. These
monuments comprise a strong visual aspect to cemeteries which is
played down in western societies today, but which can still be dis-
cerned in older cemeteries (e.g. those of the nineteenth century) and
other contemporary cultures (e.g. Mexico or Russia).
The monuments, gravestones and markers which are the crucial
part of this visual aspect and which are a widespread feature of
cemeteries make these sites places of memory and remembrance.
Indeed, this element of remembrance is so strongly tied to ceme-
teries that cenotaphs are often located in them, too, in the case of
both, modern and ancient cemeteries.8 At the individual level, this
remembrance in cemeteries takes many different forms within and
across societies. In western culture today, quiet contemplation and
subdued emotion are predominant, but small acts or rituals of
remembrance of a private or family nature have been observed in
London cemeteries.9 Other cultures put more emphasis on the expres-
sion of emotion or more outward forms of remembrance, such as
the Day of the Dead in Mexico or the Sunday family picnic at the
graveside in the Ukraine. At the societal level, monuments and grave-
stones provide an ancestral presence and an expression of origins,
lineages and biographies. They are genealogy in 3-D: a display of
descent and family links which is crucial for legitimation, but also
for the construction and re-affirmation of individual and community
identity.10 The existence of monuments also serves as a trigger for
later re-use and re-interpretation of cemeteries; without monuments,
subsequent generations or different societies might not even recog-
nize burial sites for what they are.

8
Cf. e.g. the site of Klin Yar: H. Hrke and A. Belinskij, Nouvelles fouilles
de 19941996 dans la ncropole de Klin Jar, in: M. Kazanski and V. Soupault
eds., Les sites archologiques en Crime et au Caucase durant lAntiquit tardive et le Haut
Moyen ge, Colloquia Pontica 5 (Leiden, 2000), pp. 193210.
9
D. Francis, L. Kellaher and C. Lee, Talking to people in cemeteries, Journal
of the Institute of Burial and Cremation Administration 65, 1 (1997), pp. 1425.
10
For ancestral presence, cf. J. Barrett, Fragments from Antiquity, Social Archaeology
Series (Oxford, 1994), pp. 1357, 1523; for the link between communal identity,
ancestors, and a sense of the sacred, cf. E. Durkheim, Elementary forms of the religious
life (London, 1915).
 13

3. Places of ritual

Another common feature of cemeteries is that they are places of rit-

ual. Mortuary rituals comprise a variety of ritual acts. Burial rituals
which accompany the deposition of the dead are universal in soci-
eties with cemeteries, and they frequently involve processions arising
originally from the practical need to transfer the dead from the habi-
tation area to the cemetery. By contrast, rituals of remembrance
have been observed in many societies, but have become rare in west-
ern societies, with the exception of remembrance ceremonies for the
war dead, and remembrance spectacles for some exceptional indi-
viduals (a varied group which includes Rudolf Hess, Elvis Presley
and Princess Diana).11
Mortuary rituals serve a number of well-defined purposes: they
have a psychological function, giving form to mourning and heal-
ing; they are rites de passage, accompanying and effecting the dead
persons transformation from living being to ancestor;12 and they play
a role in the reproduction of society, in the re-allocation of rites and
duties of the dead.13 Rites of passage would normally refer to reli-
gious beliefs and cosmology; and the re-allocation of rights and duties
often includes the display of genealogy (as claim to, and legitima-
tion of, power positions) and of social structures (as legitimation of
power relations). Being the place of such rituals makes cemeteries
one of the fields of discourse where societies reproduce power rela-
tions and other structures of knowledge.14
If mortuary rituals are conducted at the societal (as against indi-
vidual or family) level, they often involve ritual specialists (e.g. priests)
who would influence or control the conduct of rituals. Even atheist
burial ceremonies today are often conducted by non-congregational

11
D. Cannadine, War and death, grief and mourning in modern Britain, in
J. Whaley ed., Mirrors of mortality: Studies in the social history of death (London, 1981),
pp. 187242; J.C. Winter, Sites of memory, sites of mourning: The Great War in European
cultural history, Studies in the Social and Cultural History of Modern Warfare
(Cambridge, 1995); T. Walter ed., The mourning for Diana (Oxford, 1999).
12
A. van Gennep, Rites of passage (London, 1960) [first publ. in French 1909].
13
A.A. Saxe, Social dimensions of mortuary practices (unpubl. Ph.D. thesis,
University of Michigan, Ann Arbor, 1970).
14
J. Barrett, Fields of discourse: reconstituting a social archaeology, Critique of
Anthropology 7:3 (1988), pp. 516.
14

burial specialists who orchestrate a funeral eulogy and assist in the

consolation of the bereaved.15

4. Places of emotion

Rituals in cemeteries may also involve the expression of emotions,

but that expression is only part of a whole complex of emotions
inextricably linked with loss and bereavement. Depending on the
social and cultural context, personal grief and socially accepted forms
of mourning may be distinct, overlap or imperceptibly merge.16 In
many cases the burial place of the deceased will be the trigger and
the venue for such emotions and, where appropriate, their display
in ephemeral or material form.17 Cemeteries may, thus, also become
places of emotional healing, of coming to terms with loss.

5. Places of encounter with mortality

For most people, cemeteries provide the main encounter with death
and mortality. In cemeteries, we come face to face with the ques-
tion of the hereafter, of the place of humans in the world we
confront aspects of cosmology. Therefore, cemeteries are usually
sacred, hallowed places, distinguished from the profane surroundings
by their location and by other markers, frequently by enclosure (much
like monasteries are). By the same token, they are foci of beliefs
which today we are inclined to dismiss as superstition, like the belief
in the living dead (Wiedergnger, vampires etc.).18

15
T. Walter, Secular funerals, Theology 92 (1989), pp. 394402; T. Walter,
Funerals and how to improve them (London, 1990), pp. 217231.
16
Grief and mourning have been subjects of widespread theorising and intense
debate; for an outline of the growth of the debate, cf. L. Prior, The social organisa-
tion of death: Medical discourse and social practices in Belfast (Basingstoke, 1989), pp.
133141.
17
Because of the perceived immateriality of emotion, archaeologists have gener-
ally steered clear of research on this aspect of human behaviour. Recently, Sarah
Tarlow has drawn attention to this lacuna and proposed an archaeological approach
to cemeteries which focuses on emotion; S. Tarlow, An archaeology of remem-
bering: death, bereavement and the First World War, Cambridge Archaeological Journal
7:1 (1997), pp. 10521, part. p. 107; S. Tarlow, Bereavement and commemoration: An
archaeology of mortality, Social Archaeology Series (Oxford, 1999).
18
P. Barber, Vampires, burial and death: Folklore and reality (New Haven, 1988).
 15

6. Properties

A cemetery takes up a specific plot, be that in the landscape, on an

estate, in a town, a churchyard, a monastery, or wherever. In some
societies, a cemetery plot may be located on land without property
rights, sometimes deliberately so. But in many other contexts, it is
on community property, or on land owned or controlled by some-
body: the Church, a monastic order, or aristocratic landowners. This
means that often, there may be rights to consider, perhaps even that
somebody has control over access to, and use of, the cemetery. This
control can be meant to exclude (i.e. to reserve the cemetery for a
designated group) or include (i.e. to enforce burial in a designated
cemetery). For example, burial rights were a jealously guarded monop-
oly in parts of medieval Europe because they entailed the donation
of burial offerings (soul-scot and dying bequests) to the respective
mother church.19 In modern Europe, the use of crematoria and ceme-
teries is closely controlled by municipal authorities.

7. Localities and locales

A cemetery is always a locality in a spatial, topographical sense, and

as such, part of a constructed, cultural landscape (or townscape).
This locality becomes a locale by its link to remembrance and the
specific forms this takes in the respective society. Such a locale might
be hidden away, or it might be used for active display. Thus, Roman
cemeteries, strung out along roads for display purposes, were essen-
tially drive-through cemeteries where travellers could not avoid notic-
ing the deliberate display constructed by the community they were
approaching or leaving (fig. 1). This might also apply to prehistoric
barrows along ridgeways in western Europe, and to barrows which
themselves became markers on the featureless steppes of eastern
Europe and Central Asia.
By way of contrast, post-Roman and Christian cemeteries have
often been dead-end cemeteries: places of destination which you go
to, or away from, but rarely pass through casually. Even where the

19
Cf. e.g. B. Kemp, Some aspects of the parochia of Leominster in the 12th
century, in: J. Blair ed., Ministers and parish churches. The local church in transition,
c. 9501200, Oxbow Monographs 17 (Oxford, 1988), pp. 8395.
 16

Fig. 1. Drive-through commemoration: a Roman cemetery outside Pompeii
(photograph J. DeLaine).
 17

latter might happen regularly, for example in the case of church-

yards, there are often inhibitions or superstitions against such casual
treatment of the locality if it is still in regular use as a burial site.
This crucial aspect is highlighted every time when a municipal author-
itys plan to put a public footpath or cycle path across a cemetery
is met with controversy.20 Thus, cemeteries may constrain movement,
or mark paths, through the landscape.21

8. Expressions of mental topographies

Cemeteries may be the physical expressions of mental topographies,

although this varies markedly across cultures and periods. Such
topographies may be sacred, or they may be topographies of power,
or they may just be a reflection of domestic topography. In parts of
western Africa, on Madagascar and Borneo, tombs are houses for
the dead.22 Many Roman tombs mimick, or show paintings of, inte-
riors of the houses of the living. A prehistoric case may be the Middle
Bronze Age cremation cemetery of Down Farm (Dorset, England)
which shows a similar organization of space to the houses found
nearby.23 The Scythians had the concept of the gerrhos, a landscape
of royal barrows.24 It may be no coincidence that these steppe nomads
conceptualized their burial topography as a landscape, not a ceme-
tery. In modern Belfast, the segregation of the dead mirrors that of
the living and, thus, recreates the political geography of Northern

20
A personal experience which highlighted the issues involved and the passions
aroused was the plan of the municipal authorities at Hameln (Germany) in 1998
to put a cycle path across the Deisterfriedhof cemetery, which would run within a
few metres of the tomb of my family.
21
Barrett, Fragments from Antiquity; C. Tilley, A phenomenology of landscape: Places,
paths and monuments (Oxford, 1994); C. Tilley, The powers of rock: topography and
monument construction on Bodmin Moor, World Archaeology 28:2 (1996), pp. 16177.
22
Western Africa: S. Bhnen, Haus und Grab: gebaute Kosmologie und Ideologie,
Ethnographisch-Archologische Zeitschrift 38 (1997), pp. 45155; Madagascar: M. Bloch,
Placing the dead: Tombs, ancestral villages and kinship organisation in Madagascar (London
and New York, 1971); Borneo: P. Metcalf, A Borneo journey into death: Berawan escha-
tology from its rituals (Philadelphia, 1982); R. Huntington and P. Metcalf, Celebrations
of death (n. 6).
23
R. Bradley, Working the land: imagining the landscape, Archaeological Dialogues
4:1 (1997), pp. 3952, esp. p. 45.
24
Herodotus, The Histories, IV.71, transl. R. Waterfield, Herodotus: The Histories,
Oxfords World Classis (Oxford, 1998).
18

Ireland.25 It has also been suggested that Bronze Age barrows in

southern England created a sacred geography, and that the loca-
tions of saints tombs and the spread and redistribution of their relics
created a Christian geography from the fourth century onwards.26
Burial monuments themselves may express mental topographies.
Thus, the soil for the barrow of the Polish hero Tadeusz Ko ciuszko,
who died in the 1790s in an insurrection against Russian rule, is
said to have been carted from his battlefields to the barrow location
outside Cracow.27 Archaeological evidence has shown that the soil
for large Scythian barrows was brought from distances of several
kilometres away, suggesting the creation of a symbolic pasture for
the afterlife.28

9. Places of subjective time

The temporal dimension of cemeteries is linked to their locations,

rituals, monuments and memories. For many in western societies,
the first association with cemeteries may well be that of peace and
quiet, of an oasis of sedate pace in a hectic world. This contrast is
strongest in urban cemeteries today where visitors can be seen to
slip into that pace, to adopt an unhurried behaviour which is in
stark contrast to the time-is-money attitude outside the cemetery
walls. In rural areas, and even more so in societies without a scientific
concept of time, this contrast is bound to be less marked, but it is
still likely to be present as a different perception of time linked to
the locale. Thus, cemeteries can be argued to be islands of subjec-
tive (as against scientific, industrial) time islands where monu-
ments evoke the past, and where rituals link past, present and future.29

25
Prior, The social organisation of death, pp. 111132.
26
D. Field, Round barrows and the harmonious landscape: placing Early Bronze
Age burial monuments in South-East England, Oxford Journal of Archaeology 17:3
(1998), pp. 30926, part. p. 322; P.J. Geary, Living with the dead in the Middle Ages
(Ithaca and London, 1994), p. 166.
27
S. Schama, Landscape and memory (London, 1995), p. 26.
28
R. Rolle, The world of the Scythians (London, 1989), p. 34, on the analogy of a
recorded Hittite royal funeral.
29
The sociologist Walter has drawn on Hogans notion of everywhen, where
past, present and future are blended into a oneness, to conceptualize the dissolu-
tion of temporal and spatial boundaries after bereavement; T. Walter, On bereave-
ment: The culture of grief, Facing Death Series (Buckingham and Philadelphia, 1999),
p. 58. For concepts of time in relation to rituals and monuments, cf. R. Bradley,
 19

Cemeteries may not be a gravity-free zone,30 but they appear to

be in a different time zone.

So, where does power come into this picture of cemeteries, and how
can this power be shown in the archaeological evidence of Late
Roman and early medieval cemeteries? For this discussion, it may
be useful to distinguish between three different aspects of power in
relation to cemeteries: power of cemeteries, power in cemeteries, and
power over cemeteries. These are very broad distinctions with over-
laps, but they offer a structure for discussion.31

Power of cemeteries
One aspect is the power of the place itself. This perceived power
may be older than the cemetery, and may have been the reason
why burials were located there.32 Alternatively, burials might give
power to a location, particularly if they are burials of distinguished,
powerful individuals. Some of the power of cemeteries may also be
derived from the fact that they are liminal places, located between
the spheres of the living and of the ancestors and as Douglas has
argued: to have been in the margins is to have been in contact with

Ritual, time and history, World Archaeology 23 (1991), pp. 20919; R. Bradley, The
significance of monuments (London, 1998), part. pp. 85100; K. Mizoguchi, Time in
the reproduction of mortuary practices, World Archaeology 25:2 (1993), pp. 22335;
for a wider archaeological perspective on time, cf. G. Clark, Space, time and man: A
prehistorians view (Cambridge, 1992); more generally on concepts of time, cf. G. Debord,
La socit du spectacle (Paris, 1967); N. Elias, ber die Zeit, Arbeiten zur Wissenssoziologie
2 (Frankfurt a.M., 1984).
30
I. Wood, this volume, after van Damme.
31
The concept of power used here is deliberately broad, ranging from the con-
ventional (power over) to the vaguely Foucauldian (power in) to the metaphorical
(power of ); cf. M. Foucault, Power/knowledge: Selected interviews and other writings 19721977
(London, 1980).
32
For a prehistoric case, cf. R. Bradley, Monuments and places, in P. Garwood,
D. Jennings, R. Skeats and J. Toms eds., Sacred and profane: Proceedings of a conference
on archaeology, ritual and religion. Oxford, 1989, Oxford University Committee for
Archaeology Monograph 32 (Oxford, 1991), pp. 13540.
20

danger, to have been at a source of power.33 In reality, these aspects

are probably interrelated, and they might be difficult to distinguish
in the archaeological record, anyway.
The most obvious archaeological indication of the power of par-
ticular places is the re-use of older monuments, particularly barrows,
which has recently been discussed for the Roman Iron Age of north-
ern Germany, Roman Britain, and post-Roman England and Gaul.34
Bradley has pointed out that the distinction between natural and
cultural features may in some cases have been blurred, e.g. in the
case of megalithic graves and tors (natural stacks of granite) in south-
west England.35 This might well raise the question if those who chose
prehistoric barrows as a place for later burial recognized them for
what they were. However, this re-use is too frequent and systematic
to be dismissed as mere coincidence of prehistoric and Roman/post-
Roman burial sites. In some cases, the number of cemeteries or bar-
rows located on, or next to, older monuments can reach staggering
proportions: e.g. some 60% of known seventh-century Anglo-Saxon
cemeteries in the Upper Thames valley are found in such locations
(incl. probable and possible cases).36 Also, the very fact that early
medieval barrows were often built right next to existing, prehistoric
barrows would seem to suggest that the nature of the latter was
known to, or suspected by, the builders of the former.
The visual power of cemeteries is a much-neglected aspect of re-
search and debate because too many post-Roman monuments are

33
M. Douglas, Purity and danger: An analysis of the concepts of purity and taboo (London,
1966), p. 117.
34
E. Thte, Alte Denkmler und frhgeschichtliche Bestattungen: ein schsisch-
angelschsischer Totenbrauch und seine Kontinuitt, Archologische Informationen 19
(1996), pp. 10516; T. Eaton, A neglected landscape? A study of the value of Antiquity to
the inhabitants of Merovingian Gaul, with particular reference to the Roman rural landscape
(unpubl. M.A. thesis, University of Reading, 1995); H. Hrke, Lowbury Hill: A
context for the Saxon barrow, Archaeological Journal 151 (1994), pp. 2026; H. Hrke
and H. Williams, Angelschsische Bestattungspltze und ltere Denkmler: Bemer-
kungen zur zeitlichen Entwicklung und Deutung des Phnomens, Archologische
Informationen 20:1 (1997), pp. 257; H. Williams, Ancient landscapes and the dead:
the reuse of prehistoric and Roman monuments as early Anglo-Saxon burial sites,
Medieval Archaeology 41 (1997), pp. 132; H. Williams, The ancient monument in
Romano-British ritual practices, in C. Forcey, J. Hawthorne and R. Witcher eds.,
TRAC 97: Proceedings of the Seventh Annual Theoretical Roman Archaeology Conference (Oxford,
1998), pp. 7186; H. Williams, Monuments and the past in early Anglo-Saxon
England, World Archaeology 30:1 (1998), pp. 90108; cf. Effros, this volume.
35
R. Bradley, Ruined buildings, ruined stones: enclosures, tombs and natural
places in the neolithic of south-west England, World Archaeology 30:1 (1998), pp. 1322.
36
H. Williams, Ancient landscapes and the dead.
 21

lost or eroded, and their landscape setting has changed beyond recog-
nition.37 However, this aspect is more immediate and becomes more
understandable in the case of Roman mausolea.38 The closest equiv-
alents which the early medieval period can offer are the few burial
churches and, in particular, the more frequent and widespread bar-
rows,39 but again we see the latter in their eroded, grassed-over form,
not in their original shape, possibly with decoration and sacrifices
(fig. 2). A new Anglo-Saxon barrow of gleaming white chalk on the
green downs of England must have been an eye-catching sight, even
from a distance. The same would have been true, although perhaps
to a lesser degree, of newly erected barrows on the North German
Plain or the Ukrainian and North Caucasian steppes.
This visual power of burial places may have been enhanced by their
natural setting which, in turn, provided links to other aspects of sym-
bolic power. As Schama has observed: . . . one of our most powerful
yearnings: the craving to find in nature a consolation for our mor-
tality. It is why a grove of trees, with their annual promise of spring
awakening, are thought to be a fitting dcor for our earthly remains.40
The power of the dead has many facets: ancestral power, holy or
sacred power, and the fear of the dead. These are all aspects of sym-
bolic power, but they can be used to enhance or undermine social
power, the power of the living. They are also better illuminated
by written sources than by archaeological evidence. The symbo-
lism of swords taken from graves appears to have been particularly
powerful, in Lombard Italy as well as in Norse sagas.41 A belief
in the benevolent power of certain dead is demonstrated by the
attraction of burial ad sanctos which features most clearly in written

37
For a prehistoric case of visual power, cf. R. Bradley, Directions to the dead,
in L. Larsson and B. Stjernquist eds., The world view of prehistoric man, KVHAA
Konferenser 40 (Lund, 1998), pp. 12335.
38
I. Morris, Death ritual and society in classical Antiquity (Cambridge, 1992).
39
M. Mller-Wille, Knigsgrab und Knigskirche, Berichte der Rmisch-Germanischen
Kommission 63 (1982), pp. 349412; id., Knigtum und Adel im Spiegel der
Grabfunde, in Die Franken: Wegbereiter Europas. Vor 1500 Jahren: Knig Chlodwig und
seine Erben (Mainz, 1996), pp. 20621; R. van de Noort, The context of early
medieval barrows in western Europe, Antiquity 67 (1993), pp. 6673; M. Lutovsky,
Between Sutton Hoo and Chernaya Mogila: barrows in eastern and western early
medieval Europe, Antiquity 70 (1996), pp. 6716; Williams, Placing the dead.
40
S. Schama, Landscape and memory, p. 15.
41
Geary, Living with the dead in the Middle Ages, pp. 6167; H.E. Davidson and
P. Fisher, Saxo Grammaticus: The History of the Danes, Books IIX, vol. II (Cambridge
and Totowa, N.J., 1980), p. 69, n. 34.
 22

Fig. 2. The power of monuments: early medieval barrows and later Christian church at Old
Uppsala (lithograph by C.J. Billmark, between 1857 and 1859).
 23

sources.42 In the archaeological evidence, the enclosure of early

medieval cemeteries, e.g. in Early Christian Ireland and Scotland,43
may well be the reflection of a belief in the power of the dead, lead-
ing to a clear separation of the dead from the sphere of the living.
In the Roman world, that separation was effected by extramural
burial; it may be that the weakening of that previously strict rule in
the post-Roman period led to other forms of demarcation of the two
realms.

There are other signs of a belief in the power of the dead. While
amulets in graves are difficult to identify and interpret, other ritual
practices such as stones on buried bodies, tied hands and feet of
corpses, and post-mortem decapitation may highlight a fear of the
dead, or more specifically, the return of certain dead.44 Decapitation,
in particular, is a well-documented practice in England where it is
an element of continuity from the Roman to the Anglo-Saxon period.45
The graves of ancestors appear to have held a similar attraction,
probably where descent and genealogy were important for identity
and social status. Thus, the elite plot in the long-term cemetery of
Klin Yar (North Caucasus, Russia) was used over five centuries and
two cultural phases (Sarmatian and Alanic), with evidence of repeated
re-use of catacombs (as family vaults?), cenotaphs, and even a pos-
sible translation of a skeleton from another burial place.46 In such
cases, the archaeological inference of family relationships is a seri-
ous problem, but scientific techniques (in particular odontological
and DNA analysis) promise an improvement for the future.47

42
B. Effros, Beyond cemetery walls: early medieval funerary topography and
Christian salvation, EME 6, 1 (1997), pp. 123, part. pp. 6, 1316.
43
Ll. Laing, The archaeology of Late Celtic Britain and Ireland c. 400 1200 A.D.
(London, 1975), pp. 37780.
44
For amulets, cf. A.L. Meaney, Anglo-Saxon amulets and curing stones, BAR 96
(Oxford, 1981); for a recent review of unusual ritual practices in Early Anglo-Saxon
England, cf. H. Winskill, An analysis of Anglo-Saxon deviant burials (unpubl.
B.A. thesis, University of Reading, 1999).
45
M. Harman, T.J. Molleson and J.L. Price, Burials, bodies and beheadings in
Romano-British and Anglo-Saxon cemeteries, Bulletin of the British Museum, Nat.Hist.
(Geology) 35:3 (1981), pp. 14588.
46
See above, n. 8.
47
For a critical review of archaeological and biological approaches to the identifi-
cation of family relationships, cf. H. Hrke, Zur Bedeutung der Verwandtschaftsanalyse
aus archologischer Sicht, Germania 73 (1995), pp. 30712.
24

Power in cemeteries
While it is likely that the participants of rituals conducted in Late
Roman and early medieval cemeteries believed in the power of such
rituals, this is an intangible aspect which is difficult to demonstrate
in the archaeological record. Rituals may have been powerful sim-
ply by the impression they made on participants, by drawing them
into a ritual community, by engendering action or acceptance. But
they must also have been powerful in the sense that participants
believed that rituals had the power to effect, to achieve something
(which is one definition of rituals, and one way of distinguishing
them from ceremonies).48
The display and negotiation of social and political power appears
to have been part of mortuary rituals, and here archaeology is in a
much better position because of the material display involved (assum-
ing our interpretations of the archaeological record concerning this
aspect are correct). Roman grave monuments and mausolea clearly,
and often explicitly, display status, wealth and power.49 While there
are few such monuments extant from early medieval cemeteries, the
typical grave-goods custom of the post-Roman period may be inter-
preted as competitive display: inhumation and cremation graves of
the fifth to seventh centuries in large parts of Europe contain dress
items, jewellery, weapons, drinking vessels and other artefacts which,
because of their range and varying quantities (from poor to rich)
have traditionally been seen as a display of economic and social
power.50 Childe argued a long time ago that an elaborate grave-
goods custom is typical of societies with an unstable social hierar-
chy which leaves room for competition for status positions.51 This

48
For recent discussions, cf. F. Theuws, Introduction: Ritual in transforming
societies, in F. Theuws and J.L. Nelson eds., Rituals of power from Late Antiquity to
the Early Middle Ages, The Transformation of the Roman World 8 (Leiden, Boston
and Cologen, 2000), pp. 113; K. Nordstrm, Problems and ideas concerning ide-
ology in the construction of religion and ritual as analytical concepts, Lund
Archaeological Review 3 (1997), pp. 4957.
49
I. Morris, Death ritual and society.
50
R. Christlein, Besitzabstufungen zur Merowingerzeit im Spiegel reicher Grab-
funde aus West- und Sddeutschland, Jahrbuch des Rmisch-Germanischen Zentralmuseums 20
(1973), pp. 14780; H. Steuer, Frhgeschichtliche Sozialstrukturen in Mitteleuropa,
Abhandlungen der Akademie der Wissenschaften in Gttingen, Phil.-Hist. Klasse,
3. Folge, 128 (Gttingen, 1982); R. Samson, Social structures from Reihengrber:
mirror or mirage?, Scottish Archaeological Review 4 (1987), pp. 11626.
51
V.G. Childe, Directional changes in funerary practices during 50,000 years,
Man 45 (1945), pp. 139.
 25

would certainly explain the marked revival of the deposition of grave-

goods from the Roman to the post-Roman period in western Europe.
The interpretation in terms of competition appears confirmed for
Early Saxon England by the observation that large cemeteries have
a wider range, as well as a higher average quantity and quality, of
grave-goods than small cemeteries.52 In other words, the scale of the
competitive display during the funeral was directly related to the size
of the potential audience.
The conspicuous wealth of certain post-Roman graves, such as
those at Sutton Hoo and under the cathedrals of Cologne and Saint
Denis, have always been taken to suggest aristocratic or royal status,
in the latter case supported by the location of the burials.53 In some
cases, such as that of the Frankish king Childeric, it has been pos-
sible to identify the royal occupants of outstanding graves,54 supporting
the suggested link between funerary ostentation and political power.
But power relations in society may also be displayed in cemetery
location, lay-out and subdivisions. Thus, separate Adelsfriedhfe as well
as Adelsgrber in ordinary cemeteries have been identified across early
medieval Europe and in the preceding societies of the Roman Iron
Age.55 Differential power in gender relations may be evident in the
gender-separate cemeteries of the Older Roman Iron Age in northern

52
J.McA. King, Prestations and distance: a role for Early Saxon grave goods in
social reproduction A.D. 450600 (unpubl. M. Litt. thesis, University of Oxford, 1999).
53
R.L.S. Bruce-Mitford, The Sutton Hoo ship burial, 3 vols. (London, 19751983);
M. Carver, Sutton Hoo: Burial ground of kings? (London, 1998); P. Prin, Die Grabsttten
der merowingischen Knige in Paris, in Die Franken, pp. 41622. Cf. nn. 3 and 4.
54
M. Mller-Wille, Knigsgrab und Knigskirche; E. James, Royal burials
among the Franks, in M. Carver ed., The age of Sutton Hoo (Woodbridge, 1992),
pp. 24354; P. Prin and M. Kasanski, Das Grab Childerichs I. in Die Franken,
pp. 17382.
55
For cases in Iron Age societies outside the Roman Empire, cf. H.J. Eggers,
Lbsow, ein germanischer Frstensitz der lteren Kaiserzeit, Praehistorische Zeitschrift
34/35:2 (1949/50), pp. 58111; M. Gebhr, Zur Definition lterkaiserzeitlicher
Frstengrber vom Lbsow-Typ, Praehistorische Zeitschrift 49 (1974), pp. 82128;
W. Schlter, Versuch einer sozialen Differenzierung der jungkaiserzeitlichen Kr-
pergrbergruppe von Haleben-Leuna anhand einer Analyse der Grabfunde, Neue
Ausgrabungen und Forschungen in Niedersachsen 6 (1970), pp. 11745. Examples of early
medieval sites in western Europe: H. Ament, Frnkische Adelsgrber von Flonheim in
Rheinhessen, Germanische Denkmler der Vlkerwanderungszeit B 5 (Berlin, 1970);
U. Koch, Das frnkische Grberfeld von Klepsau im Hohenlohekreis, Forschungen und
Berichte zur Vor- und Frhgeschichte in Baden-Wrttemberg 38 (Stuttgart, 1990);
M. Martin, Das frnkische Grberfeld von Basel-Bernerring, Baseler Beitrge zur Ur- und
Frhgeschichte (Basle, 1976); P. Paulsen, Alamannische Adelsgrber von Niederstotzingen,
Verffentlichungen des Staatlichen Amts fr Denkmalpflege Stuttgart A12 (Stuttgart,
1967). For an East European case, cf. Hrke and Belinskij, Nouvelles fouilles.
26

Germany, and in the separation of genders in medieval churchyard

cemeteries on Gotland and elsewhere.56 Using a wide range of ethno-
graphic data, Derks has argued that gender-separate burial is typi-
cal of societies with low female status.57 Details of ritual practices can
also be used for the symbolic representation of power. A case in
point is the weapon burial rite which appears to have been used as
a badge of ethnic identity and social power in Anglo-Saxon society,
but there have been few other studies along these lines.58

Power over rituals themselves is another important aspect in this

context, but the nature and extent of this kind of control varies from
society to society and from religion to religion. Priests, as a distinct
class of specialists overseeing ritual, existed in Roman society and in
early medieval Christianity. Pagan priests were reported, among oth-
ers, from Anglo-Saxon England, Scandinavia and the Western Slavs.59
But it has been suggested recently that the institution of early medieval
priests among the pagan Western Slavs may have been a reaction
to contact and conflict with Christianity, and a more or less direct
copy of Christian institutions.60 Christianization is, therefore, likely
to have had profound consequences for the conduct and control of
rituals in areas where there had been no ritual specialists before.

56
H. Derks, Geschlechtsspezifische Bestattungssitten: ein archologischer Befund
und ein ethnoarchologischer Ansatz, Ethnographisch-Archologische Zeitschrift 34 (1993),
pp. 34053; J. Staecker, Searching for the unknown: Gotlands churchyards from
a gender and missionary perspective, Lund Archaeological Review 2 (1996), pp. 6386.
57
See n. 56.
58
H. Hrke, Warrior graves? The background of the Anglo-Saxon weapon
burial rite, Past & Present 126 (1990), pp. 2243; H. Hrke, Angelschsische Waffengrber
des 5. bis 7. Jahrhunderts, Zeitschrift fr Archologie des Mittelalters Beiheft 6 (Kln
and Bonn, 1992); H. Hrke, Material culture as myth: weapons in Anglo-Saxon
graves, in: C.K. Jensen and K.H. Nielsen eds., Burial and society: The chronological
and social analysis of archaeological burial data (Aarhus/Oxford/Oakville/Connecticut, 1997),
pp. 11927; cf. also N. Stoodley, The spindle and the spear: A critical enquiry into the cons-
truction and meaning of gender in the early Anglo-Saxon burial rite, BAR 288 (Oxford, 1999);
F. Theuws and M. Alkemade, A kind of mirror for men: sword depositions in Late
Antique northern Gaul, in Theuws and Nelson eds., Rituals of power, pp. 40176.
59
England: Beda Venerabilis, Historia ecclesiastica gentis Anglorum, II.13, eds.
B. Colgrave and R.A.B. Mynors, Bedes Ecclesiastical history of the English people, Oxford
Medieval Texts series (Oxford, 1969). Scandinavia: Adam of Bremen, Gesta Hamma-
burgensis ecclesiae pontificum, IV.27, transl. F.J. Tschan, Adam of Bremen, History of the
Archbishops of Hamburg-Bremen, Records of Civilization: Sources and Studies, LIII
(New York, 1959). Western Slavs: Saxo Grammaticus, Danorum Regum Heroumque
Historia, XIV.39, transl. E. Christiansen, Saxo Grammaticus, Danorum Regum Heroumque
Historia Books XXVI, 3 vols., BAR International Series 84 (Oxford, 1980).
60
H. Zoll-Adamikova, Die Einfhrung der Krperbestattung bei den Slawen an
der Ostsee, Archologisches Korrespondenzblatt 24 (1994), pp. 8193.
 27

Control over ritual in cemeteries is difficult to infer from archae-

ological evidence alone, but uniformity of grave construction and
burial ritual as well as regular lay-out of cemeteries may be possi-
ble indicators. Elements of uniformity may be seen in the lay-out of
West European Reihengrberfelder of the sixth and seventh centuries,
or in the few and standardized grave-goods of the so-called late
cemeteries of the seventh and early eighth centuries in Anglo-Saxon
England.61 The disappearance of the supposedly pagan grave-goods
custom in Western Europe has often been attributed to the grow-
ing power of the Christian church and its increasing control of the
conduct of rituals, but Christianity is difficult to demonstrate from
grave-goods or grave context alone.62 And as Geary has pointed out,
early medieval burial rites were never controlled by any ecclesiasti-
cal or lay legislation,63 which implies that uniformity was enforced
at a lower level. Because of the importance of mortuary ritual for
the reproduction of society (cf. above), it can be surmised that local
lites attempted to control its conduct. After all, the deposition of
grave-goods continued for up to two centuries after Christianisation,
in many areas until the seventh or even eighth century. Given that
early Christian teaching shows no interest in grave-goods and that
the early medieval Church neither encouraged nor prevented the
deposition of grave-goods as we find them in the graves of nomi-
nally Christian Franks, Alamanni, Goths, Lombards and others, it is
likely that the impetus for the continuation of this custom came from

61
G.P. Fehring, Einfhrung in die Archologie des Mittelalters (Darmstadt, 1987), pp.
6871; A. Boddington, Models of burial, settlement and worship: the Final Phase
reviewed, in E. Southworth ed., Anglo-Saxon cemeteries: a reappraisal (Stroud, 1990),
pp. 17799; H. Geake, The use of grave-goods in conversion-period England c. 600 c. 850,
BAR 261 (Oxford, 1997).
62
The classic study of this problem is B. Young, Paganisme, christianisation et
rites funraires mrovingiens, Archologie Mdivale 7 (1977), pp. 581; for a review
of the archaeological debate, cf. now A. Schlke, Zeugnisse der Christianisierung
im Grabbefund? Eine Forschungsgeschichte mit Ausblick, Ethnographisch-Archologische
Zeitschrift 38 (1997), pp. 45768; A. Schlke, On Christianization and grave-finds,
European Journal of Archaeology 2, 1 (1999), pp. 77106. For a historical perspective,
cf. F.S. Paxton, Christianizing death: the creation of a ritual process in early medieval Europe
(Ithaca and London, 1990), part. pp. 623.
63
Geary, Living with the dead in the Middle Ages, p. 41. The suggestion here of
(social) control of some form over burial ritual which is based on the archaeologi-
cal evidence of the sixth and seventh centuries appears to contradict the established
view, derived from the written sources, that (ecclesiastical) control over burial places
and rituals was only established during the eighth and particularly the ninth cen-
turies; cf. J.M.H. Smith, Religion and lay society, in R. McKitterick ed., The new
Cambridge medieval history vol. II (Cambridge, 1995), pp. 65478, part. 6728.
28

elsewhere, in particular from secular lites who may have found this
a useful vehicle for social competition.

Power over cemeteries

The power to choose a location for a cemetery, and to control access
to, and use of, a cemetery, are closely interrelated aspects as is
demonstrated by the control of early medieval Christian cemeter-
ies.64 Again, this is a subject which is easier to elucidate on the basis
of historical evidence, but one possible archaeological indicator might
be the location of a cemetery or burial area within the confines of
a clearly identifiable complex, such as a churchyard, a monastery or
a palace.65
Control over cemetery locations also appears to have been a ma-
jor factor in the shift of burial sites between the seventh and tenth
centuries in England. Starting in the late seventh century, burial
shifted from the late cemeteries (of possibly hybrid, pagan-Christian
nature) which had been located outside settlements in the landscape,
to the new minster churches serving several communities; these
retained the burial monopoly until the Late Saxon period when bur-
ial rights were increasingly allowed to devolve to the emerging parish
churches of that period.66
A special case is burial in a royal burial church or an Eigenkirche
(proprietary church) built and owned by an aristocratic landowner.
This was a new feature of the post-Roman period, and was estab-
lished by the end of the sixth century; the Frankish royal burials
around Paris and the high-status burials of Arlon in Belgium and
(possibly) Morken in the Rhineland appear to have been among the
earliest, archaeologically documented cases.67

Cemeteries are, thus, places of symbolic and social power: they them-
selves have power of the place; they are places of the power of the

64
Effros, Beyond cemetery walls, pp. 1121.
65
For the existence of a palace cemetery at Cordoba, cf. Chrysties, this volume.
66
Blair ed., Minsters and parish churches; R. Hodges, The Anglo-Saxon achievement
(London, 1989), pp. 1056.
67
See above, nn. 3 and 4; M. Mller-Wille, Knigsgrab und Knigskirche;
 29

dead; and they serve as an arena for the display of the power of
the living.68 Indeed, the early medieval evidence suggests that the
living used cemeteries systematically for the representation and nego-
tiation of power in a manner of speaking, they extracted via ritu-
als power for themselves from the power of the place and the power
of the dead. Some of the aspects of power of, in, and over ceme-
teries are less tangible than others. Those which are particularly
difficult to demonstrate in the archaeological record are the power
of rituals (as against their format), control over rituals, and control
over cemeteries.
The above discussion has also highlighted several aspects of con-
tinuity and transformation. There are instances of continuity of rit-
ual practices from the Roman to the post-Roman period in western
Europe, such as the re-use of older monuments, the provision of
grave-goods, and post-mortem decapitation.69 But this does not mean
that the meaning of such practices remained unchanged; certainly
the range and frequency of grave-goods seems to increase over time,
as does the location of burials at older monuments. A more pro-
found, but also more gradual transformation was the general shift
from extramural to intramural burial between the fourth and tenth
centuries across much of Europe.
But it is another transformation which can tell us more about the
different nature of power in the respective societies. While Roman
cemeteries showed a very strong element of monumental display,
often in the form of mausolea of the leading families, this element
was virtually lacking in the post-Roman cemeteries of western Europe.
In the latter, the emphasis was on the competitive display of the
wealth of the lite in the form of grave-goods. In other words: the
emphasis shifted from offering legitimation via monument construc-
tion to pressing claims via conspicuous destruction or: from geneal-
ogy to potlatch, from long-term to short-term perspective.70 The

id., Knigtum und Adel; H. Roosens and J. Alenus-Lecerf, Spultures mrovingi-

ennes au Vieux Cimetire dArlon, Archaeologia Belgica 88 (1965).
68
For the plurality of power being negotiated in one place, cf. Nelson, this vol-
ume; also, on the power of the living and royal funerals, J.L. Nelson, Carolingian
royal funerals, in: F. Theuws and J.L. Nelson eds., Rituals of power: From late Antiquity
to the early Middle Ages, TRW 8 (Leiden/Boston/Kln, 2000), pp. 13184.
69
Cf. Wood, Spultures ecclsiastiques et snatoriales, pp. 223, for the con-
tinuity of ostentation through the fifth and sixth centuries among provincial elites
of the Rhne valley.
70
An earlier change of a closely similar nature may have happened in the
30

change from Roman drive-through cemeteries to post-Roman dead-

end cemeteries may well be connected with this ritual change: geneal-
ogy can be displayed continuously, while a potlatch-style display is
a one-off affair.
It is arguable that these changes reflect the transformation of a
state society with a comparatively stable social structure, into unsta-
ble snowball societies run by competing predatory elites.71 And it
is surely no coincidence that it was with the stabilisation and cen-
tralisation of these post-Roman polities in the seventh and eighth
centuries that the lite (re)turned to monumental display while the
deposition of grave-goods (which had been going on irrespectively
of Christianisation) ceased in western Europe, but continued in north-
ern and eastern Europe where old relations of power, and tradi-
tional forms of their negotiation, survived for another two or more
centuries.

transition from the earlier to the later Bronze Age in Europe; cf. R. Bradley, The
passage of arms: An archaeological analysis of prehistoric hoards and votive deposits (Cambridge,
1990).
71
P. Wormald, The emergence of Anglo-Saxon kingdoms, in: L.M. Smith ed.,
The making of Britain: The Dark Ages (Basingstoke and London, 1984), pp. 4962, esp.
p. 59.
 TOPOGRAPHY AND THE CREATION OF PUBLIC SPACE
IN EARLY MEDIEVAL CONSTANTINOPLE*

Leslie Brubaker

Our understanding of the topography actual and symbolic and

of the notion of public space in early medieval Constantinople is
grounded in our interpretations of the transformation of urban space
during late Antiquity, a much discussed topic. As an introduction to
a consideration of Constantinople, I will therefore sketch very quickly
the basic patterns of transformation found in sites from the eastern
half of the Mediterranean basin.1 This pattern essentially involves
three key factors.
First, there is a shift away from monumental civic spaces. This
process begins as early as the third century and is nearly though
not entirely ubiquitous by the sixth. Fora could be abandoned and
the buildings pilfered for stone, or the space could be appropriated
for other uses: when the north African city of Cherchel (Roman Iol
Caesarea) was restructured in the 420s or 430s, part of the old forum
apparently became the forecourt of a new church, and part was
given over to small shops;2 at Corinth, a church and cemetery appar-
ently occupied the old forum by the mid-sixth century;3 elsewhere,

* The following paper is substantially the same as that delivered at Bellagio in

the summer of 1998, with modifications that respond to points made in discussion.
I thank the participants in the colloquium for those comments, and Mayke de Jong
for inviting me to participate. My thanks also to Gunnar Branels and Bissera
Pentcheva for discussion and comments on the text.
1
For good overviews, see: J.F. Haldon, Byzantium in the seventh century: The trans-
formation of a culture, rev. ed. (Cambridge, 1997), pp. 92124, 376402, 45961;
idem, The idea of the town in the Byzantine empire, in: G. Brogiolo and B. Ward-
Perkins eds., The idea and ideal of the town between late Antiquity and the early Middle Ages
(Leiden, 1999), pp. 123; and W. Brandes, Byzantine cities in the seventh and
eighth centuries different sources, different histories?, in the same volume, pp.
2557. A good summary of earlier literature is H. Saradi-Mendolovici, The demise
of the ancient city and the emergence of the medieval city in the eastern Roman
empire, Echos du monde classique 32 (1988), pp. 365401.
2
T. Potter, Towns in late Antiquity: Iol Caesarea and its context (Oxford, 1995), pp.
3444.
3
E. Ivison, Burial and urbanism at late antique and early Byzantine Corinth
(c. A.D. 400700), in: N. Christie and S. Loseby eds., Towns in transition: Urban
evolution in late Antiquity and the early Middle Ages (Aldershot, 1996), pp. 1045, 1112.
32

fora became residential areas (as in Luna or Rome, where ninth-

century houses have recently been excavated in the Forum of Nerva,
though the space remained open and was maintained until ca. 800)
or, as famously hypothesised by Sauvaget in 1934, were transformed
into a welter of small shops, which appeared first within the colon-
nades of classical streets and fora and then spilled out to fill the
open space itself.4
The second key feature of urban transformation was the estab-
lishment of walls as a fundamental constituent of the city. As demon-
strated by Cristina La Rocca, walls originally erected for defense
became a defining feature of urbanism during the late fourth and
early fifth centuries in cities of northern Italy;5 recent studies by Jim
Crow and Archie Dunn suggest that across the fifth and sixth cen-
turies walls could take on this same significance in the eastern
Mediterranean.6 It is not that walls were purely symbolic the walls
of Constantinople, added by Theodosios II in 412/3, for example,
extended the earlier fourth-century wall of Constantine, and ran
along the Sea of Marmara, thus providing a second line of defense
on the western edge of the city7 but that they became a marker
of urbanism as well as a protective barrier. Just how important walls
were as a marker of urban identity is clear from images, where the
shorthand for a city is a picture of a walled enclosure.8
A final key feature of urban transformation in the eastern Medi-
terranean is that there seems generally to have been a contraction

4
J. Sauvaget, Le plan de Laodice-sur-Mer, Bulletin dtudes orientales 4 (1934),
pp. 81114; see now also B. Ward-Perkins, Re-using the architectural legacy of
the past, entre idologie et pragmatisme, in: Brogiolo and Ward-Perkins, The idea
and ideal of the town, pp. 22544 esp. 2404. For Rome, R. Santangeli Valenzani,
Edilizia residenziale e aristocrazia urbana a Roma nellaltomedioevo, in: S. Gelichi
ed., I Congresso Nazionale di Archeologia medievale (Florence, 1997), pp. 6470. For Luna,
see B. Ward-Perkins, Urban continuity? in: Christie and Loseby, Towns, pp. 417,
esp. p. 8.
5
C. La Rocca, Public buildings and urban change in northern Italy in the early
medieval period, in: J. Rich ed., The city in late Antiquity (London, 1992), pp. 16180.
6
J. Crow, The long walls of Thrace, in: C. Mango and G. Dagron eds.,
Constantinople and its hinterland (Aldershot, 1995), pp. 10924, esp. 1223. A. Dunn,
The transition from polis to kastron in the Balkans (3rd8th/9th century): general and
regional perspectives, Byzantine and Modern Greek Studies 18 (1994), pp. 6080 does
not develop the point, but it is implicit in much of his discussion. See also Saradi-
Mendelovici, Demise of the ancient city, p. 397.
7
B. Meyer-Plath and A.M. Schneider, Die Landmauer von Konstantinopel (Berlin, 1943).
8
J. Deckers, Tradition und Adaption. Bemerkungen zur Darstellung der christlichen
Stadt. Mitteilungen des deutschen archaeologischen Instituts Rmische Abteilung 95 (1988), pp.
30382.
 33

of urban space.9 The contraction is often attributed to population

decline, and this may indeed often have been an issue. But we should
remember too that contraction is also a result of the more efficient
use of space: the filling in of older public spaces meant that a stable
population could fit into a smaller space. We should also remember
that walls do not lend themselves to simple, one dimensional interpre-
tations: because they have symbolic and hierarchical value in addition
to their protective role, they cannot only be understood as indica-
tive of settlement size. In Constantinople, for example, the Theodosian
wall did not only respond to population expansion: the space between
the Constantinian and the Theodosian walls was, then as now, always
at least partially devoted to market farming.10 Indeed, according to
a recent study of the city Paul Magdalinos Constantinople mdi-
vale the Byzantine capital did not suffer major population decline
in the so-called Dark Ages.11 Whether or not this was the case, the
Constantinopolitan walls remind us that the expansion of urban space
does not necessarily signal a population explosion; and the reverse
is also true: the contraction of urban space as defined by new walls
does not necessarily signal a population reduction. As Andrew Poulter
has shown for the cities of Illyricum, walled areas much smaller than
the existing settlement were erected throughout the sixth century,
but the enclosed space was largely reserved for ecclesiastical and, he
thinks, perhaps military use, with a scattering of shops; settlement
was primarily extramural.12 Poulters most recent excavation, at
Nikopolis in northern Bulgaria, actually found large areas inside the
walls left unbuilt, despite extensive extramural domestic building.13

9
See e.g. C. Foss, Archaeology and the twenty cities of Byzantine Asia,
American Journal of Archaeology 81 (1977), pp. 46986.
10
Called the second area (deuteron pempton): see C. Mango, Le dveloppement urbain
de Constantinople (IV eVII e sicles). Travaux et mmoires du CNRS, monographies 2
(Paris, 1990), pp. 4650; idem, The development of Constantinople as an urban
centre, 17th International Byzantine Congress, Main Papers (New Rochelle, 1986), p. 118
(repr. in idem, Studies on Constantinople [Aldershot, 1993], study I); J. Baldovin, The
urban character of Christian worship. The origins, development, and meaning of stational liturgy.
Orientalia christiana analecta 228 (Rome, 1987), p. 170.
11
P. Magdalino, Constantinople mdivale. Etudes sur lvolution des structures urbaines.
Travaux et mmoires du CNRS, monographies 9 (Paris, 1996), pp. 28, 4850.
12
A. Poulter, The use and abuse of urbanism in the Danubian provinces dur-
ing the later Roman empire, in: J. Rich ed., The city in late Antiquity (London, 1992),
pp. 99135. This is also a standard pattern in north Africa, e.g. at sixth-century
Timgad: J. Lassus, La fortresse byzantine de Thamugadi I (Paris, 1981).
13
Poulter presented this material at the Byzantine Seminar (University of Birming-
ham) in 1998; cf. his first report: Nicopolis ad Istrum, a Roman, late Roman and early
Byzantine city. Journal of Roman Studies monograph 8 (London, 1995).
34

This brief overview does scarce justice to a complex problem, and

I would stress especially that different geographical areas follow
different chronologies of urban spatial development. But what does
seem clear is that throughout the eastern Mediterranean basin, con-
cepts of urban space are being redefined in late Antiquity and the
early Middle Ages. Public urban space has been condensed and con-
solidated, and it clusters around religious buildings (churches or
mosques) rather than around state-dominated civic centres.14 Various
explanations for this shift have been offered, but what I am con-
cerned with here is less why urban space changed than with how
this changed urban space interacted with the rituals of urban life
with topographies of power, but also, and often at the same time,
with processes of urban integration.
It has been suggested, particularly by the late Aleksandr Kazhdan,
that the contraction and consolidation of urban public space in
the Byzantine east went hand in hand with the privatization of daily
life and the increasing importance of the nuclear family.15 There
are problems with this suggestion the urban pattern is not, for
example, confined to Byzantium but my principal worry is with
Kazhdans unexpressed assumption (and this assumption is implicit
in virtually all urban studies that focus on Byzantium) that smaller
public spaces automatically restricted public congress and effectively
killed public social life.
While the forms of ritual public behaviour in a medieval city were
different from those of a classical one,16 the changed form does not
necessarily indicate the atrophy of public life it simply indicates

14
See e.g. Potter, Towns, pp. 6373, 8090; Haldon, Byzantium in the seventh cen-
tury; G. Dagron, Le christianisme dans la ville byzantine, Dumbarton Oaks Papers
31 (1977), pp. 325, esp. 411 (repr. in idem, La romanit chrtienne en Orient [London,
1984], study IX); and for the non-Christian east, H. Kennedy, From polis to
madina: urban change in late antique and early Islamic Syria, Past & Present 106
(1985), pp. 327.
15
E.g.: In the absence of public social life, the one form of association that
flourished was the family, A. Kazhdan and G. Constable, People and power in
Byzantium. An introduction to modern Byzantine studies (Washington DC, 1982), pp. 1958,
quotation at p. 32; following, with modifications, H. Hunger, Christliches und
Nicht-christliches im byzantinischen Eherecht, sterreichisches Archiv fr Kirchenrecht 18
(1967), pp. 30525. See further Dunn, From polis to kastron, pp. 734.
16
The difference is sometimes expressed as a transition from polis to kastron (walled
town): see Dunn, From polis to kastron; A. Kazhdan, Polis and kastron in Theophanes
and in some other historical texts, in: EUCUXIA. Mlanges offerts Hlne Ahrweiler
2, Byzantina Sorbonensia 16 (Paris, 1998), pp. 34560.
 35

that ritual expressions of public life have changed. Public life is not
static, and shifting patterns of behaviour are as likely to indicate
vitality as to signal decay. In Constantinople, changes in one par-
ticular type of ritual behaviour the procession suggest that,
far from declining, the public life of the city expanded during the
early medieval period.
The public procession exemplifies the intersection of the new urban
landscape with a new form of ritual public behaviour. For early
medieval Constantinople, documents record two types of public pro-
cessions, imperial and liturgical. These develop very differently.
Imperial processions, a mainstay of Roman imperial display, became
increasingly less common: as Michael McCormick has argued in con-
nection with imperial victory processions, from the end of the fourth
century . . . ceremony tended to shift from the streets into the cir-
cus, that is, into the hippodrome or amphitheatre.17 This does not
mean that all state-sponsored processions ceased, for in victory cel-
ebrations (for example) the emperor with or without his victorious
general might still parade the streets on route to the hippodrome;
but it does mean that the procession itself became a prelude to a
more weighted event. In terms of the symbolic importance of space,
it would appear that areas associated with imperial secular public
ritual contracted, basically to the hippodrome.18 This compression of
spaces devoted to secular public ritual roughly coincides with the
contraction of the types of urban public space associated with the
Roman polis; and what this suggests is a reevaluation of civic author-
ity over ritual urban space.
At the same time, however, we begin to find widespread evidence
for a new type of public appropriation of urban space: the religious
procession. In 398 and 403 John Chrysostom records two, both
involving the translation of a martyrs relics into Constantinople. The
later account describes the arrival of relics of Phokas, a martyr from
the Pontos: Yesterday our city was aglow, radiant and famous, not
because it had colonnades, but because a martyr arrived in proces-
sion from Pontos . . . Did you see the procession in the forum? Let

17
M. McCormick, Eternal victory: Triumphal rulership in late Antiquity, Byzantium and
the early medieval West (Cambridge, 1986), quotation p. 389; discussion pp. 6479.
18
Richard Lim has argued that this represents the absorption of the secular
by the imperial: Consensus and dissensus on public spectacles in early Byzantium,
Byzantinische Forschungen 24 (1997), pp. 15979.
36

no one stay away from this holy assembly . . . for even the emperor
and his wife go with us . . ..19 It is obvious what is happening here,
at least on one level: Chrysostom is opposing the old values of the
classical polis with new ones (the city is famous not because of its
colonnades but because of its new relic); the symbolic space of the
polis is appropriated (Did you see the procession in the forum?);
and the imperial family participate in the procession rather than act-
ing as a processions goal. The fifth-century historians Sozomen and
Sokrates, the sixth-century historian Theodore Lektor, and the anony-
mous seventh-century author of the Paschal Chronicle provide additional
evidence for religious processions in Constantinople, leading John
Baldovin to remark that such processions were a part of the whole
urban pattern of worship. The liturgy in the city was the liturgy of
the city.20 This will eventually provide the context for the famous
procession around the walls of Constantinople in 626 with the
populace behind the patriarch Sergios, who held aloft a relic and a
portrait of Christ which was credited with repulsing the Avars.21
This material can be followed down a number of paths, two of
which are of particular importance here. The procession groups of
people moving from one significant point to another one in a more
or less organised manner had a long history as an important
component of religious ritual and urban integration, an obvious early
example being the panhellenic procession from Athens proper up to
the akropolis. The processional path was integrated into certain
specific Christian rituals already in the second quarter of the fourth

19
Migne PG 50, col. 699; see Baldovin, Urban character, pp. 182183.
20
Baldovin, Urban character, p. 211.
21
See e.g. J.L. van Dieten, Geschichte der Patriarchen von Sergios I. bis Johannes VI
(Amsterdam, 1972), pp. 1748. A. Cameron, Images of authority: lites and icons
in late sixth-century Constantinople, Past and Present 84 (1979), pp. 335; (repr. in
eadem, Continuity and change in sixth-century Byzantium [London, 1981], study XVIII)
cf. nn. 10, 14. Outside of Constantinople there is also abundant evidence for the
religious procession. Though liturgical processions, as such, are not recorded in
Rome until 590 when Gregory the Great, faced with an epidemic, led a pro-
cession to Sta Maria Maggiore only a year later, Gregory wrote of a proces-
sion from S Lorenzo to St Peters as if it were established practice; either the idea
became established very quickly or earlier processions are simply not recorded: see
Baldovin, Urban character, pp. 15859. In any event, the earliest western religious
procession recorded was instituted by Mamertus in Vienne ca. 470: Sidonius Apollinaris,
writing to him, congratulates him on his successful rogation (prayerful procession),
presumably around the walls of Vienne: Sidonius, Letters VII.I.2, ed. W.B. Anderson
II (Loeb), pp. 28689.
 37

century, as when the pilgrim Egeria describes the Easter procession

at the Church of the Holy Sepulchre in Jerusalem,22 but more wide-
spread application of the idea apparently waits until the very end of
the fourth or the beginning of the fifth century. In 380, the patri-
arch of Constantinople, Gregory of Nazianzus, scorned the proces-
sions of the Greeks, prompting Baldovin to conclude that none
occurred under his leadership.23 Gregorys official attitude notwith-
standing, however, he himself tells us with some enthusiasm about
heading a procession of the Nicene Christians to reclaim a church
from the heretical Arians. The weather was vile, and the Arians
lined the route and rejoiced that God was showing disfavour of the
Nicenes so overtly. But as the Nicene procession reached the church,
the sun broke through the clouds, and the crowd proclaimed a mir-
acle; Gregory preached a sermon in response.24 By ca. 400, as we
have seen, John Chrysostom speaks of two more processions, one of
which he claims has emptied the city in a torchlit procession that
stretched along the coast, making it a river of fire.25 Chrysostoms
processions, and others slightly later in the century (such as the return
of the relics of Chrysostom himself to the city in 438), are also
described by Sozomen and Sokrates, who make it clear that they
were a way of showing Nicene superiority over the Arians: the
Nicenes, favoured by the empress, were given imperial funding for
silver crosses and candles so that their processions might outdo those
of the Arians.26 The theological arguments advanced by both sides
were presumably known in some fashion to much of the populace
of the city. But this should not blind us to the use, apparently by
both sides, of liturgical processions to solidify and to create the sup-
port of the inhabitants of the city. When Chrysostom, in the pas-
sage quoted earlier, urged his listeners Let no one stay away from
this holy assembly . . . or even the emperor and his wife go with us,
he was not just asking the populace to show support for the relic,
but for the Nicene cause. The pro-Nicene faction led by Gregory
of Nazianzus and John Chrysostom was ultimately successful; we

22
See the discussion in Baldovin, Urban character, pp. 5564.
23
Baldovin, Urban character, p. 181.
24
Gregory described the circumstances in his autobiographical poem: Migne PG
37, cols. 112025 (vv. 132591).
25
Migne PG 63, col. 470; discussion in Baldovin, Urban character, p. 183.
26
See Baldovin, Urban character, pp. 183185.
38

now call them the Orthodox. In the late fourth and early fifth cen-
tury, they claimed the city, and its populace, at least in part through
the medium of the impressive and expensive liturgical procession,
which appropriated the topography of the capital through walking it.
The fifth century is also when the great chains of sacred ways
were developed to link Antioch with the monastery, now known as
Qal at Sem an, that grew up around the column where St Symeon
the Stylite sat for most of his life. Symeon died in 459; the monastery
and the routes toward it were developed in the 480s, apparently
mostly at the initiative of the emperor Zeno, who was attempting
to propitiate the hostile population of Antioch.27 Again, topograph-
ical appropriation intersects with a larger strategy of control.
The well-marked routes leading to Qal at Sem an demonstrate that
the idea of creating sacred processional spaces was not confined to
urban contexts: it extended into rural areas as well. The material
provided by John Chrysostom suggests that routes of some sort also
led into the countryside and suburban areas around Constantinople,
though no architectural markers remain to valorise the itineraries.
In a sense, and especially in the case of the Antioch routes marked
by boundary stones and columns, the city colonized the countryside
through the medium of the sacred way; but in another sense, the
rural processional route simply indicates the fragility of any urban/rural
opposition. There is, however, one difference between urban and
rural processional routes that is notable. Unlike, for example, the
Antioch sacred way, the truly urban routes were marked by monu-
ments that existed independently of the processions that used them
as signposts, and the relationship between topography and proces-
sion in such urban contexts differed from that found on the plains
of Antioch or in the suburbs of Constantinople.

There are three main sources of information about Constantinopolitan

processions. The first is the sermons and chronicles already men-
tioned, plus a few later examples such as the Chronicle of Theophanes

27
G. Tchalenko, Villages antiques de la Syrie du nord I, Institut franais darcholo-
gie de Beyrouth, Bibliothque archologique et historique 50 (Paris, 1953), pp.
20577; D. Claude, Die byzantinische Stadt im 6. Jahrhundert. Byzantinisches Archiv 13
(Mnchen, 1969), pp. 20819; J.-L. Biscop and J.-P. Sodini, Travaux Qal at
Sem an, Acts of the 11th International Congress of Christian Archaeology (Rome, 1989),
pp. 167593.
 39

the Confessor (810814).28 The second is the typikon of the Great

Church (Hagia Sophia), a book that contains liturgical instructions
for each day of the year. The earliest surviving copy dates from the
late ninth or early tenth century, but much earlier material has been
incorporated.29 The third and final source of information is the Book
of Ceremonies (De cerimoniis), a book of imperial protocol compiled from
a disparate range of earlier sources in the mid-tenth century.30
This is distinctly a mixed bag, much of it later than the early
medieval period, but liturgical scholars have come to a certain points
of agreement that are relevant to our understanding of Constantino-
politan processions. It is, for example, evident that already in the
fifth century, the main Constantinopolitan liturgy was celebrated at
different stations over the course of the year, and that processions
were an integral part of these liturgical celebrations.31 From the fifth
century on, liturgical processions moved through the major colon-
naded streets on Saturdays, Sundays and certain feast days; the peo-
ple then followed the bishop into a designated church and received
the eucharist. We do not know the exact routes used in the early
period, but we do know some of the key nodal points. In fact, the
nodal points that crop up again and again in the earliest sources
are essentially the same as those that appear in the ninth- and tenth-
century documents, suggesting that the processional routes were basi-
cally set by the sixth or seventh century.32 It would appear, in other
words, that the symbolic topography of Constantinople was estab-
lished in the early medieval period; and that, despite later topo-
graphical shifts that responded to economic incentives,33 the pattern
of symbolic space remained largely unchanged. Walking the same
processional route presumably meant something different in the twelfth
century from what it had meant in the seventh many layers of
social memory had been piled on the site, for a start, and there are
innumerable other reasons besides but the continuity is nonethe-
less worth remarking.

28
Ed. C. de Boor (Leipzig, 1883); transl. C. Mango and R. Scott, The Chronicle
of Theophanes Confessor, Byzantine and Near Eastern history A.D. 284 813 (Oxford, 1997).
29
J. Mateos, Le typicon de la Grande Eglise, 2 vols., Orientalia christiana analecta
165166 (Rome, 19621963).
30
A. Vogt ed., Constantin VII Porphyrognte, Le Livre des Crmonies, 2nd ed., 2 vols.
(Paris, 1967), with Commentaire, 2nd ed., 2 vols. (Paris, 1967).
31
Baldovin, Urban character, pp. 167226, with summaries at pp. 204, 2256.
32
So too Baldovin, Urban character, pp. 2124.
33
See Magdalino, Constantinople mdivale, pp. 5190, esp. 7885.
40

Of the eleven critical nodal points isolated in the sources up to

the tenth century, nine appear in the early sources, and most of
these date from the fifth and sixth centuries. Four are clustered into
what we might call an inner-city group, which consists of
1. Hagia Sophia: the Great Church, dedicated to Holy Wisdom,
between the palace and the patriarchate; in its present form by
532537;34
2. Hagia Eirene: the Old Church, dedicated to Holy Peace, and
perhaps, I wonder, so named in emulation of the Altar of Peace in
Rome; the site is probably pre-Constantinian, but the building was
damaged in the earthquake of 740 and reconstructed sometime after
753;35
3. The church of the Theotokos (mother of God) Chalkoprateia:
this housed the belt of the Virgin Mary and was built in the early
fifth century;36
4. The Forum of Constantine: built by Constantine ca. 330, this
was a major site along the route of more than half of all liturgical
processions. The inclusion of the Forum in the list of inner-city foci
of liturgical processions makes it clear that in early medieval Con-
stantinople ceremonial topography was not rigidly divided between
imperial and ecclesiastical sites; and it suggests that the liturgical pro-
cession subsumed and absorbed at least some of the roles of the old
civic procession.
5. The fifth nodal point appears near the Constantinian walls, on
the second main road of the city. This is the mausoleum of Constantine
and the adjacent church of the Holy Apostles: the mausoleum, built
by Constantine before his death in 337, was also used for some later
imperial burials; the church, which housed the relics of the apostles
Andrew, Luke and Timothy, was built around 360.37

34
See e.g. R. Mainstone, Hagia Sophia. Architecture, structure and liturgy of Justinians
Great Church (New York, 1988), with earlier bibliography.
35
W. George, The Church of Saint Eirene at Constantinople (Oxford, 1912); U. Peschlow,
Die Irenenkirche in Istanbul: Untersuchungen zur Architektur, Istanbuler Mitteilungen, Beiheft
18 (Tbingen, 1977), pp. 2123, 22935; R. Ousterhout, Reconstructing ninth-
century Constantinople, in: L. Brubaker ed., Byzantium in the ninth century: Dead or
alive? (Aldershot, 1998), p. 127.
36
T. Mathews, The early churches of Constantinople, architecture and liturgy (University
Park PA, 1971), pp. 2833.
37
See C. Mango, Constantines mausoleum and the translation of relics, Byzan-
tinische Zeitschrift 83 (1990), pp. 5162, 434; (repr. in idem, Studies on Constantinople
[Aldershot, 1993], study V).
 41

6. Further out still was the shrine of the Theotokos of the Blachernae,
built in the mid-fifth century to house the robe of the Virgin, the
most precious relic of the city.38 This robe, along with a portrait of
the Virgin herself, was carried in the procession around the city walls
to repulse enemies from apparently ca. 800; both the robe and the
portrait performed this role many times in subsequent centuries.
7.9. The final three nodal points are all outside the walls. The
first is the church of the Theotokos of the Source (Pege) of ca. 500,
built on the site of a healing spring.39 The final two are on the major
processional route of the city, which followed the Roman Via Egnatia
from the suburb of Hebdomon, through the Golden Gate (so called
because it was sheathed in gold), through all of the imperial fora,
and eventually to the inner-city core of the Great Palace and the
Great Church (a route of about 10 km). The Hebdomon suburb
was where the army traditionally proclaimed a new emperors acces-
sion when the army was implicated in this ritual (which happened
thirteen times between 364 and 1000),40 and it was the site of two
major churches, the church of John the Baptist (which had the
Baptists head as its major relic) and the church of John the evan-
gelist, both of the fifth century.41

The early sources are not always specific about exactly where a pro-
cession began, or where it stopped along the way. For example, we
know from the fifth-century historian Sokrates that the relics of John
Chrysostom were carried through the city in solemn procession, with
the chanting of psalms, and deposited in the church of the Holy
Apostles but we are not told the exact route.42 The emperor Marcian
(450457) participated in a liturgical procession between the palace
and Hebdomon, but neither the reason nor the precise details are
given in our source, the sixth-century writer Theodore Lektor.43 From
the same source we learn that the patriarch Timothy initiated a

38
R. Janin, La gographie ecclsiastique de lempire byzantin I, Le sige de Constantinople
et le Patriarchat oecumenique 3, Les glises et les monastres, 2nd ed. (Paris, 1969), pp.
16171.
39
Janin, Les glises et les monastres, p. 224.
40
See Baldovin, Urban character, p. 178.
41
Mathews, Early churches, pp. 5561.
42
Sokrates, History of the Church VII, 45: Migne PG 67, col. 856; see Baldovin,
Urban character, pp. 1845.
43
See Baldovin, Urban character, p. 185.
42

weekly procession to the church of the Theotokos at Chalkoprateia

sometime between 511 and 518; again, where it began is unknown,
but that it worked toward urban integration seems likely, for the
Theotokos was the protectress of Constantinople.44 Theodore Lektor
also tells us that from the time of the emperor Anastasios (ca. 496)
until his own day (pre-527) imperial legislation required all ecclesi-
astical processions to be headed by the eparch of the city, to ensure
civic control over the crowd.45 This confirms that religious proces-
sions were tightly woven into the civic fabric of urban life; and it
also underscores the issue of control that was at the heart of all
topographical progresses across the urban (and rural) landscape.
The early sources indicate the importance and the foci of liturgi-
cal processions; they do not give us the abundant detail preserved
in the ninth- and tenth-century typikon of Hagia Sophia or the Book
of Ceremonies. Using those texts as templates, routes can be mapped
out with some assurance,46 but their exact relationship with early
medieval routes is of course open to debate. From the early mate-
rial itself, however, it is clear that in Constantinople, the religious
processions normally follow one of three directional patterns: they
move from site to site, they move out of the city and then back in
again (like the relic greeting procession described by John Chrysostom),
or like the populace of Constantinople in 626 they circuit the
walls. The types of movement here vary: the first is linear, the sec-
ond is intrusive, the third is enclosing.47 All serve to unify urban
space, but the processions from site to site linked adjacent areas of
the city into a cohesive whole; the processional circuit that moved
from the city into the countryside or suburbs imposed urban author-
ity on rural sites; and the significance of the processions around the
walls as a way of asserting and ensuring control over urban bound-
aries is clear. It is also worth noting that all of the processional
routes familiar in early medieval Constantinople describe a type of

44
Ed. G. Hansen, 2nd ed. (Berlin, 1971), p. 140; Baldovin, Urban character, pp.
18586.
45
Ed. Hansen, p. 134; Baldovin, Urban character, p. 186.
46
See, e.g., G. Dagron, Empereur et prtre. Etude sur le csaropapisme byzantin (Paris,
1996), pp. 10612.
47
With better documentation, it might be possible to read meaning into each
type: for an exemplary case study see N. Zemon Davis, The sacred and the body
social in sixteenth-century Lyon, Past & Present 90 (1981), pp. 4070. I thank Chris
Wickham for bringing this article to my attention.
 43

ritual urban space that is quite distinct from that defined by the fora
of the Roman polis, though as we have seen fora could be subsumed
within processional space.
The contraction and consolidation of the use of space evident in
Constantinople and other early medieval cities suggests a reevalua-
tion of the understanding of urban topography.48 While this reeval-
uation may sometimes have been forced upon city dwellers by a
defensive or an economic necessity to draw inwards, this was cer-
tainly not always the case: when a forum falls into disuse and the
area is rebuilt as a church surrounded by a cemetery (as at Corinth)
or as a church with a courtyard and adjacent shops (as at Cherchel),
we are not necessarily dealing with wholescale urban contraction but
rather with the substitution of one type of public space for another.
As this substitution occurs, public use of urban space changes too,
and the static symbolic civic spaces of the Roman polis give way to
an apparently more diffuse and fluid use of urban space, with nodal
points defined by churches and walls but with ritual public space
constantly redefined by procession routes. Rather than a drying up
of public life in public spaces, we actually seem to see an expansion
of the public domain, from a centralised core of heavily weighted
sites around a forum to a decentralized network of sites connected
by the processional routes themselves. Though certain fixed points
recur again and again in the itineraries, the processional routes of
Constantinople were not static: different occasions were marked by
different routes punctuated by different rituals. Over the course of
a year, a high percentage of the populated area of the city was trans-
versed. This assessment makes a characterisation of the early medieval
period as dominated by a contraction of public space problematic:
symbolic and ritualisable public space, now defined by activity rather
than by static monuments, seems rather to have increased as the
centralized Roman polis was transformed and replaced by the early
medieval city. What constituted public space, and how public space
was symbolically appropriated, had changed fundamentally.

48
For a roughly parallel development in Rome, see M. Salzman, The Christiani-
zation of sacred time and sacred space, and R. Lim, People as power: games,
munificence and contested topography, both in: W.V. Harris ed., The transforma-
tions of Vrbs Roma in late Antiquity. Journal of Roman Archaeology, suppl. ser. 33
(Portsmouth, 1999), pp. 12334 and 26581.
This page intentionally left blank
 TOPOGRAPHY, CELEBRATION, AND POWER:
THE MAKING OF A PAPAL ROME IN THE EIGHTH
AND NINTH CENTURIES

Thomas F.X. Noble

It has always been easy, too easy, to think of Rome as a Christian,

or as a papal, city after the Constantinian Peace of the Church.
Within a half-century of Constantines death the imperial govern-
ment had left the city and before another century had passed the
imperial administration had all but vanished from the western half
of the empire. In just this same time frame the papal administra-
tion began to function as a licit and visible public structure and that
papal administration remains in place to this day. Yet there have
been many Romes in the last two and one-half millennia and even
though all their histories tend to overlap it is important to attempt
to identify some boundary markers between their various stages. In
this essay I wish to isolate one unique moment in the long history
of the Eternal City, a moment I have elsewhere characterized as The
Republic of St Peter.
It is my intention to show that it was during the eighth and the
ninth centuries that the popes actually turned Rome into a papal
city. My demonstration will depend on four separate, but as I hope
to prove, intimately related types of evidence: (1) The way the popes
shaped the topography and the built environment of Rome; (2) The
appearance, development, and placement of certain public, promi-
nent art works; (3) The emergence of certain revealing forms of Herr-
schaftszeichen particularly coins, titles, and techniques of dating; (4)
The intensification and elaboration of many kinds of public celeb-
rations designed to represent and articulate papal power and position.
As Federico Marazzi has felicitously put it, late antiquity witnessed
in Rome a church in the city but not yet a city of the church.1
Constantine I promoted Christianity and spent lavishly on the Church.

1
F. Marazzi, Rome in transition: economic and political change in the fourth
and fifth centuries, in: J.M.H. Smith ed., Early medieval Rome and the Christian West:
Essays in honour of A. Donald Bullough (Leiden, 2000), p. 35.
46 ..

But in confining his benefactions to the edges of the city the

Lateran and Santa Croce in Gerusalemme in the extreme south, St
Peters and St Pauls outside the city altogether Constantine
showed discretion.2 That Romes attitude toward his Christian con-
victions was rejectionist may be one part of the explanation for his
relocation to the east.3 When Constantius II visited Rome in the
spring of 357 he ignored both the Christian sites and the ecclesias-
tical personnel of the city.4 Later in that century the family of
Symmachus spent on single occasions many times more money intro-
ducing its scions into public life than the Roman Church realized
in annual revenue.5 In the fifth century, the papacy undertook its
first major building project, the erection of Santa Maria Maggiore.
But the other basilicas of the city which date to this time were with-
out exception aristocratic foundations until the construction of Santo
Stephano Rotondo. And until fairly late in the fifth century Roman
noble families were, for reasons both ideological and sentimental, as
likely to spend money on the ancient monumental heart of the city
as they were on churches.6 It seems slightly exaggerated, therefore,
for Richard Krautheimer to say, almost in the face of his own evi-
dence, that Rome by the early fifth century was a Christian city7
and with Sixtus III, if not before, the papacy took building activity
into its own hands and developed a papal building program.8 Indeed,
when Theodoric visited Rome in 500 he addressed the citizens from
the senate house, stayed on the Palatine, and invested sizeable sums
on the citys secular fabric.9 If the emperors were gone, the great

2
T.D. Barnes, Constantine and Eusebius (Cambridge, MA, 1981), p. 48.
3
R. Krautheimer, Three Christian capitals: Topography and politics (Berkeley, 1983),
pp. 289, 40.
4
Ammianus Marcellinus, The later Roman Empire, transl. W. Hamilton (Harmonds-
worth, 1986), pp. 99103.
5
J. Matthews, Western aristocracies and imperial court A.D. 364425 (Oxford, 1975),
pp. 131, 2434, 384.
6
R. Krautheimer, Rome: Profile of a city, 3121308 (Princeton, 1980), pp. 3358.
7
R. Krautheimer, Three Christian capitals, p. 94.
8
R. Krautheimer, Rome, p. 52. S. de Blaauw, Cultus et decor: liturgia e architettura
nella Roma tardoantica e medievale. Basilica Salvatoris, Sanctae Mariae, Sancti Petri (Vatican
City, 1994), p. 340 says that the creation of Santa Maria Maggiore with its nearby
tituli is the symbol of the completion of the Christianization of Rome. This seems
too early to me.
9
T.F.X. Noble, Theodoric the Great and the papacy, in: Teodorico il Grande e
i Goti dItalia, Atti del XIII Congresso internazionale di studi sullalto medioevo
 47

families were still there, and the popes were not especially promi-
nent.
Theodorics regime failed and Italy was plunged into the turbu-
lence of the Gothic Wars. Those wars occasioned great disruption
and dragged over Italy a thick veil of darkness. Textual witnesses to
central Italys fate in the period from about 550 to 700 are sparse
and reticent. Even so potentially rich a cache of material as the let-
ters of Gregory I is hard to read for lack of a controlling context.
During this period, the conventional wisdom teaches, central Italys
population sagged, its land was increasingly deserted, and its inter-
national connections diminished.10 The last pope to hail from a sen-
atorial family, Honorius I (625638), undertook such projects as
SantAgnese fuori le mura, with its beautiful apse mosaic depicting
the pope as donor, but otherwise the seventh century was without
significant papal building projects or donations. Aristocratic benev-
olence and evergetism are not in evidence either. This was a time
when there was a kind of imperial control in the city. Popes were
bullied on theological matters. Martin I was unceremoniously hauled
off to Constantinople, brutalized, and dispatched to the Chersonnesus
where he died in dreadful conditions. In 662 Emperor Constans II
visited Rome. He had an apparently cordial encounter with Pope
Vitalian and he ostentatiously visited St Peters. But he also dis-
mantled Roman buildings and carried off loot such as would have
warmed the heart of a Visigoth.11 To whom, at this time, local elites
vouchsafed their loyalty, we cannot say. If the emperor and his
exarchs were not always in complete control, they could certainly
throw their weight around in alarming and effective ways. And there
was no Symmachus or Cassiodorus to plead for the safety and well-
being of the citizens.12

(Spoleto, 1993), vol. 1, pp. 395423; J. Moorhead, Theodoric in Italy (Oxford, 1992),
pp. 605.
10
The problems here are immense. For a fine summation see P. Delogu, La
storia economica di Roma nellalto medioevo: Introduzione al seminario, in:
P. Delogu and L. Paroli eds., La storia economica di Roma nellalto medioevo all luce dei
recenti scavi archeologici, Bibliotheca di Archeologia Medievale 10 (Florence, 1993), pp.
1129.
11
L.M.O. Duchesne ed., Le Liber Pontificalis (hereafter LP) (repr. Paris, 1955), vol. 1,
p. 343 (Life of Pope Vitalian).
12
For the general themes in Romes seventh-century history see T. Noble, Rome
in the seventh century, in: M. Lapidge ed., Archbishop Theodore: Commemorative stud-
ies on his life and influence, Cambridge Studies in Anglo-Saxon England 11 (Cambridge,
1995), pp. 6887.
48 ..

In the early eighth century, things began to change. John VII, son
of the curator palatii Plato, briefly contemplated transferring the seat
of the papal government to the Palatine Hill. The church of Santa
Maria Antiqua, with its hauntingly beautiful works of art, is a mon-
ument to his plans. Plans, only. Subsequent popes stayed in the
Lateran precincts. But two generations after Romes pope and mon-
uments suffered at the hands of imperial agents, a pope had to pro-
tect an exarch from a Roman mob that wanted to tear him to pieces
for raising their taxes and trying to arrest the pope. In an ironic
twist on the lit de justice, the pope stashed him under his bed. This
was Martins revenge. Soon fiscal, diplomatic, and religious policies
formed in Constantinople caused the popes to take the lead in sev-
ering central Italys historic ties to the Roman Empire.13
Henceforth, for nearly two centuries, visitors to Rome only sought
out ecclesiastical officials and religious sites. The papal government
grew as never before, both refining institutions and projecting power
in novel ways. One critical feature of this period links it with those
that preceded and followed. By the middle of the eighth century
men whom the Liber Pontificalis labels aristocratic ascended to the
papal office and filled the offices of the papal government. This was
a new aristocracy, formed of urban and rural landholders, who
differed in degree of wealth but not in kinds of wealth from their
late antique predecessors. Until the very end of the ninth century
this aristocracy used the papal administration to create for the first
time a truly papal city. By the tenth century, however, the balance
began to tip definitively toward the aristocratic side of this papal/aris-
tocratic formula. Perhaps it would be better to say that papacy and
aristocracy began a long contest for power in Rome that was occa-
sionally interrupted by western imperial interventions, by communal
movements, and decisively by the papacys removal to Avignon.
To conclude these introductory remarks I invite you to enter St
Peters square with me. Look up to the facade of the new basilica,
to the inscription above the central portals. There you will read
Paulus Borgesius Romanus. Camillo Borghese and Symmachus had a
lot in common but they differed in important ways too. In 400 Rome
was fundamentally an aristocratic city, as it was later in 1610. But
in 1610 the aristocrats controlled the church and thereby presided

13
For basic details, T. Noble, The Republic of St Peter: The birth of the papal state,
680 825 (Philadelphia, 1984), pp. 160.
 49

over a city of the church. In Symmachus day the aristocracy and

the church had little to do with one another. It is in the eighth and
ninth centuries, as we will see, that a church in the city became a
city of the church.

T R C

In looking at the ways in which the popes could use Romes built
environment to project their power, or to create a physical envi-
ronment capable of projecting power, let us consider first their repairs
to or extensions of the citys secular fabric.14 Gregory II, Gregory
III, Hadrian (twice), Leo III, and Leo IV all made extensive repairs
to Romes Aurelian walls and towers. Surely the work of Gregory
II and III, in view of the Lombard threat and the absence of effective
imperial protection, as well as the work of Leo IV, faced as he was
by Muslim incursions and lacking Carolingian protection, make sense
in terms of military exigencies. But the work of Hadrian is intrigu-
ing. He devoted two campaigns to the walls in perhaps the high-
point of the Carolingian Pax Italiae, spending in one instance 100
lbs. of gold. I suspect that this may have been a gesture designed
to display his power and prestige.

14
The basic arguments in the next several pages depend on my Paradoxes and
possibilities in the sources for Roman society in the Early Middle Ages, in:
J. Smith ed., Early medieval Rome and the Christian West (as in n. 1), pp. 5583, where
the essential literature is cited. To avoid an endless multiplication of references here,
I shall cite only works not cited in Paradoxes or works which illuminate partic-
ular points being raised here more explicitly than there. The fundamental source
material is provided by the LP. Fundamental literature: Krautheimer, Rome, pp.
89142; R. Coates-Stephens, Dark Age architecture in Rome, PBSR 65 (1997),
177232 and Housing in early medieval Rome, A.D. 5001000, ibid. 64 (1996),
pp. 23959; L. Ermini Pani, Renovatio murorum tra programma urbanistica e restauro
conservativo: Rome e il ducato Romano, in: Committenti e produzione artistico-letteraria
nellalto medioevo occidentale, Settimane 39 (Spoleto, 1992), pp. 485530; L. Reekmans,
Limplantation monumentale chrtienne dans le paysage urbain de Rome de 300
850, in: Actes du XIe Congrs International dArchologie chrtienne, vol. 2,
Studi di antichit cristiana 41 (Rome, 1989), pp. 861915; B. Ward-Perkins, From clas-
sical Antiquity to the Middle Ages: Urban public building in Northern and Central Italy A.D.
300 850 (Oxford, 1984); S. Gibson, and Ward-Perkins, The surviving remains of
the Leonine Wall, PBSR 47 (1979), pp. 3057 and The surviving remains of the
Leonine Wall. Part II: The Passetto, ibid., 51 (1983), pp. 22239; H. Geertman,
More Veterum: Il Liber Pontificalis e gli edifici ecclesiastici di Roma nella tarda antichit e
nellalto medioevo, Archaeologia Traiectina 10 (Groningen, 1975); S. de Blaauw, Cultus
et Decor, pp. 161200, 335440, 564612.
50 ..

Hadrian also repaired four of Romes aqueducts.15 Later, Gregory

IV and Nicholas I added new work at the Aqua Sabbatina which
brought water to the region of St-Peters and to the mills on the
Janiculum. Sergius II completed additional repairs to the Aqua Jovia,
which carried water to the Baths of Caracalla. The reconstruction
of Romes aqueducts, damaged in the Gothic wars and again by the
Lombards, may be seen as no more than an attempt to make the
city more liveable, but such work may also be taken as a visible sign
of papal power and of the responsibility which the popes had assumed
for their city. I see a possible parallel between the papal concern
for Romes walls and aqueducts and the fourth- and fifth-century
aristocratic solicitude for the citys monumental center even though
the latter was almost wholly propagandistic whereas the former had
immediate, practical motivations.
Papal building of a secular sort was not confined to the city proper.
Gregory III improved the walls of Centumcellae, a strategic hamlet
in south Etruria and, years later, Leo IV completely rebuilt Centum-
cellaes fortifications and renamed the place Leopolis.16 He also
erected anew the walls and gates of Amelia and Orte. Zachary and
Hadrian built substantial buildings on at least three of the newly
founded domuscultae, although it must be admitted that we do not
know what kinds of building these were. Archaeological work dis-
cerns additional construction along the Via Flaminia and at Monte
Gelato. Because the peasants on the domuscultae could be armed,
whereas clerics and the general run of peasants could not be, this
construction in the countryside points to a real projection of papal
power in a region threatened sometimes by foreign incursions and
sometimes by local factional squabbling. Gregory IV, facing the ear-
liest Muslim attacks, fortified Ostia, renamed it Gregoriopolis and,
a generation later, Nicholas I rebuilt the fortifications there. The cre-
ation of the Leonine City by Leo IV, with its three-kilometer-long
circuit of walls, is probably the best known of all these building pro-
jects in the areas outside the city. Less well known, because the evi-

15
See P. Squatriti, Water and society in early medieval Italy, 4001000 (Cambridge,
1998), pp. 1121 for apposite comments on the practical and ideological significance
of aqueducts.
16
L. Pani Ermini, Leopoli-Cencelle: Una citt di fondazione papale, in: G. De
Boe and F. Verhaege eds., Urbanism in Medieval Europe (Zellik, 1997), vol. 1, pp.
36973 came to my attention after I had submitted Paradoxes.
 51

dence is mainly epigraphic, is the borgo constructed by John VIII

around San Paolo fuori le Mura.
In all that concerns ecclesiastical construction, the sources provide
an embarrassment of riches. For purposes of this analysis, I will
divide ecclesiastical building into two kinds. The first involves what
might be called, aptly if inelegantly, the popes home office and
the second concerns the churches and other religious establishments
of the city.

Todays visitor to the Vatican cannot help but notice the large, ram-
bling complex of buildings that form the nerve center of the papal
administration. Was there such a place in the Republic of St Peter?
We cannot say in detail what the administrative complex looked like.
Many literary descriptions are inexact and many buildings have been
overbuilt so many times that we cannot be too confident about their
original locations, shapes, and functions. And, it must be added, we
have no idea where most of the various branches of the papal admin-
istration had their precise locus of activities.17 Still, these qualifications
notwithstanding, we cannot help but be struck by the papacys atten-
tion to the Lateran and Vatican18 regions. I wish to point out, albeit
in rather summary fashion, three aspects of that attention.
First, the primary ceremonial spaces of the Lateran and Vatican
were repeatedly rebuilt, extended, and remodeled. Lets recall that
the basilica of St John Lateran was and is Romes cathedral church
and that the Lateran palace was the popes normal place of resi-
dence until after Avignon. The importance of the Vatican region
obviously derived from the presumed burial on the Vatican Hill of
St Peter whose vicar the pope was. Although many, many popes
invested heavily in Romes other patriarchal basilica Sta. Maria
Maggiore, San Paolo fuori le Mura, and San Lorenzo fuori le Mura,
their expenditures on those buildings almost pale into insignificance
before the time and trouble they lavished on the Lateran and Vatican.

17
I. Herklotz, Der Campus Lateranensis im Mittelalter, Rmisches Jahrbuch fr
Kunstgeschichte 22 (1985), pp. 142, tells most of what can be known about the dis-
position of official spaces in the Lateran.
18
L. Reekmans, Le dveloppement topographique de la rgion de Vatican la
fin de lantiquit et au dbut du moyen ge, in: Mlanges darchologie et dhistoire de
lart offerts au Professeur Jacques Lavalleye, Universit de Louvain, Recueil de travaux
dhistoire et de philologie, 4th series 45 (Louvain, 1970), pp. 195235 is a detailed
assessment of the construction activity in the Vatican region.
52 ..

The roofs of St Johns and St Peters were rebuilt several times

and the apse vaults of both churches were reconstructed. This work
may have been necessary as routine up-keep but it is worth bearing
in mind how frequently visiting dignitaries appeared at St Peters
which was of course outside the city proper worshipped there,
and then were fted in the nearby papal residences before being
marched across the city to the Lateran. Charlemagnes visit in 774
is revealing. He approached the city just before Easter, went to St
Peters and prayed, and then the king of the Franks begged the
generous pontiff to give him permission to enter the city of Rome
to fulfill his vow of praying in the various churches of God. Permission
was granted, Charles and the pope entered the city together, and
they concluded the days devotions at St-John Lateran.19 There is
archaeological evidence, moreover, that the street system through the
center of the city was repaired and rerouted to effect easier com-
munications between the Lateran and Vatican.20 There was clearly
a need to facilitate dignified celebrations.
Zachary21 added the first triclinium that is, a banqueting and
reception hall to the Lateran. He also built a tower and portico
there and restored the patriarchate whatever that means (the
patriarchate was the papal palace; that is not the problem: We
do not know what restored means). Stephen II, Hadrian I, and
Leo III built towers at St Peters. Leo III added two triclinia22 at the
Lateran one a banqueting and reception hall and the other a
council chamber and another at St Peters, where he also built
a beautiful house, surely a papal residence of some kind. Gregory
IV added yet another triclinium at the Lateran and remodeled the
residential quarters. Sergius II rebuilt the schola cantorum. Leo IV
added a dining room at the Lateran and also constructed a marble
throne in the patriarchate. Nicholas I added a very fine house of
some kind. Terminology kept pace with construction. The Lateran
complex was originally called the episcopium. In the seventh century

19
LP, 1.497.
20
D. Manacorda and E. Zanini, Sul paessagio urbano di Roma nellAlto
Medioevo, in: R. Francovich and G. Noy eds., La storia dellalto medioevo italiano
(VIX secolo) alla luce dellarcheologia, Biblioteca dellarcheologia medievale 11 (Florence,
1994), pp. 63550.
21
The following details all derive from the relevant vitae in the LP.
22
H. Belting, Die beiden Palastaulen Leos III. und die Entstehung einer pp-
stlichen Programmkunst, Frhmittelalterliche Studien 12 (1978), pp. 5583.
 53

it began to be called the patriarchium and then in the Constitutum

Constantini (cc. 13, 14), the famous Donation of Constantine forged
in, probably, the 750s or 760s, the word palatium was used for the
first time. In the 820s this word appeared in the LP.23
In the second place, the public spaces surrounding both the Lateran
and Vatican were repeatedly spruced up. Hadrian I added a gallery
and a portico at St Peters. He also fetched 12,000 tufa blocks to
rebuild the porticus from the Tiber embankment to St Peters and
he refashioned the great steps and porticoes of the basilica. Leo III
remodeled the atria. He also restored the porticoes and their roofing
at the Lateran. Sergius II reconstructed the entry zone at the Lateran
and Leo IV re-roofed the portico there. Visitors, clerics, and Romans
generally cannot have missed this constant attention to the areas sur-
rounding Romes major palaces and basilicas. These atria, porticoes,
porticus, and esplanades were precisely the places where crowds
awaited the outcomes of papal elections, through which papal funeral
processions passed, where synodal decrees were proclaimed, where
popes addressed the Romans, and where visiting dignitaries were
received.
These spaces also attracted visitors, humble pilgrims and mighty
emperors alike. This brings me to the third aspect of papal build-
ing around the Lateran and Vatican, the one devoted to caring for
Romes visitors. Beginning with Hadrian I and continuing right down
to Nicholas I, popes built or remodeled bath houses, temporary res-
idences, fountains, food distribution centers, and chapels or orato-
ries in immense numbers and then joined them with paved walkways,
covered porticoes, and flights of steps.
All of this attention to the home office, as I have called it, is
impressive in both its scale and its continuity over time. Some of
this work will naturally have been deferred maintenance inherited
from the relatively penurious sixth and seventh centuries. But much
of it is reminiscent of the creation of important public, ceremonial
spaces in Constantinople in the fourth century, of the centuries of
work on secular Romes public center, and of the better documented
construction in later Italian cities such as Florence and Venice. Such
spaces always helped to assemble and consolidate human commu-
nities, to give form and meaning to their celebrations, and to build

23
De Blaauw, Cultus et decor, p. 163. The LP reference is puzzling as it comes
from the vita of Valentine who was pope for only 40 days in 827.
54 ..

civic consciousness and pride. So it was, I believe, in the Republic

of St Peter.
In addition to work on the home office, there was general eccle-
siastical construction in the city that amounts to a veritable boom
beginning in the second half of the eighth century. This work con-
stitutes another multi-faceted aspect of the papacys projection of its
power and influence. In the years from 715 to 891 the Liber Pontificalis
supplies information on 263 separate construction projects. 174 of
these projects took place in the city, 89 of them outside. Among
these projects, only 18 were secular, although some of these were of
such a magnitude Romes walls or aqueducts as to suggest
that, as a percentage of the actual work accomplished, they really
represented a good deal more than a mere 6,4% of the total effort
and expense. The two pontificates of Hadrian I and Leo III account
for 147 of these projects, 56% of the total number but in reality a
much larger share of the actual effort and expense.
This construction work was accompanied by the donation to Roman
churches of at least 5.232 cloths of which more than 4.400 were
silk.24 Some of these were plain colored veils but many were adorned
with images, jewels, and gold embroidery. Virtually every major
church in the city got sets of intercolumnar veils for its nave, plus
altar veils and cloths. A few churches got huge curtains that hung
before their front portals or in their triumphal arches and a small
number of establishments got tapestries. In addition to all of this
cloth, popes donated precious metalwork in profusion. Lighting fixtures
and liturgical paraphernalia made up the bulk of metallic donations
but other items such as book covers, gilded railings, and repousse
images appeared as well. In all, not less than 4.480.4 lbs. of gold
and 45.867.51 lbs. of silver were donated to Roman churches.25 68%

24
In addition to my comments in Paradoxes see also P. Delogu, Limportazione
di tessuti preziosi e il sistema economico romano nel IX secolo, in: P. Delogu ed.,
Roma medievale: aggiornamenti (Florence, 1998), pp. 12341. Delogu deals with 3.738
silk and linen cloths. My totals differ from his partly because we counted a little
differently and partly because he confined himself to the pontificates running from
Hadrian I to Leo IV (772855) whereas I counted the precious cloth from LP evi-
dence across the eighth and ninth centuries.
25
Again, in addition to my comments in Paradoxes those of Delogu are crit-
ical: The rebirth of Rome in the eighth and ninth Centuries, in: R. Hodges and
B. Hobley eds., The rebirth of towns in the West A.D. 700 1050, CBA Research Report
vol. 68 (London, 1988), pp. 3342 and Oro e argento in Roma tra il VII e il IX
secolo, in: Cultura e societ nellItalia medievale: Studi per Paolo Brezzi, Istituto Storico
Italiano, Studi Storici vols. 18487 (Rome, 1988), pp. 27393.
 55

of the gold was donated by Hadrian I and Leo III and 48% of the
silver by Leo alone. These are the same two popes in whose reigns
we noticed the cluster of construction of activity.
The evidence for ecclesiastical construction and donations can only
receive a few comments here. First, although the greatest quantity
of both building and benefaction clusters between 780 and 840, each
enterprise was actually quite consistent over the whole period of the
Republic of St Peter. To be sure the 780s saw a building upsurge
that still staggers the imagination and that would test the financial
resources of almost any polity. This, I think, was Hadrians proud
gesture as much or more than this popes attempt to refurbish a
dilapidated ecclesiastical establishment. In 807 Leo III made the
largest single set of donations to Roman churches in the whole repub-
lican period, maybe in the whole Middle Ages. Let us recall that
this was a pope who was attacked violently in 799 and whose rural
estates were plundered at about the same time, as they would be
again in 815. I have a hunch that Leo was seeking to remind peo-
ple who controlled the purse strings. Leo IV showed, comparatively
speaking, more concern, in terms of both bricks and mortar and
precious gifts, to the churches of Romes countryside than any other
republican pope. I suspect that this was his way of reassuring local
populations that the papacy was still a functioning presence in Romes
territorium after the devastating Muslim raids of recent years. In sum,
the sheer consistency, dispersion, and quantity of papal building and
donation was one way of projecting power.
Second, while I do not want to suggest that the papacy instituted
a kind of New Deal, I do think that it is legitimate to point out
that all of the construction work will have provided continuous
employment for a sizeable number of people over a long period of
time. This is patronage on a pretty grand scale. Patronage is one
very important kind of power and that power was displayed with-
out interruption for a century and a half.
Third, it is worth noting that these construction projects, and their
attendant donations, extended very comprehensively over the terri-
tory of the city. Indeed, reading the LP in conjunction with a map
of Rome would be enough to convince anyone that work on the
patriarchal basilicas, the title churches, the basilican monasteries serv-
ing the major churches, and the deaconries (food distribution cen-
ters), to say nothing of the citys other main churches and monasteries,
made a mark on almost the whole area inside the walls. This topo-
graphical imperialism prompts a few remarks.
56 ..

Work will have been going on, often in many areas simultane-
ously, all over the city. Again and again we read of popes going out
personally to supervise the work. In fact, we even read that Gregory
IV took a share in the work. It appears that the papal vestararius
may, at least sometimes, have been the master builder. He regularly
visited construction sites. Such visits marked a kind of presence, a
kind of power, or at least the textual representation of such power.
The major ecclesiastical complexes tended to exist in proximate
groups: say a basilica, one or more monasteries, and a deaconry. In
neighborhoods all over the city people will have worked, or will have
watched work being undertaken, will have gotten their food, and
will have worshiped in relatively compact and carefully articulated
spaces. From one point of view it might be suggested that there were
branch offices of the home office all over the city. Putting a good
sized community into the rhythmic vibrations of its quotidian exist-
ence represents a significant kind of power.

T M A

Art talks. But to understand what it tries to say we may find some
help in the words of Clifford Geertz: The definition of art in any
society is never wholly intra-aesthetic, and indeed but rarely more
than marginally so. The chief problem presented by the sheer phe-
nomenon of aesthetic force, in whatever form and in result of what-
ever skill it may come, is how to place it within other modes of
social activity, how to incorporate it into the texture of a particular
pattern of life.26
Eighth- and ninth-century Rome witnessed the production and
display of an astonishing quantity of art that was usually of high
and sometimes of breathtaking technical mastery. Viewed from one
angle this art was decoratively pleasing. Viewed from another angle
it was spiritually uplifting. Some of that art was didactic, and some
of it was polemical. Now we must ask, with Geertz, how to place
that art into modes of social activity and how to incorporate it into
the texture of life. Images of many kinds were either repaired or
made anew in countless Roman churches. It is almost impossible to

26
C. Geertz, Art as a cultural system, in his Local knowledge: Further essays in
interpretive anthropology (New York, 1983), p. 97.
 57

imagine the richly colored environment that surrounded the daily

worshipers and occasional visitors to those churches. Huge apse
mosaics were everywhere. Triumphal arches and side walls were cov-
ered with frescoes. Altars often had cloths with images hanging from
their fronts and sides. Altar zones and the spaces between the columns
of naves were adorned with beautiful, colored silk cloths that often
had images. Some churches housed icons. Repouss images existed
in profusion. These pictures illustrated Gospel stories, proclaimed key
doctrines of the church, and reminded viewers of the holy men and
women especially commemorated in particular places. Their icono-
graphic themes reached far back into antiquity and their particular
styles drew from enough wells of inspiration to lubricate interminable
art historical investigations. Their placement in churches will undoubt-
edly have buttressed the other messages broadcast by the construc-
tion and donation records. Yet these are not the aspects of the
pictures that I want to discuss here. Instead, I want to focus briefly
on the papal portraits that festooned the churches of the city. Then
I want to turn in particular to images of Mary and of Jesus and to
the relationships between those representations and the popes who
associated themselves with them.
Based on the evidence assembled by Ladner, it seems safe to say
that popes of the republican period represented themselves some-
what more frequently that had been the case in earlier centuries.27
Often the popes depicted themselves as donors or else associated
themselves with Christ, Mary, St Peter, or particular saints. There
is nothing striking about any of this. But I think that some of these
images are legible as a series of drafts of an ideological program
while others point to very specific political situations in eighth- and
ninth-century Rome. It must be acknowledged, too, that any one
image might have multiple meanings that are not mutually exclu-
sive and that only some of those meanings may be recoverable by us.
John VII, the pope who contemplated moving the papal admin-
istration to the Palatine district, associated himself, in a fresco on
the third level of the palimpsest wall to the right of the apse in Santa
Maria Antiqua,28 at the foot of the Palatine on the forum side, with

27
G. Ladner, Die Papstbildnisse des Altertum und des Mittelalters, vol. 1 Bis zum Ende
des Investiturstreits, Monumenta di antichit cristiana, 2d series, 4 (Vatican City, 1941).
28
P. Romanelli and P.J. Nordhagen, Santa Maria Antiqua (Rome, 1964), pp. 326
for the stages in the painting of this church.
58 ..

a Maria Regina figure (fig. 1). He put himself in the same associa-
tion in the Oratory of Forty Martyrs just outside that church. He
also portrayed himself with a Maria Regina in a large mosaic in the
Marian oratory of old St Peters (figs. 2 and 3).29 John may have
been the first pope who, while he was still alive and reigning, put
up images of himself.30 The extant images may be only a small por-
tion of the ones that once existed. The Liber Pontificalis says that he
provided images in various churches so that anyone who wants to
know what he looked like will find his face depicted on them.31 John
is depicted with the square nimbus that has always been taken to
be a sign of a living person. Yet the sign is not always confined to
the living and, according to a fascinating hypothesis never refuted,
to my knowledge of John Osborne the inspiration for the square
nimbus may be Egyptian funerary plaques and the point of the image
may have been to convey a life-like representation, to show what
he looked like.32
Zachary had himself depicted with an enthroned Mary in the
chapel of Sts. Quiricus and Julitta, sometimes called the Theodotus
chapel (for reasons to which we will return), to the left of the pres-
byterium in Santa Maria Antiqua. Paul I appears with a madonna in
the apse of Santa Maria Antiqua and Hadrian I portrayed himself,
in a fresco in the atrium of the same church, with a Maria Regina.
Leo III appears with Peter and Charlemagne in the famous triclin-
ium mosaic in the Lateran, about which we will have more to say,
and also in the apse of his former title church, St Susanna, again
with Charlemagne. Paschal I appears kneeling before an enthroned
virgin in the apse mosaic of Santa Maria in Domnica (see figure 4),
with a Pantocrator in the apse of Santa Prassede, and with Christ
and various saints in the apse of Santa Cecilia in Trastevere. Gregory
IV appears with Christ and saints in the apse of San Marco and
apparently with a madonna on a magnificent altar cloth which he
donated to Santa Maria in Trastevere. Leo IV, finally, can be seen
with a madonna in the lower church of San Clemente. Apparently
he was also depicted on a silver image placed at St Peters tomb;

29
G. Ladner, Papstbildnisse, pp. 8898.
30
C. Bertelli, La Madonna di Santa Maria in Trastevere: Storia, iconografia, Stile di un
dipinto dellottavo secolo (Rome, 1961), p. 62.
31
LP, 1.385; cf. Ladner, Papstbildnisse, p. 90, fig. 88.
32
The portrait of Pope Leo IV in San Clemente, Rome: A re-examination of
the so-called square nimbus in medieval art, PBSR 47 (1979), pp. 5865.
 59

on an altar cloth at St Peters with Jesus, Peter, and Paul; on a silk

cloth donated to the oratory of St Andrew; on a cloth presented to
St Martins monastery at St Peters; on two cloths donated to SS.
Quatro Coronati depicting Jesus and the martyrs; on a silk cloth
depicting Lawrence given to San Lorenzo fuori le mura; and in a
nearby chapel of Mary on a cloth depicting the resurrection.33
Before turning to the ways in which these popes portrayed them-
selves with Mary and Jesus, there are three preliminary comments
to be made about this profusion of papal images in the eighth and
ninth centuries. First, the popes occasionally appear as donors
good examples are Paschal I in Santa Prassede and Santa Cecilia
in Trastevere or Gregory IV in San Marco. This is a very tradi-
tional kind of representation and too much should not be made of
it. But I would like to call attention once again to the contempo-
rary construction and renovation work accomplished by the popes
and to suggest that the papal images had the particular force of
reminding people just who was responsible for all of this work and
for the social and economic stimulus that it represented and pro-
moted. Construction speaks a language all its own. In important
ways that language addresses the possession and use of wealth in a
specific time and place. Wealth used on impressive public ameni-
ties and churches were surely functional substitutes for the more
obviously public baths, temples, and basilicas of antiquity was
meant to forge paternal and communal bonds.34 The papal images
were, I think, additional reminders of those bonds.
Second, the images which one pope after another distributed all
over Rome are reminiscent of the imperial images, and even of the
image-bearing consular diptychs, that had long been disseminated in
the Roman Empire. Pope Constantine in the early eighth century
refused to receive the images of the heretic emperor Philippicus
Bardanes35 and there is no secure testimony that any subsequent
imperial images were received and displayed in Rome. It seems likely
that this characteristic imperial prerogative was usurped by the popes

33
Ladner, Papstbildnisse, pp. 98152, with figs. 90, 100, 106108, 109, 110, 113,
114, 115, 116. For the textual references pertaining to Gregory IV see LP, 2.80
and for Leo IV, ibid., 2.109, 111, 113, 119, 120, 129.
34
Excellent comments in P. Brown, Power and persuasion in late Antiquity (Madison,
WI, 1992), pp. 824.
35
LP, 1.392.
60 ..

in the very years when they were achieving autonomy from the
empire.36
Third, the actual places where the extant images can be found,
or where images are known to have existed, are revealing of polit-
ical and ideological cross-currents in eighth- and ninth-century Rome.
I have already mentioned John VIIs putative transfer of the papal
administration to the Palatine. Whatever is to be thought about that,
he put the papal mark decisively and unambiguously on one of the
most Roman and imperial spots in Rome. Let us return now to
the Theodotus chapel at Santa Maria Antiqua. Zachary faced acute
Lombard pressure and had no serious prospects of outside assistance.
He and his predecessors had been steadily unraveling the ties that
had so long bound Rome to Constantinople. As the shadows length-
ened over the Byzantine rule of Rome, Roman aristocrats began to
enter the papal administration. Can it be a coincidence that Zachary
appears in an image with Theodotus, a Roman blue-blood and the
uncle of the later Pope Hadrian I? Is it surprising that Paul I, another
aristocrat whose own brother Stephen II preceded him, also appears
in Santa Maria Antiqua, a place with, by now, both historic and
social significance? Then Hadrian placed his own likeness in the
same place, a building that was becoming a papal showcase. Leos
image in his triclinium mosaic is most frequently interpreted in terms
of regnum and sacerdotium problems. We will come to those issues
shortly. But cannot it also have been a not-so-subtle reminder to
various members of the Roman elite, Leos bitter opponents in the
family of Hadrian I first of all, that Leo was not without powerful
friends? Paschal I, who experienced intense political opposition, put
his picture all over Rome. In doing so, he claimed the mantles of
Christ and of Mary. Is this a coincidence in view of his manifest
need for protection and his having to deal with the Constitutio
Romana of 824 that could have been taken as calling into question
the autonomy of papal rule in Rome?37 Gregory IV had himself
depicted in San Marco. The rich and noble Hadrian I once lived
near and served in this church. It stood at the end of the up-scale

36
Bertelli, Santa Maria in Trastevere, p. 62; H. Belting, Papal artistic commissions
as definitions of the church in Rome, in: H. Hager and S.S. Munshower eds.,
Light on the Eternal City: Observations and discoveries in the art and architecture of Rome,
Papers in Art History from the Pennsylvania State University 2. (University Park,
PA, 1987), pp. 1330.
37
On the Constitutio Romana see Noble, Republic of St Peter, pp. 30822.
 61

Via Lata district. It was the site of possible aristocratic residential

construction in the late eighth and early ninth century. Can Gregory
have been saying, in this very place, that after the faction-ridden
pontificate of Paschal, all was well again, and the great families were
once more securely in control? Leo IV placed his image in many
places but especially in the vicinity of St Peters which had been
ravaged by Muslims in 846. Not only did Leo undertake massive
construction projects in that zone, but he also put up his likeness to
remind people who their patron and protector was. These are very
interesting ways to think about how the popes could project their
power not only in general terms but also in quite specific circum-
stances.
As noted, the papal images virtually never stand alone. Most often
the popes depicted themselves with Mary or with Jesus. The over-
all scenes usually included other saints too, Peter and Paul most fre-
quently, but sometimes apostles, evangelists, and martyrs. These
images are also revealing of the techniques adopted by the popes to
represent their power and authority and point to identifiable social
and political contexts.

Let us begin with popes who appear with Mary. Because the mag-
nificent Liberian basilica, on whose triumphal arch Mary figures
prominently, was dedicated to Mary later called Santa Maria
Maggiore and because some of the work of construction and dec-
oration falls close in time to the Council of Ephesus (431) that defined
Marys role as theotokos, it was long assumed that Marian devotion
became prominent at Rome from an early date.38 Yet, without enter-
ing into the controversies, it can be stated with some assurance that
the building need not be taken as a reflection of the Ephesine decrees
and that the mosaics on the arch stress Jesus and not his mother
even though the way in which Mary was depicted on the triumphal
arch, in a Magi scene, may have inaugurated an iconographic

38
The older views are well represented in J. Wilpert, La proclamazione Efesina
e i musaici della basilica di S. Maria Maggiore, Analecta Sacra Tarraconensia 7 (1931),
pp. 197213. See now B. Brenk, Die frhchristlichen Mosaiken in Santa Maria Maggiore
zu Rom (Wiesbaden, 1975). For the building, R. Krautheimer, Corpus Basilicarum
Christianarum Romae, vol. 3 (Rome, 1967), pp. 160. The most recent discussion is
D. Russo, Les reprsentations mariales dans lart doccident: Essai sur la forma-
tion dune tradition, in: D. Iogna-Prat, . Palazzo and D. Russo eds., Marie: Le
culte de la vierge dans la socit mdivale (Paris, 1996), pp. 176203, Robert Deshman,
Servants of the Mother of God in Byzantine and Medieval Art, Word & Image, 5
(1989), pp. 3370.
62 ..

pattern with a long future ahead of it. In the sixth century, in Santa
Maria Antiqua (fig. 1), on perhaps the lowest level of the palimpsest
wall, there was an image of Mary that bears some similarity to the
image in Santa Maria Maggiore. Visible now only in fragments, this
image is of the Maria Regina type: Mary is richly garbed, wears
a jeweled crown, and sits on a jeweled throne. In the eighth and
ninth century this type of Marian image, and others that bear some
similarity to it, appeared frequently in Rome. The term Maria
Regina is not a modern confection. 39 In a sixth-century poem,
Corippus speaks of Virgo, creator genetrix sanctissima mundi, excelsi regina
poli.40 These verses are roughly contemporary with the Maria Regina
in Santa Maria Antiqua. In the late eighth century Hadrian depicted
himself with a Maria Regina figure in a fresco in the atrium of that
church. To the right of Marys head there is was, actually; the
fresco was removed by the Istituto Centrale del Restauro in 1956
a vertical inscription reading MARIA REGINA.41 In the porch of
Santa Maria in Cosmedin, not far from Santa Maria Antiqua, there
is an inscription, without its image, that might on paleographical
grounds be dated to the mid-eighth century. It reads Praeclara virgo
caelistis regina superexaltata et gloriosa domina mea Dei genetrix Maria.42 On
an enamel cross of Paschal I, Mary is called regina.43 So a dis-
tinctive type of Marian image, the Maria Regina, appeared in Rome
in the sixth century and then was highly favored in the eighth and
ninth centuries. Ursula Nilgen has called this the madonna of the
popes.

39
For studies of Maria Regina see: M. Lawrence, Maria Regina, Art Bulletin
7 (1925), pp. 15061; U. Nilgen, Maria Regina Ein politischer Kultbildtypus?,
Rmisches Jahrbuch fr Kunstgeschichte 19 (1981), pp. 333; J. Osborne, Early medieval
painting in San Clemente, Rome: The Madonna and Child in the niche, Gesta 20
(1981), 299310; R. Deshman, Servants of the Mother of God in Byzantine and
Medieval Art, Word and Image 5 (1989), pp. 3370; Russo, Les reprsentations
mariales, pp. 20317; M. Stroll, Maria Regina: Papal symbol, in: A.J. Duggan
ed., Queens and queenship in medieval Europe (London, 1997), pp. 173203.
40
Flavius Cresconius Corippus, In laudem Justini Augusti minoris, ed. A. Cameron
(London, 1976), 2, pp. 523.
41
Osborne, Early medieval painting, p. 305.
42
N. Gray, The paleography of Latin inscriptions in the eighth, ninth, and
tenth centuries in Italy, PBSR 16 (1948), p. 55; Osborne, Early medieval paint-
ing, p. 304.
43
C. Morey, The inscription on the enameled cross of Paschal I, Art Bulletin
19 (1937), pp. 5956.
 63

There are many other regina images. The icon from Santa Maria
in Trastevere, the Madonna della Clemenza is a beautiful exam-
ple (fig. 5). It probably dates from the time of Pope John VII.44
There was another, indubitably Johns, in the Marian oratory in Old
St Peters. A substantial part of it is now in Florence. This one is
unusual because it features a solitary orant Mary instead of an
enthroned virgin holding the Christ child. There might have been
one dating from Hadrian in an underground chapel dedicated to
Saints Hermes, Protus, and Hyacinth which that pope restored.45 A
Maria Regina was discovered during restorations in San Lorenzo
fuori le mura in the late nineteenth century but was destroyed.46
Finally, there is one last member of the group in the lower church
of San Clemente dating from Leo IV.47
The period from the sixth to the ninth century witnessed the pro-
duction in Rome of a number of other images of Mary. They are
sometimes hard to date. Their original iconography cannot always
be established with confidence. Repainting and clumsy restorations
render some of them very difficult to interpret. But if the regina-
type represents one way of depicting Mary, then we should at least
note the other prominent way (fig. 6). In this type of image Mary
appears wearing a maphorion (a dark blue, purple, or crimson outer
garment with a hood)48 and she is usually seated on a gilded and
jewelled throne.49 Two excellent early examples are the Turtura
image from the catacomb of Commodilla and the fresco from the

44
A shaky consensus attributes it to John: C. Cecchelli, Mater Christi, vol. 1 (Rome,
1946), pp. 545; Bertelli, La madonna di Santa Maria, pp. 8086; G. Wolf, Salus Populi
Romani: Die Geschichte rmischer Kultbilder im Mittelalter (Weinheim, 1990), pp. 12022
accepts Bertelli. P. Amato, De Vera Effigie Mariae: Antiche icone Romane (Milan, 1988),
pp. 2532 is more cautious and does not take a firm stand. M. Andoloro, La
datazione della tavola di S. Maria in Trastevere, Rivista dellistituto nazionale darche-
ologia e storia dellarte 19/20 (1972/73 [1975]), pp. 139215 holds for a sixth-century
date.
45
LP, 1. 509; E. Josi, Scoperta dun altare e di pitture nella basilica di
S. Ermete, Rivista di archeologia cristiana 17 (1940), pp. 195208; Osborne, Early
medieval painting, pp. 3056.
46
Osborne, Early medieval painting, pp. 305, 3079 (with the older literature).
47
Ibid., pp. 299303, 3079.
48
See s.v. Maphorion, Oxford Dictionary of Byzantium, vol. 2 (New York, 1991),
p. 1294.
49
Some of these will be discussed below. The oldest ones are the so-called
Pantheon Icon, the icon from Santa Maria Nova (now Santa Francesca Romana)
almost certainly once and perhaps originally in Santa Maria Antiqua, the Monasterium
tempuli image now in Santa Maria del Rosario, and perhaps the Salus Populi
Romani about which more just below. See: H. Hager, Rckgewonnene Marienikonen
64 ..

lower church of San Crisogono.50 It is primarily because Mary does

not wear a crown in these representations that causes scholars to
exclude them from the regina-type. Although I believe that the dis-
tinction is useful for strict typologies, I also think that, pushed too
hard, it obscures some features that these images tend to share with
the undoubted regina type. For instance, the magnificent apse mosaic
in Santa Maria in Domnica depicts Mary on an exquisite throne
and garbed in a luxuriously elegant maphorion.51 This image is about
as regal as one could imagine. The small marian image in the Zeno
chapel in Santa Prassede sits on a jewelled throne as does Mary,
this time from a Magi scene, in a fragment of a mosaic of John VII
now in the sacristy in Santa Maria in Cosmedin.52 These images are
remarkably similar to the famous Salus Populi Romani image of
Mary from Santa Maria Maggiore which has been confidently
dated to almost every possible period between the fifth century and
the thirteenth.53 If the Salus Populi Romani image can be placed in
the early middle ages, then it shares something very interesting with the
image in the apse of Santa Maria in Domnica, and with two unques-
tioned regina-type images, those of Santa Maria Antiqua and San
Clemente. Mary in each case holds in her right hand a mappa (or
mappula),54 which had been a consular symbol until the sixth century
when, the emperors having assumed consular authority, it became
an imperial symbol. Connecting these gorgeous, regnant Marys with

des frhen Mittelalters in Rom, Rmische Quartelschrift fr Altertumskunde und Kirchengeschichte

61 (1966), pp. 20916; Andoloro, La datazione della tavola di S. Maria in Traste-
vere, pp. 139215; Amato, De Vera Effigie Mariae.
50
E. Russo, Laffresco di Turtura nel cimiterio di Commodilla, licona di
S. Maria in Trastevere e le pi antiche feste della madonna a Roma, Bullettino del-
listituto storico italiano per il medioevo e archivio muratoriano 88 (1979), pp. 3785, 89
(1980/81), pp. 71150; A. Melograni, Le pitture del VI e VIII secolo nella basilica
inferiore di S. Crisogono in Trastevere, Rivista dellistituto nazionale darcheologia a sto-
ria dellarte, 3rd ser. 13 (1990), pp. 13961.
51
Color image (Plate XX) and full discussion in W. Oakeshott, The mosaics of
Rome (New York, 1967), pp. 2034. The most recent discussion of this image
acknowledges its regal nature but insists, unconvincingly, that it means to show
Mary bodily present in paradise: R. Wisskirchen, Santa Maria in Domnica: ber-
legungen zur frhesten apsidialen darstellung der Thronenden Maria in Rom,
Aachener Kunstbltter 61 (1995/96), pp. 38195.
52
Oakeshott, Mosaics, figs. 126, 106.
53
The most recent and authoritative discussion is Wolf, Salus Populi Romani, esp.
pp. 228 on dating questions. He says, cautiously, that it is probably late antique.
54
See s.v. Mappa, Oxford Dictionary of Byzantium, vol. 2 (New York, 1991),
p. 1294.
 65

an imperial symbol brings us back to why Marian images began

appearing rather suddenly in Rome and what they might mean.
In sixth-century Byzantium images of Mary, but not of Mary alone
to be sure, began to acquire greater prominence and Constantinople
was increasingly placed under Marys special patronage.55 While it
is possible that the earliest Maria Regina in Santa Maria Antiqua
may either fit into a broad context of intensified Marian devotion,
or else be a specifically Roman response to it, several seventh-
century developments in Rome constitute an appropriation of Mary
for the eternal city.56 Pope Boniface IV secured from Emperor Phokas
the former Pantheon which was soon rededicated as Sancta Maria
ad Martyres.57 Mary was, in a sense, now the queen of a heavenly
court whose members were Romes numerous martyrs. In the early
seventh century, too, the old Titulus Juli et Callisti came to be called
Santa Maria in Trastevere.58 Pope Sergius I implemented in Rome
the full set of four annual Marian feasts (The Visitation, 2 February;
The Annunciation, 25 March; The Assumption, 15 August; The
Nativity Marys, 8 September).59 From the time of John VII, as
we have seen, Marian images proliferated rapidly in Rome.
This concentration on Mary marked an important novelty. It may
have owed something to the Marian devotion and liturgical tradi-
tions of resident communities of Greeks and Syrians in Rome, but
local political and ideological explanations are more convincing as
an explanation. Consider the hauntingly beautiful image of Paschal
I kneeling at Marys feet in the spectacular apse of Santa Maria in
Domnica. Mary points to Paschal as her servant. This apse, Paschals
work, is in fact a quotation of a long series of similar Marian images
reaching back to John VII. These images speak the unambiguous
language of intercession. The pope stands closest to Mary who has

55
A. Frolow, La ddicace de Constantinople dans la tradition Byzantine, Revue
de lhistoire des religions 127 (1944), pp. 61127; N.H. Baynes, The supernatural
defenders of Constantinople, Analecta Bollandiana 67 (1949), pp. 16577; A. Cameron,
Images of authority: elites and icons in late sixth-century Byzantium, Past and
Present 84 (1979), pp. 335.
56
In general, T. Klauser, Rom und der Kult der Gottesmutter Maria, Jahrbuch
fr Antike und Christentum 15 (1972), pp. 12035.
57
LP, 1.317.
58
Osborne, Early medieval painting, pp. 303, 309 n. 13.
59
LP, 1.376. Discussion in: E. Palazzo and A.K. Johansson, Jalons liturgiques
pour une histoire du culte de la vierge, in: D. Iogna-Prat, . Palazzo and D. Russo
eds., Marie: Le culte de la vierge dans la socit mdivale (Paris, 1996), pp. 1536.
66 ..

the favored intercessory position in heaven. Santa Maria Antiqua

and Santa Maria in Domnica were both deaconries, food distribu-
tion centers maintained by the papal administration.60 Scenes of papal
intercession located in such places were potent reminders of vivid
daily realities. But intercession is also a matter of profound religious
and theological significance. Accordingly it is well to emphasize the
solemn, moving prayers for intercession that were publicly offered
on the great, and rather recent, Marian feasts.61
In his Marian oratory on old St Peter John placed an inscription
reading beatae Dei Genetricis Servvs (fig. 2).62 The octagonal ambo of
John VII in Santa Maria Antiqua has inscriptions in Latin and Greek
which read: Johannes Servu[s] S[an]c[t]ae M[a]riae and Ivnnou
dolou tw yeotoku.63 As Ladner points out, John was the first pope
to depict himself in a posture of submission to a heavenly figure.
The madonna has now become a ruler and the pope is her first
servitor. John appears, to Ladner, als der Knecht der Mutter Gottes.
By extension, then, John and his successors right down to Paschal
and beyond represented themselves as the dutiful servants of the
queen of heaven and not of any earthly ruler.64 As the popes were
separating themselves from Byzantium, it is striking that they very
quickly fashioned an explicit message saying that their allegiance was
not owed to anyone on earth. Twice in letters to the Frankish court
Stephen II referred to Mary as domina nostra,65 and the (probably)
eighth-century inscription from Santa Maria in Cosmedin mentioned
above says domina mea. Some years ago Guglielmo Matthiae made
the simple but important point that because so much Roman art

60
On the deaconries see below n. 147.
61
Some examples from J. Deshusses ed., Le sacramentaire grgorien: ses principales
formes daprs les plus ancient manuscrits, Spicilegium Friburgense 16 (Fribourg, 1971):
Purification (February 2): no. 126, p. 124: intercedente beata semper virgine Mariae.
Annunciation (March 25): no. 140, p. 128: ut qui vere eam genetricem dei credimus eius
apud te intercessionibus adiuvemur. Assumption (August 15): no. 660, p. 262: intercessio-
nis eius auxilio a nostris iniquitatibus resurgamus; no. 662, p. 262: genetricis filii tui domini
nostri intercessione salvemur; no. 663, p. 263: subveniat domine plebei tuae dei genetricis ora-
tio; no. 664, p. 263: a malis imminentibus eius intercessione libremur. Nativity of Mary
(September 8): no. 680, p. 268: eius intercessionibus complacatus te de instantibus periculis
eruamur.
62
G. Ladner, Papstbildnisse, p. 90.
63
Romanelli in: Romanelli and Nordhagen, Santa Maria Antiqua, p. 16.
64
Ladner, Papstbildnisse, pp. 904; Wolf, Salus Populi Romani, p. 122; Belting,
Commissions, p. 15; Lawrence, Maria Regina, p. 161.
65
Codex Carolinus, nos. 6, 10, ed. W. Gundlach, MGH Epp. 3 (Berlin, 1892), pp.
489, 502.
 67

was didactic it had the power to teach many things simultaneously,

to communicate religious truth along with social and political ideas.66
Daniel Russo has recently stressed that much of the art of eighth-
century Rome can be read as an attempt to claim legitimacy67 and
Gunther Wolf has stressed that the art in the area of Santa Maria
Antiqua, in particular, cannot be without political significance and
surely involved a certain claim of autonomy by the church of Rome.68
It was precisely that autonomy that papal art was seeking to legit-
imize by means of political pictures. And in an autonomous papal
Rome, it goes without saying, the pope was the chief patron and
intercessor in all walks of life.
It is interesting to think of Paschal in this connection too. His
potential rival was not Byzantium but Carolingian Francia. Charle-
magne had dedicated his palatine chapel at Aachen to Mary, and
after many centuries during which there was, apart from Notre-
Dame la Daurade in Toulouse, no church in the west dedicated to
Mary, Marian dedications began to spring up everywhere. What is
more, a remarkable ivory from the court of Charlemagne, called a
Virgo Militans by its most recent student, may also represent a
Carolingian effort at laying claim to Marys patronage and inter-
cession.69 If in that image Mary is called to serve the Carolingians
in war, then in another contemporary one she is figured as a priest.
The first large picture in the Sacramentary of Gellone, itself produced
in the north of France in the 790s, shows Mary holding high a cross-
staff (like the Virgo Militans and a whole series of comparanda
adduced by Lewis) but she is dressed in an ephod like a levite priest
and carries a censer.70 Paschal, whose political machinations in Rome
brought on repeated investigations and the issuance of the Constitutio
Romana, may have depicted himself as the servant of Mary, and as
the primary conduit for her benevolence, precisely to remind the
Carolingians that he stood closer to heavens queen than they did
and to remind the Romans that, Franks or no Franks, their salva-
tion depended on Marys magnanimity and the popes prayers. Leo

66
G. Matthiae, Pittura politica del medioevo romano (Rome, 1964), pp. 513.
67
Russo, Les reprsentations mariales, pp. 1736, 1915.
68
Wolf, Salus Populi Romani, pp. 120, 121.
69
S. Lewis, A Byzantine virgo militans at Charlemagnes court, Viator 11 (1980),
pp. 7193. Cf. Russo, Les reprsentations mariales, pp. 21823.
70
Liber sacramentorum Gellonensis, vol. 2, ed. Dumas, plate 1 (= Paris, BN Lat.
12048, fol. 1v).
68 ..

IVs invocations of Mary, finally, may have had the same double
audience: The Romans who had suffered Muslim attacks and the
Carolingians in the person of Louis II whose attitudes to Rome and
its region were initially undeclared. The language of this papal art
is unambiguous and its social and political contexts are not difficult
to discern.
Popes also represented themselves in connection with Jesus. For
purposes of the present study I shall confine myself to the images
erected by Leo III in his two Lateran triclinia. The construction of
triclinia was itself a political act. Zachary had built one at the Lateran,
but nothing is known about it.71 Leo then built two, one between
795 and 80072 and another later in his pontificate (fig. 7).73 A gen-
eration later Gregory IV built yet another.74 Leos first one, the aula
Leonina greater than all the other triclinia was a triconch
structure. This architectural form took its rise in the domestic archi-
tecture of fantastically rich western families in imperial times and
then became, especially in the east, a symbol of imperial rule. Between
late antiquity and the pontificate of Leo, no triconch buildings were
built in the west, as far as is known. Then, within just a few years,
Leo built two and Charlemagne built one. In the 830s, Emperor
Theophilus built one in Constantinople. These buildings were sym-
bols of sovereignty.75 No less symbolic, however, was Leos second
triclinium, the sala concilia. This apsidal structure with five flanking
conches on each side quoted the chamber of nineteen couches in
the imperial palace at Constantinople.76 It was Leo who gave to the
new papal state a proper administrative and ceremonial center. Of
the aula nothing remains today except the apse with its heavily

71
LP, 1.432.
72
Charlemagne is called rex in the inscription so the assumption has always
been that the triclinium was erected before he became emperor in 800.
73
LP, 2.34, 11.
74
LP, 2.76.
75
I. Lavin, The house of the Lord: Aspects of the role of palace triclinia in the
architecture of late Antiquity and the early Middle Ages, Art Bulletin 44 (1962), pp.
414.
76
Belting, Palastaulen, pp. 689; P. Verzone, La distruzione dei palazzi impe-
riali di Roma e la restrutturazione del palazzo lateranense nel IX secolo nel ra-
porti con quello di Costantinopoli, in: Roma e let carolingia: Atti del giornati di studio
38 maggio 1976 (Rome, 1976), pp. 407. The sala is obviously not identical to the
chamber of nineteen divans in Constantinople. The word quote is used in
Krautheimers sense: An introduction to the iconography of medieval architec-
ture, Journal of the Warburg and Courtauld Institutes 5 (1942), pp. 133.
 69

restored mosaics.77 The sala was dismantled in the sixteenth century

when Domenico Fontana constructed the present Lateran palace.
The apses in these triclinia were outfitted with mosaics whose icono-
graphic themes are at once religious and political. Leos work in cre-
ating these apse mosaics is distinctive in two ways. First, monumental
apse mosaics had not been made in Rome in many years. Leo inau-
gurated a renewal of tradition that would, in artistic terms, reach its
highpoint in the pontificate of Paschal I.78 Second, religious art with
powerful meanings had previously been confined to churches in Rome
and it was Leo who shifted the use of this art to more public spaces
where those who saw the images would be dignitaries and not wor-
shipers.79
The apsidial arch in the aula has, on its right side, the famous
investiture scene in which an enthroned St Peter presents Leo, to
his right, with a pallium, and Charlemagne, to his left, with a ban-
ner. This scene was one of a pair. The one on the left, now largely
an eighteenth-century restoration of Cardinal Francesco Barberinis
early seventeenth-century recreation of what he believed must have
been there, depicts Jesus giving keys to St Peter, or maybe to Pope
Sylvester, and a banner to Constantine. It seems likely that there
was a scene of some kind depicting Constantine, but one cannot be
too confident about just what it looked like.80 In any case, these two
scenes only take on their full meaning in conjunction with the mosaic
in the conch of the apse.
The scene in the apse shows Christ standing on the hill of par-
adise with eleven apostles arrayed to his right and left. Immediately
to Christs right the viewers left is an agitated image of Peter
holding a cross-staff. To create symmetry the mosaicist gave Peter,

77
The surviving, free-standing apse mosaics are deceptive for several reasons, not
least because Benedict XIV rotated the remnant 180o in his restoration. Belting
gives a good account of the history of the monument and its art work: I mosaici
dellaula Leonina come testimonianza della prima renovatio dellarte medievale di
Roma, in: Roma e let carolingia, pp. 16769.
78
C. Davis-Weyer, Die Mosaiken Leos III und die Anfnge der karolingischen
Renaissance in Rom, Zeitschrift fr Kunstgeschichte 29 (1966), pp. 11132. Davis
acknowledges Leos innovations but still believes that he inaugurated an adaptation
of eighth-century Greek styles that reached fruition in the time of Paschal I. Belting
refutes this view and attributes to Leo a genuine renovation of paleochristian styles:
I mosaici, pp. 16971.
79
Belting, Palastaulen, p. 73.
80
Belting, Palastaulen, pp. 656 and I mosaici, pp. 17172 makes good case
for a Constantine image of some kind.
70 ..

his staff, and his billowing garments a double space. This strategy
creates harmony between the six apostles to Christs left and the five
to his right. Christ holds an open book in which one can read Pax
vobiscum. The inscription beneath the mosaic ties the whole image
together. It is the Great Commission from Matthews Gospel (28.
19): Teach all nations, baptizing them in the name of the Father,
and of the Son, and of the Holy Spirit . . . And behold I am with
you always even to the end of the world (fig. 8).81
The meaning of the group of images is this. Christ gave his au-
thority to the apostles and his keys to Peter. Peter now transmits
authority to his vicar Leo and to his protector, Charles (as once to
Constantine). The independence of the church is proclaimed in both
spiritual and temporal terms. Constantine appears not as a ruler but
as a benefactor. Charlemagne appears not as a ruler, not as an
earthly authority to whom the pope owes obedience, but as a pro-
tector of the popes church.82 In my view, this image is, at least in
part, Leos answer to the famous letter which Charlemagne sent him
in 796.83 That letter assigned to the pope a rather passive role; he
was merely to raise his hand in prayer for the success of the Franks.
Leos triclinium mosiac assigns the pope and the papal office a more
dynamic role.
The images in the sala carry the meaning forward. This time the
apse and front wall quote the triumphal arch of San Paolo fuori le
mura. Here Peter and Paul flank Jesus, standing as representatives
of authority in the church and in doctrine, while each of the ten
conches had one apostle preaching to the nations. This set of images
completes the ones from the aula while, in Beltings words, empha-
sizing the primacy of the Roman Church in a surprisingly concrete
and also in a surprisingly political manner.84

81
Cf. Belting, Palastaulen, figs. 35.
82
Belting, Palastaulen, pp. 627; C. Walter, Papal political imagery in the
medieval Lateran palace, Cahiers Archologiques 20 (1970), pp. 17476; G. Ladner,
I mosaici e gli affreschi ecclesiastico-politici nellantico palazzo lateranense, Rivista
di archeologia cristiana 12 (1935), pp. 267, 291; K. Schatz, SJ, Knigliche Kirchen-
regierung und rmische Petrus-berlieferung im Kreise Karls des Grossen, in:
R. Berndt ed., Das frankfurter Konzil von 794: Kristallisationspunkt karolingischer Kultur,
Quellen und Abhandlungen zur mittelrheinischen Kirchengeschichte 80 (Mainz,
1997), pp. 35771.
83
Alcuin, Epistolae, no. 93, MGH Epp. 4, pp. 1368.
84
Belting, Palastaulen, pp. 6876 (quotation p. 71).
 71

Scholars have long speculated on whether or not these images

echo the Sylvester legends or the Donation of Constantine.85 This
seems entirely possible to me. I would stress that the images point
to the spiritual bases of papal power and authority. Taken by them-
selves, and taken in conjunction with other evidence, I believe that
these images Leos images with Jesus and the apostles and the
Marian images discussed above can be read as sketches of an
ideological program. In the Byzantine East, artistic and textual evi-
dence often shows two hierarchies. In one, the emperor takes a posi-
tion between Christ and the people and in another the emperor is
positioned between Mary and those who venerate her. In Rome, the
pope takes the middle position in this hierarchy. Laymen seldom
appear. To be sure, Charlemagne also kneels before St Peter in the
aula mosaic, and in likewise receiving the symbols of his rule a
banner from St Peter, he might seem to be on a par with Leo
III. But Leo is in the privileged position at Peters right hand.
Similarly, as we will see below, when emperors finally appeared on
papal coins, they were relegated to the reverse. This fresco and
mosaic evidence accords well with that of the coins that refer to
Christ or to St Peter, and with that of the voluminous textual mobi-
lizations of St Peter from this period. What I detect in the evidence
from the Republic of St Peter is a struggle to define and to articu-
late a consistent and coherent ideological program. The pope gets
his authority from God. The pope answers to God alone. The pope
intercedes with Mary on behalf of his people. The pope is the cus-
todian of Pauls teaching and Peters authority. Secular rulers are
the popes helpers.
To conclude this discussion we can return to Geertz with whom we
began. The social modes within which papal art worked can be
seen as pertaining to earthly and heavenly patronage, protection,
and intercession. As the popes separated themselves from the Byzantine
Empire and achieved autonomy, they used artistic means to com-
municate important aspects of the changed situation and to define
how, why, and for whom the new system functioned. If no single
image, style, or form of representation predominated, then the var-
ious sketches to which I referred above are nevertheless apparent
and perhaps can be explained by the very novelty of the situation.

85
One good discussion is Walter, Papal political imagery, pp. 1706.
72 ..

The pattern of life within which this papal art took a place was
religious in the first place but all-absorbing after that. Everywhere
there were reminders that the pope could carry the sins of his sub-
jects to the feet of Mary who would be asked to lay them before
Christ for forgiveness and also that the same pope could provide the
penitent with a job, a measure of grain, personal security, or social
reconciliation.

S A

Let us turn now to some of those formal marks of authority and

power that the eighth- and ninth-century popes, like all rulers, used
to express, to signify their official position. Coins will draw first atten-
tion. For any person or entity to issue coins on his or its own author-
ity is a significant claim to authentic rule. Popes began issuing coins
in the eighth century. Then we will turn to titles, including in a lim-
ited sense the ways in which popes referred to themselves and in a
larger sense to the ways in which official documents referred to other
powers, specifically to Byzantine emperors and to Frankish rulers.
Finally, we will consider dating devices in papal documents. It should
be pointed out right away that these categories of Herrschaftszeichen
overlap at several points.
It seems that a mint began operating in Rome in the time of
Emperor Constans II (r. 641668).86 Until the time of Constantine
V (r. 741775) this mint issued silver coins fairly regularly. As one
would expect, these coins bear appropriate imperial images, insignia,
and inscriptions. In the early decades of the eighth century Gregory
III and Zachary issued in their own names some square copper
objects that may be coins, weights, or pilgrims tokens.87 Not many
survive. Their intended use is hard to ascertain. And their political
significance is impossible to interpret. More intriguing are some coins

86
A.E. Bellinger and P. Grierson, Catalogue of the Byzantine coins in the Dumbarton
Oaks collection, vol. 2 in 2 parts (Washington, DC, 1968), pp. 201, 501, 501 (=
notes to nos. 1879).
87
M.D. OHara, A find of Byzantine silver from the mint of Rome for the
period A.D. 641752, Swiss Numismatic Review 64 (1985), pp. 12627 and fig. 3;
P. Grierson and M. Blackburn, Medieval European coinage, vol. 1 (Cambridge, 1986),
pp. 2623, 6445 and no. 1513. Grierson also suggests that they may be exagia,
that is weights for weighing solidi and tremisses or tokens connected somehow with
papal charitable activities.
 73

of emperors Leo III (r. 717741) and Constantine V which bear, on

their reverses, monograms of popes Gregory III and Zachary.88 This
suggests, at least, that these popes were proclaiming a kind of co-
rulership. Perhaps the enigmatic copper pieces should be taken in
the same sense. I have argued elsewhere that the popes were, from
the 730s to the 750s claiming an independence from Byzantium that
they could not yet enforce or exercise,89 and these coins may be
indicative of that ambiguous, tenuous situation.
For many years the Roman mint was inactive. Then in the 770s
Hadrian I struck some silver coins and, beginning in 781, he began
a fairly regular issue that was continued by Leo III down to Charle-
magnes imperial coronation in 800 and beyond. All of the ninth
century popes issued silver coins (fig. 9).
Hadrians earliest coins held to traditional Roman weight stand-
ards but his issues after 781 adopted the standard of the coinage
reforms of Pippin I that had been introduced into Italy in 781. After
800, Leo III switched to the new Carolingian standard promulgated
in 794. Papal coins held to this standard until late in the pontificate
of Gregory IV and then slowly declined in weight until they reverted
to approximately the original Carolingian standard.90 It is not sur-
prising that Rome would have aligned itself with the major power
in the West, indeed with its ally and protector, after liberating itself
from the clutches of Byzantium. The gradual decline in the weight
of ninth-century of papal coinage is surely an economic fact more
than a political one. That is, the relative fineness of papal coins
declined in lockstep with a discernible fall-off in papal benefactions
in precious metal and fine cloth. The popes were not devising a
standard of their own or rejecting Frankish norms. They were strug-
gling in a western Mediterranean overrun by Muslim raiders and in
a central Italy where local power struggles played themselves out in
the absence of effective Carolingian protection.
Any attempt to build the papal coinage into an economic argu-
ment crashes against two walls erected by Alessia Rovelli.91 First, the

88
OHara, Byzantine silver, pp. 113 and 124 nos. 19 and 20; Grierson and
Blackburn, Medieval European coinage, vol. 1, p. 260.
89
Noble, Republic of St Peter, pp. 1560.
90
Grierson and Blackburn, Medieval European coinage, vol. 1, pp. 25964.
91
A. Rovelli, La funzione della moneta tra lVIII e X secolo: Un annalisi
della documentazione archeologica, in: R. Francovich and G. Noy., La storia
dellalto medioevo, pp. 52137 and A. Rovelli, La moneta nella documentazione
74 ..

dissemination of this coinage is so slight as to prove minimal circu-

lation and, second, the coinage will have been too valuable for any
practical, daily use. This brings us to the inscriptions. I believe that
they point unmistakably to the issuance of this coinage as a sym-
bolic projection of power and prestige. The coins circulated in a
moral economy whose determinants were power and prestige not
supply and demand.
Weights tell one story then, but inscriptions tell another. Hadrians
earlier series of coins had, on the obverse, a long-shafted cross stand-
ing on a platform, a clear imitation of Byzantine coins, and the in-
scription or arrayed in three lines.
On the reverse these coins had , again in three lines.92
The use of the genitive here might suggest that the coinage and/or
the person issuing it are of , belong to the pope himself or St Peter.
But there does not seem to have been a precise, and thus unam-
biguous, syntax for inscriptions on coins. Hadrians second and seem-
ingly longer series had, on their obverse, surrounding
a bust image another imperial imitation and a curious (the
to the viewers left of the bust and the to the right), that defeats
some interpreters, while Grierson takes it as Iesous Basileus or else as
a date device.93 The reverse of these coins bear .
If the on the obverse does mean with Jesus as Ruler then the

altomedievale di Roma e del Lazio, in: Paroli and Delogu eds., La storia econo-
mica, pp. 33352.
92
These are discussed but not reproduced by Grierson and Blackburn, Medieval
European coinage, vol. 1, p. 638. For the coins see C. Serafini, Le monete e le bulle
plombee pontificie del medagliere vaticano (repr. Bologna, 1965), vol. 1, pp. 45 and Tavola
C, nos. 2 and 3 (genitive inscription on obverse) and P.E. Schramm, Die Anerkennung
Karls des Groen als Kaiser (bis 800), in his Kaiser, Knige, und Ppste: Gesammelte
Aufstze zur Geschichte des Mittelalters (Stuttgart, 1968), vol. 1, p. 366 fig. g (nomina-
tive inscription on obverse).
93
For the coins, Grierson and Blackburn, Medieval European coinage, vol. 1, pp.
263, 638 and nos. 1031 and 1032. For Griersons suggestions about the IB (which
Schramm, Anerkennung, p. 228 calls beachtlich) see his The coronation of
Charlemagne and the coinage of Leo III, Revue belge de philologie et dhistoire 30
(1952), pp. 82533 at 833 and Coinage, p. 638. I am unpersuaded by the argument
of M. Orlandoni, Ipotesi sul significato della sigla IB nel denaro papale di Adriano
I rinvenuto negli scavi di Aosta, Memorie dellAccademia Italiana di Studi Filatelici e
Numismatici 4 (1990), pp. 1334 to the effect that the IB refers to John the Baptist.
In an email communication Alan Stahl informs me that, much earlier, some Byzantine
coins of Alexandria had an IB that denominated twelve nummi. How such a twelve
got onto papal coins is impossible to say but perhaps it meant that the coins rep-
resented twelve of something in a local context. Unfortunately no small coinage has
been found that would qualify.
 75

imprecation for His victory on the reverse makes a coherent state-

ment. Taking together the imperial look of the coins, the call for
victory, and the (Constantinopolis Obryzum, the Byzantine
authentication symbol for pure gold), one cannot escape the con-
clusion that Hadrian was adapting aspects of Byzantiums symbolic
repertoire to his own uses.
When Leo III ascended the papal throne he issued coins whose
obverses bore , a bust image of Peter, and either a mono-
gram of the papal name or . It will be recalled that
Hadrians first coins had St Peter in the genitive on the reverse and
in some cases the popes name in the nominative on the obverse. His
second coins had, again, the popes name and perhaps Jesuss
in the nominative. Now Leo put St Peter in the nominative and
himself into the genitive. After Charlemagnes coronation, the popes
adopted a style that persisted into the tenth century. This had ref-
erences to Peter and the pope on the obverse and the name of the
emperor, often with a and/or an on the reverse. Here we
have a reversal of the coins of the early eighth century where the
popes put their monograms on the reverse of imperial coins. It is a
given of the ius monetae that the primary authority figure appears on
the obverse. Thus the popes seem to have been saying that they
shared authority in a relationship with Frankish rulers that put them
first but that ultimate legitimation came from either St Peter or Jesus.
If the popes seem to have displayed some uncertainty in articulat-
ing their message, then perhaps that is attributable to the relative
novelty of their situation and runs parallel with the similar but by
no means identical messages of papal art. Several other intriguing
and revealing manifestations of the newly emerging situation in Rome
can be discerned in various kinds of papal documents, specifically in
the ways the popes dated their documents and titled both themselves
and other rulers.
In 537 Justinian issued a novella that was promulgated in Rome
in about 550 pertaining to how documents issued anywhere in the
empire were to be dated. The ruling called for this form:
imperante domino piissimo augusto N. a Deo coronato magno imperatore.94

94
R. Schoell and W. Kroll, Corpus Iuris Civilis, vol. 3 Novellae, 5th ed. (Berlin,
1928), no. 47.
76 ..

Well into the eighth century papal letters followed this format faith-
fully. For instance, in 719 Gregory II wrote to the Anglo-Saxon mis-
sionary Boniface. The pertinent clause in his letter reads:
Data id. Maii imperante domno piissimo augusto Leone a Deo
coronato magno imperatore anno tertio, post consulatum eius anno tertio,
indictione secunda.95
In 751 Zachary wrote to Boniface and the date device in this let-
ter reads as follows:
Data IIII kalendas Novembris imperante domno piissimo augusto
Constantino a Deo coronato magno imperatore anno tricesimo sec-
ondo, post consulatum eius anno undecimo, indictione quarta.96
It must be stated right away, in connection with these letters and
with those that will follow, that very few papal letters survive com-
pletely and that, apart from a fragment from Hadrian I and a full
letter of Paschal, we have no originals.97 Thus, as we shall see vividly
just below, it is safer to talk of general trends than to try to iden-
tify decisive changes at particular moments or to speak confidently
about how things were always done. Remember that Gregory III
and Zachary put their own monograms on imperial coins and issued
those odd copper objects in their own names. Still, it seems safe to
say that the papal scrinium was conservative in its dating practices
during the very period in which the popes were slowly extricating
themselves from the institutional framework of the Byzantine Empire.
In April of 769 Stephen III convened a synod in Rome. The
opening words of its acta are distinctive:
In nomine Patris et filii et Spiritus sancti, regnante domino nostro Jesu
Christo uno ex eadem sancta trinitate cum eodem Patre et Spiritu sancto per
infinita omnia saecula, mense Aprile, indiccione septima.98
It has been customary to regard this dating clause as a dramatic
expression of the papacys rejection of Byzantine rule in Rome.99 As

95
Sancti Bonifatii et Lulli Epistolae, ed. E. Dmmler, MGH Epp. 3 (repr. Berlin,
1957), no. 12, p. 258.
96
Ibid., no. 87, p. 372.
97
R.L. Poole, Lectures on the history of the papal chancery down to the time of Innocent
III (Cambridge, 1915), p. 37. P. Jaff, Regesta Pontificum Romanorum (hereafter Jaff,
RP), vol. 1 (1885; repr. Leipzig, 1956), nos. 2462, 2551.
98
MGH Conc. 2, ed. A. Werminghoff (Hannover, 1906), no. 14, p. 79. There
are some textual problems relating to the transmission of the acts of this synod but
as they do not concern the date clause they need not detain us here.
99
H. Bresslau, Handbuch der Urkundenlehre fr Deutschland und Italien, vol. 2, pt. 2,
 77

it derives from so public and formal a setting as a synodal proto-

col, this interpretation is plausible. The absence of the emperors
name is striking but difficult to interpret. There is no other dated
document from the pontificate of Stephen III (768772) so we have
no way of saying what was then normal practice. There were, after
all, Frankish and Lombard bishops present at the synod and they
would not have welcomed an assertion or recognition of imperial
authority. Moreover, this regnante formula had appeared in a vari-
ety of other settings where no implicit rejection of imperial author-
ity was involved. Finally, such formulae had been used in the days
of imperial persecution of Christianity and just might have been used
here to signal that the authority of the heretical, that is iconoclast,
Constantine V was being ignored for a very specific reason. In other
words, this dating clause may have religious more than political sig-
nificance.100 Caution is in order.
The late 760s and very early 770s were very difficult for Rome
and the popes. There had been bloody political intrigue in the city,
indeed in the papal government itself. The Lombards had ravaged
the countryside around Rome and attempted to control the papacy.
Pippin III, the papacys ally and protector, died in 768 leaving two
sons, Charlemagne and Carloman, the former of whom was mar-
ried to a daughter of the Lombard king Desiderius, Romes neme-
sis for more than a decade.101 The very next securely dated papal
document, a privilege of Hadrian for Farfa (22 April 772), reverted
to the customary imperial-style date formula:
Data X Kalendas Martii. Imperantibus domno nostro piissimo augusto
Constantino a Deo coronato magno imperatore anno liii et post
consulatum eius anno xxxiii. Sed et Leone magno imperatore eius filio
anno xxi. Indictione X.102

4th ed. (Berlin, 1969), p. 420; Schramm, Anerkennung, p. 224; P. Classen, Karl
der Groe, das Papsttum und Byzanz: Die Begrndung des karolingischen Kaisertums, Beitrge
zur Geschichte und Quellenkunde des Mittelalters 9 (Sigmaringen, 1985), p. 11.
100
H. Fichtenau, Politische Datierungen des frhen Mittelalters, in: H. Wolfram
ed., Intitulatio II: Lateinische Herrscher- und Frstentitel im neunten und zehnten Jahrhundert,
Mitteilungen des Instituts fr sterreichische Geschichtsforschung, Ergnzungsband
24 (Wien, 1973), pp. 48589.
101
Basic details in Noble, Republic of St Peter, pp. 11232.
102
I. Giorgi and U. Balzani eds., Il regesti di Farfa, vol. 2 (Rome, 1879), no. 90,
p. 85.
78 ..

In the utter absence of comparanda it seems safest to leave on one

side Peter Classens surmise that Hadrian may have been signaling
a willingness to return to the imperial fold103 and to hew to Heinrich
Fichtenaus prudent assessment of the political situation and of the
acute predicament of the pope.104 The Lombard king was trying to
align himself with the pope and to persuade the pope to abandon
the papacys alliance with the Franks. Hadrian had no idea where
the Franks stood at that moment. Hadrian temporized and the
Lombards attacked Roman territory. In these tangled circumstances
Hadrian may have been taking some thought for the Lombard envoys
and the present uncertainty. That is, by dating the Farfa donation
according to imperial years, Hadrian avoided claiming the Duchy
of Spoleto, in which Farfa lay, as his own which the popes had
been doing for decades and left the Franks out of the picture. Then
in 773 the Lombard kingdom began breaking up in the face of a
Frankish invasion. Charlemagne had repudiated his Lombard wife
and made clear his intention to honor his fathers promise to pro-
tect Rome. Hadrian gave Spoleto a new duke, Hildeprand. In a
document for Farfa issued by Hildeprand a window opens momen-
tarily on the new situation:
In nomine Domini Dei Salvatoris Nostri Ihesu Christi. Temporibus ter beatis-
simus et coangelici domni adriani pontificis et universalis
papae.105
Whatever the emperors name meant in the Farfa document of 772,
that name vanished from all subsequent papal documents, as far as
we know. No relevant evidence survives until 781, the very year, I
repeat, in which Hadrian began his second series of silver coins. As
far as ducal documents from Farfa are concerned, Hildeprand quickly
began to date by Charlemagnes regnal years:
Regnante domno nostro Karolo viro excellentissimo rege, anno
regni eius in Dei nomine in Italia ii, mense iulii, per indictionem xiii.

103
Karl der Groe, p. 15.
104
Politische Datierungen, pp. 48992; P. Conte, Regesto delle lettere dei papi del
secolo VIII (Milan, 1984), pp. 2313.
105
Il regesto di Farfa, no. 91, p. 85. The use of Temporibus . . . is interesting
because it appears to avoid using regnal years. It is like contemporary Roman
inscriptions, of which several examples survive from the 750s and 760s: LP, vol. 1,
pp. 458 n. 27, 514 n. 2, 536 n. 69. See also Fichtenau, Politische Datierungen,
p. 490 n. 89.
 79

Regnante domno nostro Karolo excellentissimo rege Francorum

atque langobardorum, anno regni eius in Italia deo propitio ii.106
On December 1, 781 Hadrian issued a privilege for St-Denis:
Data kal. Decembris, regnante domine et salvatore nostro Iesu Christo,
qui vivit et regnat cum Deo patre omnipotente et spiritu sancto per immortalia
[infinita?] saecula, anno pontificatus nostri in sacratissima beati
apostolici Petri sub die [sede?] deo propitio decimo, indictione
quinta.107
Another document for St Apollinare in Classe from November of
782 reads:
Data kal. Novembris per manus Anastasii secundicerii, regnante domino deo
et salvatore Iesu Christo cum deo patre omnipotente et spiritu sancto per
infinita saecula, anno deo propitio pontificatus domini nostri Hadriani
in apostolica sede undecimo, indictione sexta.108
Fichtenau cites two additional letters of Hadrian, from 787 and 788,
that reveal only minor divergences from the two just quoted.109 Two
elements are of real significance in these four documents. First, the
regnante formula has been adopted on a regular basis. Jesus Christ
is claimed as direct overlord of Rome and of the lands under papal
rule. No emperor is named and, be it noted, no western ruler is put
into the place formerly held by the emperor. This system casts a
bright, retrospective light on the synodal protocol of 769 and accords
well with the otherwise anomalous IB on Hadrians coins. Second,
Hadrian dates explicitly by the years of his own pontificate. This is
a direct imitation of the way in which imperial regnal years were
cited in the formula required by Justinian and used in Rome at least
into the 750s.

106
Il regesto di Farfa, nos. 9294, 96, etc., from 775 and 776, pp. 8588.
107
Jaff, RP, no. 2435. I cite after Fichtenau, Politische Datierungen, p. 492
who quotes the document from a work of E. Baluze (Miscellanea, ed. J.D. Mansi
[Lucca, 1762], p. 3) to which I did not have access. On Hadrians dating prac-
tices see also A. Menzer, Die Jahresmerkmale in den Datierungen der Papsturkunden
bis zum Ausgang des 11. Jahrhundert, Rmische Quartalschrift fr christliche Altertumskunde
und Kirchengeschichte 40 (1932), pp. 2728, 3132. Menzer attaches more precise
significance to the date devices than Fichtenau is willing to do.
108
Jaff, RP, no. 2437. Again I cite after Fichtenau, Politische Datierungen,
p. 492 who quotes Johannes Benedictus Mirarelli, Annales Camaldulenses., vol. 1
(Venice, 1755), Appendix 12, no. 3, to which I did not have access. Cf. P. Kehr
ed., Italia Pontificia, vol. 5 (Berlin, 1911), p. 103, no. 5.
109
Politische Datierungen, p. 492.
80 ..

There is a marked convergence of elements here. Hadrians sec-

ond series of silver coins and his new and stable dating devices
appear in the early 780s. Earlier we noted that it was in the 780s
that Hadrians building boom and lavish metal and silk benefac-
tions come into view. I suggested already that some of Hadrians
building projects, particularly what I called the secular ones, may
have been proud gestures more than necessary maintenance. While
granting that we really have no idea what may have been going on
with coins and documents between 772/3 and 781 construction
work is a limited exception the obvious significance of the period
beginning in 781 can be explained. It was in 781 that Charlemagne
and Hadrian met to begin working out the precise terms of the
Italian territorial settlement that had been spelled out first in Pippin
IIIs meetings with Stephen II in Francia in 754 and then reaffirmed
in Charlemagnes visit to Rome at Easter in 774.110 Between about
760 and 780 the popes really did not know what their fate was to
be. As soon as the picture became clear, Hadrian acted decisively
on a broad front. Some years ago there was a lively debate between
Percy Ernst Schramm and Josef Der as to whether Hadrians actions
amounted to a transferral of imperial prerogatives to Charlemagne
(Schramm) or to himself (Der).111 Clearly the argument goes to the
latter. But Fichtenau is also correct in urging caution about just what
Hadrian had done. Certainly the pope had dispensed once and for
all with Byzantine imperial authority. Equally certainly, he had not
admitted Charlemagne as a new overlord. But it is probably anachro-
nistic to say that he claimed some kind of sovereignty.112 Indeed,
no pope had ever been in the exact position in which Hadrian found
himself.
With the pontificate of Leo III there are once again some new
elements. Leos letter conferring a pallium on Arn of Salzburg in
798 reads:
Data epistola XII kal. Mai. Per manum Pascalii primicerii sanctae sedis apos-
tolicae, regnante domino nostro Iesu Christo cum deo patre omnipo-
tente et spiritu sancto per infinita secula, amen; deo propitio pontificatus
domno nostro in apostolica sacratissima beati Petre sede ter-

110
Noble, Republic of St Peter, pp. 13875.
111
Schramm, Anerkennung, pp. 21563; J. Der, Die Vorrechte des Kaisers
in Rom, 772800, Schweizer Beitrge zur allgemeinen Geschichte 15 (1957), pp. 563.
112
Fichtenau, Politische Datierungen, pp. 4903.
 81

tio atque domni Caroli excellentissimi regis Francorum et

Langobardorum et patricii Romanorum, a quo coepit Italiam
anno XXV , indictione VI.113
The regnante formula and the pontificate-year dating are just what
we have seen in the time of Hadrian. It appears that some stabil-
ity has been introduced into papal practices; Hadrians system did
not die with him in 795. But there are two novelties. The first is
the inclusion of Charlemagne in the document and the second is
the queer a quo coepit Italiam formula. These elements are not easy
to interpret. Leo does place himself first. That counts for a good
deal. Leo was also in acute difficulties in Rome in the late 790s and
he may have wished to stress his relationship with his protector. The
a quo . . . formula dates from the capture of Pavia in 774. It seems
that Leo was acknowledging Charlemagnes authority without admit-
ting that the king had specific authority in Rome. Neither Leo nor
any other pope of the period ever referred to his lands as Italy so
the point of the wording seems to be that Rome did not belong to
the Italian kingdom. The protocol does mention Charlemagnes patri-
cius Romanorum title, the only credential Charlemagne might have
had to exercise power in Rome, but does not date from the con-
ferral of that particular title. The emphasis is on Leo and on the
Holy See.114
Authentic and complete papal documents from the years just after
Charlemagnes imperial coronation are exceedingly rare. In a priv-
ilege of March 803 for Fortunatus of Grado that has a dubious tex-
tual transmission we read:
Datum XII kal. April. Per manum Eustachii primicerii sanctae sedis apostolicae,
imperante nostro domino Carolo piissimo augusto a deo coro-
nato magno et pacifico imperatore anno tertio, indictione XI.115
And that is all. Leo is not named and his pontifical years are not
given. Through most of the rest of the ninth century papal letters
followed the format used in Leos letter for Fortunatus. As they are
all unremarkably similar, we need not trouble to cite them.

113
Jaff, RP, no. 2498. Salzburger Urkundenbuch, ed. W. Hauthaler and F. Martin,
vol. 2 (Salzburg, 1916), no. 2a, p. 4. See also: Fichtenau, Politische Datierungen,
p. 493; Menzer, Jahresmerkmale, pp. 30, 48.
114
Menzer, Jahresmerkmale, p. 48; Fichtenau, Politische Datierungen, pp.
49495. In the contrary sense, Schramm, Anerkennung, p. 224.
115
Jaff, RP, no. 2512. I cite after Fichtenau, Politische Datierungen, p. 495.
82 ..

But we must try to interpret them. First, let us recall that papal
coins continued to be struck with the popes name on the obverse
and the emperors on the reverse. Coins are an important kind of
Staatssymbolik. Second, it made sense for the popes to trumpet the
Carolingian imperial office. They had created it. Or at least such
was their view. They had a vested interest in assuring its continuity
and legitimacy. Moreover, Rome was the site of regular crises both
foreign and domestic right through the ninth century so interested
locals and transalpine rulers could use frequent reminders of who
was supposed to protect Rome and the popes. Third, not a single
document ever betrays the slightest intention on the part of any pope
to assign to the Carolingians actual overlordship over Rome.116 Fourth,
private documents from Rome and its region cited the emperors
name first but also included the popes name.117 Marks of papal
authority did not vanish. Fifth, papal letters, private documents, and
synodal protocols from before and after 800 began to use distinc-
tive language to refer to the papal office itself. For instance:
Anno deo propicio domini Karoli regis Francorum et Langobardorum atque patricii
Romanorum, a quo cepit Italiam, vicesimo sexto . . . presidente ter beatissimo
et coangelico Leone, summo pontifice sanctae Romanae eccle-
siae et universalis tertio papa . . . Leo sanctissimus et ter beatissimus
sanctae catholicae et apostolicae Romane ecclesiae et univer-
salis papa.118
Each aspect of this intitulation beatissimus, summus pontifex, and uni-
versalis papa had precendents, and by the last quarter of the ninth
century the formula summus pontifex et universalis papa became nor-
mal in papal documents. It is striking that the combination appears
in 798 and then continues into the next years.119 This was precisely
the time when Leo first inserted Charlemagnes name into papal
documents and coinage.
In my view, the popes were struggling to come to terms with the
wholly new situation in which they found themselves. No pope had
ever before been a temporal ruler whose position was reasonably
assured. No pope had, in a condition of fundamental autonomy,

116
Noble, Republic of St Peter, pp. 291324.
117
Fichtenau, Politische Datierungen, pp. 4957.
118
MGH Conc. 2, Concilia Aevi Karolini 1, no. 23, pp. 202, 203. See also Fichtenau,
Politische Datierungen, p. 496.
119
Ibid., pp. 4967.
 83

sought to regulate relations with an ally and protector. By laying

much greater stress on the papal office itself while experimenting
with ways to express the political situation, the popes seem to have
displayed a high tolerance for ambiguity. But the ambiguity attaches
less to the objective reality of the situation than to the intense difficulty
of articulating that reality.

Let us conclude with ritual celebrations as another kind of repre-

sentation of power and authority. Like other great urban liturgies,
Romes had been developing for centuries and would go right on
developing. I will focus on just a few revealing refinements and alter-
ations in my period. Moreover, liturgical ceremonies usually center
on the sacraments of the church. Viewed from one vantage point,
sacraments have an efficacy, a power, and a raison dtre that func-
tions, as theologians would later say, ex opere operato. But liturgies are
also visible, public, and participatory. They make and maintain com-
munities, give symbolic expression to the very real events of life, join
the present to the past in a hopeful glimpse of the future, resolve
social tensions, comment continuously on the political order, permit
visible displays of the structure of society, and permit many people
to participate in appropriate, defined ways.120
The structure of liturgies in the Sancta Romana Ecclesia was highly
articulated and complex. The basic arrangement in the eighth and
ninth centuries is less easy to perceive than it would become later.
In principle the system worked as follows. The title priests of the
twenty-five, or eventually twenty-eight title churches took it in turns,

120
For an outstanding presentation see E. Muir, Civic ritual in Renaissance Venice
(Princeton, 1981), esp. ch. 5 (A republic of processions) and ch. 6 (The ritual
occasion), pp. 185211, 21250. See also J. Baldovin, The urban character of Christian
worship: The origins, development, and meaning of stational liturgy, Orientalia Christianan
Analecta 228 (Rome, 1987), esp. pp. 105268. Still inspirational are V. Turners
works: The ritual process: Structure and anti-structure (1969; repr. New York, 1995), esp.
pp. 143 and Dramas, fields, and metaphors: Symbolic action in human society (Ithaca, 1974),
esp. pp. 2359, 23199. In a book on the priestly rituals in the Hebrew Bible
F. Gorman Jr. offers valuable definitions of and insights into the ritual process as
a priestly affair. Much of what he says bears directly on the priestly rituals of Rome:
The ideology of ritual: Space, time and status in the priestly theology, Journal for the Study
of the Old Testament, Supplement Series 91 (Sheffield, 1990), esp. pp. 1338.
84 ..

on a regular cycle, to celebrate mass in four of the patriarchal basil-

icas: St Peters, Santa Maria Maggiore, San Paolo fuori le mura,
and San Lorenzo fuori le mura. Romes suburbicarian bishops took
turns celebrating at St John Lateran. Virtually every titular and patri-
archal basilica had appointed to it one or more monasteries whose
monks saw to the laus perennis in that place. These liturgical arrange-
ments necessitated constant processions throughout the city.121
The popes celebrated most visibly in stational liturgies that arose
in antiquity and began being regularized by Gregory I. Liturgical
and other sources provide copious amounts of often conflicting evi-
dence for the organization of the stational liturgies. By the late eighth
century at least thirty-seven churches were at some time used for no
fewer than 147 separate stations. Detailed documentation can be
found for 102 stations in about 800.122 In Rome, then, there was a
stational liturgy, on average, every two to three days. John Baldovin
provides a useful, concise definition: Stational liturgy is a service of
worship at a designated church, shrine, or public place in or near
a city or town, on a designated feast, fast, or commemoration, which
is presided over by the bishop or his representative and intended as
the local churchs main liturgical celebration of the day.123
Most Roman stations were eucharistic liturgies but not all, as we
shall see below. Apart from special feasts, the more or less regular
stations were celebrated throughout the citys seven ecclesiastical
regions, which probably go back to the third century,124 on a pre-
cise weekly schedule as follows:
Sunday Third Region
Monday Fourth Region
Tuesday Fifth Region

121
Basic details and literature in Noble, Republic of St Peter, pp. 2127. See also
R. Reynolds, The organization, law, and liturgy of the Western Church, 700900,
in: R. McKitterick ed., The New Cambridge Medieval History, vol. 2 (Cambridge, 1995),
pp. 60813; de Blaauw, Cultus et decor, pp. 2272.
122
V. Saxer, Lutilisation par la liturgie de lespace urbain et suburbain: Lexemple
de Rome dans lantiquit et le haut moyen ge, Actes du XI e Congrs international
darchologie chrtienne, Studi di antichit cristiana 41 (Rome, 1989), vol. 2, pp. 9468,
10004.
123
Baldouin, Urban character, p. 37.
124
LP, 1. 64. L. Duchesne, Les rgions de Rome au moyen ge, Mlanges
darchologie et dhistoire 10 (1890), pp. 12649, at 12630; C. Pietri, Rgions ecclsi-
astiques et paroisses romaines, Actes du XI e Congrs international darchologie chrtienne,
vol. 2, pp. 103562.
 85

Wednesday Sixth Region

Thursday Seventh Region
Friday First Region
Saturday Second Region.125

The early eighth-century Ordo Romanus I provides a detailed account

of one stational liturgy.126 In summary fashion, the ceremony can be
described this way. The acolytes of the appropriate region and the
defensores of the whole city gathered at dawn at the Lateran to accom-
pany the pope on foot to the station. Lay grooms walked to the
right and left of the popes horse. In front of the pope rode regional
deacons, the primicerius and two regional notaries, and regional sub-
deacons. These officers were arranged in discrete groups and left a
good gap between themselves and the pope. Behind the pope, rid-
ing, were the vicedominus, vestararius, nomenclator, and sacellarius. One
acolyte from the stational church walked in front of the popes horse
carrying an ampulla of chrism wrapped in a mappula. Other acolytes
accompanied him. For some feasts, the Ordo says, the gospels, lit-
tle money sacks, and hand basins were carried before the pope.
Presumably at this point in the procession, in front of the pope, the
baiuli carried the wash basins, daily paten and chalice, pitchers and
vessels, other gold and silver vessels, gold and silver wine sieves, a
large silver sieve, silver cups, a gradual book, other gold and silver
vessels, and gold and silver candlesticks from the Lateran. The man-
sionarii had responsibility for all of this liturgical equipment. On spe-
cial feast days the great paten and chalice along with precious books
were let out on the signature of the vestararius, who had to enumer-
ate the number of jewels on each. Once this glittering entourage
reached the stational church it was met by local lay and ecclesias-
tical dignitaries. A lay cubicularius, who had already prepared the
popes sedan chair, walked in front of it as it entered the church.
The maiores of the church, local notables, followed the papal chair
into the church. The pope proceeded to the sacristy where he
exchanged his processional robes for liturgical vestments. Then, pre-
ceded by candles and incense, he processed to the altar for mass.

125
Baldovin, Urban character, p. 131.
126
Ordo Romanus Primus, ed. Michel Andrieu, Les Ordines Romani du haut moyen
age, Spicilegium sacrum Lovaniense 23 (Louvian, 1971), pp. 65112. See also de
Blaauw, Cultus et decor, pp. 72102.
86 ..

Additional insignia were usually used as well. The pope was pre-
ceded by two processional crosses. One was gold with purple gems.
It was given to Leo III by Charlemagne, stolen under Paschal I,
and replaced by Leo IV. Nothing is known about the other one.127
Normally seven candlesticks preceded the pope. At the stational
church the procession was met by the seven processional crosses of
the seven regions. These were stored at Santa Anastasia and carried
to the stational church without ceremony for the liturgy.128 On at
least some occasions the banners of the regions and of the scholae,
the communities of foreign residents in the city, also met the pope.129
The people of Rome, then, witnessed this grand display of the
pope himself, the most important administrators of the papal admin-
istration, and a choice sample of the treasure of the church on a
regular basis. Because privileged positions were always accorded to
the acolytes and regionary officials from the stations own area, and
because some laymen were folded into the ceremony in prominent
ways, these processions provided a constant proclamation of the pre-
cise shape of Romes governing order. But as Victor Saxer remarks,
the pope traversed the city as a sovereign, according to an etiquette
whose Byzantine origin has been underlined and which even recalls
the royal adventus.130 In the same vein, Dominic Janes notes that
Roman society was a treasure society. That is, there was a con-
stant accumulation, exchange, and display of treasure: precious met-
als, jewels, and cloth. All of this material spoke a language that
served to delineate rank, status, and power. But that language was
only comprehensible as long as some people could confine its use to
themselves while others understood its meaning.131 All of these items,
in their continuous display in the city, were signs in precisely the
sense Augustine had in mind when he said that a sign is something
that, beyond the impression it conveys to the senses, makes some-

127
Ordo XX, c. 7, ed. Andrieu, Les Ordines Romani du haut moyen age, vol. 3
(Louvain, 1974), p. 236; LP, 2.8, 110. Saxer, Lutilisation, p. 974.
128
Saxer, Lutilisation, p. 973.
129
Ibid., pp. 9757.
130
Ibid., p. 951.
131
D. Janes, God and gold in late Antiquity (Cambridge, 1998), pp. 1842. Cf. The
apposite remarks of Brenk, Early gold mosaics in Christian art, Palette 38 (1972),
pp. 1625 where gold is interpreted as meaning, in one way or another: splendor,
magnificence, opulence, fire, light, good fortune, salvation, magic, divinity, office,
prosperity, and light. He notes that the ninth-century writer Hrabanus Maurus cites
twenty different meanings of gold in his allegorical lexicon.
 87

thing else known.132 The fantastic display of wealth that accompa-

nied every stational procession might easily tempt one to advert to
Thorstein Veblens conspicuous consumption as an explanation. But
Richard Goldthwaite, speaking of Renaissance Florence in particu-
lar, argues convincingly that such an explanation will not quite do.
Instead, he says, we need an explanation that regards any particu-
lar historical configuration of consumption habits as a function of
the culture of that moment . . . Goods, in short, communicate some-
thing about culture.133 When goods are intimately connected to ritual
processes, say Douglas and Isherwood, then they are ritual adjuncts;
consumption is a ritual process whose primary function is to make
sense of the inchoate flux of events.134 The culture being commu-
nicated was majestic, serene, formal, elegant, patronal, structured,
and imperial. Thus it assigned meaning to events.
There were other regular celebrations, sometimes connected to the
stational processions, sometimes not. The four Marian feasts dis-
cussed already in connection with images were days of major pro-
cessions. Ash Wednesday, the beginning of Lent, saw another as did
November 1, the Feast of All Saints. There may have been a special
procession connected with the citys eighteen deaconries on March
25.135 Leo III introduced the Minor Litanies, essentially the Rogation
Days that appeared first in late antique Gaul. They proceeded as
follows. On the Monday before Ascension, the pope, along with all
the clergy and the whole people, departed from Santa Maria Maggiore
and processed to the Lateran basilica. On Tuesday they started from
Santa Sabina and processed to San Paolo. On Wednesday they began
at Santa Croce in Gerusalemme and wound up at San Lorenzo.136
At least eight times per year (Candlemas, Ash Wednesday, Annun-
ciation, Major Litany see below, Assumption, Nativity of the
Virgin, All Saints, and the Wednesday of Ember Days) there were
huge public gatherings, called collecta, which assembled in prescribed
places before processing to the days station. It is especially inter-
esting to note that four of these collecta gathered in the heart of the

132
De doctrina christiana, 1.2. 2.
133
R. Goldthwaite, The empire of things: Consumer demand in Renaissance
Italy, in: F.W. Kent and P. Simons eds., Patronage, art, and society in Renaissance Italy
(Oxford, 1987), pp. 15375, quotation at p. 156.
134
M. Douglas and B. Isherwood, The world of goods (New York, 1979), p. 65.
135
Saxer, Lutilisation, pp. 9519; Baldovin, Urban character, pp. 12939.
136
LP, 2.12.
88 ..

ancient city, two in the Forum Romanum, one at the foot of the
Palatine, and one near the Campus Martius. These celebrations
marked an important appropriation of the citys historical spaces.137
Two of these annual celebrations call for a few additional words.
First, there is the Litania Maior, well known because it was during
its celebration on 25 April 799 that Leo III was attacked in a street
near San Lorenzo in Lucina where the procession began. This pro-
cession went along the Via Lata, right through Romes most aris-
tocratic neighborhood, to the Flaminian Gate, stopped for a station
at San Valentino, crossed the Tiber at the Ponte Milvio, stopped
again at a no longer extant church, and stopped once again in the
atrium of St Peters before culminating with mass in the basilica.
Clergy and people participated together in this procession and a dis-
tinctive feature was the inclusion of the poor from the local xen-
odochium. Carrying painted wooden crosses, they led the procession.
Behind them came the seven processional crosses that represented
the seven regions of the city. This particular celebration, which
extended over some 10 km., incorporated the whole city.138
Leo IV led a procession with the acheiropoieta on the vigil of the
Feast of the Assumption from the patriarchate to the basilica of
Hadrian, the former Curia Senatus, and the nearby church of Santa
Maria Nova. From there the procession went to Santa Maria Maggiore
for morning mass.139 This image not made by hands is first men-
tioned by the Liber Pontificalis in the time of Stephen II.140 Whether
Stephen carried the image around on August 25 or on another

137
Saxer, Lutilisation, p. 954. The collecta assembled at San Adriano and Santi
Cosma e Damiano in the Forum, at Santa Anastasia at the foot of the Palatine,
and at San Lorenzo in Lucina near the Campus Martius.
138
The Litania Maior (so called because it was Romes longest procession) is
known from several sources: Ordo XXI, ed. Andrieu, Ordines Romani, vol. 3, pp.
2479; Deshusses ed., Le sacramentaire grgorien, vol. 1, no. 100, pp. 2113. Later writ-
ers mistakenly attributed this late eighth-century penitential liturgy to Gregory I:
see Saxer, Lutilisation, pp. 9634. As if to provide a constant reminder of this
celebration, Leo III presented St Peters with a gold-studded cloth bearing an image
of it: LP, 2.10. J. Croquison calls this a document iconographique tonnant which
might apply equally to the image itself and to Leos visibly commemorating the
most humiliating moment of his pontificate: Liconographie chrtienne Rome
daprs le Liber Pontificalis, Byzantion 34 (1964), p. 597.
139
LP, 2.110.
140
LP, 1.443: procedens in letania cum sacratissima imagine domini Dei et salvatoris nos-
tri Iesu Christi quae acheropsita nuncupatur. The pope, barefooted, carried the image on
his own shoulders to Santa Maria Maggiore.
 89

Marian feast day is hard to say and it is equally difficult to know

if the image was carried as a palladium because of Lombard threats
to the city or if the military menace and a liturgical procession just
happened to coincide.141 We cannot be certain whether or not Sergius
I introduced this specific ritual when he regularized Romes Marian
feasts. Indeed, we just do not know when the Assumption proces-
sion was set into the form which it held for centuries. It seems likely
that the image in question is the Lateran icon of Christ, long con-
served in the Sancta Sanctorum and barely visible today thanks to
Innocent IIIs silver frame. The immense range of meaning attached
to the carrying of Christs image from the popes home office to
the Senate House and Forum via the old Via Sacra cannot have
eluded anyone, however, and accords well with the construction work
and artistic patronage in this same area that were discussed above.
Our period also saw a large number of ad hoc processions and
celebrations. Gregory II led several processions after the Tiber flooded
Rome in 716.142 After a successful negotiation with King Liutprand,
Zachary returned to Rome, assembled the people at Santa Maria
ad Martyres (the Pantheon) from which they processed, singing lita-
nies, to St Peters.143 On another occasion, Zachary discovered the
head of St George in the Lateran patriarchate. He assembled this
city of Romes people and they processed to San Giorgio in Velabro
where the relic was installed.144 Paul I led a procession after the
installation of the relics of St Petronilla and another after he trans-
lated the remains of some Roman martyrs into the city.145 After the
Lateran council of 769 all of those who had participated proceeded
barefoot to St Peters.146 Hadrian I organized a procession of all his
clergy and the Roman senate from the city to his newly instituted

141
For discussion see: Saxer, Lutilisation, pp. 17880; Wolf, Salus Populi Romani,
pp. 3775.
142
LP, 1.399.
143
Ibid., p. 429.
144
Ibid., p. 434.
145
Ibid., p. 464. Translations of relics were common in the middle of the eighth
century and then again under Paschal I. These occasions were both solemn and
joyous. See J.M.H. Smith, Old saints, new cults: Roman relics in Carolingian
Francia, in: J.H.M. Smith ed., Early medieval Rome and the Christian West. Essays in
honour of Donald M. Bullough (London/Boston/Kln, 2000), pp. 31739. Saxer,
Lutilisation, pp. 9802, 10203 provides a discussion of translations that is briefer
than Smiths.
146
LP, 1.477.
90 ..

domusculta at Capracorum.147 It would be easy to go on multiply-

ing examples but I think that enough has been said to show that
virtually every part of the city, at virtually all times of the year, was
being incorporated by the papal administration into a visible pro-
jection of presence and leadership.
In 684/85 Rome saw the foundation of its first deaconry.148 By
the time of Leo III there were at least eighteen of them.149 These
were, in the simplest sense, food distribution centers. Some were
headed by laymen, some by clerics, but all heads were appointed
by the pope. The individual deaconries were staffed by monks. The
food distributed came from papal estates. For present purposes what
is striking about the deaconries is the ceremony connected with them.
On Fridays the poor of a particular area assembled at a deaconry.
From there they marched, singing psalms, to the nearest bath house.
After this ablution they returned to the deaconry for their food allot-
ment. Here again the society of the city, at several levels, was stitched
together by a single rite. The bathing, perhaps desirable from the
standpoint of hygiene, was primarily enmeshed in the web of mean-
ings spun by Christian baptism, the first and fundamental rite of ini-
tiation into the Christian faith. Perhaps too these ceremonies forged
bonds of unity with the Lateran baptistery. Every Friday a large part
of Romes population was reminded in a pointed way of its own
baptism and of its participation in a Christian society whose head
was the pope. And Romes populace was reminded of who fed it.
Hadrian indeed dispensed food to one hundred poor people every
day at the Lateran in front of a picture, which he may have put
there, of a pope feeding the poor.150 It is worth remarking yet again
how the symbols and realities in the Republic of St Peter so fre-
quently worked in the same realms of meaning, defined and affirmed

147
Ibid., p. 509.
148
Noble, Republic of St Peter, pp. 23134; Krautheimer, Rome, pp. 7477. Older
studies remain fundamental: J. Lestocquoy, LAdministration de Rome et diaconies
du VIIe au IXe sicle, Rivista di archeologia cristiana 7 (1930), pp. 26198; H.-I.
Marrou, Lorigine orientale des diaconies romaines, Mlanges darchologie et dhis-
toire 57 (1940), pp. 95142; O. Bertolini, Per la storia delle diaconie romane nel-
lalto medioevo sino alle fine del secolo VIII, in his Scritti scelti di storia medioevale,
ed. O. Banti (Livorno, 1968), vol. 1, pp. 309460; R. DAmico, Lorganizzazione
assistenziale: le diaconie, Roma e let carolingia: atti del giornata di studio 38 maggio
1976 (Rome, 1976), pp. 22936.
149
Or perhaps as many as twenty-three: Geertman, More veterum, pp. 11115; de
Blaauw, Cultus et decor, pp. 502.
150
LP, 1.502 and 518 n. 53. See Croquison, Iconographie, 56970.
 91

the same cultural norms. The lengthy and careful description in the
Liber Pontificalis of Leo IVs consecration of the Leonine City may
serve to conclude these remarks on the public, liturgical manifesta-
tions of papal power. After the walls were finished:
The blessed pope, in order that the above mentioned city, called by
its founder after his own name Leonine, might perpetually stand firm
and strong, ordered that everyone, bishops and priests, every deacon,
all the orders of the clergy of the holy, apostolic Roman church, with
great devotion and joyful hearts, after singing litanies and psalms should,
barefoot and bearing ashes on their foreheads, make a circuit of the
entire walls with him. Meanwhile, he instructed other cardinal bish-
ops to bless water so that in the midst of the procession they might,
during the offices of prayer, cast that water on the walls for the sake
of their complete sanctification.151
The pope and his companions actually stopped three times, near the
main gates, to carry out these prayers and blessings. Leo also dis-
tributed a great sum of money to the Romans as he was going to
St Peters to celebrate a mass for the safety of the city, a ceremony
reminiscent of imperial progresses and one which the Frank Clovis
mimicked, as Gregory of Tours tells us in a famous passage. He
then honored and enriched the nobles of Rome with gifts of gold,
silver, and silk. Here again we see all the orders of the city arrayed
behind the pope. He protects, blesses, and enriches them. Everyone
is there, from the highest to the lowest ecclesiastical dignitary and
from the secular nobles of Rome to the poor. This one ceremony
expresses very clearly the nature of Roman society in this dynamic
and creative time.
By the tenth century, political problems, social struggles, continu-
ing Muslim incursions, economic reversal, and the decline of the
Carolingian role in Italy all contributed to breaking down the Republic
of St Peter and, as it cracked apart, its public celebrations, con-
struction work, liberal benefactions, and Staatssymbolik were less and
less in evidence. The messages which Romes ecclesiastical leaders
had so consistently broadcast by the means available to them grad-
ually began to reach fewer people and to have less meaning. But
this does not change the remarkable degree to which the popes over
nearly two centuries gave visible, tangible expression to their power
and majesty. Between 700 and 900 Rome became a city of the
church, a papal city.

