This book presents thefirstanalytical account in English of the major devel-

opments within Byzantine culture, society and the state in the crucial for-
mative period from c. 610 to 717. Since its original publication in 1990, the
text has been revised throughout to take account of the latest research.
The seventh century saw the final collapse of ancient urban civilisation
and municipal culture, the rise of Islam, the evolution of patterns of thought
and social structure which made imperial iconoclasm possible, and the
development of state apparatuses - military, civil andfiscal- typical of the
middle Byzantine state. Over the same period, orthodox Christianity finally
became the unquestioned dominant cultural and religious framework of
belief, to the exclusion of alternative systems, which were henceforth mar-
ginalised or proscribed.
Conflicting ideas of how these changes and developments are to be under-
stood have proliferated in the pastfiftyyears. This book is the first serious
attempt to provide a comprehensive, detailed survey of all the major changes
in this period.
BYZANTIUM IN THE SEVENTH CENTURY
BYZANTIUM
IN THE SEVENTH
CENTURY
The transformation of a culture
Revised edition

J.F. HALDON
Professor of Byzantine History
Director of the Centre for Byzantine,
Ottoman and Modern Greek Studies,
University of Birmingham

CAMBRIDGE
UNIVERSITY PRESS
 Published by the Press Syndicate of the University of Cambridge
The Pitt Building, Trumpington Street, Cambridge CB2 1RP
40 West 20th Street, New York. NY 10011-4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia

© Cambridge University Press 1990

First published 1990

Reprinted 1993,1995
First paperback edition published 1997

British Library cataloguing in publication data

Haldon. J.F.
Byzantium in the seventh century: the transformation of a culture
1. Byzantine civilisation, to 717
I. Title
949.5'01

Library of Congress cataloguing in publication data

Haldon, John F.
Byzantium in the seventh century: the transformation of a culture
J. F. Haldon.
p. cm.
Includes bibliographical references
ISBN 0-521-26492-8
1. Byzantine Empire - Civilisation - 527-1081. I. Title
II. Title: Byzantium in the 7th century.
DF571.H35 1990
949.5'01-dc20 89-17309 CIP

ISBN 0 52126492 8 hardback

ISBN0 521 31917Xpaperback

Transferred to digital printing 2003

CP
ForV.J.W.
 Contents

List of plates page x

List of maps xii
Preface and acknowledgements xiii
Preface to the revised edition xv
List of abbreviations xvii
The sources xxi

Introduction 1
1 The background: state and society before Heraclius 9
2 The East Roman world c. 610-717: the politics of
survival 41
3 Social relations and the economy: the cities and the
land 92
4 Social relations and the economy: rural society 125
5 The state and its apparatus: fiscal administration 173
6 The state and its apparatus: military administration 208
7 Society, state and law 254
8 The imperial church and the politics of authority 281
9 Religion and belief 324
10 Forms of social and cultural organisation: infra-
structures and hierarchies 376
11 Forms of representation: language, literature and the
icon 403
Conclusion The transformation of a culture 436
Addendum: Further observations on the question
of the late ancient city 459
Bibliography 462
Index 482
ix
 Plates

1.1 Justinian. Copper follis page 19

1.2 Justin II. Gold solidus 32
1.3 Tiberius Constantine. Copper follis 34
1.4 Maurice. Gold solidus 34
1.5 Phocas. Gold solidus 37
2.1 Heraclius. Gold solidus 42
2.2 Constans II. Gold solidus 54
2.3 Constantine IV. Gold solidus 64
2.4 Leontius II. Gold solidus 70
2.5 Tiberius III Apsimar. Gold solidus 75
2.6 Justinian II (second reign). Gold solidus 75
2.7 Philippicus Bardanes. Gold solidus 79
2.8 Anastasius II. Gold solidus 79
2.9 Theodosius III. Gold solidus 82
2.10 Leo III. Gold solidus 84
9.1 Seventh-century icon (from a triptych) of St Theodore
Tiro 357
9.2 Sixth- to seventh-century icon of St Peter 358
9.3 Seventh- to eighth-century icon of St Athanasius and
St Basil 359
11.1 The Empress Theodora and attendants, San Vitale,
Ravenna 408
11.2(a) Late sixth- to seventh-century mosaic decoration from St 410
Demetrius, Thessaloniki (water-colour W. S. George)
(b) St Demetrius with patrons and benefactors. Mosaic of 411
post-reconstruction period, second quarter of seventh
century
11.3 Icon of the Virgin and child (detail) between St Theodore 413
and St George, with flanking angels, monastery of St
Catherine, Sinai
 List of plates xi

11.4(a) Solidus of Justinian I (obverse and reverse) 414

(b) Solidus of Constans II with Constantine, his son 414
(obverse and reverse)
11.5(a) Solidus of Justin II (obverse and reverse) 416
(b) Solidus of Maurice (obverse) 416
(c) Solidus of Phocas (obverse) 416
(d) Solidus of Heraclius and Heraclius Constantine 416
(obverse)
(e) Solidus of Constantine IV (obverse and reverse) 417
(0 Solidus of Justinian II (obverse) 417
(g) Solidus of Anastasius II (obverse) 417
(h) Solidus of Philippicus Bardanes (obverse and reverse) 417

ACKNOWLEDGEMENTS
Reproduction of the plates has been possible through the courtesy of the
following:
The Barber Institute of Fine Arts, University of Birmingham (plates 1.1-5,
2.1-10, 11.4(a) and (b) and 11.5(a)-(h))
The Michigan-Princeton-Alexandria Expedition to Mount Sinai (plates
9.1-3 and 11.3)
A. Michalakis, Athens (plates 11.1 and 2(b))
The Courtauld Institute of Art and the British School at Athens (plate
 Maps

I The empire in A.D. 565: approximate extent page 18

II Nominal extent of imperial territory in c. A.D. 600 33
III The empire in c. A.D. 650-700: the process of devastation 65
IV The empire at the accession of Leo III (A.D. 717) 83
V The Anatolian frontier region in the seventh and early 106
eighth centuries
VI Justinianic prefectures and provinces c. A.D. 565 228
VII The Anatolian themata and the late Roman provinces 230
c. A.D. 660

xu
 Preface and acknowledgements

The present work was originally conceived as a general and introductory

account of aspects of seventh-century East Roman (Byzantine) state,
society and culture. In the event, a simple survey of sources and literature
and the presentation of a general synthesis proved less and less worth-
while, or indeed desirable. It became necessary in many places to go into
considerably greater detail than planned.
Many technical matters of state and social organisation remained both
unclear and insufficiently researched; many questions of crucial import-
ance for the history of social and cultural development in the Byzantine
world of that period remained unasked. This book is, consequently, an
attempt to provide a coherent general overview by means of the analysis of
a series of specific themes and the corresponding problems which accom-
pany them. Inevitably, the themes I have discussed represent a selective
choice - I have concentrated on those aspects which I felt to be most in
need of attention, most amenable to some form of constructive solution
and most relevant to our understanding of how East Roman cultural,
social and state forms functioned and evolved as a dynamic whole
throughout the seventh century.
In writing this book, I have enjoyed the help and advice, both direct and
indirect, of many friends and colleagues. I would like in particular to thank
Wolfram Brandes, Marie-Theres Fogen, Rodney Hilton, Alexander Kazdan,
Ralph Lilie, Greg McLennan, Spiros Troianos, Chris Wickham and Fried-
helm Winkelmann, all of whom - at various times and in various places
over the last few years - have, not always knowingly, contributed in one
way or another to the formation of my views. In particular, I should like to
thank my friends and colleagues at the Centre for Byzantine, Ottoman and
Modern Greek Studies at the University of Birmingham, especially Michael
Ursinus. Under its director, Anthony Bryer, the Centre has provided the
means for fruitful research and the scholarly atmosphere in which to work.
I would also like to thank my friend and colleague at King's College,

xiii
xiv Preface and acknowledgements

London, Averil Cameron, for her constructive criticism and comments on

the penultimate version of the typescript, as well as for her interest and
encouragement in general; and my friends Leslie Brubaker, for much
fruitful discussion on art, representation and perception; Ludwig Burg-
mann and Bernard Stolte, for their willingness to read through and offer
constructive suggestions on problems relating to early Byzantine law and
legal texts; and Robin Cormack and Lucy-Anne Hunt for their help and
advice with regard to photographs. Nubar Hampartumian patiently
guided me through the extensive seventh-century Byzantine coin holdings
of the Barber Institute of Fine Arts at Birmingham; and Harry Buglass of
the Department of Ancient History and Archaeology at Birmingham is to
be congratulated on making sense of my original sketch-maps.
I should also like to thank the British Academy, the Alexander von
Humboldt-Stiftung and the Max-Planck-Institut fur Europaische Rechtsge-
schichte, under its director Dieter Simon, who all provided financial
support and an academic refuge at various times towards the completion
of the present project.
Last, but, as usual, by no means least, thanks are due to Iris Hunter of
Cambridge University Press, whose efficient, rigorous and constructive
sub-editing has turned the original sow's ear into something resembling a
silk purse.
 Preface to the revised edition

It is a great pleasure to be able to introduce this revised edition of Byzantium

in the seventh century. I have made a number of emendations and additions to
the notes and to the bibliography, and also a brief Addendum on the current
state of the question of the fate of urban centres in the seventh century. But
for technical reasons I have not incorporated all the changes I would, pre-
dictably, have wished to make, simply because that would have involved the
substantial expansion or rewriting of some sections to take into account the
work of colleagues and scholars in the various fields upon which this book
touches. This was not too difficult, however, since, six years on, I believe that
the analysis of the events which so transformed the late Roman world as I
have tried to present them here is still a valid interpretation of the material at
our disposal. Naturally, there are areas where one would wish to add
nuance, or change emphasis, but, on the whole, the broad thrust and the
direction of the interpretation remain unchanged, a point reinforced, as far
as I am able to judge it, by the work of other scholars in the field in the past
few years. The results of a good deal of work on the literary sources - Greek,
Syriac and Arabic - have now become available, for example, which adds
substantially to our knowledge of both the development of older as well as
newer genres. The cross-cultural connections between Greek and Syriac
writing, especially in the fields of historiography, hagiography and theology,
and the nature and tendency of anti-Jewish polemic, have all been studied in
greater depth, and the results of this work further enhance or modify the
conclusions reached in those chapters of this book in which these texts are
evaluated or used as sources of evidence. Similarly, the results of recent
archaeological and field survey work need to be taken into account in evalu-
ating the differential and highly localised pattern of development of urban
and fortified centres during the period in question, although the general line
of the argument presented in the relevant chapters below is confirmed.
In the process of re-evaluating the views expressed in this book, and
reassessing the sources for the analysis of the history of the East Roman
xv
xvi Preface to the revised edition

empire in the seventh century, I am grateful to many friends and colleagues

for stimulating criticism and advice. I should especially like to acknowledge
Wolfram Brandes, Paul Speck, Chris Wickham and Averil Cameron: they
almost certainly disagree with each other across a whole range of issues,
and they may not always agree with my interpretation; but I have benefited
enormously from their critical engagement, their scholarly advice and
expertise, and their friendship.

Birmingham, February 1997

 Abbreviations

AB Analecta Bollandiana (Brussels 1882-)

AHAS Ada Historica Academiae Scientiarum Hungaricae (Budapest
19 52-)
AIPHOS Annuaire de Vlnstitut de PMoTogie et d'Histoire Orientales et
Slaves (Brussels 1932-)
AvaXexra AvdXexra 'IepoaoAv/jLLTLxrjs XraxvAoyiag, I—IV, ed.
A. Papadopoulos-Kerameus (St Petersburg 1891-9)
AS Ada Sandorum (Antwerp 1643-)
B Byzantion (Brussels and Paris 1924-)
BAR British Archaeological Reports
BBA Berliner Byzantinistische Arbeiten (Berlin 1955-)
BCH Bulletin de Correspondance Hellenique (Paris 1877-)
BF Byzantinische Forschungen (Amsterdam 1966-)
BGA Bibliotheca Geographorum Araborum, ed. M.-J. De Goeje
(Leiden 18 70-). Nunc continuata consultantibus
R. Blachere (etc.) (Leiden 1938-)
BHG3 F. Halkin, Bibliotheca Hagiographica Graeca (Brussels, third
edn, 1957)
BHG, AucL F. Halkin, Auctarium Bibliothecae Hagiographicae Graecae
(Brussels 1969)
BMGS Byzantine and Modern Greek Studies (Oxford 1975-83; Bir-
mingham 1984-)
BN] Byzantinisch-Neugriechische Jahrbitcher (Berlin and Athens
1920-)
ByzBulg Byzantinobulgarica (Sofia 1962-)
BS Byzantinoslavica (Prague 1929-)
BZ Byzantinische Zeitschrift (Leipzig and Munich 1892-)
CEHE Cambridge Economic History of Europe (Cambridge 1941-)

xvii
xviii Abbreviations

CFHB Corpus Fontinum Historiae Byzantinae (Washington 1 9 6 7 -

(Series Washingtoniensis); Berlin and New York 1 9 6 7 -
(Series Berolinensis); Vienna 1 9 7 5 - (Series
Vindobonensis); Rome 1 9 7 5 - (Series Italica); Brussels
1 9 7 5 - (Series Bruxellensis))
CI Codex lustinianus
CIC Corpus luris Civilis
CM Classica et Medievalia (Copenhagen 1938-)
CMH Cambridge Medieval History (Cambridge 1913-)
CPG Clavis Patrum Graecorum, III: A Cyrillo Alexandrino ad
lohannem Damascenum, ed. M. Geerard (Turnhout 1979);
IV: Concilia Catenae, ed. M. Geerard (Turnhout 1980)
CSCO Corpus Scriptorum Christianorum Orientalium (Paris and
Louvain 1903-)
CSEL Corpus Scriptorum Ecclesiasticorum Latinorum (Vienna
1866-)
CSHB Corpus Scriptorum Historiae Byzantinae (Bonn 1828-97)
De Cer. De Cerimoniis
AIEE Askriov Trjs'Iorropixrjsxal 'EOvoAoywfjs'ETepeias
TTJS 'EMdSos (Athens 1883-)
DOP Dumbarton Oaks Papers (Cambridge, Mass., and
Washington 1941-)
AXAE AeXriov TTJS XpLcrTiavixrjsApxono\oyix/q<; 'Eraipeias
(Athens 1892-); irepio&os A', vols. 1-10 (Athens
1892-1911); ireptoSos B \ vols. 1-2 (Athens 1924-5);
irepioSos P , vols. 1-4 (Athens 1933-9); irepio8os A',
vols. Iff (Athens I960-)
EEBX 'EneTEpLs EraipsCas Bv^avnviov XTTOVSCJU (Athei is
1924-)
EHR English Historical Review (London 188 5-)
El Encyclopaedia of Islam, new edn (Leiden and London
I960-)
EO Echos d'Orient, 1-39 (Paris, Constantinople and Bucharest
1897-1941/2)
EKE See Bury, A History of the Eastern Roman Empire
FHG Fragmenta Historicorum Graecorum, ed. C. and Th. Miiller, 5
vols. (Paris 1874-85)
FM Fontes Minores (Frankfurt am Main, 19 76-)
GCS Die griechischen christlichen Schriftsteller der ersten (drei)
Jahrhunderte (Leipzig and Berlin 189 7-)
GRBS Greek, Roman and Byzantine Studies (1: Greek and Byzantine
Studies) (San Antonio, University of Mississippi,
Cambridge, Mass., and Durham 1958-)
 Abbreviations xix

HC L'Hellenisme Contemporain (Athens 1947-)

HGM Historici Graeci Minores, 2 vols., ed. L. Dindorf (Leipzig
1870-1)
IGR lus Graecoromanum
1RAIK hvestija Russkago Arkheologiceskago Instituta v
Konstantinopole, 1-16 (Odessa and Sofia 1896-1912)
JHS Journal of Hellenic Studies (London 188O-)
JIAN Journal Internationale d'Archeologie Numismatique (Athens
1897-1927)
JOB Jahrbuch der osterreichischen Byzantinistik 1 8 - (Vienna,
Cologne and Graz 1969-)
JOBG Jahrbuch der osterreichischen byzantinischen Gesellschaft
1-17 (Vienna, Cologne and Graz 1951-68)
JRS Journal of Roman Studies (London 1911-)
Klet. Phil. Kletorologion tou Philotheou (see Bibliography: Primary
sources)
LRE See Jones, The Later Roman Empire
MGH(AA) Monumenta Germaniae Historica (Auctores Antiquissimi),
edd. G. Pertz, Th. Mommsen et al. (Berlin 1877-1919)
NPB Nova Patrum Bibliotheca, ed. A. Mai, vols. I-VII (Rome
1852-4); ed. I. Cozza-Luzi, vols. VIII-X (Rome
1871-1905)
OC Oriens Christianus (Leipzig 1901-)
OCP Orientalia Christiana Periodica (Rome 1935-)
PG Patrologiae Cursus Completus, series Graeco-Latina, ed. J.P.
Migne (Paris 1857-66; 1880-1903)
PL Patrologiae Cursus Completus, series Latina, ed. J.P. Migne
(Paris 1844-1974)
PO Patrologia Orientalis, edd. R. Graffin, F. Nau (Paris 1930-)
PP Past and Present (London 1952-)
RAC Reallexikon fur Antike und Christentum, Sachworterbuch zur
Auseinandersetzung des Christentums mit der antiken Welt,
ed. Th. Klauser (Stuttgart 19 50-)
KB Reallexikon der Byzantinistik, ed. P. Wirth (ser. A, vol. I,
fascicles 1-6 only) (Amsterdam 1968-76)
RbK Reallexikon zur byzantinischen Kunst, eds. K. Wessel and
M. Restle, vols. I- (Stuttgart 1963-)
RE Paulys Realencyclopddie der classischen
Altertums-Wissenschaft, new revised edn by G. Wissowa
(vol. I/I, Stuttgart 1893-); vol. I/I (1893) to XXIII/2
(1959; with index of additions); XXIV (1963); I/Al (1914)
to X/A (1972); Supl. I (1903) to XIV (1974)
xx Abbreviations
REA Revue des Etudes Armeniennes, new series (Paris 1964-)
REB Revue des Etudes Byzantines (vols. 1-3: Etudes Byzantines)
(Bucharest and Paris 1944-)
REG Revue des Etudes Grecques (Paris 1888-)
RESEE Revue des Etudes Sud-Est Europeennes (Bucharest 1963-)
RH Revue Historique (Paris 18 76-)
RHSEE Revue Historique du Sud-Est Europeen (Bucharest 1924-46)
RJ Rechtshistorisches Journal (Frankfurt am Main 1982-)
RN Revue Numismatique (Paris 1836-)
ROC Revue de VOrient Chretien, ser. 1, vols. 1-10 (Paris
1896-1905); ser. 2, vois. 1-10 (Paris 1906-1915/17);
ser. 3, vols. 1-10 (Paris 1918/19-1935/6): vols. I-XXX
RSBN Rivista di Studi Bizantini e Neoellenici, new series (Rome
1964-)
SBB Sitzungsberichte der bayerischen Akademie der
Wissenschaften, phil.-hist. Klasse
SBN Studi Bizantini (vols. 1-2) e Neoellenici (vols. 3-10) (Rome
1925-)(cont. as RSBN)
SBO Sitzungsberichte der osterreichischen Akademie der
Wissenschaften, phil.-hist. Klasse
SOF Sudost-Forschungen (Leipzig, Vienna and Munich 1936-)
T.Usp. Taktikon Uspenskij (see Bibliography: Primary sources)
TM Travaux et Memoires (Paris 1965-)
Maria Varia Graeca Sacra, ed. A. Papadopoulos-Kerameus (St
Petersburg 1909)
VSW Vierteljahreschrift fur Sozial- und Wirtschaftsgeschichte
(Leipzig and Wiesbaden 19O3-)
VV Vizantijskii Vremmenik, vols. 1-25 (St Petersburg
(Leningrad) 1894-1927); new series (Moscow 1947-)
ZK Zeitschrift fur Kirchengeschichte (Gotha 18 7 7-)
ZRVl Zbornik Radova (vols. 1-6; Vizantoloski Institut, kn. 1-6 =
Srpska Akademija Nauka, Zbornik Radova, kn. 21, 36,
44, 49, 59, 65) VizantoloSkog Instituta (from vol. 7 (1961))
(Belgrade 19 52-)
 The sources

The sources for the period with which we shall be dealing are, in com-
parison with those for the sixth century, or the tenth and eleventh cen-
turies, for example, both limited in number and difficult to use. But I should
like to stress at the outset that these.difficulties ought not to be over-
estimated, nor should they be used as a justification for refusing to ask
questions. In fact, it is less the paucity of the sources than their nature
which is problematic. There are, effectively, only two 'histories' of the
period compiled by Byzantines, both of which date from the early ninth
century, although based in large part on earlier material. One, the Brief
History of the patriarch Nicephorus, has virtually nothing to say on the
reign of Constans II (641-68). While both use what may well be material
contemporary with many of the events they describe, it is an immensely
difficult task to sort out the different and sometimes very contradictory
traditions bound up in the two histories. The other, the Chronography of
the monk Theophanes the Confessor, was compiled between A.D. 810 and
814, and continues the Chronicle of George the Sygkellos. It is written
around a carefully worked-out chronological framework, divided into
sections by the year on an annalistic basis, at the head of each of which
Theophanes lists the year according to the age of the world, from the birth
of Christ, and according to the lengths of the reigns of the emperor, caliph,
the pope and the four patriarchs of the East. It has been shown that the
dates are one year out for the entries from the year 6102 (A.D. 609-10) to
6265 (A.D. 772-3), except for the period 6207 (A.D. 714-15) to 6218
(725-6), as the result of an incorrect division in the text. But this is the
only blemish on a fundamental text, upon which our own chronology of
the seventh and eighth centuries is based.1
1
See G. Ostrogorsky, 'Die Chronologie des Theophanes im 7. und 8. Jahrhundert1, BNJ 7
(1930), 1-56; Ostrogorsky, art. Theophanes1, in RE V/A2 (1934), 2127-32; for further
discussion and literature, see H. Hunger, Die hochsprachliche profane Literaturder Byzantiner
(2 vols. Handbuch der Altertumswissenschaft XII, 5.1 and 2 = Byzantinisches Handbuch 5, 1

xxi
xxii The sources

The Brief History of Nicephorus (patriarch from 806 until 815) covers
the period from 602 to 769, apart from the gap already mentioned for
most of the reign of Constans II. It is based in part on the same sources as
the Chronography of Theophanes. The gap seems to be the result of a loss
of some folios from the manuscript tradition. Nicephorus wrote also a
Short Chronicle which is of only limited historical value, covering the
period from Adam to 829, in which year he died.2
Last, and for the beginning of our period, there is the so-called Paschal
Chronicle, originally covering the period from Adam to the year A.D. 629
but, due to the loss of the final folios, ending in 628. It was compiled by a
priest or monk in the 630s and built around a chronological framework
intended to fix the Easter cycle and the reckoning for a purely Christian
chronology.3
These Greek sources can be supplemented from a very wide range of
other sources in other languages. And, indeed, one of the difficulties facing
the historian of this period is precisely this wide range of material, often
later in date, but again using sources contemporary with - or almost
contemporary with - the events described: chronicles, or fragments from
chronicles in Arabic, Syriac, Armenian, Coptic and Latin, for example, all
provide vital material which must be assessed. Among the most important
of these are the much later Chronicle of Michael the Syrian, Jacobite

and 2, Munich 1978), vol. 1, pp. 334-59; and also I. Rochow, 'Die monenergetischen und
monotheletischen Streitigkeiten in der Sicht des Chronisten Theophanes'. Klio 63 (1981),
669-81. On the debate over the authorship of the Chronographia, see C. Mango, 'Who
wrote the Chronicle of Theophanes?1, ZRVl 18 (1978), 578-87; and I.S. Cicurov 'Feofan
Ispovednik-publikator, redaktor, avtor?', W 4 2 (1981), 78-87. Detailed analyses and his-
torical-philological commentaries of various sections of the text of Theophanes can be found
in P. Speck, 'Die Interpretation des Beilum Avaricum und der Kater McxA-ejiiTe', in Varia II
(Poikila Byzantina VI. Bonn 1987) 37Iff., idem, Das geteilte Dossier. Beobachtungen zu den
Nachrichtenfiberdie Regierung des Kaisers Herakleios und die seiner Sohne bei Theophanes und
Nikephoros (Poikila Byzantina IX. Berlin-Bonn 1988); idem, Ich bin's nicht, Kaiser Konstantin
ist es gewesen. Die Legenden vom Enflufi des Teufels, des Juden und des Moslem auf den
Ikonoklasmus (Poikila Byzantina X. Bonn 1990), and idem, 'Der "zweite" Theophanes. Eine
These zur Chronographie des Theophanes', in Varia V (Poikila Byzantina Xm. Bonn 1994),
pp. 433-83; I. Rochow, Byzanz im 8. Jahrhundert in der Sicht des Theophanes (BBA LVII. Berlin
1991). See also LI. Conrad, 'Theophanes and the Arabic Historical Tradition: Some
Indications of InterculturalTransmission', BF IS (1990) 1-44; andL.M. Whitby, 'The Great
Chronographer and Theophanes', BMGS 8 (1982-83) 1-20. For a good, brief introduction
and assessment of the various types of source dealt with in the following, see W. Brandes, F.
Winkelmann, eds., Quellen zur Geschichte desfruhen Byzanz (BBA LV. Berlin, 1990).
2
See especially P.J. Alexander, The Patriarch Nicephorus of Constantinople. Ecclesiastical Policy
and Image Worship in the Byzantine Empire (Oxford 1958) on Nicephorus and his writings;
and Hunger, Profane Literatur, vol. 1, pp. 344-7. See also C. Mango, The Breviarium of the
Patriarch Nicephorus', in Byzantion: Tribute toAndreasN. Stratos (Athens 1986), EL pp. 545-8;
and Nicephorus, Patriarch of Constantinople. Short History, ed. and trans. C. Mango
(Washington D.C. 1990).
3
See Hunger, Profane Literatur, vol. 1, pp. 328-30.
 The sources xxiii
patriarch of Antioch, and the Chronicle falsely ascribed to Denis of Tell-
Mahre; in addition, the mostly lost chronicle which was actually composed
by the latter and upon which that of Michael the Syrian draws heavily, as
does Bar Hebraeus and one of the anonymous chronicles.4 A number of
lesser Syriac chronicles, all anonymous, compiled mostly in the tenth,
eleventh and twelfth centuries, cover events up to the years 724, 813 and
846. 5 Particularly important sources are the Armenian history compiled
by the bishop Sebeos, probably in the early 660s; and the Chronicle of
John, the bishop of Nikiu in Egypt, compiled in Coptic towards the end of
the seventh century.6 Both present events from their own, localised per-
spective, but Sebeos especially sheds much light on the political and
ecclesiastical history of the empire and the capital city. Sebeos' history is
supplemented partially by that of the later chronicler Ghevond
(Lewond).7 Of the Latin historical sources, the most useful are the chron-
icle compiled by the bishop John of Biclar, and the Liber Pontificalis,
although a number of minor chronicles and annals are also valuable.8 The
history of the Goths, Vandals and Suevi of Isidore of Seville, and its
continuations, as well as his Chronica maiora, are also valuable for the
history of the eastern Mediterranean up to the 620s, as well as for Spain
and the West.9 Finally, the Arab histories (dating mainly from the later
4
J.-B. Chabot, La Chronique de Michel le Syrien, Patriarche Jacobite d'Antioche (4 vols., Paris
1899, 1901, 1905, 1910 and 1924). On the history and structure of the text, see vol. I,
pp. xxiv-xxxvii; and J.-B. Chabot, Pseudo-Denys de Tell-Mahre', Chronique (Paris 1895). For
the Syriac material in general, see S.P. Brock, 'Syriac sources for the seventh century',
BMGS 2 (1976), 17-36; and GJ. Reinink, 'Pseudo-Methodius und die Legende vom
romischen Endkaiser', in The Use and Abuse of Eschatology in the Middle Ages, eds.
W. Verbeke, D. Verhelst and A. Welkenhuysen (Leuven 1988), 82-111, see 84ff. on the as
yet mostly unpublished but very important seventh-century history of Johannan bar
Penkaye. For editions and translations of Bar Hebraeus, see Brock, art cit, 22f. For the orien-
tal sources for the seventh century, see also the papers in Av. Cameron, L. Conrad, eds., The
Byzantine and Early Islamic Near East I: Problems in the Literary Source Material (Studies in Late
Antiquity and Early Islam 1.1. Princeton 1992).
5
See J.-B. Chabot and E.W. Brooks (eds.), in CSCO scriptores Syri, ser. 3, vol. IV (Chronica
Minora II, 4, pp. 63-119; III, 1, pp. 185-96; II, 5, pp. 123-80) and for a chronicle to the
year 1234, ed. J.-B. Chabot. CSCO scriptores Syri 56, pt. 1. In addition, there are the
chronicles of Elias of Nisibis (vol. I, ed. E.W. Brooks, CSCO scriptores Syri, ser. 3, vol. VII;
vol. II, ed. J.-B. Chabot, ibid., ser. 3, vol. VIII); and the minor anonymous Chronicon
Maroniticum, ed. and trans. E.W. Brooks and J.-B. Chabot. CSCO scriptores Syri. ser. 3, vol.
IV (Chronica Minora II, 3, pp. 35-7).
6
F. Macler, Sebeos, Histoire d'Heraclius (Paris 1904) and R.H. Charles, ed. and transl., The
Chronicle of John, Bishop of Nikiu (London 1916).
7
G. Chahnazarian, ed., Ghevond: histoire des guerres et des conquetes des Arabes en Armenie
(Paris 1856).
8
John of Biclar's chronicle is in MGH (AA) XI, 2, pp. 211-20; the Liber Pontificalis is edited
by L. Duchesne (2 vols., Paris 1884-92).
9
Isidori lunioris Episcopi Hispalensis Historia Gothorum Wandalorum Sueborum ad An. DCXXIV,
in MGH (AA) XI, 2, pp. 267-303 (text); Continuations lsidorianae Byzantina Arabica et
Hispana, in ibid., pp. 354-68; Chronica Isidori lunioris, in ibid., pp. 424-81.
xxiv The sources
eighth century and later, but often dependent upon much earlier material),
especially those of Baladhuri and Tabari. The latter are especially impor-
tant for the Muslim invasions and the occupation of the eastern territories
of the empire.10 All of these varied sources are valuable, but each must be
carefully weighed as to its textual background, its sources, the question of
contamination and interpolation - in other words, as to its value and
reliability as a historical source.
But quite apart from these more obviously 'historiographicaF forms of
evidence, there is a vast range of other material: official documents issued
or drawn up on behalf of the state (edicts and novellae, for example);11
imperial letters (sent, for example, to foreign rulers, leading secular or
ecclesiastical officials and to generals); ecclesiastical documents (patri-
archal letters, letters and documents concerning matters of dogma and
Church business, the appointment of clerics, convening of synods and so
forth, as well as the acts of the Church councils - chiefly the acts of the
Lateran council of 649, of the sixth ecumenical council of 680-1 and of
the 'Quinisext' council of 692; 12 lists of episcopal sees and descriptions of
ecclesiastical administration, and so on); 13 epic poems and encomia, such
10
See Baladhuri Kitab futuh al-Buldan. The Origins of the Islamic State, trans. Ph. Hitti
(London 1916 and Beirut 1966); and M.-j. de Goeje, ed., Annales quos Scripsit Abu Djafar
Mohammed Ibn Dfarir al Tabari cum Aliis (3 vols., Leiden 1879); and also Th. Noldeke,
Geschichte der Perser und Araber zur Zeit der Sassaniden aus der arabischen Chronik des Tabari
(Leiden 1879) for the period up to the end of the Sassanid empire. On these sources see
E.W. Brooks, The Arabs in Asia Minor (641-750) from Arabic sources', JHS 18 (1898),
182-208; E.W. Brooks, 'Byzantines and Arabs in the time of the early Abbasids', EHR 15
(1900), 728-47. On the Arab historiographical tradition, see G. Strohmaier, Arabische
Quellen', in Brandes, Winkelmann, Queilen zur Geschichte desfruhen Byzanz, pp. 234-44;
and in general A. Noth (with Lawrence Conrad), The Early Arabic Historical Tradition. A
Source-Critical Study (Princeton 1994).
11
These will be discussed in greater detail in chapter 3 (esp. in regard of the so-called
'Farmers' Law') and chapter 7 (imperial legislation and codification).
12
For the Lateran, see R. Riedinger, ed., Concilium Laterananse a. 649 Celebratum (Acta
Conciliorum Oecumenicorum, 2nd ser., vol. I. Berlin, 1984) (this edition replaces the text
in Mansi X, 863-1170). In addition, see the article by R. Riedinger, 'Die Lateransynode
von 649 und Maximos der Bekenner', in Maximus Confessor. Actes du Symposium pour
Maxime le confesseur (Fribourg, 2-5 Sept. 1980) eds. F. Heinzer and Chr. Schonborn
(Fribourg, 1982), pp. 111-21 ( = Paradosls 27); for the council of 680, Mansi XI,
190-922 and F.X. Murphy and P. Sherwood, Constantinople U et Constantinople III (Paris
1974), pp. 133-260; and for the Quinisext council, Mansi XI, 921-1005; and
V. Laurent, 'L'CEuvre canonique du concile in Trullo (691-2), source primaire du droit de
Teglise orientale', REB 23 (1965), 7-41. New edition of the acts of the council of 680:
Concilium universale Constantinopolitanum tertium, ed. R. Riedinger, 2 vols. (Acta
Conciliorum Oecumenicorum H/2.1-2) (Berlin 1990/1992).
13
In particular the description of George of Cyprus: see E. Honigmann, Le Synekde'mos
d'Hie'rokles et Vopuscule giographique de Georges de Chypre (Corpus Bruxellense Historiae
Byzantinae I. Brussels 1939); H. Gelzer, Georgii Cyprii, Descriptio Orbis Romani (Leipzig
1890); and the Pseudo-Epiphanius Notitia, in H. Gelzer, 'Ungedruckte und ungenttgend
veroflentlichte Texte der Notitiae Episcopatuum. Ein Beitrag zur byzantinischen Kirchen-
und Verwaltungsgeschichte', in Abhandlungen der bayerischen Akademie der Wissen-
 The sources xxv

as those in praise of the Emperor Heraclius by George of Pisidia, or to

celebrate and give thanks for the victory over the Persians and Avars in
626 by Theodore the Sygkellos;14 private letters, whether of laypeople or
Churchmen; semi-official documents incorporated into 'official' collections,
such as the accounts of the trials of Maximus Confessor or Pope Martin or
the account of the seizure of power by Philippicus Bardanes;15 theological
and dogmatic writings, such as the apocalyptic tracts or the collections of
'questions and answers', especially that attributed to Anastasius of Sinai;16
hagiographical writing - a particularly important source;17 epigraphic

schaften, phil.-hist. Klasse, XXI, 3 (Munich 1900). More recent edition, with analysis, of
all the Notitiae, in J. Darrouzes, Notitiae Episcopatuum Ecclesiae Constantinopolitanae (Paris
1981).
14
The poems of George of Pisidia deal with the Persian wars of Heraclius, the siege of 626,
the character and mission of the emperor himself, the role of the patricius Bonus: see
A. Pertusi, ed., Giorgio di Pisidia, Poemi I: Panegirici epici (Studia patristica et Byzantina VII,
Ettal 1959); and the discussion in ClaudiaLudwig, 'Kaiser Herakleios, GeorgiosPisidesund
die Perserkriege', in Varia HI (Poikila Byzantina XI. Bonn 1991), pp. 73-128; the sermon of
Theodore the Sygkellos, skeuophylax of the Hagia Sophia, was edited by L. Sternbach,
Analecta Avarica {Rozprawy Akademii Umiejetnosci Wydzial Filologiczny, ser. 2, vol. XV
Cracow 1900), pp. 297-334 ( = L. Sternbach, Studia philologica in Georgium Pisidam
(Cracow 1900)).
15
For Maximus: Maximi Confessoris Relatio Motionis, in PG XC, 109-29; Gesta in Primo Eius
Exsilio, in PG XC, 135-72; see C.N. Tsirpanlis, 4Acta S. Maximi', Theologia 43 (1972),
106-24. For Martin: Commemoratio, in Mansi X, 853-61 and PL CXXIX, 591-600; and
see R. Devreesse, 4Le Texte grec de l'hypomnesticum de Theodore SpoudeV, AB 53
(1935), 49-80. These slightly later accounts seem to be based on Martin's own letters 16
and 17 (PL LXXXVII, 201-4 and Mansi X, 860-4). For Philippicus, see the account of the
deacon Agathon in Mansi XII, 189C-196C.
16
In particular the Apocalypse of Pseudo-Methodius: A. Lolos, ed., Die Apokalypse des
Ps.-Methodius (Beitrage zur klassischen Philologie LXXXIII, Meisenheim am Glan 1976)
and the literature in Haldon, Ideology and Social Change, 168 and n. 74; see chapters 9 and
11 below. Further discussion in W. Brandes, 'Die apokalyptische Iiteratur', in Brandes,
Winkelmann, Queilen zur Geschichte desfruhen Byzanz, pp. 305-26. For Anastasius of Sinai,
see G. Dagron, 'Le Saint, le savant, l'astrologue. Etude de themes hagiographiques a travers
quelques recueils de "Questions et reponses" des Ve-VIF siecles', in Hagiographie, cultures et
societes (W-VIT s.) (Etudes Augustiniennes, Paris 1981), pp. 143-55 (repr. in Dragon, La
Romanite chretienne en Orient IV (London 1984). Among the most important texts of
Anastasius are the interrogationes et responsiones (PG LXXXDC, 311-384) and his third ser-
mon (PG LXXXDC, 1152-80); new edition ed. K.-H. Uthemann, Anastasti Sinaitae Opera.
Sermones Duo in Constitutionem Hominis secundum Imaginem Dei necnon Opuscula adversus
Monotheletas(CorpusChristianorumseriesGraecaXII. (Turnhout 1985), pp. 55-83). SeeJ.F.
Haldon, 'The Writings of Anastasius of Sinai: a key source for seventh-century East
Mediterranean history', in: The Byzantine and Early Islamic Near East I: Problems in the
Literary Source Materials, eds. Averil Cameron, L. Conrad (Princeton 1992), pp. 107-47.
17
Hagiography and miracles constitute a vital source for the social and cultural life of the
period. Among the many texts which are especially important are: the Lives of the
patriarchs John the Almsgiver in Alexandria, Eutychius in Constantinople, of Pope
Martin, Maximus Confessor (originally in Syriac), the Hypomnesticum or commemoration
of Theodore Spoudaios (on Pope Martin), the Lives of Theodore of Sykeon and of
xxvi The sources
material - inscriptions on city walls, for example, or on tombstones;18
numismatics; and last, but certainly not least, sigillographic material - the
seals used by both officials and private persons to validate and secure
letters or merchandise. And in addition to this predominantly 'documen-
tary' material, we must also bear in mind the considerable archaeological
evidence, particularly where settlement patterns on the one hand and
architectural forms on the other (Church buildings, fortifications and so
on) are concerned; as well as that of late Roman and Byzantine forms of
visual expression - icons, frescoes, mosaics - essential to our understand-
ing of some of the assumptions of Byzantine culture, its perception of the
world and of the relationship between both emperor and people, and
between God and humanity.19

Anastasius of Persia, the miracles of St Demetrius in Thessalonica, of Artemius and

Therapon in Constantinople, the fictional Life of Andrew the Fool, the miracles of St
Theodore at Euchaita. For literature see note 19 below; full references will be found in the
bibliography.
18
Most of the relevant material is to be found in the major collections: Corpus Inscriptiomtm
Graecarwn, ed. A. Bockh (vols. I and II) and I. Franz (vols. Hlff.) (Berlin 1828-); Monu-
menta Asiae Minoris Antiqua, eds. W.M. Calder, J. Keil et al. (Manchester 1928-62);
Inscriptions grecques et Mines de la Syrie, eds. L. Jalabert, R. Mouterde et al. (Paris 1929-);
Inscriptions grecques et Mines de la Syrie, ed. W.H. Waddington (Paris 1870 and Rome
1969 = Inscriptions grecques et Mines recueillies en Grece et en Asie Mineure III, 1/2. Paris
1870 and 1876); Supplementum Epigraphicum Graecum, eds. J.J.E. Hondius and A.G.
Woodhead (Leiden 1927-); see also F. Cumont, 'Les Inscriptions chretiennes d'Asie
Mineure', in Melanges d'archeologie et d'histoire 15 (1895); V. Besevliev, Spdtgriechische und
spdtlateinische Inschriften aus Bulgarien (Berlin, 1964); H. Gregoire, Recueils des inscriptions
grecques chretiennes d'Asie Mineure (Paris 1922); E. Popescu, lnscrip\iile Grecesti si Latine
din Secolele IV-Xlll descoperite in Romania (Bucharest 1976). For further publications, see
J. Karayannopoulos, flrj-yal TTJS Bv£avTivq<; 'Ioropias, 4th edn (Thessaloniki 1978),
pp. 195-97.
19
The critical literature on the types of sources listed here is, of course, enormous. For the
main groups of both literary and non-literary sources the following compendia, hand-
books and bibliographies will provide a basic and fairly full survey by period and genre of
the sources themselves and the problems connected with them: G. Ostrogorsky, Geschichte
des byzantinischen Staates (Handbuch der Altertumswissenschaft, XII, 1.2 = Byzantinisches
Handbuch, 1, 2, Munich, 1963), pp. 72-5; Gy. Moravcsik, Byzantinoturcica, vol. I, 3rd edn
(Berlin 1983); Hunger, Profane Literatur, vols. I and II (for historiography, secular poetry,
epistolography, military handbooks, law); J. Karayannopoulos and G. Weiss, Quellenkunde
zur Geschichte von Byzanz (324-1453) (2 vols., Wiesbaden 1982) for a survey of all the
literary genres and discussion of their value, as well as of the non-Byzantine sources, esp.
vol. II, pp. 302-18; H.-G. Beck, Kirche und theologische Literatur im byzantinischen Reich
(Handbuch der Altertumswissenschaft, XII, 2.1 = Byzantinisches Handbuch, 2, 1, Munich
1959), esp. pp. 430-73 for a detailed survey of theological and related writings in the
seventh century (dogma, polemic, liturgical poetry, sermons, hagiography, mystical and
exegetical works); on hagiography see also I. Sevcenko, 'Hagiography of the iconoclast
period', in A.A.M. Bryer and J. Herrin (eds.), Iconoclasm (Birmingham 1977), pp. 113-31,
and Sevcenko, 'L'Agiografia Bizantina dal IV al IX secolo', in La civiltd Bizantina dal IV al IX
secolo (3 vols., Bari 1977,1978 and 1982), vol. I, pp. 87-173. For a brief introduction to
the Syriac material see S.P. Brock, 'Syriac sources for seventh-century history', BMGS 2
(1976), 17-36; for numismatics, the introductory comments of M.F. Hendy, Studies in the
 The sources xxvii

The problems posed by all these sources are considerable. Many of them
are in need of modern editions and commentaries; few of them have been
studied in detail or have been internally analysed. Where written texts are
concerned in particular it must be apparent that no source can be taken at
face value. It is sometimes difficult to date a source at all - a classic example
is the very valuable but fictional account of the Life of Andrew the Fool,
dated by different scholars to both the later seventh or the later ninth
century.20 Where the date is certain, the value of the information provided
both explicitly and implicitly by a text needs to be carefully considered -
what was the purpose or context of the original compilation, for example,
and what were its possible sources of information? Often, literary texts refer
only indirectly and allusively to a particular state of affairs or development,
or use technical terms from the period of their own compilation of earlier
events - both have led, and continue to lead, to conflicting views about the
precise significance of the references in question. Similar reservations
apply to the question of the date of certain works of art, too, so that some of
the Mt. Sinai icons which are so crucial for an understanding of pre-
iconoclastic art are variously dated from the end of the sixth to the middle
of the seventh century - a period over which substantial changes occurred
- and for a knowledge of the evolution and origins of which these artifacts
are central. Likewise, the use and exploitation of sigillographic as well as
numismatic materials bring with them a number of equally formidable
problems, and it is important that the historian as well as the readers of
works of history are aware of these difficulties. For differences in interpre-
tation usually rest on two supports: conflicting views of how and what a
given source or type of source can divulge about a specific question; and
conflicting or contradictory interpretational frameworks.
This brief list of types of source highlights the mosaic-like complexity of
the historian's task. It also demonstrates the constraint upon interpreta-
tion imposed by the sources and the importance of keeping some general
principles of analysis - some theoretical guidelines - in mind. For without

Byzantine Monetary Economy, 300-1450 (Cambridge 1985), pp. 2-18: and Karayanno-
poulos and Weiss, Quellenkunde, vol. I, pp. 172-8; for seals, see the valuable comments in
the introductory section to F. Winkelmann, Byzantinische Rang- und Amterstruktur im 8.
und 9. Jahrhundert (BBA LIII, Berlin 1985), pp. 17-18; and W. Seibt, Die byzantinischen
Bleisiegel in Osterreich, I: Kaiserhof (Vienna 1978), pp. 36f.; with Karayannopoulos and
Weiss, Quellenkunde, vol. I, pp. 178-83. All these materials are surveyed in the collective
Brandes, Winkelmann, Quellen zur Geschichte desfruhen Byzanz.
20
See the conflicting views of L. Ryden, 'The Life of St Basil the Younger and the Life of
St Andreas Salos', in Okeanos. Essays presented to lhor Sevcenko on his sixtieth birthday by his
colleagues and students ( - Harvard Ukrainian Studies, 7 1983 (Cambridge, Mass. 1983)),
pp. 568-86; and C. Mango, 'The Life of St Andrew the Fool reconsidered', Rivista di Studi
Bizantini e Slavi II ( = Miscellanea A. Pertusi II, Bologna 1982), pp. 297-313 (repr. in
C. Mango, Byzantium and its Image VIII (London 1984)).
xxviii The sources

some organising framework, the 'evidence' becomes simply overwhelm-

ing; or, what is perhaps worse, too readily fitted into an ill-considered or
preconceived notion of what late Roman and early Byzantine society was
'really* like. I shall try to avoid these two extremes. But I shall also avoid a
detailed analysis of each source employed, since this would require more
than a volume to itself. Instead, I will refer, where appropriate, to relevant
discussions of the problems associated with the deployment of a particular
source or group of sources and comment only at those points where a
particular difficulty in connection with a specific source is encountered.
For the historian of the seventh century, the interrelationship between
evidence and hypothesis plays a more than usually central role. The
American philosopher W.V. Quine once wrote: 'I see philosophy and
science as in the same boat - a boat which ... we can rebuild only at sea
while staying afloat on it.' Among others, the history of the seventh
century is also in that boat.21
21
W. V. Quine, 'Natural Kinds', in Ontological Relativity and Other Essays (New York 1969),
pp. 114-38, seep. 127.
 Introduction

The seventh century was a time of fundamental transformation

throughout the eastern Mediterranean and Balkan world; and the most
powerful political entity in that world experienced a succession of major
upheavals. Entering the seventh century as the dominant political for-
mation, stretching from Spain to the Euphrates and from the Danube to the
Atlas mountains, it had been reduced by the end of our period - the year of
the accession of Leo III in A.D. 717 - to a rump of its former self: East
Roman rule in Egypt and North Africa, in Syria, Iraq and in eastern
Anatolia had been swept away by the conquests of Islam, the new and
vigorous world religion which was to provide the biggest threat to Chris-
tianity for the next thousand years. In the Balkans, Slavs and Bulgars had
reduced Roman-controlled territory over the same period to the coastal
areas and a few fortified settlements; while in Italy, the exarchate which
had been established under Tiberius Constantine on the foundations of
Justinian's reconquest was by the reign of Leo III all but extinguished. At
the same time, new and powerful foes replaced older, traditional enemies:
the expansion and the power of the caliphate centred at Damascus radi-
cally altered the balance of power in that area; in the Balkans, the
establishment and consolidation of the Bulgar khanate posed a constant
threat to Constantinople itself; while in the West, the increasingly indepen-
dent see of St Peter was compelled to loosen its ties with Constantinople in
order to preserve its own position as both leader of the Western Church
and defender of its immediate hinterland. The East Roman empire which
we observe at the beginning of the seventh century has, by the time of Leo
III, been transformed into the 'Byzantine' empire of the Middle Ages, and
along with it its institutions, its social relations and the dominant elements
of political and popular belief systems.
The seventh century, traditionally, belongs to the 'Dark Ages', in the
fullest sense of the phrase, a period when both the sources available to the
historian are fewer than in earlier or later times, and when the exigencies
2 Byzantium in the seventh century

of the struggle to survive (or, in more analytic terms, when social,

economic and cultural transformations were in full swing) made the
production of a widely based secular literature both less relevant to the
cultural identity of the dominant elite - the imperial establishment and its
bureaucratic and military representatives - and subordinate to the
demand for political and theological certainties. In fact, as a cursory
examination of the relevant bibliographies will quickly show, while there is
an absolute decline in secular literary production, this is not the case with
theological or political-theological writings, hagiography and homiletic
writings, collections of ewtapokriseis (questions and answers on everyday
or theological concerns), apocalyptic and eschatological tracts and similar
texts, all of which reflect the immediate concerns and foremost worries and
interests of seventh-century society. In this sense, at least, this is by no
means a 'dark age' - the light is there, but it shines on different areas, more
selectively. Indeed, the very term 'dark age', or 'ages', is itself suspect, as
has been noted before now. 1 Where we read in the literature about this
period of economic 'decline' or political 'collapse' we should perhaps
substitute 'political and economic transformation' - if only to avoid exam-
ining the society in question purely in terms of a reflection of the language
and perceptions of those members of it able to express themselves in
literature.
This is not a total history of Byzantine culture and society during the
period in question. Rather, I want to look at the basic 'shape' of this early
medieval social formation as it might be seen in the early seventh century
and to follow the process of change over the following one hundred or so
years. But I shall not simply describe the changes in question - as far as we
can see them through the interrogation of the available sources - I shall
also try to suggest what lay behind them, why they occurred at the time
and in the form that they did, and why they had the results which they
had. I shall be trying in the process to determine what particular char-
acteristics differentiate the later Roman and early Byzantine world from its
eastern and western neighbours, what in particular governs the develop-
ment of its specific cultural forms and modes of expression. At the same
time as trying to provide a useful survey of the Byzantine world at this
time, I shall also be concentrating on some of the key problems of Byzan-
tine history as currently identified in recent historiographical work. This
book is therefore a contemporary document, in the sense that it contains
a statement both of problems of history as they are perceived today, and of

1
See especially F.-G. Maier, 'Die Legende der "Dark Ages'", in F.-G. Maier, ed., Die Ver-
wandlung der Mittelmeerwelt (Frankfurt a.M. 1968), pp. 10ff.; and D. Talbot Rice, 'The myth
of the Dark Ages', in D. Talbot Rice, ed., The Dark Ages (London 1965), introduction.
 Introduction 3

what has been achieved; and one of historiographical desiderata - what

still remains to be done.
Because of this format, the chronological narrative has been kept to a
minimum: chapter 2 provides a brief survey of the political history of the
period, which remains the most useful measure against which to 'fix' other
developments, even if it occasionally leads to a rather inflexible schemati-
sation and even if it is often regarded as no longer a primary moment in
historical analysis; the politics of a state or its rulers are as much the
product of the cultural, social and economic relationships which determine
the shape of a society as any other aspects of its existence and evolution.
The remaining chapters - with the exception of chapter 1 (which
provides a general background for the whole) - will each deal with a
particular theme, as can be seen from the headings and sub-headings,
which will be treated both structurally and chronologically. That is to say,
the developments in time within a specific problem area will be presented
and described, while at the same time an examination of the key aspects of
the theme to be dealt with will be undertaken. In this way, I hope that I can
provide the reader with both a descriptive and an analytic account of the
movement and transformation of the early Byzantine state and its society
through time. I have also tried to make the book accessible to both
specialists and non-specialists, by incorporating within a general account
the detailed debates and analyses which will be relevant to the former.
The seventh century has received a great deal more attention over the
last twenty years than in the years prior to about 1965. This shift in
interest is partly a reflection of an awareness of the former neglect to which
the period and its problems were consigned, which in turn represented the
difficulties experienced in dealing with the sources. More importantly, it
reflects also a shift in research priorities and interests together with a
reappraisal of the methods and the theories that might be invoked in
examining such a period. Thus, if we exclude articles in journals, of the six
or so publications that have dealt specifically with the seventh century,
only two appeared before 1965: volume III of Kulakovskij's classic Byzan-
tine History, covering the period from 602 to 717; and the selection of
papers on the seventh century in volume 13 of the Dumbarton Oaks Papers.
The five-volume history of Andreas Stratos, Byzantium in the seventh
century; the Berlin symposium on the same period, out of which two major
publications appeared and a third is currently in press; and the recent book
by T.K. Louggis on Byzantium in the so-called dark ages have all appeared
since 1965. 2 But there is much disagreement between these various works
2
See J. Kulakovskij, Istoria Vizantii (Kiev 1915 and London 1973) vol. Ill; A.N. Stratos,
Byzantium in the Seventh Century (Engl. transl.) I: 602-34 (Amsterdam 1968); II: 634-41
(Amsterdam 1972); III: 642-68 (Amsterdam 1975); IV: 668-85 (Amsterdam 1978); V:
4 Byzantium in the seventh century

on fundamental questions - the nature of the social relations of production,

the role of the army in society, the relationship between the state and large
landholding, between tax and rent, and so on. In what follows, some idea
of these debates will be given and some solutions will be suggested, as part
of the process of providing a general survey of the development of East
Roman society and culture during the seventh century.
Inevitably, I have not been able to deal with everything that might seem
desirable in such a general survey. The history of Byzantine Italy and its
particular socio-cultural, economic and institutional evolution during this
period represented too different and too distant a world to be directly
relevant to the history of the central lands of the empire - although its
political and military importance, its ideological significance (Rome in
particular, of course) and its effect on Byzantine state and ecclesiastical
politics have been taken into account. In addition, of course, good surveys
of Italy in this period are available, and in several accessible languages. It
would be pointless to duplicate their results.3 Similar considerations apply
to Byzantine Africa, a political and social formation about which all too