151
LP, 2.124.
This page intentionally left blank
 MONUMENTS AND MEMORY:
REPOSSESSING ANCIENT REMAINS IN EARLY
MEDIEVAL GAUL1

Bonnie Effros

According to Gregory of Tours, following the sixth-century noble-

man Senochs consecration as a cleric, he set out to found a monastery.
Having located in the region of Tours a site at which there were a
number of ancient walls in ruins, he constructed a building and
made it ready for habitation. Gregory noted that Senoch was drawn
to the vestiges of this ancient structure not just by plentiful build-
ing materials but by the belief that Martin of Tours had recited
Mass in the remains of an oratory discovered there. After the erec-
tion of an altar and a miracle enabling a large casket of relics to
be enshrined in the small space, Gregorys predecessor, the bishop
Eufronius (d. 573) consecrated the altar and initiated the new monas-
tic community.2 A similarly propitious discovery blessed the monk
Columbanus soon after his unhappy departure from Gaul. Jonas of
Bobbio recalled that when the saint entered a fertile valley of the
Apennines to found a new monastery, he acquired there a ruined
basilica said to be dedicated to St Peter. Despite less than optimal
conditions in this mountainous landscape, divine assistance enabled
the abbot to restore the site. Gods miracles sanctified the find and
choice of this abandoned ruin, and Bobbio, built on the foundations

1
I am grateful to Bailey Young, Isabel Moreira, Nina Caputo, Pauline Head
and participants in the Topographies of power in the early medieval West con-
ference for critical insights and comments that greatly improved earlier drafts of
this essay. Research and revision of this piece were supported by a 1996 Summer
Research Fellowship from the Graduate School, Southern Illinois University at
Edwardsville, and a Berkshire Summer Fellowship at the Mary Ingraham Bunting
Institute of Radcliffe College in 1998.
2
Repperit enim infra territurii Turonici terminum parietes antiquos, quos erudirans a ruinis,
habitationes dignas aptavit. Repperitque ibi oratorium, in quo ferebatur celebre nostrum orasse
Martinum., Gregory of Tours, Liber Vitae Patrum 15.1, ed. B. Krusch, MGH SRM
1, 2 (Hannover, 1969), p. 271. E. James, Archaeology and the Merovingian monas-
tery, in: H.B. Clarke and M. Brennan eds., Columbanus and Merovingian monasticism.
BAR International Series 113 (Oxford, 1981), p. 34.
94

of the older structure, would fare well under the leadership of this
holy man.3
Whereas sixth- and seventh-century saints in the above accounts
adapted abandoned ruins, the vitae of fourth- and fifth-century bish-
ops focused on their missionary activities among pagans. Discouraging
competitors often required the destruction and immediate reoccu-
pation of ancient buildings used for unholy purposes. In his account
of the life of Maurilius of Angers (d. 453) written in the 620s, for
instance, Magnobodus of Angers described how the native of Milan
brought about divine conflagrations at two pagan temples in the civi-
tas of Angers. After such thorough cleansing by fire, Maurilius built
a church at the first and a monastery at the second of these sites.4
Venantius Fortunatus likewise related that when Paternus traveled
to Brittany, he learned of an active pagan cult. After they punished
its celebrants, Paternus and his companion Scubilio used the former
temple to hold livestock.5 Despite important differences, these vitae
shared many commonalities with the works of Gregory and Jonas,
since Magnobodus and Fortunatus indicated the practicality of uti-
lizing the building materials already present for new construction.6
Whether serendipitous discoveries as in the first two examples, or
the consequence of a saints direct action as in the second two, these
activities also symbolically evoked the pious nature of the monastic
founders. Such missionary accounts provided justification for the
incorporation of ruins during the expansion of the monastic way of
life in Merovingian Gaul.7

3
Ubi cum venisset, omni cum intentione basilicam inibi semirutam repperiens, prisco decori
renovans reddidit. In cuius restauratione mira Domini virtus panditur., Jonas of Bobbio, Vitae
Columbani discipulorumque eius libri II 1.30. ed. B. Krusch, MGH SRG 37 (Hannover,
1905), pp. 2212.
4
Vita Beati Maurilii 2 and 6, ed. B. Krusch, MGH AA 4, 2 (Berlin, 1885), pp.
8587. G. Mathon, Maurilio, Bibliotheca Sanctorum 9 (Rome, 1967), pp. 18586. I
thank Albrecht Diem for this reference.
5
. . . ut fanum profani cultus ereptis hominibus deputaretur pecoribus et insensatus locus fieret
animalium clausura., Venantius Fortunatus, Vita Sancti Paterni 410, ed. B. Krusch,
MGH AA 4, 2 (Berlin, 1885), pp. 346. J. Kitchen, Saints lives and the rhetoric of
gender: Male and female in Merovingian hagiography (Oxford, 1998), pp. 445.
6
Due to the discontinuity between the destruction of the site and the arrival of
Senoch, Percival sees this particular example as indicative of the relative scarcity
of building materials in Gaul. J. Percival, Villas and monasteries in late Roman
Gaul, Journal of Ecclesiastical History 48 (1997), p. 11.
7
Recorded miracles involving ruins appear to have been limited to male founders;
no female founders of monasteries received credit for having done the same despite
the fact that some of their houses occupied the sites of earlier remains. The monastery
 95

Keeping in mind the bias of written texts due to clerics com-

mitment to providing orthodox interpretations of the symbolism of
such ruins,8 sixth- and seventh-century modifications of crumbling
structures contributed significantly to the religious topography of
Gaul. Clerical leaders consciously adapted ancient landscapes into
Christian milieux through their exploration and excavation, followed
by their association in written works with divinely ordained events.
They might thereby verify the worthiness of such sites as holy places
to local Christian communities as well as potential pilgrims.9 This
motif apparently retained its significance for centuries. In the mid-
eighth century Vita Filiberti, Clovis II and Balthild gave a donation
of land to the saint for the foundation of Jumiges in 654. The
ancient fortification included in this bequest to Philibert enabled the
monks to transform the site into a fortress of God where those seek-
ing salvation might find shelter.10 Such depictions of ancient topog-
raphy connoted both the familiarity and sacred purpose that shaped
the identity of the monastic communities thereafter inhabiting the
older sites.
These ruins thus differed significantly from the limited knowledge
of classical monuments, the pyramids and other stone wonders of
the ancient world possessed by ecclesiastical authors such as Gregory
of Tours, Isidore of Seville and the Venerable Bede.11 For them,

of Chelles rested on the remains of structures inhabited in the Roman period but
abandoned in the fifth century. J. Ajot and D. Coxal, Chelles, Gallia informations:
Prhistoire et histoire (1993), pp. 5860. Percival, Villas and monasteries, p. 18.
8
Clerical discussions left many subjects unmentioned, including some of the
more inexplicable rites practiced in Gaul. R. Merrifield, The archaeology of ritual and
magic (New York, 1987), pp. 121. For a discussion of the possible ritual significance
of river deposits, for example, see: M. Schulze, Diskussionsbeitrag zur Interpretation
frh- und hochmittelalterlicher Flufunde, Frhmittelalterliche Studien 18 (1984), pp.
22248. On the other extreme, some clerics fabricated descriptions of antiquities.
L. Nees, Theodulf s mythical silver Hercules vase, poetica vanitas and the Augustinian
critique of the Roman heritage, Dumbarton Oaks Papers 41 (1987), pp. 44351.
9
As noted by Robert Markus, by the sixth and seventh centuries, the contem-
porary history of holy sites often mattered more than their actual past. R. Markus,
The end of ancient Christianity (Cambridge, 1990), pp. 1545.
10
Ibidem castrum condiderunt antiqui; ibi adstant in aciae nobilia castra Dei, ubi suspirantes
pro desiderium paradisi gemunt. . . ., Vita Filiberti Abbatis Gemeticensis et Heriensis 67, ed.
W. Levison, MGH SRM 5 (Hannover, 1910), pp. 58789. J.-C. Poulin, Filibertus,
in: Lexikon des Mittelalters 4 (Mnchen, 1989), pp. 4478.
11
The ancient wonders figured large among the marvels Gregory of Tours
described. Gregory of Tours, De cursu stellarum ratio 18, ed. B. Krusch, MGH SRM
1, 2, pp. 40710. For his description of the pyramids as granaries, see: Gregory of
Tours, Decem libri historiarum, 1.10, eds. B. Krusch and W. Levison, MGH SRM 1,
96

encounters with the past had more immediate applications in the

frequent interactions of Christians with the local ruins of earlier peri-
ods. The exploration of architectural remnants by late antique and
early medieval Christians shared greater affinity with the search
for the relics of holy Christians, such as Alchima and Placidinas
cemeterial excavation of the remains of Antolianus at Clermont, or
Mamertus miraculous exhumation of Ferreolus and Julian.12 Both
types of undertakings reflected some of the same reverence for and
fascination with the past. More importantly, both made the most of
the referential powers and adaptability of newly discovered physical
fragments.13 Through what might be called rituals of possession,
Christian leaders incorporated ancient ruins as well as saints relics
into their communities. Historians and hagiographers in successive
generations typically observed in these cases that their discovery and
re-use were not serendipitous but divinely ordained events. Ruins
therefore had great significance for the successful foundation of new
monasteries despite the often unknown origins of these structures.
Christian rites and the translation of relics sanctified these locations
regardless of their previous functions.14
In surviving accounts, however, only clerics of high standing might
perform the exploration of indigenous Roman structures if the activ-

1, new edition (Hannover, 1965), p. 11. Isidore, Etymologiarum sive Originum 15.11,
ed. W.M. Lindsay (Oxford, 1962). Bede mentioned the Pantheon. Bedes Ecclesiastical
History of the English people, 2.4, transl. and eds. B. Colgrave and R.A.B. Mynors
(Oxford, 1969), pp. 1489.
12
Gregory of Tours, In gloria martyrum 64, ed. B. Krusch, MGH SRM 1, 2, pp.
812. Gregory of Tours, De virtutibus S. Iuliani 2, ed. B. Krusch, MGH SRM 1, 2,
pp. 1145.
13
This essay will therefore address primarily architectural ruins and ancient arti-
facts, and not the closely related phenomenon of inventiones of saints relics, since
this topic has been dealt with extensively elsewhere: P. Brown, The cult of saints: Its
rise and function in Latin Christianity (Chicago, 1981). M. Heinzelmann, Translationsberichte
und andere Quellen des Reliquienkultes. Typologie des sources du moyen ge Occidental
33 (Turnhout, 1979). Excavation of the dead might also lead to the recovery of
the physical remains of more ordinary Christians, such as the discovery of a girls
incorrupt corpse discovered when a piece of the vaulting fell upon and shattered
the cover of her sarcophagus at the church of St Venerandus in Clermont, Gregory
of Tours insisted for some time that she was not a saint, but simply a Christian
whose body had been preserved in spices. He apparently wished to discourage a
local cult from developing around the girls remains. Gregory of Tours, De gloria
confessorum 34, ed. B. Krusch, MGH SRM 1, 2, pp. 3189.
14
Sacred spaces were created and reinforced by the habitual actions that occurred
at them. C. Gosden, Social being and time (Oxford, 1994), pp. 346; 8790. R. Bradley,
Ritual, time and history, World Archaeology 23 (1991), p. 211.
 97

ity was to be perceived as legitimate and orthodox. Gregory of Tours,

for instance, depicted excavation of ancient sites in some instances
as a sign of disbelief in the miracles of God, such as when Theodegisel,
the heretical Visigothic king, had the affrontery to dig trenches around
the site of a sacred pool in Osser (near Seville). Because he did not
believe in the presence of Gods power at the water site that filled
miraculously with holy water every Easter for the purpose of bap-
tizing local children and curing the ill, Theodegisel expected to find
that an aqueduct was the source of this occurrence. Gregory thus
observed that God punished Theodegisel with a premature death
the following year.15 In this hagiographical account, excavation of
an ancient site by this notorious heretic constituted additional proof
for readers of his unorthodoxy.

Scholars have long sought explanations for the medieval attraction

to and re-use of ancient remains.16 Historical analyses have often
traced the roots of interest in the exploration of the physical remains
of Antiquity only as far back as eleventh- and twelfth-century human-
ists.17 As early as 1844, however, Thomas Wright remarked that
inhabitants of Anglo-Saxon England were accustomed to contact
with material remains of Antiquity on a frequent basis. In fact,
he suggested that these ancient artifacts helped to shape medieval

15
Giselle de Nie has reconstructed this passage from the Liber historiarum 5.17
and 6.43, and the De gloria martyrum 234. G. de Nie, Views from a many-windowed
tower: Studies of imagination in the works of Gregory of Tours. Studies in Classical Antiquity
7 (Amsterdam, 1987), pp. 801.
16
A very helpful introduction to the medieval origins of antiquarianism is:
S. Piggott, Antiquarian thought in the sixteenth and seventeenth centuries, in: Ruins
in a landscape: Essays in Antiquarianism (Edinburgh, 1976), pp. 124. On antiquarian-
ism in general, see among others: A. Momigliano, Ancient history and the Anti-
quarian, in: Studies in historiography (London, 1966), pp. 139. K. Pomian, Collectors
and curiosities: Paris and Venice, 1500 1800, transl. Elizabeth Wiles-Portier (Cambridge,
1990).
17
M.T. Hodgen, Early anthropology in the sixteenth and seventeenth centuries (Philadelphia,
1964), pp. 1157. W.S. Heckscher, Relics of pagan antiquity in mediaeval set-
tings, Journal of the Warburg Institute 1 (193738), pp. 20420. J.B. Ross, A study
of twelfth-century interest in the antiquities of Rome, in: J.L. Cate and E.N.
Anderson eds., Medieval and historiographical essays in honor of James Westfall Thompson
(Chicago, 1938), pp. 30221.
98

consciousness, and possibly misunderstanding, of the past.18 A cen-

tury later in La fin du paganisme en Gaule, Emile Mle linked the idea
of re-use to a superficial conversion process. Pointing to the destruc-
tion of pagan religious sites by fourth-century medieval clerics such
as Martin of Tours, Mle observed that their reoccupation imme-
diately afterwards as sacred places was inconsistent with genuine
Christian faith. Based on Gregory Is instructions in 601 to the mis-
sionary Mellitus in Anglo-Saxon England to retain and consecrate
pagan structures as churches,19 Mle envisioned a process of rapid
but shallow Christianization especially in rural places, where mis-
sionaries relied upon the incorporation of traditional sites of worship
for the successful conversion of local inhabitants.20
Most recently, Bailey Young has challenged the chronology of
Mles popular vision of the early medieval re-use of rural pagan
structures. He has suggested that the destruction and transformation
of sacred places did not typically occur in rapid succession. By ques-
tioning the accuracy of the archaeological and historical evidence
presented by Mle, Young has concluded instead that whatever recla-
mation of ancient religious sites occurred, it took place only after

18
Wright noted Bedes account of the monks of Ely and the miraculous discov-
ery of a white marble sarcophagus for Aethelthryth. T. Wright, On Antiquarian
excavations and researches in the Middle Ages, Archaeologia 30 (1844), pp. 43857.
Bedes Ecclesiastical History 4.19, eds. Colgrave and Mynors, pp. 3929. D. Rollason,
Saints and relics in Anglo-Saxon England (Oxford, 1989), pp. 3441. Objects such as
Aethelthryths tomb conferred on the church in which it would be located the sta-
tus of the possession of an exclusive good, one which alluded to the Roman past
and possibly possessed amuletic powers. R. White, Roman and Celtic Objects from Anglo-
Saxon graves: A catalogue and interpretation of their use. BAR British Series 191 (Oxford,
1988), pp. 15965.
19
. . . uidelicet quia fana idolorum destrui in eadem gente minime debeant, sed ipsa, quae in
eis sunt, idola destruantur. Aqua benedicta fiat, in eisdem fanis aspargatur, altaria construantur,
reliquiae ponantur, quia, si fana eadem bene constructa sunt, necesse est ut a cultu daemonum in
obsequio ueri Dei debeant commutari, ut, dum gens ipsa eadem fana sua non uidet destrui, de
corde errorem deponat et Deum verum cognoscens ac adorans ad loca quae consuerit familiarius
concurrat., Gregory the Great, Registrum Epistularum Libri VIIIXIV, Appendix XI,
p. 56, ed. D. Norberg, CCSL 140A (Turnhout, 1982), pp. 9612. Bedes Ecclesiastical
history of the English people, 1.30, eds. Colgrave and Mynors, pp. 1069.
20
The takeover of these sites entailed continuity of tradition (though not belief )
between the pagan and Christian periods. E. Mle, La fin du paganisme en Gaule (Paris,
1950), pp. 3269. Mles conclusions have been echoed in more recent analyses
looking at Christian re-use of formerly pagan sites. M. Roblin, Fontaines sacres
et ncropoles antiques, deux sites frquents dglises paroissiales rurales dans les sept
anciens diocses de lOise, Revue dhistoire de lglise de France 62 (1975), pp. 23551.
A. Rousselle, Croire et gurir: La foi en Gaule dans lantiquit tardive (Paris, 1990), pp.
197200.
 99

intervals of a century or more. He has argued that discontinuity

between the occupation of the Roman-period sites and the re-use of
ruins hundreds of years later made this transition practical rather
than reflective of religious appropriation.21 In acknowledging exist-
ing gaps between the destruction of Roman-period sites and their
reoccupation in the early Middle Ages, however, Young has limited
his discussion to the pagan fanum, a square structure notoriously
difficult to identify as a religious center without textual references or
stray finds of cultic objects. Written works are particularly prob-
lematic in view of Christian constructs of paganism in the conver-
sion period and afterwards. Nor has Young dealt with other sorts
of ruins such as former villas and fortifications, despite the fact that
in the imperial period the former often included the equivalent of
private sanctuaries.22
Archaeologists and historians must therefore recognize the com-
plexity of the relationship between early medieval populations and
ancient landscapes. Although many early medieval structures were
built on new sites, Roman ruins played an influential role in deter-
mining the layout and perception of villages, churches and ceme-
teries both in the north and south of Gaul.23 As has been shown in
Gaul and elsewhere in early medieval Europe, physical remnants of
prehistoric activity such as burial mounds also served practical needs
such as grave sites, boundary markers and landmarks, and were
sometimes copied in contemporary funerary rites.24 Their ubiquitous
presence conceivably struck a deeper chord, shaping regional and

21
B.K. Young, Que restait-il de lancien paysage religieux lpoque de Grgoire
de Tours? in: N. Gauthier and H. Galini eds., Grgoire de Tours et lespace Gaulois:
Actes du congrs international, Tours, 35 novembre 1994, 13e supplment la Revue
archologique du Centre de la France (Tours, 1997), pp. 24150.
22
X. Lafon, A propos de Saint Ulrich: Villas et lieux de culte dans la Gaule
du Nord-Est, in: Aspects de la religion celtique et gallo-romaine dans le Nord-Est de la Gaule
la lumire des dcouvertes rcentes (Saint-Di-des-Vosges, 1989), pp. 5972.
23
J. Percival, The Roman villa: An historical introduction (Berkeley, 1976), pp. 16999.
James, Archaeology and the Merovingian monastery, p. 47. . Salin, La civilisa-
tion mrovingienne daprs les spultures, les textes et le laboratoire 2 (Paris, 1952), pp. 1222.
H. Steuer, Frhgeschichtliche Sozialstruckturen in Mitteleuropa: Eine Analyse der Auswertungsmethoden
des archologischen Quellenmaterials. Abhandlungen der Akademie der Wissenschaften in
Gttingen, philologisch-historische Klasse, dritte Folge 128 (Gttingen, 1982), p. 374.
24
On grave mounds, see: H. Ament, Merowingische Grabhgel, in: Walter
Schlesinger ed., Althessen im Frankenreich. Nationes. Historische und philologische
Untersuchung zur Entstehung der europaschen Nationen im Mittelalter 2 (Sigmarin-
gen, 1975), pp. 858. K. Sippel, Die Kenntnis vorgeschichtlicher Hgelgrber im
Mittelalter, Germania 58 (1980), pp. 13746. C. La Rocca, Using the Roman past.
100

ethnic identity by directing or transforming the memory of the ear-

lier history of a particular region.25 In the case of cemeteries, the
presence of ancient monuments may have helped contemporary
inhabitants defend claims to the land by means of reference to real
or fictive ancestors from the region.26 The few ruins identified in
hagiographical and historical compositions of the sixth and seventh
centuries in Gaul pointed to consciousness of their potential for
manipulation in clerical discourse.
As a consequence of significant tensions between written and mate-
rial sources, however, any interpretation of perceptions of ruined sites
or artifacts in early medieval Gaul must remain tentative. Although
many structures must have been abandoned for long periods of time
before their reoccupation by Christians, as demonstrated above, re-
use might also happen in close succession to destruction. In addi-
tion, the early medieval landscape included many sorts of ruins.
These buildings must have often had multiple functions which, in
any case, were not so easily discerned after the passage of a century
or more. Despite Youngs insistence that a new Christian topography
characterized Merovingian Gaul in the time of Gregory of Tours,27
both archaeological and written evidence reveal that these sacred
sites often had clear links to those existing previously. Ancient remains
held the allure of precious or scarce materials in a form ready to
use, and members of the late antique and early medieval elite often
adapted them to advance their own careers, whether in restoring a
site or pillaging it for the purpose of constructing another. Although
to a certain extent malleable, ruined monuments and ancient objects
also had powerful referential qualities, and thus might function as

Abandoned towns and local power in eleventh-century Piemonte, EME 5 (1996),

pp. 4569. H.A. MacDougall, Racial myth in English history: Trojans, Teutons, and Anglo-
Saxons (Montreal, 1982), pp. 718.
25
Patrick Prin argues that possession of ruins of former religious sites might
have also had implications for social status rather than solely religious practices.
P. Prin, Remarques sur la topographie funraire en Gaule mrovingienne et
sa priphrie: Les ncropoles romaines tardives aux ncropoles du haut-moyen ge,
Cahiers archologiques 35 (1987), pp. 201.
26
G. Halsall, Settlement and social organization: The Merovingian region of Metz (Cambridge,
1995), pp. 1815.
27
He notes: Holiness for them, was a new condition imposed on a previously
neutral spot by a sacralising Christian act, B.K. Young, Sacred topography and
early Christian churches in late antique Gaul, in: R.F.J. Jones, J.H.F. Bloemers,
S.L. Dyson and M. Biddle eds., First millenium papers: Western Europe in the first mille-
nium A.D., BAR International Series 401 (Oxford, 1988), pp. 2212; 2357.
 101

mnemonic repositories.28 As noted by Jonathan Z. Smith, holy sites

on ruins did not operate independently of the memories of the past
active among the population there, although contemporary or sub-
sequent narratives might alter the latter. Monuments instead served
to focus the rituals performed there in a very effective manner, and
hence contemporaries frequently considered ancient places sacred.29
Smiths discussion helps to modify Pierre Noras conception of lieux
de mmoire in establishing a paradigm for the process of transforma-
tion of ancient ruins into early medieval sacred space. Nora has sug-
gested that the reason for the existence of modern sites of memory
is the lack of a ritual structure of living memory that allows for the
natural continuation of traditions. In his view, in modern society,
lieux de mmoire and the rituals associated with them must be con-
sciously maintained in order to preserve a context for artifacts of
the past. These sites no longer exist spontaneously, but must be sus-
tained through the sponsorship of activities at them.30 Although Nora
believes that this process is limited to the modern world, in late
Antiquity just as in the early Middle Ages, lay and clerical author-
ities played an active and conscious role in shaping such customs.
Ancient sites provided highly valued material remains, the uses of
which were controlled by none other than fourth- and fifth-century
emperors and bishops for a variety of political and practical purposes.
Despite their rhetoric that such remnants of the ancient past were
potentially harmful if misused, ruins clearly attracted elites. Early
medieval clerical leaders appropriated and adapted the same sort of
ancient structures and fragments in the transformation of the landscape
of Gaul, albeit in a rather different fashion.31 The use in a new con-
text of such remains contributed powerfully to the construction of
Christian community and identity in early medieval Western Europe.

28
The malleability of spolia made them very attractive, since the intended use of
an object could be negated and yet the object continued to contribute to the form
and symbolism (in some cases more blantantly than others) of the new creation.
H. Westermann-Angerhausen, Spolie und Umfeld in Egberts Trier, Zeitschrift fr
Kunstgeschichte 50 (1987), pp. 3056.
29
J.Z. Smith, To take place: Toward theory in ritual (Chicago, 1987), pp. 838; 1037.
30
P. Nora, Between memory and history: Les lieux de mmoire, Representations 26
(1989), pp. 712. E. Hobsbawm, Introduction: inventing traditions, in: E. Hobsbawm
and T. Ranger eds., The invention of tradition (Cambridge, 1983), pp. 12.
31
I thus strongly disagree with Maurice Halbwachs conception of the Christian
rite as unchanging. M. Halbwachs, On collective memory, ed. and transl. L.A. Coser
(Chicago, 1992), pp. 84119.
102

Although it is difficult to ascertain why late antique and early

medieval clerics in Gaul did not find sites formerly associated with
paganism more dangerous, bishops leadership in this custom was
not so different from their innovative patronage of saints relics in
the protection of their communities. From the New Testament
onwards, demons who tempted the faithful and lured them from sal-
vation had played a central role in Christian perceptions of the cos-
mos. In Gaul, Christian rites of exorcism therefore provided a focal
point for the creation of communal integration and group identity.32
The liturgy highlighted the lives of the saints as a microcosm of the
struggles of Christian society; their personal combat against demons
demonstrated the triumph of the community of the faithful over
supernatural adversaries.33 The result of such recitations was a grow-
ing belief in the necessity of Christian holy men and women, increas-
ingly but not exclusively replaced in the West by their more reliable
relics in the care of local bishops. They provided protection for
Christian souls of the living and the dead.34
By taking a leading role in directing the laity, Merovingian bish-
ops collaborated in the sanctification of place and time by their
ritual activity.35 By the sixth century, bishops and abbots were
unsurprisingly the individuals portrayed as playing the most active
part in transforming ancient ruins. The process by which they repos-
sessed ancient structures differed significantly from the rites of exor-
cism performed by Gregory Is missionaries in England on the pagan
shrines that dotted the landscape.36 The latter groups precautions

32
P. Brown, The rise and function of the holy man in late Antiquity, in his
Society and the Holy in late antiquity (Berkeley, 1982), pp. 1236. J.M. Petersen, Dead
or alive? The holy man as healer in East and West in the late sixth century,
Journal of Medieval History 9 (1983), pp. 918. R. Van Dam, Bodily miracles, in
his Saints and their miracles in late antique Gaul (Princeton, 1993), pp. 8993.
33
M. Van Uytfanghe, Stylisation biblique et condition humaine dans lhagiographie mrovingi-
enne (650750), Verhandelingen van de Koninklijke Academie voor Wetenschnappen,
Letteren en Schone Kunsten van Belgi, Klasse der Letteren 49, 120 (Brussels,
1987), pp. 10210.
34
P. Brown, Eastern and western Christendom in late Antiquity: A parting of
the ways, in: idem, Society and the Holy, pp. 17890. A manifestation of the grow-
ing importance of the cult of saints may be measured through the practice of ad
sanctos burial in the West. Extensive documentation is provided in: Y. Duval, Auprs
des saints corps et me: Linhumation ad sanctos dans la chrtient dOrient et dOccident du
III e au VII e sicle (Paris, 1988).
35
W.E. Klingshirn, Caesarius of Arles: The making of a Christian community in late
antique Gaul (Cambridge, 1994), pp. 1549.
36
Gregorys letter did not refer specifically to ruins but rather to the pagan sanc-
 103

resembled more closely the activities of fifth-century continental saints

still facing the competition of pagan adversaries. The confidence of
Merovingian clerics in incorporating pagan remains as the founda-
tion of their building projects by contrast, indicated Christianitys
ultimate triumph over paganism in Gaul.

L R

Study of the late antique practices from which customs of re-use

arose provides a clearer understanding of the early medieval attrac-
tion to ancient sites. The late imperial tradition of excavating ancient
ruins contributed to early medieval customs of re-use, although there
would be important innovations in the practice in the context of
Gaul. Of great significance for later Christians, for instance, were
accounts of Constantines use of spolia in constructing imperial and
religious structures.37 In his search for holy sites from the earliest
days of Christianity, however, Constantine apparently did not view
pagan religious remains in such a positive light. According to Eusebius
of Caesarea, the emperor excavated and discarded the foundations
of the temple of Aphrodite prior to constructing the Church of the
Holy Sepulcher.38 Only ancient remains of Christian significance
figured large in this account.
The famous legend that arose in the late fourth and early fifth
century around Helenas supposed discovery of the True Cross at

tuaries ( fana) still operating in sixth-century England. Gregory the Great, Registrum
Epistularum XI, 56, ed. Norberg, CCSL 140a, pp. 9612. P. Brown, The rise of west-
ern Christendom: Triumph and diversity A.D. 2001000 (Malden, 1996), p. 209.
37
The earliest sites at which spolia were used extensively were the Lateran Basilica
(begun in 313) and the Arch of Constantine (consecrated in 314). B. Brenk, Spolia
from Constantine to Charlemagne: Aesthetics versus ideology, Dumbarton Oaks Papers
41 (1987), pp. 1039. According to Eusebius, the emperors attention to sites such
as the Holy Sepulcher, perceived by pagan critics as one of many memorials to
human corpses and tombs, was seen by them as unfitting and demeaning., Eusebius
of Caesarea, Oratio de laudibus Constantini 11.3, transl. H.A. Drake, In praise of Constantine:
A historical study and new translation of Eusebius Tricennial Orations (Berkeley, 1976),
p. 103.
38
Eusebius, Vita Constantini 3. pp. 257, transl. J. Wilkinson, in: Egerias travels to
the Holy Land, revised edition ( Jerusalem, 1981), pp. 1645. Brenk Spolia from
Constantine, p. 107. Smith, To Take Place, pp. 7782. Most recently on Constantines
excavation of and building campaign at the site of the holy sepulcher and Calvary,
see: E.B. Hunt, Constantine and Jerusalem, Journal of Ecclesiastical History 48 (1997),
pp. 4119.
104

Golgotha, was even more revealing of such a process. It focused on

the extraction of the precious cross and nails from ground desecrated
by a statue of Venus. These relics were transported to the capital
of the empire so that they might be venerated appropriately in honor
of the crucifixion. Ambrose thus noted that even when these remains
were amidst pagan ruins, nothing could obliterate the triumph of
Christ.39 The restoration of the True Cross granted early Christian
leaders and their successors a powerful symbol of Christs suffering,
since pieces of it might be bestowed upon well-placed Christian pat-
rons and faithful supporters throughout Christendom.40 In contrast,
if one may believe the ecclesiastical narrators, contemporaries viewed
the related pagan remains primarily as a source of pollution.
Not all fourth-century emperors exhibited positive attitudes regard-
ing the benefits of excavating ancient ruins, since they associated
these practices with illicit and dangerous magic. Ammianus Marcellinus
described the way in which those who were caught lurking near
tombs (and thus likely excavating) were assumed to be engaging in
sorcery or necromancy, both crimes punishable by death under
Constantius II.41 On 9 September 364, Valentian and Valens enacted

39
Aperit itaque hu mum, decutit pulverem, tria patibula confusa repperit, quae ruina contex-
erat, inimicus absconderat. Sed non potuit oblitterati Christ triumphus., Ambrose, De Obitu
Theodosi 45, ed. O. Faller, CSEL 73 (Wien, 1955), p. 394. English translation may
be found in: E.D. Hunt, Holy Land pilgrimage in the later Roman Empire A.D. 312 460
(Oxford, 1982), pp. 2849. For additional information on the early history of the
legend of Helena and the True Cross: J.W. Drijvers, Helena Augusta: The mother of
Constantine the Great and the legend of her finding of the True Cross (Leiden, 1992), esp.
pp. 79117.
40
Therasia, the wife of Paulinus of Nola sent a piece of the relic of the True
Cross, which she had received from Melania the Younger, to Sulpicius Severus
mother-in-law Bassula. For this event and the case of Radegund of Poitiers acqui-
sition of a piece of the True Cross: I. Moreira, Provisatrix optima: St Radegund of
Poitiers relic petitions to the East, Journal of Medieval History 19 (1993), pp. 285305.
E.G. Whatley, An early literary quotation from the Inventio S. Crucis: A note on
Baudonivias Vita S. Radegundis (BHL 7049), Analecta Bollandiana 111 (1993), pp.
8191. Gregory of Tours wrote that Helena, helped by a Jew named Judas, was
responsible for the discovery of the relic of the Holy Cross. Gregory of Tours, Decem
libri historiarum 1.36, eds. Krusch and Levison, MGH SRM 1, 1, p. 27.
41
Nam siqui remedia quartanae vel doloris alterius collo gestaret, sive per monumentum tran-
sisse vesperi, malivolorum argueretur indiciis, ut veneficus, sepulchrorumque horrores, et errantium
ibidem animarum ludibria colligens vana, pronuntiatus reus capitis interibat., Ammianus
Marcellinus, Res gestae 19.12.14, ed. and transl. J.C. Rolfe, revised edition (Cambridge,
1956), p. 541. A.A. Barb, The survival of the magic arts, in: A. Momigliano ed.,
The conflict between paganism and christianity in the fourth century (Oxford, 1963), pp. 1034.
The differences between these attitudes and those of three centuries earlier, when
 105

similar legislation banning nocturnal magic along with funereal sac-

rifices.42 The imperial custom of persecuting sorcerers in the fourth
century was commonly directed against groups not profoundly affected
by social dislocation; these accusations represented the means by
which the imperial court controlled the traditional aristocracy. Although
the laws would have a much longer legacy, as they were inherited
by jurists in the early Germanic kingdoms by means of the Theodosian
Code,43 fear of sorcery with respect to ruined sites did not continue
to play such a prominent role in succeeding centuries. While Augustine
noted that accusations of Christian sorcery included the mutilation
and burial of human remains in conjunction with curses and dia-
bolical rituals included among the magical arts, he did not link these
activities to excavation.44 By the end of the sixth century, leaders
mainly directed accusations of sorcery against political enemies or
non-Christians, in particular the Jews.45
Much of the growing interest in excavating graves in the late
fourth century stemmed from the increasing significance of the cult
of saints, in spite of the revulsion with which the remains of the
dead had traditionally been viewed by Romans. Great popular
pressure on bishops to produce the remains of indigenous martyrs
thus faced resistance in the form of imperial legislation banning relic

inhumation was rarely practiced in the Roman world, are striking. Pliny, for instance,
noted with wonder the way in which some stone sarcophagi were capable of con-
suming the bodies of the dead. Pliny, Naturalis historiae libri XXXVII 36.131, transl.
D.E. Eichholz (Cambridge, Mass., 1962), pp. 1047. A.D. Nock, Cremation and
burial in the Roman Empire, in: Z. Steward ed., Essays on religion and the ancient
world 1 (Cambridge, Mass., 1972), pp. 27880.
42
Ne quis deinceps nocturnis temporibus aut nefarias preces aut magicos apparatus aut sacrificia
funesta celebrare conetur. Detectum enim adque convictum conpetenti animadversione mactari perenni
auctoritate censemus. Codex Theodosianus 9.16.7, ed. T. Mommsen, third edition (Berlin,
1962), p. 462.
43
I.N. Wood, The Code in Merovingian Gaul, in: J. Harries and I.N. Wood
eds., The Theodosian Code (Ithaca, 1993), pp. 16169. I.N. Wood, Spultures ecclsi-
astiques et snatoriales dans la valle du Rhne (400600), Mdivales 31 (1996),
pp. 167.
44
Augustine of Hippo, De Civitate Dei 18.53, eds. B. Dombart and A. Kalb, CCSL
48 (Turnhout, 1955), pp. 6523.
45
P. Brown, Sorcery, demons, and the rise of Christianity from late antiquity
into the Middle Ages, in: M. Douglas ed., Witchcraft: Confessions and accusations
(London, 1970), p. 35. Ralph Mathisens stance is more problematic, since he calls
many acts magical when in fact contemporary clerics interpreted them as demonic
in nature. R.W. Mathisen, Crossing the supernatural frontier in Western late antiq-
uity, in R.W. Mathisen and H.S. Sivan eds., Shifting frontiers in late antiquity (Aldershot,
Hampshire, 1996), pp. 30920.
106

translations in February, 386.46 As a result, Ambrose of Milans highly

politicized excavation and transfer of the relics of Gervasius and
Protasius intra muros later that same year were the source of great
contention. His triumphant confirmation both through the miracles
of the saints and the support of the general populace allowed him
to overcome imperial opposition.47 This event represented an impor-
tant climax in rejecting legislation against this manifestation of the
cult of relics, and similar events transpired in Gaul in the following
centuries.48
In addition to banning necromancy at cemeteries, imperial legis-
lation of the 340s and 350s concerned itself with the preservation
of monuments. At risk were the precious columns of marble and
other valuable materials such as bronze and lead that adorned pub-
lic and private structures. Cut stone derived from tombs and other
monuments might be incorporated into villas, churches, or city walls,
and also represented a source of income if sold.49 Although they
banned such practices among their subjects, however, emperors con-
tinued to appropriate spolia to promote their own dignity in what
Beat Brenk has termed a new aesthetic, since these older fragments
were incorporated into more recent edifices. The confiscation of spo-
lia by anyone but the emperor was thus seen as an affront to the

46
Humatum corpos nemo ad alterum locum transferat; nemo martyrem distrahat, nemo mer-
cetur. Habeant vero in potestate, si quolibet in loco sanctorum est aliquis conditus, pro eius ven-
eratione quod martyrium vocandum sit addant quod voluerint fabricarum., Codex Theodosianus
9.17.7, p. 466, ed. Mommsen. N.B. McLynn, Ambrose of Milan: Church and court in
a Christian capital (Berkeley, 1994), pp. 20917.
47
In a letter to his sister Marcellina, Ambrose described the discovery of their
relics: Invenimus mirae magnitudinis viros duos ut prisca aetas ferebat. Ossa omnia integra, san-
guinis plurimum., Ambrose, Epistulae 77.12, ed. M. Zelzer, CSEL 82,3 (Wien, 1982),
pp. 1268. D.H. Williams, Ambrose of Milan and the end of the Nicene-Arian conflicts
(Oxford, 1995), pp. 21923. Brown, The cult of saints, p. 37.
48
This event by no means constituted the end of the struggle, as a novella of
Valentinian III in 447 was likely directed against the relic trade, and specified high
penalties for those who violated the precept. J. Harries, Death and the dead in
the late Roman Empire, in: S. Bassett ed., Death in towns: Urban responses to the dying
and the dead, 1001600 (London, 1992), pp. 615.
49
See, for instance: Codex Theodosianus 9.17.15, ed. Mommsen, pp. 4635. In
the Variae, Cassiodorus included Theoderics letter to Sabinianus (507512), which
noted the expense of the preservation of public structures and the costs incurred
by those who stole supplies reserved for upkeep. In addition, in a letter to the
Senate (510511), Theoderic deplored those who stole from buildings and diverted
the water of the aqueducts. Cassiodorus, Variarum Libri XII 1.25 and 3.31, ed. A.J.
Fridh, CCSL 96 (Turnhout, 1973), p. 33; pp. 11920.
 107

ruling authority.50 By contrast, the remains from earlier structures,

once applied to late antique religious and state buildings, served as
propaganda of imperial status, and also represented a source of civic
pride.
Scholars must therefore treat with great caution legal measures
condemning architectural applications of spolia. The re-use of ancient
remains in the late antique and early medieval West apparently accel-
erated over time. In sixth-century Italy, Theoderic pillaged ancient
structures when it was to his advantage.51 Late fifth-century Christian
bishops in the south of Gaul such as Hilarius of Arles also adopted
similar policies, but notably did not share imperial attitudes toward
the importance of civic structures. In one passage in which Honoratus
of Marseilles noted Hilarius deep involvement with his flock, he also
wrote that the bishop had harvested marble blocks to build new
basilicas from civic monuments such as Arles theater.52 These spo-
lia not only served to honor the churches but also promoted the sta-
tus of the bishop to whom the foundation of these structures was
attributed. As discussed above, these attitudes became increasingly
prevalent in the early medieval Gaul.
Some ancient sites, however, were viewed with greater trepida-
tion. In late fifth-century Gaul, Constantius of Lyons account of the
life of Germanus of Auxerre included a brief reference to the explo-
ration and re-use of a haunted house. Constantius recounted that
the bishop, directed by a ghost, combed through the rubble and
found the unburied skeletons of two chained corpses. Once interred

50
C.A. Marinescu, Transformations: Classical objects and their re-use during
late antiquity, in: Shifting frontiers, pp. 28586. Brenk, Spolia from Constantine,
pp. 1045.
51
Theoderic saw the citys fortune reflected in its buildings in a letter directed
to the landowners of Arles in the winter of 508509. . . . ut et largitatis remedio civibus
consulamus et ad cultum reducere antiqua moenia festinemus. Sic enim fiet, ut fortuna urbis, quae
in civibus erigitur, fabricarum quoque decore monstretur., Cassiodorus, Variarum Libri XII
3.44, ed. Fridh, CCSL 96, p. 127. Yet, in a letter dated 507/511, he ordered that
all buried treasure become property of the public treasury. Cassiodorus, Variarum
Libri XII 4.34, ed. Fridh, CCSL 96, p. 164. A. Schnapp, The discovery of the past:
The origins of archaeology (London, 1996), pp. 834. M. Hardt, Royal treasures and
representation in the Early Middle Ages, in: W. Pohl and H. Reimitz eds., Strategies
of distinction: The construction of ethnic communities, 300800 (Leiden, 1998), p. 275.
52
Hilarius deacon was responsible for the quarrying activities: Qui basilicis prae-
positus construendis, dum marmorum crustas et theatri proscenia celsa deponeret, fidei opere nudans
loca luxuriae, quod sanctis parabat ornatibus . . ., Honoratus of Marseille, La vie dHilaire
dArles 20, ed. and transl. P.A. Jacob, SC 404 (Paris, 1995), pp. 1347. Y. Hen,
Culture and religion in Merovingian Gaul, A.D. 481751 (Leiden, 1995), pp. 2289.
108

properly with appropriate prayers, the ghosts stopped haunting the

site and the house no longer caused any fear in the local commu-
nity. Soon afterwards, the structure could be restored and reoccu-
pied.53 Likewise, the demons disturbing the remains of former baths
in the parish of Succentriones ceased to torment local inhabitants
after Caesarius of Arles hung his staff from a wall of the structure.
The holy mans gesture caused the demons to flee, and no one in
the vicinity thereafter faced any danger in conjunction with the
ruins.54 In the eyes of both authors, the unburied dead and demons
rather than the ruins themselves represented cause for concern.
Subsequent measures in various parts of Gaul distinguished between
the re-use of ancient ruins and the destruction and spoliation of
grave sites. Early medieval clerics condemned the appropriation of
stone from monuments commemorating Christian dead not only as
unacceptable but as actions that resulted in the profanation of the
dead.55 Likewise, the Pactus Legis Salicae forbade the application of
funerary stone to the construction of a church at the same site.56 In
Visigothic Spain, the forty-sixth canon of the Fourth Council of
Toledo (633) penalized clerics partaking in the destruction of sepul-

53
Erat eminus domicilium, tectis iam pridem sine habitatore semirutus, quod etiam per incu-
riam vulgaria arbusta contexerant. . . ., Constantius of Lyon, Vie de Saint Germain dAuxerre
10, ed. and transl. R. Borius, SC 112 (Paris, 1965), pp. 13843. The account of
the saints cult being discredited as the veneration of a thief in Sulpicius Severus
Vita Sancti Martini did not involve excavation. Sulpicius Severus, Vie de Saint Martin
11.15, ed. and transl. J. Fontaine, SC 133 (Paris, 1967), pp. 2767. J.-C. Schmitt,
Les revenants: Les vivants et les morts dans la socit mdivale (Paris, 1994), pp. 434.
N. Caciola, Wraiths, revenants and ritual in medieval culture, Past and Present 152
(1996), pp. 36.
54
Mirantur et agunt Deo gratias virgamque ipsam de pariete illo suspendunt, deferentes ad
locum, ubi daemonum incursus audiebantur: statimque effugatae sunt insidiae diaboli, et ultra nulli
in loco nequissimum malum facere usque hodie adversarius ipse praesumpsit., Vitae Caesarii
Episcopi Arelatensis Libri Duo 2.22, ed. B. Krusch, MGH SRM 3 (Hannover, 1896),
p. 493.
55
Epistolae Arelatenses Genuinae 35, ed. W. Gundlach, MGH Epp. 1 (Berlin, 1892),
p. 54. For more on legislation forbidding profanation of the dead, see: B. Effros,
From grave goods to Christian epitaphs: evolution in burial tradition and the expression of social
status in Merovingian society, unpublished doctoral dissertation (Los Angeles, 1994), pp.
24350.
56
Pactus Legis Salicae (C6) 55.6, ed. K.A. Eckhardt, MGH LL 4, 1 (Hannover,
1962), p. 209. H. Nehlsen, Der Grabfrevel in den germanischen Rechtsaufzeichnungen:
Zugleich ein Beitrag zur Diskussion um Todesstrafe und Friedlosigskeit bei den
Germanen, in: H. Jankuhn, H. Nehlsen, and H. Roth eds., Zum Grabfrevel in vor-
und frhgeschichtlicher Zeit: Untersuchungen zu Grabraub und Haugbrot in Mittel- und
Nordeuropa. Abhandlungen der Akademie der Wissenschaften in Gttingen, philolo-
gisch-historische Klasse, dritte Folge 113 (Gttingen, 1978), p. 161.
 109

chers with similar measures.57 Rather than alluding to any magical

implications of these activities, bans on cemeterial spolia sought to
prevent pillaging of the stone sepulchers of the elite as well as the
lavishly outfitted tombs of the saints.58 Predictably, repeated mea-
sures against grave robbery did not enjoy any greater success in
Merovingian Gaul than they did elsewhere.59
Such attitudes did not, in contrast, extend to the early medieval
excavation of buried treasure not found in conjunction with corpses.
Gregory of Tours recounted two stories of recovered treasure hoards
without indicating any apparent revulsion from pillaging ancient prop-
erty. After the murder of Mummolus by the duke Leudegisel in 585,
the king forced Mummolus widow to reveal the source of their
wealth; she admitted that a treasure trove had been uncovered some
time earlier by her husband. The king Guntram confiscated the gold
and silver, and after dividing it with his nephew Childebert, he
donated some of his gains to the poor.60 In this text, contemporaries
did not perceive pillaged goods suspiciously or any differently than
property plundered from enemies.
Nor were unexpected finds of great wealth limited to men of the
world. Gregorys hagiographical account of the fifth-century abbot
Lupicinus involved the discovery of ancient buried treasure. After
having received divine revelation of the whereabouts of the gold and
silver, the saint went alone to this place and took back to the abbey
as much as he could carry from the hoard. By doing this annually
for some time, Lupicinus was able to provide plentiful refreshment

57
Si quis clericus in demoliendis sepulchris fuerit deprehensus, quia facinus hoc pro sacrile-
gio legibus publicis sanguine vindicatur, oportet canonibus in tali scelere proditum a clericatus
ordine submoveri et poenitentiae triennium deputari., Concilios visigticos e hispano-romanos, ed.
J. Vives (Barcelona, 1963), pp. 2078. Nehlsen, Der Grabfrevel, p. 117.
58
Saints tombs covered with gold were a constant source of temptation. Gregory
of Tours, In gloria martyrum 71, ed. B. Krusch, MGH SRM 1, 2, pp. 5356. Gregory
of Tours, De virtutibus S. Juliani 20, ed. B. Krusch, MGH SRM 1, 2, p. 573.
59
Sixth-century Salian law included measures against grave robbery; the crime
in some cases merited the harsh punishment of exile for the guilty party until the
goods were restored and fines were paid. Pactus Legis Salicae (C6) 14.10 and 55.4,
MGH LL 4, 1, pp. 689; pp. 2067. On the robbery of the grave of Gunthram
Bosos female relative at Metz, see: Gregory of Tours, Decem libri historiarum 8.21,
MGH SRM 1, 2, pp. 3878. G. Halsall, Female status and power in early
Merovingian central Austrasia: The burial evidence, EME 5 (1996), pp. 12.
60
Ferunt autem ducenta quinquaginta talenta argenti fuisse, quae in urbe relicta fuerant. Ferunt
autem ducenta quinquaginta talenta argenti fuisse, auri vero amplius quam triginta. Sed haec, ut
ferunt, de reperto antiquo thesauro abstulit., Gregory of Tours, Decem libri historiarum 7.40,
MGH SRM 1, 1, p. 363. Hardt Royal treasures, p. 275.
110

for his congregation. In this particular example, Gregory demon-

strated that it was the sanctity of the abbot that merited Gods inter-
vention in this fashion, although not all contemporaries viewed the
lavish meals of the monks as appropriate.61 In some cases, however,
Gregory gave greater attention in his writings to the negative con-
notations of searching for precious metals. Near the Rhne, for
instance, Gregory of Tours noted that a landslide killed a number
of avaricious monks seeking to gain profit by digging for bronze and
iron.62 Just as with ancient ruins, authors manipulated the interpre-
tation of buried wealth in a manner that suited their objectives.

A -

Archaeological remains have helped to document more thoroughly

the late antique and early medieval use of ruins in Gaul. Nonetheless,
the tendency to read too much into this evidence has led in some
instances to untenable conclusions, and thus finds must be assessed
cautiously. Undisputably, however, in the construction and expan-
sion of stone edifices, pagan Roman structures often provided a con-
venient source of building material from as early as the fourth century
especially south of the Loire. Yet if the example of the fourth-cen-
tury graves of six newborns, none older than six months and all
buried in a second- to third-century structure at Fleurheim (Eure)
sheds any light on this topic, it indicates that not only practicality
motivated the repossession of ancient remains.63 A later example of
a similar rite transpired at the Roman villa of Berthelming (Moselle),
rebuilt after its destruction in the late fourth century. By the early

61
Lupicinus igitur abba cum minus haberet, unde tantam susteneret congregationem, revelavit
ei Deus locum in heremo, in quo antiquitus thesauri reconditi fuerant. Ad quem locum accedens
solus, aurum argentumque, quantum levare potuerat, monasterio inferebat et, exinde coemptos cibos,
reficiebat fratrum multitudines, quos ad Dei officium congregaverat. Sicque faciebat per singulos
annos. Nulli tamen fratrum patefecit locum, quod ei Dominus dignatus est revelare., Gregory
of Tours, Vitae Patrum 1.3, pp. 2156. Schnapp, The discovery of the past, p. 87.
62
. . . triginta monachi, unde caster ruerat, advenerunt, et terram illam, quae monte deruente
remanserat fodientes, aes sive ferrum repperiunt. Quod dum agerent, mugitum montes, ut prius
fuerat, audierunt. Sed dum a saeva cupiditate retenerentur, pars illa quae nondum deruerat super
eos cecidet, quos operuit atque interfecit, nec ultra inventi sunt., Gregory of Tours, Decem libri
historiarum 4.31, p. 164. I.N. Wood, Gregory of Tours (Bangor, 1994), p. 25.
63
M.A. Dollfus and A. Guyot, Spultures de nouveau-ns dans les fouilles gallo-
romaines de Fleurheim Lyons-La-Fort (Eure), Annales de Normandie 18 (1968),
pp. 283300.
 111

medieval period, contemporaries used the structure for burials. By

the seventh century, twenty-four graves were aligned with the walls
of the villa.64 Merovingian graves sometimes also clustered at sites
of prehistoric megaliths.65 Reasons entirely undiscernable to scholars
in the twentieth century must have also existed for such practices.66
Even in the south of Gaul where there was less discontinuity from
the arrival of Germanic warriors, archaeological remains document
a process of reoccupation of civic and religious structures in the late
fifth and sixth centuries. It is not known, however, whether all already
stood in ruins at the time or whether they fell victim to various bish-
ops and abbots aggressive campaigns of church building. The cathe-
dral and baptistery of Aix-en-Provence were constructed on the site
of the citys forum, the church of Saint-tienne of Strasbourg was
built on a former Roman road, the villa of Sviac (Gers) was trans-
formed into a baptistery and small chapel, and the ecclesiastical com-
plex of Cimiez (Alpes-Maritimes) and Saint-Pierre-aux-Nonnains in
Metz were installed in former thermal baths, to name many of the
most widely accepted examples of this phenomenon.67 At Tholey
(Saarland), a small chapel of early date was built on the ruins of a
Roman villa and late Roman fortifications abandoned in the fourth
century.68 Reminiscent of the sort of continuities characteristic of the
tels of the eastern Mediterranean, the refuse of previous generations
provided an enviable location for the construction of new buildings

64
Percival, The Roman villa, p. 164; p. 184. Although his ethnic and religious
interpretations of the graves are problematic, see: M. Lutz, La villa Gallo-Romaine
et la ncropole mrovingienne de Berthelming (Moselle), Revue archologique de lEst
et du Centre-Est 1 (1950), pp. 1804. E. Delort, Berthelming (Moselle), Gallia 6
(1948), pp. 23941.
65
C. Billard, F. Carr, M. Guillon and C. Treffort, Loccupation funraire des
monuments mgalithiques pendant le haut moyen ge. Modalits et essai dinter-
prtation, Bulletin de la Socit Prhistorique Franaise 93 (1996), pp. 27986.
66
For England, see most recently: H. Williams, Monuments and the past in
Early Anglo-Saxon England, World Archaeology 30 (1998), pp. 90108. I thank
Howard Williams for allowing me to see this manuscript based upon his M.A. the-
sis (1996) at the University of Reading before its publication.
67
N. Duval, Larchitecture cultuelle, in: Naissance des arts chrtiens: Atlas des mon-
uments chrtiens de la France (Paris, 1991), pp. 1879. For a comprehensive list of pro-
posed late Roman villas that became monasteries during the middle ages (some
outside the chronological scope of this essay), see: Percival Villas and monaster-
ies, pp. 1521. Percival, The Roman villa, pp. 16999.
68
P. Van Ossel, tablissements ruraux de lantiquit tardive dans le nord de la Gaule.
51e supplment Gallia (Paris, 1992), pp. 824. Percival Villas and monasteries,
p. 14.
112

and cemeteries. Just as at Jericho, a certain sense of authenticity was

gained by coopting ancient ruins in a way that would not have been
possible at a virgin site.69
Pagan religious structures may also be shown with certainty to
have been re-used in late Antiquity and the early Middle Ages on
a few occasions in the construction of sanctuaries and larger churches.70
In the case of Roujan (Hrault), a small but classic Roman temple
became the basis for a Christian chapel of undetermined date. A
very early example of re-use was found at Civaux, thirty kilometers
southeast of Poitiers; there, J.-C. Papinot has proposed that a fourth-
century baptistery was constructed over the site of two Gallo-Roman
temples, one of the first century and one of the third. The earliest
Christian epitaph discovered in conjunction with this site dated from
the fourth century.71 Outside Gaul, in what would become the cathe-
dral of Bonn, Emile Esprandieu also noted that multiple altars for
pagan goddess cults, and a number dedicated to Mercury, consti-
tuted part of the foundations of the fourth-century church. Some of
the antiquities were carried a number of kilometers before their
deposit. Upon their arrival, a number of the fragments were also
buried deeply in wells.72 The attention with which these artifacts
were handled for various sorts of re-use suggests that the remains
were neither found there coincidentally nor did they have solely util-

69
Van Ossel interprets reuse not so much as continuity of tradition but a ten-
dency to reoccupy the same site. In the north of Gaul, due to the widespread aban-
donment of villas in the fourth and fifth centuries, he believes that reoccupation
was less common than in areas south of the Loire. Van Ossel, tablissements ruraux,
pp. 7884. For a rather different perspective on the issue of the continuity of set-
tlement in the region of Metz, see: Halsall, Settlement and social organization, pp. 1815.
70
The work of Michel Roblin on the early medieval reuse of pagan water sites
as Christian baptisteries in the Oise has relied heavily on toponyms of a much later
period. This research is thus not reliable for the period under discussion here.
Roblin Fontaines sacres et ncropoles antiques, pp. 23551.
71
J.-C. Papinot, Notices sur les vestiges archologiques de Civaux (Poitiers, 1971), pp.
320. Duval, Larchitecture cultuelle, p. 189. V.I.J. Flint, The rise of magic in early
medieval Europe (Princeton, 1991), pp. 2628. For similar examples in England, see:
M. Hunter, Germanic and Roman antiquity and the sense of the past in Anglo-
Saxon England, Anglo-Saxon England 3 (1974), pp. 358.
72
The remains were found beneath the crypt of the Bonn cathedral. . Esprandieu,
Recueil gnral des bas-reliefs de la Gaule romaine 11 (Paris, 1938), no. 77607792, pp.
77106. Indeed Merrifield notes that forty-four of the 761 Roman sculptures recorded
by Esprandieu in Roman Germany were discovered built into churches; thirty-
three statues were found in wells in Esprandieus survey. Merrifield, The archaeol-
ogy, pp. 96106.
 113

itarian purposes. Instead, ancient remains played an integral ritual

function in the foundation of Christian sites of worship.
Re-use of pagan temples occurred also in later centuries, such as
at Saint-Georges-de-Boscherville (Seine-Maritime), excavated in 1981.
Archaeologists observed that under the cloister of the Benedictine
abbey lay a pagan temple that consisted of a stone chamber 7.6
8.5 m2, at which the most recent votive deposit dated from the third
century. It appears that burials on the interior of the chamber, and
outside of it at the base of the walls, began only in the seventh cen-
tury. At this time, a new entrance was constructed in the west of
the cella and a cemetery emerged to its east consisting of forty-two
graves, followed by a nave in the Carolingian period.73 Jacques Le
Maho further estimates that in the early medieval period, local
Christians adapted for re-use as many as forty sites in Normandy,
approximately a quarter of the Gallo-Roman structures which have
now been excavated. His conclusions are similar to those made rather
more hesitantly at Anthe (Namur), since it cannot be said with cer-
tainty that the late imperial structure in the late sixth- to early sev-
enth-century cemetery was a fanum subsequently transformed into a
Christian funerary chapel.74
Former Roman sites also yielded treasures for small-time collec-
tors and thieves, such as sarcophagi, shards of glass and pottery,
coins, brooches, and even cameos, many of which have survived due
to their incorporation as grave goods in early medieval burials
in the Merovingian world. 75 Scholars have documented similar

73
The fanum itself had been constructed at an ancient site over the remains of
three small wooden temples, one of which dated to the Augustan period. J. Le
Maho, La rutilisation funraire des difices antiques en Normandie au cours du
haut moyen ge, in: Lenvironnement des glises et la topographie religieuse des campagnes
mdivales: Actes du III e Congrs International dArchologie mdivale (Aix-en-Provence, 28
30 septembre 1989). Documents dArchologie Franaise 46 (Paris, 1994), pp. 1021.
J. Le Maho, Saint-Martin-de-Boscherville (Seine-Maritime), in: M. Fixot and
E. Zadora-Rio eds., Lglise, Le terroir. Monographie du Centre des Recherches
Archologiques 1 (Paris, 1989), pp. 639.
74
A. Dierkens, Un aspect de la christianisation de la Gaule du Nord lpoque
mrovingienne. La Vita Hadelini et les dcouvertes archologiques dAnthe et de
Franchimont, Francia 8 (1980), pp. 6237.
75
Fritz Fremersdorf noted the common use of Roman remains in the Merovingian
graves at Kln-Mngersdorf. F. Fremersdorf, Das frnkische Reihengrberfeld Kln-
Mngersdorf 1. Germanische Denkmler der Vlkerwanderungszeit 6 (Berlin, 1955),
pp. 245; pp. 435. For examples of the reuse of sarcophagi, see: E. Diehl ed.,
Inscriptiones Latinae Christianae Veteres 1, new edition (Dublin, 1970), no. 149, no. 1670,
p. 39; p. 324. E. Le Blant ed., Inscriptions chrtiennes de la Gaule antrieures au VIII e
114

practices in Anglo-Saxon England.76 In the region of Hainaut in

Belgium, for instance, numerous graves in twenty-four cemeteries
from the fifth through seventh centuries contained selected objects
pillaged usually from neighboring cremation burials from the first to
third centuries. These prized articles included ceramic vases, pitch-
ers, bronze fibules, and even a bracelet of amber beads. In one
remarkable case, the late sixth- or early seventh-century female grave
#917 at Ciply (Hainaut), revealed among other artifacts a first-century
ceramic vessel which, before its re-use, had been redecorated with
motifs similar to those found on funerary vessels in a time contem-
porary to the burial.77 The common admiration for ancient wares
may also be measured through discoveries such as at Hordain (Nord),
where sixth-century vessels for offerings imitated the designs and
forms of Gallo-Roman precedents.78 Ancient objects thus adapted
and reemployed had practical, decorative and amuletic functions in
the context of these graves, as well as served to express the identity
of the interred as perceived by his or her family or religious com-
munity. Rather than being feared, ancient objects were highly val-
ued. Only in the eighth century, when Frankish clerics formulated
prayers for vessels extracted from the ground, did liturgical customs
begin to reflect greater concern with the origins of excavated objects.
The Gellone sacramentary nonetheless acknowledged these objects
as signs of Gods rewards for the faithful.79

sicle 2 (Paris, 1865), no. 545, no. 628, pp. 3002; pp. 4947. G. Henderson, Early
Medieval. Medieval Academy Reprints for Teaching 29 (Toronto, 1993), pp. 1228.
76
A.L. Meaney, Anglo-Saxon amulets and curing stones. BAR British Series 96 (Oxford,
1981), pp. 192229. R. White, Scrap or substitute: Roman material in Anglo-
Saxon graves, in: E. Southworth ed., Anglo-Saxon cemeteries: A reappraisal. Proceedings
of a conference held at Liverpool Museum 1986 (Phoenix Mill, 1990), pp. 1325. I am
especially thankful to Carol Neuman de Vegvar for sharing with me a copy of the
unpublished paper she presented at the International Medieval Conference at
Kalamazoo, 912 May, 1996, entitled, The value of recycling: Conversion and
the early Anglo-Saxon use of Roman materials.
77
G. Faider-Feytmans, Objets dpoque romaine dcouverts dans des tombes
mrovingiennes du bassin de La Haine (Belgique), in: R. Chevallier ed., Mlanges
dArchologie et dHistoire offerts Andr Piganiol 2 (Paris, 1966), pp. 10118.
78
P. Demolon, Cimetire et chapelle mrovingiens Hourdain (Nord), in:
Larchologie en Hainaut-Cambrsis, Avesnois (Valenciennes, 1981), pp. 2930.
79
The prayer Oblationes super vasa reperta in locis antiquis noted: . . . et haec vascula
que [tuae] indulgentiae piaetatis post spatia temporum a voragine terrae abstracte humanis usibus
reddedisti . . ., Liber Sacramentorum Gellonensis, ed. A. Dumas, CCSL 159 (Turnhout,
1981), p. 450. W. Krmer, Zur Wiederverwendung antiker Gefe im frhen
Mittelalter, Germania 43 (1965), pp. 3279.
 115

The interpretation of the archaeological evidence has been fraught

with difficulties especially in light of the problems of dating the re-
use of older building materials in later stuctures. As has been rightly
argued by Young and Duval, Mles view of the re-use of ancient
sites as characteristic of superficial Christian conversion is untenable
due to the infrequency of archaeological evidence of the transfor-
mation of pagan temples into churches. The majority, but far from
all, of the remains that led Young and Duval to this conclusion dated
from the sixth and seventh centuries.80 Centuries after the aban-
donment of these locations and artifacts by their original occupants,
however, ruined sites would have likely been recognized as ancient
but not necessarily associated with having ever been pagan.81 Although
in many circumstances there was no continuity of custom or mem-
ory that might explain their attraction, the referential powers of
ancient ruins and the convenience of a partially intact stone struc-
ture must have provided sufficient incentives for building there if the
population perceived their symbolic potential.
From prolific archaeological evidence pointing toward re-use of
ancient remains and the apparent lack of opposition to their repos-
session according to the written sources in Gaul, one may safely pro-
pose that clerical authorities did not challenge the transition from
ancient ruin to Christian structure as unorthodox.82 While it is pos-
sible that some priests may have taken precautions to assure them-
selves that structures were cleansed of any possible demonic presence
linked to their unknown function in previous generations, no references
survive to formal liturgical rituals for the re-use of long-abandoned
edifices in the canons of the Gallic ecclesiastical councils or the
early medieval sacramentaries. Ecclesiastical texts only addressed the

80
Young, Que restait-il, pp. 24150. Young, Sacred topography, pp. 2357.
Duval, Larchitecture culturelle, pp. 18790.
81
M. Richter, The formation of the medieval West: Studies in the oral culture of the bar-
barians (Dublin, 1994) 3542. J. Le Goff, Time, work, and culture in the Middle Ages,
transl. Arthur Goldhammer (Chicago: 1980), pp. 1568.
82
Mortuary customs amongst the mixed Merovingian population retained remark-
able continuity through the time of the conversions; clerics did not interfere with
burial rites to any significant degree prior to the late Merovingian period. O.G.
Oexle, Die Gegenwart der Toten, in: H. Braet and W. Verbeke eds., Death in
the Middle Ages. Mediaevalia Louvaniensia Series 1, Studia 9 (Louvain, 1983), pp.
4954.
116

problem of reconsecrating the structures confiscated from contem-

porary religious opponents. These included prayers for the conse-
cration of formerly heretical churches seized from Arian opponents
in the tenth canon of the Council of Orlans in 511.83 The Gelasian
Sacramentary likewise contained liturgy for the rededication of syn-
agogues as churches.84
As mentioned earlier, pope Gregory Is letter to Mellitus suggested
similar approaches to re-use of religious structures in Anglo-Saxon
England. There, the seventh-century penitential attributed to Theodore
instructed that cadavers of the pagans were to be ejected from
churches before the altars might be blessed.85 This measure, not
found in Gaul, reflected both the contemporary conversion of the
inhabitants of Anglo-Saxon England and the influence of Gregory I
who was far removed from such sites of contact. Although the so-
called penitential of Theodore demonstrated the orthodox practice
of appropriating edifices which had formerly been used for pagan
worship, it likely referred to those occupied in recent memory rather
than ancient ruins.
Rather than interpreting regular encounters between early medieval
Christians and ancient artifacts as occurring in a climate or land-
scape of fear,86 interactions between early medieval Christians and
ruins were commonplace. After the fall of the imperial administra-

83
. . . et ecclesias simili, quo nostrae innovare solent, placuit ordine consecrari., Concilia
Galliae A.511A.695, ed. C. De Clercq, CCSL 148a (Turnhout, 1963), pp. 78. The
thirty-third canon of the Council of Epao in 517 contradicted this legislation by
forbidding reconsecration to take place: Basilicas hereticorum, quas tanta execrationem
habemus exosas, ut pollutionem earum purgabilem non putemus, sanctis usibus adplicare dispicimus.
Sane quas per violentiam nostris tulerant, possumus revocare. Concilia Galliae A.511A.695,
p. 33.
84
Orationes et preces in dedicacione loci illius ubi prius fuit sinagoga., Liber Sacramentorum
Romanae Aecclesiae Ordinis Anni Circuli (Codex Vat. Reg. Lat. 316/Paris Bibl. Nat. 7193,
41/56) (Sacramentarium Gelasianum) 93.724729, eds. L.C. Mohlberg, L. Eizenhfer
and P. Siffrin, third edition, Rerum ecclesiasticarum Documenta, Series Maior,
Fontes 4 (Rome, 1981), p. 114.
85
4. In ecclesia in qua mortuorum cadavera infidelium sepeliuntur sanctificare altare non licet
sed si apta videtur ad consecrandum inde evulsa et rasis vel lotis lignis eius reaedificetur.; 5. Si
autem consecratum prius fuit missas in eo caelebrare licet si relegiosi ibi sepulti sunt., Die Canones
Theodori Cantuariensis und ihre berlieferungsformen U 2.1.45, ed. P.W. Finsterwalder
(Weimar, 1929), p. 312. For a discussion of canon 4, see: D.A. Bullough, Burial,
community and belief in the early medieval West, in: P. Wormald ed., Ideal and
reality in Frankish and Anglo-Saxon society: Studies presented to J.M. Wallace-Hadrill (Oxford,
1983), p. 189.
86
V. Fumagalli, Landscapes of fear: Perceptions of nature and the city in the Middle Ages,
ed. S. Mitchell (Cambridge, 1994), pp. 69; pp. 725. S. Semple, A fear of the
past: The place of the prehistoric burial mound in the ideology of middle and later
 117

tion, they were never actively challenged in the West as inappro-

priate except when they resembled grave robbery. Various leaders
of the Merovingian clergy acted creatively in adapting abandoned
sites, especially ones that had essentially been left untouched for cen-
turies. In making these choices, they underlined divine sanction of
such forms of re-use to promote various ideological objectives, since
the value of these artifacts was greater than the resources and labor
that went into their excavation. The public acquisition of ancient
objects, just as precious relics, enhanced the status of those who
claimed them through rituals of possession.87 This process mirrored
the literary flexibility described by Michael Hunter with respect to
the Anglo-Saxon written representation of the past; early medieval
authors did not blanch at weaving together Roman, Germanic, bibli-
cal and oral sources indiscriminately when describing their heritage.88
Rather than pointing to sharp demarcations between clerical and
so-called popular culture, the re-use of ancient remains by early
medieval clerics demonstrated that they shared the widespread attrac-
tion to ruins in Gaul.89 Indeed, archaeological evidence has attested
to that reoccupation far more frequently than written sources sup-
port this premise, although the material remains do not reveal how
the general population perceived ancient remains. Whereas one of
the most conservative of bishops, Caesarius of Arles, condemned
recurring contact with pagan sites such as fountains and sacred groves
for ritual purposes as signs of the rusticitas of the population,90 even

Anglo-Saxon England, World Archaeology 30 (1998), pp. 10926. Brown, The rise of
Western Christendom, pp. 978.
87
Pomian, Collectors and curiosities, pp. 744.
88
Hunter also refers to the imperfect awareness of the difference between the
past, however alien, and the present a trait common throughout medieval Europe,
in his discussion of anachronistic depictions on the Franks casket. Hunter, Germanic
and Roman antiquity, pp. 4657. J. Barlow, Gregory of Tours and the myth of
the Trojan origins of the Franks, Frhmittelalterliche Studien 29 (1995), pp. 8695.
89
J.-C. Schmitt, Religion populaire et culture folklorique, Annales ESC 31
(1976), pp. 9418. Neuman de Vegvar has noted that the reuse of Roman mate-
rials in Anglo-Saxon England functioned as a discourse of social reproduction which
changed in vocabulary but not in strategy with the conversions. Neuman de Vegvar,
The value of recycling.
90
Although Robert Markus points to the ways in which Caesarius attempted to
steer Christian custom in Gaul back into line with the rest of the Western churches,
he believes that the anti-pagan legislation in Merovingian councils has received too
much attention with respect to the continued existence of paganism in Gaul. He
interprets the problem as one of lax standards rather than idolatrous practices among
lay persons. R.A. Markus, From Caesarius to Boniface: Christianity and paganism
118

he chose not to include the reoccupation of ancient ruins among

taboos of pagan or pagan-like practice. Like the majority of authors
such as Gregory of Tours and later the Venerable Bede, he deemed
the timely appearance of ancient remains as a visible sign of Gods
participation in the larger scheme of universal history.91
This outlook reflected the consequence of a long evolution of atti-
tudes toward ancient remains. It contrasted greatly with late fourth-
century perceptions of re-use, a time at which no framework outside
the imperial court existed for the positive interpretation of these
activities. Imperial officials viewed the excavation of ruins, just as
holy bodies, as subversive, intended for the overthrow of the exist-
ing regime. Only with the growing importance of the cult of saints
could men such as Ambrose of Milan and Paulinus of Nola, who
sought to become representatives for the precious relics of the mar-
tyrs, challenge these laws successfully.92 The powerful bishops thereby
rendered digging among ruins more acceptable. During the sixth
and seventh centuries, monastic leaders in the countryside also began
to take up the challenge of appropriating ruins in the landscape as
ideal bases for their foundations.
In early medieval Gaul, the memory of the past evolved contin-
uously in accord with changing visions of the model Christian com-
munity. This ideology of the past reflected the material conditions
of life in that society, including ancient ruins, and subjected them
to a certain degree of adaptation. Ancient ruins thereby represented
a malleable ordering principle for the reorientation of the cultural
landscape. As circumstances changed from generation to generation,
clerical and lay elites successfully found ways in which to channel
and direct such energies. Through miracle collections and church
foundations on ancient sites, these elites thereby exercised a significant
monopoly over the memory of Antiquity. The material evidence of
the past thus continued to play a crucial role in early medieval cul-
ture, gaining new and powerful meanings reflective of the ever-
changing needs of Christian communities.

in Gaul, in: Le septime sicle: Changements et continuits (London, 1992), pp. 15468.
For a very different view of the survival of ancient rituals, especially those related
to healing, see: Rousselle, Croire et gurir, pp. 6174; pp. 185200.
91
W.D. McCready, Miracles and the Venerable Bede. Studies and Texts 118 (Toronto,
1994), pp. 1106. W. Goffart, The narrators of barbarian history (A.D. 550800): Jordanes,
Gregory of Tours, Bede, and Paul the Deacon (Princeton, 1988), pp. 2502.
92
Brown, The cult of saints, pp. 5360.
 CORDOBA IN THE VITA VEL PASSIO ARGENTEAE

Ann Christys

The history of al-Andalus to the fall of the caliphate is to a large

extent the history of Cordoba, and of that citys greatest glory.
Cordoba was first mentioned at the time of the second Punic War.
Captured by the Romans in 152 B.C., it became the capital of the
whole of Hispania Ulterior, and later of the province of Baetica. In
571 Leovigild took Cordoba from the Byzantines and established a
bishopric there, but Visigothic Cordoba was overshadowed by the
capital, Toledo. The Arab and Berber conquerors of Spain seem to
have made Cordoba their capital during the period of brief gover-
norships which followed the conquest, perhaps as early as 717,
although the Arabic sources disagree about which governor took this
initiative. The settlement of Cordoba rather than the Visigothic cap-
ital Toledo was an indication that the invaders were not confident
of their ability to control the whole of the former Visigothic realm;
Cordoba and Toledo were often in conflict. Several other cities in
al-Andalus, such as Seville, had similar Roman and Visigothic his-
tories, and must have continued to flourish after 711, because they
became the centres of the taifa kingdoms which emerged from the
break-up of al-Andalus in the eleventh century. Yet they are rarely
mentioned in the Arabic narrative histories, which are propaganda
for the Umayyad dynasty who ruled al-Andalus after 756. According
to these sources, Cordoba was without equal in the peninsula.
Although the Arabic historians recorded Umayyad campaigns
against the Christian north and against dissidents within Muslim
Spain, they focussed on life in the capital, on royal appointments
and building programmes. Cordoba reached the height of its splen-
dour in the middle of the tenth century. In 929, Abd al-Ra m n
III (912961) felt sufficiently confident of his position to cast off the
nominal allegiance which previous Umayyad emirs had offered to
the Abb sids in Baghdad. He ordered a proclamation to be made
in the great mosque in Cordoba that he should henceforward be
given the title of caliph, thus asserting spiritual as well as temporal
120

power over his subjects. Some ten years later he began to build his
palace of Mad nat al-Zahr outside the city.
At about this time, a young virgin named Argentea made her way
to Cordoba in search of a martyrs crown.1 If the sources accurately
reflect Cordobas importance as the centre of Islamic power, it is
easy to see how the city might have become the place of martyrdom
for Christians, set in the midst of a crooked and perverse nation,2
as Eulogius, the hagiographer of the ninth-century martyrs of Cordoba
had put it. In this paper I want to examine this idea, to investigate
tenth-century Cordoba as a place of power both in reality and as it
appears in the Vita Argenteae.
Remembering Cordoba of the caliphate, the Arabic histories and
the accounts of travellers to al-Andalus portrayed it as one of the
wonders of the world.3 Ibn awqal, who visited Spain in 948, wrote:
There is nothing to equal it in the whole of the Maghreb [North
Africa and al-Andalus], or even in Upper Mesopotamia, Syria or Egypt,
for the number of its inhabitants, its extent, the vast area taken up by
markets, its cleanliness, the architecture of the mosques or the great
number of baths and caravanserais. Several travellers from this city
who have visited Baghdad say that it is the size of one of the quar-
ters of that city. . . . Cordoba is not perhaps equal to half the size of
Baghdad, but is not far off being so. It is a city with a stone wall,
with handsome districts and vast squares. The ruler of this city has
reigned over it for many years, and his palace is within the city walls.4
The rulers of al-Andalus appear to have concentrated their building
programmes almost entirely on Cordoba and its immediate environs.
Above all, the rulers were celebrated by their eulogists for their
involvement in the construction of the mosque, which for modern
visitors epitomises the splendour of Umayyad Spain. One of the
longest accounts of the construction of the mosque comes from the

1
Vita Argenteae; Pasionario Hispnico, ed. A. Fabrega Grau (2 vols., Barcelona/Madrid,
19535) II, pp. 3827.
2
Eulogius, Memoriale sanctorum I, p. 30, ed. J. Gil, Corpus Scriptorum Muzarabicorum
(2 vols., Madrid, 1973), II, p. 392.
3
See e.g. the collection of traditions made by al-Maqqar in the seventeenth cen-
tury: Analectes sur lhistoire et la littrature des arabes dEspagne, ed. R. Dozy, 2 vols.
(Leiden/London, 185561) I, pp. 297362.
4
Ibn awqal, rat al-Ar , ed. J.H. Kramers (Leiden, 1939) transl. M.J. Romani
Suay, Configuracin del Mundo ( fragmentos alusivos al Magreb y Espaa), Textos Medievales
XXVI (Valencia, 1971), pp. 634.
 VITA VEL PASSIO ARGENTEAE 121

work of Ibn Idh r .5 He was writing in the fourteenth century but

claimed to be quoting from earlier authorities, particularly the tenth-
century historian al-R z , whose work does not survive. Ibn Idh r s
history of al-Andalus is often quoted, probably because of the seduc-
tive amount of detail he included, which is sometimes far more than
seems to have existed in the sources from whom he claimed to have
been quoting. Ibn Idh r said that the mosque was founded in
7856 by Abd al-Ra m n I. Abd Ra m n II added eight bays in
836. His successors carried out further improvements to the mosque
and its courtyard.6 Muhammad was said to have added a maqs ra,
the special enclosure reserved for the ruler, and restored the west
door. Al-Mundir built a treasury and Abd Ra m n III the first
minaret. The only patron whose contribution can be identified with
certainty is al-Hakam II, responsible for a further extension to the
mosque which is commemorated in an inscription around the qibla,
the niche indicating the direction of Mecca. Ibn Idh r s descriptions
of the mosque have not, so far as I am aware, been compared with
other accounts of the building. Such a study has been carried out
for the various descriptions of the palace of Mad nat al-Zahr , and
shows that they do not concur, and that they became more detailed
with each retelling. The actual building which remains is undoubt-
edly impressive, and it seems that each Umayyad ruler had to be
associated with it, in the work of the eulogists, if not in reality.
The ceremonial function of the mosque was closely linked with
that of the palace, to which it was joined by a passageway. In the
mosque, messages from the ruler were read out, announcing victo-
ries, denouncing the unorthodox, praying for rain or good fortune.
When a ruler died, the notables of al-Andalus, and it seems, a con-
siderable number of the inhabitants of the city, went to the mosque
to swear allegiance to his successor. A late copy of court annals
attributed to al-R zis son Isa described the oath of allegiance to
Abd Ra m n III, naming the important people who were there and
adding: This ceremony took all week.7 It was during Friday prayers

5
Ibn Idh r , Al-Bay n al-mughrib f akhb r al-Andalus wa-l-Maghrib, vols. III eds.
E. Lvi-Provenal and G.S. Colin (Paris, 1930), II, pp. 244 seq.
6
J.D. Dodds ed., Al-Andalus: the art of Islamic Spain (New York, 1992), pp. 1125
and 1634.
7
Una crnica annima de Abd al-Ra m n al-Nas r, transl. E. Garca Gmez and
E. Lvi-Provenal (Madrid, 1959), p. 93; M. Barcel, El califa patente: el ceremonial
122

in the mosque that the citizens of Cordoba would have heard for
the first time that they were no longer offering allegiance to Baghdad.
The size of the building was related to its function, not in the sim-
plistic way often assumed that the area of the mosque divided by
that of a prayer mat equals the male population of Cordoba but
as a reflection of its importance as a ceremonial centre. The palace
served as the main venue for the reception of embassies, (although
the mosque also doubled in this function), and it was here that feasts
were celebrated. Abd Ra m n III and his son were both interred
in the palace cemetery. This pattern, of a mosque joined to a palace,
was repeated in the building at Mad nat al-Zahr .8 The splendour
of court ceremonial at Mad nat al-Zahr was described in the Life
of John of Gorze, who went to Cordoba as an ambassador in the
950s.9 It was clearly in these two spaces, mosque and palace, that
the power of the caliphate was embodied.
It is not known how much of Cordoba, in addition to the mosque
and palace, the Umayyads developed as the concrete expression of
their dominance, nor, indeed, how big the city was. The Arabic
sources are deceptively precise about size, but frustratingly vague
about topography. Cordoba has been estimated to have as many as
a million inhabitants, and is depicted in a modern historical atlas as
being many times bigger than contemporary cities in the rest of
Europe.10 This is the result of taking literally Ibn awqals state-
ment that Cordoba was nearly half the size of Baghdad, which begs
another question; the figure of four million inhabitants often quoted
for Baghdad seems to be derived from the Arabic sources hyper-
bole. A similar process was at work in the histories of al-Andalus.
Ibn al-Khat b, in the fourteenth century, said that tenth-century
Cordoba was surrounded by a ditch and a wall of some twenty-two
kilometres long. This implies a city of approximately 5,000 hectares,
about eight times the area of the present city, although it does not
of course mean that the whole area was inhabited. The loss of
Cordobas splendour Ibn Idh r claimed that there were three

omeya de Cordoba o la escenificacin del poder, in: R. Pastor, I Kieniewicz,

E. Garcia de Enterria e.a. eds., Estructuras y formas del poder en la historia (Salamanca,
1991), pp. 5171.
8
C. Mazzoli-Guintard, Remarques sur le fonctionnement dune capitale a dou-
ble polarite: Mad nat al-Zahr -Cordue, Al-Qantara 17 (1997), pp. 4364.
9
Vita Johannis Gorzensis, MGH SS 4, pp. 33577.
10
Times Atlas of World History (3rd edn., London, 1989), pp. 1089.
 VITA VEL PASSIO ARGENTEAE 123

hundred baths and three thousand mosques,11 but only three minarets
remain is blamed on the civil wars which followed the collapse
of the caliphate. A recent article refers to the enigma of the lost
Cordoba of the caliphate. Cordoba seems to have melted away like
the cities of fable.12
Archaeologists have tried very hard to find the city of the writ-
ten sources, but succeeding generations are gradually, and with great
regret, reducing Cordoba to more modest proportions. In the nine-
teenth century, the remains of Roman aqueducts well outside the
city were thought to be the Umayyad walls, rather as many Roman
structures in Spain have been reclassified in folk memory as Moorish.
Ibn awqal claimed to have walked round the city walls in an hour,
describing a city of about the size of medieval York, and this is
probably much nearer the mark. Traces of a wall running alongside
the waterway now known as the Arroyo del Moro may date to the
ninth and tenth centuries,13 although their construction is very sim-
ilar to walls built several centuries later after the Christian Reconquest.
If these walls marked the outer limit of the city in the caliphal period,
they enclose an area of only some seven or eight hectares to the
west of the palace. The palace and mosque, together with the mar-
kets established along the Roman Cardo Maximus, occupied some 2.5
hectares. Thus the Umayyad capital seems to have been based on
the Roman city, although it was a little larger because it was extended
towards the river.14
Two arches in the western wall may mark the place where it was
crossed by an aqueduct. Abd al-Ra m n II was said to have brought
water to the outskirts of the city sometime after 756, Abd al-Ra m n
III to have extended the aqueduct two centuries later, and his suc-
cessor al-Hakam II to have brought water to the mosque. This is plau-
sible, but conflicts with statements by other authors. The elevated
water channel supplying the palace which Alvarus mentioned in the
Life of Eulogius,15 could be the one commissioned by Abd Ra m n

11
Ibn Idh r , Al-Bay n al-mughrib II, p. 247.
12
B. Pavn Maldonado, Entre la historia y la arqueologa. El enigma de la
Crdoba califal desaparecida, Al-Qantara 9 (1988), pp. 16998 and 40326.
13
Pavon, Entre la historia, p. 189.
14
P. Scales, Cordoba under the Umayyads: a Syrian garden city?, in: G. de
Roe and F. Verhaege eds., Urbanism in medieval Europe (Zellik, 1997), I, pp. 17582.
15
Alvarus, Vita Eulogii, c. 15, ed. Gil, CSM I, p. 341.
124

II,16 although al-Maqqar recorded that, during al-Mughiths attacks

on Cordoba in 711, 400 men took refuge in a church, and were
able to hold out for three months because they had fresh water from
a pipe which ran from springs situated in the foothills of the Sierra.17
One should not pick and choose between those sources which are
just credible and those that clearly are not. The Umayyads proba-
bly repaired Roman aqueducts, and perhaps built new ones, just as
the popes were doing in Rome at the same period, but since these
aqueducts do not survive, it is impossible to say whether the Arabic
sources are describing Roman structures, Muslim repair or com-
pletely new works. Just as in Ibn Idh r s description of the mosque,
so Cordobas other public building works became associated with
the rulers of Cordoba, starting from one of the first governors, al-
Sam (718721), who was said to have restored the Roman walls
and the bridge over the Guadalquivir.18
In order to vindicate Cordoba as a great metropolis it is neces-
sary to include what are always referred to as its suburbs. Ibn
Bashkuw l, in a passage quoted by al-Maqqar , listed twenty-one
such suburbs,19 and some of them are mentioned in the Calendar of
Cordoba. The recent archaeology of Cordoba has been marred by
rapid redevelopment; little time was allowed for excavation and large
areas of important Roman and Islamic remains have been destroyed.20
The full extent of Umayyad Cordoba will probably never be known.
Traces of street patterns, however, have been identified outside the
walled city, although only one of these, al-Rus fa, can be identified
with a suburb named by Ibn Bashkuw l. The rest of the citys hin-
terland was a rural landscape of gardens and orchards, dotted with
the palaces of the nobility. Indeed, one of the suburbs, Secunda, on
the other side of the Guadalquivir from the great mosque, which
later became known by antonomasia as el Arrabal, is elsewhere
referred to as a village.21 Ibn awqals picture of Cordoba should
be quoted in context, when it is clear that he too was describing a
compact city surrounded by larger area mainly given over to agri-
culture, where the nobility had their country palaces:

16
Ibn Idh r , Al-Bay n al-mughrib II, p. 93.
17
Al-Maqqari, Analectes II, pp. 68.
18
Ibn Idh r , Al-Bay n al-mughrib II, p. 25.
19
Al-Maqqari, Analectes I, pp. 3023.
20
Scales, Cordoba under the Umayyads, pp. 1778.
21
Ibn ayy n, Al-Muqtabis f bal d al-Andalus (al-Hakam II) (Al-Muqtabis VII) ed.
 VITA VEL PASSIO ARGENTEAE 125

The ruler of this city, Abd Ra m n ibn Mu ammad, founded, to the

west of Cordoba, a city which he called Zahr , on the flank of a
rocky mountain with flat summit, called Batlash; he brought markets
there and had baths, caravanserais, palaces and parks built. He invited
the people to live there and ordered that the following proclamation
should be issued throughout Spain: Whoever wished to built a house,
choosing a spot next to the sovereign, will receive 400 dirhams. A
flood of people rushed to build; the buildings crowded together and
the popularity of this city was such that the houses formed a contin-
uous line between Cordoba and Zahr .22
Even if it is taken at face value, this description of ribbon develop-
ment does not imply that the area between Cordoba and Mad nat
al-Zahr was urbanised in the commonly accepted sense. Ibn awqal
went on to describe several cities in al-Andalus, many of them, like
Cordoba, well-populated and having walls, but all were surrounded
by a vast rural area where the wealth of the city was cultivated.23
Discussing Damascus in the later middle Ages, Lapidus argued that
large villages in the agricultural hinterland of a city might be con-
sidered part of the metropolitan conglomeration.24 It is not clear that
this argument can be applied to Cordoba in the tenth century.
Although the mosque may have been grander than anything in al-
Andalus, the city itself was almost certainly little larger than several
other former Roman cities in Spain. Mrida, for example, with its
spectacular Roman buildings, many of which still survive, and the
Alcazaba by the river, must have been equally impressive. Yet Mrida
and other cities were rarely mentioned by the Arabic authors, and
their history in the early Islamic period remains obscure. Cordoba
attracted ambassadors and merchants and other visitors hopeful of
making their fortune not because of its size, nor the splendour of
its buildings, but because it was the seat of power. This may also
have continued to make the city a magnet for Christians aspiring to
martyrdom, who sought their fortune in heaven.
The Vita Argenteae has nothing to say about the size of Cordoba,
nor about its buildings. Argenteas yearning for martyrdom in the

A.A. Al-Hajji (Beirut, 1965); E. Garca Gomez, Anales Palatinos del Califa de Cordoba
al-Hakam II (Madrid, 1967), p. 211; ibid., Notas sobre la topografa cordobesa en
los Anales de Al-Hakam II por Isa R z , Al-Andalus 30 (1965), pp. 31979 at
p. 352.
22
Ibn awqal, Configuracin del Mundo, p. 64.
23
Ibn awqal, Configuracin del Mundo, p. 69.
24
I.M. Lapidus, Muslim cities in the later Middle Ages (Harvard, 1967), p. 79.
126

capital can be understood only from a close reading of the text.

Argentea was, so the Vita tells us, the daughter of a king named
Samuel and a queen named Columba of the city of Bibistrense, who
rejected the trappings of royalty and dedicated herself to chastity.
When her mother died, she refused her fathers request to carry out
the duties of a royal consort, and immured herself in a secure cham-
ber below the palace enclosure. Even this privation failed to satisfy
her aspirations and, hearing of a saint who was looking for martyr-
dom, she wrote to him asking for his advice about following this
way to perfection. He advised patience, until the right opportunity
presented itself. This seemed to come with the overthrow of the city,
when Argentea and several of her fellow-citizens made their way to
Cordoba. Argentea spent several years in Cordoba, apparently prac-
tising her vocation of chastity undisturbed. Then, the Vita records,
a Frank named Vulfura was summoned in a dream to come to
Cordoba to help her achieve her goal of martyrdom. Vulfura was
immediately arrested and thrown into prison, where Argentea vis-
ited him assiduously. During one of these visits, she was recognised
as Samuels daughter: she found herself surrounded on all sides by
pagans, and she heard, having been asked harmful questions: Are
you not, O woman, the daughter of Samuel their prince?. . . .
Therefore the blessed Argentea, wishing to be a participant in the
hoped-for passion, rejoiced, intrepidly declared that she was not only
the daughter of the aforementioned father, but truly a guardian of
the catholic faith.25 Both Argentea and Vulfura were brought before
an unnamed judge, questioned, tortured, and put to death. Argentea
was buried in Cordoba, but the fate of Vulfuras relics was unknown
even to the hagiographer.
The Vita Argenteae is usually dismissed in a few words as though
it were a mere footnote to the Cordoban martyr movement of the
850s, and Argenteas desire for a Cordoban death is seen as a
response to these events. There is, however, no direct evidence that
the hagiographer knew of the ninth-century martyrs, whose cults
were neglected in Spain. The Vita Argenteae is not modelled on Eulogius
accounts of the earlier martyrs; in particular, there are many more
circumstantial details. These details have been drawn together into
a narrative which places Argentea not only in the geographical heart

25
Vita Argenteae, 12, p. 386.
 VITA VEL PASSIO ARGENTEAE 127

of power, but in the centre of the Christian-Muslim conflict of the

tenth century. As we have seen, Argentea left her native city follow-
ing its overthrow. The nineteenth-century Dutch Arabist Dozy, an
indefatigable maker of connections between bits and pieces of evi-
dence, identified this city, which the Vita names as Bibistrense, with
a place which the Arabic sources called Bobastro.26 I cannot cor-
roborate this assumption, whose main function is to tie Argentea
in with Ibn Haf n, who led a long-running rebellion against Cordoba.
Bobastro, which is given many different but related names in the
Arabic sources, was his stronghold. He may have apostatized from
Islam to the Christian faith of his ancestors as part of his protest,
although this is not certain, as we shall see. Simonet, the author of
an influential History of the Mozarabs, stretched this association, say-
ing that Argentea was carried to Cordoba with her brother Haf
and the other citizens of Bobastro, as though they were being taken
into some Bobylonian captivity. (In fact, the passion says that cum
fratribus ceterisque concivibus Cordobensum urbem petivit which
implies that she went willingly to Cordoba, accompanied by some
of her fellow-citizens, among them perhaps monks rather than her
biological brothers.) Add to this mixture an extra-textual reference
to the Cordoban martyrs of the 850s and the result is a neat rep-
resentation of religious conflict in Spain, coming to a climax with
Argenteas death at the heart of Islamic power.
The linking of Argentea and Ibn Haf n may have deterred those
historians who have written so copiously on the ninth-century mar-
tyrs from examining the Vita Argenteae by placing it in the wrong his-
toriographical camp, linked with the Arabic sources which Latinists
consider, perhaps wrongly, to be even more impenetrable than their
own. Ibn Haf n illustrates the problems of the Arabic sources only
too well. He is well documented, but in directly contradictory accounts;
all may be late, since their manuscripts certainly are, and it is almost
impossible to arbitrate between them. By the twelfth-century, Rodrigo
Ximenez de Rada27 was already arguing about Ibn Haf ns significance
for the history of the Reconquest.28 According to Ibn Askars (1188/

26
F. Simonet, Historia de los Mozrabes de Espaa (Madrid, 1897; repr. 1984),
p. 596 n. 2.
27
R. Ximnez de Rada, Historia Arabum, in: Opera (Valencia, 1968), p. 268.
28
M. Acin Almansa, Entre el feudalismo y el Islam. Umar ibn Hafsun en los histori-
adores, en las fuentes y en la historia ( Jan, 1994).
128

91239) History of Malaga,29 he was the descendant of a prominent

Visigoth named Marcellus, but his grandfather converted to Islam.
Most of the Arabic sources relate his campaigns against Cordoba
and his alliances with neighbouring rebels, but with markedly different
emphases. It appears that Ibn Haf n alternated periods of opposi-
tion to the Umayyads with service to the regime in the army and
as a governor. His career points to a weakness of central control of
al-Andalus which otherwise emerges only rarely from the works of
Umayyad panegyrists.30 There are several similarities between the
career of Ibn Haf n and his treatment by later historians, and that
of a more famous medieval Spanish adventurer, the Cid.31 Only from
the nineteenth century has Ibn Haf n been portrayed as a focus of
the nationalist aspirations of indigenous Spaniards. The question of
his Christianity is another instance of modern historians selecting
their sources according to their own concerns.32 Ibn al-Q i a, who
was himself of Christian descent, and may be a contemporary wit-
ness (although the one surviving manuscript of his History dates from
the fourteenth century), did not mention it. He said that Ibn Haf n
died at the beginning of the reign of Abd Ra m n ibn Mu ammad
after having made friends with him and pledged his allegiance.33
The earliest account of Ibn Haf ns apostasy to Christianity appears
in a text now called the Anonymous Chronicle of Abd Ra m n al-Na ir,34
which may date from the tenth or eleventh century, although it sur-
vives in a copy of perhaps the fourteenth century. Again, it is Ibn
Idh r , who described the building of the Cordoban mosque in such
detail, who claimed to know the exact date of Ibn Haf ns apostasy.35

29
J. Vallv Bermejo, Una fuente importante de la historia de Al-Andalus. La
Historia de Ibn Askar, Al-Andalus 31 (1966), pp. 23765.
30
R. Marn Guzmn, The causes of the revolt of Umar ibn Hafsun in Al-
Andalus 880928: A study in medieval Islamic social history, Arabica 17 (1995),
pp. 180221.
31
R. Fletcher, The quest for El Cid (Oxford, 1989).
32
See e.g. A. Cutler, The ninth-century Spanish martyrs movement and the
origins of western Christian missions to the Muslims, The Muslim World 15 (1965),
pp. 32139, at p. 333: In the famous Umar ibn Haf un the dreams of the mar-
tyrs movement were perfectly realised.
33
Ibn al-Q t a, ed. and transl. J. Ribera, Historia de la Conquista de Espaa por
Abenalcotia el Cordobes (Madrid, 1926), transl. J.M. Nichols, Ibn al-Qutiyya, The history
of the conquest of Al-Andalus, unpubl. Ph.D. Diss. (University of North Carolina, 1975).
34
E. Garca Gmez and E. Lvi-Provenal eds. and transl., Una crnica annima
de Abd al-Ra m III al-Nas r (Madrid, 1950).
35
Ibn ayy n, Al-Muqtabis, transl. J. Guraieb, Cuadernos de Historia de Espaa 27
(1958), p. 172.
 VITA VEL PASSIO ARGENTEAE 129

Both Ar b ibn Sa d, quoted by Ibn Idh r , and the author of the

Anonymous Chronicle disparaged Ibn Haf n as the refuge of the unbe-
lievers and hypocrites,36 without specifying his apostasy. Archaeologists
claim to have found the church, said to have been built by his father,
where Ibn Haf n was buried, at Las Mesas de Villaverde, Ardales,
near Ronda,37 but since the identification of this site as Bobastro is
still disputed,38 this is not very likely; besides, the church has been
dated only as early medieval, and could be Visigothic. It is possi-
ble that the accusation in the Anonymous Chronicle of Abd Ra m n al-
Na ir that Ibn Haf n was an apostate was a damnatio memoriae,
especially as it is elaborated with the story that Ibn Haf ns apos-
tasy had been secret, and was revealed only when his body was
exhumed and found to be buried in the Christian manner, and that
Abd Ra m n III crucified it on the walls of Cordoba between those
of two of his sons.
To return to more immediate concerns, none of the sources say
that Ibn Haf n had a daughter, and the dates for his career are
difficult to square with those given by the Vita Argenteae. Argentea
was said to have left Bibistrense after its overthrow. Bobastro faced
a major Cordoban attack c. 923/4 and was finally overthrown c. 928,39
when Haf ibn Umar, one of several sons of Ibn Haf n, was in
control, Umar ibn Haf n himself having died c. 917.40 The Anonymous
Chronicle says that Haf ibn Umar ibn Haf n was brought before
the emir with his family and hostages, who confirmed the peace
treaty and gave them good welcome. With this good fortune, God,
with his power, put an end to the rule of the evil family of Haf n.
It is just plausible, but rather unlikely, that Argentea saw this as the
right time to be martyred. If, however, she had already left Bobastro
during one of her fathers periodic defeats, what the Vita described
as the turning cycle of her years in Cordoba must have seemed to
stretch interminably if she had to wait for martyrdom until 931
the date given in the Vita. It is impossible to mine the Vita Argenteae

36
am d al-K fir n wa-r s al-mun fiq n; Garca, Una crnica annima, p. 119.
37
R. Puertas Tricas, La iglesia rupestre de Las Mesas de Villaverde, (Ardales,
Malaga), Mainake 1 (1979), pp. 179216.
38
J. Vallv Bermejo, De nuevo sobre Bobastro, Al-Andalus 30 (1965), pp. 13974;
Acien, Entre el feudalismo y el Islam, p. 22.
39
Garca, Una crnica annima, pp. 14041, 146 and 148.
40
303/91516 according to Garca, Una crnica annima, p. 119; Ibn Idh r Al-
Bay n al-mughrib, p. 171 has 305/91718.
130

for facts in any convincing way. The whole confection that Argentea
was the daughter of a proto-nationalist Catholic rebel must be rejected,
leaving us with something much more complicated.
The Vita Argenteae is primarily a treatise on the two principal Chris-
tian virtues, described in this text as a twofold handful of flowers
the white of chastity and the purple of martyrdom and is divided
almost equally between the two aspects. Apart from the date given
for her martyrdom, the historical details are vague, and the story is
modelled on the anonymous hagiographical romances known as the
gesta martyrum41 rather than on contemporary debate with Islam.
Argenteas journey to Cordoba recalls the martyrdom of Agap, Irene
and Chion at Salonika, who when the persecution was raging under
the Emperor, . . . . abandoned their native city, their family, prop-
erty and possessions because of their love of God and their expec-
tation of heavenly things.42 When Argentea made a public declaration
of her faith,
the crowd excited into fury brought the follower of Christ before the
judge. Interrogated by the judge about the conditions of the faith, she
responded with constancy thus: Why do you exasperate me with your
questions? Have I not testified that I was a follower of the embraced
Christian faith? But because according to the apostolic dogma, which
believed in the heart leads to justice and confessed through the mouth
to deliverance, I confess before all: I believe in one God in three per-
sons, adored in indistinguishable substance, and declare the personal-
ity to be unconfused.43
Although adherence to ludicrously complicated beliefs about the
nature of God was one of the accusations made against Christians
by Muslim polemic, this passage seems rather to have been lifted
from a passion of a much earlier period, where the Trinitarian
emphasis seems to be addressing anti-Arian concerns. This impres-
sion is reinforced by the relatively good Latin of this text. Equally
anachronistic is the reference to an unnamed praeses, whose judge-
ment is enforced by lictors, surely out of place in tenth-century
Cordoba. Yet, if the hagiographer was merely reworking a conven-
tional passion for anti-Muslim polemic, the result is not a success;

41
K. Cooper, The martyr, the matrona and the bishop: networks of allegiance
in early sixth-century Rome, Journal of Roman Archaeology, forthcoming.
42
H. Musurillo, The acts of the Christian martyrs (Oxford, 1972), p. 281.
43
Vita Argenteae, c. 13; Fabrega Pasionario, p. 386.
 VITA VEL PASSIO ARGENTEAE 131

some of the new details work against the picture of Cordoba as a

city dominated by Muslim persecutors, and the focus for opposition
to Islam, that he or she may be trying to present. The text says of
Argenteas arrival in Cordoba, united with religious people in that
city, . . . . and firm in the usual fashion, she emptied herself busily
in continence, and thus for a long time she lived through the turn-
ing circle of the years. With how many and what kinds of virtues
she flourished in her fashion and was illustrious in pious acts, if we
tried to set them all out with our pen we would seem to set out what
was no less decorative and boring. The impression thus given that
there was a flourishing Christian community in Cordoba is the only
part of the whole story which seems to ring true. Some of the
churches built in the Visigothic period, including the basilica of the
Three Saints where Argentea was said to have been buried, survived
into the Reconquest period. Several inscriptions commemorated nuns
who died in the tenth century (although many survive only as copies
made in the sixteenth century).44 Christians established prominent
roles in government; the Arabic sources mention them briefly as
members of embassies to the Christian courts of northern Spain,
western Europe and Byzantium. The impression one gains from these
fragments is that Cordoba was a Christian as well as a Muslim city.
A pious woman from outside the capital would easily have found a
convent to receive her. Being martyred in the spectacular way out-
lined by the Vita might, however, have been more difficult.
The task of interpreting the Vita Argenteae is made more difficult
because it is not clear when and for whom the text was written.
The sole surviving copy of the Vita was appended to a copy of a
passionary from Cardea, near Burgos in northern Spain.45 The man-
uscript remained in Cardea until 1864, when it went to the British
Museum. A marginal note names the scribe as the priest Endura.46
This may be the same Endura who copied Cassiodorus Commentary
on the Psalms in 949 and Isidores Etymologies in 95447 and witnessed

44
J. Prez de Urbel, Los monjes espaoles en la edad media (Madrid, 1934), p. 266
mentions the nuns Ikilio, Justa and Rufina from the monastery of Santa Eulalia,
south of Cordoba, but does not give his sources.
45
British Museum Add. 25.6000; Fabrega, Pasionario 27seq.
46
f. 258v.
47
Madrid Academia de la Historia, Caradignense 76; Gaiffier B. de (1937) Les
notices hispaniques dans le martyrologe dUsuard, Analecta Bollandiana 15 (1937),
pp. 2712.
132

charters of donations to Cardea in 950, 966 and 969.48 Of the

manuscripts 269 folia, numbers 3259 are in the same hand. Towards
the end of the codex, after the Explicit (Here ends the first part of
the passionary), and written in another hand, is an appendix. It
consists of the Vita Argenteae, the passions of Cyriacus and Paula, two
African martyrs of the fourth century, and the Invention of Zoilus, a
Cordoban martyr from the same period. Many versions of the pas-
sion of Cyriacus and Paula (sometimes Paul) exist, but the Cardea
passion is very similar to a hymn in a tenth-century manuscript from
Toledo,49 which suggests that this version was known throughout
Spain. Zoilus cult was also well known, and continued to be cele-
brated in Cordoba.50 Thus, although the additions might be Cordoban
in origin, and the palaeography is said to be characteristic of man-
uscripts written in al-Andalus,51 these texts could have come to
Cardea from elsewhere in the peninsula. Two allusions in the text
might indicate that the author was a native of Cordoba it says that
Argentea made her way to Cordoba as a stranger (advena), of which
her body would soon in the future be an inhabitant, and that, after
her death, her miracles were constantly taking place among us up
to now. Yet neither of these phrases are conclusive, and Argentea
may not have been remembered in Cordoba. Her feast day is not
listed in the Calendar of Cordoba. In both the Arabic and Latin ver-
sions of the Calendar of Cordoba, October 13 is given as the feast of
three martyrs put to death in the city of Cordoba. The sepulchre is
in the District of the Tower. The Latin version adds: And their
festival is in the [church of ] the Three Saints.52 Although the saints
names are not given, this is assumed to be the church of Faustus,
Januarius and Martialis, and this is where Argentea was supposed
to have been interred. The church, which was the most important
in Cordoba after the Arab conquest, and later changed its name to

48
Fabrega, Pasionario, p. 28.
49
Madrid Biblioteca Nacional 1005.
50
B. de Gaiffier, Linventio et translatio de S. Zoilo de Cordoue, Analecta
Bollandiana 16 (1938), p. 369.
51
R. Guerreiro, Le rayonnement de lhagiographie hispanique en Gaule pen-
dant le haut Moyen ge: circulation et diffusion des Passions hispaniques, in:
J. Fontaine and C. Pellistrandi eds., LEurope Hritire de lEspagne Wisigothique (Madrid,
1992), p. 138.
52
R. Dozy and C. Pellat eds. and transl., Le calendrier de Cordoue (Leiden, 1961),
p. 151.
 VITA VEL PASSIO ARGENTEAE 133

St Peter, survived until after the fall of the Umayyads. Yet the mem-
ory of Argentea does not seem to have been preserved there.53 Since
there is no evidence for the cult of Argentea either in Cordoba or
in northern Spain, the single copy of her Vita is the only evidence
we have for this saint. Thus the Vita Argenteae may have been writ-
ten in the north, and it should be interpreted as relevant to a north-
ern-Spanish audience. Could it perhaps be a moral tale about the
duties of a princess, perhaps based on the existence of real martyrs
in Cordoba, but addressed to contemporary events in the north?
This part of the paper is highly speculative, but the details of the
argument may be rejected without invalidating its general conclu-
sions. I begin with the names of the participants in the drama which
are unusual, to put it no more strongly, and suggest that the pro-
tagonists of the Vita Argenteae are symbolic rather than real. Argenteas
name recalls that of a martyr of the ninth century, Aurea: silver and
gold. Her father Samuel does not seem to be have anything to do
with the biblical Samuel. His name might, however, bring to mind
one of the contemporary rulers of Len and Pamplona. The rulers
of Pamplona were called Sancho Garcs or Garca Sanches alter-
nately. Sancho the Fat ruled Len from 955957 and from 960967.
Bibistrense, Samuels royal city, may recall Barbastro, near Huesca,
rather than the elusive Bobastro of the Arabic sources. A female
saint called Columba was commemorated in the Spanish passionary,
but in this context, may have reminded the audience of the dove,
the Holy Spirit; it was an appropriate name for a protagonist whose
main function in the story was to die. Vulfura sounds like Wulfhere,
which is not a Frankish but an Anglo-Saxon name, but suggests a
foreigner.54 In this story his martyrdom was not described, and his
relics were apparently mislaid, so that he seems no more than a lit-
erary device, like the monk who told Argentea to wait before she
could fulfil her desire to be martyred. Perhaps Vulfura was meant
to be downplayed in comparison with Argentea. More probably, he
became superfluous to the story. Argentea and her family and com-
panions are improbable, but it is possible that their names helped
the audience of the Vita Argenteae to locate the story in northern Spain
around the end of the tenth century.

53
F. Prez, Cordue, in: DHEE 13 (Paris, 1956), p. 859.
54
But it recalls the scribe Vuilfurus of BN n.a.l. 239; see above.
134

It may be possible to pin down the genesis of the story more pre-
cisely, as a manifestation of the growth of female monasticism in
northern Spain in the tenth century. Scribes copied the lives of sev-
eral female saints, including that of Constantina, the daughter of the
emperor Constantine who, like Argentea, practised a life of chastity
rather than comply with the wishes of her father.55 Particularly pop-
ular with women was the cult of Pelagius, a boy of thirteen who,
in 926, preferred death rather than surrender his virginity to Abd
al-Ra m n III.56 Pelagius cult was associated with Elvira, sister of
Sancho the Fat, who sent an embassy to Cordoba to obtain the
relics of Pelagius for Len.57 Elvira seems, like Argentea, to have
been an exemplary princess. Perhaps others were not. When in 957,
Sancho the Fat lost his throne, the Arabic sources say that he was
restored only with the support of a Cordoban army, after he, his
grandmother Tota and Garca Snchez I of Navarre had made an
embassy to Cordoba, and surrendered a number of frontier fortresses
to the Muslims.58 Perhaps this deal involved sending a royal princess
to the Umayyad harem. It was not unknown for the Umayyads to
take Christian wives. A woman from Navarre called ub in the
Arabic histories, became prominent at the Cordoban court, after
bearing two sons to al- akam II.59 One of the boys died young, but
the other acceded to the throne as Hisham II at an early age and
his mother was involved in his regency, possibly becoming the mis-
tress of the chamberlain Ibn Ab Am r who later usurped the Umayyad
throne and took the title al-Man r.60 She is commemorated in the

55
J.E. Salisbury, Church fathers, independent virgins (London, 1991).
56
C. Rodrguez Fernndez ed., La Pasin de San Pelayo (Santiago de Compostela,
1991); transl. J.A. Bowman, Raguel, The martyrdom of St Pelagius, in: Medieval
Hagiography, an anthology, ed. T. Head (New York, 2000), pp.; M.C. Daz y Daz,
La Pasin de S. Pelayo y su difusin, Anuario de Estudios Medievales 6 (1969), pp.
97116.
57
M. Gmez Moreno ed., Chronicle of Sampiro, Boletn de la Real Academia de
la Historia 100 (1932), pp. 3378; J. Prez de Urbel ed., Historia Silense (Madrid,
1959), 28 ch. 171; ibid., Sampiro, su crnica y la monarqua leonesa en el siglo X (Madrid,
1952); the same story appears in Lucas of Tuy, Chronicon mundi (Frankfurt, 1610),
4, 85.
58
Ibn Idh r Al-Bay n al-mughrib, II, p. 251; al-Maqqar Analectes, I, p. 235.
59
Ibn al-Fara , Tar kh ulam al-Andalus eds. F. Codera and J. Ribera (Madrid,
18912), p. 10; E. Lvi-Provenal, Histoire de lEspagne musulmane, 3 vols. (2nd edn.
Paris, 1950), II, p. 173 and n. 4.
60
Ibn Idh r , Al-Bay n al-mughrib II 268.
 VITA VEL PASSIO ARGENTEAE 135

inscription on the base of a fountain erected on her orders in 977.61

She died in 999.62 ub was said to be a concubine who had been
captured in a campaign against Navarre and there is no suggestion
that she was of royal blood. Al-Man r, however, may have mar-
ried one of the daughters of Sancho, or possibly of one of the kings
of Navarre; the mother of one of his sons was the daughter of
Shanj h, the Christian, the king.63 Since the Arabic sources refer
only to the caliphs consorts who provided him with sons, there may
have been other Navarrese or Leonese princesses in his entourage
who achieved less prominence in Cordoba but whose defection to
the seductions of life with the infidel may be the subject to which
the Vita Argenteae is a counterblast. The hagiographer may be argu-
ing that, rather than travelling to Cordoba to consort with the enemy,
a Christian princess, should she find herself in that city, should pre-
pare herself by a life of chastity and prison-visiting to be worthy of
a martyrs crown.
Thus Cordoba, seen from the perspective of the Vita Argenteae is
not primarily a topos; it is not, as one might have expected from
modern reconstructions, the place of martyrdom, a far-away city
where Christian unease about Islam received its ultimate expression.
The Vita was read, and may have been written, in northern Spain.
Yet, although the hagiographer did not evoke Umayyad splendour,
not even by mentioning the great mosque, his (or her) picture of a
city where Christian activities were still openly practised, which must
come from the tenth or eleventh centuries, may be more realistic
than those drawn by the Arabic historians. The latter were remem-
bering Cordoba after its glories had passed, when they were free to
imagine how it might have been. It is almost certain, from what lit-
tle has been excavated, that their picture of Cordoba is overdrawn.
The author of the Vita Argenteae had more immediate concerns in
mind. The Vita may have been written in response to actual mar-
tyrdoms, although the circumstance of Argenteas death cannot be
disentangled from the model of the texts on which it was based, nor
from the purposes of its author. The significance of Argenteas death

61
E. Lvi-Provenal, Inscriptions arabes dEspagne (Leiden and Paris, 1931), pp. 378
no. 30.
62
Ibn al-Fara Tar kh, p. 152 no. 533.
63
Ibn Idh r , Al-Bay n al-mughrib, ed. E. Levi Provenal (Paris, 1930), III, p. 38,
cited by D.J. Wasserstein, The rise and fall of the party kings (Princeton, 1985), p. 48.
136

was that she died a virgin, and this virginity was perhaps the high-
est virtue at a time when virtue was threatened by the lure of life
under Islamic rule.64 Cordobas power lay in its ability to corrupt
Christian princes into alliances with the infidel, and their princesses
from the chastity and martyrdom of the ideal Christian life. For the
hagiographer, to go to Cordoba, either literally or by flirting with
Islam, was to place ones salvation in doubt.

64
A. Cameron, Virginity as metaphor in: eadem ed., History as text (London,
1989), pp. 184205 at p. 191.
 TOPOGRAPHIES OF HOLY POWER IN
SIXTH-CENTURY GAUL

Ian Wood

The works of Gregory of Tours have ensured that for the sixth cen-
tury the cult of the saints is better known in Merovingian Francia
than in any other area of the early medieval West. Peter Brown has
perhaps done more than anyone to draw attention to the Holy in
the pages of Gregory,1 although he was by no means the first to
emphasise the importance of those pages,2 and although others have
subsequently explored individual cults in greater detail than he has.
In particular Raymond Van Dam has written extensively on Gregorys
accounts of cults at Tours and Brioude,3 while Herbert Kessler has
usefully approached the evidence for pilgrimage to the shrine of
Martin at Tours from an art historians point of view.4 The net result
is that at Tours in particular, but also at Brioude, we can follow
Peter Brown and Raymond Van Dam in envisaging the crowds of
pilgrims turning up at the great feasts of Martin and Julian, and we
can sense something of the drama of the liturgy and of the mira-
culous, which sometimes (Gregory would like us to think regularly)
graced the festivities. From the work of Herbert Kessler and Raymond
Van Dam we can also appreciate what a literate, and perhaps some-
times slightly detatched, pilgrim might sense as he or she surveyed
the paintings and inscriptions in the church where the saint was
buried.
The evidence which allows this reconstruction of St Martins church
is not, in fact, confined to the writings of Gregory of Tours, but

1
P. Brown, Relics and social status in the age of Gregory of Tours, in: idem,
Society and the holy in late antiquity (London, 1982), pp. 22250: P. Brown, The cult of
the saints: its rise and function in Latin Christianity (Chicago, 1981).
2
Cf. C.A. Bernoulli, Die Heiligen der Merowinger (Tbingen, 1900).
3
R. Van Dam, Leadership and community in late antique Gaul (Berkeley, 1985), pp.
23055: R. Van Dam, Saints and their miracles in late Antique Gaul (Princeton, 1993),
pp. 1325, 30817.
4
H.L. Kessler, Pictorial narrative and church mission in sixth-century Gaul,
in: H.L. Kessler and M.S. Simpson eds., Pictorial narrative in antiquity and the Middle
Ages (Washington, 1985), pp. 7591.
138

comes first and foremost from a group of more or less epigrammatic

poems which adorned the walls of various of the Martinian sites of
Tours, and which were recorded in the so-called Martinellus.5 These
poems were at least in part commissioned by the sixth bishop of
Tours, Perpetuus, (46090), who requested them from the most
famous writers of late fifth-century Gaul, among others Paulinus of
Prigueux6 and Sidonius Apollinaris,7 as part of his programme of
developing the cult of Martin, which had its chief focus in the saints
funerary basilica outside the walled centre of the city, but which was
also significant both in the cathedral and, slightly further afield, at
the monastery of Marmoutier.
Following the evidence of the Martinellus the eye of the visitor
approaching the west end of the church of St Martin was directed
by the tower to contemplate the stars above, and thus to consider
the saint, already in heaven. Over the west door a representation of
the widows mite drew attention to the importance of good works.
Inside the church frescoes based on scenes from the New Testament
and the Life of Martin pointed to the mediation of heaven and earth,
and the idea was emphasised by poems of Paulinus of Prigueux,
commissioned by bishop Perpetuus. The climax of this mediation
was to be found over the arch of the apse, where the image of
Jacobs Ladder implicitly identified the spot as being an entrance to
heaven and thus, like its counterpart in the Old Testament, a
fearsome place. In Gregorys miracle stories this part of the church
is, as Raymond Van Dam has noted, a gravity-free zone, where a
madman who climbed to the roof sustained no injury when he fell.8
In the area of Martins tomb itself the pilgrim was inevitably reminded
of the presence of the saint, while his status in heaven was indicated
by the miracles worked there. What is presented is space as theol-
ogy in action, or put another way, as an electrical field energised
by liturgy and pilgrims.

5
Ed. L. Pietri, La ville de Tours du IV e au VI e sicle: Naissance dune cit chrtienne
(Rome, 1983), pp. 80212; transl. Van Dam, Saints and their miracles, pp. 3107.
6
Ed. M. Petschenig, CSEL 16, 1 (1888), p. 165: see also Van Dam, Saints and
their miracles, p. 314.
7
Sidonius Apollinaris, ep. IV 18, 5, ed. C. Luetjohann, MGH AA 8 (Berlin,
1887): see also Van Dam, Saints and their miracles, pp. 3156.
8
Gregory, De Virtutibus sancti Martini I 38, ed. B. Krusch, MGH SRM 1, 2
(Hannover, 1885): Van Dam, Leadership and community, p. 245.
 - 139

This remarkable dossier devoted to the Martinian shrines is unique,

although there are indications that other churches shared at least
some elements of the church of St Martin. Sidonius Apollinaris, for
instance, also composed verses for a church built in Lyons by bishop
Patiens9 though in this case the verse comes rather closer to
describing the physical reality of the church and its setting than do
those in the Martinellus. Indeed St Martins may well have been extra-
ordinary in the extent of its scheme of inscriptions and paintings,
and Perpetuus may have intended the shrine to be unique.
It is not just this uniqueness which ought to make the historian
of religion pause before turning the shrine of Martin into a type site
for the cult of the saints in Francia, or more generally of the early
medieval West. It is worth noting some of the oddities not just of
the basilica and cult site of Martin, but also of the town of Tours
itself. Although it was the metropolitan city of the province of
Lugdunensis III, and not merely a civitas capital, Tours was a rela-
tively undeveloped place, whose emergence as a regional centre seems
only to have dated to the fourth century.10 As late as the eighth cen-
tury Alcuin, himself abbot of St Martins, described the town, whose
monastery he knew to be a centre of religious power, as being small
and despicable with regard to your walls, and he compared it to
Poitiers, which he saw as a place of economic importance.11 This
distinction is all the more interesting in that Poitiers had its own
mighty religious topography, which Alcuin himself acknowledged,
and which boasted the shrines of Hilary and Radegund, and also
the monastery of Holy Cross, where there was even a fragment of
the True Cross itself, secured by Radegund from the Emperor Justin
and his wife.12 In Alcuins view Tours was a place of yet greater
holy power which is not surprising, given his own position as
abbot of the monastery of St Martin. Nevertheless, his description
of Tours as a place of little distinction beyond the shrine of Martin
is not obviously biased, and it has largely been supported by the
very substantial rescue archaeology of the last twenty years.

9
Sidonius, ep. II 10, 4.
10
H. Galini, Tours de Grgoire. Tours des archives du sol, in: N. Gauthier
and H. Galini eds., Grgoire de Tours et lespace gaulois (Tours, 1997), pp. 6580.
11
Alcuin, Vita Willibrordi prosaica 32, ed. A. Poncelet, AASS, November 7th, vol. 3.
12
See the comments of A. Cameron, The early religious policies of Justin II,
in: D. Baker ed., The Orthodox Churches and the West. Studies in Church History 13
(Oxford, 1976), pp. 5167: on the cult of Hilary in Poitiers, Van Dam, Saints and
their miracles, pp. 2841.
140

Tours does indeed seem to have been made up of a small walled

centre, a good proportion of which lay within what had been the
amphitheatre (one side of whose circuit wall functioned as a sub-
stantial proportion of the city wall), and a small vicus a kilometre
away, around the tomb of Martin itself (fig. 1).13 Essentially it was
a bifocal city, whose official centre lay in the walled area near the
cathedral, but whose numinous centre lay outside in what had been
one of the Roman cemeteries of the city, in the church and sur-
rounding vicus of St Martin as so often in post-Roman Gaul an
extramural tomb of one of the special dead was the magnet for the
reorientation of the city. A third focus, yet further out, and across
the river, was Martins old monastery of Marmoutier, which was
also the subject of verses in the Martinellus.14
It is against this background of a small community with three reli-
gious foci the funerary basilica of Martin, the cathedral and the
monastery of Marmoutier, all associated with the same saint that
one should read other aspects of the history of Tours. In such a
one-horse town the bishop will have been a figure of peculiar impor-
tance: his power should not be taken automatically as a model for
that of other bishops in Gaul, and, by extension, the bishops rela-
tions with the local comes should not be assumed to be normative.
The problematic nature of Tours as a type-site is further highlighted
by the fact that, according to Gregory of Tours, fourteen out of
nineteen bishops came from one family, his own15 and one should
note that the familys centre of gravity, in-so-far as we can recon-
struct it, lay to the south-east, round Clermont, Lyons and Dijon;16
an observation which complicates any understanding of the social
and political structure of the city of Tours, in that we are appar-
ently faced with a dominant episcopal dynasty which does not appear
to have been local.17 If we exclude the first bishop, Gatianus, Martin
himself, and (with less clear justification) three outsiders imposed by
queen Chrotechildis,18 then all the episcopal promoters of Martins

13
Galini, Tours de Grgoire. Tours des archives du sol.
14
Ed. Pietri, La ville de Tours du IV e au VI e sicle: naissance dune cit chrtienne; transl.
Van Dam, Saints and their miracles, pp. 3101.
15
Gregory, Decem Libri Historiarum, V 49, ed. B. Krusch and W. Levison, MGH
SRM 1, 1 (Hannover, 1951).
16
Van Dam, Saints and their miracles, pp. 5081.
17
This is the view expressed by Riculf in Gregory, Decem Libri Historiarum V 49.
18
Gregory, Decem Libri Historiarum X 31. Since Gregorys maternal ancestors came
 - 141

Tours

Rhone

+ ++
++ +++
+

+++

0 500 m

Fig. 1. Tours in the fourth to the sixth centuries.

1. ecclesia; 2. basilica of Saint-Lidoire; 3. basilica of Saint-Martin; 4. Saint-Gervais
and Saint-Protais; 5. Saint-Pierre and Saint-Paul; 6. baptistry of Saint-Martin; 7.
Saint-Venant; 8. Sainte-Marie and Saint-Jean-Baptist; 9. monastery of Ingetrude;
10. Saint-Monegonde; 11. Saint-Vincent; 12. monastery of Radegonde; 13. Saint-
Julian; 14. domus ecclesie; 15. permanent domestic occupation; 16. cemeteries (after
Galini 1997).

cult (notably Perpetuus and Euphronius) were relatives of Gregory.19

If the cult of St Martin at Tours in the fifth and sixth centuries is
correctly interpreted as being that of a single major saint in a minor
town, promoted largely by one family, it is obviously dangerous to
take that cult as the model for other cults elsewhere. In the situa-
tion envisaged here as being that at Tours, a single family was in
a better position to orchestrate a dominant regional or national cult
than would have been possible in more populous centres, where
there was more competition for urban leadership.

from Burgundy, he could have been related to Theodore, Proculus or Dinifius, who
are stated to have been from that region.
19
See R. Mathisen, The family of Georgius Florentius Gregorius and the bish-
ops of Tours, Medievalia et Humanistica 12 (1984), pp. 8395.
142

This is not to say that there are no cults in sixth-century Gaul

which can usefully be compared with that of Martin. In his hagio-
graphical writings Gregory provides evidence on an enormous num-
ber of cults, though for the most part that evidence amounts to only
a few lines on each shrine. One cult other than that of Martin clearly
stands out in Gregorys writings: that of Julian at Brioude, which
had as its centre a basilica constructed on the site of the saints
tomb, outside the walls of the castrum,20 with a subordinate but impor-
tant focus for the cult at the church of St Ferreolus, which was built
some two kilometres away, at Saint-Ferrol-les-Minimes, by the well
in which the saints severed head had been washed.21 Since the head
itself had been taken to, and remained in, Vienne, where it was
buried with Julians mentor, Ferreolus,22 this second shrine at Brioude
was a substitute for the absent relic. It also provided the opportu-
nity for processions within the area of Brioude itself.23 A third cult,
that of Benignus at Dijon, albeit the subject of only one substantial
chapter in Gregorys Liber in Gloria Martyrum,24 may usefully be con-
sidered alongside those of Martin and Julian. These cults, however,
have more in common than the fact that they receive some empha-
sis in Gregorys writings, where the cult of Julian functions very sim-
ilarly to that of Martin.25 Even more than Tours, Brioude and Dijon
were one-horse towns indeed they were castra, that is fortified
centres, and not civitates, or regional capitals and cathedral cities:26
further, the cults of Julian and Benignus were, like that of Martin,
promoted primarily by Gregorys family. His uncle, Gallus, founded
the annual pilgrimage from Clermont to Brioude,27 and the family
as a whole was so closely involved with the cult,28 that one may

20
For the funerary basilica of Julian, Gregory of Tours, De virtutibus sancti Juliani
45, 9, etc., ed. B. Krusch, MGH SRM 1, 2 (Hannover, 1885): Van Dam, Saints
and their miracles, pp. 418. See also I.N. Wood, Constructing cults in early medieval
France: saints and churches in Burgundy and the Auvergne 4001000 (forth-
coming).
21
Gregory, De Virtutibus sancti Juliani 3, 25, 26, 28.
22
Gregory, De Virtutibus sancti Juliani 12.
23
Gregory, De Virtutibus sancti Juliani 28.
24
Gregory, In Gloria Martyrum, 50, ed. B. Krusch, MGH SRM 1, 2: I.N. Wood,
Early Merovingian devotion in town and country, in: D. Baker ed., The church in
town and countryside, Studies in Church History 6 (Oxford, 1979), pp. 745.
25
Compare Gregory, De Virtutibus sancti Martini with Gregory, De Virtutibus sancti
Juliani.
26
For Dijon, see Gregory of Tours, Decem Libri Historiarum III 19.
27
Gregory, Vitae Patrum, VI 6, ed. B. Krusch, MGH SRM 1, 2.
28
Gregory, De Virtutibus sancti Juliani 235.
 - 143

wonder whether it possessed estates in the vicinity. Certainly Gregorys

father owned land in the Limagne, which lies between Brioude and
Clermont.29 In Dijon the cult of Benignus was essentially created by
Gregorys great-grandfather and namesake.30
The evidence for the origins of the cult of Benignus is worth paus-
ing over, since Gregory of Tours is particularly clear over them, and
they thus provide one model for the creation of a religious topo-
graphy. In the cemetery at Dijon was a large sarcophagus, which
attracted a certain amount of local devotion and was the scene of
at least one minor miracle. The bishop, Gregory of Langres, attempted
to stop the development of a cult, which he assumed to be pagan,
but subsequently had a vision of the martyr Benignus, who revealed
that the site was his burial place. Gregory reacted by renovating an
adjacent crypt, into which he had the sarcophagus moved. Later an
account of Benignus martyrdom was discovered in Italy, and there-
after the bishop had a large basilica built on the site. Nor was that
the end of the development of the holy place, for another church,
where the nun Paschasia was venerated, was built nearby. It was
even said that Paschasia herself had told the builders that Benignus
was supervising their work.31 This tale of a bishops attempt to stop
the development of a cult, followed by his championing of it, is all
the more interesting because Gregorys family appears to have had
land in the vicinity,32 just as Gregory of Tours Auvergnat relatives
may have owned property near St Julians at Brioude. This may
account for the familys preference for the castrum of Dijon to the
diocesan centre of Langres,33 and indeed for its promotion of the
cult of Benignus.
It may be no more than coincidence, but it is worth noting that
in championing the cult of Julian at Brioude, Gregory of Tours
family was promoting a cult site in the Auvergne which lay at the
other end of the diocese from the episcopal seat in Clermont, and

29
Gregory, In Gloria Martyrum 83: I.N. Wood, Gregory of Tours (Bangor, 1994),
p. 5.
30
Gregory, In Gloria Martyrum, 50. It is perhaps significant that the chapter on
Dijon in Gregorys Decem Libri Historiarum III, 19, is entitled De sancto Gregorio et situm
Divioninsis castri.
31
Gregory, In Gloria Martyrum 50; Gregory, In Gloria Confessorum 42, ed. B. Krusch,
MGH SRM 1, 2.
32
Gregory, Vitae Patrum VII, 23; Wood, Gregory of Tours, p. 6.
33
Wood, Gregory of Tours, pp. 67.
144

that Dijon, separated as it is from the cathedral city of Langres, pro-

vides an exact parallel. Gregory himself thought either that Dijon
should have been the civitas capital, or else that it should become
the centre of a new civitas34 the word is ambiguous, suggesting the
centre of both a secular unit and an ecclesiatical diocese. Yet one
should also remember that Langres may have been a centre of some
significance: it certainly boasted a major cult in that of three broth-
ers who were martyred in the Roman persecutions, the Tergemini
a cult about which the bishop of Tours, somewhat suspiciously, has
absolutely nothing to say, despite the fact that it was clearly well
established in his day.35 It looks, then, as if Gregory of Langres
after initial doubts and his descendents took advantage of a local
(probably non-christian) cult, which fortunately happened to be based
near one of their own properties, christianised it and promoted it to
their own benefit. Other sections of Gregory of Tours family may
have made similar use of the previously existing cults of Martin and
Julian at Tours and Brioude.
Gregory of Tours account of the discovery of the tomb of Benignus
is not our only piece of evidence relating to the cult of this more-
than-dubious martyr. One of the more remarkable works to have
survived from sixth-century Burgundy is a composite passion of a
number of martyrs from the region, the Passio sanctorum Herenei epis-
copi, Andochi presbiteri, Benigni presbiteri, Tyrsi diaconi, Felicis negotiatoris.36
This text, which survives in a single Farfa manuscript, provides a
fictitious account of early missionary work in the region of Autun,
Saulieu, Langres and Dijon. It should probably be seen as one aspect
of the cult of Benignus, and could even be the product of the house-
hold of Gregory of Langres or of one of his (related) successors
it is perhaps the Passio of Benignus supposedly found in Italy. Effectively
it provides a set of martyr acts which set Benignus and Dijon at the
very centre of the christian history of North Burgundy, and in so

34
Gregory, Decem Libri Historiarum III 19.
35
On the presence of the Tergemini in the Passio sanctorum Herenei episcopi, Andochi
presbiteri, Benigni presbiteri, Tyrsi diaconi, Felicis negotiatoris see below. In the eighth cen-
tury the cult was also of some significance in Northumbria, because abbot Ceolfrith
of Wearmouth-Jarrow died at Langres.
36
J. van der Straeten, Les actes des martyrs dAurlien en Bourgogne: le texte
de Farfa, Analecta Bollandiana 79 (1961), pp. 44768. The work is discussed at length
by van der Straeten in Les actes des martyrs dAurlien en Bourgogne: tude lit-
traire, Analecta Bollandiana 79 (1961), pp. 11544. See also, Wood, Constructing
cults in early medieval France.
 - 145

doing provides a wider context for the cult than is immediately

apparent from Gregorys remarkable chapter. Benignus is promoted
as a major saint not just for the diocese of Langres, where he is
quite clearly portrayed as superior to the local Tergemini, who he
baptises, but also for Autun and Lyons. That Benignus has to be
portrayed as the baptiser of the Tergemini is, one might note, an indi-
cation that their cult was too important to be ignored by the hagio-
grapher.

Instead of assuming that Tours, Brioude and Dijon are type-sites for
the cult of the saints, we should simply take them for what we know
them to be, that is sites of cults for which we have significant evi-
dence in Gregory of Tours writings, and which were also sites dom-
inated in the sixth century by Gregorys family. Here it is important
to remember the extent to which families and indeed individuals
were associated with specific cults.37 In all probability, when the pos-
sessed at Brioude cried out that they were being tormented by the
arrival Privatus of Javols, Ferreolus of Vienne, Symphorian of Autun
and Saturninus of Toulouse, who had come to join Julian, they were
reflecting the presence of living supporters of these additional saints.38
The arrival, as announced once again by the possessed, of Julian,
Privatus, Martin, Martial (of Limoges), Saturninus and Denis, at the
death of Gregorys friend Aridius might equally indicate the attach-
ments of the dying holy man, or of those who were to attend his
funeral.39
Given the particular nature of personal attachments to saints, and
the precise topographical contexts of the cults of Martin, Julian and
Benignus, it is dangerous to regard Tours, Brioude and Dijon as
being in any sense normative. Other sites may have been different,
and should be examined on their own terms, before we decide
whether they follow the model of Tours or not. I shall here look at
two cities whose religious topography is relatively well known, to see
how far they differ from the three centres discussed so far: Clermont
and Vienne, both of them, significantly, cities in which Gregory had
an interest, but whose cults with the possible exception of that

37
See the analysis of Gregorys attachments in Van Dam, Saints and their mira-
cles, pp. 5081.
38
Gregory, De Virtutibus sancti Juliani 30.
39
Gregory, Decem Libri Historiarum X 29.
146

of Julians master, Ferreolus, at Vienne40 are portrayed by Gregory

differently from those of Martin, Julian and Benignus, or alterna-
tively are not portrayed by the bishop of Tours at all.
Clermont appears frequently in Gregorys writings, particularly in
his narrative of the first half of the sixth century.41 It was, after all,
the city in which he was largely brought up, notably by his uncle
Gallus. Not surprisingly Gregory has some interest in some of the
saints of Clermont, especially Illidius,42 Abraham,43 Sidonius Apolli-
naris,44 Quintianus of Rodez45 and Gallus himself.46 In the cases of
the last four saints, however, Gregory is interested in their lives rather
than in their posthumous cults and three of them interest him
as bishops. He also mentions in passing a number of churches in
the city.47 In Gregorys writings, however, the most prominent reli-
gious festival in the Clermont region is the Rogation pilgrimage from
Clermont to Brioude, instituted by Gallus to avert the threat of
plague.48 Gregory, as we have seen, gives the religious topography
of the Auvergne a focus seventy kilometres to the south of Clermont
in Brioude. At first sight Clermont itself scarcely appears as a cen-
tre of holy power. Nevertheless, a close reading of Gregory does
reveal the existence of numerous churches and shrines in the city,
and, fortunately there is additional and important evidence to sug-
gest that even in Gregorys day the citys religious topography was
very different from that of Tours, Brioude or Dijon, with their

40
Gregory, De Virtutibus sancti Juliani 12.
41
For the christian topography of the town, P-F. Fournier, Clermont-Ferrand
au VIe sicle, recherches sur la topographie de la ville, Bibliothque de lcole des
Chartes, 128 (1970), pp. 273344; M. Vieillard-Troiekouroff, Les monuments religieux de
la Gaule daprs les uvres de Grgoire de Tours (Paris, 1976), pp. 85102. See also I.N.
Wood, The ecclesiastical politics of Merovingian Clermont, in: P. Wormald ed.,
Ideal and reality in Frankish and Anglo-Saxon society (Oxford, 1983), pp. 3457, and I.N.
Wood, Clermont and Burgundy: 511534, Nottingham Medieval Studies 32 (1988),
pp. 11925.
42
Gregory, Vitae Patrum II.
43
Gregory, Vitae Patrum III.
44
Gregory, Decem Libri Historiarum II, 213.
45
Gregory, Vitae Patrum IV.
46
Gregory, Vitae Patrum VI.
47
Fournier, Clermont-Ferrand au VIe sicle, recherches sur la topographie de
la ville: Vieillard-Troiekouroff, Les monuments religieux de la Gaule daprs les uvres de
Grgoire de Tours, pp. 85102.
48
Gregory, Vitae Patrum VI, 6.
 - 147

single dominant cults. Clermont, like other major cities such as Lyons
and Vienne, was almost literally ringed with shrines.49
Our main source for the religious topography of Clermont in the
early middle ages is not in fact Gregory, but rather the ninth-
century Libellus de Ecclesiis Claromontanis, a list of the churches and
altars of the city and its immediate environs.50 This document lists
fifty-four churches to be found in and around the cathedral city of
Clermont. Thirty-four of these churches were under the control of
the bishop at the time that the list was compiled; of these nine were
monastic. The twenty remaining churches were in the power of the
comes or of vassi dominici. Within these churches were one hundred
and twelve altars. There were also the bodies, or perhaps parts of
the bodies, of forty-eight named saints, and of numerous others whose
names were known to God. One church, that of saint Venerandus,
boasted six thousand two hundred of these anonymous saints. All in
all over two hundred dedications are listed, referring to close on one
hundred named saints. Even allowing for confusion between saints
of the same name and other errors in identification this provides a
remarkable sample. Moreover the picture given by this sample gives
a different impression from that presented on the surface by Gregory,
although a significant number of the cults and churches were in exist-
ence by his day, and a number of the churches and shrines are actu-
ally referred to in passing by the bishop of Tours.
Clearly the Libellus gives us a picture of an early medieval city as
it had developed by the ninth century, and Tours, Brioude and Dijon
may all have developed by that time, so that the text does not pro-
vide a perfect control for the picture given by Gregory. One should,
however, remember that Tours for Alcuin in the eighth century was
a scruffy place,51 and that it may not, therefore, have developed
much since Gregorys life-time. And one should also note that the
bishop of Tours himself names, albeit only in passing, a significant
number of the churches and saints listed in the Libellus, making it

49
Fournier, Clermont-Ferrand au VIe sicle, recherches sur la topographie de
la ville.
50
Libellus de Ecclesiis Claromontanis, ed. W. Levison, MGH SRM 7 (Hannover,
1920). See also Wood, Constructing cults in early medieval France: saints and
churches in Burgundy and the Auvergne 4001000.
51
Alcuin, Vita Willibrordi 32.
148

quite clear that even in his day Clermont had a large number of
cults, of which no single one was dominant.52
It is worth setting this information alongside what else is known
of the city. What is apparent both from Gregory and from the sev-
enth-century Passio Praeiecti and the eighth-century Vita Boniti is that
Clermont was a divided town, with a competitive upper class, which
vied for episcopal power and indeed for comital office: apart from
Gregorys own family, there were the descendents of Hortensius, and,
at the start of the sixth century if not later, the descendents of
Sidonius Apollinaris.53 This is anything but the same as Tours, with
its single dominant episcopal family, despite the challenge made to
Gregory by Leudast and Riculf,54 at least up until the end of the
sixth century. Moreover, as we have already noted with regard to
Gregorys own family, different groups favoured different saints. This
can be seen once again in the bishop of Tours writings, where the
descendents of Sidonius try to promote the cult of Anatolianus,55
while the Auvergnat cult of Stremonius, which is passed over tersely
by Gregory, was championed at Issoire by one of the btes noires of
the Decem Libri Historiarum, bishop Cautinus,56 and subsequently, in
the seventh century, by Praeiectus, who was martyrarius at the shrine
before becoming bishop of Clermont.57 Gregorys relative silence is
all the more noteworthy in that Stremonius was supposedly one of
the seven missionaries to Gaul as well as the first bishop of Clermont.
It is worth pausing just a little longer over Gregorys silence with
regard to certain cults, such as that of Stremonius or indeed of the
Tergemini. It was not only minor saints who were ignored, or nearly
so, by the bishop of Tours. Among those who receive minimal atten-
tion in Gregorys hagiography are Genovefa58 and Denis59 in the

52
See Wood, Constructing cults in early medieval France: saints and churches
in Burgundy and the Auvergne 4001000.
53
Wood, The ecclesiastical politics of Merovingian Clermont. See also P. Fou-
racre, Merovingian history and Merovingian hagiography, Past and Present 127
(1990), pp. 216; P. Fouracre and R. Gerberding, Late Merovingian France, history and
hagiography 640720 (Manchester, 1996), pp. 254300.
54
Gregory, Decem Libri Historiarum V 49.
55
Gregory, In Gloria Martyrum 64.
56
Gregory, In Gloria Confessorum 29.
57
Passio Praeiecti, 5, ed. B. Krusch, MGH SRM 5 (Hannover, 1910).
58
Gregory, In Gloria Confessorum 89; Gregory, Decem Libri Historiarum IV 1. See,
on the cult of Genovefa, M. Heinzelmann and J.C. Poulin, Les Vies anciennes de sainte
Genevive de Paris. tudes critiques (Paris, 1986).
59
Gregory, In Gloria Martyrum 72; Gregory, Decem Libri Historiarum I 30, V 32,
 - 149

Paris region, and Marcellus at Chalon-sur-Sane,60 although the last

two receive some attention in the Histories, because of the political
importance of their cult sites. In the sixth century the cults of Genovefa
and Marcellus were being championed by royalty. The same is true
of that of Denis at least by the second quarter of the seventh, and
even if it had not had much royal backing earlier, it unquestionably
had its devotees from the early sixth century onwards, not least from
the saintly Genovefa. Gregorys lack of interest in Genovefa and
Marcellus may suggest that he was inclined to overlook perhaps
for strategic reasons cults which could present a threat to that of
his beloved Martin.61 As we have seen, he could be equally selec-
tive when dealing with the potential Auvergnat rivals to Julian, or
with the cult of the Tergemini in Langres.

To return to Clermont itself: it is not unreasonable to conclude that

there is some connection between the numerous saints and saint-
cults of Clermont and the political divisions within the city and
to propose that the religious topography of Clermont was intimately
connected to the citys social and political structure. By contrast,
Tours, Brioude and Dijon were relatively simple centres, apparently
dominated by a single cult, and in the case of Tours unquestion-
ably, and of Dijon almost certainly, the cult was dominated by a
single aristocratic family, that of Gregory himself. The Tours model,
in other words, which seems to have become something of a blue-
print for scholars studying the cults of Francia, might only be expected
to hold in relatively simple urban centres. The Clermont model, by
contrast, might be expected to hold for those centres where there
was a more competitive social and political scene. Which of these
two models was predominant in Francia is open to question: a care-
ful survey of the religious topography of individual cities might be
used as an index of social competitiveness.

34, X 29: Vita Genovefae 1722, 30, ed. B. Krusch, MGH SRM 3 (Hannover, 1896).
60
Gregory, In Gloria Martyrum 52 (a miracle perhaps only recorded because of
the involvement of an Auvergnat); Gregory, Decem Libri Historiarum V 27, IX 3, 27,
X 10. Also instructive when assessing the importance of the cult is Fredegar, IV
1, 14, ed. B. Krusch, MGH SRM 2 (Hannover, 1888).
61
Gregorys commitment to Martin is still best expressed by J.M. Wallace-Hadrill,
Gregory of Tours in the light of recent research, in: idem, The long-haired kings
(London, 1962), pp. 4770.
150

One region where the Clermont model is likely to have been pre-
dominant is that of the Rhne valley with its numerous and still rel-
atively populous cities. Lyons certainly boasted both a large number
of cults and a divided political society. The cults were represented
by a great series of extra mural churches, which prompted Avitus
of Vienne to comment that the town was protected rather by basil-
icas than by defences: plus haec basilicis quam propugnaculis urbs muni-
tur.62 Several of the cults are registered clearly by Gregory himself,
and he also describes the divided nature of Lyonnais society, most
obviously in his accounts of the episcopates of Nicetius and Priscus,
bishops drawn from hostile factions, with Nicetius being, once again,
a relative of Gregory.63 Vienne is likely to have been similar, although
the political scene of the city is not recorded for this period. Certainly
it had a significant number of religious centres, apparent from Ados
Life of Theudarius,64 and reflected in the modern epithet Vienne la
sainte.65 Gregory, however, only deals with the cult of Ferreolus,
master of his beloved Julian not least because the martyrs head
was buried in his tomb.66
The fifth- and sixth-century evidence for Vienne, nevertheless,
allows a rather different approach to the question of religious topog-
raphy. Particularly important here is the evidence for the institution
by bishop Mamertus of Vienne of the Rogations, three days of pen-
itential prayers. These are first recorded in a letter written by Sidonius
to Mamertus in the spring of 473, after the former had introduced
the ceremony to Clermont to secure Christs aid in the crises (among

62
Avitus, Homily 24, ed. R. Peiper, MGH AA 6, 2 (Berlin, 1883). For an under-
standing of this particular homily it is necessary to turn to the edition by Perrat
and Audin, Alcimi Edicii Aviti Viennensis Episcopi Homilia Dicta in dedicatione
superioris basilicas, in: Studi in onore di A. Calderini e R. Paribeni, vol. 2 (Milan, 1957),
pp. 43351.
63
Gregory, Vitae Patrum VIII 5; Gregory, Decem Libri Historiarum IV 36. See also
Brown, Relics and social status in the age of Gregory of Tours, p. 245.
64
Ado, Vita Theudarii, Migne PL 123, cols. 44350. See I.N. Wood, A prelude
to Columbanus: the monastic achievement in the Burgundian territories, in: H.B.
Clarke and M. Brennan eds., Columbanus and Merovingian Monasticism, BAR International
Series 113 (Oxford, 1981), p. 61.
65
For the religious topography of Vienne, N. Gauthier and J-Ch. Picard eds.,
Topographie chrtienne des cits de la Gaule des origines au milieu du VIII e sicle, vol. 3,
Provinces ecclesiastiques de Vienne et dArles (Paris, 1986), pp. 1735. For the evidence
of the epitaphs of the bishops of the city see M. Heinzelmann, Bischofsherrschaft in
Gallien. Zur Kontinuitt rmischer Fhrungsschichten vom 4. bis zum 7. Jahrhundert. Soziale,
prosopographische und bildungsgeschichtliche Aspekte (Mnchen, 1976).
66
Gregory, De Virtutibus sancti Juliani 2, 30.
 - 151

them the Visigothic expansion) afflicting his own city.67 According

to Sidonius, Mamertus instituted three days of processions and lita-
nies as a result of a series of earthquakes. Initially Mamertus had
imposed fasts on the clergy and had incited the populace at large
to prayer. The obedience of the ordinary people shamed the rich,
who had fled the city, into returning. Thereafter the natural disas-
ters had ceased. In addition, remarks Sidonius, Mamertus had copied
Ambrose, the discoverer of the bodies of the martyrs Gervasius and
Protasius, and had translated the body of Ferreolus and, with him,
the head of Julian the martyr of Brioude, and patron of both
Sidonius and Gregory of Tours.
What is most interesting for the present argument is the evidence
for division within the city of Vienne division which is partly,
but only partly, determined by class. Mamertus can dictate to the
clergy and inspire the man in the street, but the aristocracy is ini-
tially absent, until shamed into piety. Roughly the same story, with
slight differences of emphasis, is recounted by Avitus of Vienne, in
his homily preached on the eve of the Rogation processions.68 As
the godson of Mamertus, and as his successor-but-one in the epis-
copacy of Vienne, Avitus is a well-placed witness, and he was preach-
ing to a congregation part of which, as he openly states, had lived
through the crisis, as he himself had. Avitus again situates the insti-
tution of the Rogations in a time of earthquakes, and he attributes
the idea of the processions to Mamertus, who was left standing alone
at the altar during an Easter service when a tremor, which led to
the destruction by fire of the city hall, caused the rest of the pop-
ulation to rush from the church. Avitus also recounts the opposition
of the senatorial aristocracy, noting the then size of the citys curia.
As a result Mamertus planned the institution of the Rogations with
some care. He decided that his liturgical innovation should both be
seen to be an instant success and should be habitual from the start.
To this end he softened up his congregation with prayers, and then
chose the three days between the feast of holy Ascension and the
following Sunday, as if there were a zone of special opportunity sur-
rounded by solemnities. That is he chose a timespan which was, in
a sense, a free period immediately following a major festival. On the

67
Sidonius, ep. VII, 1.
68
Avitus, Homily 6.
152

Friday he held a first day of processions and prayers, which was

organised so that there should be no stragglers. This brevis atque angusta
processio was focused on a basilica which was at the time close to the
city walls since Avitus implies that the church in question had
to be moved, it can reasonably be identified with the basilica in
which Ferreolus, together with the head of Julian, was buried, since
Mamertus had subsequently to rebuild the church on a new site,
because the original building was being undermined by the river
Rhne.69 The success of this first procession ensured that of the full
three days of prayer. These Rogations were subsequently copied
throughout Gaul as both Avitus and Sidonius bear witness.
What is clear in both narratives is the psychological astuteness of
Mamertus. He was aware that support for his idea was unlikely to
be universal, and that members of the senatorial aristocracy were
distinctly lukewarm about any novelty, so he instituted his proces-
sions in such a way that they were an instant success, and he
embarassed the sceptics into joining. Nor was he the only bishop
who resorted to tricks to entrap his congregation: better known
and less subtle is Caesarius of Arles technique of having the
church doors locked before he began his sermons, to prevent the
congregation slipping away.70
Mamertus success in social management was noted in a letter of
Sidonius Apollinaris to the aristocrat Aper.71 Sidonius, who was by
this time bishop of Clermont, had recently instituted Rogations on
the model of those of Vienne, and he wished to use them to unify
the population of Clermont. He thus wrote to Aper, who had,
significantly and not unlike the Viennese aristocracy, gone off to his
rural estate, urging him to attend. He admitted that previous days
of public prayer had been pretty lukewarm affairs. They had not
pleased everyone, for prayers for rain or sunshine could not find
favour with both the gardener and the potter at the same time.
Equally important, they had got in the way of aristocratic dinners.
Such a disengaged attitude to religious feasts was something that
Sidonius himself would have understood, for in a letter to Eriphius
recording the feast of St Justus at Lyons he noted, without criticism,

69
Gregory, De Virtutibus sancti Juliani 2.
70
Cyprianus et al., Vita Caesarii I 27, ed. B. Krusch, MGH SRM 3.
71
Sidonius, ep. V 14.
 - 153

the impromptu organisation of a ball game outside the church, while

waiting for the next service to begin.72
Sidonius at Lyons and Clermont, Gregory of Langres at Dijon
and Mamertus in the Rogation liturgy of Vienne introduce us to
some of the religious, social and political realities underlying the
development of the religious topography of Gaul. Gregory was faced
with a cult that he did not like, but exploited it to his own ends.
Mamertus was faced with a sceptical aristocracy, and forced them
to join his annual processions, in which the people of Vienne became
the reincarnation of the people of Nineveh, whose united penitence
secured divine forgiveness. This was an achievement that Sidonius
wished to emulate, but he knew only too well what difficulties
Mamertus had had to overcome in inspiring the population of Vienne
to acts of collective prayer and penance.
One might add the evidence for a slightly different category of
opposition: that of a bishop who, unlike Gregory of Langres, had
no means of controlling the development of a cult to which he took
exception. Maroveus in Poitiers, faced with Radegunds acquisition
of a relic of the True Cross, deliberately absented himself from the
city, so as to make the canonical enshrining of the relic difficult.73
He resorted to the same tactic when it came to the burial of Radegund
herself.74 He seems to have taken the presence of the relic of the
Holy Cross as a challenge to his own authority, and he clearly had
no desire to help promote the cult of the woman who had caused
him so much discomfort even though, unlike the relic of the Holy
Cross, her tomb was outside the confines of her nunnery, and thus
potentially within his control. Doubtless for him the great cult of
Poitiers was that of Hilary, theologian and, perhaps more important,
bishop75 focused, yet again, on a great extra-mural basilica.
Most of Gregory of Tours evidence on the holy topography of
Gaul relates to established cults he treats the actual process of
their establishment very much less frequently. But it is clear from
his account of events at Dijon that the institution of a cult was no
cut-and-dried issue. Even the bishop might be sceptical, or, as in the
case of Maroveus at Poitiers, downright hostile. From the writings

72
Sidonius, ep. V 17.
73
Gregory, Decem Libri Historiarum IX 40.
74
Gregory, In Gloria Confessorum 104.
75
Van Dam, Saints and their miracles, pp. 2841.
154

of Sidonius and Avitus it is clear that Mamertus was faced with a

sceptical aristocracy, and that his congregation needed to be mani-
pulated: in Clermont Sidonius himself oversaw a similar congrega-
tion. To forget the scepticism and the manipulation is to misunderstand
the complex processes which lay behind the creation of the religious
topography of sixth-century Gaul.76
This glimpse of the realities of religious enthusiasm (or its absence)
needs to be borne in mind when one considers Gregory of Tours
evidence on the cult of the saints. Indeed it should make one look
very carefully for indications in Gregorys writings that not every-
one was as enthusiastic about shrines as he was,77 and even he was
more enthusiastic about some shrines than others. There is, indeed,
a fair amount of evidence, even in Gregorys hagiography, for a less
than credulous view of the miraculous and its working at the tombs
of the special dead. At the same time, to ignore the differences
between a city like Tours, with its single episcopal family, oddly
enough imposed from outside, and a city like Vienne or Clermont,
with their sizeable aristocracies composed of potentially rival fami-
lies, is to ignore the enormous variety of circumstances which defined
the possibilities for a religious topography in any one place.
The image of the cult of the saints presented by Gregory is a
remarkable reading by a single author: it deserves to be taken seri-
ously, as indeed it has been in the last twenty years78 but it should
also be understood for what it is: one authors view, influenced largely
by his experiences at two arguably eccentric places, Brioude and
Tours. Other sixth-century cities, and other sixth-century authors,
had different stories to tell, and different types of topography to
describe. We now understand Gregorys view of the Holy well enough
to set against it the various alternative models that can be recon-
structed from other sources.

76
On this complexity, see also Brown, Relics and social status in the age of
Gregory of Tours.
77
See, I.N. Wood, How popular was early medieval devotion?, in: A.J. Frantzen
and T.N. Hill eds., Popular piety: Prayer, devotion and cult, Essays in Medieval Studies,
Proceedings of the Illinois Medieval Association 14 (1997) http://www.luc.edu/pub-
lications/medieval/vol14.
78
For an analysis of Gregorys view, however, see, crucially, Brown, The cult of
saints, esp. pp. 69127.
 MAASTRICHT AS A CENTRE OF POWER
IN THE EARLY MIDDLE AGES*

Frans Theuws

In late Antiquity Maastricht was a northern outpost of what was

then considered the civilised world, a position it had gained after
bishop Servatius died there in the second half of the fourth century.
He was buried in a cemetery on the road to Tongres, the capital
of the civitas where Maastricht was situated. Without Servatius,
Maastricht would probably have remained a place of little significance,
a small fortress on the river Meuse protecting the bridge on the road
from Tongres to Cologne. The very fact that this insignificant place
emerged as a key centre in a vast region is an indicator of the power
of the saint. Even today Servatius is vital to the self-image and iden-
tity of the citizens of Maastricht. Some see Servatius as one of the
mythical city founders of early medieval Europe.1 The early history

* This paper was written in the stimulating environment of the Netherlands

Institute for Advanced Study in the Humanities and Social Sciences (NIAS) of the
Dutch Royal Academy of Sciences. Discussions with Tom Noble and Julia Smith,
also at NIAS in 19992000, were a source of inspiration and I thank them for
sharing their insights. I would also like to thank the members of Power and Society,
Theme group 5 of the ESF programme on the Transformation of the Roman
World, for sharing their ideas. This article was written before Alain Dierkens, one
of the leading experts on the early history of the Meuse valley, published his views
on late-Roman and Merovingian Maastricht in reaction to Leupens study. We
exchanged our ideas at the ESF meeting in Nicosia, Cyprus, in March 1999, and
at the Mosa Nostra conference in Namur that same year. I would like to thank
him for his open-minded discussions. Some points in his study notably those
relating to the archaeological evidence merit further discussion. I did not include
these in the present article in the hope that we may continue the debate on
Maastricht and its related centres in the context of a future project in which the
archaeological data figure more prominently. Mayke de Jong transformed my Dutch-
English into something bearing a closer resemblance to proper English, while Annette
Visser did the final editing. Mayke de Jong, Wim Dijkman, Arnoud-Jan Bijsterveld
and Jos Bazelmans also made valuable comments on an earlier draft. Last but not
least I would like to thank Piet Leupen for his comments and for the stimulating
discussions on the subject. It was his study on the cult places of Maastricht that
sparked off a new in-depth debate on the significance of Maastricht, following a
long period in which we thought there was nothing more to be said about its early
history. He was my teacher of the first hour, and has remained my mentor until
this moment.
1
P.H.D. Leupen, Het gelijk van Sint-Servaas en Onze-Lieve Vrouwe. De
156

of Maastricht is hotly debated. The key issue is the status of its major
churches, the basilica of Our Lady in the old Roman fortress and
the cemeterial basilica of St Servatius, and their relation to the bishop.
Maastrichts complex late antique and early medieval past is worth
investigating, and not just because the fascinating local history offers
opportunities for comparison with other centres of the period. At a
more abstract level, it allows us to study the significance of a cen-
tre of power in a wider social context. We have been helped here
by twenty years of intensive archaeological research on the part of
Maastrichts municipal archaeological department, and particularly
the efforts of its archaeologists Titus Panhuysen and Wim Dijkman.
Alain Vanderhoeven from the Institute of Archaeological Heritage
of the Flemish Government in Belgium has transformed the archae-
ological landscape of Tongres in a similar fashion. The history of
late antique Maastricht cannot be understood without that of Tongres,
just as the late Merovingian history of Maastricht cannot be under-
stood without that of Lige, a new centre that emerged in the early
eighth century.
This paper does not seek to present final results. Instead, it is an
attempt to initiate a new line of research. My first objective is to
analyse the local history of Maastricht in late Antiquity and the
Merovingian period; my second is to show how this local history
relates to the debate on late antique and early medieval centres of
power and their conceptualisation. This article therefore constantly
moves between local history and theory. It contains two different
parts, each with a distinct emphasis.2 The first concentrates on
Maastrichts highly controversial local history. I have no hope of
solving existing problems and shall in fact add new ones. In this sec-
tion I intend to focus on the different actors relevant to such topo-
graphical elements of Maastricht as the castrum, cult places, portus,
and the surrounding landscape. I will devote special attention to the
significance of aristocratic families for the early development of the

Maastrichtse kapittels in de vroege Middeleeuwen, in: E.S.C. Erkelens-Buttinger

e.a. eds., De Kerk en de Nederlanden. Archieven, instellingen, samenleving (Hilversum, 1997),
pp. 2942, esp. p. 30. On the Roman history of Maastricht see T.A.S.M. Panhuysen,
Romeins Maastricht en zijn beelden (Maastricht/Assen, 1996).
2
In addition, the article contains a third strand relating to problems that are
discussed in detail. The historiography of Maastricht and the centres in the Meuse
valley forces modern scholars to explain their position in these highly specialised
discussions so that they can proceed with the main argument. Most of this discus-
sion is found in the notes.
 157

town, as this aspect has been neglected by a historiography with an

almost exclusive focus on the bishop, his office and institutions. The
early history of Maastricht is controversial, not only because of the
patchiness of extant sources and the concomitant difficulties of inter-
pretation, but because the debate has been dominated by a number
of unhelpful perspectives. Two of these we can identify easily: the
economic and the institutional. Adherents of the economic view
maintain that economic activities and, more specifically, long-distance
trade were responsible for the growth of Maastricht and Huy, Namur
and Dinant, related early medieval centres in the Meuse valley.3 This
perspective has recently been criticised.4 The development of these
centres should be interpreted in a wider context, for they were essen-
tial to the constitution of society as a coherent whole. In other words,
Maastricht did not just flourish as a result of economically favourable
conditions and long-distance trade. The economic perspective seems
to have been borrowed from the study of the development of later
towns.5 Early medieval centres, however, are different from late
antique or high medieval towns. Models useful to research into other
periods should not be applied uncritically to early medieval centres.
The institutional perspective has a different but equally problem-
atic consequence, namely, that different segments of society are stud-
ied independently. It has also led scholars of central places to neglect
the role of aristocrats. In modern historiography, the history of Maas-
tricht and related centres is one of kings, bishops and merchants.

3
One advocate of this perspective is F. Rousseau in his influential study La Meuse
et le pays Mosan. Leur importance historique avant le XIII e sicle (Namur, 1930). The pro-
duction of coins was often seen as proof of the economic importance of the power
centres in the Meuse valley. I will come back to the significance of coin produc-
tion in these centres later.
4
G. Despy, Villes et campagnes aux IXe et Xe sicles: lexemple du pays mosan,
Revue du Nord 50 (1968), pp. 14568. G. Despy, Lagglomration urbaine pendant
le haut moyen age (du VIIe sicle aux environs de 1200), Namur, le site, les hommes
de lpoque Romaine au XVIII e sicle (Bruxelles, 1988), pp. 6378. J.-P. Devroey, Villes,
campagnes, croissance agraire dans le pays Mosan avant lan mil: vingt ans aprs,
in: J.-M. Duvosquel and A. Dierkens eds., Villes et campagnes au moyen ge, Mlanges
Georges Despy (Lige, 1991), pp. 22360 (Repr. in J.-P. Devroey, Etudes sur le grand
domaine Carolingien (Aldershot, 1993), XII). A. Verhulst, The rise of cities in north-west
Europe (Cambridge, 1999), pp. 249. F. Theuws, Centres commerciaux dans la
valle de la Meuse et lconomie du Haut Moyen ge. Vers une rpresentation
conceptuelle des centres du haut moyen age, in press.
5
This of course comes as no surprise as many high medieval towns developed
out of early medieval centres that were then automatically regarded as early towns
or proto-towns and analysed from the same perspective.
158

The one exception is Kuppers study of Bishop Lamberts Life, which

does highlight the role of aristocrats in the development of these
centres and in connection with episcopal power.6 Yet this particular
study has not led to a further analysis of the nature of episcopal
power in the civitas of the Tungri. Aristocrats barely feature in the
narrative on these central places, in spite of the fact that they emerged
in the same period, and in roughly the same region, as important
aristocratic groups like the Pippinids.7 Because these groups do not
have neat institutionalised histories, they are lost sight of in an insti-
tutional perspective. What we need is a new conceptualisation of
these centres.
The second part of the paper will therefore deal with a number
of more abstract questions, and with the development of new con-
cepts. My analysis will examine not just centres of power themselves,
but the surrounding cultural landscape of which they were a part.
The ultimate aim is to arrive at a better understanding of the mean-
ing of centres of power in the early middle ages.
It is not easy to define centres of power and I will not make a
serious attempt here. A focus on power would also yield too limited
a perspective. Although power must have played a vital role in these
centres, there is more to it than that. They were places where val-
ues, norms and ideas were given form, and were reproduced, rep-
resented, interpreted and negotiated upon by different groups in society.
In other words, we are dealing with places that were essential to the
constitution of the socio-cosmological order as a whole. Any place
can function as a centre of power, from a farmstead to a large town,
for within this wide range society is constructed and reproduced,
albeit at different social levels and with different levels of intensity.
Obviously such activity had a far greater impact in a place like early
medieval Maastricht than in a farmstead.

6
J.L. Kupper, Saint Lambert: de lhistoire la lgende, Revue dhistoire ecclsi-
astique 79 (1984), pp. 549.
7
M. Werner, Der Ltticher Raum in frhkarolingischer Zeit. Untersuchungen zur Geschichte
einer karolingischen Stammlandschaft (Gttingen, 1980). R. Gerberding, The rise of the
Carolingians and the Liber Historiae Francorum (Oxford, 1987). F. Theuws, Centre and
periphery in northern Austrasia, in: J. Besteman, J. Bos and H.A. Heidinga eds.,
Medieval archaeology in the Netherlands. Studies presented to H.H. van Regteren Altena (Assen,
1990), pp. 4169. F. Theuws, Frankish transformations. Actions and thoughts of aristocrats
and dwellers in the pagus Texandri (in prep.).
 159

M (fig. 1)

Centres of power such as Maastricht derived their significance from

the complex interaction of a varied group of social actors, and from
the interaction of these actors with the supernatural world and their
physical environment.8 Until now, scholarly attention has focused on
the dynamics of social interaction and, to a lesser extent, on the

Cologne

Tongres

Saint Servatius
WIJCK
castel
lum

Jeker
0 200m
Meuse

1 2 3 4 5

Fig. 1. Maastricht in the Merovingian period, with castrum and basilica of St

Servatius (after Panhuysen 1986 and Dijkman and Ervynck 1998). 1. Roman
road from Tongres to Cologne; 2. possible roads or tracks; 3. inhabited area
with artisanal activities; 4. cemeteries; 5. reconstructed course of the Meuse.

8
I realise that this is a modern categorisation of the cosmological order.
160

relations with the supernatural. The physical setting, or more pre-

cisely, the cultural landscape inside and outside the centres, has
mainly been perceived as either a backdrop to or a product of power
relationships. While often recognising the symbolic nature of some
of these contexts, scholars have tended to view them in terms of sta-
tic representations of power. Much less attention has been given to
other questions such as how the power of geography functioned and
how the landscape was read, interpreted and negotiated upon by
different social groups and actors in early medieval places of power.9
Dominant ideas and values represented in monuments and the land-
scape may well have differed significantly from the interpretations of
those who read the monuments and the cultural landscape.10 This
complicates any analysis of the interaction between social actors, the
supernatural world and cultural space. What complicates matters still
further is the fact that the actors involved did not always distinguish
these categories (social actors, the supernatural and cultural space)
so clearly. I will deal below with some of the actors relevant to
Maastricht and comment on aspects of their presence.

M, M

The bishops of the civitas of the Tungri were very peripatetic. In the
late Roman and Merovingian period, they are believed to have
resided in Tongres, Maastricht and Lige. But what did residing
entail? The Merovingian period yields few precise indications of the
location of a sedes episcopalis. There are only indirect allusions to the
episcopal residence being associated with the church of Our Lady
in the castrum in late seventh-century Maastricht and we have no
explicit information about Tongres as the location of the sedes epis-
copalis in late Antiquity, and the data about the new centre of Lige
(c. 700) dates from the late eighth century. Some scholars are inclined
to include Huy, Dinant, Namur and the basilica of St Servatius, all
of which have been viewed as secondary residences of the bishops.

9
Recent research has addressed these problems. See for instance G.P. Brogiolo
and B. Ward-Perkins, The idea and ideal of the town between late Antiquity and the early
Middle Ages, TRW series 4 (Leiden, 1999). See also Innes in this volume.
10
H. Moore, Space, text and gender. An anthropological study of the Marakwet of Kenya
(Cambridge, 1986), pp. 1668, 1838, 18995.
 161

As we shall see, these are precisely the places that force us to adopt
a very differentiated approach to the nature of episcopal power in
the diocese of Tongres/Maastricht/Lige. Such an approach would
help resolve the tension between the ideal of a cathedra episcopalis with
a specific and stable location on the one hand, and the actual early
medieval dynamics of episcopal power on the other.
The often highly speculative chronology of successive transfers of
the episcopal see and their possible motives has attracted consider-
able attention.11 Leupen and others distinguish between the official
seat of the bishop, i.e. the place where the sedes episcopalis was located,
and the bishops residence, which could have been in a different
place.12 However, the question as to why the bishops of Tongres/
Maastricht/Lige were repeatedly on the move has not yet been ade-
quately dealt with.13
The earliest medieval texts on Maastricht were written by Bishop
Gregory of Tours (d. 594).14 Gregory devoted just a few lines to the
saintly bishop Aravatius, who modern scholars usually identify as
Servatius.15 In his Histories, Gregory wrote: At that time in the fortified
township (oppidum) of Tongres there lived an extremely saintly bishop

11
A. van Berkum, Op zoek naar de meest geschikte residentie. Geruisloze ver-
huizingen van de Episcopi Tungrorum. Een poging tot reconstructie, in: C.G. de Dijn
ed., Sint-Servatius. Bisschop van Tongeren-Maastricht. Het vroegste Christendom in het Maasland.
Handelingen van het colloquium te Alden Biesen (Bilzen), Tongeren en Maastricht 1984 (Borgloon-
Rijkel, 1986), pp. 14762. Leupen, Het gelijk van Sint-Servaas en Onze-Lieve-
Vrouwe. J.-L. Kupper, Archologie et Histoire. Aux origines de la cit de Lige
(VIIIe XIe sicle), Ontstaan en vroegste Geschiedenis van de middeleeuwse steden in de zuidelijke
Nederlanden. Een archeologisch en historisch probleem (Brussels, 1990), pp. 37789. T.A.S.M.
Panhuysen and P.H.D. Leupen, Maastricht in het eerste millenium. De vroegste
stadsontwikkeling in Nederland, Ontstaan en vroegste Geschiedenis, pp. 41155, esp.
419, 4303, and the discussion on p. 415. R. De La Haye, De bischoppen van Maastricht
(Maastricht, 1985), pp. 223 and 8597.
12
Panhuysen and Leupen, Maastricht in het eerste millenium, pp. 4301.
13
Part of the problem is of course the lack of evidence for crucial moments or
periods in the history of Tongres, Maastricht and Lige. I intend later to specify
the lacunae in our knowledge that largely determine the outcome of the debate.
14
All texts relating to Maastricht from the period 359 to 923 have been cata-
logued by H.R. van Ommeren, Bronnen voor de geschiedenis van Maastricht
(3591204). Deel 1: Bronnen betreffende het tijdvak vanaf het jaar 359 tot en met
923, Publications de la Socit Historique et Archologique de Limbourg 127 (1991), pp.
548.
15
I adhere to this identification as there are no convincing arguments for sup-
posing them to be two different bishops. On Servatius in general see: H.C. Brennecke,
Servatius von Tongern. Ein gallischer Bischof im arianischen Streit, in: C.G. de
Dijn ed., Sint-Servatius. Bisschop van Tongeren-Maastricht, pp. 1736. R. De La Haye,
De bischoppen van Maastricht, pp. 1931. For a chronology of his life, see n. 21.
162

called Aravatius, who spent his time in vigils and fasting.16 As

Aravatius was afraid that the Huns would devastate Gaul, and his
prayers at Tongres were unable to prevent this, he decided to go to
Rome and pray at the tomb of the Apostle.17 There he received the
message from the Apostle to return home:
Go home quickly and put your domestic affairs in order, prepare your
place of burial and have ready a clean shroud. You are about to leave
this earthly life, and your eyes will not see the devastation which the
Huns will cause in Gaul, for that is the decision announced by our
Lord God.
Aravatius travelled back to Gaul. A crucial passage follows:
When he reached the town of Tongres, he quickly made all the prepa-
rations necessary for his own burial. He said goodbye to his clergy
and all the other citizens of the town, and then with tears and lamen-
tation he told them that they would never see his face again. They
wept and groaned as they walked behind him, and they addressed the
following humble supplication to him: Do not leave us, holy father!
Good shepherd, do not forget us! He gave them his blessing and the
kiss of peace, and then they went back home, for with all their tears
they could not turn him back. He made his way to the town of
Maastricht (ad Treiectinsem urbem) and there he fell ill of a mild fever.
As soon as his soul had left his body he was washed by the faithful
and buried beside the public highroad. How after the lapse of many
years, his holy body was moved elsewhere I have described in my
Book of Miracles.
This text has often been used as evidence for Servatius taking up
residence in Maastricht.18 Subscribers to this view assume that Servatius
preferred the protection of the small stronghold of Maastricht to that
of the large town of Tongres. The story about the devastation of
Gaul by the Huns has captured the imagination of many scholars,
but Aravatius/Servatius must have had a different reason for depart-
ing for Maastricht. Judging by Gregorys account, the bishop first
went to Tongres after returning from Rome. It was in Tongres where

16
Gregory of Tours, Historiarum libri decem, ed. B. Krusch and W. Levison, MGH
SRM 1, II5, pp. 457; transl.: L. Thorpe, Gregory of Tours. The history of the Francs
(Harmondsworth, 1982), pp. 1145.
17
The Apostle is not mentioned by name, but it is most likely St Peter. See
R. Van Dam, transl., Gregory of Tours. Glory of the Confessors, Translated texts for his-
torians, Latin series IV (Liverpool, 1988), p. 75.
18
De La Haye, De bisschoppen van Maastricht, p. 23. Van Berkel, Op zoek naar
de meest geschikte residentie, pp. 1478.
 163

he spent time in worship before going to Rome. According to Gregory,

Tongres is also the place where the clergy and the faithful resided.
If we follow Gregorys version of events, there is no doubt that
Tongres, centre of the civitas, was the location of the sedes episcopalis
as well. It was an episcopal city where Aravatius/Servatius made
preparations for his burial and then, somewhat surprisingly, left for
Maastricht. There is no evidence that this move was motivated by
fear of the Huns, or by a desire to move the episcopal residence.
The bishop left for Maastricht to die and be buried there; that is
all that we can deduce from Gregorys narrative. To Gregory, it was
clear that Tongres was and would remain the episcopal city. Ara-
vatius/Servatius departed for Maastricht to die and be buried,
most likely in a family grave. Maastricht therefore appears to have
been his familys seat of power, the place where his relatives lived
and his ancestors were buried. It was not at all unusual for an early
medieval bishop, who was most certainly a member of an impor-
tant regional aristocratic group, to be buried in a familial context.19
By moving from Tongres to Maastricht, Aravatius/Servatius must
have acted in accordance with the interests of the aristocratic group
to which he belonged. Such behaviour, however, falls outside the
scope of the institutional perspective and its rigid separation between
the bishop and his role as a member of an aristocratic group. Thus,
the episcopal office has become the focus of attention, and Avaratius/
Servatius move to Maastricht has been interpreted in institutional
terms: moving the see and seeking refuge elsewhere. But this would
suggest that Avaratius/Servatius left his flock behind without pro-
tection, an unforgivable act for any holy man, certainly in Gregorys
view. We must look further, and elsewhere, for the answers. An epis-
copal scion of an aristocratic family was important to his clan, espe-
cially at the time of his death and burial. Gregory describes the
departure for Maastricht as a separation from the clergy of the dio-
cese: They would never see his face again. Was Avaratius/Servatius
then buried within a family context, but without any reference to
his episcopal status? Gregory is silent on this point, but it seems
highly unlikely. The funeral of an early medieval bishop would have
been an ideal opportunity for representing his dual status as both
bishop and aristocrat. Apart from the family, the clergy of the diocese

19
G. Scheibelreiter, Der Bischof in Merowingischer Zeit (Wien, 1983), pp. 2457. For
further examples, see the contributions by Wood and Le Jan in this volume.
164

were likely to attend such important burial rites. These rites created
a link between a specific place that of episcopal death and bur-
ial and different actors with a vested interest in being associated
with the saintly bishop. Servatiuss grave represented the power of
a family that staked its claim against others competing for episcopal
power.20
Aravatius/Servatiuss death is difficult to date. We only know that
he was still alive in 359.21 Not until the beginning of the sixth cen-
tury does a new and securely documented bishop re-emerge upon
the scene.22
What happened in the meantime to the bishopric and to Aravatius/
Servatiuss grave? The texts are silent and so, to a large extent, are
the material sources.23 It should be stressed that not a single cult
place emerged at Tongres, outside the walls or at one of the large
cemeteries.24 This clearly indicates that no successors to Aravatius/
Servatius were buried there, for it is inconceivable that an episco-
pal town that lacked a saint would not make the most of an epis-
copal grave. Maastricht also had no cult of a fourth- or fifth-century
bishop who was a worthy successor to Aravatius/Servatius. In my
view, this indicates that, after Aravatius/Servatius, there were no
bishops present in Tongres or Maastricht until the start of the sixth
century. This tallies with the fact that only a few early medieval
cemeteries from the Meuse valley and adjacent areas contain graves
dating from the second half of the fifth and the first half of the sixth

20
I thank Prof. P. Leupen for discussion on these matters.
21
The date of 384 has been suggested by the Bollandist Henschen. The only
certainty we seem to have is a terminus post quem of 359, which is the date of
the Synod of Rimini where Servatius was a participant. Brenecke, Servatius von
Tongern, p. 18 and p. 27. If Gregorys Aravatius is indeed the same bishop as
Servatius, Gregory obviously places him in the wrong century for the Huns sacked
Gaul in the fifth, not in the fourth century. See Van Dam transl., Gregory of Tours.
Glory of the Confessors, p. 75. See also R. De La Haye, In welke eeuw leefde Sint
Servaas?, De Maasgouw, tijdschrift voor Limburgse geschiedenis en oudheidkunde 113 (1994),
cols. 528, who places Aravatius/Servatius in the fifth century.
22
De La Haye, De bisschoppen van Maastricht, p. 39: Falco, an enigmatic bishop;
p. 401: Domitianus, two dates relate to this bishop: 535 (Council of Auvergne)
and 549 (Council of Orleans).
23
T.S.A.M. Panhuysen, Wat weten we over de continuiteit van Maastricht?,
in: C.G. de Dijn ed., Sint-Servatius, bisschop van Tongeren-Maastricht, pp. 12546, esp.
p. 126.
24
W. Vanvinckenroye, Tongeren Romeinse stad (Tielt, 1985), pp. 12532. G. de
Boe, De archeologische getuigen van het eerste christendom in de Civitas Tungrorum,
in: C.G. de Dijn ed., Sint-Servatius, bisschop van Tongeren-Maastricht, pp. 3762.
 165

century, pointing to a phase of decline and to limited resources and

population numbers.25

Servatius grave

It comes as no surprise then that there is little activity associated

with Aravatius/Servatiuss grave during this period. Notwithstanding
the subsequent historiography which has glorified Servatius, both the
grave and the cult must have had a very low profile well into the
middle of the sixth century. In his Glory of the Confessors, Gregory of
Tours devoted an intriguing passage to Aravatius/Servatiuss grave:26
Aravatius is said to have been bishop of Maastricht during the time
of the Huns, when they burst out for the invasion of Gaul. He is said
to have been buried next to the bridge of the public road. Although
snow fell around his tomb, it never moistened the marble that had
been placed on top [of the tomb]. Even when these regions were
gripped in the cold of the excessive frost and snow covered the ground
to a thickness of three or four feet, the snow never touched his tomb.
One might understand that Aravatius was a true Israelite. For as the
Israelites [passed] between the walls of water, the water was an indi-
cation not of danger but of safety; and now the snow that fell around
the tomb of this just man was an occasion not of moisture but of hon-
our. Around the tomb you might see mountains of snow heaped high,
but they never touch the edge of the tomb. We do not marvel when
the ground is covered with snow, but we do marvel that it did not
dare to touch the spot of the blessed tomb. Many times the devotion
and zeal of believers have constructed an oratory (oratorium) from wood
planks that had been planed smooth; but immediately the planks either
are snatched by the wind or collapse of their own accord. And I believe
that this continued to happen until someone came along who con-
structed a worthy building in honour of the glorious bishop. After some
time passed Monulf became bishop of Maastricht. He built, arranged,
and decorated a huge church (templum magnum) in honour of Aravatius.
His body was translated into this church with great zeal and venera-
tion, and it is now distinguished with great miracles.27

25
F. Theuws and H. Hiddink, Der Kontakt zu Rom, Die Franken, Wegbereiter
Europas. Knig Chlodwich und seine Erben (Mainz, 1996), pp. 6680.
26
Gregory of Tours, In Gloria confessorum, ed. B. Krusch, MGH SRM 1 (Hannover,
1885), c. 71, p. 340. Van Dam, Gregory of Tours. Glory of the Confessors, pp. 7576.
It would be interesting to know why Gregory thought Aravatius/Servatius worthy
of attention. As Ian Wood has pointed out, Gregory was highly selective, favour-
ing some saints and leaving out others (see Wood this volume).
27
Van Dam, Gregory of Tours. Glory of the Confessors, pp. 756.
166

It would seem that Aravatius/Servatiuss grave was situated in the

open air. But can we take the text at face value? Historiography has
until now neglected this particular aspect and concentrated instead
on the later part of the text (the passage that mentions the building
of a large church by bishop Monulphus in honour of Servatius),
which is seen as a precise description. The church replaced a wooden
oratory whose planks would not stay in place, suggesting that there
was little interest in Aravatius/Servatiuss grave until bishop Monulphus
built his magnum templum. Nevertheless his tomb was remembered,
and a humble wooden oratory was kept up. By whom? Recent
research has shown that it was not only pious Christian communi-
ties and episcopal institutions which promoted saints cults. It was
also a concern of local family politics and aristocratic networks.28
The same may be true of Aravatius/Servatiuss grave. His own fam-
ily or aristocratic group may have promoted his cult, however lim-
ited. But should we accept Gregorys account of the humble nature
of the grave architecture?
Let us turn to what archaeology can tell us about the fate of
Aravatius/Servatiuss grave. Excavations by the town archaeologist
in the basilica of St Servatius have revealed a number of late antique
and early medieval structures that are difficult to interpret. The sit-
uation is complicated still further by the fact that the remains of the
eastern parts of subsequent churches were all destroyed when vari-
ous crypts were built in the later Middle Ages. The excavator is con-
vinced, however, that he has rediscovered the remains of the cella
memoriae connected with Servatiuss grave: a small rectangular build-
ing measuring 4.3 m. by 3.9 m.29 Yet there is no clear evidence to sub-
stantiate this claim. The dating is based on two coins of Theodosius I
(379395) found on the concrete floor of another structure discov-
ered immediately to the south, but we do not know whether this
structure was erected in the same period.30 The subsequent history

28
See Wood this volume.
29
T.A.S.M. Panhuysen, Die Maastrichter Servatiuskirche im Frhmittelalter. Ein
Vorbericht ber die jngsten Grabungen des stdtisches Amtes fr Bodendenkmalpflege
Maastricht, Kunstchronik. Monatschrift fr Kunstwissenschaft, Museumswesen und Denkmalpflege
43 (1990), pp. 51433. T.A.S.M. Panhuysen, De Sint-Servaaskerk te Maastricht
in de vroege middeleeuwen. Voorlopig eindverslag van de opgravingen door de
dienst Stadsontwikkeling Maastricht in de periode 19811989, Bulletin KNOB 1991,
pp. 1524.
30
I will deal with this structure later. It is difficult to evaluate such a coin find
 167

of the small structure and its relation to later buildings is somewhat

unusual. No other building activities are recorded on the site until
the sixth century, when a church was built. The remains of its west-
ern part have been found, but these cannot be dated with any pre-
cision either. All we can say is that the new church was built after
the end of the fifth century or later. Panhuysen identifies this build-
ing as Monulphuss templum magnum, mentioned by Gregory of Tours
and built in the second half of the sixth century.31 Curiously, the
new building left the cella memoriae untouched. If this were indeed
Servatius cella memoriae, we would expect it to have been incorpo-
rated into the new building. Servatiuss body may have been trans-
ferred from this cella memoriae into the new church, as Gregorys
account seems to indicate.32 If so, it is still surprising that the cella
memoriae was left intact, standing almost directly against the western
wall, unattended until it was reused. Traces of a sarcophagus, placed
there in the sixth or seventh century, have been recovered. The cella
was demolished in the course of the seventh century, when the new
church was enlarged in a westerly direction.33 The main argument
for identifying the small building as the cella memoriae connected with
Servatiuss grave is its location. One corner of its foundations was
destroyed when a new confessio and grave chamber were built in the
eleventh century. This superposition is the main argument for iden-
tifying the small building as Servatius cella memoriae. However,
Panhuysen may well have fallen into a trap set for him by the clergy
of the eleventh century who, while perhaps digging for Servatius
grave, found these foundations and defined them as such.34 It is
difficult to know why they would have built the new confessio and
grave chamber on this very spot. There is no continuity from the
early cella memoriae to the eleventh-century grave chamber and con-
fessio. Excavations of the Carolingian basilica have yielded no archi-
tectural structures in this particular area that might be remotely
connected with Servatius grave. This is remarkable considering the

as the coins may have been circulating for a long time. The exact date of the floor
and the cella memoriae remains a moot point.
31
Panhuysen, De Sint-Servaaskerk in de vroege middeleeuwen, p. 19.
32
See n. 25.
33
The interpretation of this western extension as a west chancel seems very hypo-
thetical to me.
34
Panhuysen himself regarded this as one of the possibilities in a preliminary
report: T. Panhuysen, De archeoloog, De Sint Servaas. Tweemaandelijks restauratie-
informatiebulletin 3940, (1988), pp. 30912, esp. 312.
168

importance of the cult of Servatius in Carolingian times. Obviously,

the focus of the veneration of the saint was located elsewhere. The
cella memoriae that has been unearthed may have been one of several
already present in the late antique cemetery. Until now, we had
known of no similar examples of cellae, but this is hardly an argu-
ment for ascribing this one exception to Servatius. Only a limited
area of the cemetery, located at some distance from the Roman road
where we might expect such constructions, has been excavated to
date (see fig. 2). We should bear in mind that the original grave of
Servatius might be located at the eastern end of the church built
after the fifth century.

Tongres THEATER
Roman ro
ad

Colog
+ ne
+ ++
+
+
+ + cloister Slope

++ VRIJTHOF

+ + +
+ + +
+
++ + +
+
SINT SERVAAS
KLOOSTER
BASILICA OF SAINT SERVATIUS

SAINT JOHN

0 50 m

Fig. 2. Maastricht. The basilica of St Servatius and excavated areas (white: Theater,
Vrijthof, Basilica of Saint Servatius, Sint Servaasklooster). The locations of the ceme-
tery surrounding the basilica and the Vrijthof cemetery, as they surfaced in the
excavation trenches, are indicated by crosses. In black: structures from the tenth
and eleventh centuries and later (after various publications by Panhuysen and
Bloemers 1973).
 169

Another surprising element is the so-called piscina located to the

south of the cella memoriae. This piscina has been identified as such
on the basis of a small fragment of concrete floor bearing the imprint
of a lower step belonging to a stairway that has itself disappeared.
The excavator has subsequently termed the piscina a baptismal font,
and the building around it a baptisterium; a slightly raised floor band
running along the wall should lend support to this hypothesis. But
is this a sufficiently strong argument to reconstruct a baptisterium next
to a small cella memoriae, which has wide-ranging consequences for
the interpretation of the site? Such constellations (baptisteria in bur-
ial grounds) do exist, and baptism ad sanctos is a known feature, but
it is a highly unusual phenomenon in Gaul, and an alternative inter-
pretation will have to be found to interpret the floor.35

35
For the combination of baptisteria and memorial churches, see the extensive
reviews by J. Christern of works on early Christian cult places in Spain in the
Bonner Jahrbcher 148 (1984), pp. 75666. It is certainly not so that baptisteria only
occur near large episcopal churches. However, the examples given by Christern are
located in the Mediterranean and in Rome and include the memorial basilicas in
heremus such as the one at Qalat Siman in northern Syria. In some cases the devel-
opment is reversed, in that a saints cult with a related infrastructure is created at
a church with a baptisterium. According to Christern, the baptistries near memorial
churches were not intended for the local population but for pilgrims visiting the
saints tomb. If this is the case, a possible baptismal font near the possible grave
of Servatius cannot be used as proof that St Servatius basilica was the episcopal
church. Similar doubts were raised with regard to one of the structures found under-
neath St Severins basilica south of Cologne. Pffgen wondered whether the small
polygonal building in the cemetery, dating from the fourth century, might have
been a baptistry. After evaluating all the evidence he had to conclude that noth-
ing supported this proposition, and that it was in fact a two-storey grave monu-
ment (B. Pffgen, Die Ausgrabungen in St Severin zu Kln (Mainz, 1992), pp. 906). In
the lower chamber of this building the floor was almost entirely broken out, although
small parts remained. If I understand Pffgens description correctly, there is a band
running along the wall, comparable to the one found in Maastricht. Pffgen inter-
prets it as a band designed to prevent water from reaching the foundations from
inside. Water may have streamed in from above as the lower chamber was dug
into the subsoil (Pffgen, Die Ausgrabungen in St Severin zu Kln, p. 92). The floor of
the piscina in Maastricht seems to be rectangular (Panhuysen, De Sint-Servaaskerk
te Maastricht in de vroege Middeleeuwen), whereas the normal shape for a piscina
in a baptistry is polygonal or round. The three volumes of Les premiers monuments
chrtiens de la France (Paris, 19951998) make only two references to baptistries with
rectangular basins, one in Reims and the other in Pianottoli-Caldarello in Sicily
(see the respective entries by Neiss and Berry and Duval under these place names
in volumes 3 and 1 respectively). Also, the reconstruction implies that a piscina for
baptisms is present at places where there is no proper church. Furthermore, the
baptistry has no further architectural history and was already destroyed in the sixth
or seventh century.
170

There is more. The piscina and baptistry were thus built at a time
when the cult of Servatius was not very prominent, and when the
region had a comparatively small population. Could the so-called
piscina conceivably have been part of another cella memoriae that was
partly dug into the subsoil so that the floor could only be reached
by descending a flight of steps? Similar structures are quite common
in late antique cemeteries.36
We can now give two answers, depending on the possible loca-
tion of the grave, to our initial question about the fate of St Servatiuss
grave. If it was located in the eastern part of the later churches, we
know nothing about the graves original architecture, nor about any
building activity prior to the construction of the new post-fifth-
century church (possibly Monulphus magnum templum). If, on the other
hand, the cella memoriae bore some relationship to Servatiuss grave,
we must conclude that nothing happened in the period between the
construction of the original memoria and of the new church. Although
we cannot be certain about building activities (which cannot have
been extensive), the archaeological remains seem to fit Gregorys
description: Servatius grave architecture had a very low profile. The
grave only received attention from Servatius family or a few local
admirers from Maastricht. He was apparently not regarded as the
central saint of the civitas until the middle of the sixth century.
This supposition is confirmed by another observation. According
to Gregory of Tours, Bishop Monulphus built a magnum templum in
honour of Servatius, a fact that has held historians and archaeolo-
gists spell-bound. Yet Monulphus is not the first sixth-century bishop
we know of; he is the third. This indicates that the two previous
bishops paid no attention to the cult of Servatius or, at least, their
interest did not take the form of building activity around his grave.37
Once more, this suggests that Servatius was not considered the
official saint of the diocese in the first half of the sixth century.
Monulphus was the first to honour him, and it is no coincidence
that he and Gondulphus, who is always associated with him, are the
only Merovingian bishops to be buried near the grave of Servatius.38

36
It may well be that both the cella memoriae and the floor with a step belong to
the same building which had a sunken floor and two subsequent rooms, i.e. an
entrance room and a second room where the dead were placed.
37
Once again, we cannot say with certainty that Servatius grave was on the
east side.
38
We cannot be sure where they were buried in the sixth century. The later
 171

The lack of interest shown by the two previous bishops and the sud-
den interest of Monulphus and Gondulphus in Servatius can, I think,
best be explained by viewing these men as members of an aristo-
cratic group intent on promoting their saint. This may be an exam-
ple of the planned development of a saints cult in relation to the
interest of a specific group. This correspondence of episcopal and
local group interests would explain matters more satisfactorily than
the institutional perspective which stresses the episcopal role. The
latter perspective has led to the conclusion that the basilica of St
Servatius was an episcopal Eigenkirche, a major episcopal church (also
serving as a baptismal church and parochial church) in Maastricht,
where the bishops occasionally resided as well.39 To my mind, this
is too institutional a view of the status and function of a church that
was first and foremost a cemeterial basilica. Once again, all depends
on how we deal with concepts such as bishop and episcopal prop-
erty. I would argue that the current approach to the churchs func-
tion and the episcopal office is overly determined by modern,
Carolingian or even Roman perspectives on Merovingian bishops
and their property. The nature of episcopal property should of
course be interpreted within a medieval frame of reference, but this
was a dynamic one; the concept of episcopal property may well have
been transformed in the Carolingian period. It is worth stressing that

cenotaph is outside the Merovingian church. A Merovingian double grave has been
found underneath the cenotaph (Monulphus and Gondulphus?). Panhuysen, De
Sint-Servaaskerk te Maastricht in de vroege middeleeuwen, p. 21.
39
This is Leupens conclusion, Het gelijk van Sint-Servaas en Onze-Lieve-
Vrouwe, p. 36. He considers the church as an episcopal Eigenkirche, a concept
that is difficult to apply in these fluid situations. That this is the episcopal church
par excellence is also inferred from his conclusion that the church of our lady in
the castrum did not obtain this status until the late seventh century. Although it is
generally accepted that no episcopal church remained in Tongres, without the help
of excavations we cannot be certain. Even with the benefit of excavations, it would
be difficult to distinguish an episcopal church from another church. Leupens view
of the basilica of St Servatius as an episcopal Eigenkirche leads him to conclude
that a change of ownership took place, for in later times the Carolingians con-
trolled the basilica and monastery. However, a transfer would not have occurred if
the basilica were controlled by an aristocratic group (with or without the kings coop-
eration), such as the one Monulphus and Gondulphus belonged to (see below). In
the end, we have to admit that we have no clues as to who controlled the basilica
of St Servatius after Monulphus and Gondulphus around 600, or at any time before
the end of the eighth century when the Carolingians were in control. Nor do we
know how and when the Carolingians gained control. This is one of the great lacu-
nae that makes our reconstructions problematic.
172

the position of Merovingian bishops was a dual one, i.e. both aris-
tocratic and clerical. His Merovingian basilica of St Servatius should
therefore be viewed as an aristocratic Eigenkirche.

Episcopal graves

The bishops membership of aristocratic groups motivated their

actions, as we can also infer from the burial places of the Merovingian
bishops of the bishopric of Tongres/Maastricht/Lige. With the
exception of Monulphus and Gondulphus, not a single bishop was
buried in or around the basilica of St Servatius, or in the church
containing the sedes episcopalis in Tongres or Maastricht. Monulphus
may well have had family motives for wishing to be buried near the
grave of Servatius rather than official episcopal ones. Domitianus,
the bishop in office before Monulphus, was buried in Huy, possibly
his familys power base.40 The same may be true of the enigmatic
Bishop Perpetuus, of whom it is claimed that he was buried in
Dinant, where a cult of Perpetuus subsequently developed.41 According
to a much later source, the equally enigmatic Bishop Joannes Agnus
was buried in the church of the holy Cosmas in monte at Huy.42
Amandus, bishop for only three years, was buried in the monastery
of Elnone.43 Remaclus, of whom we do not know for certain that

40
For the burial of Domitianus at Huy, see: A. Dierkens, La ville de Huy avant
lan mil. Premier essai de synthse des recherches historiques et archologiques,
Ontstaan en vroegste geschiedenis van de middeleeuwse steden in de zuidelijke Nederlanden (S.L.,
1990), pp. 391409.
41
De La Haye, De bisschoppen van Maastricht, p. 49.
42
Huy is his home region, as he is said to have come from the nearby village
of Tihange. De La Haye, De bisschoppen van Maastricht, p. 51. See also Dierkens, La
ville de Huy avant lan mil, pp. 4045, for the debate on the status of this church
in Huy. Dierkens is correct in questioning the status of Huy and Dinant as sec-
ondary residences of the bishop of Tongres-Maastricht-Lige, for this qualification
does properly address the nature of episcopal power in the bishopric. Dierkens
believes the church of Our Lady in Huy was an episcopal church (p. 405). This
may be correct for a later period (Carolingian and later) but we should consider
another alternative for the Merovingian period, especially that of Domitianus, namely
that the church was connected with the aristocratic group of which Domitianus was
an important member, and that it gradually became incorporated into the episco-
pal sphere. I will come back to the relationship between aristocratic groups and
these centres in the Meuse valley later.
43
De La Haye, De bisschoppen van Maastricht, p. 54.
 173

he was bishop of Tongres/Maastricht, was not buried in Maastricht

but in the monastery of Stavelot/Malmdy.44 Bishop Theodardus
place of burial is not known, but later traditions maintain that Bishop
Lambert transferred Theodardus body to Lige.45 Lambert himself
was buried in a family burial ground south of Maastricht, near a
chapel dedicated to St Peter.46 His successor Hubert later transferred
his body to Lige. Hubert himself was buried in the newly-built
church of St Peters at Lige.47 These examples show that, in some
cases at least, family considerations influenced the choice of burial
location. The same may be true for all the bishops buried outside
Maastricht, and even for those (Monulphus and Gondulphus) buried
in or near the basilica of St Servatius. What we need to bear in mind
is that the burial of a bishop, an important member of an aristo-
cratic group, was not a matter of chance. Their funerals represented
different claims to power. As must have been the case with Serva-
tius grave, those of Domitianus and Perpetuus represented a dual
structure involving claims on the part of the aristocratic family
and the representation of episcopal power. It is worth considering
how such burials later helped to construct the connection of a
bishop to a given place, such as Huy, Dinant or the basilica of St
Servatius. The position of the basilica of Servatius may be excep-
tional, however, certainly when compared to other churches. It is
precisely for this reason that traditions have been kept alive, and
that we now know of these burials. They must have played a cen-
tral role in later claims by bishops to these places, especially when
the bishops were members of other competing aristocratic groups.
When assessing these claims, we must bear in mind the dynamic
nature of episcopal power, and the constant tensions between aristo-
cratic and episcopal elements in the constellation. Carolingian bishops

44
De La Haye, De bisschoppen van Maastricht, pp. 5860.
45
De La Haye, De bisschoppen van Maastricht, pp. 623. His translatio by Lambert
to Lige (not a cult place at that time) is inconceivable.
46
De La Haye, De bisschoppen van Maastricht, pp. 667. Gerberding, in The rise of
the Carolingians and the Liber Historiae Francorum, pp. 1256, considers Lambertus
to be an outsider, a courtier of Childeric II, who was appointed bishop from out-
side the region. We have to agree with Werner, Der Ltticher Raum in frhkarolingi-
scher Zeit, p. 243, that he was of local origin. Strong evidence for this is his burial
at Saint Peters church, believed to lie just south of Maastricht, where his father
was also buried.
47
De La Haye, De bisschoppen van Maastricht, p. 75. Werner, Der Ltticher Raum in
frhkarolingischer Zeit, p. 292.
174

probably differed from their Merovingian counterparts in construct-

ing their claims to places such as Huy, Dinant, Tongres and the
basilica of St Servatius. Our comparatively late sources on the epis-
copal nature of places like Huy and Dinant (the so-called Nebenresi-
denzen) most likely convey a Carolingian episcopal representation,
in which a discourse is created of episcopal Eigenkirchen such as Our
Lady in Huy. Most likely, the graves of these aristocrats/bishops
marked the beginning of an episcopal tradition, rather than that the
aristocrats/bishops were buried on existing episcopal property. After
all, the christianisation of the diocese was not merely the work of
clerics and missonaries, but just as much of aristocratic groups and
their networks.

Lige: a new Pippinid cult place

We now turn to another important instance of saint-making by a

bishop within a now familiar framework: the interests of an aristo-
cratic group. There is little doubt that in the early eighth century
Bishop Hubert wanted to create a new cult centre in Lige (24 km
south of Maastricht), and that for this reason he relocated Bishop
Lamberts remains to this villa.48 This action has been discussed with
reference to the transfer of the bishops see to Lige in the course
of the eighth century.49 Leupen proposes the following solution: the
bishops concentrated on Lige because the king was promoting the
(new) church of Our Lady inside the castrum of Maastricht, which
then led to a decline of the basilica of St Servatius, which Leupen
sees as an episcopal church. But if the basilica of St Servatius was
the main episcopal church in Maastricht, and Servatius the cham-
pion of the bishops, why did Hubert not take Servatius remains to

48
Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 280319. J.L. Kupper,
Sources crites: des origines 1185, in: M. Otte ed., Les fouilles de la place Saint
Lambert Lige (Lige, 1984), pp. 314. De La Haye, De bisschoppen van Maastricht,
pp. 8595, with in my opinion an incorrect interpretation of Werners conclusions
and incorrect conclusions regarding the date of the transfer of the see from Maastricht
to Lige in the years of Lamberts episcopate (691706). Kupper, Saint Lambert:
de lhistoire la lgende.
49
The date of the transfer and the motives are a matter of debate. Kupper,
Archologie et Histoire, p. 379. For the latest discussion see: Leupen, Het gelijk
van Sint-Servaas en Onze Lieve Vrouwe., pp. 345. It is uncertain whether there
was also a transfer of property (involving the basilica of St Servatius and the church
of Our Lady in the fortress) in the period 650750, as Leupen proposes.
 175

Lige instead of Lamberts in order to develop a new episcopal cen-

tre? Did the bishops no longer control the basilica of St Servatius?
Was it no longer in the hands of the aristocratic group to which
Hubert belonged, or did the king have some interest in it? Rather
than answer these questions, I will develop another line of investi-
gation. There is no reason to assume that Hubert was trying to
develop a new episcopal see, as has often been suggested.50 Most
likely he wished to create a new power centre with its own sacred
infrastructure,51 adapted to the needs of the aristocratic group then
in control, i.e. the Pippinids, whose sphere of influence by then
included Lige.52 This was why a new saint was needed, and the
murdered Lambert made an excellent martyr.53 The relocation of
Lambert and the rise of his cult went hand in hand with the rise
of a powerful aristocratic group.54

The bishops church in the castrum of Maastricht (fig. 3)

But what of the church of Our Lady in the castrum of Maastricht?

Not surprisingly, its status is the subject of intense debate. Some
think a church already existed there in the days of Servatius, and
that it was also the sedes episcopalis, with Servatius supposedly having
transferred the episcopal see to Maastricht. Leupen has suggested

50
See note 11 and the discussion of the problem in Kupper Saint Lambert. De
lhistoire la lgende, pp. 215. Kupper correctly opts for the second half of the
eighth century as the period in which the transfer of the see took place. See also
his Archologie et histoire: aux origines de la cit de Lige and the comments
of Genicot on his paper in La gense et les premiers sicles des villes mdivales dans les
Pays-Bas mridionaux. Un problme archologique et historique (Brussels, 1990), pp. 4502.
51
Kupper Saint Lambert. De lhistoire la lgende, p. 22.
52
It is not entirely clear when the Pippinid concentration of power developed
around Lige. Werner sees in the concentration of estates an old core of Pippinid
properties (Werner, Der Ltticher Raum in frhkarolingischer Zeit). Gerberding (The rise
of the Carolingians and the Liber Historiae Francorum) thinks that the development of
Pippinid power around Lige should be dated later i.e. the late seventh and early
eighth century.
53
We will see later that Lambert (and perhaps his family) adopted a rather inde-
pendent position in the aristocratic networks of the Meuse valley and vis--vis the
king.
54
See the section on Maastricht and the aristocrats below. We should not,
however, overlook the king. Although his role is not clear, he must have acted in
accordance with at least some of the interests of the aristocratic group. It is difficult
to disentangle the aristocratic networks in the Meuse valley in the late seventh and
early eighth century.
176

reconstructed course
of the Meuse

bridge
road
roman

0 20m

1 2 3 4 5 6

Fig. 3. Plan of the Roman fortress at Maastricht (after Panhuysen 1996). 1. observed
moat; 2. reconstructed moat; 3. foundations (fourth century and younger); 4. walls
(fourth century and younger); 5. possible location of Merovingian episcopal church;
6. reconstructed trajectories of walls.
 177

that the church (which may already have existed perhaps as a

small royal chapel) was promoted (perhaps in accordance with the
interests of the Pippinids) by the king (possibly Childeric II [672675]
and/or Clovis III [690/1694/5]) to become the new episcopal cen-
tre in the late seventh century. All this was to the detriment of the
basilica of St Servatius, which Leupen believes had until then been
the true episcopal centre in Maastricht.55
Contemporary textual sources reveal that by the late seventh cen-
tury the church of Our Lady in the castrum contained the sedes epis-
copalis: Hubert was consecrated here.56 But the question is, when did
it become so? We do not know the answer. In addition to the pos-
sibilities mentioned above the second half of the fifth century
(Servatius) and the second half of the seventh century (the king or
the Pippinids) there is a third possibility: the first half of the sixth
century, when bishops who were active in Maastricht as well as in
Tongres reappear in the sources. We should consider this possibil-
ity in the light of my earlier conclusion that there were no bishops
in the civitas of the Tungri after Servatius, and that the new bish-
ops preferred Maastricht to Tongres. As we have already seen, it
was Monulphus who showed interest in Servatius grave, not Perpetuus
and Domitianus. Was Monulphus the bishop who upgraded Maastricht
to be the episcopal see? If so, he not only created a magnum templum
over Servatius grave (perhaps controlled by his family), but he may
also have built a new episcopal church in the castrum, with the support
of the king who controlled the location. Alternatively, the king
in co-operation with or in opposition to an aristocratic group that
controlled the basilica of St Servatius may have created a new
church himself. If Monulphus upgraded Maastricht, where was the
episcopal church of Perpetuus and Domitianus located? Was it still
in Tongres, or was there no such church?
Archaeologists have not been able to shed any light on these ques-
tions. No remains of an episcopal church (with baptisterium and domus
episcopalis) have been found in the south-western corner of the fortress
at Maastricht, where such a complex might be expected. The rea-
son for this is obvious: the beautiful Romanesque church of Our

55
Leupen, Het gelijk van Sint-Servaas en Onze-Lieve-Vrouwe, pp. 357.
56
Leupen, Het gelijk van Sint-Servaas en Onze-Lieve-Vrouwe, p. 36, note 27.
The consecration thus took place after Clovis III granted immunity to the church
(see the section Maastricht and the Merovingian kings below).
178

Lady dominates this particular quarter of the town. Only the chance
find of two capitals in this area, one from the seventh century, points
to some building activity in the seventh century.57 Excavations to the
north of the church, inside the fortress, have revealed no habitation
layers dating from 400/450 to 700.58 But a new moat was dug around
the southern part of the fortress in the fifth or sixth century.59 The
latter date could match the one for the construction of a new epis-
copal church in this part of the fortress, as well as for the restora-
tion of the castrum by the bishop in co-operation with the king.60
This church must have been situated in the far south-western cor-
ner of the fortress, hard against the wall, as was often the case in
late Antiquity (fig. 3).61

57
Panhuysen, Wat weten we over de continuteit van Maastricht, pp. 12545,
esp. 132 and 13840.
58
Panhuysen/Leupen, Maastricht in het eerste Millenium, p. 430. As yet there
have been no publications on the excavations in the cloister of the Church of Our
Lady.
59
The date of this highly significant feature has not yet been confirmed. A sixth-
century date (vermoedelijk al in de zesde eeuw) is given in J.P.A. van der Vin
and T.A.S.M. Panhuysen, Romeinse en vroegmiddeleeuwse munten uit een
stadskernopgraving in Maastricht, De Beeldenaar 4 (1983), pp. 1215, esp. 122. Later,
a fifth-century date was given (vermoedelijk al in de loop van de 5e eeuw, maar
niet later dan de 6e eeuw): T.A.S.M. Panhuysen, Maastricht staat op zijn verleden
(Maastricht, 1984), p. 71. In 1986 the moat was catalogued as one of the sixth and
seventh century elements in Panhuysen, Wat weten we over de continuiteit van
Maastricht?, pp. 12546, esp. 132, whereas Van Lith suggests a date perhaps in
the fifth or at the latest, the early sixth century: S.M.E. van Lith, Late Roman
and early merovingian glass from a settlement site at Maastricht (Dutch south
Limburg), parts 1 and 2, Journal of Glass Studies 29 (1987), pp. 4759 and 30 (1988),
pp. 6276. Thus the age of the moat changes, without any accompanying publication
of coherent collections of material from different stratigraphic contexts to substan-
tiate the claims. The way in which the material has been presented to date does
not enhance our understanding of the dating of the different phases of the site.
60
Problematic here is that the excavations between Wolfstraat and Havenstraat
have revealed no trace of the moat on the western side of the castrum, north of the
gate, but only two fourth-century moats (T.A.S.M. Panhuysen, W. Dijkman and
R.A. Hulst, Archeologische kroniek van Maastricht, Publications de la Socit Historique
et Archologique dans le Limbourg 129 (1993), pp. 33580, esp. 3702). This may indi-
cate that the moat was only dug around the south-western corner of the castrum,
where the bishops church may have been built.
61
Panhuysen, Wat weten we over de continuiteit van Maastricht?, pp. 1389.
 179

A one-horse town 62

After the emergence of the cult of Servatius from the middle of the
sixth century onwards, and the construction of an episcopal church
(at whatever date), Maastricht increasingly became a bipolar centre
of power with the old castrum and the episcopal church on the one
hand and the cemeterial basilica of St Servatius on the other (fig. 1).
In this respect it is almost a copy of Tours, another bipolar town
with some distance separating the episcopal church and cemeterial
basilica.63 Like Tours, Maastricht was dominated by the cult of a sin-
gle saint. For that reason Leupen calls Servatius a mythical city
founder.64 In the more prosaic terminology of Wood, Maastricht, like
Tours, was a one-horse town.65 Yet there seems to be an impor-
tant difference between Tours and Maastricht. Tours was a small
one-horse town dominated for a long time by a single family, that
of Gregory of Tours.66 Other larger and more important towns were
the scene of competition between various aristocratic groups, with
several cults developing simultaneously.67 The evidence for Maastricht
points in yet another direction: that of a small one-horse town where,
together with a number of other places in the Meuse valley, aristo-
cratic families played out their intense rivalries. The best evidence
dates from the late Merovingian period when Bishop Lambert was
murdered in a conflict between rival aristocratic groups.68 His vita
recalls how he drew his sword when attacked in Lige. For the
hagiographer, this made him an acceptable hero. What better symbol
than the sword to characterise Lambert as bishop/aristocrat?69 He was
not the first bishop in the civitas to be murdered. His predecessor

62
See Ian Woods contribution in this volume.
63
H. Galini, Tours from an archaeological standpoint, in: C.E. Karkow, K.M.
Wickham-Crowley and B. Young eds., Spaces of the living and the dead: An archaeolog-
ical dialogue (Oxford, 1999), pp. 87105.
64
Leupen, Het gelijk van Sint-Servaas en Onze-Lieve-Vrouwe, p. 30.
65
See Wood this volume.
66
See Wood this volume.
67
See Wood this volume.
68
The historiography on Lambert and his relocation is voluminous. Recent works
include: Werner, Der Ltticher Raum, pp. 24174. De La Haye, De bisschoppen van
Maastricht, 6471. Kupper, Archologie et histoire, p. 378. Kupper, Saint Lambert.
De lhistoire la lgende.
69
Kupper, Saint Lambert. De lhistoire la lgende, p. 12. E. van Hartingsveldt,
De zeven zaken en de frankische bisschop, Revue Belge de Philologie et dHistoire 67
(1989), pp. 67794.
180

Theodardus suffered the same fate.70 Amandus, an outsider installed

by the king, was bishop for only about three years. It is said that
the clergy of Maastricht were hostile to him, or rather the local and
regional aristocracy from which the clergy were recruited. Werner
stresses that Amandus was not chosen from among the local aris-
tocracy, which to his mind signals the power of the Pippinid Grimoald
vis--vis the local aristocracy, who must have had their own candi-
date.71 Was Amandus a Pippinid candidate in a non-Pippinid cen-
tre who subsequently became involved in a fierce power struggle in
the Meuse valley? The evidence for such lite competition dates
mainly from the seventh century, but such struggles can hardly have
been a new phenomenon. They must have intensified with the
unprecedented attempts by the Pippinids to dominate the entire area,
first the Meuse valley, and then an entire kingdom. The fact that
the bishops were buried in different places support this view. The
cult of Servatius did not emerge naturally and gradually as the
result of the unrelenting piety of a faithful few. A more appropriate
image would be that of a three-stage rocket. The first stage was the
relatively low-profile creation and upkeep of his grave, probably by
the group or family Servatius belonged to; the second stage is asso-
ciated with Monulphuss efforts in the middle of the sixth century,
and the third stage relates to the rise of power of the Pippinids and
the growing influence of this group in Maastricht in the eighth cen-
tury. Each of these stages was driven by the interests of a specific
group of aristocrats.
In the final analysis, we can conclude that Maastricht was a small
one-horse town where the nature of episcopal power was largely
determined by the ever-present competition between aristocratic fam-
ilies for whom the occupation of the sedes episcopalis was just one ele-
ment in their control of the region.72 This aspect of episcopal power
is responsible for the dynamics and multi-focal history of the see of

70
De La Haye, De bisschoppen van Maastricht, p. 62.
71
Werner, Der Ltticher Raum in frhkarolingischer Zeit, p. 235.
72
Gerberding also points to this competition (The rise of the Carolingians and the
Liber Historiae Francorum), but locates it in Maastricht alone. On the basis of the
distribution of burial places of bishops, I suspect that we are dealing here with a
larger number of groups active in the Meuse valley. However, there is one notable
exception. Maastricht, Huy and Dinant are mentioned as burial places while Namur,
the power centre of the Pippinids, is absent. Does this indicate that the Pippinids
barely played a role in the competition for the bishops see?
 181

Tongres/Maastricht/Lige. We have to ask ourselves whether oppos-

ing concepts, such as wandering bishops/bishops with a fixed resi-
dence, missionary bishops/diocesan bishops, main residences/secondary
residences, and episcopal churches/private churches, are not in fact
modern constructions in an institutionally-oriented debate, which rep-
resents bishops as operating within a static, almost exclusively epis-
copal system. If we disregard the institutional perspective, we might
begin to understand why various bishops divided their time between
different places. The multi-focal activities may also be the reason
why the bishopric of Lige became so large. In the vitae of bishops
this expansion is generally ascribed to missionary activities, but it
also was the result of the geographical extension of lite networks.
Once more, we should conclude that we need more careful research
into the diffusion of Christendom through lay networks and into the
activities of the so-called clergy.73 This perspective also makes it
easier to understand why the see was moved several times, and
why Bishop Hubert was not so much moving the see as creating a
new power centre in Lige. Or perhaps we should say that this was
done by the Pippinids, who needed a new centre in a region beyond
Merovingian royal control. I agree with Leupen that the granting of
immunity to the church of Our Lady in the fortress of Maastricht
by Clovis III in the years 690 to 695 indicates of royal control.74
Leupen suspects that this immunity should also be interpreted as a
sign of Pippinid control, but I doubt this. Bishop Huberts activity
in Lige seems to tally with Pippinid initiatives, rather than with an
episcopal desire to have their own centre under their control. This
does not necessarily mean that the see was in fact moved; this would
depend on our definition of see.75 It would be helpful to know how
the various actors involved actually perceived the cemeterial basilica

73
See also Theuws, Frankish transformations, in prep.
74
Leupen, Het gelijk van Sint-Servaas en Onze-Lieve-Vrouwe, p. 36.
75
Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 3179. In Werners view,
there are hardly any political motives for Huberts actions. He was acting as bishop
of his own accord, wishing to develop the cult of Lambert that was then gaining
popularity. Werner also regards Lige as episcopal property, but there is no evi-
dence for this. He believes that Hubert was able to develop Lige because the bish-
ops exercised almost full control over it. This was indeed the case later, but we
should not assume that the same applied in the Merovingian period. It is highly
unlikely, and without precedent, for Bishop Hubert to have acted entirely on his
own, independent of the political constellation (his family, the king, the aristocrats
and the Pippinids).
182

of St Servatius, a church that ultimately became one of the pillars

of Carolingian power. All our scant evidence indicates that, towards
the end of the seventh century, the Pippinid family was unable to
win the glittering prize: the episcopal see, with the bishops church
in the castrum and the basilica of St Servatius outside.76 This means
that Maastricht had another opponent to Pippinid encroachment as
powerful as the aristocratic groups: the supposedly weak Merovingian
kings, who did have an impact on this remote part of their realm.77

Maastricht and the Merovingian kings

It is generally accepted that the Roman fortress in Maastricht, as

Roman fiscal property, was a prerogative of the Merovingian kings,
like other fortresses or state forests.78 Maastricht seems to have been
at least partly in the hands of Merovingian kings, for they honoured
it with their presence on several occasions.79 According to a decree
of King Childebert II, he and a number of optimates met in Maastricht
in 595,80 which implies that the infrastructure was sufficient to house
and feed a royal party of this nature. This in turn suggests that
Maastricht was fiscal property at the time. Records reveal another
royal visit in 669/670, by King Childeric II, when he issued a char-
ter for the monastery of Stavelot/Malmdy.81 His presence indicates

76
Anne-Marie Helvtius points out that the founding of private monasteries can
be considered an anti-episcopal act. It is then interesting to note that the Pippinids
were most active in this regard in the southern Meuse valley, a further indication
that they did not control the bishops see (see the section Maastricht and the coun-
tryside below).
77
For a different view on the weak kings, see I.N. Wood, The Merovingian king-
doms 450751 (London/New York, 1994), pp. 25572.
78
See Leupen Het gelijk van Sint-Servaas en Onze-Lieve-Vrouwe, p. 33. To
my knowledge there is as yet no adequate explanation as to how this change-over
worked and whether it went uncontested by local magnates. Cf. E. Ewig, Frhes
Mittelalter, Rheinische Geschichte in drei Bnden (Dsseldorf, 1980), p. 46.
79
Once again I rely here on the work of Van Ommeren, Bronnen voor de
geschiedenis van Maastricht (3591204).
80
Pactus legis Salicae, ed. K.A. Eckhardt, MGH LL 4 (Hannover 1962), p. 268:
(Similiter) Treiectum convenit ut . . .. Van Ommeren, Bronnen voor de geschiedenis
van Maastricht (3591204), p. 20. K.F. Drew transl., The laws of the Salian Francs
(Philadelphia, 1991), p. 157.
81
Van Ommeren, Bronnen voor de geschiedenis van Maastricht (3591204),
p. 22.
 183

that Maastricht was still firmly in the hands of the king. Late in the
tenth century Heriger recounts that the same King Childeric II was
present at Maastricht on another occasion and supported Saint
Landoaldus,82 although it is difficult to judge the reliability of such
detailed information provided three hundred years after the event.
Childeric II may have had a special interest in Maastricht. He had
local supporters in the family of Lambert, the later bishop. His par-
ents are said to have belonged to the presides, which could mean that
his father was the kings local representative.83 According to Lamberts
vita, which is one of the few texts favourable to King Childeric II,84
Lambert himself was closely associated with the king and his
entourage.85 Werner concludes that the relationship between Lambert
and Pippin II was not very close, confirming our impression that
the Pippinids did not control Maastricht at that time. Later, in the
680s, Pippin II seems to have associated himself more closely with
aristocratic groups centred around the Maastricht-Lige area.
Another important indicator of a Merovingian kings involvement
with Maastricht is the granting of immunity by Clovis III (690/691
694/695) to the church of Our Lady in the castrum at least, this is
what the author of the twelfth-century Vita Landiberti IV maintains.86
This is a rather late text, but is nonetheless accepted as reliable.87
The granting of immunity occurred in the period from 682/683, when
Lambert was again bishop, until his death sometime in the years
703 to 705. This royal act may have prevented the Pippinids from
taking control of the bishops see. Lambert, who was a somewhat

82
Herigerus, Translatio S. Landoaldi sociorumque eius, ed. O. Holder-Egger, MGH
SS 15, 2 (Hannover, 1888), pp. 599607, esp. 603. Van Ommeren, Bronnen voor
de geschiedenis van Maastricht (3591204), pp. 22 and 44.
83
See Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 2435.
84
Wood, The Merovingian kingdoms, pp. 2289.
85
For Lamberts position in the complicated political arena of that time, see
Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 25366. After the murder of
Childeric II in 675, Lambert lost the see of Maastricht. He may have belonged to
an aristocratic faction opposed to the Pippinids, but he may also have switched
allegiance to the Pippinids before Childeric IIs murder, and been deposed by
Wulfoald, a supporter of Childeric II. Later he is supposed to have belonged to
the Pippinid camp, who restored him to the position of bishop of Maastricht.
However, in later traditions Pippin II is held partly responsible for Lamberts mur-
der (see Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 1245).
86
Vita Landiberti auctore Nicolao, ed. B. Krusch, MGH SRM 6 (Hannover and
Leipzig, 1913), pp. 40729, esp. p. 411. Van Ommeren, Bronnen voor de geschiede-
nis van Maastricht (3591204), pp. 456.
87
Leupen, Het gelijk van Sint-Servaas en Onze-Lieve-Vrouwe, pp. 356.
184

fickle ally of Pippin II, seems to have followed his own, or his fam-
ilys, independent course. Like episcopal power, royal power in
Maastricht could not survive on its own, but was closely bound up
with the interests of aristocratic groups. This is unsurprising, as all
early medieval political/ecclesiastical power depended on complicated
networks, but it is nevertheless useful to bear in mind.88
The importance of the power base of the Merovingian kings in
Maastricht is also illustrated by the vastness of the royal fisc, recon-
structed by Hardenberg.89 It includes most of the later town as well
as a number of villages further to the west (fig. 4). Its western para-
meter is defined by the estates of the monastery of Munsterbilzen,
and the southern boundary by the property of the bishop. The fisc
thus comprises a large part of the fertile lands to the west of
Maastricht.90 If this reconstruction is correct and if it existed in this
form from early times, then most of the activities of bishops and
aristocrats with regard to Maastricht took place on royal property.
However, we cannot be certain that the fisc consisted of an unbro-
ken territory within its borders. It has been suggested above that the
basilica of St Servatius may have been controlled by aristocratic

88
On the relationship between aristocratic and royal power, see Gerberding, The
rise of the Carolingians and the Liber Historiae Francorum. However, the situation in
Maastricht may have been rather complicated in that different aristocratic groups
may have controlled different parts of the bipolar centre in changing combinations.
Ideally, an aristocratic group would control the basilica of St Servatius, the castrum,
the office of representative of the king and the episcopal see. I believe Lamberts
family achieved this, although there is nothing to indicate that they had anything
to do with the basilica of St Servatius. The family exercised power with the sup-
port of the king, a concentration of power that made Lambert such a controver-
sial figure, for it excluded other groups almost entirely. On the other hand, such
a situation may not have been ideal for the king, who then had to face a single
dominant group.
89
H. Hardenberg, De Maastrichtse Vroenhof , Miscellanea Trajectensia. Bijdragen
aan de geschiedenis van Maastricht uitgegeven bij gelegenheid van het 300 jarig bestaan van de
stadsbibliotheek van Maastricht 1662 31 juli 1962 (Maastricht, 1962), pp. 2953. We
should bear in mind all the uncertainties that accompany this type of historical-
geographical research. This approach tends to eliminate the geographical dynam-
ics occuring over time. Also, by adding up information from different periods in
order to reconstruct a territory, it tends to overestimate the size of the territory.
The Merovingian fisc may well have had a territorial structure that differs from
that of the Carolingian fisc.
90
It did not extend to the east of the Meuse river, where a trade and artisanal
centre probably developed. Late Merovingian pottery kilns have been found in the
southern part of this quarter (T.A.S.M. Panhuysen, W. Dijkman, R.A. Hulst,
R.G.A.M. Panhuysen, Opgravingen door het Gemeentelijk Oudheidkundig Bodemon-
derzoek Maastricht [GOBM] in het jaar 1991, Publications de la Socit Historique et
Archologique dans le Limbourg 128 (1992), pp. 25988, esp. pp. 26474).
 185

Demer

Munsterbilzen

MAASTRICHT

TONGRES

Jeker

Voer

Meuse

1 2 3

Fig. 4. The extent of the fisc of Maastricht as reconstructed by Hardenberg (1962).

The hatched line gives a crude idea of the course of the boundaries. 1. fiscal
property; 2. property of the abbey of Munsterbilzen; 3. episcopal property.

groups. It is impossible to determine the extent of the king or bishops

control, but some form of power-sharing is an attractive model for
understanding Maastrichts early history. However, both Childeric II
and Clovis III did not lose their hold on Maastricht vis--vis the
activities of local and regional aristocratic groups. It is for that rea-
son, I think, that the Pippinids decided to make Lige a centre of
their own. Aristocratic groups needed such centres with their own
sacred infrastructure, which were different from monasteries such as
Nivelles and Andenne that were created in the seventh century. This
is the assumption that underpins the next section.

Important lacunae in our knowledge of early medieval Maastricht

It might be useful to pause here and recall the gaps in our knowl-
edge of early medieval Maastricht. This would reveal the hypothet-
ical nature of our understanding of Maastrichts early medieval history.
186

The first major lacuna is the political situation surrounding the basi-
lica of St Servatius from the time of Monulphus activities (and per-
haps those of his family) in the mid-sixth century until control by
the Carolingians in the late eighth century, when a monastery was
attached to the basilica. We do not know when the Carolingians or
Pippinids gained control. Was the basilica already controlled by the
Pippinids under Pippin II, or did this not happen until after 716,
when Charles Martel could make his power felt in the region? Another
lacuna is the status of Lige at the time Hubert began to develop
it as a cult place. We do not know whether it belonged to the Pippi-
nids, to Alpaidas family (see below), to the bishops, to Huberts fam-
ily, or whether it was royal property.
A third lacuna is the location of the sedes episcopalis prior to the
end of the seventh century when it was in a church in the Maastricht
castrum. We do not know whether it was located in the basilica of St
Servatius, in the church in the Maastricht castrum, still in Tongres
of which we know so little or whether it had no fixed location
at all. A fourth lacuna is the status of Tongres from the death of
Servatius until the tenth century, when the bishops controlled a monas-
terium in the town. We do know that the king exercised some con-
trol over Tongres as well,91 but was the entire old town royal property?
Was there a Carolingian church, and if so, who controlled it? How
did the bishop regain control over the ruined old Roman city?
Perhaps because of a royal donation? There is an urgent need for
a new study of the early medieval history of Tongres up to the
eleventh century. A fifth lacuna is the status of the cult place at
Lige and its sedes episcopalis up until the eighth century. Was the
episcopal see already located there at the beginning of the eighth
century? Although these gaps in our knowledge prevent us from
arriving at firm conclusions, they must nevertheless be kept in mind.

The aristocratic groups active in the Meuse valley have been the
subject of extensive study,92 but the research has largely been ignored

91
H. Baillien, Tongeren. Van Romeinse civitas tot middeleeuwse stad (Assen, 1979), pp.
1527 and pp. 3547.
92
The standard reference work is Werner, Der ltticher Raum in frhkarolingischer
 187

by those studying power centres in the region. Or, more precisely,

these centres have not been incorporated as meaningful entities into
the history of aristocratic groups.93 The development of Merovingian
Maastricht has always been considered in relation to three other cen-
tres: Huy, Namur and Dinant. At first sight their development reveals
many parallels. Scholars have attributed this synchronous develop-
ment to their function as stations along the long-distance trade route
that ran from south to north and vice versa along the Meuse river.94
Recent criticism of this interpretation has centred on the lack of evi-
dence for long-distance trade, and on the position of these centres
some 30 kilometres apart the distance that could be sailed in one
day.95 Both Despy and Devroey have questioned the supposed par-
allel histories of Maastricht, Huy, Namur and Dinant. They suggest
that each place had its own pattern of development, which was not
necessarily synchronous with the others, and much more diverse than
has been assumed.96 However, we should question whether the
conflicting chronology of the sources does in fact reflect a different
pattern of development. The fact that the portus of Maastricht is
mentioned before 700 (in the Vita Landiberti from c. 735), that of
Huy not until 862 and those of Namur and Dinant only in the sec-
ond half of the tenth century does not mean that no boats landed
in Namur in the seventh century.97 Details may have differed, and
some elements may have been introduced earlier or later, but in
general these centres were very similar. Although we lack sufficient
archaeological data on Dinant, each centre sported a fortress, artisan
quarters, a main church dedicated to Our Lady, coin production
(for whatever purpose), an important river crossing, important burials

Zeit. Other works relevant to the study of these aristocratic groups are M. Werner,
Adelsfamilien im Umkreis der frhen Karolinger. Die Verwandtschaft Irminas von Oeren und
Adela von Pfalzel (Sigmaringen, 1982). A. Bergengrn, Adel und Grundherrschaft im
Merowingerreich (Wiesbaden, 1958). Gerberding, The rise of the Carolingians and the Liber
Historiae Francorum. R. Le Jan, Famille et pouvoir dans le monde Franc (VII eX e sicle).
Essai danthropologie sociale (Paris, 1995).
93
Elements of these places, however, such as the churches, have been studied in
relation to aristocratic groups.
94
Rousseau, La Meuse et le Pays Mosan, pp. 3781.
95
For a full discussion of these problems, see Devroey, Villes, campagnes, crois-
sance agraire dans le pays Mosan avant lan mil: vingt ans aprs, pp. 22550. See
also note 4. A. Joris, A propos de burgus Huy et Namur, Villes Affaires
Mentalits. Autour du pays Mosan (Brussels, 1993), pp. 13948.
96
Devroey, Villes, campagnes, p. 257.
97
Devroey, Villes, campagnes, p. 257.
188

(no evidence yet in Namur and Dinant) and possibly the presence
of fiscal lands (figs. 5 and 6).98 The parallels are such that we can-
not ignore them, and we have no need of economic arguments to
explain them. In each place, the seventh century saw the rise of
different aristocratic groups for whom the centre was an important
power base. The crucial elements in such a power base were the
following: one or more fortresses, landed property, a thesaurus, cult-
sites linked to the family, membership of an established noble fam-
ily, a retinue, a spouse from another powerful family, offices such
as that of bishop, count or domesticus, and proximity to the king.
Missing from this list, and therefore hardly mentioned in the litera-
ture, are the profits from exchange and trade, and the gains
both material and immaterial from artisanal production (unless
we assume a priori that aristocrats were not involved in these activ-
ities at all). Not all elements were necessarily present in all cases,
and others might well be added. The important thing to bear in
mind is that different aristocratic groups were probably active in the
power centres that boasted a castrum and a cult-site with, invariably,
Our Lady as its patron. With the exception of Dinant, we can iden-
tify seventh-century aristocratic families who turned these places into
their power base. Since the early seventh century, Namur was the
base of the Pippinids, who used its castrum as a stronghold.99 Huy
may have been connected with the family of Itta, the wife of Pippin
I: her power base lay in the Meuse valley between Namur and
Maastricht. Both the monasteries of Nivelles and Andenne may well
have been constructed on the property of her family, rather than
that of her husband.100 Finally, Maastricht was within the sphere of

98
Huy: Dierkens, La ville de Huy avant lan mil. Namur: A. Dierkens, Premiers
structures religieuses: paroisses et chapitres jusquau XIIe sicle, Namur. Le site, les
hommes de lpoque Romaine au XVIII e sicle (Bruxelles, 1988), pp. 3361. G. Despy,
Lagglomration urbaine pendant le Haut Moyen Age (du VIIe sicle aux envi-
rons de 1200), Namur. Le site, les hommes de lpoque Romaine au XVIII e sicle (Bruxelles,
1988), pp. 6378. See also the various contributions by Plumier and Plumier-Torfs
in: J. Plumier ed., Cinq annes darchologie en province de Namur 19901995 (Namur,
1996), esp. pp. 67115. Dinant: G. Despy, Lagglomration urbaine pendant le
Haut Moyen Age (du VIIe sicle aux environs de 1200), Namur. Le site, les hommes
de lpoque Romaine au XVIII e sicle (Bruxelles, 1988), pp. 6378. Dinant: Atlas du sous-
sol archologique des centres urbaines anciens. Dinant (Brussels, 1988), p. 3.
99
Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 41620.
100
Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 4014. Werner sees these
properties as proof of Pippinid presence throughout the central Meuse valley. I pre-
fer to differentiate between the properties of Pippin I and those belonging to the
 189

Fig. 5. Huy in Merovingian times with castrum, artisanal quarters, church of Our
Lady and cemeteries (after Dierkens 1990)

Fig. 6. Namur in Merovingian times with castrum, artisanal quarters, Church of Our
Lady and cemeteries (after Plumier 1996; 1996a; Plumier-Torfs and Plumier 1996)
190

influence of the aristocratic group to which Bishop Lambert belonged,

and possibly others competed for dominance as well.101 Gerberding
has correctly suggested that Alpaida, the second wife of Pippin II,
came from a family based in Maastricht and/or Lige.102
Lige and its environs probably entered the Pippinid orbit when
Pippin II married Alpaida, which would also explain why a new cult
centre was developed in Lige under Pippin II and Charles Martel,
Alpaidas son. According to Gerberding, Bishop Hubert, who cre-
ated the cult centre, had the full support of his ally Charles Martel.103
This is highly likely, but not I believe for the reasons Gerberding
mentions. Gerberding suggests that Hubert moved the episcopal see
to Lige in order to submit the bishop to Pippinid control.104 Yet
Huberts initiative was a two-stage one. Almost immediately after
Lamberts death (703/705), he constructed the basilica Sancti Landiberti,
which we know was finished before 714, as Pippin IIs son Grimoald
was murdered there.105 In other words, the creation of the new cult

aristocratic group from which Itta originated. Nivelles and Andenne might be Ittas
portion.
101
Gerberding, in The rise of the Carolingians and the Liber Historiae Francorum,
pp. 1256, regards Lambert as an outsider. But, as has been explained above, he
was a member of a very important local family.
102
Gerberding, The rise of the Carolingians and the Liber Historiae Francorum, pp.
11825. R. Gerberding, 716: A crucial year for Charles Martel, in: J. Jarnut,
U. Nonn and M. Richter eds., Karl Martell in seiner Zeit (Sigmaringen, 1994), pp.
20516, esp. 2047. Later sources suggest that Alpaida was the sister of Dodo, the
domesticus who killed Lambert, in which case it is unlikely that Alpaida and Lambert
belonged to the same family. The power base of Alpaidas family was most likely
Lige. On later traditions and their context, see the analysis by Kupper, Saint
Lambert. De lhistoire la lgende, pp. 2749. Gerberding, 716. A crucial year
for Charles Martel, pp. 206 and 214. I agree with Gerberding that Lige was not
part of the original power base of the Pippinids as Werner suggests: Gerberding,
The rise of the Carolingians and the Liber Historiae Francorum, pp. 1203.
103
However, Hubert started to create the centre while Pippin II was still in
power. The relocation of Lamberts relics to Lige took place in the year 716, 717
or 718. Werner sees no special motive, but there does seem to be a relationship
to Charles Martels rise to power. Werner, in Der Ltticher Raum in frhkarolingischer
Zeit, p. 307, is more cautious on the subject of cooperation between Pippin II and
Hubert: Dies lsst darauf schliessen dass Pippin II. der Schaffung eines mit dem
hl Lambert verbundenen religisen Zentrums in unmittelbarer Nachbarschaft seiner
bevorzugten Besitzungen Chvremont und Jupille wohlwollend gegenberstand und
sie sehr wahrscheinlich auch frderte, and a little further: drfte sich Hubert somit
auf das Einverstandnis Pippins II. Gesttzt und mglicherweise sogar dessen Interessen
entsprochen haben. Werner thus sees Huberts actions as an episcopal intitiative,
while I share Gerberdings view that they were a Pippinid activity.
104
Gerberding, The rise of the Carolingians and the Liber Historiae Francorum, p. 134.
105
Kupper, Sources crites: des origines 1185, p. 32. This murder related
the Pippinid family directly to Lambert.
 191

place was already well under way when Pippin II, as well as his
consecutive wives Plectrud and Alpaida, were still in power.106 The
actual removal of Lamberts relics took place when Charles Martel
had already taken over. Perhaps the Pippinids originally intended to
create a new centre of the type Dinant-Namur-Huy-Maastricht, i.e.
a centre with a castrum (Chvremont), a cult place (St Lamberts basi-
lica), artisanal quarters etc., but this time connected with Alpaidas
family. It is a tempting hypothesis. The rapid development of this
new centre must have been closely linked with Charles Martels vic-
tories, the direct cause of Lamberts translatio. Thus, Charles Martels
maternal family rose to great importance in a short time. Only later
the second stage was the bishops see moved to Lige. Chvremont
has never been associated with the creation of Lige and its sacred
topography because of its relatively distant location and the many
questions surrounding the chronology of its establishment.107 But was
there another hill near Lige suitable for building such a fortress? I
view Chvremont and Lige as two building blocks in a power cen-
tre typical of the Central Meuse valley, of the kind we have just dis-
cussed.108 Yet there was a major difference between the old centres
and Lige/Chvremont it was the only one beyond royal control.
I believe this was the reason why Lamberts remains, and not those
of Servatius, were moved to Lige. Servatius was associated with a
long history in which the Pippinids played no role. If we assume

106
It is uncertain whether the relocation of Lamberts relics thirteen years after
his murder was already planned from the beginning, as Werner suggests.
107
It probably dates from the early eighth century, with its construction con-
temporaneous with that of the cult-site at Lige. In the paper by A. Hoffsummer-
Bosson on the excavations in the fortress, no Merovingian pottery is mentioned or
illustrated. In the discussion on dating, she refers to the historical debate and
Werners ideas (A. Hoffsummer-Bosson, Chvremont: lapport des sources archolo-
giques, Bulletin de lInstitut Archologique Ligois 100 (1988), pp. 7187). If archaeo-
logical evidence could have established beyond doubt a seventh-century date for
the first construction activities, Hoffsummer-Bosson would surely have mentioned
it. It is referred to as Novum Castellum early in the eighth century. Werners wish to
see the Lige/Chevremont area as belonging to the Pippinid sphere of influence in
the first half of the seventh century makes him question this date of construction,
Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 41041. See also: M. Josse,
Le domain de Jupille des origines 1297 (Bruxelles, 1966) and the contributions in the
Bulletin de lInstitut Archologique Ligois 100 (1988). It would be worthwhile to carry
out new excavations in the fortress.
108
Here we should add the monastery of Saint Mary, created in the fortress of
Chvremont by Pippin II. Chvremont was clearly more than a fortress with a mil-
itary function. Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 42237.
192

that Bishop Hubert and the Pippinids did not wish to move the epis-
copal see, but were intent on creating a new aristocratic centre linked
with Alpaidas family, Lamberts relics were an excellent choice
indeed: a new saint, for a new centre.109 Only after the Pippinids
managed to break the aristocratic and royal hold on Maastricht did
they begin to control the basilica built over St Servatiuss grave and
the castrum with the bishops see.110 If this happened shortly after
Charles Martel rose to power, there was no immediate need to move
the episcopal see, but if it took longer, the move to Lige made con-
siderable sense. So who then was responsible for the move away
from a traditionally prestigious place? The creation of Lige/ Chvre-
mont shows that aristocratic groups needed composite centres such
as Dinant, Namur, Huy and Maastricht in their exercise of power,
centres that were different from monasteries such as Nivelles and
Andenne. The cult places of Maastricht (castrum), Huy, Namur, and
Dinant differed from the monasteries of Andenne, Nivelles and others
in that the latter contained the grave of a saint who was a mem-

109
A cynical element here is the possibility that Dodo, Lamberts murderer, was
a relative of Alpaida, in which case she and her family chose a martyr who was a
victim of their own family. But once again, this fits within our representation of
Lambert as a fickle ally of the Pippinids.
110
Here I wish to return to the problem of how the Pippinids/Carolingians
gained control of the basilica of St Servatius (which is not the same as gaining con-
trol of the episcopal see). I will present one of the many possibilities here. Surprisingly,
after Monulphus and Gondulphus, the bishops in the seventh and eighth centuries
do not seem to exercise control of the basilica. Although these two bishops were
buried in the basilica, no episcopal claims were made in Carolingian times, unlike
in Huy and Dinant where bishops were also buried. Is this because the basilica had
always remained firmly in the hands of an aristocratic family (e.g. that of Monulphus
and Gondulphus)? Were they able to control it for most of the seventh and early
eighth century? In Carolingian times it was again firmly in the hands of one fam-
ily: the Pippinids. The situation is thus as follows: just as the Pippinids exerted no
influence over the bishops see in the seventh and early eighth century, the bish-
ops exerted little over the basilica. Is that because bishops, like Lambert, came from
other families, centred on the castrum? The conclusion is that the Pippinids did
not gain control over the basilica via the episcopal route. That leaves a royal or
aristocratic route. However, if the basilica was controlled by an aristocratic group
the royal route would only be possible after a confiscation and donation by the
king of the basilica to the Pippinids. In that case it is astonishing that we hear noth-
ing of it. Maybe the route was a silent one. An interesting alternative could be
that Alpaidas family controlled the basilica. The Pippinids gained a strong influence
in the Maastricht-Lige area after the alliance of Pippin II with Alpaida. If Alpaidas
family controlled the basilica of St Servatius in addition to some estates around
Lige, we can understand Pippins interest in her and her family, for they provided
an important entrance to Maastricht.
 193

ber of the founding family. This seems not to be the case in cen-
tres with churches consecrated to the Virgin. Both central churches
and family monasteries had their place in the sacred infrastructure
of aristocratic families.

C, , ,

The social actors discussed so far the bishop, king and aristocrats
were not the only ones who determined the role and significance of
Maastricht, but they were without doubt the most powerful. Bishops
and aristocrats are difficult to distinguish: the Janus-like early medieval
aristocracy had both a lay and a clerical face. The king was clearly
a powerful factor in Maastricht until the end of the seventh century.
On the basis of admittedly patchy textual evidence, we now turn to
others who were important to social interaction in early medieval
Maastricht and who should not be overlooked. They are the clergy,
artisans (and builders), merchants and other inhabitants.
In the eighth century, a monasterium was attached to the cemeter-
ial basilica of St Servatius.111 Most likely the inhabitants had their
predecessors; it is difficult to imagine Christian Maastricht without
any clergy. The Life of Amandus recounts how, towards the middle
of the seventh century, the saint was driven out of Maastricht by
sacerdotes gentis illium.112 Who were these sacerdotes? Most likely they
were members of the aristocratic families who controlled the vari-
ous cult places in the Meuse valley and in the bishopric in general.
That clerics were a regular presence in Maastricht is also clear from
the account of Bishop Lamberts murder (c. 730). Lambert was sup-
posedly murdered in the company of priests and in a building con-
taining a dormitory for clerics.113 Although we know little of the
status of the Maastricht clergy, we sense their presence in the sources.

111
J. Deeters, Servatiusstift und Stadt Maastricht. Untersuchungen zu Entstehung und
Verfassung (Bonn, 1970), p. 14. Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp.
3223, n. 17.
112
This part of the vita seems to describe actual events, a conclusion that we
also draw from other texts. See Werner, Der Ltticher Raum in frhkarolingischer Zeit,
pp. 2313. De La Haye, De bisschoppen van Maastricht, p. 54.
113
Werner, Der Ltticher Raum in frhkarolingischer Zeit, p. 288. Kupper, Sources
crites: des origines 1185, p. 31.
194

To find out more about artisans in Merovingian Maastricht, we

have to turn to archaeological evidence. Several excavations in the
modern town have yielded evidence of glass-making, bone-working,
metal-working, pottery production, and coin production.114 In addi-
tion, there were building activities that required a skilled workforce.
The presence of these artisans must have been vital to the contem-
porary view of Maastricht as a true power centre.115
There is no direct textual or archaeological evidence for traders
and barge operators from the Merovingian period, but their presence
is evident from imports, some of which were building materials.
Despy and Devroey have emphasised the importance of trade, espe-
cially along old Roman roads, such as the one from Cologne, via
Maastricht to Tongres and beyond.116 However, most of the written
evidence dates from the Carolingian period, and from monks and
nuns who organised the transport of produce from outlying estates
to their abbey. We may assume that the volume of trade and trans-
port would have been considerably less at an earlier date in the far
north, before abbeys were built or when they owned less property.117
An interesting but enigmatic aspect of early medieval Maastricht
is the development of the quarter on the opposite bank of the Meuse.
Nowadays it bears the name Wijck, from the Latin vicus, often an
indication for a trading and artisanal quarter.118 Unfortunately, ero-

114
Panhuysen, Maastricht staat op zijn verleden, pp. 925. Y. Sablerolles, Made in
Maastricht: glaskralen uit de Merovingische tijd, Vormen uit vuur 155 (1995), pp. 1523.
W. Dijkman/A. Ervynck, Antler, bone, horn, ivory and teeth: The use of animal skeletal
materials in Roman and early Medieval Maastricht, Archaeologica Mosana 1 (Maastricht, 1998).
115
See elsewhere in this volume for Hedeagers contribution on Gudme and
Lundeborg, two sites that on a structural level may be highly comparable to
Maastricht.
116
Despy, Villes et campagnes aux IXe et Xe sicles: lexample du pays Mosan.
J.-P. Devroey, Les services de transport labbaye de Prm au IXe sicle, Revue
du Nord 61 (1979), 54369 (repr. in: J.-P. Devroey, Etudes sur le grand domaine caro-
lingien, (Aldershot, 1993), nr. X). Devroey, Villes et campagnes.
117
For trade and artisanal activities in general, see the recent work by Stephane
Lebecq: S. Lebecq, Pour une histoire parallle de Quentovic et Dorestat, in: J.M.
Duvosquel and A. Dierkens eds., Villes et campagnes au Moyen Age. Mlanges Georges
Despy (Lige, 1991), pp. 41528. S. Lebecq, Quentovic: un tat de question, Studien
zur Sachsenforschung 8 (1993), pp. 7382. S. Lebecq, Gregoire de Tours et la vie
dchanges dans la Gaule du VIe sicle, in: N. Gauthier and H. Galini eds.,
Gregoire de Tours et lespace gaulois. Actes du congrs internationale, Tours, 35
novembre 1994 (Tours, 1997), pp. 16976. S. Lebecq, Les changes dans la Gaule
du nord au VIe sicle, in: R. Hodges and W. Bowden eds., The sixth century. Production,
Distribution and Demand (Leiden/Boston/Kln, 1998), pp. 185202.
118
For discussion of this term, see: A.C.-F Koch, Phasen in der Entstehung von
 195

sion caused by the river has washed away most of the evidence.
It is not until the Vita Landiberti (c. 730) that we hear of a portus
at Maastricht,119 mentioned in passing in a description of Bishop
Lamberts murder thirty years earlier. His corpse was brought back
from Lige by boat. Whether the portus was situated in Wijck is not
certain, but it was clearly already in existence. Around 830, Einhard
explained in his Translatio et Miracula SS Marcellini et Petri that Maastricht
was a vicus, mainly inhabited by merchants.120
If Wijcks artisanal and trade quarter had already developed in
Merovingian times,121 and if the presence of pottery production in
its southern-most part is an indication of this early development, we
can make two interesting observations. First, there was a clear sep-
aration between this quarter and Maastrichts older centre, with its
fortress and cult place. In her contribution to this volume, Hedeager
describes craftsmen and traders as quintessential outsiders who
derived great prestige from this status. Did the same apply to
Maastricht? The second observation relates to the widely-held belief
that traders regularly operated under royal protection. Many of the
later Carolingian trade centres did enjoy some form of royal pro-
tection,122 but Maastricht seems to have been an exception. As we
have noted already, the royal fisc lay to the west of the Meuse. It
is impossible to reconstruct early medieval property relations east of
the river. Further to the south lay the estate of Breust, the late
medieval property of the chapter of St Martin in Lige, and prob-
ably episcopal or royal property at an earlier stage.123 No royal,
Pippinid or episcopal property has ever been identified immediately

Kaufmannsniederlassungen zwischen Maas und Nordsee in der Karolingerzeit,

Landschaft und Geschichte. Festschrift fr Franz Petri (Bonn, 1970), pp. 31224, esp. 318.
119
Vita Landiberti vetustissima, ed. B. Krusch, MGH SRM 6 (Hannover, 1913),
p. 371. Van Ommeren, Bronnen voor de geschiedenis van Maastricht (3591204),
p. 25.
120
Einhard, Translatio et Miracula SS Marcelini et Petri, ed. G. Waitz, MGH SS 15
(Hannover, 1887), pp. 23864, esp. 261. Translation: P.E. Dutton, Carlemagnes
courtier. The complete Einhard (Ontario, 1998), p. 122. Van Ommeren, Bronnen voor
de geschiedenis van Maastricht (3591204), pp. 312.
121
We cannot rule out the development of an artisanal quarter on the west bank,
though there is as yet no evidence for this.
122
Verhulst, The rise of cities in north-western Europe, p. 45.
123
J. Hartmann, De reconstructie van een middeleeuws landschap. Nederzettingsgeschiedenis
en instellingen van de heerlijkheden Eijsden en Breust bij Maastricht (10e-19e eeuw) (Assen/
Maastricht, 1986), p. 289.
196

to the east of Maastricht.124 This may be due to a lack of evidence,

but if an artisanal and trade quarter had developed in Wijck in
Merovingian times, this may have been precisely because of its inter-
mediary position between the various spheres of influence. It may
have been akin to Lundeborg in relation to Gudme, as Hedeager
suggests elsewhere in this volume.
By about 800, the situation had probably changed. Excavations
in Maastricht, Huy and Namur have revealed a significant new fea-
ture. Whereas there is considerable evidence of pottery, bone-work
and metal-work from the Merovingian era, comparable evidence,
although not entirely absent, is difficult to find for the Carolingian
period.125 The rather tenuous archaeological explanation for this curi-
ous phenomenon is that later activities destroyed earlier layers. But
this does not explain how a growing number of excavations have
yielded supposed identical patterns of destruction in three different
centres. It therefore seems highly unlikely, as does the notion that
all later destruction reached the precise depth of the Merovingian
layers.126 Recent excavations have been carried out in Namur at the
Grognon site and its surroundings, i.e. in the triangular area between
the Sambre and Meuse rivers, near the supposed location of the por-
tus on the Sambre. These show that in Carolingian times the ora-
tory of St Hilarius was situated on top of the previous artisanal
quarter, which would suggest that the area had a different function
then.127 Something similar may have happened in Carolingian
Maastricht, a place filled with archaeological riddles. The only clearly
identifiable element is the newly-built basilica of St Servatius, which
Panhuysen dates to roughly the middle of the eighth century.128 And
then there is the Roman castrum, whose walls were maintained till

124
Werner, Der Ltticher Raum in frhkarolingischer Zeit.
125
Maastricht: The yearly reports record activities dating from Merovingian times
at several sites encircling the old castrum, but none from the Carolingian period.
I am referring here to the following excavations: Witmakersstraat, Jodenstraat,
Pieterstraat (Rijksarchief ), Ceramique, Wolfstraat/Havenstraat, Hondstraat 1317.
Dijkman and Ervynck do not record Carolingian artisanal activities relating to bone-
working (Antler, bone, horn, ivory and teeth, p. 81). Namur: see various contri-
butions in n. 96.
126
Once again, the non-presence of material is central to understanding the
processes involved, and archaeological reports should place more emphasis on this.
127
J. Plumier, La chapelle Saint Hilaire, in: J. Plumier ed., Cinq annes darcholo-
gie en province de Namur 19901995 (Namur, 1996), pp. 958.
128
Panhuysen, De Sint-Servaaskerk te Maastricht in de vroege middeleeuwen,
p. 21.
 197

the end of the Carolingian period.129 The details elude us, but it is
clear that the nature and function of these centres changed pro-
foundly in the course of the eighth century. Perhaps artisanal activ-
ities were moved out of the old centres to quarters like Wijck or to
a quarter north of the old castrum where streets run at right angles
to the Meuse as in Dinant. Or perhaps they were moved even fur-
ther away to monasteries, which recent excavations reveal to have
been important production centres in the Carolingian period.130
Excavations around the basilica of St Servatius (fig. 2) have yet to
reveal the presence of a Carolingian artisanal quarter. This presents
us with an enigma, as we might have expected this important
Carolingian abbey to be the focus of a whole new centre, taking
over the functions of the preceding artisanal quarter on the left bank
of the Meuse around the castrum. More generally, archaeological evi-
dence has shown that places like Maastricht, Huy and Namur under-
went a major transformation after about 700. This can only be
understood within the context of the development of major monas-
tic establishments in this period, and of the complex transport sys-
tems described by Devroey.131 The changes may also have spelled a
change in status for artisans and traders, and for their position in
society in general. These are questions that we cannot fully under-
stand at present, but which we must certainly bear in mind.
As is usually the case with the early Middle Ages, the largest group
of people is the most enigmatic. We need to know who else lived
in Maastricht in the Merovingian period, what their numbers were,
where they lived, and whether they were dependent or free. We do
know of one group, the lepers of Maastricht, as Adalgisel Grimo
donated land to them in his testament of 634.132 Archaeology can
provide some answers, but first we will concentrate on a more abstract

129
Panhuysen, Maastricht staat op zijn verleden, p. 91.
130
T. Capelle, Handwerk in der Karolingerzeit, Kunst und Kultur der Karolingerzeit.
Karl der Grosse und Papst Leo III in Paderborn (Mainz, 1999), pp. 4249. R. Hodges,
A fetishism for commodities: ninth century glass-making at San Vincenzo al
Volturno, in: M. Mendera ed., Archeologia e storia della produzione del vetro preindustri-
ale (Siena, 1991), pp. 6790.
131
See n. 115.
132
Van Ommeren, Bronnen voor de geschiedenis van Maastricht (3591204),
p. 21. W. Levison, Das Testament des Diakons Adalgisel-Grimo vom Jahre 634,
Aus rheinischer und frnkischer Frhzeit. Ausgewhlte Aufstze von Wilhelm Levison (Dsseldorf,
1948), pp. 11838 (Original publication: Trierer Zeitschrift 7 (1932), pp. 6985).
Werner, Der Ltticher Raum in frhkarolingischer Zeit, pp. 3237.
198

level of analysis and examine the ways in which early medieval

Maastricht manifested itself as a centre of power.

L A

The above discussion serves as a starting point for further hypothe-

ses and is not a matter of firm conclusion. What was the chemistry
of early medieval Maastricht, and how did it function as a centre
of power? At the beginning of this paper, I maintained that this was
a matter of complex dialectics between the social actors, as well as
their interaction with the supernatural and the physical world. The
physical context cannot be considered as a mere product of the other
two components; it is important in its own right. We need to develop
a new frame of reference to replace the institutional and economic
perspectives that have so far dominated the research. Various schol-
ars have sought to reconceptualise centres of power. Devroey, for
example, has presented important research on the Meuse valley.133
He, and Despy before him, have questioned the importance attrib-
uted to long-distance trade and artisanal activities as major incen-
tives for the development of early medieval centres. Devroey stresses
the independent development of each centre but is less clear about
why they flourished, other than to discuss their integration into the
manorial economy of the early middle ages. He sees such centres as
being largely dependent on a closed regional economy, thanks to the
surplus from estates handed over in the form of taxes or rents.134
There was nothing commercial about their supply. This is an impor-
tant insight, yet Devroeys explanation remains embedded in the
economic perspective.
Martin Carver, in his Arguments in Stone, adopts another approach.135
He distinguishes between early medieval towns with a Christian-
Mediterranean background and the towns of non-Christian north-
ern and eastern Europe. He uses the twin and opposing

133
Devroey, Villes, campagnes.
134
Devroey, Villes, campagnes, pp. 251260.
135
M.O.H. Carver, Arguments in stone. Archaeological research and the european town in
the first millennium (Oxford, 1993).
 199

concepts of bureaucratic and baronial state to describe the con-

texts in which these towns flourished.136 Carver sees the differences
between the towns as largely a record of ideological investment, a
debate about alternative strategies for living.137 Carvers insistence
on ideology and the argument as a major driving force in the devel-
opment of the town has drawn criticism from Halsall and Roskams.
Halsall regards the social rather than the ideological sphere as the
key to urban survival from late Roman to early medieval times.138
What matters is a sufficiently assured social hierarchy and control
of surplus.139 This seems self-evident, but does it in itself determine
the different cultural forms that urbanisation might take? Roskams
goes even further in his criticism. Discussing the usefulness of the
search for continuity of urban life, he stresses the need to look for
diversity of urban settlements at different periods. To explain this
diversity, Roskams identifies two radically different ways, from either
an idealist or a materialist standpoint.140 He equates Carvers approach
with the idealist view, and his own explicit Marxist perspective with
its materialist counterpart. His major criticism of Carvers model is
that it leaves no room for explaining the change from one political
constellation to the next, which in turn creates a major shift in the
nature of urban communities.141 I share Roskams criticism, but his
own Marxist approach with its inevitable slave mode of production,
a tributary mode of production, localised tributary relations and
feudal relations cannot help us much either. In my opinion, if we
want to understand the significance of towns and not just early
medieval towns in a wider social context, neither the idealist nor
materialist position alone can help us. Why should we favour a per-
spective in which agencies are always seen as rooted in specifically

136
Carver, Arguments in stone, p. 18.
137
Carver, Arguments in stone, p. 57.
138
G. Halsall, Towns, societies and ideas: the not-so-strange case of late Roman
and early Merovingian Metz, in: N. Christie and S.T. Loseby eds., Towns in tran-
sition. Urban evolution in late Antiquity and the early Middle Ages (Aldershot, 1996), pp.
23561.
139
Halsall, Towns, societies and ideas: the not-so-strange case of late Roman
and early Merovingian Metz, p. 252.
140
S. Roskams, Urban transition in early medieval Britain: the case of York,
in: N. Christie and S.T. Loseby eds., Towns in transition. Urban evolution in late Antiquity
and the early Middle Ages (Aldershot, 1996), pp. 26288, esp. 278; S. Roskams, Urban
transition in North Africa: Roman and medieval towns of the Maghreb, in:
N. Christie and S.T. Loseby eds., Towns in transition, pp. 15983.
141
Roskams, Urban transition, p. 279.
200

material reality rather than in ideas when we know that such agen-
cies are rooted in both, and that practices are connected with ideas
on many levels and in many, if not all, contexts?142 Explanations
should be able to account for practices and ideas.
It is not my intention now to provide an all-embracing interpre-
tative framework. Instead, I have opted for the less ambitious course
of proposing some new areas for future research. Nor do I deal with
all themes in equal depth. This reflects my present state of mind,
rather than the full range of possible research areas. The topics I
wish to present are the following: centres of power and the articu-
lation of exchange spheres, space and the constitution of society, the
cultural appropriation of past space, and finally, power centres in
their social and symbolic landscape.

At the beginning of this paper, I criticised the dominance ascribed

to economic processes in the development of centres such as Maastricht.
Exchange did play an important role in this development, and it
should be approached with the aid of the vast body of archaeolog-
ical material and theory now at our disposal.

Maastricht and the whole transactional order

Philip Grierson in the 1950s, and then George Duby in the 1970s,
have alerted medievalists to the anthropological concepts of gift
exchange, thus questioning traditional approaches to early medieval
economies based on modern common sense.143 The debate on the

142
Roskams, Urban transition, p. 279. His examples of refuse depositions seem
to be based on the assumption that these practices inform us mainly about ancient
forms of food supply and consumption patterns. However, refuse deposition is a
cultural activity which relates to ideas about hygiene, the position of elements (ani-
mals and plants) in the cosmological order, gender relations etc., all of which affect
the nature of the data set.
143
P. Grierson, Commerce in the dark ages: a critique of the evidence, Transactions
of the Royal Historical Society 9 (1959), pp. 12340. G. Duby, The early growth of the
European economy. Warriors and peasants from the seventh to the twelfth century (Ithaca, 1989).
 201

conceptualisation of the economic among anthropologists, which

also influenced archaeology,144 has led to a distinction between soci-
eties with a gift-exchange or prestige-goods economy on the one
hand, and societies with an economy based on commodity exchange
on the other. Several authors have subsequently stressed the impor-
tance of gift exchange in the early middle ages, and have accord-
ingly suggested that the early medieval economy was a gift-exchange
economy, which supposedly developed later into a mercantile econ-
omy dominated by competing commercial centres. This is a false
dichotomy that obcures our understanding of early medieval economies,
and Barbara Rosenwein has rightly criticised it in one of her famous
studies of Cluny.145
Anthropologists have brought new perspectives to the debate on
exchange. Annette Weiner, for example, has argued that valuable
objects were kept out of the gift exchange as part of a strategy of
defining and reproducing the identity of social groups and individ-
uals.146 Weiner maintains that giving often entails keeping inalien-
able objects. Taking their cue from anthropologists, archaeologists
have tried to distinguish between prestige goods potential gifts
and mere commodities. Valuable grave goods supposedly fall into
the former category, pottery into the latter.147 But how can we make
such a distinction with only the object to go by? Over time, objects
acquire new meanings and what was once a humble pot may become
a sacred vessel.
In the early 1980s Chris Gregory helped guide the future debate
by defining gift exchange as the exchange of inalienable objects
between interdependent transactors and commodity exchange as the

Originally published as: Guerriers et Paysans, VIII eXII e sicles. Premier essor de lconomie
europenne (Paris, 1973).
144
There is a wealth of literature available. One well-known example in the field
of the exchange of prestige goods is: S. Frankenstein and M.J. Rowlands, The
internal structure and regional context of early Iron Age society in south-western
Germany, Bulletin of the Institute of Archaeology 15 (1978), pp. 73112. For a recent
discussion on the exchange of valuable objects between lords and warrior/follow-
ers, see J. Bazelmans, By weapons made worthy: lords, retainers and their relationship in
Beowulf (Amsterdam, 1999).
145
B. Rosenwein, To be the neighbor of Saint Peter, The social meaning of Clunys prop-
erty, 9091049 (Ithaca/London, 1989), pp. 12543.
146
A. Weiner, Inalienable possessions: the paradox of keeping while giving (Berkeley, 1992).
147
In archaeological literature, the terms prestige goods economy and gift
exchange are often used interchangeably.
202

exchange of alienable objects between independent transactors.148

Several anthropologists have developed this idea and provided us
with new perspectives that also apply to the early middle ages.149
Bloch and Parry no longer view gift exchange and commodity
exchange as opposing concepts. They define the exchanges of the
long-term (or the long-term transactional sphere) as transactions
concerned with the reproduction of the long-term social or cosmic
order and transactions of the short-term (or the short-term trans-
actional sphere) as transactions concerned with the arena of indi-
vidual competition.150 The emphasis here seems to have shifted from
the objects exchanged and the actors involved to the meaning of the
transaction and its perception by the actors.151 In this perspective,
long-term transactions relate to the reproduction of the social and
cosmological order, deeply affecting all of society,152 whereas short-
term transactions spring from individual behaviour and gains. Both
transactions exist side by side in many types of societies, and specific
objects are not necessarily confined to either of the spheres.153 It is
possible to transfer an object from one sphere to another, and objects
may be transformed from a commodity to a gift, or even a sacred
object and vice versa. As Bloch and Parry explain, however, such
transformations are complex affairs, for each sphere has its own
norms, values and ideas. Thus, a move from one (long or short-
term) sphere to another may involve a move from one field of val-
ues to another, with the result that objects undergo a process of
transformation and obtain new meanings. Bloch and Parry explain
that transactions can only be understood when interpreted in the
total context of contemporary forms of exchange and their con-

148
As rephrased by Bloch and Parry. C. Gregory, Gifts and commodities (London,
1982), pp. 1001.
149
A. Appadurai, Introduction: commodities and the politics of value, in:
A. Appadurai ed., The social life of things: commodities in a cultural perspective (Cambridge,
1988), pp. 363. I. Kopytoff, The cultural biography of things: commoditization
as process, in: A. Appadurai ed., The social life of things, pp. 6494. M. Bloch and
J. Parry, Introduction: money and the morality of exchange, in: J. Parry and
M. Bloch eds., Money and the morality of exchange (Cambridge, 1989), pp. 132.
150
Bloch and Parry, Money and the morality of exchange, p. 24.
151
Thus specific transactions such as paying rent or transferring surpluses can
only be qualified after an assessment of the meaning in the socio-ideological sphere.
152
Transactions also take place between members of low-ranking families.
153
For this reason it is difficult to classify societies on the basis of the dominance
of either gift or commodity exchange.
 203

comitant values.154 The two spheres of exchange and the efforts to

keep objects constitute the whole transactional order.155
Objects may move from one sphere by means of more or less
elaborate rituals. In the course of time, such objects obtain a cul-
tural biography characterised by a number of transactions and trans-
formations that add to their value and meaning.156 For archaeologists
it is important to understand that the value of an object is not just
determined by its material value but by its cultural biography, how-
ever difficult this may be to ascertain.
Nevertheless, the precise nature of the interrelationships between
the different types of exchanges and how they articulate remains
unclear. This is a central issue, for it brings us to the interconnected-
ness of different values prevalent in a given society. We must then
ask questions about the spatial dimensions of the whole transactional
order. Are there specific places where an articulation of exchange
spheres takes place involving a transformation of objects, ritually or
otherwise? Appadurais tournaments of value may be useful here:
Tournaments of value are complex periodic events that are removed
in some culturally well defined way from the routines of economic life.
Participation in them is likely to be both a privilege of those in power
and an instrument of status contests between them. The currency of
such tournaments is likely to be set apart through well understood cul-
tural diacritics. Finally, what is at issue in such a tournament is not
just the status, rank, fame or reputation of the actors, but the dispo-
sition of the central tokens of value in the society in question. Though
such tournaments of value occur at special times and places, their
forms and outcomes are always consequential for the more mundane
realities of power and value in ordinary life.157
With this modern definition in mind, we can envisage how and
where tournaments of value took place in the early middle ages.
The fair at St Denis was not just a market, but a great festival as

154
In practice this means that a study of trade that confines itself to com-
mercial transactions in the early Middle Ages (I have referred to this earlier as the
economic perspective) does not allow for a proper understanding of a place like
Maastricht.
155
Hence Maurice Godeliers choice of subtitle for the introduction to his book,
The enigma of the gift: Introduction. Concerning Things that are given, Things that
are sold, and Things that must not be given or sold, but kept. M. Godelier, The
enigma of the gift (Cambridge, 1999), transl. of Lnigme du don (Paris, 1996).
156
Kopytoff, The cultural biography of things. Bloch and Parry, Money and the
morality of exchange.
157
Appadurai, Introduction. Commodities and the politics of value, p. 21.
204

well, which bore a relationship to a cult place, a place where the

whole transactional order became visible.158 Maastricht may have cel-
ebrated its feast of St Servatius on the thirteenth of May, which was
already mentioned at the beginning of the eighth century.159 It was
a place where almost all the powerful had a foothold: the king, aris-
tocrats, bishops, artisans, merchants, clergy and above all, the saint,
who was also an actor. Thus Maastrichts position stemmed not just
from its function as a commercial centre but from the special place
it occupied in the whole transactional order, a place where the
articulation of exchange spheres took place on occasions that can be
termed tournaments of value.160 Because of the elevated social posi-
tions of the actors involved, the impact of Maastrichts tournaments
of values was considerable. Other places in the Meuse valley may
have had a similar position, adapted to the needs of the regional
and local aristocracy. The perspective of whole transactional order,
whatever its merits, at least sheds some light on the production of
gold coins (trientes) in these centres.161 Coin production played a spe-
cial part in the tournaments of value which I believe must have
taken place in early medieval Maastricht. Appadurai speaks of a cur-
rency of the tournaments of value which was defined through cul-
tural diacritics. The production of coins may have taken place in
ritualised conditions and the coins may for this reason have been used
in different contexts (i.e. in different types of transactions), but their
use in any type of transaction always contains a reference to their
special context of production. The monetarius may also have had a
special perhaps temporary position during festivals and cele-
brations that were tournaments of value. We should consider the
possibility of a special relationship between the production of coins
and the saint, the central person in the ceremonies.162 The cross on

158
I.N. Wood, The Merovingian kingdoms, p. 216.
159
The feast of St Servatius is mentioned in the Calender of St Willibrord, which
dates from the beginning of the eighth century, and in other eighth century sources.
See R. De La Haye, Het middeleeuws officie van het hoogfeest van Sint Servaas
te Maastricht, Publications de la Socit Historique et Archologique dans le Limbourg 133
(1997), pp. 93140, esp. 94.
160
As Hedeager explains in her contribution to this volume, we also need to
analyse the artisanal production of such sites as Maastricht in this light.
161
For the production of gold coins, see A. Pol, Les montaires Huy et
Maastricht. Production et distribution des monnaies mrovingiens mosanes, Bulletin
de lInstitut Archologique Ligois 107 (1995), pp. 185200.
162
This is not the same as saying that bishops were responsible for coin pro-
 205

the trientes, which appeared from the last quarter of the sixth cen-
tury onwards, marks a fundamental change in ideology with regard
to the production and use of these coins. By the end of the seventh
century, Maastrichts St Servatius had become the most important
saint in the region, with coins perhaps minted on the occasion of
his festival163 being used for other purposes later, in commercial trans-
actions and as oboli at funeral rituals.164 This varied use shows that
the value of such coins was not just derived from their material value
but from their cultural biography, including their exalted origin and
their history of circulation. In the 1950s, Grierson maintained that
gold trientes were not merely a feature of commercial exchange, a
view that has since become widely accepted. Nonetheless, the pro-
duction of such coins and their inherent value and meaning over time
are still being analysed from an institutional perspective which includes
the king, the bishop, the church and the fisc. Inspired by Bloch and
Parry, I would like to argue that minting is relevant to the whole
transactional order, involving the saint and his prestige as well.
Maastricht derived its position as a place of power precisely because
it was a point of gravity in this particular order.

In this section I will briefly deal with three aspects of the research
on places of power that relate to space. I will focus first on the
countryside surrounding Maastricht, and then on the cultural appro-
priation of past space. Finally I will deal more generally with the
meaning of space in these centres.

Maastricht and the countryside

What was Maastricht like in spatial terms in the early middle

ages? Where were the borders between this power centre and the

duction, which places too great an emphasis on the institutional role of the bishop.
Instead, values in a more general sense are the important element here. I thank
Jos Bazelmans for the discussion on this subject.
163
There may not have been a fixed atelier or workshop that produced coins
all year round like a modern business or shop.
164
See Theuws, Frankish transformations, in preparation.
206

countryside? If Maastricht was indeed a bipolar place with a castrum

and a cult place outside, how should we conceptualise the world
outside these two poles of gravity? Were the pottery kilns found to
the east of the Meuse part of the centre, or were they situated in
the countryside? This is another distinction we should avoid. When
we define a centre of power as a place, we need some idea of its
physical features, but we also need to think carefully about the actors
activities in a wider geographical network. This will result in a blurred
boundary between a centre of power and the countryside. It is a
modern distinction, based on late medieval models, and we should
not forget this.
Early medieval Maastricht was embedded in a rich agricultural
landscape.165 Recently, archaeologists have concentrated on develop-
ing a conceptual framework in order to understand the social organ-
isation of the food supply of power centres and cities. Departing
from the observed practices of stockbreeding and agriculture, they
have tried to determine the involvement of a local elite in the food
supply, and in surplus extraction and taxation. Commercial exchange
may have been of minor importance in this context,166 although some
form of commodity exchange should not be excluded. The regions
between Tongres and Maastricht, and between Maastricht and Aachen,
were dotted with Roman villas prior to the third century (fig. 7).167
There must have been some continuity in the use of this land from
the Roman period into the early Middle Ages, although there is lit-
tle solid evidence for its scale and intensity. This area of rich loam
lies between the less fertile sandy areas in the north and the forests
of the Ardennes to the south.168 From west to east, this band of fer-

165
L. Kooistra, Borderland farming. Possibilities and limitations of farming in the Roman
period and the early Middle Ages between the Rhine and the Meuse (Assen and Amersfoort,
1996).
166
Interesting comments on this type of research can be found in: Roskams,
Urban transitions in North Africa, pp. 2825. Roskams, Urban transition in
early Medieval Britain: the case of York, pp. 2805. For stock breeding and agri-
cultural practices in the vicinity of Maastricht, see: Kooistra, Borderland farming. For
consumption patterns in the Roman town of Tongres, see: A. Vanderhoeven, The
earliest urbanisation in northern Gaul. Some implications of recent research in
Tongres, in: N. Roymans ed., From the sword to the plough. Three studies on the earli-
est Romanisation of northern Gaul (Amsterdam, 1996), pp. 189260, esp. 2069.
167
A. Vanderhoeven, The earliest urbanisation in northern Gaul. Some impli-
cations of recent research in Tongres, pp. 2224. W.A. van Es, De Romeinen in
Nederland (Haarlem, 1981), p. 180.
168
N. Roymans, The sword and the plough. Regional dynamics in the
 207

Fig. 7. The distribution of Roman villas (squares) and tumuli (dots) in the area
between Tongres and Maastricht (after Vanderhoeven 1996).

tile land was intersected by the Roman road from Bavai to Cologne,
passing Tongres and crossing another major trade route, the Meuse
river at Maastricht. The road (a Chausse Brunehaut) was still in use
in the Merovingian period. The fertile lands and infrastructure were
capable of supporting large groups of people, and centres such as
Tongres, Aachen and Lige.
In addition to its ability to sustain a relatively large number of
people, the region had a social landscape which played a role in
early medieval politics and ideology.169 In early medieval Maastricht,

Romanisation of Belgic Gaul and the Rhineland area, in: N. Roymans ed., From
the sword to the plough, pp. 9126, esp. p. 62. See also the map in Werner, Der Ltticher
Raum in frhkarolingischer Zeit, Karte 14. Theuws, Frankish transformations, forthcoming.
169
For examples of an analysis of the interrelationships between social landscape,
politics and ideology, see especially M. Innes, State and society in the early Middle Ages.
208

there was a spatial debate that must have also involved the sur-
rounding countryside.170
In his study on the central Meuse region in the early Carolingian
period, Matthias Werner published an intriguing map of the prop-
erties in the countryside (fig. 8). The map differs in a significant
respect from that depicting occupation in Roman times in that it
shows control of property in addition to areas of habitation (most
villages will have existed by then). It therefore tells us much more
about the social landscape, landscapes of power or different forms
of surplus extraction than a distribution map of Roman villas could
ever do.171 One of the distinctive features is a royal landscape in
the south-western confines of the town, which dates back to the early
Carolingian period. Royal estates occupied almost the entire trian-
gle between Maastricht, Aachen and Lige. Unfortunately, we know
little about the origins of the royal estates to the east of the Meuse.
Do they predate 700/750? If there is a connection between the
creation of the new centre at Aachen and the development of
these estates, only archaeology can help us determine its nature.172
Merovingian royal property seems to have been based on ancient

The middle Rhine valley, 400 1000 (Cambridge, 2000) and Innes, Hrke and Hedeager
this volume.
170
There is meagre evidence for the Merovingian period in this regard; more
evidence exists for the Carolingian period and the tenth century. However, we
should emphasise that the Carolingian social landscape may have differed significantly
from the Merovingian. Here I can only deal superficially with the problems posed
by the evidence. See Werner, Der Ltticher Raum in frhkarolingischer Zeit. G. Rotthoff,
Studien zur Geschichte des Reichsguts in Niederlothringen und Friesland whrend der schsisch-
salischen Kaiserzeit. Das Reichsgut in den heutigen Niederlanden, Belgien, Luxemburg und
Nordfrankreich (Bonn, 1953). A central problem is our lack of information about the
antecendents of the large number of royal estates southwest of Maastricht.
171
For instance, we do not know whether each recorded site with remains of
villa-type buildings represented an estate of its own or whether several villas were
organised into larger complexes.
172
See also Nelson this volume. Archaeological research in the centre of Theux
has revealed Merovingian burials and a small cult building from this period adorned
with wall mosaics consisting of glass tesserae, which is a very rare phenomenon in
the countryside in this period. Later, in Carolingian times, a church was built on
this spot. This evidence alone is insufficient to determine whether Theux was a
royal estate in those early years, although the elaborate decoration of the chapel
suggests that it was already a high-status site. Some have suggested that Theux was
already a Merovingian fisc, carved out of the large fiscal complex of the Ardennes
forests. See P. Bertholet and P. Hoffsummer, Lglise halle des Saints Hermes et Alexandre
Theux (Dison, 1986), pp. 4382.
 209

Fig. 8. Merovingian and Carolingian property in the middle Meuse area (after
Werner, 1980). 1. Merovingian royal property; 2. Pippinid property c. 700; 3. royal
property mentioned in Carolingian times; 4. royal property mentioned in later times
(mainly after A.D. 1000).

Roman centres and the Ardennes, a huge Roman fiscal territory sur-
viving into the early middle ages.173
How did Carolingian kings transform this ancient landscape with
its long cultural biography? This remains an important question, as
does the way in which Merovingian and Carolingian kings imple-
mented their power at a local and regional level. Future research will
have to provide the answers. The presence of a royal landscape
did not mean that other social actors were completely passive. Royal

173
Innes describes a comparable situation for the central Rhine region around
Mainz and Worms. M. Innes, State and society in the early Middle Ages, and Innes this
volume.
210

estates were the stamping ground of aristocratic groups seeking to

obtain offices and other props to sustain their position. The term
royal landscape refers not only to the practicalities of the exercise
of royal power, but to the fact that it is represented as such (in the
texts!). This aspect of the representation of power and its various
interpretations is essential to any analysis of the countryside.
We find a different situation further to the west of the Meuse and
beyond, where royal property was rare. We find here instead a con-
centration of Pippinid estates, possibly originating from the family of
Alpaida, Pippin IIs second wife.174 Nor does the far west, the fer-
tile lands of Haspengouw, boast much royal property.175 In general
terms, we could say that the Meuse river is a symbolic boundary
between a royal landscape to the east and an aristocratic land-
scape to the west. The lack of overlap between royal and aristo-
cratic property supports this view, although admittedly we have no
information about the origins of the royal property east of the Meuse
and little about the early history of episcopal property south of
Maastricht. By the time the Carolingians were in power, the situa-
tion was complicated by the fact that a great deal of Pippinid prop-
erty was integrated into the pool of royal property after the ascension
of King Pippin III.
Significantly, the immediate environs of Maastricht reveal a social
landscape whose topography seems the complete reverse of the one
just described. The royal fisc of Maastricht extends only to the west
of the Meuse and is thus a bridgehead in the aristocratic landscape,
but no royal property has been identified to the east. Any such aris-
tocratic property in the east would have constituted a veritable inroad
into the royal landscape.176 This reversal is significant, for the por-

174
See above.
175
Merovingian royal estates in the countryside are almost non-existent, though
we do know of two possibilities (Halmaal and Emmeren). Royal property can later
be found at Tongeren, Ligney and Lens, and possibly at Tienen and Jandrain,
which seems insignificant in relation to the estates found east of the Meuse. For a
discussion of this property, see: Werner, Der Ltticher Raum in frhkarolingischer Zeit.
176
The division of the kingdom in the treaty of Meerssen in 870 has led to some
confusion. The districtus Trectis is mentioned as belonging to Louis the German and
the eastern kingdom, and the abbey of St Servatius as belonging to Charles the
Bald and the western kingdom. (Leupen, Het gelijk van Sint-Servaas en Onze-
Lieve-Vrouwe, pp. 401; for the general context see: J.L. Nelson, Charles the Bald
(London and New York, 1992), pp. 2246). Some identify the districtus as being the
fisc Theux because it was on the eastern side of the Meuse, whereas others inter-
pret it as the castrum Maastricht and its immediate environs. The treaty views the
 211

tus of Maastricht may have been located on the east bank of the
Meuse, i.e. outside the fisc. New research may shed more light on
the vital role of rhetorics of landscape in the creation and repre-
sentation of power relations in such regions far from the king,
yet also close by, with ever increasing proximity.
But rhetorics of the landscape also operate at a more ideologi-
cal level. The estates that were the subject of our previous discus-
sion had not just an owner, but a highly relevant spatial structure.
These properties most likely became bipartite estates in the course
of the eighth and early ninth centuries. Carolingian estates proba-
bly differed from Merovingian ones in that the latter did not have
such clearly defined parts exploited for the owner. Instead, they con-
sisted of a number of farmsteads whose inhabitants had to support
the owner when present or to pay tribute.177 I refer to these estates
as invisible because the demesne the part exploited for the land-
lord was not visible in the landscape.178 We do not know how
such early demesnes came into being but we can hypothesise that
it was through reclamation carried out by the local inhabitants as
part of their obligations to the owner. If this is the case, it is an
example of how relationships of dependence acquired visibility in the
landscape through the practices of dependent people. However,
through the process of reading this landscape and interpreting it,
relations of dependence were also given form later. These practices
may have transformed ideology. By reclaiming parts of the landscape
for the benefit of the owner, they made his presence part of the
landscape, or in other words, part of nature. The social differentiation
that accompanied it became naturalised, or in other words, part of
Gods creation and Gods given order that could not be contested
or subject to interpretation.179 Thus on a daily basis and on a micro

Meuse as a border, in which case a fisc transferred to the eastern kingdom is located
to the west of the Meuse. The treaty as well as the later discussion show that inver-
sions in the social landscape may create situations that are difficult to interpret.
177
Theuws, Frankish transformations, forthcoming.
178
Theuws, Frankish transformations, forthcoming.
179
J.-P. Devroey, Les premiers polyptyques rmois VIIeIXe sicles, reprinted
in: J.-P. Devroey, Etudes sur le grand domaine carolingien (Aldershot, 1993), n. II, esp.
pp. 937. Devroey stresses the strategy of the king and clerical elite of creating an
immobile society by dividing the landed property into two categories one exploited
by the peasants for the peasants and one exploited by the peasants for the benefit
of the elite. We can make two observations here. First, we should distinguish between
the lands of the church and the king on the one hand and those of aristocrats on
the other. Stability was not always in the best interest of this latter group, and
212

scale, the social landscape conveyed a message about power rela-

tionships and norms, values and ideas.
We must ask ourselves about the role and significance of Maastricht
in a changing social and symbolic landscape. Maastrichts role became
more enigmatic than ever in the Carolingian period as it lost some
if the charisma that was now exclusively derived from the abbey of
St Servatius. Is this transformation a coincidence caused by our pre-
sent state of knowledge, or do such places as Maastricht not fit com-
fortably within the Carolingian paradigm? Perhaps monasteries fitted
the Carolingian constellation better, as they seem to have done in
other regions,180 and if so, why were so few monasteries founded in
Maastricht and surroundings in the Merovingian and Carolingian
period? In a stimulating paper presented at the International Medieval
Conference in Leeds in 2000, Anne-Marie Helvtius explained that
the founding of a monastery or convent in the seventh and early
eighth centuries can be considered an anti-episcopal act.181 This raises
the question of whether aristocratic families who controlled the see
also founded monasteries or convents with this objective in mind.
The lack of such religious houses in and around Maastricht suggests
that this was probably not the case. The resident artistocrats were
vying for control of the episcopal office. If, on the other hand, the
Pippinids and their friends and relations founded several religious
establishments further to the south, we could view this as further

would have been difficult to achieve in view of the constant partitioning of the
property on the death of the owners. See F. Theuws, Landed property and mano-
rial organisation in northern Austrasia: some considerations and a case study, in:
N. Roymans and F. Theuws eds., Images of the past. Studies on ancient societies in north-
western Europe (Amsterdam, 1991), pp. 299407, esp. 3917. See also Innes, this vol-
ume. Second, the creation of bipartite estates and of the polyptics themselves may
have been a strategy on the part of the king and the church to immobilise or sta-
bilise society. However, evidence for this strategy does not appear until the ninth
century, whereas the creation of bipartite estates had already begun in the eighth
century. As I have said above, the ideology behind the creation of polyptics may
have developed as a result of the dialectics of reading and interpreting the land-
scape in the process of estate formation so that the new ideology may have been
handed on a platter to ninth-century authors and to the creators of polyptics. It
was probably not something they invented.
180
M. de Jong, Carolingian Monasticism: the power of prayer, in R. McKitterick
ed., The New Cambridge Medieval History, vol II c. 700 c. 900 (Cambridge, 1995), pp.
62253.
181
I wish to thank Anne-Marie Helvtius (Universit du Littoral, Boulogne sur
Mer) for allowing me to quote from her paper Gender and monasticism in the
Frankish world (unpublished). See also Dierkens, Abbayes et chapitres, pp. 31827
and De Jong Carolingian monasticism, pp. 6237.
 213

circumstantial evidence that they lacked influence over the see of

Tongres/Maastricht and were competing with other groups to con-
trol it.182 The example of Maastricht shows once again how aristo-
cratic competition for special positions within a centre affects the
structure of the cultural landscape outside it.

The cultural appropriation of past space

Early medieval cultural appropriation of past space is a much neglected

aspect of power centres in northern Gaul. Population growth after
the mid-sixth century did not occur in a vacuum, and the town-
scape of places such as Maastricht and the surrounding landscape
were not a tabula rasa to be colonised. Until the beginning of the
third century, there was a dense concentration of villas in the region
between Tongres and Maastricht, most of which later lay in ruins.
This raises a host of questions. Was the sixth-century townscape of
Tongres and the landscape between Tongres and Maastricht filled
with Roman ruins? If so, how did this affect the way in which the
early medieval inhabitants perceived the landscape? Were the ruins
in practice and symbolically still the focus of old estates, and there-
fore relevant to strategies defining local groups who might bury their
dead in such places (as was the case in Rosmeer, just west of
Maastricht, on fiscal land)?183 How did the inhabitants, the aristo-
crats, the bishop and the king use the visible past in their efforts to
give meaning to their environment and position? What was the
impact of a dead city like Tongres?184 Was it important for a bishop
seeking to define his status, power and authority to reclaim the
ancient city of Tongres? Did the presence of the castrum walls in
Maastricht and the town walls in Tongres lend prestige to these sites,
and were they important in representing power? Without doubt,

182
Dierkens, Abbayes et chapitres, p. 326, concludes that the Pippinid monasteries
he studied in the Entre-Sambre-et-Meuse were created in a region devoid of epis-
copal and royal influence. Moreover, he refers to the observation made by Van
Rey that the few episcopal foundations in the diocese of Tongres/Maastricht had
no Pippinid support .
183
H. Roosens, Het Merovingisch grafveld van Rosmeer, Archeologia Belgica 204
(Brussel, 1978), p. 6.
184
For a discussion of related problems, see C. La Rocca, Using the Roman
past. Abandoned towns and local power in eleventh century Piemonte, EME 5
(1996), pp. 4569.
214

ancient structures played a part in what Carver has called the image
war.185 Of course, the landscape of the past was also used in more
practical ways by providing buildings and building material.186 This
practical use of Roman ruins has attracted most attention, yet as
Bonnie Effros makes clear in this volume, the appropriation of ruins
laden with significance is not an unproblematic affair.187 More research
is needed, and we can profit from the recent debate on monument
re-use, which prehistorians initiated and which medieval archaeol-
ogists are only just beginning to enter into.188 A closer analysis reveals
differing perceptions of ancient ruins in different periods. The exten-
sive use of spolia from second-century funerary monuments, for
instance to repair the foundations of the Roman bridge at Maastricht
in the fourth century,189 occurred in a world where such activity
implies the involvement of the imperial government; the work on
the bridge was thus represented as an act of state. Future work on
the role and significance of a place such as Maastricht should address

185
Carver, Arguments in stone, pp. 6377.
186
B. Ward-Perkins, Re-using the architectural legacy of the past, entre idolo-
gie et pragmatisme, in: G.P. Brogiolo and B. Ward-Perkins eds., The idea and ideal
of the town, pp. 22544.
187
Effros this volume.
188
See: R. Bradley, Time regained: The creation of continuity, Journal of the
British Archaeological Association 140 (1987), pp. 117. R. Bradley, Altering the earth: The
origins of monuments in Britain and continental Europe (Edenburgh, 1993), esp. pp. 11329.
J. Thomas, Time, culture and identity. An interpretive archaeology (London and New York,
1999 [1996]). R. Bradley and H. Williams, eds., The past in the Past: the re-use of
ancient monuments (London, 1998). R. Bradley, Sacred geography, World Archaeology
28 (1998), pp, 161274. For the early Middle Ages, see: E. Thte, Alte Denkmler
und frhgeschichtliche Bestattungen: ein schsisch-angelschsischer Totenbrauch und
seine Kontinuitt. Eine vergleichende Studie, Archologische Informationen 19 (1996),
pp. 10516. H. Hrke and H. Williams, Angelschsische Bestattungspltze und
ltere Denkmler: Bemerkungen zur zeitlichen Entwicklung und Deutung des
Phnomens, Archologische Informationen 20 (1997), pp. 257. Recent work in France
includes: P.-A. Fevrier, La marque de lAntiquit tardive dans le paysage religieux
mdival de la Provence rural, in: M. Fixot and E. Zadora-Rio eds., Lenvironnement
des glises et la topographie religieuse des campagnes mdivales. Actes du III e congrs interna-
tional darchologie mdival (Aix en Provence 2830 septembre 1989), Documents darcholo-
gie franaise 46 (Paris, 1994), pp. 2735. J. Le Maho, La reutilisation funraire
des difices antiques en Normandie au cours du haut moyen age, in: M. Fixot
and E. Zadora-Rio eds., Lenvironnement des glises, pp. 1021. For a full understanding,
the interpretation of monument re-use in the Early Middle Ages should be studied
in relation to other forms of contemporary cultural appropriation of the land-
scape (Theuws, Frankish transformations, in preparation.).
189
T.A.S.M. Panhuysen, Romeins Maastricht en haar beelden (Corpus Signorum Imperii)
(Maastricht, 1996). On the re-use of spolia and the Roman state, see Effros this
volume.
 215

the ways in which past space and material culture became incorpo-
rated in the socio-cosmological order of early medieval society.

Space and the analysis of centres of power and their environment

A brief final statement of my views on the analysis of space is in

order. In my paper, I have regularly used the word representation,
a key concept in the programme on the Transformation of the
Roman World. As Ian Wood has argued, the very nature of the
sources at our disposal leads those studying the Transformation of
the Roman World to write a history of the representations of this
transformation.190 I agree, but I think this stance could be developed
further, using the insights gained by anthropologists on reading the
landscape. An important source of inspiration for me in this respect
is Henrietta Moores Space, text and gender.191
Texts and landscapes are tempting for scholars interested in the
analysis of representations. Often these representations are easily
identifiable. The fortress of Chvremont can be interpreted as part
of a defensive structure, but its role in representing power cannot
be overlooked by even the most superficial observer. To stop here
would be to ignore the dynamic ways in which such a representa-
tion works in society, and how it affects the audience. In Kenya,
Henrietta Moore studied the representations dominating the Marakwets
organisation of household space. These are clearly male representa-
tions; the visible world is a male world. Yet these dominant repre-
sentations do not remain unchallenged, for they are read by others:
young men, women in general and women who wished to be mod-
ern, and outsiders such as missionaries. Reading is not simply
adhering to or reproducing a given representation; it also leads to
new interpretations that are negotiated upon. What we can learn
from Moore is to analyse the relationship between dominant repre-
sentations and dominant interpretations and practices. Similarly, early
medieval texts and landscapes filled with dominant representations
(the fortress, villa, sacred tree, monastery, monastery fields, monu-
mental grave, etc.) were read, appropriated, interpreted, and above

190
I.N. Wood, Report: The European Science Foundations programme on the
transformation of early medieval Europe, EME 6 (1997), pp. 21727.
191
Moore, Space, text and gender. An anthropological study of the Marakwet of Kenya.
216

all, contested. When the learned men of the Carolingian age drew
up estate records, representing their society as stable and unchang-
ing, this was most likely the outcome of a long-term process of inter-
pretion and negotiation, with the landscape as its main source. It is
difficult to gain a clear idea of these matters, for this was a dialec-
tic process in which different actors had different voices. Just as the
audience of a dominant representation of power was diverse, so too
were the interpretations, as well as the processes of negotiation
between different groups about power and the basic values and ideas
in society. Because spatial elements, or the landscape as a whole,
had such a wide audience, there are considerable opportunities for
archaeologists to contribute to the debate on the constitution of soci-
ety. Our greatest challenge is to develop our discipline in such a
way that we can explain not just the practices of different groups,
but the interaction between dominant representations and dominant
interpretations as well.
 AACHEN AS A PLACE OF POWER*

Janet L. Nelson

Charlemagne chose Aachen for his principal residence in his later

years, says Einhard, because of the baths, and because he and his
Franks loved swimming indeed while Charlemagne was merely
good at hunting, at swimming he positively excelled. It may be that
not much swimming was actually done in the Aachen baths when
such a crowd of members of his entourage and military household
bathed with Charlemagne that more than a hundred men might be
in the water together.1 Alcuin interestingly recalled the baths at
Aachen when he explained to his young friend Nathaniel the mean-
ing of the draught of 153 fishes the disciples caught in the sea of
Galilee following the instructions of the risen Christ ( John 21.11): I
remember talking to my lord David about the wonderful significance
of this number, my heart hot with love, in the bath bubbling warm
from the natural springs.2 The baths, then, were also places for seri-
ous conversations, hence embraced clerics and scholars as well as
fighting men. They were a public, more or less outdoor, space clearly
marked men only: an ideal locale, then, for political sociability, and
the cultivation of health of body and mind.

* My warm thanks go to Mayke de Jong for percipient comments on a draft of

this paper, and to Carine van Rhijn for editorial help. I am also grateful to all the
colleagues who commented on earlier versions at Bellagio and Vienna, and to Paul
Fouracre whose constructive criticisms have been as helpful as ever.
1
Einhard, Vita Karoli, c. 22, p. 27. See H. Cppers, Beitrge zur Geschichte
des rmischen Kur- und Badeortes Aachen, in: idem ed., Aquae Granni. Beitrge
zur Archeologie von Aachen (Kln, 1982), pp. 175, esp. 327, 5967. For the definitional
question of capital as distinct from residence, see C.-R. Brhl, Remarques sur les
notions de capitale et de rsidence pendant le haut Moyen Age, Journal des
Savants (1967), pp. 193215, repr. in: idem, Aus Mittelalter und Diplomatik. Gesammelte
Aufstze, 2 vols. (Hildesheim/Mnchen/Zrich, 1989), vol. 1, pp. 11537, esp. p. 193
(115), stressing the conceptual link between capital in the modern sense and gov-
ernmental centre. Brhl was responding to a fundamental paper of E. Ewig,
Rsidence et capital pendant le haut Moyen Age, Revue historique 130 (1963), pp.
2572, repr. in: idem, Sptantikes und frnkisches Gallien. Gesammelte Schriften, vol. I,
Beihefte der Francia 3, ed. H. Atsma (Mnchen, 1976), pp. 362408.
2
Alcuin, Ep. 262, ed. E. Dmmler, MGH Epp. 4 (Berlin, 1895), p. 420: fer-
vente naturalis aquae balneo . . .. Cf. below, n. 84.
218 .

Baths, too, carried an aura of romanity, and a good deal of Roman

stonework evidently survived, to form the basis for a restoration.
Moreover, amongst those old stones were (whether recognised or not
by eighth-century restorers) the ruins of a little Christian church,
perhaps of the fifth or sixth century, possibly abandoned subse-
quently.3 Aachen was a site with a past, but that past must have
been fairly etiolated when some ancestor of Charlemagnes, whether
in the seventh or the eighth century is impossible to say, took pos-
session of the terrain and established a residence there.
The place first hits the early medieval historical record in the
Annales regni Francorum for 765, when Pippin is said to have stayed
in Aquis villa for the winter 7656. Then Charlemagne himself stayed
there just over two years later, significantly, in the very first winter
of his reign, in 7689.4 Two charters issued during this stay term
Aachen palatium publicum.5 Yet, with the exception of a brief stop-
over in December 777,6 Charlemagne seems not to have visited
Aachen thereafter until 788, when he wintered there and held a
large assembly in March 789 at which the Admonitio generalis was pro-
mulgated.7 This was no coincidence, for the reform project announced
in the Admonitio, and the place of that projects adoption, both rep-
resented breaks with the Frankish past and harked back to much
older pasts. At the same time, both were instruments, practical and

3
M. Untermann, opere mirabile constructa. Die Aachener Residenz Karls
des Grossen, in: 799. Karl der Grosse und Papst Leo III in Paderborn. Kunst und Kultur
der Karolingerzeit, Beitrge zum Katalog der Ausstellung (Paderborn, 1999), pp. 15264, at
152, and the invaluable study of L. Falkenstein, Charlemagne et Aix-la-Chapelle,
Byzantion 61 (1991), pp. 25264, at 2367.
4
Annales regni Francorum, ed. F. Kurze, MGH SRG (Hannover, 1895), s.a. 765,
768, pp. 223, 289. Note that the so-called original version of the Annals refers
to Aachen as Aquis villa, villa quae dicitur Aquis, while the so-called revised version
gives Aquisgrani (Aquasgrani).
5
MGH DD Karol, I, ed. E. Mhlbacher (Hannover, 1906), nos. 55, 56, pp.
813.
6
Ibid. no. 118, pp. 1646, an original (on which see now B.H. Rosenwein, Nego-
tiating space. Power, restraint and privileges of immunity in early medieval Europe (Ithaca NY,
1999), pp. 11534). Ibid. no. 152 (March 786), pp. 2057, surviving in a seventeenth-
century copy, is datelined Aachen but, as Mhlbacher notes, was more probably
issued at Attigny.
7
Annales regni Francorum, pp. 845: in Aquis palatio/in Aquisgrani palatio suo. Though
the Admonitio generalis itself contains no information on its place of issue, its com-
panion document in most manuscripts, the Duplex Legationis Edictum, opens with a
statement that it was issued in Aquis palatio publico on the 23 March 789, which
accords with the evidence of the Annales: see MGH Capit. 1, nos. 22, 23, pp. 5264,
with the reference to Aachen at p. 62.
 219

ideological, whereby Charlemagne intended to shape the future. The

789 assembly certainly foreshadowed a major change in the way of
life of the itinerant court.8 It was also, I think, the signal that such
a change had now been determined by Charlemagne, and indeed
seen as key to the projects implementation.9 Meanwhile, at Aachen,
the builders were busy.
Charlemagne wintered at Aachen again in 7945. Till that point
in his reign, Aachen had shared a prominent place on Charlemagnes
itinerary with other palaces, notably Herstal and Worms.10 Thereafter,
Aachens qualitative and quantitative growth transformed it into an
effective capital11 as well as something like a modern third-world
mushroom town. Charlemagne wintered there in 7956, 7967,
and 799800, and from winter 8012 onwards he resided there more
or less permanently, with breaks only for summer campaigns and
autumn hunting.12 According to the Annales regni Francorum for 811,
the holding of the great summer assembly at Aachen was consuetudo.13
The same source records embassies received there from Constanti-
nople, Venice and Dalmatia, Baghdad and Jerusalem, Northumbria,
Scandinavia, Saragossa and Cordoba, from Slavs and Avars, and last
but not least, from Rome: the pope came in person to celebrate
Christmas 804 at Aachen.14 These visitors were meant to be impressed

8
D. Bullough, Carolingian renewal (Manchester, 1991), p. 142 with n. 59.
9
Contrast Falkenstein, Charlemagne et Aix-la-Chapelle, p. 277, who sees
Aachens function as a capital resulting from circumstances that had become
favourable to such an evolution in the last years of the reign rather than from
any decision or programme.
10
C.-R. Brhl, Fodrum, Gistum, Servitium Regis (Kln, 1968), i, pp. 1823, esp. 22.
11
Brhl, Fodrum, pp. 212, prefers the term favourite residence; but I hope
below to defend capital. Cf. Brhl, Remarques, p. 21 (repr. p. 133). P. Bourdieu,
Rethinking the state, Sociological Theory 12 (1994), repr. in his Practical Reason (1998),
pp. 3563, examines various kinds of capital (in the sense of resources), coercive,
economic, cultural, symbolic, and statist, but does not directly discuss capitals in
the central-place sense, though he comes close to doing so in considering the French
kings of the late-twelfth and thirteenth centuries, p. 48. Cf. my forthcoming, Was
Charlemagnes court a courtly society? in: C. Cubitt ed., Court culture in the earlier
middle ages (Leiden, 2001).
12
The only winter he spent elsewhere was 8056: Annales regni Francorum, pp.
1201. Cf. J.L. Nelson, La cour impriale de Charlemagne, in: R. Le Jan ed.,
La royaut et les lites dans lEurope carolingienne (du dbut du IX e aux environs de 920) (Lille,
1998), pp. 1812, repr. in: Nelson, Rulers and ruling families in early Medieval Europe
(London, 1999), ch. XIV.
13
Annales regni Francorum, s.a. 811, p. 134.
14
For details, see Annales regni Francorum, esp. s.a. 802, 807, pp. 117, 1234, describ-
ing the embassies and exotic gifts (including the elephant Abul Abaz) from Baghdad.
220 .

by what they saw. Charlemagne had brought the treasures of the

regnum to Aachen, including, from Ravenna, the equestrian statue of
Theoderic which was re-erected near the palace.15 A whole new pub-
lic space was created with large buildings linked by a processional
way (fig. 1).16 At the summit of the highest building, the emperors
aula, was set a bronze eagle with outspread wings:17 potent symbol
of imperial power, but surely, too, of Christian evangelism.18 The
splendid new church was already largely built by 796, for Alcuin,
who left the court later that year, recalled for Charlemagnes benefit
in 798 a conversation he had had with a lady at the palace in the
course of which they had admired columns erected in that most
beautiful work and wondrous church.19 That church still stands: a
monument to Charlemagnes aspirations. Its structure, replete with
number-symbolism, and its bronze fittings, evoke, like the eagle, a
blend of meanings. One was emulation of the work of other pow-
erful rulers, past and present. San Vitale, Ravenna, provided inspi-
ration, and so too, offering an even closer parallel in terms of scale,
did Santa Sofia, Benevento, completed probably only a few years

15
Chron. Moissiacense, s.a. 796, ed. G.H. Pertz, MGH SS 1 (Hannover, 1826),
p. 302; Agnellus, Liber pontificalis ecclesiae Ravennatis, c. 94, ed. O. Holder-Egger, MGH
SRL (Hannover, 1878), p. 338; see L. Falkenstein, Charlemagne et Aix-la-Chapelle,
pp. 2478, 2501. Pace Donald Bullough, Carolingian renewal, p. 62, I think there
was more to Charlemagnes re-erection of this statue at Aachen than the sym-
bolism of the Jensen or Rolls-Royce! Magnificence was indeed the name of the
game, but it was of an imperial and Roman-Christian kind in a ceremonial setting
with shades of Ravenna.
16
See fig. 1 from Falkenstein, Charlemagne et Aix-la-Chapelle, p. 243, repro-
duced by kind permission.
17
If we can believe the late tenth-century Richer, Historiarum libri IV, III, 71, ed.
G. Waitz, MGH SRG (Hannover, 1877), p. 111: aerea aquila . . . in vertice palatii a
Karolo Magno acsi volans fixa erat.
18
A. Boureau, Laigle: chronique politique dun embleme (Paris, 1985) ranges appro-
priately widely in examining the uses of this symbol in monarchic representation.
Cf. M. Garrison, Alcuins world through his letters and verse, unpublished Ph.D.
thesis (Cambridge, 1996), pp. 196200, drawing attention especially to the eagle as
a symbol both of the evangelist John and of John the Baptist.
19
Alcuin, Ep. 149, p. 244, datable to 22 July 798. Fittings and decorations no
doubt continued to be added once the basic structure was complete. A humble
piece of wooden anchorage recently found above one of the columns in the church
has been dendrodated to 776+/10 years, which proves that new building at Aachen
predated 7889 though it doesnt, of course, supply a date for Charlemagnes
decision to build a capital. See now Untermann, opere mirabile constructa,
p. 158. Nearly half of the works listed in Untermanns bibliography of 45 items
have been published during the past ten years: a vivid indication of contemporary
interest in Aachens Carolingian past.
 221

market place

L
D

E
K

M G

A
N
J

I
H

Fig. 1. Charlemagnes Aachen.

A. The church dedicated to the virgin Mary; B. the aula regia; C. the countryard
in front of the church; DE. the covered and colonnaded processional way;
F. porch; G. northern annex; H. southern annex; I. Baths; J. main altar of the Virgin;
K. twelfth-century chapel of St John the Baptist; L. presumed location of the eques-
trian statue of Theoderic; M. baptismal font in upper storey of the church; N. altar
of the Saviour in upper storey of the church.
222 .

before work on the Aachen church began.20 No less important were

typological links, forged by the Fathers and strengthened by schol-
ars like Theodulf but surely well understood by Aachen audiences
in the 790s, between the people of the old Israel, and the baptismally
reborn gens sancta of the new Israel embodied, now, in the Church
and the peoples of Charlemagnes much expanded regnum Francorum.21
The fact that this building was, not a palatine chapel, but the bap-
tismal church of a local community, underlined its generalised, even
democratised, function, as too did the dedication of its main altars
to Christ and the Virgin. It served the community of a nascent cap-
ital, hence, like Aachen itself, had at once local and universal
significance.22
There were strategic considerations too, no doubt, for choosing
Aachen. It was geographically central on the east-west axis of Francia
itself, and it was located on a spur of the old Roman road between
Cologne and Maastricht. Militarily, once Charlemagne began his
regular Saxon campaigns (these were almost annual events between
772 and 803), Aachen was a sensible mustering-point for the host,

20
Untermann, opere mirabile constructa, p. 160; cf. J. Mitchell, Karl der
Grosse, Rom und das Vermchtnis der Langobarden, in: 799. Kunst und Kultur, pp.
95108, at 98. The Benevantan Sta Sophia model seems to me, on balance, a like-
lier one than Justinians church. Cf. Bullough, cited in next note.
21
Amid the welter of recent historiography, D. Bullough, Imagines regum and the
early medieval West, first published in 1975 and reprinted in Carolingian renewal, is
characteristically thought-provoking: Bullough insists, rightly, pp. 578, not only on
the problems caused by successive remodellings of the interior of the Aachen church,
but on the need to focus on demonstrably original features of structure and fittings,
whose importance was indicated by Einhard himself, Vita Karoli, c. 26, pp. 301.
For Theodulf, see E. Dahlhaus-Berg, Nova Antiquitas et Antiqua Novitas: typologische
Exegese und isidorianische Geschichtsbild bei Theodulf von Orleans (Kln, 1975), and now
A. Freemans introduction to her fine edition of the Libri Carolini, Opus Caroli regis
contra synodum (Libri Carolini ), MGH Conc. 2, Supplementum I (Hannover, 1998).
For the Franks as the chosen, and the echo of 1 Peter 2:910, see J.L. Nelson,
The Lords anointed and the peoples choice: Carolingian royal ritual, in: D. Can-
nadine and S. Price eds., Rituals of royalty: Power and ceremonial in traditional societies
(Cambridge, 1987), reprinted in J.L. Nelson, The Frankish world (London, 1996), pp.
99132, at 10812, and now the excellent systematic study of this theme by
M. Garrison, The Franks as the New Israel? Education for an identity from Pippin
to Charlemagne, in: Y. Hen and M. Innes eds., The uses of the past in the early Middle
Ages (Cambridge, 2000), pp. 11461, with a brief, illuminating, discussion of Aachen
at 1546.
22
Cf. my own remarks (a shift of register from private to public), in: Carolingian
royal funerals, in: F. Theuws and J.L. Nelson eds., Rituals of power: From late Antiquity
to the early Middle Ages (Leiden, 2000), p. 147.
 223

with a direct route leading east towards the heart of Saxony.23 To-
wards the west, it was only 17 km. from the Meuse and a busy quay
at Maastricht.24 In every direction, substantial concentrations of Caro-
lingian family lands lay not far from Aachen.

Within only a few years, Aachen had become a hub: continuously

attracting crowds of litigants and seekers of justice, who caused mag-
num impedimentum in palatio,25 clients, royal servants, beggars, whores.26
There was a weekly market; there was a mint, which in a political
sense, if not in terms of volume of output, was the most important
in the realm.27 Around the royal palace and the cubiculum regium,28
where Charlemagne and his closest entourage were lodged, there
clustered the mansiones of the ministri and royal officers; the houses of
the servants of all those officers; the houses of the nobility. Notker has
a story which may be apocryphal (he was after all writing seventy
years after Charlemagnes death) but vividly evokes conditions of life
at Aachen. After praising the fine buildings built by Charlemagne,
following Solomons example, Notker mentioned the mansiones of the
men of various ranks: constructed all around the palace, large and
lofty, but not so lofty as the emperors residence: Charlemagne, very
clever as he was ( peritissimus), could look out from the windows of
his apartments, over the mansiones of the nobles and see all they did,
all their comings and goings, without any of them realising it.29 Thus,
Charlemagne having made his sedes the more or less permanent cen-
tre of an empire, was in a position to shape those who dwelled

23
E.M. Wightman, Gallia Belgica (London, 1985), Map at pp. xiixiii; cf. J.W.
Bernhardt, Itinerant kingship and royal monasteries in early Medieval Germany, 9361075
(Cambridge, 1993), Map 7, p. 322.
24
CF. Einhard, Translatio SS Marcellini et Petri, IV, c. 13, ed. G. Waitz, MGH
SS 15, 1 (Hannover, 1888), p. 261. See A. Verhulst, The rise of cities in North-West
Europe (Cambridge, 1999), pp. 478; and for the regional context, F. Theuws,
Landed property and manorial organisation in Northern Austrasia: some consid-
erations and a case study, in: N. Roymans and F. Theuws eds., Images of the past.
Studies on ancient societies in Northwestern Europe (Amsterdam, 1991), pp. 299407.
25
MGH Capit. 1, eds. A. Boretius and V. Krause (Hannover, 1883), no. 64,
p. 153.
26
De disciplina palatii Aquisgranesis (?820), MGH Capit. I, no. 146, pp. 2978, trans-
lated in Appendix I, below.
27
Cf. S. Coupland, In palatio nostro: les monnaies palatines de Charlemagne,
Bulletin de la socit franaise de numismatique 41 (1986), pp. 87-9.
28
inhard, Translatio SS Marcellini et Petri, II, 1, ed. G. Waitz, MGH SS 15, 1
(Hannover, 1888), p. 245.
29
Notker, Gesta Karoli I, 30, ed. H.F. Haefele, MGH SRG (Berlin, 1962), p. 41.
224 .

therein into a self-conscious elite. Einhard, who later in life would

himself keep a townhouse at Aachen, wrote of the houses where
conversations took place.30 Aulici, comites, actores, mansionarii, camerarii,
the officers of the count of the palace, were showing, and showing
off, their new administrative talents.31 In the light of the above, and
using Martin Biddles archaeologists check-list of urban traits,32
Aachen can be appraised as an urban site. It may never have had
defences, and the existence of a planned street-system is question-
able.33 But it certainly did have a market; a mint; a role as a cen-
tral place; a relatively large and dense population; a diversified
economic base; houses of urban type; social differentiation; complex
religious organisation; and a judicial centre. Particularly compre-
hensive evidence for market, population, houses, and economic role
is to be found in the Capitulary on palace discipline, and though
this was produced early in Louis the Piouss reign, it surely reflects
conditions that had come into being during the latter part of
Charlemagnes.34
Charlemagnes choice, in effect, was for a new place, and a new
kind of place. The very fact that it was new, at least in the sense
of hitherto insignificant, meant that it was untramelled by a past. It
could be made the focus of new traditions, and embrace a variety
of traditions35 without being the prisoner of any one of them.36 The
fact, for instance, that it was not the seat of a bishopric meant that
it could be designed as a new kind of religious centre, a theatre for

30
Einhard, Translatio II, 3, 4; IV, 7, pp. 248, 258 (the domus Einhardi); Vita Karoli,
c. 32, p. 37: in domibus ubi conversabatur. For the extension of the originally monas-
tic sense of conversatio to lay people, including those at Charlemagnes court, see J.L.
Nelson, The voice of Charlemagne, in a forthcoming Festschrift (Oxford, 2001).
31
J. Fleckenstein, Die Struktur des Hofes Karls des Grossen im Spiegel von
Hinkmars De ordine palatii, Zeitschrift des Aachener Geschichtsvereins 83 (1976), pp. 522.
See further, below; and also my forthcoming paper, Was Charlemagnes court a
courtly society?
32
M. Biddle, Towns, in: D.M. Wilson ed., The archaeology of Anglo-Saxon England
(London, 1976), pp. 99150, at 100.
33
Falkenstein, Charlemagne et Aix-la-Chapelle, p. 272; Untermann, opere
mirabile constructa, p. 162.
34
Cf. above, n. 26, and Appendix A, below.
35
Including Roman ones: W. Jacobsen, Die Pfalzkonzeptionen Karls des Grossen,
in: L.E. Saurma-Jeltsch ed., Karl der Grosse als vielberufener Vorfahr (Sigmaringen, 1994),
pp. 2348.
36
Untermann, opere mirabile constructa, p. 162: Nicht ein realistisches Bild
von Rom prgt die Pfalz Aachen, sondern ein eklektische, idealierende Vorstellung
von angemessener, der Residenz eines Knigs wrdiger Reprsentation.
 225

the rites of rulership. Bishops were there in plenty, and that very
plenty underlined the rulers dominance.37 The climaxing of the five
great reform councils of 813 in a general meeting at Aachen put
the trumps in Charlemagnes hands.38 No wonder that for at least
the latter part of his reign, Charlemagne, changing earlier plans, had
indicated that the Aachen church was where he wished to be buried.39
Three further types of evidence can be brought into the picture.
First, an examination of the scribes responsible for royal charters
suggests that the staffing of Charlemagnes writing-office increased
strikingly from 800 onwards, as compared with earlier phases of the
reign. In these years, no fewer than nine new charter-scribes were
active, in addition to two new notaries for the count of the palace
who was responsible for hearing law-suits and complaints, and had
his own writing-office.40 In the pre-imperial years, the numbers of
active scribes of royal charters were much smaller, starting from the
early 770s, when Hitheriuss absence in Italy apparently meant that
no charters could be issued for the best part of a year,41 and sett-
ling down to a complement of two or three in the 780s and 790s.
A royal Judgement of 806 (one of only two extant from the years
800814), showing no fewer than three deputy counts of the palace,
could suggest a parallel increase of personnel in that department.42
The second type of evidence is that of capitularies. Thanks to
Hubert Mordek, the substantial impression given by the old MGH
edition can be both amplified and given more precise contours.43 It

37
Contrast the reading of Brhl, Capitale, p. 21 (p. 133).
38
F.-L. Ganshof, Note sur les Capitula tractanda de causis cum episcopis et abbatibus
tractandis de 811, Studi Gratiani 13 (1967), 125, at 159, stresses the limits to
Charlemagnes control, where I would see a consistent policy pursued effectively
throughout the imperial years, notably from 811 through to 813: Nelson, The
voice of Charlemagne, forthcoming.
39
Nelson, Carolingian royal funerals, pp. 144, n. 48, 146, 1513; cf. Einhard,
Vita Karoli, c. 31, p. 35.
40
See Appendices II and III.
41
Bullough, Carolingian renewal, p. 127.
42
MGH DD Karol, I, no. 204, pp. 2734.
43
H. Mordek, Karolingische Kapitularien, in: idem ed., berlieferung und Geltung
normativer Texte des frhen und hohen Mittelalters, Quellen und Forschungen zum Recht
im Mittelalter 4 (Sigmaringen, 1986), pp. 2550; cf. idem, art. Kapitularien, in:
Lexikon des Mittelalters 5 (1990), cols. 9436; and, on the stages of a capitularys cre-
ation, idem, Recently discovered capitulary texts belonging to the legislation of
Louis the Pious, in: P. Godman and R. Collins eds., Charlemagnes heir. New per-
spectives on the reign of Louis the Pious (Oxford, 1990), pp. 43753, and idem, Kapitularien
und Schriftlichkeit, in: R. Schieffer ed., Schriftkultur und Reichsverwaltung unter den
226 .

may be, as F.-L. Ganshof long ago argued, that Charlemagne returned
from Italy in 801 fired with the idea of an imperial programme
entailing not just revisions of and additions to the leges, the codes of
Frankish law, but also a raft of new administrative legislation in the
form of capitularies.44 The profusion of such documentation in the
imperial years is, at all events, remarkable. Not only were there
large-scale efforts, like the capitulary of 802 dignified by Ganshof as
programmatic, or the 806 Divisio regni, or the huge combined out-
put of the reform councils of 813 which was collected and promul-
gated at Aachen later that year: there were also literally dozens of
brief sets of rules and regulations, as often as not simple lists of
points, which interested parties that is, former assembly-attenders
drew up for themselves as aides-mmoires for what had been pro-
jected, discussed and agreed. All this suggests considerably larger
numbers of active participants in the business of administrative reform
than before the Aachen years.
The third sort of evidence is a single piece: the De ordine palatii.
Often referred to as Hincmars, and undoubtedly surviving in a
form revised and augmented by the archbishop of Rheims at the
very end of his life, in 882, this extraordinary text originated, as
Hincmar himself said, in a work by Adalard of Corbie (+823).
Hincmar refers to Adalard as an old and wise kinsman of the elder
Charles, the emperor, and first amongst his first counsellors, adding
that he himself, as a youth, had seen Adalard, and that he had read
and copied his treatise.45 Brigitte Kasten, building on the work of

Karolingern, Abhandlungen der Nordrhein-Westflischen Akademie der Wissenschaften

97, 1996), pp. 3466; and for the manuscripts and their contents idem, Bibliotheca
capitularium regum Francorum manuscripta. berlieferung und Traditionszusammenhang der
frnkischen Herrscherlasse, MGH Hilfsmittel 15 (Mnchen, 1995). See further A. Bhler,
Capitularia Relecta. Studien zur Entstehung und berlieferung der Kapitularien Karls
des Grossen und Ludwigs des Frommen, Archiv fr Diplomatik 32 (1986), pp. 305501;
J.L. Nelson, Literacy in Carolingian government, in: R. McKitterick ed., The uses
of literacy in early Medieval Europe (Cambridge, 1990), pp. 25896, repr. in: J.L. Nelson,
The Frankish world (London, 1996), pp. 136; and now the excellent discussion in:
P. Wormald, The making of english law (Oxford, 1999), Part I, chapter 2, esp. pp.
4553.
44
Ganshof, Charlemagnes programme of imperial government, in his The
Carolingians and the Frankish monarchy, transl. J. Sondheimer (London, 1971), pp. 5585.
45
De ordine palatii, eds. T. Gross and R. Schieffer, MGH Fontes Iuris Germanici
Antiqui 3 (Hannover, 1980), c. III, p. 54 [c. 12 in Krauses edition in MGH Capit.
2 (Hannover, 1897), p. 522]: Adalardum senem et sapientem domni Karoli magni impera-
toris propinquum . . . inter primos consiliarios primum, in adolescentia mea vidi. Cuius libellum de
ordine palatii legi et scripsi. I give references to both these editions chapter-divisions
 227

C.-R. Brhl and others, has persuasively argued that it was written
for Bernard of Italy, c. 810814 (which I would narrow further to 812),46
when Adalard had gone back to Italy with the young king in the
role of advisor and guide: guidance which included the finding of a
suitable bride.47 While Hincmar made additions (the preface addressed
to King Carloman (87984) and the ensuing chapters I to III;48 the

(Gross and Schieffers first, followed by Krauses in square brackets), but note that
the text survives in just one, early sixteenth-century, manuscript which lacks any
chapter-divisions, except the word capi (with superscript suspension-mark) at the
beginning of Gross and Schieffers c. IV, p. 56 and p. 18, n. 34: see below, n. 48.
B. Kasten, Adalhard von Corbie (Dsseldorf, 1986), pp. 7284, offers an invaluable
discussion of arguments for Adalards authorship of the De ordine, gives good grounds
for dating the work vor 814 (p. 76), and, following C.-R. Brhl, Hinkmariana
1, Deutsches Archiv 29 (1964), pp. 4877, at 54, suspects that Adalard wrote for
Bernard of Italy.
46
Kasten, Adalhard, p. 79, suggests 810814, but the likely link with Bernards
coming of age and endowment with a kingdom by his grandfather suggests to me
812: see below, p. 231. I think Kastens alternative possibility of soon after 781,
with Pippin as addressee, can be excluded: see below, p. 229. P. Depreux, Prosopographie
de lentourage de Louis le Pieux (781840) (Sigmaringen, 1997), p. 77, without com-
mitting himself to a precise date, accepts a link between Adalards writing of the
De ordine palatii and his role as Bernards counsellor.
47
Translatio sancti Viti, c. 3, ed. I. Schmale-Ott, Verffentlichungen der Histori-
schen Kommission fr Westfalen 41, Fontes Minores 1 (Mnster, 1979), p. 38; cf.
B. Kasten, Adalhard, p. 72, The notable role assigned to the queen at De ordine, c. V,
pp. 724 [c. 22, p. 525], might be linked with Bernards marriage which accord-
ing to the Translatio sancti Viti was arranged by Adalard; cf. P. Depreux, Das
Knigtum Berhards von Italien, Quellen und Forschungen aus Italienischen Archiven und
Bibliotheken 72 (1992), pp. 125, at 7, and idem, Prosopographie, pp. 1345, with fur-
ther references. The hypothesis raises a difficulty, however (unnoted by Depreux,
Das Knigtum): it seems that Bernard never formally married his partner Cunigunda,
in other words that no full marital endowment was made in 812/3, nor was she
ever queen: see C. La Rocca, Les reines dans le royaume dItalie du IXe sicle,
in: F. Bougard and R. Le Jan eds., Les douaires (Lille, forthcoming), for penetrating
discussion. It could well be that it was intended that the relationship should be
legalised once Cunigunda had borne children, but Bernards rebellion in 817 and
death in 818 precluded such developments. Cunigundas case could then be com-
pared with that of Angilberga, Louis IIs acknowledged partner from 851 but not
endowed and married till 860 after the birth of children: see the dower-charter of
5 October (860), ed. K. Wanner, Ludovici II Diplomata, Instituto Storico Italiano per
il Medio Evo (Rome 1994), no. 30, pp. 1257. I intend to explore elsewhere the
implications of such provisional relationships for the management of royal families
in the early Middle Ages. The relevance of all this to the authorship of the De
ordine palatii would be that Adalard in 812 expected Cunigunda to become queen
in fairly short order, and hence signalled her future role in the palace. For the
basis of Adalards depiction in his personal acquaintance with the queens role at
the courts of Pippin and Charlemagne, see below, pp. 2312.
48
The presence in the unique manuscript of the single indication of a new chap-
ter at Gross and Schieffers c. IV could represent the join between Hincmars new
prefatory material and Adalards original text: see above, n. 45. Further Hincmarian
228 .

references to the apocrisiar,49 and, possibly, to twice-yearly assem-

blies),50 and while some of the text, for instance c. VIs stress on
counsel and references to seniores and minores, could work equally well
for 881 or 812, certain features can be claimed as symptomatic of
Adalards core-text. First, this is a work of political science, that is,
it is not primarily concerned, as are other exemplars of the Mirror
of Princes genre, with christian moralising but with a cool exposi-
tion of structures and processes and an appreciation of the socio-
logical context in which political relations operated.51 Its relevant
here to recall that, unlike Hincmar who was a former child-oblate,
and monk for many years, before accepting episcopal office, Adalard
had been an adult conversus, having spent his earlier years as a young
warrior at the courts of Pippin and then, till 771/2, Charlemagne.
His knowledge was of the experiential kind, sufficiently so to have
evoked both an appropriate nickname, Antony (who had also been
a conversus), among the court-circle of Charlemagne, and a (perhaps
defensively) favourable comment from his biographer.52 Then there
are stylistic traits: the lack of biblical quotations from c. IV on, which
could again reflect Adalards relative illiteracy, and contrasts strik-
ingly with Hincmars writing; the use of phrases to link chapters or
sections, which is a feature, too, of Adalards Statutes for Corbie;53
and certain touches, notably the use of scapoardus to mean pincerna
(butler), which seem indicative of the authors Italian governmental
experience and Italian addressee.54 But Adalards experience of

additions were interpolations, detectable only by the sharp eyes of H. Lwe and
others. Few if any scholars nowadays accept L. Halphens argument that Adalards
entire libellus was Hincmars invention: cf. L. Halphen, Le De ordine palatii dHincmar,
Revue historique 183 (1938), pp. 19.
49
In cc. IV, V, VI, pp. 56, 62, 6870, 88 [cc. 13, 16, 20, 32, pp. 5224, 528],
see H. Lwe, Hinkmar von Reims und der Apokrisiar. Beitrge zur Interpretation
von De ordine palatii , Festschrift fr H. Heimpel, 3 (Gttingen, 1972), pp. 197225.
50
De ordine palatii, c. VI, pp. 826 [cc. 2930, p. 527], see Lwe, Hinkmar,
pp. 2212; Kasten, Adalhard, pp. 767.
51
J.L. Nelson, Charles the Bald (London, 1992), pp. 438 (carelessly attributing the
whole work to Hincmar). Cf. P. Kirn, Die mittelalterliche Staatsverwaltung als
geistesgeschichtliches Problem, Historisches Vierteljahrschrift 27 (1937), pp. 52348, at
5336.
52
See below, n. 61. For further comments on monks easily traversing the space
between palace and monastery, see Mayke de Jongs chapter in this volume.
53
See Kasten, Adalhard, p. 74, citing the dissertation of J. Schmidt which I have
unfortunately not been able to consult.
54
De ordine palatii, c. V, p. 64 [c. 17, p. 523], see Brhl, Hinkmariana, pp.
524, and Kasten, Adalhard, p. 78 with n. 209, citing Paul the Deacons gloss,
Historia Langobardorum, V, 2: pincerna quem vulgo scaffardum dicimus.
 229

Carolingian government was by no means confined to Italy: Hincmar

rightly identified his role as Charlemagnes kinsman and first coun-
sellor. Though he served two important stints in Italy in the 780s
and in 8104, the bulk of his political career was played out at, or
in close contact with, Charlemagnes court, specifically between 790
and 810. In other words, Charlemagnes Aachen years were also
very largely Adalards. When Alcuin wanted to get a message to his
former student Angilbert at court in 790, he wrote to Adalard.55
When Alcuin wanted to know about the events of summer 799, he
wrote to ask Adalard.56 It was perhaps at Paderborn that Adalard
impressed Pope Leo III so greatly that at a subsequent meeting
between the two in Rome, the pope jokingly told him: O Frank,
be sure that if I find you to be other than I think you are, no other
Frank had better come here expecting me to trust him!57 When in
809 Charlemagne wanted to guide discussion at Aachen of astro-
nomical matters bearing on the date of Easter, he seems to have
asked Adalard for a position paper.58 Later that same year, a synod
summoned to Aachen to pronounce on the filioque question produced
a statement which was taken to Rome early in 810 by Adalard who
had evidently taken a leading part in the discussions.59 But perhaps
the most telling evidence for Adalards frequent residence at court
is the fact that Alcuin gave him a nickname, Antony. The allusion
here was to the celebrated fourth-century desert hermit, and it is
hard not to discern a little irony here, at Adalards expense.60 Word-

55
MGH Epp. 4, no. 9, p. 35.
56
MGH Epp. 4, no. 181, pp. 299300. The Paderborn meeting followed a family-
conclave at Aachen: cf. Nelson, Charlemagne, pater optimus?, in: J. Jarnut, Am
Vorabend der Kaiserkrnung: Karl der Grosse und Leo III (forthcoming, 2001). Other let-
ters addressed to Adalard are nos. 175, 176, 220, 222, 237 of which only 220
(c. 801), p. 364, and 237 (801), pp. 3812, and just possibly 175, pp. 2901, look to
me as if they might have been addressed to Adalard at court rather than at Corbie.
In no. 237, p. 382, the phrase ubicumque vadas suggests an allusion to Adalards trav-
els, cf. also no. 222, p. 365, addressing Adalard as mansionarius, which might allude
to his earthly, and courtly rather than monastic, as well as hoped-for heavenly, role
as a dweller in many mansions and an assigner of appropriate residence rights to
others.
57
Paschasius Radbertus, Vita Adalardi, c. 17, PL 120, col. 1517. For Adalards
subsequent visit to Rome, see below, n. 58.
58
MGH Epp 4, Epistolae variorum no. 42, the replies of the compotist(a)e, where
c. 9, p. 566, follows the lectio composed by Adalard. Cf. Kasten, Adalhard, p. 63.
59
MGH Conc. 2, 1, ed. A. Werminghoff (Hannover, 1906), no. 33, pp. 23544;
Annales regni Francorum 809, p. 129.
60
For nicknames at Charlemagnes court in general, see M. Garrison, The emer-
gence of Latin literary culture and the court of Charlemagne, in: R. McKitterick
230 .

play and ironic mixes of affection and critique were symptomatic of

the close personal relationships generated in the intimate yet also
highly competitive atmosphere of the court.61
The De ordine palatii describes a system whose salient features
arguably reflect Adalards court experiences, above all, at Aachen.
This was the venue of nearly all the great assemblies of the years
after 794, and the De ordines details about procedural arrangements,
involving a complex geography of separate sittings, messengers, and
then collective decision-making, can be substantiated in part through
careful weighing of the multifarious documentation that capitularies
offer for the years from 802 onwards.62 Aachen, again, was the hub
of multiple givings, receivings, and redistributions of gifts; and where
such transactions are scrappily documented in prose and poetry, the
De ordines several references accurately depict the variety and signi-
ficance of these various transactions: gifts in kind and cash flowed,
both regularly, on an annual basis, and irregularly, as time and cir-
cumstance allowed, from king and queen to unbeneficed milites in
the palace service, and into the palace at assemblies generaliter, that
is, from churches and from lay attenders too.63 Aachen, again, was
the destination of foreign embassies whose origins were, as the De
ordine says, diverse, and whose incidence multiplied, from 802, when
Byzantine envoys arrived near the beginning of the year, and Isaac
the Jew (whom Charlemagne had sent to the East four years before)
arrived home on 20 July, along with one envoy from the calif Harun

ed., Carolingian culture: Emulation and innovation (Cambridge, 1993), pp. 11140; idem,
The social world of Alcuin: nicknames at York and at the Carolingian court, in:
L. Houwen and A. MacDonald eds., Alcuin of York and his influence on European cul-
ture, Germania Latina III (Groningen 1998), pp. 5979.
61
In her soon to be published Ph.D. thesis, ch. 7, pp. 1812, Garrison notes
that Antony was cited by Augustine, De doctrina christiana, proemium 4, ed. J. Martin,
CSEL 32 (Turnhout, 1962), pp. 23, as an example of an illiterate who learned
the scriptures by heart and learned to understand them by deep reflection, and that
Paschasius, Vita Adalardi, c. 61, PL 120, col. 1539, echoed the contrast between
Antony/Adalards lack of formal training and his natural gift for understanding.
Garrisons perceptive observations offer possibilities for new insights into Adalards
life and, through that, into the De ordine palatii.
62
De ordine palatii, c. VI, pp. 8296 [cc. 2936, pp. 3279]. Cf. J.L. Nelson, The
voice of Charlemagne, forthcoming.
63
De ordine palatii, cc. V, VI, pp. 72, 802 [cc. 22, 2728, pp. 525, 526]. The
whole subject was illuminatingly discussed by T. Reuter, Plunder and tribute in
the Carolingian empire, Transactions of the Royal Historical Society 35 (1985), pp. 7594,
at 807; cf. idem, Assemblies, in: P. Linehan and J.L. Nelson eds., The medieval
world (London, 2001), forthcoming.
 231

al-Rashid and another from Abraham amir of Kairouan, with the

elephant and with gifts, until 810, when Charlemagne returned
home from Saxony to receive envoys from Constantinople and from
Cordoba.64 The phrase return home assumes a more or less fixed
capital; and the De ordines observations on the management of the
palace economy tend to confirm this,65 as also do references to the
mansiones of chief ministers and to the presence of discipuli at what
Kasten terms a sort of civil service college (Verwaltungshochschule).66
Further governmental features noted in the De ordine are depicted
as arising from realms vast extent: the need to have farflung regna
represented, as it were, by careful distribution of offices at court; the
need for good intelligence on frontier regions, given their sensitivity;
the necessity of delegating responsibility, so that regional officers only
forward matters to the ruler for decision when these cannot be dealt
with locally.67 This clear perception of the geography of power and
its implications might be seen as the hallmark of someone who had
experience of government in a major province (Italy) but also at the
centre.
Lastly, I return to the De ordines comments on the role of the
queen as not only highly relevant to Bernards situation in 812, when
Adalard had brought Cunigunda to be the young kings bride, but
also deriving from Adalards mixed experience of the court of
Charlemagne. Adalard, originally a tiro in the household of Charle-
magnes parents, had first withdrawn from Charlemagnes court, and
(temporarily) from royal favour in 772 as a consequence of the repu-
diation of the Lombard queen in whose selection and approval
Adalard had been personally involved.68 Subsequent experience of

64
Annales regni Francorum 802, p. 117; 810, p. 132: domum revertitur; cf. De ordine
palatii, c. V, p. 74 [c. 22, p. 525].
65
Though Kasten, Adalhard, p. 78, finds here die organisatorische Seite der
wirtschaftlichen Versorgung eines reisenden Knigshofes, no very sharp distinction
can be made between a seasonally mobile court spending lengthy winter stays in
one spot, and an (only partly) itinerant court.
66
Kasten, Adalhard, p. 82.
67
De ordine palatii, cc. IV, p. 66 [c. 18, pp. 5234], VII, pp. 946 [c. 36,
p. 529], VI, p. 88 [c. 31, pp. 5278], VI, p. 84 [c. 30, p. 527]. Cf. K.-F. Werner,
Missus marchio comes, in: W. Paravicini and K.F. Werner eds., Histoire com-
pare de ladministration, Beiheft der Francia, 9 (1980), pp. 191239, esp. 2123.
68
J.L. Nelson, Famille, p. 205, suggesting, perhaps, Adalards prior closeness
to Bertrada; cf. now J.L. Nelson, Making a difference in ninth-century politics: the
daughters of Desiderius, in: A. Murray ed., After Romes fall: Narrators and sources of
232 .

Charlemagnes court before 800 might well have confirmed Adalards

confidence in the importance of queenly patronage and power. The
queenless years which came after 800, however well Charlemagnes
daughters worked as advisers, and despite Adalards own personal
standing with and participation in that regime, could never have
been treated as a normal or even desirable state of affairs: hence De
ordines recommendation of more conventional arrangements. Adalard
hoped that Cunigunda could assume the role once played, in his
view successfully, by Charlemagnes Lombard queen, by Hildegard,
and by Fastrada.69 This was not a role that Adalard could have
observed during his years governing Italy.
All told, the above features of the De ordine seem to reflect not (or
not just) Adalards experience of the Lombard kingdom, but his years
at the Aachen court which had left him just as keenly aware of
the importance of the on-the-spot networking implied by the invi-
tations issued by capitanei ministeriales to the young palace milites to
dine with them in their mansiones.70
The variety of senses in which Charlemagne used Aachen as a
theatre of power can be illustrated by two episodes, one apparently
small-scale, the other of vast political import. The first is described
by Charlemagne himself in a letter to Bishop Ghaerbald of Lige.
Recalling orders to clergy that no-one should presume to lift some-
one from the font at baptism before they can recite the Lords Prayer
and the Creed before you or one of your ordained ministers,
Charlemagne reports that just recently, there were many in our
presence (apud nos) at Epiphany whom we found wished to receive
infants from the baptismal font, and we ordered them to be indi-
vidually and carefully tested and examined as to whether they knew
and had by heart the Lords Prayer and the Creed, and many of
them could not remember them at all. We ordered them not to pro-
ceed, therefore, and not to presume to raise any infant from the
baptismal font until they were able to know and recite the Prayer

barbarian history. Essays presented to Walter Goffart (Toronto, 1998), pp. 17190.
69
This now seems to me a preferable explanation of De ordines depiction of the
queens role to the alternatives I offered in: La cour impriale, p. 178, n. 8.
70
De ordine palatii, c. V, p. 80 [c. 27, p. 526]: . . . illos [milites] praefati capitanei
ministeriales certatim de die in diem, nunc istos, nunc illos ad mansiones suas vocabant et non
tam gulae voracitate, quam verae familiaritatis seu dilectionis amore, proutque cuique possibile erat,
impendere studebant. Cf. Dhuoda, Liber manualis, ed. P. Rich (Paris, 1975), III, 11,
p. 194, advising her son to dine as often as possible with good priests.
 233

and Creed. They blushed deeply [valde erubescentes fuerunt] because of

this, and willingly promised that if they were allowed, they would
be able in time to remove this shortcoming from themselves.71 Did
the inadequate would-be godparents perhaps include some of the
Aachen lite along with humbler people? If so, their public naming
and shaming here by the ruler himself showed his determination that
the laity should practise the fundamental requirements of Christianity
regardless of rank. The second, much more famous, scene shows
power that likewise transcends modern distinctions between public
and private, religious and secular. In September 813, Charlemagne
presided over a great assembly at Aachen, where he asked every-
one, from the greatest to the least, if it pleased them that he should
hand over the imperial dignity to his son Louis, and they replied
enthusiastically that it was Gods choice. The Sunday following, 11
September, when all had processed into the church, Charlemagne
asked Louis if he would agree to abide by a series of rules of just
government, and when Louis agreed, he was told to take the crown
from the altar and place it on his own head.72 The power, again,
is Gods, though everyone is involved in rulerships transmission,
witnessing what is in effect a coronation oath. But Charlemagnes
power and responsibility as Gods agent on earth are also impres-
sively on display. Aachen, in short, was the place where collective
power was mobilised by a ruler exceptionally aware of his respon-
sibility. This is what gave his project its lan.
Aachen was to remain throughout the ninth century a favoured
site for ritual inaugurations and acknowledgements and transfers of
regnal power in Francia. When Louiss own turn came to pass ruler-
ship on to his son Charles in 837, he too chose the Aachen palace
as the ritual site.73 After Louiss death, and months of inter-Carolingian
conflict, Aachen, the sedes regni, was abandoned by Lothar in March

71
MGH Capit. I, no. 122, p. 241. The letters date is uncertain but cant be
from long before 810. For the wider context of Charlemagnes efforts to get the
practice and meaning of baptism right, and to enforce the obligations of god-
parents, see S.A. Keefe, Carolingian baptismal expositions: a handlist of tracts and
manuscripts, in: U.-R. Blumenthal ed., Carolingian essays (Washington DC, 1983),
pp. 169237, showing that no fewer than 61 responses to Charlemagnes enquiry
are still extant; J.L. Nelson, Parents, children and the church in the earlier Middle
Ages, Studies in Church History 31 (1994), pp. 81114.
72
Thegan, Vita Hludowici imperatoris, ed. E. Tremp, MGH SRG 54 (Hannover,
1995), c. 6; cf. Annales regni Francorum 813, p. 138.
73
Annales de Saint-Bertin 837, ed. F. Grat et al. (Paris, 1964), p. 22.
234 .

842: Lothars younger brothers Louis and Charles immediately made

for Aachen, held an assembly there, and their bishops pronounced
Lothars forfeiture of the regnum on account of his lack of the knowl-
edge of ruling the state. Everyone then asked Louis and Charles
if they were willing to rule according to Gods will, and when they
agreed, ordered them to receive the regnum.74 Aachen was still a
central place in the minds of competing Carolingians until the late
870s, with successive rival kings staging rituals of power in the church
and the palace.75 Conversely, in 881, the author of the Annales Fuldenses
recorded the desecration of Aachen by Northmen who stabled their
horses in the church and ransacked the palace.76 This was the last
mention of Aachen in these Annals; and in the Chronicle of Regino
too, there are no further records of ritual events at Aachen after
881. These silences of the late ninth century shout the demise of a
place of power in a certain sense. When Aachen was used again
by Ottonians and others, down to the twentieth century, it was no
longer as a base of power. By the year 1000, even the exact loca-
tion of Charlemagnes tomb had been forgotten.77
Nevertheless Aachen, especially during the lifetime of Charlemagne,
had been a place of power in more than one sense. A story recorded
in 885 about the mid-eighth-century origins of the Aachen palace
may well have begun life, and circulated for decades, among aulici.78
The story-teller is Notker the Stammerer, and though, on his own
admission, he never visited Francia,79 hes already been mentioned
in this paper as the recipient and purveyor of an anecdote about
Charlemagnes control of Aachens inhabitants that depends for its
effect on detailed knowledge of the places topography.80 In this other
story, Aachen appears as a place of power of a different kind:

74
Nithard, Historiarum Libri IV, III, 7, ed. E. Mller, MGH SRG (Hannover,
1907), p. 39, IV, 1, p. 40.
75
Annales Fuldenses 870, ed. F. Kurze, MGH SRG (Hannover, 1891), pp. 712;
876, p. 87; 877, p. 90; Annales de Saint-Bertin 869, pp. 1678; 870, pp. 171, 175;
876, p. 210.
76
Annales Fuldenses 881, p. 97.
77
K. Grich, Otto III. ffnet das Karlsgrab in Aachen. berlegungen zu
Heiligerverehrung, Heiligsprechung und Traditionsbildung, in: G. Althoff and
E. Schubert eds, Herrschaftsreprsentation im Ottonischen Sachsen (Sigmaringen, 1998),
pp. 381430.
78
Cf. M. Innes, Memory, orality and literacy in an early medieval society, Past
and Present 158 (1998), pp. 336.
79
Notker, Gesta Karoli, I, 34, p. 47.
80
Above, p. 223 and n. 29. For other stories about Aachen, Notker, Gesta Karoli,
 235

[Pippin] once fought an unheard-of battle against spiritual wickedness

(Ephes. 6, 12). Since the thermal baths had not yet been built at
Aachen, the hot and most health-giving springs simply bubbled forth
there. Pippin ordered his chamberlain to make sure that the springs
were clean and that no unknown person had been allowed in. When
this had been done, the king picked up his sword in his hand and
hurried off to the bath in his shirt and slippers. Suddenly the Old
Enemy attacked him and made as if to kill him. The king protected
himself with the sign of the Cross and drew his sword. Sensing a shape
in human form, he struck his invincible sword so firmly into the ground,
that it was only after a long struggle that he managed with great
difficulty to draw it out again. But the shape had such grossness to it
that it befouled all the springs with filth and blood and revolting slime.
Not even this could trouble the unconquerable Pippin, however. Dont
be upset by whats happened, said he to his chamberlain. Let the
filthy water flow away, and as soon as it runs clean again, then I will
have my bath at once.81
This seriously entertaining story was intended for a Carolingian ruler,
Charles the Fat, of whom Notker still, at the time he wrote this part
of the Gesta Karoli, had high hopes for a renovation of the regnum
Francorum. The story focuses first on dynastic anxiety, generated by
the idea of the dynastys founder exposed to diabolical threat. The
apt symbol of this threat was diabolical pollution of springs that were
an inexplicable natural force, canalisable, as it were, but hardly con-
trollable by man. In a place that should have been health-giving,
the innocent would-be bather had encountered the slime of lurking
pre-Christian, anti-Christian, power. The storys second motif is
confidence: the waters would flow pure again, and Pippins victory
was also Christs. No Christian story-teller or audience could miss
the message that associated the health-giving spring with the salvific
waters of baptism. Just as Charlemagne (and, driven by his deter-
mination, Aachen assemblies and congregations) wrestled with the
problems of correct baptismal performance and the Christian for-
mation of the young baptised, so the story recorded, but surely not

I, 27, p. 38; I, 31, 32, pp. 425; II, 8, p. 59; II, 21, p. 92. For Book Is group of
Aachen stories, and for the dates of writing of the two books, see now Simon
MacLean, Charles the Fat, University of London Ph.D. thesis, 2000. For Notkers
humour, see D. Ganz, Humour as history in Notkers Gesta Karolii Magn, in: E.B.
King, J.T. Schaefer and W.B. Wadley eds., Monks, nuns, and friars in medieval society
(Sewanee TE, 1989), pp. 17183.
81
Notker, Gesta Karoli, II, 15, p. 80. The translation is mine, but with some debt
to that of L. Thorpe, Two Lives of Charlemagne (Harmondsworth, 1969), pp. 1601.
236 .

invented, by Notker was that, with Gods help, a courageous Caro-

lingian king (and, by extension, his fideles) could vanquish the dan-
gers in chthonic powers. Pippins initial purification of the natural
springs had made possible their canalising by his son, the builder of
the Aachen baths. The story, whose projection back onto Charle-
magnes father made it the more effective as a myth of dynastic
legitimation, explained and justified the centring of Charlemagnes
whole project at Aachen. Literally and metaphorically, Aachen was
not just to be represented as, but to be, a place of cleansing for a
people assured that they were Gods, a place where power was har-
nessed for multiple human applications in the service of God.
Einhards muscularly romanist (and Suetonius-inspired) account of
Charlemagnes choice of Aachen underestimated, if it did not wholly
misrepresent, the significance for Charlemagne of Aachens bubbling
springs.82 Alcuin, recalling for Charlemagnes benefit the Aachen
baths as the entirely appropriate location for his own exposition of
John 21.11 as a prototype of apostolic Church-making, evoked
his lords divinely-appointed task.83 Notkers story needs to be read,
or heard, alongside Alcuins commentary, and as a counterweight
to Einhards rationalising. What Aachens hot springs meant for
Charlemagne and his court in some ways resembled their meaning
for their nineteenth-century devotee the Empress Elisabeth after whom
they are named today, yet in other ways (and these were the ways
of Alcuin and Notker as well as Einhard) meant things that were
quite different, indeed absolutely of their own Carolingian moment.
What was inevitably presented, given a culture deriving all legiti-
macy from the past, as renewal was in fact new. Geo-politics and

82
Part of its point, no doubt, was to contrast Charlemagne with Titus, yet the
context in both is secular and political: cf. J.L. Nelson, The Lords Anointed and
the peoples choice: Carolingian royal ritual, in: D. Cannadine and S. Price, Rituals
of royalty (Cambridge, 1987), pp. 13780, repr. in: J.L. Nelson, The Frankish world
(London, 1996), pp. 99132, esp. 1123.
83
Alcuin, Ep. 212 (800/1), p. 353, to the arch-chaplain Riculf, reminded him
of the same interpretation of John 21, 11. Cf. Alcuin, Ep. 113 (796), p. 163, to
Arn of Salzburg, with a similar exegesis. In his Commentary in Joh., vii, 43, Alcuin
explained that the unbroken net with its huge load of fish signified the Holy Church
of the elect undamaged by any dissensions, PL 100, col. 996. See Garrison, Alcuins
world, pp. 198, 2024, for insights on the associative implications of Alcuins
thoughts on fishing and episcopal duties. A further echo is surely to be found in
Annales Sancti Amandi 800, ed. G.H. Pertz, MGH SS I (Hannover, 1826), p. 10. A
serious Sunday-lunchtime conversation with Billy Nelson helped me think laterally
about Notkers story.
 237

geo-ideology had everything to do with this. Aachens construction

as a political centre was, in direct and obvious ways, made possible
by Charlemagnes political achievement, and while his regime, like
all regimes, was transient though it lasted longer than many
it proved to be ideologically indefinitely renewable. Yet, in still more
fundamental ways, and by a thought-provoking paradox, the pre-
condition for that achievement, as well as its sign, was the creation
of a new place of power.
238 .

A I

Translation of Capitulare de disciplina palatii Aquisgranesis (?820), MGH Capit.

I, no. 146, pp. 297-8, from MS Paris BN lat 4788, fol. 114

Capitulary on discipline in the palace at Aachen

1. Every palace officer (ministerialis) should make inquiry most carefully first
of all among his own men and then among their fellows ( pares) if any
of them, or of their people (sui) should be found hiding a whore or a
man. Any man who had such a man or woman with him, if he refuses
to pay emendation, he is to be kept an eye on in our palace. Likewise
it is our will that the household-officers (ministeriales) of our beloved queen
and of our children do the same.

2. Ratbert the estate-manager (actor) is to have a similar inquiry made in

his area of office ( per suum ministerium) that is in the houses (mansiones) of
our servants both in Aachen and in the nearby estates belonging to
Aachen. Peter and Gunzo must make a similar inquiry throughout the
personal rooms (scruae) and houses of all our estate-managers, and Ernald
through the houses of all traders whether they trade in the market (in
mercato) or elsewhere, and whether they are Christians or Jews. The
deputy-seneschal (mansionarius) is likewise to make inquiry with his junior
officers through the houses of bishops and abbots and counts, those who
are not themselves estate-managers, and our vassals (vassi ), at a time
when those lords (seniores) are not there in those houses.

3. We will and command that none of those who serve us in our palace
should presume to harbour any man on account of his having com-
mitted theft, or any homicide or adultery or any crime whatsoever and
who has come to our palace for this reason and has tried to go into
hiding there. If any free man who violates this order and harbours such
a man, let him know that he will have to carry on his shoulders that
man who was found in his house, first around the palace and then to
the jail in which that criminal must be put. If the man who despises
our order is a slave, he is likewise to carry the malefactor on his own
shoulders to jail, and then he himself is to be lead into the marketplace
and there flogged according to his deserts. It is our will likewise, con-
cerning trollops ( gadales) and whores, that in whosesoevers houses they
may be found they should likewise by carried by those men to the mar-
ketplace where those same women are to be flogged or, if he refuses, it
is our will that he, along with her shall be beaten in the same place.
 239

4. Whoever finds men brawling (rixantes) in the palace and could make
peace between them but refuses to do that, let him know that he will
be [treated as] as partner in the crime (damnum) done by the brawlers
between them. If he sees them brawling and cant make peace between
them, and refuses to make known to the brawlers that he will report
them, we likewise will that he should share in [paying] the composition
for the crime those two have committed.

5. Whoever receives or harbours or refuses to make arrangements to expel

a man coming from anywhere whatsoever to our palace who has com-
mitted any crime in our palace, let him present him [i.e. the criminal]
or, if he cannot present him, let him pay compensation for the crime
that [the criminal] did.

6. Our palatine counts are to apply all their attention to ensuring that
appellants whose statements they have received should not remain in
our palace after they [i.e. those counts] have taken their statement (indicu-
lum) from them.

7. Supervisors (magistri ) are to be set up over beggars and poor people with
the task of giving them care and provision so that ?tricksters [lacuna]
and frauds may not hide themselves among them.

8. Every week, on every Saturday, our officials and servants are to report
what they have done about making those inquiries, and in cases that
have been reported to us, whether they have diligently and truthfully
held a further inquiry and investigated it, so that, if it pleases us, they
can affirm in our hands [in manu nostra i.e. their hands in ours?] that
they have reported nothing but the truth to us.
240 .

A II: C

The evidence is culled from Mhlbachers introduction and edition of

Charlemagnes chartes, MGH DD Karol. I. Dates and charter (D) num-
bers give earliest and latest evidence for an individuals activity.

a. Chancellor Hitherius ([760, D Pippini 13] 768, D 55777,

D 114)
Notaries: Rado (772, D 67777, D 114)
Wigbald (774, D 84786, D 154)

b. Chancellor Rado (777, D 116799, D 188)

Notaries: Wigbald (774, D 84786, D 154)
Ercambald (778, D 119799, D 188)
Giltbert (778, D 120798, D 179)
Optatus (779, D 122788, D 162)
Widolaicus (781, D 136794, D 178)
Jacob (787, D 157792, D 175)

c. Chancellor Ercambald (799, D 187812, D 217)

Notaries: Ermin (799, D 187)
Amalbert (799, D 189807, D 205)
Genesius (799, D 190802, D 198)
Hagding (803, D 200)
Aldric (807, DD 206, 207)
Blado (808, D 208)
Ibbo (809, D 209810, D 210)
Suavis (811, D 213, 215)
Guidbert (812, D 217)

d. Under Jeremiah (8134, D 218)

Notary: Witherius diaconus (813, D 218)
 241

A III. C
(with thanks to Billy Nelson for bar-charting)

24

20

16

12

0
769
771
773
775
777
779
781
783
785
787
789
791
793
795
797
799
801
803
805
807
809
811
813
This page intentionally left blank
 CONVENTS, VIOLENCE, AND COMPETITION FOR
POWER IN SEVENTH-CENTURY FRANCIA*

Rgine Le Jan

The monasteries founded in such numbers in northern Gaul in the

seventh century, in the wake of preachers and missionaries, made a
powerful contribution to the christianianisation of the countryside.1
Most of the time, however, the missionaries were not operating in
virgin territory among populations that were still largely pagan. The
journey-routes, and the staging-posts, of the holy men, of Columbanus
and his disciples, or, a little later, of Amandus, were not the result
of chance: they were organised, prepared and supported by groups
of aristocratic kin and by kings and their families.2 The missionaries
were expected, and no doubt even summoned, by the aristocrats
themselves, and the missions succeeded in the end only where they
were favoured by religious, social and political conditions. Monastic
foundations multiplied when these served the interests of aristocratic

* I should like to thank Jinty Nelson for translating this paper. [ Translators note:
While the word monastery strictly speaking refers to a religious community whether
of men or women (cf. French monastre), in modern colloquial English it tends
to be reserved for male institutions, with convent the usual word for womens
ones. While the existence of double monasteries in the early Middle Ages compli-
cates the picture, the word convent is used in what follows to denote a commu-
nity, wholly or largely consisting of women, of which an abbess was in charge.
Monastery (cf. monasticism) is used either in a general sense for any religious house,
or to denote a male community. For the sake of stylistic variety, the terms mens
house and womens house are also used.]
1
There is a vast historiography on this subject. See the classic study of F. Prinz,
Frhes Mnchtum im Frankenreich. Kultur und Gesellschaft in Gallien, den Rheinlnden und
Bayern am Beispiel der monastischen Entwicklung (4. bis 8. Jht) (Mnchen and Wien,
1965), and the works of J. Semmler, especially Mission und Pfarrorganisation in
den rheinischen, Mosel- und Maaslndischen Bistmers (510. Jht), Cristianizzatione
e organizzazione ecclesiastica delle campagne nellalto medioevo e resistenze, Settimane 28 ii
(Spoleto, 1982), pp. 81388. For two regional studies, see A. Dierkens, Abbayes et
chaptres entre Sambre et Meuse (VII eXII e sicle) (Sigmaringen 1985), and A.-M. Helvetius,
Abbayes, vques et laques. Une politique du pouvoir en Hainaut au Moyen-Age (VII eXI e
sicles), (Brussels, 1994).
2
R. Le Jan, Famille et pouvoir dans le monde franc (VII eX e sicle). Essai danthropolo-
gie sociale (Paris, 1995), pp. 1256, 387400.
244

families. Reducing the phenomenon of monasticism to its religious

aspect alone would mean seriously misunderstanding a basic feature
of the way in which early medieval society worked, namely, the inter-
penetration of the religious and the secular, the sacred and the pro-
fane, the public and the private. In the seventh and eighth centuries,
founding a monastery was as much a political as a religious act.
Monasteries were places of prayer and asceticism, but at the same
time they were places of power. That power was spiritual by virtue
of the tombs of the founding saints and the relics piously and jeal-
ously preserved in these places; it was economic thanks to the landed
capital invested in them, and the revenues extracted therefrom; and
it was cultural because of the education provided there for the young
oblates, and the making of copies of authoritative books, as well as
the production of new works, by the artisans and artists who worked
there. In a wider sense, monasteries were key centres of political
power, where networks of clientage and fidelity could be constructed
and reinforced, and new bonds created. Everything converged at
these sites. In a society soaked in the irrational, the possession of
precious relics made it possible to attract the gifts of the faithful and
to group friends around the monastery. Recent work has shown how
important were the kind of links created by gifts in the tenth and
eleventh centuries,3 and these findings are equally valid for earlier
periods too. It was by such means that monastic communities were
inserted into local communities, took a leading hand in the game of
inheritances, and became major players in the complex mechanisms
by which land-ownership circulated.4 Monasteries themselves were
true microcosms reflecting the social form and internal structure of
aristocratic groups. Last but not least, monasteries were the only
institutions whose capital, both symbolic and real, and regarded as
something sacred, grew of its own accord, following the rhythm of
vocations and oblations, gifts, and spiritual and material exchanges.
In the seventh century, founding a monastery in the countryside
thus became an aristocratic familys means of expressing its power
while at the same time reinforcing it by appropriating the sacred
power conveyed in relics. Yet the monasteries in whose founding

3
See especially B. Rosenwein, To be the neighbor of Saint Peter. The social meaning of
Clunys property, 9091049 (Ithaca and London, 1989).
4
On this cluster of problems, see Les transferts patrimoniaux en Europe occidentale
VIII eX e sicle, Mlanges de lcole Franaise de Rome 1112 (1999).
 , , 245

aristocratic kin-groups took part were not all of the same type, and
had varied social and political functions. Some were founded on
fiscal land given by the king, and hence were placed de facto under
royal patronage even if the influence of the founding group might
still predominate there. Others belonged to the founding family as
its own property: cases in point were the convents to be considered
in this paper. It is true that the concept of the Eigenklster is applic-
able to these places only to a certain extent, in so far as in the
Merovingian period every monastery was subject to episcopal author-
ity, unless an exemption had been obtained.5 Every one of these
monasteries had been founded with the support of the local bishop.
Yet it is not exemption that defines a private monastery, but rather
the fact that the foundation had been the work of a noble family,
and involved the earmarking of patrimonial land, and, further, that
the family kept certain rights over the monastery, particularly that
of influencing the choice of the abbot, and, above all, of guaran-
teeing the places tuitio, its protection. The monastery was therefore
part of the familys honor without being subject to the traditional rules
of inheritance. The foundation of a private monastery thus strength-
ened the founding group at the expense of its rivals: so much was
this the case that the Eigenkloster swiftly became major objects in a
competition for power which involved the use of force, directly and
indirectly, outside and inside monastic communities.
Seventh-century monasticism had a pronounced female character:
for it was women in particularly large numbers who founded con-
vents into which they retired to abandon the world and vow them-
selves to a life of prayer and divine service.6 Hence it was womens

5
J. Semmler, Episcopi potestas und karolingische Klosterpolitik, in: A. Borst ed,
Mnchtum, Episkopat und Adel zur Grndungszeit des Klosters Reichenau, Vortrge und Forschungen
20, (Sigmaringen 1974), pp. 30594.
6
S.F. Wemple, Women in Frankish society. Marriage and the cloister, 500 to 900
(Philadelphia, 1981), pp. 12688. The chronology of monastic foundations for women
later in the medieval period is discussed by B.L. Venarde, Womens monasticism in
the central Middle Ages (Ithaca and London, 1997), figure 2, p. 8. On reasons mo-
tivating women to choose the monastic life, see Wemple, Women, pp. 190ff., and
H.-W. Goetz, Frauen im Frhmittelalter. Frauenbild und Frauenleben im Frankenreich,
(Weimar/Kln/Wien, 1995). A critical discussion of recent historiography, see
A.-M. Helvetius, Virgo et virago: reflexions sur le pouvoir du voile consacr daprs
les sources hagiographiques de la Gaule du Nord, in: S. Lebecq, A. Dierkens,
R. Le Jan and J.-M. Sansterre eds., Femmes et pouvoirs des femmes Byzance et en
Occident (VI eXI e sicle) (Lille, 1999), pp. 1978.
246

houses which, in the seventh century, were the specific targets of

violence, of actual attacks, of iniuriae committed by the abbesses ene-
mies. True, Merovingian society was violent generally,7 and these
convents, mostly open to their surroundings, were ready prey to the
wicked and the greedy. Abbesses welcomed pilgrims, sick people, vis-
itors, royal envoys, refugees; abbesses sent messengers, kept up rela-
tions with other abbesses, with their diocesan bishops, and with the
king. But no more so than did abbots. Besides, these convents were
double monasteries, with a small male community living alongside
the main, female, community, under the abbesss authority. Such
institutions, founded under the influence of Irish monasticism, met
the needs of society by affording greater security for the nuns in a
social context where effective protection could not be guaranteed at
local level, in the countryside.
An abbess might be attacked the moment her family became weak-
ened, but the deeper reasons for such acts of violence have be sought
at the level of the symbolic system. In a society where social repro-
duction was achieved by means of mechanisms of exchange and the
creation of bonds of friendship, and through highly codified rituals,
iniuria caused a breach of equilibrium which entailed an exchange
of violence in order that peace could be re-established.8 In so far as
violence was controlled and led eventually to pacification, it was part
of the normal mechanisms of social regulation, and was closely linked
with honour as studied recently by anthropologists like Julian Pitt-
Rivers,9 sociologists like Pierre Bourdieu10 and historians like Klaus
Schreiner.11
Attacks on abbesses and their communities of nuns were not a
matter of blind violence, then, and their explanation lies elsewhere
than in mere greed for profit. In order to highlight more clearly the
function of a convent as place of power, I want now to examine

7
On early medieval violence, see G. Halsall ed., Violence and society in the early
medieval west (London, 1997).
8
Roger Verdier, Le systme vindicatoire, in: idem ed., La vengeance. tudes dth-
nologie, dhistoire et de philosophie, vol. 1 (Paris, 1980), pp. 1342.
9
J. Pitt-Rivers, Anthropologie de lhonneur. La msaventure de Sichem (Paris, 1983).
10
P. Bourdieu, Le sens de lhonneur, in: idem, Esquisse dune thorie de la pra-
tique, prcde de trois tudes dthnologie kabyle (Geneva, 1972), pp. 1344.
11
K. Schreiner, Verletzte Ehre, in: K. Schreiner and G. Schwerhoff eds.,
Verletzte Ehre. Ehrkonflikte in Gesellschaften des Mittelalters und der frhen Neuzeit (Norm und
Struktur) (Kln/Weimar/Wien, 1995), pp. 128.
 , , 247

the causes and nature of acts of violence against abbesses. I shall

look at three cases-studies in detail, Faremoutiers, Nivelles and St-
Jean de Laon. All three are well-documented in hagiographical sources
that are either strictly contemporary or were written only slightly
after the events they record: for Faremoutiers, the Vita Columbani,
written by Jonas of Bobbio c. 640;12 for Nivelles, the Vita Geretrudis
prima, written c. 670, and the Additamentum Nivialense de Fuliano;13 and
for St-Jean, the Vita Sadalbergae and the Vita Anstrudis,14 the last of
these probably a ninth-century text but certainly composed on the
basis of an earlier version written in the early eighth century, shortly
after Anstrudes death.15
Pressure could often be exerted even before a convent was founded
or in its earliest days. According to the Vita Geretrudis, after the death
of Pippin I, his widow Itta was staying at her estate of Nivelles, in
the south of Brabant, with her daughter Gertrude, and wondering
what was the best course of action to take for herself and the girl:
at this point, Amandus, who had probably just become bishop of
Tongres-Maastricht with the help of Grimoald, son of Pippin and
Itta,16 paid Itta a visit, and advised her to found a convent for her-
self and her daughter.17 Of course this story is a reconstruction of
the past in typical hagiographic style, and modern scholars have had
good reason to surmise that Grimoald was behind the decision to
found a convent at Nivelles,18 on family land.19 Indeed by 648/9,

12
Jonas of Bobbio, Vita Columbani abbatis et discipulorum eius libri duo, ed. B. Krusch,
MGH SRM 4 (Hannover 1902), pp. 1152; partial transl. P. Fouracre and R. Ger-
berding, Later Merovingian France (Manchester 1996).
13
Vita Geretrudis prima, ed. B. Krusch, MGH SRM 2 (Hannover, 1887), pp.
45263, transl. J.A. McNamara and J. Halborg with E. Gordon Whatley, Sainted
women of the Dark Ages (Durham and London 1992), pp. 22034; Additamentum Nivialense
de Fuliano, ed. Krusch, ibid., pp. 44951.
14
Vita Sadalbergae abbatissae Laudunensis, ed. B. Krusch, MGH SRM 5 (Hannover,
1910), pp. 4066, transl. McNamara et al., Sainted Women, pp. 17694; Vita Anstrudis,
ed. W. Levison, MGH SRM 6 (Hannover, 1913), pp. 6678, partial transl. Fouracre
and Gerberding, Later Merovingian France.
15
Levison, introduction to the Vita Anstrudis, ed. cit., p. 64.
16
M. Werner, Der ltticher Raum in frhkarolingischer Zeit. Untersuchungen zur Geschichte
einer karolingischen Stammlandschaft (Gttingen, 1980), p. 235; A. Dierkens, Saint-
Amand et la fondation de labbaye de Nivelles, in: Saint-Gry et la christianisation
dans le Nord de la Gaule, V eIX e sicles, Revue du Nord 69 (1986), p. 329.
17
Vita Geretrudis, c. 2, p. 455.
18
Werner, Der ltticher Raum, p. 234.
19
A. Bergengruen, Adel und Grundherrschaft im Merowingerreich (Wiesbaden, 1958),
p. 109.
248

his position was secure enough for what was no doubt a long-
cherished project to be put into effect.20 Grimoald had had to bide
his time. In 640, on the death of his father Pippin, he had been
driven out of the post of mayor of the palace and replaced by Otto,
former guardian of the young king Sigebert III; and it was only
three years later that Grimoald returned to the political centre-stage
after Ottos murder by Duke Leuthar.21 Thereafter Grimoald strength-
ened his hand, and took a large part in the foundation of the
monastery of Cugnon in 646, and in Remaclus foundation of Stavelot-
Malmdy shortly after.22 But these two houses, often wrongly thought
to have been Pippinid foundations, were neither of them private
monasteries. They had been founded on royal land thanks to the
co-operation of certain nobles and the king. The case of Nivelles
was quite different. Assured of the support of Amandus, bishop-elect
of Tongres-Maastricht, Grimoald was able to found a private con-
vent23 on land that apparently belonged to his mother Itta, though
its not quite clear whether the property was part of her own inher-
itance or belonged to her dower. She assumed the leadership of the
community, as widows often did. The hagiographer did not fail to
note, however, that the project had aroused fierce opposition from
wicked men who caused Itta and Gertrude to suffer injuries, insults
and privations.24 Who were these opponents? Itta came originally
from Brabant and had links with families who, like the Pippinids,
had supported Amanduss missionary work and who, shortly after
Nivelles foundation, founded their own monasteries of Maubeuge,
Mons and Soignies. These families were based in the border-lands
of Neustria and Austrasia and had links with both regna. The foun-
dation of Nivelles, following the other foundations inspired or encour-
aged by Amandus, must have brought an increase in contacts and
hence consolidated the Pippinids position in this strategic region.
Their enemies were comparable to those who emerged after Remacluss
foundation of Cugnon25 in the shape of rival and neighbouring groups

20
Dierkens, Saint Amand, p. 331.
21
R. Schieffer, Die Karolinger (2nd ed., Stuttgart/Berlin/Kln, 1997), p. 18.
22
For the foundation of Stavelot-Malmdy, see Werner, Der ltticher Raum, p. 355.
23
R. Gerberding, The rise of the Carolingians and the Liber Historiae Francorum
(Oxford, 1987), p. 96; J. Semmler, Episcopi potestas und karolingische Klosterpolitik,
in Borst ed., Mnchtum, p. 387.
24
Vita Geretrudis, c. 2, pp. 4556.
25
According to Heriger in his chronicle of the bishops of Lige, Remaclus left
 , , 249

ever ready to block their opponents efforts to strengthen their posi-

tion by founding a new monastery. Its worth recalling that after
Grimoalds fall, between 656 and 662, there was no new Pippinid
foundation until 691, when Pippin IIs power had become firmly
established in Austrasia. Only then was his mother Begga, a widow
for thirty years, able to found on her estate near the Meuse at
Andenne a private convent where she could shelter for the remain-
der of her widowhood.26 Once the opposition had been overcome,
Nivelles grew rapidly and Gertrude succeeded her mother without
difficulty. Three months before her own death, she left the leader-
ship of her convent to her niece Wulftrude, Grimoalds daughter,
who had been dedicated since childhood to the religious life.