686-711 (Amsterdam 1980); F. Winkelmann, H. Kopstein, H. Ditten and I. Rochow,

Byzanz im 7. Jahrhundert. Untersuchungen zur Herausbildung des Feudalismus (BBA
XXXXVIII, Berlin, 1978) and H. Kopstein and F. Winkelmann, eds., Studien zum 7. Jahrhun-
dert in Byzanz. Probleme der Herausbildung des Feudalismus (BBA XXXXVII, Berlin 1976); for
the Dumbarton Oaks sumposium, see DOP 13, (1959). See further T.K. Louggis, Aoxiyno
yia TTJP xoivwvLxrj egeAigr) orrj bidpxeia TWV keyojievcw axoieivaiv auovcov (Athens
1985), a somewhat reductionist interpretation which takes issue with a number of
traditional interpretations without really demonstrating a convincing alternative (and
which seems to assume that all historians in the Soviet Union or Eastern Europe are
'Marxist', whereas those elsewhere are not - see pp. 87ff.)- In addition, the introduction (by
V. Vavfinek) and contributions to the collection From Late Antiquity to Early Byzantium, ed.
V. Vavfinek (Prague 1985), provide useful guides to recent debates and literature among
both Soviet/East European and West European/North American historians. Other works
might be mentioned which deal only partly with aspects of seventh-century history. These
will be referred to as they become relevant and are listed in the bibliography. For a recent
treatment of the sources for the period, see the important collective work Quellen zur
Geschichte des friihen Byzanz ed. F. Winkelmann (BBA LV, Berlin, 1989); and for the
seventh-century city, W. Brandes, 'Die byzantinische Stadt Kleinasiens im 7. Jahrhundert -
ein Forschungsbericht', Klio 70 (1988), 176-208; W. Brandes, Die Stddte Kleinasiens im 7.
und 8. Jahrhundert (BBA LVI, Berlin 1989).
3
Apart from the older works, see T.S. Brown, Gentlemen and Officers. Imperial Administration
and Aristocratic Power in Byzantine Italy A.D. 554-800 (Rome 1984); and more recently
T.S. Brown, 'The interplay between Roman and Byzantine traditions and local sentiment in
the Exarchate of Ravenna', in: Bisanzio, Roma e Vltalia nell'alto Medioevo (Settimane di
Studio del Centro Italiano di Studi sull'alto Medioevo XXXIV, Spoleto 1988), 127-60. CJ.
Wickham, Early Medieval Italy. Central Government and Local Society, 400-1000 (London
1981); A. Guillou, Regionalisme et independance dans Vempire byzantin au VIIe siecle: Yexemple
de Vexarchat et de la pentapole d'ltalie (Istituto storico italiano per il medio evo, Studi storici
75-6, Rome 1969); 0. Bertolini, 'Riflessi politici del'controversie religiose con Bisanzio
nelle vicende del sec. VII in Italia', in Caraterri del secolo VII in Occidente (Spoleto 1958),
pp. 733-89 (= Settimane di studio del Centro Italiano di Studi sull'alto Medioevo V
(1957)).
 Introduction 5

little is known after the middle of the seventh century. In this case,
however, the only major work on the subject is the now very old, but still
essential, survey of Charles Diehl, whose study on the exarchate of Africa
was first published in 1896. 4 This can be supplemented by a number of
works dealing with particular aspects of the history of North Africa, but a
detailed analysis of the literary, religious and social-economic history of
Byzantine Africa from the reign of Heraclius up to the fall of Carthage in
the last decade of the seventh century is still not available. Fortunately,
recent work promises to fill some of these gaps, although there is still a
great deal of detailed research, especially archaeological work, to be done
before the later history of the isolated Latin culture of North Africa
becomes reasonably clear; and there remains a dismal lack of source
material. Again, Byzantine Africa was an important consideration in the
eyes of the government at Constantinople, which did its best, with limited
resources, to maintain its political and military hold, as well as its ideo-
logical authority. It consumed wealth in the form of military and naval
resources, although the degree of its contribution to the fisc in general is
unclear; and it remained a part of the Byzantine world, from the Constanti-
nopolitan perspective, until the end - the expedition to retake Carthage in
698 is demonstration enough of this. In this respect, and in as far as the
general course of North African history is concerned, it will be dealt with in
this book. Anything more would involve a study in its own right. 5
4
Ch. Diehl, L'Afrique byzantine: histoire de la domination byzantine en Afrique (533-709) (2
vols., Paris 1896). See also R. Goodchild, 'Byzantines, Berbers and Arabs in seventh-
century Libya', Antiquity 41 (1967), 114-24; R. Goodchild, 'Fortificazione e palazzi bizan-
tini in Tripolitania e Cirenaica', in XIII Corso di Cultura sull'Arte Ravennate e Bizantina
(Ravenna 1966), pp. 225-50; Ch.A. Julien, Histoire de VAfrique du Nord des origines a la
conquete arabe, 2nd edn (Paris 1951); M. Restle, art. 'Byzacena', in RbKl (1966), 837-66;
also E. Kirsten, Nordafrikanische Stadtbilder, Antike und Mittelalter in Libyen und Tunisien
(Heidelberg 1966); and P. Goubert, Byzance avant llslam II/2 (Paris 1965), pp. 185-236
(for the period up to 610 only).
5
See the various articles published in connection with the University of Michigan excavati-
ons at Carthage, esp. vol. VII (1982), notably the important contribution of Averil
Cameron, 'Byzantine Africa - the literary evidence', ibid., 29-62, with the older literature.
See also D. Pringle, The Defence of Byzantine Africa from Justinian to the Arab Conquest (BAR,
Oxford 1981); N. Duval, 'Influences byzantines sur la civilisation chretienne de 1'Afrique du
Nord', KEG 84 (1971); W.H.C. Frend, 'The Christian period in Mediterranean Africa
c. A.D. 200-700', in Cambridge History of Africa, vol. II (Cambridge 1978), pp. 410-89;
W.H.C. Frend, 'The end of Byzantine North Africa. Some evidence of transitions', Bulletin
archeologique du Comite des Travaux Historiques et Scientifiques, new series, 19 (1985),
387-97 (IP Colloque international sur l'histoire et l'archeologie de 1'Afrique du Nord,
Grenoble, 5-9 Avril, 1983); M. Brett, 'The Arab conquest and the rise of Islam', in
Cambridge History of Africa, vol. II (Cambridge 1978), pp. 490-555, esp. pp. 490-513. For
further literature and discussion, see N. Duval, 'L'Afrique byzantine de Justinien a la
conquete musulmane. A propos de travaux recents', Moyen-Age 89 (1983), 433-9; as well
as the work of J. Durliat, Recherches sur l'histoire sociale de VAfrique byzantine: le dossier
epigraphique (533-709) (These du troisieme cycle, Universite de Paris I, Paris 1977); Les
Dedicaces d'ouvrages de defense dans VAfrique byzantine (Collections de l'ecole francaise de
6 Byzantium in the seventh century

There are other topics which I have not dealt with in depth, chiefly
because they are adequately treated elsewhere. The (primarily theological)
literature of the seventh century as literature has been left to others, for
example; similarly, the history of the art and architecture (again, the
surviving material is almost entirely religious in character) of the period,
which presents certain very important characteristics and shifts, has been
dealt with predominantly on the basis of work done by others - although,
for reasons which will become clearer in the relevant discussion (in
chapter 11), a great deal more emphasis has been placed upon visual
representation than on architecture.
Historians rarely preface their work with statements of theoretical intent
- perhaps much to the relief of many readers, but this is not necessarily
always a good thing. For every work of historiography relies on sets of
assumptions; and 'theories', however implicit they might be, are inescapa-
ble. I will mention here some of my own basic assumptions.
The main point to make is that this book is conceived and written within
a historical materialist framework - that is to say, it is written from a
'Marxist' perspective. I place the word Marxist in quotation marks
advisedly: the word can mean, and is regularly in debate within the social
and historical sciences used to refer to, such a wide variety of subtly or
not-so-subtly differentiated views and approaches, that to refer to oneself
as a Marxist is of only limited help in determining which of a variety of
perspectives within a range of possibilities is actually meant. For while
Marxism has a relatively short history as a philosophical and political
movement, it is nevertheless immensely ramified and has been enormously
influential.
One may ask, of course, why it should be at all necessary to justify one's
terms of reference or indeed to situate oneself in a particular historiogra-
phical tradition. Surely it should be enough to base oneself firmly in the
sources and to apply one's historical common sense to their interpretation
and to the possible shape of a given set of historical developments? The
answer, and the justification, is not difficult to grasp. Theories are, in effect,
sets of premises - whether they are implicit or explicit is unimportant at
this point - which condition both the mode of interpretation as well as
(crucially) the mode of appropriation of knowledge (in other words, the
very way in which we permit ourselves to 'know' something). Such
premises or assumptions are, as I have said, implicit in every piece of

Rome, Rome 1981); 'Les Finances municipales africaines de Constantin aux Aghlabides',
in Bulletin Archeologique du Comite des Travaux Historiques et Scientifiques, n.s. 19 (1985),
377-86 (as above); 'Les Grands proprietaires africains et l'etat byzantin', Cahiers de
Tunisie XXIX (1981), 517-31; 'I/Administration civile du diocese byzantin d'Afrique
(533-703)', Rivista di studi bizantini e slavi 4 (1984), 149-78.
 Introduction 7

analysis, whether it be of literary texts or of historical sources. Theory, in

this sense, is inescapable; and there is no use in appealing to an objective,
fact-based history, for such does not, and indeed cannot, exist. It is better to
admit that this is the case and to make these underlying assumptions
explicit, for this allows at least the possibility of seeing where inconsist-
encies and contradictions might lie, inconsistencies which are inevitable in
any process of intuitive and ad hoc reasoning. Theory is, from this point of
view, highly desirable, since it is impossible to analyse that of which we
have no, or only indirect, experience (either personally or culturally)
without setting up a theoretical framework within which w e can justify
and direct our evaluation of the data. 6 1 can do no better than to quote the
linguist Noam Chomsky:
The search for rigorous formulation has a much more serious motivation than
mere concern for logical niceties or the desire to purify well-established methods of
... analysis. Precisely constructed models ... can play an important role, both
negative and positive, in the process of discovery itself. By pushing a precise but
inadequate formulation to an unacceptable conclusion, we can often expose the
exact source of this inadequacy and, consequently, gain a deeper understanding of
the ... data ... Obscure and intuition-bound notions can neither lead to absurd
conclusions nor provide new and correct ones ... those ... who have questioned
the value of precise and technical development of theory may have failed to
recognise the productive potential in the method of rigorously stating a proposed
theory and applying it strictly to [the] material with no attempt to avoid unaccept-
able conclusions by ad hoc adjustments or loose formulation.7
Of course, Chomsky is talking here of socio-linguistic data, the analysis
and interpretation of which is subject to different principles from those
applied to historical evidence. But the point he is making is equally
relevant.
The search for rigorous formulation is not, therefore, a pointless exer-
cise. Indeed, if we are to be able to deal effectively with our data, in such a
way that we can order it according to the structure and function of the
argument we wish to make, then an awareness of our basic assumptions is
essential. For historians within the Marxist tradition, the need for rigorous
6
For the foundations of my approach, see my remarks in 'Ideology and social change in the
seventh century: military discontent as a barometer', Klio 68 (1986), 139-90, esp.
142-55. For a good general survey of current Marxist historical thinking from an explicitly
realist and materialist epistemoldgical standpoint, see G. McLennan, Marxism and the
Methodologies of History (London 1981). See further the comments in J.F. Haldon,
'"Jargon" vs. "the facts"? Byzantine history-writing and contemporary debates', BMGS 9
(1984-5), 95-132, esp. 96, 101f.; and idem, The State and the Tributary Mode of Production
(London 1993), esp. 1-109,140-58.
7
N. Chomsky, Syntactic Structures (Janua Linguarum, series minor IV, The Hague and Paris
1976), p. 5. Compare the remarks of the Marxist historian Eric Hobsbawn: one must 'try to
structure (memory) by fitting it into an explanatory or theoretical scheme. It's one of the
advantages of being a Marxist ... Things which might otherwise be "trivial pursuit"
actually get fitted into a pattern ...' (The Guardian, Friday 26 Feb. 1988, p. 25).
8 Byzantium in the seventh century

formulation and careful application of theoretical principles is crucial; and

this involves determining the content and theoretical weight of the con-
cepts invoked, either as abstract or concrete descriptive categories. For it is
precisely through such categories that causal relationships - whether of
socio-economic relationships or political ideological developments - can be
specified and understood, and tied in to longer-term transformations.
Technical terms such as mode of production, social formation, symbolic
universe and so on will thus occur from time to time in this book more
especially in the context of analyses of the ways in which Byzantine society
actually worked and changed. These, and a number of other terms, are
part of a wider theoretical framework, a set of assumptions about the
fundamental social, economic and cultural relationships which provide the
dynamic for every human society, the generative syntactic structures, to
borrow once more from linguistics, which constitute the grid within which
social-cultural causation is to be understood. Such terms also imply a
particular epistemology, of course, in this case a realist and materialist
philosophy which provides a framework for and limitations to the ways in
which knowledge of the world, past or present, can be appropriated and
employed. But that is another story.
 CHAPTER I

The background: state and society before

Heraclius

It is all too easy to forget that every human society, whatever its achieve-
ments and the ways in which it impresses itself upon its natural context, is
closely bound to its geographical and climatic conditions of existence in
ways which may at first seem insignificant, or so obvious as to need no
further consideration. Not only methods of agriculture and the production
of social wealth, but modes of dress and the technology of clothing, for
example, are subject to these conditions, although it is not always possible
to determine exactly how the relationship operates. This is no less the case
for the late Roman state and for late Roman society and culture, which
occupied and dominated the east Mediterranean basin, the Balkans and
much of the north African littoral up to the middle of the seventh century.
The many local cultures and their histories which were incorporated into
that state and were subject to its administrative and political machinery
consequently varied very greatly in their appearance one from another,
inasmuch as the geographical and climatic features of the zones in which
they were located varied. It is useful to emphasise this perhaps obvious
point at the outset. For while we shall be dealing with the general social,
economic and cultural history of the Roman and Byzantine world during
the seventh and early eighth centuries, it is important to remember that
beneath the uniformity often imposed by contemporary descriptions, the
apparatuses of the state and its official ideology, there continued to exist a
variety and range of local cultures and of ways of doing and thinking
which, while they may not necessarily have been in conflict with the
umbrella of Roman and state culture, official orthodoxy and imperial
ideology, were nevertheless often very different from one another and from
the urban society and culture of the capital.1 It is perhaps easier to see this
1
See for example P. Charanis, 'Observations on the demography of the Byzantine empire',
Xlllth International Congress of Byzantine Studies, Main Papers XIV (Oxford 1966), pp. 1-19;
and the essays in P. Charanis, Studies on the Demography of the Byzantine Empire (London
1972). See also the important discussion in J. Koder, Der Lebensraum der Byzantiner
(Darmstadt 1984).
10 Byzantium in the seventh century

during the sixth century, when Egypt, along with the wealthy elite cities of
the east Mediterranean, such as Antioch or Tyre, clearly represented
(beneath their common Hellenistic veneer) cultures of very different origins
and appearance. But even in the later seventh century, when the empire
consisted of little more than central and western Anatolia and Thrace,
local differences and cultural variation continued to play a role.
Climatic variation within the empire was considerable. The political
world of the later Roman state was dominated by three land masses:
the Balkans, Asia Minor and the Middle East zone as far east as the
Euphrates and as far south as Egypt; by the two seas which both
separated and united them, the east Mediterranean and the Aegean, and
the Black Sea; and by the geographically, but by no means politically,
peripheral areas of Italy and the North African possessions of the empire.
From the middle of the seventh century it is Anatolia and the Balkans
that dominate, as the remaining territories are progressively lost; and
from this time Italy and the Ionian Sea, on the one hand, and the Crimea
and its hinterland, on the other, set the outer limits to what was there-
after 'Byzantine'.
Before the loss of these eastern lands, the empire as a whole was
relatively rich in exploitable arable land and had access to regular and
(usually) sufficient supplies of grain. Egypt, Thrace and the coastal plains of
north-west and west Anatolia were all grain-producing areas.2 But the loss
of Egypt involved a major re-adjustment and an increased dependency of
Constantinople on Thrace and the Anatolian sources, supplemented by
produce from the city's own immediate hinterland. Of the three land
masses referred to already, therefore, it is the Balkans and Anatolia which,
after the loss of Egypt, determine the parameters within which the Byzan-
tine economy, in the widest sense of the word, had to function. A recent
estimate has put the total revenue loss to the empire in the seventh century
at something in the order of seventy-five per cent of its sixth-century
income - a dramatic loss which must have affected all aspects of the state
administration fundamentally.3 It was from this much-reduced area that
the state now drew its revenues and manpower resources, and it was these
areas - the Balkans and Anatolia - which had to be defended. It was the
requirements of defence and revenue extraction within these areas which
moulded the administrative apparatus of the state and which we shall look
at in greater detail below.

2
Procopius. Historia Arcana XXII, 137 (Thrace, Bithynia, Phrygia); cf. M. Rostovtzeff, art.
•frumentum', RE VII/1, 126-87, esp. 129 and 137.
3
M.F. Hendy, Studies in the Byzantine Monetary Economy c. 300-1450 (Cambridge 1985),
pp. 619. 625-6.
 State and society before Heraclius 11

Roman and Byzantine society was predominantly rural and agricultural.

Settlements - villages and cities - were supplied almost without exception
from their immediate hinterlands as far as all essential items - food, drink,
clothing, the raw materials of everyday life, housing - were concerned.
Only the larger cities, or the wealthiest individuals, had the resources to
import goods, and these were chiefly luxury items for the landed elite who
dwelt there.4 Only Rome and Constantinople imported foodstuffs over long
distances and on a large scale, and this was possible only because the state
took complete responsibility for the financial burden entailed. Before the
loss of Egypt, Constantinople drew its grain supplies almost entirely from
that region, occasionally turning to Thrace or the nearest regions of
Anatolia. After the 640s and 650s, it has been shown that Thrace,
Paphlagonia and to a degree the Pontus, and north-west Asia Minor -
Bithynia - were the chief sources of supply. Similarly, livestock, vegetables
and other commodities were procured from these areas; although it is
highly likely that, with the exclusion of its grain supplies, Constantinople
could support itself from its immediate hinterland during much of the
seventh century.
The nature and extent of long-distance trade was not dependent on the
demand from a few major urban centres alone, however, and the problem
of the extent of a trans-Mediterranean commercial network has led to a
rich discussion among economic and social historians of the middle and
later Roman periods. Wine, oil and ceramics (the latter often for functional
reasons of transportation) were widely exported in the late Roman period
from North Africa, for example. And these exports depended to a degree, at
least, on commercial demand and competitive prices. But it seems that the
extent of this trade was also related to the extent of imperially funded
shipping and transport; so that although an established pattern of com-
mercial exchange in certain goods (which may also have included textiles
and clothing, for example) did exist, it too was in several ways parasitic
upon the activities of the state.5
4
See A.H.M. Jones, The Later Roman Empire 284-602: A Social Economic and Administrative
Survey (3 vols., Oxford 1964), vol. I, p. 465, vol. II. pp. 770, 840f., 850; A.H.M. Jones, The
Greek City from Alexander to Justinian (Oxford 1940), pp. 259ff.; E. Kirsten, 'Die byzantin-
ische Stadt', in Berichte zum XL International Byzantinisten-Kongress (Munich 1958),
pp. 10f.; D. Claude. Die byzantinische Stadt im 6. Jahrhundert (Byzantinisches Archiv XIII,
Munich 1969), pp. 176f. For a general discussion with literature, see J.F. Haldon, 'Some
considerations on Byzantine society and economy in the seventh century', BF 10 (1985),
75-112, see 78-80; note also J. Teall, -The Byzantine agricultural tradition', DO? 25
(1971), 34-59; and for a detailed analysis, J. Durliat, De la ville antique a la ville byzantine. Le
probleme des subsistances (Collection de l'ecolefrancaisede Rome 136. Rome 1990).
5
See the detailed summary with literature in R.-J. Lilie, Die byzantinische Reaktion auf die
Ausbreitung der Araber (Miscellanea Byzantina Monacensia XXII, Munich 1976),
pp. 201-27; Hendy, Studies, pp. 49, 51 and esp. 561-4; J. Teall, The grain supply of the
Byzantine empire', DO? 13 (1959), 87-139. For the middle and early empire, see
especially K. Hopkins, 'Taxes and trade in the Roman Empire (200 B.C.-A.D. 200)', JRS,
12 Byzantium in the seventh century

This pattern of supply, of course, reflects the costs and availability or

otherwise of transport, as well as its efficiency. Transport by sea was
considerably less expensive than by land; it was also generally much
faster. This in itself can explain the reliance on Egyptian corn. But while
the exigencies of transport by land meant that long-distance conveying of
foodstuffs, either by cart or on pack-animals, was usually out of the
question, it does not mean that the empire was poor in such resources.
Rather, the expense of transport meant that settlement patterns tended to
conform to the limits of what was locally available - although this then
depended upon a number of variables, such as security, disease and
epidemic, climatic fluctuations (drought, flooding) and so on. Only the
major cities, with outlets to sea-borne transport, such as Constantinople or
Antioch, could extend beyond this.6 Population density was similarly
subject to these considerations.
Resource availability affected directly the distribution of military forces,
of course: where local resources were inadequate, or where the presence of
soldiers posed a clear threat to the livelihood of a tax-paying population
(and where this was perceived), supplies had to be brought in, at consider-
able expense, from further afield. Hence Justinian's establishment of the
Quaestura exercitus, which provided the Danubian forces, by sea, with
supplies from the Aegean islands and western Anatolia; and hence Mau-
rice's ill fated attempts to have the same forces winter across the Danube in
hostile territory and away from the stricken and impoverished frontier
regions of Illyricum.7 As we shall see, the provision of supplies for large
concentrations of soldiers was always problematic, and the state had to
develop new ways of dealing with the difficulties this posed during the
seventh century.
The provision of livestock for transport and for military purposes also
70 (1980), 101-25; and for a summary of some aspects of the debate for the period from
the fourth to the sixth century, see CJ. Wickham, in ]RS 78 (1988), 183-93 (review of
A. Giardina, ed., Societd romana ed impero tardoantico III. he merci. Gli insediamenti (Rome
and Ban 1986); and the essays in Hommes et richesses dans VEmpire byzantin 1: IV-VIF siecle
(Paris 1989), and in The Byzantine and Early Islamic Near East V: Trade and exchange in the late
antique and early Islamic Near East, eds. L.A. Conrad, G.R.D. King (Studies in Late Antiquity
and Early Islam 1, V. Princeton forthcoming).
6
Jones, LRE, vol. II, pp. 712ff. and map V; pp. 841-7; Haldon, Some Considerations, 79-80.
For the distribution of cities relative to their geographical hinterland, see Hendy, Studies,
pp. 78-90 (Balkans) and pp. 90fT. (Anatolia). The technology of agricultural production is
equally important in this context, of course, although I will not discuss it in detail here. For
good general surveys, see K.D. White, Agricultural Implements of the Roman World (Cam-
bridge 1967); K.D. White, Roman Farming (London 1970); and esp. A.A.M. Bryer, 'Byzan-
tine agricultural implements: the evidence of medieval illustrations of Hesiod's Works and
Days', ABSA 81 (1986), 45-80. See also Jones, LRE, vol. II, pp. 767ff.; Koder, Der
Lebensraum der Byzantiner, pp. 55-7.
7
Hendy. Studies, p. 404: Jones, LRE, vol. I, p. 482, vol. II, pp. 661, 844 (quaestura exercitus),
678 (Maurice); see also W.E. Kaegi, jr., Byzantine Military Unrest 471-843: An Interpreta-
tion (Amsterdam 1981), p. 110.
 State and society before Heraclius 13

played a significant role. Mules and horses were crucial to the imperial
forces at all times, although little is known of the ways in which they were
raised, maintained or supplied. Imperial estates in Asia Minor, particularly
in the provinces of Phrygia, Lydia, Asia and Cappadocia, were important
in the ninth century and after, as well as in the later Roman period. Both
types of animal were raised and, in the ninth and tenth centuries at least,
for which the evidence is reasonably clear, the regulations governing their
issue were complex and strict. But the state relied heavily from the seventh
century, if not already before, on the raising of animals by levy on
landowners and estates. From the later seventh century (at the earliest),
soldiers supported by state subsidies (fiscal exemptions, for example) of one
sort or another, provided their own mounts. Since both horses and
donkeys, and consequently mules, were indigenous throughout the
empire, this will have been a ready source of such animals in times of
need.8
Apart from its resources in arable products and livestock, and the
derivatives therefrom (leather, horn, glue and felt, as well as food products)
the lands of the later Roman state also provided a variety of metal ores,
both precious and base. In Anatolia, the Caucasus region and to a degree
also the Taurus, along with the eastern Pontus, were sources of gold,
silver, lead and iron, as well as copper; in the Balkans, especially in the
inner mountain regions, the same ores were to be found. But the evidence
is scarce for the Byzantine period proper, and while iron, for example, was
certainly extracted from several areas throughout the period with which
we are concerned, political control over these areas was often uncertain.
The availability of iron for weapons and farming implements may at times
have been very problematic; and Byzantine regulations concerning the
export and import of gold are to be seen also in this context, as well as in
the wider context of international supply and demand in precious metals
and the demands of state finance.9

Late Roman society presents an often confusing picture of landowners,

senators, freedmen, slaves, semi-servile and free peasants, state and
Church officials, and soldiers of differing statuses. It is possible to describe
8
For the stud-farms in Cappadocia, see Jones, LRE, vol. II, p. 768; for imperial stud-farms
in Asia, Phrygia and Lydia in the eighth century, see J.F. Haldon, Byzantine Praetorians:
An Administrative, Institutional and Social Survey of the Opsikion and Tagmata, c. 580-900
(Poikila Byzantina HI Bonn 1984), p. 597 and n. 988 (also for raising animals through
compulsory levies in the same period). Note also Hendy, Studies, pp. 311, 610-12.
9
The most accessible general surveys are those of Lilie, Die byzantinische Reaktion, pp. 255fT.
with literature and esp. the statistical chart at pp. 258-62; and Sp. Vryonis, jr., 'The
question of the Byzantine mines', Speculum 37 (1962), 1-17. See on later supplies for the
army Haldon, Byzantine Praetorians, p. 321 and n. 978 (pp. 593-4). For late Roman
sources and evidence, Jones, LRE, vol. II, pp. 838-9. On gold see most recently Hendy,
Studies, pp. 257(T.; and Koder, Der Lebensraum der Byzantiner, pp. 59-61.
14 Byzantium in the seventh century

this picture in several different ways, and each of these will have a certain
validity according to the purpose of the description. The most obvious
approach is to begin from the sources themselves, and to see how these
different groups were described in juridical terms, in the legal sources, for
example. What definitions are used, how are these related to one another
and how are they related to the laws of property, the functioning of
Church and state administration, and to the emperor. Another approach
would be to look at the relationship of the various groups within society
to land and to property, and the ways in which these relationships deter-
mined their status. These two approaches are, of course, quite compatible,
indeed necessarily so if we are to have any chance of understanding how
the society actually functioned.10
In the following pages, I will look at the general shape of the late
Roman social formation and at the contradictions and tensions within it
which provide the foundation for the developments of the seventh
century.
The state itself was a complex structure of interlocking administrative
functions whose primary purpose, as it had developed over the preceding
centuries, was to defend and where possible extend imperial territory; to
provide for the maintenance and equipping of the soldiers and officials of
the administration and the imperial household; and to extract the neces-
sary resources in the form of taxation in kind, in cash and in services for
the fulfilment of these functions. At its head stood the emperor, the state
embodied and, in Christian terms, God's chosen representative on earth.
From the point of view of the 'official' ideology of state and Church, the
state represented also the realm of the Chosen People on earth, the Chris-
tians who had hearkened to the message of Jesus Christ. The emperor,
therefore, played a crucial role as leader and guide of this people, as
defender of right belief or orthodoxy, as intercessor with Christ for his
people and as protector of the rights of Christians everywhere. It was also
incumbent upon him, where conditions permitted, to extend the territory
of the empire and thus promote the orthodox faith; where physical poli-

10
The best general survey remains that of Jones, LRE; see also F.H. Tinnefeld, Die friihbyzan-
tinische Gesellschaft. Struktur - Gegensdtze - Spannungen (Munich 1977); and for the fiscal
administration of the state, see Hendy, Studies. Other accounts from different perspectives
include the excellent survey of P. Brown, The World of Late Antiquity (London 1971); the
essays in A. Momigilano, ed., The Conflict between Paganism and Christianity in the Fourth
Century (Oxford 1963); A. Piganiol, L'Empire chretien I (325-395). Histoire romaine,
vol. IV, part 2 (Paris 1947); E. Stein, Histoire du Bas-Empire, vol. I (Paris and Bruges 1959
and Amsterdam 1968). Stein's work parallels Jones, LRE, but provides a much fuller
political history and goes into a number of detailed questions more deeply; see also J. Vogt,
The Decline of Rome (London 1965) and P. Brown, The later Roman empire', in Brown,
Religion and Society in the Age ofSt Augustine (London 1972), pp. 46-73.
 State and society before Heraclius 15

tical incorporation was impractical, conversion of those outside the

bounds of the empire was an alternative strategy. 11
During the sixth century, the Emperor Justinian had, of all the emperors
since the extinction of the Western empire in the later fifth century, been
able to put the key elements of that political ideology into effect. The
reconquest of North Africa from the heretical Vandals, of Italy from the
Ostrogoths (short-lived though these successes ultimately proved to be) and
of parts of south-eastern Spain from the Visigoths were symbols of the
might of the Roman state and its ruler at their most powerful. They were
also clear expressions, and to some degree a vindication, of Roman claims
to world empire and to orthodoxy. 12 But they also reflected the different
fates of the eastern and the western halves of the empire during the fifth
century. The dominant position which the East Roman state enjoyed
throughout the sixth century, in spite of the constant conflicts along its
frontiers, was a result of very much longer-term developments, just as the
political and social structure of the Eastern empire represented the evolu-
tion of several centuries. During the fifth century the unified Roman world
empire, which stretched from Scotland to the Sahara and from Gibraltar to
the Euphrates, was replaced by a poly centric system of barbarian successor
kingdoms in the West, facing and contrasting with the cultural and
political unity, at least superficially, of the Hellenistic world of the East
Roman empire, in which the old order was still represented, however

11
For late Roman imperial ideology, see Jones, LRE, vol. I, pp. 321-9; 0. Treitinger, Die
ostromische Kaiser- und Reichsidee nach ihrer Gestaltung im hofischen Zeremoniell (Jena
1938); the essays in H. Hunger, ed., Das byzantinische Herrscherbild (Wege der Forschung,
vol. CCCXXXXI, Darmstadt 1975); J.F. Haldon, 'Ideology and social change in the seventh
century: military discontent as a barometer', Klio 68 (1986), 139-90, esp. 139-61;
F. Dvornik, Early Christian and Byzantine Political Philosophy (2 vols., Washington D.C.
1966), vol. II, pp. 614ff., 652f.; A. Pertusi, Insegne del potere sovrano e delegato a
Bisanzo e nei paesi di influenza bizantina', in Simboli e simbologia nell alto medioevo
(Settimane di Studio del Centro Italiano di Studi sull'alto Medioevo XXIII, Spoleto 1976),
pp. 481-563; N.H. Baynes, 'Eusebius and the Christian empire', A1PH0S 2 (1934),
13-18 (repr. in Baynes, Byzantine Studies and Other Essays (London 1955), pp. 168-72).
12
See Jones, LRE, vol. I, p. 270. But this is not to say that Justinian's reign was free of
cultural and ideological conflict and contradictions. On the contrary, while acceptance of
the imperial system provided the elements of uniformity and harmony at some levels,
there existed a range of critical and oppositional elements which were clearly expressed
through the literature of the period. Although such criticism seems to have been limited,
on the whole, to literate elements of society, especially among those connected with the
administration of the state or the palace (such as Lydus or Procopius, for example), it did
on occasion find echoes in popular hostility, as in the Nika riots; or less openly, in the
rumours and stories about Justinian being either a secret heretic or indeed a tool of the
devil himself. For the best modern account of some of these tendencies, see Cameron,
Procopius, pp. 242ff.; and the detailed analysis of Chr. Gizewski, Zur Normativitdt und
Struktur der Verfassungsverha'ltnisse in der spdteren romischen Kaiserzeit (Munchener
Beitrage zur Papyrusforschung und Antiken Rechtsgeschichte LXXI, Munich 1988),
pp. 131-47.
16 Byzantium in the seventh century

dimly. The reasons for the ultimate survival of the eastern half of the
empire and the disappearance of the western half are complex and still
debated. But the higher degree of urbanisation in the East, its apparently
greater resources in both materials and in manpower, its greater agri-
cultural and commercial wealth, its more deep-rooted, albeit very diverse,
cultural identity and, not least, the less disruptive and factional effects of
the senatorial aristocracy in the government, and on the economy as a
whole, must lie at the heart of any answer which we might offer. A more
effective bureaucracy, more able efficiently to extract the resources and
revenues necessary for the state's survival, was also a fundamental factor.
Thanks to its favourable cultural and political situation and a sound
economic base, the Eastern empire was able to withstand and to stave off
the disasters that befell the West. State unity and political stability were
preserved.13
It was on this foundation that Justinian was able to launch his massive
programme of reconquest. His achievements cannot be denied. The empire
extended at the end of his reign once more across the whole Mediter-
ranean; the Church of the Holy Wisdom marks one of the first and greatest
achievements of Byzantine art and architecture; the Corpus Iuris Civilis
established the basis for the development of European law. The essential
prerequisites for these achievements were, as we have seen, the strong
economic base of the Eastern empire, the effective repulse of the barbarians
to north and west, together with the fragile internal stability of the
successor kingdoms. But Justinian cannot be mentioned without reference
also to his gifted subordinates and associates - Belisarius, Narses, Tribo-
nian, John the Cappadocian and last, but certainly not least, his consort,
the Empress Theodora. Her dubious background, her intrigues and her
nepotism, quite apart from her questionable and, indeed, potentially trea-
sonable monophysite tendencies, were enough to inspire the historian
Procopius to write one of his most vicious caricatures in the Secret History.
But even he could not deny her political insight and ability, nor her

13
The most useful account of these developments from a political history viewpoint is Jones,
LRE, vol. I, pp. 173ff., esp. also pp. 1025ff.; see also Jones, The Roman Economy: Studies in
Ancient Economic and Administrative History, ed. P. Brunt (Oxford 1974), esp. chapters 4
('Over-taxation and the decline of the Roman empire', originally in Antiquity 33 (1959),
39-43), 8 (Taxation in antiquity') and 9 (Inflation under the Roman empire', originally
in Econ. Hist. Review 5 (1953), 293-318); and for a useful general survey, P. Anderson,
Passages from Antiquity to Feudalism (London 1974), pp. 97-103 with literature. See also
Brown, The World of Late Antiquity, esp. pp. 42-4; and W.H.C. Frend, 'The monks and the
survival of the East Roman empire in the fifth century', Past and Present 54 (1972), 3-24;
and N.H. Baynes, 'The decline of the Roman power in western Europe: some modern
explanations', JRS 33 (1943), 29-35 (repr. in Baynes, Byzantine Studies and Other Essays,
pp. 83-96); also E. Stein, 'Paysannerie et grands domains dans l'empire byzantin',
Recueils de la societe Jean Bodin II: Le Servage (Brussels 1959), pp. 129-33.
 State and society before Heraclius 17

stubborn determination and pride, a pride which saved the emperor

himself and his authority at the height of the Nika riot: 'For one who has
been an emperor', she is reported to have said, It is unendurable to be a
fugitive. May I never be separated from this purple, and may I not live that
day on which those who meet me shall not address me as mistress.'14
Justinian and Theodora are two among many examples of social mobility
within late Roman society.15 Thanks to his uncle Justin (518-27) Justi-
nian received the best of educations and, during the reign of the former, an
intimate acquaintance with diplomacy, statecraft and palatine politics. He
mastered the intricacies of the bureaucracy with an eye for detail and with
great energy. His insistence on determining personally every aspect of state
policy, down to the tiniest detail, was even in his own time a legend. He
was known as 'the emperor who never sleeps'. He was an autocrat, and he
tried to realise the ideals of imperial office and the imperial ideology.16
Justinian's great ambition was the renovatio imperil the restoration of
the world empire of Rome. The requirements for this task were consider-
able: the reconquest of lost territories; the establishment of Chalcedonian
orthodoxy and religious unity throughout the empire; the reorganisation
of the administrative and judicial machinery; a planned economic policy
designed to support the increased demands of his military undertakings;
and a grand scheme of building and renewal in both civil, ecclesiastical
and military spheres. Such policy aims were inevitably quite unrealisable
in a number of respects. The partial reorganisation of the administration
and the fiscal establishment of the state, the partially successful policy of
reconquest, the vast expense of the long-drawn-out wars, especially in
Italy, and the partial nature of the building programmes, all point to
ultimate failure. The resources available were simply not adequate to the
task. Apart from this, the practical realities of sixth-century politics and the
actual strengths of the successor kingdoms in the West cannot be ignored.
The fundamental principle enunciated from Constantinople, of a single
legitimate emperor and empire, was as taken for granted as the notion of a
single Christian Church. Even Germanic rulers recognised the emperor at
Constantinople as the highest source of legitimacy. But their idea of
imperial authority did not involve actual reincorporation into the political

14
Procopius, De hello Persico I, 24.331". (trans. H.B. Dewing).
15
Numerous accounts of Justinian's reign have appeared. See in particular Stein, Bas-
Empire, vol. II, pp. 275-780; and Jones, LRE, vol. I, pp. 269-302, for detailed analyses
with sources and literature. See also R. Browning, Justinian and Theodora (London 1971);
B. Rubin, Das Zeitalter lustinians I (Berlin 1960); J.W. Barker, Justinian and the Later Roman
Empire (Madison 1960).
16
See, for example, Procopius, Historia arcana XII, 27; xm, 28flf. On Justinian's character and
personality see the short account in Stein, Bas-Empire, vol. II, pp. 275-83; Jones, LRE,
vol. I, pp. 269ff.; and M. Maas, John Lydus and the Roman Past (London 1992).
Map I The empire in A.D. 565: approximate extent
 State and society before Heraclius 19

Plate 1.1 Justinian. Copper follis

framework of the Roman state, and Justinian's attempts to turn ideological

theory into pragmatic politics met with universal, if not always par-
ticularly successful, opposition. A second element of the ideological equa-
tion was likewise not to be realised in practical terms; for while Justinian
saw his role and that of any Roman emperor as entailing the liberation of
orthodox subject-populations in the West from the rule of heretics, the
practicalities of finance and politics meant that he was unable to put his
theory into practice. But his position was made abundantly clear in several
contemporary statements, notably at the beginning of the Corpus Iuris
Civilis:
Through the might of God we rule the empire, which was passed on to us by the
heavenly majesty; we wage wars with success, secure peace, and we maintain the
edifice of the state. We likewise exalt our spirit in contemplation, through the help
of the almighty Godhead, so that we place our trust not in our arms, nor in our
soldiers, nor yet in our own abilities; but we found all our hopes in the all-seeing
protection of the all-highest Trinity alone, from which derive all elements of the
Universe which lend order to the whole world.17
It was the precarious stability of the successor kingdoms which gave
Justinian the opportunity to promote his policy of renovatio on the foreign
political plane. Both the Vandal and the Ostrogothic kingdoms were rent
by actual or latent conflict, and they were in any case unable to offer any
co-ordinated opposition to Eastern Roman attack.18 But the empire faced
17
Justinian, Digesta, 'Constitutio Deo Auctore' (CIC I, p. 8).
18
On the successor kingdoms, see esp. E.A. Thompson, 'The barbarian kingdoms in Gaul
and Spain', Nottingham Medieval Studies 7 (1963); C.J. Wickham, Early Medieval Italy
20 Byzantium in the seventh century

powers other than the relatively recently formed states of the West. The
Sassanid Persian empire presented a constant threat to the provinces of the
Roman East. Persian political ideology likewise provided its rulers with an
instrument and an excuse for aggression, namely the long-nurtured hope
of recovering all the territories which had once belonged to the great
empire of the Achaemenids destroyed by Alexander.19 After many years of
relative peace, the reign of Justinian saw an increase in hostilities between
the two powers. Chosroes I (531-79), Justinian's contemporary and equal
in statecraft and organisational enterprise, pushed through a series of
administrative and military reforms that brought the Sassanid state to the
height of its power and influence. War broke out in the East in 527, but
lasted only briefly until in 532 an 'everlasting' peace was agreed. This was
crucial to Justinian's plans, for East Roman resources were not adequate to
the task of funding warfare in East and West simultaneously. Roman
diplomacy played a crucial role in averting this eventuality - through
'subsidies', tribute, territorial or trading concessions, as well as through
espionage, intrigue, the fomenting of unrest among client states and so on.
Buffer states and client princes - from the Ghassanids in the Syrian desert
to the Christian Armenian principalities in the Caucasus - were supported
by border garrisons and settlements and protected the length of the
north-eastern and eastern frontiers, from the Black Sea down to the
Arabian desert. But the 'eternal' peace did not last long: in 540 renewed
Sassanid incursions took place, aimed primarily at extracting tribute and
booty from the cities of the eastern provinces. Those that refused to pay up
- such as Antioch - were sacked. Chosroes was eventually bought off in
545; and a second truce was signed in 551, although desultory conflicts
continued in the northern zone (where Romans and Persians were keen to
control the Armenian principalities and the districts of Lazica and Suania,
out of both strategic interest and on account of the mineral wealth of the
region) until in 561 a ten-year truce was arranged. This was not to last
much longer than Justinian himself.
The course of the wars of reconquest is too well known to need more
than a brief survey here.20 Between 533 and 534 Vandal Africa was
(London 1981); J.M. Wallace-Hadrill, The Barbarian West 400-1000 (Oxford 1966); and
J.M. Wallace-Hadrill, The Long-Haired Kings (London 1962); M. Wes, Das Ende des Kaiser-
tums im Westen des romischen Reiches (The Hague 1967); E.A. Thompson, The Goths in
Spain (Oxford 1969); ^.A.B. Llewellyn, Rome in the Dark Ages (London 1971). There is, of
course, a much wider literature, but further bibliography will be found in these works. See
also C. Courtois, Les Vandales et VAfrique (Paris 1955), pp. 353ff.
19
See A. Christensen, L'Iran sous les Sassanides, second edn (Copenhagen and Paris 1944);
and R. Frye, The Heritage of Persia (London 1963) for a general survey.
20
Jones, LRE, vol.1, pp. 2 7 1 - 8 , 287-94; and Stein, Bas-Empire, vol.11, pp. 283-96,
311-18, 3 3 9 - 6 8 , 485ff., 560ff., provide detailed narrative accounts. See further
J.B. Bury, The Later Roman Empire 395-565 (2 vols., London 1923), pp. 124ff.
 State and society before Heraclius 21
returned to imperial control. But these wealthy provinces were to prove
almost as difficult a problem as were the Vandal rulers who had been
overthrown. Constant attacks from the Berber tribes in the interior meant
constant military preparedness, the construction of forts and a consider-
able drain on resources.
In 533 Belisarius began the reconquest of Italy from the Ostrogoths.
Sicily fell quickly. But the Romans underestimated both Ostrogothic oppo-
sition and the disastrous effects of continuous warfare on the Italian
countryside and population; apart from this, the constant need for troops
on the eastern front, deliberate underfunding and undermanning by Justi-
nian for fear of rebellion by over-successful generals, particularly Belisa-
rius, and lack of resources and irregular pay for the troops meant that the
war dragged on until the 550s. Indeed, the last Gothic garrisons surren-
dered or were destroyed only in 554. Simultaneously with these last
actions, Roman forces wrested control of south-east Spain from the Visi-
goths (in 552), thereby increasing the empire's strategic control of the
western Mediterranean, but extending also the limited resources at its
disposal.21
A possible third front in the Balkans did not materialise until towards the
end of Justinian's reign, in 559 and after.22 The raids of Slavs and of Turkic
peoples across the Danube, which threatened Constantinople itself, seemed
to be little more than short-term problems. The arrival of the Avars north
of the Danube could not yet be perceived for the threat it was to become.
But the regular withdrawal of troops from Thrace and Illyricum to serve on
the eastern front or in Italy or Africa left these areas without adequate
protection. In consequence, the whole Balkan region, from the Adriatic to
the Black Sea and south to Thessaloniki and even Constantinople, was
subject to regular devastation - the economic consequences for the state
are apparent.23
For contemporaries, however, whatever the criticisms voiced by Proco-
pius or a faction of the senate at Constantinople, the brilliance of Justi-
nian's achievements and the scope of his programme concealed the dis-
crepancy between reality and ideological wish-fulfilment.24 The real

21
See P. Goubert, 'Byzance et l'Espagne wisigothique', REB 2 (1944), 5-78; P. Goubert,
'1/Administration de l'Espagne byzantine I\ REB 3 (1945), 127-42; 'IT, REB 4 (1946),
71-134. See also F. Gorres, 'Die byzantinischen Besitzungen an den Kiisten des spanisch-
westgotischen Reiches', BZ 16 (1907), 515-38.
22
See, for example, Stein, Bas-Empire, vol. II, pp. 535-40.
23
See Stein's account, Bas-Empire, vol. II, pp. 521-5, 541-5; Jones, LRE, vol. I, pp. 293.;
Obolensky, The Byzantine Commonwealth, pp. 44ff.; G. Ostrogorsky, Geschichte des byzanti-
nischen Staates (Munich 1963), pp. 68-70.
24
See F. Tinnefeld, Kategorien der Kaiserkritik in der byzantinischen Historiographie (Munich
1971), pp. 191ff., and J. Irmscher, 'Justinianbild und Justiniankritik im friihen Byzanz', in
22 Byzantium in the seventh century

situation became clear to most only during the reigns of his successors,
who had to struggle with a situation they had inherited and on the whole
did so with only limited success. The reconquest of the West had far-
reaching consequences for the history of Western Europe and Italy as well
as for Byzantium. The continued Byzantine presence in Italy, for example,
the destruction of the Ostrogothic kingdom, the arrival of the Lombards
(which the Byzantines, to a certain extent, brought upon themselves), all
contributed towards the development of the medieval papacy. Most impor-
tantly for Byzantium, however, the state which Justinian bequeathed to his
successors had only limited resources to cope with the enormous problems
it now faced in its over-extended imperial possessions. It was inevitable
that the structure should collapse at some point.

The state and its apparatus

Justinian's internal policies were, no less than his reconquests, determined
largely by his ideology of renewal. But behind the facade of world conquest
lay considerable social and economic difficulties. The Nika riot in 532,
whatever its immediate background and the political motives of those
members of the senate who tried to use the hippodrome factions for their
own ends, is evidence of longer-term and deeper-seated social discontent
and alienation. 25 Justinian was certainly aware of some of these problems.
His intention was to reform the administration, strengthen the rural
economy and establish unity of belief. As his many novellae and edicta
illustrate, he knew of the failings and conflicts within the administrative
machinery of the state. He clearly felt that part of his task lay in the
establishment of a just society for his subjects, a task with which he had
been privileged by God.26 But he did no more than reform - the system of
government and administration in all its complexity and with all its
weaknesses was the system he inherited from the past. His task was merely
to make it function properly, justly, efficiently; not to change it in any
fundamental way.
The late Roman state had the structure of an autocratic absolutism, a
structure which had evolved gradually during the long history of the
principate, culminating with the establishment of the dominate under
Diocletian and his successors. Nothing, according to Justinian's express
H. Kopstein and F. Winkelmann, eds., Studien zum 7. Jahrhundert in Byzanz. Probleme der
Herausbildung des Feudalismus {BBA XXXXVII, Berlin 1976), pp. 1 3 1 - 4 2 .
25
E. Patlagean, Pauvrete economique et pauvrete sociale a Byzance, 4e-7e siecles (Paris 1977),
pp. 215ff.; Stein, Bas-Empire, vol. II, pp. 449ff. for the Nika riot itself.
26
Jones, LRE, vol. I, pp. 2 6 9 - 7 0 ; Stein, Bas-Empire, vol. II, pp. 277ff.; and esp. M. Maas,
' R o m a n history and Christian ideology in Justinianic reform legislation', DOP 4 0 (1986),
17-31.
 State and society before Heradius 23

view, was greater or more sacred on earth than the imperial majesty.27
Titles, symbols of office and the imperial court ceremonial served to express
and to enhance the God-given nature of imperial authority and power. The
ideological system in which the formal political system was given its
rationale had grown slowly out of a synthesis of Christian and Roman
imperial traditions, both in their turn determined by the philosophical and
symbolic traditions of the late ancient world. The 'objective idealism' of late
ancient philosophical thought, which can be followed in the thinking of
Plotinus and Philon through to the hierarchical world-view of Pseudo-
Dionysius, together with the conflict between both Christian theology and
Hellenistic philosophy on the one hand, and within the Church itself on the
other, constitute the background and determining context for this evolu-
tion. 28 Court, civil administration, justice and the army depended upon the
emperor, from whom they received their legitimation and their com-
petence. The political system, with it formal ideology and its assumption of
God-given jurisdiction provided a focus for unity in a culturally, linguisti-
cally and economically diverse world, in a way that few autocracies have
succeeded in doing. 29 The 'price' of this, of course, lay in an absolute
political orthodoxy which permitted no open criticism of the system or its
principles as such, and which also involved a far-reaching state interven-
tion in economic and demographic matters in the interests of the mainte-
nance of resources and the ability of the state to defend its territories and
revenues.
The complex administrative establishment, together with the military
establishment and the armies, constituted the heart of the state struc-
ture. 30 The system was based upon the principle of centralisation, of a
2
7 CII, 14.12.
28
See especially the discussion with literature of Averil Cameron, Images of authority: elites
and icons in late sixth-century Byzantium', Past and Present 84 (1979), 3-35, see 6ff.; also
0. Kresten, 'Iustinianos I, der "Christusliebende" Kaiser', Romische Historische Mitteilun-
gen 21 (1979), 83-109; S.S. Averincev, in chapter 2 of Z. Udal'cova, ed., Kul'tura
Vizantii, IV-pervaja polovina VII v. (Moscow 1984); and esp. J. Shiel, Greek Thought and the
Rise of Christianity (London 1968), pp. 18ff., 3 9 - 8 9 .
29
See the discussion and comments of C. Capizzi, 'Potere e ideologia imperiale da Zenone a
Giustiniano (474-527)', in L'Imperatore Giustiniano, storia e mito, ed. G.G. Archi (Circolo
Toscano di Diritto Romano e Storia del Diritto V, Milan 1978), pp. 3-35 with further
literature.
30
The best detailed surveys of the administrative and fiscal structures of the late Roman
state can be found in Stein, Bas-Empire, vol. II, esp. pp. 419ff., 735ff.; Jones, LRE, vol. I,
pp. 2 7 8 - 8 5 , with 32 Iff., a general analysis of the state's government and administrative
structures from the fourth to the sixth centuries; see further Hendy, Studies, pp. 371ff. and
notes 1 and 2 with literature; A.E.R. Boak, The Master of Offices in the Later Roman and
Byzantine Empire (New York 1919), in A.E.R. Boak and J.E. Dunlop, Two Studies in Late
Roman and Byzantine Administration (New York 1924); M. Clauss, Der Magister Officiorum
in der Spdtantike (4.-6. Jahrhundert): das Amt und sein Einfluss auf die kaiserliche Politik
(Munich 1980).
24 Byzantium in the seventh century

division between civil and military spheres of competence, and of a merit-

ocracy in its staffing. Its officials underwent a regulated training (although
in the very loosest sense), and its personnel occupied posts which were
arranged in a fixed hierarchy of authority and status, with exact func-
tions, competences and duties attached to each position. The sixth-century
writer John Lydus has left us a vivid picture of the operation of this system,
of its evolution according to contemporary understanding, and of the
esprit de corps (or lack of it) among its members. His writings represent
probably the best and most illuminating examples of an insider's view on
the Roman state administrative machinery, with all its strengths and
weaknesses, and they provide invaluable evidence for the ways in which
the whole structure actually worked. 31

The state and the Church

Justinian's interest in religious and Church affairs forms an integral part of
his approach to governing the empire. The successes of the state depended,
as we have seen, upon God's favour and, therefore, upon the strict obser-
vation of right belief in all senses: whether in the administration of Church
and clergy, or in the observance of the Chalcedonian creed and the appli-
cation of the principles enunciated through it. Unorthodox belief was a
direct challenge to the emperor's authority and position, as well as to God;
it was likewise a threat to the stability of the state. 32 Throughout his reign,
Justinian attempted to bring conformity of belief and creed to the empire,
both by persuasion as well as by force - conversion, threat of permanent
exile, death. Clashes of opinion were often resolved ultimately in the
crudest of ways, as was demonstrated in the so-called Three Chapters con-
troversy, in which Justinian eventually got his way through bullying and
bribery. 33 The greatest problem, however, one which no emperor suc-
ceeded in solving, was that of the Eastern monophysites, who formed a
large part of the populations of Eygpt and Syria, but existed throughout
the east, and whose adherence to their beliefs constituted a direct chal-
lenge to imperial authority. Persecutions and forced conversions failed dis-
mally to resolve the issue, and it ceased to be directy relevant only when

31
loannis Lydi De Magistratibus Populi Romani Libri Tres, ed. R. Wiinsch (Leipzig 1903); and
the commentary of T.F. Carney, Bureaucracy in Traditional Society: Romano-Byzantine
Bureaucracies Viewed from Within (Lawrence Kan 1971).
32
See, in general, Jones, LRE, vol. I, pp. 2 8 5 - 7 ; Stein, Bas-Empire, vol. II, pp. 3 6 9 - 4 0 2 ,
623-83 for a detailed treatment.
33
See esp. Stein, Bas-Empire, vol. II, pp. 632ff.; Jones, LRE, vol. I, pp. 2 9 6 - 7 ; D.J. Constante-
los, 'Justinian a n d t h e Three Chapters controversy', Greek Orthodox Theological Review 8
(1962-3).
 State and society before Heraclius 25

those territories were conquered by Islam in the middle of the seventh

century. 34
Justinian's interest and concern, while it may have been more intense
than that of many emperors, is certainly not to be seen as unusual,
however. For it is worth emphasising that it was only in and through the
vocabulary of Christianity, which described the 'symbolic universe' of the
East Roman world, that Justinian and his contemporaries were able to
apprehend their world and act in and upon it. Their symbolic universe -
their 'thought-world' - was by definition a 'religious' one, in which
human experience and perception of their world, both secular and
spiritual, had necessarily to be expressed through this religious vocabu-
lary. Politics are thus always 'religious', and religion is always 'political',
however implicit this may be; and for a Roman emperor, religious
matters were as much a key element of the everyday political and civil
administration of the empire as were the affairs of the fisc or the imperial
armies. It is important to stress this perhaps rather obvious point, since it
is all too easy to impose a division between 'religious' and 'political' or
'secular' in modern terms which bears little or no relation to either the
theory or the practice and understanding of late Roman or Byzantine
culture, and serves only to obfuscate the underlying structures of its belief
and practice.

The 'Corpus luris Civilis'

Only in the field of the law did Justinian succeed in carrying out one of his
major projects. The codification of the civil law of the empire, compiled in
the years 528-33, replaced all earlier codifications. The first section, the
Codex lustinianus, contained all imperial edicts and promulgations from
Hadrian to 533 which were regarded by the compilers as relevant; Justi-
nian's promulgations after this date are separately collected in the volume
of Novellae Constitutiones, drawn up after his reign. The second section, the
Digest or Pandects, consisted of a revised selection of comments from
Roman jurists, intended to complement the codex; while the third section,
the Institutiones, represented a sort of handbook and reference work for the

34
The best brief survey of religious politics, the relationship between state and Church, and
the role of the emperor in this sphere is to be found in F. Winkelmann, Die ostlichen Kirchen
in der Epoche der christologischen Auseinandersetzungen (5. bis 7. Jahrhundert) (Kirchenge-
schichte in Einzeldarstellungen 1/6, Berlin 1980), esp. pp. 103ff. Note also W.H.C. Frend,
'Old and new in Rome in the age of Justinian', in D. Baker, ed., The Relation between East
and West in the Middle Ages (Edinburgh 1973), pp. 11-28; also J. MeyendorfT, 'Justinian,
the empire and the Church', DOP 22 (1968).
26 Byzantium in the seventh century

imperial jurists who formed the backbone of the central administrative

apparatus. 35
The Corpus luris Civilis embodies in many respects the ideal world which
Justinian sought to achieve through his expansionist and reformist poli-
cies. The continued emphasis on the absolute power of the emperor,
holding his authority through God's will and grace - as well as the stress
on the Christian idea of the all-powerful ruler within yet prescribing the
law, in contrast to the classical tradition - is a clear demonstration of this.
In particular, the theories underlying much of Justinian's reforming and
administrative activity stand out. It is perhaps ironic that the theory of
renovatio survived in a practical form to influence the development of
European legal and constitutional history so deeply, yet its political results
failed to survive much more than a generation after the death of one of its
most successful exponents.