The Vita Sadalbergae provides another instance of the problems that

new convents could run into. Sadalberga, daughter of Duke Gunduin
and his wife Saretrude, had been unable to follow the vocation
aroused in her in childhood when Columbanuss disciple Eustasius
has visited her parents.27 She had been married off, and quickly wid-
owed, then made to accept a second marriage with Blandin, or Baso,
a close counsellor of King Dagobert. The couple had five children,
and then, once those were grown up, Sadalberga got her husbands
agreement to found a convent at Langres, forty miles from Luxeuil,
on land inherited from her father Gunduin. This, then, was another
case of a private convent, founded with the support of Sadalbergas
brother Leuduin, bishop of Toul. The new house seemed destined
to flourish, but according to the hagiographer, the site eventually
proved too exposed to the attacks of enemies, and c. 650 Sadalberga
had to abandon her paternal lands and her patria (Austrasia?), and
move to Laon to found another house on the frontier between
Austrasia and Neustria. According to the Vita, she chose Laon because
of the sites defensive advantages, but other sources reveal that she
had some supporters there. One of her kinsmen, Chagnoald, had
been bishop of Laon, and Bishop Attelanus, who had succeeded
Chagnoald soon after 632, backed the new convent. It flourished so
well that before she died in 665, Sadalberga could have the body

Cugnon to escape, inter alia, the vexations of neighbouring lords, Gesta Pontificum
Leodiensium, c. 47. This account is accepted by Werner, Der ltticher Raum, p. 367.
26
Werner, Der ltticher Raum, pp. 4014; Schieffer, Die Karolinger, p. 22.
27
Vita Sadalbergae, c. 3, p. 57.
250

of her brother Leuduin/Bodo brought there from Toul, and she

bequeathed the convents leadership to her daughter Anstrude.
Once the foundations initial difficulties had been overcome, both
Nivelles and St-Jean, Laon, were able to develop quickly, as did
Eboriacum/Faremoutiers, founded in the Brie a few decades earlier,
c. 620. Yet the abbesses of all three convents, Fara at Faremoutiers,
Wulftrude at Nivelles, and Anstrude at Laon, were victims, one after
the other, of iniuriae, violent attacks by their enemies who tried to
drive them out and to take over control of their foundations. As we
shall see, however, these acts of violence cant be understood unless
they are linked with three murders of which kinsmen of each of the
three abbesses had been victims not long before. The iniuriae suffered
by the abbesses thus belong within the framework of feuds or power-
struggles that pitted rival groups against one another: they were the
sign and expression, in each case, of a crisis of familial power.
The case of Eboriacum/Faremoutiers reveals something of how
complex were these struggles and how high were the stakes. The
convent had been founded c. 620 by Chagneric and his daughter
Burgundofara on one of their estates, within an environment of
Columbanan influence. Jonas of Bobbio in his Vita Columbani says
nothing of any subsequent hostility to the new foundation, but the
communitys rapid success must surely have aroused envy.28 At any
event, towards the end of the 630s, the mayor of the palace Aega
was a savage adversary of Faremoutiers: he continually violated its
boundaries and kept persecuting its dependants ( familia) living in the
neighbourhood every time he got the chance.29 Aega was of Neustrian
origin, and is first documented in the Testament of Bishop Bertramn
of Le Mans as the buyer of a colonica near that civitas.30 After Dagoberts
reunification of the kingdom in 629, Aega became a powerful man
at his court, and one of the kings chief counsellors.31 A few days

28
On Faremoutiers development in the Merovingian period, see J. Gurout, art.
Faremoutiers, DHGE 16, cols. 5347.
29
Vita Columbani II, 67, p. 137: Erat enim adversarius monasterii Ega nomine, vir in
seculo sublimis, cui Dagobertus moriens filium Chlodoveum cum regno commendaverat. His ergo
adversabatur supradicto coenubio terminosque violabat omnemque familiam eius circummanentem
quacumque poterat occasione persequebatur.
30
H. Ebling, Prosopographie der Amtstrger des Merowingerreiches von Chlothar II (613)
bis Karl Martell (741), Beihefte der Francia 2 (Sigmaringen, 1974), no. XII, p. 38.
31
Fredegar, Chronicon IV, c. 62, ed. B. Krusch, MGH SRM 2, p. 151, transl.
J.M. Wallace-Hadrill, The fourth book of the Chronicle of Fredegar and its continuations
(London, 1960), p. 51.
 , , 251

before his death, Dagobert entrusted Aega with the care of the palace
and the regnum, in the name of his young son Clovis and Queen
Nanthild. As mayor of the palace of Neustria-Burgundy, Aega acted
effectively as regent until his death in 641.32 Fredegar paints a flattering
portrait of him, only criticising his tendency to avaritia,33 and says
nothing about his hostility to Faremoutiers. Yet the mention of Aegas
avaritia may be a coded reference to his attacks on the convent.
Aegas hostility to Faremoutiers certainly predated his mayoralty, but
Jonas seems to limit his acts of violence to that very period, when
his power was at its height. Burgundofaras brother Chainulf seems
to have been count of Meaux during those years,34 and Aega must
have used either his own lands or royal lands assigned to him (the
two possibilities are obviously not exclusive) as the base from which
to attack the convent. These hostilities constituted a grave affront to
Faras family, particularly to her brothers Bishop Burgundofaro and
Count Chainulf. Did they decide to bring the case before the royal
tribunal? Certainly it was in the course of a placitum at Augers-en-
Brie, in full view of the court, that Chainulf was assassinated by
Aegas own son-in-law.35 Aegas assaults on Faremoutiers, and Chainulfs
murder by Ermenfred, were clearly linked. Here was a case of a
feud that pitted against each other two groups competing for power
in Neustria. The Faronids and Aegas family in fact belonged to two
distinct factions at the Neustrian court. The first, represented by
Chagneric, his brother Chagnoald and their kinsman Autharius, had
earlier served the Austrasian king Theodebert II. He had installed
them in the Brie, and for obvious reasons of political survival, they

32
Ebling, Prosopographie, no. XII, p. 39.
33
Fredegar, Chron. IV, 80, p. 161 (transl. Wallace-Hadrill, p. 68): Aega vero inter
citiris primatebus Neustreci prudentius agens et plenitudinem pacienciae inbutus, cumtis erat pre-
cellentior. Eratque genere nobele, opes habundans, iusticiam sectans, aeruditus in verbis, paratus
in rispunsis; tantummodo a plurimis blasphemabatur, eo quod esset avariciae deditus.
34
The link between Burgundofara and Chainulf emerges from the abbesss tes-
tament, ed. J. Gurout, Le testament de sainte Fare. Matriaux pour ltude cri-
tique de ce document, Revue dHistoire Ecclsiastique 60 (1965), pp. 761, where she
refers to property to be bequeathed to her brothers Chagnulf and Burgundofaro
and to her sister Chagnetrada. The onomastic evidence further indicates this rela-
tionship, with the name-element Chagn- present in Chagneric, Chagnulf, Chagnoald,
Chagnetrada: Le Jan, Famille, pp. 38893.
35
M. Lecomte, Lextention sud-est du pagus meldensis ou civitas meldorum au
VIIe sicle, Le Moyen Age 8 (1895), pp. 16. A placitum was an assembly at which
the king and/or some of his leading men dealt with, inter alia, legal business, includ-
ing the hearing of disputes.
252

subsequently rallied to Clothar II of Neustria when he reunited the

kingdom.36 This group also had interests at Soissons and at Laon:
to that extent, it remained orientated towards Austrasia, where, at
the young Dagoberts Austrasian court, Chagnerics son Burgundofaro
duly appeared, together with his kinsmen Ado and Dado/Audoin,
the sons of Autharius. After Clothar IIs death in 629 and Dagoberts
accession to kingship over the entire regnum Francorum, the group rose
to power in eastern Neustria and embarked on a systematic exten-
sion of territorial control there by amassing the chief honores in that
region. By the close of the 630s, the Faronids held not only the bish-
opric (Burgundofaro) and countship (Chainulf ) of Meaux,37 but had
also managed to create a monastic network within the wider Luxeuil
orbit for Columbanus in his time had had the support of this
family. First of all Chagneric had founded Faremoutiers for his daugh-
ter c. 620, and then all three of his sons, in their turn, had founded
monasteries: the eldest, Ado, founded in solo proprio a convent for
nuns at Jouarre c. 635, his brother, Rado, a monastery for men at
Rueil in patrimonio proprio, while the youngest, Dado/Audoin, founded
Rebais c. 636/7 with the help of King Dagobert who this time pro-
vided the land.38 The groups rapid ascent in eastern Neustria could
not fail to have aroused the hostility of rival groups, especially of
old Neustrian families, and not least those opposed to Columbanan
monasteries and their supporters. Jonas says that Aega was among
those who opposed the monks of Luxeuil.39 For his part, the king
supported the Faronid foundations, but his attitude here was not
without self-interest. He certainly wanted to control and counter-
balance, if not limit, the Faronids power in the Brie. In providing
the estate on which Authariuss sons would build their last founda-
tion, Dagobert was inserting himself into the network of founders
and de facto placing the monastery under his protection. Almost imme-

36
Le Jan, Famille, p. 394.
37
Burgundofaro became bishop of Meaux between 629 and 637, J. Gurout, art.
Faron, in: DHGE 16, col. 654. He succeeded Gundoald who was probably his
kinsman: Le Jan, Famille, p. 394. Chagneric had perhaps been count of Meaux
before his son Chagnulfus. Chagnoald, the eldest of Chagnerics sons, was pro-
bably born in the Portois; after becoming a monk at Luxeuil, he was bishop of
Laon by 626/7, and died c. 632/4, Gurout, art. Fare, DHGE 16, col. 515.
38
Vita Agili abbatis Resbacensis, ed. J. Mabillon, AA SS OSB 2, p. 321. For Jouarre,
see J. Gurout, Les origines et le premier sicle de labbaye, in: LAbbaye Notre-
Dame de Jouarre (1961), pp. 167.
39
Jonas, Vitae Columbani II, c. 17, MGH SRM 4, p. 269.
 , , 253

diately, he granted immunity to Rebais, and the following year got

Bishop Burgundofaro to grant it an exemption. The royal grant needs
now to be reconsidered in the light of Barbara Rosenweins recent
study of immunity. As she has so beautifully demonstrated, immu-
nity and exemption in the Merovingian period could not be disso-
ciated and were always granted to monasteries under royal patronage.
In this period, in other words, there were no royal monasteries in
the Carolingian sense of the term. Such privileges can be seen as a
flexible instrument that enabled more than one power-sharer to derive
benefits in terms of prestige and social connexions.40 Here immunity
and exemption, negotiated between the king and the group that
founded Rebais (Audoin), and held the bishopric of Meaux (Burgun-
dofaro), the county of Meaux (Chagnulf ) and perhaps already the
abbacy (if Agil was abbot by the time of the grant), reinforced the
bonds between the various parties. The prohibition on entering
the monastery tended above all publicly to affirm the kings role
as the one who directly or indirectly had control over sacred space.
The Vita Agili, moreover, late though the text is, seems to convey
an echo of tension between Audoin and the king: tension resolved
by the monasterys coming under royal authority.41 It seems note-
worthy that of the four monasteries founded by this group (Fare-
moutiers, Jouarre, Rueil, Rebais), none of the three private houses
(Faremoutiers, Jouarre, Rebais) obtained exemption and immunity,
and in particular that Bishop Burgundofaro and Count Chainulf,
who thus forbade free access to Rebaiss septa secreta, were kinsmen
of the founder and first abbot. It therefore seems likely that the king
was anxious to avoid Faronid exploitation of Rebais as a private
monastery, hence as a focal point for their political activities. This
hypothesis is strengthened by Aegas designation as mayor of the
palace a mere two years after the exemption grant, and just before
the kings death in 639. It is tempting to link Dagoberts attitude
towards the Faronids with his and his son Sigibert IIIs concern to
limit Pippinid power in Austrasia. For while Dagobert had Pippin I
follow him into Neustria after the reunifying of the regnum in 629,
as guardian for his young son Sigebert of Austrasia he chose Otto
who belonged to a kin-group consisting of the Pippinids rivals and

40
B.H. Rosenwein, Negotiating space. Power, restraint, and privileges of immunity in early
medieval Europe (Ithaca and London, 1999), esp. pp. 5996.
41
Vita Agili, c. 17, p. 323.
254

enemies. Further, when Pippin I died in 640, Sigebert chose Otto

to succeed him, thus overlooking Pippins son Grimoald. When
Grimoald, having finally become mayor of the palace, founded
Cugnon and Stavelot-Malmdy c. 646, Sigebert took action in his
turn to provide the requisite lands for the new foundations, just as
his father had done for Rebais some years before.

Clearly there were several groups in Neustria who were rivals for
power. The Faronids were one of the strongest of them, firmly based
in the Brie, and with Audoin as their man at court. Yet King
Dagobert had succeeded in maintaining a balance between the var-
ious groups. His death unleashed a crisis for the Faronids, both in
the Brie and at court. In the Brie, there began a period of violence
marked by the iniuriae against Faremoutiers and by Chainulfs mur-
der which, as Fredegar strongly indicates, only makes sense if dated
within Aegas lifetime. Aegas death changed the situation: the
dowager-queen Nanthild allowed Chainulfs kin to seek revenge. To
escape the faidosi (pursuers of the feud), the murderer had to flee to
Austrasia and took refuge at Rheims in the church of St-Rmi.42 Thus
Fara was able to withstand the assaults of her enemies, protected by
the mercy of the Lord and the prayers of the saints . . . so much so
that all those whose greed had made them her ravagers and accusers
in the past subsequently became generous and well-disposed defend-
ers. Fara kept Faremoutiers until her death. The Faronids had been
able to maintain their position only with the queens suppport, how-
ever, and Audoins appointment to the see of Rouen in 641 was a
further sign of their loss of influence at court. In the end, Faremou-
tiers came under royal control, when two Anglo-Saxon princesses
succeeded Fara as joint-abbesses: Sthryth, daughter-in-law of Anna
queen of the East Angles and thelburh, Annas illegitimate daugh-
ter.43 Jean Gurout saw here the influence of Queen Balthild, who
was also of Anglo-Saxon origin. But the presence of the Anglo-Saxon

42
Fredegar, Chron. IV, 83, p. 163 (transl. Wallace-Hadrill, p. 70): Ante paucis
diebus, Ermenfredus, qui filiam Aegane uxorem acceperat, Chainulfo comiti in Albiodero vico in
mallo interfecit. Ob hanc rem gravissema stragis de suis rebus iussionem et permissum Nantilde
a parentebus Ainulfi et populum pluremum fiaetur. Ermenfredus in Auster Remus ad baseleca
sancte remediae fecit confugium; ibique diebus plurimis hanc infestacionem devitando et rigio temore
residit.
43
Ibid., and Gurout, art. Fare, DHGE 16, col. 527. Thus, under royal pres-
sure, the convent was removed from Faronid control.
 , , 255

princesses at Faremoutiers should more probably be linked with the

crisis the convent underwent c. 641 and the likely intervention of
Erchinoald who succeeded Aega as mayor of the Neustrian palace.
Bede says that while Fara was abbess, an Anglo-Saxon princess was
alrready at Faremoutiers: this was Ercangota, daughter of King
Erconbert of Kent (64064) and Sexburh, daughter of King Anna
of the East Angles. Bede adds that the kingdom of the Angles had
few monasteries at this time, and that many of the inhabitants of
Britannia had become accustomed to enter Frankish monasteries to
practise the monastic life there. They also sent their daughters so
that they could be educated and married to the Heavenly Bridegroom,
and especially to the convents of Brie (Faremoutiers), Chelles and
Andelys-sur-Seine.44 Ian Wood has also pointed out that the royal
families of Kent and East Anglia seem to have been related to
Erchinoalds family, for the mother of King Erconberht of Kent was
probably of Frankish origin, a kinswomen of Erchinoald, mayor of
the palace.45 Erchinoald had contacts of his own with England: it
was he who offered King Clovis II his future wife, a slave-woman,
or captive, named Balthild. There cannot be much doubt that
Erchinoald pointed Ercangota in the direction of Faremoutiers, in
order to strengthen his own as well as royal influence there, nor that
the arrival of the two Anglo-Saxon princesses ought to be seen as
part of the aftermath of the crisis of 641. The Faronids could resist
their enemies successfully only with Queen Nanthilds help and also
with the support of the new mayor, Erchinoald. Since he belonged
to a different faction from that of Audoin and his kin,46 Erchinoald,
keen to enhance his own influence at Faremoutiers, caused the two
Anglo-Saxon royal women, who were also his own kinswomen, to
enter the convent so that he could prepare the succession to Fara
and hence the convents transfer to royal control. Its worth noting,
too, that Jouarre also came under the control of the royal convent
of Chelles.47

44
Bede, Historia ecclesiastica gentis Anglorum III, 8, transl. J. McClure and R. Collins
(Oxford, 1989), pp. 1224.
45
I. Wood, The Merovingian kingdoms 450751 (London, 1994), pp. 1989; see also
Fouracre and Gerberding, Later Merovingian France, pp. 1045.
46
Fouracre and Gerberding, Later Merovingian France, p. 105.
47
J.L. Nelson, Queens as Jezebels. The careers of Brunhild and Balthild in
Merovingian history, in: D. Baker ed., Medieval women (London, 1978), pp. 3177
at 702 (repr. in L.K. Little and B.H. Rosenwein eds, Debates on Medieval history
(Oxford and New York, 1999).
256

The Vita Geretrudiss account of the attacks on Nivelles throws more

light on the familial context of such acts of violence. The author
describes how Wulftrude, having succeeded her aunt Gertrude as
head of the convent, had to undergo violent pressure: It happened . . .
out of hatred of her father, that kings, queens, even priests, through
the Devils envy, wished to drag her away from her place, at first
by persuasion and later by force, so that they might evilly possess
the property of God which the blessed girl oversaw.48 Scholarly opin-
ion is divided on the date of the assaults on Wulftrude, because of
the disputed chronology of the fall of Grimoald (whether before 657,
or in 662), but all are agreed in identifying the persecutors of
Grimoalds daughter as Kings Clothar III and Childeric II along
with their mother Queen Balthild, widow of Clovis II, and Queen
Chimnechild, widow of Sigebert III. The attacks on Nivelles clearly
postdated the death of Childebert the Adopted, Grimoalds fall, and
Childerics assumption of the Austrasian throne. Childeric represented
the Neustrian branch of the royal dynasty, and he was summoned
to the throne by enemies of the Pippinids. Thus behind the kings
and queens who attacked Wulftrude lay the shadows of Wulfoald,
the new mayor of the Austrasian palace, and Gunduin, his son-in-
law. It was Gunduin who, shortly before, had murdered Ansegisil,
Grimoalds brother-in-law. This was the heart of the struggle which
over at least two generations had pitted the Wulfoald-Gunduin group,
and more broadly the Agilolfings too, against the Arnulfings and
Pippinids. Wulftrude, like her father and uncle before her, was a
victim of this hatred (odium). Her enemies used first persuasion, then
force, to drive her out of Nivelles and thus strike a blow to Pippinid
power that was intended to be fatal. Yet Wulftrude fought back.
When she died in 669, she had named her successor as Agnes, a
young nun whom Gertrude herself had brought up in the convent.

Wulftrudes adversaries found more ammunition for their assault on

the Pippinids: Childeric IIs charter for Stavelot, dated 6 September

48
Transl. slightly adapted from that of Fouracre and Gerberding, Later Merovingian
France, pp. 578, from Vita Geretrudis, c. 6, p. 460: Contigit autem ex odio paterno, ut
reges, reginae, et etiam sacerdotes per invidiam diaboli illam de suo loco primum per suasionem,
postmodum vellent per vim trahere, et res Dei, quibus benedicta puella praeerat, iniquiter pos-
siderent.
 , , 257

670,49 meant a substantial reduction in the southerly part of the royal

estates granted to Stavelot-Malmdy,50 while at the same time pre-
senting the original foundation as having been the work of Sigebert
III.51 Furthermore, this document mentioned Gunduin, along with
the two queens, Chimne-child and her daughter Bilihild, wife of the
young Childeric II. Then Bishop Theotard of Tongres-Maastricht,
a kinsman of the Pippinids, was murdered, and replaced, sometime
between 669/670 and mid-675, by Lambert, one of the Pippinids
opponents.52 But the worst was over. The set-back to the Pippinids
adversaries at Nivelles confirms the impression that after 657/662,
the Pippinids had succeeded in preserving the core of their Austrasian
power-base,53 and that at the heart of Pippinid power, Nivelles
remained one of the familys safest anchorage-points. Wulfoalds
death, followed by Childeric IIs in 675, allowed Pippin II to return
to the centre-stage of Austrasian politics. His marriage to Plectrude
had already brought him important resources around Cologne and
in the Eifel region. It was, however, in c. 679, in avenging the mur-
der of his father Ansegisil, that Pippin gave proof positive of his vir-
tus by striking down Gunduin with his own hand, as David slew
Goliath.54
As a final instance, I take the case of Abbess Anstrude, who
encountered such difficultes at Laon after founding the monastery
of St-Jean. The Vita Anstrudis tells at length of the iniuriae inflicted
on the abbess. Her enemies deployed against her all sorts of wicked
ploys, accusing her to the king of false witness so as to induce him
to expel her from the convent founded by her mother Sadalberga.55

49
MGH DD Merovingorum, ed. G. Pertz, no. 29. On this charter, see the recent
analysis of T. Klzer, Merowingerstudien I, MGH Studien und Texte 21 (Hannover,
1998), pp. 239.
50
See now R. Noel, Moines et nature sauvage dans lArdenne au haut Moyen
Age, in: J.-M. Duvosquel and A. Dierkens eds, Villes et campagnes au Moyen Age.
Mlanges G. Despy (Brussels, 1991), pp. 56297 esp. 580.
51
Sigeberts charters for Stavelot-Malmdy show that the king had given the
royal land, but that Grimoald had founded the monastery suo opere: cf. Werner,
Der ltticher Raum, pp. 105, 364.
52
Werner, Der ltticher Raum, p. 256.
53
M. Becher, Der sogenannte Staatsreich Grimoalds. Versuch einer Neubewer-
tung, in: J. Jarnut, U. Nonn, M. Richter eds., Karl Martell in seiner Zeit (Sigmaringen,
1994), p. 142.
54
Annales Mettenses priores, ed. B. von Simson, MGH SRG (Hannover-Leipzig,
1905), p. 2, transl. Fouracre and Gerberding, Later Merovingian France, p. 351.
55
Vita Anstrudis, c. 11, p. 71.
258

The chronology of these events is worth recalling. Before her death

in 655, Sadalberga had decided to pass on control of the convent
to her daughter Anstrude as soon as she should reach the age of
twenty. With this in mind, the foundress had taken advice from the
bishop of Laon and also from the local aristocracy, and she had
secured the nuns agreement. Having obtained royal permission,
Anstrude then received from the bishop her consecration as abbess.56
Thus all possible steps had been taken to assure the convents sta-
bility. Yet, a few years later, probably c. 679, Baldwin, Anstrudes
brother, was murdered. The hagiographer recounts in detail the cir-
cumstances of the crime and the abbesss grief. The names of the
murderers are not given, though, lest insult be offered to their fam-
ily: Baldwin is simply said to have fallen into a trap laid by false
friends. Summoned to a placitum at an estate in the Laonnois, Baldwin
had gone taking only two horsemen with him, and there, at the very
place which should have been one of peace, he was slain by those
whom he had believed to be his and Anstrudes friends.57 The cir-
cumstances of this murder should be set alongside those of Chainulfs
death as discussed above. The Vitas reference to peace-making (sub
falso colore pacis) and the mention of friends and faithful ones, seem
to indicate that the meeting was intended to put an end to a dis-
pute that involved Baldwin, and that the friends and faithful ones
who betrayed him were the very people who were supposed to nego-
tiate a settlement. The Vita shows Anstrude directly involved in the
dispute, for it was she who had sent Baldwin to the placitum to defend
their interests: Alas, why did I send you on the advice of the cru-
ellest of men? Would that it had pleased Heaven that this dispute
had never arisen since it was for this that I decided that you should
go and be murdered in so cowardly a fashion, you who were the
staff of my weakness!58 Baldwin was thus the convent of St-Jeans

56
Vita Anstrudis, c. 4, p. 68.
57
Vita Anstrudis, c. 5, p. 68: Erat namque ei frater carissimus nomine Balduinus; adver-
sus hunc, stimulante diabolo, concipiunt et meditantur dolos, qui videbantur esse amici et fideles
huius virginis et fratris . . . Tenerrime autem dilectus a sorore Balduinus in simplicitate cordis sui
dolose cum peste invidiae vocatur ad placitum a ministris satanae in quadam villa Laudunensis
provinciae; qui sumens secum duos tantummodo equites, tertius confestim ivit ut causa placiti ad
hostes crudeles, omnium hominum quos terra sustinet sceleratissimos. Sub falso colore pacis decep-
tus est et confossus gladiis, sicque apparuerunt impii in occisione iusti, qui videbantur esse amici;
quorum nomina et stirpem dicere, iniuriam esse putamus.
58
Vita Anstrudis, c. 6, p. 69.
 , , 259

defender, just as Chainulf had been the defender of Faremoutiers

and Grimoald of Nivelles. After a long account of the grief of the
abbess and her nuns, the hagiographer also mentions how grieved
were the inhabitants of Laon whose tutor et pastor Baldwin had been.
He was probably archdeacon of Laon, and it may well have been
planned that he should become bishop eventually: after all, the
Faronid-Gunduin group wielded control over the bishopric, and it
had always had the firm support of the abbesses Sadalberga and
then Anstrude. Yet the events described in the Vita show that Anstrudes
family was now in difficulties at Laon itself and in the Laonnois.
After Baldwins murder, the Vita says that Anstrude suffered all kinds
of iniuriae from her enemies, and that those enemies were influential
at the Neustrian court where they had gained the support of Ebroin,
at that time mayor of the palace under Theuderic III. On a visit
to Laon by the mayor and the king, Ebroin directly accused the
abbess of wrong-doing, though without making any charges stick.
Even the nuns were apparently beginning to have doubts about their
abbess, when the assurances offered to them by a man named Aglibert,
together with a miracle, finally convinced Ebroins entourage, and
then the mayor himself, of Anstrudes sanctity!

In the troubled circumstances of the years 675680, a good deal of

political reconfiguration was under way and bonds of friendship
broke, turning friends into enemies. In Neustria-Burgundy, Ebroin
was the new mayor of the palace, while in Austrasia, the Pippinids
had resurfaced after Childeric IIs death, with the return of Dagobert
II, Pippins king.59 While it is impossible to date Baldwins death
precisely, the presence at Laon of Theuderic III and Ebroin is surely
to be linked with the battle of Bois-des-Fays (Lucufao), where the
Neustrians took on the Austrasians led by Duke Martin and Pippin
II, and defeated them. This Neustrian triumph must have taken place
between 675 and 679, when Dagobert II died.60 In any event, the
political upheavals of these crucial years certainly brought changes
of loyalty in their wake. Anstrudes family was allied with the
Wulfoald-Gunduin group,61 and must surely have benefited from

59
Gerberding, The rise of the Carolingians, p. 77.
60
Here I follow Gerberdings reconstruction of events, The Rise of the Carolingians,
pp. 7884, which seems to me convincing. It is also followed by Schieffer, Die
Karolinger, p. 23. Dagobert II was indeed Pippins king.
61
Le Jan, Famille, pp. 3901.
260

Childeric IIs reunification of the kingdom to strengthen its position

in the Laonnois. Conversely, the disappearance of the Austrasian
mayor, Ebroins return to Neustria, and the revival of the Pippinids
fortunes in Austrasia,62 must inevitably have had repercussions at
Laon, weakening the grip of Anstrudes family. At the time of the
events recounted in the Vita, some of the amici et fideles of Anstrude
and her brother had clearly changed sides, and friends had become
enemies. It would be going beyond the evidence to claim that Ebroin
and his supporters were responsible for Baldwins murder, but the
attacks on Anstrude herself prove, in any case, that towards the close
of the 670s, she was not among the friends of the mayor of the
palace.
What explanation can be offered for Ebroins and Theuderics
dramatic change of attitude towards Anstrude? According to the Liber
Historiae Francorum and the Continuator of Fredegar, Duke Martin
took refuge at Laon after his defeat at the hands of Ebroin and
Theuderic. Ebroin then sent Aglibert and Bishop Reolus of Rheims
to Laon to persuade Martin to come out and surrender to the king.
It was a trap: Martin emerged from the stronghold, only to be cut
down.63 Anstrude was in the thick of these events. Perhaps she had
at first given a welcome to Martin, for he certainly seems to have
found a secure refuge at Laon. Did Anstrude then make a deal with
Aglibert, Ebroins envoy: she would hand over Martin in exchange
for the mayors protection? The Vita breathes not a word. But the
intermediary role of Aglibert, and the dramatic change of heart of
Ebroin, suddenly transformed into an amicus religionis, make this sce-
nario a highly plausible one. In any case, Ebroins death in 680
loosed a new volley of assaults on Anstrude. The assailant this time
was a man named Harvey (Chariveus), who is said by a later source
to have been count of Laon.64 He pursued her with murderous intent
right into the church of Ste-Marie, but was suddenly stricken with

62
The Vita Anstrudis shows that at the beginning of the eighth century, Anstrude,
subjected to persecution by Bishop Madelgaud of Laon, appealed to her consan-
guineus Count Wolfold of Verdun, who intervened on her behalf with Pippin II.
Count Wolfold had evidently rallied to Pippin by then.
63
Liber Historiae Francorum, c. 46, p. 320; Continuator of Fredegar, c. 3, p. 170
(Wallace-Hadrill, p. 83).
64
L. Halkin and G. Roland eds., Chartes de labbaye de Stavelot-Malmdy 1 (Brussels,
1909), n. 12 (692).
 , , 261

remorse and died. He was buried in the church.65 Anstrude was also
threatened by a man called Ebrohard. He accused her of having
offered a refuge in her convent to Gislehard against whom he was
conducting a feud, and he demanded that she open the convent
gates to him. Once inside the holy precinct, Ebrohard was struck
by divine justice and died a horrible death. The hagiographer is
specific about Ebrohards accusation: the abbess was his adversary
(adversatrix) and had favoured the side of men who spurned his lord-
ship.66 These episodes confirm both Anstrudes involvement in the
political affairs of Laon, and the fragility of her position there in
the 680s. The situation changed with Pippin IIs reunification of the
kingdom, after 687. She was at loggerheads with Bishop Madelgar,
whose persecution took the form of trying to usurp (usurpare) her
convent. After many confrontations, Anstrude appealed to the prin-
ceps Pippin, using the mediation of one of her kinsmen, Count
Wulfoald of Verdun, a former enemy of the Pippinids who had now
rallied to the triumphant mayor. Pippin received the abbesss envoys
kindly, and sent his son Grimoald to Laon with orders to tell the
bishop to stop troubling the abbess.67 The hagiographer here puts a
bland construction on a major political shift: the coming of St-Jean
into the hands of the Pippinids.68 The Gunduin group had just
definitively lost control of their foundation.

The acts of violence analysed above were certainly not gratuitous,

nor were they directed solely against the abbesses in question. The
motives behind them were far bigger, and their context was a series
of bitter power-struggles which inevitably threatened the honour of
families, because competing groups were bent on strengthening their
own position at others expense and hence, on undermining their
rivals position on their home territory. Revenge imposed its own
imperatives; and the iron laws of politics did not lose their force at
convent-gates.
In the three cases discussed here, acts of violence against abbesses
followed immediately on the deaths of their father or brother. The
sudden absence of a defender weakened the abbesss position and

65
Vita Anstrudis, c. 14, p. 72.
66
Vita Anstrudis, c. 15, p. 73: . . . quod eius fuerat adversatrix, et quod faverat partibus
eorum, qui suum despexerunt dominium.
67
Vita Anstrudis, c. 16, p. 73.
68
Semmler, Episcopi potestas, p. 312.
262

enabled her enemies to assault her directly, hence to strike her fam-
ily at the very heart of its symbolic power by seizing control of the
sacra which, up to then, had been jointly guarded by the abbess and
her kinsman-protector. In this society, politics turned on the control
and manipulation of the sacred:69 aristocratic families had to make
good their claims to a share of that key resource in order to sacralise
their own power. Churches and family monasteries, in my view,
were, along with terra salica and honores, the two wings, sacred and
profane, of a single capacity for domination. To capture the sacred
forces of rural areas in the process of christianisation, noble families
established churches on their estates, while the mightiest of them
founded monasteries and convents where the relics they had acquired
were jealously guarded. Thus was their power itself made sacred.
Convents of women were not the only type of family monastery, but
I think they constituted the crucial sacralising element of familial
power. A convent was a place where sacred and profane converged
and fused around the two poles of the family the masculine and
the feminine. This is what the foundation-stories in the Vitae of
women saints, and also the same sources accounts of attacks on
abbesses, are really all about.
The hagiographic texts on which modern historians must rely are
part of a system of representation that conveys a model of Christian
life adapted to the mentalits of the early medieval period: a heroic
model, articulated around the conflict of good and evil. The voca-
tions of women are presented in the Vitae, far more often than voca-
tions in the Vitae of male saints, as the result of a radical conversion,
which arouses the familys opposition, creates a sharp rupture with
the womans original milieu, and entails a rejection of social con-
ventions. Seen in this light, hagiographic topoi are part and parcel
of the construction of familial power, and determine the specific
place of womens foundations at the heart of family honour. Two
themes are central here: confrontation between a girl and her fam-
ily, and an ensuing reconciliation.
The theme of confrontation is often associated with that of refusal
to marry, and that is one of the commonest topoi in womens Vitae.70

69
S. Boesch Gajano, Reliques et pouvoirs, in: E. Bozoki and A.-M. Helvetius
eds., Les reliques. Objets, cultes, symboles (Turnhout, 1999), p. 263.
70
Compare the incomprehension aroused by the young Trudos refusal to go
hunting with other noble youths, Vita Trudonis, c. 4, ed. W. Levison, MGH SRM
 , , 263

The foundresses of Faremoutiers, Nivelles, and St-Jean, Laon, were

cases in point: all three were among those women depicted as hav-
ing struggled against their families to escape marriage, the normal
destiny of the female sex, and instead to vow their virginity to Christ.
In each of the three cases, the young saint discovers a precocious
vocation, and for each, this is linked with the visit of a missionary
who blesses her: Columbanus does this for Fara, Eustasius for
Salaberga, and Amandus for Gertrude. There is no reason to doubt
the reality of these missionaries influence on the minds of young
women. Yet in all three cases, the father opposed the vocation, while
the mother defended it. Though there are variants in other Vitae
(Aldegunds mother, for instance, opposed her daughters vocation),71
the classic model is that of paternal opposition and maternal sup-
port. These girls, then, are shown in the Vitae refusing to submit to
the laws of their feminine condition and the rules governing the
exchange of women. Did their conduct create real tensions within
aristocratic society? Clearly, female honour was closely bound to sex-
uality, to marriage and reproduction. Only widowhood might be said
to lend itself much better to the monastic life; and many a pious
matron was able to enter a convent after her husbands death, or
even after she had done her duty as a woman, that is, borne and
brought up children. Pious spouses are sometimes said to have cho-
sen to separate so that both could retire into monastic foundations,
as Sadalberga and Blandin/Baso did, or Waltrude and Vincent/
Madalgar.72 But did young girls refusals to obey the rules constraining
womankind seriously threaten a demographic and social equilibrium?
At first glance, a positive answer to that question might seem sug-
gested by several Vitae (those of Sadalberga, Gertrude, and Rictrude)
which show kings putting pressure on girls to marry. Dagobert him-
self wanted to marry Gertrude,73 and he also imposed a second mar-
riage on Sadalberga,74 and tried to make Rictrude remarry.75 Other

6 (Hannover, 1913), p. 278, or the family hostility provoked by a boys religious

vocation, Vita Leutfredi abbatis Madriacensis, c. 1, ed. B. Krusch, MGH SRM 7
(Hannover, 1920), p. 9. But the topos is less frequent in male Lives.
71
Hucbald of St-Amand, Vita Aldegundis virginis, c. 2, Migne PL 132, col. 862.
72
Vita Aldegundae prima, c. 4, ed. Levison, MGH SRM 6, pp. 878.
73
Vita Geretrudis, c. 1, p. 459.
74
Vita Sadalbergae, c. 6, p. 53.
75
Hucbald of St-Amand, Vita Rictrudis abbatissae Marcianensis, c. 8, ed. J. Mabillon,
AA SS OSB 2, pp. 9423.
264

sources show Merovingian kings of the first half of the seventh cen-
tury quite powerful enough to intervene in family strategies and
decree marital exchange. If the Vitae are to be credited, and the king
really did make girls and young women marry, was his aim not to
prevent families founding convents which would later turn out to
have such political importance? If so, there is a significant implica-
tion, namely, that girls who rejected marriage in favour of founding
convents did not really reduce their familys symbolic capital by caus-
ing them to lose opportunities to make new alliances and perpetu-
ate the dynastic line: instead, the founding families actually retrieved
what they gave up when a daughter chose to become a nun, or
when they offered their own children as oblates, or when they gave
land to their own monasteries. Why would aristocrats have welcomed
missionaries into their halls, and given support to their activities, had
those activities really endangered the mechanisms of social exchange?
The truth was that when young women chose the monastic life, they
served the interests of their families. Refusal of marriage was only
a topos deployed by the authors of Vitae to exalt virginity, hence to
enhance the sacredness of what were in fact joint-foundations of
father and daughter.76
What mattered in these Vita-accounts was the representation of
opposition, then of the restoration of concordia, between the mascu-
line and feminine poles of the family. The Vita Odiliae is a fine exam-
ple. It records events that happened in the late seventh century, even
though it postdates them by some eighty years.77 Odilia was the
daughter of Adalric/Eticho, duke of Alsace and his wife Berthswind.78
Odilia was born blind, and her father rejected her. He gave orders
for her to be killed or at any rate removed from the family scene.
It was Odilias mother who intervened at this point. She handed the
baby over to a wet-nurse who cared for her for a year. Odilias
identity was then discovered, the nurse had to give her up, and the

76
For a later period, that is, the central Middle Ages, see P.D. Johnson, Equal
in monastic profession. Religious women in medieval France (Chicago, 1993), pp. 145.
77
Vita Odiliae abbatissae Hohenburgensis, ed. W. Levison, MGH SRM 6 (Hannover,
1913), pp. 2450. See F. Cardot, Le pouvoir aristocratique et let sacr au haut
Moyen-ge: Saint Odile et les Etichonides dans la vita Odiliae, Le Moyen ge 89
(1983), pp. 17393.
78
There is a large literature on Adalric/Eticho and his family: see especially
F. Vollmer, Die Etichonen. Ein Beitrag zur Frage der Kontinuitt frher Adelsfami-
lien, in: G. Tellenbach ed., Studien und Vorarbeiten zur Geschichte des grossfrnkischen und
frhdeutschen Adels (Freiburg, 1957), pp. 13784, esp. 1417.
 , , 265

baby was sent to a monastery in the Jura. Bishop Ehrard of Regens-

burg, after a vision had revealed Odilias situation to him, went to
the monastery, baptised the child and thus cured her blindness. From
that moment, she advanced in holiness, with increased prayer, fast-
ing and almsgiving. Adalric/Eticho learnt of his daughters cure, but
still refused to acknowledge her. Again the bishop intervened, this
time appealing to the father directly by telling him that the dis-
sension between him and his daughter was the work of the Devil
and that he must restore concordia. For the moment, the bishop
pleaded in vain. But a day came when Odilia, threatened by enmi-
ties within the monastery where she resided, sought the help of her
brother, who arranged for her return to her natal family. The father
finally gave way, and founded the convent of Hohenburg for his
daughter. In this case, its clear that the theme of refusal of mar-
riage was not essential to the hagiographic plot, and that the key
element was the daughters heroic struggle, often with her mothers
help, against her father. In every instance mentioned above, the
father put up a lengthy resistance, even after a miracle (in Odilias
case, the curing of her blindness) had manifested the divine will. For
in conflicts that symbolised the battle of good and evil, paternal resis-
tance was as necessary as daughterly intransigence in order for her
to triumph in the end. Female sanctity, then, was represented in
action in the daughters struggle against her father and against the
social pressures he embodied. The struggle was heroic and also
unequal: the young woman nearly always emerged victorious and
this enabled her to transcend, through the sacred, the natural weak-
ness of her sex: she became a virago, one who wielded genuine power.79
Fundamental here is the theme of reconciliation and co-operation
between the two poles. It is this theme that locates the founding of
a convent within the crucible of the family, and thus enables famil-
ial honour to be sacralised. The very moment of foundation is placed
at the point of intersection between the two phases of opposition
and co-operation between the historical individuals, who are female
and male. Only the restoration of concordia allows the hagiographer
to present the foundation as the collaborative work of father (or
brother) and daughter. The relationship between the two individuals is
transformed: the daughter runs the convent, and is its gubernatrix,

79
Helvtius, Virgo et virago, pp. 189204.
266

while the father protects it, as its defensor. The bonds of kinship, far
from being weakened, are actually strengthened here, through the
co-operation of the masculine and feminine poles of the family.
Womens houses of this type are thus, far more than male houses
can be, family monasteries in the true sense of the term.
Abbesses remained closely and permanently linked to their fam-
ily,80 not so much by being subordinated to it, but because that was
their function. Abbots, by contrast, were more readily detached from
their family, and became linked into other networks. The seventh-
century abbess quite directly served her familys interests, first by
gathering and bringing up daughters, nieces, female kin, as Gertrude
brought up her niece Wulftrude as well as the future abbess Agnes,81
Sadalberga her daughter Anstrude,82 Odilia her nieces Eugenia, Atala
and Gundlinda.83 The abbess also opened her convent-gates to the
men of her family, and in so doing identified herself as wholly
involved in the play of family politics. Gertrude and Wulftrude were
in direct contact with Grimoald, who employed Nivelles in his polit-
ical strategy. The Additamentum Nivialense records a visit of Grimoald
and Bishop Dido of Poitiers to Nivelles to visit the holy places.84
The meeting certainly took place, even if there is dispute over its
date (651, or 656) and its purpose (to concert policy before Sigeberts
death, or to plan a response to it). It also shows the mayor of the
Austrasian palace having close political links with Neustrians (the see
of Poitiers was part of the Neustrian kingdom at that time). Its
already been observed that Grimoald chose to meet Dido at Nivelles
because he thought his sisters convent the most secure place in
which to devise a common political strategy at a critical conjunc-
ture. No doubt similar factors were at work when Burgundofaro and
Chagnulf were at home at Faremoutiers, or Baldwin at St-Jean,
Laon, or Odilias brother Adalbert at Hohenburg. Its no less cer-
tain that convent-gates were opened only to friends of the abbess and
her family. When Jonas of Bobbio says that Aega violated the bound-
aries (termini ) of Faremoutiers, or when the hagiographer tells of

80
M. Gaillard, Les fondations dabbayes fminines dans le nord et lest de la
Gaule, de la fin du VIe la fin du Xe sicle, Revue dHistoire de lglise de France
76 (1990), pp. 520; Helvetius, Virgo et virag, p. 198.
81
De virtutibus sanctae Geretrudis, c. 6, ed. Krusch, MGH SRM 2, p. 467.
82
Vita Anstrudis, c. 3, p. 67.
83
Vita Odiliae, c. 19, p. 47.
84
Additamentum Nivialense de Fuilano, MGH SRM 4, p. 451.
 , , 267

Ebrohards demands to Anstrude for entry to St-Jean, it is evident

that the abbesses personally controlled access to their convents and
that they reserved access to the septa secreta for their kinsmen and
friends. When the Vita Geretrudis reports Grimoald and Dido meet-
ing at Nivelles to visit the holy places, the septa secreta of this Pippinid
private monastery, it should therefore be taken absolutely at its word.
The problem of access to monasteries is linked to that of immu-
nities and exemptions, referred to above in the context of Rebais.
In the Merovingian period, so far as the extant sources go (and some
may of course have been lost), no private monastery, whether the
house was for men or for women, ever received privileges of immu-
nity or exemption. Those monasteries that did receive privileges were
all royal foundations, or founded with royal collaboration. The might-
iest aristocratic families of the seventh century gained access to the
sacred by means of the convents they founded, and, within the frame-
work of the family, the sacred seems to have located itself on the
side of the women, or, to be specific, the consecrated women. Under
the tuitio of father, brother or son, these women exercised the sacred
part of familial power and guarded the septa secreta, reserving access
for friends of the family.

This is not the place for pinning down more precisely what the septa
secreta consisted of.85 The Additamentum Nivialense refers to holy places
(loci sancti ). Relics come to mind as linked to power in their very
essence. Sofia Bosch Gajano has recently highlighted the point that
the community responsible for a relic-cults organisation transformed
the passive object of the cult into a place where miracle-working
force was concentrated.86 The relic not only had a power in itself,
which was displayed in miracles, but it also conferred a power on
whoever possessed it. The Vita Geretrudis shows how Nivelless relics,
and especially those of Gertrude herself after her death, sanctified
the community of nuns, and sacralised the power of its abbesses and,
through them, the power of the Pippinid family. Relics were among
the familys collective possessions and they were inalienable.87

85
But see the contributions to this volume of Rosenwein and De Jong.
86
Gajano, Reliques et pouvoirs, p. 259.
87
Cf. A. Weiner, Inalienable possessions. The paradox of keeping-while-giving (Berkeley,
1992); M. Godelier, Lnigme du don (Paris, 1996).
268

When Pippin IIs mother Begga was finally able to enter her con-
vent at Andenne, she sought spiritual support (adiutorium de causa spir-
ituale) from the nuns of Nivelles, and their abbess Agnes. They sent
her relics, books of Holy Scripture, and some nuns of advanced age
capable of teaching the discipline of life according to a Rule and of
setting up the new community properly. They also sent the bed in
which her sister Gertrude had died. Begga had the relics, and the
bed, solemnly carried into the convent church and placed on the
altar of St Genevieve.88 In acting thus, the nuns of Nivelles had not
alienated their relics, but simply transferred the use of them to
other members of the familia. The gift established the two commu-
nities belonging to a single familia, and a single sacred power.
To relics were soon added the tombs of foundresses. Womens
houses had a commemorative function which mens houses perhaps
never had in such an intense form. The foundress-saints were buried
in the main convent church: Fara at Faremoutiers; Itta, Gertrude
and Wulftrude at Nivelles; Sadalberga and Anstrude at St-Jean, Laon.
The commemorative function extended beyond this, however, for
these convents also found place for the tombs of bishops (or future
bishops) who were the foundresses close kin: in the crypt at Jouarre,
the tomb of Abbess Theodechild had alongside it the tombs of
Abbesses Balda and Agilberta, her kinswomen, and that of her brother
Bishop Agilbert of Paris.89 Sadalberga had the body of her brother
Leuduin/Bodo, bishop of Toul, brought to Laon so that it could be
reinterred in her convent, where Anstrude would later have her
brother Baldwin buried. Abbesses guarded the tombs of those of
their male relatives who had been, as they themselves were, conse-
crated to God. At the heart of the family convent, then, in their
septa secreta, was the place where familial power itself became sacred.
Its no coincidence that the Pippinids enemies attacked Nivelles
rather than Fosses in Brabant, also founded by Itta but a male
monastery.90
In a society of intense competition for power, and in which the
categories of profane and sacred, like those of public and private,
were deeply intertwined, family convents served to locate familial

88
De virtutibus, c. 10, p. 469.
89
G.A. de Rohan-Chabot, marquise de Maill, Les cryptes de Jouarre (Paris, 1971).
90
Dierkens, Abbayes, pp. 705.
 , , 269

power within the sacred. Consecrated women were the intermedi-

aries of this sacralisation. They could also be its victims. It is hardly
surprising that all the great womens houses discussed in this chap-
ter eventually passed into royal control.
This page intentionally left blank
 ONE SITE, MANY MEANINGS:
SAINT-MAURICE DAGAUNE AS A PLACE OF POWER
IN THE EARLY MIDDLE AGES1

Barbara H. Rosenwein

Great is the power (virtus) at the tombs of the [Theban] martyrs

wrote Gregory of Tours about the site of the saints buried at the
monastery of Saint-Maurice dAgaune.2 What sort of power did they
have? The answer is hardly simple. Nor is it the same for all time.
Since the power at the tombs was believed to be the power of God,
it demanded recognition, deference, and monumentalization. These
acts in turn intensified and complicated the ways in which the sites
power was understood. And, in yet another turn, those who tapped
into and associated themselves with the place hoped to and appar-
ently did enhance their own power and prestige. This paper is
an exploration of the changing ways in which beliefs about the power
at Agaune led people rulers, bishops, monks, and occasionally
ordinary people to reorganize and make use of the site and to
model other institutions upon some of the key features of the monastery
built there.
Agaunes very emplacement made it powerful for worldly reasons.
High upon a rocky cliff close to the Rhne river, about 40 km from
the Great Saint Bernard pass, it was a strategic point between Italy
and the north. This meant different things at different times: in the
Roman period, Agaune was a toll-collection center;3 in the Burgundian

1
I am grateful to the members of the Bellagio workshop, and in particular to
Albrecht Diem, Janet L. Nelson, Julia M.H. Smith, Ian Wood and, above all, Mayke
de Jong for their important suggestions. I thank Alessandra Antonini, Charles Bonnet,
and Franois Wibl for their generous help in introducing me to the archaeology
of Agaune. Christian Sapin supplied much-needed general orientation. Loyola Univer-
sity Chicago Research Services awarded me a Summer and Research Grant in
1998, which made it possible to for me to visit the site and write up my findings.
Loyola University Center for Instructional Design (LUCID) ably drew the figure.
2
Gregory of Tours, Liber in gloria martyrum, c. 75, B. Krusch ed., MGH SRM 1,
pt. 2 (Hannover, 1885, rev. 1969), p. 87: Magna est enim virtus ad antedictorum mar-
tyrum sepulchra.
3
M. Zufferey, Die Abtei Saint-Maurice dAgaune im Hochmittelalter (830 1258),
272 .

and Merovingian periods, it was a demarcation point, separating the

Italian south from Europes northern kingdoms; in Charlemagnes
empire, it was a symbolic hinge, connecting the conquered peninsula
of Italy to the kingdom of the Franks. Its geo-political position was
always the backdrop to Agaunes numinous powers.
Unexpected springs of water spout from crags on the site. Pagan
Romans, who incorporated aquae into their topographies of power,
dedicated Agaune to the nymphs.4 The Christians followed suit in
their own way. The Passio Acaunensium martyrum by Eucherius, bishop
of Lyon (d. 450/4), gives us the fifth-century version of the Christian-
ization of Agaune: Roman imperial troops from Thebes, called up
by emperor Maximian at the end of the third century, and led by
their commander Maurice, were decimated near Agaune for refus-
ing to kill Christians in the vicinity.5 Their bodies were discovered
about a century later, Eucherius tells us, by Theodore, bishop of Mar-
tigny (then known as Octodorum). Theodore, disciple of Ambrose, and
not to be outdone by the latters appropriation of Saints Gervasius
and Protasius,6 constructed a basilica at Agaune in honor of Maurice
and his associates now nestled against the looming rock, leaning
against it on just one side.7 As we shall soon see, this description in
the Passio inspired a compelling but wrong modern interpretation
of the archaeological evidence.
A whole monastic complex, probably dual sex, grew up around
the tombs of the martyrs and served the pilgrims who came there.
We know about this community, however, only from Eucheriuss
Passio, the anonymous Life and Rule of the Jura abbots, and the equally
anonymous Vita of Saint Severinus.8 Subsequent written sources made
a conscious effort to suppress or in any event keep mute about
this early monastery.

Verffentlichungen des Max-Planck-Instituts fr Geschichte 88 (Gttingen, 1988),

p. 29, n. 26.
4
An altar dedicated to the nymphs dating from the third century was found on
the spot. It remains today in the vestibule of the abbey.
5
Eucherius, Passio Acaunensium martyrum, c. 2, B. Krusch ed., MGH SRM 3
(Hannover, 1896), p. 33. Eucherius was almost certainly substituting Maximian for
Constantius Chlorus, father of Constantine, thus writing the latter out of the story
of Christian persecution.
6
P. Brown, The cult of the saints: Its rise and function in Latin Christianity (Chicago,
1981), pp. 367.
7
Eucherius, Passio Acaunensium martyrum, c. 16, p. 38: basilica, quae vastae nunc adi-
uncta rupi, uno tantum latere adclinis iacet.
8
Vita Severini abbatis Acaunensis, B. Krusch ed., MGH SRM 3 (Hannover, 1896),
 - 273

In 515 Sigismund, a Burgundian prince fairly newly converted

from Arianism to Catholic Christianity, rebuilt and reorganized the
site with the help of his episcopal advisors.9 To Agaunes fame as
the place of martyr-soldiers, new sources of power were now added.
They included the new monasterys extraordinary day-and-night liturgy,
which the monks carried out in relay; its symbolic embodiment
of an episcopal-royal alliance; and, soon, its stewardship of the first
royal saint in the West. For, about ten years after Sigismund and
his wife and sons were killed by the Franks in 523, the abbot of
Saint-Maurice retrieved their bodies and buried them in a church
near his monastery.10 By 590, according to Gregory of Tours, the
king was consorting with saints and Masses said in his honor were
working miracles.11
Saint-Maurice thereafter became a model monastery for Burgundian
kings. When King Guntram founded Saint-Marcel de Chalon, he
had Saint-Maurice in mind; and when Dagobert reformed his favorite
monastery, Saint-Denis, he too had the exemplar of Agaune in view.12
The Carolingian period brought changed status to the monastery
when, in the ninth century, Saint-Maurice became a house of canons.
Nevertheless, it still maintained a reputation for holy power, and in
888 Rudolf chose the spot for his coronation as king of Burgundy.13
What was so compelling about the place? Certainly Frederick
Paxton is not wrong to stress the cult of Sigismund, healer of fevers
and the first patron saint of an illness. Nor is Friedrich Prinz wrong

pp. 16870; Vie des Pres du Jura, F. Martine ed., SC 142 (Paris, 1968). On dating
the latter text to just before 515 and on the general silence regarding this pre-515
community, see: F. Masai, La Vita patrum iurensium et les dbuts du monachisme
Saint-Maurice dAgaune, in: J. Autenrieth and F. Brunhlzl eds., Festschrift Bernhard
Bischoff zu seinem 65. Geburtstag (Stuttgart, 1971), pp. 4369.
9
Nevertheless, the title of Avituss Homily 24, Dicta in basilica sanctorum Acaunensium,
in innovatione monasterii, shows that the church was being revived, not founded; see
Avitus of Vienne, Homilia 24, in: U. Chevalier ed., Oeuvres compltes de saint Avit vque
de Vienne (nouv. ed., Lyon, 1890), p. 337.
10
Gregory of Tours, Decem libri historiarum, III, c. 5, B. Krusch and W. Levison
eds., MGH SRM 1, pt. 1 (Hannover, 1951), p. 101.
11
See F.S. Paxton, Power and the power to heal: the cult of St Sigismund of
Burgundy, EME 2 (1993), pp. 95110; Gregory of Tours, Liber in gloria martyrum,
c. 74, p. 87.
12
For Guntrum, see below, at note 36; for Dagobert, see note 39.
13
Zufferey, Die Abtei Saint Maurice dAgaune, p. 95. Rudolf I was lay abbot of Saint-
Maurice before becoming king of Burgundy; see Regum Burgundiae e stirpe Rudolfina
Diplomata et Acta, eds. T. Schieffer and H.E. Mayer = Die Urkunden der Burgundischen
Rudolfinger (Mnchen, 1977), p. 93, no 1.
274 .

to highlight the monasterys extraordinary liturgy and its popular

martyr-saint, Maurice himself.14 All of this is true; but it is not the
whole truth. Agaune was powerful because it meant more than one
thing, and sometimes different things in different contexts. The remain-
der of this paper will elaborate on this point, highlighting those few
moments when the sources both material and written seem to
cluster closely enough to allow us to say something reasonable about
them. These moments are: the time of Sigismunds foundation; the
reign of King Guntram; the mid-seventh century; and the early
Carolingian period.

The foundation of Saint-Maurice marked the start of an orthodox

(i.e. catholic) royal-episcopal alliance. Indeed, I have argued else-
where that the monastery was in many ways a creation of the epis-
copacy.15 In that same study, I argue that Agaunes liturgy, dubbed
the laus perennis by modern commentators, has been wrongly attrib-
uted to the model of the staunchly orthodox Akoimetoi monks at
Constantinople. The only man associated with Agaune who might
have been in a position to know about those monks was Avitus,
bishop of Vienne and advisor to Sigismund and his father, King
Gundobad. But Avituss letters show him to be thoroughly confused
as to what was and was not orthodox at Constantinople, especially
regarding liturgical practices. On the other hand, Avitus and the
other bishops involved with Sigismund in Agaunes foundation need
have looked no further than the practices of bishops and monks in
the Rhne Valley to find extraordinary liturgical innovation. There
was precedent right at home for a liturgy of day-and-night psalmody.
When Sigismund founded Agaune, he had good political reasons
for favoring the Theban martyrs. His cousin Sedeleuba and his aunt
Theudelinda had already founded churches near the city of Geneva

14
Paxton, Power and the power to heal, esp. pp. 1059; F. Prinz, Frhes
Mnchtum im Frankenreich. Kultur und Gesellschaft in Gallien, den Rheinlanden und Bayern am
Beispiel der monastischen Entwicklung (4. bis 8. Jahrhundert) (2nd ed., Mnchen, 1988),
pp. 10212.
15
B.H. Rosenwein, Perennial prayer at Agaune, in: S. Farmer and B.H. Rosen-
wein eds., Monks and nuns, saints and outcasts: Religion in medieval society (Ithaca, N.Y.,
2000), pp. 3756.
 - 275

dedicated to Ursus and Victor, martyrs associated with Maurice.16

Sigismund thus may have been, in the words of Ian Wood, intent
on eclipsing the works of his relatives.17 In this he resembled the
bishops of Geneva, who also took keen interest in the cult. Bishop
Domitianus was known for having transferred Victors bones to
Geneva as well as for his discovery and invention of the relics of
Innocentius, another Theban martyr.18 And Bishop Maximus, ac-
cording to the Vita abbatum Acaunensium, was the advisor who incited
Sigismund to oust the vulgar crowd at Agaune and reorganize the
site in a way suitable for the martyrs, even though Agaune was not
in Maximuss diocese.19 This suggests that the bishop of Geneva may
have been interested in redrawing jurisdictional as well as spiritual
boundaries.
Bishops and prince together, then, reconfigured the topography of
the holy, setting up a monastery that suppressed an older commu-
nity of worshippers at the martyrs tombs while drawing upon a
large local repertory of cults and cultic practices for the new monas-
tic ordo there. The new monks still tended the relics of the Theban
martyrs, but they did so in a different way, and in entirely new
buildings.
This was not recognized in the 1940s and 1950s, when the archae-
ologist Louis Blondel confidently asserted that he had found the pre-
Sigismund edifices: a chapel, later enlarged into a basilica built over
the martyrs tombs against the rock; an attached baptistery; and a
hospice for pilgrims. He also found structures that he took to be
Sigismunds basilica, and he interpreted a ramp that led around the
southern and western walls of that new basilica as a pathway by

16
For Theudelinda, see: Passio S. Victoris et Sociorum, c. 2, AASS, September VIII,
p. 292; for Sedeleuba, see: Fredegar, Chronica, IV, c. 22, B. Krusch ed., MGH SRM
2 (Hannover, 1888), p. 129; L. Blondel, Le prieur Saint-Victor. Les dbuts du
christianisme et la royaut burgonde Genve, Bulletin de la socit dhistoire et
darchologie de Genve 11 (1958), pp. 21158; I.N. Wood, Avitus of Vienne: Reli-
gion and culture in the Auvergne and the Rhne Valley, 470530 (Ph.D. diss.,
Oxford, 1980), pp. 2089.
17
Wood, Avitus of Vienne, p. 217.
18
Passio S. Victoris et sociorum, c. 2, p. 292; Eucherius, Passio Acaunensium martyrum,
appendix 2, p. 41; Wood, Avitus of Vienne, p. 209.
19
Vita abbatum Acaunensium, c. 3, B. Krusch ed., MGH SRM 3 (Hannover, 1896),
p. 176: Maximus Genavensis urbis antistes . . . ad hanc devotionem Sigismundi praecordia inci-
tavit, ut de loco illo, quem pretiosa morte Thebaei martyres et effusione sanguinis . . . ornaverant,
promiscui vulgi commixta habitatio tolleretur, et . . . nitor habitantium remearet.
276 .

rocky bluff
Blondels
pre-515 basilica

Blondels
pre- 515 baptistry

rock and earth fill

passage way Sigismunds church
wall with arcosolium

ramp

tombs beneath

Carolingian church Carolingian church

western apse eastern apse

Fig. 1. Saint-Maurice dAgaune, fifth to the ninth centuries (schematised plan,

adapted from Blondel 1948, figs. 2, 3 and 5).

which pilgrims gained access to the old basilica, under which was
the primitive mausoleum (fig. 1).20
We have to thank several trees and their destructive roots for caus-
ing an emergency that sent new archaeologists led by Hans-Jrg
Lehner to the site in 1995/96.21 It should be said from the outset
that their reassessment is extremely preliminary, mostly unpublished,
and, above all, incomplete. They were able to revisit only the site
that Blondel had chosen to excavate, and that is probably too re-
stricted. The present church and buildings of the complex take up
a good deal of space below and to the southeast of Blondels exca-
vation area. But when the present community at Saint-Maurice
decided to expand their church in the 1940s, Blondel observed very

20
Blondel published numerous articles on Saint-Maurice beginning in the 1940s
and continuing into the 1960s. The most important include: L. Blondel, Les anci-
ennes basiliques dAgaune. tude archologique, Vallesia 3 (1948), pp. 957 and
Aperu sur les difices chrtiens dans la Suisse occidentale avant lan mille, in:
Frhmittelalterliche Kunst in den Alpenlndern Art du haut moyen ge dans la rgion alpine
Arte dellalto medioevo nella regione alpina, Actes du III e congrs international pour ltude du
haut moyen age, 9 14 septembre 1951 (Olten, 1954), pp. 271308, at 2839; on the
ramp, see Blondel, La rampe daccs la basilique dAgaune. Une rectification,
Vallesia 22 (1967), pp. 13.
21
H.-J. Lehner, Saint-Maurice, in: F. Wibl ed., Chronique des dcouvertes
archologiques dans le canton du Valais en 1995, Vallesia 51 (1996), pp. 3414.
 - 277

clearly (before it was covered over by modern structures once again)

the remains of an ancient (perhaps sixth-century) baptistery and an
eighth-century tomb.22 It is thus very likely that the early architectural
group at Agaune was considerably larger and more complex than
the one now on view, extending into the area now covered by mod-
ern edifices. Saint-Maurice was, after all, meant to showcase the
piety of a king and his bishops at a time when the nearby bishop
of Martigny had a huge ecclesiastical complex boasting two large
churches side-by-side,23 and when the bishop at Geneva presided
over a still more impressive compound, with two differently organized
basilicas marking out its northern and southern flanks; a baptistery;
and a grand episcopal reception hall fitted out with a magnificent
mosaic tile floor.24 It is unlikely that a royal scion and his episcopal
advisors envisaged less for their common enterprise.25
So what we see is hardly what was there. And yet what we see
is enormously suggestive. It suggests, in the first place, that nearly
everything that had been built before 515 was obliterated, whether
by chance or design. What Blondel took as the evidence for the pre-
515 basilica an apse wall turns out, upon modern inspection, to
be a Romanesque or, more likely, Gothic building. The baptistery
that Blondel had identified off the south wall of the basilica turns
out to have been a sacristy attached to the later Carolingian church
built on the site; while the mausoleum that Blondel had thought
contained the tombs of the martyrs, over which the first chapel had

22
See the state of the question and discussion of the date of this baptistery in
F.-O. Dubuis and A. Lugon, Les premiers sicles dun diocse alpin: recherches,
acquises, et questions sur lvch de Sion, pt. 3: Notes et documents pour servir
lhistoire des origines paroissiales, Vallesia 50 (1995), pp. 1359.
23
H.-J. Lehner and F. Wibl, Martigny VS: De la premire cathdrale du Valais
la paroissiale actuelle: la contribution de larchologie, Helvetia Archaeologica 25
(199498), pp. 5168, esp. 6064.
24
C. Bonnet, Les fouilles de lancien groupe piscopal de Genve (19761993), Cahiers
darchologie genevoise 1 (Geneva, 1993). I thank Professor Bonnet for a splendid
tour of the excavations. In general, the Burgundian region including the sees of
Vienne, Valence, and Lyon boasted tri-partite episcopal compounds (i.e., two
cathedrals plus baptistery) at this period. See J.-F. Reynaud, Lyon (Rhne) aux pre-
miers temps chrtiens. Basiliques et ncropoles. Guides archologiques de la France 10
(1986), esp. pp. 2930, 8997.
25
This may be especially true given the association of Theodore with Ambrose:
Ambroses Milan had a double cathedral, as did many late antique Lombard epis-
copal sees; see P. Piva, Le cattedrali lombarde. Ricerche sulle cattedrali doppie da SantAmbrogio
allet romanica (Quistello, 1990), esp. chap. 2.
278 .

been erected, in fact contains only three tombs: one dates from before
515 while the others are later (though pre-Carolingian). Some few
walls that Blondel considered to be the pre-515 hospice for pilgrims
are (according to the observations of Lehner and his group) prop-
erly to be dated to that period; but their function is uncertain. The
only buildings that Blondel plausibly got right are the church of
Sigismund (with its subsequent expansions) and the Carolingian church
with its eastern and western apses.
Nothing, therefore, suggests the pre-515 community or any of its
structures, and it is tempting to think that the site was in fact reor-
ganized as completely as possible in the time of Sigismund. Certainly
the texts about the new foundation hid and obfuscated the existence
of the earlier monks who lived on the site. But need the architec-
ture mirror the texts? Let us put the matter in its simplest form: on
the basis of our present knowledge of the site, there is no trace of
an early basilica against the rock. It is possible that we have been
looking in the wrong place; in that case, we might say simply that
Sigismund and his advisors erected a prestigious new church with-
out reference to the old. Alternatively, it is possible that the first
basilica was where Blondel sought it, but was obliterated deliberately
by Sigismunds architecture. Finally, it is just possible that the early
basilica never in fact existed.
Whatever the case, the absence of the oldest structures does not
obviate the fact that, as Charles Bonnet pointed out to me, there is
evidence of long-term continuity at the site. In particular, two hot
zones may be discerned. One ran north-south along a line marked
by the eastern apses of a sequence of churches built on the site up
to and including the Carolingian period. (It was signaled, as well,
by the Gothic apse next to the rock). The other ran parallel to, but
to the west of, the first. For the earliest period, this western axis is
clearly represented by only the smallest bit of wall. But we can see
it as a focal point in the Carolingian period, when an entire west-
ern apse was built, into which was placed the tomb of Saint-Maurice
within an arcosolium, a rectangular space topped by an archway. This
western axis may not have been neglected by Sigismunds church
either, as we shall see.
We can associate that latter church with the constructions that
Lehner calls phase 2, which appear to date from the sixth century.
Blondel confidently spoke of Sigismunds church, and in this instance
he may not be wrong. We know that Sigismunds church was ded-
 - 279

icated in 515, and we have a homily written by bishop Avitus for

the occasion.26 There Avitus calls attention to the psalmody of the
monks; he even makes up a word, psalmisonum, to emphasize the
solemn tones of the day and night liturgy there. Unfortunately, he
does not describe the church in which this liturgy took place; but
Blondel was right to think that it was a simple basilica with one
fairly elongated eastern apse.27 Blondel also thought that flanking it
to the south and gradually rising along its west end was a ramp
leading to the old basilica of the pre-515 monks. It now seems more
likely that the ramp led right into Sigismunds church. Why would
pilgrims care to enter there? I have two complementary suggestions.
First, they entered to marvel at the monks non-stop liturgy. It is
striking that no barrier has been discovered between the area around
the altar and the rest of the basilica, for such structures were com-
mon in churches of this period.28 Indeed, Gregory of Tours suggests
that laypeople were welcome to come to Agaune and listen to the
monks. His first illustration of the virtus of the tombs of the martyrs,
quoted at the beginning of this paper, features Saint Maurice com-
forting a mourning mother on the spot by inviting her to rise for
matins the next day to listen for the voice of her dead son among
the chorus of psalm-singing monks. She could do this every day of
[her] life if she liked.29
Second, pilgrims entered only after going round the west end of
the church. This may have been to allow them to visit, in some way
we cannot now determine, the relics (if such there were) along that
western axis. Certainly it is striking that a bit later another corridor
was built, wrapping itself right around the ramp, under which are
tombs that appear to date from the eighth and ninth centuries. This
was clearly a privileged burial place. But here we are getting ahead
of our story.
To return to the foundation of 515, then, the sources, both tex-
tual and material, suggest two preoccupations. One was to forget

26
On the date of the foundation, see J.-M. Theurillat, LAbbaye de St-Maurice
dAgaune des origines la rforme canoniale, 515830 environ = Vallesia 9 (1954).
27
Lehner, Saint-Maurice, modifies the sequence of apse construction, however.
28
See Bonnet, Groupe piscopal de Genve, pp. 389, and F. Oswald,
L. Schaefer, and H.R. Sennhauser eds., Vorromanischen Kirchenbauten. Katalog der Denkmler
bis zum Ausgang der Ottonen, 2 vols. (Mnchen, 19661971, reprint 1991).
29
Gregory of Tours, Liber in gloria martyrum, c. 75, p. 88.
280 .

the first community that had tended the relics of the martyrs. The
other was to create through episcopal ingenuity and royal power
a spectacularly long liturgy that would express the piety of the bish-
ops while according appropriate deference to the site and glory to
the king.

T K G

The texts for King Guntram show that the alliance between bish-
ops and kings persisted when the Merovingians took over Burgundy
and that Agaune remained a potent symbol of close royal-episcopal
relations.30 Gregory of Tours could hardly mention Agaune without
invoking the piety of kings. In his Histories, he associated the site
with the remorse of King Sigismund, penitent murderer of his own
son. In the Liber in gloria martyrum, he linked it as well to King
Guntram, Frankish king of Burgundy (56192). Indeed, in Gregorys
second (and last) illustration of the virtus of the tombs of the Theban
martyrs, he dwelt on Guntrams spiritual activities, his renunciation
of earthly pomp, and his gifts to the monks at Agaune.
For Gregory, Guntram was a bishop manqu. Indeed, he was another
Mamertus, the bishop of Vienne who (as Ian Wood describes in this
volume) created a new kind of rogation liturgy in the face of nat-
ural disasters:
as if a good bishop [Gregory writes] providing the remedies by which
the wounds of a common sinner might be healed, [Guntram] ordered
all the people to assemble in church and to celebrate Rogations with
the highest devotion. . . . For three days, his alms-giving flowing more
than usual, he was so anxious about all the people that he might well
have been thought not so much a king as a bishop of the Lord.31
At the Council of Valence in 585, Guntrams bishops met on account
of the complaints of the poor to decide what would be best for the
safety of the king, the salvation of his soul, and the state of reli-
gion.32 It is clear by the end of the document that the poor were
the monks of royal monasteries; or, more precisely, they were the

30
Pace Paxton, Power and the power to heal, p. 107.
31
Gregory of Tours, Decem libri historiarum, IX, c. 21, p. 441.
32
Concilium Valentinum, in: C. de Clercq ed., Concilia Galliae, a.511a.695,
CCSL 148A (Turnhout, 1980), p. 235.
 - 281

monks of monasteries favored by King Guntram, Queen Austrechildis,

and their two daughters, the latter three now deceased. The coun-
cil confirmed the gifts to loci sancti by these royal personages and,
calling its assent not simply worthy of bishops but a matter of divine
inspiration, turned to consider how to protect the basilicas of Saint-
Marcel and Saint-Symphorian and other places endowed by royal
largesse. It declared that whatever the royal family had given or
would give to these places whether in the ministry of the altar
or in gold and silver ornaments (speciebus) for the divine cult was
not in future to be diminished or taken away either by the local
bishop or by royal power ( potestas regia), on pain of perpetual anath-
ema. Here the alliance of king and bishops had become so intense
as to lead them to proclaim a mutual and complementary self-
restraint. In the mid-seventh century, under the impetus of a reform
movement spearheaded by the disciples of Columbanus, this self-
restraint would come to be interpreted as formal exemption.33

T -

In the mid-seventh century Agaune had two related meanings: litur-

gical and juridical. It was lauded for its liturgy and renowned for
its monastic exemption, which betokened its excellent relations with
kings.
The liturgy is easy to deal with: Fredegar, for example, writing
in the mid-seventh century,34 speaks of the psalmody of Saint-Denis
ad instar on the model of Agaune.35 The long day and night
liturgy at Agaune continued to exert its magnetic attraction.
Exemption is more complicated and involves a new kind of royal
model, though one not entirely divorced from that of Sigismund.
The new model, however, went beyond linking king to episcopacy:
the king himself became pliant and bishop-like. When Fredegar tells
us that in 584 King Guntram founded Saint-Marcel de Chalon on
the model of Saint-Maurice, what he means (as he goes on to say)

33
On these disciples, see B.H. Rosenwein, Negotiating space: Power, restraint, and priv-
ileges of immunity in early medieval Europe (Ithaca, N.Y., 1999), chap. 3.
34
W. Goffart, The Fredegar problem reconsidered, in: idem, Romes fall and
after (London, 1989), p. 322, dates Fredegar c. 658.
35
Fredegar, Chronica, IV, c. 79, p. 161.
282 .

is that Guntram, thinking like a bishop, called a council of bishops

to carry forward the task.36 Fredegar had in mind the Council of
Valence. But by Fredegars day, some of the words of this council
had been incorporated into the first charter of exemption, that for
Rebais. And Saint-Marcel itself was understood to be a precedent
for the Rebais exemption. The privilege for Rebais, drawn up c. 640,
presents Burgundofaro, bishop of Meaux, as initiating a series of
provisions directed against his own diocesan powers of jurisdiction
over the monastery of Rebais. In the first of these provisions, as
Albrecht Diem pointed out to me, the ideas and even the vocabu-
lary of the Council of Valence appear: whatever is given to the
monastery, whatever, that is, that pertains to the divine cult or func-
tions as offerings for the altar, is not to be usurped or diminished
by bishops or kings (regalis sublimitas). Diem concludes that Rebais
is a sort of mega-extension of Valence.37
Rebaiss is the first extant privilege containing an episcopal exemp-
tion. Indeed, it is probably the first ever drawn up. Yet it places
itself within a venerable monastic tradition that begins with Saint-
Maurice. Its provisions of exemption (libertas), it declares, arise not
from mere impulse (instinctu) but rather from the norms of the holy
places of Agaune, Lrins, Luxeuil, and Saint-Marcel of Chalon.
How can this be? There are no charters of exemption for any of
these monasteries prior to the one given to Rebais. I suggest that
the new mid-seventh century understanding of the right relations
between a special, royal monastery and the king and his bishops was
read back to the time of Agaunes foundation. Fredegar thought that
Guntram followed the model of bishops when he was with bishops:
sacerdus ad instar.38 He noted that Guntram called a synod of forty
bishops ad instar institucionis monasterii sanctorum Agauninsum, that is,
following the example of the foundation of Agaune, which (this was
Fredegars point), in the time of Sigismund was confirmed by Avitus
and other bishops upon the orders of the prince. For Fredegar, then,
Saint-Marcel followed the model of Agaune because it involved a
king who acted according to the model of bishops and who ratified
his foundation through them. Agaune became a type of exempt

36
Fredegar, Chronica, IV, c. 1, p. 124.
37
Quoted from a private E-mail communication. For the Council of Valence see
note 32 above.
38
See Goffart, Fredegar problem, p. 343.
 - 283

monastery through a chain of associations that led from Rebais back

to Saint-Marcel and the Council of Valence, and from thence back
to Agaune.
In 654, just a few years before Fredegar was writing, Clovis II
issued a diploma for Saint-Denis that neatly tied together the new-
style royal patronage with both exemption and the non-stop liturgy
at Agaune.39 Itself a confirmation of an episcopal exemption that
must have been very close to Burgundofaros for Rebais, the Saint-
Denis privilege linked the kings interests with those of the bishops,
and it ended with a reminder that King Dagobert, Cloviss father,
had instituted at Saint-Denis psalmody per turmas, just as it is prac-
ticed at the monastery of Saint-Maurice dAgaune. These various
ideas came together because of the mid-seventh century conviction
that episcopal exemption freed the monastery to carry out its litur-
gical round for the stability of the kingdom. The emphasis in the
mid-seventh century was on the king as an associate of episcopal
sponsors who guaranteed the monastic liturgical enterprise by stay-
ing clear of the monastery.
This perspective is echoed as well in the papal exemption for
Agaune, issued in the mid-seventh century by Pope Eugenius I
(654657). In this text, as reconstructed by Anton, the pope writes
at the behest of Clovis II (postulavit a nobis Chlodoveus), presenting his
words as confirmation of the statutes and privileges of King Sigismund
and the kings who came after him.40 The right of the brethren to
choose their own abbot is affirmed, and the pope prohibits the dioce-
san bishop from extending his ditio or potestas over the monastery;
nor may the diocesan even enter it unless invited by the abbot to
celebrate Mass; nor may he take away any of the alms given to it
by the faithful; nor, finally, may he carry off the tithes which, says
the privilege, were given to the monastery by the founder, now styled
Saint Sigismund.
By the late seventh or early eighth century, the idea that Agaune
was hands off to its diocesan bishop was enshrined in the formu-
lary of Marculf, where Agaune again paraded with Lrins and Luxeuil

39
Chartae Latinae Antiquiores: Facsimile-edition of the Latin charters prior to the ninth cen-
tury, eds. A. Bruckner and R. Marichal, Part XIII, France I, H. Atsma and J. Vezin
eds., [henceforth ChLA] vol. 13 (Dietikon-Zrich, 1981), pp. 367, no 558.
40
H.H. Anton, Studien zu den Klosterprivilegien der Ppste im frhen Mittelalter. Beitrge
zur Geschichte und Quellenkunde des Mittelalters 4 (Berlin, 1975), pp. 12 and 115.
284 .

as precedents for episcopal exemption.41 Indeed, the association be-

came general and routine. In Charlemagnes first diploma for Farfa,
for example, the monastery received a privilege said to be on the
model of those for Lrins, Agaune, and Luxeuil, namely (in the words
of the charter):
that no bishop should receive a gift for the election of the abbot; nor
have power to carry away from the monastery the crosses, chalices,
patens, books or anything else pertaining to the ministry of the church;
nor have the least power to subject the monastery to princely taxa-
tion; nor, finally, be able to exact tribute or a census from that monas-
tery of theirs.42
Thus, in the mid-seventh century Agaune was a place of power not
so much at its site on the Rhne as at the Frankish royal court,
where its neat dual symbolism as exemplar of effective liturgy
and as model of episcopal and royal synergy gave it particular
panache when kings and bishops were creating the first charters of
exemption and immunity.
This view of royal/episcopal/monastic relations did not last; the
privilege for Farfa marks the last gasp of the Merovingian tradition
of according episcopal exemptions to monasteries. Already by the
mid-eighth century bishops had virtually stopped giving out episco-
pal exemptions. In the Carolingian period, kings gave out both
exemptions and immunities, but they changed their character. By
the addition of tuitio (protection), they asserted not as in the
Merovingian period a hands-off policy but rather their very active
hands-on control over the monasteries of the empire.43

T C

Meanwhile a different aspect of Agaune was gaining new emphasis:

the organization of the monks into turmae (companies) to carry out
41
Marculf, Formulae, I in: A. Uddholm ed., Marculfi formularum libri duo (Uppsala,
1962), p. 20, no 1.
42
Diplomata Karolinorum, E. Mhlbacher ed., MGH DD 1 (2d ed., reprint Berlin,
1956), p. 141, no 98 (775): . . . ut nullus piscoporum pro electione abbatis dationem accipere
debeat et potestatem non habeat de ipso monasterio auferre cruces calices patenas codices vel reli-
quas quaslibet res de ministerio cclesi nec ipsum monasterium sub tributo ponere principum potes-
tatem minime haberet nec denuo tributum aut censum in supradicto monasterio eorum exigere
debeat . . .
43
See Rosenwein, Negotiating space, Part 2.
 - 285

their day-and-night liturgy in a church which itself had become all

the holier by organizing in focused fashion the translated relics of
the martyrs. The sources here are both material and textual. They
include the new Carolingian structures at Agaune itself and the so-
called foundation charter of King Sigismund, which Theurillat has
shown was a forgery of the late eighth/early ninth centuries. The
Passio sancti Sigismundi and the Vita Sadalbergae I also take to be
Carolingian. Though based on sixth-century materials, they may help
us to assess the meaning of Agaune in the late eighth and early
ninth centuries.
In the early Carolingian period, Sigismunds church, which had
meanwhile undergone several changes and expansions at its east end,
was knocked down and subsumed into a far larger basilica with two
apses, the western one of which had a crypt below.44 The focus of
this western end was the tomb of Saint Maurice, which was placed
in an arcosolium within its western wall. Though Lehners findings
suggest that the old access ramp was destroyed at this time, it seems
that the passage-way that limned it, while no longer opening onto
the new church, was nevertheless fitted out with windows or aper-
tures. These might have provided pilgrims with contact of some sort
with the relics along the western axis. Entry and egress was pro-
vided for the western crypt by a set of stairs. This church was more
clearly focalized and organized than its predecessor.
Efficient organization was also the theme of the Carolingian texts
concerning Agaune. Consider the so-called foundation charter of
Sigismund. Theurillat has argued plausibly that this source was forged
whole cloth in the late eighth or early ninth century.45 It is of rather
little value for the sixth century, but no one has yet bothered to put
it into its Carolingian context. It is worthwhile to make the attempt
here.
The text falls into two parts: first is the account of a huge coun-
cil purportedly taking place at Agaune in 515 consisting of 40 bish-
ops, 40 counts, and a very pliant King Sigismund; second is the
kings donation charter, which focuses on the royal properties given

44
Blondel spoke of a crypt in the eastern apse as well, but Lehner, Saint
Maurice, observed no evidence for this.
45
Theurillat, LAbbaye de St-Maurice dAgaune, p. 63. Dubuis and Lugon, Les pre-
miers sicles dun diocse alpin, pp. 1289, cite an alternative date: during the
reign of Rudolf III of Burgundy (9931032).
286 .

to the monastery. The council is presented as a dialogue between

bishops and king. The counts are there only for show. Four bish-
ops dominate: Maximus, Theodore, Victor, and Viventiolus. Histori-
cally, this is utterly impossible: Theodore was long dead by the time
of Sigismund. Rhetorically, it is extremely effective. The tone of the
proceedings is set from the start when the king abjures the Arian
heresy and asks the bishops to instruct him in the true religion. The
four do not hesitate to do so. After evoking some general principles
(for example, to live justly), Theodore gets down to brass tacks. The
immediate and pressing question is what to do about the bodies of
the Theban martyrs. Who will build churches for them? Unmentioned,
of course, is the church that we know had been built for those relics
and tended by a group of monks. The king volunteers to do what
is necessary. Theodore advises him on how to dispose of the relics:
put the ones that can be associated with a specific name Maurice
himself, Exupery, Candidus, Victor in an ambitus of the basilica
(this is, surely, a reference to the wall of the Carolingian church in
which is the arcosolium) and put the others in a well-fortified place
so that they cannot be stolen. Then have the monks there carry on
an office of perpetual psalmody day and night.
The latter is institutionalized through the prescriptions of Bishops
Victor and Viventiolus. There are to be eight groups here they
are called normae (a common term for monastic community), in most
of the other sources of the period, turmae to succeed one another
in relay for the various hourly offices. An abbot presides over all;
deacons preside over each norma. The monks are freed from man-
ual labor; their clothing, drink, and food are prescribed. They are
to sleep in one dormitory, eat in one refectory, warm themselves in
the same warming room.
The kings role, says the Viventiolus of this account, is to endow
the monastery. If the abbot runs into any problems, he is to betake
himself to the Holy See and seek help there. This is an extraordinary
suggestion: such right of appeal was a provision of only the rarest
and most up-to-date privileges of the eighth century, such as the one
Stephen II gave Fulrad of Saint-Denis in 757.46 It is so unusual that

46
P. Jaff et al. eds., Regesta Pontificum Romanorum, 2 vols. (2nd ed., Leipzig,
188588; reprint Graz, 1956), no 2331. There are two versions, of which the first,
A, is largely authentic. See A. Stoclet, Fulrad de Saint-Denis (v. 710784), abb
et archiprtre de monastres exempts, Le Moyen Age 88 (1982), pp. 20535.
 - 287

it suggests that the creation of the foundation charter of Sigismund

might reasonably be placed at the time of Fulrad. Indeed, we know
from the Liber Pontificalis that Agaune is where Stephen and Fulrad
met in 753, on Stephens trans-alpine journey to ask for Pippins aid
against the Lombards.47
However, there is another context for the text as well, one that
is broader and may be more important: the monastic reform move-
ment of the Carolingian period. Three decades ago Franois Masai
already noticed that the foundation charter of Sigismund echoed two
rules.48 One, the so-called Rule of Four Fathers, has Serapion, Paph-
nutius, and two fathers both named Macarius together in council,
each taking turns in a sort of dialogue in which they dictate their
rule. The other is the Rule of St Benedict. Both were collected in
the Carolingian reformer Benedict of Anianes Codex Regularum.49 It
is clear, however, that a monastic reformer need not necessarily have
been at Aachen or Inden to have been preoccupied with cleaning
up untidy monastic practices and making all orderly, regular, and
uniform. Indeed, it is rather likely that Agaune was the place where
the foundation charter of Sigismund was drawn up.
Although the turmae of the monks at Agaune were mentioned in
Dagoberts charter for Saint-Denis in 654, in the phrase psallencius
per turmas, this use of the term remained an isolated instance until the
Carolingian period.50 None of the sources that may be associated
with the foundation of 515 not the writings of Avitus of Vienne,
not the Vita abbatum Acaunensium, not the additions made after the
death of Sigismund to the text of the Passio Acaunensium martyrum, not
the writings of Gregory of Tours nor even the later chronicle of
Fredegar say a word about turmae. These sources certainly stress
the day and night psalmody carried out by the monks; but they are
unconcerned about its practical organization. By contrast, the
Carolingian sources can almost be so identified because of their use

47
Liber Pontificalis, c. 94, 24 in: L. Duchesne ed., Le Liber Pontificalis. Texte, intro-
duction et commentaire, 2 vols. (188692), 1, p. 447.
48
Masai, La Vita patrum iurensium, pp. 513.
49
Benedict of Aniane, Codex regularum, part 1, Migne PL 103, coll. 43542 for
the Rule of the Four; the Benedictine rule is not printed in sequence in the PL ed.
but it constituted the first Western Rule in Benedict of Anianes collection. The
most recent ed. of the Regula IV Patrum is J. Neufville, Rgle des IV Pres et
Seconde Rgle des Pres. Texte critique, Revue bndictine 77 (1967), pp. 47106.
50
ChLA 13, p. 37, no 558.
288 .

of the term. In the so-called Chronicle of the Ninth Century, the

monks are organized in nine turmae to chant their psalms.51 In the
Vita Sadalbergae, Sadalbergas nuns are distributed per turmas, ad instar
Agaunensium.52 In the Carolingian Gesta Dagoberti I, Dagoberts reform
of Saint-Denis, on the model of Agaune and (in this case) Tours as
well, has the monks chanting the psalms turmatim.53 And in the Vita
Amati, which, however, may possibly be a Merovingian text, the saint
organizes his house at Remiremont per septem turmas, presumably
inspired by Agaune, where he had once spent time as a monk.54
When a donor named Ayroenus gave a donation to the monastery
of Saint-Maurice in 765, he did so to the sacred place or indeed
to the turma Valdensis, where the monk Matulphus, the turmarius, is
seen to preside.55 Historians have interpreted this as a vestige of
the original organization at Agaune; but it might as easily mark a
newly reformed organization there.
More than mere interest in organization may be involved. The
word turma had primarily military associations.56 In Eucheriuss Passio
of the Theban martyrs, the impious squadrons who carried out
emperor Maximians evil persecutions were organized in turmae.57 In

51
For the text, see Theurillat, LAbbaye de St-Maurice dAgaune, p. 55.
52
Vita Sadalbergae, c. 17, B. Krusch ed., MGH SRM 5 (Hannover, 1910), p. 59.
53
Gesta Dagoberti I regis francorum, c. 35, B. Krusch ed., MGH SRM 2 (Hannover,
1888), p. 414.
54
Vita Amati, c. 10, B. Krusch ed., MGH SRM 4 (Hannover, 1902), p. 218. On
the possibility of this as a Merovingian text, see: I.N. Wood, Forgery in Merovingian
hagiography, in: Flschungen im Mittelalter. Internationaler Kongress der MGH, Mnchen,
16.19. September 1986, Pt. 5: Fingierte Briefe, Frmmigkeit und Flschung, Realienflschungen,
MGH Schriften 33.V (Hannover, 1988), pp. 3701.
55
M. Besson, La donation dAyroenus Saint-Maurice (mardi 8 octobre 765),
Zeitschrift fr schweizerische Kirchengeschichte/Revue dhistoire ecclsiastique suisse 3 (1909),
pp. 2946.
56
The Vulgate provides a quick overview of turmas semantic field. In Gen. 32:78,
Jacob divides his people and herds into duae turmae. They are his company, to be
sure, but one that is decidedly unarmed. In Exod. 6:26, God commands Moses
and Aaron to lead the children of Israel out of Egypt per turmas suas. Here turma
is used in place of cognatio. Nevertheless here we are not far from military mean-
ing, for these same cohorts will (in Num 1:52) pitch their camp per turmas et cuneos
atque exercitum suum. In 1 Chron. 27 the kings army is organized in turmae in com-
panies of 2400 men. Nevertheless, in 2 Chron. 35:10 the Levites stand in turmis to
take part in the rites of Passover. Clearly the meaning of turma deserves special
study; but from the evidence here adduced, we may say fairly certainly that it
implies more than a simple group: it is an organized band, under a leader, and,
while not necessarily armed, it can quickly become so.
57
Eucherius, Passio Acaunensium Martyrum, c. 1, p. 33.
 - 289

Avitus of Viennes poem on the deeds of the Jews, the word is equiv-
alent to an army cohort.58 In Prudentiuss Psychomachia the virtues
are drawn up in turmae.59 In Gregorys Moralia in Job the Chaldaean
army forms three turmae.60
Using the word turma thus gave a particularly militant cast to the
efficacy and singleness of purpose of monastic psalmody. In the
Carolingian period it became a kind of shorthand for the monastic
corporation as a whole: Lorsch was a monachorum turma in a charter
of protection issued by Charlemagne c. 772/3, and Fuldas abbot
Sturm presided over turmae monachorum in a charter of 779.61 It is
useful to note in this regard that visual representations of the sol-
dier-martyr Saint-Maurice began to be produced only in the ninth
century.62 In the Passio Sigismundi regis, the king sets up his choirs of
psalm-singers at Agaune ad instar caelestis militiae.63 Monasteries had
always been understood as a kind of religious army, but in the
Carolingian period the liturgy itself was militarized. This may be
connected with its renewed emphasis on prayer for the dead.64
Via a rapprochement of material and written sources, we have
seen that Agaune was a powerful holy place in part meaning a
model holy place for a very long time. The king, his bishops,
and the military martyrs they honored there were always paramount
in the power that it exerted. What is more interesting is that those
elements were paired with different ones, hence given different mean-
ings, at different times. In 515 and shortly thereafter, they were tied

58
Avitus, Poematum libri VI, bk. V, in: Chevalier, Oeuvres compltes, p. 78: Post quos
belliferae disponunt arma cohortes,/Ducunt et validas instructo robore turmas.
59
E.g. Prudentius, Psychomachia, l.14, J. Bergman ed., CSEL 61 (Wien, 1926),
p. 170: ipse salutiferas obsesso in corpore turmas depugnare iubes.
60
Gregory I, Moralia in Job, II, c. 15, M. Adriaen ed., CCSL 143 (Turnhout,
1979), p. 75.
61
MGH DD 1, p. 105, no 72 (Lorsch); p. 177, no 127 (Fulda).
62
D. Thurre, Culte et iconographie de saint Maurice dAgaune: bilan jusquau
XIIIe sicle, Zeitschrift fr schweizerische Archaeologie und Kunstgeschichte/Revue suisse dart
et darchaeologie/ Rivista svizzera darte e darcheologia 49 (1992), pp. 718.
63
Passio Sigismundi regis, c. 6, B. Krusch ed., MGH SRM 2 (Hannover, 1888),
p. 336.
64
On Carolingian prayer for the dead, see: M. Lauwers, La mmoire des anctres,
le souci des morts. Morts, rites et socit au moyen ge (diocse de Lige, XI eXIII e sicles)
(Paris, 1997), pp. 94100. The writings of Gregory the Great already expressed
some themes, especially regarding the efficacy of prayer for the dead, that were
later picked up by the Carolingians. On Gregorys views, see: J. Ntedika, Lvocation
de lau-del dans la prire pour les morts. tude de patristique et de liturgie latines (IV eVIII e
sicles) (Louvain, 1971), pp. 5960, 10510.
290 .

to an emphasis on episcopal liturgical innovation. In the mid-

seventh century, they suggested a model of freedom from episcopal
control. In the early ninth century they were harnessed to an ideal
of organization and militant liturgy.
These were not contradictory representations: episcopal will and
freedom from episcopal control went hand in hand in the mid-
seventh century, and liturgy by turmae was itself a reflection of episco-
pal creativity and royal and soldierly militancy. There is every reason
to think that at some level all these facets coexisted at Agaune from
the time of its reorganization in 515. Nevertheless, the reason that
it is important to tease out various emphases and subtleties of mean-
ing is to quell our impulse to generalize. If we read that a monastery
was set up on the model of Agaune, we should not jump to the
conclusion that such a monastery carried out the laus perennis. (Indeed,
it most certainly did not carry out the laus perennis.) If we are speak-
ing about a mid-seventh century monastery, it is very much more
likely that the place had an episcopal exemption or wanted one.
If our source is from the late eighth century, the monastery in ques-
tion was probably organized by turmae and performed an aggressive
non-stop liturgy or at least hoped to do so.
All this leads to a final, general hypothesis. It is that for a place
of power to be lasting, it must have the same sort of complexity as
a great piece of music, so that in each era new maestri can tease out
different timbres and themes. So it was with Agaune in the early
middle ages.
 MONASTIC PRISONERS OR OPTING OUT?
POLITICAL COERCION AND HONOUR IN THE
FRANKISH KINGDOMS*

Mayke de Jong

In 818, in the aftermath of Bernard of Italys revolt against Louis

the Pious, the initial death sentence for Bernard the emperors
nephew and other ringleaders was converted into blinding, a pun-
ishment which cost Bernard his life.1 But as the Royal Frankish
Annals put it, most of Bernards followers were treated more leniently.
All the bishops involved were deposed, and mancipated to monas-
teries as a matter of course; as for the laity, those most guilty were
exiled, while those deemed to be more innocent were to be ton-
sured, to live in monasteries.2 Only three years later, however, the
tide had turned. At the assembly of Diedenhofen in 821 the emperor
declared a general amnesty, allowing all insurgents to leave their
monasteries as well as their involuntary clerical state. Most of the
former rebels probably availed themselves of this opportunity, although
one author noted that some chose to stay, now giving freely to God
what they had been forced to offer ignominiously and against their
will.3 The rebellion of 830 elicited a similar response. According to

* This chapter is dedicated to my friend and colleague Piet Leupen: a belated

gift for his 60th birthday, and a contribution to the Festschrift he did not want us
to write. But something of the sort emerged all the same; see also Frans Theuws
chapter in this book. Furthermore, I am grateful to Barbara Rosenwein, Albrecht
Diem and Rosamond McKitterick for their helpful comments on earlier drafts.
Rosamond McKitterick also kindly checked my English. With regard to the rhetoric
of the sources I have learned a lot from Philippe Buc.
1
About this revolt, see J. Jarnut, Kaiser Ludwig der Fromme und Bernhard
von Italien. Die Versuch einer Rehabilitierung, Studi Medievali 30 (1989), pp. 63748.
2
Annales regni Francorum, ed. R. Rau, Quellen zur karolingischen Reichsgeschichte
I (Darmstadt, 1974), s.a. 818, p. 148: . . . coniurationis auctores . . . iussit orbari, episco-
pos synodali decreto depositos monasteriis mancipari, caeteros, prout quisque vel nocentior vel inno-
centior apparebat, vel exilio deportari vel detondi atque in monasteriis conversari. Cf. Anonymus
(Astronomer), Vita Hludowici, c. 30, ed. E. Tremp, MGH SRG 54 (Hannover, 1995),
p. 386.
3
Paschasius Radbertus, Vita Adalhardi, c. 50, Migne PL 120, col. 1534: Tum
deinde quorumdam tonsura propter furoris saevitiam illata transiit ad coronam, et dant Deo sponte,
quod dudum inviti quasi ad ignominiam susceperant.
292

the Astronomer, Louis the Pious displayed a truly imperial leniency

towards his opponents: He ordered the laymen to be tonsured in
appropriate places, the clerics to be locked up in suitable monas-
teries.4 Only three months later, however, Louis decided to become
even more lenient, returning their property to the insurgents and
giving those already tonsured a choice between remaining clerics or
returning to the lay state.5
The use of religious communities as places of internal exile was
no Carolingian novelty. For centuries Frankish kings had despatched
dangerous rebels to monasteries: sons, relatives, bishops and lay aris-
tocrats. In German legal history this phenomenon has become known
as Klosterhaft (monastic imprisonment), Zwangstonsur (forced ton-
sure), or politische Mnchung (making someone a monk for polit-
ical reasons).6 Behind these overlapping concepts is the idea that
monasteries were the prisons avant la lettre of early medieval states,
and that kings controlled monastic space, to the extent that they
could turn them into something resembling a prison. There is a gen-
eral assumption that the monasteria that harboured such political exiles
were royal ones, even to the extent that monastic exile has become
one of the ways of identifying royal monasteries.7
Monastic imprisonment is a misleading notion: this is the point
I want to get across in this chapter. This holds true even for the
Carolingian age, when rulers and bishops had more control of mon-
astic space than ever before. Heavy-handed Carolingian protection

4
Anonymus, Vita Hludowici, c. 45, p. 464: . . . sed usus, ut multis visus est, leniori
quam debuit pietate (. . .), laicos quidem praecepit locis opportunis attundi, clericos vero in con-
venientibus itidem monasteriis custodiri.
5
Anonymus, Vita Hludowici, c. 46, p. 338: Ipso denique tempore consuetae non immemor
misericordiae, quae sicut de se ait Iob, ab initio crevit cum illo, et de utero matris videtur cum
ipso egressa, eos quos dudum exigentibus meritis per diversa deputaverat loca, evocatos bonis pro-
priis restituit; et si qui attonsi fuerant, utrum sic manere, an in habitum redire pristinum vellent,
facultatem contribuit.
6
K. Sprigade, Die Einweisung ins Kloster und in den geistlichen Stand als politische
Massnahme im frhen Mittelalter (Heidelberg, 1964); W. Laske, Das Problem der Mnchung
in der Vlkerwanderungszeit; Rechtshistorische Arbeiten 2 (Zrich, 1973); idem, Zwangsau-
fenthalt im frhmittelalterlichen Kloster. Gott und Mensch im Einklang und Wider-
streit, Zeitschrift der Savignystiftung fr Rechtsgeschichte, kanonistische Abteilung 95 (1978),
pp. 32130; K. Bund, Thronsturz und Herrscherabsetzung im Frhmittelalter; Bonner
Historische Forschungen 44 (Bonn, 1979).
7
Cf. I.N. Wood, The Merovingian kingdoms, 450 751 (London, 1994), p. 195;
F. Prinz, Frhes Mnchtum im Frankenreich. Kultur und Gesellschaft in Gallien, den Rheinlanden
und Bayern am Beispiel der monastischen Entwicklung (4. bis 8. Jahrhundert) (2nd ed.,
Darmstadt, 1988), p. 155.
 293

(tuitio) of royal monasteries implied their self-evident use as places

where political opponents might be banished, and with greater expec-
tations of permanence, for monastic exile was now perceived as a
public penance, which, according to some, entailed life-long obligations.
As becomes clear from the reports on the rebellion of 818 just cited,
however, older and more more flexible forms of monastic exile con-
tinued to exist in the ninth century, so even for this period, the con-
cept of monastic imprisonment remains an inadequate one. It evokes
an eminently modern institution,8 and, moreover, it takes it for
granted that early medieval rulers possessed the powers of coercion
usually associated with the modern state. But what kept these pris-
oners inside the monastic confines? Unless we imagine abbots and
monks as jailers, rattling their keys, or members of the abbots sec-
ular retinue keeping constant guard, some measure of co-operation
from those exiled to religious communities must have been involved.
What made monastic space suitable as a location for inner exile?
We shall see that this was connected not to the development of monas-
teries as royal prisons but rather as sacred places. Their sacrality
was formally recognized, at least at times, by royal immunities and
episcopal exemptions. These created hands-off zones that enhanced
royal authority by allying the king to sacred spaces out of bounds
to secular power, including the representatives of royal might.9 If
monasteries enjoying a royal immunity or royal protection (tuitio)
immunitys Carolingian successor were indeed the ones where
prominent political opponents ended up, and this seems to have been
the case, this is difficult to reconcile with the notion of Klosterhaft.
There are other problems. The implicit assumption of those writing
about politische Mnchung is that prominent rebels exiled to monas-
teries received a clerical tonsure as a preparation for monastic vows
or a promotion to higher ecclesiastical orders; their clerical tonsure
was meant to be irrevocable, and their return to the world there-
fore amounted to apostasy.10 Clerical tonsure was a more open-ended
affair, however. Clerici, the lowest order of the ecclesiastical hierarchy,
were betwixt and between; as long as they were not admitted into

8
M. Foucault, Surveiller et punir. La naissance de la prison (Paris, 1975).
9
The two recent fundamental discussions are: W. Davies and P. Fouracre eds.,
Property and power in the early Middle Ages (Cambridge, 1995); B.H. Rosenwein, Negotiating
space. Power, restraint and privileges of immunity in early medieval Europe (Ithaca, 1999).
10
Sprigade, Einweisung, pp. 445; Bund, Thronsturz, pp. 3423.
294

higher orders or made monastic vows, a return to the world outside

remained possible. To my mind, this flexibility was an integral part
of a practice that I would prefer to call monastic exile rather than
monastic imprisonment. The high-born happy few that were allowed
to leave the political arena unscathed, retreating into monastic space,
were never meant to become monks immediately, or, for that mat-
ter, monastic prisoners. Both parties kept their options open; this
was the aim of the operation.
Monastic exile represented the honourable option out of a polit-
ical predicament, on both sides. Recent historiography has concen-
trated on the opposition between political and religious motives. Did
the powerful who withdrew into a monastery do so because of polit-
ical coercion, acting under pressure, or was it a matter of a reli-
gious conversion? This either/or question has proven notoriously
difficult to answer, for often early medieval authors are far from
unanimous on this point. There is a reason for this: early medieval
authors reconstructed or construed such events post-hoc, and they
had honour on their minds. Whether a retreat into a monastery was
voluntary or not mattered deeply to the reputation of the political
actors involved. After the event, reputations might be made or bro-
ken by portraying it either as a voluntary and honourable decision,
or as its despicable opposite. The rhetoric of early medieval authors
intent on portraying monastic exile, one way or the other, should
be taken into account. It will bring us no closer to what actually
happened or to the actors personal motivation, but it will shed
some light on the values religious and secular that informed
monastic exile, and on monastic space as a political time-out zone
allowing conflicts in the outside world to be suspended or resolved.

C ?

The discussion about so-called monastic imprisonment intersects with

a debate concerning supposedly decreasing levels of political violence
in the Frankish kingdoms. Were Carolingian rulers less inclined to
kill their political opponents than their Merovingian predecessors?
Jrg Busch has recently answer this question in the affirmative.11

11
J.W. Busch, Von Attentat zur Haft: Die Behandlung von Konkurrenten und
Opponenten der frhen Karolinger, Historische Zeitschrift 263 (1996), pp. 56188.
 295

From Charles Martel onwards, Busch argues, rulers were wary of

killing competitors and powerful opponents, resorting to Klosterhaft
instead. According to Busch, this increasing leniency cannot be ex-
plained by the fact that that many Carolingian insurgents came from
the ruling family itself, for Merovingians had their relatives killed
without any qualms.12 Supposedly, it was the increasing Christian-
isation of actual relations of power that was the main cause of the
Carolingian mildness towards opponents.13 In Buschs view, monas-
teries played a crucial role in this process, not only as the recipients
of monastic prisoners, but also from Charles Martels sons
onwards as the educators of young princes and the mediators of
Christian norms.14 In other words, Carolingian kings were less vio-
lent because they were better Christians than their predecessors.
In any long-term view, the notion of ever decreasing levels of
violence due to Christianisation, civilisation, or both is unten-
able. It derives from nineteenth-century evolutionist dreams of peace
and order that were rudely interrupted by the massive state-directed
violence of the twentieth century.15 If Carolingians used less politi-
cal violence, what about subsequent and presumably even more
Christian medieval dynasties? Furthermore, those who credit Christian-
ity with autonomous powers capable of containing political violence
seem to treat a complex and historical religion as a supra-historical
and unchanging phenomenon. To love ones neighbour as oneself is
indeed a central Christian tenet, but so is the command to fight
valiantly against Israels enemies, an injunction foremost in the minds
of those ruling the New Israels of the early medieval West.16 Many

12
Busch, Vom Attentat zur Haft, p. 571.
13
Busch, Vom Attentat zur Haft, p. 576: . . . eine zunehmende Verchristlichung
des tatschlichen Herrscherverhaltens und nicht blo eine Verchristlichung der ein-
schlgigen Normen. In a similar vein, ibid., p. 584: Verchristlichung und damit
schlielich einhergehend Verrechtlichung sind allgemeine Phnomene des 8. Jahr-
hunderts.
14
Busch, Vom Attentat zur Haft, pp. 5845.
15
This kind of evolutionism is certainly a problem in Norbert Eliass ber den
Prozess der Zivilisation, 2 vols. (Bern/Mnchen, 1969). This was the second edition;
a first had appeared in 1939, but remained largely unnoticed. For a thoughtful cri-
tique, see B.H. Rosenwein, Controlling paradigms, in eadem ed., Angers past. The
social uses of an emotion in the Middle Ages (Ithaca, 1998), pp. 23347.
16
About the ramifications and limits of the Frankish New Israel, see M. Garrison,
The Franks as the New Israel? Education for an identity from Pippin to Charle-
magne, in: Y. Hen and M. Innes eds., The uses of the past in the early Middle Ages
(Cambridge, 2000), pp. 11461.
296

different Christianities have existed, and few of those happened to

comply with modern notions of what real Christianity should look
like. If the Carolingians indeed killed fewer of their powerful oppo-
nents than their Merovingian predecessors had, explanations should
be looked for in a different direction. To begin with, which options
did rulers have to suppress insurgence, and what was to be gained
from either killing political opponents or from sparing their lives?
The answer varies according to specific political circumstances. Exec-
ution or murder was not necessarily the safest method of eliminat-
ing the competition, certainly not if rulers needed to win the loyalty
of the aristocratic backers of the insurgents. A need to accommo-
date, and a sharp eye for the long-term consequences of an irrevo-
cable killing may have guided the decision to refrain from eliminating
the opposition. Paul Fouracre has suggested that the Carolingian
rulerss more lenient treatment of political opponents might be ex-
plained by a lingering precariousness in their position as kings, which
made violence especially within the royal family particularly
dangerous.17 After all, the Carolingians were usurpers. One might
perhaps add that within this huge empire a loyal Reichsaristokratie was
a scarce commodity, so it may have been expedient to spare insurgents
who in due course might become loyal fideles once more. Yet the
Carolingian rulers themselves did not think they had moved beyond
killing their opponents, including members of their own family
or should we take the Divisio regnorum of 806 as an example of the
self-control of Christian kings, as Busch seems to do?18 This capitu-
lary shows what Charlemagne thought his sons capable of: he for-
bade them to kill, blind, maim or forcibly tonsure their younger
kinsmen.19 These remained the options open to any early medieval
ruler, the Carolingians not excepted; if the latter resorted to politi-
cal coercion by means of forcibly tonsuring their enemies and exil-
ing them to monasteries, this must have been politically expedient,
even if this happened to conform to Christian principles of forgive-
ness. To complicate the argument even further: in the Divisio regnorum

17
P. Fouracre, Attitudes towards violence in seventh- and eighth-century Francia,
in: G. Halsall ed., Violence and society in the early Medieval West (Woodbridge, 1998),
pp. 6075, esp. p. 70. Fouracre has also noticed a decrease of political violence at
the highest political level, but attempts to connect this to changing literary tradi-
tions and political structuresa much more sensible approach.
18
Busch, Vom Attentat zur Haft, p. 575.
19
Divisio regnorum (806), c. 18, MGH Cap. I, no. 45, pp. 12930.
 297

of 806, forcible tonsure figures as the weakest form of agression

against royal kinsmen, but it also could be the result of an act of
royal pardon that enhanced the kings authority; as such, monastic
exile functioned in the aftermath of the revolts against Louis the
Pious or, at least, in the court-oriented historiography portraying
Louiss leniency in this light. Similarly, in 788 at Ingelheim the
Franks unanimously (una voce) sentenced Duke Tassilo of Bavaria to
death for treason. Charlemagne, however, moved by mercy and love
of God and conscious of the fact that Tassilo was his kinsman,
decided otherwise. As the Royal Frankish Annals put it, Charlemagne
asked Tassilo what he wished to do; the latter requested to be given
leave to be tonsured, enter a monastery and do penance for his
many sins, so he might save his soul.20 This rendering of the event
enhanced the kings reputation in various ways, according to values
that were not wholly identical, but happened to correspond and
mutually reinforce each other: mercy and the love of God, an emi-
nently royal self-control, the capacity of the king to pardon where
others had called for revenge, restraint in dealing with ones kinsmen,
respect for a high-born adversary, hope of salvation.21 Contradictory
biblical precepts were embedded in new and complex cultural con-
texts; rather than invoking Christianisation as the inevitable agent
of civilisation, we should concentrate on historical varieties of Chris-
tianity in short, on Christianities, plural.

M,

Without any doubt there were royal prisons in the Frankish king-
doms; these tend to surface in hagiographical texts, for the liberation
of prisoners was one of the favourite miracles of Merovingian saints.
Three decades ago, Frantisek Graus gathered a wealth of texts doc-
umenting these liberation miracles, a body of evidence that still awaits

20
Annales regni Francorum s.a. 788, p. 56: Ille vero postolavit, ut licentiam haberet sibi
tonsorandi et in monasterio introeundi et pro tantis peccatis paenitentiam agendi et ut suam sal-
varet animam. About political violence in the aftermath of insurgence against Charle-
magne, see Fouracre, Attitudes towards violence, pp. 6870.
21
These dynamics are fully present in the difficult years 828833, and in Louis
the Piouss dealings with prominent enemies, such as his kinsmen Adalhard and
Wala; these issues will be explored in my forthcoming book The penitential state.
298

further analysis.22 We should be wary of turning a hagiographical

clich into a straightforward report on early medieval prison condi-
tions, but some features are too recurrent to be merely stereotypical.
Merovingian royal prisons were located in cities (civitates), and they
contained the more deplorable specimens of humankind: helpless
men in chains who were about to be executed, without honour or
recourse to powerful friends except the saints to whom they cried
out from their dungeons. This must have been not unlike the dis-
honourable imprisonment against which Visigothic members of the
political lite vociferously protested in 683: they did not deserve dis-
honourable incarceration and humiliation, and demanded the libera
custodia, either at the court or in the monastery, that was fitting to
their high status.23 By early medieval definitions, imprisonment in a
carcer was a grievous defamation. When St Paul remained in Rome
for two years, he enjoyed libera custodia, that is, a liberty of move-
ment within certain restrictions; he could not leave the city, but he
could write his letters and lived in the style befitting an apostle. This
was Bedes view, and it was reiterated by Hrabanus Maurus.24 It is
difficult to get a good idea of the restraints suffered by high-born
monastic exiles, but these comments admittedly from an entirely
different context on the honourable nature of libera custodia are
more informative in this respect than the vast hagiographical dossier
on prisoners liberated by saints. For the Carolingian aristocrat Nithard,
for example, the notion of libera custodia was intricately connected
with a retreat to the monastery, and the same was true of Hincmar
of Rheims.25

22
F. Graus, Die Gewalt bei den Anfngen des Feudalismus und die Gefange-
nenbefreiung der merowingischen Hagiographie, Jahrbuch fr Wirtschaftsgeschichte 1
(1961), pp. 61156.
23
XIII Conc. Tolet. c. 2, ed. J. Vives, Concilios visigticos e hispano-romanos (Barcelona,
Madrid, 1963), p. 417: . . . hos sine aliquo vinculorum vel inuriae damno sub libera custo-
dia consistere oportebit . . .
24
Hrabanus Maurus, Enarrationes in Epistolae B. Pauli, Migne PL 111, col. 1378D.
Also: Haymo of Halberstadt, In epistolam II ad Thimotheum, Migne PL 117, col. 810B:
Nam cum venisset Romam, duobus annis mansit in libera custodia, et in suo conductu, et postea
transivit ad alias nationes quae erant in circuitu Romae. Nam quando ista scribebat, in libera
custodia erat: et quia statim ut adductus est, non est interfectus idcirco dicit se liberatum.
25
Nithard, Historiae I, c. 2, ed. Ph. Lauer (Paris, 1964), p. 8: Hinc autem metuens
ne post dicti fratres populo sollicitato eadem facerent, ad conventum publicum eos venire praecepit,
totondit, ac per monasteria sub libera custodia commendavit; ibid., I, c. 3, p. 10 Et Lodharius
quidem eo tenore republica adepta, patrem et Karolum sub libera custodia servabat. Cum quo
monachos, qui eidem vitam monasticam traderent, et eamdem vitam illum assumere suaderent, esse
 299

In the Frankish kingdoms, a small number of religious communi-

ties for example, St Marcel in Chalon and St Symphorian in
Autun, St Maurice dAgaune, Luxeuil, St Wandrille, St Denis and
Chelles became places where powerful political opponents might
be sent to or retreat to of their own volition in the expecta-
tion that their lives would be spared as long as they remained within
the monastic confines. Yet such expectations are only possible if
monastic space is perceived by all concerned as a separate territory
with clear boundaries. Royal immunity and/or episcopal exemptions
helped to reinforce this sense of integrity, but before monastic com-
munities could become the beneficiaries of such privileges guaran-
teeing the inviolability of monastic space, they first had to become
identified with well-defined places that enjoyed a measure of stability
through time. A monastic community moving elsewhere to retain its
ascetic standards, leaving its unsatisfactory abbot behind,26 was of no
use to the rulers and bishops granting such privileges. They had
sacred places in mind, not saintly people.
This type of monasticism, so familiar to those dealing with the ninth
century and beyond, gradually emerged in the West from the mid-
sixth century onwards. It was a monasticism that did not favour free-
floating ascetiscism, but instead, well-defined sacred places with an
army of prayer that was guaranteed to remain in situ for the duration
of its members lives.27 This place-bound religious life, enhanced by
the presence of powerful relics within the monastic confines, enabled
the powerful to have a stake in sanctity, and to control these precious
resources to a greater or lesser extent by showering monasteries and

praeceperat. Cf. also Hincmar, Consilium de poenitentia Pippini regis, Migne PL 125, col.
1122B: Reconciliatus autem benigne tractetur, et tali loco sub libera custodia misericorditer cus-
todiatur, ut custodes monachos ac bonos canonicos habeat, qui eum exhortentur, et quorum doct-
rina et exemplo bene de caetero vivere et praeterita peccata plangere discat.
26
Regula cuiusdam patris ad monachos c. 20.5, ed. F. Villegas, La Regula cuiusdam
patris ad monachos. Ses sources littraires et ses rapports avec la Regula monachorum
de Columban, Revue de lhistoire de la spiritualit 49 (1973), pp. 336cf. p. 26.
27
M. de Jong, Carolingian monasticism: the power of prayer, in: R. McKitterick
ed., The New Cambridge Medieval History II, c. 700 c. 900 (Cambridge, 1995), pp.
62253; for a perceptive discussion about the emergence of this kind of place-bound
sanctity, see P. Fouracre, The origins of the Carolingian attempt to regulate the
cult of the saints, in: J. Howard-Johnston and P.A. Hayward eds., The cult of the
saints in late Antiquity and the early Middle Ages. Essays on the contribution of Peter Brown
(Oxford, 1999), pp. 14365. About very different patterns of insular and Breton
sanctity, see J.M.H. Smith, Saints, miracles and relics in Brittany, Speculum 65
(1990), pp. 30943.
300

nunneries with lavish gifts, exemptions, immunities and protection

(tuitio). The members of religious communities became sanctified by
the purity of the sacred location they had entered, rather than by
their individual asceticism. The rise of child oblation as the pre-
dominant way of recruiting new monks and nuns is the most telling
symptom of this development. The innocent children that entered a
locus sanctus were pure, and expected to retain this purity as long as
they remained isolated within the monastic confines, guarded against
contaminating contact with the world outside.28 This type of monas-
ticism, with its sharp boundaries delineating the inner sanctum of the
monastery and the liminal zones surrounding this so-called claustrum,
became increasingly harnessed to the salvation of the world outside.
The rulers were the first to avail themselves of the benefits of these
stable sacred resources; grants of immunity or exemption were aimed
at safeguarding the sanctity of a specific place, where the purity of
prayer might enhance the wellbeing of the rulers and their families,
and the stability of the realm.
There is nothing self-evident about this development, except with
the hindsight informed by the Carolingian order, which had come
to depend on the power of monastic prayer to an even greater extent.
The background and context of royal immunities and episcopal
exemptions has recently been explored and re-interpreted by Barbara
Rosenwein; Albrecht Diem has now provided us with a detailed view
from within, that in many ways complements Rosenweins analysis.29
Diem has charted the transformation of monasticism in late antique
Gaul and the Frankish kingdoms, revealing how informal congrega-
tions of individual ascetics striving for personal salvation gradually
turned into sacred places (loci sancti) filled with monks and nuns
who had entered the community in childhood. These places retained
their purity as long as their inner domain (septa secreta) remained inac-
cessible to the laity. This inviolabitity of the inner domain ensured
the efficacy of a monastic prayer mediating between God and those
who supported the community and its resident saints by their gifts
and protection. Distance was therefore not a geographical concept,

28
M. de Jong, In Samuels Image. Child oblation in the early medieval West (Leiden
etc., 1996), esp. pp. 12655.
29
Rosenwein, Negotiating space; A. Diem, Keusch und Rein. Eine Untersuchung zu den
Ursprngen des frhmittelalterlichen Klosterwesens und seinen Quellen (Amsterdam, 2000). See
also Rosenweins contribution to this volume.
 301

for such sacred places were situated within a world that increasingly
came to depend on monastic mediation; so it was all the more essen-
tial to preserve the separateness of monastic space, lest the power
of prayer would be lost. Privileged outsiders first bishops, then
kings protected the inner domain from contamination; by impos-
ing restrictions of physical access on themselves as well, they ensured
direct access to the benefits of monastic prayer.
The Council of Chalcedon (451) firmly put the monks of each
city or region under episcopal authority,30 but where Gaul was con-
cerned, this largely remained wishful thinking on the part of bishops;
monasticism never became fully integrated into diocesan structures.
Bishops did play an important part in creating the place-bound
monasticism just mentioned, however. One decisive stage in this
process occurred after the death of Bishop Caesarius of Arles (524);
his powerful model of secluded and cloistered female monasticism was
extended to male communities as well, by bishops intent on safe-
guarding their foundations against malicious gossip in the world out-
side.31 It was this kind of monasticism that first attracted royal interest:
in 547, under Caesarius successor Aurelian, a male monastery
well furnished with relics was dedicated in Arles with the sup-
port of King Childebert and his wife Ultrogotha.32 A next crucial
phase occurred in the seventh century, in the wake of Columbanus
brief but important impact on Frankish monasticism. A rigid control
of speech, thought and dreams became the hallmark of cloistered
life, with daily confession and penance operating as a self-cleansing
mechanism that ensured the virtus of prayer and the purity of what
had become a sacred place.33
In order to discipline its members, religious communities developed
elaborate strategies for punishing wayward monks and nuns,34 but
these centered upon excluding the culprits from prayer and communal

30
Cf. Rosenwein, Negotiating space, p. 33.
31
About Merovingian bishops and their strategies to safeguard the reputations
of religious communities, see Diem, Keusch und Rein, pp. 174183; for later develop-
ments, M. de Jong, Imitatio morum. The cloister and clerical purity in the Carolingian
world, in: M. Frassetto ed., Medieval purity and piety. Essays on medieval clerical celibacy
and religious reform (New York/London, 1998), pp. 4980.
32
All this is an all too brief summary of Diems extensive analysis; see also W.E.
Klingshirn, Caesarius of Arles. The making of a Christian community in late antique Gaul
(Cambridge, 1994), p. 263.
33
Diem, Keusch und Rein, pp. 19698, 22429.
34
Diem, Keusch und Rein, pp. 21629.
302

life, rather than on disciplinary incarceration. Significantly, the first

explicit mention of a monastic carcer in Frankish sources dates from
the 840s. By then, the increasing differentiation of monastic space
had yielded the architectural concept of the claustrum, the inner space
only accessible to the members of the community itself and a few
privileged outsiders.35 Also, Benedicts Rule had become the text gov-
erning monastic life; for his own sixth-century Italian monastery,
Benedict had still reckoned with the need to exile monks from the
community, but a Carolingian commentator on the Rule amended
this in a significant way. The nutriti, that is, those who had grown
up within the monastic confines since childhood, had never been of
the world and should therefore not be sent back to it; instead, they
should be incarcerated, to better their lives.36
This is a monastic prison indeed, but one meant for internal use,
and emerging much later than monastic exile with its enduring con-
notations of honourable libera custodia.37 Where did monastic exiles
reside, once they had entered the monastic confines: within the septa
secreta itself, or within a more liminal zone of the community? As
far as I can see at present, it was the monastery in its entirety,
including its inner and liminal zones, that might serve as a location
of exile and/or refuge. Merovingian asylum, another understudied
topic, of course comes into it. Those who ran to monasteries to save
their lives may well have made a beeline for the most sacred part
of the monastic precincts, the main altar and its resident saint, but
detailed descriptions of the kind Gregory of Tours furnished with
regard to those seeking the protection of St Martin are lacking when
it comes to monastic exile and/or asylum.38 Gregory wrote about

35
M. de Jong, Internal cloisters: The case of Ekkehards Casus sancti Galli, in:
W. Pohl and H. Reimitz eds., Grenze und Differenz im frhen Mittelalter, Forschungen
zur Geschichte des Mittelalters 1 (Wien, 2000), pp. 20929.
36
Hildemar of Corbie/Civate, Expositio regulae S. Benedicti, ed. R. Mittermller,
Expositio regulae ab Hildemaro tradita et nunc primum typis mandata (Regensburg etc., 1880),
c. 28, p. 363: si autem ille, qui expellendus est, fuit ab infantia in monasteria, sicut diximus,
nutritus, non debet expelli, sed magis in carcerem mitti, quoadusque malum suum emendet et
bonum etiam facere vellet. See also ibid., c. 2, p. 109 and c. 71, p. 627; M. de Jong,
Growing up in Carolingian monastery: Magister Hildemar and his oblates, Journal
of Medieval History 9 (1983), pp. 12223.
37
There is only one Frankish exception: the Regula cuiusdam patris ad monachos
(c. 6.2, p. 15) did envisage a carcer, but as Diem explains (Keusch und Rein, pp. 230
236), this rule is very quite exceptional, and probably represents a version of Irish
monasticism that was not integrated into the mainstream of Frankish religious life.
38
For a summary of the passages in Gregory of Tours work referring to asy-
lum, see M. Weidemann, Kulturgeschichte der Merowingerzeit nach den Werken Gregors von
Tours I (Mainz, 1982), pp. 3069.
 303

the episcopal church in Tours, a public place by definition; monas-

tic exiles moved into a monastic inner sanctum, a place inaccessible
to outsiders or to be treated with circumspection, also by the authors
writing about such a delicate transition. This is my explanation for
the time being, but there may be better ones.

From the early sixth century onwards, bishops in Gaul had disci-
plined incorrigible clergy including fellow-bishops by dispatching
them to a monastery in order to do penance.39 In fact, a deposition
followed by penance in the monastery was the punishment par excel-
lence for higher clerics guilty of severe crimes. These men and some-
times women, widows of clerics were to be spared the shame of
submitting to an excommunication or of performing a penance under
public scrutiny, which would reflect badly on the entire clergys rep-
utation, and therefore did so secretius, in the secrecy of the monas-
tic confines. This practice not only required a measure of episcopal
control over monasteries, but also that such communities were per-
ceived as secret places, removed from the public gaze. Significantly,
it was an episcopal council in 533 chaired by Caesarius of Arles, a
bishop deeply involved in the creation of monastic separateness, that
condemned Bishop Contumeliosus of Riez for sexual misconduct and
alienation of church property, sending him to a monastery to do
penance. Predictably, Caesarius maintained that Contemuliosus was
deposed, and should remain in in Casinenso monasterio (location unknown)
forever; other bishops begged to differ, however, and the question
of the duration of Contumeliosus penance was left undecided.40
For centuries, monasteries were to remain the secluded locations
where bishops, priests and deacons guilty of capital crimes made
their amends in relative secrecy. Yet the debate and incertainty about

39
K.L. Noetlichs, Das Kloster als Strafanstalt im kirchlichen und weltichen
Recht der Sptantike, ZRG, kan.Abt. 80 (1994), pp. 1840; C. Vogel, La discipline
pnitentielle en Gaule des origines la fin du VII e sicle (Paris, 1952), pp. 13940; M. de
Jong, What was public about public penance? Paenitentia publica and justice in the
Carolingian world, La giustizia nell alto medioevo (secoli IXXI) II, Settimane 42,
(Spoleto, 1997), pp. 8756.
40
M. de Jong, Transformations of penance, in: F. Theuws and J.L. Nelson
eds., Rituals of power. From late Antiquity to the early Middle Ages (Leiden/Boston/Cologne,
2000), pp. 2001; Klingshirn, Caesarius, pp. 24749.
304

the duration of such amends continued as well. Had such penitents

actually converted to monasticism forever, or was their penitence
a temporary affair? Given that those who had committed grave and
notorious sins increasingly gave satisfaction by retreating into a reli-
gious community, this question was difficult to answer. When in 834
Archbishop Ebo of Rheims, the scapegoat of the rebellion against
Louis the Pious, had performed seven years of penance in Fulda, he
felt this sufficiently exonerated him to return to his former see; his
political opponents, however, insisted that Ebo had been irrevoca-
bly deposed, and that his penance in fact amounted to a conversion
to monastic life. The public penance imposed on Louis the Pious in
833, for that matter, gave rise to a similar debate.41
Bishops were the first to avail themselves of monasteries as means
to discipline their clergy, but kings soon followed suit, usually with
episcopal support. Bishop Gregory of Tours saw nothing wrong with
King Guntram (561592) resorting to monastic exile in his efforts
to curb the insurgence of two bloodthirsty brothers, Bishops Salonius
of Embrun and Sagittarius of Gap. The king was the one to con-
vene the synod that deposed the two culprits, but then granted them
an appeal to Pope John III, and accepted the papal verdict that they
should be restored to their bishoprics. When the brothers proved
incorrigible and once more showed contempt of the king, Guntram
had them shut up in monasteries far removed from each other, to
do penance; he instructed his counts to keep the twosome under
armed guard, and not to allow them any visitors.42 During a subse-
quent trial at the council of Chalon (579), some bishops felt the
brothers had performed a sufficient penance and should be restored
to their former office; it took fresh charges of offending the king
and betraying the patria to get a new verdict of deposition. The
two criminal bishops were deposed once more, and taken into cus-

41
R. Kottje, Die Bubcher Halitgars von Cambrai und des Hrabanus Maurus. Ihre ber-
lieferung und ihre Quellen, Beitrge zur Geschichte und Quellenkunde des Mittelalters
8 (Berlin/New York, 1980) pp. 21640; M. de Jong, Paenitentia publica and jus-
tice, pp. 8857. About Louis the Piouss public penance, see M. de Jong, Power
and humility in Carolingian society: The public penance of Louis the Pious, EME
1 (1992), pp. 2952.
42
Gregory of Tours, Decem libri historiarum, MGH SRM 5, c. 20, p. 228: His
auditis, rex commotus valde, tam equos quam pueros vel quaecunque habere poterant abstulit;
ipsosque in monasteriis a se longiori accessu dimotis, in quibus paenitentiam agerent, includi prae-
cepit, non amplius quam singulos eis clericos relinquens: iudices locorum terribiliter commonens, ut
ipsos cum armatis custodire debeant, ne cui ad eos visitandos ullus pateat aditus.
 305

tody in the basilica of St Marcel in Chalon (in basilicam beati Marcelli

sub custodia detruduntur). This site was obviously considered a more
effective place to contain the likes of Sagittarius and Salonius than
the two monasteries, far removed from each other, where they had
been kept earlier but St Marcel in Chalon was no foolproof
prison either, for the two escaped, to become wanderers on the
face of the earth.43
On the one hand, Gregorys pages about Sagittarius and Salonius
reveal a king who high-handedly used monastic space in order to
control two powerful clerical opponents, just as bishops had been in
the habit of doing; on the other, this is a story of surprising royal
lenience, of which Gregory patently disapproved. The two bishops
received chance upon chance to rehabilitate themselves, but sank
ever further into a morass of sin. Guntrams counts were make sure
the brothers remained inside the nameless monasteries where they
had been initially sent, checking on their visitors as well; by implica-
tion, this duty could not be left to the responsible abbots. Presumably
Sagittarius and Salonius were equally well guarded in St Marcel, but
all the same, they managed to escape. What turned this basilica into
a place where King Guntram thought he might safely confine two
formidable political opponents? St Marcel had been Guntrams foun-
dation, favoured by gifts from the king, his wife and his daughters,
but before we turn St Marcel into an evident state prison, it is
worth pointing out that only four years or more after the trial of the
two criminal bishops, Guntram confirmed his gifts to St Marcel and
to St Symphorian, safeguarding the property of these two commu-
nities against possible violations by the bishops and royal power.
He did so at the Council of Valence (583583), with seventeen bish-
ops co-signing the conciliar record, including the bishop of Chalon
himself.44 Guntrams gesture are an early sign of the self-imposed
limits bishops and kings set against their own encroachment upon
monastic space, but it did not yet amout to a full-blown immunity
or exemption.45 Did such privileges, when they emerged in the course

43
Gregory, Decem libri historiarum, V, c. 27, p. 233. Cf. De Jong, Transformations
of penance, pp. 2102.
44
See about this important immunity avant la lettre A. Diem, Was bedeutet
regula Columbani?, in: M. Diesenberger and W. Pohl eds., Integration und Herrschaft.
Ethnische Identitten und soziale Organisation im Frhmittelalter (Wien, 2001; forthcoming);
also, cf. Barbara Rosenweins chapter in this volume.
45
Rosenwein, Negotiating space, p. 45.
306

of the seventh century, turn monastic communities into places even

better equipped to lock up royal prisoners? Or did a royal immu-
nity instead preclude the matter-of-fact use of monastic space with
which Gregory of Tours credited King Guntram? I would argue for
the latter.
There is one powerful layman in Gregories Histories who counts
as a monastic prisoner: Merovech (d. 578), the rebellious son of King
Chilperic I (561584). As Friedrich Prinz expressed it, the fact that
Merovech was tonsured, made a priest and packed off to St Calais
at his royal fathers command no doubt meant that this monastery
had close connections with the royal house, otherwise it could not
have served as a kind of state prison.46 But according to Gregory,
Merovech never made it to to St Calais. Only lightly guarded, he
escaped en route, put on secular clothes, and sought asylum in the
church of St Martin in Tours, where Gregory just happened to be
celebrating mass. Some state prison! The passage about Merovechs
so-called imprisonment deserves some closer scrutiny.47 Held in cus-
tody by his father, he was tonsured and had his clothes changed for
those customarily used by clerics, then ordained a priest; he then
was sent to St Calais near Le Mans in order to be instructed in the
way of life of a priest (regula sacerdotalis). In other words, the court
or wherever Chilperic happened to be was the place of custodia,
not the monastery. At the court, Merovech was instantly transformed
into a priest, with the necessary earlier stages the reception of a
clerical tonsure and habit thrown in for good measure. But it was
Merovechs priestly consecration that mattered most, for at least
in theory this was an irrevocable measure disqualifying the prince
from the throne forever.48 Of course Merovech had no idea what it
took to be a priest or perhaps a bishop, eventually so he was
sent to St Calais for further instruction. Was he meant to remain

46
Prinz, Frhes Mnchtum im Frankenreich, p. 155.
47
Gregory, Decem libri historiarum, V, c. 14: Post haec Merovechus cum omni custodia
a patre retineretur, tonsuratus est, mutataque veste, qua clericis uti mos est, presbyter ordinatur,
et ad monasterium Cenomannicum, quod vocatur dirigitur, ut ibi sacerdotali erudiretur regula.
48
For Carolingian example of royal son who was gradually excluded from the
throne, see Charles the Balds treatment of his son Carloman, discussed by De
Jong, In Samuels Image, pp. 2578. Carloman first became a child oblate, then a
deacon; upon his revolt and bid for power in 870 he lost ecclesiastical rank and
was imprisoned in the castrum of Senlis. Further rebellion led to his blinding and
exile to Corbie. From there, Carloman fled to his uncle, Louis the German; he
died as the abbot of Echternach in 881.
 307

in this community? There is no way to tell, but surely Gregorys

neutral dirigitur should not be translated tendentiously as packed off;
according to Gregory, Merovech travelled with a minimal guard and
easily escaped, an indication perhaps of some confidence in
the effectiveness of the new priests ordination. From whichever side
one looks at this story, it does not qualify as an instance of Klosterhaft.

D : VITA COLUMBANI

The seventh century was the heyday of Merovingian immunities,

as Barbara Rosenwein expressed it,49 and therefore an age of even
stricter and subtler definitions of the boundaries of sacred
space. In his celebrated Life of Columbanus, written between 639 and
643, Jonas of Bobbio spelled out the rules of the game according to
the monastic point of view: these boundaries were not to be set by
rulers. One of Jonas stories to this effect revolves around interde-
pendent places of power and their limits: the sacred space repre-
sented by Columbanus foundation Luxeuil, as opposed to the unholy
space of a contaminated court. At the centre of this episode is Queen
Brunhild, the Jezebel of Jonas hagiography. After the death of her
son Childebert II (596) she ruled together with her grandsons
Theudebert II (592612) and Theuderic II (596613). The latter is
the other villain of the piece; together with his grandmother, Theuderic
drove Columbanus from Luxeuil and sent him into exile. The story
discussed here leads up to these dramatic events.50
In the royal villa Bruyres-le-Chatel, Brunhild presented Columbanus
with the illegitimate sons of King Theuderic II, ordering him to give
then his blessing, but the saint flatly refused and left; when he crossed
the threshold of the court (aula regia), a terrifying earthquake occurred.
In revenge, Queen Brunhild instructed Luxeuils neighbours to stop
any monk who wished to leave the monastery, and to withold all
material sustenance and other support from the community. Colum-
banus retaliated by going to Theuderics court, residing in another
villa, but refusing to enter the royal establishment; he planted him-
self outside, and refused to budge. Theuderic, wishing to keep his

49
Rosenwein, Negotiating space, pp. 7496.
50
Jonas, Vita Columbani I, c. 19, ed. B. Krusch, MGH SRM 4, pp. 8790;
Rosenwein, Negotiating space, pp. 703.
308

peace with the redoubtable man of God, ordered his servants to

bring Columbanus food and gifts fit for a king (regio cultu oportuna).
But Columbanus, surveying this splendour, declared that they were
an abomination: the servants of God should not contaminate their
mouth with food sent by those who barred them from entering their
own place and that of others. All the vessels broke, wine and food
dripped to the floor, and all who witnessed this miracle were fright-
ened out of their wits. As a result, Theudebert and Brunhild came
to the saint, asking for forgiveness and promising to reform their
lives. Pacified by these promises, the saint returned to the monastery.
This is a story about a saint establishing his authority by defining
the nature of monastic and royal space, turning the tables on his
royal adversaries by means of inversion. Brunhild did not dare to
interfere with Luxeuils inner domain; instead, she attempted to sever
the monastery from the indispensable support of its neighbours
(vicini ). This was an effective threat to Luxeuils existence, for no
religious community could survive in total isolation. But there was
another side to this coin: a royal court (aula regia) cut off from the
benefits of prayer and blessing was also in danger of disintegration.
By refusing to enter, Columbanus declared the aula regia to be out
of bounds, a contaminated place. Theuderic then attempted to shift
the boundaries of the court; by giving the saint a royal welcome in
a place Columbanus had defined as being outside the court, the
king tried to include the saint in royal space by extending its para-
meters. These tactics miserably misfired; even when Theuderic moved
the aula regia to the saint the mountain came to Mohammed
it remained a source of pollution by which a man of God should
not contaminate himself. Of course Columbanus emerged victorious
from this battle of wits, for Jonas was intent to show that the saint
was the one who most effectively controlled the nature of royal and
monastic space; to make his point, Jonas implicitly contrasted the
foot-loose nature of the royal court with Luxeuils stability. The scene
is set in two royal villae, both explicitly mentioned by name, to which
the saint travels; the court can be here, there and everywhere, but
at the background hovers the monasterium, always in one place. It is
Columbanus who has freedom of movement and entry, inflicting a
terrifying tremor on one aula regia, and refusing to set foot in the
other. Conversely, all that Brunhild and Theuderic could manage
was some unsuccessful manipulation in the margins of places with
boundaries over which they had no control. The implication of the
 309

story is that monastic space was firmly integrated into the topogra-
phy of political power, and would sustain and strengthen the might
of kings as long as the latter respected the integrity of the monas-
tic confines. If the they violated the integrity of the locus sanctus, how-
ever, the aula regia would suffer as well measure for measure.
But Jonass narrative about defining royal and monastic space did
not end here. Theuderic persisted in his concubinage, and a renewed
confrontation with the saint ensued, with Brunhild still at the back-
ground as the evil genius inspiring strife: she intimated to bishops
that Columbanus had polluted the very rule he had instructed his
monks to live by. Significantly, it was now the king who travelled
to Luxeuil, to berate Columbanus for departing from the custom of
allowing all Christians access to the very secret enclosure (septa secre-
tiora). What better way to ridicule a hated king than by putting patently
ridiculous words in his mouth? By the time Jonas wrote his Life of
Columbanus, any statement about such a custom must have seemed
ludicrous. From the mid-630s onwards, the saints disciples began to
issue episcopal exemptions, in unison with kings granting immuni-
ties;51 regardless of such privileges, it had already been self-understood
for several generations that one could not simply enter the very
secret enclosure of any religious community. But here is a king who
declares the monastic enclosure accessible to all, thus being turned
into a laughing stock for Jonas contemporaries: the man obviously
had no idea what he was talking about. Columbanus, who did know
how to organise a proper monastery, countered that the dwellings
of the Lords servants were out of bounds for laymen and those
unfamiliar with religious life, but of course the monastery had suit-
able quarters where guests received a warm welcome. In other words,
monastic space was made up of concentric circles, with an inner
enclosure surrounded by a liminal area accessible to outsiders. This
spatial division had become customary in prominent monasteries with
royal connections in Jonas day and age, as is revealed by the next
stupid move from a king oblivious to custom. Briefly put, their alter-
ceration amounted to the following. Theuderic said to Columbanus,
open up your monastery, or you will receive no gifts or support from
me. This yielded Columbanus predictable reply: if you violate our

51
Rosenwein, Negotiating space, p. 66, rightly concludes that Jonas Vita Columbani
represents the views of the next generation, i.e., that of Columbanus pupils.
310

inner space, I shall not accept your gifts or support. And if you have
come to destroy our community, rest assured that your kingdom and
progeny will be destroyed as well. This was a powerful malediction,
and also a crystal-clear statement of the kind of interdependence
between monastic prayer and royal power Jonas believed in. What
Theuderic should have said, of course, was: please make sure your
monastery is sacred and inaccessible, and therefore worthy of my
patronage. Jonas made him say precisely the opposite, and carried
his portrait of a king who did not know custom to an even more
vicious level. Theuderic tried to enter the refectory, very much a
part of the septa secretiora, but, terrified by the saints malediction, he
retreated. The king could not resist a taunt and a threat: you prob-
ably hope to get your martyrs crown through me, dont you?.
Theuderic went back to the court, but made another attempt at iso-
lating Luxeuil, inciting his leading men to declare they wanted no
community in their territory did not make everyone welcome. Colum-
banus repaid this in kind: he would no longer leave the confines of
his community (caenubii septa), unless they dragged him out by force.
Jonas story still continues. Columbanus was left in Luxeuil, in the
custody of one of the kings proceres, who made him leave the monastery
for Besanon. Here the saint was to remain until a royal verdict had
been pronounced. This ignominous episode called for yet another
installment of Columbanus turning the tables on the king: the saint
went out into the city and liberated all the kings prisoners, accord-
ing to time-honoured hagiographical principles.52 Good Merovingian
saints freed prisoners by droves, but in Jonas narrative, this familiar
topos took on a new meaning. Theuderic restricted Columbanus free-
dom of movement, so the saint retaliated by setting the kings pris-
oners free. As usual, chains dissolved miraculously, but Jonas version
of the saintly liberation of prisoners had some special features:
Columbanus called upon the kings prisoners to do penance, guid-
ing them to a church with locked doors. These were duly opened
by divine power, but then closed once more, in front of the amazed
guards chasing their fugitive charges.
Several important messages were transmitted at once. To begin
with, there was no question of Columbanus himself being anything
remotely resembling a prisoner; the saint moved about the city

52
Graus, Die Gewalt bei den Anfngen des Feudalismus, passim.
 311

freely, in search of prisoners to liberate. Second, this particular pas-

sage is the only instance where Jonas uses the words carcer or ergas-
tulum; locking up miserable prisoners was the business of kings,
mercifully freeing them that of saints. Third, against a royal model
of incarceration, Columbanus posited his superior means of punish-
ment: sinners should do penance. Given Jonas definition of the
medicamenta paenitentiae elsewhere in his Life of Columbanus,53 this meant
as much as, come to the monastery and save your souls. And last
but not least, any king attempting to wrest a saint from his locus
sanctus would be confronted with doors that opened and closed at
unpredictable moments, guided by God rather than by royal com-
mand. Columbanus triumphantly returned to his monastery through
the centre of town, without anyone hindering his passage.
To the two rulers whose reputation Jonas intended to dismantle
as thoroughly as possible, this was a challenge they could not allow
to go unnoticed; the truth still did not sink in. Brunhild and Theuderic
sent soldiers who entered Luxeuils inner domain (septa), where the
saint sat in the atrium of the church, quietly reading a book. The
soldiers rushed past Columbanus, occasionally touched him, even
stumbled over him, but they could not see him. It was a lovely
scene, Jonas said.54 The man in charge, however, looking through a
window, suddenly saw the saint sitting there, and realised he was in
the presence of a miracle. Was there better proof of this place being
sacred? From then on, the narrative develops, revolving around a
saint who did not wish to leave his community, but did not want
to endanger the lives of the soldiers who came to carry him off, and
the soldiers themselves, torn between obedience to their king and
deep fear for the divine punishment they might incur because of
their subservience to earthly powers.
This is Jonas background to Columbanus exile in 610: a truly
impressive hagiographical narrative, meant to explain to the next
generation Columbanus pupils why the saint who had created a
sacred domain deeply respected by kings had nonetheless been chased
out of his septa secretiora, leaving Luxeuil and part of his monastic flock
behind. By damning Brunhild and her progeny, Jonas implicitly
extolled the virtues of other rulers, the good kings (and bishops) who

53
De Jong, Transformations of penance, pp. 2157.
54
Jonas, Vita Columbani I, c. 20: ipsum nequaquam viderent, eratque expectaculum pul-
cherrium.
312

respected and protected monastic space, making better use of its

divine uses than Brunhild/Jezebel and her bungling son had done.55
Jonas was writing in better times, at the very start of the heyday of
Merovingan immunities. His Life of Columbanus is very much a doc-
ument that belongs to his own day and age, when the interdepen-
dence between royal and monastic space and the separateness
essential to their mutual benefits were first formulated in the
juridical documents known as immunities. Jonas elaborate narra-
tive reveals a topography of power into which Luxeuil was fully inte-
grated, provided it remained a locus sanctus. The monastery as a
prison does not not enter into the discussion. On the contrary, the
real carcer was located in the city (urbs) and guarded by the kings
soldiers; the monastery was a place where prisoners might go vol-
untarity, to do penance, after having been liberated by a saint.

O ?

In a stimulating article on kings who opted out Clare Stancliffe

argued that the involuntary retreat to a monastery was typical of
the Romanised continent. In Merovingian circles the tonsuring of
a prince or king and his confinement within a monastery was simply
a political act, designed to remove a rival king.56 By contrast, in the
early medieval insular world kings opted out of their own accord,
freely exchanging their royal office for the monastery or a life-long
pilgrimage. This difference, Stancliffe maintains, can be explained
by different religious traditions. Irish ascetism took a bleak view of
the possibilities of rulers to achieve Christian goals while remaining
in the secular domain; on the Continent, however, a more optimistic
attitude to Christian kingship prevailed. In other words, Continental
rulers had no reason or incentive to renounce kingship of their own
volition: they could be kings and good Christians at the same time.

55
Cf. J.L. Nelson, Queens as Jezebels: the careers of Brunhild and Balthild in
Merovingian history, Studies in Church History, Subsidia I (1978), pp. 3177; repr.
in eadem, Politics and ritual in early medieval Europe (London/Ronceverte, 1986), pp.
148. For an in-depth exploration of Jonas Life of Columbanus, see I.N. Wood,
The Vita Columbani and Merovingian hagiography, Peritia 1 (1982), pp. 6380.
56
Clare Stancliffe, Kings who opted out, in P. Wormald (with D. Bullough
and R. Collins) ed., Ideal and reality in Frankish and Anglo-Saxon society. Studies presented
to J.M. Wallace-Hadrill (Oxford, 1983), pp. 15476 (esp. p. 158).
 313

A decade earlier, Karl-Heinz Krger had discussed the puzzling con-

version of two rulers opting out, the Frankish maior domus Carloman
(747) and the Langobard king Ratchis (749), who both retreated to
Monte Cassino. Krger came to the conclusion that these two con-
versions followed Anglo-Saxon models.57 In Carlomans case, Krger
argues, it was clearly the Anglo-Saxon missionary Boniface who was
behind his conversion, but even for Ratchis, Krger manages to
come up with some Anglo-Saxon connections: Bishop Wilfrid of York
and King Caedwalla of Mercia had once visited the Langobard court
in Pavia.58
Did it take Anglo-Saxons to inspire eighth-century Frankish and
Langobard rulers with religious fervour? This view seems a bit insular,
to say the least, but there is a more fundamental problem. Throughout,
the question is whether a given conversion was either voluntary or
forced. Though well aware of the contradictory nature of the sources,
Krger feels compelled to decide in favour of Carlomans religious
motivation, and then proceeds to build a context for Carlomans
and Ratchis supposedly voluntary entry into Monte Cassino.59 For
Stancliffe, Carloman and Ratchis are the two exceptions that prove
the rule: kings voluntarily abandoning secular power are only found
in England.60 An essential element is left out of the argument, how-
ever: the particular rhetoric of the sources that describe these events.
The obvious disagreement among early medieval authors about what
exactly happened to Carloman and Ratchis should be taken into
account. Was there perchance a debate after the event, and a wish
to portray a retreat into monastic space as an eminently honourable
conversion or as its opposite? And what about the sources offering
proof that the tonsuring and monastic confinement of continental
princes or kings was merely a political political act?
We are back with Jonas Life of Columbanus once more, for this
contains the key text upon which such arguments are built: that the
kings on the Continent did not opt out voluntarily, that no religious
motivation was involved until Boniface and so-called insular inspi-
ration came along, and that Merovingians aimed for murder, not

57
K.-H. Krger, Knigskonversionen im 8. Jahrhundert, Frhmittelalterliche Studien
7 (1983), pp. 169222.
58
Krger, Knigskonversionen, pp. 21617.
59
Krger, Knigskonversionen, p. 185.
60
Stancliffe, Kings who opted out, p. 158.
314

for monastic exile. But as we have seen, Jonas cleverly constructed

narrative was meant to destroy the reputation and honour of Brunhild
and Theuderic, the evil instruments of the saints exile from his locus
sanctus and Jonas managed quite successfully. This is not a text
from which to glean straightforward information about Merovingian
custom of any sort. After his departure from Luxeuil, Columbanus
was welcomed in the realm of Theudebert II, the other of Brunhilds
grandsons he once had refused to bless. Theuderic then attacked his
half-brother, and the two kings fought a war aimed at mutual destruc-
tion, vaunting the invinciblity of their peoples. It is at this point that
Columbanus prevailed upon Theudebert the most sensible of the
warring parties to give up the superciliousness of his arrogance
and to become a cleric in ecclesia positus, lest he lose eternal life
along with his earthly realm. This earned Columbanus the ridicule
of the king himself as well as of the royal entourage: They had
never heard of a Merovingian who had been raised over a kingdom
becoming a cleric of his own free will.61 This is the passage on
which historians have concentrated: Merovingian kings were no wimps
opting out. But the story goes on. Jonas has Columbanus exclaim:
If he does not accept the honour of being tonsured voluntarily, he
will briefly be a cleric against his will!.62 This is another turning
point in the narrative, and one more maledictio, a malicious prophecy
by a powerful saint, which of course came true. The two half-broth-
ers fought a ferocious battle at Zlpich, and Theudebert had to flee.
Columbanus, sitting on a rotting tree trunk reading a book, had a
vision of Theuderic calling him, asking him for help. Columbanus
refused: this was not what the Lord had intended when He ordered
us to pray for our enemies, he answered. Theudebert was duly cap-
tured and taken to his grandmother Brunhild, who full of wrath
( furens) had him made a cleric; only a few days afterwards she ruth-
lessly (impie) ordered him to be killed.

61
Jonas, Vita Columbani I, c. 28, p. 105: Inter quae vir Dei ad Theudebertum accedit
eumque suadet, ut coepte arrogantiae supercilium deponeret seque clericum faceret, et in ecclesia
positus, sacre subderetur religione, nec simul cum damna presentis regni aeternae pateretur vitae
dispendia. Quod et regi et omnibus circumadstantibus rediculum excitat, aientes, se numquam aud-
isse, Merovengum, in regno sublimatum, voluntarium clericum fuisse.
62
Ibid.: Detestantibus ergo omnibus beatus Columba ait: Si voluntarius nullatenus clerica-
tus honorem sumat, in brevi invitus clericus existat. His ergo dictis, vir Dei ad cellolam remeat,
moxque prophetici dicti eventum res non diu dilata adfirmat.
 315

As an example of a typically Merovingian practice of forced ton-

sure and monastic confinement, this story presents some problems,
if only because there is not a monastery in sight. According to Jonas,
Brunhild requested that Theudebert was made a cleric, thereby
implicitly granting him his life, but shortly afterwards she had him
put to death. This is no doubt part of Jonass narrative tour de force63
by which he succeeded in perpetually damning Brunhilds memory.
We have only Jonass word for it that Theudebert was murdered at
Brunhilds behest, the but moral of the tale is clear: killing an oppo-
nent who had just been tonsured was impious. As for Theudebert
and his entourage ridiculing Columbanus suggestion that the should
become a cleric of his own volition, this was just one more instance
of Jonas cleverly highlighting the difference between a wise saint and
a stupid king. Theudeberts stupidity (and superbia) consisted of not
opting out at a moment of total defeat, saving his life and honour.
In other words, a real king would have withdrawn from the politi-
cal arena, accepting the clericatus honor. Theudebert refused to do so,
and got his just deserts: the shame of a forced tonsure, and of death
at the hands of a wicked woman. This was the view Jonas wished
to get across. There is no way to tell whether it was shared by all
the powerful of his day, but neither is there any reason to jump to
the opposite conclusion.
By the mid-seventh century, rulers exchanging secular power for
ecclesiastical honour were by no means exceptional on the Continent.
In Visigothic Spain, the honourable way out for a king was to adopt
the status of a public penitent. Chindasuinth (653),64 Wamba (680)65
and Erwig (687)66 all gave up royal office by becoming penitents,
thus avoiding a bloody conflict and making room for their successors;
a similar scenario was probably intended in 833, when Louis the
Pious submitted to a public penance. In Wambas case, this has been
construed as the result of an evil plot the king had supposedly
been drugged and turned into a penitent but there is no evidence
for this tall story invented in the late-ninth century. It is much more
likely that these kings decided to become penitents of their own

63
Woods expression, in The Merovingian kingdoms, 450751 (London, 1994), p. 131.
64
Fredegarii continuationes, IV, c. 82, ed. A. Kusternig (Darmstadt, 1994), p. 256.
65
Twelfth Council of Toledo (681), ed. Vives, Concilios, p. 386.
66
Laterculus Visigothorum, no. 4647. ed. Th. Mommsen (1898), MGH Auct. Ant.
13, p. 468.
316

accord, guided by a mixture of political acumen and religious fer-

vour.67 We cannot be certain, but we do know that contemporary
authors chose to depict these royal conversions as a highly com-
mendable gesture, even in Wambas case.68 This amounted to more
than a mere cover-up. Queen Balthilds withdrawal into her own
foundation, the nunnery Chelles sometime between August 664
and August 665 was consistently presented presented as a vol-
untary action. That the queen actually needed to opt out because
of factional strife is clearly hinted at in various sources,69 yet authors
favourable to Balthild subtly enhanced her reputation by stressing
that she entered Chelles of her own volition. This certainly holds
true for the author of the first Vita Balthildis, written before 690.70
The queen fervently wished to retreat to Chelles, but for the love
of her she was delayed by the Franks who ultimately permitted
her to leave because they were afraid of the formidable Lady Balthild:
But the lady considered it Gods will that this was not so much their
decision as a dispensation of God that her holy desire had been fulfilled,
through whatever means, with Christ as a guide. And having been
escorted there by certain noblemen, she came to her above-mentioned
monastery at Chelles, and there, as is fitting, she was honourably and
very lovingly received into the holy congregation by the holy maidens.
Then, however, she had a complaint of no mean size against those
whom she had kindly nurtured, because they had erroneously consid-
ered her suspect and even repaid her with evil for good deeds. But,
discussing this quickly with the priests, she kindly forgave them every-
thing and asked them to forgive her the disturbance of her heart. And
thus, with the Lord as provider, peace was fully restored among them.71

67
Cf. M. de Jong, Adding insult to injury: Julian of Toledo and his Historia
Wambae, in: P. Heather ed., The Visigoths from the migration period to the seventh century.
An ethnographic perspective (Woodbridge, 1999), pp. 373422, esp. pp. 3734, with ref-
erences to earlier literature.
68
The Twelfth Council of Toledo merely stated that Wamba received his penance
in the event of inevitable need, designating his successor: Idem enim Wamba prin-
ceps dum inevitabilis necessitudinis teneretur eventu, suscepto religionis debitu cultu et venerabili
tonsurae sacro signaculo, mox per scribtuarum definitionis suae hunc inclytum dominum nostrum
Ervigium post se praeelegit regnaturum, et sacerdotali benedictione unguendum; cf. Vives ed.,
Concilios, p. 386.
69
See Fouracre and Gerberding, Late Merovingian kingdoms, pp. 97118 for a dis-
cussion of the sources for Balthilds life; about her reputation, Nelson, Queens as
Jezebels. About Balthild and her policy of immunities and exemptions, see Wood,
Merovingian kingdoms, pp. 197222; Rosenwein, Negotiating space, pp. 7481.
70
Fouracre and Gerberding, Late Merovingian kingdoms, pp. 1145.
71
Vita Balthildis c. 10, MGH SRM 2, p. 495. I follow the translation of the Life
by Fouracre and Gerberding, Late Merovingian France, pp. 11832.
 317

This is as good an example as any of the way in which the theme

of force versus choice might be exploited to bolster a reputation.
The author cleverly managed to convey two conflicting messages:
factional strife forced Balthild to retreat into her monastery, but
this nevertheless was her own choice, guided by Christ. She there-
fore entered Chelles in a most honourable fashion, receiving a royal
reception. Afterwards, however, days of political activity were not
quite over, for a grievous dispute still had to be settled. Peace was
the result, at a price. Balthild did not become a nun; by emphasiz-
ing her holy desire (sancta devotio, the expression is used twice) to
enter Chelles, the author turned Balthilds retirement into a matter
of choice and honour.72 Once inside, her behaviour befitted that of
the most worthy of nuns, as the author of the Life is at pains to
point out, but nonetheless Balthild remained betwixt and between,
out of the political arena but with all her royal potential intact. Her
Life provides a revealing vignette of queenly diplomatic activity within
the cloister, with Balthild suggesting to the abbess that they should
constantly visit the king and queen and the palace nobles in befitting
honour with gifts, as was the custom, so that the house of God
would not lose the good reputation with which it had begun, but
would always remain more fully in the affection of love with all its
friends and more strongly in the name of God in love, as it is writ-
ten: It is necessary to have a good report of them who are without.
(I Tim. 3:7).73
As a former outsider who was now within, Balthild was in an
excellent position to appreciate her monasterys need to keep up
good relations with the palace, and to protect the blameless repu-
tation which would attract gifts and protection. Such tell-tale details
suggest that Bathilds life as a politically active member of the royal
family did not end once she retired to Chelles. She was a queen
lying low, but she remained a queen.

72
Vita Balthildis c. 10, ed. B. Krusch, MGH SRM 2, p. 495: Erat enim eius santa
devotio, ut in monasteriam, quem prediximus, religiosarum foeminarum, hoc est in Kala, quam
ipsa edicavit, conversare deberet. The expressions devotio and conversare both belong to
monastic usage, but vow is only one of its possible translations (others are reli-
gious intention or pious wish). If one wishes to translate Balthilds devotio sancta as
a vow at all, one should not treat this as a formal monastic vow of any sort, but
rather as a personal and informal promise to God.
73
Vita Balthildis c. 12, p. 498.
318

This brings us to the flexibility and open-endedness of monastic exile.

The usual assumption is that monastic exiles became full members
of a religious community and were therefore were politically kalt-
gestellt. But early medieval exilers and exiles must have known from
experience that monastic confinement was rarely permanent, that
hair once tonsured grew back, that veils could be taken off, and that
a political following might as well be built up within as without a
monastery. This becomes apparent from the cluster of texts dealing
with the tempestuous events of 673675, which entailed the monas-
tic exile in Luxeuil of two adversaries, the maior domus Ebroin and
Bishop Leudegar of Autun. Paul Fouracre has brilliantly examined
the evidence, so I shall not repeat his analysis here.74 Just a brief
reminder of the main events: after having engineerd Ebroins down-
fall and exile to Luxeuil in 673, Bishop Leudegar ended up there
himself only two years later. Both men left Luxeuil again to return
to power, with all the risks involved: in 678 Leudegar was grue-
somely executed as an accessory to regicide, and Ebroin was mur-
dered two years later.75
The rhetoric of voluntary/involuntary exile had its impact on these
texts as well. The first Passion of Leudegar, written shortly after the
saints death, contains a remarkable story about Theuderic III, Balt-
hilds youngest son who was put onto the throne by Ebroin. Upon
Ebroins fall from power King Theuderic had his hair cut off by
certain men who were seen to be leaders in the kingdom and wished
by flattery to persuade Childeric not to shed blood.76 They presented
him shaven to his victorious brother, but the way this confrontation
unfolded owes a lot to the fact that the first Passion of Leudegar
was written at a time when Theuderic had returned to power. Asked
by his brother what fate he preferred, he [Theuderic] would only
say this: that he had been unjustly cast down from the throne, and
he declared that he was anticipating a swift judgement from God in
his favour. It was then ordered that he should remain in the monastery

74
P. Fouracre, Merovingian history and Merovingian hagiography, Past and
Present 127 (1990), pp. 338.
75
See also Wood, The Merovingian kingdoms, pp. 2348.
76
Passio Leudegarii, c. 6, ed. B. Krusch, MGH, SRM 5, p. 288; transl. Fouracre
& Gerberding, Late Merovingian France, p. 223.
 319

of the martyr St Denis and be protected there until he grew his

hair, which they had cut off . As an example of representing monas-
tic exile as a result of an honourable choice, this passage equals that
of the Life of Balthild. As for Leudegar himself (still according to the
first Passion), he was first taken to Luxeuil to stay there until all
had taken counsel together what should be done with a man of such
great reputation, but was subsequently sentenced to remain in Luxeuil
in perpetual exile. Yet the author explicitly indicates what kind of
leeway such a perpetual banishment might offer: This judicial decree
was quickly confirmed, with some of the priests and bishops feeling
that presently they would bring him back from the kings anger.77
In this narrative, Luxeuil has many functions. It was a place where
one could bring a political hot potato like this powerful bishop while
deciding what to do with a man with so many allies. The allies
themselves seem to have been content with the verdict of perpetual
banishment, banking on the kings wrath to subside, so they might
bring him out again.

All this hinges on a kind of double entendre: that monastic exile was
perpetual, but exiles might be pardoned and leave again; that monas-
teries were both within withdrawn from the world; that tonsured
exiles were immune to violence, but that they might be killed if one
managed to get them outside the monastic precincts. As the First
Passion of Leudegar (c. 12) has it: Again and again Hermenar threw
himself at the kings feet, imploring him to let Leudegar remain in
Luxeuil and not to order him to be led out to a certain death accord-
ing to the cruel ones whom the devil had stirred up in a rage against
him.78 Apart from glossing over Hermenars dubious role in Leudegars
martyrdom, this passage suggests that the king was eminently cap-
able of extracting Leudegar from Luxeuil if he wished. All the same,
the author of Leudegars first Passion is extremely vague about the
way Leudegar left his place of internal exile. Two dukes were charged
with this task; one of their men was to kill Leudegar as soon as he
saw the servant of God outside Luxeuil. But the villain took fright
and begged the saint for mercy; the next thing we know is that
Leudegar was indeed outside Luxueil, still in the custody of the two

77
Passio Leudegarii, c. 12, p. 294.
78
Passio Leudegarii, c. 12, p. 294.
320

dukes, who now had become his firm allies. All this prevarication
suggests that the crossing of Luxeuils boundaries in either direction
was a sensitive issue.79
In the first Passion, Leudegars entry into Luxeuil is unequivocally
presented as a sentence of exile, the result of an act of mercy on
the part of King Childeric and his leading men.80 By the mid-eighth
century, when an author from Poitiers rewrote the tale of Leudegars
sufferings, the story had become a different one. Leudegar now took
the initiative himself, requesting the kings permission to leave the
world and devote himself to God. Once in Luxeuil, he encountered
his fellow-exile Ebroin, who received a similar uplifting treatment
from this hagiographer. The two men confessed their mutual guilt,
obeyed the abbot when he imposed a penance on them, and did
their utmost to live forever within the monastic community as if they
were monks.81
This is the view of a man writing more than 70 years afterwards,
who could safely credit the protagonists with lofty penitential motives.
It is a far cry from the seventh-century versions of Ebroins exile.
The Liber Historiae Francorum (727) merely stated that the enraged
Franks tonsured Ebroin and sent him to the monastery of Burgundy
in Luxeuil, after having cut off Theudeberts hair.82 The Continuator
of Fregedars Chronicle related these events in an equally dead-pan
fashion, but with the addition of a tell-tale detail: Ebroin was taken
to Luxeuil against his will (invitum).83 This one little word reflects two

79
Cf. Passio Leudegarii, cc. 14 and 16, pp. 29698, about the duces who had
been ordered to take Leudegar out of Luxeuil. They suddenly appear in c. 14,
without any indication that they did follow up their orders, and suddenly appear
to have done so in c. 16.
80
Also in Leudegars case: Passio Praeiecti, c. 26, MGH SRM 5, p. 241: Leodegarius
vero penitentia ductus et exilium Luxovio trusus.
81
Ursinus, Passio altera Leudegarii, c. 4, MGH SRM 5, p. 327.: Et iuxta iussum
regis, ipsumque pontificem deprecantem, Luxovio coenuvio, ut ei liceret, relicto saeculo, vacare Deo,
humili poposcit prei se dirigendum; quem protinus illic ire non distulit. Qui festinus in monasterio
perveniens, ibidem Ebruinum iam clericum invenit; dicens se aliquid in eo peccasse, veniam vicissim
petentes, steterunt concordes. Tamen ab abbate seiuncti, aliquod spatium temporis uterque peni-
tentiam agentes, inter contubernia monachorum strinue habitare quasi perpetue monachi conati sunt.
82
Liber Historiae Francorum, c. 45, ed. B. Krusch, MGH SRM 2, p. 317: Eo tem-
pore Franci adversus Ebroinum insidias praeparant, super Theudericum consurgunt eumque de
regno deiecunt, crinesque capitis amborum vi abstrahentes, incidunt. Ebroinum totundunt eumque
Luxovio monasterio in Burgundia dirigunt.
83
Fredegarii continuationes c. 2, ed. Kusternig, p. 272: Eo tempore Franci adversus Ebroi-
num insidias praeparant, contra Theudericum consurgunt eumque a regno deiciunt. Crines capitis
eius abscidentes tutunderunt Ebroinumque et ipsum tutundent et in Burgundia Luxovio monaste-
rio invitum dirigunt.
 321

generations of literary agitation against Ebroin. In the first Passio

Leudegarii Ebroin became the villain of the piece, exiled to do penance
for abominable sins ( facinora), a perfidious Julian who, once in Luxeuil,
merely feigned to lead the life of a monk a true apostate.84 This
dim view of Ebroin would eventually prevail. Yet however diver-
gent, all accounts agree in one respect: whether Ebroins efforts to
behave like a monk were presented as sincere or hypocritical, he
never became a monk. At best he was tonsured, in monastic habit,
or living as if he would be a monk forever. It was nothing sur-
prising, then, for Ebroin to gather a gang (comitatus) of friends and
servants around him at Luxeuil, and to leave the place when he
thought the time was ripe.85 For such exiles, the monastery was a
place where one might keep ones options open. Ebroin only became
a real monk in the early ninth century, from a typically Carolingian
perspective. The Annales Mettenses Priores, a hagiographical piece of
propaganda for the Carolingian dynasty composed in 806 or shortly
thereafter,86 made the most of Ebroins departure from Luxeuil. Given
that most Merovingian kings were not deemed worthy of more than
a passing remark, the authors digression about Ebroin, a cruel man
and prone to several vices may have been induced by contemporary
controversies about converts to monastic life leaving their commu-
nities once more, against all the rules that had been set by this time:
At one point he, forced by certain circumstances, joined the monastery
which is called Luxeuil, and there, the hair of his head having been
cut, he, with a vow, took up the habit of monastic life. But with the
passing of years, when another king who had been friendly to him
took up the rule of the Franks, he, leaving the monastery behind and
keeping the habit only so far as his tonsure, abandoned the monastic
habit, took a wife, and again snatched up the position of mayor of
the palace. But compared with the irregular and illicit way he suc-
ceeded to the management of his office, he excercised his rule even
more perversely and wickedly.87

84
Passio Leudegarii, c. 6, MGH SRM 5, p. 288: Episcopis tunc quibusdam interceden-
tibus et praecipue intervento antistitis Leodegarii eum non interficiunt, sed Luxovio monasterio diri-
gitur in exilium, ut facinora, quae perpetraverat, evadisset penitendo; ibid., c. 13, p. 296: In
illis igitur diebus adhuc exsul in Luxovio resedebat Ebroinus monachali habitu tonsuratus, simu-
latam gerens concordiam, quasi dum uterque unam, sed disparem exilii accepissent sententiam, con-
cordem ducerent vitam; Ibid. c. 16, p. 298: Iuliano similis, qui vita fincta monachorum tenuit.
85
Passio Leudegarii, c. 16, p. 298.
86
For a recent appraisal of this text, with full references to earlier literature, see
Y. Hen, The Annals of Metz and the Merovingian past, in: Y. Hen and M.
Innes eds., The uses of the past in the early Middle Ages (Cambridge, 2000), pp. 17590.
87
Annales Mettenses priores, s.a. 688, ed. B. von Simson, MGH SRG 10 (Hannover/
322

This verdict is the result of a cumulative literary defamation,88 but

the Annals of Metz introduced a new element: Ebroin had taken
monastic vows (votum), then became an apostate and returned to
power; to make matters worse, he took a wife as well. This was a
real accusation in society busily ordering its monastic life, including
the irrevocability of monastic vows, but to earlier authors, this whole
business about broken vows would not have been understandable;
Ebroin had never become a full member of a religious community,
and neither did any of his fellow exiles, Balthild included.
This distinction did matter, not only when Louis the Pious in 833
staunchly refused to take voluntary monastic vows under duress, as
a follow-up to his public penance,89 but also in the world of Ebroin
and Leudegar. The subtle expressions used by contemporary authors
speak for themselves: these were still political actors who only became
like monks. The same holds true for the monastic exile imposed in
788 and 794 on Charlemagnes brilliant adversary Tassilo, duke of
Bavaria. The Royal Frankish Annals portrayed Tassilo as an oath-
breaker and a fickle ally, but when it came to his monastic exile,
they left his reputation intact. Charlemagne, moved by the love of
God and conscious that Tassilo was his kinsman, decided to spare
his life, and asked him what he wished to do; Tassilo asked leave
to be tonsured and do penance for his many sins, so that he might
save his soul.90 This much resembles the treatment Theudebert III
supposedly met at the hands of his brother, with one important
difference. For the author of the Royal Frankish Annals, monastic
exile and public penance had become identical, with the result that

Leipzig, 1905), pp. 56; transl. Fouracre and Gerberding, Late Merovingian France,
p. 353.
88
Vita Anstrudis, cc. 1113, ed. B. Krusch and W. Levison, MGH SRM 6, pp.
712; cf. J.A. McNamara and John E. Halborg with E. Gordon Whatley eds. and
transl., Sainted women of the Dark Ages (Durham/London, 1992), pp. 29697. In this
preseumably eighth-century text, Ebroin became the quintessential violator or the
cloister and Anstruds exiler until the communitys terrifying prayer and a mir-
acle turned him into the abbess devoted supporter. For more Ebroin-bashing, see
the Vita Philiberti cc. 247, ed. B. Krusch, MGH SRM 5, pp. 5969.
89
Annales S. Bertiniani s.a. 834, ed. R. Rau, Quellen zur karolingischen Reichs-
geschichte II (Darmstadt, 1972), p. 22; De Jong, In Samuels Image, p. 262.
90
De Jong, Paenientia publica and justice, pp. 88082, for a compilation of the
relevant sources. But in the Annales Petaviani s.a. 788, ed. G.H. Pertz, MGH SS 1,
p. 17, Tassilos first disappearance into a monastery presumably Jumiges
was turned into a dishonourable and coercive affair: et Taxilo dux tonsus est, retrususque
Gemetico monasterio. For a context to this passage, see Garrison, The Franks as the
New Israel?, p. 152.
 323

monastic exile became more like a definitive departure from the

world; this was the way in which a true penitent should atone for
his sins. But then again, he or she might not. Penance became
a theme that could be fully exploited in the literary battle fought
over the memoria of those who had left the political arena. Had it
been voluntary and therefore honourable, or a matter of despicable
coercion? This was the question. Meanwhile, the ordinary kind of
monastic exile by means of a mere clerical tonsure continued
to exist. Louis the Pious fully availed himself of this method of cor-
recting and pardoning his opponents in the wake of the revolts of
818 and 833, in the full understanding than none of these men
would enter monastic life forever.