Social structure and economic relations

As has been stated many times, the basis of the later Roman state and its
economy was agricultural production. The greater part of the population
were rural, agricultural and represented a subsistence peasant economy.
Something like ninety-five per cent of the state's income was probably
derived from tax or other expropriations on the land. Agriculture was the
basis of state, Church and private wealth. Since the costs of inland trans-
port were high, trade in agricultural produce, except on the most highly
localised basis, was virtually non-existent. The only exception was for the
state-funded movement of grain, for example. Otherwise, local self-
sufficiency was the rule, and only where cheaper forms of transport were
available and a reasonable livelihood based on market exchange could be
assured was commercial monoculture practised - the production of oil or
wine, for example, where access to sea-borne trade was easy. It is in this
respect that the importance of the North African export of oil and wine, for
example, must be taken into account, as well as that of Egyptian goods
accompanying the state-funded shipping of grain to Constantinople.36
35
For general a n d m o r e detailed treatments, see Jones, LRE, vol. I, pp. 278ff. a n d Stein,
Bas-Empire, vol. II, pp. 4 0 2 - 1 7 ; a n d esp. P.E. Pieler, 'Byzantinische Rechtsliteratur', in
Hunger, Profane Literatur, vol. II, pp. 3 4 3 - 4 8 0 , see 4 1 Iff.
36
Jones, LRE, vol. I, p. 4 6 5 ; vol. II, pp. 7 6 9 - 7 0 for tax; a n d o n the agricultural population,
ibid., pp. 7 6 7 - 8 2 3 . For transport a n d trade, ibid., pp. 827ff., a n d Hendy, Studies,
pp. 554ff. For the Mediterranean trade in wine, oil a n d related products, see note 5 above.
There is a h u g e literature on the subject of late R o m a n social-economic structures.
General surveys in Jones, LRE a n d Tinnefeld, Die fruhbyzantinische Gesellschaft, pp. 1 8 - 5 8
(landowners a n d agricultural producers) a n d 5 9 - 9 9 (the senate); see also Stein, Bas-
Empire, vol. II. For a m o r e recent Marxist assessment, see G.E.M. de Ste Croix, The Class
Struggle in the Ancient Greek World (Oxford 1981), pp. 453ff. Debate on the role a n d
 State and society before Heraclius 27

The class structure of the late Roman social formation is, in spite of the
often confusing array of contemporary technical terms, reasonably clear.
The contrast between humiliores and potentiores, however nuanced by
social change or legal context, appears throughout the legislation of the
fifth and sixth centuries. 37 The dominant social group was represented by
a numerically relatively small class of land-owning magnates, for the most
part members of the senatorial aristocracy. Together with the Church,
which by Justinian's time had become a substantial landowner, and the
fisc - which in the form of imperial and state lands was also a major
landowner - this class exercised a more or less complete control over the
means of production of the empire. In opposition stood the vast mass of the
rural and urban populations, the greater part of which stood in relation-
ships of varying degrees of dependence or subservience to the landowners,
decline of slavery as a significant feature of Roman and late Roman relations of produ-
ction, the development of the colonate, of large senatorial estates or latifundia, and the role
of the state in the economy, has been lively, in both Soviet (and Eastern European) and
Western (both Marxist and non-Marxist) historiography. Soviet historians in particular
have been keen to emphasise the causes of the decline of the slave mode of production, the
rise of the colonate and dependent peasantry, and of feudal or proto-feudal relations of
production in the Roman period (second to fourth centuries and beyond), although the
debate within Soviet historiography has itself been lively - one school of thought
preferring to locate the origins of feudal relations of production in the later Roman period,
another placing it later, in the Byzantine period proper (ninth and tenth centuries). For
representative surveys, see E.M. Staermann, Krizis rabovladel'ceskovo stroya v zapadnykh
provinciyakh Rimskoi imperii (Moscow 1957) (German trans. W. Seyforth, Die Krise der
Sklavenhalterordnung im Westen des romischen Reiches (Berlin 1947)); Staermann, Krizis
anticnoi kul'tury (Moscow 1975), arguing for a synthesis of Roman and Germanic cultural
forms and social relations. See also the literature cited in Haldon, Some considerations, 99
and n. 59; and esp. Z.V. Udal'cova and K.A. Osipova, 'Tipologiceskie osobennosti feoda-
lizma v VizannT, in Problemy social'noi struktury i ideologii srednevekovogo obscestva I
(Leningrad 1974), pp. 4-28. This perspective has been criticised on the one hand by those
historians who reject the notion of synthesis altogether, such as M.Ya. Siuziumov,
'Zakonomernii perehod k feodalizmu i sintez', Anticnaya drevnost'i srednie veka 12 (1975),
33-53, and by those who see the structural impasse between ancient slavery and
developing feudal relations as being resolved by synthesis in the West and by the
continued existence of the centralised state in the East, such as G.G. Litavrin, Vizantiiskoe
obscestvo i gosudarstvo v X-Xl vv (Moscow 1977), and esp. A.P. Kazdan, Vizantiiskaya
kul'tura (Moscow 1968) (German trans. Byzanz und seine Kultur (Berlin 1973)). For the
Western literature, see esp. M.I. Finley, Ancient Slavery and Modern Ideology (London
1980); Finley, The Ancient Economy (London 1973); R. Remondon, La Crise de Vempire
romain (Paris 1964); the essays of M. Bloch, Slavery and Serfdom in the Middle Ages, Eng.
trans. W.R. Beer (Berkeley 1975); K. Hopkins, Conquerors and Slaves (Cambridge 1978);
M. Mazza, Lotte sociali e restaurazione autoritaria, second edn (Bari 1973), esp.
pp. 119-216; Anderson, Passages from Antiquity to Feudalism, pp. 18-103; and the
critiques of P.Q. Hirst, 'The uniqueness of the West', Economy and Society 4 (1975),
446-75 and CJ. Wickham, 'The other transition; from the ancient world to feudalism',
Past and Present 103 (1984), 3-36; see also 0. Patterson, 'On slavery and slave formati-
ons', NewLeft Review 117 (Sept.-Oct. 1979), 31-67; and de Ste Croix, The Class Struggle in
the Ancient Greek World, esp. pp. 205ff. and 453ff.
37
Patlagean, Pauvrete economique, pp. llff. with literature, discusses the terms and the
changes in meaning they underwent during the fourth to the sixth centuries.
28 Byzantium in the seventh century

whether institutional or individual. Of the agricultural population, the

great majority were coloni of varying status, tied in a more or less rigid way
to their holdings. Communities of free peasants continued to exist, often in
substantial numbers, especially in certain regions of the Balkans, Anatolia
and Syria. During the fourth, and particularly the fifth, centuries many
specific groups within the empire's population had been incorporated into
what has been termed the Zwangswirtschaft of the later Roman empire, that
is, the compulsory tying of individuals according to their status and
occupation to their functions. This affected particularly the urban curial
class, soldiers, members of some trading and craft occupations, who
became hereditary members of their various collegia or corporations. And
although the state seems in fact to have had only limited success in making
these restrictions and regulations work (for it is doubtful whether the
legislation, which seems anyway to have been intended for the West more
often than the East, could be effectively applied outside the state apparatus
itself) they have reinforced the impression that late Roman society was
tantamount to a caste system, a position which has now generally been
abandoned. 38 Certainly by Justinian's reign some of the better-known fixed
relationships - in the army, for example - seem to have been dissolved or
were ignored by the state. There is no evidence, for example, that the
comitatenses or field troops were still conscripted on this basis; although
those in the limitanei, where service in the unit was still viewed as a
privilege, were. 39
The taxes and revenues needed by the state were extracted in a variety of
ways. Extraordinary levies on foodstuffs or raw materials formed an
important element, especially in respect of the supplying and provisioning
of garrisons or of troops on the move. Similarly, corvees or compulsory
labour duties might be imposed on a community (and taken into account
in the regular tax-assessment) for the maintenance of bridges, roads and so
on. But the chief form of revenue extraction, carried out by officials of the
prefecture at the provincial tier, was the land-tax, based on the amount
and nature of the land, and the number of persons cultivating it. Referred
to in general terms in its late third- and fourth-century form as the annona
or, on the principle of assessment (area of land, number of persons farming
it) as the capitatio/iugatio assessment, this was an inclusive tax to which all
cultivated land was subject. The principles of assessment were straight-
forward. A unit of land, or iugum, was taxed only if a corresponding unit of
38
See Jones, LRE, vol. I, pp. 68-70, vol. II, pp. 861ff., and also Jones, 'The caste system in
the later Roman empire', Eirene 8 (1970), 79-96 (repr. in The Roman Economy,
pp. 396-418, with emendations and additions), and R. MacMullen, 'Social mobility and
the Theodosian code', JRS 54 (1964), 49ff.
39
J.F. Haldon, Recruitment and Conscription in the Byzantine Army c. 550-950: A Study on the
Origins of the stratiotika ktemata, in SBO 357 (Vienna 1979), pp. 20-8.
 State and society before Heraclius 29

manpower (caput) was available to cultivate it. Since both iugum and caput
were, in the first instance, notional units of assessment, and caput, indeed,
signified originally merely the liability for assessment, the two were some-
times used interchangeably. Contrary to the views of some historians,
however, neither caput nor capitatio denoted a poll tax, as has now been
convincingly demonstrated. 40 The system was not very flexible, and a
number of measures had to be introduced to deal with the problems which
resulted from demographic decline and consequent desertion of agri-
cultural land. The slow process of binding the agricultural labour force to
the land was one solution. Another was to make individuals and commu-
nities collectively responsible for adjacent land or land within their fiscal
census district, and consequently for the revenues due from such land, if it
were deserted or abandoned. This system, known as adiectio sterilium, or
emPoXr) (TCOV cnropcov) in Greek, was increasingly invoked as the sixth
century progressed, so that Procopius claims (albeit with some exagger-
ation) that the landowners of the empire were entirely ruined as a result. 41
Capitatio/iugatio assessment also provided the basis for the calculation of
the rate of conscription of peasants for the army. By the middle of the fifth
century, and certainly throughout most of the sixth century, the impo-
sition was commuted into a tax payable in gold, although it was not
necessarily always collected in this form. It was assessed and collected from
all free owners of land, whatever their wealth or status. Landlords were
held to be fiscally responsible for their slaves and for the adscripted or
semi-free tenants on their land (on these categories, see below). Increas-
ingly, as the colonate spread, and as more and more agricultural commu-
nities became tenants on larger estates, tax was paid to the landlord first,
from whom the state then collected it. During the sixth century, the state
regularly delegated some of its functions to private persons, who were
expected to carry out the appropriate tasks. The collection of the revenues
due from large estates seems regularly to have been attributed to the
landlords of such properties, including even the lands of the Church.
Similarly, there is some evidence to suggest that soldiers were often

40
The debate is complex because the technical terminology of the codes is often obscure and
contradictory. See W. Goffart, 'Caput' and Colonate: Towards a History of Late Roman
Taxation (Phoenix suppl. vol. XII, Toronto 1974) for a detailed discussion. While I do not
accept all of Goffart's conclusions, he quite rightly questions the traditional interpretation
of caput/capitatio as a revised equivalent of an early poll-tax.
41
See Procopius, Historia arcana xxm, 15-16; and the description and analysis of Jones,
LRE, vol. II, pp. 813-15. Fiscal responsibility was allocated by Justinian's time on the
basis of the land belonging, or having belonged, originally to a single owner or to the
same census district (eiripoXfi djjioSoiJXwv and emPoA/f) 6|xoxr|vaa)v respectively).
See esp. Justinian, Nov. 128, 7 and 8 (a. 545). For the technical terms see Jones, LRE,
vol. II, p. 815 n. 105; M. Kaplan, 'Les Villageois aux premiers siecles byzantins (Vleme-
Xeme siecle): une societe homogene?', BS 43 (1982), 202-17, see 206f.
30 Byzantium in the seventh century

maintained on a similar basis, their supplies and costs being deducted from
the tax revenue collected by the landlord responsible for the lands or
districts in which they were based. Other taxes - on trade and on commer-
cial activity - brought in only very limited revenue. As we have seen,
Jones' calculation of something like ninety-five per cent of total state
income from the land, while it may only be an approximate guess, gives
some idea of the dominant and crucial role of agricultural production to
the late Roman state. 42
The economic class-divisions within late Roman society are not difficult
to perceive. In contrast to the vast mass of the population, peasants
carrying on a subsistence agriculture (or, in coastal regions, for example,
fishing and related activities) with limited freedom of movement and with
limited rights of alienation in respect of their holdings, whether subject
directly to the fisc or to their landlords, stood the dominant class of landed
magnates and the service elite of the state bureaucracy, alongside the
Church and the state itself. The cities, while they certainly contained a
large population of free persons, offered only limited opportunities to those
of humble social origins, chiefly as craftsmen, occasional labourers or as
beggars - which, as in any pre-industrial social formation with a relatively
advanced social division of labour, was the regular occupation of a
number of people. The one obvious route out of poverty was to enter at
some level or other the state apparatus or the Church; or to win a position
in the retinue of a powerful person. And it is worth making the point here
that while a measure of social mobility did exist in late Roman and early
Byzantine society, it has often been given more prominence than it really
deserves: the individuals who succeeded in rising to positions of power and
authority received the attentions of their contemporaries precisely because
they were exceptions. Their total number is tiny. Late Roman society
represents a social formation in which class divisions were otherwise fairly
firmly drawn, and in which access to power through wealth and state
service excluded the great mass of the population from anything but the
subsistence economy of the peasantry.
It must be said, however, that the oppression of the agricultural popu-
lation and indeed the polarisation of interests between exploiting and
exploited classes was much clearer in the West than in the East. Again, the
42
In general, see Jones, LRE, vol. I, pp. 449-62, 464-9; vol. II, pp. 773-5; Jones, 'Capitatio
and iugatio', JRS 47 (1957), 88-94 (repr. in The Roman Economy, pp. 280ff.); Goffart,
'Caput' and Colonate; A. Cerati, Caractere annonaire et assiette de I'impot fancier au bas-empire
(Paris 1974); the essays in C.E. King, ed., Imperial Revenue, Expenditure and Monetary
Policy in the Fourth Century A.D. (BAR Int. Series 76, Oxford 1980); Stein, Bas-Empire,
vol. I, pp. 74ff., 441-3 and note 44. See also the comments of A. Guillou, 'Transformati-
ons des structures socio-economiques dans le monde byzantin du VIe au Vine siecle', ZRVI
19 (1980), 71-8, see 72ff. with sources and literature cited.
 State and society before Heraclius 31

reasons for this are complex and have received much attention. Peasant
revolts in the West in the fifth century are nowhere paralleled in the East
and do seem to reflect the higher degree of impoverishment and oppression
of the Western colonate, both at the hands of their landlords and of the
state. The Eastern landed elite, powerful as it was, never dominated both
urban and rural life and economic relations to the extent that seems to
have occurred in the West; power in the East remained more diffuse, the
state always managed to retain an effective fiscal control, and the rural
population, while oppressed and exploited, remained both more diverse
and heterogeneous in social and economic terms than in the West. 43 This
is not to say that peasant opposition to the state or to private oppression
was absent in the East. On the contrary, the existence of the so-called
Scamares in the Balkans, for example, in an economic and physical context
not too different from that of the Gallic Bacaudae (that is, state oppression
together with military and economic insecurity on a generalised scale), is
illustrative. Brigandage as a widespread form of communal opposition to
both state and private oppression is well attested during the sixth century
in Lycaonia, Phrygia, Pisidia and the Pontus. 44 While it may not reflect a
direct reaction to the oppression of a landlord, it does demonstrate the fact
that many communities regarded acceptance of state authority and the
consequent fiscal demands as unacceptably oppressive.45

Justinian's successors
The contradictions and weaknesses within this vast social and political
structure were made apparent by the last years of Justinian's reign and
during the reigns of his successors Justin II (565-78), Tiberius Constantine
(578-82), Maurice (582-602) and Phocas (602-10). 46 In the first place,
the conflict of interest between Chalcedonian orthodoxy, represented by
the state along with the Western and Anatolian provinces, and the mono-
physitism of a large part of the population of Syria, Egypt and eastern
43
See esp. E.A. Thompson, 'Peasant revolts in late Roman Gaul and Spain', Past and Present
2 (1952), 11-23 (repr. in M.I. Finley, ed., Studies in Ancient Society (London 1974),
304-20); de Ste Croix, The Class Struggle in the Ancient Greek World, pp. 474ff., a good
general survey of peasant and 'popular' opposition and of the degree of exploitation of the
rural population. See also note 4 above.
44
See Justinian, Nov. 24, 1 (a. 535, for Pisidia); Nov. 25, 1 (a. 535, for Lycaonia); Edict. 8,
proem.; 3. 1 (a. 548, for Pontus). Note also Justinian, Nov. 30, 7.1 (a. 536, for Cap-
padocia).
45
On the Scamares, see esp. A.D. Dmitrev, 'Dvizenie Skamarov', VV 5 (1952), 3-14; Jones,
LRE, vol. I, p. 294; vol. n, p. 656 on Anatolia.
46
For the political history of the period, see Jones, LRE, vol. I, pp. 303-17; Ostrogorsky,
Geschichte, pp. 66-72; Cambridge Medieval History, vol. iv, revised edn (2 parts, Cambridge
1966), I, pp. 28-30; P. Goubert, Byzance avant Vlslam (3 vols., Paris 1951-65); and Stein,
Studien.
32 Byzantium in the seventh century

Plate 1.2 Justin II. Gold solidus

Anatolia, was brought into the open more clearly than ever as the
ultra-orthodox Justin II - quite 'correctly' adhering to fundamental
elements within the imperial ideology - tried to impose orthodoxy through
mass persecutions and forced conversions. 47 Under Justinian, especially
while Theodora lived, imperial policy had been less strongly delineated,
although formal state opposition to monophysitism was quite clear. Now
the conflict broke out into the open again, revealing how precarious the
ideological unity of the state actually was.
In the second place, Justin's notion of his position and that of the empire
failed to take into account the real threat posed by the Avars in the north
and the Persians in the East. In his haughty refusal to continue paying the
subsidies to the Avar khagan which Justinian had consented to, he earned
the hostility of a powerful and extremely dangerous foe who, at the head of
large numbers of migrating Slav communities, was over the following
forty years to destroy much of Roman civilisation and political authority in
the Balkans and most of what is now modern Greece. While hostilities did
not commence immediately, and while Justin, through deft political
manoeuvring at the expense of the Gepids (settled in Pannonia Secunda,
south of the Danube) was able to recapture the Danubian fortress-city of
Sirmium, lost some thirty years earlier, the arrival of the Avars in the
Balkan-Danubian political forum had ultimately disastrous conse-
quences. 48 For the Lombards, settled to the north-west of the Gepids, had
asked the Avars for their assistence in their war with the latter. The result
was the destruction of the Gepids; but the Lombards, frightened by their
dangerous new allies, decided to emigrate westwards. In 568 they,
together with a number of other Germanic groups, marched into north-
east Italy, quickly conquering and occupying the plain of Venetia and most

47
See in particular Averil Cameron, 'The early religious policies of Justin IF, Studies in
Church History XIII (1976), 51-67; I. Rochow, 'Die Heidenprozesse unter den Kaisern
Tiberios I. Konstantinos und tylaurikios', in Studien zum 7. Jhdt, pp. 120-30.
48
For a useful summary, see A^Avenarius, Die Avaren in Europa (Bratislava 1974), esp.
pp. 67ff.
 300miles
Subject to Slav immigration
and Avar attack

1 Exarchate of Ravenna 6 Duchy of Rome

2 Liguria 7 Perugia
3 Duchy of Naples 8 Pentapolis
4 Calabria 9 Exarchate of Carthage
5 Bruttium

Map II Nominal extent of imperial territory in c. A.D. 600

34 Byzantium in the seventh century

Plate 1.3 Tiberius Constantine. Copper folhs

Plate 1.4 Maurice. Gold solidus

of Liguria. By 572 they were masters of much of northern Italy and

threatened to destroy the newly established prefecture and the results of
Justinian's expensive reconquests.49
In Africa, serious Berber revolts had meant considerable temporary
losses and great expense to the treasury;50 while in Spain Visigothic forces
began the gradual reconquest of the territories expropriated by Justinian.
By 629 these Spanish possessions had been lost entirely. Justin then
refused to pay the Persian subsidy due in 572; and in spite of initial
successes, Roman forces suffered several defeats in the ensuing warfare,
and a number of cities were sacked, notably Apamea in 573. The empire
was thus engaged in conflict on several fronts at once, and the one crucial
weakness in Justinian's strategy must have become apparent: the terri-

49
See I. Dujcev, 'Bizantini e Longobardi', in Atti del Convegno Internationale sul tema: la civiltd
del Longobardi in Europa (Rome 1974), pp. 45-78.
50
See in general, D. Pringle, The Defence of Byzantine Africa from Justinian to the Arab Conquest
(BAR, Oxford 1981); P. Goubert, Byzance avant Vlslam, vol. II, 2: Rome, Byzance et
Carthage; and the literature in notes 4 and 5 in the Introduction, above.
 State and society before Heraclius 35

tories were simply too widely dispersed and too extended, and the
resources needed to defend them under such conditions were simply not
available. The war in the East continued to drain the treasury until 591,
when civil war in Persia enabled Maurice to intervene to help Chosroes II
and at the same time to conlcude an extremely favourable peace agree-
ment. The war in the Balkans dragged on, the Romans registering some
successes after 592. But the whole area had from the 580s been deeply
infiltrated by migrating Slav settlers, and the indigenous population in the
countryside was often either overwhelmed by these newcomers or driven
out. While a number of cities held out, their existence was precarious; and
once Roman military support had been removed, they could no longer
survive. Only coastal cities such as Thessaloniki survived, but even here
under considerable military and economic pressure. 51 In 602 the mutiny
that led to Maurice's deposition took place; and during the reign of Phocas
the strategy which had helped to stabilise the situation in the Balkans to
Roman advantage was effectively abandoned. The whole area north of the
Thracian plain, with the exception of the stronger fortresses and cities and
coastal tracts, was lost to the authoritiy of the empire.
Maurice's deposition and murder in 602 gave the Persian Great King
Chosroes II the chance he needed; and on the pretext of avenging his
benefactor, his forces invaded the eastern provinces, ostensibly in support
of anti-Phocas elements in the eastern field armies. Thus began a war
which lasted until 626-7 and deprived the Roman state of huge areas -
eventually including all of Egypt and Syria - for many years. 52
The new and threatening situation which developed after the death of
Justinian demanded new methods of dealing with it. Under the Emperor
Maurice two new administrative circumscriptions appear, the exarchates
of Ravenna and Carthage. These were effectively militarised districts, the
old pretorian prefectures, in which for reasons of defence and military
security, and of resource allocation, supreme civil and military authority
was vested in a single military governor, or exarch. In many respects, the
principle of their organisation foreshadows the later Byzantine themata (or

51
Obolensky, The Byzantine Commonwealth, pp. 48ff.; G. Gomolka, 'Bemerkungen zur Situa-
tion der spatantiken Stadte und Siedlungen in Nordbulgarien und ihrem Weiterleben am
Ende des 6. Jahrhunderts', in Studien zum 7. Jhdt, pp. 35-42 with literature; B. Zasterova,
'Zu einigen Fragen aus der Geschichte der slawischen Kolonisation auf dem Balkan', ibid.,
pp. 59-65; H. Ditten, 'Zur Bedeutung der Einwanderung der Slawen', in F. Winkeimann,
H. Kopstein, H. Ditten and I. Rochow, Byzanz im 7. Jahrhundert. Untersuchungen zur
Herausbildung des Feudalismus {BBA XXXXVHI, Berlin 1978), pp. 7 3 - 1 6 0 , esp. 84ff. a n d
98ff. with extensive literature; also P. Lemerle, 'Invasions et migrations d a n s les Balkans
depuis la fin de l'epoque r o m a i n e j u s q u ' a u Vffle siecle', RH 78 (1954), 2 6 5 - 3 0 8 .
52
See the account in Stratos, Byzantium in the Seventh Century, vol. I, pp. 5 8 - 6 8 .
36 Byzantium in the seventh century

themes) but equally, precedents for them can be found in Justinianic

legislation to deal with internal security in provinces such as Phrygia, for
example. Institutionally, therefore, they are hardly new and seem simply
to represent the obvious functional response of the state and its resources
to a specific military threat. The reasons for their establishment at this time
are not hard to see: the changed situation in Italy brought about the
arrival and initial successes of the Lombards; and the constant threat and
damaging attacks of the Berbers in North Africa.53
The economic situation of the empire also deteriorated after Justinian.
Justin seems to have achieved a reasonable reserve in the treasury, but at
the cost of cutting down his expenditure on the army and fortifications and
ceasing the subsidies to the Avars and Persians. Tiberius went to the other
extreme, although his policies are perhaps easier to understand. On his
formal accession in 578 he remitted an entire year's taxes (by reducing the
usual demand by twenty-five per cent over the next four years). He spent
money freely on the army and carried through a vast recruitment cam-
paign. He also renewed the payment to the Avars, which bought peace for
a few years. But events in Italy demonstrated the scarcity of resources
under which the empire suffered. In spite of pleas for military aid in 578
and 580 from the senate in Rome and later from the pope, Tiberius merely
sent some cash and suggestions for diplomatic action to stem the incur-
sions of the Lombards. 54 Neither cash nor troops were available in great
supply, primarily because of the need to keep large armies in the Balkans
and in the East. Even here, however, resources were scarce. Maurice tried
to reduce the regular pay of the field units by twenty-five per cent which
led to a mutiny in 588; and although the situation was recovered, similar
attempts to save cash and resources by ordering the Danube forces to
winter over the river in hostile territory led to further mutinies, until in
602 the troops, led by their elected leader Phocas, marched on Constantin-
ople. Maurice fled, and after some days of confusion, Phocas was eventu-
ally proclaimed emperor. 55
The reign of Phocas, which lasted until his overthrow by Heraclius in
610, was both militarily and economically a disaster. Persian successes in
the east meant that the state was able to exact with only the greatest
difficulty the revenues from Syria, Phoenicia and neighbouring districts.
Asia Minor was raided and economic and social life disrupted, again with
53
For the exarchates, see Jones, LRE, vol.1, pp. 312f.; Brown, Gentlemen and Officers,
pp. 48ff.; Goubert, Byzance avant l'lslam; Ostrogorsky, Geschichte, p. 6 8 a n d note 1; a n d
Ch. Diehl, Etudes sur Vadministration byzantine dans VExarchat de Ravenne (Bibliotheque des
ecoles francaises d'Athenes et de Rome LIII, Paris 1888), esp. the opening chapter.
54
See Menander Protector, frg. 4 9 (Exc. de Leg., II, p. 469); Haldon, Recruitment and
Conscription, pp. 25f., 2 8 .
55
See Kaegi, Military Unrest, esp. pp. 6 6 - 7 3 a n d 1 0 1 - 1 9 .
 State and society before Heraclius 37