It is within this complicated context voluntary/involuntary, monk/

cleric, good/bad penitent that Carlomans retreat into monastic
life in 747 should be situated. The locus classicus is Einhards Life of
Charlemagne.91 For reasons unknown, but probably because of his
love for the contemplative life, Carloman relinquished his royal office
and went to Rome says Einhard. Having discarded his secular
clothes, Carloman became a monk in St Silvester in Monte Soracte,
his own foundation. Here he remained for a few years, but because
of the hustle and bustle of a steady stream of eminent Frankish pil-
grims to Rome, eager to visit their former lord, Carloman withdrew
to the peace and quiet of Monte Cassino, where he led a religious
life until he died. To my mind, the question whether Carloman opted
out for religious or political reasons is une question mal pose: religious
fervour and political expediency were not necessarily at odds. It looks
as if Carloman remained politically active once he retreated from
the world like Balthild, Ebroin, Leudegar and others.92 Many

91
Einhard, Vita Karoli, c. 2, ed. R. Rau, Quellen zur karolingischen Reichsgeschichte 1
(Darmstadt, 1974), p. 168. For the many instances in which Einhards account has
been taken at face value, Bund, Thronsturz, pp. 3678 may serve as a pars pro toto.
But questions about historiographical complications surrounding the accession of the
Carolingians are now finally being raised: see R. McKitterick, The illusion of royal
power in the Carolingian Annals, The English Historical Review 115 (2000), pp. 120.
92
Liber Pontificalis, nr. 94, Vita Stephani II, c. 30, ed. L. Duchesne, Le Liber Pontificalis
I (Paris, 1955), pp. 4489; R. Davis, transl. and ed., The lives of the eighth-century popes
(Liber Pontificalis) 94.30 (Liverpool, 1992), p. 65. Here Carloman is accused of co-
operating with the unmentionable Aistulf. But God was propitious and Carloman
totally failed to divert to his purpose the steadfast heart of his brother the christ-
ian Pepin king of the Franks. I follow Davis translation. Cf. T.F.X. Noble, The
republic of St Peter: The birth of the papal state, 680825 (Philadelphia, 1984), p. 82.
324

important visitors paid their respects to their former ruler now in

Monte Soracte, conveniently nearby when they made a pilgrimage
to Rome, so the accusation that Carloman was busily plotting with
the Lombards against his brother Pippin is not entirely surprising.
Great men (and women) who disappeared into monastic space did
not entirely relinquish their power and status; by voluntarily leaving
the political arena, such qualities might even be enhanced. Instead,
some questions should be asked about the rhetoric of the sources
usually cited to prove Carlomans exclusively religious motives. There
are two nearly contemporary ones. A version of Annales Petaviani
attributed Carlomans conversion to his compunction about a crush-
ing defeat in Alemannia, where many thousands of men met their
death.93 This hardly amounted to praise. The same holds true for
the Continuation of Fredegars Chronicle, which took Pippins line
entirely: Carloman, burning with religious desire, committed his
realm and sons into Pippins hands and went off to Rome, to become
a monk forever. This was the text upon which Einhard relied. It
left the initiative and honour to Carloman, but it also hammered
home an important point: Carloman left for Rome in order to per-
severe in the monastic order.94 In other words, Carloman had safely
left the corridors of power forever; this was not a story with an open
end. But what if his brother Pippin had been killed in a hunting
accident, like the Langobard King Aistulf in 756? Would Carloman
then have emerged from Monte Cassino, as Aistulf s brother Ratchis
did, in order to take charge of a chaotic situation?95 There are strik-
ing similarities between Carloman and Ratchis: both had a power-
ful brother, both became pilgrims to Rome, both ended up in Monte
Cassino as servants of Christ. But there are also differences, and
the most conspicuous one is that Ratchis departure from the polit-
ical scene or his reappearance, for that matter was not a con-
tentious issue for contemporary authors. The Liber Pontificalis reported
that Ratchis, accompanied by his wife and sons, had been made a
cleric, and then with his entire family received the monastic

93
Annales Petaviani s.a. 746, ed. G.H. Pertz, MGH SS 1, p. 12; cf. Krger,
Knigskonversionen, p. 186, n. 85.
94
Fredegarii continuationes, c. 30, ed. J.M. Wallace-Hadrill, The fourth book of the
Chronicle of Fredegar and its continuations (London, 1960), pp. 1002; Annales Petaviani
s.a. 746, MGH SS 1, p. 11.
95
Bund, Thronsturz, pp. 2149.
 325

habit.96 Intimations of political controversy only surfaced a century

later.97 But Carlomans opting out caused real embarassment. Carol-
ingian historians went out of their way to make it clear that Carloman
had become a real monk, rather than simply submitting to clerical
tonsure; the mans religious fervour was harnessed to this image-
building as well. Thus, Carloman became the one pious exception
proving the rule of hard-nosed Frankish Klosterhaft.
Einhard did much to reinforce this view, claiming that he did not
really know what was behind Carlomans decision. But it appeared,
said Einhard, that Carloman gave up his secular reign because he
burned with longing for the contemplative life. Then, in much detail,
Einhard described the various stages by which Carloman had left
the corridors of Frankish power: he exchanged his secular clothes
for clerical ones and became a monk in Monte Soracte, but when
the world impinged on him too much, he retreated to the most
sacred monastic space in the Carolingian empire: Benedicts very
own Monte Cassino. Was there a more definitive way to leave the
world? Not for Einhard, surely, who wrote several generations later
and had been raised in the Benedictine monastery of Fulda.98 For
Ratchis, however, in 756 the road from Monte Cassino back to Pavia
was but a short one, as it had been for an Anglo-Saxon king recalled
from the monastery to lead his people into battle.99 The same may
have held true for Carloman with regard to Aachen, until in due
course he realised that he had only one option left: to remain in
Monte Cassino, leading the religious life Einhard credited him with.
All that remained by the 820s was a series of skilfull accounts editing
out the tension of the late 740s, crowned by Einhards disclaimer: he
had no idea what had moved Carloman, if it was not genuine piety.
It is worth pointing out that Einhards chapter on Carloman is
preceded by his extremely effective defamation of the do-nothing
Merovingian dynasty, and also by his story about the most infamous

96
Liber Pontificalis, Vita Zachariae c. 23, ed. Duschesne, I, p. 434.
97
Benedicti S. Andreae monachi chronicon, c. 16, ed. G.H. Pertz, MGH SS 3, p. 702.
98
J. Fleckenstein, Einhard, seine Grndung und sein Vermchtnis in Seligenstadt,
in: K. Hauck ed., Das Einhardkreuz (Gttingen, 1974), pp. 96121; De Jong, In
Samuels Image, pp. 233234.
99
Stancliffe, Kings who opted out, p. 154 (Bede, Historia ecclesiastia, II, c. 15
and III, c. 18): Sigebert, the king of the East Angles, who in c. 631 was enticed
back onto the battlefield. This, however, is Bedes vignette of a saintly kind who
would not fight with the sword, for he had become a monk; he wielded a stick
instead.
326

monastic exile of all: Childeric III, the last Merovingian ruler who
was deposed at the Roman pontiff Stephanus orders, and tonsured,
and thrust into a monastery.100 This brief sentence is packed with
powerful statements: all that happened to the hapless last Merovingian
king was done by force, proving the point that Childeric had lost
all claims to royal honour. The only eighth-century source, the Royal
Frankish Annals, say no more than that Childeric was tonsuratus et in
monasterio missus in the year 750. But along comes Einhard. Apart
from crediting Pope Stephanus II consecrated on 26 March 752
with the kings deposition,101 Einhard introduced a much stronger
expression: Childeric was in monasterium trusus once more, a use
of force and loss of honour is implied. Which monastery? Einhard
did not bother to mention a particular place, either from lack of
interest or, more likely, as part of his strategy of defamation. Useless
kings, justly deposed, went to nameless monasteries.102 Conversely,
when it came to Carlomans various stages of monastic retreat,
Einhard named every relevant place: Rome, Monte Soracte, Monte
Cassino. In other respects as well, the two chapters seem to be linked
by their contrast. Whereas the last Merovingian ruler was dragged
off to a nameless monastery, Carloman was merely driven by reli-

100
Einhard, Vita Karoli, c. 1, p. 166: qui iussu Stephani Romani pontificis depositus ac
detonsus atque in monasterium trusus est.
101
Likewise: Breviarium Erchanberti, MGH SS 2, p. 328. Cf. R. McKitterick, The
illusion of royal power in the Carolingian Annals, English Historical Review 115
(2000), pp. 120.
102
The claim that it was St Bertin that served the last Merovingian as his prison
crops up in the Gesta of the Abbots of St Wandrille, in an entry written before 830.
Einhard had a lot of influence in this monastery, so it is not surprising that the
same strong terminology is used; cf. Gesta abbatum Fontanellensium, ed. S. Loewenfeld,
MGH SRG 24 (1886), p. 43; ed. F. Lohier and F. Laporte (Paris, 1938), p. 77. A
privilegium from Pippin III for St Wandrille is cited, and firmly dated as having been
issued on 6 June 750: Quo anno idem gloriosus Pippinus ex consultu beati Zachariae papae
urbis Romae a Bonifacio archiepiscopo unctus, rex constitur Francorum, ablato principis nomine.
Unde rumor potentiae eius et timor virtutis in universas transiit terras. Et Hildericus rex, Meroingorum
ex genere ortus, depositus tonsusque in monasterio Sancti Audomari qui dicitur Sidiu trusus est.
The author adds: Cuius filius nomine Theodericus in hoc monasterio anno sequenti clericus
effectus collocatus est. This had apparently become the official version of the Carolingian
usurpation, pointedly ante-dating Pippins assumption of royal office. However,
Folcuin (d. 990), the author of the Gesta of the abbots of St Bertin recalled that
Childeric was buried in St Bertin, and included a royal charter of 743, confirming
earlier royal privileges, but did not mention Childeric having been thrust (trusus)
into St Bertin; cf. Folcuin, Gesta abbatum Sithiensium, ed. B. Gurard, Cartulaire de lab-
baye de Saint-Bertin (Paris, 1840), cc. 31 and 34, pp. 513. Cf. K.-H. Krger, Saint-
Bertin als Grablege Childerichs III und der Grafen von Flandern, Frhmittelalterliche
Studien 8 (1974), pp. 7180.
 327

gious zeal, making his way gradually, and of his own volition
to the then most sacred places in Western Christendom. Could it
be that these two famous chapters attempted to seal off any troubling
memories of two rulers who chose the honourable option out
withdrawal into a monastery while keeping their options open,
at least in the first years of their monastic exile? If any such mem-
ories were still alive when Einhard wrote, he made sure to bury
them by contrasting Childeric, the powerless victim of coercion, with
Carloman, consumed by religious fervour. In both cases, the result
was the same: two rulers disappeared into monastic space forever.

This has been an exploration of two interconnected issues. On the

one hand, there is the integration of monastic space into the topog-
raphy of political power in the Frankish kingdoms. By the early sev-
enth century, the purity of monastic prayer represented a crucial
source of spiritual power to Merovingian rulers; control over these
precious resources of prayer was to be preserved by royal restraint.
A group of royal monasteries emerged, the recipients of immuni-
ties and exemptions: sacred islands where royal might was both
restricted and reinforced. These monasteries were certainly closely
linked to royal power, but they were by no means royal prisons.
Such loci sancti became places of internal exile to the powerful need-
ing to escape from the political arena if they were to save their lives.
They could fulfill this role beause sacred space had become indis-
pensable to the powers that were as a hands-off zone. This is
not to say that rulers did not invade the cloister, but if they did, it
might cost them their reputation in the long run. On the other hand,
we have to face the problems posed by the rhetoric of the sources.
What seems at first sight to be a straightforward description of a
withdrawal to the monastery, may turn out to be a literary strategy
aimed at preserving or destroying the subjects honour. Reputations
were made or broken by means of depicting monastic exiles either
as masters of their own fate, religiously motivated, or as involuntary
captives in the monastic precincts. In view of this ambivalence
and often, contention in the texts informing us about the power-
ful who disappeared into monastic space, the concept of monastic
prisons is not very helpful. The boundaries between sacred and
328

secular space were reinforced in literary battles between saints and

kings, but the fury about invasion of the cloister helped to distance
worlds that came very near, all the time and of necessity. A reli-
gious community without links to the outside world was doomed to
disappear; the same held true for a monastery that did not know
how to guard the separateness and purity that inspired confidence
from outsiders. Monastic exiles could exist by virtue of the proximity
of monastic and royal precincts, but their lives depended upon the
integrity of sacred space; if such boundaries, however undefined, dis-
appeared, monastic exile was no longer possible. A retreat into monas-
tic space might be both temporary and permanent, depending on
how matters developed. All of this tension between versatile polit-
ical futures and the place-bound ideals of religious life surfaces
in texts dealing with so-called monastic prisoners. Such texts are
never explicit about the precise limits and extent of sacred space. In
fact, early medieval authors tended to be vague about the spatial
details of the loci sancti that became integrated in the topography of
political power. The contours of these sacred spaces do become vis-
ible, however, not because we have any reliable description of the
concentric circles of which monasteries consisted, from inner sanctum
to outer precincts, but because early medieval authors were extremely
sensitive to the issue of border-crossing. Whenever sacred space was
infringed upon, an outburst of protest followed. Through such instances
of indignant outrage, but even more by what was left unsaid, a land-
scape emerges. It is dominated by the secular powerful moving about,
and by saints and their communities staying in place; sacred space
was defined as a hands-off zone where the normal rules of power
politics, honour and revenge no longer pertained. All the same, these
places were closely connected to the secular centres of power. In fact,
they lent a measure of stability to volatile political communities. The
monastic exiles temporarily became part of the stable world, but
then returned to the political arena, where they might survive, or be
killed. Gradually, the boundaries between these worlds became less
permeable. By the ninth century, the former fluidity and double entendre
of monastic exile tended to harden into a strictly regulated public
penance or as a fully fledged monastic profession, but rulers still
needed their time-out zones and used them in more traditional
ways. I have only scratched the surface of these issues, and intend
to return to them in the near future.
 MONASTERIES IN A PERIPHERAL AREA:
SEVENTH-CENTURY GALLAECIA

Pablo C. Daz

This chapter discusses the pivotal role of a special kind of monastic

organisation in a peripheral area (Gallaecia) during a period in which
this region was affected by change. As I shall argue, the monaster-
ies depicted in the seventh-century Regula communis functioned as
places of power, in that they enabled the inhabitants of Galicia to
resist these changes and preserve older patterns of community life
within the monastic confines. These monasteries were able to with-
stand the might of bishops and local aristocratic landowners, well
into the era of the so-called Reconquest. But first we should answer
an essential question: what is a peripheral area? Or, more specifically,
why do we characterize late ancient and early medieval Gallaecia as
a peripheral area?

The first and commonsensical meaning of periphery is that of an

area closest to the outer of a given space perceived as the centre:
an external boundary or region. In this physical sense, Gallaecia,
turned into a province by Diocletians reform, was situated on the
furtherst borders of the Roman world. Gallaecia was a finis terrae in
the west of the Empire, the most remote province of Hispania, which
itself was seen by imperial authors as an outermost boundary.1
However, this concept of remoteness and distance, of eccentricity,

1
J. Arce, Orbis Romanus y Finis Terrae, in: C. Fernndez Ochoa ed., Los finisterres
atlnticos en la Antigedad. poca prerromana y romana (Madrid, 1996), p. 73, with ref-
erences to Silius Italicus, Punica 17.637 (terrarum fines Gades); Plinius maior, Naturalis
historia 5, 76 (Gadibus extra orben conditis); Expositio totius mundi 59 (est ibi finis mundi );
M. Barahona Simoes, Finis Terrae: the land where the Atlantic Ocean begins,
History of the european ideas 15, 46 (London, 1992), pp. 8538. Of a general nature,
C. Nicolet, Linventaire du monde: geographie et politique aux origines de lempire romain (Paris,
1988).
330 .

was not only perceived by those writing in the centre of power, be

it in Rome or in the East. Gallaecias inhabitants also felt that they
were part of a distant world. After the middle of the fifth century,
Hydatius began his Chronicle by saying that he was writing from
the end of the earth (extremus plage), in the context of relating some
events that occurred within Gallaecia, at the edge of the entire world
(intra extremam uniuersi orbis Galleciam).2 This same sense of remoteness
was expressed by Pope Vigilius, who in 538 replied to a series of
letters sent from Braga by Bishop Profuturus. From the perspective
of Rome, the Profuturus had the care of Christs flock in the outer
parts of the world (extremis mundi partibus).3 This perhaps stands to
reason, but a few years later, in 561, Lucrecius of Braga, in the
opening speech of a council held in this metropolitan see, referred
to Gallaecia as in ipsa extremitati mundi.4 Already in the seventh cen-
tury, and in the monastic context upon which this article centres,
Valerius of Bierzo confirmed that feeling of geographical isolation:
. . . in ista ultimae extremitatis occiduae partis confinia.5 This is also an
acknowledgement of a cultural and political remoteness, not only
with respect to an Empire long forgotten, or to an apostolic see that
by the seventh century had practically lost all contact with Hispania,
but even with regard to the centre of Visigothic political power then
located in Toledo.6
Gallaecia was therefore situated within the geographical limits of
the known world, but its isolation was not merely a geographical
matter. In fact, the Gallaecia depicted by seventh-century sources was
a Visigothic province largely unaffected by the developments in the
rest of the kingdom. Isolated politically during the period of Sueve
control, its integration within the structures of the Visigothic kingdom
was slow and probably never complete. Its administrative structure

2
Hydatius, Chronica, Intr. 1 and 6. The references are from the edition by R.W.
Burgess, The Chronicle of Hydatius and the Consularia Constantinopolitana. Two con-
temporary accounts of the final years of the Roman Empire (Oxford, 1993).
3
Vigilius Papa, Epistula ad Profuturum, Intr., Migne PL 84, col. 82932.
4
Ed. J. Vives, Concilios visigticos e hispano-romanos (Barcelona-Madrid, 1963), p. 66.
5
Valerius abbas Bergidensis, De genere monachorum 1, 910; also in Epistola de beatis-
simae Aetheriae 1, 1113: . . . huius occiduae plagae sera processione tandem refulsisset extre-
mitas . . . and 4, 89: . . . extremo occidui maris Oceani litore exorta . . .
6
J. Mattoso, Les Wisigoths dans le Portugal medieval: tat actuel de la ques-
tion, in: J. Fontaine and Ch. Pellistrandi eds., LEurope heritire de lEspagne Wisigothique
(Madrid, 1992), p. 325; cf. J. Leyser, Concepts of Europe in the early and high
Middle Ages, Past & Present 137 (1992), pp. 2831.
 331

was not adapted until the middle of the seventh century, in the
period following 652, when Recesvinth carried out a general admin-
istrative reorganization.7 At time many mints of that territory were
apparently closed, with only three left to function, while the origi-
nal provincial borders were redrawn. All this become clear from the
acts of the Council of Mrida in 666.8 It is hard to say whether
Gallaecia was thought of as similar to the rest of the territory once
subject to Sueve rule. Visigothic sources seem to attribute to Gallaecia
a different status and its own idiosyncrasies, as is shown by literary
texts9 as well as by ecclesiastical10 and legislative documents.11 Whether
or not these references imply anything more than a stock phrase for
defining this part of the kingdom we do not know; if so, this would
mean that the once the Sueve kingdom was subdued, there was a
tacit agreement that a distinct Sueve-Galician aristocracy still existed.
This could explain why Gregory the Great refered to Reccared as
king of the Goths and the Sueves (rex Gothorum atque Sueuorum).12
Most likely this special political status of the former Sueve king-
dom was the result of a de facto situation, not of any formal or legal
recognition. The story of the conquest of Gallaecia by Leovigild leaves
no doubt as to the method used: King Leovigild laid waste to
Gallaecia, took the kingdom from King Audeca, who was taken pris-
oner and subjected the people, the treasure and the fatherland of
the Sueves to his power, making it a province of the Goths.13 The
Sueve kingdom consisted of its treasure, an inalienable property asso-
ciated with the monarchy; of the patria, the territory over which the

7
Cf. E.A. Thompson, The Goths in Spain (Oxford, 1969), pp. 2117.
8
Concilium Emeritensis, a. 666, c. 8: . . . suggerente sanctae memoriae sanctissimo viro
Orontio episcopo, animus eius (Reccesuinth) ad pietatem moverit, ut terminos huius provinciae
Lusitaniae . . . sedem reduceret et restauraret. (. . .) hoc etiam adiciens ut de [eo] id unde ad
Galliciae metropolim diocesis sui fuerat possessum ille reciperet, quamvis longa post tempora, quae
parrochiae // suae fuerant debita.
9
John of Biclar, Chronica, a. 590, 1: Sancta Synodus episcoporum totius Hispaniae
Galliae, et Galleciae in urbe toletana . . .
10
Concilium III Toletanum, a. 589, c. 2: . . . omnes ecclesias Spaniae, Galliae vel Gallaeciae
Concilium XIII Toletanum, a. 683, Tomus: . . . in provincia, Galliae vel Galliciae atque in
omnes provincias Hispaniae.
11
Leges Visigothorum IX, 2, 8: . . . quilibet infra fines Spanie, Gallie, Gallecie vel in cunc-
tis provinciis . . .
12
Epistula IX, 229.
13
John of Biclar, Chronica, a. 585, 2: Leovegildus rex Gallaecias vastat, Audecanem regem
comprehensum regno privat, Suevorum gentem Thesaurum et patriam in suam redigit potestatem et
Gothorum provinciam facit.
332 .

monarchy reigned, and which is specified as Gallaecia; and of its peo-

ple, probably not only Sueves, but all the subjects of the kingdom.
All this was subjected by right of conquest. However, in the same
year a certain Malaricus, probably a Sueve aristocrat, assumed tyran-
nical power and almost managed to reign.14 It is very likely that
such situation instability characterized the entire Visigothic period in
these northwestern border areas of the realm.
An isolated reference from the Asturian period tells us that, while
his father was alive, Wittiza lived in Tude, present-day Tuy.15 The
chronicle of Alfonso III implies that this happened on the instruc-
tions of his father: Wittiza, with whom the king while alive shared
the realm, and ordered to live in the city of Tude, so that the father
had the kingdom of the Goths, and the son, that of the Sueves.16
We do not know whether this decision was motivated by some polit-
ical conflict and the weakness of the kingdom in its final years, when
centrifugal forces threatened to disintegrate the realm. This disinte-
gration is often ascribed to conflicts between monarchs and nobles,17
but we should not rule out separatist tendencies in Gallaecia,18 still
identified by this text as the kingdom of the sueves. Its distinct sta-
tus apparently continued to be recognized. The choice of Tude as
Wittizas seat of power is equally peculiar. Probably the city had
gained a certain prominence within Gallaecia, for after Chindaswinths
reign, only Braga, Lucus and Tude continued to be mints. The city
may have been chosen and promoted by the Visigothic authorities
as an alternative to Braga, which was the centre of the old Sueve
aristocracy and therefore a possible stronghold of resistance. These
references are brief, but no briefer than those concerning the rest
of the peninsula. Except for this intriguing passage about Wittiza

14
John of Biclar, Chronica, a. 585, 6: Malaricus in Gallaecia tyrannidem assumens quasi
regnare vult.
15
Chronica Albendensia XIV, 33: Uittizza rg. An. X. Iste in uita patris in Tudense Hurbe
Gallicie resedit See J. Gil, J.L. Moralejo, J.I. Ruz de la Pea, Crnicas asturianas
(Oviedo, 1985), pp. 15188.
16
Adefonsi Tertii Chronica 4 (Rotensis): (Wittiza) quem rex in uita sua in regno participem
fecit et eum in Tudensem ciuitatem auitare precepit, ut pater teneret regnum Gotorum et filius
Sueuorum. Crnicas asturianas, pp. 11349.
17
L.A. Garca Moreno, El fin de reino visigodo de Toledo. Decadencia y catstrofe. Una
contribucin a su crtica (Madrid, 1975), pp. 140212.
18
E. Ewig, Residence et capitale pendant le haut Moyen Age, Sptantikes und
frnkisches Gallien. Beiheften der Francia 3 (Sigmaringen, 1976), p. 369, considers that
there was a lively autonomism of the Sueve kingdom, fueled by the tradition of the
old sedes regia.
 333

and Tude, we have no information about later Sueve attempts to

regain independence, or about the need for a military occupation.
Suffice it to say that in the seventh century the region was undeni-
able in the political periphery.

Its geographical location and political isolation are two aspects of

the peripheral status of Gallaecia. Of greater interest to us is the
regions closely related social and cultural separateness. Its analysis
takes us directly to the role played by the monasteries in this region
in the seventh century. On the one hand, this was a result of the
peculiar way in which Christianity came to be implanted in the
Hispanic North-West, and on the other, of the impact Christianity
appears to have had on the social structures in a region where prim-
itive economic relations and group associations still predominated.
The introduction of Christianity in Gallaecia was a relatively late
and difficult process. It is almost unanimously accepted that the
North and North-West of Hispania were pockets of paganism until
practically the end of the Visigothic period; already in 561 Lucrecius
of Braga observed that the most isolated regions of Gallaecia still had
not embraced the true faith.19 Whether this was a matter of the sur-
vival of specific beliefs, or, more accurately in my view, a continuity
of ancestral ways of life deeply rooted in rural society, is a matter
for discussion. The way in which historians sometimes jump from
so-called superstition to well-defined forms of paganism is not sus-
tained by the sources.20 The texts simply yield insufficient information
to warrant such conclusions. We are dealing with a province of the
Empire with relatively low levels of urbanization, despite the sur-
vival of pre-Roman housing structures. Hence the ecclesiastic organ-
ization of the countryside occurred very late. Until the end of the
sixth century, no episcopal and hierarchical structures were established.

19
Concilium I Bracarensis, a. 561, Incipit: . . . in ultimis huius provinicae [regionibus] con-
stituti aut exiguam aut pene nullam rectae eruditionis notitiam contingerunt.
20
Cf. J.N. Hillgarth, Popular religion in Visigothic Spain, in: E. James ed.,
Visigothic Spain: New approaches (Oxford, 1980), pp. 360; M. Sotomayor, Penetracin
de la Iglesia en los medios rurales de la Espaa tardorromana y visigoda, Cristianiz-
zazione ed organizzazione ecclesiastica delle campagne nellAlto Medioevo, (Spoleto, 1982), pp.
1753.
334 .

Instead, the bishops had to fight against great landowners who built
churches on their estates. Around this time the Parrochiale Suevum
describes a system of churches depending on a group of episcopal
sees which were being formed simultaneously, thanks to the concil-
iar action initiated by Martin of Braga, the missionary responsible
for the conversion of the Sueves; he seems to have been behind the
councils held in this city in the years 561 and 572.21
For the rest, the first diffusion of Christianity in important areas
of Gallaecia at the end of the fourth century and the beginning of
the fifth was the work of Priscillianism.22 The success of the followers
of Priscillian among the Galician population was largely due to their
acquiescence with regard to indigenous practices of magic and proph-
esying. This is revealed by the acts of the Council of Braga in 561,
and by the testimony of the accusers of Priscillians adherents,23 even
to the extent that the persistence of such practices and the success of
Priscillianism have been seen as two facets of the same reality: the
survival of indigenous culture.24 The Priscillianist ideas were theolo-
gically poor, but not necessarily heretical;25 all the same, the practices
Priscillians followers expressed a radical and challenging rejection of
all worldliness. Their taste for isolated religious services of men and
women in small convents, where consecrated wine and bread was
kept to be used during Mass, and some other customs such as going
barefoot, which gave rise to the suspicion of superstition and mag-
ical arts (superstitio extitiabilis, arcanis oculta secretis26 or magicarum artium

21
P. David, Lorganisation ecclsiastique du royaume suve au temps de Saint
Martin de Bracara, Etudes historiques sur la Galice et le Portugal du VI e au XII e sicle
(Lisboa-Paris, 1947), pp. 182; P.C. Daz, El Parrochiale Suevum: organizacin ecle-
sistica, poder politico y poblamiento en la Gallaecia tardoantigua, in: J. Alvar ed.,
Homenaje a Jos M . Blzquez VI (Madrid, 1998), pp. 3547.
22
Cf. M.V. Escribano, Igrexa e Herexia en Gallaecia: O Priscilianismo, in:
G. Pereira-Menaut ed., Galicia fai dous mil anos. O feito diferencial galego I. Historia 1,
(Santiago de Compostela, 1997), pp. 278321.
23
J. Cabrera, Estudios sobre el priscilianismo en la Galicia antigua (Granada, 1983),
pp. 2213. L. Cracco Ruggini, El xito de los Priscilianistas: a propsito de cul-
tura y fe en el siglo IV d.C., in: R. Teja and C. Prez eds., Congreso Internacional
la Hispania de Teodosio, v. 1 (Valladolid-Segovia, 1997), p. 41.
24
Cf. C. Mole, Uno storico del V secolo. Il vescovo Idazio (Catania, 1978), p. 124.
25
A.B.J.M. Goosen, Algunas observaciones sobre la neumatologa de Priscilianus,
Primera reunin gallega de estudios clsicos (Santiago de Compostela, 1981), pp. 23742;
more extensively, see idem, Achtergronden van Priscillianus christelijke ascese (Nijmegen,
1976), 2 vols.
26
Sulpicius Severus, Chronicorum II, CSEL 1, p. 46, 1.
 335

profana secreta),27 together with their anti-hierarchical conception of

the episcopal office, would soon lead to their relentless persecution
on the part of the bishops.28
The success of Priscillianism in the Hispanic North-West has been
explained by its sympathetic attitude to indigenous religious practices.
Whatever the case, its loss of influence from the middle of the sixth
century did not necessarily the eradication of the older beliefs with
which Priscillianism had been identified. When, in the mid-seventh
century, Braulius of Zaragoza warned Fructuosus of Braga to beware
of the corrupt doctrine of Priscillian,29 he may have been more wor-
ried by its reputed condoning of traditional practice than by any
well-defined dogma. The writings of Martin of Braga, especially his
treaty De correctione rusticorum, the acts of the Second Council of Braga
of 572 and, even at the end of the seventh century, the writings of
Valerius of Bierzo, reveal the resilience of such traditional religious
practices.30
Martin of Braga, a Pannonian monk who came to Gallaecia from
the East, is remembered for his missionary work and his fight against
paganism.31 However, his many activities also included the founding
of monasteries, especially the abbey-bishopric of Dumio, on the out-
skirts of Braga, which was to become an important centre of monastic
culture in the next century. From its foundation onwards, the monastery
at Dumio was closely connected with the Sueve monarchy. Probably
Martin earned his episcopal office as a reward from the Sueve church
for his success in converting the king, and, finally, the Sueve people.
There was no vacant see, and a location as close as possible to Braga
and the Sueve monarch was considered appropriate for the man
who had converted the royal family.32 Hence, the monastery at Dumio

27
Leo I Papa, Epistula 15, 16, 1.4. See B. Vollman, Studien zum Priszillianismus.
Die Forschung, die Quellen, der fnfzehnte Brief Papst Leos des Grossen (St. Ottilien, 1965),
pp. 12238.
28
M.V. Escribano, Iglesia y Estado en el certamen priscilianista. Causa ecclesiae e iudi-
cium publicum (Zaragoza, 1988), pp. 24659.
29
Epistula 44, 75. Cf. L. Riesco Terrero, Epistolario de San Braulia Introducin,
edicin critica, traduccin (Sevilla, 1975), p. 171.
30
S. McKenna, Paganism and pagan survivals in Spain up to the fall of the Visigothic
kingdom (Washington, 1938); M. Meslin, Persistances piennes en Galice, vers la
fin du Vie sicle, Hommages Marcel Renard 2 (Bruxelles, 1969), pp. 51224.
31
A. Ferreiro, The missionary labors of St Martin of Bracara in 6th century
Galicia, Studia Monastica 23 (1981), pp. 1126.
32
Cf. P.C. Daz, El reino suevo de Hispania y su sede en Bracara, in: G. Ripoll
and J.M. Gurt eds., Sedes regiae (ann. 400800) (Barcelona, 2000), pp. 40423.
336 .

became an episcopal see, but without any churches or territory over

which to exercise its jurisdiction. The sphere of influence of the
bishop of Dumio was defined by the monastic family,33 which ulti-
mately gained a tremendous authority. Dumio possessed an undeni-
able moral authority resulting from Martins prestige. Doubtless the
sees legal status was exceptional ,34 and its position with respect to
the bishop of Braga, in whose diocese Dumio had been installed,
was equally unusual.35 On the death of the metropolitan bishop
Lucrecius of Braga, Martin came to occupy the latters see without
abandoning his bishopric Dumio. Initially this did not create a
precedent, for it was only shortly before 656 that Fructuosus once
more exercised episcopal authority over both Dumio and Braga.

Dumios monastic influence was to be far-reaching. Whether bish-

ops managed to conquer rural areas is a matter of debate, but the
monasteries authority and success in this respect is indisputable.
Whereas the representatives of the diocesan clergy could only resort
to the preaching of the gospel, hoping that their churches would
become centres of faith and religious practice, the introduction of
the monasteries offered a model of organization which, as we shall
see, enabled the integration of traditional ways of life into new sys-
tems of values. The priests, with their scant theological training, as
the councils repeatedly pointed out,36 easily adapted to local syn-
cretistic traditions,37 but the monasteries were capable of imposing

33
Parrochiale Sueuum 6: Ad Dumio familia servorum, CCSL 175, 41120.
34
P.R. Oliger, Les evques rguliers (Paris-Louvain, 1958), p. 20, attributes an Irish
origin to this type of bishopric, considering that they were not known on the con-
tinent except in the north of Armorica.
35
An interesting comparison can be made with the seventh-century Meuse val-
ley, where aristocratic monastic foundations, often with their own Klosterbischof,
helped to contain episcopal power. See A. Dierkens, Abbayes et chapitres entre Sambre
et Meuse (VII e-XI e scicles): Contribution lhistoire religieuse des campagnes du haute moyen
ge. Beiheften der Francia 14 (Sigmaringen, 1985); M. de Jong, Carolingian monas-
ticism: the power of prayer, in: R. McKitterick ed., The New Cambridge Medieval
History II, c. 700-c. 900 (Cambridge, 1995), pp. 6258.
36
J. Fernndez Alonso, La cura pastoral en la Espaa romanovisigoda (Rome, 1955),
pp. 69118.
37
Cf. H. Kuhn, Das Fortleben des germanisches Heidentums nach der Christian-
iserung, La conversione al cristianesimo nellEuropa dellAlto Medioevo (Spoleto, 1967), pp.
 337

more or less uniform standards of behaviour, supported by rigid pen-

itential systems. In many places remote from the episcopal centre,
the maintenance of the churches had become a problem, to the
extent that buildings fell into ruin and the daily liturgy was no longer
carried out, as King Egica stated in the Tomus presented before the
sixteenth Council of Toledo in 693.38 The monasteries, however,
fulfilled a role they also assumed elsewhere in Europe. These were
self-sufficient units in a rural landscape, capable of attending to their
own spiritual needs and those of their surroundings; these centres of
religious power were therefore well placed for spreading the gospel
to the remotest corners of rural areas.39 As we shall see, this effectiveness
was ensured by respect for a monastic rule and a ready submission
to the bishop who endorsed it.
The extraordinary proliferation of monasteries in seventh-century
Gallaecia is usually explained by the missionary zeal of the two found-
ing fathers, Martin and Fructuosus of Braga. However, practical
considerations will have mattered as well. Large rural monasteries
were more capable of meeting the demands of peasant communities
than urban bishops. But the latter reasserted themselves in due course.
As a result, the documents of the second half of the seventh century
testify to concerted attempts to reinstate some kind of canonical order
in monasteries accused of laxity.
It is not the object of this paper to study all the monastic mani-
festations of the Hispanic North-West, and neither shall we deal here
with those communities which, with greater or lesser flexibility,
adapted to the canonical and monastic systems which had prolifer-
ated north of the Alps, in response to Eastern models. For example,
in Gallaecia Bishop-Abbot Fructuosus (d. c. 665) founded a number of
monastic communities known from the Vita Fructuosi.40 He also estab-
lished a monastic rule perfectly suited into these traditions, a for-
mula which probably lasted throughout the century. The foundation

74357; R. Knzel, Paganisme, syncretisme et culture religieuse populaire au haut

Moyen Age. Rflexions de mthode, Annales ESC 47 (1992), pp. 105569.
38
Cf. Vives, Concilios visigticos e hispano-romanos, pp. 4845.
39
Cf. G. Gonzlez Echegaray, El monacato en la Espaa nrdica en su con-
frontacin con el paganismo (ss. VIVII), Semana de Historia del monacato cntabro-
astur-leons (Oviedo, 1982), pp. 3556; Hillgarth, Popular religion in Visigothic
Spain, p. 38, A. Barbero and M. Vigil, La organizacin social de los cntabros
y sus transformaciones en relacin con los orgenes de la Reconquista, Sobre los
orgenes sociales de la Reconquista (Barcelona, 1974), pp. 18892.
40
M.C. Daz y Daz, Vita Sancti Fructuosi (Bracara, 1974).
338 .

of the monastery of Samos shortly after 650, on the initiative of

Ermefredus of Lucus, probably followed a similar pattern.41 But here
we are interested in another category of texts. These reveal not only
the variety of responses to monastic initiatives in Gallaecia, but also
ways in which the monasteries served as a focus of regional tradi-
tions, harnessing and adapting them to new beliefs. Religious and
social conflict was important during this period, and should be taken
into account.
The first document which merits our attention is the famous Regula
communis, or Rule of the Abbots. Of uncertain authorship, although
attributed by some scholars to Fructuosus, the Regula communis was
probably a document arising from specific circumstances. It was a
series of instructions issued by an assembly of abbots who attempted
to canonize the many spontaneous monastic initiatives just men-
tioned. The Regula communis is generally thought to have been written
around 660.42 Apparently it should offer an alternative for the rigid
monastic system outlined by Fructuosus in his earlier monastic rule
(c. 645), which had been put into practice in Fructuosus own foun-
dations. The inflexibility of this system was unacceptable to the
monasteries organized outside the monastic traditions represented by
Fructuosus. We get to know more about such unruly monasteries
from the two first chapters of the Regula communis; the first chapter
is of particular interest in this respect:
Indeed, some have the custom of organizing monasteries in their own
homes, through fear of hell, and of joining together in a community
with their wives, children, serfs and neighbours under the steadfast-
ness of an oath, and of consecrating churches in their own homes with
the names of martyrs, and of calling them monasteries. But we do not
call those dwellings monasteries, but rather the ruin of souls and the
perversion of the church. Hence the origin of heresy and the schism
and great controversy over the monasteries.43

41
A. Mund, La inscripcin visigoda del monasterio de Samos, Studia Monastica
3 (1961), pp. 15764, although he believes that the disciplinary reference would
refer to the Regula communis.
42
J. Campos and I. Roca, Santos Padres espaoles II (Madrid, 1971), p. 166.
43
Regula communis 1: Solent enim nonnulli ob metum gehennae in suis sibi domibus monas-
teria componere et cum uxoribus filiis et seruis atque uicinis cum sacramenti conditione in unum
se copulare et in suis sibi ut diximus uillis et nominee martyrum ecclesias consecrare et eas tale
nominee monasteria nuncupare. Nos tamen haec non dicimus monasteria sed animarum perditionem
et ecclesiae subuersionem. Inde surrexit haeresis et schisma et grandis per monasteria controuersia.
See: Campos and Roca, Santos Padres espaoles II, (Madrid, 1971), pp. 172208.
 339

Ch.J. Bishko44 believes that these monasteries, like the presbyterial

ones cited in the second chapter of the Regula communis, were fos-
tered by a segment of the secular clergy strongly opposed to the
Bishop-Abbot of Dumio. Bishkos argument rests on the reference
to heresy and schism in the first chapter, as well as on chapter 20
which speaks of nostra ecclesia.45 He assumes that the Bishop-Abbot
of Dumio was the episcopus sub regula mentioned in the Regula com-
munis, and that the monasteries and monks under his rule lived in
constant struggle against other communities supported by another
part of the clerical hierarchy mentioned in Rules second chapter:
Some presbyters [priests] have the custom of feigning holiness, and
they do so, not precisely for the sake of eternal life, but rather serve
the church as paid workers, and with the pretext of holiness seek the
emoluments of wealth (. . .). They did not live an industrious life in
the monastery (. . .), they preach what they do not observe and follow
the common way of secular bishops, of worldly princes or of the peo-
ple. (. . .). These, just as they rejoice in their advantages, congratulate
themselves on our misfortunes, and plot with full intention so as to
divulge falsely what they have not heard against us and spread and
maintain publicly in the squares what we have not committed, as if
we had been caught in misdeed. Furthermore, those who leave the
monastery because of their own vices are received with applause, pro-
tected and defended by them.46
With regard to the control of monastic life, these texts supposedly
reflect the confrontation of two divergent ethical conceptions, cher-
ished by opposing factions in the Galician bishopric: an ascetic move-
ment pitted against a more worldly group, which was denounced by
the Third Council of Braga held in 675. There a whole series of
disciplinary errors were noted: acceptance of anticanonical liturgical
practices, worldliness, abuse of episcopal power, simony, etc.47 But
the first chapters of the Regula communis may also reflect a conflict
of jurisdiction between a monastic congregation sponsored from

44
Ch.J. Bishko, The pactual tradition in Hispanic monasticism, Spanish and
Portuguese monastic history, 6001300 (London, 1984), p. 22.
45
Regula communis 20 [De fugitiuis]: Quod si et ipsi laici suo eum recipierint consortio et
pariter cum eo contra monasterium exarserint in contumeliam, cuncti a nostra ecclesia expellantur
et nullo nobiscum karitatis foedere copulentur, quousque ueritatem cognoscant, et nobiscum stantes
iniurias ecclesiae uindicantes pari deuotione consurgant.
46
Regula communis 2.
47
See, specially, Incipit and cc. 1, 2, 4, 7 and 8. Vives, Concilios visigticos e hispano-
romanos, pp. 3718.
340 .

Dumio, with substantial influence in other dioceses, and the claims

of bishops who felt their rights were infringed upon. The latter must
have considered Dumios pre-eminance as a flagrant violation of their
episcopal prerogative. Furthermore, as is revealed in the second chap-
ter, conflict also raged over the control of donations and the man-
agement of charities.
If we think of the Regula communis as an initiative of Fructuosus,
and of a monastic congregation backed by Dumio, we must keep in
mind that shortly before 656 Fructuosus once more joined the see
of Dumio to that of Braga, as Martin had done a century before.
The Tenth Council of Toledo (656) was attended by Fructuosus in
his capacity as the metropolitan of Gallaecia. It issued a decree depos-
ing Potamio, the bishop of Braga, who had confessed to fornication,
and declared:
This is what the council decreed concerning Bishop Potamio: with the
full agreement of all of ours we have decided that the venerable
Fructuosus, bishop of the church of Dumio, should be in charge of
the church of Braga, so that by assuming leadership of the church of
Braga he will thus hold together and preserve the entire metropolitan
province of Gallaecia and all its bishops and peoples, as well its pas-
toral care and the administration of ecclesiastical property.48
Fructuosus was probably chosen for this exalted role by virtue of his
prestige, which had given Dumio renewed influence. When Fructuosus
gained the metropolitan see and thus became the head of the Galician
church, this created a favourable climate for the proliferation of
monasteries following the Regula communis. However, we do not know
when Fructuosus died; this must have happened before 675, when
Leudigisus was the metropolitan bishop presiding over the Third
Council of Braga. From then on, the subscriptions to ecclesiastical
councils seem to indicate a conflict, possibly over efforts on the part
of Braga to absorb Dumio. After all, Fructuosus position as a bishop
in two sees depended on his personal prestige. The continuation of
this situation must have been subject to the approval of the monks

48
Concilium X Toletanum, a. 656, Item Decretum pro Potamio episcopo in eodem concilio:
venerabilem Fructuosum ecclesiae Dumiensis episcopum conmuni omnium nostrorum electione con-
stituimus ecclesiae Bracarensis gubernacula continere, ita ut omnem metropolim provinciae Gallaeciae
cunctosque episcopos populosque conventus ipsius omnemque curam animarum et rerum Bracarensis
ecclesiae gubernanda suscipiens ita conponat atque conservet . . .. Vives, Concilios visigticos e
hispano-romanos, p. 321.
 341

of Dumio, who would not renounce their episcopal rights unless they
had a good relationship with their metropolitan bishop. This may
not always have been the case. In 683 Liuva subscribed the acts of
the Thirteenth Council of Toledo as bishop of Braga and Dumio,
whereas two years earlier he had done so exclusively for Braga. In
684 another council was held in Toledo and the see was represented
by an abbot, but in 688, when the Fifteenth Council of Toledo gath-
ered, Dumio was represented by Vincentius and Braga by Faustinus.
After thirty years of unification, a bishop of Dumio appeared on the
scene once more. This marked a climax in the ongoing tension
between episcopal jurisdiction on the one hand, and the monastic
confederation led by Dumio on the other; the conflict, once Liuva
had died, may well have induced the monks to demand, once more,
a bishop of their own.
Subsequently, Faustinus was transferred to Hispalis (Seville), in the
wake of the removal of Sisibert of Toledo, and the bishop of Porto,
Felix, was appointed as metropolitan of Braga.49 We do not know
whether these moves were used to resolve the conflict mentioned
above; when the Sixteenth Council of Toledo (693) confirmed these
new appointments, Felix signed as bishop of Braga, making it clear
that he was also bishop of Dumio. The solution of the problem can-
not have benefited Dumio, for we know that the bishops of this gen-
eral council took a dim view of a monastery that was also an episcopal
see.50 However, it is also clear that at this stage the monastic con-
gregation at Dumio was an extremely powerful organization; for the
see of Braga, losing control of it meant relinquishing authority in a
substantial part of the Galician church.

We have paid some attention to these debates, for they loom large
in present-day ecclesiastical history, but they are in fact only of sec-
ondary importance. In our view, the tension between different types

49
Concilium XVI Toletanum, a. 693, Decretum iudicii ab universis editum. Vives, Concilios
visigticos e hispano-romanos, pp. 5135.
50
Concilium XII Toletanum, a. 681, c. 4, revoked a decision from the times of
Wamba, taken on the initiative of the King, by which the Monastery of Aquis had
become an episcopal see. Vives, Concilios visigticos e hispano-romanos, p. 392.
342 .

of monasteries, and above all the appearance of those monasteries

created by an accumulation of families and neighbours, can be
explained differently. In its first two chapters the Regula communis
describes two kinds of reprehensible monasteries: those built in the
cities by priests with a view to profit, and others that apparently
emerged spontaneously. The latter, it was said, diverged from the
essential principles of Christian charity; they had no stable leader-
ship, sought worldly profit, and its members were solely concerned
with maintaining a wife and children. Everyone should stay away
from these communities, and neither should they be imitated. To
end this situation an attempt was made to subject the spontaneous
monasteries to a discipline, putting them under the guidance of a
bishop who lived by the Rule (per Regulam uiuit or who sub Regula
uiuit probably the bishop of Dumio, as we have said),51 or per-
haps under the supervision of more than one bishop.52 This disci-
pline was to be guaranteed by monthly synods of abbots, where the
abbots gathering in different regions.53
The Rule criticized two kinds of monasteries, yet when we unravel
the disciplinary content of its subsequent chapters we find that its
most important target consisted of the apparently very threatening
monastic associations of families, neighbours and their servants. To
counter these, a rigid abbatial authority was put into place. To
oppose those who, living as they pleased, did not want to be sub-
jected to any superior, and elected as abbot someone who would
allow them their whims,54 the authority of a superior with due
qualifications was to be established. (cc. 3, 5, 10 and 14). In order
to prevent indiscriminate access to monastic life, it was decided that
only free men or ex-serfs with a charter of liberty could enter the

51
Cf. J. Orlandis, El movimiento asctico de San Fructuoso y la congregacin
monstica dumiense, Estudios sobre instituciones monsticas medievales (Pamplona, 1971),
p. 77; idem, Las congregaciones monsticas en la tradicin suevo-gtica, Estudios
sobre instituciones monsticas medievales, p. 102, who follows I. Herwegen, Das Pactum
des Hl. Fruktuosus von Bracara Ein Beitrag zur Geschichte des Suevisch-Westgotischen
Mnchtums und seines Rechtes (Stutttgart, 1907), pp. 5560.
52
Cf. Ch.J. Bishko, Episcopus sub regula or episcopi sub regula? St Fructuosus
and the monasticized episcopate in the peninsular West, Bracara Augusta 21 (1967),
pp. 634; idem, The pactual tradition in Hispanic monasticism, pp. 1920.
53
Regula communis 10: Secundo ut per capita mensium abbates de uno confinio uno se copi-
lentur loco, et mensuales laetanias strenue celebrent . . . These assemblies of abbots were
not a novelty but acquired genuine significance here, cf. A. Mund, Les nciens
synodes abbatiaux et les Regulae SS. Patrum, Studia Anselmiana 44 (1959), pp. 10725.
54
Regula communis 1: . . . quia suo arbitrio uiuunt nulli senioreum uolunt esse subiecti (. . .)
talem praeesse sibi abbatem desiderant ut ubi se uoluerint conuertere quasi cum benedisctione suas
uoluntates faciant.
 343

monastery, who, moreover, would submit themselves obediently to

the discipline of the abbot (cc. 4 and 5). But it soon became clear
that the Rule imposed a discipline that was at loggerheads with more
traditional models of monastic life, something which, as we have
pointed out, also held true of the Regula Fructuosi that was so severe
with anti-canonical practices.55
The sixth chapter of the Regula communis foresaw the possibility of
men entering the monastery with their wives and children: How
men should live without danger with their wives and children in the
monastery (Qualiter debeant uiri cum uxoribus ac filiis absque periculo uiuere
in monasterio). Throughout this long and detailed chapter, the sepa-
ration of sexes is regulated as well as the relationship between par-
ents and children under seven years old. The latter were expected
to be instructed in the Rule so they may be prepared, whether they
are boys or girls, for the monastery where they will live in the future
(ut siue sint pueri siue puellae monasterio prouocentur, ubi habitare futuri erunt).
Clearly boys and girls will live in the same monastery. In later chap-
ters (cc. 15, 16 and 17) some practical arrangements for the sepa-
ration of the sexes are worked out: the conditions under which men
and women can see each other, their placing and conduct when
meeting at a religious service, and so on. It is evident that these reg-
ulations were meant for an environment that accepted the monas-
tic profession of entire families; the continuation of family ties within
the religious community was taken into account, albeit within severe
limits. Ultimately, this system would produce so-called double monas-
teries.56 Moreover, not only families in the sense of parents and small
children entered these monasteries, but also larger groups of rela-
tions and dependents, including serfs. According to the Rule, the
regular meetings should first strive to correct the many men who
showed concern for their wives and children and even for other rela-
tions.57 In a context where the illness of a monk was foreseen it was
established:

55
A. Linage Conde, En torno a la Regula Monachorum y su relacin con otras
reglas monsticas, Bracara Augusta 21 (1967), pp. 12363.
56
J. Orlandis, Los orgenes del monaquismo dplice en Espaa, Estudios sobre
instituciones monsticas medievales, p. 34. Cf. Fernndez Alonso, La cura pastoral en la
Espaa romanovisigoda, p. 492, who understands that the Regula communis was already
in its origins, destined to these double monasteries.
57
Regula communis 13: Solent nonnulli prosuis uxoribus atque filiis aut etiam quibusque
propinquis curam habere more pietatis plerique uero qui non sunt in talibus implicati pro alimento
sunt solliciti.
344 .

We order that all sick monks should lie in male monasteries and we
order that neither mother, nor sister, nor wife, nor daughter, nor rela-
tion, nor a female stranger or a female slave nor any other kind of
woman should care for the men during their illness. If one of the
aforementioned women should happen to be sent with the abbess with
some medicine, she shall not be able to visit him without a male nurse,
nor remain next to him. We order the same regarding the men.58
The first chapter of the Rule criticized the often less than voluntary
monastic professions made by serfs, a theme also taken up by Valerius
of Bierzo (d. c. 695).59 Yet as far as we can see, the Regula communis
by no means tried to eradicate these collective professions. Instead,
the text sought to canonize them. But who were the men and
women who made these collective professions? Once more, we turn
to the Rule and related documents.

When the first chapter of the Regula communis condemns local monas-
teries set up without control, it emphasizes a particular and pecu-
liar aspect of those making their profession:
Estranged from the very neighbours to whom they had bound them-
selves by oath, they separate with intense quarrels and disagreements.
And not simply, but with insults, they carry off each others goods
that, carried away by an illusory idea of charity, had previously put
together to use in common. But if one of them should suffer weak-
ness, they resort to the relations that they left in the outside world for
them to help them with weapons, sticks and threats.60
This situation, presented as characteristic of these arbitrarily estab-
lished monasteries, also occurred among those reorganized by the
Rule, and furthermore, it did so in a way very similar to the one
just observed:
We have found that in not very cautious monasteries those who entered
with their goods, having later lost their fervour, seek with great infamy
the world they left and return to it like dogs to vomit, and try, together

58
Regula communis 17.
59
De genere monachorum 1: . . . tolluntur ex familiis sibi pertinentibus subulci, de diversisque
gregibus dorseni, atque de possessionibus parvuli, qui pro officio supplendo inviti tondentur et nutri-
untur per monasteria, atque falso nomine monachi nuncupantur.
60
Regula communis 1.
 345

with their relations, to snatch back what they had taken to the monastery,
calling upon secular judges and devastating the monasteries with men
of arms.61
This situation is described again with regards to fugitive monks
(c. 20), who, probably with the intention of re-appropriating what
they had contributed, relied on the support of their relations to
recover it. Despite the brevity of these references, it seems clear that
those making their profession in this type of monastery were not
really aware from the start that a monastic profession implied a re-
nunciation of worldly property. Although the Rule points out (c. 18)
that those who wish to enter the monastery must first distribute all
their riches among the poor, this does not seem to have been com-
mon practice. It is very likely that the contribution of each of the
members who entered the community was essential for its mainte-
nance; these were not large foundations with an abundant original
patrimony based on a large estate. Rather these were communities
situated in marginal areas, in rough and steep mountains and inac-
cessible valleys ( fragosa et abrupta montium et inaccessibilia uallium), as
we read in the text. Here agriculture yielded only a meagre liveli-
hood, sufficient for scarcely three months, the Rule says, so it should
be supplemented with livestock.62
Furthermore, these newly professed monks who renounced their
vows seem to have harboured notions of ownership that were very
different from those we know from late Roman law or the Leges
Visigothorum, by which most ecclesiastical institutions and more cen-
tral monasteries lived. The Regula communis evokes conceptions of
property involving extensive communal rights. These may have been
connected with the agricultural or grazing practices of local com-
munities, or by the collective use of undivided public areas. Later
tradition in Galicia considered the family house as something sacred,
a part of a patrimony handed over through the generations; its main-
tenance was more important than the private interest of the family
member who benefited from it by association.63 Something similar
may have pertained in an earlier age.

61
Regula communis 18.
62
Regula communis 9: . . . et insuper uix tribus mensibus per pleraque monasteria abundar-
entur, si sola cotidiana fuissent paxamacia in hac prouincia plus omnibus terris laboriosa.
63
J. Garca Fernndez, Sobre los orgenes del paisaje agrario gallego, Estudios
geogrficos 129 (1972), pp. 75363. Of a more general nature A. Gurevic, Reprsenta-
tions et attitudes lgard de la proprit pendant le haut moyen ge, Annales ESC
27 (1972), pp. 52347.
346 .

As P.D. King maintained, in the Iberian peninsula of the seventh

century the individual could no longer count on an effective net-
work of kinship.64 This observation is correct if we analyse the con-
tents of the great majority of Visigothic sources, especially the legislative
ones. In the Regula communis we find several references to this sub-
ject, embedded in more general statements about the patrimonial
interests of the monastery, monastic discipline and the effective exer-
cise of abbatial authority. We have already seen some examples of
this, such as a concern that those who wish to enter the monastery
should leave their goods to the poor and not to any relative. Apart
from being more virutuous, a donation to the poor would be a more
effective way to renounce ones property than handing it over to a
family member. (c. 4). Likewise there is mention of the elderly who
retire to the monastery, so as to avoid affection for ones relatives
(propinquitatis affectum, c. 8); or, in a more general way, when the sins
and wickedness of monks is discussed, it is said that many are in
the habit of caring for their wives, children and other relatives,
according to customary affection (solent nonnulli pro suis uxoribus atque
filiis aut etiam quibusque propinquis curam habere more pietatis).65 These
attachments are very human, and induced compassion, nostalgia, or
a desire to see their relations once more in those who had converted
to monastic life. Yet in the world we encounter in the Regula com-
munis, with its irregular neighbourhood monasteries, Kings obser-
vation about increasing individualism does not seem to hold true.
Here, ties and loyalties of kinship still generated lasting solidarity,
and most likely communities of patrimonial interests tended to extend
beyond the restricted family circle, and beyond affective emotions.
On at least three occasions the Rule describes what these ties of
kinship were like. A first reference lists those to whom a professing
monk must not leave his goods: He should not give it to his father,
nor mother, nor brother, nor relative, nor blood relation, nor adop-
tive son, nor wife, nor children (non dedit patri, non matri, non fratri,
non propinquo, non consanguineo, non filio adoptiuo, non uxori, non liberis . . .,
c. 4). Elsewhere (c. 13) the Rule discusses the monks vanity about
their family connections: one boasts about the nobility of his geneal-
ogy and lineage, and others make similar claims for their parents,
their cousins, their relatives, their brothers and blood relations and

64
P.D. King, Law and society in the Visigothic kingdom (Cambridge, 1972), p. 222.
65
Regula communis 13.
 347

the like (alius de genealogia et de sua gente fatetur esse princeps, alius de par-
entibus, alius de germanis, alius de cognatis, alius de fratribus et consanguineis
et idoneis). When making it clear that a sick monk should not be
attended by any woman, the Rule sums up the possible relationships
that might be involved: mother, full sister, wife, daughter, relative,
stranger, slave (mater, germana, uxor, filia, propinqua, extranea, ancilla)
(c. 18). Likewise, the Pactum, a document transmitted together with
the Regula communis with which we will deal below, a more general
list of forbidden connections is mentioned: cum parentibus, germanis,
filiis, cognatis uel propinquiis . . ., which covers all possibilities.
The terminological problems are evident. These terms can have
had a special meaning in local tradition which now eludes us, but
clearly that the concept of the late Roman family is not very use-
ful in the Galician context. Part of the terminology conforms to that
of that found in the Leges Visigothorum; at times, however, the meaning
of terms denoting family seems to have a broader scope. A case in
point is propinquus, especially if it is used in opposition to extraneum.66
This broad spectre is best expressed in the already mentioned ref-
erence to someone taking pride in his genealogy and people, a
phrase which perhaps indicates the more encompassing solidarity
leading to a predominance of collective interests. It should also be
noted that the Regula communis has a more elaborate terminology for
degrees of kinship than any other Western monastic rule; whereas
the consanguinei can still be found occasionally, and the expression
propinquus is used once by Isidore of Seville and by Aurelian and
Caesarius of Arles (as opposed to eight times in the Regula communis
and another mention in the Pactum), genealogia does not appear in
any other Rule.67 This is probably because the underlying social real-
ity the Regula communis reflects diverged from the more general pat-
tern in the Iberian peninsula.68

66
This opposition is likewise found in Regula Isidori, whose context, that of the
south of Hispania, is totally different: . . . parentibus uel extraneis . . . (c. 19); . . . propin-
quum uel extraneum . . . (c. 24).
67
Cf. J.M. Clement, Lxique des anciennes rgles monastiques occidentales, 2 vols.
(Steenbrugge, 1978).
68
Cf. D.A. Bullough, Early medieval social groupings: The terminology of kin-
ship, Past & Present 45 (1969), pp. 112, who shows how cognatio, genealogia and con-
sanguinitas are terms used in the sense of kin-group, or kindred, while the persons
who collectively form the group are known as propinqui or parentes. On the termi-
nology of kinship, within the context of Gaul, see R. Le Jan, Famille et pouvoir dans
le Monde Franc (VII e X e sicle). Essai danthropologie sociale (Paris, 1995), pp. 15978.
348 .

One might wonder whether these relationships were a matter of

communities defined by kinship, or of villages with strong neigh-
bourly interests. The existence of such villages is confirmed in the
Hispanic north in the early middle ages.69 Relatives and uicini joined
together to form part of the monasteries denounced in the first chap-
ter of the Rule. In this respect, one should realize that for a long
time in rural areas relations and friends tended be confused.
Members of a village community considered themselves related to
each other, which, given the level of endogamy, was often true.70
To sum up, in a traditional society the only true friends a man can
have are those linked to him by ties of blood.71

This seems to have been the situation that the Regula communis bears
testimony to, and it probably corresponded to a specific level of
social development. When in the fifth century Hydatius wrote his
chronicle, there were indigenous groups in Gallaecia, the Auregenses72
and Aunonenses,73 which were strong enough, according to the chron-
icler, to maintain a prolonged and even successful armed conflict
with the Sueve kings, as was the case of the Aunonenses with whom
the Sueve kingdom was forced to sign a peace treaty. They also had
the ability to get into contact with the Visigoth king in Gaul. These
peoples are not mentioned again, but in the following century the
Sueves still faced the Runcons.74 Martin of Braga observed how
deeply rooted traditional beliefs were, and there existed a very pow-
erful indigenous substratum, as revealed in the already mentioned
Parroquiale suevum.75 This text gives a total of 132 place names, 13
episcopal sees and 119 churches; from the analysis of 89 of these,
Piel deduced that 11 were personal names, 27 Roman-Latin generic

69
J.A. Garca de Cortazar, Les communauts villageoises du Nord de la Pninsule
Ibrique au Moyen Age, Flaran 4 (1982), pp. 5577.
70
J.L. Flandrin, Orgenes de la familia moderna (Barcelona, 1979), pp. 4850.
71
Bullough, Early medieval social groupings: The terminology of kinship, p. 12.
72
Hydatius, Chronica 197.
73
Hydatius, Chronica 229, 235 and 243.
74
John of Biclar, Chronica, a. 572, 3; Isidore of Seville, Historia gothorum, vandalo-
rum et suevorum 91.
75
CCSL 175, ed. P. David, 411.20.
 349

names and 51 pre-Roman names without etymology.76 The first

impression is that the rural area and its social reality to which the
text refers had a mainly non-Romanized substrate, and furthermore
that at least 19 of these names, together with 11 mediaeval inter-
polations, have ethnic origins. In these regions, if a new church had
to be named, the name of a people (or of the segment of a people)
was preferred over a place name.
The family and social relationships mentioned in the Regula communis
confirm these observations. The members of the monastic communities
the Rule attempted to regulate, and at least the neighbourhood
monastery denounced in the first chapter, seem to have been part of
a rural population inhabiting marginal areas, where relatively prim-
itive social relations still predominated.77 These were communities in
which social, economic and family life had never adapted to Roman
influence. This of course does not only hold true for Gallaecia alone,
but for any extensive and largely inaccessible mountain area relying
on hunting and livestock for its sustenance.78 In Galicia, however, this
situation prevailed, to continue throughout the post-Roman centuries.79

We have tried to sketch a context for the people who organized

themselves in the communities regulated by the Regula communis. But

76
J.M. Piel, ber die Namen der sog. Divisio Theodemiri , Romanische Forschungen
71 (1959), pp. 1607.
77
This strength of the elements that we have called primitive is not unanimously
accepted. Cf. G. Pereira-Menaut, Cambios estructurales versus romanizacin con-
vencional. La transformacin del paisaje poltico en el Norte de Hispania, in:
J. Gonzlez and J. Arce eds., Estudios sobre la Tabula Siarensis (Madrid, 1988), pp.
24559; idem, Callaecia, in: F. Coarelli, M. Torelli and J. Uroz Sez eds., Conquista
romana y modos de intervencin en la organizacin urbana y territorial. Dialoghi di Archeologia
12 (Roma, 1992), pp. 31925.
78
Cf. J. Goody, The development of family and marriage in Europe (Cambridge, 1983),
pp. 156.
79
Cf. J. Gaudemet, Les communauts familiales (Paris 1963), pp. 85100; E. Hinojosa
y Naveros, La comunidad domstica en Espaa durante la Edad Media, Obras
II (Madrid, 1955), pp. 33145; J. Martnez Gijn, La comunidad hereditaria y la
particin de la herencia en el derecho medieval espaol, Anuario de Historia del
Derecho Espaol 27 (1957), pp. 221302; L. Garca de Valdeavellano, La comunidad
patrimonial de la familia en el derecho espaol medieval, Estudios medievales de
Derecho Privado (Sevilla, 1977), pp. 295321; T.F. Glick, Islamic and Christian Spain in
the early Middle Ages: Comparative perspectives on social and cultural formation (Princeton,
1979), p. 137; E. Montanos Ferrn, La familia en la alta edad media espaola (Pamplona,
1980), pp. 161328.
350 .

why did these monasteries play such a prominent role? Only by

addressing this question we can understand the place of these pow-
erful communities in the world of late antique and early medieval
Gallaecia and their significance as places of power.
We began our argument by pointing out the marginal or periph-
eral nature of Gallaecia. This marginality did not, however, imply
stagnation; it did not mean that there was no familiarity with clas-
sical forms of social and spatial organization, or involvement in long-
distance trade and cultural contacts, or that there was no similarity
to processes of development known in the rest of the Iberian penin-
sula, and the West in general. In fact, our brief survey of indige-
nous Galician culture from the fifth to the seventh centuries reveals
qualitative and quantitative changes. Hyadatus Aunonenses can prob-
ably be identified with the church of Aunone, mentioned in the
Parrochiale among the churches of the diocese of Tude. This may
indicate a process of territorialization. By the beginning of the sev-
enth century such ethnic groups seem to have disappeared. Valerius,
who gave by far the best description the rural environment of Gallaecia,
no longer mentions them. Whereas in the fifth century such peo-
ples still represented military force and autonomous political capac-
ity, their names had turned into a mere identifying labels by the
sixth century. By the time Valerius wrote, a brief description of
ancestral and popular practices sufficed. Most likely, the traditional
society was breaking down, and the Regula communis, as well as the
organization of family and neighbourhood communities in monastic
style, was a sign of this transition. These monasteries provided new
forms of social solidarity in a rapidly changing world.
On the other extreme of the social spectrum, among the large
land-owning aristocracy, the development that took place from the
fifth to the seventh century seems to have followed generally known
patterns. The invasions of the fifth century probably threw the Galician
aristocracy into a temporary state of confusion, but apparently the
ensuing conflict of loyalties did not affect the aristocracys power and
status. Although some events might lead us to think the Sueves
attacked or killed the representatives of this social group,80 other
events make it clear that some Galician aristocrats were willing to
cooperate with the Sueves.81 In the following centuries the large

80
Hydatius, Chronica 191, 194 and 225.
81
Hydatius, Chronica 240.
 351

landowners occupied an essential position in the social network of

the kingdom. Their unquestionable economic autonomy was accom-
panied by an undeniably important role in religious affairs; they built
churches on their lands and had so much power that they came
into conflict with the bishops, since they administrated these churches
and managed them, ignoring diocesan discipline.82 They probably
also had a large measure of autonomy in the fiscal and political
domain. When Leovigild was carrying out his campaigns against the
frontier of the Sueve kingdom, he came up against a certain Aspidius,
whom he captured with his wife and children; the chronicler defined
him as the lord of the place (. . . loci seniorem . . .).83 Whether the
man in question was a large Galician-Roman landowner or an indige-
nous aristocrat, his power must have been based on this wealth and
the accumulation of landed property. The church itself controlled
large areas of land; its wealth included large monasteries, among
which Dumio was particularly conspicuous. Its huge fortune and the
size of its properties can be deduced from the will of Abbot Ricimiro,
revoked by the Tenth Council of Toledo held in 656.84
Evidently in this period the accumulation of land on the part of
the local aristocracies and the church was accomplished at the expense
of the small landowners and the peasant communities. This process,
which is evident from the writings of Valerius of Bierzo,85 hit a hard
blow at traditional arrangements of property. Most likely it meant
the appropriation of common pastures, and ultimately the absorp-
tion of entire peasant communities into the landowners patronage,
as well as the breaking up of the undivided patrimonies which were
essential for maintaining the peasant community and the solidarity
of the kin-group.
These conflicts are implicit in the text of the Regula communis.
Disturbances were caused by relatives who came to recover prop-
erty contributed by the newly professed, and the first chapter of the
Rule reveals the instability of these pseudo-monasteries where, upon
leaving, the run-away monks fought over patrimonies they had donated

82
Concilum II Bracarensis, cc. 5 and 6. Vives, Concilios visigticos e hispano-romanos,
p. 83.
83
John of Biclar, Chronica, a. 575, 3: Leovegildus rex Aregenses montes ingreditur, Aspidium
loci seniorem cum uxore et filiis captivos ducit opesque eius et loca in suam redigit potestatem.
84
Vives, Concilios visigticos e hispano-romanos, pp. 3224.
85
R. Frighetto, Panorama econmico-social del NO. de la Pennsula Ibrica en poca visigoda.
La obra de Valerio del Bierzo (Tesis Doctoral. Salamanca, 1996), forthcoming.
352 .

to the monastery. The sudden and violent attacks suffered by the

communities under the Regula consensoria monachorum86 were all signs
that previous patterns of property ownership were breaking down,
a process to which the monastic organisation contributed as well.
After all, the very notion of irrevocable donations to a monastic com-
munity by an individual was directly opposed to the concept of undi-
vided peasant patrimonies.
The peasant communities organized themselves as monasteries in
an attempt to preserve their integrity. This a priori needs to be
demonstrated. A close reading of the Regula communis and the criti-
cism it met, as well as of the De genere monachorum by Valerius and
the Consensoria, reveals that the motive to found these monasteries
was not primarily a religious one, at least not religious in the way
a monastic profession required. The first criticism of the authors of
Regula communis was that some had joined forces for fear of hell, and
thus sought to gain more than was possible in the world outside;
their behaviour was comparable to that of lay people and worldly
princes (c. 1). Of those who professed to the Rule, it was observed
that many did not come to the monasteries through love of Christ,
but were driven by weakness, rather than by religious considerations
(c. 9). Allegedly, they were scared of impending death and anguished
by illness, acting not from love of Heaven but out of fear of the
punishments awaiting them in Hell (c. 18). From the more general
chapters of the Rule it becomes clear that economic order was both
of major importance and discipline difficult to impose. It was nec-
essary to accommodate elements that did not fit monastic tradition:
cohabitation of the sexes, family life, and the care of the helpless
elderly of the community. It also transpires that the monastic space
envisaged was not a unity, but a dispersed group of buildings
probably a village converted into a monastery.87

86
Regula consensoria monachorum 7: . . . incursio repentina aut hostilitas . . ., Migne PL
66, cols. 9936. Ch.J. Bishko, The date and nature of the Spanish Consensoria
Monachorum, American Journal of Philology 69 (1948), pp. 3823, felt that this attack
would respond to the same cause reflected in chapter 18 of the Regula communis,
where the relatives of a monk tried to recover by force what he had brought to
the monastery; or even more clearly in the reference in chapter 3 where the pos-
sibility is posed that an enemy of the monastery should appear and try to take
something and carry it away by violence . . . (Si certe aliquis insequutor monasterii accesserit
et aliquid auferre conauerit . . .).
87
P.C. Daz, Formas econmicas y sociales en el monacato visigodo (Salamanca, 1987),
pp. 904.
 353

What possible advantages could this have? In principles the arbi-

trarily formed monasteries are presented as a failure. However, the
procedure seems to have become acceptable once some order was
imposed. The formation of a monastic congregation fulfils this require-
ment in several ways. On the one hand, the acceptance of a disci-
pline set by a bishop afforded legitimacy; this, and the Rule sustaining
it, had the support of the clerical hierarchy. Given the fact that the
diocesan bishop was also the bishop of Dumio, it was a system both
favoured by tradition and by an ecclesiastical structure with tremen-
dous economic power. To the church of Dumio, this was a conve-
nient base for spreading the faith; submission to the Rule and a
regular supervision of activities and discipline to some extent safe-
guarded unity of purpose and orthodoxy. To those who were part
of these monasteries, organization meant stability. The support of
Dumio was a guarantee against attempts at annexation on the part
of the diocesan church and laymen alike (c. 3). Meanwhile, the old
structure of kinship and even of neighbourhood could be maintained.
Furthermore, and this was essential, these communities operated on
a contractual basis. They were not governed by an externally imposed
discipline, but by a discipline sustained by mutual agreement, which
included the possibility of expulsion.
To all this we must add a phenomenon that may have been essen-
tial for the continuity of the community and the preservation of tra-
ditional concepts of property: the Christianisation of rural areas had
a direct influence on customs relating to wills. The request that faith-
ful Christians should yield a part of their inheritance to the Church
Christs share as St Augustine called it88 or that the son who
was a priest or monk should not be forgotten in the will, are to be
found everywhere in late ancient Christian literature. We should also
remember that a voluntary donation, seeking divine favour, would
soon become a universal custom, which in turn was to become one
of the main mechanisms for accumulating property on the part of
churches and monasteries.89 This undoubtedly altered strategies of

88
Augustinus, Sermo 355, 4: Sed plane, si faciat quod saepe hortatus sum: unum filium
habet, pentet Christum alterum; duos filios habet, putet Christum tertium; decem habet, Christum
undecimum faciat, et suscipio. Cf. J. Gaudemet, LEglise dans lEmpire Romain (IVV sicles)
(Paris, 1958), pp. 2958, with reference also to Salvien of Marseilles (Timothee ad
ecclesiam Libri IV ).
89
Daz, Formas econmicas y sociales en el monacato visigodo, pp. 457, where the case
of Hispania is analyzed.
354 .

inheritance; the church demanded documents proving rights of prop-

erty, at times infringing upon local customs. By encouraging these
kind of donations the church stemmed the flow of such legacies to
the family. In practice, it broke up family property structures with
a broad collective base. The foundation of monasteries based on
family or neighbourhood groups halted this process. It created a
closed circuit of transfer of property and inheritance, thus impeding
the disintegration of traditional structures and safeguarding the integrity
of patrimonial property. This is why abandoning the monastery
became an act especially condemned by the Regula communis as well
as the Pactum and the Consensoria. The monastery became a defen-
sive structure of the peasant community against innovations dictated
by the ecclesiastical hierarchy.
The Pactum is crucial to this argument. This was a contractual
document given to those who professed to monastic life, freely commit-
ted themselves on entering the monastery, which limited the author-
ity of the communitys superior, and therefore also that of the group
of supervising abbots. The Pactum became a substitute for the cus-
tomary law that governed the peasant community and gave it cohe-
sion. In the model of such a pactum transmitted together with the
Regula communis we find some examples of the original community
rule inspiring these contracts. For example, after affirming that they
would humbly accept their abbots discipline in accordance with the
Rule, and promising that they would follow his teaching, the text
states that if someone grumbling against the Rule and your author-
ity should be stubborn, disobedient or perverting the law, then we
shall all have the power to meet in an assembly, and, after having
read the Rule in the presence of everyone, to prove his guilt in pub-
lic.90 To call upon a meeting of the entire community, on the ini-
tiative of the rank and file of the monks and in cases when the
authority of the abbot was called into question, was not common in
monastic rules, which usually upheld the principle of obedience and
the disciplinary superiority of the abbot or his representatives. However,
the Pactum did authorize the superior to act against any of the mem-
bers of the community who, with the aid of a relative or another
monk, had plotted secretly contra regulam; he could impose a solitary

90
Pactum, ll. 6835: aliquis ex nobis contra regulam et tuum praeceptum murmurans, con-
tumax, inoboediens, uel calumniator, existeret tunc habeamus potestatem omnes in unum congre-
gare, et lecta coram ommnibus regula culpam publice probare.
 355

confinement of six months, with the monk being shorn and clad in
penitential garb.91 If the culprit did not accept this, he was to be
stripped, receiving 72 lashes, and, after having taken off his monas-
tic habit, so he is undressed from [the clothes] he was dressed in
upon his entry, he should be cut off from and expelled from the
community with public shame (deposita ueste monasterii, indutus quod in
introitu exutus est scissum notabili cum confusione a coenobio expellatur).92
Expulsion was also prescribed by the Regula communis for the ex-
communicated who continuously persisted in their errors. This was
expressed in similar terms: Taken into the meeting he should be
divested of his monastic habit and clothed in the dress he once
brought when he came from the world; had he should be expelled
from the monastery with shameful notoriety (in conlatione deductus exu-
atur monasterii uestibus et induatur quibus olim adduxerat saecularibus; et eum
confusionis nota a monasterio expellatur . . .).93 Given that a monastic pro-
fession was meant to be irrevocable, an expulsion was a drastic pun-
ishment in monastic tradition. It is contemplated, for example, two
times in the Regula Benedicti (cc. 28 and 71), but it is unknown in
Hispania out of the North-West pactual context.94 Here, expulsion
was a matter of the community protecting itself against errant mem-
bers, according to the rules which had governed their peasant com-
munities. Expulsion from the monastery resembled the exclusion from
the neighbourhood group, or from the kin-group within which the
individual received protection.
The opposite case, that of a monk who decides to leave the com-
munity because of some vice, is dealt with in a similar way. Once
apprehended by the civil authorities he must immediately be sub-
jected to the discipline of the Rule by his superior; if he seeks refuge
elsewhere, he will be excommunicated without ever being reconciled,
not even on his deathbed.95 Significantly, those supposedly could give

91
Pactum, ll. 6902: . . . per sex menses indutus tegmine raso aut cilicio, discinctus et dis-
calceatus in solo pane et aqua in cella obscura exerceat quodlibet opus excommunicatis.
92
Pactum, ll. 6946.
93
Regula communis 14. The text foresees that the guilty party may try to defend
himself with the aid of his relatives (et cum propinquis se uindicare maluerit), which
agrees with the idea that expulsion is a punishment, separation from the commu-
nity and its protection.
94
The Consensoria also stated that monks could be expelled for compelling rea-
sons. Regula consensoria monachorum 4: Sed si contingerit ut aliquis ex qualibet causa neces-
sitatis a monasterio fuerit abstractus . . .
95
Pactum, ll. 696705.
356 .

protection to the runaway monk were the bishop, or someone under

his authority, or a layman (episcopus uel eius qui sequitur ordo, aut laicus),
which once more indicates a tension between a monastic congrega-
tion depending on a bishop-abbot, presumably the one of Dumio,
on the one hand, and the rest of the bishops or at least some of
them on the other. The contractual nature of the power structure
within the community is also revealed at the end of the Pactum. If
the abbot treats any of the monks unjustly, with pride or anger, or
if he is guilty of favouritism, the monks have a right to be heard,
and the superior must bow to the Rule (in communi regula ceruicem
humiliare et corripere et emendare), and if he has no intention of mending
his ways,
we shall also have the power to bring in the other monasteries, or at
least to call to our meeting to our congress the bishop who lives accord-
ing to the rule, or the count who is a catholic defender of the church,
so that you shall mend your ways in their presence.96
The fact that this Pactum has been transmitted together with the
Regula communis has been taken as proof that this was the very pactum
referred to in chapters 8 and 18 of the Rule. Whether the system
of signing the contractual agreement of stability was already in exist-
ence when the Rule was drawn up, or whether it arose from some
immediate need, or under pressure from rival forms of monasticism,
is a matter for debate.97 There is also doubt as to whether the text
known as Regula consensoria monachorum is another model of pactual
contract, and if so, which monasteries it applied to. For a while this
text was considered a rule of Priscillianist monks. However, its struc-
ture, its references to the agreements entered into for the sake of
stability and for the preservation of the order in the case of exter-
nal attack, as well as the explicit reference that the text makes to
itself as a pactum,98 have lead scholars to think this was another con-
tractual model that was not necessarily associated with the Dumian
community. Bishko considered this text to be connected with a specific

96
Pactum, ll. 705717: tunc habeamus et nos potestatem cetera monasteria commouere aut
certe episcopum qui sub regula uiuit uel catholicum ecclesiae defensorem comitem et aduocare ad
nostram conlationem ut coram ipsis te corripias.
97
A. Linage Conde, La autoridad en el monacato visigodo, Ligarzas 7 (1975),
pp. 224; Bishko, The pactual tradition in Hispanic monasticism, pp. 203.
98
Regula consensoria monachorum 6: quia non poterit proprie retinere quod per pactum ad
omnes pertinere.
 357

type of consensorial monastery,99 which was perhaps on good terms

with those represented by the Regula communis. After all, the latter
was not condemned in the Regula consensoria, and the two texts have
been transmitted in the same collections of monastic rules.100 According
to Bishko, the consensorial monasteries represented an independent
phenomenon, unrelated to the presbyterial or neighbourhood monas-
teries criticized in the Regula communis.101 These communities probably
also had some pact; as such, Bishko identified the iuramentum or sacra-
menti conditio mentioned in the first chapter of the Regula communis.102
In Bishkos view these monastic pacts were yet another example of
the various ways in which monastic orthodoxy adapted to indige-
nous custom and vice versa,103 but in no case should such oaths be
treated as an example of Germanism coming from Sueve or Visigothic
influences, as many scholars are inclined to, taking Herwegens lead.104
The Regula communis and the Pactum both clarify the nature of the
relationship between monastic communities proper, and those super-
vised by the bishop-abbot and the congress of abbots. The latter
represented an agreement between powers and a mutual acceptance
of discipline and doctrine, in exchange for the maintenance, in as
far as possible, of older conceptions of power, based on the struc-
tures of kinship and the social environment from which such com-
munities arose. This type of monastic association would enable local
communities to preserve some of their former group cohesion, with
the support of the ecclesiastical authorities of Dumio or of Dumio/
Braga and the lay powers in the region.
The chapters of the Rule reveal how collective tasks, previously
taken care of by the rural community, were now undertaken by the
monastery. The Rule foresees, as an important task, the reception

99
Bishko, The date and nature of the Spanish Consensoria Monachorum, pp.
37795.
100
Jordanus of Quedlinburg, Iordani, eds. R. Arbesmann and W. Humfner, Iordani
de Saxonia Liber Vitas Fratrum (New York, 1943), pp. lxxvilxxvii. The editors attribute
the Consensoria to Fructuosus of Braga. However, G. Turbessi, Regole monastiche antiche
(Roma, 1974), pp. 2957, includes it in the monastic legislation of St Augustine.
101
Herwegen, Das Pactum des Hl Fruktuosus von Bracara, pp. 768, identified it as
the pact of the monasteries condemned in chapter 1 of the Regula communis.
102
Bishko, The date and nature of the Spanish Consensoria Monachorum, p. 392;
idem, The pactual tradition in Hispanic monasticism, p. 21.
103
Bishko, The pactual tradition in Hispanic monasticism, pp. 234, although
he considers that the nature of this indigenous substratum has not yet been determined.
104
Herwegen, Das Pactum des Hl Fruktuosus von Bracara, p. 26, nn. 1 and 2.
358 .

and care of the elderly and sick, the care of children, and further-
more unusual tasks such as the redemption of captives, which should
probably be seen in the light of the climate of violence that the Rule
itself reflect, and especially the Consensoria.105
Seen from this perspective, the monasteries of the Regula communis
incorporated earlier structures of peasant power, which were then
associated in a monastic confederation and sponsored by the abbey/
bishopric of Dumio in what the Rule calls nostra ecclesia (c. 20), a
device which, despite certain restrictions, allowed these communities
to keep their own identity, turning them into a powerful network
within Gallaecia in the second half of the seventh century. This con-
federate and pactual structure enabled these members of these
monastic communities and the members of former peasant commu-
nities to withstand the might of the large landowners and the dioce-
san church, which, by contrast, founded the presbyterial monasteries
denounced in the Rules second chapter. According to this text, the
presbyterial monasteries encouraged defections and would receive
and protect those who had abandoned the monastic communities
subject to the discipline of the Sancta communis regula. This conflict
probably reflects the crisis and social polarisation of Visigothic soci-
ety during those final years of the monarchy.106
It is rather difficult to determine the success of the monasteries
following the Regula communis. Towards the end of the seventh cen-
tury, Valerius at great length condemned monastic communities con-
sisting of families and their serfs. Most likely he not only had the
monasteries criticized by the Regula communis in mind, but also those
sponsored by this very Rule. Despite occasional disagreement, the
strategy of turning local communities into monasteries seems to have
been sufficiently effective to withstand ecclesiastical resistance and

105
P.C. Daz, Redimuntur captiui. A propsito de Regula Communis IX , Gerion
10 (1992), pp. 28793. About the liberation of captives as a constant theme in
Merovingian and early Carolingian hagiography, see F. Graus, Die Gewalt bei
den Anfangen des Feudalismus und die Gefangenenbefreiung der merowingischen
Hagiographie, Jahrbuch fr Wirtschaftsgeschichte 1 (1961), pp. 61156. About children
and their care in Western monasteries, see Mayke de Jong, In Samuels Image. Child
oblation in the early Medieval West (Leiden etc., 1996).
106
Bishko, The pactual tradition in Hispanic monasticism, p. 22, who also
believes that the neighbourhood monasteries were encouraged by the diocesan clergy
at odds with Dumio, which is more difficult to sustain.
 359

local aristocracies. In the whole of northern Hispania, neighbour-

hood and family monasteries organized in this way proliferated dur-
ing the Reconquest, which probably slowed down the introduction
of a manorial system.107

107
Ch.J. Bishko, Gallegan pactual monasticism in the repopulation of Castille,
Estudios dedicados a Menndez Pidal, vol. II (Madrid, 1951), pp. 51331; published with
an Additional note in Spanish and Portuguese monastic history (London, 1984), pp.
51331 and 532A36A; idem, The pactual tradition in Hispanic monasticism,
pp. 2543; J. Orlandis, Los monasterios familiares en Hispania durante la Alta
Edad Media, Estudios sobre instituciones monsticas medievales, pp. 12564. More recently
J.M. Mnguez, Ruptura social e implantacin del feudalismo en el Noroeste penin-
sular (Siglos VIIIX), Stvdia Historica. Historia Medieval 3 (1985), pp. 732.
This page intentionally left blank
 AEDIFICATIO SANCTI LOCI: THE MAKING OF A
NINTH-CENTURY HOLY PLACE

Julia M.H. Smith

It happened at that time that Tethuiu was sent out under obedience
by the holy father of the monastery of Redon. A powerful man named
Ronuuallon had given his house, built of planks of wood, to the holy
monks for the sake of his soul, and so the aforementioned monk had
been sent to collect it and bring it to the monastery with carts and
oxen. Having done this, he came towards the monastery with the build-
ing materials and carts. But when they came down from the top of
the hill which overlooks the monastery, one of the carts broke loose,
careered away and knocked down one of the servants, named Ioucum,
so that his hips and arms seemed to be shattered. When the man of
God saw this, he stood stunned, expecting nothing other than the
death of his servant. He therefore began to pray hard to God for his
recovery. As he was praying to God in this way, his servant quickly
rose from the place where he had lain, safe and unharmed. In this
event, the mercy of God did not fail his servant.1
Institutionalised monasticism was such an integral aspect of Carolingian
society that we run the risk of taking it for granted. Landowners on
a massive scale, home to many hundreds of monks, powerhouses of
prayer and politics, the major monasteries of the Carolingian empire
transformed the landscape as much by their politics and religiosity
as by their splendid buildings and vast agricultural estates. Part of
their success lay in their ability to exploit the potential of the writ-
ten word by developing archives of written documentation and by
projecting powerful, self-justifying images in a wide variety of tex-
tual formats. Throughout the ninth, tenth and eleventh centuries,
textuality became so implicated in the ideology of monastic reform

1
Gesta sanctorum Rotonensium, II.8, ed. C. Brett, The monks of Redon: Gesta sanctorum
Rotonensium and Vita Conuuoionis (Woodbridge, 1989), p. 169, adapted from Bretts
translation. Hereafter GSR.
Whilst writing this paper I have benefited greatly from the support, specialist
advice and comments on drafts of Donald Bullough, Julia Crick, Mayke de Jong,
Janneke Raaijmakers, Barbara Rosenwein, Simon Taylor and Frans Theuws, all of
whom I thank warmly.
362 ..

that the achievement of an enduring corporate existence presupposed

a written collective memory. In general, the religious communities
which lack documentation are those which failed to establish an insti-
tutional existence, were denigrated or silenced for not conforming
to the high standards of reforming ideals, or which fell outside the
hegemonic power structures of the day. Such communities ascetics
living within the family home, clergy in loose forms of community,
womens monasteries remain houses without history, undocu-
mented because powerless.2 These places have benefited from recent
interest in marginal communities of all sorts: their existence serves
to highlight the documentary strategies which accompanied the
monastic bid for power.3
How, then, did a new religious community build an enduring
existence? How was a corporate identity achieved, maintained, re-
corded and transmitted to posterity? How did the use of the written
word contribute to this? This paper asks these questions with respect
to one particular ninth-century monastery, Redon (Ille-et-Vilaine).
It answers them by deploying two late ninth-century sources, the
monasterys archive and an anonymous work on its history and holi-
ness, commonly known as the Gesta sanctorum Rotonensium. The tale
of Tethuius runaway wagon occurs in the latter and adumbrates
the three themes I shall pursue: place, holiness, and aedificatio in its
dual senses of material construction and spiritual improvement.4 It
hints too at the wider environment in which this project of aedificatio
took place one of donors, material resources, labourers, prayer
and intercession, and of course abbatial authority. Just as the planks
of Ronuuallons house were disassembled, transported and reused to

2
P. Stafford, Queens, nunneries and reforming churchmen: gender, religious
status and reform in tenth- and eleventh-century England, Past & Present 163 (1999),
pp. 335 at pp. 145.
3
G. Declercq, Originals and cartularies: the organization of archival memory
(ninth to eleventh centuries), in: K. Heidecker ed., Charters and the use of the written
word in medieval society (Turnhout, 2000), pp. 14770, at p. 165, quoting O. Guyotjeannin.
I am extremely grateful to Marco Mostert and Karl Heidecker for allowing me
access to this volume in advance of its publication.
4
The dual senses of aedificatio are contained within the text of the GSR: the pref-
ace to book II (pp. 1457) speaks of both the aedificatio mentium credentibus and the
aedificationem sancti loci with planks and beams. The association of spiritual formation
with physical place is also stresssed in III.4: de multis pauca uobis referam, ad aedificationem
uestram et ad firmamentum sancti loci (p. 199). For further uses of aedificio, aedificare in
both senses see GSR I.1, II.9 (pp. 109, 171).
 - 363

make a completely different building in the monastic precinct, so

this paper dismantles Redons ideological self-image, examines its
constituent parts and reassembles them into something very different,
an exploration of the role of record-keeping, rhetoric and ideology
in the building of a ninth-century holy place.

T R

Redons project of textual aedificatio forms the second part of this

paper, for it is best assessed against the background of a more con-
ventional account of its institutional development. This first section
tells a more familiar tale of land, politics and patronage, but in an
unfamiliar environment. It also supplies the context within which the
monks who contributed to the written project lived and worked. As
will become clear, by commencing outside the cloister and then nar-
rowing attention to the texts produced or kept within it, my strategy
parallels that of Redons house author.5
Redon was founded in 832 at the confluence of the river Vilaine
and one of its tributaries, the Oust, on a plot of land just within
the diocese of Vannes which Ratuili, a major local landowner, gave
to Conuuoion, founder and first abbot.6 The Vilaine and its tribu-
taries form the major arterial waterways of a wide region; located
near major Roman roads and within the rivers tidal reaches, Redon
enjoyed a strategic location (fig. 1).7 Until 851 the river also formed
the boundary between Carolingian Neustria and Brittany, a province
where Carolingian control was always indirect and usually contested.8
In this sense, Redon was marginal, a frontier location whether viewed
from a Frankish or a Breton perspective. But the political frontier
paid scant attention to cultural or linguistic difference and indeed,
the basin of the Vilaine and its affluents provided the means to cre-
ate economic, tenurial and familial networks which criss-crossed its

5
Below, p. 378.
6
Cartulaire de Redon, ed. A. de Courson (Paris, 1863), no. 1. Hereafter cited as
CR and the appendix as CR A.
7
N.-Y. Tonnerre, Les pays de la Basse Vilaine au haut moyen ge, Mmoires
de la Socit dHistoire et dArchologie de Bretagne 63 (1986), pp. 2972; G. Astill and
W. Davies, A Breton landscape (London, 1997).
8
Cf. J.M.H. Smith, Province and empire: Brittany and the Carolingians (Cambridge,
1992); A. Chdeville and H. Guillotel, La Bretagne des saints et des rois, V eX e sicle
(Rennes, 1984).
364 ..

Fig. 1. Redon and environs. Small caps: mentioned in the Gesta sanctorum
Rotonensium; lower case: not mentioned in the GSR.

banks. Redon fully exploited these possibilities. In terms of regional

topography, therefore, Redon was a central place.9 In historiographical
terms, Redon has been central to early medieval Breton history, but
entirely peripheral and usually ignored from a Frankish per-
spective. Its centrality in Breton terms depends, however, on a pat-
tern of documentation more characteristic of Frankish than of Breton
churches; this in turn urges a reassessment of its peripheral status in
Carolingian terms.10

9
On the politics and perceptions of Redons frontier location, see J.M.H. Smith,
Confronting identities: the rhetoric and reality of a Carolingian frontier, in: W. Pohl
and M. Diesenberger eds., Integration und Herrschaft. Ethnische Identitten und soziale
Organisation im Frhmittelalter (Wien, forthcoming).
10
On the distinctiveness of Breton literacy and documentary traditions see N.-Y.
Tonnerre, Celtic literary tradition and the development of a feudal principality in
Brittany, in: H. Pryce ed., Literacy in medieval Celtic societies (Cambridge, 1998), pp.
16682; H. Guillotel, Cartulaires bretons mdivaux, in: O. Guyotjeannin, L. Morelle
and M. Parisse eds., Les Cartulaires. Actes de la table ronde organise par lEcole National
des Chartes Mmoires et documents de lEcole des Chartes 39 (Paris, 1993), pp.
 - 365

We should also pay attention to the date of Redons foundation.

The great tide of late seventh- and eighth-century monastic foun-
dations had long since waned by the time Redon was established in
832. By then, almost all Frankish monasteries were established insti-
tutions which had largely completed assembling their landed endow-
ments and building their networks of patronage; only minor differences
in chronology differentiate northern Gaul, the Rhineland or Bavaria
in this respect.11 In Neustria, many of the communities which housed
the most important relic shrines of the region had an even longer
history, having originated as extramural episcopal basilicas which
only later become monasteries or communities of regular canons.12
Although small aristocratic family monasteries were founded here
and there in the Frankish lands during the ninth century, only in
newly-converted Saxony did royal or episcopal initiative foster new
communities.13 In Breton terms, Redon was also a latecomer on the
scene. Although a precise chronology is impossible, several major
monasteries certainly existed in pre-Carolingian Brittany and at least
two had already attracted the patronage of Carolingian rulers.14 No
other monasteries are known to have been established in the Carol-
ingian period. Also, the landscape around Redon was dotted with
tiny religious communities and hermitages, monasteriola, abbatiolae and
minihis.15 Redons foundation thus fits no general rhythm of institu-
tionalised monastic foundation.

32541; J.M.H. Smith, Oral and written: saints, miracles and relics in Brittany,
c. 8501250, Speculum 65 (1990), pp. 30943. For a parallel subversion of the cen-
tre/periphery distinction on the eastern margin of the Carolingian world in the
context of a major churchs textual self-presentation, see S. Airlie, True teachers
and pious kings: Salzburg, Louis the German and Christian order, in: R. Gameson
and H. Leyser eds., Belief and culture in medieval Europe (Oxford, 2001).
11
Cf. M. Innes, State and society in the early Middle Ages: the middle Rhine valley,
400 1000 (Cambridge, 2000), p. 17. In general, see M. de Jong, Carolingian
monasticism: the power of prayer, in: R. McKitterick ed., New Cambridge Medieval
History II: c. 700c. 900 (Cambridge, 1995), pp. 62253.
12
See also G. Oexle, Forschungen zu monastischen und geistlichen Gemeinschaften im west-
frnkischen Bereich, Mnstersche Mittelalter-Schriften 31 (Mnchen, 1978).
13
The chronology and context of Saxon foundations are sketched by C. Carroll,
The bishoprics of Saxony in the first century after Christianization, EME 8 (1999),
pp. 21945 at pp. 2257.
14
Smith, Province and empire, pp. 712.
15
Some 17 such places are known from the Redon charters (CR 11, 13, 57, 97,
106, 141, 152, 154, 181, 225, 233, 247, 272, 274, 276, A4, A26, A31, A40, A45,
A53, A54, and H. Morice, Mmoires pour servir de preuves lhistoire ecclsiastique et civile
de Bretagne, 3 vols. (Paris, 17426), vol. 1, col 265). There are also 24 abbots
366 ..

Every attempt to found a monastery was inevitably a speculative

venture; success or failure contingent and unpredictable. In Redons
case, the immediate political context of its foundation and the details
of Conuuoions career underscore this. Notably, the resentments
within the Frankish aristocratic elite which boiled over into open
rebellion against Louis the Pious in 830 and 8334 had their epi-
centre in western Neustria and the Breton border region and
these revolts followed hard on the heels of repeated Breton efforts
to throw off Carolingian hegemony. Had Breton warlords and Frankish
rebels made common cause in the early 830s, the Vilaine valley
around Redon would in all likelihood have been a battleground or,
at the very least, a region of contestation and anarchy.16 832 was
thus hardly an auspicious moment to found a monastery right on
the Franco-Breton border. Indeed, efforts to win imperial approval
met with immediate rejection.17 But when Nominoe, the Breton
regional leader, gave his support to the new monastery and asked
the fledgling community to pray for the Carolingian emperor in his
troubles, the political context was transformed. From 834 onwards,
Redon developed rapidly under the joint patronage of Breton principes
and Carolingian rulers.18

unconnected to Redon, only one of whom is certainly associated with one of these
monasteriola (CR 21, 28, 47, 52, 64, 109, 116, 129, 138, 139, 142, 143, 150, 153,
154, 159, 160, 171, 198, 233, 241, 247, 255, 265, 266, 268, 276, 281, A41). A
minihi (from monachia) was an ecclesiastical site with rights of sanctuary by the eleventh
century. A. Chdeville and N.-Y. Tonnerre, La Bretagne fodale, XI eXIII e sicle (Rennes,
1987), pp. 3547; cf. W. Davies, Adding insult to injury: power, property and
immunities in early medieval Wales, in: W. Davies and P. Fouracre eds., Property
and power in the early Middle Ages (Cambridge, 1995), pp. 13764.
16
A glance at the politics of the reign of Charles the Bald affirms the plausibil-
ity of such a scenario, for in his reign Franks and Bretons did indeed meet in bat-
tle twice, very close to Redon, and Breton leaders and Frankish rebels did ally in
revolt. Smith, Province and empire, pp. 6085 on Breton revolts in the early years of
Louis the Piouss reign and on the intertwining of Breton and Frankish politics in
the 830s; ibid. pp. 85115 for the Breton role in the politics of Charles the Balds
reign.
17
GSR, I.89, pp. 1337.
18
For details of Carolingian patronage, see J.M.H. Smith, Culte imprial et
politique frontalire dans la valle de la Vilaine: le tmoignage des diplmes car-
olingiens dans le cartulaire de Redon, in: M. Simon ed., Landvennec et le monachisme
breton dans le haut moyen ge (Landvennec, 1985), pp. 12939. The exact nature of
Carolingian patronage remains problematic; see the discussion of W. Davies, Small
worlds: the village community in early medieval Brittany (London, 1988), p. 193 and n. 23.
For princely patronage, see Smith, Province and empire, pp. 12930.
 - 367

Redons first abbot was no modest ascetic. Of noble origin,

Conuuoion seems to have been an abbot with enough political acu-
men to remain on good terms with both Frankish and Breton lead-
ers even when they were in direct conflict with each other.19 The
men who entered the monastic life with him included men from
noble families, friends and associates of the Carolingian magnates
whom Louis the Pious had appointed to office in eastern Brittany.
Very probably bilingual, some of them definitely literate priests, these
men were ideally placed to act as political mediators.20 They were
adult converts, some of them certainly aristocrats who brought with
them into the monastery a mastery of secular politics. Not mission-
aries, not oblates, quite detached from their families, their radical
break with their secular past was similar to that of Benedict of Aniane.
As for Conuuoion in particular, he had been educated among the
cathedral clergy of Vannes and, according to his eleventh-century
vita, had been promoted to deacon by Bishop Raginarius.21 Probably
an appointee of Louis the Pious, Raginarius had spoken out against
the foundation of a monastery at Redon in Louis presence. By per-
sisting in establishing his community on the site, Conuuoion thus
directly opposed the bishop whom he had formerly served.22 Contro-
versy thus marked Conuuoions career and Redons establishment
alike. Conwoions skills in negotiation and mediation must have con-
tributed substantially to Redons ability to survive on its strategic
frontier location.
Indeed, it did more than survive: it thrived. From 834 onwards,
local landowners made numerous property donations. The region
west of the Vilaine was predominantly a land of peasant proprietors

19
Cf. GSR I.11, ed. Brett, pp. 1413 on Conuuoions activities during the Franco-
Breton conflict in 836/7; that Charles the Bald issued a diploma for Redon at the
height of his confllict with Nominoe also suggests Conuuoion performed a delicate
political balancing act. G. Tessier, Recueil des actes de Charles II le Chauve, 3 vols.
(Paris, 194355), vol. 1, pp. 34851, no. 132 (3 August 850).
20
GSR I.1, ed. Brett, pp. 10713 and my comments in Confronting identities.
The writing abilities of Conuuoion and his companion Condeloc is attested by their
role as charter scribes. CR 5, 128, 212, 177, 179, A4.
21
Vita Conuuoionis, 2, ed. Brett p. 229. It would make more sense of Conuuoions
career if he had been archdeacon of the diocese; since GSR II.1 (ed. Brett, pp.
14751) depicts him celebrating mass, he must have been ordained a priest at some
point.
22
GSR I.8, p. 133; Smith, Province and empire, p. 76; H. Guillotel, Le manuscrit,
in: Cartulaire de labbaye Saint-Sauveur de Redon [facsimile edition] (Rennes, 1998),
p. 21.
368 ..

and of machtierni, small-time landlords who nevertheless functioned

as a local ruling elite; Redons first property acquisitions lay here.23
In the absence of anything comparable to the generous royal endow-
ments of fiscal land which had established the economic security of
seventh-century monasteries in northern Gaul, it took many indi-
vidual acts of generosity to build even a modest landed endowment
field by field. East of the Vilaine, estates were considerably larger
but nevertheless still small in comparison with the massive scale of
Frankish aristocratic landed wealth in Austrasia and the Rhineland:
here too Redon gradually acquired substantial holdings.24 And the
abbey was not merely the passive, if grateful, beneficiary of these
donations: on occasion Conuuoion directly approached landowners
and persuaded them to relinquish key properties,25 or bought up
lands directly adjacent to estates already in monastic ownership.26 In
addition, the monastery gained jurisdictional rights over several com-
munities in the vicinity; these formed the nucleus of the monasterys
seigneurial powers.27 It also absorbed many of the tiny monasteries
in the region; previously independent local priests also joined the
monastery in increasing numbers.28 The monasterys foundation thus
transformed local tenurial, political and ecclesiastical traditions.
As its landed endowment grew, so did the buildings. Conuuoion
oversaw the building of two churches, the main abbatial basilica ded-
icated to Christ the Saviour and another dedicated to the Virgin;
we hear also of the monastic dormitory, gatehouse, domus peregrin-
orum, hospitale pauperum, a hospitium for the external canons attached to
the monastery as well as the monks cemetery and garden.29 In 863
he also turned the princely residence at Pllan into a dependent cell

23
On this distinctive society see Davies, Small worlds; Astill and Davies, Breton
landscape, pp. 91115.
24
Smith, Province and empire, pp. 3543.
25
CR 23.
26
This is sometimes explicit: a deed of sale to Redon specifies that the land lay
circumcincta a terra mea et a terra supradicti emptoris per botinas fixas per loca designata (CR
125) and sometimes can be deduced (e.g. boundary clauses indicate that the land
purchased in CR 148 lay adjacent to that gifted five years previously by CR 12).
Other purchases in areas where Redon already had substantial landholdings are
CR 38, 39, 40, 80, 81, 82, suggesting a calculated strategy of estate accumulation.
Further purchases are recorded in CR 209, 244, A19.
27
Redons seigneurie is discussed by Davies, Small worlds pp. 188200.
28
Davies, Small worlds, pp. 1902.
29
CR 28, 234, 325, A35; GSR II.1, 2, 3, 4, 6, 7, pp. 149, 153, 155, 157, 167.
 - 369

and built a stone church there.30 The size of the community also
expanded: Conuuoions original six followers doubled to twelve within
six months and then increased to eighteen by 854, and to twenty-
five by 870.31 Still tiny in comparison with major Carolingian monas-
teries, Redon had nevertheless become of major regional significance.32
By the time the aging Conuuoion resigned as abbot in January or
February 867, Redon had impressive buildings and a secure exist-
ence, funded by its regional patrimony and sustained by Breton
princely and Carolingian royal approval.33 Under the rule of an abbot
who was as much local lord as spiritual father, Redon had com-
pletely altered the traditional topographies of power in southeastern
Brittany.
Unsurprisingly, contestation and conflict accompanied this trans-
formation. Repeated lawsuits and disputes over property rights and
ownership brought Conuuoion or his successor, Ritcant, before infor-
mal mediators, local tribunals or the Breton princely court to defend
Redons title: in this respect too, building a monastic endowment
could be fraught and uncertain work.34 Unlike the endowments of

30
CR 78, 82, 241. The church (S. Maxent) had architectural features in com-
mon with many other Carolingian churches. R. Couffon, LArchitecture religieuse
en Bretagne, VeXe sicles, Mmoires de la Socit dHistoire et dArchologie de Bretagne,
23 (1943), pp. 140 at pp. 315.
31
CR 1, A2, A37, 224.
32
Redon was comparable in size to St Peters, Gent which had 24 regular canons
during Einhards lay abbacy (81540) and Niederaltaich which was founded c. 741
for 20 monks but had apparently dropped to 16 under Abbot Gozbald (82555).
Major Carolingian monasteries usually housed upwards of one hundred monks, the
largest could support three or four hundred. Fulda reached over 600 under Hrabanus
Maurus. Summary figures derived from U. Berlire, Le Nombre des moines dans
les anciens monastres, Revue Bndictine 41 (1929), pp. 23161 and 42 (1930), pp.
1942. For St Peters Gent see the Ratio fundationis seu aedificationis Blandiniensis coeno-
bii, ed. M. Gysseling and A.C.F. Koch, Diplomata Belgica ante annum millesimum cen-
tesimum scripta, 2 vols. (Brussels, 1950), vol. 1 no 49 at p. 125; for Niederaltaich see
the breviarius of Urolf and a charter of Gozbald, Monumenta Boica XI (Mnchen,
1771), pp. 146, 10913; for Fulda, K. Schmid, Mnchslisten und Klosterconvent
von Fulda zur Zeit der Karolinger, in: K. Schmid ed., Die Klostergemeinschaft von
Fulda II/2, Mnstersche Mittelalter-Schriften 8 (Mnchen, 1978), pp. 6112.
33
For the chronology of Conuuoions resignation and death a year later, see
A. de La Borderie, La chronologie du cartulaire de Redon, Annales de Bretagne 5
(1890), pp. 535630 at pp. 6117.
34
On disputes and disputing procedures in the region see W. Davies, Disputes,
their conduct and their settlement in the village communities of eastern Brittany in
the ninth century, History and anthropology 1 (1985), pp. 289312; eadem, People
and places in dispute in ninth-century Brittany, in: W. Davies and P. Fouracre
eds., The settlement of disputes in early medieval Europe (Cambridge, 1986), pp. 6484. A
370 ..

the major monasteries of Austrasia and the Rhineland which had

been developed in the eighth century in the context of strong regional
networks of aristocratic power and patronage, Redon was located in
an area where social power remained incoherent, opaque and open
to challenge.35 No regionally dominant family claimed it as its Eigen-
kloster; rather, Redons vigorous land grab cut right across local famil-
ial traditions with their presumptions of relatives interest in the lands
of their close kin. Although Redon certainly cultivated links with
donors families across the generations, this was not always enough
to ensure support.36 There is no evidence that the immunity and
protection granted by Charles the Bald in 850 meant anything in
practical or political terms; even the support of the Breton principes
who succeeded Nominoe was not always forthcoming.37
Conuuoions death on 5 January 868 proved to be a turning point
in Redons history. Thereafter, the flow of donations dwindled to a
trickle. And in 874, another death accelerated the pace of change.
This was that of Salomon, the most vigorous and successful of ninth-
century Breton principes (857874), who had established a quasi-royal
rule throughout the peninsula and western Neustria. His assassina-
tion plunged Brittany into a political crisis from which his succes-
sors never fully recovered, quite apart from the Viking attacks which
strained both resources and leadership. After 874, virtually no-one
except the principes bothered to support Redon any more. By the
early tenth century, Redon still received lands when local landowners
entered the monastery to die in monastic dignity, but the commu-
nity seems not to have prospered.38 With the Vilaine valley too vul-
nerable to Viking attack and even the inland refuge at Pllan which
Salomon had given them not safe, the remaining monks packed their

list of all forty-nine records of disputes is provided at n. 10, p. 68 of the latter arti-
cle; Redons property rights and claims were at issue in two-thirds of these.
35
Cf. Innes, State and society, pp. 1393 on the patronage networks which under-
pinned the property acquisitions of Lorsch and Fulda.
36
As is demonstrated by the relationship between Redon and the family of Ratuili,
the monks first benefactor. Ratuili had four sons: he gave one, Liberius, to Redon
as an oblate a year after his founding grant; another, Catuuoret also gave land and
a son to Redon; the other two, Ratfred and Ratuili (II) were in dispute with Redon
about land they claimed as their inheritance, and threatened to burn the monastery
down. CR 3, 4, 105, 215, GSR I.3, pp. 1159. On other activities of Ratuili and
his family, see Davies, Small worlds, pp. 17981.
37
CR 235 for Alan Is restoration in 878 of a grant cupiditate aliorum principum
ablata est a Sancto Salvatori.
38
CR 270, 278, 280.
 - 371

bags in 919 and decamped, reaching Auxerre before 921 but finally
settling in Poitou in 924.39
Almost a century had passed since Conuuoion and his compan-
ions first adopted the ascetic life on their plot of land in the lee of
the hill that overlooked the confluence of the Oust and the Vilaine.
Decades of growth, propsperity and influence had proven unsus-
tainable. Uncertainty, even a mood of retrenchment and decline must
have been inevitable. In this changed atmosphere, the monks began
their project of textual aedificatio, to which we now turn.

In the early Middle Ages, a powerful connection existed between

landholding and literacy. Late Roman procedures for guaranteeing
title to property had depended heavily on written documentation.
Early medieval kings and churches were keen, if for rather different
reasons, to adapt late Roman practices to their changing institutional
and legal environment and to exploit the administrative and ideo-
logical potential of the carta and notitia. Documentary traditions also
continued in use in local, lay society. These can be traced through
the indirect testimony of laws or formularies, and occasionally directly
through surviving archives.40 Yet the administrative framework within
which such documents were used was shifting, and although the late
Roman procedure of registering title in municipal archives, the gesta
municipalia, lingered on here and there into the early seventh century,
the onus for redacting and preserving texts gradually passed every-
where to individual or institutional proprietors.41 In the transformed
environment of the early Middle Ages, careful documentation of
property claims and entitlements nevertheless remained crucial.

39
CR 283 for the narrative. The date of departure from Redon is supplied by
a fragmentary set of annals apparently composed by a member of the Redon
community in the first half of the tenth century. Annales Rotonenses (um 919), ed.
B. Bischoff, in: Analecta Novissima. Texte des vierten bis sechzehnten Jahrhunderts (Stuttgart,
1984), pp. 1025. The monks return at the end of the tenth century and Redons
continuous existence thereafter down until its dissolution in 1790 is only rele-
vant here inasmuch as it ensured the survival of the ninth-century textual corpus.
40
P. Classen, Fortleben und Wandel sptrmischen Urkundenwesens im frhen
Mittelalter, in: idem ed., Recht und Schrift im Mittelalter, Vortrge und Forschungen
23 (Sigmaringen, 1977), pp. 1354.
41
P. Johanek, Zur rechtlichen Funktion von Traditionsnotiz, Traditionsbuch und
frher Siegelurkunde, in: Classen ed., Recht und Schrift, pp. 13162, esp. pp. 13645.
372 ..

Italy is well-known as a region of particularly strong continuity of

documentary forms and procedures.42 Another was western Neustria.
Here, as in Italy, registration in the gesta municipalia persisted into
the early seventh century, and the evidence of regional formularies
confirms that a sub-Roman documentary tradition remained strong
throughout the Merovingian period.43 Charter evidence also indicates
that local scribes remained in the habit of drawing up formal records
for lay landowners in the dioceses of Rennes, Nantes, Angers and
Vannes. Around 800, in both urban centres and rural villages in this
region, local scribes some certainly clerical, others conceivably
lay drew up documents recording land transfers and reporting
the settlement of disputes between private individuals in a tradition
evincing clear debts to the Merovingian usages of the region.44 These
scribes were the recognised redactors of formal documents and
employed authoritative forms of words for the purpose, although
whether they ever functioned in association with comital lawcourts
or episcopal households cannot be ascertained.45 Whatever the pre-
cise local arrangements for having a document drawn up may have
been, these local scribes were heirs to a tradition stretching back to
the public procedures of late Antiquity. Ninth-century landowners in
this region clearly placed great value on written documentation of
their property rights and also took considerable pains to preserve
their own charters.
When an unknown scribe used the traditonal local formulae of
the Vannetais to draw up the document recording Ratuilis grant of
the site of Redon,46 he unwittingly recorded not only the beginning
of the tenurial revolution which Redons foundation brought about
but also the beginning of a documentary and archival evolution. For

42
N. Everett, Scribes and charters in Lombard Italy, Studi Medievali 3e ser. 41
(2000), pp. 3983.
43
I.N. Wood, Disputes in late fifth- and sixth-century Gaul: some problems,
in: Davies and Fouracre eds., The settlement of disputes, pp. 722 at pp. 124; idem,
Administration, law and culture in Merovingian Gaul, in: R. McKitterick ed.,
The uses of literacy in early medieval Europe (Cambridge, 1990), pp. 6381.
44
M. Planiol, Histoire des institutions de la Bretagne, 5 vols., (Mayenne, 19814), vol.
II, pp. 15360; Davies, Small worlds, pp. 1348; eadem, People and places in dis-
pute, pp. 6870 with map of distribution of these local scribes on p. 69.
45
For local scribes elsewhere in the Carolingian empire and the discussion over
whether they had any official status, see Johanek, Zur rechtlichen Funktion,
pp. 1403; R. McKitterick, The Carolingians and the written word (Cambridge, 1989),
pp. 11534; Innes, State and society, pp. 1138 and Everett, Scribes and charters in
Lombard Italy, pp. 4255.
46
CR 1.
 - 373

Redon rapidly developed its own documentary forms to supplement,

even replace, traditional local ones. Its monks also functioned as
scribes, usually but not always anonymously for transactions to which
the abbey was party.47 As Redons landholdings grew, scribal activ-
ity became increasingly centred on the monastery.
Naturally, the monks kept these vital legal and administrative
records, but they were also keen to acquire any existing documen-
tation which pertained to a piece of land prior to their acquisition
of it.48 Thus, from the founding moment onwards, Redon built up
its archive, a form of institutional memory which interrupted indi-
vidual or familial record-collecting. By Conuuoions death it con-
tained several hundred documents, recording transactions which
predated Redons acquisition of particular lands by a generation or
more, grants to Redon by local landowners, the outcome of disputes
and lawsuits which the monks had fought and the benefactions of
Breton princes and Carolingian sovereigns.49
Only east of the Rhine did Carolingian monasteries decide to pre-
serve their property records by copying out the many individual
pieces of parchment into books, in the process ordering them by
geographical or chronological schemes, or a mixture of both. The
idea was slow to spread west of the Rhine: except for a couple of
experiments in cartulary construction in tenth-century Flanders, it
was not until the eleventh century that churches in West Francia/
France began to follow suit. That is not to say, however, that ninth-
and tenth-century churches west of the Rhine made do with an
unsorted muddle of loose documents. Rather, there is good evidence
both for the careful endorsing of the loose originals and for the com-
pilation of dossiers of selected royal diplomata and papal or episco-
pal privileges in the context of specific legal or political battles.50

47
W. Davies, The composition of the Redon cartulary, Francia 17/i (1990), pp.
6990, esp. pp. 779.
48
For example, when Redon acquired an estate at Lusanger (dioc. Nantes) in
864, it took over a family archive of six charters stretching back to 816 and detail-
ing complex transactions within three generations of one family. CR 57, 225, 226,
227, 229, 230, 231. Cf. Smith, Province and empire, pp. 389.
49
Around 350 documents survive from the period before 924, of which 283 are
in Redons main cartulary. However, some time between 1773 and 1856 forty-two
folios vanished from the section of the cartulary which contains charters from the
first twenty-five years of the abbeys history. These will have contained approxi-
mately one hundred charters, of which only about half survive in seventeenth-
century transcripts and eighteenth-century histories. For details of Redons cartu-
laries, see below, nn. 52, 54.
50
Declercq, Originals and cartularies, in: Heidecker ed., Charters and the use of
374 ..

At Redon, Conuuoions successor Ritcant (86771) took the step

of sorting the extensive monastic charter collection into an ordered
archive. He arranged the documents in several groups, the largest
of which contained documents produced and retained at Redon dur-
ing Conuuoions abbacy, filed in approximately chronological order.
Then he gathered together charters produced elsewhere, in the local-
ities. These were less systematically arranged, although he did keep
together several groups of documents all relating to the same place.
His third bundle comprised the documents produced during his own
abbacy, more or less in the order in which they were produced.51
Another group probably contained princely charters, royal diplomata,
episcopal and papal letters.52 Ritcants arrangement, in whatever may
have been the ninth-century equivalent of filing boxes, was carefully
maintained by his successors and they added their own documents
to the stack containing the records from their own period in office.53
This large collection of loose documents survived the tenth-century
hiatus in Redons history to be copied in the order Ritcant had insti-
tuted into the cartulary compiled during the abbacy of Almod
(10621084). Unfortunately, we cannot know how selective Almod
may have been in copying out the ninth-century archive, although
on philological grounds it is clear that he (or his copyist) did not
alter the ninth-century orthography of the documents in front of
him.54 The rush to compile cartularies in the eleventh century reflected

the written word; P. Geary, Entre gestion et gesta, in: Guyotjeannin et al. eds., Les
cartulaires, pp. 1324.
51
Thus far I follow the analysis of Davies, Composition, pp. 802.
52
For examples of Carolingian and post-Carolingian separate classification of
diplomata and important privileges and their copying into separate dossiers, see
Declercq, Originals and cartularies, pp. 1513. I suggest that Redon was fol-
lowing this common ninth- and tenth-century practice in this respect for two rea-
sons: (a) all the surviving Carolingian diplomata and several of the ninth-century
princely charters have a different textual tradition from the rest of the charters and
(b) a second cartulary, the mid-twelfth century petit cartulaire, included a collec-
tion of royal diplomata, princely charters and ecclesiastical privileges and corre-
spondence. Only six leaves survive which, inter alia, contain important princely
charters; an eighteenth-century description made before the mutilations notes
that it also included letters between the princeps Salomon and Pope Hadrian II.
H. Guillotel, Les cartulaires de labbaye de Redon, Mmoires de la Socit dHistoire
et dArchologie de Bretagne 63 (1986), pp. 2748 at pp. 42, 45; Brett, introduction to
GSR, p. 27. I have not seen the extant folios.
53
Davies, Composition.
54
See now, in addition to the edition by de Courson (as n. 6 above) the fac-
simile, Cartulaire de labbaye de Saint-Sauveur de Redon, Amis des archives historiques
du diocse de Rennes, Dol et Saint-Malo (Rennes, 1998). For discussions of its
 - 375

a widespread concern for reforming the past in light of present

needs,55 and Almod responded to the political challenges of eccle-
siastical property owning in the same way as many of his contem-
poraries. It is from this eleventh-century cartulary (which received
additions into the middle of the twelfth century) that our knowledge
of Redons properties and local Breton society in the ninth century
is derived.
Like Almod, Ritcant adopted the prevalent attitude of the day to
the challenge of archival preservation. Each of them thereby offered
an image of the monasterys past, as well as of its property claims
and rights. In Ritcants case, that past stetched back just under forty
years a span encompassed within the living memory of at least one
member of the community.56 This forty years had witnessed Redons
establishment of a secure existence despite the climate of Franco-
Breton tensions and the repeated challenges and claims mounted by
aggrieved local landowners. But Ritcant made his compilation just
as the monks must have been struggling to come to terms with the
falling level of local benefaction after Conuuoions death. Ordering a
charter collection was not, then, simply an adjunct to efficient prop-
erty management (although it is clear that Ritcant was an energetic
administrator and landlord): it was also a textual representation of
the monasterys past and his predecessors great achievements. As
such, its purpose must have been instructive edifying, even to
remind older monks and inform new recruits of Redons commem-
orative obligations to local families, Breton princes and Carolingian
kings, and to proclaim its importance and wealth. Ritcant not only
consolidated Redons property claims, he consolidated its institutional
memory.
He may also have tinkered with that memory. Some of Redons
ninth-century charters show clear evidence of subsequent emend-
ation, which, as Wendy Davies has argued, very probably occurred
in the ninth century.57 None of them offer massive revisions of the

compilation and history see Guillotel, Les cartulaires, and his prefatory essay to
the new facsimile edition.
55
P.J. Geary, Phantoms of remembrance: Memory and oblivion at the end of the first mil-
lennium (Princeton, 1994), p. 87.
56
Leuhemel was one of the original six monks in 832, became prior and was
still alive to witness a charter in 870 (CR 224).
57
W. Davies, Forgery in the Cartulaire de Redon, in: Flschungen im Mittelalter,
MGH Schriften, 6 vols. (Mnchen, 19889), vol. IV, pp. 26574.
376 ..

past, merely enough to protect Redons title from the counter-claims

of a donors heir. Whether or not the responsibility is attributable to
Ritcant, the monks of Redon had become aware of the malleability
of the past, and of the potential for rewriting it to suit contemp-
orary needs. Their archive offered an ordered, somewhat manipulated
view of the past, designed for the needs of a community which may
already have been aware that its glory days were fast ebbing away.

As Redon faced mounting difficulties in securing support and patron-

age in the generation following Conuuoions death, another monk
took a different approach to the project of textual aedificatio.58 This
man had entered the monastery as a small child, presumably as an
oblate,59 had known the founding fathers personally, and, indeed,
had been Conuuoions servant.60 Writing for his fellow monks whilst
some but not all of the original six were still alive, he produced the
work known as the Gesta sanctorum Rotonensium. Mabillons editorial
title is misleading: it is certainly not in any conventional sense the
gesta of Redons monks.61 Rather, its subject is the place itself, and
anything which demonstrated its holiness, including both its living
members and the long-dead saints whose relics they acquired. He
set himself to represent Redons history in sacred terms, to offer a
historical account of the aedificatio sancti loci. In so doing, he con-
structed a vision of Redons past to reassure monks and patrons alike
of the monasterys continuing importance.

58
For arguments as to the date of the GSR, see F. Lot, Mlanges dhistoire bretonne
(Paris, 1907), pp. 517 cf. Bretts introduction to her edition, pp. 810. My own
view is that it is more likely to have been written sooner rather than later after
Conuuoions death.
59
GSR II.5, p. 163: denique cum essem iuuenculus in monasterio positus . . ..
60
GSR II.1, p. 149: Testor uobis, fratres carissimi, quia ego eram illo tempore illius sancti
[Conuuoion] minister.
61
Cf. M. Sot, Gesta abbatum, gesta episcoporum, Typologie des sources du moyen
ge occidental 37 (Turnhout, 1981); P. Rich, En relisant LHistoire des Saints
de Redon, in: Simon ed., Landvennec et le monachisme breton, pp. 1318. As F. Lifshitz
has pointed out, efforts such as these to classify such works as either history or
hagiography have had a deleterious effect on the understanding of their nature and
function. F. Lifshitz, Beyond positivism and genre: hagiographical texts as his-
torical narrative, Viator 25 (1994), pp. 95113. Note that the (?) eleventh-century
 - 377

The manuscript tradition of the Gesta Sanctorum Rotonensium is entirely

separate from that of the charters.62 Unfortunately, it is also defec-
tive and does not allow us to establish exactly what the author wished
to say about his own endeavour. The extant text lacks the opening
preface, the ending of the first chapter and one or more chapters
from the end of the third and last book. The loss of the initial pre-
face is particularly unfortunate, for this was the conventional place
for an author to declare his purpose, and self-consciously to situate
himself within the literary traditions he was manipulating. Nevertheless,
the prefaces to books two and three allow us to go some way towards
establishing what the author was seeking to do. The preface to book
two provides a retrospective indication of the subject matter of the
preceding book: de situ et constructione sancti loci.63 This he subdivides
into three themes: the struggle of the holy monks against their wicked
enemies, the building of the holy place and a brief inquiry into
monastic property.64 He also announces the purpose of this second
part, to elaborate the life and manner of living of the holy monks
who served Christ the Lord faithfully in that most sacred place right
until the end of their life, which he also refers to as their gesta et
facta.65 The third book, on the other hand, concentrates on mirabilia,
virtutes and prodigia performed at the sanctus locus, especially those
worked by relics obtained by Conuuoion. The author announces that
at its end he will deal with Conuuoions death, but this is the miss-
ing section.66
In both these prefaces, the unifying theme of the sanctus locus is
prominent. But subject matter also relates to avowed purpose. In
contrast to the antique custom of emperors committing accounts of
their victories to writing and to annals, it is far more important, he
states, to set down in writing the struggles of the holy men who

author of the Vita Conwoionis regarded the GSR as a liber miraculorum. VC 5, ed. Brett,
p. 233.
62
Outlined by Brett in her introduction, pp. 20934 with the important addi-
tional remarks of J.-C. Poulin, Le dossier hagiographique de Saint Conwoion de
Redon. A propos dune dition rcente, Francia 18/i (1991), pp. 13959.
63
GSR II. pref., p. 145.
64
GSR II. pref., p. 147: certamen malorum hominum cum sanctis uiris descripsimus, et
aedificationem sancti loci, et inquisitiones rerum ad eum pertinentium succincte deprompsimus.
65
GSR II. pref., p. 145: uitam et conuersationem sanctorum monachorum qui in eodem
sacratissimo loco usque ad finem uitae suae fideliter Christo Domino mancipauerunt Gesta et
facta: ibid., p. 147.
66
GSR III. pref., p. 185, 187.
378 ..

battled day and night with the invisible enemy.67 In this way, his
task is to edify. When these things are read, the memory of the
saints is evoked, edification of the mind is provided for the faithful,
honour is shown to the monks.68 In distancing himself from classical
historiographical traditions, he presents himself as the mouthpiece of
Christian, salvific history. His selection of miracle stories told to him
by the elders of the community is narrated with Gods help in order
to confirm the faith of believers, because it was most harmful that
all the stories should be consigned to oblivion. . . . Who knows how
many miracles and how many marvels were shown forth through
them, if not the author of all mankind, Jesus Christ our Lord, who
everywhere rules and guards his saints and makes their merits known
among their peoples?69 His purpose is thus to anchor Redon within
the overall history of Christian redemption. In this way, an account
of the building of the monastery and the deeds of its inhabitants
becomes a work of edification for the faithful.
Three aspects of the authors working methods elucidate his pro-
ject of textual aedificatio. First, his division of the subject matter into
three distinct books, each with its own theme, also has a topo-
graphical dimension. The struggles against wicked men of the first
book occur in widely separated places, ranging from Redon to Louis
the Piouss imperial court at Charmont-en-Beauce, Tours and Thion-
ville. This is the only book which has a roughly chronological order-
ing. The second narrows the focus to the monastery and its immediate
vicinity, narrating anecdotes about the monks vita et conversatio which
took place at indeterminate times in the monastic fields and precincts.
The third looks into the abbey church itself, recording the miracles
which happened at its relic shrines and recounting the preservation
of the church from Viking destruction. The three tableaux thus move
from the profane world of courtly politics through monastic space
and finally to the place of most intense sacrality. The holy place,
the sanctus locus, is thus situated in a landscape whose contours are

67
GSR II. pref., ed. Brett, p. 145: Quanto magis nunc debemus certamina sanctorum
uirorum litteris tradere, qui incessanter die ac nocte dimicauerunt cum inuisibili hoste?
68
Ibid.: Et cum ista leguntur, memoria sanctorum colligitur, aedificatio mentium credentibus
traditur, honor monachis exhibetur.
69
GSR III. pref., p. 185: ad corroborandum fidelium fidem_ quia ualde perniciosum erat,
ut haec omnia traderentur obliuioni . . . Quis enim scit quantae uirtutes et quanta prodigia per eos
ostensa sunt, nisi auctor hominum, Iesus Christus dominus noster, qui ubique regit et custodit
sanctos suos, et merita eorum per populos suos demonstrat?
 - 379

simultaneously geographical, historical and sacred. This, then, is the

architectural principle of this work of textual aedificatio.
Second, the segmental structure of all three books is important
within this overall framework. The author presents each episode as
a discrete textual unit. Almost every chapter is deliberately homiletic,
framed by opening and closing biblical quotations which serve to
emphasise the moral lessons to be drawn from the incident narrated.
Read on its own, any individual chapter fulfills the authors edify-
ing purpose. Together with his simple, usually unadorned Latin style,
this format suggests that the text may have been intended for monas-
tic lections. Some chapters preach a message of more general applic-
ability or are particularly suited for a lay audience, emphasising that
right religious conduct means honouring Redon and its monks.70
Since laymen clearly did on occasion attend monastic services, the
message matched the audience.71 For both monastic and lay hearers
then, the Gesta sanctorum Rotonensium offered a view of Redons his-
tory intended to strengthen the minds of the faithful. Its modular
structure enabled individual passages to stand alone, yet combined
them all together into a structured whole.72
Completing this survey of the textual construction of the Gesta
sanctorum Rotonensium, we should note the authors choice of building
materials. As a contribution to the literature of monastic spiritual
edification, the Gesta belong to a literary tradition stretching back to
the works of Sulpicius Severus, Cassian and Gregory the Great. And
the decision to reach out to the laity as well may reflect the grow-
ing Carolingian concern to adapt this advice to provide moral instruc-
tion for those in secular life. It is no surprise, then, that the classics
of this literature supplied some of the individual building blocks:

70
For example, the narration of the illness of the founder Ratuili, his cure by
the monks and, as a thank offering, his oblation of his son Liberius and a further
grant of land in GSR I.3 is opened by Iac. V.145 on priests anointing the sick
and terminated by Lc. VI.21, Mt V.6, V.3, XXVIII.20 on the blessed.
71
For a lay penitent attending the night vigil, see GSR III.8, ed. Brett, pp. 20712.
72
At both Farfa and Mont-Saint-Michel, ninth-century texts on the foundation
of the monastery were also used for monastic lections, albeit in the eleventh cen-
tury. Constructio monasterii Farfensis, ed. U. Balzani, Il Chronicon Farfense di Gregorio di
Catino, Fonti per la Storia dItalia 3334, 2 vols., (Rome, 1903, reprint 196972),
vol. 1, pp. 323; Apparitio S. Michaelis archangeli in Monte Tumba in Gallia, Migne PL
96 cols. 138994 and comments of N. Simonnet, La fondation du Mont-Saint-
Michel daprs la Revelatio ecclesiae sancti Michaelis, Annales de Bretagne et des Pays de
lOuest 106 (1999), pp. 723 at pp. 78.
380 ..

Sulpicius Severus Vita Martini and Dialogues and also the Dialogues of
Gregory the Great underpin parts of the text. Whereas the former
provided a repertoire of words and phrases to be appropriated, the
latter also supplied a template for how to write about holiness in a
way which contributed substantially to the edifice as a whole. The
author was also familiar with at least one Merovingian vita, and one
section of the Liber Pontificalis.73 His textual sources also extended to
Redons own archive, for he demonstrably had certain charters in
front of him as he wrote. On occasion, he simply extracted details
of property granted;74 elsewhere he presented accounts of some of
Redons conflicts with local landowners that differed substantively
from his archival sources.75 Finally, he supplemented the archives with
stories told to him by other monks, establishing an easy congruence
between textual and oral material, as did other ninth-century writers
on the fairly recent past.76 Although his indebtedness to his oral
informants is explicit, his textual sources remain undisclosed, apart
from biblical citations.77 Just as the planks from Ronuuallons house
that were reused in building the abbey buildings would not have
proclaimed their source, so the anonymous authors textual borrow-
ings and allusions remain covert.
At its most straightforward level, the Gesta sanctorum Rotonensium
thus indicate the level of cultural achievement in this monastery
towards the end of the ninth century. Conuuoions building work
had not been limited to the monastic buildings, but also to its library
and the minds of the boys in the monastic classroom. From oblate
to house author, the Redon anonymous had himself participated in
that. And in his writing, he pursued the work of construction in new
directions, shaping the monasterys self-representation and speaking to
the textual community which he hoped his words would encourage.

73
Summary analysis of textual borrowings and stylistic influences is provided by
Brett, GSR, pp. 649. Her list is incomplete: for additions see Poulin, Dossier
hagiographique, pp. 1445 and C. Stancliffes review of Bretts edition in Journal
of Ecclesiastical History 42 (1991), pp. 4735. A further textual borrowing is identified
at n. 86 below; others almost certainly remain to be identified.
74
Poulin, Dossier hagiographique, pp. 1467.
75
Below, pp. 3901.
76
As is well demonstrated by M. Innes, Memory, orality and literacy in an
early medieval society, Past & Present 158 (1998), pp. 336.
77
For example GSR II. pref., 8, pp. 185, 209.
 - 381

Part of the charm of the Gesta sanctorum Rotonensium lies in the texts
apparent simplicity, with its tales of Conuuoions struggles against
rapacious local landowners, sick men being healed by the monks
miracles, and penitents with clanking chains and suppurting sores
reconciled at the shrines of Redons potent relics. Yet in practice
hagiographical works are rarely simple and straightforward. This is
evident in the case of many vitae, whose rhetorical strategies and
political agendas have been the subject of numerous recent studies.
Artful simplicity is also to be found in collections of miracle stories.
Characteristically, these proclaim the power of the saint or relics
through a series of disjoint vignettes, each telling another miracle
but yet making the same point.78 Though their opening may well
be an account of the death of the holy man or woman, or of the
translation of his/her relics, their closure is often arbitrary or incom-
plete, thus making either excerpting or continuation by a later author
easy. The anecdotes generally have little or no explicit temporal rela-
tionship to any other events and only a loose chronological artic-
ulation with each other. Their episodic structure gives early medieval
libri miraculorum a scrap-book quality which contributes to the definition
of the genre.
Close reading suggests that underneath this characteristic present-
ation, libri miraculorum nevertheless comprise carefully selected mate-
rial which may construct highly tendentious or partial images. That
they are also revelatory of their authors ideologies and presumptions
is little surprise, but they may also encode coherent polemical positions
and arguments. Structured as much by exclusion as inclusion of per-
sons, issues or themes, a liber miraculorum offers an image of the world
as its author wished it might be but not necessarily as it was.79
Viewed from this perspective, the Gesta sanctorum Rotonensium become
a sophisticated presentation of Redons past designed to meet the
anxieties of a monastic community left increasingly insecure as bene-
factors turned their backs and Vikings advanced. This textual aedificatio

78
The best survey of the genre remains M. Heinzelmann, Translationsberichte und
andere Quellen des Reliquienkultes, Typologie des sources du moyen ge occidental 33
(Turnhout, 1979).
79
An exemplary study from which I have learned much is K. Ashley and
P. Sheingorn, Writing Faith: Text, sign and history in the miracles of Sainte Foy (Chicago,
1999).
382 ..

is, in effect, a pice justificative for the monasterys continuing exist-

ence. The concluding section of this paper examines how the Redon
author deployed his literary resources to this end, and deconstructs
his carefully built text to reveal its inner polemics.
The authors choice of subject immediately becomes significant.
In most cases where a house author took up the pen within living
memory of a revered founder or first abbot/abbess, the result was
a vita. Eigils Life of Sturm, Liudgers Life of Gregory, Ardos Life
of Benedict of Aniane, Agiuss Life of Hathumoda or Rimberts Life
of Anskar are all ninth-century vitae written by a disciple or close
personal associate of the deceased. Although Conuuoions life from
the foundation of Redon until his death provided the framing chronol-
ogy and the subject matter for the opening and closing paragraphs,
and although the Gesta hail Conuuoion as the fundator et constructor
under whose guidance the holy place attained the peak of monastic
perfection,80 the work in no way conforms to the conventions of a
founders vita. Why might the author have decided to reject this
hagiographical tradition? It is tempting to speculate that Conuuoion
had been far too politically controversial. Redons establishment had
been marked by the conflict between founder and diocesan bishop;
Conuuoions retirement from the abbacy a year before his death
may suggest an equally fraught ending to his career. His aggressive
strategies for land acquisition and the long sequence of litigation in
which he was embroiled may well have left him more feared than
admired by the laity.
How, then, does the Gesta argue for Redon as a place distinguished
by its holiness? We may answer with respect to place and person.
The site itself, strategically located, had doubtless been chosen care-
fully. What interested the author of the Gesta, however, was its suit-
ability in terms of the western ascetic tradition. From Evagriuss
translation of Athanasiuss Life of Antony onwards, the ideal monas-
tic site was one which resembled a desert, a dangerous place to tem-
per and test the ascetic spirit. But, as the vita Antonii also made clear,
it should also be a place inhabited by God, a paradise-like place of
fresh waters, spreading trees and verdant meadows. Merovingian
hagiography developed the themes, balancing rhetorical traditions

80
GSR III. pref., p. 187: quia ipse sanctus pater fundator et constructor sancti Rotonensis
loci ab initio exstitit, et usque ad summum perfecte perduxit.
 - 383

and actual topography.81 This is exactly what the Redon author did.
The site chosen by Conuuoion was a locus desertus.82 Here the devil
lurked to tempt the weaker monks and to rouse evil laymen to attack
the community.83 But it was not at all like the rugged offshore islands
or coastal caves where dragons lurked which earlier generations of
Breton holy men had sought out. Rather, it was a place known to
God and revealed by him to the holy,84 a place where wonderful
charity, great austerity, the utmost humility shines forth, chastity
above all.85 As such, it deserved to be described in words redolent
of paradise. To do so, the Redon author simply lifted verbatim the
description in the vita Filiberti of the site of Jumiges on the fertile
banks of the Seine.86 His purpose was not simply to capture the lush-
ness of the place, but rather to emphasise the sacred qualities that
enabled the monks to pursue their vocation there. Here, in the divine
presence, this locus sanctus provided a glimpse of paradise.
The authors insistence that Redon is a locus sanctus is unremit-
ting.87 Although the original Christian loca sancta were the sites in

81
D. von der Nahmer, ber Ideallandschaften und Klostergrndungsorte, Studien
und Mitteilungen zur Geschichte des Benediktinerordens und seiner Zweige 84 (1973), pp.
195270; M.-E. Brunnert, Fulda als Kloster in eremo. Zentrale Quellen ber die
Grndung im Spiegel der hagiographischen Tradition, in: G. Schrimpf ed., Kloster
Fulda in der Welt der Karolinger und Ottonen (Frankfurt, 1996), pp. 5978.
82
GSR I.1, 2, pp. 109, 113.
83
GSR II.5, 6, III.6, pp. 163, 165, 203 (the devil and the monks), ibid., I.5, 6,
7, III.7, pp. 1213, 125, 12731, 205 (the devil and laymen).
84
GSR I.2, pp. 1135; p. 115: peruenit ad locum sibi a Deo reuelatum.
85
GSR I.3, p. 119: Caritas ibidem fulget mira, abstinentia magna, humilitas summa, casti-
tas ante omnia.
86
GSR I.3, p. 119, whose italicised portion is taken directly from vita Filiberti
ch. 7, ed. W. Levison, MGH SRM 5, pp. 5889: Uere digna etymologia nominis Roton
nuncupatur, quia diuerso uernat more gemmarum decore; hinc frondium coma siluestris, hinc mul-
tiplices arborum fruges, illinc placet uberrima tellus, istinc uirentia prata graminibus, hinc horto-
rum odoriferi flores, hinc uinearum abundant butriones; cuncta undique aquis irrigata; inclita
coespis pastui pecorum congrua fundens frugem lactiferam; nunc ascendens mare eructat, nunc ad
sinum rediens aquarum impetus manat; compendium nauium apta; nihil paene indigens ex eo
quicquid ministratur uehiculis pedestribus, plaustris equinis etiam atque ratibus. Ibi adstant in
acie milites Christi, ubi suspirantes pro desiderio paradisi gemunt. Levison noted the exact
parallel, Poulin inexplicably denied it (Dossier, p. 145), Bretts edition fails to note
it. See the discussion of representations of the Seine in the vita Filiberti and other
sources in M. Diesenberger, Wahrnehmung und Aneignung der Natur in den Gesta
abbatum Fontanellensium, in: C. Egger and H. Weigl eds., Text Schrift Codex.
Quellenkundliche Arbeiten aus dem Institut fr sterreichische Geschichtsforschung, Mitteilungen
des Instituts fr sterreichische Geschichtsforschung Ergngzungsband 35 (Wien,
2000), pp. 933.
87
I count thirty-two instances of the expression or its very close equivalent (sacratis-
simus locus, sanctum monasterium etc.).
384 ..

Palestine hallowed by Christs presence and death, the spread of the

cult of martyrs in late Antiquity had been accompanied by an accep-
tance that sanctity extended from the person to embrace the place
where his/her tomb lay.88 Then, as confessors came to be hailed
alongside martyrs, it became a hagiographical assumption by the end
of the sixth century that any site associated with such a person must
be a locus sanctus. The Redon anonymous thus tapped into an estab-
lished hagiographical convention. His repetitious use of the phrase
perhaps also owes something to specifically Breton toponymic con-
ventions for monastic sites,89 and it certainly had strong legal overtones
too. From c. 700 onwards, royal grants of immunity to a church or
monastery made it clear that the place thus privileged was a locus
sanctus.90 Such a church stood in a special relationship to the king
and his officials, a place whose sacrality made it so powerful that
even royal power there was constrained. Immunities were one of the
chief instruments through which the interaction of secular and sacred
power was negotiated and, by the Carolingian period, conflated.91
Redon was no different from many other Carolingian monasteries
in this respect, for in 850, Charles the Bald had issued a diploma
granting immunity and protection, thereby recognising that Redon
was an established site of political and sacred significance in the
topography of the Carolingian Reichskirche.92

88
R.A. Markus, How on earth could places become holy? Origins of the Christian
idea of holy places, Journal of Early Christian Studies 2 (1994), pp. 25771; L. Pietri,
Loca sancta: la gographie de la saintet dans lhagiographie gauloise (IV eVI e s.),
in: S. Boesch Gajano and L. Scaraffia eds., Luoghi sacri e spazi della santit (Turin,
1990), pp. 2335.
89
In addition to the common usage of locus to refer to a monastery (A.-M.
Dimier, Le mot locus employ dans le sens de monastre, Revue Mabillon 62 (1972),
pp. 13354), there is a distinctive Breton place-name element loc- (from locus) mean-
ing monastery, holy place found from the eleventh century at the latest. Whether
it was used with that meaning in earlier centuries remains disputed. See R. Largillire,
Les saints et lorganisation chrtienne primitive dans lArmorique breton (Rennes, 1925), pp.
1727; P. Quentel, Toponymie bretonne. Chronologie des noms en loc-, Revue
Internationale dOnomastique 14 (1962), pp. 818; F. Gourvil, Les loc- dans lhagio-
toponymie bretonne, ibid. 15 (1963), pp. 4752; L. Fleuriot, Les origines de la Bretagne
(Paris, 1980), p. 80.
90
See, for example, the use of the phrase in Marculf, Formulae I.3, 4, 15, 16,
35; Formulae Imperiales e curia Ludovici Pii 4, 28, Collectio Sangallensis 3, ed. K. Zeumer,
Formulae Merovingici et Karolini Aevi, MGH LL sectio V, pp. 43, 44, 53, 534, 65,
290, 306, 3978.
91
See B.H. Rosenwein, Negotiating space: Power, restraint and privileges of immunity in
early medieval Europe (Ithaca, NY, 1999), esp. pp. 74134.
92
Tessier, Actes de Charles II, vol. I, pp. 34851, no. 132.
 - 385

For all its juridical implications, a locus sanctus remained a place of

nearness to God, where he and his saints were venerated. Conuuoions
new church was consecrated on the feast of the apostles Simon and
Jude (28 October) in 832/3, and dedicated to Christ the Saviour.93
Common in late Antiquity, dedications to sanctus Salvator continued
to be made in the early Middle Ages. Sometimes, they deliberately
invoked Rome, where the Lateran basilica was known as the basilica
Salvatoris.94 On other occasions, the dedication expressed the nature
of a church as a house of God, as it had done in early Christian
times. As such, it continued in use in the eighth and ninth centuries
for new churches at Fulda, Prm, Eichsttt, Paderborn, Aniane and
elsewhere.95
Although the dedication of a church commonly involved placing
relics in the altar, all that was technically required was a mass of
consecration. Relics remained optional, if desirable, into the tenth
century.96 Whether Carolingian churches dedicated to Christ the
Saviour contained modest relics from the start is unclear. In many
cases however, the acquisition of major relics enshrined at the main
altar tended to push the initial dedication into the background. Such
shrines brought enhanced sacrality, encouraged pilgrims and hence
both income and reputation.
By the ninth century, Roman relics had an especial prestige, and
Redon joined the competition for them.97 The Gesta sanctorum Rotonensium
describe at some length Conuuoions theft of the relics of the late
fourth-century bishop of Angers, Hypothemius, before narrating how

93
CR 137, 277 for the day; the year of consecration must precede the begin-
ning of the charter habit of making grants to Sanctus Salvator in Rotono of which the
earliest are dated 1 May 834 (CR 128, 130).
94
Cf. Bede, Historia ecclesiastica gentis Anglorum, I.33, ed. B. Colgrave and R.A.B.
Mynors (Oxford, 1969), p. 114. On this see N. Brooks, Canterbury, Rome and
English identity, in: J.M.H. Smith ed., Early medieval Rome and the Christian West:
Essays in honour of Donald A. Bullough (Leiden, 2000), pp. 22146 at pp. 2248.
95
Cf. E. Ewig, Kathedralpatrozinien im rmischen und im frnkischen Gallien,
in: idem, Sptantikes und frnkisches Gallien, Beihefte der Francia 3, 2 vols. (Mnchen,
1979), vol. II pp. 260317 at p. 263 who overemphasises the Anglo-Saxon influence
behind these continental dedications; also P. Le Matre, Image du Christ, image
de lempereur. Lexemple du culte du Saint Sauveur sous Louis le Pieux, Revue
dHistoire de lEglise de France 68 (1982), pp. 20112.
96
On the various western rites for consecrating a church in the early Middle
Ages, see G.G. Willis, Further essays in early Roman liturgy (London, 1968), pp. 13373.
97
J.M.H. Smith, Old saints, new cults: Roman relics in Carolingian Francia,
in: eadem ed., Early medieval Rome, pp. 31739.
386 ..

in 848/9, Pope Leo IV gave him the relics of his martyred third-
century predecessor, Marcellinus.98 The miracles worked by the relics
then form the main theme of the third book. These relic acquisi-
tions were not only central to Conuuoions strategy for building up
Redons reputation as a sacred place, they are crucial to the anony-
mous authors textual representation of it. Through them, he situated
Redon in the topography of Christian holiness. This had a regional
aspect, and the Gesta duly describe pilgrims who found healing at
Redon who had originated at Lehon (northern Brittany), Nantes,
Beauvoir (northern Poitou) and Glanfeuil (near Angers).99 In report-
ing the visionary visit of three famous holy men to one of these pil-
grims, the Gesta claimed too that Redon outranked the archiepiscopal
shrines of western Francia, Samson at Dol, Martin at Tours and
Hilary at Poitiers.100
More ambitious, however, was the universal dimension of the sta-
tus claimed for Redon as a result of its possession of the relics of
Pope Marcellinus. The monks and all the Bretons were delighted
with great joy that they had been found worthy to receive the vicar
of the holy apostle Peter in their province. In the same place the
Lord Jesus Christ reveals the merits of his martyr everywhere.101 As
a papal shrine, Redon was proclaiming its affiliation with the uni-
versal Roman church and its recognition of the authority of the vica-
rius Sancti Petri apostoli.102 Pilgrims from as far afield as Spoleto and
Lotharingia confirmed the monasterys status as an international cen-
tre of healing.103 Moreover, Redons relics healed those who had

98
GSR II. 9, 10, pp. 17182. The Hypothemius of the GSR is presumably
the Apodemius who is listed as the second bishop of Angers in episcopal lists.
L. Duchesne, Fastes piscopaux de lancienne Gaule, 3 vols. (Paris, 18941915), vol. 2,
pp. 3523. References to the presence of Hypothemius in the charters show that
his relics were there by 842/3 at the latest. On relic thefts in general, see P.J.
Geary, Furta sacra: Thefts of relics in the central Middle Ages, 2nd edn. (Princeton 1990),
which does not however refer to this particular theft.
99
GSR II.1, III.3, 5, 6, pp. 14751, 1957, 2013.
100
GSR III.3, p. 197.
101
GSR II.10, p. 181: gauisus est populus gaudio magno, quia meruerant accipere uica-
rium sancti Petri apostoli in sua prouincia; atque in eodem loco Dominus Iesus Christus merita
martyris sui ubique demonstrat.
102
GSR II.10, p. 181, describing Marcellinus. There is an important political con-
text here too, for the later years of Conuuoions abbacy were dominated by the
attempts of some of the Breton bishoprics to throw off their Carolingian archbishop
(Tours), and reject the jurisdiction of the Frankish church and of Rome. Smith,
Province and empire, pp. 15461; K. Herbers, Leo IV. und das Papsttum in der Mitte des
9. Jahrhunderts (Stuttgart, 1996), pp. 32036.
103
GSR III.1, 8, pp. 18993, 20713.
 - 387

failed to find a cure elsewhere. Frotmund and his brother had killed
two relatives in a feud over inheritance, and a Lotharingian synod
sent them bound in pentitential chains to Rome. According to the
Redon anonymous, they moved on to Jerusalem, Egypt, Africa, and
back to Rome. But the pope sent them back to the Holy Land,
where their wanderings took them to Cana, the Red Sea, Mount
Ararat and Mount Sinai before returning to Rome, crossing the Alps
and finally reaching Brittany. Frotmunds brother died at Rennes,
but he continued to Redon.104 That his chains finally fell off and his
wounds healed at the shrine of Marcellinus not only suggests that
Redon regarded itself as an outpost of papal influence, but also that
it claimed to be the most potent shrine in all Christendom.105 Where
Rome, Jerusalem and Mount Sinai failed, Redon succeeded.
In locating Redon in the topography of Christendom, the author
of the Gesta was also carefully omitting material which did not fit
his purpose. For Conuuoion had acquired other relics too. Maxentius
was a Poitevin saint whose relics had reached Brittany by unknown
means and which Conuuoion had enshrined in Redons dependent
cell at Pllan. Charters make reference to this saint; several high sta-
tus people, Conuuoion included, had been buried next to his shrine.106
Redon also possessed a copy of the vita Maxentii, but the Gesta author
ignores this.107 Redons fourth saint, Melor, is much more shadowy.
A Brittonic saint whose cult centre in Brittany had been Lanmeur,
three charters note his presence at Redon itself from 849 onwards.108
In stressing Redons Angevin and Roman allegiances, the Gesta con-
veniently forgot the Breton and Poitevin saintly heritage.
Just as the Redon archive presented the monastery as the focus
of a topography of power carefully built up by Conuuoion, so too
the Gesta presented it as the central place in a topography of Christian
holiness. Commencing with the consecration of the church, Redons
sacrality had gradually increased as relics were translated there and

104
GSR III.8, pp. 20713. For an identification of Frotmund and the political
context of his sentencing, see Poulin, Dossier, pp. 1545.
105
For the papal aspect of the shrine of Marcellinus, see Herbers, Leo IV, pp.
3738.
106
CR 52, 236, 241, 283. Cf. G. de Poerck, Les reliques des saints Maixent et
Lger aux IXe et Xe sicles et les origines de labbaye de lEbreuil en Bourbonnais,
Revue Bndictine, 67 (1962), pp. 6195.
107
CR 241.
108
CR 59, 269, A36. On Melor, see F. Duine, Mmento des sources hagiographiques
de lhistorie de Bretagne (Rennes, 1918), no. 84, pp. 99101.
388 ..

legal immunity acquired. One of the most sacred spots in Christendom,

Redons house author presented his readers and hearers with a holy
place so suffused with divine presence that it resembled paradise.
Redon doubtless had an urgent need to make this claim. Even in
its immediate vicinity, tiny monasteries tended the cults of local saints,
unknown to us by not thereby unimportant to their neighbours.109
At a regional level, other, more famous saints attracted pilgrims and
worked miracles at their well-endowed shrines. Eastwards, up the
Loire valley, Martins famous and rich shrine at Tours was 200 km
away; Maurus, disciple of Benedict, lay somewhat nearer to hand at
Glanfeuil, and Philibert had been translated in 836 from the island
of Noirmoutier to Grandlieu, only 75 km south-east of Redon (fig. 1).
Even closer, the shrine of Ermenland on the island of Indre down-
stream from Nantes also healed pilgrims sent on from Rome.110 To
the north, 100 km or as away, two outcrops off the Breton coast
both housed active shrines. By the 860s the bishops of Alet were
energetically promoting the relics of St Malo on his island hermitage,
whilst on a rocky knoll in the marshes of the Cousnon estuary, the
recently-arrived relics of St Michael were already demonstrating their
ability to attract pilgrims from all over the Frankish world.111 The
guardians of Hypothemius and Marcellinus had to compete in an
environment already pullulating with miracle-working shrines: Redons
claims reflect the magnitude of the challenge.

109
Martiris Sergii monasterium: CR 247; monasteriolum quod vocatur Sent Ducocan: CR
247, A4; Sancta Leupherina in monasterio Conoch: CR 152, 154; monasteriolum quod vocatur
Sent Tovi: CR A40, A45; monasterium Sancti Toinani: CR 276.
110
On Martins shrine as a centre of pilrimage, see P. Gasnault, Le tombeau
de Saint Martin et les invasions normandes dans lhistoire et dans la lgende, in:
Mmorial de lanne martinienne. Seizime centenaire de labbaye de Ligug, centenaire de la dcou-
verte du tombeau de saint Martin Tours (Paris, 1962), pp. 5166 at pp. 523. No liber
miraculorum records miracles being worked at St Martins, however. For miracles
recorded in the mid- to late-ninth century at Glanfeuil and at Grandlieu, see Odo
of Glanfeuil, Historia translationis S. Mauri, chs. 56, AA SS Jan. I, pp. 105160 at
pp. 10579 and Ermentarius, De translationibus et miraculis Santi Filiberti, ed. R.
Poupardin, Monuments de lhistoire des abbayes de Saint-Philibert (Paris, 1905), pp. 1970.
For a pilgrim sent from Rome to Indre, see Donatus, Vita Ermenlandi abbatis Antrensis,
ch. 27, ed. W. Levison, MGH SRM 5, pp. 674710 at p. 708.
111
On the cult of Malo and its promotion see Smith, Oral and written, pp.
3314. For the arrival of Michaels relics at Mont-Saint-Michel, see the Apparitio S.
Michaelis archangeli (as n. 72); for Frankish pilgrims there see Itinerarium Bernardi, ed.
T. Tobler, Descriptiones Terrae Sanctae ex saeculo VII, IX, XII, XV (Leipzig, 1874), pp.
8599 and Adso, Vita Frodoberti abbatis Cellensis, chs. 312, ed. W. Levison, MGH
SRM V, pp. 6788 at pp. 857.
 - 389

One way in which a holy place is distinguished from its milieu is

by rituals and codes of conduct which emphasise its purity. Early
medieval monastic texts make much of this; abundant legistation con-
trols monastic behaviour, restricts access to the laity and eliminates
pollutants from ritual actions and spaces.112 The Redon author was
sensitive to this: he depicts individuals behaving in ways which are
either clearly pure and proper, or improper and thus polluting. In
this way, he demonstrates that Redon is a place where full purity is
maintained. To demonstrate that Redon was a place of purity, the
Gesta author judged people not according to whether they were monks
or clerics but instead in terms of their support for Redon and its
ideals. Pious lay benefactors who gave land and offered their sons
as oblates enhanced the blessedness of Redon;113 hermits and monks
from other monasteries who spent a period of time at Redon con-
tributed to its sacrality.114 These included the Angevin hermit Gerfred
who instructed the fledgling community in the rule of St Benedict.115
But when brother Osbert left the angelic life, he fell into the wost
possible form of error, falling prey to demons and lost all discipline
in his conduct, as an untamed horse without a rider rushes any-
where headlong.116 Adherence to the Rule of Benedict thus forms
an important definition of Redons purity.117 In a landscape dotted
with tiny shrines, and the face of the regional competition of such
prestigious cults as St Martin or St Michael, emphasis on regular
life distinguished Redon and reinforced its allegiance to Carolingian
reform ideology.
Notable among these images of the pure and impure are the bish-
ops of Vannes. Raginarius is mentioned only for his opposition to
Redons foundation; his successor Susannus (838848) only for his

112
A. Diem, Keusch und rein. Eine Untersuchung zu den Ursprngen des frhmittelalter-
lichen Klosterwesens und seine Quellen (Amsterdam, 2000); M. de Jong, Internal clois-
ters: the case of Ekkehards Casus Sancti Galli , W. Pohl and H. Reimitz eds., Grenze
und Differenz im frhen Mittelalter. Forschungen zur Geschichte des Mittelalters 1 (Wien,
2000; forthcoming).
113
GSR I.3, pp. 1159.
114
GSR I.2, III.3, 5, pp. 1135, 1957, 2013.
115
Gerfred, a monk of St-Maur de Glanfeuil, was a hermit in western Brittany
at Coat-Guinec au Huelgoat (Finistre). Poulin, Dossier, p. 149.
116
GSR II.5, pp. 1615; quotations from p. 163.
117
Similarly, charters drafted at Redon put emphasis on its Benedictine observ-
ance, frequently referring to grants made ad illos monachos habitantes et exercentes regulam
Sancti Benedicti in monasterio quod vocatur Roton. For example: CR 2, 3, 4, 5, 6, 7, etc.
390 ..

improper conduct, which Conuuoion denounced to the pope.118 The

Gesta create the impression that the monastery functioned without
any reference whatsoever to its diocesan bishop. The solemn litur-
gies to enshrine the relics of Hypothemius and Marcellinus are the
litmus test: Carolingian legislation had stipulated that relic cults should
remain under firm royal or episcopal control. Many Frankish trans-
lationes take care to emphasise the bishops officiating role, yet the
Gesta remain silent about who said mass at the installation of the
relics of Hypothemius and Marcellinus.119 Furthermore, we know
from a letter which survives among the Redon charters that the bish-
ops of Vannes did have a claim to authority over the monastery, at
least for ordaining its monks to the diaconate and priesthood but
that Bishop Courantgen (850868) relinquished the right.120 Whilst
it may be going to far to characterise the bishops of Vannes as the
villains of the piece, their rhetorical role as the antithesis of Redons
holiness hints at tense relations between monks and bishops and
underscores the political subltety of the Gesta.
More useful in narrative terms were Redons secular opponents,
both Louis the Pious with his entourage and the local landowners
who opposed Redon and tried to recover lands which their relatives
had donated to the monks.121 Conuuoion had won imperial support
after two years of adroit politicking, but for the Gesta audience the
moral was rather that all that will live godly in Christ Jesus shall
suffer persecution.122 Redons monks thus approximate to living mar-
tyrs, the soldiers of Christ [who] stand in battle order.123 The theme

118
GSR I.8, II.10, pp. 133, 179. The context of the charges against Susannus in
GSR II.10 was highly political. This and all other accounts of the events of 8489
are highly tendentious. See my comments in Province and empire, pp. 1546. Guillotel
has recently made the important suggestion that GSR II.10 may have been rewrit-
ten in the late eleventh or early twelfth century. Cartulaire de Redon: le manu-
scrit, p. 20.
119
Carolingian legislation is summarised by Heinzelmann, Translationsberichte,
p. 36. For examples of bishops authorising and presiding at relic translations in the
830s see Historia translationis s. Balthildis, AASS Jan., II.7479; Translatio sancti Viti mar-
tyris, ch. 5, ed. I. Schmale-Ott, Verffentlichungen der Historischen Kommission
fr Westfalen (Mnster, 1979), pp. 468.
120
CR A46. Courantgen surrendered the right to ordain Redons monks on the
grounds that Viking raids were making the journey to Vannes approx (60 km) too
difficult.
121
GSR I. 1, 5, 6, 7, pp. 109, 12131.
122
II Tm 3:12 quoted in GSR I.7 and 11, pp. 127, 141. Translations of biblical
quotations given in the Douai-Rheims version.
123
GSR I.3, p. 119: ibi adstant in acie milites Christi.
 - 391

is pursued through the accounts of the monks dealings with local

tyrants. Whereas the charters show Conuuoion resorting to lawcourts,
negotiation and pay-offs to defend his property and rights, the Gesta
reassures the monks that trust in God would bring victory over their
enemies, and that blessed are they that suffer persecution for jus-
tice sake: for theirs is the kingdom of heaven.124 As virtual martyrs,
Redons monks live in a virtual paradise, for Redon itself was the
site where the army of God was arrayed.125
Outward persecution was not the only or even the best mark of
the purity of Redons monks. In presenting their inner holiness, the
author of the Gesta had a persuasive model to hand in the Dialogues
of Gregory the Great.126 Writing in 5934, Gregory had offered
sketches of many Italian holy men as a reassurance that, with the
age of actual martyrdom now past, good men performing miracles
did still exist.127 Gregory presented a cogent portrait of holiness pre-
senting the quotidian virtues of Christian life: charity, obedience and
the discipline of the flesh.128 Whether monk, bishop or even layman
committed to a Christian life, Gregorys saint was the vir Dei, vir ven-
erabilis, vir sanctus, famulus Dei.129 His saints sanctity was rarely spec-
tacular, and always the external mark of the interior spiritual condition.
It thus adapted the literature of asceticism to the conditions of post-
Roman Italy. Gregorys account of everyday sanctity soon established
itself as one of the most influential and popular monastic texts of
the Middle Ages.130
The highly institutionalised Benedictine monasticism of the Carol-
ingian empire, with its emphasis on intercessory liturgy and corpo-
rate existence was a far cry from Gregorys understanding of the
ascetic life. So too was the tendency for monks to be priested, for
child oblation to provide a major source of recruits for the monastic

124
Mt 5:10 quoted in GSR I.9, p. 135.
125
See quotation linking the paradise-like site of Redon and its battle lines in
n. 86 above.
126
Direct quotations from the Dialogues are noted by Brett, p. 65 and Poulin,
Dossier hagiographique, p. 144. For the Dialogues influence on another ninth-
century hagiographical text, see H.E. Stiene, Gregors des Grossen Dialogi und die
Vita Goaris Wandelberts von Prm, Mittellateinisches Jahrbuch 18 (1983), pp. 5162.
127
Dialogues, ed. A. de Vog, SC 251, 260, 265 3 vols. (Paris, 197880), I. pref.
7, vol. I, p. 14.
128
C. Straw, Gregory the Great: Perfection in imperfection (Berkeley, 1988), p. 69.
129
Epithets assembled by Straw, Gregory the Great, p. 71.
130
De Vog, introduction, vol. 1, pp. 1413, 1648.
392 ..

life, for clerical canons to be attached to the monastery and for com-
memorative prayer for patrons and benefactors to command much
time.131 In all these respects, Redon was a typical ninth-century
monastery.132 Yet the author of the Gesta sanctorum Rotonensium found
no discordance between the monastic life he experienced directly
and that portrayed by Gregory. Indeed, the Dialogues provided the
template for demonstrating the ritual purity of Redons monks and
enabled him to stress the efficacy of their prayers and masses.
With his emphasis on vita et conversatio,133 he transformed Redons
monks into holy men just like Gregorys. The latter had emphasised
his subjects holiness by recording the miracles which were the outer
signs of their vocation and monastic way of living, their conversatio
and their vita.134 In adopting this correlation between purity of life
and miraculous actions, the Redon author remained unaffected by
the deep mistrust of living holy men and thaumaturges which dis-
tinguished Carolingian Christianity from its late antique roots.135 His
monks are the living saints of Redon, for the sancti monachi and their
sanctus ac venerabilis . . . abbas136 heal the sick, walk on water, have
visions, identify and defeat the devil.137 Just as for Gregory the Great,
such miracles were important inasmuch as they provided the exter-
nal signs of inner spiritual grace. Condeloc, for example, once a
wealthy layman, became the monastic gardener. A vir simplex et rectus,
atque omnibus bonis adornatus, his new life was led in summa sanctitate.
He maintained his chastity; innocens, he trusted everyone; his eyes
flowed with tears. Because he flourished with virtues Conuuoion

131
Cf. De Jong, Carolingian monasticism.
132
Monks as priests: of the 28 witnesses of CR 224, twenty were priests; monks
celebrating daily mass: GSR II.1, 2, 3, pp. 1475; oblates: CR 27, 51, GSR I.3, pp.
1159; clerical canons: GSR II.6, p. 165; commemorative prayer: CR 28, 52, 69,
80, 236, 238, 255, A44. In addition, dozens of pro anima mea charters imply prayers
for the souls of living donors.
133
Above, n. 65.
134
For example, on Sabinus, bishop of Canosa, Gregory says: sed talis eiusdem viri
vita perhibetur, ut qui conversationem agnoverit, virtutem non debeat mirari. Dialogues, III.5.5,
ed. de Vog, II.276.
135
J.M.H. Smith, The problem of female sanctity in Carolingian Europe,
c. 780920, Past & Present 146 (1995), pp. 337; P. Fouracre, The origins of the
Carolingian attempt to regulate the cult of saints, in: J. Howard-Johnston and P.A.
Hayward eds., The cult of saints in late antiquity and the early Middle Ages (Oxford, 1999),
pp. 14365.
136
GSR II. Pref., 1, pp. 145, 1479.
137
GSR II.1, 2, 4, 5, 8 pp. 14753, 15765, 16771.
 - 393

put him in charge of the monastic garden. His inner holiness man-
ifested itself when, faced with an infestation of caterpillars, he expelled
them by invoking the Trinity.138
In recounting Redons foundation, the author had already sum-
marised the virtues for which Redon was important: caritas . . . mira,
abstinentia magna, humilitas summa, castitas ante omnia.139 His miracle sto-
ries add substance to the assertion and affirm that humble monks
contributed as much as their influential abbot to Redons holiness.
Since the miracles worked by the monks were not theirs but Gods,
they furnished a further proof that Redon was a place touched by
Gods presence.140 The anonymous authors message for his readers
is clear: See, dearest brothers, by what ways and means the Lord
makes his saints wonderful everywhere in heaven and on earth.141
As Christ had said to his disciples, Without me you can do noth-
ing.142 Redons monks should count themselves among the saints,
and know that Christ was in their midst.
Among ninth-century works of monastic edification, the Gesta sanc-
torum Rotonensium is highly distinctive for retaining the Gregorian readi-
ness to associate purity of monastic life with miracle-working. The
author developed the theme so that, by affirming Gods presence at
Redon, he could argue that the sanctity of this holy place inhered
not only in its relics but in its living inhabitants and the pure liturgy
which they maintained in the face of persecution. As a rhetorical
strategy, it enabled him to present Redon as a worthy rival to the
established holy places of Francia and Christendom, equally potent
for its lived sanctity and its relics.
That the holiness of Redons monks demonstrated the holiness of
the place is the message of the last chapter of the Gesta in its extant,
truncated form. Here, the author narrates with brio the activities of
two Viking fleets in the mouth of the Loire in 852 and then the
move of one of them up the Vilaine. Before fleeing, the monks
prayed that their monastery would be spared the pollution of the
pagans. A storm panicked the Vikings, who vowed not to desecrate

138
GSR II.3, pp. 1535.
139
As n. 85 above.
140
Mirabilia domini nostri Iesu Christi . . . per famulos suos. GSR II.4, p. 157.
141
Videte, fratres carissimi, qomodo uel qualiter Dominus sanctos suos in caelo et in terra
ubique mirificat. GSR II.4, p. 159.
142
Io 15:5, quoted in GSR II.2, p. 151.
394 ..

the holy place of God and instead lit candles around the altar.
Only those Vikings who slunk into the sacristy to drink the mass
wine died, having gone instantly mad. The miracle manifested the
transformative powers of Redons mass wine; the Vikings who sur-
vived were those who honoured the sanctity of the monastic altar.
Here, finally, the author made the parallel between Jerusalem and
Redon explicit. As Sennacherib, king of the Assyrians prepared to
attack Jerusalem but was deflected by the God of the Israelites, so
God thwarted the Viking attack on Redon.143 As the sacrosanct space
of Redons altar lay at the heart of the cloister, the place furthest
from the outside world where priestly purity contributed to the efficacy
of the sacraments, so this episode lay at the heart of the Gesta.
Postponed until nearly the end of the work, the safe-keeping of the
Breton Jerusalem is the narrative culmination of everything which
precedes it.

Redon was on the outermost periphery of the networks of associa-

tion, friendship and patronage which bound together many of the
major monasteries of the Carolingian empire. Unmentioned in the
extant libri memoriales, its abbot and monks lacked links to court cir-
cles, the texts that circulated through such networks or the aristo-
cratic patronage that came with them. Although it participated fully
in the culture of Carolingian monasticism it nevertheless lacked the
means to integrate itself in that wider monastic world. Nevertheless
Redon enjoyed direct, if tenuous, links with the Carolingian court,
for Louis the Piouss precious diplomata legitimised its existence and
bound it into the community of prayer which upheld imperial author-
ity. Charles the Bald, furthermore, incorporated it into the network
of juridical holy places which formed his state church. To construe
Redon as a marginal place is thus overly reductive. Rather, its his-
tory points to the flexibility of state-sponsored Benedictine monasti-
cism, whose ideologies and practices only spread successfully insofar
as individual monastic communities found it worth their while to tap
into them. Redon was authentically Carolingian-Benedictine, but on
its own terms.
From a local point of view, Redon was an upstart and powerful
presence, which rode rough-shod over traditional ways of conduct-

143
GSR III.9, pp. 21319.
 - 395

ing life but brought spiritual benefits to the region. That it devel-
oped as an important central place in its own right is an example
of a monastery functioning as an agent of social and political change,
of monastic engagement with the secular world. It also reminds us
forcibly of the fluidity of political and cultural topography of early
medieval imperial peripheries. The twelfth-century abbey church that
replaced Conuuoions original buildings is today stands at the cen-
tre of a prosperous regional market town, a mark of the enduring
significance of these early medieval transformations to topographies
of power.
The moment of Redons foundation had hardly been opportune.
Under the shrewd leadership of Conuuoion, the community had nev-
ertheless prospered and obtained widespread support. It had forged
an institutional existence which exploited local possibilities, princely
patronage and distant imperial ideologies alike. Conuuoions death
in 868 perturbed its search for stability however. Since the Carolingian
state church and Breton princely power both crumbled in the last
quarter of the ninth century, his demise ushered in a period of
retrenchment from which Redon never recovered in the early Middle
Ages. In this changed environment, Redon needed a defender, a
proponent, an advocate. It needed to justify its existence, to show
that there was only one safe haven in a world where power was
always in contestation, whether between pagans and Christians, Franks
and Bretons, abbot and laymen, God and the devil. One anony-
mous monk rose to the challenge. He argued Redons case: that it
was Jerusalem in Brittany, paradise on earth, home of miracle-work-
ing monks, a veritable holy place. It needed to be: how else could
a new monastery transform the ecclesiastical topography of western
Gaul whose main contours stretched back five hundred years to the
days of Martin of Tours? And how else could it justify itself in the
straightened and insecure circumstances of the late ninth century?
Perhaps because he was spared the straitjacket of an institutional
history reaching back far into the distant past and replete with its
own canonical traditions, perhaps because he lacked access to the
most up-to-date norms of Carolingian historical and hagiographical
literature, perhaps because of a sturdily independent habit of thought,
the Redon anonymous produced a work of considerable originality
and effectiveness. Carefully selecting a range of topographical, ideo-
logical and ascetic themes which served his purpose, he envisioned
Redon as none of the monks can ever quite have experienced it.
396 ..

With unadorned Latin but skilful rhetoric he constructed a place of

such unimpeachable holiness as to obfuscate the degree of special
pleading involved. That the Gesta are essentially tendentious emerges
only when they are read against Redons charters on the one hand,
and Latin hagiographical conventions on the other. This interpreta-
tion reveals that whatever late ninth-century Redon lacked in insti-
tutional stability, it acquired in textual representation.
In arguing the case for Redon, the Gesta sanctorum Rotonensium may
seem grandiose and unrealistic. But that is to regard the monasteries
with enduring links to the Merovingian or Carolingian ruling dynas-
ties communities such as Saint-Denis, Fulda, Prm or Lorsch
as the norm. They were not. They enjoyed exceptional wealth, priv-
ilege and prominence, and left ample written evidence to tell us so.
The typical early medieval monastery was one which left scrappy,
partial documentation, endured cyles of prosperity and decay, and
enjoyed only a regional significance. Redon represents this. Its ex-
ceptionalism lies in the conjunction of a large archive and a historical-
cum-hagiographical self-portrait of overlapping content. The Redon
cartulary and the Gesta each offer a profoundly different perspective
on what was entailed in building a holy place in the ninth century.
We should not fall into the trap of thinking that one is somehow
more real than the other: rather, they present complementary iden-
tities, each constructed within very different textual traditions and
strategies. Each puts the other into perspective. Taken together, they
demonstrate the textual bid for monastic power at its most persua-
sive and cogent.
 PEOPLE, PLACES AND POWER IN
CAROLINGIAN SOCIETY*

Matthew Innes

A C : ,

In the middle of August 795, a public meeting was held in the woods
which covered the area where the valley of the Rhine met the rugged
hills of the Odenwald, at a tumulus called the Walinehoug close by the
villa of Heppenheim. Heppenheim was the site of a sizeable and
strategically placed estate centre which had been granted to the
nearby, and newly founded, abbey of Lorsch by Charlemagne him-
self in 773, immediately after he had assumed lordship over the
monks. Evidently the precise bounds of the extensive rights over the
woodland and hamlets of the surrounding area that went with
Charlemagnes gift were a matter of dispute between the monks and
their neighbours, for one or other party petitioned Charlemagne him-
self about their extent. Hence the calling of the 795 meeting through
a precept of King Charles, and the subsequent establishment of the
boundaries of property rights within the Odenwald through the
judgement and testimony of illustrious men.1
Over 1200 years on, a local meeting in the woods that line the
Rhine valley as it meanders its way north might seem impenetrably

* I would like to thank fellow participants in the Topographies of Power con-

ference, Marios Costambeys, Paul Fouracre and Simon MacLean for discussion of
issues covered in this paper. For the wider political and social history of the area
discussed, see M. Innes, State and society in the early Middle Ages: the Middle Rhine Valley,
400 1000 (Cambridge, 2000); the current discussion aims to investigate the rela-
tionship between space, place and power in the hope of opening a different, but
complimentary, perspective on the basic problems of early medieval society to that
adopted in my book, but some self-citation has proved unavoidable in covering
background information.
1
Codex Laureshamensis, ed. K. Glckner, 3 vols. (Darmstadt, 192936) [hereafter:
CL plus charter number], vol. 1, no. 6a: De marcha Hephenheim . . . . Warinus ex pre-
cepto Karoli regis anno XXVII regni eius, mediante mense augusto placitum in eadem silva ad
tumulum qui dicitur Walinehoug habuit, et cum illustrium virorum iudicio et testimonio terminum
et divisionem eius faciens, eam a silva que pertinet ad Moynecgowe, et ab omnibus circumpositis
marchis, sub certis et designatis limitbus disterminauit.
398

obscure, and the actions of those who attended of little moment. But
by analysing moments such as this, when latent conflicts within the
rural communities that made up the Carolingian Empire emerge
into the light of day, we can examine some fundamental aspects of
social organisation.2 Despite its inherent difficulty and modesty, the
notice of the 795 enquiry is thus an excellent starting point for an
investigation of the relationships between place and power in the
Carolingian world. I thus begin with a microcosm made possible by
the 795 meeting. Within this small world power the ability to
affect the strategies of others, either actively or passively was
embedded in interpersonal relationships, but exercised in public; like
an electric current, it flowed from person to person at those places
where individuals came together, and local conflicts were processed
and consensus created. Using the very full documentary record from
the middle Rhine and lower Main valleys, it is possible to trace the
development of the circuits of power over time. This essay argues
that, at the beginning of the Carolingian period, the making of those
contacts which allowed power to be exercised locally remained essen-
tially a matter of social interaction, which was patterned around a
settlement hierarchy ultimately determined by, but not identical to,
the Roman civitas system, which were the ultimate places of power
through which rulers plugged into local currents. The ninth century
saw changes in these local patterns of interaction, and the creation
of fixed and insulated connections between the circuits of local power.
These changes enabled both kings and elites to plug into the cur-
rents of local power more efficiently from a greater distance, and
thus increased the social and political distance between rulers and
localities. But they also led to more direct competition between kings
and elites for access to local circuits of power. Changing spatial pat-
terns of social action can thus reveal the complexity of the evolving
symbiotic relationship between kings, elites and localities in the
Carolingian period (fig. 1).
From the outset, it is vital to remember that the sources from
which we work did not stand outside of the relationships they recorded.

2
For this approach see e.g. W. Davies and P. Fouracre eds., The settlement of dis-
putes in early medieval Europe (Cambridge, 1986). For the value of the charter mate-
rial in building a bottom up picture of early medieval society see C. Wickham,
Rural society in Carolingian Europe, in: R. McKitterick ed., New Cambridge Medieval
History II (Cambridge, 1997), pp. 51037, and the model case study, W. Davies,
Small worlds: the village community in early medieval Brittany (London, 1989).
 , 399

Fig. 1. The middle Rhine valley and the major places mentioned in the text.

In particular, the documentary riches on which this essay is based

rehearse the property claims of the church in the language of the
church. Unlike an ethnographer able potentially at least to talk to
all the parties involved, we have only a textual presentation, and
one produced by a particular interested party and designed to bol-
ster their particular understanding of property relationships at that.3

3
But for a stimulating ethnographic study which considers the active relation-
ship between local knowledge in the field and official documentation, see P. Gow,
400

Nonetheless, even such a partial record allows us to view the mech-

anisms through which local conflicts were processed, and if we recre-
ate the context of our text carefully enough we may even begin to
hear an echo of the loud debate that must have taken place.4 Indeed,
the account of the August 795 meeting is one of a particularly intrigu-
ing, yet almost wholly neglected, class of documents recording the
local enactment of royal commands. It sheds a penetrating light on
the actual impact of the royal will here itself transmitted in writ-
ten form in the localities, in other words the basic mechanisms
of Carolingian government.5

The record of the 795 meeting is a difficult text or rather, series

of texts, for the woodland rights that were at stake were of such
importance to the Lorsch monks that the boundaries established by
the 795 meeting were amplified and edited by successive monastic
scribes to support current claims. By the twelfth century, when the
surviving Lorsch cartulary was compiled and the surviving account
of the 795 meeting copied into the introductory chronicle as an
appendix to Charlemagnes gift of Heppenheim to the monks, it was
one of the foundation charters of the territorial lordship the monks
exercised in their immediate vicinity.6 It is therefore composed of a
number of layers, and the work of successive generations of modern
antiquarians, diplomatists and local historians have gone some way
towards elucidating its textual archaeology. The account of the 795
meeting, so circumstantial, unformulaic and unconcerned with self-
verification and thus so unlike a forgers concoction, is accepted as
essentially genuine, not least as the procedures and personnel it details
are so utterly consistent with those recorded in the thousands of

Land, people and paper in Central Amazonia, in the useful anthology edited by
E. Hirsch and M. OHanlon, The anthropology of landscape: Perspectives on place and space
(Oxford, 1995), pp. 4362.
4
Cf. R. Balzaretti, The monastery of Sant Ambrogio and dispute settlement
in early medieval Milan, EME 3 (1994), pp. 118.
5
See M. Innes, What was Carolingian government?, in: J. Storey ed., Charlemagne
(forthcoming). It is worth noting that the almost wholesale ignorance of such doc-
uments is a direct result of the ingrained statism of modern scholarship, and par-
ticularly the rigid application of nineteenth-century bifurcations of public and
private: local documents responding to royal commands like the Heppenheim
placitum have been left among the private charters and thus dissociated from
official and governmental material and left to local historians, social historians
and prosopographers.
6
Innes, State and society, pp. 24650.
 , 401

other transmitted documents from this region at this time; but the
boundaries it describes are to be regarded with the utmost suspicion.7
We are given the names of no fewer than thirty seven men who
served as witnesses and made these boundaries: they are listed as
representatives of the geographical units, or pagi, into which the
region was divided, with a dozen men from the Lobdengau (the
lower Neckar valley around Ladenburg), eight from the Wingarteiba
(a small area deep within the Odenwald), and seventeen, led by one
Count Rupert, coming jointly from the Rheingau (the western part
of the Odenwald down to the banks of the Rhine) and the Maingau
(the lower Main valley and the neighbouring parts of the Odenwald).
The identities of these individuals can be fleshed out from the wealth
of documentary evidence for land-holding and monastic patronage
in the area; and the lists are circumstantially plausible in the light
of the rest of the surviving documentation, much of which did not
survive through the Lorsch archive. These men property-holders
of some social distinction, the grandest of them not without illustri-
ous connections or hopes of rising through royal service were
representatives of their localities, leaders of their communities. Their
coming together to reach a collective judgement and determine right
both mobilised the local knowledge necessary to settle the claims and
counter-claims they heard, and underwrote the acceptability and
validity of the settlement.