Plate 1.5 Phocas. Gold solidus

the inevitable effects on the ability of the state to raise revenue. Pay and
supplies for the army must also have been affected. The withdrawal of
large numbers of troops, albeit temporarily, from the Danube front, gave
the Avars and their Slav confederates and vassals a free hand to establish
themselves securely south of the Danube, to devastate the countryside,
isolate the cities and towns, and generally render the whole area untena-
ble. Civil strife following Maurice's death, constant plots against Phocas in
the capital, the resulting repression and the discontent of large sections of
the populace of Constantinople, meant that Phocas was quite unable to
maintain a uniform policy to deal with these problems. 56
The extent to which the wars in the east and the Balkans influenced the
basic structure of East Roman society is difficult to determine, of course.
Undoubtedly, massive insecurity in the Balkans meant that regular agri-
cultural production must have suffered considerably. Similarly, large
tracts of Anatolia must have been affected by Persian military operations.
Political repression and the expropriation of senators under Phocas must
also have meant the accumulation in state hands, or the redistribution, of
considerable amounts of landed property in the provinces. All these devel-
opments heightened people's awareness and perception of their situation
in significant ways, as they likewise affected the administrative structure of
the empire in these regions. Fundamental features of a Byzantine, rather
than a late Roman culture and ideology had begun to form already,
features which these diasters highlighted, but which were apparent even in
the last years of Justinian's reign and throughout the reigns of his
successors.
From the last years of Justinian the empire entered what might be
perceived as a period of ideological reorientation. Late Roman culture
demonstrates at this time a loss of confidence or trust in the traditional
symbols of authority and the establishment, a drift away from the sym-
56
Ostrogorsky, Geschichte, pp. 70-2.
38 Byzantium in the seventh century

bolism of the heaven-endowed earthly empire - the imperial cult and

hierarchies of state and Church - and its apparent fallibility, towards
embodiments of heavenly power of a less fallible nature: the cults of saints,
the cult of the Virgin, the icon; symbols, in short, of heavenly intercession.
This change in emphasis affected equally the imperial establishment,
which promoted aspects of the change as a means of refocusing the
divergent trends within the symbolic universe of late Roman society
around the specifically imperial, Christian and orthodox ideology of the
state, centred about the Christ-loving, God-protected rulers at Constantin-
ople. 57
Late Roman society, thoroughly Christian, became increasingly intro-
verted. Marginal groups were no longer tolerated as they had been, in spite
of the persecutions which occasionally occurred, in the earlier sixth centu-
ry. 58 At the same time, Constantinople, the city of the court and the
bureaucracy, became more and more the focal point of the empire poli-
tically and culturally, a process which was accelerated after the loss of
Antioch, Jerusalem, Alexandria and, lastly - and much later - Carthage,
during the course of the seventh century. The decline and eventual
collapse of the older provincial municipal institutions both administra-
tively and socially completed this process.59 It is possible to detect a
transfer of social and economic interest away from the provincial cities to
Constantinople, the former having lost not only their corporate and
administrative identity, but thereby also their social relevance and attrac-
tion. It was this new emphasis upon the centre, Constantinople, which
began to emerge in the late sixth century and which the emperors used as
a means of reinforcing their own authority. 60
The reorientation of this period appears most clearly in the official stress
upon the heavenly guardians of the emperor and the state, and imperial
ceremonial, which focused in a particularly evocative way on the figures
of the emperor and the symbolism of imperial authority. In the changed
situation of the later sixth century, one of the results of this new emphasis,
as we shall see in chapter 9, served also to point out the frailty of earthly
power and to direct attention away from God's representative upon earth
57
For a summary and discussion of these developments, see Haldon, 'Ideology and social
change', 1 6 1ff.,and Cameron, 'Images of authority', 4ff.
58
See especially P. Brown, Religion and Society in the Age of St Augustine (London 1972),
p. 55; J. Nelson, 'Symbols in context: rulers' inauguration rituals in Byzantium and the
West in the early Middle Ages', Studies in Church History XIII (1976), 9 7 - 1 1 9 , see 115f.
59
J.F. Haldon and H. Kennedy, 'The Arab-Byzantine frontier in the eighth and ninth cen-
turies: military organisation and society in the borderlands', ZRVI 19 (1980), 7 9 - 1 1 6 ,
see 87ff.; Haldon, 'Some considerations', 8 6 - 8 ; H. Hunger, Reich der neuen Mitte. Der
christliche Geist der byzantinischen Kultur (Vienna, Graz and Cologne 1965), pp. 42ff.
60
Haldon, 'Ideology and social change', 1 7 1 - 3 , for a summary of these developments with
further literature.
 State and society before Heraclius 39

to God Himself.61 Evidence for local and civic saints' cults increases sharply
at this time and, together with the developments noted already, suggests
strongly that ordinary people were transferring their attention away
from the worldly and physical authority of the emperor and the state,
distant and ineffective as it often was, towards a more immediate and
tangible power, a power and authority invested in heavenly guardians and
intercessors directly by God. Given the close links deliberately fostered by
the emperors themselves between their authority and its sources, such a
development is not really surprising in the context of shifting political
horizons and military and social vulnerability.62 These developments had
important repercussions in the middle and later seventh century.
The later sixth and early seventh century witnessed a series of dramatic
changes in the late Roman world: politically the empire suffered a series of
major blows to its prestige and authority; economically, it was able to
maintain its resources, although much reduced, only with considerable
difficulty. Socially, it was rent by divisions within the Church, while the
vast majority of the producing population was maintained at a subsistence
level for the benefit of the state and its bureaucracy, and the wealthy social
elite which dominated it. Many contemporaries were fully aware of the
dangers of the situation, and debates about the likely outcome of the
long-term political crisis which engulfed the Roman world during and after
the reign of Phocas were not unusual. As Jacob, the Palestinian Jew who
was forcibly baptised at Carthage under Heraclius in 634, remarked, the
future of the empire seemed most uncertain:
From the Ocean, that is of Scotia [i.e. Ireland], from Britannia, Hispania, Francia,
Italia, Hellas and Thrace and Egypt and Africa and upper Africa, the boundaries of
the Romans and the statues of their emperors were seen until our times; for all
peoples submitted to them at God's command. But now we see Romania reduced
and humbled.63
A telling picture of the situation after Maurice's death is given in the
collection of the miracles of St Demetrius, written down not long after the
events they claim to portray:
For you all know what clouds of dust the Devil stirred up during the years of the
successor of the emperor Maurice, of blessed memory, when he smothered love,
and sowed mutual hatred in all of the East, in Cilicia, in Asia, in Palestine and the
neighbouring lands up to Constantinople itself. The demes were no longer content
merely to spill the blood of their comrades on the streets; they broke into one
61
See Cameron, 'Images of authority', 1 5 - 2 1 .
62
Averil Cameron, T h e Theotokos in sixth-century Constantinople: a city finds its symbol',
JThS 2 9 (1978), 7 9 - 1 0 8 , see 102ff.
63
Doctrina lacobi nuper Baptizati (ed. N. Bonwetsch, in A b h a n d l u n g e n der koniglichen
Gesellschaft der Wissenschaften zu Gottingen, phil.-hist. Klasse, XII, 3, Berlin 1910),
p. 62. 4-12.
40 Byzantium in the seventh century

another's homes and slew the occupants mercilessly. Women and children, old and
young, those who were too weak to save themselves byflight,they hurled from the
windows of the upper floors; like barbarians they plundered their fellow citizens,
their acquaintances and their relatives, and put their homes to the flame.64
The picture may be overdrawn, but it neatly encapsulates one aspect of the
mood of the Eastern empire during the early years of the seventh century.
64
Miracula S. Demetrii (ed. Lemerle), 82 (I, 112.11-113.7). For the general development of
late Roman society through the period from the late fourth to later sixth centuries, see the
useful survey of Averil Cameron, The Mediterranean World in Late Antiquity AD 395-600
(London 1993), in which a wide range of work up to the date of publication on the literary
sources and the archaeological material is cited and taken into account.
 CHAPTER 2

The East Roman world c. 610-717: the

politics of survival

HERACLIUS 6 I O - 6 4 I

The reign of the 'tyrant' Phocas demonstrates the degree to which the
Justinianic expansion of the sixth century had over-extended the resources
of the state, and how crucial the stability of the central authority was to the
well-being of the empire as a whole. Phocas, a subordinate officer from the
Danube forces, seems to have had neither the ability nor the experience
needed of a ruler in the situation in which the late Roman state found itself.
His reign is remarkable chiefly for the plots and attempts on his life which
he managed to avoid during the eight years of his reign in Constantinople,
and for the disastrous collapse of the empire's defences, especially in the
East. A long series of unsuccessful, senate-inspired plots eventually culmi-
nated in the expedition sent by Heraclius the Elder, the exarch of Africa
based at Carthage, under his son Heraclius and his nephew Nicetas; the
former with a squadron of ships with Byzantine and Moorish troops to
Constantinople, the latter marching in 608 via Egypt, which quickly joined
the rebellion.1 Heraclius with his fleet appeared off Constantinople on 3
October in the year 610; he had been greeted enthusiastically en route
wherever he had stopped; and the story was repeated at Constantinople.
Phocas' supporters deserted him, Heraclius was let into the city, and the
tyrant was executed. His last words testify both to the situation of the
empire after eight years of directionless government, and to Phocas' own
desperate incompetence. As Heraclius confronted him before his execution
he asked: 'Is it in this manner that you have governed the state?' Phocas
replied: 'Will you be able to do any better?'2
Heraclius was crowned by the patriarch Sergius in the chapel of St
Stephen in the palace on 5 October. His accession was acclaimed by the
1
On the reign of Phocas, see Ostrogorsky, Geschichte, pp. 70-2; Stratos, Byzantium in the
Seventh Century, vol. I, pp. 48-91 for a detailed political history; and J.Herrin, The
Formation of Christendom (London 1987), pp. 187ff. for a summary. A detailed account: D.
Olster, The Politics of Usurpation in the Seventh century: rhetoric and revolution in Byzantium
(Amsterdam 1993).
2
John of Antioch,frg.5, 38; Exc. de Insid., 150.

41
42 Byzantium in the seventh century

Plate 2.1 Heraclius. Gold solidus

demes, the palace guards and by the senate according to tradition; and
there is no doubt that it was a popular accession. Phocas had been almost
universally detested by the end of his reign; and it is interesting that the
one quarter where his rule was popular was in the papal chancery at
Rome. Since the early sixth century, the patriarchs of Constantinople had
been wont to call themselves also ecumenical, a title to which Pope
Gregory I took fierce objection. While the Emperor Maurice stayed out of
the debate, Phocas quickly made clear his support for the papal position
and in 607 issued a document which recognised the apostolic Church of St
Peter as the head of all the Churches. His popularity in Rome is therefore
easily understood; but it was the only foundation of such popularity.3
Heraclius is popularly, and probably quite justifiably, regarded by both
modern historians and by Byzantines as one of the empire's greatest rulers.
But his accession saw no immediate change for the better in the situation of
the state. Indeed, things became decidedly worse, so much so that in 618
the emperor planned to abandon Constantinople, which seemed too open
to attacks by Avars and Slavs in the north and the Persians from the east,
and move the seat of empire to his old home, Carthage. 4 But the response
to this plan from the populace of the city, and the arguments of the
patriarch Sergius, dissuaded him from this action. Instead, he began to
develop plans to restore the situation. In the East as in the West, however,
he met with only limited success and a number of failures.
In 611, the Persians were driven out of Caesarea in Cappadocia; but
Byzantine counter-attacks in Armenia and Syria were unsuccessful, and
after defeats in 613, the Persians marched once more into the Anatolian
provinces.5 They also occupied Damascus, and in 614 Jerusalem itself was

3
For the edict, see Dolger, Regesten, no. 155; and on the disagreement over titles, see
E. Caspar, Geschichte des Papsttums (2 vols., Tubingen 1935), vol. II, pp. 452ff.; V. Laurent,
'Le Titre de patriarche oecumenique et la signature patriarcale', REB 6 (1948), 5f.
4
Nicephorus, 12, 10.
5
See esp. W.E. Kaegi, jr., 'New evidence on the early reign of Heraclius', BZ 66 (1973),
308-30, see 313ff.; Sebeos, 65.
 The East Roman world: the politics of survival 43

taken, a particularly heavy blow to Byzantine morale: not only was the
Church of the Holy Sepulchre destroyed, but the True Cross was carried off
to Ctesiphon. At the same time, the Persians partly occupied Cilicia, and
the important city of Tarsus was taken. Armenia likewise fell to Persian
attack, and the Byzantine forces of the magister militum per Armeniam were
pushed westwards and northwards. In 615 further attacks into Asia Minor
took place, and in 616 Persian troops reached the Bosphorus.6 Byzantine
troops seem to have retained some cohesion, however. The armies of the
magister militum per Armeniam were still operational; forces of the magister
militum in praesenti and of the magister militum per Orientem seem to have
been active in Cilicia and Isauria. 7 Troops from Africa and Egypt under
Heraclius' cousin Nicetas were operating in Syria and in Egypt itself;
although Persian invasion and occupation of these territories between 616
and 620, followed by the consolidation of Persian authority in Syria and
Palestine, terminated this oppositional activity.8 This was again a major
blow for the Byzantines, for the 'granary of the empire' and the chief
supplier of grain to Constantinople was now lost to the enemy.
In the north and west, meanwhile, the empire slowly lost control over
much of the Balkan region. The partial withdrawal of troops from the
Danube front in 602 has traditionally been taken as the moment from
which this process began in earnest. It has been assumed that there was
thereafter little to hinder the Slav populations north of the Danube from
moving south and overrunniung considerable areas. But, in fact, the
evidence suggests that throughout most of the reign of Phocas only minor
incursions took place - it was the civil wars between Heraclius and the
supporters of Phocas which actually necessitated the withdrawal of sub-
stantial forces from the Balkans and which expedited the real collapse. This
was especially the case after 613 when the war with the Persians took a
turn for the worse.9 The actual occupation of the Balkans seems to have
6
See the account of Theophanes on Persian inroads from the year 612: 299.32-3, 300.1-6,
20-21 (for 614); 3OO.3Osq. (615); 301.9sq. (616-17).
7
See Haldon, Recruitment and Conscription, p. 29, note 29; and for a temporary mint
established first at Seleucia (615-17) and then at Isaura (617-18) see Ph. Grierson,
Catalogue of the Byzantine Coins in the Dumbarton Oaks Collection and in the Whittemore
Collection, vol. II: Phocas to Theodosius III, 602-717 (Washington B.C. 1968), pp. 327f. It
must reflect the minting of coin specifically to pay the troops of the region. An otherwise
unattested arms factory at Seleucia is also evidence of local military operations. See
G.Zacos and A. Veglery, Byzantine Lead Seals, vol. I, parts 1-3 (Basel 1972), no. 1136.
8
See the account of Stratos, Byzantium in the Seventh Century, vol. I, pp. 113f.; F. Winkel-
mann, 'Agypten und Byzanz vor der arabischen Eroberung', BS 40 (1979), 161-82, esp.
169-70 with literature. Winkelmann shows that, pace Stratos and others, the Persians
were probably not greeted with open arms, nor hailed as liberators of the Coptic and
monophysite population.
9
See esp. the discussion of R.-J. Lilie, 'Kaiser Herakleios und die Ansiedlung der Serben', SoF
44 (1985), 17-43 and the literature cited. For the traditional position, see P. Charanis,
44 Byzantium in the seventh century

been a very slow process, lasting somefiftyyears: in 582, the Avars - who
exercised a political-military hegemony over the numerous Slav clans and
tribal groups - captured Sirmium and occupied the surrounding region, an
event which effectively opened a door into Byzantine lands south of the
Danube. In 626, after the defeat of the Avaro-Slav attack on Constantin-
ople, Avar domination of the Slav peoples began to loosen, and the
independent Sklaviniai of the Balkans henceforth figure as the main oppo-
nents of Byzantine rule in the area, at least until the arrival of the Bulgars
in the 680s. Both the literary sources and the numismatic evidence - finds
of hoards which represent a response to invasions and attacks - provide a
general framework for the chronology of these events. Between about 576
and 586/7 a number of invasions and raids penetrated into the south
Balkans, past Thessaloniki and into the Peloponnese.10 Again, from about
609 until the 620s, further hoards together with less precise literary
references - that is, of dubious or inexact chronology - attest to raids deep
into Greece and the Peloponnese.11 The pattern suggests an initial series of
raids and invasions, beginning probably before 577, but ending tempo-
rarily in the late 590s, until the last year of Phocas, when a second wave
began. The permanent settlement of Slavs in the north and central Balkans
begins already in the 570s, of course, on a very small scale; the Miracles of
•Ethnic changes in the Byzantine empire in the seventh century', DO? 13 (1959), 23-44,
see 37fT.
10
For the literary sources, see Menander Protector, frgs. 47 (p. 209.3sq.) and 48
(p.468.36sq.); John of Ephesus, III, 25 (from 577/8 to 583/4). Note also the reference to
clerics having to abandon their sees as a result of barbarian attacks in the letters of
Gregory I. See E. Chrysos, 'SV^POAT) <rrf)v icrropCa TTJS 'Hiretpov xai-a rr|v
irpwTOpv^avTivT) eiroxt) (A' - XT'aC.)', 'HireLpcjnxrj Xpovixr\ 23 (1981), 72-7, with
literature. Note also the Chronicle of Monemvasia, which mentions the conquest of the
Peloponnese and the expulsion of its indigenous population in 587/8 - something of an
exaggeration: P. Lemerle, 'La Chronique improprement dite de Monemvasie. Le contexte
historique et tegendaire', REB 21 (1963), 5-49. In spite of the Chronicle's claims,
Monemvasia seems in fact to have been established in 582/3. See P. Schreiner, 'Note sur la
fondation de Monemvasie en 582-583', TM 4 (1970), 4711T.; P. Schreiner, ed., Die
byzantinischen Klemchroniken (2 vols., CFHB XII1 and 2, Vienna 1975 and 1977), vol. I,
p. 319, nos. 41a and b; vol. II, pp. 77f. For the numismatic material see B. Athanasso-
poulou-Penna, 'G^cravpos vojiiajjudTcov 6OU aUbva (i.X. diro TT)V TrepioxTl T<bv Oilpwv',
Apxaiokoyixri ecpri^epig (1979), 200-13; D.M. Metcalf, 'The Slavonic threat to Greece
circa 580: some evidence from Athens', Hesperia 31 (1962), 134-41; T.L. Shean, 'The
Athenian Agora: excavations of 1972', Hesperia 42 (1973), 395-8; and esp. the general
survey of A. Avramea, 'Noiuaixcmxoi "{hjactvpoi" xai \xs\iovoi\xsva vo|juarp,aTot dird
TTJV rieXoTTOvvTicro (XT'-Z' ca.)\ XvfifieLXTa 5 (1983), 49-89. See also M. Nystazopoulou,
'X\)(xf3o\f| els rf)v xPovo^6*YT|oriv TO>V AfJapixwv xai a\apixa>v emBpojjuov em
MavpixCov (582-602), Xv/xfieixra 2 (1970), 149; and Th. Olajos, 'Contribution a la
chronologie des premieres installations des Slaves dans 1'empire byzantin', B 55 (1985),
506-15.
11
Literary evidence: see P. Lemerle, Les plus anciens recueils des miracles de St. Demetrius', I. le
Texte (Paris 1979), II. Commentate (Paris 1981), see 179-94 (1,175. lsq.); II, pp. 85f. for
an invasion of 614-16; and for coin finds, see Avramea, 'S-qaavpoC, 77f.; D. M. Metcalf,
'The Aegean coastlands under threat: some coins and coin hoards from the reign of
Heraclius', ABSA 57 (1962), 14-23.
 The East Roman world: the politics of survival 45

St Demetrius attest to permanent Slav groups in the region of Thessaloniki

in the early seventh century; but the penetration and occupation of the
Peloponnese, for example, seems only to have commenced on a permanent
basis after 609/10. 12
The immediate danger to the Byzantine state, however, came from the
Avars, who by the time of Heraclius exercised authority over a consider-
able area, from the Danube plain eastwards into the south Russian steppes,
ruling a confederacy or empire which had considerable resources in
manpower at its disposal. Already in the years 615-16, as both Persians
and Avars approached Constantinople in a joint effort to force the surren-
der of the City, the extreme danger was apparent. In 617, in an attempt to
meet the Avar khagan, Heraclius was ambushed near the Long Walls in
the area of Heraclea and nearly captured.13 By 619, however, he had been
able to arrange a truce with the khagan, which enabled him to transfer
troops from Europe to Asia Minor. In a series of campaigns, beginning 622
and lasting until 628, Heraclius was able to outmanoeuvre the Persian
forces in Anatolia and Armenia and take the war to the heart of the
Sassanid empire.14 The Avars, who were bought off again in 623, bided
their time until 626, when a Persian counter-attack and a major Avar
offensive against Constantinople endangered the capital of the empire
itself. Not to be distracted from the campaigns in the East, Heraclius
12
See Lemerle, Us plus anciens recueils, 100-15 (1,124.1sq.) II, p. 69ff.; and Sp. Vryonis, jr.,
The evolution of Slavic society and the Slavic invasions in Greece. The first major Slavic
attack on Thessaloniki, A.D. 597', Hesperia 50 (1981), 378-90; Vryonis, 'Review essay of
Michael W. Weithmann, Die slawische Bevolkerung aufder griechischen Halbinsel (Munich
1978)', Balkan Studies 22 (1981), 405-39; and for the occupation, M.Kordoses,
M
HaXapLXT| eiroixtiaTi VTX\V IleXoTTOvv^ao \ie $acn\ Ta aXapixd TOirwvujiia',
Aa)8(ov7) 10 (1981), 388-421; and see also J.Ferluga, 'Untersuchungen zur byzantini-
schen Ansiedlungspolitik auf dem Balkan von der Mitte des 7. bis zur Mitte des 9.
Jahrhunderts', ZRVl 23 (1984), 49-61; and the older work of M. Vasmer, Die Slawen in
Griechenland (Berlin 1941 and Leipzig 1970), esp. pp. 118ff. Tn general, see Ostrogorsky,
Geschichte, pp. 77tt.\ Obolensky, Byzantine Commonwealth, pp. 52ff.; Ditten, 'Zur Bedeu-
tung der Einwanderung der Slawen', p. 95f.; idem, Ethnische Verschiebungen zwischen der
Balkanhalbinsel und Kleinasien vom Ende des 6. bis zur zweiten Hdlfte des 9. Jahrhunderts (BBA
UK. Berlin 1993), pp. 45ff.; Stratos, Byzantium in the Seventh Century, vol. I, pp. 118ff.; and
the most recent brief survey, M. Nystazopoulou-Pelekidou, 'Les Slaves dans l'empire byzan-
tin', in 17th International Byzantine Congress, Main Papers (Washington D.C. 1986), pp.
345-67, see 346-51. For another view, see M.W. Weithmann, Die slawische Bevolkerung auf
der griechischenHalbinsel{Munichl97S).
13
See N.H. Baynes, 'The date of the Avar surprise', BZ 21 (1912), 110-28; also Stratos,
Byzantium in the Seventh Century, vol. I, pp. 145ff., who argues for a later date, namely
623.
14
See the brief summary in Ostrogorsky, Geschichte, pp. 86f., and N. Oikonomides, 'A
chronological note on the first Persian campaign of Heraclius (622)', BMGS 1 (1975),
1-9; A.Pernice, L'lmperatore Eraclio (Florence 1905), pp. 11 Iff.; most recently Stratos,
Byzantium in the Seventh Century, vol. I, pp. 135-44, 151-72, 197-234; and note
Haldon, Byzantine Praetorians, pp. 169-71 with notes 333-41; and C. Zuckermann, 'The
reign of Constantine V in the Miracles of St Theodore the Recruit (BHG 1764)', REB 46
(1988), 191-210, see 206ff.
46 Byzantium in the seventh century

reinforced the Constantinoplitan garrison with troops from the field forces
under his brother Theodore; and with the help of an imperial fleet, which
destroyed the Slav attack on the sea-walls of the city, the siege was
defeated and the Avar khagan, followed shortly thereafter by the Persian
forces, withdrew. 15 The retreat of the Persian forces under Sahrbaraz to
Syria, and the defeat of a second Persian column under Sahin, shortly
before this, and at the hands of Heraclius' brother Theodore, signalled the
final failure of Persian attempts to force the surrender of Constantinople.
The war ended effectively in early 628, when, after the crushing defeat of
his forces in the Caucasus and near Nineveh in 627, Chosroes was deposed
and murdered and replaced by his son Kavadh (called Siroes in the Greek
sources), who immediately negotiated a peace with the Romans. The
agreement of 591 was brought back into force, and over the following year
the Persian army withdrew from Egypt and all the territories it had
occupied since the opening phases of the wars. 16 The course of events was
complicated and influenced in addition by Byzantine plans to bring about
the conversion of the Persians, initially in the context of the war through
the general Sahrbaraz, a plan which achieved a certain degree of success
before the Arab conquests put an end to Byzantine hopes. 17 Following on
his triumphal return to Constantinople in 628, when he was met at Hiereia
on the Asia Minor coast by the senate, clergy, the patriarch Sergius, his son
Constantine and sections of the populace, the crowning achievement of the
reign was the return of the True Cross to the city of Jerusalem on 21 March
in 630. Thus ended what has sometimes been seen as the first crusade,
symbolised in ceremonial enactment. 18
In the West, the Avars' retreat in 626 had equally far-reaching con-
sequences. Uprisings of subject Slav peoples soon followed, and within a
few years an independent west Slav confederacy had appeared, under the
leadership of a certain Samo, a Frankish merchant by origin. This was a
direct result of the revolt against Avar domination and the diplomatic

15
See F. Barisic, 'Le siege de Constantinople par les Avares et les Slaves en 626', B 24
(1954), 371-95; and in particular P. Speck, ed., Zufdlliges zum Bellum Avaricum des
Georgios Pisides (Miscellanea Byzantina Monacensia XXIV, Munich 1980), with literature
and discussion; V. Grumel, 'La Defense maritime de Constantinople du cote de la Come
d'Or et le siege des Avares', BS 25 (1964), 217-33; Herrin, Formation of Christendom,
pp. 198f.
16
See the account of Stratos, Byzantium in the Seventh Century, vol. I, pp. 210ff.; and the
literature in note 14 above.
17
See C. Mango, 'Deux etudes sur Byzance et la Perse Sassanide, II: Heraclius, Sahrvaraz et
la vraie croix', TM 9 (1985), 105-18
18
Cf. A. Frolow, 'La Vraie croix et les expeditions d'Heraclius en Perse', REB 11 (1953),
88ff.; Frolow, 'La Dedicace de Constantinople dans la tradition byzantine', Revue de
Vhistoire des religions 127 (1944), 6Iff.; Stratos, Byzantium in the Seventh Century, vol. I,
237ff.,esp. 252-5.
 The East Roman world: the politics of survival 47

activity and support of Constantinople.19 At about the same time, accord-

ing to later Byzantine tradition, Heraclius reached an agreement with two
Slav groups, the Croats and the Serbs, inviting them to attack the Avars in
the west Balkans, south of the Danube, and officially granting them
permission to occupy the regions they were able to recapture for the
empire. They were also converted to Christianity, according to the tenth-
century account of the Emperor Constantine VII, although this was clearly
a short-lived state of affairs. Nevertheless, according to the later tradition,
it enabled the Byzantines to reassert, formally and in theory, their authority
over these regions. In fact, it has been shown that Heraclius was able to do
very little to recover the situation in the Balkans, and it was really during
his reign that Byzantine power in that region effectively disappeared.20
A third blow to Avar hegemony was the rebellion some time between
619 and 635 - probably in the 630s - of the Onogur 'Huns' or Bulgars
under their leader Kovrat, who with his uncle Organa had visited Con-
stantinople in 619, where he had been baptised and where the Byzantines
mediated in establishing a treaty with the Avars. The Onogur revolt was
certainly supported from Constantinople, and in return for confirming the
treaty he had witnessed in 619, Kovrat received the title of patrikios. The
Onogur-Bulgars occupied an area in and around the Kuban region and on
the east coast of the Sea of Azov, their territory being referred to as 'old
Great Bulgaria'. While their successful rebellion against the Avars was
welcome news in Constantinople, the later history of this people, in
particular the branch led by one of Kovrat's sons, Asparuch, was to have a
crucial impact on Byzantium.21
By the early 630s, therefore, the empire had been able to restore its
eastern territories in full and, to a limited extent at least, to stabilise the
19
See esp. A. Avenarius, Die Awaren in Europa, pp. 123-38; Ditten, 'Zur Bedeutung der
Einwanderung der Slawen', pp. 127ff. with discussion and full bibliography; V. Chalou-
pecky, 'Considerations sur Samon, le premier roi des Slaves', BS 11 (1950), 223-39;
Obolensky, The Byzantine Commonwealth, p. 59; Obolensky, in CMH IV.l, p. 482; Stratos,
Byzantium in the Seventh Century, vol. I, p. 316.
20
Avenarius, Die Awaren in Europa, pp. 1 3 8 ^ 7 ; Obolensky, The Byzantine Commonwealth,
pp. 59f.; Stratos, Byzantium in the Seventh Century, vol. I, pp. 325-37; Ditten, Zur Bedeu-
tung der Einwanderung der Slawen, p. 129 with literature; Ostrogorsky, Geschichte, pp. 87f.;
V. Popovic, 'Aux origines de la Slavisation des Balkans; la constitution des premieres
Sklayinies macedoniennes a la fin du sixieme siecle', Compte-rendu de VAcad. des Inscripti-
ons et Belle-Lettres (Paris 1980), pp. 230-57; and the literature and discussion in Lilie,
Kaiser Herakleios und die Ansiedlung der Serben, esp. 18ff. and note 3, and 29ff. Note also
J. Cangova, 'Les Slaves aux environs de Preslav aux VIF-VIIP siecles d'apres les donnees
archeologiques', Studia in Honorem Veselin Besevliev (Sofia 1978), pp. 363-8; J.Ferluga,
'Untersuchungen zur byzantinischen Ansiedlungspolitik auf dem Balkan von der Mitte des
7. bis zur Mitte des 9. Jahrhunderts', ZRVl 23 (1984), 49-61.
21
Nicephorus, 12.20; 24.9; Avenarius, Die Awaren in Europa, pp. 153-7; Obolensky, The
Byzantine Commonwealth, pp. 62-3; R. Grousset, L'Empire des Steppes (Paris 1939),
pp. 229-30; Ostrogorsky, Geschichte, p. 87 and note 3.
48 Byzantium in the seventh century

position in the Balkans. The Avar power was destroyed; and while the Slav
population of the Balkans and the Peloponnese made Byzantine authority
rather an empty concept, a more-or-less peaceful situation had been won,
which gave the land and population time to recover from the years of
warfare and devastation.