7
The bibliography is vast, but Glckners notes in his edition of the text remain
a good starting-point, and H.-P. Lachmann, Die frhmittelalterlichen Marken zwi-
schen Rhein und Odenwald unter besondere Bercksichtigung der Mark Heppenheim,
Berichte zur deutsche Landeskunde 49 (1975), pp. 2737, the best treatment, which I
largely follow; see also C. Wickham, European forests in the early Middle Ages:
Landscape and land clearance, reprinted from Settimane 37 (1989), pp. 479548,
in Wickhams collected essays, Land and power: Studies in Italian and European social his-
tory, 4001200 (London, 1994), pp. 155200. Note that the two critical key studies
of the middle Rhenish material, M. Gockel, Karolingische Knigshfe am Mittelrhein,
Verffenlichungen der Max-Planck Instituts fr Geschichte 31 (Gttingen, 1970) and
F. Staab, Untersuchungen zur Gesellschaft am Mittelrhein in der Karolingerzeit, Geschichtliche
Landeskunde 11 (Wiesbaden, 1975) both likewise regard CL 6a as an interpolated
account of a genuine meeting. There is a good discussion of the significance of the
interests at and around Heppenheim, again accepting the account of the 795 meeting,
by the doyen of mittelrheinische Landesgeschichte, H. Bttner: Die politische Krfte zwi-
schen Rhein und Odenwald bis zum 11. Jht., in his collected essays, ed. A. Gerlich,
Zur frhmittelalterlichen Reichsgeschichte am Rhein, Main und Neckar (Darmstadt, 1975), pp.
25366.
402

Not only did the tradition of processing conflict through public

meetings ensure both local knowledge and local agreement, but also
the precise makeup of the 795 meeting underlined a concern with
reaching a broadly based and extensive local consensus. The villa of
Heppenheim, and the vast woodland mark which was attached to
it, lay unambiguously within the Rheingau, but representatives of
the surrounding pagi as well as the illustrious of the Rheingau met
at the Walinehoug (fig. 2). Doubtless this was partly because at places
the boundary of the Heppenheim mark were coterminous with the
boundaries of the Rheingau. But this was essentially an assembly of
the local elite as a whole to settle a highly contentious dispute, which
determined the distribution of economic, social and political power
within their area. After all, it was not Count Rupert, the royal official
who led the contingent of Rheingau and Maingau men, who presided,

Fig. 2. The mark of Heppenheim with its boundaries in 795 (after Lachmann)
and the later interpolations to the west and south.
 , 403

but another count, Warin, most of whose activities in royal service

were focussed slightly further to the south, along the Neckar. It is
more or less impossible to make sense of the precise personnel and
procedures used in terms of administrative or jurisdictional bound-
aries: this was a meeting whose rationale was social and political,
not a court with a defined legal competence.8
Warin brandished the royal precept which called the meeting and
authorised it to determine the boundaries of the Heppenheim mark
not as count of the locality, but on account of his special link with
the place under discussion, for the notice rehearses the history of
the Heppenheim estate before Charlemagne gave it to Lorsch. It
recounts how Heppenheim was held by Warins father, Wegelenzo,
in beneficium, then by Warin himself as a count in ministerium ad opus
regis, and then by Count Baugolf.9 Deciding on the current state of
property rights involved establishing an agreed view of their history;
the recitation of the estates genealogy established that the 795 meet-
ing was establishing an order which had, according to the scribe,
existed since ancient time, and which could be traced back through

8
In an important article J. Hannig, Zentrale Kontrolle und regionale Macht-
balance. Beobachtungen zum System der karolingischen Knigsboten am Beispiel
des Mittelrheingebietes, Archiv fr Kulturgeschichte 66 (1984), pp. 146 esp. pp. 2733,
argues that the Heppenheim case and other surviving placita deal with particularly
difficult or sensitive cases which could not be handled through the normal courts;
but given the total lack of evidence for the existence of a normal system of neat
territorial jurisdictions it may be more likely that the kind of local meetings revealed
in the placita were the norm see Innes, State and society, pp. 11824.
9
Again, a history which finds striking corroboration elsewhere in the docu-
mentary evidence: see Innes, State and society, pp. 1802, with references to earlier
discussions, for similar changes at Lorsch, and note also Die Urkunden Pippins, Karlmanns
and Karls des Groen, ed. E. Mhlbacher, MGH DD 1 (Berlin, 1906), Charlemagne
charter no. 63 for Charlemagnes similar intervention, on formerly fiscal land at
nearby Umstadt in favour of Fulda against a local landowner. We are clearly wit-
nessing a wholesale working of local patronage networks, involving a thorough look
at local landowning particularly on fiscal land. For Heppenheim see Innes, State and
society, p. 190 and for the family involved Gockel, Knigshfe, pp. 2625. This Baugolf
was a close kinsman of Warins, and probably identical with the eponymous abbot
of Fulda. Note that the language used to explain Heppenheims status first as
a beneficium held by Wegelenzo, who was not a count, then a ministerium held by a
local count is recognisably that of the early Carolingian period, whilst the
reflection that the woodland mark had been attached to Heppenheim semper ex tem-
pore antiquo sub ducibus et regibus again shows a full understanding of local history
that would have escaped a later forger (in the late seventh and early eighth cen-
tury the region up to the Rhine had been effectively ruled by more or less inde-
pendent duces based at Wrzburg: Innes, State and society, p. 176).
404

three generations of property-holders.10 An intimate link between

property rights, the people who had exercised them, and the places
at which they were exercised runs through the entire procedure: the
meeting, led by Warin, former holder of Heppenheim, took place
within the woods under scrutiny themselves, concluded with a per-
ambulation of the boundaries which it had established. A written
account complemented the memory of those who took place.
The account of the 795 meeting records the creation of a topog-
raphy of ownership. The boundaries of the Heppenheim mark were
consciously embedded in local topography. The long account of these
boundaries, notwithstanding the accretions of later confection, serves
as a reminder that rights over land were understood first and fore-
most as experienced practices. Because they were practical and real,
they were etched into the landscape through topographical markers
and place-names. Both could be used actively, to make claims, as
well as serving as a passive witnesses of exercised rights. The sur-
viving account describes the bounds of the Heppenheim mark being
physically marked, with the construction of tumuli at key points by
local property-holders, some of whom doubled as royal agents. Place-
names similarly served to articulate the property claims of land-
owners whose interests abutted onto this contested boundary with its
predatory monastic owner: Albwines sneida, Hildegeresbrunno, Manegoldescella
and so on. The description of the bounds thus evokes a world where
rights and relationships were set in a living landscape, and place was
central to identity.11 But we should also remember that such detailed
and elaborate accounts of estate boundaries are extremely unusual
amongst the voluminous Carolingian material. Evidently in normal
transfers of property, concerning discrete units of agricultural exploita-
tion or settlement, such details needed no rehearsal: where so-and-

10
For the ongoing debate about the applicability of notions of customary law
to the early middle ages see the characteristically sensible overview of S. Reynolds,
Kingdoms and communities in Western Europe, 900 1300 (Oxford, 1984), pp. 1266. The
Heppenheim document presented its conclusions as the declaration of a self-evident
order rooted in past practice, but these findings were debated and constructed, and
clearly in actuality anything but long-established: antiquity is here a legitimating
ideology within which the meeting operated.
11
Thus P.J. Geary, Land, language and memory in Europe, 7001100, Transactions
of the Royal Historical Society 9 (1999), pp. 16984, drawing particular attention to
the frequent use of the vernacular in boundary clauses right across western Europe,
and whilst not discussing the Heppenheim document drawing on evidence from
Fulda and Wurzburg which is very close in date and context.
 , 405

sos holding ended was known to all, and all that was recorded was
the owner of neighbouring owners. The Heppenheim description,
though, settled boundaries of a different significance, which marked
off a huge area of wooded hills, complete with scattered dwellings,
peasant holdings, hunting and woodland rights exercised by a vari-
ety of individuals and institutions. Such rights were anything but self-
evident or rooted in everyday practice: they established essentially
political control on a level divorced from the direct exploitation of
the countryside. Hence the care with which the highly contentious
and actively contested rights were recorded, and the very need for
a written record in the first place.12
Note that the meeting was held out-of-doors, away from the main
core of settlement in Heppenheim, at a tumulus whose name may
have commemorated an earlier holder of the villa, and which cer-
tainly seems to have served as an identifiable marker at which locals
could meet.13 Whatever private negotiations went on before the 795
meeting, decisions were reached not at the royal court, not at Lorsch,
nor at the residence of either of the counts present, but at the place
under discussion itself: rights were inextricably tied to the places
where they were exercised, and the landscape itself was witness to
rights over the land. The findings of the meeting may have been
recorded in writing and preserved at Lorsch, but in 795 agreement
was reached face-to-face and in public, and hence out-of-doors. On
this most local of levels, rights were inscribed in the landscape,
through the actions of human inhabitants who shaped and labelled
it: landscape functioned as a mnemonic for property rights and the
relationships between people, which they created.

Whilst place and landscape were important in the settlement of the

boundaries of Heppenheim, they were not sufficient on their own.
This was a dispute over the local distribution of power, and its reso-
lution took force because it was agreed and established by illustrious

12
These concerns are also present in the other detailed boundary descriptions in
the Lorsch and Fulda material; for the argument that the Fulda material actually
relates to a significant shift in the organisation of agrarian exploitation, see Innes,
State and society, pp. 7381. We cannot therefore see such accounts, whether vernacular
or Latin, as straightforward evidence for an organic link between land and identity.
13
Indeed, if the element Waline- which serves to identify the tumulus relates to a
personal name, as seems likely to this untutored eye, it is striking how closely it
relates to names in the family of those local nobles who held Heppenheim from
the king before 773.
406

men in a meeting sanctioned by the king and summoned through

his written order carried by one of his local agents. This meeting
was thus public simultaneously in several different senses. It was set
in motion by an authoritative royal order, chaired by his agent on
account of that order, and thus legitimated by royal power. But its
findings gained force from the presence of the local elite, and its
staging out of doors, in the collective gaze; its doings were thus open,
transparent and manifest. In this sense, the meeting involved an
implicit process of representation, with local elites acting as de facto
representatives of the wider community because they acted before
that community with royal blessing. Their judgement was therefore
able to determine the fate of the peasants who had interests in
Heppenheim and the neighbouring woodland.14
As the subsequent alteration and interpolation of the written account
of boundaries reminds us, however, the decisions inscribed in the
landscape through word and deed in 795 did not stay constant for
all time. The settlement of 795 was essentially an agreement about
the local distribution of power, and the relationship between Lorsch
and its neighbours at a particular point in time; the story of the suc-
cessive reworkings of the details of the 795 agreement by successive
generations of Lorsch monks is the story of the changing balance of
power in the locality. Interestingly, here writing, operating in the
context of non-written forms of legal memory, proved malleable, and
was adapted to reflect changing realities, just as the landscape fea-
tures recorded in the successive versions of the bounds had their
meaning altered to legitimate new social realities: we can see no
hard-and-fast divide between oral and written legal practices, but
rather see both serving to underline the status quo and to legitimate
current claims. Writing did not alter social practice in and of itself:
the extension of the boundaries of Lorschs rights as recorded in
writing seems to have followed Lorschs acquisition of the dominant
property interests in the settlements around Heppenheim in the ninth
century, whilst physical force was deployed to good effect in assert-
ing and defending claims in the tenth century. Nonetheless, the
monks control of the written record regarding the Heppenheim mark
gave them a strategic advantage over their neighbours, particularly

14
Cf. J. Habermas, The structural transformation of the public sphere, transl. T. Berger
(London, 1989), esp. pp. 113.
 , 407

as local power came to be abstracted from social relationships on

(and with) the ground. Lorschs interests in the Heppenheim wood-
lands were slowly reified into a series of jurisdictional rights that
existed independently of the people over whom they were exercised,
until by the eleventh century they were the basis of the monasterys
territorial lordship.

Thanks to the richness of the Lorsch archive in which this document

was transmitted, and that of the monastery of Fulda, it is possible
to build up a very detailed picture indeed of patterns of social action
in Carolingian middle Rhine and lower Main: around 5,000 records
of legal transactions are transmitted from the region, most concern-
ing the second half of the eighth century and the first decades of
the ninth. Ploughing through the Lorsch and Fulda evidence cer-
tainly is a pretty rebarbarative exercise, but the charters represent
total social occasions that were marked by public rituals, and so they
supply information of the utmost significance.
In the second half of the eighth century, the charters suggest the
existence of a hierarchy of venues for public meetings and legal
action. At the apex of this hierarchy were cities: the former Roman
civitates of Mainz and Worms, now episcopal seats where kings stayed
when they were in the region, plus the former Roman stronghold
of Ladenburg on the lower Neckar, an important ecclesiastical out-
post of Worms east of the Rhine which early medieval scribes like-
wise referred to as a civitas.15 The significance of the cities as the
places par excellence for public action is clearest at Mainz: through the
760s and 770s it is difficult to find a donation of land to Fulda
which was not made at Mainz. The historical links between Fulda
and Mainz through the person of St Boniface might partly explain

15
See Staab, Gesellschaft, pp. 11836; H. Bttner, Zur Stadtentwicklung von
Worms im Frh- und Hochmittelalter, in: Aus Geschichte und Landeskunde: Festschrift
F. Steinbach (Bonn, 1960), pp. 389407; idem, Ladenburg am Neckar und das
Bistum Worms bis zum Ende des 12. Jahrhunderts, Archiv fr Hessische Geschichte
und Altertumskunde 28 (1963), pp. 8398; idem, Mainz im Mittelalter: Gestalten und
Probleme, in: his collected essays, ed. A. Gerlich, Mittelrhein und Hessen: Nachgelassene
Studien, (Stuttgart, 1989), pp. 150; L. Falck, Mainz im frhen und hohen Mittelalter
(Mitte 5. Jht. bis 1244) (Dsseldorf, 1972).
408

this prominence, as they undoubtedly do account for the fact that

Fuldas most important patrons in this period were the elite of the
Mainz area who expressed their devotion to Bonifaces memory by
giving land to the resting place of his body at Fulda. But there is
enough evidence from Fulda, Lorsch and Wissembourg16 for the
importance of Worms, Ladenburg, and, further south, Strasbourg,
to assure us that the Boniface connection simply made Mainzs cen-
trality especially visible, not exceptional in itself. The presence of
civic notaries, who used styles from the late Roman legal tradition
as transmitted through Merovingian Gaul, might even suggest that
the high profile of cities in the earliest charters owes something to
the continuation of ultimately Roman city-based administrative tra-
ditions, albeit in a piecemeal manner and a radically changed social
and political context.17
Certainly the Roman past was central to the definition of these
civitates above all through the Roman walls that were the main
marker, which differentiated them from the countryside. But these
settlements were far from being ghost towns, not least given their
natural domination of river routes along the Rhine and Main and
so patterns of inter-regional contact. The Roman heritage, indeed,
consolidated this domination, in that the Roman road system remained
central to land travel in the early Middle Ages. Ecclesiastical organ-
isation, centred on bishoprics located in these sites, in spite of some
disruption in the post-Roman period, also contributed to the local
centrality of the civitates. About them as functioning settlements, char-
ters and a little archaeology allow some suggestions. We can say
most about Mainz, which was an agglomeration of more or less self-
contained elite compounds, some secular, some ecclesiastical (the line
between the two blurred).18 Already in the seventh century its cen-
trality was such that the leading men of the surrounding area could
be collectively referred to as the Macanenses, the men of Mainz.19

16
A. Doll and K. Glckner, Traditiones Wizenburgenses: Die Urkunden des Klosters
Weissenburg, 661864 (Darmstadt, 1979).
17
Staab, Gesellschaft, pp. 13753, and cf. Innes, State and society, pp. 11118.
18
In addition to bibliography from the previous note see K. Weidemann, Die
Topographie von Mainz in der Rmerzeit und des frhen Mittelalters, Jahrbuch
des Rmisch-Germanisches Zentralmuseums Mainz 15 (1968), pp. 14699; E. Wamers, Die
frhmittelalterlichen Lesefunde aus der Lhrstrae (Baustelle Hilton II) in Mainz (Mainz, 1994).
19
J.M. Wallace-Hadrill ed., The fourth book of the chronicle of Fredegar with its contin-
uations (Oxford, 1960), IV:87, pp. 734.
 , 409

Those at the apex of the regional elite had town houses here: not
necessarily places where they resided the year round, but pieds-a-
terre where they stayed when in the city on business. When the
young Hraban Maur was given over to Fulda, his parents gave the
monks a stone town house in Mainz, reserving life-interests for them-
selves and Hraban. Hrabans father was often present at Mainz, wit-
nessing charters there for a whole range of kin and contacts with a
wide geographical range.20 What marked out Mainz and the other
civitates was a concentration of high status people: bishops, kings and
royal officials as well as members of local elites. Part and parcel of
this was a concentration of churches too, their dedications reflecting
the citys importance as a regional focus, and as a point of contact
between the region and the wider world. One grouping founded a
church dedicated to St Lambert of Lige, one of the classic politi-
cal saints of the late Merovingian period, at Mainz at some point
in the first decades of the eighth century, and by the ninth century
we also find a church dedicated to the Bavarian missionary, Emmer
am of Greensburg.21
It was presumably precisely this kind of centrality, plus the nat-
ural advantages of its location, that made Mainz so economically
important once exchange down the Rhine to the North Sea took
off in the seventh and eighth centuries. The archaeology, however,
strongly suggests that Mainz was a regional centre before it became
the economic valve linking the regions economy to the North Sea
via the Rhine. The earliest of the remarkable finds of craftwork and
coin from the city are of high status goods (predominantly ornate
womens brooches) from the very beginning of the seventh century,
and we know from written sources (Gregory of Tours, Venantius
Fortunatus, Jonas of Bobbio) that Mainz was the political, social and
cultural centre of the middle Rhine and Main valleys from at least

20
Townhouse: E.E. Stengel ed., Urkundenbuch des Klosters Fulda (2 vols., Marburg,
191358) [hereafter: UBF] charter no. 177. On this charter see F. Staab, Wann
wurde Hrabanus Maurus Mnch in Fulda? Beobachtungen zur Anteilnahme seiner
Familie an den Anfngen seiner Laufbahn, in: R. Kottje and H. Zimmermann
eds., Hrabanus Maurus: Lehrer, Abt und Bischof (Wiesbaden, 1982), pp. 75101 (the best
discussion of Hrabans family), and cf. M. De Jong, In Samuels Image: Child oblation
in the early medieval West (Leiden/New York/Kln, 1996), pp. 737. Hrabans father
Walaram witnessed no fewer than 41 charters dealing with property in the Mainz
area between 754 and 802: see UBF passim.
21
St Lamberts: CL 19667, 196972, 1974 and see Gockel, Knigshfe, pp. 23858.
Church dedications: Falck, Mainz, pp. 123.
410

the late sixth century. The earliest status symbols found at Mainz
are best seen as evidence for the concentration of moveable wealth
there, and may even suggest that the city functioned as an arena
for conspicuous display amongst the elite. In any case, these certainly
predate the creation of a regularised network of administered exchange
along the Rhine, associated with the foundation of emporia, notably
Dorestad near the Rhine delta, in the later seventh century a
development paralleled at Mainz with the demolition of part of the
Roman walls and the creation of an enterprise zone along the
Rhines banks circa 700, and so vividly illuminated by recent finds.
Mainz was exceptional here, as the centre for long-distance exchange
within the region: the situation at Worms, which seems to have been
an important centre within the local economy but not a supra-
regional trading centre on any scale before the ninth century, may
have been more normal.22 But the central role of the civitates within
the regional economy, particularly before the economic changes of
the late ninth century, serves only to underline their general impor-
tance within their localities.
None of this is to deny that the eighth-century elite was ruralised
in both their residences and interests. To stick with Hraban and his
family, it is clear that an estate complex at Hofheim, a few miles
down the Nibelungenstrasse towards Worms from Heppenheim, served
as their country seat, complete with a newly-founded proprietary
church dedicated to St Boniface that doubled as a family mausoleum.
Whilst Hofheim was clearly in a very real sense the place from which
this family ran their interests, it is striking that we never see them
exercising power from Hofheim direct: none of the fifty or so doc-
uments in which Hrabans family figured were actually transacted
there. Hofheim may well have been a place for meeting friends and
clients, doing deals and wielding influence, but public business was
formally conducted elsewhere, in full view of the collective gaze.
Thus even when they wished to conduct property dealings concerning
Hofheim itself, Hrabans family travelled to the appropriate civitas,
at Ladenburg, to do so in public.23

22
Mainzs economic success story deserves a fuller study, but the basic archae-
ological data is discussed and analysed by Wamers, Die frhmittelalterliche Lesefunde,
and on the regional economy see the important study of W. He, Geldwirtschaft
am Mittelrhein in karolingischer Zeit, Bltter fr deutsche Landesgeschichte 98 (1962),
pp. 2663.
23
UBF 38, and on Hofheim see also E.F.J. Dronke ed., Codex Diplomaticus Fuldensis
 , 411

This hierarchy of venues reflected and reinforced a social hierar-

chy. This is clear if we look at charter witnesses from Mainz. Counts
were often present more often than in transactions that took place
outside the city but this simply reflected the citys social and polit-
ical importance. There is no hint that these counts had any official
function in Mainz. Rather, the most striking point about the char-
ter witness lists is their domination by the great and good of the
region. The witness lists of charters redacted within Mainz tend to
be dominated by a small, compact and interrelated group, regard-
less of the identity of the donor and recipient of the transaction
recorded. Although or perhaps because they were dominated
by the local elite, cities were defined public spaces, for public in con-
temporary usage carried with it connotations of the visible, open,
manifest, that which was acknowledged in the eyes of the locality.
Because of their control of the levers of patronage, Hrabans kin
might dominate the locality, but this did not place them outside of
local traditions of collective action: they were powerful because they
could deliver results through public meetings. When, on occasion,
rural groups who had little contact with the city came to Mainz to
have a charter drawn up, the local elite still witnessed: there were
no formal barriers of access, precisely because this social hierarchy
presented itself as a natural order (as social hierarchies which are
deeply entrenched and largely unthreatened are wont to do).

The key to eighth-century society is the relative lack of differentiation

of social action, and of specialisation of social function. Mainz or
Worms, or Ladenburg was where you went for everything: to visit
powerful relatives, to buy and sell luxury goods, to petition an official,
to encounter the sacred, to see and be seen by the cream of regional
society. In other words, thanks to the all-round dominance of a local
elite systems of cultural, political, social and economic power were
mutually reinforcing. These different spheres of activity were embed-
ded in, and thus determined by, patterns of elite sociability.24

(Kassel, 1850) [hereafter: CDF] charter no. 487, and Hraban, Carmina no. 86, ed.
E. Dmmler, MGH Poetae Latinae aevi Karolini 2 (Hannover, 1881), p. 238, and
Gockel, Knigshfe, pp. 412 n. 42.
24
Cf K. Polanyi, The economy as instituted process, in: idem et al. eds., Trade
and market in early empires: Economics in theory and practice (New York, 1957), pp. 24357,
for a classic discussion of this phenomenon in the context of economic systems.
412

The nature of the charter evidence from the middle Rhine, and
particularly the unique position of Mainz within it, may accentuate
the position of cities in the region. Certainly we should not forget
that our surviving charter traditions relate to a relatively late, Carol-
ingian phase of monastic foundation, and thus give a very particu-
lar perspective on social change. Already in the seventh century we
find the first rural monastic foundations in the area Tholey in
the Hunsruck and Wissembourg on the Saar and in the first half
of the eighth century the wave of monastic foundations supported
by the local elite such as Hornbach on the Blies, Amorbach on the
Main grows. We would expect, thinking about what we know of
seventh-century developments further west, in both Neustria and cen-
tral Austrasia, that such foundations would act as rural focal points
for elite families in the seventh century. Certainly there are hints at
the familial role of rural monastic foundations in our region in the
late Merovingian period, and the underlying dynamic in the chang-
ing geography of power was similar (although the eighth-century evi-
dence also suggests some genuine differences between regions, with
rural family monasteries coming relatively late to the Rhine valley).
It is clear that the scale and scope of the monasticism offered by
the foundations of the mid-eighth century, above all by Lorsch and
Fulda, marked the advent of something new. Lorsch was able to
swallow up earlier family foundations, such as Baumerlenbach and
Roden in the Odenwald, and Fulda likewise received some gifts of
earlier foundations. Whereas the rationale of these smaller houses
remained resolutely familial, reminiscent of those seventh-century
foundations of the Neustrian and Austrasian heartlands, monasteries
like Lorsch and Fulda offered spiritual patronage on a supra-familial
scale.25 The emergence of Lorsch and Fulda as providers of spiri-
tual patronage in the second half of the eighth century went hand
in hand with their acquisition of titles to land, and therefore social
muscle, to an unprecedented extent, in a region which was only just
being fully integrated into the Frankish kingdom. The extraordinary
scale of monasticism east of the Rhine in the later eighth century
must be explained by the conjunction of political and religious cen-
tralisation, as direct rule was imposed by the Carolingians on a region
which had been under only indirect Frankish overlordship under the

25
For seventh-century monasticism and familial strategy see Rgine Le Jans
paper in this volume; a comparison between her discussion and mine must make
clear the shift in emphasis between the late Merovingian and Carolingian periods.
 , 413

Merovingians, whilst the work of Anglo-Saxon and other missionar-

ies bound the regions religious life to the Frankish church. Local
elites bought into the monastic foundations of these missionaries and
their Frankish allies to cement their place in a new, supra-regional,
network of spiritual and political patronage.26
The surviving charters present a dramatic as we shall see, an
over-dramatic picture of the implications of this process for local
patterns of power. It is particularly evident in the Lorsch material.
The monastery of Lorsch was founded in 764, on what had been
an aristocratic complex not unlike that of Hrabans family at Hofheim.
The first series of donations of land to the new monastery, in 766,
open a window onto a complex settlement hierarchy in the sur-
rounding area, with Ladenburg at its apex and larger rural settle-
ments like Mannheim and Weinheim, both on roads and administrative
boundaries, operating as more local centres. But the emergence of a
monastic centre at Lorsch transformed this pattern. By the 770s when
land in the locality was given to the monks, the donors visited Lorsch
itself to have a charter drawn up, usually by a Lorsch monk.27 A
similar pattern emerges in the Fulda donations. The earliest dona-
tions tend to be made at Mainz, but in the 780s and 790s we begin
to have gifts of middle Rhenish land to Fulda being written up at
Fulda, where the donors, their relatives, clients and friends travelled.
A neat example is provided by charters from 788, 791 and 801 asso-
ciated with the young Hraban, a child oblate at Fulda: here his kin
and their closest contacts travelled to Fulda and made pious gifts of
interests in Mainz.28 Those who sought the patronage of the monks
of Gorze, near Metz, began to make journeys of a similar length to
transact their gifts at this date too.29 These were not just family get-
togethers; they mark the emergence of new patterns of sociability.
That patrons of monks were increasingly prepared to travel to
monastic complexes to make their pious gifts is hardly surprising,
but what is telling is that monastic complexes were increasingly seen

26
On patronage of Lorsch see M. Innes, Kings, monks and patrons: Political
identity at the abbey of Lorsch, in: R. Le Jan ed., La royaut et les lites dans lEurope
Carolingienne (Lille, 1998), pp. 30124, expanded in Innes, State and society, pp. 1350.
27
See CL, and on Lorschs estates see F. Knpp ed., Die Reichsabtei Lorsch, 2 vols.
(Darmstadt, 19747), and D. Neundrfer, Studien zur ltesten Geschichte des Kloster Lorsch
(Berlin, 1920).
28
UBF 177, 178, 219, on which see Staab, Wann wurde Hrabanus Maurus
Mnche im Fulda?, and De Jong, In Samuels Image, pp. 737.
29
A. dHerbomez ed., Cartulaire de labbaye de Gorze (Paris, 1902), nos. 70, 71.
414

as legitimate stages for such property transactions, witnessed by friends

and neighbours. Such journeys never wholly dominate the record
going to Gorze or Fulda from the middle Rhine valley was not a
trip undertaken lightly and the majority of gifts to Fulda and, for
that matter, Gorze, continued to be written up at settlements local
to the donors. Nonetheless, the temporal patterning of such journeys
is striking, given the way in which the earliest donations had been
made not at the monasteries themselves, but at established places for
legal action. Evidence from other great Carolingian monasteries
Echternach and Gorze in the Moselle valley, Wissembourg in the
Saarland suggests that the Lorsch and Fulda evidence may give
a particularly clear view of a more widespread phenomenon.
The increasing prominence of monasteries as social centres whilst
obviously overstated by the surviving evidence, which, after all, was
transmitted through monastic archives was linked to real processes
of institutional development and social change. After all, the thou-
sands of Lorsch and Fulda charters recorded the transfer of prop-
erty rights to the new monasteries which became landowners on an
unprecedented scale. And this change in patterns of land tenure was
not just quantitative. The acquisition of rights over land by monas-
teries changed the rules of landownership. Monastic control of land
created continuous, indeed more or less immutable, bonds between
monasteries and localities where they held land, and thus between
monasteries and the kin-groups which defined themselves in rela-
tionship to that land. Monastic networks were different from earlier
systems of patronage because they were embedded in rights over
land, rights which were not redistributed through inheritance, or as
needs waxed and waned in the course of the familial cycle, but which
were constants. This transfer of land funded monasticism on a new
scale, with hundreds of monks inhabiting large, stone-built complexes.
Rebuilding at Lorsch in 767774 necessitated a shift of location,
from the islet in the Weschnitz at the heart of a high status secu-
lar site to a natural mound above the river with more room for
expansion; at Fulda, building campaigns culminated in the career of
Abbot Ratgar in the 810s. These new-style monastic complexes were
doubtless much more impressive, much more inviting, and better
able to house a group of patrons, than their predecessors.30

30
For a stimulating discussion of the development of ecclesiastical architecture
as motivated by competition and imitation see W. Jacobsen, Was is die karolingi-
 , 415

The new prominence of monastic complexes should, then, be taken

seriously. It marked a clear departure from the characteristic inher-
ited patterns of social action, with their emphasis on the public and
collective. Gifts to the monks took on a wholly different aura from
public business, charters placed on the altar, in the church, before
a restricted and very special group, with the legal action taking place
in an enclosed and controlled space. Monasteries were social cen-
tres of a very different type to cities. They were not defined by their
visibility to the locality, nor even by the concentration of the interests
of the local great and good. They were by definition sacred places,
articulating the hierarchical relationship between God and this world,
in which power descended from above. The need to reconcile this
with their centrality to patterns of landholding and allegiance led to
a complex and ambivalent set of practices: the establishment of a
firm division between the claustrum, cut off from the secular world,
and the rest of the complex; and the development of ideas about
immunity and sacred space.31 Charters very rarely described actions
which took place in monastic complexes as public, because the term
continued to carry implications of secular collective action which was
not appropriate to a monastic context.
Although monastic complexes were quite clearly differentiated from
open, public stages for the transaction of public business, they should
be taken seriously as centres of social action. After all, the visits
recorded in the charters are only the tip of an iceberg, and they
make clear that we are dealing with groups of kin, clients and allies
descending en masse, often to visit kin who were inmates of the
monastery. When we imagine such visits, we should not forget the
splendid hospitality which worried some Carolingian commentators
on the Benedictine Rule, and made guesthouses such an important
facet of Carolingian monasticism.32 When Hrabans father and his

sche Renaissance in der Baukunst?, Zeitschrift fr Kunstgeschichte 3 (1988), pp. 31347;

I would suggest that changing social functions of monasticism encouraged the
dynamic. Note that the new monastic complex at Lorsch played an important role
in the development of a standard Carolingian monastic plan, notably in the cen-
trality of the cloister. Cf. S. Rabe, Faith, art and politics at St-Riquier (Philadelphia,
1995).
31
See M. De Jong, Carolingian monasticism: the power of prayer, in:
R. McKitterick ed., The New Cambridge Medieval History II (Cambridge, 1995), pp.
62253 esp. pp. 63540 and B.H. Rosenwein, Negotiating space: power, restraint and
privileges of immunity in early medieval Europe (Chicago, 1998), esp. pp. 99115.
32
De Jong, Carolingian monasticism, pp. 633, 6378.
416

friends, family and followers visited Fulda in 791, for example, the
charter recording their gift even drops the characteristic honorific
used to describe the abbot as most venerable, and calls Baugolf
most jocund; in the 810s the monks of Fulda complained about the
raucous secularity of Abbot Ratgars lay familiars.33 Sustained work
on the chronological patterning of such visits as revealed in the char-
ters, their relation to patronal Saints days, and to seasonal festivals
such as Easter, could shed real light on the new patterns of socia-
bility which emerged, and on the functioning of monastic networks.
Nor should we forget that monasteries were centres of production,
their landed muscle and dedication to functional specialisation lead-
ing to the foundation of monastic workshops. This also made monas-
teries centres of redistribution, as the swords, cloaks and horses given
out by monks to their benefactors in a handful of charters suggest.34
When we think of monasteries as social centres we should imagine
laymen hunting in the abbots wood, feasting on his game and enjoy-
ing the wine which Lorsch and Fulda specialised in producing.
For all the significance of the emergence of rural monasteries as
places of power, they did not carry all before them, but rather slot-
ted into, and in the process transformed, existing social patterns. For
one thing, even in the charters it is essentially only one type of trans-
action pious gifts of land which frequently took place in monas-
tic complexes. Sales and exchanges of land, in contrast, continued
to take place at the kind of important rural centres that they had
done before the foundation of Lorsch and Fulda, even when they
involved the monks. Charter formulae, indeed, clearly recognised a
distinction between sales and exchanges on the one hand, and pious
gifts on the other, as types of social action: the Saint was the legal
actor in the latter, whereas it was merely the abbot in the former.35

33
UBF 178; Supplex Libellus cc. 3, 15, ed. J. Semmler, in: K. Hallinger ed., CCM
I (Siegburg, 1963), pp. 31927.
34
F. Schwind, Zu karolingerzeitlichen Klstern als Wirtschaftsorganismen und
Sttten handwerklicher Ttigkeit, in: L. Fenske e.a. eds., Institutionen, Gesellschaft und
Kultur im Mittelalter: Festschrift J. Fleckenstein (Sigmaringen, 1984), pp. 10123, esp.
p. 115 on the charter evidence; from our region e.g. CL 508, CDF 336, CDF 471.
35
The evidence in CL, CDF and UBF is legion: see e.g. Glckners chronolog-
ical register to the Lorsch material in vol. 1 of CL, where it emerges particularly
clearly. This is not to deny that there are problems is classifying the type of reci-
procal exchange encouraged by the monasteries, and that it might be possible to
see some charters as gifts as, in content, being hidden sales and so on: but it is
clearly important that contemporaries attempted to distinguish gifts from sales, and
 , 417

The reflection of this in the choice of stages surely reflects a wider

perception of a distinction between transactions involving formal
entry into the spiritual patronage of the Saint, and more secular
property deals which happened to involve the monastery as a land-
owner. That is, whilst pious gifts and related types of sociability
which primed networks of saintly patronage networks were separated
from the public world of collective secular action, that world itself
continued.

The emergence of rural monasteries was not simply a spiritual ini-

tiative, but also a political strategy on the part of local elites. Lorsch,
for example, was founded on the site of the rural residence of one
of the most powerful families in the region, and on its foundation
was given to Chrodegang of Metz, a kinsman of the family which
owned the Lorsch estate, who installed his brother as the first abbot.36
Through Chrodegang, Lorsch was part of a web of monasteries
which spread from the Meuse to the Rhine and was centred on the
bishopric of Metz. Lorsch was situated only a kilometre or two from
the place with which we began, Heppenheim. Control of Heppenheim,
a fiscal estate held through the eighth century by members of one
local landowning family, was the key to local political dominance.
Lorsch was initially not seen as a centre of settlement, but as an
outlying site within the area dependant of Heppenheim: in the par-
lance of these scribes it was a locus which lay on the edge of the
marca attached to the villa of Heppenheim, between Heppenheim and
the neighbouring villa, Brstadt. Chrodegang held the important fiscal
church at Heppenheim before Lorschs foundation, and so we could
see his interest in Lorsch as a continuation of these earlier interests.
Fiscal churches at this date in this area served as centres for the col-
lection of royal dues.37 So the initial foundation was a very real bid

that this affected not only charter formulae but also the places at which transac-
tions took place.
36
What follows expands upon Innes, Kings, monks and patrons and has
benefited in particular from discussion with Marios Costambeys. For the early his-
tory of the Lorsch estate cf. H.-P. Wehlt, Reichsabtei und Knig. Dargestellt am Beispiel
der Abtei Lorsch mit Ausblicken auf Hersfeld, Stablo und Fulda, Verffentlichungen der
Max-Planck Instituts fr Geschichte 28 (Gttingen, 1968), pp. 1625 (with a more
rigorously fiscal interpretation than mine).
37
Chrodegang and Heppenheim: CL 429. Fiscal churches and royal dues: see
the two closely related confirmations, P. Kehr ed., Die Urkunden Arnulfs, MGH
Diplomata regum Germaniae ex stripe Karolinorum 3 (Berlin, 1940) nos. 67 and
418

to institutionalise local political power, led by Chrodegang in alliance

with his kinsmen who supplied the land at Lorsch, and the family
which controlled Heppenheim. When Chrodegang had Roman relics
translated to the new monastery at Lorsch in 765, Count Cancor,
Lorschs owner, and Count Warin, who controlled the Heppenheim
estate, led the watching crowds.38
Charlemagnes 773 gift of Heppenheim to Lorsch was therefore
in part a recognition that the two neighbouring sites formed one
complex of power. But Charlemagnes gift only came after he had
successfully gained a hold of Lorsch, using a family dispute between
the heirs of Chrodegang and those of Lorschs founders as an oppor-
tunity to intervene, and establishing his lordship over Lorsch; the
lavish gift of Heppenheim only served to firm his control. As Josef
Semmler has shown, events at Lorsch are paralleled right across the
newly acquired regions of Charlemagnes Empire in the last decades
of the eighth century: east of the Rhine and south of the Loire
Charlemagne pursued a systematic policy of imposing himself over
and above local elites by establishing royal lordship over key rural
monasteries.39 Part and parcel of this process was the dismantling of
the regional monastic constellations like Chrodegangs; at a similar
date, the foundations associated with Boniface, Fulda included, were
disentangled. Royal lordship did not dislodge the networks of power
which emanated from rural monasteries, but bound them into a sys-
tem of royal patronage which prevented their becoming the nodal
points of freestanding political configurations.40

69, based on charters going back to Pippin, confirming the bishop of Wrzburgs
possession of a number of these churches and the appended rights.
38
CL c. 6 (this section of the cartulary-chronicle based on a lost Carolingian
account of Nazarius miracles).
39
J. Semmler, Die Geschichte der Abtei Lorsch von der Grndung bis zum
Ende der Salierzeit, 764 bis 1125, in: Knpp ed., Die Reichsabtei Lorsch I, pp. 75173
at pp. 7880.
40
For events at Lorsch and Fulda as indicative of the impact of Carolingian
reform in the localities, and particularly the relationship between monasteries and
the episcopal hierarchy, see Innes, Kings, monks and patrons, pp. 31011 (a link
given further confirmation by Charlemagnes gift of a very rare total immunity to
the church of Metz at precisely the juncture at which the standing of Lorsch was
clarified: Mhlbacher ed., Die Urkunden Pippins, Karlmanns und Karls des Groen Charle-
magne charter 66). Rosenweins discussion of Chrodegangs role here (Negotiating
space, pp. 99115) is crucial and its implications for our understanding of Carolingian
reform need pursuing.
 , 419

And by establishing his lordship over the monastery, Charlemagne

was also able to establish direct control over the key estates of
Heppenheim and Lorsch. Before the establishment of royal lordship
of Lorsch, the precise tenurial status of Heppenheim and Lorsch was
ambivalent. Heppenheim was technically public land, but was passed
on within one family, and even in the ninth century, after the
Heppenheim mark as a whole had been granted to the Lorsch monks,
Warins son retained possession, and claimed ownership, of land
there; Lorsch likewise was described as public (it was a locus publi-
cus), but was also seen as being owned outright by the family of
Cancor.41 So by placing the lot soundly in the hands of monks under
monastic lordship, Charlemagne was disembedding the families who
had enjoyed a de facto monopoly on these public lands, and plac-
ing these key resources under direct royal control.

These political changes were compounded, in the middle Rhine, by

geopolitical change. Until the middle of the eighth century, rulers vis-
ited the region which was effectively the frontier of direct Frankish
power only intermittently. The increased eastward scope of Frankish
politics under the Carolingians brought kings to the Rhine more fre-
quently, until by the 790s it was established as one of the key areas
of royal residence within the Empire as a whole. Initially, when kings
were present in the middle Rhine they resided in urban palaces at
Mainz and Worms which had been used by their Merovingian prede-
cessors.42 Fire may have destroyed Charlemagnes favourite royal

41
Earlier historians such as Wehlt have tended to see the description of Lorsch
as public as indicative of fiscal interest, but it is difficult to make such an inter-
pretation stick, and it may make much more sense to see such labels as evidence
for the ambivalence surrounding notionally public resources whose effective con-
trol was embedded in one family. A direct parallel is CL 22, a court case of 819
trying to unravel the ultimate ownership of another piece of land attached to a
fiscal church which had been controlled by Warin and given to Lorsch, but which
Warins son likewise continued to control. Marios Costambeys forthcoming mono-
graph on the Farfa material, based on his unpublished 1998 Cambridge Ph.D. the-
sis, documents very similar strategies of monastic foundation and political control
in eighth-century central Italy.
42
C.-R. Brhl, Knigspfalz und Bischofstadt in frnkischer Zeit, Rheinische
Vierteljahrsbltter 23 (1958), pp. 161274 esp. pp. 22936 (Mainz) and 25967 (Worms);
420

residence, that at Worms, in 790, but in the following decades a

real shift occurs, with the development of new rural palaces.
Again, comparison suggests that, whilst we must acknowledge pecu-
liarities in the experience of this particular region, the pattern of
change is one that can be paralleled across Francia as a whole: one
thinks of the move from primarily urban patterns of royal itiner-
ancy, and the increasingly central role of suburban and rural palaces,
in Merovingian Gaul in the long seventh century, and the devel-
opments of the late eighth century in our area could be seen as a
relatively late and compressed articulation of the same dynamic.
Certainly, the foundation of Frankfurt and Ingelheim in our region
in the 790s directly mirrors the simultaneous building of Aachen, a
custom-built rural palace for the whole Empire.43 These were dedi-
cated royal centres, founded on exclusively royal sites cut off from
the hustle and bustle of everyday social exchange. This was the pro-
jection of the royal household on a grand scale, so as to encourage
the kings political constituency to look on themselves as members
of the royal familia. The palaces themselves were built in a monu-
mental style which articulated a specifically royal agenda and evoked
ancient models. They were seen as sacred places, but their charisma
was not a matter of the presence of bishops, monks or relics, but
the moral seriousness of Christian rulership: the frescoes around the
chapel and great hall at Ingelheim depicted the Carolingians as
champions of the faith, converting the heathen, and in doing so cap-
ping the history of salvation.44

and in general see P. Classen, Bemerkungen zur Pfalzenforschung am Mittelrhein,

in: Deutsche Knigspfalzen I, Verffentlichungen der Max-Planck Instituts fr Geschichte
11/i (Gttingen, 1963), pp. 7596. This is not to deny that rural royal residences
had an important role to play as early as the sixth century; but the overall pattern
of residence, and the function of different kinds of residences within the itinerary,
does seem to me to change quite markedly in the seventh and eighth centuries.
43
For this crucial period see D. Bullough, Aula Renovata: the Carolingian court
before the Aachen palace, in: idem, Carolingian renewal: Sources and heritage (Manchester,
1991), pp. 12360. Ingelheim: P. Classen, Die Geschichte der Knigspfalz Ingelheim
bis zur Verpfndung an Kurpfalz 1375, in: J. Autenrieth ed., Ingelheim am Rhein:
Forschungen und Studien zur Geschichte Ingelheims (Ingelheim, 1964), pp. 87146 and for
archaeology W. Sage, Die Ausgrabungen in der Pfalz zu Ingelheim am Rhein,
196070, Francia 4 (1976), pp. 14160. Frankfurt: M. Schalles-Fischer, Pfalz und
Fiskus Frankfurt. Eine Untersuchung zur Verfassungsgeschichte des frnkisch-deutschen Knigtums,
Verffentlichungen der Max-Planck Instituts fr Geschichte 30 (Gttingen, 1969).
On Aachen see now Janet L. Nelsons chapter in this volume.
44
P. Rich, Les reprsentations du palais dans les textes littraires du Haut
Moyen Age, Francia 4 (1976), pp. 16171, and see Ermold, In honorem Hludovici Pii,
 , 421

The creation of a royal landscape in the middle Rhine and lower

Main valleys took place in three critical stages: first, in the 790s,
with the realisation that the increased scope of Frankish power east
of the Rhine made the region a key point in the expanded king-
dom; second, in the 820s as Louis the Pious increasingly cultivated
support east of the Rhine; and third, in the 850s with the crystalli-
sation of an east Frankish kingdom, and the resultant shift of empha-
sis from the line of the Rhine itself to the Main corridor, and so
from Ingelheim to Frankfurt. Geopolitical change, and above all
direct rule east of the Rhine in middle of eighth century, and emerg-
ence of a new kingdom whose centre of gravity lay east of Rhine
in course of the ninth, explains the timing of these shifts, but they
marked fundamental alterations in the exercise of political power.
But these political changes themselves led to socio-economic change
in the localities: after all, the new palace complexes and the great
and good who haunted them needed feeding. Fiscal land in the
region, before the late eighth century, seems to have been utilised
only indirectly, granted out to local families who could effectively do
as they pleased on the ground. The creation of a royal landscape
changed all this, with royal rights reorganised so as to be exploited
directly. These developments supply an important context for the
Heppenheim dispute with which we began, not least as the Oden-
wald was a royal forest devoted to royal hunting, and the 795 meet-
ing was above all concerned with property boundaries within the
Odenwald. This context also explains the intensive manorialisation
of royal land in the region, evident in the remarkable ninth-century
survey of it which survives.45
The new palaces of the decades around 800 were not isolated:
they were the prime sites in a consciously planned royal landscape,
which was multi-centred and predicated on movement between a
series of complementary centres. Smaller residences such as Trebur
or Brstadt were sites for less solemn or formal occasions than the
epic backdrop of the great palaces: hence when Louis the German
wished to emphasise the subordinate nature of the kingships he had

extract transl. in P. Godman, Poetry of the Carolingian Renaissance (Oxford, 1985), pp.
2505.
45
The polyptych is CL 36714, and it is the mainstay of Gockels study, Knigshfe,
which also has an important discussion of the management of forest rights in the
Odenwald.
422

granted his sons he had them judge cases at Brstadt, not Frankfurt.46
These smaller palaces were also conveniently sited for the pleasures
of the hunt in the royal forest: they were sites for the intimate rather
than the sacred face of kingship.47 Royal churches, and particularly
favoured monasteries, were also important. Already in the 770s the
Anglo-Saxon holy woman Leoba was implored to reside close to the
royal centres of Mainz and Worms.48 In the 790s St Albans at Mainz
emerged as the burial place of one of Charlemagnes wives and an
important political centre, coinciding with the emergence of nearby
Ingelheim as a royal residence.49 In the 820s the Roman relics
acquired by the courtier and royal advisor Einhard would not set-
tle until they found their way to Seligenstadt, a convenient distance
up the Main from the recently redeveloped palace at Frankfurt.50

46
Annales Fuldenses [hereafter AF], ed. F. Kurze, MGH SRG (Hannover, 1891);
transl. T. Reuter, The Annals of Fulda (Manchester, 1992), s.a. 873, ed. Kurze, p. 78.
Brstadt: Gockel, Knigshfe, pp. 258310. Trebur: M. Gockel, Die Bedeutung
Treburs als Pfalzort, Deutsche Knigspfalzen III, Verffentlichungen der Max-Planck
Instituts fr Geschichte 11/iii (Gttingen, 1980), pp. 86110.
47
On the political significance of the royal hunt see R. Hennebicque [Le Jan],
Espaces sauvages et chasses royales dans le Nord de la France, Revue du Nord 62
(1980), pp. 3557; on the social significance of hunting see J. Jarnut, Die frhmitte-
lalterliche Jagd unter rechts- und sozialgeschichtelichen Aspekten, Settimane 31
(1985), pp. 765808; on the relationship between hunting resources and palaces in
addition to Le Jan see K. Bosl, Pfalz und Frsten, and K. Hauck, Tiergrten im
Pfalzbereich, in Deutsche Knigspfalzen I, Verffentlichungen der Max-Planck Instituts
fr Geschichte 11/i (Gttingen, 1963), pp. 129, 3074. For the complementary
intimate and sacred faces of kingship see J.L. Nelson, The Lords anointed and the
peoples choice: Carolingian royal ritual, in: D. Cannadine and S. Price eds, Rituals
of royalty: Power and ceremonial in traditional societies (Cambridge, 1987), pp. 13780.
48
Rudolf of Fulda, Vita Leobae abbatissae Biscofesheimensis, ed. G. Waitz, MGH SS
15.1 (Hannover, 1887), c. 20, p. 130, and see J.L. Nelson, Women at the court
of Charlemagne: a case of monstrous regiment? in her The Frankish world, 750900
(London, 1996), pp. 22342 at p. 234.
49
See F. Staab, Die Knigin Fastrada, in: R. Berndt ed., Das Frankfurter Konzil
von 794: Kristallisationspunkt karolingische Kultur, 2 vols. (Mainz, 1997), vol. I, pp.
183217.
50
Einhard, Translatio et miracula sanctorum Marcellini et Petri, ed. G. Waitz, MGH
SS 15.1 (Hannover, 1887), pp. 23864. J. Fleckenstein, Einhard, seine Grndung
und sein Vermchtnis im Seligenstadt, in: K. Hauck ed., Das Einhardskreuz: Vortrge
und Studien der Munsteraner Diskussion zur arcus Einhardi (Gttingen, 1974), pp. 96121,
in a seminal and stimulating paper saw Einhard creating a new basis for his power
in his relics following his retirement from the court, but the proximity of Seligenstadt
to Frankfurt is particularly significant when placed alongside the evidence for
Einhards continued political involvement with the court in the 830s (see M. Bondois,
La translation de Saints Marcellin et Pierre. tude sur Einhard et sa vie politique de 827 834,
(Paris, 1907); the crucial evidence is in Einhards letters): I hope to return to the
issue of Einhards retirement and his political role in the 830s at length elsewhere.
 , 423

Foundations such as Lorsch and Fulda, though they predated the

creation of a royal landscape, were integrated into it in recognition
of their spiritual and political power: Lorsch was conveniently sited,
and with the increased importance of Frankfurt as principal seat of
the eastern kingdom from the middle of the ninth century it was
redeveloped as a mausoleum for the east Frankish royal line, a spe-
cial place of dynastic charisma; Fulda became a favourite Lenten
visiting place of kings coming from or going to Frankfurt at a sim-
ilar date.51 Even the cities of Mainz and Worms, the traditional cen-
tres of regional society, were a part of the system, with their own
special functions within it: royal assemblies for the regions along and
east of the Rhine were habitually held at Mainz and Worms through
the ninth century and beyond, the places of consensus within the
royal landscape.52
Political topography is not just a matter of places: space, and the
links between places, are also central in the functioning of a land-
scape. On a micro-level, motion took the form of ceremonies and
processions within individual complexes, which could transmit pow-
erful messages.53 But movement also took place between a series of
royal places within a given region, as well as around the kingdom
between different complexes of royal interests. About movement
between the different sites we are less than fully informed, but sail-
ings along the Rhine and Main which bore the hallmarks of a cer-
emonial progress may have played an important role.54 Franz Staab

51
On royal lordship at Fulda and Lorsch see Wehlt, Reichsabtei und Knig; on
royal redevelopment of Lorsch in particular see Innes, Kings, monks and patrons,
pp. 3123, and on Lenten prayer see R. Kottje, Knig Ludwig der Deutsche und
die Fastenzeit, in: H. Roman and J. Ratzinger eds., Mysterium der Gnade: Festschrift
J. Auer (Regensburg, 1975), pp. 30711, which reminds us just how much there is
to be done on the cultural and social aspects of royal itineraries. Frankfurt as prin-
cipal seat: Regino of Prm, Chronicon, ed. F. Kurze, MGH SRG (Hannover, 1890),
s.a. 876, p. 121.
52
Demonstrated for Arnulf s reign by E.-M. Eibl, Zur Stellung Bayern und
Rheinfranken im Reiche Arnulfs von Krnten, Jahrbuch fr Geschichte des Feudalisimus
8 (1984), pp. 73112.
53
See for example the processions and ceremonies accompanying Harald Klaks
baptism at Ingelheim described in Ermold, In honorem Hludovici, ed. E. Faral, Ermold
le Noir: Pome sur Louis le Pieux (Paris, 1964), or in Notker of St Gallens vivid evo-
cation of Easter ceremonies, Gesta Karoli, ed. H.F. Haefele, MGH SRG (Berlin,
1959), II:21.
54
Ceremonial sailings: examples in the various annals are numerous, see e.g. AF
s.a. 840, p. 31, s.a. 874, p. 83.
424

has delineated the infrastructure of roads and provisions which under-

pinned the royal iter, and along which royal servants and political
news travelled. Strategically-placed monasteries and royal villae were
the stepping stones on which the royal household paused and caught
its breath before moving on, and the rich polyptych evidence from
our region shows how the demands of the royal iter shaped the serv-
ices demanded from the local peasantry.55 The monastic cell of St
Goar, a dependence of the great royal abbey of Prm, was one of
these stepping stones, strategically placed downriver from Bingen in
the Rhine gorge: a written collection of the miracles associated with
the shrine give a flavour of the intermittent royal stopovers, with its
stories of royal goodwill and munificence.56 The kings progress from
one such site to another was marked by rituals of arrival which
underlined the ties between the different aspects of the royal progress.
Movement over space had an important temporal element too, deter-
mined by, and in turn defining, the political calendar, which was
itself shaped by the liturgical year and the seasons public meet-
ings in spring, prayer at Lent, regal celebration at Easter, positive
action in the summer, more exclusive counsel in the winter.

The Carolingians did not seek to inscribe royal power into the land-
scape of the entire Empire. The middle Rhine valley may have been
a royal heartland, but the political topography of neighbouring areas,
whose elites visited kings at Frankfurt or Ingelheim, looked very
different, as kings had to acknowledge on the rare occasions that
they entered them.57 Hence in 852, when Louis the German made
a grand tour through Saxony an event which was worthy of
extended comment and special explanation in the eyes of contem-

55
Staab, Gesellschaft, pp. 32106 and see S. Airlie, Bonds of power and bonds
of association in the court circle of Louis the Pious, in: P. Godman and R. Collins
eds., Charlemagnes heir: New perspectives on the reign of Louis the Pious (814 40) (Oxford,
1989), pp. 191204 esp. pp. 1956, for non-royal travel along the iter and the travel
of political news.
56
See Wandelbert of Prm, Miracula s. Goaris, ed. O. Holder-Egger, MGH SS
15:1 (Hannover, 1887), pp. 36173.
57
This phenomenon is best studied for the Ottonian period: see particularly
E. Mller-Mertens, Die Reichsstruktur im Spiegel der Herrschaftspraxis Ottos des Groen
(Berlin, 1980) and for the Anglophone, J.W. Bernhardt, Itinerant kingship and royal
monasteries in Germany, 9191056 (Cambridge, 1994). For Carolingian itineraries see
C.-R. Brhl, Fodrum, Gistum, Servitium Regis: Studien zu den wirtschaftlichen Grundlagen des
Knigtums im Frankenreich und in den frnkischen Nachfolgestaaten Deutschland, Frankreich und
Italien vom 6. bis zur Mitte 14. Jahrhunderts, 2 vols. (Kln and Graz, 1968).
 , 425

porary commentators he stayed in individual holdings (mansiones)

as the opportunity permitted, holding assemblies at Minden and
Erfurt, both the sites of bishoprics.58 It was the secular and ecclesi-
astical elite whose interests informed the political topography of this
region, and in the normal run of things they sought out kings in
the royal heartland of the middle Rhine: given the importance of
royal patronage and extensive links, the court was a crucial point of
reference even for those who ruled the roost with limited interfer-
ence in their home areas.
The elites of particular regions interacted with kings at particular
complexes of royal power, movements between court and locality
determined by an annual political rhythm, as is neatly shown by the
denouement of an east Frankish political crisis in 874. Early in the
year, Louis the Younger, the ambitious son of the elderly east Frankish
ruler Louis the German, held a secret meeting with certain of his
fathers counsellors at Seligenstadt. A sense of crisis may have been
in the air, as contemporary annals record a prodigiously bitter win-
ter and the ravaging of rebellious Slav neighbours, and Louis the
Younger was evidently seeking to use this to strengthen his position
and slip the tight leash on which his father kept him. But Louis the
German acted decisively to defuse the potential crisis, coming to
Frankfurt, just down the Main from Seligenstadt, around February
1, where he took counsel with his faithful men about the peace and
state of his kingdom. The contrast between the sons rebellious secret
meeting and the fathers legitimate and public counsel-taking is strik-
ing. Louis the German stayed in the region, continuing to disarm
potential discontents, breaking off secular affairs and spending Lent
in prayer, before travelling to Fulda to celebrate Easter and then
holding a general assembly at the royal villa of Trebur. Here we
clearly see the political importance of the royal iter, and the unwrit-
ten norms and expectations that surrounded it.
Louis conspicuous piety over Lent and Easter is a particularly
striking example of the power of place and its political significance.
Whilst in prayer in Lent Louis dreamt that he had seen his own
dead father, Louis the Pious, asking that prayer be said for the good
of his soul: Louis the German played the dutiful son and did so. In

58
AF s.a. 852, ed. Kurze, pp. 423, using Reuters translation. And see the com-
ments of T. Reuter, Germany in the early Middle Ages, 800 1056 (London, 1991), pp.
859.
426

the eyes of one contemporary commentator from this it may be

understood that although [Louis the Pious] had done many praise-
worthy things pleasing to God, nevertheless he allowed many things
against Gods law in his kingdom, notably in failing to observe the
warnings of the Archangel Gabriel which Abbot Einhard wrote down
in twelve chapters and offered to him.59 Louis the Germans dream
clearly echoed another dream, nearly fifty years earlier, in which the
Archangel Gabriel had appeared to a monk at Seligenstadt and urged
that reform was the response to political crisis, a message passed on
to Louis the Pious by Einhard, Seligenstadts founder.60 The attention
attracted by Seligenstadt surely relates to its high profile in the con-
temporary crisis involving tensions between a Carolingian father and
son, Louis the German and Louis the Younger. Louis the German,
in making the dream public and acting upon it, was demonstrating
his own filial piety, a demonstration which was a clear message to
his own rebellious and eponymous son. By linking himself to Einhards
pleas of reform to Louis the Pious he was demonstrating to his fol-
lowers that the proper response to political crisis was piety to pro-
priate God, as he himself practised, and so to disarm those counsellors
who had met in secret with his son, alarmed at the portents of cri-
sis. And place played an important role in this strategy, linking the
current crisis to one of half a century earlier, with a connection
which was self-evident to contemporary commentators.

Sites of royal power played a crucial role in political strategies. The

role of particular complexes of royal centres as focal points for
regional elites made them crucial stages for rulers negotiating with
those elites for political support. Indeed, from the very beginning
Carolingian palaces may have been designed to function as such.
Charlemagnes choice of Frankfurt as a palace site may have owed
much to the influence of the kin of Fastrada, his wife at the time
that the centre was developed, whose interests were focussed in the
lower Main region; Ingelheim, interestingly, lies close to several impor-

59
AF s.a. 874, ed. Kurze, pp. 813, transl. Reuter.
60
Einhard, Translatio, III:15, ed. Waitz, pp. 2523.
 , 427

tant residences of another of Charlemagnes wives, Hildegard, and

one of Hildegards descendants, Odo, began a stellar political career
as administrator of the palace at Ingelheim.61 Certainly by the reign
of Louis the Pious we can begin to trace specific elite groupings
enjoying particular ties to individual royal centres. The marriage of
Louis the Pious eldest son, Lothar I to the daughter of Hugh, count
of Tours, but also the most powerful aristocrat along the upper Rhine
in Alsace and Alemannia, was celebrated at Worms, and evidently
marked the consummation of a crucial political alliance between
Lothar and Hughs friends and clients.62 It is striking how many
episodes of the subsequent conflict between Lothar and his father
and brothers centre on the complex of royal palaces around Worms
and Lothars alliance with the aristocracy of the neighbouring regions,
the political constituency of this palace complex. Louis the Pious
temporary deposition in 833 came after his celebration of Easter at
Worms, as he headed south into Hughs backyard to a confronta-
tion with his sons at which Lothar was able to command aristocratic
loyalties. In both 833 and following his death in 840 Lothars ini-
tial actions on claiming Imperial status were to attempt to rule from
the royal heartland of the middle Rhine, winning over the support
of powerful backers from the surrounding regions who came to meet
him there. In 841, attempting to rally aristocratic support against
his brothers, he ordered his followers to meet him at Mainz and
Speyer, before marrying off his daughter to a prominent aristocrat
and celebrating the wedding at Worms.63 Note throughout these

61
On both Fastrada and Hildegard see now Franz Staabs excellent study of
their careers, Die Knigin Fastrada, in: Berndt ed., Das Frankfurter Konzil von 794,
I. The suggestion about the link between Fastradas interests and Frankfurt was
made by Janet L. Nelson; the links between Ingelheim and Hildegards kin both
fit the model, and can be documented continuing through the plentiful charter evi-
dence in UBF and CDF plus Ermolds testimony as to Odos official position at
court at Ingelheim.
62
Annales Regni Francorum, ed. F. Kurze, MGH SRM (Hannover, 1895), s.a. 821.
63
For Lothars activities in 834 and 841 see J.F. Bhmer with E. Mhlbacher,
Regesta Imperii I. Die Regesten des Kaiserreiches unter den Karolingern 751918 (2nd edn.,
Innsbruck, 1908). For the marriage of Lothars daughter, AF s.a. 841, ed. Kurze
p. 33, and for the type of political concessions wrung from Lothar by prospective
supporters see Innes, State and Society, pp. 21112. Note the long high profile stay
in the area by Charles the Bald and Louis the German once they had driven Lothar
out, which clearly was important in winning over aristocratic support: see above all
Nithard, Historiae, ed. P. Lauer (Paris, 1926), III:6, pp. 11012, and note Charles
marrying into the regional aristocracy at this point (Innes, State and society, p. 209).
428

episodes the significance of cities, traditional sites of royal assemblies,

as the points of negotiation between kings and elites: different kinds
of centres were suitable for different kinds of political activity.
Aristocrats were also well aware of the potential of these sites as
ways of attracting royal backing at crucial moments. In 855, for
example, the princes and leading men of the kingdom of Lothar I
were anxious to secure the succession of Lothar II on their rulers
death, and worried at the ambitions of the western king, Charles
the Bald, and the claims of another of Lothar Is son, Charles of
Provence: they therefore brought [Lothar II] to Louis, king of the
eastern Franks and his uncle, in Frankfurt [and] with Louis agree-
ment and support they agreed that he [Lothar II] should rule them.64
These actions had been prefaced by less public negotiations using
other sites within the political topography, reflected in the charter
evidence for the visit of the leading magnate of Lothars kingdom
to the royal abbey of Lorsch whilst his master lay on his deathbed.65
Fourteen years later, on Lothar IIs death, Charles the Bald invaded
Lotharingia and had himself crowned king at Metz whilst Louis the
German lay ill so much that the doctors despaired of saving his
life. But when in early 870 Louis recovered and travelled to Frankfurt,
where he arrived on the feast of the Purification of the Virgin Mary,
there he found many of the leading men of Lothars kingdom, who
had waited for him there a long time.66
The integration of royal sites into elite political strategies thus
emerges from the narrative sources: whilst high politics remained
court-centred and Knigsnhe was the trump card, the topography of
royal power remained a constant point of reference for local elites
who wished to exercise something more than local power. But, on
a level which barely finds mention in the narrative sources, local
networks of kin, friends and clients which underpinned elite strate-
gies at court needed constant maintenance. Overlapping and inter-
acting with the topography of royal power were a multiplicity of
local landscapes determined by kinship, land, and patronage of the
church; an itinerant lifestyle was necessary to hold on to scattered
estates and to guard diverse interests, whilst visits to court, to kin,

64
AF s.a. 855, ed. Kurze, p. 55, transl. Reuter.
65
CL 1522.
66
AF s.a. 870, ed. Kurze, p. 70, transl. Reuter.
 , 429

to patrons and to favoured monasteries ensured that contacts with

backers and brokers were continued.
The will of one woman from the circles which sought out Louis
the German at Frankfurt in 855 and 870 gives a priceless insight
into the topography of elite power. Ercanfrida, widow of an influential
and well-connected count of Trier, made her will in 853: it centred
on the grant of the estate which she had been given as a morning-
gift after her marriage to the monks of St Maximians, Trier, who
were to celebrate her and her husbands memory with an annual
feast, and a series of gifts to kin who were to distribute gifts for
Ercanfridas soul to a series of twenty-one churches down the Moselle
and Rhine. These favoured monasteries were the nodal points of
Ercanfridas world, like the kin who were to make gifts to them
points of contact with local networks of power; St Maximians was
the most powerful house in Trier and was thus central to the fam-
ilys power in the area that Ercanfridas husband had ruled. The
whole transaction was carried out before a series of aristocratic wit-
nesses, included her husbands erstwhile patron, in the public court
(mallus) of the Trier region, which was held in an open-air site beyond
the city itself. Ercanfridas will was carried out in the forum through
which local power was exercised, and its contents acknowledged the
network of family estates and favoured monastic houses through
which bonds with kin and friends were maintained. Its very exist-
ence makes Ercanfridas will unusual: her lack of direct heirs made
the fate of her legacy unsure, and it has been suggested that the will
itself is particularly sensitive to the need to ensure the support of
surviving kin and patrons for its contents. This pattern of giving was
probably only possible in this kind of exceptional context. It is pre-
cisely because of this very peculiar context that the document is able
to reveal so vividly the contacts which made up an aristocratic world,
and there is no reason to see these patterns as unusual (fig. 3).67

67
C. Wampach ed., Urkunden- und Quellenbuch zur Geschichte der altluxembourgischen
Territorien, vol. 1 (Luxembourg, 1935) no. 87, and on Ercanfrida see J.L. Nelson,
The wary widow, in: W. Davies and P. Fouracre eds., Property and power in the
early Middle Ages (Cambridge, 1995), pp. 82113; R. Le Jan, Famille et pouvoir dans le
monde Franc, VIII eX e sicles (Paris, 1995), pp. 6970; B. Kasten, Erbrechtliche
Verfgungen des 8. und 9. Jhts. Zugleich ein Beitrag zur Organisation und zur
Schriftlichkeit bei der Verwaltung adeliger Grundherrschaft am Beispiel des Grafen
Heccard aus Burgund, Zeitschrift der Savigny-Stiftung fr Rechtsgeschichte, Germanische
Abteilung 107 (1990), pp. 236338, esp. pp. 24084.
430

Fig. 3. Beneficiaries of Ercanfridas will (After Le Jan).

These complex spatial patterns may be usually invisible in the sources,

but the incandescent outbreaks of violent political conflict which
flamed through the last decade of the ninth century shed the occa-
sional brilliant light on the topography of elite power.68 Indeed, the

68
The fundamental discussion of this outbreak of conflict is G. Althoff, Amicitiae
und Pacta. Bndnis, Einung, Politik und Gebetsgedenken im beginnenden 10 Jht., Schriften
 , 431

sudden visibility marks a new stage in the relationship between the

topographies of royal and elite power. In our area, conflict broke
out with quite literally a vengeance in the 890s. In 892 Count
Megingoz, the dominant figure in the middle Rhine and Moselle
region, was murdered by his rival Alberich at the small monastery
of Rettel, on the Moselle just north of Trier; Alberich was in turn
struck down by Megingozs followers in 895 (again on a religious
festival in an ecclesiastical context, which is suggestive of the polit-
ical significance of visits to favoured churches on holy days).69 The
Carolingian ruler, Arnulf, had attempted to pacify the region and
assert his control of it by visiting monasteries and bishoprics for the
sake of prayer and making large gifts to the bishops in 893; in 894,
he had tried in vain to have his illegitimate son Zwentibald accepted
as king of Lotharingia, which for quarter of a century had been ruled
by east Frankish kings who had been happy to leave regional govern-
ment in the hands of Megingoz and his clients.70
The geography of these events highlights the complexity of the
spatial interaction between kings and elites. An urban centre and
the seat of the most powerful churchman in the region, Trier was
not used to sustained royal presence, and the lay-abbacies of a series
of royal monasteries which were the key to local dominance had
traditionally been granted to local leaders, most recently Megingoz
whose control of the region was centred on the monastery of St
Maximian at Trier. Arnulfs actions acknowledged this, entering the
Moselle as a peacemaker and the guest of local bishops in 893, and
subsequently working from the traditional places of negotiation between
east Frankish kings and Lotharingian aristocrats, the royal centres of
the middle Rhine: conciliation in 893 culminated in an assembly,
and lavish royal gift-giving, at Worms, whilst in 894 Arnulfs failed
attempt to have Zwentibald accepted as king in a meeting at Worms
was followed by more exclusive counsel at Lorsch, before a synod
at Trebur and another Worms assembly finally secured Lotharingian
acknowledgement of Zwentibald in 895.71

der MGH 37 (Hannover, 1992); the best political narrative is E. Hlawitschka,

Lothringen und das Reich an der Schwelle der deutschen Geschichte, Schriften der MGH 21
(Stuttgart, 1968); on the implications for local power see Innes, State and society, pp.
22241.
69
Regino, Chronicon, s.a. 892, 896, ed. Kurze, pp. 140, 144.
70
Arnulfs tour: AF s.a. 893, ed. Kurze p. 122, transl. Reuter p. 124; Regino
s.a. 893, 894, ed. Kurze, pp. 1412.
71
Trier: Regino s.a. 892, p. 140, and for events in 893, 894, 895, Regino ed.
Kurze, pp. 1413.
432

But once Zwentibalds kingship in Lotharingia was accepted, it

was necessary to re-establish a landscape of royal power within the
region. Aachen was traditionally the key to Lotharingia, and it had
been the favoured residence of Lothar and Lothar II.72 But Zwentibald
chose to rule from Trier. This may have made strategic sense, as
Zwentibald was worried by the support attracted by Rudolf of
Burgundy in Burgundy. It was also a political necessity, as the power
vacuum in the Moselle following Megingozs murder was the only
available point of entry into Lotharingian politics. The key to
Zwentibalds efforts to establish himself was his control of what had
been Megingozs honores, and particularly the lay-abbacy of St Maxi-
mians at Trier. But the very act of holding these honores brought
Zwentibald into conflict with Megingozs kin and clients: it is no
accident that the latter opened a feud by killing Megingozs assas-
sin immediately after Zwentibalds arrival in Lotharingia. Later accu-
sations that Zwentibald struck the archbishop of Trier on the head
with his own pastoral staff, and that he played hard and loose with
church land and lay honores, reflect the lack of a royal base which
necessitated intrusion in the key city of Trier, for a generation the
preserve of bishop and count.73 In 897 the leading aristocrats, stripped
by Zwentibald of their honores, marched on the king at Trier and
forced him to flee to Worms, where Arnulf acted as conciliator
between his son and the Lotharingian aristocracy.74 Zwentibalds sub-
sequent actions brought him into conflict with his erstwhile allies,
the faction which dominated the Meuse valley, as he sought to avoid
offending his former opponents on the Moselle. Whereas Carolingians
had previously played off these geographically distinct factions, each
of which sought control of the regnum as a whole through Knigsnhe,
Zwentibald as ruler of Lotharingia was more directly implicated in
regional politics and thus encountered direct local opposition. This
was not a move against kingship, or the Carolingian family, as the
respect shown to Arnulf at Worms demonstrates: it was a refusal to
allow the creation of a royal toehold in Lotharingia to disrupt the
networks of power which had been built up over the previous decades.
Conflict in Lotharingia culminated in Zwentibalds assassination by

72
Aachens role under the later Carolingians is discussed by L. Falkenstein, Otto
III und Aachen, MGH Studien und Texte 22 (Hannover, 1998), pp. 27.
73
AF s.a. 900, ed. Kurze, p. 134.
74
Regino, Chronicon s.a. 897, ed. Kurze, p. 145.
 , 433

his aristocratic enemies in 900. Even this was not an attempt to opt
out of Carolingian kingship: the Lotharingians loyalty was prob-
lematical precisely because they had access to alternative sources of
royal power in the west, whilst Zwentibald was finally abandoned
only when a more pliable Carolingian who was conveniently kin of
some of their number became available in the shape of Louis the
Child, Arnulfs successor in the east.75
Political conflict in Lotharingia in the 890s was played out through
competing topographies of power. As western and eastern Frankish
kingships had stabilised, the former heart of the Empire became the
scene of competing and shifting loyalties. The second half of the
ninth century had seen the emergence of a system of power based
on mediation between regional elites and the eastern court, and so
on dialogue between the regional networks emanating from Trier
and the royal sites of the Rhine valley. The effort to establish Zwenti-
bald in Lotharingia, and so have an adult Carolingian dominating
the geographical heart of the Empire, threatened this arrangement,
and the conflict between Zwentibald and the Lotharingian elite turned
precisely on issues of access and presence.
But in the accounts of the conflict we also witness an emerging
stress on the political significance of fortifications.76 In 906 contin-
ued conflict over dominance in Lotharingia still centred on domi-
nance of Trier through the key lay-abbacy of St Maximians, which
was violently invaded by one faction: the denouement of the conflict
came, however, as they marched into their rivals backyard, the Blies
valley, ravaging, and laid siege to a fortification owned by their
rivals.77 In a similar feud for regional dominance which raged through
the 890s and 900s in Thuringia, the fortified stronghold of one fac-
tion at Bamberg similarly emerges as the key to their power: from

75
The feud, indeed, was all along a matter of international relations, in that
Zwentibald, whose death began it, was a nepos of the western king, Odo (and note
that another Lotharingian nepos of Odos, Waltgar, was also involved in violent
conflict with his local opponents in 892: Regino, p. 140). The resolution of the cri-
sis, and the elevation of Zwentibald to the Lotharingian throne, similarly depended
on negotiations with other rulers (Louis of Provence at Lorsch in 894, Odo at
Worms in 893 and 895), whilst Zwentibalds fall followed alliances between the
Lotharingian aristocracy, Charles the Simple and Louis the Child.
76
See e.g. Regino, Chronicon, s.a. 898, ed. Kurze, pp. 1456, for the stronghold
of Zwentibalds opponent Reginar at Doveren; and s.a. 901, p. 149, for Zwentibalds
killer, Stephen, falling to his death in lurid circumstances from his residence.
77
Regino, Chronicon, s.a. 906, ed. Kurze, pp. 1501.
434

it, they could control the region and drive rivals beyond the Spessart.78
Unfortunately, the archaeological evidence for these fortifications is
still patchy, and its periodisation unclear, but defensible sites were
nothing new; some prominent fortified sites of the tenth century,
such as Chvremont, originated in the late Merovingian period.79
The precise balance between royal oversight and local control of
fortifications is difficult to unpick. Kings did attempt to regulate for-
tification, with Charles the Balds famous legislation for the destruc-
tion of all unlicensed castella in all probability the tip of an iceberg
of royal bargaining with aristocrats over when and where such seats
could be built: even in the tenth century, there is documentary evi-
dence suggesting royal oversight of castle-building in the Moselle.
Monasteries, indeed, may have played an important role in mediat-
ing the control of fortifications: the Moselle document concerns a
castellum built by three laymen but given by them to St Maximians
at Trier with royal blessing.80
Fortifications did not sweep away previous patterns of power. The
charters show a vigorous tradition of local collective action which
was not yet enveloped into lordship, and the central places of local
networks remained the cities such as Trier and Worms, and monas-
teries like St Maximians and Lorsch.81 If fortified residences remained
bolt-holes for times of troubles, not political centres in their own
right, their increased prominence in the atmosphere of political vio-
lence of the decades around 900 need not be seen as a revolution-
ary change. Nonetheless, in the context of a changing balance between
royal and elite power non-royal residences came to have a more vis-
ible place in the political landscape. In 930 Henry I, the first king
of a new, Saxon, dynasty, had to court the bishops, counts and other
nobles of the region by visiting them one by one, in person in their

78
See especially Regino s.a. 902, 903, 906, ed. Kurze, pp. 149, 149, 1513, for
the importance of the fortress at Bamberg for control of Thuringia.
79
On Chvrement see Falkenstein, Otto III und Aachen, pp. 428.
80
For Carolingian policy towards fortifications see C. Coulson, Fortresses and
social responsibility in late Carolingian France, Zeitschrift fr die Archologie des
Mittelalters 4 (1976), pp. 2937. The charter is H. Beyer et al. eds., Urkundenbuch zur
Geschichte der jetzt die Preuischen regierungsbezirke Coblenz und Trier bildenden mittelrheinischen
Territorien I (Koblenz, 1860), no. 167, discussed both by Coulson and by K.-U.
Jschke, Burgenbau and Landesverteidigung um 900, Vrtrge und Forschungen 16
(Sigmaringen, 1975). Jschke links the charter to Henry Is policy of building manned
fortifications against the Magyars; interestingly, at Hersfeld this was also imple-
mented via the agency of a royal monastery.
81
Innes, State and society, pp. 2313, 2401.
 , 435

homes (domi ) and creating bonds of friendship through gift-giving

and the sharing of banquets: the residences of the elite were the cen-
tral places in the political topography of the region.82
Henry I in 930 was in a particularly weak position, an outsider
who had to come to terms with deeply entrenched local families,
but by the reign of his son Otto I, kings were once again accentu-
ating their social distance from their erstwhile peers. But Ottonian
kingship rested on a different form of power-sharing between kings
and elites from that attempted by the Carolingians. Otto allowed
local landowners such as Conrad the Red to rule the region via the
cities of Worms and Trier and via key monasteries, and Conrads
possession of fortified residences at strategic points such as the
tower from which his son dominated Worms a generation later, on
the site of the old Carolingian royal palace underwrote his local
power.83 Royal patronage (in Conrads case marriage to Ottos daugh-
ter), political ritual and the royal itinerary thus took on a height-
ened significance as the integrative factor in the polity, binding
together the regional elites. And it was through the royal itinerary
that the relationship between regnal and regional power was nego-
tiated. In the great revolt of 9534, Otto lost the support of Conrad,
who fell in with Ottos rebellious brother Liudolf. Otto found him-
self unable to celebrate Easter at Aachen, as he had intended, or at
Mainz, and was sent scurrying back to Saxony. The proper perform-
ance of political ritual at these sites was crucial in relating kingship
to the surrounding regions.84

82
Adalbert of Trier, continuation of Reginos Chronicle, ed. F. Kurze, MGH
SRG (Hannover, 1890), s.a. 931, p. 159, and see Althoff, Amicitiae und Pacta.
83
Innes, State and society, pp. 2359, 2435. The classic description of the Salians
fortifications allowing them to dominate Worms: Vita Burchardi, ed. G. Waitz, MGH
SS 4 (Hannover, 1841), cc. 69, pp. 8357. This vivid description needs handling
with care.
84
The events of 9534, and Widukinds celebrated account thereof, have been
much discussed. The classic analysis of the centrality of ritual to Ottonian rule is
K. Leyser, Rule and conflict in an early medieval society: Ottonian Saxony (Oxford, 1979).
For Ottonian itinerancy see above all Mller-Mertens, Reichsstruktur, and Bernhardt,
Royal monasteries and itinerant kingship; on Ottonian government and its differences from
its Carolingian predecessor see H. Keller, Grundlagen ottonischer Knigsherrschaft,
in: K. Schmid ed., Reich und Kirche vor dem Investiturstreit. Festschrift G. Tellenbach
(Sigmaringen, 1985), pp. 1734; idem, Zum Charakter der Staatlichkeit zwischen
karolingischer Reichsreform und hochmittelalterliche Herrschaftsausbau, Frhmittelal-
terliche Studien 32 (1989), pp. 24864; idem, Reichsorganisation, Herrschaftsformen
und Gesellschaftsstrukturen im Regnum Teutonicum, Settimane 38 (1990), pp.
15995.
436

This essay is not an attempt to advance any paradigmatic claims for

the region it discusses; indeed, by definition the richness of the evi-
dence makes it exceptional. Proper generalisation should proceed not
from granting normative status to one area and then cataloguing
deviation. Rather, we must acknowledge regional variety and then
seek to explain it. If we avoid the temptation of dismissing difference
and instead try to assess the reasons for it, we are in a position to
detect underlying structural characteristics and their dynamics, and
to identify the crucial variables which determined divergent regional
experiences. Clearly, geopolitical change the more easterly ori-
entation of the Carolingian kings when compared to their Merovingian
predecessors, and the eventual splitting of the Carolingian Empire
into an eastern and western kingdom with a central zone typified
by divided and shifting allegiances played a central role in the
developments traced above. But that does not negate their significance.
We are used to thinking of society in static, diagrammatic terms,
as a two-dimensional hierarchy. But in a world where power was
embedded in personal relationships, space and movement across space
were a fundamental element of social structure; patterns of mobil-
ity, movement and presence are central to the sociology of power.
Throughout the early Middle Ages power remained relatively unin-
stitutionalised, difficult to abstract from the small worlds of status,
honour and collective action which we saw so vividly in the Walinehoug
meeting. Exercising power at a distance was the fundamental prob-
lem faced by both elites and kings. Elite power strategies thus cen-
tred around the need to plug into as many of these small worlds as
possible, through the creation and maintenance of an extensive cir-
cuit of contacts and interests: the successful pursuit of power was
possible only through the careful manipulation of patterns of move-
ment and residence. Royal power, too, aimed at overcoming space,
through movement around and between a series of royal heartlands,
where complexes of rights and property sustained residences, and to
which the aristocracies of neighbouring regions came.
In the Carolingian period we see a dramatic increase in patterns
of social complexity, allowing a greater distance to emerge between
elites and the localities that they ruled, as more forms of specialised
social centres, through which both elites and kings could plug into
local circuits of power, were created. One driving force for political
 , 437

development was, indeed, a dialectic between elites and kings in

attempting to control the interface between regional and regnal power.
But kings and elites remained interdependent. And the highest ech-
elons of elite society the so-called Imperial aristocracy were
always in danger of becoming disembedded from the worlds whence
they sprang. If entry into this charmed circle whose Knigsnhe placed
them on a different political plane remained the goal of regional
elites, the price was a reliance upon royal patronage which made
for insecurity. The crisis of the Carolingian Empire was essentially
a crisis caused by increasing political complexity, as the social distance
between courts and regions increased, and court elites competed with
one another to access regional support. The outcome was a simpler
system in which regional elites were left more or less to their own
devices and kingship fulfilled an essentially integrative function.
This page intentionally left blank
 THE REGIA AND THE HRING BARBARIAN
PLACES OF POWER

Walter Pohl

Then to Hrothgar was granted glory in battle,

mastery of the field; so friends and kinsmen
gladly obeyed him, and his band increased
to a great company. It came to his mind
that he would command the construction
of a huge mead-hall, a house greater
than men on earth ever heard of,
and share the gifts God had bestowed on him
upon its floor with folk young and old . . .1

Imagining barbarian residences, we may picture the tall-gabled mead

halls in Beowulf: the laughter of heroes, harp-music, scops deliver-
ing poems, ceremonial gift-giving and a lady with a mead cup enter-
taining the lords retainers.2 In fact, sophisticated models of early
medieval social interaction and rituals of power have been based
upon the Beowulf poets vivid descriptions.3 But there are some
methodological problems involved. We have to keep in mind that
Scandinavian society as described in the Old English Beowulf is styl-
ized in a particular way.4 Whether cleric or layman, the author
describes places of power with a great deal of empathy and delight.
But perhaps his suggestive text tells us more about what, in his view,

1
Beowulf, vv. 6573; transl. Michael Alexander (Harmondsworth, 1973), p. 53.
2
E.g. Beowulf vv. 613ff., p. 70.
3
M. Enright, The lady with the mead cup (Dublin and Portland, 1996); J. Bazelmans,
Beyond power. Ceremonial exchanges in Beowulf in: F. Theuws ed., Rituals of
power from late Antiquity to the early Middle Ages, TRW 7 (Leiden/Boston/Kln, 2000),
pp. 31176.
4
F.C. Robinson, Beowulf , in: The Cambridge companion to Old English literature,
eds. M. Godden and M. Lapidge (Cambridge, 1991), pp. 14259; P. Wormald,
Anglo-Saxon society and literature, ibid., pp. 122, esp. pp. 9f.; R. Frank,
Germanic legend in Old English literature, ibid., pp. 88106, esp. pp. 97ff.; M.E.
Goldsmith, The mode and meaning of Beowulf (London, 1970); M.A. Parker, Beowulf and
Christianity (New York, 1987).
440

a barbarian residence should (or, in some cases, should not) be like

than about what really happened there. It is not unlikely that many
barbarian residences were wooden halls occasionally filled with drunken
warriors bragging about their heroic deeds. Therefore, general anthro-
pological models in many respects tend to conceptualize the obvious.
But the more specific (and therefore, interesting) they become, the
less can they be applied to barbarian places of power in general. It
is perfectly possible to study the cosmological significance of gift
exchange, the role of women in a warrior society, or the composition
and function of the kings retinue in Beowulf, and such studies may
sharpen our eyes for the problems at stake. However, their results
can hardly be generalized for the pre-Carolingian period when we
have no descriptions of barbarian places of power in the west at all.5

R G M R

There is, as it seems, only one general observation about barbarian

places of power that can be made on the basis of our written evi-
dence: From Augustus to Charlemagne, hardly any author bothered
to write about the residences of Germanic kings and dukes.6 This
silence even holds true for basic topography and names where
were the residences of the kings of the Thuringians, the Heruls, the
Gepids or the Lombards? Where did the reges of the Alamanni so
vividly described by Ammianus Marcellinus reside? Perhaps on some
of the many hillforts that have been excavated, for instance the
Runder Berg at Urach or the Zhringer Burgberg (and in some
cases, archaeology provides good indications), but Ammianus does
not care to tell us.7 Only a few scraps of information have been
transmitted in writing. In the first century A.D., the residence of

5
Role of women: Enright, The lady with the mead cup. Retinue: J. Bazelmans,
Conceptualising early Germanic political structure: a review of the use of the con-
cept of Gefolgschaft, in: N. Roymans and F. Theuws eds., Images of the past. Studies
on ancient societies in Northwestern Europe (Amsterdam, 1991), pp. 91130. Cf. W. Pohl,
Die Germanen (Mnchen, 2000), pp. 6972.
6
W. Pohl, Germania Herrschaftssitze stlich des Rheins und nrdlich der
Donau, in: J.M. Gurt and G. Ripoll eds., Sedes regiae (400800 D.C.), Reial Acadmia
de Bones Lletres de Barcelona (Barcelona, 2000), pp. 40423.
7
H. Steuer, Die Alamannen auf dem Zhringer Burgberg (Stuttgart, 1990); idem,
Herrschaft von der Hhe, Die Alamannen, Archologisches Landesmuseum Baden-
Wrttemberg, Ausstellungskatalog (Stuttgart, 1997), pp. 14963.
 REGIA HRING 441

Marbod, king of the Marcomanni, seems to have been quite extra-

ordinary. Tacitus recounts that when Marbod was dethroned by
Catualda, the latter inrumpit regiam castellumque iuxta situm.8 This means
that the residence, the regia, was next to the fortress, and not inside
it. Strabo, in his Geography, still highlights t to Marobdou bas-
leion, Marbods residence, and Ptolemy lists the toponym Marboudon.9
In the sixth century, Venantius Fortunatus alludes to the royal hall
of the Thurinigians that had been burnt down by the Franks, but
does not mention the name, the location or any details (although
Queen Radegund must have known it well).10 In 639, the Thuringian
duke Radulf was besieged in an unnamed wooden fortress at the
river Unstrut by King Sigibert II. But that was not, properly speak-
ing, a fortified residence Radulf had specifically constructed it
when he heard that the Frankish army was approaching.11 The fort
of Wogastisburc where Samo, king of the Vinedi, resisted Dagoberts
army in 631/32 is one of the rare places of power in the Merovingian
periphery of which we know the name.12 There is a little more in
Frankish sources about Frisian residences in the time of Radbod,
but the Frisians lived much closer to the Frankish heartlands, and
participated in the new urbanism at Dorestad, and the remains of
the old at Utrecht.13 Only Carolingian writers began to be interested
in places of power of the migration period, their distant past, spot-
ting, for instance, King Wachos residence in the land of the Beovinidi
(Bohemia), and a Harilungoburg near the Danube.14
This scarcity of information is not simply due to the lack of writ-
ten souces. We have extensive accounts of, for instance, the wars

8
Tacitus, Annales 2, 62.
9
Strabo, Geographica 7, 1, 3; Ptolemy, Opera 2, 14, and the map from Codex
lat. VF. 32, Biblioteca Nazionale di Napoli, reproduced in Claudii Ptolemei Cosmographia,
Tavole, ed. L. Pagani (Torriana, 1990), tav. V.
10
Venantius Fortunatus, Opera poetica, ed. F. Leo, MGH AA 4, 1 (Berlin, 1881),
Appendix 1, 5f.; cf. Appendix 3.
11
Fredegar, Chronica IV, 87: Radulfus haec cernens, castrum lignis monitum in quodam
montem super Unestrude fluvio in Toringia construens.
12
Fredegar, Chronica IV, 68; cf. W. Pohl, Die Awaren (Mnchen, 1988), p. 260.
13
Even before Radbod, Aldgisls palatium is mentioned in the Vita Wilfredi, c. 26.
Cf. I. Wood, The Merovingian Kingdoms 450751 (London and New York, 1994),
p. 269; p. 297 (Radbod in Utrecht and Dorestad).
14
Wachos residence: Historia Langobardorum codicis Gothani, c. 2, eds. L. Bethmann
and G. Waitz, MGH SRL (Hannover, 1878), p. 6. Harilungoburg: MGH DD.LD
8 (832 6). H. Wolfram, Salzburg, Bayern, sterreich. Die Conversio Bagoariorum et
Carantanorum und die Quellen ihrer Zeit, MIG Erg. Bd. 31 (Wien, 1995), p. 54.
442

fought by Arminius, leader of the Cherusci, against the Romans and

even against members of his own family, but we do not hear about
his residence.15 The Vita Severini contains a wealth of circumstan-
tial information about life along the Norican Danube in the second
half of the fifth century; we know that Severinus could easily reach
the residence of the kings of the Rugi across the Danube, but
Eugippius does not give any specific information about it, apart from
the legend that Queen Giso detained some barbarian goldsmiths,
aurefices barbaros, to make them produce royal prestige goods, pro fab-
ricandis regalibus ornamentis.16 Barbarian residences only become appar-
ent once they have been transferred to Roman cities: Frankish Cologne
or Tournai, Gepid Sirmium or Bavarian Regensburg, which Arbeo
of Freising, in the eighth century, calls metropolis huius gentis in arce,
the walled capital of this people, built in stone, defended by towers
and adorned by many fountains.17 In the forests of Germania, we
hear more about holy places: the sacred groves of the Marsi, Semnones
or Naharnavali in the classical period, or the Saxon Irminsul in the
eighth century.18 Obviously, whoever knew Roman and, later, Frankish
towns, palaces and roads, was not very impressed by centres of power
beyond the old frontiers.

T P : H A

There is one exception, however, and this is what this paper will
concentrate on: the places of power of steppe peoples. Byzantine
diplomats occasionally described such steppe residences. The eye-
witness report of life at the court of Attila around 450 written by
Priscus is the most famous of these descriptions, but there is also
Menanders account of a visit to the khagan of the Turks in the late
sixth century.19 Byzantine sources about the Avar wars in the last

15
E.g. Tacitus, Annales, cc. 5571; Pohl, Die Germanen, pp. 1416.
16
Eugippius, Vita Severini, c. 8; in general, see Eugippius und Severin. Der Autor, der
Text und der Heilige, eds. W. Pohl and M. Diesenberger (Wien, forthcoming).
17
Arbeo, Vita Haimhrammi, cc. 4; 6. Wolfram, Salzburg, Bayern, sterreich, p. 12.
18
Tacitus, Annales 1, 51; Tacitus, Germania, cc. 39, 40, 43; Annales regni Francorum
a. 772; Pohl, Die Germanen 83f.
19
Priscus, ed. R.C. Blockley, The fragmentary classicising historians of the later Roman
Empire: Eunapius, Olympiodorus, Priscus and Malchus (Liverpool, 1983), pp. 222378; or:
Fragmenta historiae Byzantinae, Fragmenta historicorum Graecorum 4, ed. K. Mller (Paris,
1851), pp. 69110; Menander, ed. R.C. Blockley, The history of Menander the Guardsman
 REGIA HRING 443

third of the sixth century, chiefly Menander and Theophylactus

Simocatta, detailed as they are about many aspects of wars and
negotiations, contain long rhetorical speeches by Roman ambassadors
and Avar khagans, but hardly any circumstantial evidence about the
setting of these meetings.20 A few details about the hring of the late
Avar khagans emerge from contemporary Frankish sources, and
Notker later gives an elaborate and exaggerated description.21 Steppe
rulers residences continued to inspire western phantasies, not only
in the Nibelungenlied, but also in the books of Plano Carpini or
William of Rubruck, at which I shall take a few glimpses for com-
parison, and many other travel reports, notably Marco Polos.22
Similar descriptions appear in Chinese sources, of which I will briefly
consider one of the Turkish court in the early seventh century by
the Buddhist monk Hsan-Tsang. Most of the texts have the advan-
tage of being (or being based on) eye-witness reports. The vivid
descriptions they offer can, however, be misleading; the method-
ological implications of their genre and the intentions of their authors,
and the problems of comparison they raise, will be addressed below.
The report by Priscus of his visit to Attilas court in 449/50 is
the most detailed description of a late antique barbarian place of
power.23 The king of the Huns had several residences, but chose to
receive the Byzantine ambassadors in the largest and most beautiful
one, situated in a big village:
A village like a very large city, in which we found wooden walls made
of smoothed planks. They were joined together to suggest solidity in
such a way that even by looking hard one could scarcely see the joints.
You might see dining rooms of great dimensions and colonnades laid
out with every form of decoration. The area of the courtyard was
enclosed by a circuit wall of high extent so that its size might show

(Liverpool, 1985); or: ed. K. Mller, Fragmenta historicorum Graecorum 4 (Paris, 1851),
pp. 20069.
20
Cf. W. Pohl, Konfliktverlauf und Konfliktbewltigung: Rmer und Barbaren
im frhen Mittelalter, Frhmittelalterliche Studien 26 (1992), pp. 165207.
21
Notker, Gesta Karoli magni imperatoris, ed. H.H. Haefele, MGH SRG, NS 12
(Berlin, 1959); overview of other sources: W. Pohl, Die Awaren, pp. 3068.
22
J. Larner, Marco Polo and the discovery of the world (New Haven and London,
1999).
23
W. Pohl, La sfida attilana: Dinamica di un potere barbaro, in: S. Blason
Scarel ed., Attila Flagellum Dei? (Rome, 1994), pp. 6991, and now in: W. Pohl,
Le origini etniche dellEuropa (Rome, 2000), pp. 24162; E.A. Thompson, The Huns,
2nd edn. (Oxford, 1996), pp. 10436; G. Wirth, Attila Das Hunnenreich und Europa
(Stuttgart/Berlin/Kln, 1999), pp. 7986.
444

that this was a royal palace. This was the seat of Attila (. . .); this was
the dwelling he preferred to the cities which he had captured.24
The palace was built in a slightly higher position than the rest of
the village and surrounded by a decorated fence with towers that
were built with an eye not to security but to elegance.25 It does not
seem that the residence had any defences at all. In this compound,
there were a number of wooden buildings, partly covered with carved
decoration and panelled walls, the floors covered with carpets. In
front of the building where Attila lived, the king received his sub-
jects to settle their disputes, as Priscus could observe: As I was stand-
ing in the midst of the whole throng (for I was known to Attilas
guards and followers, and no one hindered me), I saw a group of
persons advancing and heard murmuring and shouts around the
place, since Attila was coming out. He came out of the house swag-
gering and casting his eyes around. When he had come out, he
stood with Onegesius in front of the building, and many persons
who had disputes with one another stepped forward and received
his judgement. Then he re-entered the house and received the bar-
barian envoys who had come to him.26 In front of the house, Attila
was also greeted by singing girls dressed in white, and offered a wel-
come drink when he returned from a military expedition.
In the centre of Attilas compound, there was a large banquet hall
where Priscus repeatedly attended ceremonial banquets: All the seats
were arranged around the walls of the building on both sides. In
the very middle of the room Attila sat upon a couch. Behind him
was another couch, and behind that, steps led up to Attilas bed,
which was screened by fine linens and multicoloured ornamental
hangings like those which Greeks and Romans prepare for wed-
dings.27 Attila rested on a kline, a Roman bed; one of his sons had
the privilege of sitting on its edge. Unlike his guests, the king was
dressed in an exceedingly simple fashion, and ate from a wooden

24
Priscus, ed. Blockley 11, 3, p. 281 (= Jordanes, Getica, c. 34).
25
Priscus, ed. Blockley 11, 2, p. 265. O. Maenchen-Helfen, Die Welt der Hunnen
(Wien, Kln and Graz, 1979), p. 137, argues that Attilas wooden halls must have
been built by Gothic carpenters. But rather than simply the employment of foreign
labour the wooden buildings represented the process of acculturation of the Huns,
who had already lived for at least three generations in an environment where wood
was much easier to get by than in Central Asia.
26
Priscus, ed. Blockley 11, 2, p. 277.
27
Priscus, ed. Blockley 13, p. 285.
 REGIA HRING 445

plate, obviously to demonstrate that he did not need any precious

materials to demonstrate his superiority. The guests sat on the chairs
along the walls according to their honour and prestige. The right
side was considered more honourable, whereas the Roman envoys
had been placed along the left wall. This order of prestige was
enacted in an elaborate drinking ritual.
Other buildings belonged to Attilas wives. Hereka (Kreka) inhab-
ited a compound of her own: Inside the wall there was a large clus-
ter of buildings, some made of planks carved and fitted together for
ornamental effect, others from timbers which had been debarked
and planed straight. They were set on circular piles made of stones,
which began from the ground and rose to a moderate height.28 In
Herekas house, the floor was covered with woollen felt-rugs for
walking upon, the queen was reclining on a soft couch when Priscus
entered.29 The road to Attilas palace passed through the compound
that belonged to his chief advisor Onegesius: The buildings of
Onegesius were second only to those of the king in magnificence,
and they too had a circuit wall made of timbers but not embell-
ished with towers, as was Attilas. Not far from this wall was a bath
which Onegesius had built, fetching stones from Pannonia.30 It was
the only stone building in the whole village and had been constructed
by a captive Roman architect from Sirmium who then, to his dis-
may, had to serve Onegesius as a bath attendant.
We know much less about the residence of the Avar khagans that
must have been in the same region, most likely, in the plains between
the Danube and the Tisza.31 Byzantine sources give no information
about it. Frankish annals that report the conquest of the Avar kha-
ganate by the armies of Charlemagne only briefly mention the Hunorum
regia, quae hringus vocabatur.32 The Lorsch annals call it locum, ubi reges
Avarorum cum principibus suis sedere consueti erant, quem in nostra lingua hringe

28
Priscus, ed. Blockley 11, 2, p. 275.
29
Priscus, ed. Blockley 11, 2, p. 275.
30
Priscus, ed. Blockley 11, 2, p. 265.
31
The only hint for the location of Attilas main palace is the route that Priscus
embassy took: north from Naissus, then across the Danube somewhere east of
Belgrade, and then across some navigable rivers of which the last was the Tiphisas,
most probably, the Tisza (Priscus 11, 2, p. 261). Pippins army in 796 came from
Italy and had to cross the Danube to reach the residence of the khagan (Pohl, Die
Awaren, pp. 3068).
32
Annales qui dicitur Einhardi a. 796.
446

nominant.33 One redaction of the Royal Annals speaks of eorum regia,

quae, ut dictum est, hringus, a Langobardis autem campus vocatur.34 This evi-
dence has provoked some debate. The ring seems to indicate that
the Avar residence was round, the campus, that it was situated in a
plain and made the impression of a mobile settlement. The image
of the ring is enlarged into bizarre dimensions by Notker of St Gall,
who claimed to rely on a story he had heard as a child from Adelbert,
a veteran of the Avar wars.35 According to him, the country of the
Huns was enclosed by nine concentric circles, separated from each
other by a distance as from Zurich to Constance. Each of them was
20 feet wide and twenty feet high, built from timber and filled with
stones and mud. These images are probably derived from the Avar
defences, munitiones, built in several places in the western periphery
of the Avar empire which Charlmagnes army encountered in 791.
But nothing suggests that the khagans residence itself was fortified;
at least, no resistance was offered there in 795 and 796.36 Contemporary
sources and Notkers account agree in one point: that is the mar-
vellous treasure that many generations had heaped up at the Avar
court, and which Charlemagne distributed freely after getting hold
of it in 795/96. The hoard found at Nagyszentmikls, east of the
Tisza, almost exactly one thousand years later (in 1799) and now
kept in Viennas Kunsthistorisches Museum may be a part of the Avar
treasure that was hidden before the Franks could get hold of it.37

E C A

Huns and Avars in Europe lived in the periphery of the Eurasian

steppe and got accustomed to sedentary ways of life fairly soon.38
But on the other hand, they remained part of a cultural universe

33
Annales Laureshamenses a. 796.
34
Annales qui dicuntur Einhardi a. 796.
35
Notker, Gesta Karoli II, 1.
36
W. Pohl, Die Awaren, pp. 31220.
37
G. Lszl and I. Rcz, Der Goldschatz von Nagyszentmikls (Budapest, 1977) (with
the outdated opinion that this was a Hungarian treasure). A publication edited by
Falko Daim is in preparation.
38
W. Pohl, The role of steppe peoples in eastern and central Europe in the
first millennium A.D., in: P. Urbanczyk ed., Origins of Central Europe (Warsaw, 1997),
pp. 6578.
 REGIA HRING 447

that extended to the Inner Asian frontiers of China.39 Therefore, it

may be worth looking for comparison to reports from the courts of
medieval Inner Asian rulers. In August 569, the Byzantine ambassador
Zemarchos departed on a long journey to the residence of the Turkish
khagan Sizabulos (Istmi) in the Altai mountains.40 When Zemarchos
and his companions reached the place where Sizabulos was presently
staying in a valley of the so-called Golden Mountain having
arrived there, they were summoned and immediately came in Sizabulos
presence. He was in a tent, sitting upon a golden throne with two
wheels, which could be drawn when necessary by one horse. The
envoys offered their gifts and were greeted by the khagan. Then
they turned to feasting and spent the rest of the day enjoying lav-
ish entertainment in the same tent. It was furnished with silken hang-
ings dyed without skill in various colours. On the next day, they
were received in another hut, which was similarly decorated with
multicoloured silken hangings. In it stood statues of different shapes.
Sizabulos sat there on a couch (kline) made completely of gold. In
the middle of the building there were golden urns, water-sprinklers
and also golden pitchers. On the third day, they came to another
dwelling in which there were gilded golden pillars and a couch of
beaten gold which was supported by four golden peacocks. In front
of this dwelling were drawn up over a wide area wagons contain-
ing many silver objects, dishes and bowls, and a large number of
statues of animals, also of silver and in no way inferior to those
which we make; so wealthy is the ruler of the Turks.41 Sizabulos
presented the Byzantine envoys with a notable display of prestige
goods this is as good a description of a barbarian treasure, and
the way in which it was displayed in public, as we get in the early
middle ages.42 Then, the Byzantines were dismissed with presents;
Zemarchos received a female slave.

39
O. Lattimore, Inner Asian frontiers of China (New York, 1940).
40
For the context, W.-E. Scharlipp, Die frhen Trken in Zentralasien (Darmstadt,
1992), esp. pp. 1929; D. Sinor, The establishment and dissolution of the Trk
Empire, in: D. Sinor ed., The Cambridge History of Early Inner Asia (Cambridge, 1990),
pp. 285316; or in: idem, Studies in medieval inner Asia (Aldershot, 1997), n. I.
41
Menander, ed. Blockley, fr. 10, 3, pp. 11921. The Greek words for the
dwellings vary; sometimes Menander uses skhnh (tent), sometimes kalubh (hut).
42
See M. Hardt, Royal treasures and representation in the early middle ages, in:
W. Pohl and H. Reimitz eds., Strategies of distinction. The construction of ethnic communities,
300 800, TRW 2 (Leiden/Boston/Kln, 1998), pp. 25581; and his dissertation on
early medieval treasures (Marburg, 1999) which he is currently preparing for print.
448

Sixty years later, in 629/30, a traveller from a very different cul-

tural background arrived in Central Asia at the court of another
Turkic ruler. It was the buddhist monk Hsan-Tsang who, on his
way to India, encountered the khagan Tung shih-hu near the Issik-
Kl, north of the Tien-shan mountains.43 This was not far from
where Zemarchos had met Sizabulos; the region of the Issik-Kl
served as winter quarters, as that lake does not freeze over because
of its warm sources.44 The khagan inhabited a spacious tent adorned
with golden flowers whose splendour blinded the eyes. His officers
sat in front in two rows on long mats that had been spread for
them; all of them wore splendid costumes of embroidered silk. The
khagans body guard stood behind them. Although he was a bar-
barian ruler who lived in a felt tent, one could only regard him with
a feeling of admiration and respect. When Chinese envoys arrived,
the khagan offered them wine and drank with them. Then one
could see the increasingly merry guests trying to outdo each other
drinking, clinking their cups that they filled and emptied in turn,
while the music of the barbarians of the south and north, east and
west kept ringing out with its noisy chords. Quarters of sheep and
calves were heaped up in front of the guests, only Hsan-Tsang
received pure food according to his monastic diet.45 The khagans
respect for the monk was also expressed by the fact that he was
allowed to sit in an iron chair, whereas all other guests sat on mats
on the floor.
Another Chinese author, Sung-yn, had described the residence
of the Hephthalites in 519: The Hephthalites do not live in cities
surrounded by walls; they have their residence in a mobile camp.
Their dwellings are made of felt (. . .) The king has put up for him-
self a large felt tent of about forty square feet; the walls are made
of woollen rugs. He wears garments of embroidered silk; he sits on
a golden couch supported by four golden phoenixes. There follows
a description of the kings favourite wife, who wears a long silk robe
and has her hair set in an eight-foot long horn adorned with pre-
cious stones in five colours.46

43
Scharlipp, Die frhen Trken, pp. 28f.
44
For the source and its context, I have to rely on R. Grousset, Die Reise nach
Westen, oder wie Hsan-Tsang den Buddhismus nach China holte (Kln, 1986).
45
Grousset, Die Reise nach Westen, pp. 78f.
46
Grousset, Die Reise nach Westen, pp. 76f. (with n. 1).
 REGIA HRING 449

M : P C W R

Extensive reports by Europeans who visited Central Asian courts are

only preserved again at a much later date, from the thirteenth cen-
tury onwards.47 In the meantime, the world of the steppe had changed
to considerable extent.48 Therefore, thirteenth-century sources should
not be used to shed light on those aspects of early medieval steppe
rulers residences that remain obscure in the earlier reports ( just as
the Edda cannot be used to reconstruct ancient Germanic religion).49
However, those elements that appear both in early medieval and
thirteenth-century century texts may be taken as an indication that
they may have had a more general significance. Of course, this type
of comparison always risks projecting intertextual relations onto the
object under scrutiny. In our case, it is rather unlikely that western
friars who travelled to the Mongol court knew the Excerpta de lega-
tionibus, and unthinkable that they knew any of the Chinese sources.
It is not impossible, though, that some information from Byzantium
available to Greek diplomats had reached them indirectly. We know,
for instance, that William of Rubruck met envoys from the emperor
at Nikaia, Vatatzes, at the Mongol court. As our aim here is not to
analyse all the details of barbarian residences, but only to draw a
general picture of the way in which they were perceived by western
envoys, a cautious look far beyond the time limits of this volume
may be permissible.
In Iohannes de Plano Carpinis Historia Mongalorum, there is a
description of his visit to the court of the Mongol Khan Gyk in
the 13th century; the friar travelled there as papal envoy in 1245.50

47
Les prcurseurs de Marco Polo, ed. A. tSerstevens (Paris, 1959); J.-P. Roux, Les
explorateurs au Moyen Age (Paris, 1985); D. Sinor, Le Mongol vue par lOccident, in:
idem, Studies in Medieval Inner Asia, n. IX; Die Mongolen und ihr Weltreich, ed. A. Eggebrecht
(Mainz, s.a.), esp. M. Weiers, Westliche Boten und Reisende zu den Mongolen
im 13. und 14. Jahrhundert (pp. 18595); V. Reichert, Begegnungen mit China: Die
Entdeckung Ostasiens im Mittelalter (Sigmaringen, 1992); Larner, Marco Polo.
48
For instance, K.A. Wittfogel, China und die osteurasische Kavallerie-Revolution,
Ural-Altaische Jahrbcher 49 (1977), pp. 12140, interprets the rise of the Mongols in
the context of an eastern Eurasian cavalry revolution.
49
Cf. the review of J. Laszlovszky ed., Tender meat under the saddle: Customs of eat-
ing, drinking and hospitality among conquering Hungarians and nomadic people, Medium Aevum
Quotidianum, Sonderband 7 (Krems, 1998) by N. Berend in The Medieval Review
(online), February 2000.
50
Iohannes de Plano Carpini, Historia Mongolorum (esp. c. 9, 2935, pp. 317322);
see also ibid., L. Petech, Introduzione (pp. 146); E. Menest, Giovanni di Pian
450

Like some of the earlier travellers, he was impressed by the num-

ber of lavish tents of state. The tent used for meetings and banquets
could, according to Plano Carpinis reckoning, accommodate more
than 2000 people, was made of purple fabric and surrounded by
wooden panels on which various images were painted. In these pan-
els, there were two doors, one of which was exclusively reserved for
the khan. There was another tent called orda aurea, which rested on
gilded columns: Tentorium illud erat positum in columnis, que aureis laminis
erant tecte, et clavis aureis cum aliis lignis erant affixe et de baldachino erat
tectum superius et interius parietum, sed exterius alii erant panni. Here, Gyk
was sitting on a throne to receive homage, after which a banquet
started with drinking and eating that continued all day.51 A third
tent that the papal envoys were taken to later was the most mar-
vellous of all, totum de purpura ruffa. When they entered, they were
offered wine and beer, and had the opportunity to admire the splen-
did ebony throne:
Solariulum unum de tabulis erat alte preparatum, ubi tronus imperatoris erat posi-
tus. Tronus autem erat de ebore, mirabile sculptus. Ibi etiam erat aurum et lapi-
des pretiosi, si bene meminimus, et margarite. Et per gradus ascendebatur illud
quod rotundum erat parte posteriori. Banci etiam erant positi in circuitu sedis, ubi
domine sedebant in scamnis a parte sinistra, a dextris autem nemo sedebat superius,
sed duces sedebant in bancis inferius, in medio, et alii sedebant post eos.52
These three tents were the largest ones; but the tents of the khans
wives were also of considerable size and splendour. Arrangements at
the court of Batu that they passed on the way were slightly more
modest, but otherwise similar.53
In the years 125355, the friar William of Rubruck travelled to
the Mongol empire, and then wrote a report to Louis IX of France.54
He visited several khans on his journey eastward: first, Sartach/Sartaq
in his encampment three days journeys west of the Volga; then,
Batus mobile camp on the eastern bank of the Volga, and finally,

di Carpine: da compagno di Francesco a diplomatico presso i Tartari, pp. 4968;

Larner, Marco Polo, pp. 1823. For the Mongol context of his report, see C.M.
Halperin, Russia and the golden horde (Bloomington, 1987), esp. pp. 3335; D. Morgan,
The Mongols (Malden and Oxford, 1987), esp. pp. 13644.
51
Iohannes de Plano Carpini, c. 9, 32, p. 320.
52
Iohannes de Plano Carpini, c. 9, 35, p. 322.
53
Iohannes de Plano Carpini, c. 9, 17, p. 311f.
54
The mission of Friar William of Rubruck. His journey to the Great Khan Mngke, 125355,
ed. and transl. P. Jackson (London, 1990); cf. Larner, Marco Polo, pp. 2325.
 REGIA HRING 451

the winter residence of the great khan Mangu/Mngke, between the

Altai and the Tien-shan mountains. Batus camp was moved up and
down along the Volga following the rhythm of the seasons; William
was impressed by its size and the careful lay-out:
His own (= Batus) dwelling had the appearance of a large city stretch-
ing far out lengthways and with inhabitants scattered around in every
direction for a distance of three or four leagues. And just as the peo-
ple of Israel knew on which side of the tabernacle to pitch their tent,
so these people know on what side of the residence (curia) to station
themselves when they are unloading their dwellings. For this reason
the court is called in their language orda, meaning the middle, since
it is always situated in the midst of his men, except that nobody takes
up his station due south, this being the direction towards which the
doors of the residence open.55
The residences of the khans all looked more or less like Mngkes
residence, as described by William on the occasion of their first audi-
ence.56 At the entrance, there was a bench with some comos.57 The
envoys were told to sit on a bench. The interior of the dwelling
was completely covered in cloth of gold, and in a little hearth in
the middle there was a fire burning, made from twigs and the roots
of wormwood, which grows there to a considerable size, and also
from cattle-dung. He (the khan) was sitting on a couch, dressed in
a fur which was spotted and very glossy like sealskin (. . .) He then
asked us what we should like to drink: wine, or terracina (rice wine),
or caracomos (clear mares milk), or bal (honey mead).

What do these texts, from Priscus report to that by William of

Rubruck, tell us about places of power in the Steppe? In many
respects, the perceptions they relate are surprisingly similar. Almost
all of them centre around a large tent, or a wooden hall, in which
audiences, meetings and banquets took place. Authors (for instance
Menander) do not always distinguish clearly between tents and wooden
buildings; large audience-halls with wooden pillars and panels,

55
William of Rubruck, c. 19, 4, p. 131. One league is about 1500 paces.
56
William of Rubruck, c. 28, 14, p. 177.
57
As in Sartaqs residence: William of Rubruck, c. 15, 6, p. 117.
452

decorated with textiles, were often simply described as dwellings.58

Sometimes there were several tents used for banquets or receptions,
of different sizes and distinguished by colour (silver/gold, white/
purple). Although detailed descriptions are rare, we often get some
information on materials and decoration: wood or felt, purple and
gold, elaborate pillars, wood carvings, painted panels, golden flowers,
embroidered silken hangings, rugs or carpets. In the centre, often in
an elevated position, or sometimes on one side, there is a throne or
a royal couch, which, as a rule, is briefly sketched in our reports: a
simple wooden chair in Attilas tent,59 two couches in his banquet-
ing hall; a golden throne on two wheels and a couch of beaten gold
which was supported by four golden peacocks in Sizabulos tents; a
golden couch resting on four golden phoenixes at the Hephthalite
court; an ebony throne in Gyks residence. In front of the throne
or along the side walls, chairs or mats were arranged on which aris-
tocrats, ambassadors and leaders of subject peoples sat in a carefully
arranged order of precedence.60 This order of precedence may have
been underlined by drinking rituals, as described by Priscus at Attilas
banquet.61 The Secret History of the Mongols records a quarrel
between the Khans Gyk and Batu about which of them was
allowed to drink first at a ceremonial dinner.62 The ambassadors
reports from the Mongol court underline the symbolic significance
of the doorways: there was a taboo on touching the threshold; a
reserved door for Khan Gyk; and inside, a table with kumys next
to the entrance.
Social interaction at court figures prominently in most narratives.
A lot of emphasis is put on banquets that usually went on all day.
The day was spent drinking; drinks, most of them alcoholic, differed
widely: wine, beer, mead, kamon (according to Priscus, a drink made
from barley),63 and in Central Asia, also kumys (an alcoholic drink

58
See above, n. 41. Cf. M. Balzer, . . . et apostolicus repetit quoque castra suorum.
Vom Wohnen im Zelt im Mittelalter, Frhmittelalterliche Studien 26 (1992), pp. 20829.
59
Priscus, fr. 11, 2, p. 255.
60
For similar arrangements at the Byzantine court, see D. Smythe, Why do
barbarians stand around the emperor at diplomatic receptions?, in: J. Shepard and
S. Franklin eds., Byzantine diplomacy (Aldershot, 1992), pp. 30512.
61
In general, R. Bleichsteiner, Zeremonielle Trinksitten und Raumordnung bei
den turko-mongolischen Nomaden, Archiv fr Vlkerkunde 6/7 (1951/52), pp. 181208.
62
Ed. W. Heissig, Dschingis-Khan ein Weltreich zu Pferde. Das Buch vom Ursprung
der Mongolen (Kln, 1985), c. 12, pp. 168f.
63
Priscus, fr. 11, 2, p. 261, where he also mentions medos.
 REGIA HRING 453

made from mares milk), rice wine, milk. At Mngkes court, the
envoys were offered a choice of several drinks.64 Treasures, and their
public display, sometimes only to a restricted audience, play an impor-
tant part, for instance at the court of Sizabulos. Gift giving was an
essential part of diplomatic contacts, but also of the relationship
between subordinate princes and the rulers of steppe empires.65 What
gift exchange in places of power in the steppe meant for barbarian
society can only be guessed at from chance remarks.66 When William
of Rubruck and his companions left Mngkes court and refused gifts
of gold and silver, the khans secretaries insisted that they took a
tunic each: You are not willing to accept gold and silver, and you
have been here a long time praying for the khan. He requests each
one of you to accept at least a simple garment, so that you do not
leave him empty-handed.67 As far as the relationship between Byzan-
tium, or the European powers of the thirteenth century, and steppe
empires was concerned, the distinction between gifts and tribute
became rather blurred. Priscus observed that the Western Emperor
had conferred upon Attila the title of magister militum, thus conceal-
ing the word tribute. As a result, the payments were issued to him
disguised as provisions issued to the generals.68
Noble women, and especially the rulers wives, come into focus
to a differing degree. In several residences, they had their own build-
ings near those of their husband. This was the case in the palace
of Attila, where Priscus paid a visit to Attilas main wife Hereka.
For Sartaqs court, William of Rubruck is very specific: His camp
(curia) struck us as extremely large, since he has six wives, and his
eldest son, who is with him, two or three; and to each woman belong
a large dwelling and possibly two hundred wagons. 69 Later, at
Mngkes court, William visited some of the womens houses.70 Gyks
mother Tregene, who had been regent after the death of his father
gedei, is exceptional for taking her share in affairs of state.71 There

64
William of Rubruck, c. 28, 15, p. 178.
65
E.g. Iohannes de Plano Carpini c. 9, 31 and 34, pp. 319 and 321. Cf. Pohl,
Die Awaren, pp. 20915.
66
I have discussed that in Pohl, La sfida attilana, for the Huns, and in Pohl,
Die Awaren, pp. 17885, for the Avars.
67
William of Rubruck, c. 36, 15, p. 251.
68
Priscus, fr. 11, 2, p. 279.
69
William of Rubruck, c. 15, 1, p. 114.
70
William of Rubruck, c. 29, 3243, pp. 19599.
71
Denis Sinor, The making of a Great Khan, Studies in Medieval Inner Asia,
454

was a tent that was divided equally between Gyk and his mother
for holding court.72 Whether women were present at receptions and
banquets is not always clear. Priscus does not mention them at Attilas
banquet, and as he goes to some lengths in describing which of
Attilas sons and relatives sat next to him, there may not have been
any. Neither are ladies mentioned in the residence of Sizabulos. At
the courts of Mongol rulers, the wives of the Khan seem to have
been present. In Gyks purple tent, the dominae sat to his left on
the platform, and the noblemen to his right below, in bancis inferius.
Et omni die veniebat multitudo maxima dominarum; probably all of them
were his wives.73 At Batus court, one of the khans wives was sitting
next to him. Apart from that, men were sitting around to his right
and the ladies on his left, and the space on their side not taken up
by women since the only ones there were Batus wives was filled
by men.74 William of Rubruck then encountered a similar arrange-
ment in Mngkes residence, where the envoys were made to sit in
front of the women. Beside the khan, again there sat a young wife,
and a full-grown daughter of his, a very ugly girl named Cirina, was
sitting on a couch behind them with the other children. The dwelling
had belonged to a Christian wife to whom he had been very attached
and who had borne him his daughter. And although he had brought
in the young wife as well, nevertheless the daughter was mistress of
the entire establishment (curia) that had belonged to her mother.75
The audience therefore took place in one of the womens compounds;
William later visited it again when the khan was not present.76
It is quite remarkable that women get more attention in the reports
from Mongolic residences than in our early medieval texts. But it is
doubtful whether any general conclusion can be drawn, apart from
difference in the authors interests. John and William were friars who

n. XIV, esp. pp. 25153. The ascension of Gyk and his short reign are also
featured prominently in Rashid al-Din, J.A. Boyle ed., The successors of Genghis Khan
(New York, 1971), pp. 178203.
72
Iohannes de Plano Carpini c. 9, 36, p. 322: Ibidem divisi fuerunt ab invicem imper-
ator et mater eius: et mater imperatoris ivit in unam partem et imperator in aliam, ad iudicia
facienda.
73
Iohannes de Plano Carpini c. 9, 35, p. 322.
74
William of Rubruck, c. 19, 6, p. 132.
75
William of Rubruck, c. 28, 15, p. 178.
76
William of Rubruck, c. 29, 32, p. 195.
 REGIA HRING 455

went to Mongol courts with an eye to mission, in which women

(some of whom were Christians) could perhaps support them, and
were therefore keenly observed. Priscus and Zemarchos (on whose
report Menander relies) were Byzantine diplomats who were more
interested in hierarchies and informal networks of power. Still, Priscus
also found it advisable to establish contacts with women who had a
say in politics: most importantly, Attilas wife Hereka who not only
received presents but also arranged a dinner in the house of her
master of affairs where the Roman envoys met some of the lead-
ing men of the nation.77 The wives or widows of kings also had
their own economic base, for instance one of Bledas widows in
whose village the Roman envoys had to seek shelter because of a
storm.78 The system of providing for (at least some of ) the many
wives of the ruler who had their own houses seems to have been
similar to that observed by William of Rubruck in the Mongol empire.
In the case of the Avars, a queen is mentioned only once in a polit-
ical role: the catuna mulier, the khagans wife, accompanied her hus-
band when he submitted to Pippin in 796, where an Avar renegade,
who perhaps had suffered the effects of her power, abused her.79
Power was, to a large extent, exercised in the hall.80 What is, how-
ever, largely absent from all accounts is an active or ceremonial role
of women at court assemblies, receptions or banquets. There is no
evidence for a lady with a mead cup entertaining the steppe emperors
retinue, neither in the early middle ages nor at the Mongol court.
Women did however play an important role in rituals of welcome.
When Attila returned to his favourite residence, the wife of Onegesius
came out to meet him with her attendant, serving food and wine
which he consumed in the saddle. Both in Priscus and in William
of Rubrucks account, the ruler is also welcomed by singing girls.
In this village, as Attila was entering, young girls came to meet him

77
Priscus, fr. 13, p. 291.
78
She ruled the village: Priscus, fr. 11, 2, p. 261.
79
De Pippini regis victoria Avarica, MGH Poet.Lat. 1 (Berlin, 1881), p. 116, 6; Pohl,
Die Awaren, p. 305. The title katun for the rulers wife is rather common among
steppe peoples, also among the Mongols (William of Rubruck, c. 29, 19, p. 189).
80
J.L. Nelson, Queens as Jezebels: the careers of Brunhild and Balthild in
Merovingian history, in: D. Baker ed., Medieval women (Oxford, 1978), pp. 3177,
who underlines the strong role of Frankish queens in this exercise of power in the
hall.
456

and went before him in rows under cloths of white linen, which
were held up by the hands of women on either side . . . There were
many such rows of women under cloths, and they sang Scythian
songs.81 And 800 years later: Semper, quando Cuyuc de tentorio exibat,
cantabatur eidem et cum quibusdam virginibus pulchris, que in summitate lanam
habebant coccineam, inclinabant ei, quod nulli alii duci fiebat, quousque exterius
morabatur.82
What should we make of the things that we might expect to find
in these descriptions but that are scarcely present? Several examples
can be noted. First, religious sites, rituals and practices are virtually
absent from our early medieval reports. Temples, priests, sacred
images, rites and ceremonies, magic, divination, taboos, are a recurrent
feature in texts about barbarians; Tacitus Germania, for instance, is
quite rich in detail about Germanic religion.83 The thirteenth-century
friars, whose reports are cushioned in ethnographic material, do
describe some religious practices at steppe residences.84 William of
Rubruck is not only very accurate in his observations about Manich-
eans, Armenians and other Christians he encounters, closely observ-
ing ways in which they had adapted Christian ritual to pagan practices;
he also records in detail rituals of divination from the charred shoul-
der blades of sheep at Mngkes court, and repeatedly remarks that
touching the threshold of the residence was taboo.85 The only hint of
religious beliefs in Priscus is the story about the sword of Mars found
by accident, and which Attila considered as an omen.86 Menander
describes with gusto a shamanistic purification rite practiced on
Zemarchos and his companions when they entered the Turkish realm,
with fire, incense, barbarian chants, bells and drums, trance and
exorcism.87 Plano Carpinis party had to undergo a somewhat sim-

81
Priscus, fr. 11, 2, p. 265.
82
Iohannes de Plano Carpini c. 9, 31, p. 319.
83
Pohl, Die Germanen, pp. 7885.
84
Esp. Iohannes de Plano Carpini c. 3, pp. 23544; William of Rubruck c. 25,
pp. 15356. In general, see Jean P. Roux, La religion des Turcs et des Mongols (Paris,
1984); C.C. Mller, Die Religion der Mongolen, in: A. Eggebrecht ed., Die Mongolen
und ihr Weltreich (Mainz, s.a.), pp. 16982.
85
William of Rubruck, c. 29, 2628, pp. 19294; c. 35, pp. 24045. For a sim-
ilar oracle in the Secret history of the Mongols, see Das Buch vom Ursprung, ed.
Heissig, c. 22, p. 166.
86
Priscus, fr. 12, 12, pp. 280f.
87
Menander, fr. 10, 3, p. 119. Cf., in general, M. Eliade, Schamanismus und archa-
ische Ekstasetechnik (Frankfurt/Main, 1975; French original: Paris, 1951), esp. pp.
177248; I.M. Lewis, Ecstatic Religion. An anthropological study of spirit possession and
 REGIA HRING 457

ilar ritual of purification when they crossed the border, in terre finibus
Comanorum, on the way to Batus court, when they had to pass between
two fires.88 Menander, however, does not mention any religious prac-
tices at the khagans residence itself. His silence cannot be due to a
lack of interest or the systematic disregard for pagan practices by a
Christian author, otherwise he would have omitted the shamanistic
ritual as well. On the other hand, it was not part of the mission of
Byzantine diplomats to describe in detail what was to be expected,
among which would have been the pagan religion of their hosts.
Few early medieval authors cared much about the exact forms of
barbarian paganism, and many texts rely on stereotypes of classical
religion or even descriptions from the Old Testament when they
describe pagan practices.89 On the other hand, we may conclude
that religious ritual was not an unmistakeable part of the setting in
which ambassadors were received by steppe emperors. If cosmology
governed life at barbarian courts,90 its impact goes rather unnoticed
in the reports we have, even in the practices described by Plano
Carpini and William of Rubruck.
Secondly, the same can be said for the sense of the past. Barbarian
envoys who arrived at Rome or Constantinople were (as far as they
understood them) overwhelmed by representations of the past: statues,
inscriptions, triumphal arches, cenotaphs, mosaics, and so on. The
Beowulf poem is permeated by recapitulations of a (often not-so-
distant) past that serves as an index of prestige, and as a matrix for
the future. The reports from steppe residences have little that is com-
parable. Plano Carpini prefixes a short history of the Mongols to
his account, and William of Rubruck indulges in historical digres-
sions. Both also use information they have obviously collected on
the spot. But the traces of the staging of memory as a part of the

shamanism (Harmondsworth, 1971); W. Heissig, Schamanen und Geisterbeschwrer in der

stlichen Mongolei (Wiesbaden, 1992); J. Gieauf, Ekstatische Seelenreisende. Die
nordeurasischen Schamanen, Mitteilungen der Grazer Morgenlndischen Gesellschaft 6
(1997), pp. 3962.
88
Iohannes de Plano Carpini c. 9, 14, p. 310; cf., however, c. 3, 4, p. 343 (walk-
ing between two fires as a ritual of submission). D. Sinor, Diplomatic practices in
medieval Inner Asia, Studies in Medieval Inner Asia, n. XVI, p. 348.
89
For instance, in the case of the Lombards: W. Pohl, Deliberate ambiguity.
The Lombards and Christianity, in: I. Wood ed., Conversion in the Middle Ages. IMR
series (Turnhout, forthcoming).
90
As could be expected according to the model proposed by Bazelmans, Beyond
power.
458

representation of rulership, or of ethnic identity, are few. Plano

Carpini relates that idols of Chingis-Khan were venerated, and saw
one in front of Gyks residence.91 A rare exception in the early
medieval texts is Priscus remark that at Attilas banquet, when night
fell, two barbarians came in and stood before Attila and chanted
songs which they had composed, telling of his victories and his deeds
of courage in war. Reactions were emotional: The guests fixed their
eyes on the singers: some took pleasure in the verses, others recall-
ing the wars became excited, while others, whose bodies were enfee-
bled by age and whose spirits were compelled to rest, were reduced
to tears.92 The scene may have been similar when a bard was bid-
den to sing/a hall-song for the men on the mead-benches in Beowulf;
repeatedly, those songs were in praise of somebody who was pre-
sent, too.93 We know from Old Turkic inscriptions that a rulers
exploits mattered for the ideology of steppe empires (as could be
expected).94 The Secret History of the Mongols appropriately begins
with Gengis Khans lineage, going back to a grey wolf.95 Past vic-
tories played a role in many negotiations. But formal statements, or
ritual invoking the deeds of the ruler, let alone his lineage or the
origin myth of his people, do not occur in our texts. Western ambas-
sadors seem to have missed, or failed to record such occasions, and
the uses of the past for steppe empires went unnoticed.96
Related to this lack of attention towards barbarian memoria, there
is, thirdly, a surprising vagueness about names and locations. None
of the residences has a name in our sources, apart from being called
Attilas palace or the hring of the Avars. This corresponds to the
lack of names for the residence of the Thuringians or other Germanic
peoples, so it is not simply due to the mobile and temporary charac-
ter of the encampments of a steppe ruler and in any case Attilas
wooden palace or the Avar ring were permanent settlements. Even
Beowulf usually gives no names for the mead-halls, but only attrib-

91
Iohannes de Plano Carpini c. 3, 3, p. 237; Die Mongolen und ihr Weltreich, p. 182;
p. 203.
92
Priscus, fr. 13, 1, p. 287.
93
Beowulf vv. 106768.
94
V. Thomsen, Alttrkische Inschriften in der Mongolei, Zeitschrift der Deutschen
Morgenlndischen Gesellschaft 78 (1924/25), pp. 12175.
95
Das Buch vom Ursprung, ed. Heissig, c. 1, p. 9.
96
Cf., in general: Y. Hen and M. Innes eds., Using the past in early medieval Europe
(Cambridge, 2000).
 REGIA HRING 459

utes them to a king or a people (the great exception being Heorot).

The lack of place-names does not reflect a general disregard for bar-
barian geography. Roman and Byzantine, and later, Carolingian
authors fill their accounts with barbarian names. The topography of
barbarian lands strove to define the vast spaces beyond the civilised
world using names of rivers, mountains, and peoples, in some cases
also long lists of barbarian civitates (as in Ptolemaios, or later, the
Bavarian Geographer). These ethnic topographies were a means of
cognitive appropriation that might facilitate political control.97 It seems
that barbarian residences were only indirectly linked with this matrix,
by being connected with names of a people and a ruler; they were
not, as might be expected, knots in the net of political geography
of barbarian countries. Even their location was not described with
much exactitude. It is almost a paradox that on the basis of writ-
ten accounts distant Central Asian residences can be traced more
easily than those in the Carpathian basin, or in the forests of Germania,
a few dozen miles from the Roman frontier.
Fourth, but not least, much pragmatic information is absent from
the early medieval accounts, notably as regards the economy and
infra-structure. Among the few exceptions, there are the goldsmiths
at the Rugian court, the maids of Attilas wife Hereka stitching
embroidery in her house, and Onegesios Roman architect. The size
of the population, their living conditions, the role of artisans and
trading activities remain largely obscure. One might expect some
interest from diplomats in the production and exchange of weapons,
horse-breeding or the general level of technology, but little of this
is attested in the early medieval texts used here (quite unlike, for
instance, the reports of Arabian geographers or the deeply puzzling
Cosmography of Aethicus Ister).98 Plano Carpini and William of
Rubruck provided much fuller accounts of the living conditions of
the Mongols in the context of their ethnographic chapters, although
much of that was from second-hand knowledge.99

97
H. Reimitz, Grenzen und Grenzberschreitungen im karolingischen Mittel-
europa, in: W. Pohl and H. Reimitz eds., Grenze und Differenz im frhen Mittelalter
Forschungen zur Geschichte des Mittelalters 1 (Wien, 2000), pp. 10566.
98
I.N. Wood, Aethicus Ister: an exercise in difference, in: W. Pohl and H. Reimitz
eds., Grenze und Differenz im frhen Mittelalter, Forschungen zur Geschichte des Mittelalters
1 (Wien, 2000), pp. 197208.
99
Iohannes de Plano Carpini, esp. c. 2, pp. 23235; c. 4, 711, pp. 24851;
William of Rubruck, cc. 27, pp. 7292.
460

On the basis of these observations, we can return to the fundamental

methodological question raised above. How do we deal with these
reports, and how cautious do we have to be about the information
they convey? The vivid stories, told by an eye-witness in the first
person singular, produce a strong narrative effet de verit. The
genre of diplomatic travel reports with its attention for details, its
sobriety, its aloofness creates a further impression of credibility.100
And indeed, as compared to the collection of prejudice and stereo-
types about the Huns that we find in Ammianus Marcellinus,101 the
report of Priscus is a unique source of information. However, we
have to be cautious; even Priscus cannot be regarded as a straight-
forward first-hand witness of life at the court of Attila. We have to
take several filters into account through which information about the
court of Attila has been passed down to us. First, we do not get the
full diplomatic report delivered by the ambassadors on their return,
but a stylized literary version directed at a general Roman audience,
in which a lot of specific information which was of interest to the
courtiers and generals of Theodosius II may have been omitted. The
emperor Marcian, who came into power in 450, quite soon after
the embassy had returned, may have used some of the information
brought back when he made the hazardous decision to stop pay-
ments to Attila.102 But most of it became useless after Attila died in
453. Furthermore, we do not have the work of Priscus (or that of
Menander) as a whole, but only fragments preserved in the Excerpta
de legationibus in the tenth century. That Constantine Porphyrogennetos,
and others, took such interest in the material is the only reason why
we know anything at all about the places of power of those steppe
rulers with whom the Byzantines had to deal with.103 But we only
have the texts that were selected in the tenth century, and therefore
see the courts of Attila and Sizabulos through Constantines eyes.

100
See, in general: J. Shepard and S. Franklin eds., Byzantine diplomacy (Aldershot,
1992); esp. M. Mullett, The language of diplomacy, ibid. pp. 20316.
101
Ammianus Marcellinus 31, 2, transl. W. Hamilton (Harmondsworth, 1986),
pp. 41114.
102
Wirth, Attila, p. 88.
103
The Excerpta de legationibus are the only completely transmitted part of the
Eklogai, the encyclopedia that Constantine had commissioned, a series of thematic
selections from ancient and Byzantine writers; cf. K. Belke and P. Soustal eds., Die
Byzantiner und ihre Nachbarn (Wien, 1995), p. 46.
 REGIA HRING 461

A second filter is constituted by the eyes of the diplomat, or, in

the case of the thirteenth-century friars, the eyes of the missionary-
diplomat. He basically writes what pertains to his mission, and his
account centres on diplomatic exchange, political rhetoric and polite
conversation (or religious argument); all of that fills many pages in
our texts, and the passages that I have considered only constitute
the setting for these negotiations, a few sketches on the margin of
a political agenda. It is no coincidence that most of the missions
from which reports have survived were failures. The plot against
Attilas life prepared by members of the Priscus embassy was dis-
closed, bringing the envoys into an embarrassing and dangerous sit-
uation. The Turks of Sizabulos did not attack the Avars from the
rear as the Byzantines had hoped, and when the Byzantine envoys
returned later, they were bitterly reproached by the successor of
Sizabulos for having concluded a treaty with the Avars. The Mongols
were not receptive to the friars missionary zeal: Plano Carpini
came back to the pope with one of the most ominous documents
ever received by a western potentate,104 and the joint attack on the
Muslims in the Holy Land hoped for by William Rubruck was never
launched. Diplomatic reports, and their literary elaborations, were
mostly intended to clear the envoys of the suspicion of having done
a bad job. Acute observers like Priscus may have added some descrip-
tions and anecdotes to the accounts of their diplomatic business, but
most of those belong to the level of literary stylization, and do not
necessarily represent what they had observed with the cold eye of
the diplomat. Both Plano Carpini and William of Rubruck garnished
their reports with valuable ethnographic material, but that was of
course a mix of observations, hearsay and literary borrowings.
A further filter is the most interesting one, for it takes us into the
heart of barbarian politics. What ambassadors observe at a foreign
court is for the most part specifically addressed to them. The reports
I have quoted prove that steppe rulers were experts in the art of
dealing with diplomats. Attila or Mngke received the envoys with
a carefully calculated blend of hospitality and intimidation, curiosity
and neglect, condescension and confidentiality, generosity and greed.105

104
P. Jackson and D. Morgan, Introduction, The mission of Friar William of
Rubruck, pp. 28f.
105
Sinor, Diplomatic practices in medieval Inner Asia. Cf. Pohl, Konfliktverlauf ,
with further literature.
462

Next to sending armies, dealing with envoys was the principal way
in which a steppe empire happened. This was a privileged way of
symbolic communication in which submission, alliance and enmity
were continually renegotiated. In this way, frontiers of significance
were being drawn and redrawn:106 inside and outside, war and peace.
No fundamental difference was made between the envoys of subject
peoples and of exterior powers, a fact that Byzantine envoys ascribed
to the hubris and arrogance of the barbarians, but one that puzzled
Plano Carpini and William of Rubruck. Whoever entered into gift
exchange with a steppe ruler recognised his power, and became part
of a sophisticated ceremonial hierarchy of foreign envoys. However,
there was a considerable difference between the steppe empires in
the Byzantine periphery and the Mongols. Huns and Avars relied
on the massive import of Byzantine prestige goods, and diplomatic
relationships were a way to negotiate this flow of gold and silver,
silk and crafts from Byzantium.107 For the Mongols, western Europe
was ephemeral. That did not make much of a difference in the cer-
emonial setting. William of Rubruck sat on the less prestigious left
side next to the women at Mngkes reception; but the Byzantines
had to sit on the left side in Attilas banquet hall, too.
In both cases, the envoys played a double role in the representa-
tion of barbarian rulership. On the one hand, they were the audi-
ence at which speeches and rhetoric exchanges, displays of treasure,
gestures of intimidation, and the representation of rulership at cer-
emonial banquets were directed. On the other hand, they were them-
selves part of the staging of barbarian rulership. The reception of
ambassadors was a ritual of power directed at the aristocrats pre-
sent at court, and designed to demonstrate the high esteem and
unequalled power their ruler enjoyed. In fact, this was not so different
in Constantinople, and at the courts of Charlemagne or Frederick
Barbarossa. The number of foreign envoys, their submissiveness, and
at times the contempt with which he treated them, considerably
enhanced the prestige of a ruler.108 Ambassadors are not external
observers of barbarian places of power, they are a key to their func-

106
Pohl, Soziale Grenzen.
107
Pohl, Die Awaren, pp. 17885.
108
See, for example, the Panegyric by Corippus on Justin II (In laudem Iustini
Augusti Minoris, ed. A. Cameron [London, 1976]) in which Avar envoys are treated
with extreme contempt; or the story in Notker, Gesta Karoli, 2, 6, about ambas-
sadors who were plunged into confusion by being successively led to several mag-
nificently dressed dignitaries they mistook for Charlemagne.
 REGIA HRING 463

tion. The representation of a barbarian ruler is no one-way show-

off, it is a communication between competitors in a game of power.
For the envoy, being part of this game may obscure many of the
things that happen in and around a barbarian residence, and he
may not even fully understand his part in it. But his perceptions fol-
low the logic of the game, and help us to reconstruct it, even though
we only have access to it through filters the communication
between the author and his public, and between the diplomat and
those who sent him.
A further element of interaction between the diplomat and the
barbarian state is the control of movement. A considerable propor-
tion of the reports deals with the way in which the envoys were told
which route to take and what to do. Priscus narrative shows just
how much coordination was necessary: The Byzantines meet Attila
in his camp near the frontier, they follow him, then take a different
route, converge with a West Roman embassy, have to wait for Attila
in a specified village and then follow him to his main residence.
There is an element of taboo in these travel arrangements. The
Byzantines are not allowed to arrive at the residence before Attila,
just as they are not allowed to pitch their tent on higher ground
than the king. While near the court, all the movements of the envoys
are under control, and they frequently receive indications what to
do from more or less high-ranking messengers. Finally, they are
escorted out of the country again. These arrangements are an exam-
ple of the subtle control of spatial movements in space common in
steppe empires. Even Priscus observed some of the constant com-
ings and goings of Attilas envoys entrusted with the control of groups
of warriors and the sedentary part of the population somewhere on
the periphery of the empire (for instance, the Akatzirs north-east of
the Carpathian mountains). Maintaining the hold on all these groups
required the constant exchange of information and a coordination
of movements that was complicated by the fact that the residence
itself was constantly on the move. Ethnological studies of nomad
societies have demonstrated how such a complex coordination of
movements can be achieved.109 However, the early medieval regna

109
F. Barth, Nomads of South Persia (New York, 1965); Pastoral production and society
(Cambridge, 1979); esp. J. Legrand, Conceptions de lspace, division territoriale
et division politique chez les Mongols de lpoque post-imperiale (pp. 15569);
P. Bourdieu, Entwurf einer Theorie der Praxis (Frankfurt, 1979); E. Turri, Gli uomini
delle tende (Milano, 1983).
464

in the west faced similar problems. For instance, legislation by the

Lombard kings Ratchis and Aistulf shows the attempt to control all
traffic across the kingdom, not least to in order to curb the exchange
of envoys between the pope and the Franks.110

The Mongols seem to have developed the control of space beyond

what Attila or the ancient Turks had achieved. Whereas Attila seems
to have conferred the task of communicating with foreign envoys ad
hoc and regarded this as a favour and a sign of personal trust, the
Mongols had set up clear spheres of competence. William of Rubruck
repeatedly had to deal with officials whom he, perhaps mistakenly,
calls iam-iam, and who were responsible for dealing with arriving
envoys.111 And like all other travellers in the Mongol realm, he could
use the yam, the system of relay stations that facilitated and con-
trolled travel throughout the steppe.112 Mongol rulers also cooper-
ated more directly with cities than early medieval steppe empires
had done. The Roman cities that Attilas armies occupied in 447,
for instance Naissus, were practically deserted when Priscus saw them
two years later. Attila could have shifted his residence to the former
imperial capital Sirmium (as the Gepids did later), but he showed
no intention to do so. When the Avars together with the Persians
besieged Constantinople in 626 the arrangement was that the Avars
would plunder the city when it was taken, whereas the Persians
would receive the population; the khagan did not plan to maintain
urban life in the city, let alone move his residence there.113 The
Mongols controlled a number of important cities along the silk route.
One of Mngkes residences was outside the walls of Karakorum.114
And Marco Polo, a few decades later, encountered Kublai Khan in
his urban residence of Khanbaliq in China.
Diplomats descriptions of the residences of steppe rulers are traces
of a particular form of communication, very different, for instance,
from the perspective of the Secret History of the Mongols, written

110
W. Pohl, Frontiers in Lombard Italy: the laws of Ratchis and Aistulf , in:
W. Pohl, I. Wood and H. Reimitz eds., The transformation of frontiers. From late
antiquity to the Carolingians (Leiden/Boston/Kln, 2000), pp. 11742.
111
For instance, William of Rubruck c. 15,1, p. 114 (with n. 3).
112
For the institution of the Mongolic postal system, see the Secret history of
the Mongols (Das Buch vom Ursprung, c. 12, ed. Heissig, pp. 175f.).
113
Pohl, Die Awaren, p. 251.
114
William of Rubruck c. 32, pp. 221f.
 REGIA HRING 465

from a Mongol perspective in 1240.115 They privilege those elements

where an intense interaction between the ruler and the ambassador
took place. This was not only an exchange of words, but also of
gestures, symbols, gifts, threats and commands. What this commu-
nication meant for the inner cohesion of a steppe empire remained
largely obscure to its visitors. But that does not make those obser-
vations which allow us to reconstruct their role at a barbarian court
any less valuable. There is much less evidence about other aspects
of barbarian residences in these reports. The envoys did have a vital
interest in gathering information about the delicate balance of power
between the rulers sons, his chief advisors and, perhaps, his wives,
as far as it mattered for their purpose. But in this respect, their inter-
est differed from ours. They wanted to identify individuals whose
influence as advisors and possibly as intercessors they might use, or
had to fear. But they were less interested in the complex system of
ritual exchange between the ruler and his aristocrats as a whole, and
hardly capable of deciphering the cosmological framework that might
have regulated these relations.
Of course, the envoys had some interest in the different symbolic
systems that they encountered. There was always the risk of comit-
ting a fatal mistake. William of Rubrucks companion once stum-
bled over the threshold of the khans tent and as a consequence,
was almost put to death; William had to swear they had not known
about the taboo.116 Ambassadors noticed restricted and open spaces
and the hierarchical order of accessibility. But again, that was part
of the social field in which they themselves had to learn to move.
Their texts have to be understood in relationship to this social space.
Read systematically, they can yield considerable information about
it. On the other hand, especially in Byzantium, the texts contributed
decisively towards shaping the reality that they described. The col-
lection of the Excerpta de legationibus only represents a late stage in
the collection of a growing body of texts about missions to foreign
powers which generations of diplomats used as a guideline. Such
interests obviously were less important in diplomatic contacts with
barbarian residences in the west. Even though Roman diplomacy

115
Das Buch vom Ursprung, ed. Heissig.
116
William of Rubruck, c. 29, 29 and 37, pp. 19496. I would like to thank
Herwig Wolfram and Brigitte Resl for suggestions, and Ian Wood and Ann Christys
for also revising my English.
466

had acquired a thorough knowledge of ways to deal with barbar-

ians, writers obviously were not very keen on expounding it for a
general public. Steppe residences were the only examples of bar-
barian large-scale organisation that were conspicuous enough for
early medieval authors to write about.
On the whole, these descriptions do not highlight what was exotic,
bewildering or incomprehensible about barbarian places of power.
Apart from some atmospheric detail, many features that emerge from
the texts were reasonably familiar. Above all, one thing seemed quite
natural to these visitors that came from far away, the one thing that
they had in fact come for: power. We may wish that they had been
more curious to ask what it was, how it worked, and in what ways
it shaped social relations. But they made up for this lack of analyt-
ical insight by leaving an account of their own movement through
the gravitational field of a distant place of power, almost a social
experiment by which they tested its force and impact.
 ASGARD RECONSTRUCTED?
GUDME A CENTRAL PLACE IN THE NORTH*

Lotte Hedeager

Gudme/Lundeborg on the Danish island of Funen was excavated

during the 1980s and early 1990s,1 and has been interpreted as a
unique trading and production site that flourished from the third to
the sixth/seventh centuries. The site has mainly been understood in
terms of long-distance trade, economy, control, production, gold,
hall, richness, gods, sacred, and power. Although these key-
words are significant, they have never been included in a coherent
model of explanation, neither in relation to Gudme/Lundeborg itself,
nor to the entire group of Scandinavian sites that could be classified
as central places.2
Initially, Gudme was considered unique, but nowadays it is regarded
as one of several rich archaeological settlements in South Scandinavia
dating from the third to the sixth/seventh centuries.3 Because of its

* I am indebted to all members of the ESFs theme group on Power and

Society for stimulating and productive meetings, always supporting an interdisci-
plinary approach. This paper on Gudme was first presented at our meeting in
Ravenna 1995, and then, in a more elaborate version, in Bellagio in 1998. I am
grateful for all critical as well as encouraging comments during this long time of
formation. I am a prehistoric archaeologist whose main field is the pagan North
and its archaeology, not textual evidence or Christianity, so I am grateful to Mayke
Jong and Frans Theuws for their support. Needless to say, any remaining deficiencies
are my own responsibility. Last but not least, Mayke has transformed my English
text into the English you are about to read.
1
P.O. Nielsen, K. Randsborg and H. Thrane eds., The archaeology of Gudme and
Lundeborg (Copenhagen, 1994).
2
E.g. L. Larsson and B. Hrdh eds., Centrala platser, centrala frgor. Acta Archaeologica
Lundensia, Ser. In 8, no. 28 (Lund, 1998).
3
I.e. C. Fabech, Organising the landscape. A matter of production, power, and
religion, in: T. Dickinson and D. Griffiths eds., The making of kingdoms. Anglo-Saxon
Studies in Archaeology and History 10 (Oxford, 1999), p. 37; U. Nsmann, The
ethnogenesis of the Danes and the making of a Danish kingdom, in: T. Dickinson
and D. Griffiths eds., The making of kingdoms. Anglo-Saxon Studies in Archaeology
and History 10 (Oxford, 1999), p. 8.
468

size and the nature of its finds, studying Gudme/Lundeborg can give
us more insight in the characteristics of a central place, and also
provide a model of comparison that is helpful to the interpretation
of other sites in Scandinavia.
This paper deals with Gudme/Lundeborg as a place that has been
constructed, maintained and transformed over centuries, for purposes
other than strictly economic and political ones. First and foremost,
Gudme was a ceremonial centre, where ancient beliefs were articu-
lated in rituals and performances. In the opening section, I will dis-
cuss Gudme as a place where foreign objects from the outside world
were acquired (trade) and transformed into prestige objects (pro-
duction) embedded in the cosmological order. Using data from
anthropological research as an explanatory framework, I will pay
special attention to the importance of skilled crafting and skilled
metal work in particular as an activity fundamental to the process
of transformation. To broaden the context, I will also look at the
role of smiths and the significance of gold in Old Norse sources. All
this will reveal that metallurgy, skilled metal work and gold were
crucial concepts in northern cosmology. Finally, I will focus on Gudme
and the surrounding landscape as a sacred place a representa-
tion of the centre of the world along the lines of northern mythol-
ogy. The article is organised in an inductive manner, beginning with
the archaeological evidence and finishing with the interpretative topo-
graphic model of Gudme as a paradigmatic model of Asgard home
of the Asir gods.
This approach is not unproblematic. The Old Norse sources orig-
inate from early Christian times, that is, the early thirteenth cen-
tury, and are therefore not to be treated as a reflection of genuine
paganism. It would go too far to discard all written texts as useless
to our endeavour, however. This would leave archaeologists without
any relevant written evidence from the North; if used carefully, the
Old Norse texts yield valuable information. Similarly, an anthropo-
logical approach based on non-western, pre-industrial societies, fur-
nishes archaeologists with a general theoretical framework, enabling
them to get beyond the archaeological and textual evidence. Lacking
the modern separation of economic, political and symbolic institu-
tions, pre-Christian Scandinavia can be compared to traditional com-
munities; in both cases the world view of a given society tends to
fuse these separate domains into a coherent whole. Since much cos-
 469

mological information is thought to be contained in myths,4 special

attention will be paid to the myths of Old Norse literature. If one
focusses on Gudme as a symbolically constructed place which rep-
resents notions of the cosmological order in connection with social
power, the spatial organisation of the place can no longer be inter-
preted as a mere expression of the practicalities of power, or as a
simple reflection of economic activities, including production and/or
trade. Instead, such activities are to be included in a coherent model
of explanation, which should also become part of a more general
discussion of other central places in the North.

T G 5

Gudme is a complex of sites in a limited area on the east coast of

the island Funen. Apart from the locality Gudme itself, which is the
inland settlement area, there are also Lundeborg, a trading port sit-
uated on the coast, and the extensive cemetery of Mllegrdsmarken
(fig. 1).6

4
J. Weiner, Myth and metaphor, in: T. Ingold ed., Companion Encyclopedia of
Anthropology (London, 1999), p. 591.
5
Neither Gudme nor Lundeborg have been published in monographic form.
The publication from 1994, Nielsen e.a., The archaeology of Gudme is the last extended
contribution including a variety of articles by different authors about the
investigation, 198491 and the find material. Afterwards the big hall and other
houses have been excavated. These are published by P. stergaard Srensen, Gud-
mehallerne, Nationalmuseets Arbejdsmark, 1994 (Copenhagen, 1994). This article
apart a lot of minor articles have been published in a variety of contexts and by
a variety of authors too. The latest example is H. Thrane, Materialien zur Topografie
einer eisenzeitlichen Sakrallandschaft um Gudme auf Ostfnen in Dnemark, in:
A. Wesse ed., Studien zur Archologie des Ostseeraumes. Von der Eisenzeit zum Mittelalter.
Festschrift fr Michael Mller-Wille (Neumnster, 1998) and H. Thrane, Gudme,
Reallexikon der Germanischen Altertumskunde 13 (Berlin, 1999), pp. 1428. The Gudme/
Lundeborg area occupied a prominent place in the Late Roman import into
Scandinavia. U. Lund Hansen, Rmischer Import im Norden (Copenhagen, 1987).
6
In this paper I want to focus on the overall structure of the area in the fourth
to the sixth century. The present material indicates a maximum for Gudme in the
fourth to the sixth century and for Lundeborg in the third and fourth century.
However, none of the sites have been fully excavated; only 10% of the site at Lun-
deborg have so far been investigated and the huge settlement area of Gudme has
been excavated in selected plots. Still, the dating is a matter of debate; the general
impression is that the two places have functioned roughly during the same period.
 470

Fig. 1. The research area of Gudme. The cultural landscape is reconstructed on the basis
of the topographical maps of c. 1800.
1. Gudme III, 2. Settlement (indicated by Sehested as Mltidsplads), 3. Mllegrdsmarken,
4. Broholm gold hoard, 5. Lang cemetery, 6. Lundeborg. settlement, D bronze statue,
hoard, + graves, church, x stray find. Heights are in meters above sea level. (After
H. Thrane, Das Gudmeproblem und die Gudme-untersuchungen, Frhmittelalterliche Studien
21 (1987), p. 36.
 471

The settlement of Gudme7 is located approximately four kilome-

ters inland. While the surrounding area was inhabitated from around
100 B.C. until the early middle ages, this particular settlement, of
which about 500.000 m2 has been documented, existed from c. 200
600 A.D.8 A range of small excavations in the central area reveal a
settlement organized in fenced farmsteads. These are similar to other
farmsteads in Denmark from the Late Roman and Migration peri-
ods. The number of contemporary farms is unknown, but judging
by the estimated size of the settled area there must have been at
least 50 farms around 400 A.D., the peak of activity of the settle-
ment. Besides common farmsteads, two exceptional houses have been
excavated.9 The largest building a hall10 was nearly 50 meters
long and 10 meters wide with eight sets of posts, each with a diam-
eter of more than 80 centimeters, to support the huge roof construc-
tion (the size of these posts may indicate a second floor).11 Two wide
doorways in the centre of the building led into a particularly large
room. According to the excavator the house was a long construc-
tion building (cogging). One smaller house nearby was constructed
in the same manner. It had a length of c. 25 meters, and a width
of nearly 10 meters. Compared to the normal farmsteads with a
maximum length of 35 meters and a standard width of 67 meters,
the size of the hall (and the width of the small house too) is unknown
in Denmark before the Viking Age. The construction is remarkable
as well, for nothing like it is known in Denmark before the early
middle ages. Apparently, these houses represent a tradition and a
technical knowledge that has no background in local traditions.

7
P.O. Nielsen, The Gudme-Lundeborg project. Interdisciplinary research, 1988;
P. Vang Petersen, Excavations at sites of treasure trove finds at Gudme; P. ster-
gaard Srensen, Houses, farmsteads and settlement pattern in the Gudme area;
L. Jrgensen, The find material from the settlement of Gudme II composition
and interpretation, all in: P.O. Nielsen e.a. eds., The archaeology of Gudme and Lundeborg
(Copenhagen, 1994).
8
L. Jrgensen, The warrior aristocracy of Gudme: the emergence of landed
aristocracy in Late Iron Age Denmark?, in: H.G. Resi ed., Produksjon og Samfund.
Varia 30, Universitetets Oldsaksamling (Oslo, 1995), p. 215 and Fig. 14.
9
stergaard Srensen, Gudmehallerne.
10
The foundation of the hall is dated to A.D. 300400, i.e. in the early stage
of the settlement: Thrane, Gudme, p. 145, with references.
11
This is my suggestion; the issue has not yet been discussed by archaeologists.
472

Gudme

In Gudme itself, but also along the coast and scattered in the whole
area, numerous hoards, treasures, and single finds of precious met-
als, especially gold, have been found. All kinds of Roman coins (gold,
silver and copper)12 were also present in amazing numbers, either as
hoards or as stray finds. Hoards containing scrap silver (amongst
others Byzantine tablewares cut into pieces), scrap bronze such as
part of a Byzantine statue, and even scrap gold, all indicate an exten-
sive forging and casting activity. This impression is supported by
finds of drops of melted gold, silver, bronze and glass as well as
hammer scales of iron slag and small strips of gold foil. Among the
finds from the plough soil were a Frankish silver bird fibula and
some gilded fragments of a Roman bronze helmet. Other hoards or
treasures contain golden jewelry, necklaces, bracteates etc., mostly of
local Scandinavian origin, but also some of Byzantine provenance.
In his study of gold bracteates Morten Axboe, agreeing with Karl
Hauck, has drawn attention to the fact that the Gudme area played
an important part in the development of gold bracteates and their
iconography, although it cannot have been the only place in Scan-
dinavia that played such a role.13 In addition, the largest14 Danish
gold hoard from the Migration period was recovered in the same
area: the hoard from Broholm with more than four kilos of golden
arm- and necklaces, bracteates and suchlike. Besides the jewelry, neck
rings, arm rings, finger rings, bracteates, fibula with inlayed garnets,
etc., the hoards contain scabbard mouths, some with marvellous
filigree work, some shaped as a gold spiral, as well as intertwined
pair of rings for pommels of the so-called ringswords which have
been interpreted as the customary insignia for members of the royal
Frankish hird (retinue of warriors).15 Although the settlement has not

12
A. Kromann, Gudme and Lundeborg The coins, in: Nielsen e.a. eds.,
The archaeology of Gudme, pp. 647.
13
M. Axboe, Gudme and the gold bracteates, in: Nielsen e.a. eds., The archae-
ology of Gudme, pp. 6877, and K. Hauck, Gudme as Kultort und seine Rolle beim
Austausch von Bildformularen der Goldbrakteaten, in Nielsen e.a. eds., The archae-
ology of Gudme, pp. 7888.
14
With the exception of two golden horns from Gallehus in Southern Jutland
(stolen and remelted early in the ninetheenth century).
15
H. Steuer, Helm und Ringschwert. Prunkbewaffnung und Rangabzeichen ger-
manischer Krieger. Ein bersicht, in: Studien zur Sachsenforschung 6 (1987), pp.
189236; H. Steuer, Archaeology and history: proposals on the social structure of
 473

yet been fully excavated,16 the results are already impressive: a huge
settled area with a monumental building at its heart a hall measur-
ing around 500 m2, like the royal palace in Copenhagen today
and with an overwhelming amount of metal finds, scrap metal as
well as masterpieces, in and around the central settlement area.

The coastal site of Lundeborg

Lundeborg17 is situated on the coast, on both sides of the mouth of

Tange , a small stream connecting Lundeborg with its hinterland
around Gudme. The Iron Age finds recovered there were distrib-
uted in an area of some 900 meters along the coast by 30 to 75
meters inland. There was activity at Lundeborg from the third cen-
tury (late Roman Period) until 700/800 (early Viking Age). Only
10% of the site has been excavated; the objects found in this area
mostly date from the third and fourth centuries. The objects found
at Lundeborg stem from both trading activities and crafts. The
commodities are mainly Roman glass beads, glass vessels and terra
sigillata. Germanic brooches, arrowhead and metal fittings for swords
and scabbards have been found as well. The other remarkable fea-
ture of Lundeborg is the convincing evidence of extensive craft activ-
ities. All known Iron Age crafts are represented by their tools: those
of carpenters, bronze-, silver- and goldsmiths, blacksmiths as well as
craftsmen working with amber, bone and antler. Small weights, sil-
ver denarii and gold pieces have been found as well. Iron ingots seem
to have been imported from Poland. More than 8000 unused and
broken iron rivets from ships have been found. No doubt, extensive
shipbuilding and ship repairs have taken place at the coast during
the whole period. So far no traces of proper houses have been recov-
ered. There are only traces of small structures (c. 4 5 meters),
interpreted as huts for seasonal use, and dating to the early phase,

the Merovingian empire, in: K. Randsborg ed., The birth of Europe: Archaeology and
social development in the first millennium A.D. Analecta Romana Instituti Danici, Supple-
mentum XVI (Roma, 1989), pp. 10022. L. Jrgensen, The warrior aristocracy
of Gudme.
16
The latest updated map of the excavated areas is shown in P. stergaard
Srensen, Gudmehallerne, fig. 2.
17
P.O. Thomsen, Lundeborg an early port of trade in Southeast Funen, in:
Nielsen e.a. eds., The archaeology of Gudme, pp. 239; P.O. Thomsen et al., Lundeborg
en handelsplads fra jernalderen (Svendborg, 1993).
474

i.e. the third and fourth centuries. According to the excavator, the
many activities at Lundeborg have probably been limited to specific
periods during the year.18

The cemetery of Mllegrdsmarken

Mllegrdsmarken19 is the huge cemetery situated between Lundeborg

and Gudme close to the stream Tange . About 2300 graves have
been excavated and many more have been destroyed by ploughing.
The graves are primarily cremation graves, with many urn graves
among them; these have been dated from the first century B.C. up
to the early fifth century. There are no traces of later burials; the
same holds true for all the other known cemeteries in the Gudme
region. Thus far, this picture corresponds with burial practices from
all over Denmark: there is a hiatus from the fifth until the eight
century (early Viking Age).20 Still, cremations at the cemetery may
have continued during the fifth and sixth centuries. Mllegrdsmarken
is unique in many respects, mostly because of the number of graves
and the amount of Roman objects, but also in view of the crema-
tion graves which predominate. These are often richly furnished but
never contain weapons.
Furthermore, recent excavations at Mllegrdsmarken have pro-
duced new evidence concerning the organisation of the cemetery,
such as possible small houses for ritual purposes and parts of tracks
running through the area.21 Although Mllegrdsmarken ceased to
be in use, the extensive cemetery located at the brink of Tange ,
halfway between Gudme and Lundeborg, must have been clearly
visible in the Migration period landscape. The prevailing impression
of Gudme-Lundegrd-Mllegrdsmarken is that of an important area
in which specialized production of luxury goods took place, aimed
as we shall see at the rituals of life and death typical of Germanic
gift-giving societies.22

18
Thomsen, Lundeborg, p. 23.
19
E. Albrectsen, Fynske Jernaldergrave IV, Bd.12 (Odense, 1971); J. Christoffersen,
Mllegrdsmarken Struktur und Belegung eines Grberfeldes, Frhmittelalterliche
Studien 21 (1987), pp. 85100.
20
L. Hedeager, Iron-Age societies (Oxford, 1992).
21
C. Madsen and H. Thrane, Mllegrdsmarkens veje og huse, Fynske Minder
(1995), pp. 7791.
22
M.J. Enright, Lady with a mead cup (Dublin and Portland, 1996); M. Hardt,
 475

Apart from being an important archaeological site, the Gudme area

also contains a significant number of place-names with allusions to
pre-Christian religion. Many of these place-names are holy, and on
the basis of such toponymic evidence the conclusion can be drawn
that this region also had religious significance. Gudme itself means
the home of the gods, i.e. the place where the ancient god/gods
were thought to live. At a distance of 1.5 to 2.5 kilometers to the
north, west and south of Gudme, there are three hills with significant
names: Gudbjerg to the west means the hill of the god/gods, Albjerg
to the south means the hill of the shrine and Galbjerg to the north
has a less clear meaning, but has been interpreted as the hill of
sacrifice,23 although an explanation of the word gal as galdr may
be more plausible. Galdur should probably rather be translated with
magic; the word galdr in Old Icelandic refers to sorcerers songs
in the sense of charm or spell. Galdur was an important element
in Icelandic witchcraft, which was practiced as late as the seventeenth
century; it indicates that in Iceland, the spoken word was the most
important instrument of supernatural power.24 In Nordic mythology
Odin himself was the great sorcerer in control of sejd, the technique
of the shamanic soul journey, who also mastered oral magic, galdra.25
This is yet another reference to chants and spells being central to
witchcraft; it also testifies to the power embedded in metric forms.26

Royal treasures and representation in the early Middle Ages, in: W. Pohl and
H. Reimitz eds., Strategies of distinction. The construction of ethnic communities, 300 800
(Leiden, 1998), pp. 25580; J. Bazelmans, Beyond power. Ceremonial exchanges
in Beowulf , in: F. Theuws and J.L. Nelson eds., Rituals of power. From late Antiquity
to the early Middle Ages. (Leiden, 2000), pp. 31176.
23
J. Kousgrd Srensen, Gudhem, Frhmittelalterliche Studien 19 (1985), pp. 1318.
24
K. Hastrup, Iceland: sorcerers and paganism, in: B. Ankerloo and G. Henning-
sen eds., Early Modern witchcraft. Centres and peripheries (Oxford, 1990), p. 387.
25
D. Strmbck, Sejd. Textstudier i nordisk religionshistoria (Stockholm and Copenhagen,
1935); M. Eliade, A history of religious ideas, vol. 2 (Chicago, 1984), p. 160; L. Hedeager,
Odins offer. Skyggger af en shamanistisk tradition i norsk folkevandringstid, Tor
29 (1997), pp. 26578; L. Hedeager, Myth and art: a passport to political author-
ity in Scandinavia during the Migration Period, in: T. Dickinson and D. Griffiths
eds., The making of kingdoms. Anglo-Saxon Studies in Archaeology and History 10
(Oxford, 1999), pp. 1516; L. Hedeager, Scandinavia (c. 500700 A.D.), in:
P. Fouracre ed., New Cambridge Medieval History I (Cambridge, forthcoming); B. Solli,
Odin the queer? Om det skreive i norrn mytologi, Universitetets Oldsaksamling,
rbok, 1997/1998 (Oslo, 1998), pp. 742.
26
K. Hastrup, Iceland: sorcerers and paganism, pp. 3889.
476

All this indicates that Galbjerg means the hill of galdring, i.e. the
hill of magic. It is of course impossible to date these place-names
with any precision, but according to philologists, the Gudhem-names
stem from the Migration and Merovingian periods. This is consis-
tent with the archaeological record.27
Gudmes great wealth suggests that this site was not just a cen-
tral place for trade and production, but one with sacred connota-
tions; a place where master artisans transformed bars, ingots, and
coins of gold into symbolic objects like bracteates and ornamented
scabbard mounts. Against this background, and also with the sacred
toponomy in mind, Karl Hauck has argued that the iconography of
the gold bracteates points to the establishment of an Odin cult in
Gudme, connected with sacred kingship.28 A motif resembling the
archetypal representation of a shaman presumably Odins jour-
ney to the Other World is the most common one on these
bracteates, but other motifs known from Nordic mythology are also
found (fig. 2).29 Initially the Scandinavian gold bracteates imitated
the Byzantine Emperors medallions, symbolically connecting Roman
Emperors and Asir kings.
I do not wish to claim that myths and stories from the Poetic Edda
or in Snorris Edda have remained unchanged until the thirteenth
century, when they were written down. In their present form, these
texts represent a pre-Christian universe that has been incorporated
into a literary Christian tradition, in the complicated way of any

27
H. Thrane, Das Gudme-Problem und die Gudme-Untersuchungen, Frhmittelal-
terliche Studien 21 (1987), p. 40.
28
K. Hauck, Gudme in der Sicht der Brakteatenforschung, Frhmittelalterliche
Studien 21 (1987), pp. 14781; K. Hauck, Gudme as Kultort und seine Rolle beim
Austausch von Bildformularen der Goldbrakteaten, in: Nielsen e.a. eds., The archae-
ology of Gudme, pp. 7888.
29
D. Ellmers, Zur Ikonographie nordischer Goldbrakteaten, Jahrbuch des Rmisch-
Germanischen Zentralmuseums Mainz 17 (1970), pp. 202, 210, 220; K. Hauck, Gtterglaube
im Spiegel der goldenen Brakteaten, in: C. Ahrens ed., Sachsen und Angelsachsen.
Verffentlichungen des Helms-Museums 32 (Hamburg, 1978), p. 210; K. Hauck,
Gudme als Kultort und seine Rolle beim Austausch von Bildformularen der Gold-
brakteaten, pp. 7888; E. Oxenstierna, Die Goldhrner von Gallehus (Liding, 1956),
p. 36; Although the time-difference between the written text and the iconography
is several hundred years, it does not seem too daring to regard the gold bracteates
as confirmation of the central religious complex and the central myths of the fifth
and sixth century traced in the Old Norse literature from the early Middle Ages,
as argued by L. Hedeager, Skandinavisk dyreornamentik, in: I. Fuglestvedt,
T. Gansum and A. Opedal eds., Et Hus med Mange Rom. Vennebok til Bjrn Myhre p
60rsdagen. AmS-Rapport 11A (Stavanger, 1999), pp. 21938.
 477

Fig. 2. Gold bracteates from the Gudme area (after K. Hauck IK 2, Tafel 229b,
IK I Tafel, 75,3b).
478

such tradition reflecting on its past from a new perspective. There

is no doubt, however, that the Poetic Edda and Snorris Edda give us
a good idea of the nature and scope of Old Norse mythology.30 Any
mythological system provides a conceptual support for social life in
traditional societies, preserving the integrity of society, 31 and the
Northern mythological system was no exception.
If Gudme was indeed the main home of the Odin cult, as has
been maintained, the central area framed by the sacred hills would
have been a place of display and communication, at the social level
as well as with the supernatural world. In this place the represen-
tation of world was given a concrete form by specialists in control
of the production process by which metal was transformed from one
shape (scrap metal, ingots, coins etc.) into another (bracteates, fittings
for swords etc.).

For the Nordic realm before 800 there is no textual evidence of any
specific locations of religious or political power, such as monasteries
or other sacred sites, cities, or royal palaces, so the archaeological
sources and the toponymic evidence provide the only basis for
analysing the concept of places of power in this area. Still, the Old
Norse literature does throw some light on certain essential compo-
nents of places of power in Scandinavia. For example, the hall
assumes great importance in the ideological universe represented in
these texts.32 Given the prominent role of the hall in Old Norse lit-
erature, it is remarkable that the word hall hardly ever turns up
in Scandinavian place-names. The reason may be that the Scandinavian
language of the time used another word, such as sal, as in Uppsala,
Onsala, Odensala or just Sal (a): the god whose name is compounded

30
P. Meulengracht Srensen, Om eddadigtenes alder, in: G. Steinsland, U.
Drobin, J. Pentikinen and P. Meulengracht Srensen eds., Nordisk Hedendom. Et sym-
posie (Odense, 1991), pp. 21728; M. Clunies Ross, Prolonged echoes. Old Norse myths
in medieval northern society, vol. 1 (Odense, 1994), vol. 2 (Odense, 1998).
31
Weiner, Myth and metaphor, pp. 5912.
32
F. Herschend, Livet i Hallen (Uppsala, 1997); ibid., The Idea of the Good (Uppsala,
1998); ibid., Halle, Reallexikon der Germanischen Altertumskunde 13 (Berlin, 1999), pp.
41425; Enright, Lady with a mead cup.
 479

with sal is always Odin, the king of the gods.33 The word sal is
often linked with zulr (thyle), the term for a particular type of leader
or priest. The thyle is regarded as a poet, i.e. a skald or storyteller:
in other words, the person who preserves the treasure hoards of mys-
tical and magical knowledge that was essential to understand the
Eddic poems. He was the cult leader who understood the cult activ-
ities and uttered the proper magical words. His main function was
to speak, whether this was to recite the verses and sacred stories, to
know and to declare the laws, or to function as the spokesman of
a king or earl during a feast, a cult festival or a legal moot.34 The
thyle was one of the specialists who controlled societys esoteric
knowledge on the kings behalf.
Apparently sal means the kings and earls assembly hall, cult
hall or moot hall: the place in which the functions of theatre, court,
and church were united.35 The sal or the hall was the centre of
the human microcosmos, the symbol of stability and good leader-
ship. The hall was also the location where communal drinking took
place, which had the purpose of creating bonds of loyalty and fictive
kinship; liquor was the medium through which one achieved ecstasy,
and thus communion with the supernatural.36 The high seat, that is,
the seat with the high-seat posts, served as the channel of commu-
nication with the supernatural world. Since the hall with the high
seat served as the geographical and ideological centre of leadership,
it is understandable why the earls and kings, as the literature tells
us, could suppress and ruin each other by simply destroying their
opponents hall.37
The multifunctional role of the hall thus extended beyond the site
itself. The hall was at the centre of a group of principal farmsteads;
it was the heart of the central places from the later part of the Iron
Age,38 which existed all over Scandinavia, as is now increasingly rec-
ognized. Apart from Gudme/Lundeborg one might mention Sorte
Muld on Bornholm, Lejre, Boeslunde, Jrlunde, Kalmargrd, Nrre

33
S. Brink, Political and social structures in early Scandinavia, Tor 28 (Uppsala,
1996), pp. 23581.
34
Brink, Political and social structures, pp. 2567.
35
See the comprehensive account in Herschend, The idea of the good.
36
Enright, Lady with a mead cup, p. 17.
37
F. Herschend, Hus p Helg, Fornvnnen 90 (Stockholm, 1995), pp. 2218;
ibid. Livet i Hallen.
38
A possible ranking of this places can be found in U. Nsman, The etnogen-
esis of the Danes and the making of a Danish kingdom, pp. 110.
480

Snede, Stentinget, Drengsted and Ribe in Denmark; Trondheim,

Borre, Kaupang and Hamar in Norway; Slinge, Helg, Birka, Up-
pkra, V, Gamla Uppsala, Hgum, Vendel and Valsgrde in
Sweden.39 Characteristicially, many of these sites are located a few
kilometres inland, relying on one or more landing places or ports
situated on the coast.40 Although this is still a matter of debate, I
believe that such central places served as a basis for some form of
politcal or religious control excercised over a larger area; the radius
of their influence went well beyond the site itself.41
In his innovative analysis of the toponymic evidence Stefan Brink42
has argued that rather than being a precisely defined site, such cen-
tral places should be understood as a somewhat larger area encom-
passing a number of different but equally important functions and
activities. Both toponymic evidence and archaeological finds suggest
that this was a recurrent pattern. This means that it is inadequate
to refer to these sites as trading sites, cult sites, meeting or thing
places, emphasizing only one of their many functions. Instead, these
locations should be perceived as multifunctional and composite sites.
In addition to their official function as trading- and market sites,
and as centres where laws were made and cults were established,
these central places were probably also associated with special func-
tions such as the skilled crafting of jewellery, weapons, clothing, and,
furthermore, with special cultic activities performed by religious spe-

39
For further details see L. Jrgensen, Stormandssder og skattefund i 3.12.
rhundrede, Fortid of Nutid 2 (1995), pp. 83110; Brink, Political and social struc-
tures, and Larsson and Hrdh eds., Centrala platser, centrala frgor.
40
Fabech, Organising the landscape, p. 43.
41
The social geographer Mats Wiedgren has discussed the concept of central
places and criticized the Scandinavian Iron Age archaeologists for having returned
to their classical hunting grounds: imports and precious metals. By focussing on
gold and its connection with aristocratic power, archaeologists have neglected the
economic and social organization of the agrarian landscape. M. Wiedgren, Kultur-
geografernas bnder och arkeologernas guld finns det ngon vg til syntes?, in:
L. Larsson and B. Hrdh eds., Centrala platser, cantrala frgor (Lund, 1998), pp. 28196.
42
Brink, Political and social structures, pp. 2358. In several articles C. Fabech
has developed this model in archaeological case studies; most recently Fabech,
Organising the landscape, and idem, Kult og samfund i yngre jernalder
Ravlunda som eksempel, in: L. Larsson and B. Hrdh eds., Centrala platser, centrala
frgor (Lund, 1998), pp. 14764. However, the model of ritual depositions in the
cultural landscape, which plays an important part in Fabechs general model, has
been the subject of debate; see: L. Hedeager, Sacred topography. Depositions of
wealth in the cultural landscape, in: A. Gustafsson and H. Karlsson eds., Glyfer och
arkeologiska rum en vnbok til Jarl Nordbladh (Gteborg, 1999), pp. 22952.
 481

cialists. These places were also the residence of particularly privi-

leged warriors or housecarls. Archaeological research has revealed a
whole range of activities in Gudme, but the picture becomes more
clear by taking the place-names into account. These demonstrate
the presence of a pagan priesthood, of military units and warriors,
of the most prominent smiths, and so on. Some of these central
places, such as Gudme/Lundeborg, originated in the Late Roman
period (200400), but the majority only emerged after 400. In fact,
it looks as if most of these multifuntional sites did not predate the
seventh century. Many of these sites remained centres of power and
of economic activity well into the middle ages.
In some respects, Gudme fits the general model of a central place
as we have defined it, but in others, it diverges. First, Gudme is
among the earliest of these places, and may even be the earliest, for
it already gained its central position during Late Roman period.
Second, Gudme is bigger and the settlement area more extended
than that of any of the other central places hitherto found in South
Scandinavia;43 its great hall, situated in the centre, is unique because
of its size and its construction. Third, the sheer amount of archae-
ological finds from the area is overwhelming; this goes especially for
the number of gold finds and superb jewellery produced by skilled
craftsmen. Fourth, the evidence of place-names connected with the
sacred is more persuasive in Gudme than anywhere. We have to
consider the possibility that Gudme was perceived as the foremost
residence of the pre-Christian god(s), and that it may have occu-
pied a unique place in the cosmology of the Nordic realm in the
middle of the Iron Age.44

G: A

To understand the role of Gudme as the central place in South

Scandinavia, in the sense that it was the place receiving most of the
desirable objects from the world outside, we also have to take another

43
Jrgensen, The warrior aristocracy of Gudme, fig. 14.
44
Gudme is suggested as the dominant centre in South Scandinavia during the
Migration Period by J. Ringtved, The geography of power: South Scandinavia
before the Danish kingdom, in: T. Dickinson and D. Griffiths eds., The making of
kingdoms. Anglo-Saxon Studies in Archaeology and History 10 (Oxford, 1999), pp.
4964.
482

problem into account. How could such objects be reinterpreted and

transformed into so-called prestige objects, that might subsequently
become embedded in the local cultural system? We should therefore
consider the possible symbolic structure underlying the production
and acquisition of valuable goods. In other words, the association of
elites with crafting and long-distance trade must not merely be under-
stood as a materialistic and economic phenomenon, but also in terms
of qualities and values prevailing within a cosmological scheme.45
Long-distance trade and exchange in the Roman and Migration
Periods have been discussed from different perspectives, such as
trade, economy, prestige-goods, gifts, war booty, payment,
and so on. Initially, Roman tablewares were manufactured in the
Empire and brought to the Nordic area either by long-distance
traders or as ceremonial gifts in the context of inter-elite alliances.
In the the Migration period, however, gold came into the barbarian
world primarily as a military payment by the Romans, to be trans-
formed into prestigious objects essential for Germanic gift-giving.
This represents an essential shift in the relations Germanic societies
entertained with the outside world a change that can be docu-
mented in Gudme.
It is highly unlikely that any prehistoric society ever saw activities
and objects associated with remote distances in a neutral light. The
common denominator of objects deriving from outside the local cul-
tural system and serving as gravegoods, is prestige goods, or lux-
ury goods. Elite graves from all over prehistoric Europe contain
foreign objects as a significant part of their display, be they from
Hallstat and La Tne (with Greek and Etrurian imports), from
the Roman period (imports from Greek and Italy), or from the
Migration and Merovingian period (with Byzantine and Frankish
objects). Elite graves in Scandinavia followed these patterns as well.
In prehistoric Europe foreign objects were closely connected with
social prestige at a more general level. How can this be explained?

Objects obtained from distant places have two things in common.

First, whether crafted or uncrafted, they all had to pass the bound-
ary between the unfamiliar world outside and the familiar inner

45
M. Helms, Craft and the kingly ideal. Art, trade and power (Austin, Texas, 1993);
E. DeMarrais, L.J. Castillo and T. Earle, Ideology, materialization, and power
strategies, Current Anthropology 37 (1996), pp. 1531.
 483

world of a given society. Furthermore, they represented unequal

access to symbols of status and authority. Mary Helms, an anthro-
pologist who has investigated how various traditional cultures inter-
pret space and distance in cosmological terms,46 has argued that
social and political power are associated with information about
strange places, peoples, and things. In all non-Western native cos-
mologies, people and things coming from afar are invested with a
range of symbolic meanings, and with qualifications such as superi-
ority, inferiority, or danger.47 In well-functioning societies, bad things
are banished and good things are acquired from outside. In her
impressive Craft and the Kingly Ideal (1993) Mary Helms argues that
places out there are represented by two axes, one situated hori-
zontally (the geographical distance) and the other vertically (the cos-
mological distance). A morally informed and cultured social group
representing the centre is surrounded on all sides by a cosmological
realm out there, which, as Helms expresses it, is believed to con-
tain all manner of visible, invisible and exceptional qualities, ener-
gies, beings, and resources, some harmful, some helpful to those at
the centre.48 When geographical and supernatural distance are thought
of as corresponding and closely related to one another, a horizon-
tal movement, away from the social centre, is also a departure into
an area that is seen as increasingly different, and therefore increas-
ingly supernatural, mythical, and powerful. Often objects acquired
from geographically distant places carry associations with ancestors
and cultural heroes.49
The point of Helms argument is that objects acquired from the
world beyond the confines of a traditional society are considered to
be powerful, and potentially beneficial either to society at large or
to its leaders. In order to comprehend how a Roman vessel or
a cowri shell may become a more powerful object than a locally
manufactured iron spear, we have to regard power not merely as a
function excercised by people, but also as an entity or quality that
may be acquired or accumulated, and as an existential reality that

46
M. Helms, Ulysses sail: An ethnographic odyssey of power, knowledge and geographical
distance (Princeton, 1988); ibid. Craft and the kingly ideal; ibid., Access to origins. Affines,
ancestors, and aristocrats (Austin, 1998).
47
Helms, Ulysses sail; Helms, Craft and the kingly ideal. The following is based pri-
marily on these two works.
48
Helms, Craft and the kingly ideal, p. 7.
49
Helms, Ulysses sail.
484

may be connected to wealth or weapons, but also to other objects.

In this sense, power is a spiritual energy enabling an individual to
interact with the forces of the natural and supernatural world. Objects
obtained from outside tend to channel and concentrate such energy,
and the individual in possession of such goods will become associ-
ated with the power with which these objects are infused.50
The fundamental quality attributed to such prestige goods or
luxury goods is that of having encapsulatd the forces present in a
dangerous universe. Long-distance travellers may have chosen par-
ticular places, such as Lundeborg, to cross the barrier between the
outside world and the inner space of society, bringing Roman table-
ware, Byzantine jewellery, and metals such as gold, silver and bronze.
Helms sees similarities between travellers, who communicate with
the dangerous but potentially powerful outside world, and religious
practitioners shamans, priests and diviners because the latter
also explore, control, transform, and exploit a realm situated outside
human society. She also includes artisans in this category, for high-
level crafting is an act of creation, equally involving some kind of
communication with the outside world; craftsmen, therefore, tend to
be liminal figures credited with supernatural powers. This must have
been particularly true if objects representing spiritual power were
crafted, as was the case in Gudme, where gold bracteates and other
emblems of religious leadersip were produced.
Locally crafted goods as well as those acquired from distant coun-
tries could thus become the tangible embodiement of numinous pow-
ers. Both the skilled metal-worker and the skilled trader acts as
intermediaries between human society on the one hand, and a greater
realm outside characterised by what Helms calls the intangible ener-
gies that must be tapped and transformed via the artisans various
skills into cultural formats that encode qualities identified with what-
ever is thought to constitute true humanness.51 By obtaining such
goods, the elite is involved in a process by which resources from
outside are brought into their society, where they are subsequently
transformed, both materially and symbolically, in order to meet local

50
D. Carmichael, Places of power: Mescalero Apache sacred sites and sensitive
areas, in: D.L. Carmichael, J. Hubert, B. Reeves and A. Schanche eds., Sacred sites,
sacred places. One World Archaeology (London, 1994), pp. 8998; E.W. Herbert,
Iron, gender and power (Indiana, 1993), pp. 15.
51
Helms, Craft and the kingly ideal, p. 15.
 485

ideological needs. Likewise, the organisation of expeditions and exper-

tise concerning shipbuilding and navigation do not only reflect prac-
tical considerations; such skills also represent the ability to mediate
successfully between ones own society and the outside world. For
obvious practical reasons, the final stages of shipbuilding often take
place on the beach; but the beach is also a liminal zone, separat-
ing land and sea, the gateway to a dangerous world beyond the own
group, but full of desirable objects.52
The argument developed by Helms is of course a matter for fur-
ther discussion, but I still find her approach stimulating and good
to think with in relation to Gudme/Lundeborg, for it offers a new
paradigm for understanding the cosmology of traditional societies,
that of pre-Christian Scandinavia not excepted. If we follow Helms,
Gudme can be understood as a central place where Roman table-
wares and Byzantine gold were acquired and transformed, while
Lundeborg, a site of shipbuilding and other kinds of skilled crafting,
served as the liminal zone between inland society and the great and
unknown world across the sea.

One of Gudmes most striking characteristics is the overwhelming

evidence of intensive crafting activities, especially those of jewellers
and blacksmiths. Metal production and craftmanship in Scandinavia
during the Iron Age are usually regarded as a neutral or even sec-
ondary affair, but to my mind, metallurgy and skilled crafting were
in fact closely connected to what in these societies was conceived of
as the quality of power. The role of metal-workers especially
blacksmiths and jewellers deserves special attention,53 for the tech-
nicalities of metallurgy and metalwork included a symbolic and rit-
ual element,54 which gave the practicioners as special status. Mastering
metallurgy meant controlling a transformation: from iron ingots to

52
Ibid., p. 21.
53
Weavers for example can be seen as skilled artisans as well, but their activi-
ties are difficult to trace at Gudme.
54
M. Rowlands, The cultural economy of sacred power, in: Les princes de la
protohistoire et lmergence de ltat; Actes de la table ronde internationale organise par le Centre
Jean Brard et lcole francaise de Rome; Naples, 2729 octobre, 1994. (Naples and Rome,
1999), pp. 16572.
486

the tools for agricultural production and the weapons on which pro-
duction, fertility, and protection or aggression depended; from ingots,
bars, and items of gold and silver into ritual objects central to the
symbolic universe of a given society.55 Blacksmiths and jewellers in
traditional societies are usually associated with power56 because they
forge the implements by which the natural and social world may be
dominated; futhermore, they create objects that mediate between
mankind and the supernatural.
To be a specialist of this kind demands not only superb skills, but
often also the possession of magic power.57 The smiths work requires
the esoteric kind of knowledge enabling him to manipulate the dan-
gerous forces unleashed in the process of transforming shapeless metal
into a finished product; this especially holds true when sacred objects
are cast, or specific types of jewellery associated with status and/or
ceremonial use. Because of the secret knowledge inherent to such
activities, smiths were specialists who were both powerful and feared.58
As Eugenia Herbert explains,59 in traditional societies the nature of
their work set smiths apart from other people. Often they are eth-
nically different, or at least regarded as others, precisely because
they mediate between the natural and the supernatural. The smith
has magical powers, often holding a high position in society.60 Although
these insights are derived from traditional African societies of the
past centuries, they seem to be of general validity for most tradi-
tional societies.61

55
In this particular case I refrain from discussing iron technology and the extrac-
tion of iron ore as such although this must have been of major importance in an
Iron Age society.
56
P. Maret, The smiths myth and the origin of leadership in Central Africa, in:
R. Haaland and P. Shinnie eds., African iron working (Oslo, 1985), pp. 7387; E. Her-
bert, Red gold of Africa (Madison, 1984); ibid. Iron, gender and power; Rowlands, The
cultural economy of sacred power, pp. 16572. However, metallurgy and skilled
metal work represent important values in the central myths of any traditional soci-
ety, not only in Africa. This counts for the Indo-European cultures as well. Judging
by Homer, classical authors, and the Celtic myth Cycles, the smiths had a high
position, however different these societies (e.g. M. Rowlands, The archaeological
interpretation of prehistoric metalworking, World Archaeology 3 (1971), p. 216).
57
Herbert, Red gold of Africa, p. 33.
58
M. Eliade, The forge and the crucible (2nd ed., Chigago, 1978), p. 99.
59
Ibid. and Herbert, Iron, gender and power.
60
As argued convincingly by Herbert, Red gold of Africa, p. 33; Herbert, Iron,
Gender and power; Helms, Craft and the kingly ideal; Rowlands, The archaeological
interpretation of prehistoric metal working, pp. 2156.
61
Ibid.
 487

In order to get a hold of the metal, the artisan often has to take
part in trading activities.62 Together with poets, troubadours, carvers,
and musicians, smiths constitute a group of specialists whose frequent
long-distance travel associates them with spatial distance and foreign
places. As such, they might gain great reputations; as Helms argues,
artisans coming from outside were often believed to be superior.
Such specialists, as well as travelling religious experts come to embody
the supernatural qualities of the world beyond the settlement. They
roam between cultivated and settled space and the wild and dan-
gerous territories beyond its pale.63
These views from anthropology and ethno-history furnish a tempt-
ing frame of reference for the interpretation of the archaeological
evidence from Gudme presented in the first section of this chapter,
and I shall pursue my argument along these lines. In Gudme as
well, artisan smiths, shamans and long-distance travellers may have
functioned as specialists in distance, concentrated in what consti-
tuted a multifuncional central place.

Given the importance of smithing and jewellery associated not only

with Gudme, but with any central settlement and big farm from the
fifth century until the late Viking Age in Scandinavia, such activi-
ties must have served a purpose. This problem may of course be
approached from a functional perspective: all big farms needed tools
and weapons, and smithing activities must have been an essential
part of day-to-day work in all non-urban, pre-industrial societies.
Obviously weapons and iron tools were primarily manifactured to
meet practical demands, but this is not true of items of gold and
silver, which met social requirements. Keeping this in mind, smithing
and the manufacture of jewellery can be expected to have a place
in the mythological world of pre-Christian Scandinavia.
This brings us back to the problem of the written sources. The
lack of Scandinavian written records before c. 1200 means that any

62
Maret, The smiths myth and the origin of leadership in Central Africa,
p. 76.
63
Helms, Craft and the kingly ideal, pp. 359.
488

conclusions drawn about pre-Christian beliefs are speculative. However,

the Old Norse myths incorporated in the Poetic Edda and in Snorris
Edda may well provide valuable information, if treated with extreme
caution; these texts of course do not necessarily fit the material evi-
dence from the Iron Age, but they may create a plausible setting.64
Central myths representing the wisdom and knowledge of the pre-
Christian world may contain core elements that remain in place
through time, although encrusted in new layers of meaning, and
adapted to new contexts. As Kirtsen Hastrup maintains, an exten-
sive pagan wisdom still existed in pre-Reformation Icelandic society,
in the guise of traditional esoteric knowledge, which then was trans-
formed into Satan-worship and witchcraft in the seventeenth cen-
tury.65 Galdur (galdring), galdraastafir (magic staves), runa (secret or
occult), skaldskapur (poetry), and hamrammr (a person who could change
shape at will), were too much rooted in everyday life and thought
to lose their role as meaningful concepts. The runic staves from
Bryggen in Bergen also indicate that the Eddic poems were still well
known in Norway by the late middle ages.66

From another perspective Margareth Clunies Ross67 has convincingly

argued that mythological texts such as the Poetic Edda, some scaldic
verses, and Snorris Edda, incorporate myths that explore fundamental
ideas with regard to life and death in Nordic societies. These ideas
retained their relevance when the majority of the population of
Iceland had become Christian. The myths were simply very good
to think with and good to use as schemas for the representation of
the major concerns of human life.68 Snorris role in writing the Edda
has been a matter of debate since long. Some scholars have seen
him as a creative literary artist, others as a far more reliable com-

64
G. Steinsland, Det Hellige Bryllup og Norrn Kongeideologi (Oslo, 1991); L. Hedeager,
Cosmological endurance: pagan identities in early Christian Europe, European
Journal of Archaeology 1 (1998), pp. 38296; ibid., Skandinavisk dyreornamentik.
L. Hedeager, Sacred topography. Depositions of wealth in the cultural landscape.
65
Hastrup, Iceland: sorcerers and paganism, p. 401
66
P. Meulengracht Srensen, Om eddadigtenes alder, in: G. Steinsland e.a.
eds., Nordisk Hedendom (Odense, 1991), p. 219; Clunies Ross, Prolonged echoes I, the
Myths, p. 139.
67
Clunies Ross, Prolonged echoes vol. 1; ibid., Prolonged echoes vol. 2, The reception of
Norse myths in medieval Iceland.
68
Clunies Ross, Prolonged echoes vol. 2, p. 191.
 489

municator of pre-Christian mythology,69 who had no obvious reason

to disparage the faith of his ancestors, although he probably did not
share it himself.70
Although Snorris work has been influenced by Christianity for
he wrote in a Christian society one may well wonder whether
he could manipulate old and familiar myths to the extent of giving
a false image of the pre-Christian cosmology still known to his audi-
ence. Within a Christianised society, pre-Christian traditions were
still kept alive, possibly perceived through a Christian perspective,
and therefore distorted. Also, we should not discount the possibility
of some invention of tradition, and of authors like Snorri being
eager to incorporate traditional myths within a new framework, stress-
ing the continuity between old times and the new world they inhab-
ited. By the way, Snorris narratives correspond in many details to
other early medieval texts from different parts of Scandinavia. In
spite of all these problems, we cannot afford to disregard Snorris
informations and the informations given by the Poetic Edda.
If one pursues these arguments, it follows that Christianity was a
historical phenomenon, not something transcending time and change.
Christianisation also meant that prior beliefs and concepts helped
to shape new varieties of Christianity. Magic, be it of a Christian
type, or of the sort defined by Christians as pagan, remained
omnipresent in the daily lives of medieval people.71 Yet I would go
further than this. We are dealing not merely with folk magic, but
with a longue dure of fundamental myths and systems of beliefs in
Iceland, which should not be regarded as the mere product of the
late pre-Christian era in Scandinavia. On the contrary, its roots must
be deeply anchored in the traditional pagan universe.72 Even if they
do not fit Gudme chronologically, it may give us some ideas about
the beliefs that informed the construction of Gudme as a sacred site.

69
In general modern scholars represent the positive opinion, i.e. J. de Vries,
Altgermanische Religionsgeschichte (Berlin, 1970); A. Holtsmark, Studier i Snorres Mytologi
(Oslo, 1964); E.O.G. Turville-Petre, Myth and religion of the North (Westport, 1975).
70
E.g. A. Hultgrd, Fornskandinavisk hinsidestro i Snorre Sturlusons spegling,
in: U. Drobin ed., Religion och Samhlle i det Frkristna Norden (Odense, 1999), pp.
10924.
71
M. de Jong, Religion, in R. McKitterick ed., The Oxford Short History of Europe
II (4001000) (Oxford, 2001), pp. 13168.
72
Meulengracht Srensen, Om eddadigtenes alder, pp. 21728.
490

The description of smithing and gold may in Eddic poetry shed some
light on the social importance of metal work in the late Iron Age.

T O N

In the Poetic Edda as well as in Snorris Prose Edda metallurgy and

skilled metal work were closely associated with dwarfs who were
imagined to mine and manufacture underground. In the worlds first
age, the happy Golden Age, the gods had special talent for skilled
metallurgy. But when this talent was destroyed by the arrival of giant
women from Utgard, the gods had to create the dwarfs and place
them in the outside underground world, among stones and
cliffs, where they controlled precious metals and produced much cov-
eted objects. Subsequently the dwarfs became the gods craftsmen,
creating technical wonders for their masters, sometimes willingly and
sometimes under duress. However, the gods remained dependent on
the dwarfs, who crafted the precious objects ensured success in the
gods struggle against the Giants: Odins spear Gungnir and his
golden ring Draupnir, Thors hammer Mjollnir, and Freyjas golden
necklace. Moreover, the dwarfs were credited with magical powers.73
Like the Asir in the Golden Age, the dwarfs constituted a male soci-
ety unable to reproduce itself.
This is how smiths, forgers, and jewellers are represented in the
northern mythology. They were all dwarfs, they lived apart, they
were in possession of magical powers, and they formed a male soci-
ety. Although Snorri designates Odin and his priests as forgers of
songs,74 neither Odin nor any of the other Asir gods were in com-
mand of forging. But the Old Norse texts also contain other smiths.
The most famous text in this respect is Volund the Smith, a lay in the
Poetic Edda.75 This is an Old Norse version of the widely known story
of the master smith, adapted to the code of the Nordic apophthegm.
Volund is the tragic figure of the hero-smith, captured and mauled
by the king, robbed of his gold and sword, held prisoner and forced
to create high quality weapons and jewellery for his captor. With

73
R. Simek, Dictionary of Northern mythology (Cambridge, 1993), p. 68.
74
Eliade, The Forge and the crucible, p. 98.
75
The Poetic Edda, transl. L.M. Hollander (Austin, 1994).
 491

revenge as its central theme, the poem must have provided a logi-
cal and intelligible story line for its Scandinavian audience.
The ability to grow wings and fly like the wind to escape the
greedy king, as Volund did, is typical of the master smith who could
change shape like the shaman to mediate between human society
and the supernatural world.76 Volunds pedigree and family relations
are a good illustration of the smiths position in the cosmological
world of the Old Norse texts. As son of a Finnish king his origin
was clearly defined as out there; in the Old Norse sources a Finnish
(or Saami) background always indicated someone who represented
dangerous magical forces from outside. Volund, who is called king
of the elves, was married to a valkyrie, a giant woman from the
outside world. She was a skilled weaver, herself daughter of a king
and in control of shape changing. Although Volund is not a dwarf,
he is no human being either; he is most at home in the outside and
dangerous world from where he was captured by a human king and
brought into society. His forge is situated on an isolated islet, and he
himself is a feared person in control of the gold.77 Although married,
he has no children, so he does not belong to any family group as
a human being, set apart from society. As the master smith in con-
trol of gold as well as skilled crafting, he fabricates prestigious objects
essential for the kingly ideal. Like the Asir gods, the worldly king is
dependent on the smith to come across these emblems of royal power.
In other words, the king depends on Volund the Smith, his captive,
to retain his royal power.
One more personified smith is known from the mythological cir-
cle of the Poetic Edda, namely Regin from the lay Reginsml. This is
part of the great epic cycle of the Volsunga, which tells the story of
the fall of the Burgundians after the attack by the Huns in 437.
Known from a number of Old Norse Sources, the Volsunga Saga
became the core of the Niebelungenlied in a Christianised German ver-
sion from around 1200. In this epic cycle about Odins grandchild
Volsunga and his descendants, Regin the Smith is an important,
although subordinate character. His family was composed of a father
(no mother is mentioned) and two brothers (no sisters); the father,

76
An element of shamanism was present on Iceland only in the late Middle Ages
(Hastrup, Iceland: Sorcerers and paganism, pp. 3889).
77
A. Bksted, Nordiske Guder og Helte (Copenhagen, 1990), pp. 21620.
492

Hreidmar, was an odd person who knows magic; one brother, Utter,
had the shape of an otter (and was killed by the god Loki), and the
second, Ffnir, changed himself into a dragon to guard the gold
treasure. In the story Regin acts like a human being and travels,
like human smiths were supposed to do, to a foreign king to become
his masters smith. Later he went on to another ruler, Volsungs son
Sigurd, who was a famous war-king. Regin is the only one who
knows how to forge a sword with the necessary (magical) power to
kill Ffnir, and he knows the right magical acts to perform before
the fight becomes succesful. With this sword named Gram, Sigurd
was able to kill the dragon Ffnir, Regins brother, and lay his hands
on the gold.
Although Regin at first sight behaves like a human being, he is
not an integrated member of human society. He is a long-distance
traveller and a skilled artisan smith, he travels between realm of
kings, he masters magic, and his brothers master shape changing.
Even the strongest king is dependent on him. Furthermore, there
are no women present in his family, neither mother, nor sister or
wife, and he has no children. He is a stranger among humans, a
liminal figure who partly belongs to the world outside.
To sum up, such skilled smiths, whether dwarfs or men, have cer-
tain specific traits in common. They all belonged to the realm out-
side human society; they were all males and they were for social,
not biological reasons unable to reproduce themselves. By way
of magic, the objects they forged were essential to the power posi-
tion of the elite, whether gods or human kings. Last but not least
the smiths were, in one way or another, skilled long-distance trav-
ellers; they mediated between the settled heartland of human soci-
ety and the dangerous outside world. In all, they seem to represent
structures and concepts specific to Nordic mythology.

G O N

In Volsunga Saga treasures of gold generate the greed that constitutes

the main story-line. In the Old Norse sources gold and gold treasures
regularly play a central role in the construction of stories. Time and
again we meet the disastrous greed for gold as an archetypal theme
in myths and stories; here and in other heroic tales, such as Beowulf,
 493

Saxos Gesta Danorum, and Snorris Ynglinga Saga, the highly ritualised
competitive gift-giving system endows the gold with authority and
power.78 Gold itself is personified in the name Gullveig, which means
golden-drink, golden-intoxication or golden-power; comprehen-
sively, it means as much as the the personified greed for gold.79
Gold was a potent vehicle of cultural values. Within the same con-
ceptual framework gold could function as a medium of power, of
art, and of exchange.80 The amount of gold treasures from the fifth
century in Scandinavia appears that if it confirms this general approach.
The Golden Age of Scandinavia is the Migration Period. Immense
quantities of gold were deposited in the fifth and sixth centuries, in
the course of only a few generations.81 The written sources, whether
the Old Norse ones or texts from continental early medieval Europe,
yield the impression that gift-giving was the crucial instrument in
creating and upholding political alliances. Movable wealth with strong
symbolic connotations were the most prestigious gifts in this highly
ritualised process.82 Much gold and silver, swords and other presti-
gious good must have circulated as gifts without leaving any traces
in the archaeological record.83 If the strategy of gift-giving included
an element of competitive display, however, gift-giving was more

78
Cf. M. Mauss, The gift (London, 1990 [1950]), pp. 17, 603; Enright, Lady
with a mead cup; Herschend, The idea of the good; Bazelmans, Beyond power. Ceremonial
exchange in Beowulf , pp. 31176, and H. Hrke, The circulation of weapons in
Anglo-Saxon societies, in: F. Theuws and J.L. Nelson eds., Rituals of power. From
late Antiquity to the early Middle Ages (Leiden, 2000), pp. 37799; latest: J. Bazelmans,
By weapons made worthy. Lords, retainers and their relationship in Beowulf (Amsterdam, 1999).
79
The name Gulleveig, however, is known exclusively from Voluspa (21 and 22)
in the Poetic Edda.
80
Herbert, Red gold of Africa, pp. 3012.
81
An equal depositional pattern characterise the Viking Age where the hoards,
however, mainly consist of silver objects (Hedeager, Sacred topography. Depositions
of wealth in the cultural landscape, pp. 22952).
82
Cf. J. Bazelmans, The gift in the Old English epic Beowulf . Lecture given
on: Theory and method in the study of material culture, Leiden 31/82 /9 1992; ibid.,
Beyond power. Ceremonial exchange in Beowulf ; R. Le Jan, Frankish giving of
arms and rituals of power: continuity and change in the Carolingian Period, in:
F. Theuws and J.L. Nelson eds., Rituals of power. From late Antiquity to the early Middle
Ages (Leiden, 2000), pp. 281309; Enright, Lady with a mead cup; Herschend, The idea
of the good.
83
Cf. the discussion in F. Theuws and M. Alkemade, A kind of mirror for men:
Sword depositions in late antique Northern Gaul, in: Theuws and Nelson eds.,
Rituals of Power, pp. 40176.
494

likely to play a central role in political strategies; in these cases, we

should expect to find some evidence of the ritualised use of these
artefacts in hoards and in graves.84
According to the early written evidence gold treasures and the
powerful enchantments were associated with members of the upper
social stratum; conversely, folkloristic treasure legends from later peri-
ods feature people of a much lower social standing. These later tales
contain an element of ludicrousness never encountered in the Scan-
dinavian legends from the early middle ages and the late Iron Age,
where the value to the treasures is bound up with the notion of
faith. Gold represented its owners honour and riches, and as such
it was equivalent to happiness. Stealing a treasure did not only mean
robbing someone of his riches, but also to steal his good fortune,
and thus condemning him to a dismal fate. For this reason, those
who managed to steal a treasure were struck by dire punishment.85
To sum up, objects of gold were central to political strategies pri-
marily because such treasures had been acquired by honourable and
daring acts performed in far-away places. In the late Iron age and
early Middle items of gold represented the honour and respectabil-
ity of the owner. To secure or maintain dominance in the social
hierarchy of early medieval societies, gold had to be appropriated
and controlled by the elite. By the added value of highly qualified
artisans, however, gold was transformed into something that embod-
ied values crucial to elite identities in the Nordic realm.

A central place with sacred functions represents the whole universe

in symbolic form; it is deliberately constructed as the centre of the
universe, be it a Christian cathedral or a pagan cult site organized
around a sacred pool, a world tree or the like, as Mircea Eliade
made clear in several publications.86 Byzantine churches, it has been

84
J.C. Barrett, R. Bradley and M. Green eds., Landscape, monuments and society.
The prehistory of Cranborne Case (Cambridge, 1991), p. 240.
85
T. Zachrisson, Grd, Grns, Gravflt. Sammanhang kring delmetalldeper och runste-
nar frn vikingetid och tidigmedeltid in Uppland och Gstrikland. Stockholm Studies in
Archaeology vol. 15. (Stockholm, 1998), ch. III.
86
M. Eliade, Patterns in comparative religion (London, 1997); ibid., The sacred and the
profane. The nature of religion (New York and London, 1987).
 495

argued, embodied all the features of the Christian universe. According

to Eliade, citing historians of church architecture, the four parts of
the interior of the church symbolise the four cardinal directions. The
interior of the church is the universe. The altar is Paradise, which
lay in the east. The imperial door to the altar was also called the
Door of Paradise. During Easter week, the great door to the altar
remains open during the entire service; the meaning of this custom
is clearly expressed in the Easter Canon: Christ rose form the grave
and opened the doors of Paradise unto us. The west, on the con-
trary, is the realm of darkness, of grief, of death, the realm of the
eternal mansions of the dead, who await resurrection of the flesh
and the last judgement. The middle of the building is the earth. As
Eliade argues, the earth is rectangular and is bounded by four walls,
which are surmounted by a dome.87
Eliades views on Byzantine churches may be useful to our ques-
tion: how can a sacred place be organized to repeat the paradig-
matic work of the god(s)? In Eliades terminology, a Byzantine church
was a central place for rituals, incorporating an image of the cos-
mological world, as sacred placeces always do, be they pagan or
Christian. All the constructions associated with sacrality symbolize
the entire universe, and this symbolism also extends to the appar-
ently secular part of the settlement.88 In Lund, the see of the Danish
archbishop in Scania from the twelfth century onwards, the whole
Christian world was deliberately replicated in the city. The topog-
raphy of the churches built after Lund became an archbishopric in
1104 mirrored the supposed location of important saints graves in
the Christian world. The cathedral was situated in the centre of the
city (like Jerusalem in the Christian world). To the east, churches
were built that were dedicated to patron saints from Asia; in the
western part of the city the patron saints were European ones; in
the northern part of the city the main patron saint was St Olav,
buried in Trondheim, in the far north of the Christian world. Thus,
Lund was constructed as a sacred city, a microcosmos of the Chris-
tian world.89

87
M. Eliade, The sacred and the profane, pp. 612, citing H. Sedlmayr, Die Entstehung
der Kathedral (Zrich, 1950), p. 119.
88
M. Eliade, Patterns in comparative religion, pp. 36785.
89
And the following: A. Andrn, Vrlden frn Lunds horisont, in: C. Wahl
496

A Christian topography apparently emerged all over the Christian

world. Pilgrimage in late antiquity and the early middle ages inte-
grated the Holy Land and its sacred sites into a new Christian land-
scape. In this landscape the places of the saints (loci sanctorum) were
the real landmarks, because they established a connection with the
heroic history of the Church.90 This organisation of the landscape
according to a Christian cosmology has also been investigated in
connection with the Burgundian monastery of Cluny. Its Christian
landscape was no neutral configuration. Instead, it was charged with
affective significance; within this particular landscape Saint Peter was
a familiar and regular landowner.91
Christianity established itself in a Roman world full of sacred places
and sacred spaces, and had to compete by creating its own sacred
topography.92 In early medieval Scandinavia Christianity was grad-
ually incorporated into the cosmology of a pagan Northern world,
after centuries of contact and mutual influence prior to the begin-
nings of a formal conversion around the year 1000. As Anders Andrn
has noted, in the course of this process the whole landscape, includ-
ing the parishes with their churches and manor houses, became
organised according to a Christian topography.93
The creation of sacred places in pre-Christian Scandinavia must
have followed the pre-Christian cosmology, of which, however, very
little is known. In a society without any form of central public power,
such as pre-Christian Scandinavia, where a precarious peace had to
be constantly negotiated, the most important institutions were the
home, the hall, and the thing, where social and legal negotiations
took place. According to the sagas, these institutions were the sacred
foundations of society, the focal points in the topographical structure

ed., Metropolis Daniae. Ett stycke Europa (Lund, 1998), pp. 11730; ibid., Frn antiken
till antiken, in: S. Thorman and M. Hagdahl eds., Staden, himmel eller helvete (Stockholm,
1998), pp. 14293; ibid., Landscape and settlement as utopian space, in: C. Fabech
and J. Ringtved eds., Settlement and landscape. Proceedings of a conference in rhus, Denmark,
May 47 1998 (rhus, 1999), pp. 38394.
90
De Jong, Religion.
91
B. Rosenwein, To be the neighbor of Saint Peter. The social meaning of Clunys prop-
erty, 9091049 (Ithaca and London, 1989), p. 203.
92
Cf. especially R.A. Markus, The end of ancient Christianity (Cambridge, 1990);
also De Jong, Religion.
93
A. Andrn, Vrlden frn Lunds horisont; ibid., Frn antiken till antiken,
in: S. Thorman and M. Hagdahl eds., Staden, himmel eller helvete (Stockholm, 1998),
pp. 14293; ibid., Landscape and settlement as utopian space.
 497

of the Icelandic universe in the early middle ages. The most common
locations for fights and open attack were outside the precinct of the
farms (utangards): in woods, near rivers and in areas bordering on the
wild interior, that is, the dangerous world outside. An attack on
the hall or the home, or to take up arms at the thing all counted
as the worst possible outrage in Icelandic society. The world depicted
in the sagas knew a distinction between the sacred and the profane,
but it was radically different from that of the modern secularized
world view.94 Spatial organization in early medieval Scandinavia was
never perceived as neutral. On the contrary, land and settlement
were organized in accordance with a clear perception of centres of
the universe. This is to be supported by Frands Herschends study
of archaeological and textual evidence of the hall in late Iron Age
Scandinavia, including Beowulf and The Fight at Finnsburg, as well as
the sagas.95 These sources all reveal that royal power could not be
exercised without a hall, however temporary such a structure may
have been. To some degree, the struggle for power among the lead-
ing families revolved around destroying each others halls, and there-
fore the backbone of the familys power. This was political rather
than economic warfare, with destruction as its main purpose, not
plunder.96

To sum up, landscapes and settlements in the early middle ages and
the late Iron Age, be they (arch)bishoprics, churches or manor houses/
halls, were no neutal configurations, but organized according to a
specific symbolic meaning. This fits the general explanation of sacred
places and sacred spaces offered by Mircea Eliade. A sacred space,
whether it is a city, a temple, a church, a monastery, a hall, or sim-
ply a house, can be a transcendent space constructed as the centre
of the universe; it was thus sharply differentiated from the profane
world surrounding it.97
In a tentative manner for our insight into pre-Christian religion
remains hypothetical it has been suggested that Gudme was the

94
V. Iason, Topography and world view, in: Njls saga, in: S. Hansson and
M. Malm eds., Gudar p jorden. Festskrift till Lars Lnnroth (Stockholm, 2000), pp.
1315.
95
F. Herschend, Striden i Finnsborg, Tor 29 (1997), pp. 295333; ibid. The
idea of the good, pp. 1462.
96
Herschend, The idea of the good, p. 37.
97
Eliade, Patterns in comparative religion, pp. 37980.
498

main centre of the Odin cult during the period when this site
flourished, or at least during the fifth and sixth centuries. Three
sacred hills the hill of the god/gods, the hill of the shrine, and
what has been interpreted as the hill of magic (galdring) frame
the central locality, a place that like Christian churches served
as a channel to the sacred.98 Still following Eliade, the construction
of Gudme as a sacred place reflects, and, to some extent, also reit-
erates the creation of the world.

A: H

It is far from clear what the pre-Christian universe in Scandinavia

looked like, but there are some common features attested in the Poetic
Edda as well as in Snorris Edda that are worth exploring, however
tenuous the connection with Gudme itself may be. Yet we should
not discard what we know from textual evidence; it may offer a
plausible context, if not a certain one, for our interpretation of
Gudmes significance.
At the heart of the Norse literature and its view of the world is
the world-tree, Yggdrasill; it is an evergreen ash whose roots reach
out to the ends of the inhabited universe. Mankind lived under one
root, the giants under another, and Hel, the realm of the dead, is
located under the third. Beneath Yggdrasil, according to Voluspa (19)
lies Urds well, where the gods should meet and hold council before
Ragnarok;99 according to Snorri100 this is the location for Mimirs
well, which hides all wisdom and understanding in the world.101 All
nine worlds are mentioned more than once, but never clearly specified.
The Asir gods lived in Asgard, the Vanir in Vanheim, and mankind
inhabited Midgard. Beyond the human world lies Utgard, home of the
giants and demons; in the east, separated by rivers, lies Jotunheim,
the world of the giants. In the north and beneath Midgard, is Hel,
the residence of the those who died on land of illness or of old age.
Whereas Alfheim is the world of the elves, earth is that of the dwarfs.

98
Ibid., pp. 36785.
99
Grimnismal 29, 30.
100
Gylfaginning 14.
101
The descriptions in the Poetic Edda and in Snorris Edda are not completely
identical; for juxtaposition see for example: Simek, Dictionary of Northern Mythology.
 499

There are two further worlds, possibly, one inhabited by the dead
heroes in Valhall and the other by the most powerful gods, called
the Holy and Mighty Ones in Alvissml in the Poetic Edda. It is not
clear, however, whether all these various worlds were oriented towards
the world-tree, and it has proven impossible to come up with a con-
vincing blue-print of Scandinavian views of the other world.102
Furthermore, the two axes of the cosmic world, one horizontal and
one vertical, are a matter of debate.103
In old Norse texts the representation of Asgard, home of the gods,
yields many problems of interpretation. Snorri is the one who fre-
quently mentions Asgard and gives the most detailed description in
Gylfaginning in his Edda, and in Ynglinga Saga.104 Apart from being
part of a didactic work about the art of scaldic poetry, the Gylfaginning
is also a systematic presentation of pre-Christian mythology, as I
argued above. In the following I shall briefly describe this cosmic
world of the North.
Although in this elusive Nordic cosmology the Yggdrasill is the
undisputed centre of the universe, Asgard figures as the home of the
gods and the residence of the Asir. A giant built Asgard on Idavoll;
in Asgards centre lies Hlidskjalf, Odins high seat,105 from where he
overlooked the whole world. The gods had a temple, Gladsheim, and
a separate hall for the female Asir, Vingolf. Gladsheim, the bright
home was Odins residence,106 and maybe also that of Hlidskjalf;
furthermore it harboured Valhall, where Odin gathered the warriors
slain in battle. In Gylfaginning (13) Snorri says Gladsheim was the
temple of Odin and twelve other gods; inside and outside, it was
made of gold, and it was the best and greatest building in the world.
Another crucial element of Idavoll and the only other building men-
tioned was the forge. In the beginning, hammers, anvils, and tongs
were created. From then onwards, the gods themselves were able to
produce all the implements they needed. They forged iron ore, made
woodcarvings and had sufficient gold to contruct their dwellings, and
even their furniture, with gold. As I said earlier, this age is called

102
H. Ellis Davidson, The lost beliefs of Northern Europe (London, 1993), p. 69.
103
J.P. Schjdt, Horizontale und vertikale Achsen in der vorchristlichen skan-
dinavischen Kosmologie, in: T. Ahlbck ed., Old Norse and Finnish religions and cul-
tic place-names (Stockholm, 1990), pp. 3557.
104
Snorris Edda, Gylfaginning 2, 8, 9, 41; Snorri Sturluson, Ynglinga Saga 2, 5 and 9.
105
According to the introduction to Grmnismal, Skrnismal and Gylfaginning 16, 49.
106
According to Grmnismal 8.
500

the Golden Age, the happy first age of the world, before the arrival
of women from the dangerous outside world of Utgard, which meant
that the gods lost their skills as artisans, and their control over the
precious metals.
In a manner of speaking, Asgard is a subordinate centre within
the Scandinavian view of the universe. By settling on Idavoll, where
a giant built Asgard, the gods founded the world of the Asir. As I
have argued earlier, the skilled and powerful carpenter who created
Asgard belonged to the world outside. Judging by its impressive hall,
Asgar was also the focus of ideals of kingship. From his High Seat,
the link between earth and heaven, Odin, the hall-owner, was in
contact with the outside world through his shamanistic helping spir-
its, the two ravens. Asgard was also a place where skilled crafting
took place, particularly metal work; at first the gods had unlimited
time for it, and also boundless access to gold. On top of this, Valhall
is the place for Odins hird (armed followers) of human heroes. The
hall is covered by a roof of spears and shields, and armour is piled
on its benches.107
According to Old Norse tradition, Asgard lost its Paradise-like sta-
tus after the war that ended its Golden Age. From then on, the Asir
lost control of the highly skilled crafting that had been their mono-
poly; they also lost direct access to all precicious metals, including
iron, copper, silver, and gold. As a remedy, the myths explain, the
gods created the dwarfs, who were now to become the skilled arti-
sans in charge of iron and precious metals. The Golden Age became
what it is throughout human history: a paradise lost, and the object
of intense nostalgia. Asgards original splendour was eclipsed forever,
for the Asir themselves had become dependent on dwarfs for gold,
skilled crafting, and magical treasures. Although its status declined,
however, Asgard remained the central part of the universe accord-
ing to Northern mythology.

G: A

In the Christian world of the middle ages, Palestine, Jerusalem, and

the Temple represented the centre of the world; the rock on which

107
Grmnismal 810, 1826; Gylfaginning 3740.
 501

the temple of Jerusalem was built, was the navel of the earth. Sacred
places in Christian Western Europe all had an inner sacred space,
inaccessible to the uninitiated, such as the altar in any church, or,
in a monastery, the claustrum, i.e. the secluded space only acces-
sible to munks/nuns. Jerusalem/Paradise represented a central ideal;
in the ninth-century Plan of St Gall, the monastic choir was called
Paradisum.108
A twelfth-century visitor to Jerusalem, the Icelandic pilgrim Nicolas
of Thverva, wrote on the Holy Sepulcher: The centre of the world
is there; there, on the day of the summer solstice, the light of the
sun falls perpendicularly from Heaven.109 The Apocalypse, widely
read and commented upon in medieval Europe, gives a detailed and
enticing description of the heavenly city of Jerusalem: So in the
Spirit he carried me away to a great high mountain, and showed
me the holy city of Jerusalem, coming down out of heaven from
God . . . it had the radiance of some priceless jewel, like a jasper,
clear as crystal. It had a great, high wall, with twelve gates . . . The
wall was built of jasper, while the city itself was of pure gold, brought
as clear glass . . . The twelve gates were twelve pearls, each gate being
made from a single pearl. The streets of the city were of pure gold,
like translucent glass.110
In some ways, Asgard may have been Scandinavias heavenly
Jerusalem in the late Iron Age, an ideal world that had once been
lost, but which also might be retrieved. If Gudme was a sacred place,
the home of the gods, as we have argued earlier, it may indeed have
been constructed to represent the centre of the world and a cosmic
moral order, with the Asir gods in mind.

If we pursue this argument, a possible context for Gudme begins to

emerge. Something resembling the centre of Asgard Gladsheim,
according to Snorri the best and greatest building in the world and
the hall of Odin may have been on the minds of those who built
the central hall of Gudme. With its 500 square meters it is the largest

108
Cf. W. Horn and E. Born, The plan of St Gall. A study of the architecture and econ-
omy of, and life in a paradigmatic Carolingian monastery, 3 vols. (Los Angeles/London,
1979), with elaborate reproductions of the Plan of St Gall.
109
L.I. Ringbom, Graltempel und Paradise (Stockholm, 1951), p. 255, quoted after
Eliade, The sacred and the profane. The nature of religion, p. 40.
110
Parts of Apocalypse 21, 1021, cited from A. Andrn, Landscape and set-
tlement as utopian space, p. 351.
502

building known from Denmark before the Viking Age, constructed

with a measure of technical knowledge without any precedent in
local tradition. Together with a smaller hall (Vingolf, the halll of the
female Asir?) it represents a complex and extremely accomplished
building that was most likely created by skilled craftsmen who were
outsiders as also held true of the mythological Gladsheim. Glads-
heims centre was Odins High Seat, from where he surveyed the
entire world. In Gudme, the High Seat in the hall must have been
a similar centre, which connected divine and royal power. From this
elevated place, the king had a privileged view of the supernatural
world, and access like Odin to the secret knowledge essential
to his authority.111
The hall in Gudme is situated in a location held by archaeolo-
gists to be the workshop area because of the many finds of work-
shop material, especially from metal work. 112 In a traditional
archaeological view such workshop areas and workshop produc-
tion are treated as marginal to social and political life, but to my
mind, this interpretation is too narrow. Skilled crafting, especially
forging and the work of jewellers and probably woodcarving as
well were the hallmark of political and ideological authority in
the traditional societies I have discussed earlier. In this process the
ideal of cosmic order was re-created and re-expressed in a tangible
form.113 For this very reason, Old Norse mythology situated the work-
shop area close to the hall. Highly skilled metal work was not merely
a craft; it was an integral part of political and religious power, and
something closely linked to ideals of royal authority.
The excavations in Gudme have shown that the big hall and the
workshop area were located in the central and southern part of the
settlement; the dwellings of the high-ranking warriors, however, were
situated to the north of this area.114 In the Old Norse mythology
Odins hird of (dead) human heroes lived in a separate hall, Valhall,

111
This is widely accepted among Scandinavian archaeologista and historians of
religion. It was first invented by Gro Steinsland (in historiy of religion) and Frands
Herschend (in archaeology), based on textual and archaeological evidence. G. Steins-
land, Det Hellige Bryllup og Norrn Kongeideologi (Oslo, 1991), ibid. Eros of Dd
de to hovedkomponenter i norrn kongeideologi, in: Studien zum Altgermanischen.
Festschrift fr Heiko Uecker (Berlin, 1994), pp. 62647. Herschend, Livet i Hallen,
pp. 4959; ibid. The idea of the good, pp. 2562.
112
Jrgensen, The warrior aristocracy of Gudme.
113
Eliade, The forge and the crucible; Helms, Craft and the kingly ideal, p. 180.
114
Jrgensen, The warrior aristocracy of Gudme, p. 212.
 503

situated in that part of Asgard which is close to Gladsheim. Although

this is highly speculative, Valhall may be located to the north, for
this was where Norse mythology situated the realm of the dead.115
The high-ranking warriors living in Gudme may have been dedi-
cated to Odin, as high-ranking warriors from the Viking Age are
known to have been.
Continuing this attempt to make sense of the topography of Gudme,
the next element to be metioned is the lake is in the western part
of the central settlement, and some springs connecting Gudme lake
with Gudbjerg to the west and Galbjerg to the north. Careful inves-
tigation has yielded no indication whatsoever that the lake was used
for sacrificial purposes. In the Old Norse mythology, the springs
reflected the significance of the mythical springs to which Urds well
(the well of fate) and Mimers well (the spring of wisdom) count,
rising from below the roots of Yggdrasil and may as such belonge
to the centre of the cosmic world. This is the place where the gods
hold council, and Mimers well is known as the source for Odin to
achieve his wisdom.
There are other streams in the Gudme area, however. Tange
arises near the sacred hill of Albjerg, to the south of the central set-
tlement area. It passes Mllegrdsmarken cemetery on its way to the
coast at Lundeborg. This cemetery, which is by far the largest in
Denmark in prehistoric times, is located halfway between Gudme
and the coast, on the northern bank of Tange . Keeping Nordic
mythology in mind, such a great cemetery must have been associ-
ated with the realm of the dead, the world of Hel, where those who
died on land, of natural causes, were buried. Snorri situated it some-
where in the north, separated from Midgard by rivers, so one needed
to cross a bridge in order to get there.116 In his Edda Snorri identifies
Niflheim with Hel,117 a mythical place in the icy north. From this
perspective, Mllegrdsmarken is located between the centre of the
world (Gudme) on the one hand, and the outside realm, where
Utgard is to be found, on the other.
Lundeborg on the coast, the transitional zone between civilisation
and a threatening world out there of giants, demons and chaos,

115
I.e. Gylfaginning 48; in some early texts, however, Valhall was thought of as
part of Hel: Simek, Dictionary, p. 54.
116
Gylfaginning 48.
117
Gylfaginning 33.
504

was the place where long-distance travellers entered inner space, the
domain of the familiar. It was the transformative, liminal zone between
land and sea where prestige goods from beyond entered society as
well as a place where specific kinds of skilled crafting took place,
such as extensive repairs to ships.118 Organising expeditions and mas-
tering shipbuilding and navigation are all prerequisites for skilled
long-distance travelling, and therefore part of the process of bring-
ing resources of ultimate cosmological qualities into society.119
To enter Gudme from the coast, from Utgard, may have entailed
a process of initiation. Gudme, as a sacred place associated with
myths concerning the home of the gods, must have been anxiously
guarded against unwanted incursions. A sacred place like Gudme
was both accessible and inaccessible, a place of great repute, that
was also forbidden to the uninitiated, and for this very reason a
powerful model to emulate; this is a characteristic that Gudme shares
with many other sacred places, such as the monasteries discussed in
this volume. The entrance to this secluded zone may have been the
stream Tange passing through the realm of the dead on the north-
ern bank, and with its source close to the sacred mountain Albjerg,
the hill of the shrine, south of Gudmes central area.

Myths and symbols, rituals and processions and people in need

of these were all part of what kept sacred places alive. Some of
these stressed the difficulties of entering the centre without coming
to grief, others made the area accessible.120 Like any other sacred
place, Gudme must have been a scene where cultic dramas were
played out, encapsulating the hopes, desires and expectations of those
involved. This left some tempting traces in later literature, of which
I am admittedly making the most.121 There is no contemporary doc-
umentation for such cultic dramas, but Old Norse literature at least
gives us a plausible, albeit problematic, frame of reference to imag-
ine such rituals. Thinking along these lines, I would say that enter-
ing Gudme was a passage through the entire cosmic landscape that

118
P.O. Thomsen, et al., Lundeborg en handelsplads fra jernaldere. (Svendborg,
1993), p. 73.
119
Helms, Craft and the kingly ideal, p. 21.
120
Eliade, Patterns in comparative religion, pp. 3824.
121
W. Grnbech, Kultur und Religion der Germanen, Bd. 2 (Stuttgart, 1954), pp.
264341.
 505

ranged between Utgard and Asgard, the outside and the inside. Put
differently, those who arrived in Lundeborg, after a long and ardu-
ous voyage across the sea, were then taken, by gradual stages, to
the impressive hall in Gudme, the home of gods and kings.
Setting apart a location as a sacred place implies a sudden inter-
ruption between the sacred and the profance; by definition, a sacred
place is unlike its surroundings, though its Hinterland is usually
connected with the centre. Men and women constructed such sacred
spaces according to patterns they believed to be divinely inspired.122
In this sense, Gudme may have been a reconstruction of Asgard;
the enormous amount of gold found in Gudmes centre as well as
in its surroundings suggests that those who built this complex cen-
tral place perceived it as a sacred place, and possibly as a replica-
tion of Asgard, its divine counterpart. Treasure played an important
role in all this. According to the Old Norse mythology, gold and
silver deposited in the Underworld were not meant to remain out
of sight forever; these were divine treasures, furnishing jewellers with
the necessary gold for their artisan work, the tangible symbols of
power. In the human world of Gudme, something similar may have
pertained: treasures were deposited, that is, put underground; how-
ever, they may still have been visible, at least to the initiated.123 If
so, these treasures mediated between the Other World and humankind,
generation after generation, forming the cosmic landscape and pro-
tected it from the dangerous forces of the outside world.124 These
treasures could as well be meant for the skilled work of jewellers,
transforming gold into objects who were essential to the power posi-
tion of the elite.

C: A

In this chapter I have developed a tentatative model that will hope-

fully add to a better understanding of Gudmes underlying structure,
and of the complexity of such a central place in Scandinavia dur-
ing the late Iron Age. Too little imaginative thought has hitherto

122
Eliade, The sacred and the profane, pp. 2067.
123
Hedeager, Sacred topography.
124
G. Wiker, Gullbrakteatene i dialog med naturkreftene. Unpublished ph.d. thesis
Oslo University (Oslo, 2000).
506

been devoted to the implications of Gudmes toponomy, gold finds

and other extraordinary features. By focusing on Gudme as a sym-
bolic constructed place that represented specific concepts of cosmo-
logical order, I have tried to extend the explanation beyond the
traditional references to trade, power, richness, and so on. In
order to do so, I have had to face a basic problem, namely that the
archaeological evidence and the written sources are not contempo-
raneous. But does this mean that all texts should consequently be
jettisoned as unreliable? I believe not, provided one realises that in
their written form they only originated in Scandinavias early Christian
period, and therefore do not reflect anything like an original and
untainted paganism. The introduction of anthroplogical perspectives
on the nature of trade and crafts in traditional, non-industrialized
societies, also furnishes the archaeologist with some helpful parallels.
I am well aware that what I have performed is a highly speculative
operation, but I am equally convinced that much is gained by also
applying our well-informed imagination to the interpretation of com-
plex sites such as Gudme. We urgently need to get beyond the tra-
ditional circular arguments about gold meaning power and vice versa.
On the one hand I have discussed Gudme as an extraordinary
place; on the other I have stressed that it has many features in com-
mon with other places in Scandinavia that have also been called
central places or places of extraordinary power. Gudme may in
fact be the key to a better understanding of comparable sites, for
this archetypal sacred place, embodying the nostalgia for Asgrad,
is likely to have served as a model for emulation throughout Scan-
dinavia, albeit with more humble results. All these different versions
of sites inspired by Gudme fall into the category of what archaeol-
ogists today call central places.125 These sites can be regarded as par-
adigmatic models of the cosmic world, deriving their structure and
organisation from archetypal sacred places126 such as Gudme on
Funen, and probably also from contemporary important sites such
as Helg (i.e. holy island) in the Mlar area.127 These are archae-
ologically well-defined settlement areas, which I have classified as
multifunctional and composite central places because they combine

125
See a.o. Larsson and Hrdh eds., Centrala platser, centrala frgor.
126
Eliade, Patterns in comparative religion, pp. 3825.
127
A. Lundstrm, Helg als frhmittelalterlicher Handelsplatz in Mittelschweden,
Frhmittelalterliche Studien 2 (1968), pp. 27890.
 507

the function of trading sites, cult-sites, production places, the hall

(or sal), gold finds etc. within a limited area.128 To some extent,
the puzzle of such complex central places in the late Iron Age of
South Scandinavia can be solved by a comparison with the cathe-
drals and monasteries in the middle ages. All were places of power,
created to be paradigmatic models of the universe, be they pagan,
Christian, or anything in between.

128
L. Jrgensen, Stromandssder og skattefund i 3.12. rhundrede, Fortid og
Nutid 20 (1995).
This page intentionally left blank
 THE LOWER VISTULA AREA AS A REGION OF
POWER AND ITS CONTINENTAL CONTACTS

Przemyslaw Urba czyk

Because of its important position in the early history of the Baltic,

the lower Vistula region has attracted much scholarly attention. The
regions many contacts with distant parts of Europe, and the well-
documented changes in the archaeological record often interpreted
in terms of ethnicity and migrations made it a focus of research
for both historians and archaeologists. However, a truly long-term
perspective still eludes us, for those attempting to generalize about
the developments in the region have limited their observations to
either the first or the second half of the first millenium A.D.1 Tra-
ditionally the first millenium is divided in two halves, i.e. the Roman
period and early Middle Ages. This artificial boundary is seldom
crossed, partly to avoid the difficult problems of interpretation con-
nected with the transformation that took place in the middle of the
first millennium. Another drawback of current research is its ten-
dency to focus on relatively small areas, with the result that archae-
ologically observed changes appear to be sudden or radical, and
therefore difficult to explain. Moreover, Roman written sources tend
to be used as haphazard illustrations of archeologically observed
trends, instead of as an integral element of interpretative arguments.
As a consequence, hardly any attention has been paid to the long-
term significance of the area, a significance that persisted through-
out a millennium, regardless of important cultural and geopolitical
changes. My aim here is to sketch a continental background to the
changes that took place in the lower Vistula region during the first
millennium. Any regional analysis should be related to a broader
geographical perspective, which often helps to understand local

1
J. Okulicz-Kozaryn, Centrum kulturowe z pierwszych wiekw naszej ery u
uj cia Wis y, Barbaricum 2 (1992), pp. 138156. M. Jagodzi ski, Rejon Wis y we
wczesnym redniowieczu. Struktura zasiedlenia I stosunki etniczne, in: M. Jagodzi ski
ed., Pogranicze polsko-pruskie w czasach w. Wojciecha (Elbl[g, 1999), pp. 3580.
Whenever possible I give references to texts published in West-European languages.
510

processes; this is the core of my argument. Since Roman times

different parts of Europe were linked in a complex network of ver-
satile relations: economic, cultural and geopolitical. Changes taking
place in one part of the continent provoked echos in distant regions.
The lower Vistula region is a textbook example of this.
The Romans were familiar with the region around the delta of
the Vistula river, situated far beyond the northern Danube frontier
of the Empire. A number of texts mention this part of the conti-
nent; they suggest that it belonged to the best-known areas of the
Barbaricum,2 which implies the existence of lively contacts between
the Empire and the shores of the Baltic.
The presence of a great river leading to the Northern Ocean3
and its coast rich in amber seemed to be known already to learned
Greeks several centuries before Christs birth.4 Its name, Vistula,
appears for the first time on a map prepared for the emperor Augustus
by his friend Marcus Vipsanius Agrippa. It was unveiled in Rome
on the porticus vipsaniae in 12 A.D.5 This map, along with Agrippas
Commentarii, greatly influenced the development of Roman geographical
knowledge; here the Vistula figured as the river dividing Germania
and Dacia. Later Roman authors also treated the Vistula as an impor-
tant boundary, but in their eyes, the river separated Germania and
Sarmatia, or Germania and Scythia.
Expeditions to the northern seas during Augustus reign, as well
as the expansion across the Danube under Trajan in 101106, must
have provided the Romans with new information about the Baltic
shores. Yet it was the official expedition organised by Nero in ca.
63 which brought about a decisive change in the contacts with the
lower Vistula region.6 This campaign was a part of the emperors
ambitious plan to explore the limits of the world. To this end he

2
J. Kolendo, Les connaissances des Anciens sur les lieux de provenance de
lambre jaune, Archeologia 34 (1985), pp. 115. J. Kolendo, Rejon uj cia Wis y w
oczach staro ytnych, in: Badania archeologiczne w wojewdztwie elbl[skim w latach
1980 1983 (Malbork, 1987), pp. 193206.
3
This was a general notion of the waters that surrounded the northern edge of
the world known to the Romans. Scandinavia was considered an island.
4
Cf. D. Timpe, Entdeckungsgeschichte, in Reallexikon der Germanische Altertumskunde
7, 2nd edn. (Berlin, New York, 1989), pp. 30789.
5
P. Trousset, La carte dAgrippa: nouvelle proposition de lecture, Dialogues
dHistoire Ancienne 19, 2 (1993), pp. 13757.
6
J. Kolendo, A la recherche de lambre baltique. Lexpedition dun chevalier
romains sous Nron, Studia Antiqua 4 (1981), p. 0000.
 511

sent expeditions to the south (to search for the sources of the river
Nile), to the east (to find Jasons Colchis) and to the north (to the
coast of the Northern Ocean). The information brought back by
the special envoy (eques R[omanus] ) along a route running all the way
from Rome to the Baltic led to an intensified trade contacts between
Italy and the eastern border of the lower Vistula. Above all, the
Romans were after amber, the raw material for incense and the dec-
orative objects that were so popular with the Roman elite.7
According to Pliny the Elder (ca. 2373 A.D.) this highly prized
amber was brought to the province of Pannonia by Germani.8 These
were probably middlemen, active in the regions between the south-
east Baltic coast and the Roman border. There is no textual evi-
dence as to the precise identity of those people. Archaeology can
only suggest that they may be identified with the producers of the
archaeologically defined Przeworsk culture.9 This culture (also
known as Venedian or Vandal culture) was dominant in the Polish
lowlands from the second century B.C. until the fifth century A.D.
The attempts to interpret this culture in ethnic terms, however,
connecting it with Vandals, Silingians, Herulians, Burgundians or
Rugians, are not convincing.
By 98 A.D. the amount of information had grown significantly,
so in his Germania Tacitus could include not only geographic data
but also some ethnographic indications. He is the first to mention
Gotones and Aestii living at the northern end of the amber route, who,
in Tacitus view, had different languages and a disctinct material
culture.10 It was the Aestii who collected amber on the coast and
exchanged it with merchants coming from the Empire.11 Ptolemy of
Alexandria, writing in the second quarter of the second century A.D.,
maintained that the Vistula divided Germania from Sarmatia. Along its
northern stretch he located the Gythones12 as well as other peoples

7
F. Braemer, Lambre lpoque romaine. Problmes dorigine, de commerce
par terre et par mer et de lieu de faonnage des objets, notamment figurs, in:
Colloque international sur les ressources minrales et lhistoire de leur exploitation (Grenoble,
1983), pp. 36181.
8
Pliny, Historia Naturalis XXXVII, 43.
9
P. Wielowiejski, Bernstein in der Przeworsk-Kultur, Bericht der Rmisch-
Germanischen Kommission 77 (1996), pp. 215342, esp. p. 217.
10
Tacitus, Germania 45.
11
Tacitus, Germania 456.
12
Ptolemy, Geographia II, 11.
512

including some Osioi (= Ostoi?).13 Historians seem to agree on the

identification of these peoples with, respectively, the Goths and
the western Balts.14 Archaeologists connect these ethnic labels to the
Wielbark culture and the West Baltic grave-mound culture.15
This basic division in two distinct groups observed both by mod-
ern archaeologists as well as ancient historians dominate the ethno-
graphic picture of the south-east Baltic region during the first half
of the first millennium. It is not the place here to discuss the com-
plex problems of ethnicity, which is in recent research considered to
be circumstantial, negotiable rather than a given structure, subjec-
tive rather than universal, emotional rather than rational.16
Regardless of all these difficulties, I will stick to the traditional
names used by historians and archaeologists. I do so for pragmatic
reasons, and also because such notions may serve as a starting point
for a more general discussion of the problems pertaining to ethnicity.
In other words, I am aware that equations of the type of Goths =
Wielbark culture are too simple, but at this point I would like to
avoid endless deliberations about the ethnic meaning of particular
elements of the material culture.
The so-called Wielbark culture is archaeologically visible since the
mid-first century A.D. in the regions to the west of the lower Vistula.
Excavations have yielded the cemeteries thought to be characteris-
tic for this culture, and also numerous Roman imports to the south-
east Baltic region. The notion Wielbark culture stems from a modern
search for material traces of the Goths, who, according to much
later textual sources, had been present in this region after they
migrated from Scandinavia. The first name given to these lost Goths

13
Ptolemy, Geographia III, 5, 9.
14
H. Wolfram, Die Goten. Von den Anfngen bis zur Mitte des sechsten Jahrhunderts.
Entwurf einer historischen Ethnographie, 3rd edn. (Mnchen, 1990), pp. 4766.
15
Cf. V. Bierbrauer, Die Goten vom 1.7. Jahrhundert n. Chr.: Siedelgebiete
und Wanderbewegungen aufgrund archologischer Quellen, Peregrinatio Gotica 3,
Universitetets Oldsaksamlings Skrifter NY rekke 14 (Oslo, 1992), pp. 943; also,
V. Bierbrauer, Archologie und Geschichte der Goten von 1.7. Jahrhundert.
Versuch einer Bilanz, Frhmittelalterliche Studien 28 (1994), pp. 15171. For an inter-
esting linguistic perspective, see D.H. Green, Linguistic evidence for the early
migrations of the Goths, in: P. Heather ed., The Visigoths from the migration period to
the seventh century (Woodbridge, 1999), pp. 1132.
16
R. Jenkins, Rethinking ethnicity. Arguments and explorations (London, 1997). S. Jones,
The archaeology of ethnicity. Constructing identities in the past and Present (London, 1997).
P. Urba czyk, Archeologia etniczno ci fikcja czy nadzieja?, in: A. Buko and
P. Urba czyk eds., Archeologia w teorii i w praktyce (Warsaw, 2000), pp. 13746.
 513

retrieved by archeaologists was Gotho-Gepidic culture. For decades

such historical reconstructions clearly determined the outcome of
archaeological interpretations. The uncritical acceptance of the Ostro-
gothic oral tradition, supposedly recorded by Cassiodorus and Jordanes
in the sixth century, resulted in the formulation of concepts that
aimed to establish a perfect fit between written texts and archaeo-
logical data.17 Everyone was happy: historians had their archaeological
evidence confirming their textual sources, and conversely, archaeol-
ogists could put a name to the culture they excavated.
This situation, seemingly advantageous to both disciplines, lasted
until recent years when new interpretations of both the texts and
the material evidence were proposed. According to this new research
Gothic ethnicity was formed at the very beginning of our era as a
result of a transformation of the local Oksywie culture.18 There was
no invasion of Scandinavian immigrants. Instead, local populations
initiated lively contacts with distant lands. These connections trig-
gered of a process of social and cultural transformation. Most impor-
tant was the trade with Roman Italy, where the elite admired the
amber from the Baltic shores.19 The inflow of silver, gold and rare
objects, as well as the need to institutionalise these regular contacts,
fostered an increasing social hierarchy in the societies around the
lower Vistula. This process is well attested for the area east of the
lower Vistula, where an elite developed that demonstrated its status
by burying their dead in ostentatiously rich graves. The cemetery of
Weklice near Elbl[g provides a good example of this.20 This display
of wealth featured exclusive Roman imports (glass vessels, medallions,
glazed pottery, precious artefacts), all crucial to the representation
of social prestige.
The emergence of this new social hierachy depended on the control
of the northern part of the amber route (fig. 1).21 This was no longer

17
For a fundamental critique of the use of Jordanes as a straightforward histor-
ical record, see W. Goffart, The narrators of barbarian history (A.D. 550800). Jordanes,
Gregory of Tours. Bede and Paul the Deacon (Princeton, 1988), pp. 20111.
18
P. Heather, The Goths (Cambridge, 1996). P. Urba czyk, The Goths in Poland
where did they come from and when did they leave?, European Journal of Archaeology
1 (1998), pp. 397415.
19
Tacitus, Germania, 456. The centre of the amber industry was located in
Aquilea where raw amber was reworked before entering the Roman markets.
20
J. Okulicz-Kozaryn, Das Grberfeld von Weklice. Zur Besiedlungsgeschichte des
Weichselsraums in der rmischen Kaiserzeit, Archeologia 40 (1991), pp. 11527.
21
See also: Urba czyk, The Goths in Poland . . ., fig. 2.
514

Fig. 1. A reconstruction of the main direction of the Amber Route in the first
century (after Urba czyk 1998, fig. 2).
 515

a road in the Roman sense of the word, that is, a one-way street
along which precious goods flooded to Rome, but a route along
which various goods were transported in both directions. Elaborate
constructions existed only at some strategic points. One such locality
was discovered and excavated near wi ty Gaj (former Heiligenwalde)
south of Elbl[g. There a huge road construction made of oak and
sand was laid out across the marshes of the Dzierzgo river valley.
Built in the second half of the first century B.C., this construction
was functional, with repairs and reconstructions, until the early third
century A.D.22
The size of the track (ca. 1.300 metres long and up to 4 metres
wide) is sufficient proof of its great importance. Clearly visible traces
of cart wheels and scattered pieces of raw amber leave no doubt as
to the material that was transported over this particular road. A geo-
logical study allowing for a reconstruction of the coast-line in early
Roman times clarifies the strategic value of the northernmost stretch
of the main route over land along the south-east Baltic coast. This
reconstruction also shows the ecological differences between the west-
ern and eastern zones of the Wielbark culture, separated by sea
inlet deeply cutting into the land, and by the broad marshy valley
of the Dzierzgo river.23
This situation did not change much during next hundred and fifty
years, when Wielbark culture populations thrived under climatic
optimal conditions. Their unchallenged domination of an area with
many natural resources, attracting foreign traders as well, allowed
their culture to spread towards the south and south-east. This expan-
sion was probably of a mixed character, for both migration and the
adoption by others of what we call Wielbark-Gothic culture played
a role. The Marcomannic wars troubling the imperial marches in
166180 A.D. had no impact on this development. On the contrary:
the richest Wielbark graves date from the late second century and
beyond, when close contacts with South Scandinavia are clearly
attested (fig. 2).24

22
J. Sadowska-Topr, Staroytne drogi w dolinie rzeki Dzierzgo w wietle bada arche-
ologicznych (Warsaw, 1999).
23
M. Kasprzycka, T o paleogeograficzne osadnictwa u aw Elbl[skich w pierwszym tysi[cle-
ciu naszej ery (Warsaw, 1999).
24
Okulicz-Kozaryn, Centrum kulturowe z pierwszych wiekw naszej ery u uj cia
Wis y, p. 137.
516

Fig. 2. The concentrated distribution of identical Roman imports in the lower

Vistula region and in south Scandinavia (after Okulicz 1992, fig. 6).
 517

The problems of sustaining the traditional contacts with the Roman

empire were solved by activating trade with the north-western Roman
provinces, to which not only raw amber, but also finished products
(mainly amber beads) were exported. The large-scale production of
characteristic wheel-turned and octogonal beads took place some-
where in the lower Vistula region. These became fashionable imports
in large areas of the Barbaricum as well as in the West- and Central-
European provinces of the Empire.25
Various types of data suggest that the situation changed sometime
around the turn of the second century and the early third century
A.D. Archaeology shows that most of the classic Wielbark cemeter-
ies to the west of the lower Vistula were abandoned by this time.
This coincided with the appearance of the Goths in the Black Sea
area, so some have taken Jordanes claim that all Goths migrated
to the south-east on the order of their king Berig literally.26 In other
words, a change in material culture is explained by an uncritical
application of data from the written sources. This is typical of
archaeologists dealing with the early Gothic history. An easy solu-
tion is looked for, and it is near at hand: didnt Jordanes say so
himself ? However, the reasons for changes in material culture are
much too complex for this haphazard combining of archeological
and textual data. First, we need a more coherent picture explaining
the circumstances, reasons and results of the processes reflected in
material culture as well as those suggested by the texts.
To begin with, we have to consider the claim that the turn of the
second century saw a deterioration of the climate in the Baltic.27
Recent paleo-climatic studies indicate, however, that such a deteriora-
tion only started at the turn of the third century, although during
the previous two hundred years there were short periods in which
the average temperature dropped.28 Two or three bad harvests could
have been dangerous for an agricultural society especially in a region
with rather unfertile sandy soils. In the face of failing crops and

25
Wielowiejski, Bernstein in der Przeworsk-Kultur.
26
Jordanes, Getica, 26ff.
27
H. Zabehlicky, Kriegs- oder Klimafolgen in archologischen Befunden?, in:
H. Friesinger, J. Tejral and A. Stupner eds., Markomannenkriege. Ursachen und Wirkungen
(Brno, 1994), pp. 4639.
28
M. Kasprzycka, Zmiany klimatyczne i procesy osadnicze w rejonie dolnej Wis y 25001000
lat temu [manuscript submitted to Archeologia Polski ].
518

persistent bad weather, some may have decided to look for better
soils, as Jordanes said.29
These deteriorating natural conditions coincided with what histo-
rians of the Roman Empire consider to be a period of economic
crisis lasting from 235 to 284 A.D.30 The barbarians took advantage
of the weakness of the Romans by attacking the border provinces
in the third quarter of the century. These fifty years of decline, albeit
temporary, must have had a long-term negative impact on the fre-
quency and predictability of the long-distance contacts an trade.
Indeed, the number of Roman coins found in the south-east Baltic
region is significantly lower for this period.31
Such a combination of economic and climatic factors may have
struck the Wielbark elite hardest, which depended on substantial
surplusses and exclusive goods for consumption and redistribution.
A reduced agricultural production and a decrease in the volume of
long-distance trade must have resulted in an intensification of the
competition for prestige and status. Sustaining a high social position
became more difficult than ever before. Several possible crisis strate-
gies could be followed in such a situation.
The first one, the military alternative, entailed the internal use of
physical power by the elite in order to coerce members of the soci-
ety and/or its neighbours to provide important goods and to have
their power accepted. However, a militarised economy is self-consuming
and unsustainable without a developed system of taxation, and the
application of physical force is often a short-term solution. In soci-
eties without a developed state political power usually relies on accep-
tance and consensus, attained by persuasion and by attracting followers,
rather than by coercion. Another type of military solution is exter-
nal aggression. Raiding and looting rich neighbours by small but
determined war parties was often a strategy of societies on the periph-
ery of empires that had problems with their distant marches. However,
the Wielbark Goths had no chance to raid the distant Roman empire
and its precious resources they depended on for sustaining their social
prestige.

29
Jordanes, Getica 96.
30
T. Kotula, Le dbut de la crise du IIIe sicle de lEmpire romain: depuis
quand? (summary), in: A. Bursche, M. Mielczarek and W. Nowakowski eds., Nunc
de Svebis dicendum est (Warsaw, 1995), p. 156.
31
A. Bursche, Later Roman-Barbarian contacts in Central Europe. Numismatic evidence
(Berlin, 1996), pp. 89 ff.
 519

The alternative was an economic strategy, such as consuming

and redistributing treasure, but it is unlikely that these tactics were
applied in the transformation of the Baltic regions, in the void left
by Roman power. Economies based on redistribution had no power
centres with an independent economic base, and had to rely on the
support of its followers eager for gifts. Thus, in times of crisis there
were no reserve supplies to support the system. If such a crisis sit-
uation continued, the elite was in danger. The alternative for the
leaders was either to dissolve into a society with an organisation
slowly developing into one more egalitarian social structures, or to
find better luck by initiating a risky journey. When there were no
sufficient resources to be mustered locally, the only choice was to
move the settlement nearer to the area where such resources were
abundant. Will we ever know the circumstances of such dramatic
decisions to leave a familiar region? Did the dominant lineage decide
to migrate towards the Roman colonies on the Black Sea coast near
the fertile soils of the Ukraine, or was there a conflict between com-
peting lineages, of which only one left the Vistula homeland with
its supporters? Or, was there no clear decision and is the story of
the heroic migration a later invention constructed to legitimize the
Amals domination of the Pontic Goths?
Whatever is the answer, each of these situations must have involved
an ideological change, with leading families developing and pro-
moting a myth of ancient glory and success, won by real or invented
ancestors. This helped to symbolically connect the prosperity of the
society to its leaders, who had inherited good luck from their ances-
tors. Such constructions could not solve the fundamental social and
economic problems determined by external factors, but they could
legitimize dramatic decisions taken by the leaders whose positions
were endangered and who were in desperate need of support.
The narrative of the early Gothic history is full of dramatic events.
The heroic crossing of a sea (= the Baltic) recorded by Jordanes
might just have been ideological camouflage to show how mindful
leadership helped to turn a crisis situation into succes. Berig, the
king who had to persuade his followers to undertake a migration
from the lower Vistula to the Black Sea could have been at its origin.
Within this context, one might also argue that the story of the organ-
ised migration from the north was invented in the Ostrogothic king-
dom, once the attack by the Huns in 375 led to the dramatic decision
to leave the Pontic zone, and to flee for safety into the Roman empire.
520

Roman texts first mention the presence of the Goths near the
Danube limes in 238 A.D. This fits very well with the argument
that the Goths alleged departure from the Baltic homeland coin-
cided with the crisis in the Empire that started in 235 A.D.32
This journey was a real success, and its glorious memory was later
promoted by the Amal dynasty, which needed a myth of an heroic
past with which everybody could identify him/herself. The creation
of an identity by the Gepids, who were first observed near the
Danube limes during the reign of the Emperor Probus (276282
A.D.),33 was of a similar nature.
We do not know whether this was the only migration, for only
those leaders who managed to build stable domains in the new lands
managed to create the type of dynastic tradition later recorded by
Cassiodorus and Jordanes. Those who did not succeed disappeared
from historys view. To my mind, we should be suspicious of records
and reconstructions of so-called primary migrations involving a small
number of participants who were later followed by subsequent larger
waves, or by a stable inflow of people moving along the lines join-
ing the Baltic and Pontic regions. Archaeological data do not sup-
port the notion of a depopulation of the lower Vistula area. The
so-called Wielbark culture did not disappear in this region, and its
tradition can be traced until the fifth or even the sixth century, when
local populations developed a new material culture and merged with
recently arrived newcomers.
Neither was there any break in contacts between those who migrated
on the one hand, known archaeologically as the Cherniakhov cul-
ture, and those who stayed in the south-east Baltic area. Both pop-
ulations communicated with each other, which led to the formation
of a large zone of cultural contact that stretched between the two
seas where Wielbark and Cherniakhov cultures were located. The
linear distribution pattern of many artefacts typical for this zone
clearly shows the main axis of these contacts (fig. 3). The northern
end of this axis was the region of lower Vistula.

32
Cf. Wolfram, Die Goten, pp. 5364; Heather, The Goths, pp. 3843.
33
See Heather, The Goths, pp. 435; W. Pohl, Die Gepiden und die Gentes
an der mittleren Donau nach dem Zerfall des Attilareiches, in: H. Wolfram and
F. Daim eds., Die Vlker an den mittleren und unteren Donau im fnften und sechsten Jahrhundert,
Denkschriften der sterreichischen Akademie der Wissenschaften, Phil.-Hist. Kl.
145 (Wien, 1980), 239305.
 521

Fig. 3. An example of the linear distribution pattern of specific types of finds

between the Baltic and Black seas (after Kokowski 1995, map 56, cf. similar dis-
tribution patterns in his maps 6, 9, 26, 48 and 65.)
522

This long-distance communication between various Gothic groups

added much to Roman knowledge of the area. Basic geographic and
ethnographic information survived in the history of the Goths writ-
ten by Cassiodorus and rewritten by Jordanes in the first half of the
sixth century. Jordanes mentioned the river Vistula/Viscla six times.
He also correctly pointed to its sources in the Carpathians and
described its tripartite delta.34 He thought of the Vistula as the bor-
der between Germania and Scythia.35 There Jordanes situated some
mysterious Vidivarii and the Hestii.36 From the period of Theoderics
reign (493526) we have also proof of some direct contacts between
Ostrogothic Italy and the south-east Baltic area. It is a concept of
a letter sent in 523526 by Theodoric to those Hestii who lived on
the coast rich in amber.37
Four hundred years passed since the lower Vistula was mentioned
for the first time in written sources, and still it was the amber that
was on the mind of the powerful in the Mediterranean.38 However,
the intermediaries changed: in the first century the Germani exported
the amber from the north,39 in the first half of the sixth century we
hear of the Hestii assuming this role.40 One might argue that such
accounts are untrustworthy because other independent information
is lacking, but this scant information is not contradicted by the archae-
ological data, so it may be accepted on a provisional basis.
Archaeologists date the westward expansion of the West-Baltic cul-
ture towards the lower Vistula to the late fifth century.41 The material
evidence of this process is usually connected with the Aestii/Osioi
(= Ostoi?)/Hestii who were often mentioned in Roman sources. In
earlier times, they kept their settlements well away from the area where
the Wielbark culture is found, creating a kind of no mans land in
between. What caused the expansion of that characteristic tradition
around the middle of the first millennium? Why did the character-
istic features of the Wielbark culture reach other areas by the third

34
Getica 17 and 36.
35
Jordanes, Getica 17 and 31.
36
Jordanes, Getica 36, 96, and 120.
37
Cassiodorus, Variae V, 2.
38
J. Kolendo, Nap yw bursztynu z p nocy na tereny Imperium Rzymskiego w
IVI w.n.e., Prace Muzeum Ziemi 41 (1991), pp. 91100.
39
Pliny, Historia Naturalis XXXVII, 43.
40
Cassiodorus, Variae V, 2.
41
Jagodzi ski, Rejon we wczesnym redniowieczu. Struktura zasiedlenia
i stosunki etniczne.
 523

and fourth century? The explanation that the Wielbark people left
the area, making room for new settlers who occupied their deserted
lands, is too simple. I think the more likely explanation is that of
new cultural codes being accepted by those who lived there.
This merging of various traditions must be seen against the back-
ground of continental developments. Once more north and south
appear to be closely connected. The final collapse of Attilas polity
in 453 created a power vacuum along the northern frontier of the
Roman Empire. This resulted in geopolitical disorder, and led to
conflicts among ambitious leaders who, until then, had been kept in
check by the mighty Hun. The disappearance of a major power that
had dominated multi-ethnic masses opened the way for many to
strive for supreme prestige, territorial control and privileged relations
with Constantinople. This spiral of conflicts must have produced
many unlucky leaders, who had to look for better chances in regions
not claimed by the winners. They had to seek new territories to pre-
sent their claims for power.
Roman historians exclusively recorded migrations of groups who
moved around the Mediterranean zone. However, there must have
been migrations to the north as well; a good example is that of the
Heruli, who set a course northwards in 512. Thus, it seems highly
probable that some desillusioned groups followed the ancient route
established between the Black Sea and the Baltic, that is, between
the Cherniakhov and Wielbark cultures or, phrased differently,
between the Pontic domains of the Goths and their original home-
land. This direction was attractive for those intent on military and
political expansion, for relations with the home territory had been
kept up throughout the centuries. Cultural similarities between the
two regions made direct contact more easy, and presumably facili-
tated the adaptation to a culture that was, after all, a familiar one.
We will never really know whether any re-migration to the Baltic
zone actually occurred, for those who left the sphere of Roman civil-
isation simply disappeared from the focus of literate observers. Only
archaeology furnishes some data to support this hypothesis.
In the late fifth and early sixth century, numerous late Roman
and Byzantine golden solidi as well as valuable objects suddenly
appeared in the eastern part of the lower Vistula region.42 This

42
M. Jagodzi ski, Archeologiczne lady osadnictwa mi dzy Wis [ a Pas \k[. Katalog
524

increase in war gains stemming from robbery, war booty and

tribute may indicate the presence of powerful men who brought
these prestige goods with them, once they re-migrated to the north.
If so, these new powerful settled in what had earlier been the richest
area of the Wielbark culture, taking advantage of the strategic position
of this region. What we are left with is a new material culture appar-
ently emerging in the lower Vistula during the sixth and seventh
centuries, a phenomenon that requires some kind of explanation.
The archaeological evidence of this period consists of a complex
mixture. Rich cremation graves often situated on top of horse burials,
and the presence of characteristic sax-type swords in decorated scab-
bards,43 as well as other finds, all indicate a combination of Germanic,
nomadic and Baltic traditions. Apart from these southern and east-
ern elements there are also connections with the Merovingian king-
dom, as well as with Scandinavia.44 All this material evidence suggests
that various tribes merged into a new entity to which Jordanes in
the mid-sixth century gave the name Vidivarii.45 Some archeaologists
have argued in favour of a continuation of local Wielbark traditions
in this region,46 something they consider proven by are similar tech-
niques used in the production of pottery.47
The geographical range of these cultural phenomena extends to
the west as far as the Drun o estuary and Dzierzgo river, the tra-
ditional divide of the Wielbark culture. Further to the west a new
no-mans land) was established during the fifth century. It reached
to the western (main) branch of the lower Vistula. This situation
presumbably lasted until the eighth century, when the area between

stanowisk (Warsaw, 1997). M. Jagodzi ski, Archeologiczne lady osadnictwa mi dzy

Wis [ a Pas \k[ we wczesnym redniowieczu. Komentarz do katalogu stanowisk,
in: P. Urba czyk ed., Adalbertus, vol. 1 (Warsaw, 1998), pp. 15997, esp. p. 166.
43
P. Urba czyk, Geneza wczesno redniowiecznych metalowych pochew broni
bia ej ze stanowisk kultury pruskiej, Przegl[d Archeologiczny 26 (1978), pp. 10745.
44
A. Bitner-Wrblewska, The Southeastern Baltic zone and Scandinavia in the
early migration period, Barbaricum 2 (1992), pp. 24566. W. Duczko, Scandinavians
in the Southern Baltic between the 5th and the 10th centuries A.D., in: P. Urba czyk
ed., Origins of Central Europe (Warsaw, 1997), pp. 191211.
45
Jordanes, Getica 36 and 96.
46
Bitner-Wrblewska, The Southeastern Baltic zone, p. 262. Jagodzi ski,
Archeologiczne lady osadnictwa mi dzy Wis [ a Pas [k[ we wczesnym redniowieczu.
Komentarz do katalogu stanowisk, p. 164.
47
A. Moszczy ski, Wczesno redniowieczne szlaki komunikacyjne Starego
Mazowsza, in: P. Urba czyk ed., Adalbertus, vol. 1 (Warsaw, 1998), pp. 24162,
esp. table 3.
 525

the main branch of the Vistula and the Dzierzgo river was still
almost depopulated.48 To the west of this area the Wielbark tradi-
tions vanished completely, to be replaced, from the early seventh
century onwards, by groups with a radically different material culture,
commonly identified as early Slavs.49 They ousted the Scandinavians
who had penetrated the south Baltic coast during the Migration
Period.50
Thus, the lower Vistula was once more an ethnic and cultural
border area, of the kind defined earlier by Marcus Vipsanius Agrippa
and Ptolemy. Since the seventh century the lower Vistula divided
Pomerania, settled by the Slavs, from the region inhabited by the
Estii/Balts. This division lasted for several centuries, which does not
mean that the situation was stable. A new archaeological landscape
emerged in the eighth and ninth centuries, when many Scandinavian
artefacts and also graves with evidence of Scandinavian ritual
testify to trans-Baltic contacts.51 These contacts were so intensive that
several emporia were founded under the aegis of Scandinavian settlers.
One of them was Truso, known from the report of the Anglo-
Saxon merchant-sailor Wulfstan on the south-east Baltic coast, possibly
made to King Alfred the Great (871899) himself. When the king
of Wessex had Orosius Historia adversus paganos (417) translated into
Old English by an anonymous West Saxon, Wulfstans report was
included.52 Wulfstan accurately described the route from Hedeby/
Haithabu to Truso via the Estian Gulf (= the Vistula Bay), but
only recently, when geologists reconstructed the coastal area east of
the lower Vistula, the significance of the last part of Wulfstans
itinerary has been fully understood (fig. 4). Wulfstan also confirmed
the rivers persistent function as an important border, when he ob-
served that it divided the Slavs from the land settled by the Estii.

48
Jagodzi ski, Archeologiczne lady osadnictwa mi dzy Wis [ a Pas [k[ we
wczesnym redniowieczu. Komentarz do katalogu stanowisk, pp. 16770.
49
Maybe these were the Slavs settled near the Western Ocean, who were inter-
rogated by Emperor Maurikios ca. 595 A.D. (Teophylact Simokatta, Historiae VI,
2, 1016).
50
Duczko, Scandinavians in the Southern Baltic, pp. 195200.
51
B. von zur Mhlen, Die Kultur der Wikinger in Ostpreussen, Bonner Hefte zur
Vorgeschichte 9 (Bonn, 1975). Duczko, Scandinavians in the Southern Baltic.
52
Translated in M. Swanton ed. and transl., Anglo-Saxon prose (London, 1975),
pp. 3237; some helpful comments in M. Lapidge and S. Keynes, Alfred the Great.
Assers Life of King Alfred and other contemporary sources (Harmondsworth, 1983), pp.
33 and 258.
526

Fig. 4. The reconstruction of the Vistula estuary at the end of the tenth century
(after Kasprzycka 1998, fig. 6).
 527

Archaeological investigations have enabled a more precise location

of this border: once more, it was the Dru no estuary and Dzierzgo
river.
These two ethnic regions developed differently. The western one
was densely settled since the ninth century by people with cultural
affiliations to Slavic Pomerania, Kulm Land and Masovia. People
over here lived in settlements concentrated around large central
strongholds. Their material culture was not very sophisticated or rich,
which is characteristic of most Slavic populations during this period.
There is no evidence that they developed any territorial political
organisation comparable to the one in central Poland, where early
state formation processes started by the turn of the ninth century.
Meanwhile, the eastern zone kept up its cultural relations with the
Baltic tribes. Here archeaologists have also observed a relatively poor
and uniform material culture.53 In most cremation graves only few
grave goods were present, and there were no more horse burials, so
typical of an earlier period. Strongholds were built there, but they
were different: small, and situated on the boundaries of local settle-
ments. Apparently, these strongholds had another function than their
Slavic counterparts. These were local watchtowers, not central places.
Wulfstans report is the last historical source in which the thou-
sand-years-old name Estii was recorded. Since the mid ninth cen-
tury the new ethnonym of Burus (= Pruss) became popular. It was
first used by the so-called Bavarian Geographer, when he wrote in
ca. 848 of the eastern neighbours of Louis the Germans east-Frankish
kingdom. Subsequently the new name appeared in the Arab sources
as well; by the tenth century it was generally accepted as that of the
West-Baltic tribes. It survived until 1945, when the German provinces
of Ost- and Westpreussen disappeared from the map.54
The precise location of the site of Truso was finally identified in
1982, on the eastern coast of Lake Dru no, the last remnant of the
southern part of the former Vistula Bay.55 This ten hectare large set-
tlement, with a layout of regular streets and a rampart, was inhabited

53
Jagodzi ski, Rejon Wis y we wczesnym redniowieczu. Struktura zasiedlenia
i stosunki etniczne.
54
It may be interesting to note here that river Dzierzgo and lake Druno marked
the border line between those two provinces.
55
M. Jagodzi ski and M. Kasprzycka, The early medieval craft and commer-
cial centre at Janw Pomorski near Elbl[g on the South Baltic Coast, Antiquity 65
(1991), pp. 696715.
528

by Balts, Scandinavians and Slavs, who had workshops producing

jewellery, pottery as well as glass, iron, antler and amber artefacts.
Coins, weights and scales, and imports from Scandinavia and west-
ern Europe indicate that trade was their main activity. Numerous
finds of Arabic dirhams from the early ninth century56 testify to the
importance of Truso for the Baltic exchange system.
Early in the tenth century Truso went up in flames; it was never
rebuilt again. In the mid-tenth century its function of a coastal port
of trade was taken over by Gda sk, a town promoted by the nascent
Polish state. This replacement was not an accident. It should be
interpreted against the larger background of important contemporary
changes in the way maritime trade was organized. Truso disappeared
from the Baltic stage just like many other Viking Age emporia (e.g.
Haithabu in Denmark, Menzlin in Germany, Wiskiauten in Lithuania).
These were replaced by newly founded towns enjoying the protection
of the kings in charge of early states. These new political powers
managed to muster an effective control over their newly subordinated
territories, and to expel Viking chiefs and entrepreneurs who had
refused to submit to a new regime of taxation. The new ports were
located near the old trade centres, usually at the mouth of big rivers
which allowed ships and merchants to penetrate landinwards.
This particular transformation was also induced by changes in
shipbuilding technology. Traditional flat-bottomed ships were replaced
by those with deep hulls, which offered more space for cargo. They
were not pulled onto land out of the shallow bays, but needed deep
ports with special waterfront constructions. These technological changes
contributed to the decline of Truso, which now was overtaken by
Gda sk, just as Haithabu was replaced by Schleswig, or Hamwih
by Southampton.
However, there is more to this transformation. Archaeological data
clearly suggest substantial changes in the economic significance of
the regions on either side of the lower Vistula. In the ninth century
it was the coastal area east of the river where the majority of the
deposits of dirhams are found (fig. 5). Was it again amber, or rather
furs and slaves that were sought by the entrepreneurs who kept up
the trade contacts between the Baltic and the lower Volga? Whatever

56
M. Czapkiewicz, M. Jagodzi ski and Z. Kmietowicz, Arabische mnzen aus
einer frhmittelalterlichen Handwerker- und Handelsiedlung in Janw Pomorski,
Gem. Elbl[g, Folia Orientalia 25 (1988), pp. 157169.
 529

Fig. 5. The distribution pattern of finds of dirhams issued before A.D. 900 (after
Brather 1996, fig. 4).
530

it was, this trade lost its importance during the next century, when
the opposite situation emerged: all treasures with dirhams from that
period are found to the west of the lower Vistula. This change in
the geography of coin deposition patterns gives the impression that
the Prussian lands lost their economic significance. It is even more
striking after 970, when coin finds are only known from Slavic
Pomerania (fig. 6).
The Prussian side of the lower Vistula seems to have lost its con-
tacts with the outside world, to become an isolated province in the
rapidly changing Baltic basin. The port in Gda sk built in the sec-
ond quarter of the tenth century was the Baltic vanguard of the
Polish state. The Vistula became the spine of the expanding state
of Mieszko I, who converted to Christianity in 966, thus founding
a stable dynastic state. He tried to subjugate groups in the estuar-
ies of both large rivers Vistula and Oder which ran from his
territory to the Baltic. His son Boleslav the Great (9921025) enlarged
his domain in all directions, with only one exception: the Prussian
lands, where he gained no success whatsoever. The reason for this
was the different territorial organisation of the Prussians, who had
no political centres and developed a system of boundary defence that
was difficult to break. They themselves actively raided the Polish ter-
ritory. They rejected the ideological and organisational characteris-
tics of their neighbours, including the Christian faith. Therefore, the
Polish monarch applied a strategy of partial measures. He tried to
promote indirect cultural and ideological expansion by organising
Christian missionary activities. The first missionary campaign was
led by Bishop Voitech-Adalbert, who came from Rome especially for
this purpose.57 He went from Boleslavs court by boat to Gda sk, and
subsequently to the Prussian lands east of the lower Vistula, where
he was killed on 23 April 1997. All that is meticulously described
in St Adalberts Life written in 999 in Rome by his friend, the
Benedictine monk Johannes Canaparius.
With this note on the tragic death of Adalbert I will end this
paper, which began and ended with texts written in Rome. During
the millennium between the first information recorderd by Marcus
Vipsanius Agrippa, and the Life of Adalbert of Prague, the region

57
P. Urba czyk, St Adalbert-Voitech missionary and politician, in: P. Urba
czyk ed., Early Christianity in Central and East Europe (Warsaw, 1997), pp. 15562, esp.
p. 161.
 531

Fig. 6. The distribution pattern of finds of dirhams issued after A.D. 970 (after
Brather 1996, fig. 11).
532

of the lower Vistula was part of a transformation affecting large areas

of Europe, which brought about numerous changes in the material
culture, observed by archaeologists. There were also flexible ethnic
affiliations among the inhabitants of the lower Vistula, which have
been mainly traced by historians. Regardless of all these cultural,
political and social changes, the rivers Dzierzgo and the Dru no
estuary retained their function as a significant border from the sec-
ond centuy A.D. until the year 1000.
 TOPOGRAPHIES OF POWER:
SOME CONCLUSIONS*

Mayke de Jong
Frans Theuws

How did people construct places of power, and how did such places,
in turn, create powerful people? This is the central question of this
book, discussed by historians and archaeologists. Those who contrib-
uted to this volume have focussed on the high-level power exercised
by early medieval political and religious elites and its topographical
setting, and on the transformation of this relationship in the post-
Roman kingdoms. Did a Roman world, where cities were the obvi-
ous central places providing the stage for rituals of power, give way
to a multitude of places and spaces where political and religious
power might be represented? If so, what was different about the
interaction between people and places in this new context? Where
and how was political and religious power or the combination
thereof concentrated and expressed, and what can we learn about
the ways in which audiences recognised and interpreted such places
of power? These are some of the issues featuring in the papers gath-
ered in this book. Given the expertise present in our group, the
Frankish kingdoms get a large share of our attention, but the chang-
ing topography of power in the old centres of the Roman world,
Rome and Constantinople, also enters upon the scene. Furthermore,
there are chapters on what centres of power may have meant in
the steppes of Inner Asia, Scandinavia or the lower Vistula, where
political power was even more mobile and decentralised than in the
post-Roman kingdoms. Monasteries and their integration into early
medieval topographies power also loom large in this book. There
are many areas we did not manage to cover, and we are well aware
of only having scratched the surface of a vast area of research
but let us concentrate on what we have to offer.

* Barbara Rosenwein and Matthew Innes sent us stimulating comments on an

earlier draft of this conclusion; Bonnie Effros and Walter Pohl kindly checked the
final version. We are most grateful for all this assistance; any remaining errors are
our own.
534

The notion of weak post-Roman states, mentioned in Chris Wick-

hams introduction a discussion paper which was our point of
departure turned out to be the kind of Weberian ideal type
Wickham meant it to be: a concept useful to work with, precisely
because it concentrates the mind on the ways in which early medieval
polities did not fit this definition. Certainly, political power and its
location became more decentralised and less institutionalised in the
post-Roman West, but this did not necessarily entail any weakness.
Instead, we are dealing with power assuming different forms and fol-
lowing different rules. Early medieval power-relations became inter-
personal and complex; they were situated topographical contexts one
cannot classify as either political, religious, institutional, social or eco-
nomic. They were all of this at the same time. Hence, whenever the
expression central places is used in this volume, it is emphatically
not with the exclusively institutional and economic meanings which
have burdened this concept.
Some examples might clarify this point, and they take us straight
into the contributions of this book. In the Frankish kingdoms the
possible meanings of an episcopal town are not exhausted by stat-
ing that this was an episcopal see, dominated by a bishop; this derives
from an institutional perspective which may be relevant to later
Middle Ages, but not to earlier periods. As Frans Theuws explains
with regard to Maastricht, this episcopal see was not fixed in one
location; episcopal power depended on a network of places of power
in the Meuse Valley, and Maastricht itself, embedded in this net-
work, was a centre in which kings, artistoctrats and (aristocratic) bish-
ops all had a competitive stake. It was a multivalent place, to which
all those competing in this arena assigned a different significance,
but these mutiple meanings reinforced each other. Something simi-
lar holds true for monasteries. These were sacred places, first and
foremost, but it was precisely the religious nature of monastic sites
that turned them into places with a significance that went far beyond
modern or early Christian notions of a religious community.
In early Christianity God was thought to be present wherever a
Christian community happened to meet, but in the post-Roman
world Christianity became a religion revolving around well-defined
places: the loci sancti empowered by the relics of saints and enriched
by the patronage the laity.1 The powerful kings, bishops and lay

1
For a still fundamental discussion of this development, see R.A. Markus, The
end of ancient Christianity (Cambridge, 1990), pp. 13955.
 535

aristocrats claimed the biggest stake in these sacred domains. A

specific Frankish model emerged, which is explored by Le Jan, Rosen-
wein, Innes, and De Jong. Monasteries and their relics became crucial
nodes in familial, episcopal and royal networks. By the middle of
the seventh century, a limited number of monasteries were singled
out by episcopal privileges of exemption and royal grants of immu-
nity. This meant that such monastic sites became hands-off areas,
sacred spaces where the ordinary rules of political practice prevailing
in the world outside were suspended; as such, these loci sancti were
invested with an inherent power to which only those who had granted
these hands-off privileges had direct access. In other words, these
sacred places were gradually integrated in the topography of royal
and episcopal power, becoming central by virtue of their liminal-
ity. By the late eighth century, such favoured religious communities
were called royal abbeys. Carolingian royal protection (tuitio) turned
them into communities of prayer closely connected with the court.
This is not to say that all Frankish monasteries fell into this cat-
egory. Many smaller monastic houses which did not belong to the
happy few of those singled out by kings and bishops fixing sacred
space and depending on monastic prayer. Other religious commu-
nities remained local and unknown places that only left haphazard
traces, or none at all. With regard to religious foundations in Brittany
and Galicia, Julia Smith and Pablo Diaz remind us that there were
other ways of creating sacred spaces, in which kings, bishops or aris-
tocrats played only a minor role, or where, according to Diaz, entire
villages, threatened in their existence by royal or episcopal power,
transformed themselves into monastic communities, preserving older
social patterns. Still, the alternative monastic strategies investigated
by Diaz and Smith help to bring out the peculiarities of the Frankish
model and its alliance between monastic space and political power
a phenomenon that by no means is an exclusively Frankish one. In
other post-Roman kingdoms not represented in this book, such as
Lombard Italy and Anglo-Saxon England, political power also increas-
ingly depended on monastic prayer.2
In seventh-century Francia, aristocratic families located familial
power within the sacred by founding female religious communities
in which abbesses and nuns retained intricate links with the world

2
Cf. M. de Jong, Religion, in R. McKitterick (ed.), The Short Oxford History of
Europe II (Oxford, 2001), pp. 13164.
536

they came from, as Rgine Le Jan explains. Consecrated women

were the intermediaries of this sacralisation, but also its extremely
vulnerable victims. This was a precarious balance. As Le Jan says,
it is hardly surprising that all the great womens houses discussed
in this chapter eventually passed into royal control.3 In Matthew
Innes chapter, this particular stage of the process is outlined: monas-
teries once dominated by local aristocrats became royal ones. But
Innes also shows how the control of such sacred sites allowed new
rulers the Carolingians to gain a foothold in regional power
structures. Monasteries such as Fulda and Lorsch were central places
where the members of the elite flocked when there was important
business at hand; a significant scene already existed, into which the
kings inserted themselves. The integration of sacred loci into a polit-
ical topography is also central to the papers written by Barbara
Rosenwein and Mayke de Jong. Rosenwein shows how one partic-
ular site, Saint-Maurice dAgaune, could assume many different mean-
ings over time, moving gradually into a royal orbit. But kings did
by no means dictate the terms of the sacred, not even in royal
monasteries. As De Jong argues, such hands-off areas played an
important role in situations of high-level political conflict, allowing
the powerful to withdraw gracefully into sacred space, until the tide
had turned; so-called monastic imprisonment was a flexible and
open-ended strategy.
Monastic stabilitas thus supported the stability of the new polities,
but unlike the sacred sites they founded, protected and relied upon,
those exercising political power be they kings, bishops or nobles
were not place-bound. The powerful moved about, displaying their
might in a public arena which included towns, royal and aristocratic
residences, and, increasingly, monasteries. By their presence, they
helped to build up places of power. Kings visiting with their entour-
age, bishops and abbots meeting for a synod, aristocrats gathering
to transact their business they all created central places over time.
Such important events and visits were remembered, discussed and
written about. A trail of royal charters springs to mind, issued in a
multitude of dispersed locations in the Frankish kingdoms. Centrality
consisted of a network of significant places where rituals and cere-
monies were enacted and remembered. Or, as Matthew Innes expressed

3
See above, p. 269.
 537

it: The centre was a locus, not a location, a kind of forcefield bend-
ing and warping physical space, as kings wove a giant web around
their kingdoms.4
With regard to ancient cities remaining undisputed central places,
the image of webs being woven, connecting new significant places
to older ones, is also an appropriate metaphor. Constantinople re-
mained the centre of an empire, a real capital in the Roman sense
of the word, but as Leslie Brubaker argues, the shift away from
monumental civic places that already had started in the third cen-
tury had become ubiquitous by the early Middle Ages. In the Eastern
Mediterranean, public space had contracted, and it now clustered
around religious buildings rather than around civic ones. This did
not mean that public life and its rituals declined or disappeared.
Public life had lost nothing of its vitality, but it followed a different
pattern. Whereas the space associated with secular public ritual seems
to have contracted to the hippodrome, religious processions gained
in importance, linking the new foci of public space: churches, mostly,
but not exclusively, for there was no rigid separation between eccle-
siastical and imperial sites. As Brubaker notes, this new symbolic
topography, established in the early medieval period, was still largely
in place in the twelfth century a remarkable case of longue dure.
To compare the topography of an ancient city like Constantinople
to webs woven in the post-Roman kingdom takes a helicopter-view,
but from this perspective, one might say that in both East and West,
spaces and places defined as sacred became the scene on which
the important rituals of power were enacted; the other side of this
coin is that throughout the early medieval world, royal (and impe-
rial) ritual became sacralized.
Early medieval Rome, however, was a city without a resident em-
peror or king. Who represented power to whom in Rome? Thomas
Nobles chapter deals with the emergence of Rome as a papal city
in the eighth and early ninth centuries, and with the articulation of
an ideological programme expressed in papal patronage and gifts:
buildings, mosaics, precious textiles, gold and coins, to mention just
a few items. In Rome as well, liturgical processions connected the
significant churches within the city, redefining space; popes led the
way, appropriating public space and turning it into St Peters Republic.

4
A reaction per e-mail of 6 December 2000 to an earlier draft of this conclusion.
538

The audience for this display of papal power did not merely con-
sist of the people of Rome in general, or its aristocratic lay and
clerical elites in particular. There was also a Rome in the mind
above the Alps, a font of authenticity and orthodoxy upon which
new political identities in the north such as the Carolingian em-
pire were built. Romes symbolic topography, explored by Noble,
had huge ramifications, for Franks, Anglo-Saxons and other peoples
in the north harboured great expectations of this source of authen-
ticity and orthodoxy. The famous papal approval of the Carolingian
bid for royal power may well have been a Frankish invention of tra-
dition,5 but this does not detract from the importance attached by
the Franks and other northerners to papal authority and Roman
authenticity. The tide of relics of Roman martyrs that flooded into
Carolingian royal abbeys meant that the recipient sites became more
sacred, for they were now connected with Rome, and with the
authentic Christian past now embodied by an eminently papal Rome.6
This was a two-way street. Papal claims to being a summus pontifex
et universalis papa were supported by emperors who visited Rome and
inspected the city and its expressive rituals they depended on. In
other words, Rome was not only a papal city; it was also a theatre
where rulers who based their authority on a close connection with
Rome manifested themselves as humble pilgrims, surely, but also
expressing their power through gestures of humility.
In 804 Pope Leo III travelled to Aachen to celebrate Christmas.
By then, Aachen had become established as the privileged seat of
an emperor a capital, even, and an urban site, as Janet L. Nelson
argues. Her chapter vividly evokes the growth of royal and imper-
ial business which made Aachen the hub of the Carolingian empire.
A comparison between Aachen and Rome (and Constantinople, for
that matter) yields a whole range of interesting differences. To begin
with, Aachen was a new centre. As Nelson puts it, what was inevitably
presented, given a culture deriving all legitimacy from the past, as
renewal was in fact new.7 Aachen was not a tabula rasa: there were

5
Cf. R. McKitterick, The illusion of royal power in the Carolingian Annals,
The English Historical Review 115 (2000), pp. 120.
6
J.M.H. Smith, Old saints, new cults: Roman relics in Carolingian Francia,
in: eadem ed., Early medieval Rome and the Christian West: Essays in Honour of Donald
Bullough (Leiden/Boston/Kln, 2000), pp. 31739.
7
See above, p. 236.
 539

Roman remains which were re-used in the construction of the palatium

and its famous chapel, though Nelson doubts whether these inspired
the eighth-century builders with any concrete sense of romanity.
The same holds true for the baths where Charlemagne and his
courtiers swam. As had been the case with the Roman baths of the
past, those in Aachen were the locus of male conviviality, a place
where the Frankish king and his inner circle relaxed and discussed
important business. But Nelsons chapter also explores new mean-
ings of the baths that were at Aachens heart, and they grew over
time. Aachen was not just to be represented as, but to be, a place
of cleansing for a people assured that they were Gods, a place where
power was harnessed for multiple human applications in the service
of God.8 It could gain this position because there was no serious
ecclesiastical competition. There was no bishop in Aachen, although
droves of bishops flocked to the place, underlining its importance
and centrality; hence, this new centre could become loaded with reli-
gious connotations, but also figure as the theatre for eminently sec-
ular rites of rulership not unlike Constantinople, but with a very
different starting point and development. From its foundation, Aachen
was a resolutely royal centre, with a religious significance that owed
as much to the more recent as to the distant past. Aachen embod-
ied the link between the old Israel and the new, the baptismally
reborn gens sancta of the new Israel embodied, now, in the Church
and the peoples of Charlemagnes much expanded regnum Francorum.9
For this very reason, Aachen was called the sacred palace (sacrum
palatium) in the generation following Charlemagne.
Aachen was a place with a Roman past, but as Nelson explains,
it is impossible to say whether those who reconstructed the place in
the eighth century were aware of this. What was the symbolic poten-
tial of visible remains from the glorious past? Did the ruins of ancient
monuments still have a meaning to later generations? Did they impose
contraints on new mental maps, and did their presence, however
ephemeral, guide the selection of apparently new places of power?
These questions are confronted in Bonnie Effros chapter on the re-
use of Roman ruins, but they are also at the back of Frans Theuws
mind, when he discusses the complexity of Maastricht as an emerging

8
See above, p. 236.
9
See above, p. 222.
540

place of power in the eighth century, within the Meuse Valley web
of which Tongres, Namur and Lige were also an integral part. Theuws
rejects the notion of a central place in this case, a city merely
defined in economic and institutional terms. Bishops, as members of
aristocratic groups, gained a foothold in places dispersed over an
large but integrated region, studded with places with an ancient past,
and this mattered. When it comes to determining how this past was
remembered, however, historians and archaeologists alike are faced
with limits and constraints of their sources. Discussing early medieval
Gaul, Effros admits that there is a tension between written and mate-
rial sources which makes any effort to determine past perceptions of
ruined sites or artifacts tentative at best. Once written sources
are absent, it is even harder to assess what the physical presence of
the past may have meant for those living with it. As Heinrich Hrke
remarks in his discussion of cemeteries, the visual power which ancient
remains may have exerted over later spectators eludes us; too many
post-Roman monuments have disappeared, and the landscape is
changed beyond recognition. Yet however elusive, the presence of
the past clearly weighted upon the present. Sometimes the written
sources allow a glimpse into the possible reactions to ancient mon-
uments and ruins; these might be rejected and destroyed, but also
integrated into Christian communities by what Effros calles rituals
of possession. Such rituals also were also enacted in texts. Some
might depict the foundation of their monastery as having occurred
in the wilderness, a virgin territory without a past, while others
incorporated the presence of ancient ruins into their narrative. In
his Life of Columbanus, Jonas of Bobbio combined these two themes.
When searching for a suitable place in the wilderness (in heremo) to
found a monastery, Columbanus and his monks hit upon an ancient
fortification called Luxeuil, the former site of great Roman baths;
images of pagan idols still littered the surrounding forest, now the
territory of wild beasts.10 This is an image of idolatry having been
conquered by wilderness, but nonetheless, the presence of the past
looms large in Jonas narrative. Luxeuil was founded on royal land.
Did the remains of a glorious Roman past make this land more
royal, and was this perceived as such by the king and his entourage

10
Jonas of Bobbio, Vita Columbani I, c. 10, ed. B. Krusch, MGH SRM 4 (Hannover,
1902), p. 76.
 541

who supported Luxeuils foundation? It is hard to say, for Jonas

wrote a generation later, but these questions should be asked.
Places of power, be they monasteries, palaces or entire regions
studded with central places, had many meanings to different people.
The perceptions of the past, however elusive they may be to modern
scholars, were part of this spectrum of meaning this much is
clear. It is their very complexity that determines the symbolic poten-
tial of such places, as well as their endurance. As Barbara Rosenwein
expressed it, for a place of power to be lasting, it must have the
same sort of complexity as a great piece of music, so that in each
era new maestri can tease out different timbres and themes.11 In less
poetic words, places of power were constructed, re-constructed and
manipulated over time. Texts played an important role in this process,
but so did human action. The group of nobles flocking to Lorschs
villa of Heppenheim (discussed by Innes) went to a place that to
them held significance as a site where important public business
might be transacted a public place. Thus they helped to build
Lorschs reputation in this respect. So did the charter recording this
event, for that matter, but these men conducting their affairs in a
monastic site were no figment of the scribes imagination, and neither
are the people who turned cemeteries into much more than a place
where one might commemorate the dead, as Hrke reminds us.
This is where Appadurais concept of tournaments of value,
invoked by Frans Theuws, comes into its own. Theuws uses this
notion to clarify how one central place Maastricht and its ramifica-
tions became a focus for many actors and activities, ranging from
political competition to the production of valuable artefacts. Was the
value of the latter enhanced because of their connection with places
which also had sacred connotations? Did sacred places attract pre-
cious metals, which were then transformed into artefacts with great
symbolic value? These questions crop up on Theuws discussion of
Maastricht, but they are central to Lotte Hedeagers chapter on
Gudme, a home of the gods in South Sandinavia. Recent excava-
tions in this area have yielded a stupendous amount of treasure, as
well as buildings of a formidable size. This was a composite site
flourishing between the third and seventh centuries, but what did it
mean to those who built and sustained it? Hedeager wrestles with a

11
See above, p. 290.
542

lack of total textual evidence, although she tentatively uses informa-

tion yielded by thirteenth-century epic material. Her analysis mainly
rests on a combination of archaeological data and insights gained
from anthropology. As Hedeager argues, Gudme was a place attract-
ing precious goods from the world outside, which were then reshaped
and transformed, gaining new meanings intricately associated with
Gudmes status as a sacred place. Hedeager broaches an issue also
relevant to contexts other than pre-Christian Scandinavia. If cen-
tral places were built up by means of precious metals and artefacts
imported from distant and exotic places, how did perceptions of this
great world outside contribute to notions of centrality on the inside?
Or, to put it differently, when an elephant arrived in Aachen in 802,
as a gift from Harun-al-Rashid, did this enhance Aachens profile as
a central place? Judging by the way the author of the Royal Frankish
Annals dwelled on the elephants difficult journey to the north, this
was indeed the case.12 Precious objects and animals arrived
from faraway and mysterious regions via tortuous routes.
In the case of the elephant, we can see how a historiographer
made the most of his journey and arrival, but where Gudme is con-
cerned, it remains a matter of carefully assessing archaeological data
and tentatively exploring a possible context. The same holds true
for Przemek Urba czyks contribution on the region of the Lower
Vistula. This was another region attracting a concentration of human
activity and productivity marking then out as central. What about
all these dirhams and other exotic items turning up in excavations
in the Lower Vistula why there, and not anywhere else? This
was region of rich economic resources of which various powers availed
themselves, the Roman empire to begin with but many followed
suit, and this region remained important over the centuries. The
Lower Vistula was a highly valued source of amber in the Roman
era, but it subsequently became a depository of wealth that must
have had its impact on the ruling elites in situ, as Urba czyk argues.
Hedeagers reflections on Gudme as a privileged location for high-
status artisanal production yield some relevant insights for scholars
thinking about monasteries in the early medieval West. These were
not only sacred places, but also sites where prestigious artisans were

12
Annales regni Francorum s.a. 801802, ed. R. Rau (Darmstadt, 1974), pp. 768;
cf. also p. 94, on the death of the elephant in 810.
 543

at work. Was this a matter of monks just having superior resources

the traditional narrative cherished by historians or did the sacred-
ness of these production site also come into it? This is a question
we need to pursue in the future one of many.
Tournaments of value also happened at the level of texts. It is
misleading to say that some places happened well-documented, while
others just disappeared without a trace. In the post-Roman king-
doms, the written word was an instrument of power, which went a
long way towards constructing places of power. To be undocumented
meant not having the wherewithal to establish a lasting existence.
This is one of the conclusions of Julia Smiths investigation of Redon,
a marginal place, whether viewed from a Frankish or Breton per-
spective. Smiths contribution makes it clear that Redons aedificatio
was a matter of texts, and the same pertains to Frankish royal monas-
teries, but the authors who thus built Redon drew upon textual tra-
ditions that diverged from mainstream Carolingian monasticism.
From a different perspective, but with a similar emphasis on the
rhetoric of texts, De Jong argues that the withdrawal of high-level
political players into monastic space was never an affair represented
in a neutral fashion by the authors who recorded such events. They
might represent this opting out as a voluntary and therefore hon-
ourable action, or as a the result of dishonourable coercion. Great
reputations were constructed or destroyed in debates about the right
interpretation raging long after the events in question. The only part
of this debate we can still grasp is the one which was fixed in texts
which still survive but these went a long way towards establish-
ing the views that came to dominate both early medieval and mod-
ern historiography.
Those dealing with texts have to be aware that they are entering
the vociferous debates of the past, with authors making a strong case
for some places being more important than others. But there are
also meaningful silences, difficult to identify as such, and efforts to
relegate places that did hold significance to contemporaries to a
minor status, or even to oblivion. Textual aedificatio of places of power
should not be taken at face value. Ann Christys account of Cordoba
is a worrying case in point. This was a town enthusiastically and
admiringly written up in Arabic sources listing its vast number of
baths and caravanserais, the focus of Al-Andalus identity, but when
it comes to excavated remains, there is not much to be found. Should
one hope that more of the Cordoba existing in early medieval minds
544

may be unearthed in due course? Or will Cordoba go the way of

Tours, a one-horse town, as Ian Wood calls it? The great reputa-
tion of this one-horse town was constructed by its bishop Gregory,
whose sweeping and compelling narrative pointedly excluded other
centres representing possible competition for Tours (and St Martins)
pre-eminence. By the 1970s, when historians became seriously inter-
ested in the intricate connection between early medieval bishops and
their saints, Tours became the model of an episcopal town, and the
paradigm for the cult of the saints in early medieval Gaul. Woods
chapter offers a fundamental critique of a model relying almost exclu-
sively on Gregorys writings, and reminds us of topographies of power
about which this influential author remained silent.
In other chapters, the possibilities and limits of texts also a cen-
tral issue. As Walter Pohl says in his discussion of the elusive courts
of Inner Asia, there are no textual sources giving any insight into
the pre-Carolingian hall and what went on in there, and Beowulf s
much-cited account is highly stylised and idealised. Drunken war-
riors did most likely boast about their heroic deeds in such places,
but as accounts of such sessions become more detailed and high-
lighted in anthropologically inspired research such accounts
become less amenable to generalisation. So heroic tales were told,
and then what? This is a view which may displease archaeologists
who have tried to make sense of the rituals in the hall featuring in
Beowulf and other epics, aided by anthropological theory. Pohls
approach is a different one, but equally valid. He scrutinises the
accounts of the diplomats and voyagers coming from places they
themselves perceived as central, who tried to make sense of the vast
and unknown world they encountered. Pohls chapter should be read
alongside Hedeagers; in very different ways, Hedeager and Pohl
both shed light on the possible impact of places out there on cen-
trality defined from inside.
In short, texts helped to shape and articulate places and relations
of power, or, as Matthew Innes reminds us, from the outset, it is
vital to remember that the sources from which we work did not
stand outside of the relationships they recorded.13 The written sources
present us with clearcut topographies of power, for those in charge
of the written word were instrumental in the aedificatio of central

13
See above, p. 398.
 545

places, enhancing their importance over time or allowing them to

be forgotten. The authors of the sources we work with had a par-
ticular agenda, and we need to be aware of this, if we are to make
the most of the rich and complex texts at our disposal. Archaeological
data often yield a very different picture. The tensions between writ-
ten and material sources are part and parcel of this book. We do
not argue that archaeological and historical sources are in any way
complementary, in the sense that historians can rely on the appar-
ent solidity of excavation reports, or that archaeologists can make
sense of material culture by taking recourse to the enticing details
offered by the written record. These are false securities, for mater-
ial and written sources pose their own problems, and they played
their own role in strategies of aedificatio, along with a variety of
human activity, such as the performance of rituals and ceremonies,
or the naming of places. Yet places of power are not only created
by authors, but also by readers who interpreted, contested and
changed their meaning. This ongoing process of aedificatio and inter-
pretatio is an challenging area of future research where archaeologists
and historians can fruitfully join forces. Some first attempts in this
direction have inspired this book; we hope this becomes evident to
those perusing its contents.
This page intentionally left blank
 BIBLIOGRAPHY

Adalbert of Trier, Continuation of Reginos Chronicon, ed. F. Kurze, MGH SRG

20 (Hannover, 1890).
Adam of Bremen, Gesta Hammaburgensis ecclesiae pontificum, transl. F.J. Tschan, Adam
of Bremen, History of the Archbishops of Hamburg-Bremen. Records of Civilization: Sources
and Studies LIII (New York, 1959).
Additamentum Nivialense de Fuliano, ed. B. Krusch, MGH SRM 2 (Hannover, 1888),
pp. 44951.
Adefonsi Tertii Chronica, in: J. Gil, J.L. Moralejo and J.I. Ruz de la Pea eds., Cronicas
asturianas (Oviedo, 1985), pp. 11349.
Ado, Vita Theudarii, Migne PL 123, col. 44350.
Adso, Vita Frodoberti abbatis Cellensis, ed. W. Levison, MGH SRM 5 (Hannover,
1910), pp. 6788.
Agnellus, Liber pontificalis ecclesiae Ravennatis, ed. O. Holder-Egger, MGH SRL
(Hannover, 1878).
Alcuin, Commentaria in sancti Joannis Evangelium, Migne PL 100, col. 7337.
, Epistulae, ed. E. Dmmler, MGH Epp. 4 (Berlin, 1895).
, Vita Willibrordi prosaica, ed. A. Poncelet, AASS, November 7th, vol. 3.
Alvarus, Vita Eulogii, ed. J. Gil, CSM I.
Ambrose, De obitu Theodosi, ed. O. Faller, CSEL 73 (Wien, 1955).
, Epistulae, ed. M. Zelzer, CSEL 82, 3 (Wien, 1982).
Ammianus Marcellinus, Res gestae, ed. and transl. J.C. Rolfe (revised edition, Cambridge,
1956).
, The later Roman Empire, transl. W. Hamilton (Harmondsworth, 1986).
Analectes sur lhistoire et la littrature des arabes dEspagne, ed. R. Dozy, 2 vols. (Leiden/
London, 185561).
Annales s. Amandi, ed. G.H. Pertz, MGH SS I (Hannover, 1826), pp. 610 and
1214.
Annales s. Bertiniani, ed. R. Rau, Quellen zur karolingischen Reichsgeschichte II (Darmstadt,
1972); eds. F. Grat et al., Annales de Saint-Bertin (Paris, 1964).
Annales qui dicuntur Einhardi, see Annales regni Francorum.
Annales Fuldenses, ed. F. Kurze, MGH SRG (Hannover, 1891), pp. 1107; transl.
T. Reuter, The Annals of Fulda (Manchester, 1992).
Annales Laubienses, ed. G.H. Pertz, MGH SS 4 (Berlin, 1841), pp. 920.
Annales Laureshamenses, ed. G.H. Pertz, MGH SS 1 (Hannover, 1826), pp. 1939.
Annales Mettenses priores, ed. B. von Simson, MGH SRG 10 (Hannover, 1905), pp.
198.
Annales Petaviani, ed. G.H. Pertz, MGH SS 1 (Hannover, 1829), pp. 158.
Annales regni Francorum, ed. F. Kurze, MGH SRG 6 (Hannover, 1895), pp. 3115;
ed. R. Rau, Quellen zur karolingischen Reichsgeschichte I (Darmstadt, 1974).
Annales Rotonenses, ed. B. Bischoff, Annales Rotonenses (um 919), in: Analecta novis-
sima. Texte des vierten bis sechzehnten Jahrhunderts (Stuttgart, 1984), pp. 1025.
Anonymus (Astronomer), Vita Hludowici, ed. E. Tremp, MGH SRG 54 (Hannover,
1995).
Apparitio S. Michaelis archangeli in Monte Tumba in Gallia, Migne PL 96, cols. 138994.
Arbeo of Freising, Vita vel passio Haimhrammi episcopi et martyris Ratisponensis, ed.
548

B. Krusch, MGH SRG 13 (Hannover, 1920), pp. 2699; ed. B. Bischoff, Leben
und Leiden des Hl. Emmeram (Mnchen, 1953).
Augustine of Hippo, De Civitate Dei, eds. B. Dombart and A. Kalb, CCSL 48
(Turnhout, 1955).
, De doctrina christiana, ed. J. Martin, CSSL 32 (Turnhout, 1962).
Obras completas de San Agustn XXVI. Sermones (6 ) 339396, ed. P. de Luis (Madrid,
1985).
Avitus of Vienne, Epistolae, ed. R. Peiper, MGH AA 6, 2 (Berlin, 1883), pp. 117269.
, Opera omnia, ed. U. Chevalier, Oeuvres compltes de saint Avit vque de Vienne,
(Lyon, 1890).
Baluze, E., Miscellanea, ed. J.D. Mansi (Lucca, 1762).
Beda Venerabilis, Historia ecclesiastica gentis Anglorum, eds. B. Colgrave and R.A.B.
Mynors, Bedes Ecclesiastical History of the English People. Oxford Medieval Texts
series (Oxford, 1969).
Bede, The Ecclesiastical History of the English People, eds. and transl. J. McClure and
R. Collins (Oxford, 1969).
Benedict of Aniane, Codex regularum, Migne PL 103, col. 7171380.
Benedicti S. Andreae monachi chronicon, ed. G.H. Pertz, MGH SS 3 (Hannover, 1829),
pp. 695719.
Beowulf, transl. Michael Alexander (Harmondsworth, 1973).
Sancti Bonifatii et Lulli Epistolae, ed. E. Dmmler, MGH Epp. 3, Epp. Merovingici
et Karolini Aevi 1 (repr. Berlin, 1957).
Braulio of Saragossa, Epistularium, ed. L. Riesco Terrero, Epistolario de San Braulio.
Introducin, edicin crtica, traduccin (Sevilla, 1975).
Le Calendrier de Cordoue, eds. and transl. R. Dozy and C. Pellat (Leiden, 1961).
Die Canones Theodori Cantuariensis und ihre berlieferungsformen, ed. P.W. Finsterwalder
(Weimar, 1929).
Cartulaire de labbaye de Gorze, ed. A. dHerbomez (Paris, 1902).
Cartulaire de labbaye de Saint-Sauveur de Redon. Amis des archives historiques du diocse
de Rennes, Dol et Saint-Malo (Rennes, 1998).
Cartulaire de Redon, ed. A. de Courson (Paris, 1863).
Cassiodorus, Variarum Libri XII, ed. A.J. Fridh, CCSL 96 (Turnhout, 1973).
Chartae Latinae Antiquiores: Facsimile-edition of the Latin charters prior to the ninth century,
eds. A. Bruckner and R. Marichal, part XIII, France I, ed. H. Atsma and
J. Vezin, vol. 13 (Dietikon/Zrich, 1981).
Chronica Albendensis, eds. J. Gil, J.L. Moralejo and J.I. Ruz de la Pea, Crnicas asturi-
anas (Oviedo, 1985), pp. 15188.
Chronicon Moissiacense, ed. G.H. Pertz, MGH SS 1 (Hannover, 1826), pp. 282313.
Chronicle of Sampiro, ed. M. Gmez Moreno, Boletn de la Real Academia de la
Historia 100 (1932), pp. 3378.
Codex Carolinus, ed. W. Gundlach, MGH Epp. 3 (Berlin, 1892), pp. 476657.
Codex Diplomaticus Fuldensis, ed. E.F.J. Dronke (Kassel, 1850).
Codex Laureshamensis, ed. K. Glckner, 3 vols. (Darmstadt, 192936).
Codex Theodosianus, ed. Th. Mommsen (3rd edition, Berlin, 1962).
Concilia Galliae, a. 511a. 695, ed. C. de Clercq, CCSL 148A (Turnhout, 1980).
Concilios visigticos e hispano-romanos, ed. J. Vives (Barcelona/Madrid, 1963).
Constantius of Lyon, Vita Germani, ed. and transl. R. Borius, Vie de Saint Germain
dAuxerre, SC 112 (Paris, 1965).
Constructio monasterii Farfensis, ed. U. Balzani, Il Chronicon Farfense di Gregorio di Catino.
Fonti per la Storia dItalia 3334, 2 vols. (Rome, 1903, reprint 196972).
Conte, P., Regesto delle lettere dei papi del secolo VIII (Milan, 1984).
Flavius Cresconius Corippus, In laudem Iustini Augusti Minoris, ed. A. Cameron (London,
1976).
Corpus Iuris Civilis, vol. 3 Novellae, eds. R. Schoell and W. Kroll (5th edition Berlin,
1928).
 549

Corpus Scriptores Ecclesiasticorum Latinorum 16, ed. M. Petschenig (1888).

Crnicas asturianas, eds. J. Gil, J.L. Moralejo, J.I. Ruz de la Pea (Oviedo, 1985).
Cyprianus et al., Vita Caesarii, ed. B. Krusch, MGH SRM 3 (Hannover, 1896), pp.
457501.
Davis, R., transl. and comm., The lives of the eighth-century popes (Liber Pontificalis)
(Liverpool, 1992).
Dhuoda, Liber manualis, ed. P. Rich, SC 225 (Paris, 1975).
Donatus of Metz, Vita Ermenlandi abbatis Antrensis, ed. W. Levison, MGH SRM 5
(Hannover, 1910), pp. 674710.
Drew, K.F., transl., The laws of the Salian Franks (Philadelphia, 1991).
Duchesne, L., Fastes piscopaux de lancienne Gaule, 3 vols. (Paris, 18941915).
Dutton, P.E., Carlemagnes courtier. The complete Einhard (Ontario, 1998).
Egerias travels to the Holy Land, transl. J. Wilkinson (revised edition, Jerusalem, 1981).
Einhard, Translatio et Miracula SS Marcelini et Petri, ed. G. Waitz, MGH SS 15
(Hannover, 1887), pp. 23864.
, Vita Karoli Magni, ed. O. Holder-Egger, MGH SRG 25 (Hannover, 1911);
ed. R. Rau, Quellen zur karolingischen Reichsgeschichte 1 (Darmstadt, 1974).
Epistolae Arelatenses Genuinae, ed. W. Gundlach, MGH Epp. 1 (Berlin, 1892), pp.
183.
Epistolae decretales ac rescripta Romanorum pontificum, ed. F. Antonio Gonzlez (Madrid,
1821).
Erchanberti breviarium regum Francorum, ed. G. Pertz, MGH SS 2 (Hannover, 1829),
pp. 3279.
Ermentarius, De translationibus et miraculis Santi Filiberti, ed. R. Poupardin, Monuments
de lHistoire des Abbayes de Saint-Philibert (Paris, 1905), pp. 1970.
Ermoldus Nigellus, In Honorem Hludovici Pii, ed. E. Faral, Ermold le Noir: Pome sur
Louis le Pieux (Paris, 1964).
Eucherius, Passio Acaunensium martyrum, ed. B. Krusch, MGH SRM 3 (Hannover,
1896), pp. 2041.
Eugippius, Vita Severini, ed. R. Noll, Eugippius: Das Leben des Heiligen Severin (Passau
1963); ed. Th. Mommsen, MGH SRG 26 (Berlin, 1898).
Eulogius, Memoriale sanctorum, ed. J. Gil, CSM, 2 vols. (Madrid, 1973).
Eusebius of Caesarea, Oratio de laudibus Constantini, transl. H.A. Drake, In praise of
Constantine: A historical study and new translation of Eusebius Tricennial Orations (Berkeley,
1976).
Flavius Cresconius Corippus, In laudem Justini Augusti minoris, ed. A. Cameron (London,
1976).
Folcuin, Gesta abbatum Sithiensium, ed. B. Gurard, Cartulaire de labbaye de Saint-Bertin
(Paris, 1840).
Fragmenta historiae Byzantinae, Fragmenta historicorum Graecorum 4, ed. K. Mller (Paris,
1851), pp. 69110.
Fredegarii Chronicarum libri IV cum continuationibus, ed. B. Krusch, MGH SRM 2
(Hannover, 1888), pp. 1193; ed. A. Kusternig, Ausgewhlte Quellen zur deutschen
Geschichte des Mittelalters 4a, 2nd edn. (Darmstadt, 1994), pp. 1325; ed. M. Wallace-
Hadrill, The Fourth Book of the Chronicle of Fredegar with its Continuations (London,
1960).
Gesta Dagoberti I regis francorum, ed. B. Krusch, MGH SRM 2 (Hannover, 1888), pp.
399425.
Gesta sanctorum Rotonensium, ed. C. Brett, The monks of Redon: Gesta sanctorum Rotonensium
and Vita Conuuoionis (Woodbridge, 1989).
Giorgi, I. and U. Balzani eds., Il regesti di Farfa, vol. 2 (Rome, 1879).
Gregory of Tours, De cursu stellarum ratio, ed. B. Krusch, MGH SRM 1, 2 (revised
edition, Hannover, 1969), pp. 40422.
, De virtutibus S. Martini, ed. B. Krusch, MGH SRM 1, 2 (Hannover, 1885),
pp. 584661.
550

, Decem libri historiarum, eds. B. Krusch and W. Levison, MGH SRM 1, 1

(Hannover, 1951; revised edition, Hannover, 1965).
, Glory of the confessors, transl. R. Van Dam. Translated texts for historians, Latin
series IV (Liverpool, 1988).
, De virtutibus sancti Martini, ed. B. Krusch, MGH SRM 1, 2 (Hannover, 1885),
pp. 58466; revised edition (Hannover, 1969), pp. 134211.
, De virtutibus sancti Juliani, ed. B. Krusch, MGH SRM 1, 2, (revised edition,
Hannover, 1969), pp. 11234.
, In gloria confessorum, ed. B. Krusch, MGH SRM 1, 2 (revised edition, Hannover,
1969), pp. 294370.
, In gloria martyrum, ed. B. Krusch, MGH SRM 1, 2 (revised edition, Hannover,
1969), pp. 34111.
, Vitae patrum, ed. B. Krusch, MGH SRM 1, 2, (revised edition, Hannover,
1969), pp. 21194.
, The history of the Franks, transl. L. Thorpe (Harmondsworth, 1982).
Gregory the Great, Epistola, ed. D. Norberg, CCSL 140A, Registrum epistolarum, libri
VIIIXIX, Appendix (Turnhout, 1982), pp. 80511.
, Dialogi, ed. A. de Vog, SC 251, 260, 265 3 vols. (Paris, 197880).
, Moralia in Job, ed. M. Adriaen, CCSL 143 (Turnhout, 1979).
, Registrum epistolarum libri VIIIXIV, ed. D. Norberg, CCSL 140A (Turnhout,
1982).
Haymo of Halberstadt, In epistolam II ad Thimotheum, Migne PL 117, col. 797810.
Herigerus, Translatio s. Landoaldi sociorumque eius, ed. O. Holder-Egger, MGH SS 15,
2 (Hannover, 1888), pp. 599607.
Herodotus, Historiae, transl. R. Waterfield, Herodotus: The Histories. Oxfords World
Classics (Oxford, 1998).
Hildemar of Corbie/Civate, Expositio regulae s. Benedicti, ed. R. Mittermller, Expositio
regulae ab Hildemaro tradita et nunc primum typis mandata (Regensburg etc., 1880).
Hincmar of Rheims, Consilium de poenitentia Pippini regis, Migne PL 125, col. 111922.
, De ordine palatii, eds. T. Gross and R. Schieffer, MGH Fontes Iuris Germanici
Antiqui 3 (Hannover, 1980); ed. V. Krause, MHG Capit. 2 (Hannover, 1897),
pp. 518530.
Historia Langobardorum codicis Gothani, eds. L. Bethmann and G. Waitz, MGH SS Rer.
Lang. (Hannover, 1878).
Historia Silense, ed. J. Prez de Urbel (Madrid, 1959).
Historia translationis s. Balthildis, AASS Jan., II, pp. 7479.
Honoratus of Marseilles, Vita Hilarii, ed. and transl. P.-A. Jacob, La vie dHilaire
dArles, SC 404 (Paris, 1995).
Hrabanus Maurus, Carmina, ed. E. Dmmler, MGH Poet. Lat. aevi Karolini 2
(Hannover, 1881), pp. 154258.
, Enarrationes in Epistolae B. Pauli, Migne PL 111, col. 12731616.
Hucbald of St Amand, Vita Aldegundis virginis, Migne PL 132, col. 862.
, Vita Rictrudis abbatissae Marcianensis, ed. J. Mabillon, AASS O.S.B. II, pp. 94243.
Hydatius, Chronica, ed. R.W. Burgess, The Chronicle of Hydatius and the Consularia
Constantinopolitana. Two contemporary accounts of the final years of the Roman Empire
(Oxford, 1993), pp. 70123.
Ibn al-Fara , Tar kh ulam al-Andalus, eds. F. Codera and J. Ribera (Madrid,
18912).
Ibn al-Q a, ed. and transl. J. Ribera, Historia de la Conquista de Espaa por Abenalcotia
el Cordobes (Madrid, 1926).
Ibn awqal, rat al-Ar , ed. J.H. Kramers (Leiden, 1939) transl. M.J. Romani
Suay, Configuracin del Mundo ( fragmentos alusivos al Magreb y Espaa). Textos Medievales
XXVI (Valencia, 1971).
Ibn ayy n, Al-Muqtabis f bal d al-Andalus (al-Hakam II) (Al-Muqtabis VII), ed. A.A.
Al-Hajji (Beirut, 1965).
 551

, Al-Muqtabis, transl. J. Guraieb, Cuadernos de Historia de Espaa 27 (1958).

Ibn Idh r, Al-Bay n al-mughrib f akhb r al-Andalus wa-l-Maghrib, eds. E. Lvi-Provenal
and G.S. Colin (Paris, 1930).
Iohannes de Plano Carpini (Giovanni di Pian del Carpine), Historia Mongolorum (Storia
dei Mongoli ), ed. P. Daffin e.a. (Spoleto, 1989).
Inscriptiones Latinae Christianae Veteres 1, ed. E. Diehl, new edition (Dublin, 1970).
Isidore of Seville, Etymologiarum sive Originum, ed. W.M. Lindsay (Oxford, 1962).
, Historia vel Origo Gothorum, ed. C. Rodrguez Alonso, Las Historias de los godos,
Vndalos y suevos de Isidoro de Sevilla. Estudio, edicin crtica y traduccin (Len, 1975).
, Regula sancti patris Isidori episcopi, eds. J. Campos and I. Roca, Santos Padres
espaoles II (Madrid, 1971), pp. 90125.
Itinerarium Bernardi, ed. T. Tobler, Descriptiones Terrae Sanctae ex saeculo VII, IX, XII,
XV (Leipzig, 1874), pp. 8599.
John of Biclar, Chronica, ed. Th. Mommsen, Continuatio Isidoriana Hispana, MGH AA
11 (Berlin, 1961), pp. 21120.
Jonas of Bobbio, Vita Columbani abbatis discipulorumque eius libri duo, ed. B. Krusch,
MGH SRM 4 (Hannover, 1902), pp. 1152; ed. B. Krusch, MGH SRG 37
(Hannover, 1905), pp. 1152.
Jordanus of Quedlinburg, Iordanis de Saxonia Liber Vitas Fratrum eds. R. Arbesmann and
W. Humfner (New York, 1943).
Laterculus Visigothorum, ed. Th. Mommsen, MGH AA 13 (Hannover, 1898), pp. 4649.
Le Blant, E. ed., Inscriptions chrtiennes de la Gaule antrieures au VIII e sicle 2 (Paris,
1865).
Leges Visigothorum, ed. K. Zeumer, MGH LL 1 (Hannover/Leipzig, 1902), pp. 34456.
Leo I Papa, Epistula 15 (ad Turibium Asturicensem Episcopum), ed. B. Vollmann, Studien
zum Priszillianismus. Die Forschung, die Quellen, der fnfzehnte Brief Papst Leos des Grossen
(St. Ottilien, 1965), pp. 12238.
Libellus de Ecclesiis Claromontanis, ed. W. Levison, MGH SRM 7 (Hannover, 1920).
Le Liber Pontificalis. Texte, introduction et commentaire, ed. L. Duchesne, 2 vols. (Paris,
188692).
Liber sacramentorum Gellonensis, ed. A. Dumas, CCSL 159 (Turnhout, 1981).
Liber sacramentorum romanae aeclesiae ordinis anni circuli (Cod.Vat.Reg.Lat. 316/Paris Bibl.Nat.
7193, 41/56), eds. L.C. Mohlberg, L. Eizenhfer and P. Siffrin. Rerum Ecclesias-
ticarum Documenta, Series Maior Fontes 4 (Rome, 1960).
Libri Carolini, see Opus Caroli regis contra synodum.
Liutprand of Cremona, Antapodosis, ed. J. Becker, MGH SRG 40 (Hannover, 1915).
Lucas of Tuy, Chronicon mundi (Frankfurt, 1610).
Ludovici II Diplomata, ed. K. Wanner, Instituto Storico Italiano per il Medio Evo (Rome,
1994).
Marculf, Formulae, ed. K. Zeumer, Formulae Merovingici et Karolini Aevi, MGH LL 5
(Hannover, 1886), pp. 36106; ed. A. Uddholm (Uppsala, 1962).
McNamara, J.A. and John E. Halborg with E. Gordon Whatley eds. and transl.,
Sainted women of the Dark Ages (Durham/London, 1992).
Menander, ed. K. Mller, Fragmenta historicorum Graecorum 4 (Paris, 1851), pp. 20069.
, ed. R.C. Blockley, The History of Menander the Guardsman (Liverpool, 1985).
Mirarelli, Johannes Benedictus, Annales Camaldulenses 1 (Venice, 1755).
Monumenta Boica XI (Mnchen, 1771).
Nithard, Historiae, ed. Ph. Lauer, Nithard: Histoire des fils de Louis le Pieux (Paris, 1926);
ed. E. Mller, MGH SRG 14 (Hannover, 1907).
Notker, Gesta Karoli magni imperatoris, ed. H.H. Haefele, MGH SRG, NS 12 (Berlin,
1959).
Odo of Glanfeuil, Historia translationis S. Mauri, AASS Jan. I, pp. 105160.
Oeuvres compltes de saint Avit vque de Vienne, ed. U. Chevalier (new edition Lyon, 1890).
Opus Caroli regis contra synodum. (Libri Carolini), ed. A. Freeman, MGH Conc. II,
Supplementum I, (Hannover, 1998).
552

Pactum s. Fructuosi, eds. J. Campos and I. Roca, Santos Padres espaoles II (Madrid,
1971), pp. 20811.
Pactus Legis Salicae, ed. K.A. Eckhardt, MGH LL 4, 1 (Hannover, 1962).
Parochiale suevum, ed. P. David, CCSL 175 (Tunhout, 1965), pp. 41120.
Paschasius Radbertus, Vita Adalhardi, Migne PL 120, col. 150782.
Passio Leudegarii, ed. B. Krusch, MGH SRM 5 (Hannover, 1910), pp. 282322.
Passio Praeiecti, ed. B. Krusch, MGH SRM 5 (Hannover, 1910), pp. 22548.
Passio s. Victoris et Sociorum, ed. J. Mabillon, AASS September VIII, p. 292.
Passio Sigismundi regis, ed. B. Krusch, MGH SRM 2 (Hannover, 1888), pp. 32940.
Pliny, Naturalis historiae libri XXXVII, transl. D.E. Eichholz (Cambridge, 1962).
The Poetic Edda, transl. L.M. Hollander (Austin, 1994).
Priscus, ed. R.C. Blockley, The fragmentary classicising historians of the later Roman Empire:
Eunapius, Olympiodorus, Priscus and Malchus (Liverpool, 1983).
Blockley, R.C. ed., The History of Menander the Guardsman (Liverpool, 1985).
Prudentius, Psychomachia, ed. J. Bergman, CSEL 61 (Wien, 1926).
Claudius Ptolemeaus, Opera, ed. J.L. Heilberg (Leipzig, 18981908).
, CosmographiaTavole geografiche di Tolomeo, ed. L. Pagani (Torriana, 1990).
Raguel, The Martyrdom of St. Pelagius, transl. J.A. Bowman, in: T. Head ed.,
Medieval Hagiography, an Anthology (New York, 2000).
Rashid al-Din, ed. J.A. Boyle, The successors of Genghis Khan (New York, 1971), pp.
178203.
Ratio fundationis seu aedificationis Blandiniensis coenobii, ed. M. Gysseling and A.C.F. Koch,
Diplomata Belgica ante annum millesimum centesimum scripta, 2 vols. (Brussels, 1950).
Regesta Pontificum Romanorum, eds. P. Jaff et al., 2 vols. (2d edition, Leipzig, 188588;
reprint Graz, 1956).
Regino of Prm, Chronicon, ed. F. Kurze, MGH SRG 50 (Hannover, 1890).
Regula communis, eds. J. Campos and I. Roca, Santos Padres espaoles II (Madrid, 1971),
pp. 172208.
Regula consensoria monachorum, Migne PL 66, col. 9936.
Regula cuiusdam patris ad monachos, ed. F. Villegas, La Regula cuiusdam patris ad
monachos. Ses sources littraires et ses rapports avec la Regula monachorum
de Columban, Revue de lhistoire de la spiritualit 49 (1973), pp. 336.
Regula quattuor patrum, ed. J. Neufville, Rgle des IV Pres et Seconde Rgle des
Pres. Texte critique, Revue bndictine 77 (1967), pp. 47106.
Regum Burgundiae e stirpe Rudolfina Diplomata et Acta, eds. T. Schieffer and H.E. Mayer
= Die Urkunden der Burgundischen Rudolfinger (Mnchen, 1977).
Richer, Historiarum libri IV, ed. G. Waitz, MGH SRG 51 (Hannover, 1877).
Rimbert, Vita Anskarii, ed. G. Waitz, MGH SRG 55 (Hannover, 1884).
Rudolf of Fulda, Vita Leobae abbatissae Biscofesheimensis, ed. G. Waitz, MGH SS 15,
1 (Hannover, 1887), pp. 11831.
Salzburger Urkundenbuch, eds. W. Hauthaler and F. Martin, vol. 2 (Salzburg, 1916).
Santos Padres espaoles II, eds. J. Campos and I. Roca (Madrid, 1971).
Saxo Grammaticus, Danorum Regum Heroumque Historia, transl. E. Christiansen, Saxo
Grammaticus, Danorum Regum Heroumque Historia Books XXVI, 3 vols. BAR International
Series 84 (Oxford, 1980).
Sidonius Apollinaris, Carmina et epistolae, ed. C. Luetjohann, MGH AA 8 (Berlin,
1887).
, Epistolae, transl. O.M. Dalton, The letters of Sidonius, 2 vols. (Oxford, 1915).
, Letters, ed. and transl. W.B. Anderson. The Loeb classical library (London
and Cambridge Mass., 193665).
Sokrates Appolinaris, History of the Church, PG 67, cols. 29842.
Snorres Edda, transl. B. Collinder (Stockholm, 1983).
Snorre Sturluson, Nordiska Kungasagor I. Frn Ynglingasagan till Olav Tryggvasons
saga, transl. K.G. Johansson (Stockholm, 1991).
 553

Strabon, Geographica, ed. W. Aly, 4 vols. (Bonn, 1957).

Sulpicius Severus, Chronicorum, ed. C. Halm, CSEL 1 (Wien, 1866), pp. 3105.
, Vita s. Martini, ed. and transl. J. Fontaine, Vie de Saint Martin SC 133 (Paris,
1967).
Supplex Libellus, ed. J. Semmler, in: K. Hallinger ed., CCM 1 (Siegburg, 1963), pp.
31927.
Swanton, M., ed. and transl., Anglo-Saxon Prose (London, 1975).
Tacitus, Annalium libri XVI, ed. H. Furneaux (Oxford, 1965); ed. H. Heubner
(Stuttgart, 1983).
Tacitus, Germania, eds. M. Winterbottom and R.M. Ogilvie, Opera minora (Oxford,
1975); ed. A.A. Lund (Heidelberg, 1988).
Tessier, G., Recueil des actes de Charles II le Chauve, 3 vols. (Paris, 194355).
The Chronicle of Theophanes the Confessor. Byzantine and Near Eastern History A.D. 284 813,
ed. C. de Boor (Leipzig, 1883), transl. C. Mango and R. Scott (Oxford, 1997).
Thegan, Vita Hludowici imperatoris, ed. E. Tremp, MGH SRG 54 (Hannover, 1995).
Theophanes the Confessor, Chronicle, ed. G. Hansen (2nd edition Berlin, 1971).
Thorpe, L., Two Lives of Charlemagne (Harmondsworth, 1969).
Traditiones Wizenburgenses: Die Urkunden des Klosters Weissenburg, 661864, eds. A. Doll
and K. Glckner (Darmstadt, 1979).
Translatio sancti Viti, ed. I. Schmale-Ott. Verffentlichungen der Historischen Kommis-
sion fr Westfalen 41, Fontes Minores 1 (Mnster, 1979).
Una crnica annima de Abd al-Ra m n al-Na r, eds. and transl. E. Garca Gmez and
E. Lvi-Provenal (Madrid, 1959).
Die Urkunden Arnulfs, ed. P. Kehr, MGH DD 3 (Berlin, 1940).
Die Urkunden Pippins, Karlmanns and Karls des Groen, ed. E. Mhlbacher, MGH DD
1 (Berlin, 1906).
Urkundenbuch zur Geschichte der jetzt die Preuischen regierungsbezirke Coblenz und Trier bilden-
den mittelrheinischen Territorien vol. I, ed. H. Beyer e.a. (Koblenz, 1860).
Urkundenbuch des Klosters Fulda, ed. E.E. Stengel (2 vols., Marburg, 191358).
Urkunden- und Quellenbuch zur Geschichte der altluxembourgischen Territorien, ed. C. Wampach,
vol. 1 (Luxembourg, 1935).
Ursinus, Passio altera Leudegarii, ed. B. Krusch, MGH SRM 5 (Hannover, 1910), pp.
32356.
Valerius abbas Bergidensis, De genere monachorum, ed. M.C. Daz y Daz, Ancdota
Wisigothica, I. Estudios ediciones de textos literarios menores de poca visigoda (Salamanca,
1958), pp. 5661.
, Epistola de beatissimae Aetheriae laude, ed. A. Arce, Itinerario de la virgen Egeria
(Madrid, 1980), pp. 816.
Venantius Fortunatus, Opera poetica, ed. F. Leo, MGH Auct. Ant. 4,1 (Berlin, 1881).
, Vita s. Paterni, ed. B. Krusch, MGH AA 4, 2 (Berlin, 1885), pp. 337.
Vigilius Papa, Epistula 1 (ad Profuturum), ed. F. Antonio Gonzlez, Epistolae decretales
ac Rescripta Romanorum Pontificum (Madrid, 1821), pp. 15456.
De virtutibus s. Geretrudis, ed. B. Krusch, MGH SRM 2 (Hannover, 1888), pp. 46474.
De virtutibus s. Martini, ed. B. Krusch, MGH SRM 1,2 (Hannover, 1985).
Vita abbatum Acaunensium, ed. B. Krusch, MGH SRM 3 (Hannover, 1896), pp. 32936.
Vita s. Agili abbatis Resbacensis, ed. J. Mabillon, AASS O.S.B. II, p. 321.
Vita Aldegundae prima, ed. W. Levison, MGH SRM 4 (Hannover, 1902), pp. 878.
Vita Amati, ed. B. Krusch, MGH SRM 4 (Hannover, 1902), pp. 21521.
Vita Anstrudis abbatissae Laudunensis, ed. W. Levison, MGH SRM 4 (Hannover, 1902),
pp. 6678.
Vita Argenteae; Pasionario Hispnico, ed. A. Fabrega Grau (Barcelona/Madrid, 195355).
Vita Audoini, ed. B. Krusch, MGH SRM 5 (Hannover, 1910), pp. 53667.
Vita Balthildis, ed. B. Krusch, MGH SRM 2 (Hannover, 1888), pp. 475508.
Vita beati Maurilii, ed. B. Krusch, MGH AA 4, 2 (Berlin, 1885), pp. 84101.
554

Vita Burchardi, ed. G. Waitz, MGH SS 4 (Hannover, 1841), pp. 82946.

Vita Danielis, ed. H. Delehaye, Les saints stylites (Paris, 1923).
Vita Filiberti abbatis Gemeticensis et Heriensis, ed. W. Levison, MGH SRM 5 (Hannover,
1910), pp. 568606.
Vita s. Fructuosi, ed. M.C. Daz y Daz (Bracara, 1974).
Vita Genovefae, ed. B. Krusch, MGH SRM 3 (Hannover, 1896), pp. 20438.
Vita Geretrudis prima, ed. B. Krusch, MGH SRM 2 (Hannover, 1888), pp. 45263.
Vita Johannis Gorzensis, MGH SS 4 (Hannover, 1841), pp. 33577.
Vita Landiberti auctore Nicolao, ed. B. Krusch, MGH SRM 6 (Hannover and Leipzig,
1913), pp. 40729.
Vita Landiberti vetustissima, ed. B. Krusch, MGH SRM 6 (Hannover, 1913), pp.
35384.
Vita Leutfredi abbatis Madriacensis, ed. B. Krusch, MGH SRM 7 (Hannover, 1920),
pp. 718.
Vita Odiliae abbatissae Hohenburgensis, ed. W. Levison, MGH SRM 6 (Hannover, 1913),
pp. 2450.
Vita Sadalbergae, ed. B. Krusch, MGH SRM 5 (Hannover, 1910), pp. 4966.
Vita Severini abbatis Acaunensis, ed. B. Krusch, MGH SRM 3 (Hannover, 1896), pp.
16870.
Vita Trudonis, ed. W. Levison, MGH SRM 4 (Hannover, 1902), pp. 27398.
Vitae Caesarii episcopi Arelatensis libri duo, ed. B. Krusch, MGH SRM 3 (Hannover,
1896), pp. 433501.
Vitae patrum Jurensium, ed. F. Martine, Vies des Pres du Jura, SC 142 (Paris, 1968).
Vitae sanctorum patrum Emeritensium, ed. A. Maya Snchez (Turnhout, 1992).
Wandelbert of Prm, Miracula s. Goaris, ed. O. Holder-Egger, MGH SS 15, 1
(Hannover, 1887), pp. 36173.
The Mission of Friar William of Rubruck. His Journey to the Great Khan Mngke, 125355,
ed. and transl. by Peter Jackson (London, 1990).

Acin Almansa, M., Entre el feudalismo y el Islam. Umar ibn Hafsun en los historiadores,
en las fuentes y en la historia ( Jan, 1994).
Actes du XI e Congrs international darchologie chrtienne. Studi di antichit cristiana 41
(Rome, 1989).
Airlie, S., Bonds of power and bonds of association in the court circle of Louis
the Pious, in: P. Godman and R. Collins eds., Charlemagnes heir: New perspectives
on the reign of Louis the Pious (81440) (Oxford, 1989), pp. 191204.
, True teachers and pious kings: Salzburg, Louis the German and Christian
order in: R. Gameson and H. Leyser eds., Belief and Culture in Medieval Europe
(Oxford, 2001), forthcoming.
Ajot J. and D. Coxal, Chelles, Gallia informations: Prhistoire et histoire (1993), pp.
5860.
Albrectsen, E., Fynske Jernaldergrave IV, Bd. 12 (Odense, 1971).
Althoff, G., Amicitiae und Pacta. Bndnis, Einung, Politik und Gebetsgedenken im beginnen-
den 10. Jht. Schriften der MGH 37 (Hannover, 1992).
Amato, P., De vera effigie Mariae: Antiche icone Romane (Milan, 1988).
Ament, H., Frnkische Adelsgrber von Flonheim in Rheinhessen. Germanische Denkmler
der Vlkerwanderungszeit B 5 (Berlin, 1970).
, Merowingische Grabhgel, in: W. Schlesinger ed., Althessen im Frankenreich.
Nationes: Historische und philologische Untersuchung zur Entstehung der euro-
paschen Nationen im Mittelalter 2 (Sigmaringen, 1975), pp. 6393.
DAmico, R., Lorganizzazione assistenziale: le diaconie, in: Roma e let carolingia:
Atti del giornati di studio 38 maggio 1976 (Rome, 1976), pp. 22936.
 555

Andoloro, M., La datazione della tavola di S. Maria in Trastevere, Rivista del-

listituto nazionale darcheologia e storia dellarte 19/20 (1972/73 [1975]), pp. 139215.
Andrn, A., Guld og makten tolkning av de skandinaviska guldbrakteaternas
funktion, in: C. Fabech and J. Ringtved eds., Samfundsorganisation og Regional
Variation. Jysk Arkeologisk Selskabs Skrifter XXVII (rhus, 1991), pp. 24555.
, Vrlden frn Lunds horisont, in: C. Wahl ed., Metropolis Daniae. Ett stycke
Europa (Lund, 1998), pp. 11730.
, Frn antiken till antiken, in: S. Thorman and M. Hagdahl eds., Satden,
himmel eller helvete (Stockholm, 1998), pp. 14293.
, Landscape and settlements as utopian space, in: C. Fabech and J. Ringtved
eds., Settlement and landscape. Proceedings of a Conference in rhus, Denmark, May 47
1998. (rhus, 1999), pp. 38394.
Anton, H.H., Studien zu den Klosterprivilegien der Ppste im frhen Mittelalter. Beitrge zur
Geschichte und Quellenkunde des Mittelalters 4 (Berlin, 1975).
Appadurai, A., Introduction: commodities and the politics of value, in: A. Appadurai
ed., The social life of things: Commodities in a cultural perspective (Cambridge, 1988),
pp. 363.
Arbesmann R. and W. Humfner, Iordanis de Saxonia Liber Vitas Fratrum (New York,
1943).
Arce, J., Orbis Romanus y Finis Terrae, in: C. Fernndez Ochoa ed., Los finisterre
atlnticos en la Antigedad. poca preromana y romana (Madrid, 1996), pp. 714.
Arrhenius, B., Ein Goldschmiedegrab von Hovgrdsberg Vendel, Uppland, Sweden,
Frhmittelalterliche Studien 13 (1979), pp. 393414.
Ashley, K. and P. Sheingorn, Writing Faith: Text, sign and history in the miracles of Sainte
Foy (Chicago, 1999).
Astill, G. and W. Davies, A Breton landscape (London, 1997).
Atlas du sous-sol archologique des centres urbains anciens. Dinant (Bruxelles, 1988).
Axboe, M., Guld og guder i folkevandringstiden, in: C. Fabech and J. Ringtved
eds., Samfundsorganisation og Regional Variation. Jysk Arkeologisk Selskabs Skrifter
XXVII (rhus, 1991), pp. 187200.
, Gudme and the gold bracteates, in: P.O. Nielsen e.a. eds., The archaeology
of Gudme and Lundeborg (Copenhagen, 1994), pp. 6877.
Baillien, H., Tongeren. Van Romeinse civitas tot middeleeuwse stad (Assen, 1979).
Baldovin, J., The urban character of Christian worship. The origins, development, and mean-
ing of stational liturgy. Orientalia christiana analecta 228 (Rome, 1987).
Balzaretti, R., The monastery of SantAmbrogio and dispute settlement in early
medieval Milan, EME 3 (1994), pp. 118.
Balzer, M., . . . et apostolicus repetit quoque castra suorum. Vom Wohnen im Zelt im
Mittelalter, Frhmittelalterliche Studien 26 (1992), pp. 20829.
Barahona Simoes, M., Finis Terrae: the land where the Atlantic Ocean begins,
History of the European Ideas 15, 46 (London, 1992), pp. 8538.
Barb, A.A., The survival of the magic arts, in: A. Momigliano ed., The conflict
between paganism and christianity in the fourth century (Oxford, 1963), pp. 10025.
Barber, P., Vampires, burial and death: Folklore and reality (New Haven, 1988).
Barbero A. and M. Vigil, La organizacin social de los cntabros y sus transfor-
maciones en relacin con los orgenes de la Reconquista, in: Sobre los orgenes
sociales de la Reconquista (Barcelona, 1974), pp. 13995.
Barcel, M., El califa patente: el ceremonial omeya de Cordoba o la escenificacin
del poder, in: R. Pastor, I. Kieniewicz, E. Garcia de Enterria e.a. eds., Estructuras
y formas del poder en la historia (Salamanca, 1991), pp. 5171.
Barlow J., Gregory of Tours and the myth of the Trojan origins of the Franks,
Frhmittelalterliche Studien 29 (1995), pp. 8695.
Barnes, T.D., Constantine and Eusebius (Cambridge, MA, 1981).
Barrett, J., Fields of discourse: reconstituting a social archaeology, Critique of
Anthropology 7, 3 (1988), pp. 516.
556

, Fragments from Antiquity. Social Archaeology Series (Oxford, 1994).

Barrett, J.C., R. Bradley and M.Green eds., Landscape, monuments and society. The pre-
history of Cranborne Chase (Cambridge, 1991).
Barth, F., Nomads of South Persia (New York, 1965).
, Pastoral production and society (Cambridge, 1979).
Bassett, S. ed., Death in towns: Urban responses to the dying and the dead, 1001600
(Leicester, 1992).
Baynes, N.H., The Supernatural Defenders of Constantinople, Analecta Bollandiana
67 (1949), pp. 16577.
Bazelmans, J., Conceptualising early Germanic political structure: a review of the
use of the concept of Gefolgschaft, in: N. Roymans and F. Theuws eds., Images
of the past (Amsterdam, 1991), pp. 91130.
, The gift in the Old English epic Beowulf Unpublished lecture given at the
symoposium Theory and Method in the Study of Material Culture, Leiden 31./8.
2./9.1992.
, Beyond power. Ceremonial exchanges in Beowulf , in: F. Theuws and J.L.
Nelson eds., Rituals of power from late Antiquity to the early Middle Ages, TRW 8
(Leiden/Boston/Kln, 1999), pp. 31176.
, By weapons made worthy: Lords, retainers and their relationship in Beowulf (Amsterdam,
1999).
Bksted, A., Nordiske Guder og Helte (Copenhagen, 1990).
Becher, M., Der sogenannte Staatsstreich Grimaolds. Versuch einer Neubewertung,
in: J. Jarnut, U. Nonn, M. Richter eds., Karl Martell in seiner Zeit (Sigmaringen,
1994), pp. 11947.
Belke, K. and P. Soustal eds., Die Byzantiner und ihre Nachbarn (Wien, 1995).
Bellinger, A.E. and P. Grierson, Catalogue of the Byzantine Coins in the Dumbarton Oaks
Collection. Vol. 2 in 2 parts (Washington, DC, 1968).
Belting, H., I mosaici dellaula Leonina come testimonianza della prima renovatio
dellarte medievale di Roma, in: Roma e let carolingia: Atti del giornati di studio
38 maggio 1976 (Rome, 1976), pp. 167182.
, Die beiden Palastaulen Leos III. und die Entstehung einer ppstlichen
Programmkunst, Frhmittelalterliche Studien 12 (1978), pp. 5583.
, Papal artistic commissions as definitions of the Church in Rome, in:
H. Hager and S.S. Munshower eds., Light on the Eternal City: Observations and dis-
coveries in the art and architecture of Rome. Papers in Art History from the Pennsylvania
State University 2 (University Park, PA, 1987), pp. 1330.
Berg, S., R. Rolle and H. Seemann, Der Archologe und der Tod. Bucher Report Series
(Mnchen/Luzern, 1981).
Bergengrn, A., Adel und Grundherrschaft im Merovingerreich (Wiesbaden, 1958).
Berkum, A. van, Op zoek naar de meest geschikte residentie. Geruisloze ver-
huizingen van de Episcopi Tungrorum. Een poging tot reconstructie, in: C.G. de
Dijn ed., Sint-Servatius. Bisschop van Tongeren-Maastricht. Het vroegste Christendom in het
Maasland. Handelingen van het colloquium te Alden Biesen (Bilzen), Tongeren en Maastricht
1984 (Borgloon-Rijkel, 1986), pp. 14762.
Berlire, U., Le nombre des moines dans les anciens monastres, Revue Bndictine
41 (1929), pp. 23161 and 42 (1930), pp. 1942.
Bernhardt, J.W., Itinerant kingship and royal monasteries in early medieval Germany, 9361075
(Cambridge, 1993).
Bernoulli, C.A., Die Heiligen der Merowinger (Tbingen, 1900).
Bertelli, C., La Madonna di Santa Maria in Trastevere: Storia, Iconografia, Stile di un dip-
into dellottavo secolo (Rome, 1961).
Bertholet, P. and P. Hoffsummer, Lglise halle des Saints Hermes et Alexandre Theux
(Dison, 1986).
Bertolini, O., Per la storia delle diaconie romane nellalto medioevo sino alle fine
 557

del secolo VIII, in: O. Banti ed., Scritti scelti di storia medioevale (Livorno, 1968),
pp. 309460.
Besson, M., La donation dAyroenus Saint-Maurice (mardi 8 octobre 765), Zeit-
schrift fr schweizerische Kirchengeschichte/Revue dhistoire ecclsiastique suisse 3 (1909), pp.
2946.
Bierbrauer, V., Die Goten vom 1.7. Jahrhundert n. Chr.: Siedelgebiete und Wan-
derbewegungen aufgrund archologischer Quellen, Peregrinatio Gotica 3. Universitetets
Oldsaksamlings Skrifter NY rekke 14 (Oslo, 1992), pp. 943.
, Archologie und Geschichte der Goten von 1.7. Jahrhundert. Versuch einer
Bilanz, Frhmittelalterliche Studien 28 (1994), pp. 15171.
Biddle, M., Towns, in: D.M. Wilson ed., The archaeology of Anglo-Saxon England
(London, 1976), pp. 99150.
Billard, C., F. Carr, M. Guillon and C. Treffort, Loccupation funraire des mon-
uments mgalithiques pendant le haut moyen ge. Modalits et essai dinterpr-
tation, Bulletin de la Socit Prhistorique Franaise 93 (1996), pp. 27986.
Biscop, J.-L. and J.-P. Sodini, Travaux Qalat Seman, Acts of the 11th International
Congress of Christian Archaeology (Rome, 1989), pp. 167593.
Bishko, Ch.J., The date and nature of the Spanish Consensoria Monachorum, American
Journal of Philology 69 (1948), pp. 37795.
, Gallegan pactual monasticism in the repopulation of Castille, Estudios de-
dicados a Menndez Pidal II (Madrid, 1951), pp. 51331 (Published with an Additional
Note in: idem, Spanish and Portuguese monastic history, 6001300 (London, 1984),
pp. 51331 and 532A36A).
, Episcopus sub regula or episcopi sub regula? St. Fructuosus and the monasticized
episcopate in the peninsular west, Bracara Augusta 21 (1967), pp. 634.
, The pactual tradition in Hispanic Monasticism, in: idem, Spanish and Portuguese
monastic history, 6001300 (London, 1984), pp. 143.
Bitel, L.M., Isle of the saints. Monastic settlement and Christian community in early Ireland
(Cork, 1990).
Bitner-Wrblewska, A., The Southeastern Baltic zone and Scandinavia in the early
Migration Period, Barbaricum 2 (1992), pp. 24566.
Blaauw, S. de, Cultus et Decor: liturgia e architettura nella Roma tardoantica e medievale.
Basilica Salvatoris, Sanctae Mariae, Sancti Petri, 2 vols. (Vatican City, 1994).
Blair, J. ed., Minsters and parish churches: the local church in transition, c. 950 1200.
Oxbow Monograph 17 (Oxford, 1988).
Bleichsteiner, R., Zeremonielle Trinksitten und Raumordnung bei den turko-mon-
golischen Nomaden, Archiv fr Vlkerkunde 6/7 (1951/52), pp. 181208.
Bloch, M., Placing the dead: tombs, ancestral villages and kinship organisation in Madagascar
(London and New York, 1971).
Bloch, M. and J. Parry, Introduction: money and the morality of exchange, in:
J. Parry and M. Bloch eds., Money and the morality of exchange (Cambridge, 1989),
pp. 132.
Blockley, R.C., The fragmentary classicising historians of the later Roman Empire: Eunapius,
Olympiodorus, Priscus and Malchus (Liverpool, 1983).
, ed., The History of Menander the Guardsman (Liverpool, 1985).
Bloemers, J.H.F., Twenty five years of ROB Research in Roman Limburg, Berichten
van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 23 (1973), pp.
237258.
Blondel, L., Les anciennes basiliques dAgaune. tude archologique, Vallesia 3
(1948), pp. 957.
, Aperu sur les difices chrtiens dans la Suisse occidentale avant lan mille,
in: Frhmittelalterliche Kunst in den AlpenlndernArt du haut moyen ge dans la rgion
alpineArte dellalto medio evo nella regione alpina, Actes du III e congrs international pour
ltude du haut moyen age, 914 septembre 1951 (Olten, 1954), pp. 271308.
558

, Le prieur Saint-Victor. Les dbuts du christianisme et la royaut burgonde

Genve, Bulletin de la socit dhistoire et darchologie de Genve 11 (1958), pp.
21158.
, La rampe daccs la basilique dAgaune. Une rectification, Vallesia 22
(1967), pp. 13.
Boddington, A., Models of burial, settlement and worship: the final phase reviewed,
in: E. Southworth ed., Anglo-Saxon cemeteries: A reappraisal (Stroud, 1990), pp. 17799.
Boe, G. de, De archeologische getuigen van het eerste christendom in de Civitas
Tungrorum, in: C.G. de Dijn ed., Sint-Servatius, bisschop van Tongeren-Maastricht. Het
vroegste Christendom in het Maasland. Handelingen van het colloquium te Alden Biesen (Bilzen),
Tongeren en Maastricht 1984 (Borgloon, Rijkel, 1986), pp. 3762.
Boesch Gajano, S., Reliques et pouvoirs, in: E. Bozoki and A-M. Helvetius eds.,
Les reliques. Objets, cultes, symboles (Turnhout, 1999), pp. 25569.
Bhmer, J. F. with E. Mhlbacher, Regesta Imperii I. Die Regesten des Kaiserreiches unter
den Karolingern 751918 (2nd edn., Innsbruck, 1908).
Bondois, M., La translation de Saints Marcellin et Pierre. tude sur Einhard et sa vie poli-
tique de 827 834 (Paris, 1907).
Bonnet, C., Les fouilles de lancien groupe piscopal de Genve (19761993). Cahiers
darchologie genevoise 1 (Geneva, 1993).
Borderie, A. De La, La chronologie du cartulaire de Redon, Annales de Bretagne
5 (1890), pp. 535630.
Bosl, K., Pfalz und Forsten, in: Deutsche Knigspfalzen I. Verffentlichungen der
Max-Planck Instituts fr Geschichte 11/i (Gttingen, 1963), pp. 129.
Bourdieu, P., Le sens de lhonneur, in: idem, Esquisse dune thorie de la pratique,
prcde de trois tudes dethnologie kabyle (Genve, 1972), pp. xxx.
, Entwurf einer Theorie der Praxis (Frankfurt, 1979).
, Rethinking the State, Sociological Theory 12 (1994), repr. in: idem, Practical
Reason (1998), pp. 3563.
Boureau, A., Laigle: chronique politique dun embleme (Paris, 1985).
Bradley, R., Time regained: the creation of continuity, Journal of British Archaeological
Association 140 (1987), pp. 117.
, The passage of arms: An archaeological analysis of prehistoric hoards and votive deposits
(Cambridge, 1990).
, Monuments and places, in: P. Garwood, D. Jennings, R. Skeats and
J. Toms eds., Sacred and profane: Proceedings of a conference on archaeology, ritual and reli-
gion. Oxford, 1989. Oxford University Committee for Archaeology Monograph 32
(Oxford, 1991), pp. 13540.
, Ritual, Time and History, World Archaeology 23 (1991), pp. 20919.
, Altering the earth: The origins of monuments in Britain and Continental Europe (Edinburgh,
1993).
, Working the land: imagining the landscape, Archaeological Dialogues 4, 1
(1997), pp. 3952.
, The significance of monuments (London, 1998).
, Ruined buildings, ruined stones: enclosures, tombs and natural places in the
neolithic of south-west England, World Archaeology 30, 1 (1998), pp. 1322.
, Directions to the dead, in: L. Larsson and B. Stjernquist eds., The world
view of prehistoric man. KVHAA Konferenser 40 (Lund, 1998), pp. 12335.
, Sacred Geography, World Archaeology 28 (1998), pp. 161274.
Bradley, R. and H. Williams, eds., The past in the Past: The re-use of ancient monuments
(London, 1998).
Braemer, F., Lambre lpoque romaine. Problmes dorigine, de commerce par
terre et par mer et de lieu de faonnage des objets, notamment figurs, in:
Colloque international sur les ressources minrales et lhistoire de leur exploitation (Grenoble,
1983), pp. 36181.
 559

Brandes, W., Byzantine cities in the seventh and eighth centuriesdifferent sources,
different histories?, in: G. Brogiolo and B. Ward-Perkins eds., The idea and ideal
of the town between late Antiquity and the early Middle Ages, TRW 4 (Leiden/Boston/Kln,
1999), pp. 2557.
Brather, S., Frhmittelalterliche DirhamSchatzfunde in Europa. Probleme ihrer
wirtschaflichen Interpretation aus archologischer Perspektive, Zeitschrift fr Archologie
des Mittelalters 23/24 (1996), pp. 73153.
Brenk, B., Early gold mosaics in Christian Art, Palette 38 (1972), pp. 1625.
, Die frhchristlichen Mosaiken in Santa Maria Maggiore zu Rom (Wiesbaden, 1975).
, Spolia from Constantine to Charlemagne: Aesthetics versus ideology, Dumbarton
Oaks Papers 41 (1987), pp. 1039.
Brennecke, H.C., Servatius von Tongern. Ein gallischer Bischof im arianischen
Streit, in: C.G. de Dijn ed., Sint-Servatius. Bisschop van Tongeren-Maastricht. Het vroeg-
ste Christendom in het Maasland. Handelingen van het colloquium te Alden Biesen (Bilzen),
Tongeren en Maastricht 1984 (Borgloon-Rijkel, 1986), pp. 1736.
Bresslau, H., Handbuch der Urkundenlehre fr Deutschland und Italien 2, 2. (4th edition
Berlin, 1969).
Brink, S., Political and social structures in Early Scandinavia, Tor 28 (Uppsala,
1996), pp. 23581.
Brogiolo, G.P. and B. Ward-Perkins, The idea and ideal of the town between late Antiquity
and the early Middle Ages. TRW 4 (Leiden/Boston/Kln, 1999).
Brown, P., Sorcery, demons, and the rise of Christianity from late Antiquity into
the Middle Ages, in: M. Douglas ed., Witchcraft: Confessions and Accusations (London,
1970), pp. 1745.
, The rise and function of the holy man in late Antiquity, Journal of Roman
studies 61 (1971), pp. 80101; repr. in: idem, Society and the holy in Late Antiquity
(Berkeley, 1982), pp. 10352.
, The cult of saints: Its rise and function in Latin Christianity (Chicago, 1981).
, Eastern and western Christendom in late Antiquity: a parting of the ways,
in: idem, Society and the Holy in Late Antiquity (Berkeley, 1982), pp. 16695.
, Relics and social status in the age of Gregory of Tours, in: idem, Society
and the holy in late Antiquity (London, 1982), pp. 22250.
, Power and persuasion in Late Antiquity (Madison, WI, 1992).
, The rise of western Christendom: Triumph and diversity A.D. 200 1000 (Malden,
1996).
Bruce-Mitford, R.L.S., The Sutton Hoo ship burial, 3 vols. (London, 19751983).
Brhl, C.-R., Knigspfalz und Bischofstadt in frnkischer Zeit, Rheinische Vierteljahrsblt-
ter 23 (1958), pp. 161274.
, Hinkmariana 1, Deutsches Archiv 29 (1964), pp. 4877.
, Fodrum, Gistum, Servitium Regis: Studien zu den wirtschaftlichen Grundlagen des Knigtums
im Frankenreich und in den frnkischen Nachfolgestaaten Deutschland, Frankreich und Italien
vom des bis zur Mitte des 14. Jahrhunderts, 2 vols. (Kln and Graz, 1968).
, Remarques sur les notions de capitale et de rsidence pendant le haut
Moyen Age, Journal des Savants, 1967, pp. 193215, repr. in: idem, Aus Mittelalter
und Diplomatik. Gesammelte Aufstze, 2 vols. (Hildesheim/Mnchen/Zrich, 1989),
vol. 1, pp. 11537.
Brunnert, M.-E., Fulda als Kloster in eremo. Zentrale Quellen ber die Grndung
im Spiegel der hagiographischen Tradition in: G. Schrimpf ed., Das Kloster Fulda
in der Welt der Karolinger und Ottonen (Frankfurt, 1996), pp. 5978.
Bhler, A., Capitularia Relecta. Studien zur Entstehung und berlieferung der Kapi-
tularien Karls des Grossen und Ludwigs des Frommen, Archiv fr Diplomatik 32
(1986), pp. 305501.
Bhnen, S., Haus und Grab: gebaute Kosmologie und Ideologie, Ethnographisch-
Archologische Zeitschrift 38 (1997), pp. 45155.
560

Bullough, D.A., Early medieval social groupings: the terminology of kinship, Past
and Present 45 (1969), pp. 318.
, Burial, Community and belief in the early medieval West, in: P. Wormald
ed., Ideal and Reality in Frankish and Anglo-Saxon Society: Studies Presented to J.M. Wallace-
Hadrill (Oxford, 1983), pp. 177201.
Bullough, D., Carolingian renewal: Sources and heritage (Manchester, 1991).
, Aula renovata: the Carolingian court before the Aachen palace, in: D. Bullough,
Carolingian renewal: Sources and heritage (Manchester, 1991), pp. 12360.
Bund, K., Thronsturz und Herrscherabsetzung im Frhmittelalter. Bonner Historische
Forschungen 44 (Bonn, 1979).
Burgess, R.W., The Chronicle of Hydatius and the Consularia Constantinopolitana.
Two contemporary accounts of the final years of the Roman Empire (Oxford, 1993).
Bursche, A., Later Roman-Barbarian contacts in Central Europe. Numismatic evidence (Berlin,
1996).
Busch, J.W., Vom Attentat zur Haft: die Behandlung von Konkurrenten und
Opponenten der frhen Karolinger, Historische Zeitschrift 263 (1996), pp. 56188.
Bttner, H., Zur Stadtentwicklung von Worms im Frh- und Hochmittelalter, in:
Aus Geschichte und Landeskunde: Festschrift F. Steinbach (Bonn, 1960), pp. 389407.
, Ladenburg am Neckar und das Bistum Worms bis zum Ende des 12.
Jahrhunderts, Archiv fr Hessische Geschichte und Altertumskunde 28 (1963), pp. 8398.
, Die politische Krfte zwischen Rhein und Odenwald bis zum 11. Jht., in
his collected essays, ed. A. Gerlich, Zur frhmittelalterlichen Reichsgeschichte am? Rhein,
Main und Neckar (Darmstadt, 1975), pp. 25366.
, Mainz im Mittelalter: Gestalten und Probleme, in his collected essays, ed.
A. Gerlich, Mittelrhein und Hessen: Nachgelassene Studien (Stuttgart, 1989), pp. 150.
Cabrera, J., Estudios sobre el priscilianismo en la Galicia antigua (Granada, 1983).
Caciola, N., Wraiths, revenants and ritual in medieval culture, Past and Present
152 (1996), pp. 345.
Cameron, A., The early religious politics of Justin II, in: D. Baker ed., The ortho-
dox churches and the West. Studies in Church History 13 (Oxford, 1976), pp.
5167.
, Images of authority: lites and icons in late sixth-century Constantinople,
Past and Present 84 (1979), pp. 335, repr. in eadem, Continuity and change in sixth-
century Byzantium (London, 1981), study XVIII.
Cameron, A., Virginity as metaphor, in: eadem ed., History as text (London, 1989).
Cannadine, D., War and death, grief and mourning in modern Britain, in:
J. Whaley ed., Mirrors of mortality: Studies in the social history of death (London, 1981),
pp. 187242.
Capelle, T., Handwerk in der Karolingerzeit, Kunst und Kultur der Karolingerzeit. Karl
der Grosse und Papst Leo III in Paderborn (Mainz, 1999), pp. 4249.
Cardot, F., Le pouvoir aristocratique et le sacr au haut Moyen-Age: Sainte Odile
et les Etichonides dans la vita Odiliae, Le Moyen Age 89 (1983), pp. 17393.
Carmichael, D., Places of power: Mescalero Apache sacred sites and sensitive
areas, in: D.L. Carmichael, J. Hubert, B. Reeves and A. Schanche eds., Sacred
Sites, Sacred Places. One World Archaeology (London, 1994), pp. 8998.
Carroll, C., The bishoprics of Saxony in the first century after Christianization,
EME 8 (1999), pp. 21945.
Carver, M.O.H., Arguments in Stone. Archaeological Research and the European Town in the
first Millennium (Oxford, 1993).
Carver, M., Sutton Hoo: Burial ground of kings? (London, 1998).
Cecchelli, C., Mater Christi 1 (Rome, 1946).
Chadwick, N.K., The Age of the saints in the early Celtic Church (Oxford, 1961).
Chapman, R., The emergence of formal disposal areas and the problem of mega-
liths in prehistoric Europe, in: R. Chapman, I. Kinnes and K. Randsborg eds.,
The archaeology of death (Cambridge, 1981), pp. 7181.
 561

Chdeville, A. and H. Guillotel, La Bretagne des saints et des rois, V eX e sicle (Rennes,
1984).
Chdeville, A. and N.-Y. Tonnerre, La Bretagne fodale, XI eXIII e sicle (Rennes, 1987).
Childe, V.G., Directional changes in funerary practices during 50,000 years, Man
45 (1945), pp. 139.
Christlein, R., Besitzabstufungen zur Merowingerzeit im Spiegel reicher Grabfunde
aus West- und Sddeutschland, Jahrbuch des Rmisch-Germanischen Zentralmuseums
20 (1973), pp. 14780.
Christoffersen, J., MllegrdsmarkenStruktur und Belegung eines Grberfeldes,
Frhmittelalterliche Studien 21 (1987), pp. 85100.
Clark, G., Space, time and man: A prehistorians view (Cambridge, 1992).
Classen, P., Bemerkungen zur Pfalzenforschung am Mittelrhein, in: Deutsche
Knigspfalzen I. Verffentlichungen des Max-Planck Instituts fr Geschichte 11/i
(Gttingen, 1963), pp. 7596.
, Die Geschichte der Knigspfalz Ingelheim bis zur Verpfndung an Kurpfalz
1375, in: J. Autenrieth ed., Ingelheim am Rhein: Forschungen und Studien zur Geschichte
Ingelheims (Ingelheim, 1964), pp. 87146.
, Fortleben und Wandel sptrmischen Urkundenwesens im frhen Mittelalter,
in: idem ed., Recht und Schrift im Mittelalter. Vortrge und Forschungen 23 (Sig-
maringen, 1977), pp. 1354.
, Karl der Groe, das Papsttum und Byzanz: Die Begrndung des karolingischen Kaisertums.
Beitrge zur Geschichte und Quellenkunde des Mittelalters 9 (Sigmaringen, 1985).
Claude, D., Die byzantinische Stadt im 6. Jahrhundert. Byzantinisches Archiv 13 (Mnchen,
1969), pp. 20819.
Clement, J.M., Lexique des anciennes rgles monastiques occidentales, 2 vols. (Steenbrugge,
1978).
Clunies Ross, M., Prolonged echoes. Old Norse myths in medieval Northern society 1: The
Myths (Odense, 1994); 2: The reception of Norse myths in medieval Iceland
(Odense, 1998).
Coates-Stephens, R., Housing in early medieval Rome, A.D. 5001000, PBSR 64
(1996), pp. 23959.
, Dark Age architecture in Rome, PBSR, 65 (1997), pp. 177232.
Costambeys, M., An aristocratic community on the northern Frankish frontier,
690726, EME 3 (1994), pp. 3962.
, Piety, property and power in eighth-century central Italy (unpublished Ph.D. thesis,
Cambridge, 1998; forthcoming in book form).
Conte, P., Regesto delle lettere dei papi del secolo VIII (Milan, 1984).
Cooper, K., The martyr, the matrona and the bishop: networks of allegiance in
early sixth-century Rome, Journal of Roman Archaeology, forthcoming.
Couffon, R., LArchitecture religieuse en Bretagne, VeXe sicles, Mmoires de la
Socit dHistoire et dArchologie de Bretagne 23 (1943), pp. 140.
Coulson, C., Fortresses and social responsibility in late Carolingian France, Zeitschrift
fr die Archologie des Mittelalters 4 (1976), pp. 2937.
Coupland, S., In palatio nostro: les monnaies palatines de Charlemagne, Bulletin de
la socit franaise de numismatique 41 (1986), pp. 879.
Cracco Ruggini, L., El xito de los Priscilianistas: a propsito de cultura y fe en
el siglo IV d.C., in: R. Teja and C. Prez, Congreso Internacional la Hispania de
Teodosio (Valladolid/Segovia, 1997), pp. 3947.
Croquison, L., Liconographie chrtienne Rome daprs le Liber Pontificalis,
Byzantion 34 (1964), pp. 534606.
Crow, J., The long walls of Thrace, in: C. Mango and G. Dagron eds., Constantinople
and its hinterland (Aldershot, 1995), pp. 10924.
Cppers, H., Beitrge zur Geschichte des rmischen Kur- und Badeortes Aachen,
in: idem ed., Aquae Granni. Beitrge zur Archeologie von Aachen (Kln, 1982), pp.
175.
562

Cutler, A., The ninth-century Spanish martyrs movement and the origins of west-
ern Christian missions to the Muslims, The Muslim World 15 (1965), pp. 32139.
Czapkiewicz, M., M. Jagodzinski and Z. Kmietowicz, Arabische Mnzen aus einer
frhmittelalterlichen Handwerker- und Handelsiedlung in Janw Pomorski, Gem.
Elblog, Folia Orientalia 25 (1988), pp. 15769.
Dagron, G., Le christianisme dans la ville byzantine, Dumbarton Oaks Papers 31
(1977), pp. 325. (repr. in idem, La romanit chrtienne en Orient (London, 1984),
study IX).
, Empereur et prtre. Etude sur le csaropapisme byzantin (Paris, 1996).
Dahlhaus-Berg, E., Nova Antiquitas et Antiqua Novitas: typologische Exegese und isido-
rianische Geschichtsbild bei Theodulf von Orleans (Kln, 1975).
Dahn, F., Die Knige der Germanen (Mnchen, 18611901).
Danforth, L., The death rituals of rural Greece (Princeton, N.J., 1982).
David, P., Lorganisation ecclsiastique du royaume suve au temps de Saint Martin
de Bracara, in: idem, Etudes historiques sur la Galice et le Portugal du VI e au XII e
sicle (Lisboa/Paris, 1947), pp. 182.
Davidson, H.E. and P. Fisher, Saxo Grammaticus: The History of the Danes, Books IIX,
vol. 2 (Cambridge/Totowa, N.J., 1980).
Davies, W., Disputes, their conduct and their settlement in the village communities
of eastern Brittany in the ninth century, History and Anthropology 1 (1985), pp. 289312.
, and P. Fouracre eds., The settlement of disputes in early medieval Europe (Cambridge,
1986).
, People and places in dispute in ninth-century Brittany in: W. Davies and
P. Fouracre eds., The settlement of disputes in early medieval Europe (Cambridge, 1986),
pp. 6484.
, Small worlds: The village community in early medieval Brittany (London, 1988).
, Forgery in the Cartulaire de Redon, in: Flschungen im Mittelalter. MGH
Schriften IV (Mnchen, 198889), pp. 26574.
, The composition of the Redon cartulary, Francia 17/i (1990), pp. 6990.
, and P. Fouracre eds., Property and power in the early Middle Ages (Cambridge,
1995).
, Adding insult to injury: power, property and immunities in early medieval
Wales in: W. Davies and P. Fouracre eds., Property and power in the early Middle
Ages (Cambridge, 1995), pp. 13764.
Davis-Weyer, C., Die Mosaiken Leos III. und die Anfnge der karolingischen
Renaissance in Rom, Zeitschrift fr Kunstgeschichte 29 (1966), pp. 11132.
De La Haye, R., De bischoppen van Maastricht (Maastricht, 1985).
, In welke eeuw leefde Sint Servaas? De Maasgouw, tijdschrift voor Limburgse
geschiedenis en oudheidkunde 113 (1994), col. 528.
, Het middeleeuws officie van het hoogfeest van Sint Servaas te Maastricht,
Publications de la Socit Historique et Archologique dans le Limbourg 133 (1997), pp.
93140.
Debord, G., La socit du spectacle (Paris, 1967).
Deckers, J., Tradition und Adaption. Bemerkungen zur Darstellung der christlichen
Stadt, Mitteilungen des deutschen archaeologischen Instituts, Rmische Abteilung 95 (1988),
pp. 30382.
Declercq, G., Originals and cartularies: the organization of archival memory (ninth
to eleventh centuries) in: K. Heidecker ed., Charters and the use of the written word
in medieval society (Turnhout, 2000), pp. 14770.
Der, J., Die Vorrechte des Kaisers in Rom, 772800, Schweizer Beitrge zur allge-
meinen Geschichte 15 (1957), pp. 563.
Deeters, J., Servatiusstift und Stadt Maastricht. Untersuchungen zu Entstehung und Verfassung
(Bonn, 1970).
Delogu, P., The rebirth of Rome in the eighth and ninth centuries, in: R. Hodges
 563

and B. Hobley eds., The rebirth of towns in the West A.D. 7001050. CBA Research
Report 68 (London, 1988), pp. 3342.
, Oro e argento in Roma tra il VII e il IX secolo, in: Cultura e societ nellItalia
medievale: Studi per Paolo Brezzi. Istituto Storico Italiano, Studi Storici 184187
(Rome, 1988), pp. 27393.
, and L. Paroli eds., La storia economica di Roma nellalto medioevo all luce dei recenti
scavi archeologici. Bibliotheca di Archeologia Medievale 10 (Florence, 1993).
, La storia economica di Roma nellalto medioevo: Introduzione al seminario,
in: idem and L. Paroli eds., La storia economica di Roma nellalto medioevo all luce dei
recenti scavi archeologici. Bibliotheca di Archeologia Medievale 10 (Florence, 1993),
pp. 1129.
, ed., Roma medievale: Aggiornamenti (Florence, 1998).
, Limportazione di tessuti preziosi e il sistema economico romano nel IX
secolo, in: idem ed., Roma medievale: Aggiornamenti (Florence, 1998), pp. 12341.
Delort, E., Berthelming (Moselle), Gallia 6 (1948), pp. 23941.
DeMarrais, E., L.J. Castillo and T. Earle, Ideology, materialization, and power
strategies, Current Anthropology 37, 1 (1996), pp. 1531.
Demolon, P., Cimetire et chapelle mrovingiens Hourdain (Nord), in: Larchologie
en Hainaut-Cambrsis, Avesnois (Valenciennes, 1981), pp. 2930.
Depreux, P., Das Knigtum Berhards von Italien, Quellen und Forschungen aus
Italienischen Archiven und Bibliotheken 72 (1992), pp. 125.
, Prosopographie de lentourage de Louis le Pieux (781840) (Sigmaringen, 1997).
Derks, H., Geschlechtsspezifische Bestattungssitten: ein archologischer Befund und
ein ethnoarchologischer Ansatz, Ethnographisch-Archologische Zeitschrift 34 (1993),
pp. 34053.
Deshman, R., Servants of the Mother of God in Byzantine and medieval art,
Word and Image 5 (1989), pp. 3370.
Deshusses, J. ed., Le sacramentaire grgorien: Ses principales formes daprs les plus ancient
manuscrits. Spicilegium Friburgense 16 (Fribourg, 1971).
Despy, G., Villes et campagnes aux IXe et Xe sicles: lexemple du pays Mosan,
Revue du Nord 50 (1968), pp. 14568.
, Lagglomeration urbaine pendant le haut moyen age (du VIIe sicle aux
environs de 1200), Namur, le site, les hommes de lpoque romaine au XVIII e sicle
(Bruxelles, 1988), pp. 6378.
Devroey, J-P., Les services de transport labbaye de Prm au IXe sicle, Revue
du Nord 61 (1979), pp. 54369 (repr. in: J.-P. Devroey, Etudes sur le grand domaine
Carolingien, (Aldershot, 1993), nr. X.
, Villes, campagnes, croissance agraire dans le pays Mosan avant lan mil:
vingt ans aprs, in: J.-M. Duvosquel and A. Dierkens eds., Villes et campagnes au
moyen ge, Mlanges Georges Despy (Lige, 1991), pp. 223260 (repr. in J.-P. Devroey,
Etudes sur le grand domaine Carolingien (Aldershot, 1993), nr. XII.
, Les premiers polyptyques rmois VII eIXe sicles, in: La grande domaine aux
epoques Merovingiennes et Carolingiennes (Gand, 1985), pp. 7897 (repr. in: J.-P. Devroey,
Etudes sur le grand domaine Carolingien (Aldershot, 1993), n. II.
Daz, C., Formas econmicas y sociales en el monacato visigodo (Salamanca, 1987).
Daz, P.C., Redimuntur captiui. A propsito de Regula Communis IX, Gerion 10
(1992), pp. 28793.
, El Parrochiale Suevum: organizacin eclesistica, poder politico y poblamiento
en la Gallaecia tardo-antigua, in: J. Alvar ed., Homenaje al Prof. J.M . Blzquez VI
(Madrid, 1998), pp. 2537.
, El reino suevo de Hispania y su sede en Bracara, in G. Ripoll and J.M.
Gurt eds., Sedes regiae (ann. 400800), (Barcelona, 2000), pp. 40423.
Daz y Daz, M.C., La Pasin de S. Pelayo y su difusin, Anuario de Estudios
Medievales 6 (1969), pp. 97116.
564

, Vita Sancti Fructuosi (Bracara, 1974).