The monophysite problem and the rise of Islam

Heraclius' achievement was remarkable. The result of these years of

constant warfare and effort was a restored empire. But it was a greatly
weakened empire. All the resources of Church and state had been swal-
lowed up in the war effort; and while the eastern provinces were safe and
could once more provide the revenues needed to maintain the imperial
armies and the corn to feed Constantinople, Thrace and what other Balkan
possessions remained in Byzantine hands took much longer to recover.
Heraclius was further burdened with the considerable debt he owed the
Church, for it was only through borrowing and confiscating considerable
amounts of silver and gold plate, bronze statues, and other forms of wealth,
that he had been able to raise the cash with which to pay the troops. 22
Despite the ending of hostilities, a substantial army still had to be main-
tained; imperial generosity in the rebuilding of cities and fortresses and the
rewarding of the soldiers who had fought against the Persians meant
further drains on limited resources; while the disruption of warfare in parts
of Asia Minor must have seriously affected the ability of many regions to
pay the taxes demanded from them after their restoration to Imperial
control. The Persians had pillaged many of the cities of the East and had
removed the treasures of the Church and the reserves of local administra-
tive capitals, such as Alexandria. 23 In short, and in spite of external
appearances, the empire was economically in a parlous condition.
The situation was not helped by Heraclius' efforts to solve the question of
the split between the Chalcedonian and monophysite communities. The
varied cultural and linguistic make-up of the empire as well as the
demands of the imperial ideology made unity in orthodoxy an essential
and urgent matter of practical state politics. Together with the patriarch
Sergius and the Chalcedonian archbishop of Alexandria, Cyrus (from
631), the formula of the single energy was worked out, intended to provide
a bridge for Chalcedonian theology to approach the monophysite position.
In the context in which it was invoked and elaborated, it functioned as a
compromise intended to meet the criticisms which the monophysites had
raised against the traditional dyophysitism of Chalcedon, according to
22
Hendy, Studies, pp. 2 3 1 , 4 9 4 - 5 ; Ostrogorsky, Geschichte, p. 8 3 .
23
W i n k e l m a n n , Agypten und Byzanz, 1 6 9 .
 The East Roman world: the politics of survival 49

which the two natures of Christ, human and divine, operated through a
single energy. But it is important to note that the debate was not begun by
this political need. On the contrary, discussion over the nature of the
energies of Christ had begun among theologians already in the sixth
century, albeit with the intention of bringing the neo-Chalcedonian and
the monophysite positions closer.24
But even though Heraclius led the negotiations himself, only partial and
temporary successes were gained. Pope Honorius lent his support to the
project; but even the agreement reached by Cyrus in 633 with the Theodo-
sianites, a group of monophysites in Egypt, was short-lived. Once again,
the imperial government was impelled to employ force in order to gain
acceptance for its views. But Chalcedonians themselves soon began to
express doubts about the theological validity of the argument, doubts
expressed particularly strongly by Sophronius, patriarch of Jerusalem from
634. The result was that an alternative, emended position was developed,
primarily, it would seem, by the patriarch Sergius. Partly encouraged by
the rather reserved position of the pope, he argued that the crucial element
was not the single energy, but the single will - thelema - of Christ. In 638,
Heraclius issued the famous Ekthesis, in which further discussion of the
problem of one or two energies was forbidden, and the new, monothelete,
formula was set out. It was put up for all to see in the narthex of the Hagia
Sophia. But the new policy was likewise rejected by a large number of
Chalcedonian churchmen, particularly in the West, as well as by the
monophysite Churches. And although Sergius was replaced after his death
in 638 by the keen monothelete Pyrrhus, it continued to meet with
opposition, now from the successor of Pope Honorius in Rome also. The
compromise had merely caused another division within the Church. But it
also had important consequences for the imperial ideology itself and for the
position of the emperors, as we shall see. And in the meantime, both Syria
and Palestine had been conquered by the Arabs. 25
The arrival on the historical stage of Islam and its initial bearers, the

24
See F. Winkelmann, 'Die Quellen zur Erforschung des monenergetisch-monotheletischen
Streites', Klio 6 9 (1987), 5 1 5 - 5 9 , esp. 555f., following S.Helmer, Der Neuchalkedonismus.
Geschichte, Berechtigung und Bedeutung eines dogmengeschichtlichen Begriffes (Bonn 1962),
see p. 2 2 3 .
25
For the background to these developments see Ostrogorsky, Geschichte, pp. 9 0 - 1 ; Winkel-
m a n n , Die ostlichen Kirchen, pp. 1 0 6 - 7 ; Haldon, 'Ideology a n d social change', 16Iff.;
and esp. V.Grumel, 'Recherches sur l'histoire d u Monothelisme', EO 27 (1928), 6 - 1 6 ,
2 5 7 - 7 7 ; 2 8 (1929), 2 7 2 - 8 3 ; 2 9 (1930), 1 6 - 2 8 ; H.-G. Beck, Kirche und theologische
Literatur im byzantinischen Reich, 2nd edn (Munich 1977), pp. 292-4; Herrin, Formation of
Christendom, pp. 206ff.; the source analysis of J.-L. Van Dieten, Geschichte der Patriarchen
yon Sergios 1. bis Johannes VI (610-715) (Amsterdam 1972), pp. 179ff. ('Die griechische
Uberlieferung liber die Anfange des Monotheletismus'); and esp. that of Winkelmann, 'Die
Quellen'.
50 Byzantium in the seventh century

nomadic tribesmen of the Arabian peninsula, demonstrated just how weak

the long-drawn-out conflict between the two great powers had left the
empire. Syria was finally lost in 636, Palestine in 638, Mesopotamia in
639/40, Armenia in the same year and Egypt by 642. Initially intending to
direct the destruction of the invaders from his headquarters at Antioch,
Heraclius abandoned the struggle in 636 after the disastrous defeat at the
battle of the Yarmuk and returned to Constantinople.26 Byzantine forces
held out most effectively in Palestine, where the patriarch Sophronius
inspired the defence of Jerusalem. Elsewhere, however, it seems that an
underlying apathy in the bulk of the monophysite population to Byzantine
rule - in Egypt and Syria especially - deprived the defenders of any
otherwise committed popular support they might have hoped for and also
made the Islamic invaders less unacceptable. This is not to say that they
were welcomed with open arms. Merely that the morale of the imperial
troops was low, the population not always particularly supportive, and
resources were anyway overstretched.27 In addition, it seems clear that
neither the Persians nor the Byzantines fully realised the nature of the
danger facing them. In spite of their larger than usual scale, it must have
seemed difficult to believe that these raids were anything more than a
temporary occurrence, producing temporary set-backs. That the Arabs
were here to stay, however, soon became apparent; for already by the time
of their attack on Egypt, the Sassanid empire had been virtually destroyed.
The new Arab empire was in the making.28
26
The best and most recent account is that of Lilie, Die byzantinische Reaktion, pp. 4 0 - 6
(Syria, Palestine, Mesopotamia); 4 6 - 5 2 (Egypt); 5 2 - 6 (Armenia). See also F.Gabrieli,
Muhammad and the Conquests of Islam (London 1968), pp. 1 4 3 - 7 4 with literature; and
Winkelmann, 'Agypten und Byzanz', esp. 1611T.; W.E. Kaegi, jr., 'Heraklios and the
Arabs', Greek Orthodox Theol. Review 27 (1982), 109-33; F.Donner, The Early Islamic
Conquests (Princeton 1981), pp. 55IT.; and, for a general survey, A.G.K. Savvides, To
olxovfxevLxo Bv^avnvb xpctros xal rj eyupdvicrn TOV IaAdfjL. 518-717 /JL.X. (Athens
1985).
27
The question of monophysite 'nationalism' and hostility to Constantinople in the relatively
rapid Arab conquest of the Eastern provinces has been much discussed. See the comments
of Winkelmann, 'Agypten und Byzanz', 162IT., 176-8; and in general W.H.C. Frend, The
Rise of the Monophysite Movement. Chapters in the History of the Church in the fifth and sixth
Centuries (Cambridge 1972); Frend, 'Heresy and schism as social and national move-
ments', Studies*in Church History IX (1972), 37-56, see esp. 45ff.; and the summary of
Stratos, Byzantium in the Seventh Century, vol. II, pp. 1 1 7 - 3 3 , who surveys the modern
literature. In fact, as Winkelmann's careful analysis suggests, internal weakness as a
result of social tension (between big landowners and peasantry, for example) together
with administrative and military incompetence, were as much to blame. For further
discussion, see F. Winkelmann, 'Die Stellung Agyptens im ostromisch-byzantinischen
Reich', in Graeco-Coptica: Griechen und Kopten im byzantinischen Agypten, ed. P. Nagel (Halle
1984), pp. 11-35, see pp. 14fT.; also Kaegi, 'Heraklios and the Arabs', 127; and J.Moor-
head, 'The monophysite response to the Arab invasions', B 51 (1981), 5 7 9 - 9 1 .
28
For the Arabian background to the expansion of Islam, and for the life of Muhammad, see
Gabrieli, Muhammad and the Conquests of Islam, pp. 2 5 - 9 0 ; and the brief summaries in
Ostrogorsky, Geschichte, p. 92; Lilie, Die byzantinische Reaktion, pp. 34-9. For the conquest
 The East Roman world: the politics of survival 51

Heraclius' last years mark a sad end to what had been a glorious reign.
The Arab victories deprived the empire once again of territories that had
only recently been won back with such difficulty and sacrifice. He suffered
from a nervous disorder which meant that he could not bear to look upon
water. His natural son John Athalarich was involved in plots to remove
him from the throne; 29 and while the conspirators were betrayed and
punished, the plot itself was symptomatic of the strong hostility felt by the
populace of the city and many people in the ruling circle towards Hera-
clius' second wife, his niece, Martina. The marriage had been unpopular
from the start, regarded as incestuous and contrary to the civil law of the
state; and in spite of Martina's courage in accompanying the emperor on
his campaigns against the Persians, her interest in securing the succession
for her own sons with Heraclius, as well as for his son from his first
marriage, Constantine, seems to have aroused a great deal of suspicion.
The events in Constantinople surrounding the last years of the old Hera-
clius are confused, and the sources are often contradictory. 30 But in order
to secure the succession fairly for the children of both first and second
marriages, Heraclius raised his eldest son with Martina to the imperial
dignity on 7 July 638, making David, the next son with Martina, Caesar.
On Heraclius' death on 11 February 641, therefore, he was succeeded by
the fifteen year old Heraclius, known as Heracleonas, and the twenty-eight
year old Constantine. Martina was explicitly given a role to play, in so far
as she was to be regarded by both half brothers as mother and empress.
But the arrangement was destined not to work. To begin with, a strong
opposition, expressed particularly vocally through the senate, to Heraclius'
will quickly became apparent which, while accepting the half-brothers as
co-emperors, rejected Martina as unworthy of representing the empire or

of Persia, see Gabrieli, Muhammad and the Conquests of Islam, pp. 118-42. On the Arab
conquests in general for this period, see Lilie, Die byzantinische Reaktion, pp. 60-162;
H.Ahrweiler, 'L'Asie Mineure et les invasions arabes', RH 227 (1962), 1-32; E.W.
Brooks, 'The struggle with the Saracens 717-867', in CMH IV, 1st edn (Cambridge
1923), pp. 119-38; M. Canard, 'Byzantium and the Muslim world to the middle of the
eleventh century' in CMH vol. IV. 1, 2nd edn (Cambridge 1966), pp. 696ff., see 698-9;
D.R. Hill, The Termination of Hostilities in the Early Arab Conquest, A.D. 634-656 (London
1971); J. Wellhausen, Die Kdmpfe der Araber mit den Romdern in der Zeit der Umaijaden, in
Nachrichten der konigl. Gesellschaft der Wissenschaften zu Gottingen, phil.-hist. Klasse IV
(1901), pp. 1-34. For further literature see D.J. Constantelos, 'The Muslim conquest of the
Near East as revealed in the Greek sources of the seventh and eighth centuries', B 42
(1972), 325-57; and Donner, The Early Islamic Conquests.
29
See Stratos, Byzantium in the Seventh Century, vol. I, p. 219; vol. II, p. 136; Ostrogorsky,
Geschichte, p. 93; for the plot involving Athalaric and a number of the Armenian officers in
Heraclius' entourage, see Stratos, Byzantium in the Seventh Century, vol. II, pp. 137ff. with
literature.
30
The most useful general survey of these years, although not always the most objective
interpretation, is that of Stratos, Byzantium in the Seventh Century, vol. n, pp. 139ff., 176ff.
On the texts for these events, see esp. Speck, Das geteilte Dossier; and idem, 'Der Kater
e', 374ff.
52 Byzantium in the seventh century

of receiving foreign ambassadors, since she was a woman. 31 Two hostile

factions developed around the two half-brothers, but with Constantine's
death on 25 May 641 (according to one rumour, by poison), Heracleonas
and Martina seemed to have won the day. 32 Constantine's supporters were
now banished, but opposition seems only to have been strengthened. The
senatorial establishment, as well as the population of the city, and sig-
nificant elements of the army, were opposed to Martina's regency. The
rumours that Constantine had been poisoned received widespread accept-
ance, and there was a demand that his son, also Constantine, usually
named Constans, should be made co-emperor with Heracleonas. As troops
of the Anatolian armies appeared at Chalcedon, Heracleonas was forced to
accede to this demand. But even so, Heracleonas and Martina, along with
her other sons, David Tiberius and Marinus, were deposed, mutilated and
exiled (the first two to Rhodes) in September 641. Constans II began to rule
as sole emperor. 33
Heraclius' reign thus ended in confusion and political and military
uncertainty. But in spite of the catastrophes of his last years, his achieve-
ments remain clear. His great wars of reconquest restored the empire in the
East to the borders of Maurice's day, greater even than in the time of
Justinian; and while the effective loss of the Balkans to Slav peoples
changed the centre of gravity within the empire, Heraclius was neverthe-
less able through skilful diplomacy to reassert Byzantine imperial auth-
ority, however superficially, in a way that might have seemed impossible
at the end of the reign of Phocas.
Heraclius seems also to have carried through a number of administra-
tive reforms, chiefly connected with the needs or with the results of his
wars with Persia. 34 Changes in military organisation reflected both the
needs of the Persian wars on the one hand and the transfer of troops to
meet the demands of the war with the Arabs on the other. 35 By the same
token, a fiscal reform was undertaken, which reflected the beginnings of
the process through which the system familiar from the eighth century and
later came to be established.36 These changes again seem to reflect the
needs of the 'war economy' of much of Heraclius' reign. In this connection,
of course, Heraclius' introduction of the silver hexagram must be con-
sidered, part of his overall response to the dramatic scarcity of cash

31 See Nicephorus, 2 7 . 1 3 ; 2 8 . 5 .
32
Ostrogorsky, Geschichte, p. 94; Stratos, Byzantium in the Seventh Century, vol. II,
pp. 179-85.
33
Ostrogorsky, Geschichte, p. 95; Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 2ff.;
Herrin, Formation of Christendom, pp. 2 1 3 - 1 7 .
34
These will be discussed in detail in chapter 5.
35
Haldon, Byzantine Praetorians, pp. 1 4 2 - 5 0 , 1 6 4 - 8 2 .
36
Hendy, Studies, pp.417ff.
 The East Roman world: the politics of survival 53

resources during the first fifteen to twenty years of his reign, an innovation
which, as has also been noted, served its immediate purpose but, owing to
the inflexibility of the relationship between gold and silver which was
operated by the state, did not in the end prove stable. 37
Apart from this material and Heraclius' interest in the administrative
affairs of the Church and in canon law - reducing and subjecting to a more
rigorous central control the clergy of the larger churches, and especially of
the Hagia Sophia 38 - there is very little evidence to support the often
repeated notion of a major administrative reform during his reign. Numer-
ous administrative changes there certainly were, and most seem to have
been the result of considered policy plans designed to facilitate his military
operations or to alleviate the chaos of the years before 628. But there is no
reason to think that he introduced the 'theme system' or was responsible
for a major shift in recruitment policy for the army. I will return to these
questions in the following chapters.

CONSTANS II, 6 4 I - 6 6 8
Constans II was only eleven years old when he became sole ruler, and it is
clear that he had to rely upon the senate, both formally and informally,
during his first years. The senate itself, representative of the Constantino-
politan bureaucracy and the provincial landowning class, seems to have
experienced a considerable increase in its influence and prestige during
these years, a development which contrasts sharply with the rather sub-
servient position it occupied during much of the sixth century. Its power
may have been concealed in constitutional and legislative matters during
the sixth century, because the economic interests of its members were not
threatened; but it is interesting to note that it was at times of political and
economic insecurity or chaos that the senate as a body reasserted itself -
during the reign of Phocas, for example, and during the last years of
Heraclius, the reign of Heracleonas and Constantine, and the opening part
of the reign of Constans II. As has been shown, however, the senate
assumed a new position in the affairs of the state from this time on, as a
higher court and as a source of advice and advisers to the emperors too.
The reasons for this must be sought in both the political situation and the
prevailing ideological tendencies of the time, as well as in the actual
make-up of the senate and the social origins of its members. 39

37
See esp. Hendy, Studies, p. 494ff.
38 See J. Konidares, 'Die Novellen des Kaisers Herakleios', FM V (1982), 3 3 - 1 0 6 .
39 Ch. Diehl, 'Le Senat et le peuple byzantin a u x VII e et VIII e siecles', B 1 (1924), 2 0 1 - 1 3 ;
H.-G. Beck, Senat und Volk von Konstantinopel. Probleme der byzantinischen Verfassungsge-
schichte, in SBB VI (1966), 1-75; Ostrogorsky, Geschichte, p. 95f.
54 Byzantium in the seventh century

Plate 2.2 Constans II. Gold solidus

Constans' first years were dogged by the political difficulties inherited

from the reign of his half-uncle Heracleonas and his step-grandmother
Martina. The general Valentinus, who seems to have held the command of
the Opsikion forces, but may also have exercised the functions of an overall
commander of operations on the eastern front, had been made count or
comes of the excubitores, a small, elite palace guards unit, by Heracleonas
and Martina in an attempt to buy his loyalty; and there is some evidence
that he also had himself proclaimed co-emperor or Caesar (or that he tried
to do so). Although he had been one of the prime movers in Martina's
downfall, he was clearly keen to maintain his position of authority under
Constans II. But in attempting to impose his will in Constantinople
through his troops, he sparked off a popular riot which resulted in the
routing of his soldiers and his own lynching, probably in 644 or 645.
Constans was left in sole charge, aided and guided by his own officials and
the members of the senate. 40
He did not have a particularly auspicious start to his reign. The external
situation of the empire was once again precarious. Byzantine troops
evacuated Alexandria in September 642 as part of the agreement reached
by the patriarch Cyrus with the Arabs, on Martina's instructions, accord-
ing to which Byzantine troops and others who wished to leave would do
so over a specific period. On 29 September the victorious Arab leader
c
Amr entered the city and began systematically from that time on to
extend Muslim authority along the North African coast westwards. The
Pentapolis quickly followed and, in 643, Tripolis also. And in spite of a
temporary Byzantine reoccupation of Alexandria in 645, following c Amr's
recall, the Arabs were able to expel the Byzantines for good in 646. The
return of Byzantine authority and official Chalcedonian policy had not
been entirely welcome - indeed, the monophysite patriarch Benjamin
greeted the Arab leaders on their return at the head of a rejoicing proces-

por Valentinus and his role, see Haldon, Byzantine Praetorians, pp. 178ff., and Stratos,
Byzantium in the Seventh Century, vol. Ill, pp. 11-13.
 The East Roman world: the politics of survival 55

sion of Copts; and Egypt was henceforth to remain permanently in Muslim

hands. 41
Under a leader just as able as cAmr, Arab forces began in the early 640s
to raid Anatolia. From 642 to 643 raids into Armenia began in earnest;
and in 647 Mucawiya raided Cappadocia and besieged Caesarea. Phrygia
was raided from there, and Amorium was also attacked. No attempt to
hold key points or establish permanent bases was made; but enormous
booty was taken back to Damascus, where the future caliph received a
hero's welcome.42 Up until this point, and indeed until 655, Byzantine
control of the sea had been unchallenged. As they reached the Mediter-
ranean, however, the Arab leadership realised that to pre-empt Byzantine
counter-attacks and also to extend their conquests they would need a fleet.
Mucawiya was the first who seems to have recognised this, and in 649,
after the construction of a small fleet, the first Arab naval expedition took
place, directed against Cyprus. The capital Constantia was taken by storm;
and although a three-year truce was negotiated, Arab naval construction
proceeded apace. In 654 Rhodes was devastated (the fallen Colossus being
sold off to a merchant from Edessa); Cos was taken; Crete was pillaged. In
655, the Byzantine fleet, in an effort to put an end to this naval threat, and
under the personal command of the emperor, was decisively defeated by
Muc awiya's navy, and Constans barely escaped with his life.43
The civil wars which now developed upon the murder of c Othman in 656
between Mucawiya, proclaimed caliph in Syria, and cAli, in Medina, the
son-in-law of Muhammad, gave the Byzantines a breathing-space. In order
to concentrate his energies in defeating cAli's party, Mucawiya agreed to a
truce with the Byzantines in 659, which lasted until 661/2. 4 4 And while
hostilities, regular raids and attacks upon Byzantine fortresses and settle-
ments in Asia Minor continued thereafter with equal fierceness and reg-
ularity throughout the 660s, this truce had important consequences. To
begin with, it enabled Constans to turn his attention to the Balkans, where
41
On Cyrus and Benjamin, see Winkelmann, Agypten und Byzanz, 170-5, with literature;
and for the attempted reconquest under the general Manuel (launched from the Byzan-
tine base in Rhodes), see Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 35-8; AJ.
Butler, The Arab Conquest of Egypt and the last thirty years of the Roman Dominion (Oxford
1902), pp. 194ff.; Lilie, Die byzantinische Reaktion, pp. 4 8 - 5 2 .
42
See Lilie, Die byzantinische Reaktion, pp. 54ff., for a detailed account of these raids, their
frequency, direction and effects.
43
See Theophanes, 346.9sq.; Stratos, Byzantium in the Seventh Century, vol. Ill, p p . 4 8 - 5 5 ;
Lilie, Die byzantinische Reaktion, pp. 6 7 - 8 . On Arab sea-power see Stratos, vol. Ill, pp. 38ff.;
Lilie, pp. 64ff.; and esp. E. Eickhoff, Seekrieg und Seepolitik zwischen Islam und Abendland
(Berlin 1966); A.M. Fahmy, Muslim Sea-Power in the Eastern Mediterranean (London
1950).
44
For the civil war and its background, see Gabrieli, Muhammad and the Conquests of Islam,
pp. 9 4 - 8 ; and for its results for Byzantine strategy, Stratos, Byzantium in the Seventh
Century, vol. Ill, pp. 1 3 5 - 4 1 , 187ff.
56 Byzantium in the seventh century

Byzantine forces had been on the defensive since the beginning of his reign.
In 658, as internal strife among the Arabs relieved the pressure on the
eastern front, Constans undertook an expedition into the Sklaviniai, the
regions of the Slavs, where - according to Theophanes - he defeated many
tribes and took many prisoners. 45 The result was the transportation of
several thousands of prisoners with their families to devastated or unoccu-
pied areas of Anatolia and the probable recruitment of Slav soldiers into
the Byzantine forces there, 46 beginning a policy which was to be pursued
with vigour by both Constantine IV and Justinian II. It was also part of a
wider change in the constitution of the population of Asia Minor itself and,
more particularly, it hints at significant changes in methods of recruitment
and maintenance of the imperial forces in the provinces.