Diem, A., Keusch und rein. Eine Untersuchung zu den Ursprngen des frhmittelalterlichen
Klosterwesens und seine Quellen (Ultrecht, 2000).
, Was bedeutet regula Columbani?, in: M. Diesenberger and W. Pohl eds.,
Integration und Herrschaft. Ethnische Identitten und soziale Organisation im Frhmittelalter,
Forschungen zur Geschichte des Mittelalters 3 (Wien, 2001; forthcoming).
Dierkens, A. Un aspect de la christianisation de la Gaule du Nord lpoque
mrovingienne. La Vita Hadelini et les dcouvertes archologiques dAnthe et de
Franchimont, Francia 8 (1980), pp. 61328.
, Abbayes et chapitres entre Sambre et Meuse (VII eXII e sicle) (Sigmaringen, 1985).
, Saint Amand et la fondation de labbaye de Nivelles, Saint-Gry et la chris-
tianisation dans le nord de la Gaule, V eIX e sicle, Revue du Nord 69 (1986), pp. 32534.
, Premiers structures religieuses: paroisses et chapitres jusquau XIIe sicle,
Namur. Le site, les hommes de lpoque romaine au XVIII e sicle (Bruxelles, 1988), pp.
3361.
, La ville de Huy avant lan mil. Premier essai de synthse des recherches
historiques et archologiques, Ontstaan en vroegste geschiedenis van de middeleeuwse ste-
den in de zuidelijke Nederlanden (S.L., 1990), pp. 391409.
Diesenberger, M., Wahrnehmung und Aneignung der Natur in den Gesta abbatum
Fontanellensium in: C. Egger and H. Weigl eds., TextSchriftCodex. Quellenkundliche
Arbeiten aus dem Institut fr sterreichische Geschichtsforschung. Mitteilungen des Instituts
fr sterreichische Geschichtsforschung, Ergngzungsband 35 (Wien, 2000), pp.
933.
Dieten, J.L., Geschichte der Patriarchen von Sergios I. bis Johannes VI. (Amsterdam, 1972).
Dijkman, W and A. Ervynck, Antler, bone, horn, ivory and teeth: The use of animal skele-
tal materials in Roman and early medieval Maastricht. Archaeologica Mosana 1 (Maastricht,
1998).
Dimier, A.-M., Le mot locus employ dans le sens de monastre, Revue Mabillon
62 (1972), pp. 13354.
Dodds, J.D. ed., Al-Andalus: The Art of Islamic Spain (New York, 1992).
Dollfus, M.A. and A. Guyot, Spultures de nouveau-ns dans les fouilles gallo-
romaines de Fleurheim Lyons-La-Fort (Eure), Annales de Normandie 18 (1968),
pp. 283300.
Doppelfeld, O. and R. Pirling, Frnkische Frsten im Rheinland. Schriften des Rheinischen
Landesmuseums Bonn 2 (Dsseldorf, 1966).
Douglas, M., Purity and danger: An analysis of the concepts of purity and taboo (London,
1966).
Douglas, M. and B. Isherwood, The world of goods (New York, 1979).
Drijvers, J.W., Helena Augusta: The mother of Constantine the Great and the legend of her
finding of the True Cross (Leiden, 1992).
Dubuis, F.-O. and A. Lugon, Les premiers sicles dun diocse alpin: recherches,
acquises, et questions sur lvch de Sion, pt. 3: Notes et documents pour
servir lhistoire des origines paroissiales, Vallesia 50 (1995), pp. 1359.
Duby, G., The early growth of the European economy. Warriors and peasants from the seventh
to the twelfth century (Ithaca, 1989). Originally published as: Guerriers et Paysans,
VIII eXII e sicles. Premier essor de lconomie europenne (Paris, 1973).
, Ideologies in Social History, in: J. Le Goff and P. Nora eds., Constructing
the Past: Essays in Historical Methodology (Cambridge, 1985), pp. 15165.
Duchesne, L., Les rgions de Rome au moyen ge, Mlanges darchologie et dhis-
toire 10 (1890), pp. 12649.
Duczko, W., Scandinavians in the Southern Baltic between the 5th and the 10th
centuries A.D., in: P. Urbanczyk ed., Origins of Central Europe (Warsaw, 1997),
pp. 191211.
Duine, F., Mmento des sources hagiographiques de lHistorie de Bretagne (Rennes, 1918).
 565

Dunn, A., The transition from polis to kastron in the Balkans (3rd8th/9th cen-
tury): general and regional perspectives, Byzantine and Modern Greek Studies 18
(1994), pp. 6080.
Durkheim, E., Elementary forms of the religious life (London, 1915).
Duval, N., Larchitecture cultuelle, in: Naissance des arts chrtiens: Atlas des monuments
chrtiens de la France (Paris, 1991), pp. 186219.
Duval, Y., Auprs des saints corps et me: Linhumation ad sanctos dans la chrtient dOrient
et dOccident du III e au VII e sicle (Paris, 1988).
Eaton, T., A neglected landscape? A study of the value of Antiquity to the inhabitants of
Merovingian Gaul, with particular reference to the Roman rural landscape (unpublished M.A.
thesis, University of Reading, 1995).
Ebling, H., Prosopographie der Amtstrger des Merowingerreiches von Chlothar II (613) bis
Karl Martell (741). Beihefte der Francia 2 (Sigmaringen, 1974).
Effros, B., From grave goods to Christian epitaphs: Evolution in burial tradition and the expres-
sion of social status in Merovingian society. Unpublished doctoral dissertation (Los
Angeles, 1994).
, Beyond cemetery walls: early medieval funerary topography and Christian
salvation, EME 6, 1 (1997), pp. 123.
Eggebrecht, A. ed., Die Mongolen und ihr Weltreich (Mainz, s.a.).
Eggers, H.J., Lbsow, ein germanischer Frstensitz der lteren Kaiserzeit, Praehistorische
Zeitschrift 34/35, 2 (1949/50), pp. 58111.
Eibl, E.-M., Zur Stellung Bayern und Rheinfranken im Reiche Arnulfs von Krnten,
Jahrbuch fr Geschichte des Feudalisimus 8 (1984), pp. 73112.
Ellmers, D., Zur Ikonographie nordischer Goldbrakteaten, Jahrbuch des Rmisch-
Germanischen Zentralmuseums Mainz 17 (1970), pp. 20184.
Ellis Davidson, H., The lost beliefs of northern Europe (London, 1993).
Eliade, M., Schamanismus und archaische Ekstasetechnik (Frankfurt/Main, 1975; orig.
Paris, 1951).
, The forge and the crucible (second edition Chicago, 1978).
, A history of religious ideas, 3 vols. (Chicago, 19781985).
, The sacred and the profane. The nature of religion (New York,London, 1987).
, Patterns in comparative religion (London, 1997).
Elias, N., ber den Prozess der Zivilisation, 2 vols. (Bern/Mnchen, 1969).
, ber die Zeit. Arbeiten zur Wissenssoziologie 2 (Frankfurt a.M., 1984).
Enright, M., Lady with a mead cup. Ritual, prophecy and lordship in the European warband
from La Tne to the Viking Age (Dublin/Portland, 1996).
Ermini Pani, L., Renovatio murorum tra programma urbanistica e restauro conser-
vativo: Rome e il ducato Romano, in: Committenti e produzione artistico-letteraria nel-
lalto medioevo occidentale. Settimane 39 (Spoleto, 1992), pp. 485530.
, LeopoliCencelle: Una citt di fondazione papale, in: G. de Boe and
F. Verhaege eds., Urbanism in Medieval Europe 1 (Zellik, 1997), pp. 36973.
Es, W.A. van, De Romeinen in Nederland (Haarlem, 1981).
Escribano, M.V., Iglesia y Estado en el certamen priscilianista. Causa ecclesiae e iudicium
publicum (Zaragoza, 1988).
, Igrexa e Herexia en Gallaecia: O Priscilianismo, in: G. Pereira-Menaut
ed., Galicia fai dous mil anos. O feito diferencial galego I. Historia 1 (Santiago de
Compostela, 1997), pp. 278321.
Esprandieu, ., Recueil gnral des bas-reliefs de la Gaule romaine 11 (Paris, 1938).
Everett, N., Scribes and charters in Lombard Italy, Studi Medievali 3e ser. 41
(2000), pp. 3983.
Ewig, E., Rsidence et capital pendant le haut Moyen Age, Revue historique 130
(1963), pp. 2572; repr. in: idem, Sptantikes und frnkisches Gallien. Gesammelte Schriften,
vol. I, Beihefte der Francia 3, 2 vols., ed. H. Atsma (Mnchen, 1976), pp. 362408.
, Kathedralpatrozinien im rmischen und im frnkischen Gallien in: idem,
566

Sptantikes und frnkisches Gallien, vol. II, Beihefte der Francia 3, 2 vols., ed. H. Atsma
(Mnchen, 1979), pp. 260317.
, Frhes Mittelalter. Rheinische Geschichte in drei Bnden (Dsseldorf, 1980).
Fabech, C., Kult og samfund i yngre jernalderRavlunds som eksempel, in:
L. Larsson and B. Hrdh eds., Centrala Platser, Centrala Frgor (Lund, 1998), pp.
14764.
, Organising the Landscape. A matter of production, power, and religion,
in: T. Dickinson and D.Griffiths eds., The Making of Kingdoms. Anglo-Saxon Studies
in Archaeology and History 10 (Oxford, 1999), pp. 3747.
Faider-Feytmans, G., Objets dpoque romaine dcouverts dans des tombes mrovingi-
ennes du bassin de La Haine (Belgique), in: R. Chevallier ed., Mlanges dArchologie
et dHistoire offerts Andr Piganiol 2 (Paris, 1966), pp. 10118.
Falck, L., Mainz im frhen und hohen Mittelalter (Mitte 5. Jht. bis 1244) (Dsseldorf,
1972).
Falkenstein, L., Charlemagne et Aix-la-Chapelle, Byzantion 61 (1991), pp. 23189.
, Otto III und Aachen. MGH Studien und Texte 22 (Hannover, 1998).
Fehring, G.P., Einfhrung in die Archologie des Mittelalters (Darmstadt, 1987).
Fernndez Alonso, J., La cura pastoral en la Espaa romanovisigoda (Rome, 1955).
Ferreiro, A., The missionary labors of St. Martin of Bracara in sixth-century
Galicia, Studia Monastica 23 (1981), pp. 1126.
Fevrier, P.-A., La marque de lAntiquit tardive dans le paysage religieux mdi-
val de la Provence rural, in: M. Fixot and E. Zadora-Rio eds., Lenvironnement
des glises et la topographie religieuse des campagnes mdivales. Actes du III e congres inter-
national darchologie mdival (Aix en Provence 2830 septembre 1989). Documents
darchologie franaise 46 (Paris, 1994), pp. 2735.
Fichtenau, H., Politische Datierungen des frhen Mittelalters, in: H. Wolfram
ed., Intitulatio II: Lateinische Herrscher- und Frstentitel im neunten und zehnten Jahrhundert.
Mitteilungen des Instituts fr sterreichische Geschichtsforschung, Ergnzungsband
24 (Wien, 1973), pp. 453540.
Field, D., Round barrows and the harmonious landscape: placing Early Bronze
Age burial monuments in South-East England, Oxford Journal of Archaeology 17, 3
(1998), pp. 30926.
Flandrin, J.L., Orgenes de la familia moderna (Barcelona, 1979).
Fleckenstein, J., Einhard, seine Grndung und sein Vermchtnis im Seligenstadt,
in: K. Hauck ed., Das Einhardskreuz: Vortrge und Studien der Munsteraner Diskussion
zur arcus Einhardi (Gttingen, 1974), pp. 96121.
, Die Struktur des Hofes Karls des Grossen im Spiegel von Hinkmars De
ordine palatii , Zeitschrift des Aachener Geschichtsvereins 83 (1976), pp. 522.
Fletcher, R., The quest for El Cid (Oxford, 1989).
Fleuriot, L., Les origines de la Bretagne (Paris, 1980).
Flint, V.I.J., The rise of magic in early medieval Europe (Princeton, 1991).
Foss, C., Archaeology and the twenty cities of Byzantine Asia, American Journal
of Archaeology 81 (1977), pp. 46986.
Foucault, M., Surveiller et punir. La naissance de la prison (Paris, 1975).
, History of sexuality I (London, 1979).
, Power/knowledge: Selected interviews and other writings 19721977 (London, 1980).
Fouracre, P., Merovingian history and Merovingian hagiography, Past and Present
127 (1990), pp. 338.
, and R. Gerberding, Late Merovingian France. History and hagiography 640 720
(Manchester, 1996).
, Attitudes towards violence in seventh- and eighth-century Francia, in:
G. Halsall ed., Violence and society in the early medieval West (Woodbridge, 1998), pp.
6075.
, The origins of the Carolingian attempt to regulate the cult of the saints,
in: J. Howard-Johnston and P.A. Hayward eds., The cult of the saints in late Antiquity
 567

and the early Middle Ages. Essays on the contribution of Peter Brown (Oxford, 1999), pp.
14365.
Fournier, P.-F., Clermont-Ferrand au VIe sicle, recherches sur la topographie de
la ville, Bibliothque de lcole des Chartes 128 (1970), pp. 273344.
Francis, D., L. Kellaher and C. Lee, Talking to people in cemeteries, Journal of
the Institute of Burial and Cremation Administration 65, 1 (1997), pp. 1425.
Francovich, R. and G. Noy eds., La storia dellalto medioevo italiano (VIX secolo) alla
luce dellarcheologia. Biblioteca dellarcheologia medievale 11 (Florence, 1994).
Frank, R., Germanic legend in Old English literature, in: M. Godden and
M. Lapidge eds., The Cambridge Companion to Old English literature (Cambridge, 1991),
pp. 88106.
Frankenstein, S. and M.J. Rowlands, The internal structure and regional context
of early Iron Age society in south-western Germany, Bulletin of the Institute of
Archaeology 15 (1978), pp. 73112.
Fremersdorf, F., Das frnkische Reihengrberfeld Kln-Mngersdorf 1. Germanische Denkmler
der Vlkerwanderungszeit 6 (Berlin, 1955).
Frighetto, C., Panorama econmico-social del NO. de la Pennsula Ibrica en poca visigoda.
La obra de Valerius of Bierzo (Doctoral thesis, Salamanca, 1996), forthcoming.
Frolow, A., La ddicace de Constantinople dans la tradition Byzantine, Revue de
lhistoire des religions 127 (1944), pp. 61127.
Fumagalli, V., Landscapes of fear: Perceptions of nature and the city in the Middle Ages
(Cambridge, 1994).
Gaiffier, B. de, Les notices hispaniques dans le martyrologe dUsuard, Analecta
Bollandiana 15 (1937), pp. 2712.
, Linventio et Translatio de S. Zoilo de Cordoue, Analecta Bollandiana 16
(1938).
Gaillard, M., Les fondations dabbayes fminines dans le nord et lest de la Gaule,
de la fin du VIe la fin du Xe sicle, Revue dHistoire de lEglise de France LXXVI
(1990), pp. 520.
Galini, H., Touis de Grgoire. Tours des archives du sol, in: N. Gauthier and
H. Galini eds., Grgoire de Tours et lespace gaulois (Tours, 1997), pp. 6580.
, Tours from an archaeological standpoint, in: C.E. Karkow, K.M. Wickham-
Crowley and B. Young eds., Spaces of the living and the dead: An archaeological dia-
logue (Oxford, 1999), pp. 87105.
Ganshof, F.-L., Note sur les Capitula tractanda de causis cum episcopis et abbatibus trac-
tandis de 811, Studi Gratiani 13 (1967), pp. 125.
, Charlemagnes programme of imperial government, in: idem, The Carolingians
and the Frankish monarchy, transl. J. Sondheimer (London, 1971), pp. 5585.
Gansum, T., Mythos, Logos, Ritus. Symbolisme og gravskikk i lys av gudediktene
i den eldre Edda, in: I. Fuglestvedt, T. Gansum and A. Opedal eds., Et Hus
med Mange Rom. Vennebok til Bjorn Myhre p 60-rsdagen. AmS-Rapport 11B (Stavanger,
1999), pp. 439504.
Ganz, D., Humour as history in Notkers Gesta Karoli Magni , in: E.B. King, J.T.
Schaefer and W.B. Wadley eds, Monks, nuns, and friars in medieval society (Sewanee
TE, 1989), pp. 17183.
Garca de Cortazar, J.A., Les communauts villageoises du Nord de la Pninsule
Ibrique au Moyen Age, Flaran 4 (1982), pp. 5577.
Garca de Valdeavellano, L., La comunidad patrimonial de la familia en el dere-
cho espaol medieval, in: idem, Estudios medievales de Derecho Privado (Sevilla, 1977),
pp. 295321.
Garca Fernndez, J., Sobre los orgenes del paisaje agrario gallego, Estudios
geogrficos 129 (1972), pp. 75363.
Garca Gomez, E., Notas sobre la topografa cordobesa en los Anales de Al-
Hakam II por Isa R z , Al-Andalus 30 (1965), pp. 31979.
, Anales Palatinos del Califa de Cordoba al-Hakam II (Madrid, 1967).
568

Garca Moreno, L.A., El fin del reino visigodo de Toledo. Decadencia y catastrofe. Una con-
tribucin a su crtica (Madrid, 1975).
Garrison, M., The emergence of Latin literary culture and the court of Charlemagne,
in: R. McKitterick ed., Carolingian culture: Emulation and innovation (Cambridge, 1993),
pp. 11140.
, Alcuins world through his letters and verse. Unpublished Ph.D. thesis (Cambridge 1996).
, The social world of Alcuin: Nicknames and the Carolingian court, in: Alcuin
of York and his influence on European culture. Germania Latina III (Groningen, 1999),
pp. 5779.
, The Franks as the New Israel? Education for an identity from Pippin to
Charlemagne, in: Y. Hen and M. Innes eds., The uses of the past in the early Middle
Ages (Cambridge, 2000), pp. 11461.
Gasnault, P., Le Tombeau de Saint Martin et les invasions normandes dans lhis-
toire et dans la lgende in: Mmorial de lanne martinienne. Seizime centenaire de lab-
baye de Ligug, centenaire de la dcouverte du tombeau de saint Martin Tours (Paris, 1962),
pp. 5166.
Gaudemet, J., LEglise dans lEmpire Romain (IVV sicles) (Paris, 1958).
, Les communauts familiales (Paris, 1963).
Gauthier, N. and J-Ch. Picard eds., Topographie Chrtienne des cits de la Gaule des ori-
gines au milieu du VIII e sicle 3, Provinces ecclesiastiques de Vienne et dArles (Paris, 1986),
pp. 1735.
Geake, H., The use of grave-goods in conversion-period England c. 600c. 850. BAR 261
(Oxford, 1997).
Geary, P.J., Furta sacra: Thefts of relics in the central Middle Ages (2nd ed., Princeton
1990).
, Phantoms of remembrance: Memory and oblivion at the end of the first millennium
(Princeton, 1994).
, Living with the dead in the Middle Ages (Ithaca and London, 1994).
, Land, language and memory in Europe, 7001100, Transactions of the Royal
Historical Society 9 (1999), pp. 16984.
Gebhr, M., Zur Definition lterkaiserzeitlicher Frstengrber vom Lbsow-Typ,
Praehistorische Zeitschrift 49 (1974), pp. 82128.
Geertman, H., More Veterum Il, Liber Pontificalis e gli edifici ecclesiastici di Roma nella
tarda antichit e nellalto medioevo. Archaeologia Traiectina 10 (Groningen, 1975).
Geertz, C., The Interpretation of cultures (New York, 1973).
, Art as a cultural system, in: idem: Local knowledge: Further essays in interpre-
tive anthropology (New York, 1983).
Genicot, J., La gense et les premiers sicles des villes medivales dans les Pays-Bas meridionaux.
Un problme archologique et historique (Brussels, 1990).
Gennep, A. van, Rites of passage (London, 1960) [orig. 1909].
George, W., The Church of Saint Eirene at Constantinople (Oxford, 1912).
Gerberding, R., The rise of the Carolingians and the Liber Historiae Francorum (Oxford,
1987).
, 716: A crucial year for Charles Martel, in: J. Jarnut, U. Nonn and
M. Richter eds., Karl Martell in seiner Zeit (Sigmaringen, 1994), pp. 20516.
Gibson, S. and B. Ward-Perkins, The surviving remains of the Leonine wall,
PBSR 47 (1979), pp. 3057.
, The surviving remains of the Leonine wall. Part II: The Passetto, PBSR
51 (1983), pp. 22239.
Gieauf, J., Ekstatische Seelenreisende. Die nordeurasischen Schamanen, Mitteilungen
der Grazer Morgenlndischen Gesellschaft 6 (1997), pp. 3962.
Glick, T.F., Islamic and Christian Spain in the early Middle Ages: Comparative perspectives
on social and cultural formation (Princeton, 1979).
Gockel, M., Karolingische Knigshfe am Mittelrhein. Verffentlichungen der Max-Planck
Instituts fr Geschichte 31 (Gttingen, 1970).
 569

, Die Bedeutung Treburs als Pfalzort, Deutsche Knigspfalzen III. Verffent-

lichungen der Max-Planck Instituts fr Geschichte 11/iii (Gttingen, 1980).
Godelier, M., Lnigme du don (Paris, 1996).
, The enigma of the gift (Cambridge, 1999; orig. Paris, 1996).
Godman, P., Poetry of the Carolingian Renaissance (Oxford, 1985).
Goetz, H.W., Frauen im Frhmittelalter. Frauenbild und Frauenleben im Frankenreich (Weimar/
Kln/Wien, 1995).
Goffart, W., The narrators of barbarian history (A.D. 550800): Jordanes, Gregory of Tours,
Bede, and Paul the Deacon (Princeton, 1988).
, The Fredegar problem reconsidered, in: idem, Romes fall and after (London,
1989).
Goldsmith, M.E., The mode and meaning of Beowulf (London, 1970).
Goldthwaite, R., The empire of things: consumer demand in Renaissance Italy,
in: F.W. Kent and P. Simons eds., Patronage, art, and society in Renaissance Italy
(Oxford, 1987), pp. 15375.
Gonzlez Echegaray, G., El monacato en la Espaa nrdica en su confrontacin
con el paganismo (ss. VIVII), in: idem, Semana de Historia del monacato cntabro-
astur-leons (Oviedo, 1982), pp. 3556.
Goody, J., The development of family and marriage in Europe (Cambridge, 1983).
Goosen, A.B.J.M., Achtergronden van Priscillianus christelijke ascese, 2 vols. (Nijmegen,
1976).
, Algunas observaciones sobre la neumatologa de Priscilianus, in: Primera
reunin gallega de estudios clsicos (Santiago de Compostela, 1981), pp. 23742.
Grich, K., Otto III. ffnet das Karlsgrab in Aachen. berlegungen zu Heiliger-
verehrung, Heiligsprechung und Traditionsbildung, in: G. Althoff and E. Schubert
eds., Herrschaftsreprsentation im Ottonischen Sachsen (Sigmaringen, 1998), pp. 381430.
Gorman, F., The ideology of ritual: Space, time and status in the priestly theology. Journal
for the Study of the Old Testament, Supplement Series 91 (Sheffield, 1990).
Gosden, C., Social Being and Time (Oxford, 1994).
Gourvil, F., Les loc- dans lhagio-toponymie bretonne, Revue Internationale dOnomastique
15 (1963), pp. 4752.
Gow, P., Land, people and paper in Central Amazonia, in: E. Hirsch and
M. OHanlon eds., The Anthropology of landscape: Perspectives on place and space (Oxford,
1995), pp. 4362.
Graus, F., Die Gewalt bei den Anfngen des Feudalismus und die Gefangenen-
befreiung der merowingischen Hagiographie, Jahrbuch fr Wirtschaftsgeschichte 1
(1961), pp. 61156.
Gray, N., The paleography of Latin inscriptions in the eighth, ninth, and tenth
centuries in Italy, PBSR 16 (1948), pp. 1189.
Green, D.H., Linguistic evidence for the early migrations of the Goths, in:
P. Heather ed., The Visigoths from the Migration Period to the seventh sentury (Woodbridge,
1999), pp. 1132.
Gregory, C., Gifts and commodities (London, 1982).
Grieg, S., Smedverkt_y i Norske Gravfunn. Oldtiden 9 (Oslo, 1922).
Grierson, P., The coronation of Charlemagne and the coinage of Leo III, Revue
Belge de Philologie et dHistoire 30 (1952), pp. 82533.
, Commerce in the Dark Ages: a critique of the evidence, Transactions of the
Royal Historical Society 9 (1959), pp. 12340.
, and M. Blackburn, Medieval European coinage 1 (Cambridge, 1986).
Grnbech, W., Kultur und Religion der Germanen (Stuttgart, 1954).
Grousset, R., Die Reise nach Westen, oder wie Hsan-Tsang den Buddhismus nach China
holte (Kln, 1986).
Guerout, J., art. Fare, DHGE 16 (Paris, 1967). col. 515.
, art. Faremoutiers, DHGE 16, (Paris, 1967), cols. 53437.
, art. Faron, DHGE 16 (Paris, 1967), col. 654.
570

, Les origines et le premier sicle de labbaye, in: Labbaye Notre-Dame de

Jouarre (Paris, 1961), pp. 167.
Guerreiro, R., Le rayonnement de lhagiographie hispanique en Gaule pendant le
haut Moyen ge: circulation et diffusion des Passions hispaniques, in: J. Fontaine
and C. Pellistrandi eds., LEurope Hritire de lEspagne Wisigothique (Madrid, 1992).
Guillotel, H., Les cartulaires de labbaye de Redon, Mmoires de la Socit dHistoire
et dArchologie de Bretagne 63 (1986), pp. 2748.
, cartulaires bretons mdivaux, in: O. Guyotjeannin, L. Morelle and M. Parisse
eds., Les cartulaires. Actes de la table ronde organise par lEcole National des Chartes.
Mmoires et documents de lEcole des Chartes 39 (Paris, 1993), pp. 32541.
, Le manuscrit, in: Cartulaire de labbaye Saint-Sauveur de Redon [facsimile edi-
tion] (Rennes, 1998).
Gurevic, A., Reprsentations et attitudes lgard de la proprit pendant le haut
moyen ge, Annales ESC 27 (1972), pp. 52347.
Habermas, J., The structural transformation of the public sphere, transl. T. Berger (London,
1989).
Hager, H., Rckgewonnene Marienikonen des frhen Mittelalters in Rom, Rmische
Quartalschrift fr Altertumskunde und Kirchengeschichte 61 (1966), pp. 20916.
Halbwachs, M., On collective memory, ed. and transl. L.A. Coser (Chicago, 1992).
Haldon, J.F., Byzantium in the seventh century: the transformation of a culture (rev. ed.
Cambridge, 1997).
, The idea of the town in the Byzantine empire, in: G. Brogiolo and B. Ward-
Perkins eds., The idea and ideal of the town between late Antiquity and the early Middle
Ages, TRW 4 (Leiden/Boston/Kln, 1999), pp. 123.
Halperin, Ch. M., Russia and the Golden Horde (Bloomington, 1987).
Halphen, L., Le De ordine palatii dHincmar, Revue historique 183 (1938), pp. 19.
Halsall, G., Settlement and social organization: The Merovingian region of Metz (Cambridge,
1995).
, Early medieval cemeteries, New light on the Dark Ages 1 (Glasgow, 1995).
, Towns, societies and ideas: the not-so-strange case of late Roman and early
Merovingian Metz, in: N. Christie and S.T. Loseby eds., Towns in Transition.
Urban evolution in late Antiquity and the early Middle Ages (Aldershot, 1996), pp. 23561.
, Female status and power in early Merovingian Central Austrasia: The bur-
ial evidence, EME 5 (1996), pp. 124.
, ed., Violence and society in the early medieval West (London, 1997).
Hannig, J., Zentrale Kontrolle und regionale Machtbalance. Beobachtungen zum
System der karolingischen Knigsboten am Beispiel des Mittelrheingebietes, Archiv
fr Kulturgeschichte 66 (1984), pp. 146.
Hardenberg, H., De Maastrichtse Vroenhof , Miscellanea Trajectensia. Bijdragen to de
geschiedenis van Maastricht uitgegeven bij gelegenheid van het 300 jarig bestaan van de stads-
bibliotheek van Maastricht 1662 31 juli 1962 (Maastricht, 1962), pp. 2953.
Hardt, M., Royal treasures and representation in the early Middle Ages, in:
W. Pohl and H. Reimitz eds., Strategies of distinction: The construction of ethnic com-
munities, 300800, TRW 2 (Leiden/Boston/Kln, 1998), pp. 25580.
Harries, J., Death and the Dead in the Late Roman Empire, in: S. Bassett ed.,
Death in Towns: Urban Responses to the Dying and the Dead, 1001600 (London, 1992),
pp. 5667.
Hartingsveldt, E. van, De Zeven Zaken en de frankische bisschop, Revue Belge
de Philologie et dHistoire 67 (1989), pp. 67794.
Hastrup, K., Iceland: sorcerers and paganism, in: B. Ankarloo and G. Henningsen
eds., Early Modern witchcraft. Centres and peripheries (Oxford, 1990), pp. 383401.
Hauck, K., Tiergrten im Pfalzbereich, in: Deutsche Knigspfalzen I. Verffentlichungen
der Max-Planck Instituts fr Geschichte 11/i (Gttingen, 1963), pp. 3074.
, Gtterglaube im Spiegel der goldenen Brakteaten, in: C. Ahrens ed., Sachsen
 571

und Angelsachsen. Verffentlichungen des Helms-Museums 32. (Hamburg, 1978),

pp. 185218.
, Gudme in der Sicht der Brakteatenforschung, Frhmittelalterliche Studien 21
(1987), pp. 14781.
, Gudme as Kultort und seine Rolle beim Austausch von Bildformularen der
Goldbrakteaten, in: P.O. Nielsen e.a. eds., The Archaeology of Gudme and Lundeborg
(Copenhagen, 1994), pp. 7888.
Hrke, H., Warrior graves? The background of the Anglo-Saxon weapon burial
rite, Past and Present 126 (1990), pp. 2243.
, Angelschsische Waffengrber des 5. bis 7. Jahrhunderts. Zeitschrift fr Archologie
des Mittelalters Beiheft 6 (Kln/Bonn, 1992).
, Lowbury Hill: A context for the Saxon barrow, Archaeological Journal 151
(1994), pp. 2026.
, Zur Bedeutung der Verwandtschaftsanalyse aus archologischer Sicht,
Germania 73 (1995), pp. 30712.
, Material culture as myth: weapons in Anglo-Saxon graves, in: C.K. Jensen
and K.H. Nielsen eds., Burial and society: The chronological and social analysis of archae-
ological burial data (Aarhus, Oxford/Oakville, Connecticut, 1997), pp. 11927.
Harke, H. and Williams, H., Angelschsische Bestattungspltze und ltere Denkmler:
Bemerkungen zur zeitlichen Entwicklung und Deutung des Phnomens, Archologische
Informationen 20 (1997), pp. 257.
Hrke, H., The circulation of weapons in Anglo-Saxon societies, in: F. Theuws
and J.L. Nelson eds., Rituals of Power. From late Antiquity to the early Middle Ages,
TRW 8 (Leiden/Boston/Kln, 2000), pp. 37799.
Hrke, H. and A. Belinskij, Nouvelles fouilles de 19941996 dans la ncropole de
Klin Yar, in: M. Kazanski and V. Soupault eds., Les sites archologiques en Crime
et au laucase durant lAntiquit tardive et le slaut Mogen Age. Colloquia Pontica 5 (Leiden,
2000), pp. 193210.
Harman, M., T.J. Molleson and J.L. Price, Burials, bodies and beheadings in
Romano-British and Anglo-Saxon cemeteries, Bulletin of the British Museum, Nat.Hist.
(Geology) 35, 3 (1981), pp. 14588.
Harries, J., Sidonius Apollinaris and the fall of Rome, A.D. 407485 (Oxford, 1994).
Hartmann, J., De reconstructie van een middeleeuws landschap. Nederzettingsgeschiedenis en
instellingen van de heerlijkheden Eijsden en Breust bij Maastricht (10e-19e eeuw) (Assen/
Maastricht, 1986).
Heather, P., The Goths (Cambridge, 1996).
Hedeager, L., Iron-Age societies (Oxford, 1992).
, Odins offer. Skygger af en shamanistisk tradition i nordisk folkevandringstid,
Tor 29 (Uppsala, 1997), pp. 26578.
, Cosmological endurance: pagan identities in early Christian Europe, European
Journal of Archaeology 1, 3 (1998), pp. 38296.
, Myth and art: a passport to political authority in Scandinavia during the
Migration Period, in: T. Dickinson and D. Griffiths eds., The Making of Kingdoms.
Anglo-Saxon Studies in Archaeology and History 10 (Oxford, 1999), pp. 15156.
, Sacred topography. Depositions of wealth in the cultural landscape, in:
A. Gustafsson and H. Karlsson eds., Glyfer och arkeologiska rumen vnbok til Jarl
Nordbladh (Gteborg, 1999) pp. 22952.
, Skandinavisk dyreornamentik. Symbolsk reprsentation af en for-kristen kos-
mologi, in: I. Fuglestvedt, T. Gansum and A. Opedal eds., Et Hus med Mange
Rom. Vennebok til Bjorn Myhre p 60rsdagen. AmS-Rapport 11A (Stavanger, 1999),
pp. 21938.
, Europe in the Migration Period. The formation of a political mentality,
in: F. Theuws and J.L. Nelson eds., Rituals of power. From late Antiquity to the early
Middle Ages, TRW 8 (Leiden/Boston/Kln, 2000), pp. 1557.
572

, Scandinavia (c. 500700 A.D.), in: P. Fouracre ed., New Cambridge Medieval
History I (Cambridge, forthcoming).
Heckscher, W.S., Relics of pagan Antiquity in mediaeval settings, Journal of the
Warburg Institute 1 (193738), pp. 20420.
Heinzelmann, M., Bischofsherrschaft in Gallien. Zur Kontinuitt rmischer Fhrungsschichten
vom 4. bis zum 7. Jahrhundert. Soziale, prosopographische und bildungsgeschichtliche Aspekte
(Mnchen, 1976).
, Translationsberichte und andere Quellen des Reliquienkultes. Typologie des sources du
Moyen ge Occidental 33 (Turnhout, 1979).
, and Poulin, J.-C., Les Vies Anciennes de sainte Genevive de Paris. tudes critiques
(Paris, 1986).
Heissig, W. ed., Dschingis-Khanein Weltreich zu Pferde. Das Buch vom Ursprung der
Mongolen (Kln, 1985).
, Schamanen und Geisterbeschwrer in der stlichen Mongolei (Wiesbaden, 1992).
Helms, M., Ulysses sail: An ethnographic Odyssey of power, knowledge and geographical dis-
tance (Princeton, 1988).
, Craft and the kingly ideal. Art, trade and power (Austin, Texas, 1993).
Helvtius, A-M., Abbayes, vques et laques. Une politique du pouvoir en Hainaut au Moyen
Age (VII eXI e sicle) (Bruxelles, 1994).
, Virgo et virago: rflexions sur le pouvoir du voile consacr daprs les sources
hagiographiques de la Gaule du Nord, in: S. Lebecq, A. Dierkens, R. Le Jan,
J-M. Sansterre eds., Femmes et pouvoirs des femmes Byzance et en Occident (VI eXI e
sicle) (Lille, 1999), pp. 189204.
Hen, Y., Culture and religion in Merovingian Gaul, A.D. 481751 (Leiden/New York/Kln,
1995).
, and M. Innes eds., The uses of the past in early medieval Europe (Cambridge, 2000).
, The Annals of Metz and the Merovingian past, in: Y. Hen and M. Innes
eds., The uses of the past in the early Middle Ages (Cambridge, 2000), pp. 17590.
Henderson, G., Early medieval. Medieval Academy Reprints for Teaching 29 (Toronto,
1993).
Hennebicque [Le Jan], R., Espaces sauvages et chasses royales dans le Nord de
la France, Revue du Nord 62 (1980), pp. 3557.
Herbers, K., Leo IV. und das Papsttum in der Mitte des 9. Jahrhunderts (Stuttgart, 1996).
Herbert, E., Red Gold of Africa (Madison, 1984).
Herbert, E.W., Iron, gender and power (Indiana, 1993).
Herklotz, I., Der Campus Lateranensis im Mittelalter, Rmisches Jahrbuch fr Kunstgeschichte
22 (1985), pp. 142.
Herschend, F., Hus p Helg, Fornvnnen 90 (Stockholm, 1995), pp. 2218.
, Livet i Hallen (Uppsala, 1997).
, Striden i Finnsborg, Tor 29 (1997), pp. 295333.
, The idea of the good in late Iron Age society (Uppsala, 1998).
, Halle, Reallexikon der Germanischen Altertumskunde 13 (Berlin, 1999), pp. 41425.
Herwegen, I., Das Pactum des Hl. Fruktuosus von Braga: ein Beitrag zur Geschichte des sue-
visch-westgotischen Mnchtums und seines Rechtes (Stutttgart, 1907).
He, W., Geldwirtschaft am Mittelrhein in karolingischer Zeit, Bltter fr deutsche
Landesgeschichte 98 (1962), pp. 2663.
Hillgarth, J.H., Visigothic Spain and Early Christian Ireland, Proceeding of the Royal
Irish Academy 52, C, 6 (1962), pp. 16794.
, Popular religion in Visigothic Spain, in: E. James ed., Visigothic Spain: New
approaches (Oxford, 1980), pp. 360.
Hinojosa y Naveros, E., La comunidad domstica en Espaa durante la Edad
Media, Obras 2 (Madrid, 1955), pp. 33145.
Hirsch, E. and M. OHanlon eds., The anthropology of landscape: Perspectives on place and
space (Oxford, 1995).
 573

Hlawitschka, E., Lothringen und das Reich an der Schwelle der deutschen Geschichte, Schriften
der MGH 21 (Stuttgart, 1968).
Hobsbawm, E., Introduction: inventing traditions, in: idem and T. Ranger eds.,
The invention of tradition (Cambridge, 1983), pp. 114.
Hodgen, M.T., Early anthropology in the sixteenth and seventeenth centuries (Philadelphia,
1964).
Hodges, R., The Anglo-Saxon achievement (London, 1989).
, A fetishism for commodities: Ninth-century glass-making at San Vincenzo
al Volturno, in: M. Mendera ed., Archeologia e storia della produzione del vetro prein-
dustriale (Siena, 1991), pp. 6790.
Hoffsummer-Bosson, A., Chvremont: lapport des sources archologiques, Bulletin
de lInstitut Archologique Ligois 100 (1988), pp. 7187.
Holtsmark, A., Studier i Snorres Mytologi (Oslo, 1964).
, Fornnordisk Mytologi (Lund, 1992).
Hultgrd, A., Fornskandinavisk hinsidestro i Snorre Sturlusons spegling, in: U. Dro-
bin ed., Religion och Samhlle i det Frkristna Norden. Et symposium. (Odense, 1999),
pp. 10924.
Hunger, H., Christliches und Nichtchristliches im byzantinischen Eherecht, ster-
reichisches Archiv fr Kirchenrecht 18 (1967), pp. 30525.
Hunt, E.B., Constantine and Jerusalem, Journal of Ecclesiastical History 48 (1997),
pp. 40524.
Hunt, E.D., Holy Land pilgrimage in the later Roman Empire A.D. 312460 (Oxford, 1982).
Hunter, M., Germanic and Roman Antiquity and the sense of the past in Anglo-
Saxon England, Anglo-Saxon England 3 (1974), pp. 2950.
Huntingdon, R. and P. Metcalf, Celebrations of death: The anthropology of mortuary ritual
(Cambridge, 1979).
Lason, V., Topography and world view in Njls saga, in: S. Hansson and
M. Malm eds., Gudar p jorden. Festskrift till Lars Lnnroth. (Stockholm, 2000),
pp. 13141.
Innes, M., Kings, monks and patrons: Political identity at the Abbey of Lorsch,
in: R. Le Jan ed., La Royaut et les lites dans lEurope Carolingienne (Lille, 1998),
pp. 30124.
, Memory, orality and literacy in an early medieval society, Past and Present
158 (1998), pp. 336.
, State and society in the early Middle Ages: The middle Rhine valley 400 1000
(Cambridge, 2000).
, What was Carolingian government?, in: J. Storey ed., Charlemagne (forth-
coming).
Iogna-Prat, D., . Palazzo and D. Russo eds., Marie: Le culte de la vierge dans la socit
mdivale (Paris, 1996).
Ivison, E., Burial and urbanism at late antique and early Byzantine Corinth
(c. A.D. 400700), in: N. Christie and S. Loseby eds., Towns in transition: Urban
evolution in late Antiquity and the early Middle Ages (Aldershot, 1996), pp. 99125.
Jackson, P. and D. Morgan, eds. and transl., The mission of Friar William of Rubruck.
His journey to the Great Khan Mngke, 125355 (London, 1990).
Jacobsen, W., Was is die karolingische Renaissance in der Baukunst?, Zeitschrift
fr Kunstgeschichte 3 (1988), pp. 31347.
, Die Pfalzkonzeptionen Karls des Grossen, in: L.E. Saurma-Jeltsch ed., Karl
der Grosse als vielberufener Vorfahr (Sigmaringen, 1994), pp. 2348.
Jagodzi ski, M., Archeologiczne lady osadnictwa mi\dzy Wis [ a Pas \k[. Katalog stanowisk
(Warsaw, 1997).
, Archeologiczne lady osadnictwa mi\dzy Wis [ a Pas \k[ we wczesnym red-
niowieczu. Komentarz do katalogu stanowisk in: P. Urba czyk ed., Adalbertus,
vol. 1 (Warsaw, 1998), pp. 15997.
574

, Rejon Wis y we wczesnym redniowieczu. Struktura zasiedlenia i stosunki

etniczne in: M. Jagodzi ski ed., Poggranizce polsko-purskie w czasach w. Wojciechna
(Elbl[g, 1999), pp. 3580.
and M. Kasprzycka, The early medieval craft and commercial centre at
Janw Pomorski near Elbl[g on the South Baltic Coast, Antiquity 65, no. 248
(1991), pp. 696715.
James, E., Archaeology and the Merovingian monastery, in: H.B. Clarke and
M. Brennan, Columbanus and Merovingian monasticism. BAR International Series 113
(Oxford, 1981), pp. 3355.
, Royal burials among the Franks, in: M. Carver ed., The age of Sutton Hoo
(Woodbridge, 1992), pp. 24354.
Janes, D., God and gold in late Antiquity (Cambridge, 1998).
Janin, R., La gographie ecclsiastique de lempire byzantin I, Le sige de Constantinople et le
Patriarchat oecumenique 3, Les glises et les monastres (2nd ed., Paris, 1969).
Jarnut, J., Die frhmittelalterliche Jagd unter rechts- und sozialgeschichtlichen
Aspekten, Luomo di fronte al mondo animale nellalto medioevo, Settimane 31 (Spoleto,
1985), pp. 765808.
Jschke, K.-U., Burgenbau and Landesverteidigung um 900. Vrtrge und Forschungen
16 (Sigmaringen, 1975).
Johanek, P., Zur rechtlichen Funktion von Traditionsnotiz, Traditionsbuch und
frher Siegelurkunde, in: P. Classen ed., Recht und Schrift, pp. 13162.
Johnson, P.D., Equal in monastic profession. Religious women in medieval France (Chicago,
1993).
Jong, M. de, Growing up in a Carolingian monastery: Magister Hildermar and
his oblates, Journal of Medieval History 9 (1983), pp. 99128.
, Power and humility in Carolingian sociey: the public penance of Louis the
Pious, EME 1 (1992), pp. 2952.
, Carolingian monasticism: the power of prayer, in: R. McKitterick ed., The
New Cambridge Medieval History II, c. 700c. 900 (Cambridge, 1995), pp. 62253.
, In Samuels Image: Child oblation in the early medieval West (Leiden/ New York/Kln,
1996).
, What was public about public penance? Paenitentia publica and justice in the
Carolingian world, La giustizia nellalto medioevo (secoli IXXI) II, Settimane 42
(Spoleto, 1997), pp. 863902.
, Imitatio morum. The cloister and clerical purity in the Carolingian world,
in: M. Frassetto ed., Medieval purity and piety. Essays on medieval clerical celibacy and
religious reform (New York/London, 1998), pp. 4980.
, Adding insult to injury: Julian of Toledo and his Historia Wambae, in
P. Heather ed., The Visigoths from the Migration Period to the seventh century. An ethno-
graphic perspective (Woodbridge, 1999), pp. 373422.
, Transformations of penance, in: F. Theuws and J.L. Nelson eds., Rituals of
power. From late Antiquity to the early Middle Ages, TRW 8 (Leiden/Boston/Kln,
2000), pp. 184224.
, Internal cloisters: the case of Ekkehards Casus sancti Galli, in: W. Pohl
and H. Reimitz eds., Grenze und Differenz im frhen Mittelalter. Forschungen zur
Geschichte des Mittelalters 1 (Wien, 2000), pp. 20921.
, Religion, in: R. McKitterick ed, The Short Oxford History of Europe II (Oxford,
2001), pp. 13164.
Joris, A., A propos de burgus Huy et Namur, VillesAffairesMentalits.
Autour du pays Mosan (Brussels, 1993), pp. 13948.
Josse, M., Le domain de Jupille des origines 1297 (Bruxelles, 1966).
Josi, E., Scoperta dun altare e di pitture nella basilica di S. Ermete, Rivista di
archeologia cristiana 17 (1940), pp. 195208.
Jrgensen, L., The find material from the settlement of Gudme IIcomposition
 575

and interpretation, in: P.O. Nielsen e.a. eds., The archaeology of Gudme and Lundeborg
(Copenhagen, 1994), pp. 5363.
, The warrior aristocracy of Gudme: The emergence of landed aristocracy in
Late Iron Age Denmark?, in: H. Gostein Resi ed., Produksjon og Samfund. Varia
30, Universitetets Oldsaksamling (Oslo, 1995), pp. 20520.
, Stormandssder og skattefund i 3.12.rhundrede, Fortid of Nutid 2 (1995),
pp. 83110.
Kasprzycka, M., Mierzeja Wi lanapowstanie, rozwj, cie niny, in: P. Urba czyk
ed., Adalbertus 1 (Warsaw, 1998), pp. 3548.
, T o paleogeograficzne osadnictwa zu aw Elbl[skich w pierwszym tysi[cle-ciu naszej ery
(Warsaw, 1999).
, Zmiany klimatyczne i procesy osadnicze w rejonie dolnej Wis y 25001000 lat temu,
[manuscript submitted to Archeologia Polski].
Kasten, B., Adalhard von Corbie (Dsseldorf, 1986).
, Erbrechtliche Verfgungen des 8. und 9. Jhts. Zugleich ein Beitrag zur
Organisation und zur Schriftlichkeit bei der Verwaltung adliger Grundherrschaft
am Beispiel des Grafen Heccard aus Burgund, Zeitschrift der Savigny-Stiftung fr
Rechtsgeschichte, Germanische Abteilung 107 (1990), pp. 236338.
Kazhdan, A. and G. Constable, People and power in Byzantium. An introduction to mod-
ern Byzantine studies (Washington DC, 1982).
, Polis and kastron in Theophanes and in some other historical texts, in:
EUCUXIA, Mlanges offerts Hlne Ahrweiler 2. Byzantina Sorbonensia 16 (Paris,
1998), pp. 34560.
Keefe, S.A., Carolingian baptismal expositions: a handlist of tracts and manu-
scripts, in: U.-R. Blumenthal ed., Carolingian essays (Washington DC, 1983), pp.
169237.
Kehr, P. ed., Italia Pontificia 5 (Berlin, 1911).
Keller, H., Grundlagen ottonischer Knigsherrschaft, in: K. Schmid ed., Reich und
Kirche vor dem Investiturstreit. Festschrift G. Tellenbach (Sigmaringen, 1985), pp. 1734.
, Zum Charakter der Staatlichkeit zwischen karolingischer Reichsreform und
hochmittelalterliche Herrschaftsausbau, Frhmittelalterliche Studien 32 (1989), pp.
24864.
, Reichsorganisation, Herrschaftsformen und Gesellschaftsstrukturen im Regnum
Teutonicum, in: Il secolo di ferro: mito e realt de Secolo X, vol 1. 38 (1990), pp.
15995.
Kemp, B., Some aspects of the parochia of Leominster in the 12th century, in:
J. Blair ed., Minsters and parish churches: the local church in transition, c. 9501200.
Oxbow Monograph 17 (Oxford, 1988), pp. 8395.
Kennedy, H., From polis to madina: urban change in late antique and early Islamic
Syria, Past and Present 106 (1985), pp. 327.
Kessler, H.L., Pictorial narrative and church mission in sixth-century Gaul, in:
H.L. Kessler and M.S. Simpson eds., Pictorial narrative in Antiquity and the Middle
Ages (Washington, 1985), pp. 7591.
King, J. McA., Prestations and distance: A role for Early Saxon grave goods in social repro-
duction A.D. 450600 (unpubl. M.Litt. thesis, University of Oxford, 1999).
King, P.D., Law and society in the Visigothic Kingdom (Cambridge, 1972).
Kirn, P., Die mittelalterliche Staatsverwaltung als geistesgeschichtliches Problem,
Historisches Vierteljahrschrift 27 (1937), pp. 52348.
Kitchen, J., Saints lives and the rhetoric of gender: Male and female in Merovingian hagio-
graphy (Oxford, 1998).
Kitzinger, E., Mosaic pavements in the Greek east and the question of a renais-
sance under Justinian, Actes du VI e Congrs International dEtudes Byzantines 2 (Paris,
1951), pp. 20923, repr. in: idem, The art of Byzantium and the medieval west: selected
studies (Bloomington, 1976), essay III.
576

Klauser, T., Rom und der Kult der Gottesmutter Maria, Jahrbuch fr Antike und
Christentum 15 (1972), pp. 12035.
Klingshirn, W.E., Caesarius of Arles: The making of a Christian community in late antique
Gaul (Cambridge, 1994).
Knpp, F. ed., Die Reichsabtei Lorsch, 2 vols. (Darmstadt, 19747).
Koch, A.C.-F., Phasen in der Entstehung von Kaufmannsniederlassungen zwischen
Maas und Nordsee in der Karolingerzeit, in: Landschaft und Geschichte. Festschrift
fr Franz Petri (Bonn, 1970), pp. 31224.
Koch, U., Das frnkische Grberfeld von Klepsau im Hohenlohekreis. Forschungen und
Berichte zur Vor- und Frhgeschichte in Baden-Wrttemberg 38 (Stuttgart, 1990).
Kokowski, A., Grupa Mas om\cka (Lublin, 1995).
Kolendo, J., A la recherche de lambre baltique. Lexpedition dun chevalier romain sous Nron,
Studia Antiqua 4 (Warsaw, 1981).
, Les connaissances des Anciens sur les lieux de provenance de lambre jaune,
Archeologia 34 (1985), pp. 115.
, Rejon uj cia Wis y w oczach staro ytnych, in: Badania archeologiczne w wojew-
dztwie elbl[skim w latach 19801983 (Malbork, 1987), pp. 193206.
, Nap yw bursz\ynu z p nocy na tereny Imperium Rzymskiego w IVI w.n.e.,
Prace Muzeum Ziemi, vol. 41 (1991), pp. 91100.
Klzer, T., Merowingerstudien I. MGH Studien und Texte 21 (Hannover, 1998).
Kooistra, L., Borderland farming. Possibilities and limitations of farming in the Roman period
and the early Middle Ages between the Rhine and the Meuse (Assen and Amersfoort,
1996).
Kopytoff, I., The cultural biography of things: commoditization as process, in:
A. Appadurai ed., The social life of things: Commodities in a cultural perspective (Cambridge,
1988), pp. 6494.
Kottje, R., Knig Ludwig der Deutsche und die Fastenzeit, in: H. Roman and
J. Ratzinger eds., Mysterium der Gnade: Festschrift J. Auer (Regensburg, 1975), pp.
30711.
, Die Bubcher Halitgars von Cambrai und des Hrabanus Maurus. Ihre berlieferung
und ihre Quellen. Beitrge zur Geschichte und Quellenkunde des Mittelalters 8
(Berlin/New York, 1980).
Kotula, T., Le dbut de la crise du IIIe sicle de lEmpire romain: depuis quand?
(summary) in: A. Bursche, M. Mielczarek, W. Nowakowski eds., Nunc de Svebis
dicendum est (Warsaw, 1995), p. 156.
Krautheimer, R., An Introduction to the Iconography of Medieval Architecture,
Journal of the Warburg and Courtauld Institutes 5 (1942), pp. 133.
, Corpus Basilicarum Christianarum Romae 3 (Rome, 1967).
, Rome: Profile of a city, 3121308 (Princeton, 1980).
, Three Christian capitals: Topography and politics (Berkeley, 1983).
Krmer, W., Zur Wiederverwendung antiker Gefe im frhen Mittelalter, Germania
43 (1965), pp. 3279.
Kromann, A., Gudme and Lundeborg The coins, in: P.O. Nielsen e.a. eds.,
The archaeology of Gudme and Lundeborg (Copenhagen, 1994), pp. 647.
Krger, K.-H., Saint-Bertin als Grablege Childerichs III und der Grafen von
Flandern, Frhmittelalterliche Studien 8 (1974), pp. 7180.
, Knigskonversionen im 8. Jahrhundert, Frhmittelalterliche Studien 7 (1983),
pp. 169222.
Kuhn, H., Das Fortleben des germanisches Heidentums nach der Christianiserung,
La conversione al cristianesimo nellEuropa dellAlto Medioevo (Spoleto, 1967), pp. 74357.
Knzel, R., Paganisme, syncretisme et culture religieuse populaire au haut Moyen
Age. Rflexions de mthode, Annales ESC 47 (1992), pp. 105569.
Kupper, J.L., Sources crites: des origines 1185, in: M. Otte ed., Les fouilles de
la place Saint Lambert Lige (Lige, 1984), pp. 314.
 577

, Saint Lambert: de lhistoire la legende, Revue dhistoire ecclesiastique 79 (1984),

pp. 549.
, Archologie et Histoire. Aux origines de la cit de Lige (VIIIeXIe sicle),
in: Ontstaan en vroegste geschiedenis van de middeleeuwse steden in de zuidelijke Nederlanden.
Een archeologisch en historisch probleem (Brussels, 1990), pp. 37789.
La Rocca, C., Public buildings and urban change in northern Italy in the early
medieval period, in: J. Rich ed., The city in late Antiquity (London, 1992), pp. 16180.
, Using the Roman past. Abandoned towns and local power in eleventh-
century Piemonte, EME 5 (1996), pp. 4569.
, Les reines dans le royaume dItalie du IXe sicle, in: F. Bougard and
R. Le Jan eds, Les douaires (Lille, forthcoming).
Lachmann, H.-P., Die frhmittelalterlichen Marken zwischen Rhein und Odenwald
unter besondere Bercksichtigung der Mark Heppenheim, Berichte zur deutsche
Landeskunde 49 (1975), pp. 2737.
Ladner, G., I mosaici e gli affreschi ecclesiastico-politici nellantico palazzo later-
anense, Rivista di archeologia cristiana 12 (1935), pp. 26592.
, Die Papstbildnisse des Altertum und des Mittelalters 1: Bis zum Ende des Investiturstreits.
Monumenta di antichit cristiana, 2d series, 4 (Vatican City, 1941).
Lafon, X., A propos de Saint Ulrich: Villas et lieux de culte dans la Gaule du
Nord-Est, in: Aspects de la religion celtique et gallo-romaine dans le Nord-Est de la Gaule
la lumire des dcouvertes rcentes (Saint-Di-des-Vosges, 1989), pp. 5972.
Laing, L., The archaeology of late Celtic Britain and Ireland c. 4001200 A.D. (London, 1975).
Lapidge, M and S. Keynes, Alfred the Great. Assers Life of King Alfred and other con-
temporary sources (Harmondsworth, 1983).
Lapidus, I.M., Muslim cities in the Later Middle Ages (Harvard, 1967).
Largillire, R., Les saints et lorganisation chrtienne primitive dans lArmorique breton (Rennes,
1925), pp. 1727.
Larner, J., Marco Polo and the discovery of the world (New Haven/London, 1999).
Larsson, L. and B. Hrdh eds., Centrala Platser, Centrala Frgor. Acta Archaeologica
Lundensia, Ser. In 8, no. 28 (Lund, 1998).
Laske, W., Das Problem der Mnchung in der Vlkerwanderungszeit. Rechtshistorische
Arbeiten 2 (Zrich, 1973).
, Zwangsaufenthalt im frhmittelalterlichen Kloster. Gott und Mensch im Ein-
klang und Widerstreit, Zeitschrift der Savignystiftung fr Rechtsgeschichte, kanonistische
Abteilung 95 (1978), pp. 32130.
Lassus, J., La fortresse byzantine de Thamugadi I (Paris, 1981).
Lszl, G. and I. Rcz, Der Goldschatz von Nagyszentmikls (Budapest, 1977).
Laszlovszky, J. ed., Tender meat under the saddle: Customs of eating, drinking and hospitality
among conquering Hungarians and nomadic people. Medium Aevum Quotidianum,
Sonderband 7 (Krems, 1998).
Lattimore, O., Inner Asian frontiers of China (New York, 1940).
Lavin, I., The House of the Lord: Aspects of the role of palace Triclinia in the
architecture of late Antiquity and the early Middle Ages, Art Bulletin 44 (1962),
pp. 414.
Lauwers, M., La mmoire des anctres, le souci des morts. Morts, rites et socit au Moyen
ge (diocse de Lige, XI eXIII e sicles) (Paris, 1997).
Lawrence, M., Maria Regina, Art Bulletin 7 (1925), pp. 15061.
Le Goff, J., Time, work, and culture in the Middle Ages, transl. Arthur Goldhammer
(Chicago, 1980).
Le Jan, R., Famille et pouvoir dans le monde Franc (VII eX e sicle). Essai danthropologie
sociale (Paris, 1995).
, Frankish giving of arms and rituals of power: Continuity and change in the
Carolingian Period, in: F. Theuws and J.L. Nelson eds., Rituals of Power, TRW
8 (Leiden/Boston/Kln), 2000) pp. 281309.
578

Le Maho, J., La rutilisation funraire des difices antiques en Normandie au cours

du haut moyen ge, in: Lenvironnement des glises et la topographie religieuse des cam-
pagnes mdivales: Actes du III e Congrs International dArchologie mdivale (Aix-en-Provence,
2830 septembre 1989). Documents dArchologie Franaise 46 (Paris, 1994), pp.
1021.
, Saint-Martin-de-Boscherville (Seine-Maritime), in: M. Fixot and E. Zadora-
Rio, Lglise, Le Terroir. Monographie du Centre des Recherches Archologiques
1 (Paris, 1989), pp. 639.
Le Matre, P., Image du Christ, image de lempereur. Lexemple du culte du Saint
Sauveur sous Louis le Pieux, Revue dHistoire de lEglise de France 68 (1982), pp.
20112.
Lebecq, S., Pour une histoire parallle de Quentovic et Dorestad, in: J.M. Duvosquel
and A. Dierkens eds., Villes et campagnes au Moyen Age. Melanges Georges Despy (Lige,
1991), pp. 41528.
, Quentovic: un tat de question, Studien zur Sachsenforschung 8 (1993), pp.
7382.
, Gregoire de Tours et la vie dchanges dans la Gaule du VIe sicle, in:
N. Gauthier and H. Galini eds., Gregoire de Tours et lespace gaulois. Actes du con-
gres internationale, Tours, 35 novembre 1994 (Tours, 1997), pp. 16976.
Lebecq, S., Les changes dans la Gaule du nord au VIe sicle, in: R. Hodges
and W. Bowden eds., The sixth century. Production, distribution and demand, TRW 6
(Leiden/Boston/Kln 1998), pp. 185202.
Lecomte, M, Lextension sud-est du pagus meldensis ou civitas meldorum au VIIe
sicle, Le Moyen Age 8 (1895), pp. 16.
Lehner, H.-J. and F. Wibl, Martigny VS: De la premire cathdrale du Valais
la paroissiale actuelle: la contribution de larchologie, Helvetia Archaeologica 25
(199498), pp. 5168.
Lehner, H.-J., Saint-Maurice, in: F. Wibl, ed., Chronique des dcouvertes ar-
chologiques dans le canton du Valais en 1995, Vallesia 51 (1996), pp. 34144.
Les premiers monuments chretiens de la France (Paris, 19951998).
Les transferts patrimoniaux en Europe occidentale VIII eX e sicle, Mlanges de lEcole Franaise
de Rome 111, 2 (1999).
Lestocquoy, J., Administration de Rome et diaconies du VIIe au IXe sicle, Rivista
di archeologia cristiana 7 (1930), pp. 26198.
Leupen, P.H.D., Het gelijk van Sint-Servaas en Onze-Lieve-Vrouwe. De Maastrichtse
kapittels in de Vroege Middeleeuwen, in: E.S.C. Erkelens-Buttinger e.a. eds., De
Kerk en de Nederlanden. Archieven, instellingen, samenleving (Hilversum, 1997), pp. 2942.
Lvi-Provenal, E., Inscriptions arabes dEspagne (Leiden/Paris, 1931).
, Histoire de lEspagne Musulmane 3 vols. (2nd edition Paris, 1950).
Levison, W., Das Testament des Diakons Adalgisel-Grimo vom Jahre 634, Aus
rheinischer und frnkischer Frhzeit. Ausgewhlte Aufstze von Wilhelm Levison (1948, Dssel-
dorf ), pp. 118138 (Original publication: Trierer Zeitschrift 7, (1932), pp. 6985).
Lewis, I.M., Ecstatic Religion. An anthropological study of spirit possession and shamanism
(Harmondsworth, 1971).
Lewis, S., A Byzantine Virgo Militans at Charlemagnes court, Viator 11 (1980), pp.
7193.
Leyser, J., Concepts of Europe in the early and high Middle Ages, Past and Present
137 (1992), pp. 2547.
Leyser, K., Rule and conflict in an early medieval society: Ottonian Saxony (Oxford, 1979).
Lifshitz, F., Beyond positivism and genre: hagiographical texts as historical nar-
rative, Viator 25 (1994), pp. 95113.
Lim, R., Concensus and dissensus on public spectacles in early Byzantium,
Byzantinische Forschungen 24 (1997), pp. 15979.
Lim, R., People as power: games, munificence and contested topography, in:
 579

W.V. Harris ed., The transformations of Vrbs Roma in late Antiquity. Journal of Roman
Archaeology, suppl. ser. 33 (Portsmouth, 1999), pp. 26581.
Linage Conde, A., En torno a la Regula Monachorum y su relacin con otras reglas
monsticas, Bracara Augusta 21 (1967), pp. 12363.
, La autoridad en el monacato visigodo, Ligarzas 7 (1975), pp. 624.
Lith, S.M.E. van, Late Roman and early Merovingian glass from a settlement site
at Maastricht (Dutch South Limburg), parts 1 and 2, Journal of Glass Studies 29
(1987), pp. 4759 and 30 (1988), pp. 6276.
Lot, F., Mlanges dhistoire bretonne (Paris, 1907).
Lwe, H., Hinkmar von Reims und der Apokrisiar. Beitrge zur Interpretation
von De ordine palatii , in: Festschrift fr H. Heimpel 3 (Gttingen, 1972), pp. 197225.
Loyer, O., Les chrtients celtiques (Paris, 1965).
Lund Hansen, U., Rmischer Import im Norden. Nordiske Fortidsminder ser. B, bd. 10
(Copenhagen, 1987).
Lundstrm, A., Helg as frhmittelalterlicher Handelsplatz in Mittelschweden,
Frhmittelalterliche Studien 2 (1968), pp. 27890.
Lutovsky, M., Between Sutton Hoo and Chernaya Mogila: barrows in eastern and
western early medieval Europe, Antiquity 70 (1996), pp. 6716.
Lutz, M., La villa Gallo-Romaine et la ncropole mrovingienne de Berthelming
(Moselle), Revue archologique de lEst et du Centre-Est 1 (1950), pp. 1804.
MacDougall, H.A., Racial myth in English history: Trojans, Teutons, and Anglo-Saxons
(Montreal, 1982).
MacLean, S., Charles the Fat. University of London Ph.D. thesis (London, 2000).
Madsen, C. and H. Thrane, Mllegrdsmarkens veje og huse, Fynske Minder 1995
(Odense, 1995), pp. 7791.
Maenchen-Helfen, O., Die Welt der Hunnen (Wien/Kln/Graz, 1979).
Magdalino, P., Constantinople mdivale. Etudes sur lvolution des structures urbaines. Travaux
et mmoires du CNRS, monographies 9 (Paris, 1996).
Mainstone, R., Hagia Sophia. Architecture, structure and liturgy of Justinians Great Church
(New York, 1988).
Mle, ., La fin du paganisme en Gaule (Paris, 1950).
Manacorda, D. and E. Zanini, Sul paessagio urbano di Roma nellAlto Medioevo,
in: R. Francovich and G. Noy eds. La storia dellalto medioevo italiano (VIX secolo)
alla luce dellarcheologia. Biblioteca dellarcheologia medievale 11 (Florence, 1994),
pp. 63550.
Mango, C., The development of Constantinople as an urban centre, in: 17th
International Byzantine Congress, Main Papers (New Rochelle, 1986), p. 118; repr. in
idem, Studies on Constantinople (Aldershot, 1993), nr. I.
, Le dveloppement urbain de Constantinople (IV eVII e sicles). Travaux et mmoires
du CNRS, monographies 2 (Paris, 1990).
, Constantines mausoleum and the translation of relics, Byzantinische Zeitschrift
83 (1990), pp. 5162, 434; repr. in: idem, Studies on Constantinople (Aldershot, 1993),
study V.
Marazzi, F., Rome in transition: economic and political changes in Rome, 4th6th
centuries, in: J.M.H. Smith ed., Early medieval Rome and the Christian West. Essays
in honour of Donald A. Bullough (Leiden/Boston/Kln, 2000), pp. 2141.
Maret, P. de, The smiths myth and the origin of leadership in Central Africa,
in: R. Haaland and P. Shinnie eds., African Iron Working (Olso, 1985), pp. 7387.
Marn Guzmn, R., The Causes of the Revolt of Umar ibn Hafsun in Al-Andalus
880928: a study in medieval Islamic social history, Arabica 17 (1995), pp.
180221.
Marinescu, C.A., Transformations: Classical objects and their re-use during late
Antiquity, in: R.W. Mathisen and H.S. Sivan eds., Shifting frontiers in late Antiquity
(Aldershot, 1996), pp. 28598.
580

Markus, R.A., The end of ancient Christianity (Cambridge, 1990).

, From Caesarius to Boniface: christianity and paganism in Gaul, in: Le sep-
time sicle: Changements et continuits (London, 1992), pp. 15472.
, How on earth could places become holy? Origins of the Christian idea of
holy places, Journal of Early Christian Studies 2 (1994), pp. 25771.
Marrou, H.-I., Lorigine orientale des diaconies romaines, Mlanges darchologie et
dhistoire 57 (1940), pp. 95142.
Martnez Gijn, J., La comunidad hereditaria y la particin de la herencia en el
derecho medieval espaol, Anuario de Historia del Derecho Espaol 27 (1957), pp.
221302.
Masai, F., La Vita patrum iurensium et les dbuts du monachisme Saint-Maurice
dAgaune, in: J. Autenrieth and F. Brunhlzl eds., Festschrift Bernhard Bischoff zu
seinem 65. Geburtstag (Stuttgart, 1971), pp. 4369.
Marquise de Maill, Les cryptes de Jouarre (Paris, 1971).
Martin, M., Das frnkische Grberfeld von Basel-Bernerring. Baseler Beitrge zur Ur- und
Frhgeschichte (Basel, 1976).
Mateos, J., Le typicon de la Grande Eglise, 2 vols. Orientalia christiana analecta 165166
(Rome, 19621963).
Mathews, T., The early churches of Constantinople: Architecture and liturgy (University Park
PA, 1971).
Mathisen, R., The family of Georgius Florentius Gregorius and the bishops of
Tours, Medievalia et Humanistica 12 (1984), pp. 8395.
Mathisen, R.W., Crossing the supernatural frontier in Western late Antiquity, in:
idem and H.S. Sivan eds., Shifting frontiers in late Antiquity (Aldershot, 1996), pp.
30920.
Mathon, G., Maurilio, Bibliotheca Sanctorum 9 (Rome, 1967), pp. 1856
Matthews, J., Western Aristocracies and Imperial Court A.D. 364425 (Oxford, 1975).
Matthiae, G., Pittura politica del medioevo romano (Rome, 1964).
Mattoso, J., Les Wisigoths dans le Portugal Medieval: tat actuel de la question,
in: J. Fontaine and Ch. Pellistrandi, LEurope Heritire de lEspagne Wisigothique
(Madrid, 1992), pp. 3259.
Mauss, M., The Gift. The form and reason for exchange in archaic societies (London, 1990)
transl. W.D. Halls.
Mazzoli-Guintard, C., Remarques sur le fonctionnement dune capitale a double
polarite: Mad nat al-Zahr -Cordue, Al-Qantara 17 (1997), pp. 4364.
McCormick, M., Eternal victory: Triumphal rulership in late Antiquity, Byzantium and the
early medieval West (Cambridge, 1986).
McCready, W.D., Miracles and the Venerable Bede. Studies and Texts 118 (Toronto,
1994).
McKenna, S., Paganism and pagan survivals in Spain up to the fall of the Visigothic king-
dom (Washington, 1938).
McKitterick, R., The illusion of royal power in the Carolingian Annals, The English
Historical Review 115 (2000), pp. 120.
McLynn, N.B., Ambrose of Milan: Church and court in a Christian capital (Berkeley, 1994).
McNamara, J.A. and John E. Halborg with E. Gordon Whatley eds. and transl.,
Sainted women of the Dark Ages (Durham/London, 1992).
Meaney, A.L., Anglo-Saxon amulets and curing Stones. BAR British Series 96 (Oxford,
1981).
Melograni, A., Le pitture del VI e VIII secolo nella basilica inferiore di S. Crisogono
in Trastevere, Rivista dellistituto nazionale darcheologia a storia dellarte, 3rd ser. 13
(1990), pp. 13961.
Meneghini, R. and R. Santangeli Valenzani, Sepolture intramuranee e pesaggio
urbano a Roma tra V e VII secolo, in: L. Paroli and P. Delogu eds., La Storia
economica di Roma nellallto Medioevo alla luce dei recenti scavi archaeologici (Florence,
1993), pp. 89111.
 581

Menzer, A., Die Jahresmerkmale in den Datierungen der Papsturkunden bis zum
Ausgang des 11. Jahrhundert, Rmische Quartalschrift fr christliche Altertumskunde und
Kirchengeschichte 40 (1932), pp. 27103.
Merrifield, R., The archaeology of ritual and magic (New York, 1987).
Meslin, M., Persistances piennes en Galice, vers la fin du VIe sicle, in: Hommages
Marcel Renard 2 (Bruxelles, 1969), pp. 51224.
Metcalf, P., A Borneo journey into death: Berawan eschatology from its rituals (Philadelphia,
1982).
Meulengracht Srensen, P., Om eddadigtenes alder, in: G. Steinsland, U. Drobin,
J. Pentikinen and P. Meulengracht Srensen eds., Nordisk Hedendom. Et symposie
(Odense, 1991), pp. 21728.
Meyer-Plath, B. and A.M. Schneider, Die Landmauer von Konstantinopel (Berlin, 1943).
Mnguez, J.M., Ruptura social e implantacin del feudalismo en el Noroeste penin-
sular (Siglos VIIIX), Studia Historica. Historia Medieval 3 (1985), pp. 732.
Mitchell, J., Karl der Grosse, Rom und das Vermchtnis der Langobarden, in:
799. Karl der Grosse und Papst Leo III in Paderborn. Kunst und Kultur der Karolingerzeit,
Beitrge zum Katalog der Ausstellung (Paderborn, 1999), pp. 95108.
Mizoguchi, K., Time in the reproduction of mortuary practices, World Archaeology
25, 2 (1993), pp. 22335.
Mole, C., Uno storico del V secolo. Il vescovo Idazio (Catania, 1978).
Momigliano, A., Ancient history and the antiquarian, in: Studies in historiography
(London, 1966), pp. 139.
Montanos Ferrn, E., La familia en la alta edad media espaola (Pamplona, 1980).
Moore, H., Space, text and gender. An anthropological study of the Marakwet of Kenya
(Cambridge, 1986).
Moorhead, J., Theodoric in Italy (Oxford, 1992).
Mordek, H., Karolingische Kapitularien, in: idem ed., berlieferung und Geltung nor-
mativer Texte des frhen und hohen Mittelalters. Quellen und Forschungen zum Recht
im Mittelalter 4 (Sigmaringen, 1986), pp. 2550.
, art. Kapitularien, in: Lexikon des Mittelalters 5 (1990), cols. 9436.
, Recently discovered capitulary texts belonging to the legislation of Louis the
Pious, in: P. Godman and R. Collins eds., Charlemagnes heir. New perspectives on
the reign of Louis the Pious (Oxford, 1990), pp. 43753.
, Traditionszusammenhang der frnkischen Herrscherlasse. MGH Hilfsmittel 15 (Mnchen,
1995).
, Kapitularien und Schriftlichkeit, in: R. Schieffer ed., Schriftkultur und Reichsver-
waltung unter den Karolingern. Abhandlungen der Nordrhein-Westflischen Akademie
der Wissenschaften 97 (1996), pp. 3466.
Moreira, I., Provisatrix optima: St. Radegund of Poitiers Relic Petitions to the East,
Journal of Medieval History 19 (1993), pp. 285305.
Morey, C., The inscription on the enameled cross of Paschal I, Art Bulletin 19
(1937).
Morgan, D., The Mongols (Malden/Oxford, 1987).
Morice, H., Mmoires pour servir de preuves lhistoire ecclsiastique et civile de Bretagne,
3 vols. (Paris, 17426).
Morris, I., Death ritual and society in classical Antiquity (Cambridge, 1992).
Moszczy ski, A., Wczesno redniowieczne szlaki komunikacyjne Starego Mazowsza,
in: P. Urba czyk ed., Adalbertus, vol. 1 (Warszawa, 1998), pp. 24162.
Mhlen, B. von zur, Die Kultur der Wikinger in Ostpreussen. Bonner Hefte zur Vorgeschichte
9 (Bonn, 1975).
Muir, E., Civic ritual in Renaissance Venice (Princeton, 1981).
Mller, C.C., Die Religion der Mongolen, in: Arne Eggebrecht ed., Die Mongolen
und ihr Weltreich (Mainz, s.a.), pp. 16982.
Mller-Mertens, E., Die Reichsstruktur im Spiegel der Herrschaftspraxis Ottos des Groen
(Berlin, 1980).
582

Mller-Wille, M., Knigsgrab und Knigskirche, Berichte der Rmisch-Germanischen

Kommission 63 (1982), pp. 349412.
, Knigtum und Adel im Spiegel der Grabfunde, in: Die Franken: Wegbereiter
Europas. Vor 1500 Jahren: Knig Chlodwig und seine Erben (Mainz, 1996), pp. 20621.
Mullett, M., The language of diplomacy, in: J. Shepard and S. Franklin eds.,
Byzantine Diplomacy (Aldershot, 1992), pp. 20316.
Mund, A., Les nciens synodes abbatiaux et les Regulae SS. Patrum, Studia Anselmiana
44 (1959), pp. 10725.
, La inscripcin visigoda del monasterio de Samos, Studia Monastica 3 (1961),
pp. 15764.
Musurillo, H., The Acts of the Christian martyrs (Oxford, 1972).
Nahmer, D. von der, ber Ideallandschaften und Klostergrngungsorte, Studien
und Mitteilungen zur Geschichte des Benediktinerordens und seiner Zweige 84 (1973), pp.
195270.
Nsmann, U., The Ethnogenesis of the Danes and the making of a Danish kingdom,
in: T. Dickinson and D. Griffiths eds., The Making of Kingdoms. Anglo-Saxon Studies
in Archaeology and History 10 (Oxford, 1999), pp. 110.
Nees, L., Theodulf s mythical silver Hercules vase: poetica vanitas and the Augustinian
critique of the Roman heritage, in: Dumbarton Oaks Papers 41 (1987), pp. 44351.
Nehlsen, H. and H. Roth eds., Zum Grabfrevel in vor- und frhgeschichtlicher Zeit:
Untersuchungen zu Grabraub und Haugbrot in Mittel- und Nordeuropa. Abhandlungen
der Akademie der Wissenschaften in Gttingen, philologisch-historische Klasse,
dritte Folge 113 (Gttingen, 1978), pp. 10768.
Nelson, J.L., Queens as Jezebels: the careers of Brunhild and Balthild in Merovingian
history, Studies in Church History, Subsidia I (1978), pp. 3177; repr. in eadem,
Politics and ritual in early medieval Europe (London/Ronceverte, 1986), pp. 148.
, Parents, children and the church in the earlier Middle Ages, Studies in
Church History 31 (1994), pp. 81114.
, Literacy in Carolingian government, in: R. McKitterick ed., The uses of lit-
eracy in early medieval Europe (Cambridge, 1990), pp. 25896, repr. in: Nelson, The
Frankish world (London, 1996), pp. 136.
, The Lords anointed and the peoples choice: Carolingian royal ritual, in:
D. Cannadine and S. Price eds., Rituals of royalty (Cambridge, 1987), pp. 13780,
repr. in: eadem, The Frankish world (London, 1996), pp. 99132.
, Charles the Bald (London, 1992).
, The wary widow, in: W. Davies and P. Fouracre eds., Property and power
in the early Middle Ages (Cambridge, 1995), pp. 82113.
, Women at the court of Charlemagne: a case of monstrous regiment? in:
J.C. Parsons ed., Medieval Queenship (Stroud, 1993), repr. in eadem, The Frankish
World, 750900 (London, 1996), pp. 22342.
, Making a difference in ninth-century politics: the daughters of Desiderius,
in: A. Murray ed., After Romes fall: Narrators and sources of barbarian history. Essays
presented to Walter Goffart (Toronto, 1998), pp. 17190.
, La cour impriale de Charlemagne, in: R. Le Jan ed., La royaut et les lites
dans lEurope carolingienne (du dbut du IXe aux environs de 920) (Lille, 1998), repr. in:
eadem, Rulers and ruling families in early Medieval Europe (London, 1999), ch. XIV.
, Carolingian royal funerals, in: F. Theuws and J.L. Nelson eds, Rituals of
power: From late Antiquity to the early Middle Ages, TRW 8 (Leiden/Boston/Kln,
2000), pp. 131184.
, Charlemagne, pater optimus?, in: J. Jarnut, Am Vorabend der Kaiserkrnung: Karl
der Grosse und Leo III (forthcoming, 2001).
, Was Charlemagnes court a courtly society?, in: C. Cubitt ed., Court cul-
ture in the earlier Middle Ages (Leiden, 2001), forthcoming
, The voice of Charlemagne, forthcoming (Oxford, 2001).
 583

Neundrfer, D., Studien zur ltesten Geschichte des Kloster Lorsch (Berlin, 1920).
Nichols, J.M., transl., Ibn al-Qutiyya, The History of the Conquest of Al-Andalus, unpubl.
Ph.D. Diss. (University of North Carolina, 1975).
Nicolet, C., Linventaire du monde: Geographie et politique aux origines de lEmpire romain
(Paris, 1988).
Nie, G. de, Views from a many-windowed tower: Studies of imagination in the works of Gregory
of Tours. Studies in Classical Antiquity 7 (Amsterdam, 1987).
Nielsen, P.O., The Gudme-Lundeborg project. Interdisciplinary research 1988,
in: idem e.a. eds., The archaeology of Gudme and Lundeborg (Copenhagen, 1994), pp.
1622.
, K. Randsborg and H. Thrane eds., The archaeology of Gudme and Lundeborg
(Copenhagen, 1994).
Nilgen, U., Maria ReginaEin politischer Kultbildtypus?, Rmisches Jahrbuch fr
Kunstgeschichte 19 (1981), pp. 333.
Noble, T.F.X., The Republic of St. Peter: The birth of the Papal State, 680825 (Philadelphia,
1984).
, Theodoric the Great and the Papacy, in: Teodorico il Grande e i Goti dItalia.
Atti del XIII Congresso internazionale di studi sullalto medioevo 1 (Spoleto,
1993), pp. 395423.
, Rome in the seventh century, in: M. Lapidge ed., Archbishop Theodore:
Commemorative studies on his life and influence. Cambridge Studies in Anglo-Saxon
England 11 (Cambridge, 1995), pp. 6987.
, Paradoxes and possibilities in the sources for Roman society in the early
Middle Ages, in: J.M.H. Smith ed., Early medieval Rome and the Christian West.
Essays in honour of Donald A. Bullough (Leiden/Boston/Kln, 2000), pp. 5583.
Nock, A.D., Cremation and burial in the Roman Empire, in: Z. Stewart ed.,
Essays on Religion and the Ancient World 1 (Cambridge, Mass., 1972), pp. 277307.
Noel, R., Moines et nature sauvage dans lArdenne au haut Moyen Age, in:
J-M. Duvosquel and A. Dierkens eds., Villes et campagnes au Moyen Age. Mlanges
G. Despy, (Lige, 1991), pp. 56397.
Noetlichs, K.L., Das Kloster als Strafanstalt im kirchlichen und weltichen Recht
der Sptantike, ZRG, kan. Abt. 80 (1994), pp. 1840.
Nordstrm, K., Problems and ideas concerning ideology in the construction of
religion and ritual as analytical concepts, Lund Archaeological Review 3 (1997),
pp. 4957.
Noort, R., van de, The context of early medieval barrows in western Europe,
Antiquity 67 (1993), pp. 6673.
Nora, P., Between memory and history: Les Lieux de Mmoire, Representations 26
(1989), pp. 725.
Ntedika, J., Lvocation de lau-del dans la prire pour les morts. tude de patristique et de
liturgie latines (IV eVIII e sicles) (Louvain, 1971).
OHara, M.D., A Find of Byzantine Silver from the Mint of Rome for the Period
A.D. 641752, Swiss Numismatic Review 64 (1985), pp. 10539.
Oakeshott, W., The mosaics of Rome (New York, 1967).
Oexle, G., Forschungen zu monastischen und geistlichen Gemeinschaften im westfrnkischen
Bereich. Mnstersche Mittelalter-Schriften 31 (Mnchen, 1978).
Oexle, O.G., Die Gegenwart der Toten, in: H. Breat and W. Verbeke eds., Death
in the Middle Ages. Mediaevalia Louvaniensia Series 1, Studia 9 (Louvain, 1983),
pp. 4954.
Okulicz, J., Das Grberfeld von Weklice. Zur Besiedlungsgeschichte des Weich-
selsraums in der rmischen Kaiserzeit, Archeologia 40 (1991), pp. 11527.
, Centrum kulturowe z pierwszych wiekw naszej ery u uj cia Wis y, Barbaricum
2 (1992), pp. 13756.
Oliger, P.R., Les vques rguliers (Paris, Louvain, 1958).
584

Ommeren, H.R. van, Bronnen voor de geschiedenis van Maastricht (3591204).

Deel 1: Bronnen betreffende het tijdvak vanaf het jaar 359 tot en met 923,
Publications de la Socit Historique et Archologique de Limbourg 127 (1991), pp. 548.
Orlandis, J., Los orgenes del monaquismo dplice en Espaa, in: idem, Estudios
sobre instituciones monsticas medievales (Pamplona, 1971), pp. 1734.
, El movimiento asctico de San Fructuoso y la congregacin monstica
dumiense, in: idem, Estudios sobre las organizaciones monsticas medievales (Pamplona,
1971), pp. 6982.
, Las congregaciones monsticas en la tradicin suevo-gtica, in: idem, Estudios
sobre las organizaciones monsticas medievales (Pamplona, 1971), pp. 95123.
, Los monasterios familiares en Hispania durante la Alta Edad Media, in:
idem, Estudios sobre instituciones monsticas medievales (Pamplona, 1971), pp. 12564.
Orlandoni, M., Ipotesi sul significato della sigla IB nel denaro papale di Adriano
I rinvenuto negli scavi di Aosta, Memorie dellAccademia Italiana di Studi Filatelici e
Numismatici 4 (1990).
Osborne, J., The portrait of Pope Leo IV in San Clemente, Rome: a re-examination
of the so-called square nimbus in medieval art, PBSR 47 (1979), pp. 5865.
, Early medieval painting in San Clemente, Rome: the Madonna and Child
in the niche, Gesta 20 (1981), pp. 299310.
, The Roman catacombs in the Middle Ages, Papers of the British School at
Rome 53 (1985), pp. 278328.
Ossel, P. van, tablissements ruraux de lantiquit tardive dans le nord de la Gaule. 51e sup-
plment Gallia (Paris, 1992).
Oswald, F., L. Schaefer and H.R. Sennhauser eds., Vorromanischen Kirchenbauten. Katalog
der Denkmler bis zum Ausgang der Ottonen, 2 vols. (Mnchen, 19661971, reprint
1991).
Ousterhout R., Reconstructing ninth-century Constantinople, in: L. Brubaker ed.,
Byzantium in the ninth century: Dead or alive? (Aldershot, 1998), pp. 11530.
Oxenstierna, E., Die Goldhrner von Gallehus (Liding, 1956).
Oxford Dictionary of Byzantium 2 (New York, 1991).
stergaard Srensen, P., Houses, farmsteads and settlement pattern in the Gudme
area, in: P.O. Nielsen e.a. eds., The Archaeology of Gudme and Lundeborg (Copenhagen,
1994), pp. 417.
, Gudmehallen. Kongeligt byggeri fra jernalderen, Nationalmuseets Arbejdsmark
1994 (Copenhagen, 1994), pp. 2539.
Pffgen, B., Die Ausgrabungen in St. Severin zu Kln (Mainz, 1992).
Palazzo, E. and A.K. Johansson, Jalons liturgiques pour une histoire du culte de
la vierge, in: D. Iogna-Prat, . Palazzo and D. Russo eds., Marie: Le culte de la
vierge dans la socit mdivale (Paris, 1996), pp. 1536.
Panhuysen, T.A.S.M., Maastricht staat op zijn verleden (Maastricht, 1984).
, Wat weten we over de continuiteit van Maastricht?, in: C.G. de Dijn ed.,
Sint-Servatius, bisschop van Tongeren-Maastricht. Het vroegste Christendom in het Maasland.
Handelingen van het colloquium te Alden Biesen (Bilzen), Tongeren en Maastricht 1984
(Borgloon-Rijkel, 1986), pp. 125146.
, De Archeoloog, De Sint Servaas. Tweemaandelijks restauratie-informatiebulletin
3940, (1988), pp. 30912.
, Die Maastrichter Servatiuskirche im Frhmittelalter. Ein Vorbericht ber
die jngsten Grabungen des stdtisches Amtes fr Bodendenkmalpflege Maastricht,
Kunstchronik. Monatschrift fr Kunstwissenschaft, Museumswesen und Denkmalpflege 43 (1990),
pp. 51433.
, De Sint-Servaaskerk te Maastricht in de vroege middeleeuwen. Voorlopig
eindverslag van de opgravingen door de dienst Stadsontwikkeling Maastricht in
de periode 19811989, Bulletin KNOB (1991), pp. 1524.
 585

, and Leupen, P.H.D., Maastricht in het eerste millenium. De vroegste stads-

ontwikkeling in Nederland, Ontstaan en vroegste geschiedenis van de middeleeuwse steden
in de zuidelijke Nederlanden. Een archeologisch en historisch probleem (Brussel, 1990), pp.
41155.
Panhuysen, T.A.S.M., W., Dijkman, R.A., Hulst and R.G.A.M., Panhuysen, Opgra-
vingen door het Gemeentelijk Oudheidkundig Bodemonderzoek Maastricht [GOBM]
in het jaar 1991, Publications de la Socit Historique et Archologique dans le Limbourg
128 (1992), pp. 25988.
Panhuysen, T.A.S.M., W., Dijkman and R.A., Hulst, Archeologische kroniek van
Maastricht, Publications de la Socit Historique et Archeologique dans le Limbourg 129
(1993), pp. 33580.
Panhuysen, T.A.S.M., Romeins Maastricht en zijn beelden. Roman Maastricht reflected in
Stones (Maastricht/Assen, 1996).
Papinot, J.-C., Notices sur les vestiges archologiques de Civaux (Poitiers, 1971).
Parker, M.A., Beowulf and Christianity (New York, 1987).
Paulsen, P., Alamannische Adelsgrber von Niederstotzingen. Verffentlichungen des Staatlichen
Amts fr Denkmalpflege Stuttgart A12 (Stuttgart, 1967).
Pavn Maldonado, B., Entre la historia y la arqueologa. El enigma de la Crdoba
califal desaparecida, Al-Qantara 9 (1988), pp. 16998 and 40326.
Paxton, F.S., Christianizing death: The creation of a ritual process in early medieval Europe
(Ithaca and London, 1990).
, Power and the power to heal: the cult of St. Sigismund of Burgundy, EME
2 (1993), pp. 95110.
Percival, J., The Roman villa: An historical introduction (Berkeley, 1976).
, Villas and monasteries in late Roman Gaul, Journal of Ecclesiastical History
48 (1997), pp. 121.
Pereira-Menaut, G., Cambios estructurales versus romanizacin convencional. La
transformacin del paisaje poltico en el Norte de Hispania, in: J. Gonzlez and
J. Arce eds., Estudios sobre la Tabula Siarensis (Madrid, 1988), pp. 24559.
, Callaecia, in: F. Coarelli, M. Torelli and J. Uroz Sez eds., Conquista romana
y modos de intervencin en la organizacin urbana y territorial. Dialoghi di Archeologia 12
(Roma, 1992), pp. 31925.
Piel, J.M., ber die Namen der sog. Divisio Theodemiri , Romanische Forschungen 71
(1959), pp. 1607.
Prez, F., Cordue, DHGE 13 (Paris, 1956), 859.
Prez de Urbel, J., Los monjes espaoles en la edad media (Madrid, 1934).
, Sampiro, su crnica y la monarqua leonesa en el siglo X (Madrid, 1952).
Prin, P., Remarques sur la topographie funraire en Gaule mrovingienne et
sa priphrie: Les ncropoles romaines tardives aux ncropoles du haut-moyen
ge, Cahiers archologiques 35 (1987), pp. 930.
, Die Grabsttten der merowingischen Knige in Paris, in: Die Franken:
Wegbereiter Europas. Vor 1500 Jahren: Knig Chlodwig und seine Erben (Mainz, 1996),
pp. 41622.
Prin, P. and M. Kazanski, Das Grab Childerichs I., in: Die Franken: Wegbereiter
Europas. Vor 1500 Jahren: Knig Chlodwig und seine Erben (Mains, 1996), pp. 17382.
Peschlow, U., Die Irenenkirche in Istanbul: Untersuchungen zur Architektur. Istanbuler
Mitteilungen. Beiheft 18 (Tbingen, 1977).
Petersen, J.M., Dead or alive? The Holy Man as healer in East and West in the
late sixth century, Journal of Medieval History 9 (1983), pp. 918.
Pietri, C., Rgions ecclsiastiques et paroisses romaines in: Actes du XI e Congrs
international darchologie chrtienne (Rome, 1989), pp. 103562.
Pietri, L. ed., La ville de Tours du IV e au VI e sicle: Naissance dune cit chrtienne (Rome,
1983).
586

, Loca sancta: la gographie de la saintet dans lhagiographie gauloise (IVeVIe

s.) in: S. Boesch Gajano and L. Scaraffia eds., Luoghi sacri e spazi della santit
(Turin, 1990), pp. 2335.
Piggott, S., Antiquarian thought in the sixteenth and seventeenth centuries, in:
Ruins in a landscape: Essays in antiquarianism (Edinburgh, 1976), pp. 124.
Pitt-Rivers, J., Anthropologie de lhonneur. La msaventure de Sichem (Paris, 1983).
Piva, P., Le cattedrali lombarde. Ricerche sulle cattedrali doppie da SantAmbrogio allet roma-
nica (Quistello, 1990).
Planiol, M., Histoire des institutions de la Bretagne, 5 vols., (Mayenne, 19814).
Plumier, J. ed., Cinq annes darchologie en province de Namur 1990 1995 (Namur, 1996).
, La chapelle Saint Hilaire, in: idem ed., Cinq annes darchologie en province de
Namur 19901995 (Namur, 1996a), pp. 958.
Plumier-Torts, S. and J. Plumier, Lpoque mrovingienne in: J. Plumier ed., Cinq
annes darchologie en province de Namur 19901995 (Namur, 1996), pp. 9194.
Poerck, G. de, Les reliques des saints Maixent et Lger aux IXe et Xe sicles et
les origines de labbaye de lEbreuil en Bourbonnais, Revue Bndictine 67 (1962),
pp. 6195.
Pohl, W., Die Gepiden und die Gentes an der mittleren Donau nach dem Zerfall
des Attilareiches, in: H. Wolfram and F. Daim eds., Die Vlker an den mittleren
und unteren Donau im fnften und sechsten Jahrhundert. Denkschriften der sterreichischen
Akademie der Wissenschaften, Phil.-Hist. Kl. 145 (Wien, 1980), pp. 239305.
, Die Awaren (Mnchen, 1988).
, Konfliktverlauf und Konfliktbewltigung: Rmer und Barbaren im frhen
Mittelalter, Frhmittelalterliche Studien 26 (1992), pp. 165207.
, La sfida attilana: Dinamica di un potere barbaro, in: S. Blason Scarel ed.,
AttilaFlagellum Dei ? (Rome, 1994), pp. 6991.
, The role of steppe peoples in eastern and central Europe in the first mil-
lennium A.D., in: P. Urbanczyk ed., Origins of Central Europe (Warsaw, 1997), pp.
6578.
, Die Germanen (Mnchen, 2000).
, Frontiers in Lombard Italy: the laws of Ratchis and Aistulf , in: W. Pohl,
I. Wood and H. Reimitz eds., The transformation of frontiers. From late Antiquity to the
Carolingians (Leiden/Boston/Kln, 2001), pp. 117141.
, Deliberate ambiguity. The Lombards and Christianity, in: I. Wood ed.,
Conversion in the Middle Ages. IMR series (Turnhout, forthcoming).
, GermaniaHerrschaftssitze stlich des Rheins und nrdlich der Donau,
in: J.M. Gurt and G. Ripoll eds., Sedes regiae (400800 D.C.). Reial Acadmia de
Bones Lletres de Barcelona (Barcelona, 2000), pp. 30518.
, Le origini etniche dellEuropa (Roma, 2000).
Pohl, W., and M. Diesenberger, Eugippius und Severin. Der Autor, der Text und der Heilige,
(Wien, forthcoming).
Pol, A., Les montaires Huy et Maastricht. Production et distribution des mon-
naies mrovingiens mosanes, Bulletin de lInstitut Archologique Ligois 107 (1995),
pp. 185200.
Polanyi, K., The Economy as instituted process, in: idem e.a. eds., Trade and mar-
ket in early empires: Economics in theory and practice (New York, 1957), pp. 24357.
Pomian, K., Collectors and curiosities: Paris and Venice, 1500 1800, transl. Elizabeth
Wiles-Portier (Cambridge, 1990).
Poole, R.L., Lectures on the History of the Papal Chancery Down to the Time of Innocent III.
(Cambridge, 1915).
Potter, T., Towns in late Antiquity: Iol Caesarea and its context (Oxford, 1995).
Poulin, J.-C., Filibertus, Lexikon des Mittelalters 4 (Mnchen, 1989), pp. 4478.
, Le Dossier hagiographique de Saint Conwoion de Redon. A propos dune
dition rcente, Francia 18,1 (1991), pp. 13959.
 587

Poulter, A., The use and abuse of urbanism in the Danubian provinces during
the later Roman empire, in: J. Rich ed., The city in late Antiquity (London, 1992),
pp. 99135.
, Nicopolis ad Istrum, a Roman, late Roman and early Byzantine city. Journal of Roman
Studies monograph 8 (London, 1995).
Prinz, F., Frhes Mnchtum im Frankenreich. Kultur und Gesellschaft in Gallien, den Rheinlnden
und Bayern am Beispiel der monastischen Entwicklung (Mnchen and Wien, 1965; 2nd
ed. Mnchen, 1988).
Prior, L., The social organisation of death: Medical discourse and social practices in Belfast
(Basingstoke, 1989).
Puertas Tricas, R., La iglesia rupestre de Las Mesas de Villaverde (Ardales, Malaga),
Mainake 1 (1979), pp. 179216.
Quentel, P., Toponymie bretonne. Chronologie des noms en loc-, Revue Internationale
dOnomastique 14 (1962), pp. 818.
Rabe, S., Faith, art and politics at St-Riquier (Philadelphia, 1995).
Reece, R. ed., Burial in the Roman world. Council for British Archaeology Research
Report 22 (London, 1977).
Reekmans, L., Le dveloppement topographique de la rgion de Vatican la fin
de lantiquit et au dbut du moyen ge, in: Mlanges darchologie et dhistoire de
lart offerts au Professeur Jacques Lavalleye. Universit de Louvain, Recueil de travaux
dhistoire et de philologie, 4th series, 45 (Louvain, 1970), pp. 195235.
, Limplantation monumentale chrtienne dans le paysage urbain de Rome
de 300 850, in: Actes du XI e Congrs International dArchologie chrtienne (Rome,
1989), pp. 861915.
Reichert, V., Begegnungen mit China: Die Entdeckung Ostasiens im Mittelalter (Sigmaringen,
1992).
Reimitz, H., Grenzen und Grenzberschreitungen im karolingischen Mitteleuropa,
in: W. Pohl and H. Reimitz eds., Grenze und Differenz im frhen Mittelalter (Wien,
2000), pp. 105166.
Reuter, T., Plunder and tribute in the Carolingian Empire, Transactions of the Royal
Historical Society 35 (1985), pp. 7594.
, Germany in the early Middle Ages, 8001056 (London, 1991).
, Assemblies, in: P. Linehan and J.L. Nelson eds, The Medieval World (London,
2001), forthcoming.
Reynaud, J.-F., Lyon (Rhne) aux premiers temps chrtiens. Basiliques et ncropoles. Guides
archologiques de la France 10 (1986).
Reynolds, R., The organization, law, and liturgy of the Western Church, 700900,
in: R. McKitterick ed., The New Cambridge Medieval History 2 (Cambridge, 1995),
pp. 587621.
Reynolds, S., Kingdoms and communities in Western Europe, 9001300 (Oxford, 1984).
Rich, P., Les reprsentations du palais dans les textes littraires du Haut Moyen
Age, Francia 4 (1976), pp. 16171.
Richter, M., The formation of the medieval West: Studies in the oral culture of the barbarians
(Dublin, 1994).
Ringbom, L.I., Graltempel und Paradise (Stockholm, 1951).
Ringtved, J., The geography of power: South Scandinavia before the Danish king-
dom, in: T. Dickinson and D. Griffiths eds., The Making of Kingdoms (Oxford,
1999), pp. 4964.
Robinson, F.C., Beowulf , in: M. Godden and M. Lapidge eds., The Cambridge
Companion to Old English Literature (Cambridge, 1991), pp. 14259.
Roblin, M., Fontaines sacres et ncropoles antiques, deux sites frquents dglises
paroissiales rurales dans les sept anciens diocses de lOise, Revue dhistoire de
lglise de France 62 (1975), pp. 23551.
Rodrguez Fernndez, C. ed., La Pasin de San Pelayo (Santiago de Compostela, 1991).
588

Rollason, D., Saints and relics in Anglo-Saxon England (Oxford, 1989).

Rolle, R., The world of the Scythians (London, 1989).
Roma e let carolingia: Atti del giornati di studio 38 maggio 1976 (Rome, 1976).
Romanelli, P. and P.J. Nordhagen, Santa Maria Antiqua (Rome, 1964).
Roosens, H., and J. Alenus-Lecerf, Spultures mrovingiennes au Vieux Cimetire
dArlon, Archaeologia Belgica 88 (1965).
Roosens, H., Het Merovingisch grafveld van Rosmeer. Archaeologia Belgica 204 (Brussel,
1978).
Rosenwein, B.H., To be the neighbor of Saint-Peter. The social meaning of Clunys property,
909 1049 (Ithaca, 1989).
, Controlling Paradigms, in idem ed., Angers Past. The Social Uses of an Emotion
in the Middle Ages (Ithaca, 1998), pp. 23347.
, Negotiating space: Power, restraint, and privileges of immunity in early medieval Europe
(Ithaca, N.Y., 1999).
, Perennial prayer at Agaune, in: S. Farmer and B.H. Rosenwein eds., Monks
and nuns, saints and outcasts: Religion in medieval society. (Ithaca, N.Y.), pp. 3656.
Roskams, S., Urban transition in early medieval Britain: the case of York, in:
N. Christie and S.T. Loseby eds., Towns in transition. Urban evolution in late Antiquity
and the early Middle Ages (Aldershot, 1996), pp. 26288.
, Urban transition in North Africa: Roman and medieval Towns of the
Maghreb, in: N. Christie and S.T. Loseby eds., Towns in transition. Urban evolu-
tion in late Antiquity and the early Middle Ages (Aldershot, 1996), pp. 15983.
Ross, J.B., A study of twelfth-century interest in the antiquities of Rome, in: J.L.
Cate and E.N. Anderson eds., Medieval and historiographical essays in honor of James
Westfall Thompson (Chicago, 1938), pp. 30221.
Rotthoff, G., Studien zur Geschichte des Reichsguts in Niederlothringen und Friesland whrend
der schsisch-salischen Kaiserzeit. Das Reichsgut in den heutigen Niederlanden, Belgien, Luxemburg
und Nordfrankreich (Bonn, 1953).
Rousseau, F., La Meuse et le Pays Mosan. Leur importance historique avant le XIII e sicle
(Namur, 1930).
Rousselle, A., Croire et gurir: La foi en Gaule dans lantiquit tardive (Paris, 1990).
Roux, J.-P., La religion des Turcs et des Mongols (Paris, 1984).
, Les explorateurs au Moyen Age (Paris, 1985).
Rovelli, A., La moneta nella documentazione altomedievale di Roma e del Lazio,
in: Paroli and P. Delogu eds., La storia economica, pp. 33352.
, La funzione della moneta tra lVIII e X secolo: Un annalisi della docu-
mentazione archeologica, in: R. Francovich and G. Noy eds., La Storia dellAlto
Medioevo italiano (VIX secolo) alla luce dellarcheologia. Biblioteca dellarcheologia
medievale 11 (Florence, 1994), pp. 52137.
Rowlands, M.,The archaeological interpretation of prehistoric metal working, World
Archaeology 3 (1971), pp. 21024.
, The cultural economy of sacred power, in: Les princes de la Protohistoire et
lmergence de ltat. Actes de la table ronde internationale organise par le Centre
Jean Brard et lcole francaise de Rome; Naples, 2729 octobre 1994 (Naples-
Rome 1999), pp. 16572.
Roymans, N., The sword and the plough. Regional dynamics in the Romanisation
of Belgic Gaul and the Rhineland area, in: idem ed., From the sword to the plough.
Three studies on the earliest Romanisation of northern Gaul (Amsterdam, 1996), pp. 9126.
Russo, D., Les reprsentations mariales dans lart doccident: Essai sur la forma-
tion dune tradition, in: D. Iogna-Prat, . Palazzo and D. Russo eds., Marie: Le
culte de la vierge dans la socit mdivale (Paris, 1996), pp. 176203.
Russo, E., Laffresco di Turtura nel cimiterio di Commodilla, licona di S. Maria
in Trastevere e le pi antiche feste della madonna a Roma, Bullettino dellistituto
storico italiano per il medioevo e archivio muratoriano 88 (1979), pp. 3785, 89; (1980/81),
pp. 71150.
 589

Sablerolles, Y., Made in Maastricht: glaskralen uit de Merovingische tijd, Vormen

uit vuur 155 (1995), pp. 1523.
Sage, W., Die Ausgrabungen in der Pfalz zu Ingelheim am Rhein, 196070,
Francia 4 (1976), pp. 14160.
Salin, ., La civilisation mrovingienne: daprs les spultures, les textes et le laboratoire 2 (Paris,
1952).
Salisbury, J.E., Church Fathers, independent Virgins (London, 1991).
Salzman, M., The Christianization of sacred time and sacred space, in: W.V.
Harris, The transformations of Vrbs Roma in late Antiquity. Journal of Roman archae-
ology, suppl. ser. 33 (Portsmouth, 1999), pp. 12334.
Samson, R., Social structures from Reihengrber: mirror or mirage?, Scottish
Archaeological Review 4 (1987), pp. 11626.
Santangeli Valenzani, R., Edilizia residenziale e aristocrazia urbana a Roma nel-
laltomedioevo, in: S. Gelichi ed., I Congresso Nazionale di Archeologia medievale
(Florence, 1997), pp. 6470.
Saradi-Mendolovici, H., The demise of the ancient city and the emergence of the
medieval city in the eastern Roman empire, Echos du monde classique 32 (1988),
pp. 365401.
Sauvaget, J., Le plan de Laodice-sur-Mer, Bulletin dtudes orientales 4 (1934), pp.
81114.
Saxe, A.A., Social dimensions of mortuary practices (unpublished Ph.D. thesis, University
of Michigan, Ann Arbor, 1970).
Saxer, V., Lutilisation par la liturgie de lespace urbain et suburbain: Lexemple
de Rome dans lantiquit et le haut moyen ge, Actes du XI e Congrs international
darchologie chrtienne. Studi di antichit cristiana 41 (Rome, 1989), pp. 9171033.
Scales, P., Cordoba under the Umayyads: a Syrian garden city?, in: G. Roe and
F. Verhaege eds., Urbanism in Medieval Europe (Zellik, 1997), pp. 17582.
Schalles-Fischer, M., Pfalz und Fiskus Frankfurt. Eine Untersuchung zur Verfassungsgeschichte
des frnkisch-deutschen Knigtums. Verffentlichungen der Max-Planck Instituts fr
Geschichte 30 (Gttingen, 1969).
Schama, S., Landscape and memory (London, 1995).
Scharlipp, W.-E., Die frhen Trken in Zentralasien (Darmstadt, 1992).
Schatz K., SJ, Knigliche Kirchenregierung und rmische Petrus-berlieferung im
Kreise Karls des Grossen, in: R. Berndt ed., Das frankfurter Konzil von 794: Kristal-
lisationspunkt karolingischer Kultur. Quellen und Abhandlungen zur mittelrheinischen
Kirchengeschichte 80 (Mainz, 1997), pp. 35771.
Scheibelreiter, G., Der Bischof in Merowingischer Zeit (Wien, 1983).
Schieffer, R., Die Karolinger (2nd edition, Stuttgart/Berlin/Kln,1997).
Schjdt, J.P., Horizontale und vertikale Achsen in der vorchristlichen skandi-
navischen Kosmologie, in: T. Ahlbck ed., Old Norse and Finnish Religions and Cultic
Place-Names (Stockholm, 1990), pp. 3557.
Schlter, W., Versuch einer sozialen Differenzierung der jungkaiserzeitlichen Krper-
grbergruppe von Haleben-Leuna anhand einer Analyse der Grabfunde, Neue
Ausgrabungen und Forschungen in Niedersachsen 6 (1970), pp. 11745.
Schmid, K., Mnchslisten und Klosterconvent von Fulda zur Zeit der Karolinger
in: K. Schmid ed., Die Klostergemeinschaft von Fulda, II/2, Mnstersche Mittelalter-
Schriften 8 (Mnchen, 1978), pp. 571639.
Schmitt, J.-C., Religion populaire et culture folklorique, Annales ESC 31 (1976),
pp. 94153.
, Les revenants. Les vivants et les morts dans la socit mdivale (Paris, 1994).
Schnapp, A., The discovery of the past: The origins of archaeology (London, 1996).
Schramm, P.E., Die Anerkennung Karls des Groen als Kaiser (bis 800), in:
idem, Kaiser, Knige, und Ppste: Gesammelte Aufstze zur Geschichte des Mittelalters 1
(Stuttgart, 1968), pp. 21563.
Schreiner, K., Verletzte Ehre, in: idem and G. Schwerhoff eds., Verletzte Ehre.
590

Ehrkonflikte in Gesellschaften des Mittelalters und der frhen Neuzeit (Norm und Struktur)
(Kln/Weimar/Wien, 1995), pp. 128.
Schulze, M., Diskussionsbeitrag zur Interpretations frh- und hochmittelalterlicher
Flufunde, Frhmittelalterliche Studien 18 (1984), pp. 22248.
Schlke, A., Zeugnisse der Christianisierung im Grabbefund? Eine Forschungs-
geschichte mit Ausblick, Ethnographisch-Archologische Zeitschrift 38 (1997), pp. 45768.
, On Christianization and grave-finds, European Journal of Archaeology 2:1 (1999),
pp. 77106.
Schwind, F., Zu karolingerzeitlichen Klstern als Wirtschaftsorganismen und Sttten
handwerklicher Ttigkeit, in: L. Fenske e.a. eds., Institutionen, Gesellschaft und Kultur
im Mittelalter: Festschrift J. Fleckenstein (Sigmaringen, 1984), pp. 10123.
Sedlmayr, H., Die Entstehung der Kathedral (Zrich, 1950).
Semmler, J., Episcopi potestas und karolingische Klosterpolitik, in: A. Borst ed.,
Mnchtum, Episkopat und Adel zur Grndungszeit des Klosters Reichenau. Vortge und
Forschungen 20 (Sigmaringen, 1974), pp. 30594.
, Die Geschichte der Abtei Lorsch von der Grndung bis zum Ende der
Salierzeit, 764 bis 1125, in F. Knpp ed., Die Reichsabtei Lorsch vol. I (Darmstadt,
1977), pp. 75173
, Mission und Pfarrorganisation in den rheinischen, Mosel- und Maaslandischen
Bistmer (5.-10. Jahrhundert), in: Cristianizzazione e organizzazione ecclesiastica delle
campagne nellalto medioevo e resistenze, Settimane 28 (Spoleto, 1982), pp. 81388.
Semple, S., A fear of the past: The place of the prehistoric burial mound in the
ideology of middle and later Anglo-Saxon England, World Archaeology 30 (1998),
pp. 10926.
Serafini, C., Le monete e le bulle plombee pontificie del medagliere vaticano 1. Repr. (Bologna,
1965).
Serstevens, A. t., ed., Les prcurseurs de Marco Polo (Paris, 1959).
Shepard, J. and S. Franklin eds., Byzantine diplomacy (Aldershot, 1992).
Shephard, J., The social identity of the individual in isolated barrows and barrow
cemeteries in Anglo-Saxon England, in: B.C. Burnham and J. Kingsbury eds.,
Space, hierarchy and society. BAR International Series 59 (Oxford, 1979), pp. 4779.
Simek, R., Dictionary of Northern Mythology (Cambridge, 1993).
Simonet, F., Historia de los Mozrabes de Espaa (Madrid, 1897, repr. 1984).
Simonnet, N., La Fondation du Mont-Saint-Michel daprs la Revelatio ecclesiae sancti
Michaelis, Annales de Bretagne et des Pays de lOuest 106 (1999), pp. 723.
Sinor, D., The establishment and dissolution of the Trk Empire, in: D. Sinor
ed., The Cambridge History of Early Inner Asia (Cambridge, 1990), pp. 285316.
, Studies in medieval Inner Asia (Aldershot, 1997).
, Le Mongol vue par lOccident, in: idem, Studies in medieval Inner Asia
(Aldershot, 1997), nr. IX.
, The making of a Great Khan, in: idem, Studies in medieval Inner Asia (Aldershot,
1997), nr. XIV.
, Diplomatic practices in medieval Inner Asia, in: idem, Studies in medieval
Inner Asia, nr. XVI.
Sippel, K., Die Kenntnis vorgeschichtlicher Hgelgrber im Mittelalter, Germania
58 (1980), pp. 13746.
Smith, J.M.H., Culte imprial et politique frontalire dans la valle de la Vilaine: le
tmoignage des diplmes carolingiens dans le cartulaire de Redon in: M. Simon
ed., Landvennec et le monachisme breton dans le haut moyen ge (Landvennec, 1985),
pp. 12939.
, Oral and written: saints, miracles and relics in Brittany, c. 8501250,
Speculum 65 (1990), pp. 30943.
, Saints, miracles and relics in Brittany, Speculum 65 (1990), pp. 30943.
, Province and empire: Brittany and the Carolingians (Cambridge, 1992).
 591

, Religion and lay society, in: R. McKitterick ed., The new Cambridge medieval
history vol. II (Cambridge, 1995), pp. 65478.
, The problem of female sanctity in Carolingian Europe, c. 780920, Past
and Present 146 (1995), pp. 337.
, ed., Early medieval Rome and the Christian West: Essays in Honour of Donald A.
Bullough (Leiden/Boston/Kln, 2000).
, Old saints, new cults: Roman relics in Carolingian Francia, in: eadem ed.,
Early medieval Rome and the Christian West: Essays in Honour of Donald A. Bullough
(Leiden/Boston/Kln, 2000), pp. 31739.
, Confronting identities: the rhetoric and reality of a Carolingian frontier in:
W. Pohl and M. Diesenberger eds., Integration und Herrschaft. Ethnische Identitten und
soziale Organisation im Frhmittelalter, Forschungen zur Geschichte des Mittelalters 3
(Wien, 2001; forthcoming).
Smith, J.Z., To take place: Toward theory in ritual (Chicago, 1987).
Smythe, D., Why do barbarians stand around the emperor at diplomatic recep-
tions?, in: J. Shepard and S. Franklin eds., Byzantine Diplomacy (Aldershot, 1992),
pp. 30512.
Solli, B., Odinthe queer? Om det skreive i norr_n mytologi, Universitetets
Oldsaksamling, rbok 1997/1998 (Oslo, 1998), pp. 742.
Srensen, P. and J. Kousgrd, Gudhem, Frhmittelalterliche Studien 19 (1985), pp.
13138.
Sot, M., Gesta abbatum, gesta episcoporum, Typologie des sources du moyen ge occi-
dental 37 (Turnhout, 1981).
Sotomayor, M., Penetracin de la Iglesia en los medios rurales de la Espaa tar-
dorromana y visigoda, in: Cristianizzazione ed organizzazione ecclesiastica delle cam-
pagne nellAlto Medioevo, Settimane 28 (Spoleto, 1982), pp. 1753.
Sprigade, K., Die Einweisung ins Kloster und in den geistlichen Stand als politische Massnahme
im frhen Mittelalter (Heidelberg, 1964).
Squatriti, P., Water and society in early medieval Italy, 4001000 (Cambridge, 1998).
Staab, F., Untersuchungen zur Gesellschaft am Mittelrhein in der Karolingerzeit, Geschichtliche
Landeskunde 11 (Wiesbaden, 1975).
, Wann wurde Hrabanus Maurus Mnch in Fulda? Beobachtungen zur An-
teilnahme seiner Familie an den Anfngen seiner Laufbahn, in: R. Kottje and
H. Zimmermann, eds., Hrabanus Maurus: Lehrer, Abt und Bischof (Wiesbaden, 1982),
pp. 75101.
, Die Knigin Fastrada, in: R. Berndt ed., Das Frankfurter Konzil von 794:
Kristallisationspunkt karolingische Kultur I (2 vols., Mainz, 1997), pp. 183217.
Staecker, J., Searching for the unknown: Gotlands churchyards from a gender
and missionary perspective, Lund Archaeological Review 2 (1996), pp. 6386.
Stafford, P., Queens, nunneries and reforming churchmen: gender, religious status
and reform in tenth- and eleventh-century England, Past and Present 163 (1999),
pp. 335.
Stancliffe, C., Kings who opted out, in: P. Wormald (with D. Bullough and
R. Collins) ed., Ideal and reality in Frankish and Anglo-Saxon Society. Studies presented to
J.M. Wallace-Hadrill (Oxford 1983), pp. 15476.
Starume, E., Smeden i jernalderen, bofastikke bofast, h_y eler lav status,
Universitetets Oldsaksamling rbok 1984/85 (Oslo, 1986), pp. 4558.
Steinsland, G., Det Hellige Bryllup og Norr_n Kongeideologi (Oslo, 1991).
Steuer, H., Frhgeschichtliche Sozialstruckturen in Mitteleuropa: Eine Analyse der Auswertungs-
methoden des archologischen Quellenmaterials. Abhandlungen der Akademie der Wissen-
schaften in Gttingen, philologisch-historische Klasse, dritte Folge 128 (Gttingen,
1982).
, Archaeology and history: proposals on the social structure of the Merovingian
Empire, in: K. Randsborg ed., The Birth of Europe: Archaeology and Social Development
592

in the First Millenium A.D. Analecta Romana Instituti Danici, Supplementum XVI
(Rome, 1989), pp. 10022.
, Die Alamannen auf dem Zhringer Burgberg (Stuttgart, 1990).
, Herrschaft von der Hhe, Die Alamannen. Archologisches Landesmuseum
Baden-Wrttemberg, Ausstellungskatalog (Stuttgart, 1997), pp. 14963.
Stiene, H.E., Gregors des Grossen Dialogi und die vita Goaris Wandelberts von
Prm, Mittellateinisches Jahrbuch 18 (1983), pp. 5162.
Stoclet, A., Fulrad de Saint-Denis (v. 710784), abb et archiprtre de monastres
exempts, Le Moyen Age 88 (1982), pp. 20535.
Stoodley, N., The spindle and the spear: A critical enquiry into the construction and meaning
of gender in the early Anglo-Saxon burial rite. British Archaeological Reports 288 (Oxford,
1999).
Straeten J. van der, Les actes des martyrs dAurlien en Bourgogne: tude lit-
traire, Analecta Bollandiana 79 (1961), pp. 11544.
, Les actes des martyrs dAurlien en Bourgogne: le texte de Farfa, Analecta
Bollandiana 79 (1961), pp. 44768.
Straw, C., Gregory the Great: Perfection in imperfection (Berkeley, 1988).
Stroll, M., Maria Regina: Papal Symbol in: A.J. Duggan ed., Queens and Queenship
in Medieval Europe (London, 1997), pp. 173203.
Strmbck, D., Sejd. Textstudier i nordisk religionshistoria (Stockholm/Copenhagen, 1935).
Tarlow, S., An archaeology of remembering: death, bereavement and the First
World War, Cambridge Archaeological Journal 7, 1 (1997), pp. 10521.
, Bereavement and commemoration: an archaeology of mortality. Social Archaeology Series
(Oxford, 1999).
Tchalenko, G., Villages antiques de la Syrie du nord I. Institut franais darchologie de
Beyrouth, Bibliothque archologique et historique 50 (Paris, 1953), pp. 20577.
Thte, E., Alte Denkmler und frhgeschichtliche Bestattungen: ein schsisch-angel-
schsischer Totenbrauch und seine Kontinuitt, Archologische Informationen 19
(1996), pp. 10516.
Theurillat, J.-M., LAbbaye de St-Maurice dAgaune des origines la rforme canoniale,
515830 environ = Vallesia 9 (1954).
Theuws, F., Centre and periphery in northern Austrasia, in: J. Besteman, J. Bos
and H.A. Heidinga eds., Medieval archaeology in the Netherlands. Studies presented to
H.H. van Regteren Altena (Assen, 1990), pp. 4169.
, Landed property and manorial organisation in Northern Austrasia: some
considerations and a case study, in: N. Roymans and F. Theuws eds., Images of
the past. Studies on ancient societies in Northwestern Europe (Amsterdam, 1991), pp. 299407.
, and H. Hiddink, Der Kontakt zu Rom, Die Franken, Wegbereiter Europas. Knig
Chlodwich und seine Erben (Mainz, 1996), pp. 6680.
, Frankish transformations. Actions and thoughts of aristocrats and dwellers in the pagus
Texandri (in prep.).
, Centres commerciaux dans la valle de la Meuse et lconomie du haut
Moyen Age.Vers une rpresentation conceptuelle des centres du haut Moyen
Age, forthcoming.
, Introduction: Ritual in transforming societies, in F. Theuws and J.L. Nelson
eds., Rituals of power from Late Antiquity to the Early Middle Ages. TRW 8
(Leiden/Boston/Cologne, 2000), pp. 113.
Theuws F. and M. Alkemade, A kind of mirror for men: sword depositions in
Late Antique northern Gaul, in: F. Theuws and J.L. nelson eds., Rituals of power
from Late Antiquity to the Early Middle Ages. TRW 8 (Leiden/Boston/Cologne, 2000),
pp. 40176.
Thomas, J., Time, culture and identity. An interpretive archaeology (London/New York,
1999 [1996]).
Thompson, E.A., Britonia, in: M.W. Barley and R.P.C. Hanson eds., Christianity
in Britain, 300700 (Leicester, 1968), pp. 2015.
 593

, The Goths in Spain (Oxford, 1969).

, The Huns (Oxford, 2nd edition. 1996).
Thomsen, P.O., B. Blsild, N. Hardt and K.K. Michaelsen, Lundeborgen handel-
splads fra jernalderen (Svendborg, 1993).
Thomsen, P.O., Lundeborgan early port of trade in South-East Funen, in: P.O.
Nielsen e.a. eds., The Archaeology of Gudme and Lundeborg (Copenhagen, 1994), pp.
239.
Thomsen, V., Alttrkische Inschriften in der Mongolei, Zeitschrift der Deutschen
Morgenlndischen Gesellschaft 78 (1924/25), pp. 12175.
Thrane, H., Das Gudme-Problem und die Gudme-Untersuchungen, Frhmittelalterliche
Studien 21 (1987), pp. 148.
, Materialien zur Topografhie einer eisenzeitlichen Sakrallandschaft um Gudme
auf Ostfnen in Dnemark, in: A. Wesse ed., Studien zur Archologie des Ostseeraumes.
Von der Eisenzeit zum Mittelalter. Festschrift fr Michael Mller-Wille (Neumnster, 1998),
pp. 23547.
, Gudme, Reallexikon der Germanischen Altertumskunde 13 (Berlin, 1999), pp. 1428.
Thurre, D.,Culte et iconographie de saint Maurice dAgaune: bilan jusquau XIIIe
sicle, Zeitschrift fr schweizerische Archaeologie und Kunstgeschichte/Revue suisse dart et
darchaeologie/Rivista svizzera darte e darcheologia 49 (1992), pp. 718.
Tilley, C., A phenomenology of landscape: places, paths and monuments (Oxford, 1994).
, The powers of rock: topography and monument construction on Bodmin
Moor, World Archaeology 28:2 (1996), pp. 16177.
Times Atlas of World History (3rd edn., London, 1989).
Timpe, D., Entdeckungsgeschichte, Reallexikon der Germanische Altertumskunde 7, 2nd
edition. (Berlin/New York, 1989), pp. 307389.
Tonnerre, N.-Y., Les pays de la Basse Vilaine au haut Moyen ge, Mmoires de
la Socit dHistoire et dArchologie de Bretagne 63 (1986), pp. 2972.
, Celtic literary tradition and the development of a feudal principality in
Brittany in: H. Pryce ed., Literacy in medieval Celtic societies (Cambridge, 1998), pp.
16682.
Tovar, A., Un obispo con nombre britnico y los orgenes de la dicesis de
Mondoedo, Habis 3 (1972), pp. 1558.
Toynbee, J.M.C., Death and burial in the Roman world. Aspects of Greek and Roman
Life (London, 1971).
Trousset, P., La carte dAgrippa: nouvelle proposition de lecture, Dialogues dHistoire
Ancienne19, 2 (1993), pp. 13757.
Turbessi, G., Regole monastiche antiche (Rome, 1974).
Turner, V., The ritual process: structure and anti-structure (1969; repr. New York, 1995).
, Dramas, fields, and metaphors: symbolic action in human society (Ithaca, N.Y, 1974).
Turri, E., Gli uomini delle tende (Milan, 1983).
Turville-Petre, E.O.G., Myth and religion of the North (Westport, 1975).
Ullrich, H., Totenriten und Bestattung im Palolithikum, in: F. Horst and H. Kei-
ling eds., Bestattungswesen und Totenkult in ur- und frhgeschichtlicher Zeit (Berlin, 1991),
pp. 2334.
Untermann, M., Opere mirabile constructa. Die Aachener Residenz Karls des Gros-
sen, in: 799. Karl der Grosse und Papst Leo III in Paderborn. Kunst und Kultur der Karo-
lingerzeit, Beitrge zum Katalog der Ausstellung (Paderborn, 1999), pp. 15264.
Urbanczyk, P., Geneza wczesno redniowiecznych metalowych pochew broni bia ej
ze stanowisk kultury pruskiej, Przegl[d Archeologiczny 26 (1978), pp. 10745.
, St. Adalbert-Voitechmissionary and politician, in: idem ed., Early Christianity
in Central and East Europe (Warsaw, 1997), pp. 15562.
, The Goths in Polandwhere did they come from and when did they leave?,
European Journal of Archaeology 1, 3 (1998), pp. 397415.
Uytfanghe, M. van, Stylisation biblique et condition humaine dans lhagiographie mrovingi-
enne (650750). Verhandelingen van de Koninklijke Academie voor Wetenschappen,
594

Letteren en Schone Kunsten van Belgi, Klasse der Letteren 49, 120 (Brussel,
1987).
Vallv Bermejo, J., De nuevo sobre Bobastro, Al-Andalus 30 (1965), pp. 13974.
, Una fuente importante de la historia de Al-Andalus. La Historia de Ibn
Askar, Al-Andalus 31 (1966), pp. 23765.
Van Dam, R., Leadership and community in late antique Gaul (Berkeley, 1985).
, Saints and their miracles in Late Antique Gaul (Princeton, 1993).
, Bodily miracles, in: idem, Saints and their miracles in late antique Gaul (Princeton,
1993), pp. 82115.
Vanderhoeven, A., The earliest urbanisation in northern Gaul. Some implications
of recent research in Tongres, in: N. Roymans ed., From the sword to the plough.
Three studies on the earliest Romanisation of northern Gaul (Amsterdam, 1996), pp.
189260.
Vang Petersen, P., Excavations at sites of treasure trove finds at Gudme, in: P.O.
Nielsen e.a. eds., The archaeology of Gudme and Lundeborg (Copenhagen, 1994), pp.
3040.
Vanvinckenroye, W., Tongeren Romeinse stad (Tielt, 1985).
Veit, U., Studien zum Problem der Siedlungsbestattung im europischen Neolithikum. Tbinger
Schriften zur ur- und frhgeschichtlichen Archologie 1 (Mnster/New York, 1996).
Venarde, B.L., Womens monasticism in the central Middle Ages (Itahca, 1997).
Verdier, R., Le systme vidicatoire, in: idem ed., La vengeance. Etudes dethnologie,
dhistoire et de philosophie vol. 1 (Paris, 1980), pp. 1342.
Verhulst, A., The rise of cities in North-West Europe (Cambridge, 1999).
Verzone, P., La distruzione dei palazzi imperiali di Roma e la restrutturazione del
palazzo lateranense nel IX secolo nel raporti con quello di Costantinopoli, in:
Roma e let carolingia. Atti del giornati di studio 38 maggio 1976 (Rome, 1976), pp.
407.
Vestergaard, E., Vlsunge-Nibelungen Traditionen. Antropologiske studier i en episk traditions
transformation i forhold til dens sociale sammenhng Unpublished Ph.D. thesis, University
of rhus (1992).
Vieillard-Troiekouroff, M., Les monuments religieux de la Gaule daprs les oeuvres de Grgoire
de Tours (Paris, 1976), pp. 85102.
Vin, J.P.A. van der, and T.A.S.M. Panhuysen, Romeinse en vroegmiddeleeuwse
munten uit een stadskernopgraving in Maastricht, De Beeldenaar 4 (1983), pp. 1215.
Vogel, C., La discipline pnitentielle en Gaule des origines la fin du VII e sicle (Paris, 1952).
Vogt, A. ed., Constantin VII Porphyrognte, Le Livre des Crmonies, 2nd edition. 2 vols.
(Paris, 1967), with Commentaire, 2nd edition. 2 vols (Paris, 1967).
Vollman, B., Studien zum Priszillianismus. Die Forschung, die Quellen, der fnfzehnte Brief
Papst Leos des Grossen (St. Ottilien, 1965).
Vollmer, F., Die Etichonen. Ein Beitrag zur Frage der Kontinuitt frher Adels-
familien, in: G. Tellenbach ed., Studien und Vorarbeiten zur Geschichte des grossfrnki-
schen und frhdeutschen Adels (Freiburg, 1957), pp. 13784.
Vries, J. de, Altgermanische Religionsgeschichte (Berlin, 1970).
Wallace-Hadrill, J.M., Gregory of Tours in the light of recent research, in: idem,
The long-haired kings (London, 1962), pp. 4970.
Walter, C., Papal Political Imagery in the Medieval Lateran Palace, Cahiers
Archologiques 20 (1970).
Walter, T., Secular funerals, Theology 92 (1989), pp. 394402.
, Funeralsand how to improve them (London, 1990).
, On bereavement: The culture of grief. Facing Death Series (Buckingham/Philadelphia,
1999).
, ed., The mourning for Diana (Oxford, 1999).
Wamers, E., Die frhmittelalterlichen Lesefunde aus der Lhrstrae (Baustelle Hilton II) in
Mainz (Mainz, 1994).
 595

Ward-Perkins, B., From classical Antiquity to the Middle Ages: urban public building in
Northern and Central Italy A.D. 300850 (Oxford, 1984).
, Re-using the architectural legacy of the past, entre idologie et pragma-
tisme, in: G.P. Brogiolo and B. Ward-Perkins eds., The idea and ideal of the town
between late Antiquity and the early Middle Ages, TRW 4 (Leiden/Boston/Kln, 1999),
pp. 22544.
, Urban Continuity, in: N. Christy and S.T. Loseby eds., Towns in Transition.
Urban evolution in Late Antiquity and the Early Middle Ages (Aldershot, 1996), pp. 417.
Wasserstein, J.D., The rise and fall of the party kings (Princeton, 1985).
Wehlt, H.-P., Reichsabtei und Knig. Dargestellt am Beispiel der Abtei Lorsch mit Ausblicken
auf Hersfeld, Stablo und Fulda. Verffentlichungen des Max-Planck Instituts fr
Geschichte 28 (Gttingen, 1968).
Weidemann, K., Die Topographie von Mainz in der Rmerzeit und der frhen
Mittelalter, Jahrbuch des Rmisch-Germanisches Zentralmuseums Mainz 15 (1968), pp.
14699.
Weiers, M., Westliche Boten und Reisende zu den Mongolen im 13. und 14.
Jahrhundert, in: A. Eggebrecht ed., Die Mongolen und ihr Weltreich (Mainz s.a.),
pp. 18595.
Weiner, A., Inalienable possessions. The paradox of keeping-while-giving (Berkeley 1992).
Wemple, S.F., Women in Frankish society. Marriage and the cloister, 500 to 900 (Philadelphia,
1981).
Werner, J., Frankish royal tombs in the cathedrals of Cologne and Saint-Denis,
Antiquity 38 (1964), pp. 20116.
Werner, K.-F., Missusmarchiocomes, in: W. Paravicini and K.F. Werner eds.,
Histoire compare de ladministration. Beiheft der Francia 9 (1980), pp. 191239.
Werner, M., Der Ltticher Raum in frhkarolingischer Zeit. Untersuchungen zur Geschichte
einer karolingischen Stammlandschaft (Gttingen, 1980).
, Adelsfamilien im Umkreis der frhen Karolinger. Die Verwandtschaft Irminas von Oeren
und Adela von Pfalzel (Sigmaringen, 1982).
Westermann-Angerhausen, H., Spolie und Umfeld in Egberts Trier, Zeitschrift fr
Kunstgeschichte 50 (1987), pp. 30536.
Whatley, E.G., An early literary quotation from the Inventio S. Crucis: A Note on
Baudonivias Vita S. Radegundis (BHL 7049), Analecta Bollandiana 111 (1993), pp.
8191.
White, R., Roman and Celtic objects from Anglo-Saxon graves: A catalogue and interpretation
of their use. BAR British Series 191 (Oxford, 1988).
, Scrap or substitute: Roman material in Anglo-Saxon graves, in: E. Southworth,
Anglo-Saxon cemeteries: A reappraisal. Proceedings of a Conference held at Liverpool Museum
1986 (Phoenix Mill, 1990), pp. 12552.
Wibl, F. ed., Chronique des dcouvertes archologiques dans le canton du Valais
en 1995, Vallesia 51 (1996).
Wickham, C., European forests in the early Middle Ages: landscape and land clear-
ance, in: Lambiente vegetale nellalto medievo, Settimane 37 (Spoleto, 1989), pp.
479548; repr. in: Wickham, Land and Power: Studies in Italian and European Social
History, 4001200 (London, 1994), pp. 155200.
, Rural society in Carolingian Europe, in: R. McKitterick ed., New Cambridge
Medieval History II (Cambridge, 1997), pp. 51037.
, Aristocratic power in eighth-century Lombard Italy, in A.C. Murray ed.,
After Romes fall: Narrators and sources of early medieval History (Toronto, 1998), pp. 15270.
Wiedgren, M., Kulturgeografernas bnder och arkeologernas guldfinns det ngon
vg til syntes?, in: L. Larsson and B. Hrdh eds., Centrala Platser, Cantrala Frgor.
Acta Archaeologica Lundensia, Ser. In 8, no. 28 (Lund, 1998), pp. 28196.
Wielowiejski, P., Bernstein in der Przeworsk-Kultur, Bericht der Rmisch-Germanischen
Kommission 77 (1996), pp. 215342.
596

Wightman, E.M., Gallia Belgica (London, 1985).

Williams, D.H., Ambrose of Milan and the end of the Nicene-Arian conflicts (Oxford, 1995).
Williams, H., Ancient landscapes and the dead: the reuse of prehistoric and Roman
monuments as early Anglo-Saxon burial sites, Medieval Archaeology 41 (1997), pp.
132.
, The ancient monument in Romano-British ritual practices, in: C. Forcey,
J. Hawthorne and R. Witcher eds., TRAC 97: Proceedings of the Seventh Annual
Theoretical Roman Archaeology Conference (Oxford, 1998), pp. 7186.
, Monuments and the past in early Anglo-Saxon England, World Archaeology
30,1 (1998), pp. 90108.
, Placing the dead: investigating the location of wealthy barrow burials in sev-
enth century England, in: M. Rundkvist ed., Grave matters: eight studies of First
Millennium A.D. burials in Crimea, England and southern Scandinavia. BAR International
Series 781, (Oxford, 1999), pp. 5786.
Willis, G.G., Further essays in early Roman liturgy (London, 1968).
Wilpert, J., La proclamazione Efesina e i musaici della basilica di S. Maria
Maggiore, Analecta Sacra Tarraconensia 7 (1931), pp. 197213.
Winskill, H., An analysis of Anglo-Saxon deviant burials (unpubl. B.A. thesis, University
of Reading, 1999).
Winter, J.C., Sites of memory, sites of mourning: The Great War in European cultural history.
Studies in the Social and Cultural History of Modern Warfare (Cambridge, 1995).
Wirth, G., AttilaDas Hunnenreich und Europa (Stuttgart/Berlin/Kln, 1999).
Wisskirchen, R., Santa Maria in Domnica: berlegungen zur frhesten apsidialen
darstellung der Thronenden Maria in Rom, Aachener Kunstbltter 61 (1995/96),
pp. 38195.
Wittfogel, K.A., China und die osteurasische Kavallerie-Revolution, Ural-Altaische
Jahrbcher 49 (1977), pp. 12140.
Wolf, G., Salus Populi Romani: Die Geschichte rmischer Kultbilder im Mittelalter (Weinheim,
1990).
Wolfram, H., Die Goten. Von den Anfngen bis zur Mitte des sechsten Jahrhunderts. Entwurf
einer historischen Ethnographie, 3rd edition. (Mnchen, 1990).
, Salzburg, Bayern, sterreich. Die Conversio Bagoariorum et Carantanorum und die Quellen
ihrer Zeit, MIG Erg. Bd. 31 (Wien, 1995).
Wood, I.N., Early Merovingian devotion in town and country, in: D. Baker, ed.,
The Church in town and countryside. Studies in Church History 6 (Oxford, 1979), pp.
7475.
, Avitus of Vienne: Religion and culture in the Auvergne and the Rhne Valley, 470530
(Ph.D. diss., Oxford, 1980).
, A prelude to Columbanus: the monastic achievement in the Burgundian ter-
ritories, in: H.B. Clarke and M. Brennan, eds., Columbanus and Merovingian monas-
ticism. BAR International Series 113 (Oxford, 1981), pp. 332.
, The Vita Columbani and Merovingian hagiography, Peritia 1 (1982), pp.
6380.
, The ecclesiastical politics of Merovingian Clermont, in: P. Wormald, ed.,
Ideal and Reality in Frankish and Anglo-Saxon Society (Oxford, 1983), pp. 3457.
, Clermont and Burgundy: 511534, Nottingham Medieval Studies 32 (1988), pp.
11925.
, Forgery in Merovingian Hagiography, in: Flschungen im Mittelalter. Internationaler
Kongress der MGH, Mnchen, 16.19. September 1986 5: Fingierte Briefe, Frmmigkeit und
Flschung, Realienflschungen. MGH Schriften 33.V (Hannover, 1988), pp. 36984.
, Administration, law and culture in Merovingian Gaul in: R. McKitterick
ed., The uses of literacy in early Medieval Europe (Cambridge, 1990), pp. 6381.
, The Code in Merovingian Gaul, in: J. Harries and I. Wood eds., The
Theodosian Code (Ithaca, 1993), pp. 16177.
 597

, Gregory of Tours (Bangor, 1994).

, The Merovingian Kingdoms 450751 (London/New York, 1994).
, Spultures ecclsiastiques et snatoriales dans la valle du Rhne (400600),
Mdivales 31 (1996), pp. 1327.
, How popular was early medieval devotion? in: A.J. Frantzen and T.N. Hill
eds., Popular piety: Prayer, devotion and cult. Essays in Medieval Studies, Proceedings
of the Illinois Medieval Association 14 (1997).
, Report: The European Science Foundations programme on the transfor-
mation of early medieval Europe, EME 6 (1997), pp. 21727.
, Aethicus Ister: an exercise in difference, in: W. Pohl and H. Reimitz eds.,
Grenze und Differenz im frhen Mittelalter. Forschungen zur Geschichte des Mittelalters
1 (Wien, 2000), pp. 197208.
, Constructing cults in early medieval France: saints and churches in Burgundy
and the Auvergne 4001000 (forthcoming).
Wormald, P., The emergence of Anglo-Saxon kingdoms, in: L.M. Smith ed., The
making of Britain: the Dark Ages (Basingstoke and London, 1984), pp. 4962.
, Anglo-Saxon society and literature, in: M. Godden and M. Lapidge eds.,
The Cambridge Companion to Old English Literature (Cambridge, 1991), pp. 122.
, The making of English Law (Oxford, 1999).
Wright, T., On Antiquarian excavations and researches in the Middle Ages,
Archaeologia 30 (1844), pp. 43857.
Ximnez de Rada, R., Historia Arabum (Valencia, 1968).
Young, B., Paganisme, christianisation et rites funraires mrovingiens, Archologie
Mdivale 7 (1977), pp. 581.
, Que restait-il de lancien paysage religieux lpoque de Grgoire de Tours?,
in: N. Gauthier and H. Galini eds., Grgoire de Tours et lespace Gaulois: Actes du
congrs international, Tours, 35 novembre 1994. 13e supplment la Revue archologique
du Centre de la France (Tours, 1997), pp. 24150.
, Sacred topography and early Christian churches in late antique Gaul, in:
R.F.J. Jones, J.H.F. Bloemers, S.L. Dyson and M. Biddle eds., First Millenium
Papers: Western Europe in the First Millenium A.D. BAR International Series 401
(Oxford, 1988), pp. 21940.
Zabehlicky, H., Kriegs- oder Klimafolgen in archologischen Befunden?, in:
H. Friesinger, J. Tejral and A. Stupner eds., Markomannenkriege. Ursachen und Wirkungen
(Brno, 1994), pp. 4639.
Zachrisson, T., Grd, Grns, Gravflt. Sammanhang kring delmetalldeper och runstenar frn
vikingetid och tidigmedeltid in Uppland och Gstrikland. Stockholm Studies in Archaeology
15. (Stockholm, 1998).
Zemon Davis, N., The sacred and the body social in sixteenth-century Lyon, Past
and Present 90 (1981), pp. 4070.
Zoll-Adamikova, H., Die Einfhrung der Krperbestattung bei den Slawen an der
Ostsee, Archologisches Korrespondenzblatt 24 (1994), pp. 8193.
Zufferey, M., Die abtei Saint-Maurice dAgaune im Hochmittelalter (8301258), Verffent-
lichungen des Max-Planck-Instituts fr Geschichte 88 (Gttingen, 1988).
This page intentionally left blank
 INDEX
Aachen 2068, 217ff.
as a centre 432, 538
as an urban site 2256, 233
as palatium publicum 218
assemblies at 2189
baths at 2179, 2356, 539
bishops at 539
buildings at 2202
Charlemagne at 217ff., 539
church at 2202
decline of 234
Einhard on 217, 224, 236
elephant at 542
Notker Balbulus of St Gall on
2236
palace at 420, 539
pope Leo III at 538
rituals at 2334
ruins at 218
strategic importance of 2223
abbeys see: monasteries
Abd al-Ra m n III 119, 123, 129
Adalard of Corbie 2269
at the court 22931
Addidamentum Nivialense de Fuliano 247
Admonitio Generalis 2189
Ado 252. See: Faronids
aedificatio 3623, 3712, 3769, 383,
543
Al-Andalus 119ff.
Alcuin, on Tours 138, 147
Almod, abbot of Redon 3756
Alpaida, wife of Pippin II 1902
Alvarus 1234
Amandus, bishop of
Tongres-Maastricht 172, 180, 243,
247
amber route 511, 5135, 517, 522, 542
Ambrose of Milan 106
Ammianus Marcellinus 104
Andelys-sur-Seine, monastery of 255
Andennes, monastery of 1856
Annales Mettenses priores 3212
Annales Petaviani 324
Annales regni Francorum 2189, 322, 326
Anonymous chronicle of Abd Ra m n
al-Na ir 1289
Anstrud, abbess of St Jean (Laon)
25762. See: Baldwin, Vita Anstrudis
Appadurai, A. 2024
aquaducts 50
Aravatius: See: Servatius
Arians 37
Asgard, Home of the Gods 498505
assemblies, at Aachen 2189. See:
councils
asylum 302, 306
Attelanus, bishop of Laon 249
Attila 523
court of 4435, 4601, 463
Priscus on 4425
Aula Leonina 6870
Austrechildis, queen 281
Avar khagans, court of 4423, 445
Avitus, bishop of Vienne 151, 274,
279
Baldovin, J. 84
Baldwin, brother of Anstrud 258
Balthild, queen 95, 254, 3167
See: Vita Balthildis
banquets 4445
barbarian royal residences
diplomats at 442ff.
drinking rituals at 4434, 448, 451,
453, 455
image of 43940
of Attila 4434, 4601, 463
of Avar khagans 4423, 445
of Frisians 441
of Marcomanni 441
of Thuringians 441
of Turkish khagans 4423, 445, 447
women at 445, 448, 4535, 459
See: hall, palaces
baths at Aachen 2179, 2356, 539
Bede 255
Begga 249, 268
Benedict of Aniane. See: Codex regularum
Benedict. See: Regula Benedicti
Benignus, cult of at Dijon 1425
Beowulf 439, 457, 492, 497, 544
Bernard, king of Italy 227, 231, 291
Bibistrense, city of 127, 129, 133
600
Bishko, Ch.J. 339, 357
bishops
and ruins 1023, 117
and saints cults 137ff.
and towns 137ff., 534
at Aachen 539
episcopal see: 160ff., 175186
graves of 1724
of Tours 1401
power of 173ff., 275ff.
See: exemptions
See: Amandus of Tongres-Maastricht,
Attelanus of Laon, Avitus of
Vienne, Burgundofaro of Meaux,
Caesarius of Arles, Chagnoald of
Laon, Chrodegang of Metz,
Contumeliosus of Riez, Ebo of
Rheims, Eucherius of Lyon,
Fructuosus of Braga, Ghaerbald
of Lige, Gregory of Langres,
Gregory of Nazianzus, Gregory
of Tours, Hincmar of Rheims,
Hubert of Lige, Lambert of
Lige, Leudegar of Autun,
Leudegisus of Braga, Mamertus
of Vienne, Martin of Braga,
Monulphus of Tongres-Maastricht,
Potamio of Braga, Sagittarius of
Gap, Salonius of Embrun,
Servatius of Tongres-Maastricht
blinding 291, 296
Bloch, Maurice 202, 205
Blondel, L. 2759
Bonnet, C. 278
Book of Ceremonies 39, 42
Bosch Gajano, S. 267
Bourdieu, P. 246
Bradley, R. 20
Brenk, B. 106
Brink, S. 480
Brioude, cult of Julian 1425
Brittany 36196
Brown, P. 137
Brunhild, queen 30715
Burgundofara, abbess of Faremoutiers
250. See: Chagneric
Burgundofaro, bishop of Meaux
2503, 2823. See: Chagneric
burial, secondary 112
Busch, J.W. 2946
Byzantine Empire
and Rome 60, 66, 71, 73, 767, 80
as a state 3
Caesarius, bishop of Arles 152, 301
and ruins 1178
Capitulare de disciplina palatii
Aquisgransensis 2389
capitularies 2256
Carloman, king
conversion of 313, 323ff.
Carolingian government, mechanisms
of 400ff.
cartularies 3735
of Redon 3745
Carver, M. 1989, 214
Cassiodorus 513, 522
castle-building, royal control over 434
See: fortifications
cemeteries 9ff.
and graves of ancestors 23
and mental topographies 178
and power 1928
as disposal areas 11
as places of memory 112
as places of ritual 134, 247, 29
control over 278
in the landscape 15, 17, 29
visual aspect of 12, 21
See: Maastricht, Mllegrdsmarken
Chagneric (father of Burgundofara/o),
2513
Chagnoald, bishop of Laon 249
Chaignulf, count of Meaux 251
Charlemagne 284
and Lorsch 4189
and political opponents 2967
and pope Hadrian I 80
and pope Leo III 81
and Rome 52, 78
at Aachen 217ff., 539
chancery staff 240
charters 241
letter to bishop Ghaerbald of Lige
2323
See Einhard, Vita Karoli
Charles the Bald, king
and Redon 370, 384, 394
in Lotharingia 428
Charles the Fat, king 235
charters
Charlemagnes 241
Childeric II for Stavelot 2567
from Lorsch 4134
from the Middle Rhine 4126
of Redon 8745
Chelles, monastery of 3156

Cherniakhov culture 520, 523
Chvremont 1902
Childe, V.G. 24
Childeric II, king of the Franks 177,
1823, 2567
Childeric III, king 3267
Chindasuinth, king, penance of 315
Chlotar II, king 252
Chrodegang, bishop of Metz 4178
cities
as central places 533, 537
as places of power 67
public meetings at 4079, 411, 428
walls 32
See: Aachen, Bibistrense, Brioude,
Chvremont, Clermont, Cologne,
Constantinople, Cordoba, Dijon,
Dinant, Frankfurt, Huy, Langres,
Lige, Maastricht, Mainz, Namur,
Rome, Toledo, Tours, Trier,
Vienne, Worms
civitates, Roman past of 408
Clermont
and Gregory of Tours 1469
as a model 14950
as cult centre 1469
churches at 148
cults and politics at 1489
Clovis II, king 95, 253
Clovis III, king 177, 183
Clunies Ross, M. 4889
Codex regularum 287
coins 166
and papal authority 725, 78, 82
papal, and economy 734
production 2045
Cologne 194, 207
Columbanus 243, 281, 301. See: Jonas
of Bobbio, Vita Columbani
Constans II, emperor 47
Constantine, emperor 103
Constantinople 31ff., 48, 537
architecture 53
church of Holy Apostles 40
church of the Theotokos
Chalkoprateia 40
church of Theotokos of the Source
41
Forum of Constantine 40
Golden Gate 41
Hagia Eirene 40
Hagia Sophia 3942
Hebdomon 41
601
mausoleum of Constantine 40
processions at 35ff., 537
shrine of the Theotokos of the
Blachernae 41
symbolic topography of 3941
Contumeliosus, bishop of Riez,
penance of 303
Conuuoion, abbot of Redon 363ff.
convents see: monasteries, female
conversion, royal 3125
Anglo-Saxon 313
See: Carloman, Ratchis
Cordoba
as place of power 119ff., 5434
mosque of 1202
palace of 1212
cosmological order 20115
councils
Arles (533) 303
Braga (561) 334
Braga (572) 335
III Braga (675) 339
Merida (666) 331
X Toledo (656) 340, 351
XIII Toledo (683) 341
XV Toledo (688) 341
XVI Toledo (693) 337, 341
Valence (585) 280, 2823, 305
court
and sacred space 3068
nicknames at 22830
women at 2312
See: Attila, Avar khagans, barbarian
royal residences, Hephthalite
court, Mongol court, palaces,
Turkish khagans
craftsmen, significance of 4845, 487
See: smiths, dwarfs
Cugnon, monastery of 248
Cunigunda, queen 2312
Dado/Audoin 252, 254. See:
Faronids
Dagobert I, king 2523
Daniel the Stylite 5
Davies, W. 375
De ordine palatii 2268, 2302
deaconry (Rome) 90
dead, power of the 21, 23
Derks, H. 25
Despy, G. 187, 194, 198
Devroey, J.-P. 187, 194, 1978
Diem, A. 282, 300
602

Dijon to the monastery 291ff.

cult of Benignus 1425
religious topography of 143
Dinant 157ff.
Divisio regnorum (806) 2967
Dorestad, emporium of 410, 441
Douglas, M. 1920
dreams as political instrument 4256
drinking rituals
at barbarian court 4434, 448,
451, 453, 455
in Scandinavia 479
Duby, G. 200
Dumio, abbey and bishopric 335ff.
dwarfs
and mining 490
as smiths 490, 492, 500
Ebo, archbishop of Rheims, penance
of 304
Ebroin
exile of 31821
maior domus 259
penance of 320
Eddic poetry as a source 487, 4902
Effros, B. 214
Eigenkirche 171, 174
burial in 28
Eigenkloster 245
Einhard 442
on Aachen 217, 224, 236
on Caroloman 3235
on Childeric III 3267
See: Vita Karoli and Translatio et
miracula SS. Petri et Marcellini
elephant at Aachen 542
Eliade, M. 4945, 497
emporia 525.
disappearance of 529
See: Dorestad, Truso
Ercanfrida, will of 42930
Erwig, king, penance of 315
Esprandieu, E. 112
Eucherius, bishop of Lyon. See: Passio
Acaunensium martyrum
Eugenius I, pope 283
Eulogius 126
Excerpta de legationibus 460, 465
exemptions
episcopal 245, 253, 2824, 299300
papal 283
See: Rebais
exile
from the monastery 3556
voluntary/involuntary 31821
Faremoutiers, monastery of 247ff.
See: Burgundofara
Farfa, monastery of 284
Faronids 2514
Fastrada, queen 232, 426
Ferreolus 142, 150, 152
Fichtenau, H. 789
fortifications
in lower Vistula area 527
royal control over 434
Fosses, monastery of 268
Foucault, M. 5
Fouracre, P. 256, 318
Frankfurt, palace at 4223, 426
Fredegars Chronicle 251, 254
Continuations 260
on liturgy 2813
Fructuosus, bishop of Braga 335ff.
See: Vita Fructuosi
Fulda, monastery of 304, 407, 4124,
423, 536
abbot Ratgar 416
Funen 467ff.
Gallaecia 329ff.
christianization 3335, 353
isolation of 32933, 35051
monasteries in 329ff.
social change 3489
See: Priscillianism
Ganshof, F.-L. 226
Geary, P.J. 27
Geertz, C. 56, 71
Gerberding, R. 190
Gertrude, abbes of Nivelles (Life of
247)
Gesta Dagoberti I 288
Gesta sanctorum Rotonensium 376ff. See:
Redon
Ghaerbald of Lige, bishop 2323
gift-giving 545, 244, 281
debates on 2015
See: exemptions, immunities
gold
at Gudme/Lundeborg 468, 472,
476, 482, 484, 4924, 505
in lower Vistula area 513
Gorze, monastery of 4134
Gothic ethnicity, formation of 513
Gothic Wars 47
Graus, F. 297

graves
gravegoods 246, 29
of ancestors 23
of bishops 1724
spoliation of 1089
Gregory I, pope 98, 37980, 3913.
See: Moralia in Job
Gregory IV, pope 58
Gregory, bisshop of Langres 1434,
153
Gregory, bishop of Nazianzus 37
Gregory, bishop of Tours 93ff.,
137ff., 544
and Clermont 1469
and excavations 97
and Vienne 145, 1501
family of 1419
on Aravatius/Servatius 1616
on asylum 302
on Merovech 306
on St Martin 45
on St Maurice dAgaune 271, 273,
27980
on Tours 179
on treasure 10910
Gregory, C. 201
Grierson, Ph. 200
Grimoald, mayor of the palace 180,
190, 2479. See: Wulftrude
Gudme/Lundeborg 467ff.
archaeology of 46974
as centre of the world 5012
compared with Maastricht 1956
gold at 468, 472, 476, 482, 484,
4924, 505
hall at 502, 505
Odin cult at 476, 478, 498
sacred features of 4758
treasure at 472, 479, 5412
Gundobad, king 274
Guntram, king 2734, 2802, 3046
Hadrian I, pope 4951, 58, 62, 74,
7880
and Charlemagne 80
Hagia Sophia, typicon of 39, 42. See:
Constantinople
hall
at Gudme 502, 505
importance of 47880, 4967, 500
Halsall, G. 199
Hauck, K. 472, 476
Hedeager, L. 1956
Helena and the True Cross 1034
603
Helms, M. 4835, 487
Helvtius, A.-M. 212
Henry I, king, and local nobility
4345
Hephthalite court, Sung-yn at 448
Heppenheim
dispute over 3978, 4006, 421
local importance of 4179, 541
meeting at (795) 3978, 4006, 421
Hildegard, queen 232, 427
Hincmar, archbishop of Rheims
2269
Historia Mongolorum see: Iohannes de
Plano Carpini
Hohenburg, monastery of 265
Honorius I, pope 47
honour
depiction of in texts 298, 319327
familial 2612
female 263
Hrabanus Maurus, family of 4101,
413, 4156
hring 443, 4456
Notker Balbulus of St Gall on 446
Hsan-Tsang at Turkish court 448
Hubert, bishop of Lige 1735
Huy 157ff.
Hyadatus 3301, 350
Ibn Askar 1278
Ibn al-Khatib 122
Ibn al-Qutiya 128
Ibn Haf n 1279
Ibn awqal 120, 1225
Ibn Idh r 121,1289
immunities 253, 284, 299300,
3056, 312, 535
imprisonment 292ff.
Ingelheim, palace at 297, 420, 4267
Iohannes de Plano Carpini at Mongol
court 443, 44950
iter, royal 4245, 428
Itta 188, 247
Janes, D. 86
John Chrysostom 35ff.
John of Gorze, Life of 122
John VII, pope 48, 578, 66
Jonas of Bobbio, Vita Columbani 247,
250, 307315, 5401
Jordanes 513, 51730, 522, 524
Jouarre, monastery of 252
Julian, cult of at Brioude 1425
Jumiges, monastery of 95
604
Kasten, B. 2267
King, P.D. 346
kings
See: Bernard of Italy, Charlemagne,
Charles the Bald, Charles the Fat,
Childeric II, Childeric III,
Chindasuinth, Chlotar II, Clovis
II, Clovis III, Carloman, Dagobert
I, Erwig, Gundobald, Guntram,
Henry I, Lothar I, Lothar II,
Louis the German, Louis the
Pious, Marbod, Otto I, Sisigmund,
Theodegisl, Theuderic III,
Theodoric, Theudebert, Wamba,
Wittiza, Zwentibold
kinship
bonds of 3448, 3512
marriage politics. See: honour
Knigsnhe, importance of 428, 432
Krautheimer, R. 46
Kupper, J.L. 158
La Rocca, C. 32
Ladner, G. 66
Lambert, bishop of Lige 173ff. See: Vita
Lamberti vetustissima, Vita Lamberti IV
landscape
adaptation of 95
symbolism of 208ff.
transformation of 97103
Langres, cult of Tergemini 1445,
1489
Laon 269ff.
Lapidus, I.M. 125
lay abbacy of St Maximians (Trier)
4323
Le Maho, J. 113
Leges Visigothorum 345
Lehner, H.-J. 276, 278, 285
Leo III, pope 58, 68, 71, 75, 801,
538
and Charlemagne 81
Leo IV, pope 58
Leonine City (Rome) 50, 91
Lrins, monastery of 2823
Leudegar, bishop of Autun
exile of 31820
penance of 319
See: Passio prima Leudegarii, Passio
secunda Leudegarii
Leudegisus, bishop of Braga 340
Leupen, P.H.D. 1749
Libellus de ecclesiis Claromontanis 1478
Liber historiae Francorum 260, 320
Liber pontificalis 48, 58, 289, 324, 380
Libera custodia 298, 302
Libri miraculorum 381
Lige 157ff.
as centre of Pippinid power 17393
See: Chvremont
litanies See: liturgy
liturgy
litanies (Rome) 878
monastic 284ff.
stational (Rome) 836
See laus perennis, processions, St
Maurice dAgaune, Rogations,
turmae
local power, exercise of 398ff.
Lombard king and pope 78
Lorsch, abbey of 397, 400, 4123,
417, 4234, 536, 541
and Charlemagne 4189
charters 4134
conflict involving 397, 4005
monastic network of 4178
Lothar I, king, marriage of 427
Lothar II, king, succession of 428
Louis the German, king
and political crisis 4245
in Saxony 4245
Louis the Pious
and Redon 366, 390, 394
coronation of 233
public penance of 313, 322
rebellion against 291, 297, 366,
370
Luxeuil, monastery of 2823, 30714,
31822
Maastricht
and Merovingian kings 1826
and Pippinid power 17395, 2134
artisanal quarters 1948
as centre 155ff., 534, 53950
basilica of Our Lady 1758
basilica of St Servatius 164ff.
castrum 19698, 206
cemeteries 160174
compared with Gudme 1956
episcopal graves 1724
episcopal see (sedes episcopalis) 17598
excavations at 16672, 1967
portus 187, 211
royal estates 20813
See: Amandus, Hubert, Lambert,
Dinant, Huy, Meuse Valley,
Monulphus, Servatius, Wijck

Mainz
as a centre 40713, 423
economic importance of 40911
Mle, E. 98
Mamertus, bishop of Vienne 1504
Marazzi, F. 45
Marbod, king of the Marcomanni
441
Marco Polo 443
Marian feasts (Rome) 87, 89
Martellinus 13840
Martin, bishop of Braga 334
Martyrs, Theban 272, 2869
Mary, images of 618
Matthiae, G. 667
Maubeuge, monastery of 248
Megingoz, count, murder of 4312
Mellitus, missionary 98
Merovech, son of Chilperic I 306
Meuse valley 155ff.
Mllegrdsmarken, cemetery of 469,
474, 503
monasteries
and public business 4156
neigbourhood monasteries 3439
and aristocratic families 243ff.
and royal authority 281
and stabilitas 299301, 536
and texts 361ff.
as prisons 292ff.
as places of power 1, 5
as production site 1967, 543
as sacred places 271ff., 3068,
38194, 5345, 5423
as social centres 4125
carcer 302
child oblation 264, 300, 302
double monasteries 246
Eigenklster 245
familia 268
female 245ff.
legal activities at 4125
liturgy 284ff.
penance at 3037
political significance 244ff.
relics 267
royal patronage of 272ff., 4189
ruins 937
sacred space 267, 281, 3001
See: exemptions, exile, gift-giving,
immunities, penance, septa secreta
See: Andelys-sur-Seine, Andennes,
Chelles, Cugnon, Dumio,
Faremoutiers, Farfa, Fosses, Fulda,
605
Gorze, Hohenburg, Jouarre,
Jumiges, Lrins, Lorsch, Luxeuil,
Maubeuge, Mons, Monte Cassino,
Munsterbilzen, Nivelles, Rebais,
Redon, Rueil, Seligenstadt,
Soignies, St Calais, St Denis, St
Jean (Laon), St Marcel (Chalon),
St Maurice dAgaune, St
Symphorian (Autun), St
Maximians (Trier), St Calais,
Stavelot-Malmdy
Mongol court
Iohannes de Plano Carpini at 443,
44950
William of Rubruck at 443,
44951, 453
Mons, monastery of 247
Monte Cassino, monastery of 313,
3237
Monulphus, bishop of
Tongres-Maastricht 16474, 180
monuments, re-use of 20, 93ff., 214
Moore, H. 215
Moralia in Job 289
Munsterbilzen, monastery of 184
Namur 157, 160ff.
Nanthild, queen 2545
Neustria (Carolingian) 363
Nicenes 37
nicknames at the court 22830
Nilgen, U. 62
nimbus, square 58
Nivelles, convent of 185, 188, 247ff.
See: Wulftrude
Nora, P. 101
Notker Balbulus of St Gall
on Aachen 2236
on the hring 446
Odilia, abbess of Hohenburg 2645
Odin cult at Gudme/Lundeborg 476,
478, 498
Ordo Romanus I 85
Osborne, J. 58
ossuaries 112
Otto I, king and local nobility 435
Pactum 3547
palaces
rural 4201
See: Aachen, Cordoba, Frankfurt,
Ingelheim. See also: barbarian royal
residences
606
Panhuysen, T. 156, 167, 196
papal portraits 57ff.
and papal ideology 712
categories of 5961
Gregory IV 58
Hadrian I 58, 62
John VII 578, 66
Leo III 58, 68, 71
Leo IV 58
Paschal I 58, 65, 67
Paul I 58
with Jesus 6872
with Mary 618
Zachary 58
Paris as a centre 78
Parochiale Suevum 350
Parry, J. 202, 205
Paschal Chronicle 36
Paschal I, pope 58, 65, 67
and Francia 678
Passio Acaunensium martyrum 272, 287
Passio Praeiecti 148
Passio prima Leudegarii 31821
Passio secunda Leudegarii 320
Passio Sisigmindi regis 285, 289
Paternus 94
Paul I, pope 58
Paxton, F. 273
Pelagius 134
penance
as punishment 303ff.
in the monastery 3037
public 304
See: Chindasuuninth, Contumeliosus
of Riez, Ebo of Rheims, Ebroin,
Erwig, Leudegar of Autun, Louis
the Pious, Sagittarius of Gap,
Salonius of Embrun, Tassilo,
Wamba
Phokas (martyr) 356
Pippin I, mayor of the palace 188,
247
Pippin II, mayor of the palace
1836, 249, 257
Alpaida, wife of 1902
Pippinids, rising power of 158,
17593, 253, 25570
Pitt-Rivers, J. 246
Plectrude, wife of Pippin II 191
Pllan, cell of 368, 370, 387
popes
administration 45, 51
and building 49ff.
and Carolingian emperors 802
and documents 7583
and Lombard king 78
and patronage 55
and topographical imperialism 556
as temporal rulers 823
at Aachen 21920, 538
coins 725, 78, 82
exemptions 283
insignia 86
northern perception of 538
symbols of authority 7283
See: Eugenius I, Gregory I, Gregory
IV, Hadrian I, Honorius I, John
VII, Leo III, Leo IV, Paschal I,
Paul I, Stephanus II, Zachary
Potamio, bishop of Braga 340
prestige goods, meaning of 200ff.,
482ff.
priests as ritual specialists 26
Prinz, F. 273
Priscillianism 334ff., 356
Priscus on Attilas court 4425
prisons
liberation miracles 2978, 310
monasteries as 292ff.
royal 2978, 310
See: libera custodia
processions 7
Constantinople 35ff., 537
Frankish 423
Rome 8391, 537
Scandinavia 504
Vienne 1503
property rights 4045
punishment
of political opponents 2917
See blinding, exile, imprisonment,
penance, tonsure
queens see: Austrechildis, Balthild,
Brunhild, Cunigunda, Fastrada,
Hildegard, Nanthild
Rado 252. See: Faronids
Ratchis, king, conversion of 313, 324
Ratgar, abbot of Fulda 416
Rebais, monastery of 252
exemption for 2823
Redon, monastery of 362ff., 543
Almod, abbot of 3756
and Charles the Bald 370, 384,
394
and Louis the Pious 366, 390, 394
and Rome 3867

archive of 3716, 380, 387
as centre 3645
as holy place 38194
buildings at 3689, 385, 395
cartulary of 3745
charters 3745
community at 367, 369
Conuuoion, abbot of 363ff.
endowment of 36372
foundation of 36372
Gesta sanctorum Rotonensium 376ff.
relics at 38194
Ritcant, abbot of 3746
Regula Benedicti 287, 302, 355
Regula communis 329ff.
Regula consensoria monachorum 352
relics 267, 275, 280, 285
and ruins 96, 1056
at Redon 38194
Roman 538
re-use
of monuments 20, 93ff., 214
of objects 1134
of pagan religious structures 1124
of ruins 935
Rictrude, abbess of Marchiennes 263
Ritcant, abbot of Redon 3746
Rogations 280
at Vienne 1502, 280
Rome 45ff., 538
and art 56ff.
and Byzantine Empire 60, 66, 71,
73, 767, 80
and Charlemagne 52, 78
and Redon 3867
as papal city 45ff.
church decoration 54ff.
deaconries 90
ecclesiastical building projects 54
Leonine city 50, 91
litanies at 878
Marian feasts at 87, 89
papal gifts to churches 545
processions at 8391, 537
relics from 538
stational liturgy 836
Synod of (769) 767
Rosenwein, B.H. 201, 300
Roskams, S. 199
Rovelli, A. 734
Rueil, monastery of 252
ruins
and bishops 1023, 117
and Caesarius, bishop of Arles 1178
607
and elites 1013
and memory 1001
and monastic foundation 937
and relics 96, 1056
and rituals of possession 96, 117
at Aachen 218
attitudes towards 10310
perception of 100, 540
reuse of 935
Roman 99100, 539
sacrality of 101
Sadalberga, abbess of St Jean (Laon)
24950, 257
family of 24950
See: Vita Sadalbergae
Sagittarius, bishop of Gap, penance of
304
Sala concilia 68, 702
Salonius, bishop of Embrun, penance
of 304
Sancho the Fat 1334
Sauvaget, J. 32
Saxer, V. 86
Schama, S. 21
Schreiner, K. 246
Seligenstadt, abbey of 426
Semmler, J. 418
septa secreta 300, 3101
Servatius, bishop of Tongres-Maastricht
1616
Sidonius Apollinarus 1389, 148,
1504
silk decorations
Rome 54, 537
Turkish court 447, 452
Simonet, F. 127
Sisigmund, king and saint 27378,
2837. See: Passio Sisigmundi regis
Smith, J.Z. 101
smiths 468, 4857, 4902
dwarfs as 490, 492, 500
Soignies, monastery of 247
Sokrates (5th C. historian) 367, 41
Sozomen 367
spolia 103, 1079
St Calais, monastery of 306
St Denis, monastery of 203, 273, 281
St Jean (Laon), monastery of 247ff.
St Marcel (Chalon), monastery of
273, 2823
St Martin, power of 45
St Maurice dAgaune, monastery of
27190
608

architecture 2759 Avar 446

Gregory of Tours on 271, 273, display of 867, 447, 453
27980 Gregory of Tours on 10910
liturgy at 274ff. triclinia 6872
St Maximians (Trier), monastery of Trier as a centre 4315
429, 431, 434 True Cross, and Helena 1034
lay abbacy of 4323 Truso, emporium of 525, 5278
St Servatius, basilical monasterium of Turkish khagans, court of 4423, 445,
193ff. 447
St Symhporian (Autun), monastery of decorations at 447, 452
281 Hsan-Tsang at 448
Staab, F. 4234 Zemarchos at 447
states, types of early medieval 34 turmae 28690
Stavelot-Malmdy 248, 257 Typicon, of Hagia Sophia 39, 42
Stephanus II, pope 326
Subh, Woman and Cordoban Court Umayyad dynasty 11920, 128
1345 urban space
Sulpicius Severus 37980 consolidation of 34
Sung-yn at Hephthalite court 448 contraction of 323, 43
Symeon the Stylite 38 redefinition of 34
Synod of Rome (769) 767 transformation of 313

Tacitus 441, 456, 458, 511 Valierius of Bierzo 330, 335, 345,
Tassilo, duke of Bavaria, penance of 3502
297, 322 Venantius Fortunatus 94
Tergemini, cult of at Langres 1445, Vienne
1489 as cult centre 145, 1501
Theban Martyrs 272, 2869 Avitus, bisshop of 151, 274, 279
Theodegisel, king 97 Gregory of Tours and 145, 1501
Theodore Lektor 36, 412 processions at 1503
Theodoric, king 467 Rogations at 1502, 280
Theodosius I, emperor 166 violence
Theophylactus Simocatta 443 against female monasteries 243ff.
Theudebert II, king 251, 307, 3145 against political opponents 2947
Theuderic III, king 318 Vistula area, lower 509ff.
Toledo 119 early knowledge about 5102,
tonsure 5223
clerical 2934 fortifications 527
forced 294ff. gold in 513
Tours 13942 Vita abbatum Acaunensium 287
Alcuin on 138, 147 Vita Agilis 253
as a model 140, 149, 544 Vita Amandi 193
bishops of 1401 Vita Anskarii 2
cult of St Martin 1412, 544 Vita Anstrudis 247, 257
Gregory of Tours on 179 Vita Antonii 382
St Martins church 1379 Vita Argenteae 12536
trade, long-distance 2067, 4823, audience of 1312
487. See: amber route authorship of 132
Translatio et miracula SS. Petri et Marcellini date of 1312
195 on virtues 1301
treasure 4934, 519 provenance of 132
at Gudme/Lundeborg 472, 479, Vita Balthildis 316, 319
5412 Vita Boniti 148
 609

Vita Columbani 5401. See Jonas of Wijck (near Maastricht) 1945

Bobbio William of Rubruck at Mongol court
Vita Eulogii 1234 443, 44951, 453
Vita Filiberti 95 witchcraft, Icelandic 475, 488
Vita Fructuosi 337 Wittiza, king of the Sueves 332
Vita Karoli (Einhard) 3237 Wood, I. 179, 215, 255, 275
Vita Landiberti vetustissima 187 Worms, as a centre 423, 427, 431,
Vita Landiberti IV 183 4345
Vita Odiliae 264 Wright, T. 978
Vita Severini 442 Wulftrude, daughter of Grimoald,
Vita Sadalbergae 247ff., 285, 288 abbess of Nivelles 249

Wamba, king, penance of 3156 Yggdrasill, the World-Tree 4989, 503

Warin, count 4034, 418 Young, B. 98100
Weiner, A. 201
Wielbark Zachary, pope 58
culture 5123, 515, 5178, 520, Zemarchos at Turkish court 447
5224 Zeno, emperor 38
elite, and crisis 51820 Zwentibold, king 4313