The monothelete controversy

Constans' internal policies and administration were dogged throughout
his reign by the problems inherited from his grandfather with regard to the
doctrines of monenergism and monotheletism. At his accession the Ekthe-
sis of Heraclius was still displayed in the narthex of the Great Church and
continued to represent the official policy of the state. Monotheletism had
already met with almost universal rejection in the West, both in Rome and,
much more significantly as it turned out, in Africa, where one of the most
important theologians of the orthodox Church, and certainly the most
important in his own time, Maximus the Confessor, was active in the
debate. In Syria and Palestine, in contrast, the neo-Chalcedonian commu-
nities seem to have espoused the formula of monotheletism with some
enthusiasm, as later evidence would suggest, and it is clear that this fact in
itself must have been a significant factor in persuading Constans and his
advisers to maintain the policy as the official line. 47
Early in 646, a number of local synods were held in the cities of the
exarchate of Carthage, probably in response to Maximus' campaign
against monotheletism; and the result was the condemnation of the
45
Theophanes, 3 4 7 . 6 .
46
For example, Theophanes, 3 4 8 . 1 8 (for 665): a group of 5,000 Slavs which deserted to the
Arabs in t h a t year and which m a y h a v e consisted of soldiers. See Lilie, Die byzantinische
Reaktion, pp. 70, 237fF.; Ditten, 'Zur Bedeutung der Einwanderung der Slawen', pp. 147ff.;
Ditten, 'Slawen im byzantinischen Heer', in Studien zum 7. Jhdt., pp. 7 7 - 9 1 , see 86ff.;
P. Charanis, 'The transfer of population as a policy in the Byzantine empire', Comparative
Studies in Society and History 3 (The Hague 1961), 1 4 0 - 5 4 , see 1 4 3 ; and see also Lemerle,
Les plus anciens recueils II, pp. 130f.
47
The adherence of the patriarch of Antioch Makarios, along with a seemingly large
n u m b e r of his clergy a n d the non-monophysite population to monotheletism after the
Arab conquest and u p to the sixth council of 6 8 0 (in w h i c h Makarios w a s formally
deposed), is good evidence for this. See Winkelmann, 'Die Quellen', 5 5 5 - 6 and sources.
 The East Roman world: the politics of survival 57

imperial policy as heresy. The exiled monothelete patriarch of Constantin-

ople, Pyrrhus, debated publicly with Maximus, under the chairmanship of
the exarch himself, Gregory, but had to acknowledge defeat; and it may
have been the solidarity of the North African bishops and population
against the imperial policy that encouraged the exarch to declare his
opposition to Constans' rule in 646/7 and have himself proclaimed
emperor - perhaps in conscious imitation of the great Heraclius. This
short-lived attempt met with failure when in the following year Gregory
was killed fighting Arab raiders near Sufetula.48 The raiders withdrew, but
imperial authority was quickly reasserted, although this had little effect
upon the internal political and ecclesiastical situation. In an effort to reach
yet another compromise, the patriarch Paul issued the famous Typos in
648, and in the name of the emperor, in which it was decreed that the
Ekthesis should be removed from the narthex of Hagia Sophia, but that at
the same time no further discussion of the issue was to take place: neither
was the question of the single will or energy of Christ to be debated, on pain
of punishment. A prohibition on discussion, however, designed to prevent
further disagreement, and to reinforce imperial authority, which was by
now deeply bound up with the monothelete doctrine, was unlikely to lead
to an acceptable solution. And indeed, at the Lateran council held in Rome
in 649, under the newly elected Pope Martin, the 105 bishops (chiefly from
the West) condemned both Ekthesis and Typos, along with the three
patriarchs Sergius, Pyrrhus and Paul (then occupying the patriarchal
throne). The results were circulated throughout the Church, in East and
West, as well as to the emperor himself. Maximus even described the
council as the sixth ecumenical council - -\ direct challenge to the emper-
or's traditional authority as convenor of such meetings; and recent
research has made it clear that the acts of the council were carefully
written up by Maximus and his supporters with the express intention of
achieving the greatest possible propaganda effect. The council itself seems
to have been orchestrated and dominated by Maximus and his confeder-
ates, and there can be little doubt that the later (anti-monothelete)
orthodox tradition relied for its acounts of the events and the history of the
controversy on these 'biased' reports.
The challenge to imperial authority could not be overlooked. It was a
total rejection of the imperial policy - and by implication therefore of the
emperor's orthodoxy - and his fitness to rule. Constans ordered Olympius,

48
For Gregory's rebellion and the opposition of the African clergy to monotheletism, see
Averil Cameron, 'Byzantine Africa - the literary evidence', pp. 56-9; I. Rochow, 'Zu
einigen oppositionellen religiosen Stromungen', in Byzanz im 7. Jhdt., pp. 225-88, see
263-4; Diehl, L'Afrique byzantine, pp. 555ff.; Stratos, Byzantium in the Seventh Century,
vol. Ill, pp. 62-73.
58 Byzantium in the seventh century

the exarch of Ravenna, to arrest Martin and to compel the bishops then
assembled in Rome to accept and sign the Typos. Like Gregory in Africa,
however, Olympius seems to have found it more acceptable to rebel and to
have himself proclaimed emperor. The imperial government, thoroughly
involved in the worsening situation in the East, did nothing. But, also like
Gregory, Olympius' independence was short-lived. He died in 652, and in
June 653 the new exarch with his troops was able to carry out Constans'
orders and arrest the pope. Martin arrived in Constantinople in December,
when he was immediately arraigned before the senate. The case against
him was based on political evidence - Olympius' rebellion providing the
most damning proof of his treasonable conduct. In spite of his protestations,
Martin was refused permission or opportunity to speak on the subject of
the Typos. He was found guilty and condemed to death, later commuted by
Constans to exile in Cherson, where he died - he had already been ill at the
time of his arrest - in 656. Maximus suffered a similar fate, arrested and
brought from Italy, imprisoned, tried and eventually - after the failure of
successive efforts to change his mind, in which the desire of the imperial
government for a public recognition from Maximus that the emperor's
authority was absolute was made strikingly clear - mutilated and exiled,
dying in 662 in the fortress of Schemarion in Lazica, in the Caucasus. 49
The imperial victory which was thus secured treated some of the
symptoms only, however, of a much deeper malaise within the political-
ideological world of East Rome. Imperial authority had been forcefully and,
in the eyes of many, effectively placed under a question mark. The rights of
emperors to define dogma and to approve or disapprove of synodal and
other ecclesiastical gatherings of similar status had been queried. But at an
even deeper level, the public political debate represented by the mono-
thelete controversy reflected also widespread shifts within the whole
framework of the formal imperial ideology and indeed beyond this, within
the symbolic universe of Byzantine culture itself. The central position of the
emperor in respect of certain key questions had been thrown into relief; his
49
For Olympius and Italy, see Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 104-11;
Stratos, 'The exarch Olympios and the supposed Arab invasion of Sicily in A.D. 652', JOB
25 (1976), 63-73; and for the history and development of the monothelete controversy,
the roles of Martin and Maximus, and the outcome of their trials, see esp. Beck, Kirche,
pp. 430-73; Van Dieten, Geschichte der Patriarchen, pp. 179-218; P.Verghese, The
monothelite controversy - a historical survey', Greek Orthodox Theol. Review 13 (1968),
196-211; V.Grumel, 'Recherches sur l'histoire du monothelisme', EO 27 (1928), 6-16,
257-77; 28 (1929), 19-34, 272-83; 29 (1930), 16-28; Herrin, Formation of Christen-
dom, pp. 255-9, 263-5; and the fundamental studies of R. Riedinger, 'Aus den Akten der
Lateransynode von 649', BZ 69 (1976), 17-38; 'Griechische Konzilsakten auf dem Wege
ins lateinische Mittelalter', Annuarium Historiae Conciliorum 9 (1977), 253-301; 'Die
Lateransynode von 649. Ein Werk der Byzantiner um Maximos Homologetes', Byzantina
13 (1985), 519-34; with further literature and discussion in Winkelmann, 'Die Quellen',
515-59, see 538.
 The East Roman world: the politics of survival 59

relationship to God and his function as God's vicegerent had also, in some
respects, been re-examined. By the later 640s, of course, the territories for
whose predominantly monophysite populations the imperial monothele-
tism of the court and the Constantinopolitan patriarchate had been origi-
nally intended were lost to the empire. No doubt it was felt that attempts
might be made in the future to bring them back into the empire, and so
monotheletism retained its political-religious relevance in this respect. But
essentially, monotheletism, and more particularly the efforts of the
emperor to impose a monothelete imperial policy on the Chalcedonian
clergy, were responses to a different set of developments. As we shall see in
a later chapter, these developments and the tendencies noted above were
fundamental to the later social and political history of the Byzantine state
in the later seventh and eighth centuries. 50

The Sicilian connection

In 654, Constans had crowned his son Constantine co-emperor; in 659 his
two younger sons, Heraclius and Tiberius, were likewise raised to the
imperial dignity. With this move Constans took the final step in passing
over his younger brother Theodosius; and in 660 he had Theodosius
tonsured and shortly afterwards murdered. It may well be that Theodosius
was involved in a plot of anti-monothelete groups to depose Constans.
However that may be, his treatment of Martin and Maximus, and finally of
his brother, seems to have roused the hostility and dislike for him of the
population of Constantinople. He was branded as a second Cain; and
whether or not he had also long-term strategic reasons for the move, this
seems to have encouraged him to transfer the seat of his government from
Constantinople to the West. It is indeed interesting, for - as Ostrogorsky
noted - the idea does seem to represent a continuity with the plans of both
Maurice and Heraclius. 51 Whether this was conscious, or reflected merely
the common-sense strategic requirements of a government under intense
pressure, must remain uncertain. What it does illustrate is that there was
clearly no thought that the West at this time was in some way less central
or important to the empire. Indeed, it suggests on the contrary that
Constans himself saw the West as both a strategically and politically safer
50
Maximus w a s the first to h a v e argued t h a t the emperor, as a layperson, had no
jurisdiction in questions of belief: see, e.g., PG XC, col. 117B-C. This was, of course, to
prove a n extremely important a r g u m e n t in later debates involving the relationship
between secular a n d ecclesiastical authority in both East and West. For the implications of
this controversy and a n analysis of its context, see Haldon, 'Ideology and social change',
esp. 166ff.
51
Ostrogorsky, Geschichte, p. 1 0 1 . On the m u r d e r of Theodosius, see Stratos, Byzantium in
the Seventh Century, vol. Ill, pp. 191ff.; and Theophanes, 3 4 7 and 3 5 1 .
60 Byzantium in the seventh century

seat of government and administration. The empire was still in the eyes of
its subjects and its rulers the Roman empire - Rome itself and Italy (in spite
of the Lombards) were a fundamental element of that empire - and, in spite
of Islamic naval threats, the empire was still united in its parts by the
Mediterranean Sea. Constans' move is a striking testimony to that set of
beliefs - even if it is also a testament to the beginning of the disintegration
of that pan-Mediterranean world and its culture, and an attempt to
reaffirm what was all too rapidly becoming a facet of the glorious past. 52
It is clear from Constans' activities in Italy, however, that he intended
originally to subjugate the Lombards and at the same time to reorganise
the defence and administration of Africa against impending Arab attack.
He left Constantinople in 661/2, stopping at Thessaloniki, marching on to
Athens and Corinth and, embarking again from Corinth or possibly Patras,
sailed to Otranto. From here he opened his campaign against the Lom-
bards, at first with considerable success, although failing to take Ben-
evento. As his forces began to suffer from lack of supplies, he eventually
had to abandon the siege and retire to Naples, whence he briefly visited
Rome. For twelve days he participated in celebrations and prayer, greeted
and entertained by Pope Vitalian. Thereafter he returned to Naples and
finally established his headquarters in Sicily, at Syracuse. From that point
his intention was to set up the defences of the island and of southern Italy
against Arab sea-borne attack, and to reassert imperial authority in Africa,
which may still have been at that time hostile to his rule, following
Gregory's failed rebellion.53
Constans' rule in Sicily rapidly became extremely unpopular. He seems
to have been in considerable financial difficulties and had to ask for money
from the Church of Ravenna. In 666 the see of Ravenna was granted
independence from Rome - autocephaly - by Constans (partly, it has been
suggested, in response to the generosity of the Church there). But the
presence of large numbers of troops of the imperial armies was an enor-
mous burden on the resources of the island and indeed of all southern Italy.
While the strategy which led him there seems to have been entirely
logical, his desire to transfer his headquarters there on a permanent basis
met with a great deal of opposition. His wife and three sons were prevented
from joining him by officers of state and by members of the demes. Finally,
in 668, he became the victim of an assassination plot. On 15 September he
was murdered in his bath by a cubicularius, and shortly thereafter the
52
A good description of the campaigns u p to the emperor's establishment in Italy can be
found in Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 197ff.
53
The m a r c h to Italy, the campaigns undertaken en route, a n d Constans' stay in Sicily have
been the subject of m u c h discussion. See Ostrogorsky, Geschichte, pp. 1 0 1 - 3 ; Stratos,
Byzantium in the Seventh Century, vol. Ill, pp. 2 0 9 - 1 7 , vol. IV, pp. 8f. a n d esp. P. Corsi, La
spedizione Italiana di Costante 11 (Bologna 1983)
 The East Roman world: the politics of survival 61

Armenian general Mzez Gnouni, or Mizizios, as he is called in the Greek

sources, count of the Opsikion, the praesental army of the emperor, was
acclaimed emperor. He seems to have had little actual or serious support,
for the arrival of the exarch Gregory with his troops put an end to the
usurpation. Mzez was executed.54 Constans' body was returned to Con-
stantinople, where it was interred in the Church of the Holy Apostles.
The coup which ended the reign of Constans, when he was only thirty
years of age, was in itself not an entirely unexpected turn of events. As we
have seen, the treatment he meted out to his brother and to Martin and
Maximus and their followers had not added to his popularity. In addition,
there surfaced from time to time in the background of Constans' reign a
faction or factions within the senate and the ruling military circles which
were hostile to him. He had already at the beginning of his reign been
threatened by the power of Valentinus; but the latter seems to have been
checked by the senate and other factions and interests in the capital. He
had been fortunate in the failure of first the exarch Gregory and then
Olympius, in Africa and Italy respectively. Both were areas where the
combination of civil and military authority in the hands of a single
governor made opposition to Constantinople a possibility; both were areas
where cultural and linguistic differences made the development of a
localised and occasionally - depending upon the political and ideological
context - anti-Constantinopolitan feeling inevitable.55 Already in 616 and
619, and again in 642, there had been mutinies or rebellions in Italy; and
while the first seems to have been little more than a mutiny over pay in the
army, those of 619 and 642 were clearly of a more political nature. 56
Equally problematic for Constans seems to have been the Armenian
military faction, whose members dominated the chief military posts at this
period. A plot in 652 involving members of the senate and of the Armenian
military aristocracy, some of whom held commands in the forces then in
Thrace, was betrayed to Constans while he was himself in Armenia, and he
was able to arrest the conspirators in good time. 57 This plot in particular
may represent senatorial hostility to the no longer malleable young
54
See Haldon, Byzantine Praetorians, p. 1 7 9 a n d note 3 7 7 for Constans' troops in Italy a n d
Sicily. For Mzez, see ibid., p. 3 5 9 a n d note 1 0 9 1 ; a n d Stratos, Byzantium in the Seventh
Century, vol. IV, pp. 8 - 1 4 ; Ostrogorsky, Geschichte, pp. 1 0 2 - 3 ; Guillou, Regionalisme,
p. 160; T.S. Brown, 'The Church of R a v e n n a a n d the imperial administration in the
seventh century', EHR, 9 4 (1979), 1 - 2 8 , esp. 17; for t h e g r a n t of autocephaly, ibid., llff.
55
See Guillou, Regionalisme, esp. pp. 231ff. a n d 236ff. comparing Egypt, Africa a n d Italy;
a n d t h e c o m m e n t s of T.S. Brown, 'The Interplay between R o m a n a n d Byzantine Traditi-
ons', 148ff. a n d 1 5 1 ^ .
56
See Ch. Diehl, VExarchat de Ravenne, pp. 340f.; Guillou, Regionalisme, pp. 204ff.; Stratos,
Byzantium in the Seventh Century, vol. I, pp. 1 2 1 - 2 , vol. Ill, pp. 76-9; Brown, Gentlemen
and Officers, pp. 1 5 9 - 6 3 .
57
See Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 190-1.
62 Byzantium in the seventh century

emperor, who had been under the influence of his advisers since his
accession, but who was by this time clearly beginning to assert his
independence with increasing confidence.
Constans seems throughout his reign to have been the target of conspi-
racies, and these were, in many respects - and as I have argued elsewhere
- a reflection or symptom of deeper changes in attitudes to the person of
the emperor and to his role, as well as of the often confused situation in
which the empire found itself at this period. Rebellions in Armenia were an
important political factor for the government, for the nationalism and the
clan traditions of the powerful Armenian princely and noble houses meant
that the area was easily alienated. The continued conflict between Chalce-
donian and monophysite Churches there, and the vested interests of both
the caliphate (which now replaced the Sassanids in this role) and the
Byzantines, meant that Armenia was a constant source of conflict and a
potential threat to imperial security. In 648 or 649 Constans issued an
order that the Armenian Church should accept its subordination to the
Church of Constantinople and the Chalcedonian creed. In response, the
Armenian clergy and many princes, including the Byzantine governor,
Theodore Rstuni, met at the synod of Dvin in 649, where the imperial
order was condemned and rejected.58 The result was an agreement
between Armenia and the Arabs by which the former threw off Byzantine
authority and accepted Muslim overlordship. Constans was persuaded to
march against the rebellious princes and, in a short campaign, suceeded in
forcing the rebels to withdraw. But he had to return to the West in 652 in
order to deal with the plot in Thrace in that year, leaving as commander-
in-chief the general Marianus who, after the Armenian rebels had received
Arab reinforcements, was defeated and his troops routed. Armenia
remained semi-independent and under Muslim domination until after the
murder of the Caliph Othman; upon which the Armenian Prince Hama-
zasp Mamikonian once again brought Armenia back into the Byzantine
fold.59 This was not the end of the problem, however. A rebellion of the
strategos of the Armeniakon district, the Armenian Sahpur (Saborios), took
place in 668 while Constans was in the West. Again, the issue was
resolved by the untimely death of the rebel, whereupon the troops of the
region reaffirmed their loyalty to the empire. But it is once more sympto-
matic of the sort of difficulties facing the government at this time in respect

58
For a s u m m a r y of Byzantine-Armenian relations a n d the Arab attacks in t h e Caucasus
see Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 19ff.; a n d for the attempt at
union a n d subsequent developments, ibid., 2 6 - 3 1 ; Winkelmann, Die ostlichen Kirchen,
pp. 129-30.
59
On these events see Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 2 7 - 3 1 ; Kaegi,
Military Unrest, pp. 1 6 2 - 3 .
 The East Roman world: the politics of survival 63

of the complex interrelationship between local identities, state politics and

religious issues. 60
Constans' reign is in many ways the most crucial of the seventh century,
although we are often able to interpret with difficulty the developments
reflected in the rather limited source materials. It was during his reign that
the limits of Muslim expansion into Anatolia were reached and that what
was to become for the next two centuries the 'frontier zone' was attained. It
was during his reign, in all probability, that the basic elements of thematic
provincial and military administration came into being. He was the first
emperor to realise the potential of transplanting large numbers of Slav
prisoners to devastated areas of Anatolia, both to provide manpower for
the army and to repopulate the regions most affected by Arab raiding. He
had to deal with the Arab threat on both a naval and military basis; and it
was also during his reign that the importance of the Armenian noble clans
and of Armenians in Byzantine service takes on a real significance. Most
importantly, perhaps, Constans represents in many respects the ruler who
epitomised the struggle of the emperors to retain their absolute authority in
both Church and state affairs on the Justinianic and traditional model.
Thereafter, the role of the emperors in Church matters and the position of
the emperor in the relationship between God and humanity, as theorised in
formal ideology and popular conception, underwent a series of subtle
changes which were to have important consequences for his successors.

CONSTANTINE IV, 6 6 8 - 6 8 5
Constans was succeeded by his son Constantine, during whose reign the
first great siege of Constantinople took place. The Arab offensive had been
renewed in 663, shortly after Muc awiya's victory over cAli, and this time
the general outline of a coherent strategy became apparent: together with
constant raids along the frontier and raids deep into the Anatolian
provinces, designed to disrupt economic activity and the defensive capabi-
lities of the Byzantine armies, Mu c awiya's ultimate aim was clearly the
capture of Constantinople itself and the overthrow of Islam's only
remaining foe of any consequence in the Mediterranean. 61 Cyprus,
Rhodes, Cos and Cyzicus had all been occupied by Arab naval forces by
670, thus securing a firm base from which to mount the final assault on
the city. In 672 another squadron occupied Smyrna and set up another,
temporary, base; and in 674 the main action commenced. A large Arab
60
See Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 236ff.; Kaegi, Military Unrest,
pp. 166f.
61
For a detailed analysis, see Lilie, Die byzantinische Reaktion, pp. 69-97; and note Gabrieli,
Muhammad and the Conquests of Islam, pp. 2 3 2 - 3
64 Byzantium in the seventh century

Plate 2.3 Constantine IV. Gold solidus

fleet arrived off Constantinople and blockaded the city. Each summer for
the next four years the same fleet blockaded Constantinople, withdrawing
to shelter off Cyzicus for the winter season; and each year the defence held
firm. During the final naval battle the Byzantine fleet used its terror
weapon, mentioned in the sources for the first time, the deadly Greek fire.
Supposedly invented by the Syrian architect and mechanic Callinicus, who
had fled from his homeland to Constantinople, this seems to have consisted
essentially of crude oil, possibly with a few extra ingredients (various
inflammable and 'sticky' resins, for example) which when heated was
expelled through a siphon at enemy vessels. It was, in effect, a sort of
primitive napalm. The result was a major Byzantine victory. At the same
time, Byzantine land forces were able to surprise and defeat one of the
major Arab columns in Anatolia; and Mucawiya was forced to withdraw
his armies and ships and sue for peace. A thirty-year peace was signed, and
the caliph undertook to make a yearly payment of 3,000 gold pieces,
together with fifty Byzantine prisoners and fifty stallions.62
The effects of this defeat were enormous. It signalled the end of Muca-
wiya's plans to take Constantinople and incorporate Rum into the cali-
phate. It immediately increased Byzantine prestige in the Balkans and the
West, so that both the khagan of the Avars (now confined effectively to the
plain of Hungary) and the princes and chieftains of the Balkan Slavs sent
ambassadors to Constantinople bearing gifts and recognising Byzantine
supremacy. 63 In the East, peace with the caliphate meant that the Byzan-
tines could now concentrate properly on their northern and western
fronts. At the same time, internal dissension within the caliphate after
Mucawiya's death in 680 and the accession of Yazid I meant the end, for

62
For the siege a n d the eventual Byzantine victory, see Lilie, Die byzantinische Reaktion,
pp. 7 6 - 8 2 ; Dolger, Regesten, no. 2 3 9 ; Stratos, Byzantium in the Seventh Century, vol. IV,
pp. 2 9 - 3 9 ; Ostrogorsky, Geschichte, pp. 1 0 3 ^ . On Greek fire see J.F. Haldon and
M.Byrne, 'A possible solution to the problem of Greek fire', BZ 70 (1977), 9 1 - 9 .
63
Cf. Theophanes, 356.2sq.; Nicephorus, 33.6sq.
 300 miles
Themata
1 Opsikion ^00 kilometres
2 Thrakesion
3 Karabisianoi
4 Anatolikon
5 Armeniakon

" Lost to Arabs c. 670-698 /. Approximate extent of

imperial territory c. 650-700
^ . i,-/
• •' O c c u p i e d by Arabs c. 7 0 0 - 7 2 0 1 1 1 1 R e g i o n s constantly raided
'.'.'. a n d after MM

Regions regularly devastated

or made insecure

Map III The empire in c. A.D. 650-700: the process of devastation

66 Byzantium in the seventh century

the time being, of the Arab threat. 64 The Anatolian provinces were given a
short time to recover from the economic and demographic devastation
they had suffered, while the empire was able to consolidate the administra-
tive and organisational changes that had taken place since the 640s.

The Bulgars

In the Balkans, however, the empire was soon faced by a new and equally
dangerous enemy. While a notional suzerainty was still exercised by the
Byzantines over much of the peninsula south of the Danube, this was
limited in real terms to some coastal and riverine settlements and fortresses
and certain littoral strips, particularly in Greece and the Peloponnese, but
also in Dalmatia. The extent inland of actual Byzantine authority remains
unclear, although the subject is still hotly debated. Numerous attacks were
directed by the Slavs against Byzantine fortresses and towns, and in
particular on Thessaloniki, which continued throughout the century to
lead a precarious existence. It was besieged in 675-6 and 677; while Slav
pirates raided other Byzantine coastal lands in 678-9, for example. 65 The
number of punitive expeditions mounted by the emperors in the last third
of the seventh century suggests that imperial control can have been
neither very great nor at all secure. 66 The situation changed dramatically,
however, with the advent of the Bulgars. The khanate of old Great
Bulgaria, a confederation of Onogur-Bulgar clans, had already entered
into friendly relations with Byzantium in the first half of the seventh
century, as we have seen; and it was with Byzantine support that its ruler,

64
Lilie, Die byzantinische Reaktion, pp. 97ff. The question of the involvement of the Mardaites
in compelling the Arabs to treat with the Byzantines is also relevant, of course. These
m o u n t a i n brigands - w h e t h e r or not they were encouraged by official Byzantine policy -
were a real threat to the internal security of the Lebanon a n d n o r t h Syria; and it seems
t h a t the Byzantines were able to use t h e m as a further inducement for the caliphate to
come to terms. See Stratos, Byzantium in the Seventh Century, vol. IV, pp. 39ff.; Lilie, Die
byzantinische Reaktion, pp. 9 9 , lOlf. On the civil w a r in the caliphate, see G. Rotter, 'Die
Umayyaden u n d der zweite Biirgerkrieg ( 6 8 0 - 6 9 2 ) ' , A b h a n d l u n g e n fur die Kunde des
Morgenlandes, XXXXV, 3 (Wiesbaden 1982).
65
See Ditten, 'Zur Bedeutung der E i n w a n d e r u n g der Slawen', pp. 149f.; and P. Lemerle,
'Invasions et migrations d a n s les Balkans depuis la fin de l'epoque r o m a i n e j u s q u ' a u VHP
siecle', RH 2 1 1 (1954), 2 6 5 - 3 0 8 , see 301f.; Stratos, Byzantium in the Seventh Century,
vol. IV, pp. 63ff.
66
See the excellent discussion with literature of Ditten, 'Zur Bedeutung der Einwanderung
der Slawen', pp. 1 3 2 - 6 with 1 1 3 - 1 9 . Note also Obolensky, The Byzantine Commonwealth,
p. 59; Stratos, Byzantium in the Seventh Century, vol. I, pp. 332ff.; Ostrogorsky, Geschichte,
pp. 8 7 - 8 ; a n d Lemerle, Les plus anciens recueils, II, p. 1 2 8 . Important gains were made, of
course. After the expedition of Constantine IV in 6 7 8 some degree of Byzantine control
seems to h a v e been accepted by Slav tribes in the Thessaloniki region. The political events
of the period 6 8 2 - 4 a n d the Kouber/Mauros episode seems to confirm this. See Lemerle,
Les plus anciens recueils, II, pp. 1 3 8 - 6 2 (and I, pp. 2 2 2 - 3 4 for the text).
 The East Roman world: the politics of survival 67

Kovrat, threw off the Avar yoke. About the middle of the century this loose
amalgam was broken up by the arrival of the Chazars. Some of the Bulgars
were incorporated into the Chazar khanate; other groups fled, mostly
westwards. Among them was a group under the leadership of Asparuch,
one of the sons of Kovrat, who appeared north of the Danube delta in the
years after 670, intending to settle in 'Byzantine' territory across the
Danube and to exploit the fertile pasturelands to the south. The Byzantines
certainly recognised no threat in the Bulgars' arrival; but they were
nevertheless unwilling to permit them to enter what was, in political
theory if not in actuality, Byzantine territory. In 680 a fleet was despatched
with troops to the mouth of the Danube, while a cavalry force marched up
from Thrace, intending to expel the Bulgars from their stronghold in the
delta. The Bulgars avoided open battle, but were able to take advantage of
a Byzantine withdrawal to take the imperial forces by surprise and inflict a
substantial defeat upon them. In 681 Constantine IV concluded a treaty
with Asparuch which recognised the Bulgar occupation of the territories
already held and agreed to an annual tribute or subsidy.67 As a result of
this arrangement, the Byzantines lost control of a number of Slav groups
who had hitherto recognised Byzantine overlordship in the area about the
lower Danube from the Dniester to the Balkan range itself, including part of
the plain of Walachia, south Bessarabia, the Dobrudja, and the older
province of Moesia Inferior. The so-called 'seven tribes', the Severi, and
several other groups in these regions now came under Bulgar over-
lordship, and from this time the development of a Bulgaro-Slav state in the
north-eastern Balkan zone can be followed. The capital of this new
khanate was established at Pliska, strategically placed to control the
approaches to the Dobrudja, and the route from the Danube via Anchialus
to Constantinople itself. From this time on, the existence of an independent
and often hostile Bulgar power was to be the cause of some of the greatest
difficulties faced by the empire and was to have a decisive influence on the
course of Byzantine history. 68

The end of monotheletism

These new developments in the Balkans coincided with a major shift in
Byzantine policy regarding monotheletism. As we have seen, Constans had
enforced the observance of his Typos of 648 and had eventually bullied the
67
See Ostrogorsky, Geschichte, p. 1 0 5 ; Stratos, Byzantium in the Seventh Century, vol. IV,
pp. 93ff., 1 0 1 - 1 3 ; Avenarius, Die Awaren in Europa, pp. 171ff.; Lilie, Die byzantinische
Reaktion, pp. 9 9 - 1 0 0 .
68
See Obolensky, The Byzantine Commonwealth, pp. 63f.; Ditten, Zur Bedeutung der Einwan-
d e r u n g der Slawen, pp. 1 3 5 a n d 150; Stratos, Byzantium in the Seventh Century, vol. IV,
pp. lOlff.; Ostrogorsky, Geschichte, p. 1 0 6 .
68 Byzantium in the seventh century

orthodox opposition to his policies into silence. With his death, and more
importantly with the pragmatic recognition that the relevance of this
compromise was now lost, along with the majority of the monophysite
provinces, while Rome and the West retained their political and strategic
significance for the empire, Constantine IV decided that the time was ripe
for a reconciliation between the imperial and the Western Churches. With
the agreement of Rome, the sixth ecumenical council was convoked and
met from 3 November 680 until 16 September 681. Its main task was to
rescind the doctrines of monenergism and monotheletism and return the
Christian world to doctrinal unity. Those who were held responsible for its
introduction and spread were anathematised: the patriarchs Sergius,
Pyrrhus and Paul, together with Pope Honorius. Altogether the council
held eighteen sittings, mostly presided over and led by the emperor himself;
and at its final meeting he was hailed as the 4new Marcian' and the 'new
Justinian', the destroyer of heretics.69 The sixth council represents an
important moment in East-West relations and in the history of the Church;
it represents also a recognition on the part of Constantine that the split in
the Church, which had without any doubt been promoted by his father,
was injurious to his own position, as orthodox ruler and defender of the
faith, and his authority in theoretical and theological terms. In practical
respects, of course, this was unquestioned. But as long as an argument
based upon his 'heretical' stand as a monothelete could be voiced, his
position was threatened.70 Constantine took some time to reach his final
decision on the convening of a council, however; and in his letter to the
patriarch George I (679-86), with whom he had replaced the less mallea-
ble Theodore I in 679, he explains the long delay in the calling of a synod
as a result of the many cares and problems he had had to deal with in view
of the hard-pressed military situation.71 While this was certainly the case,
it is reasonable to suppose that Constantine deliberated over his change of
policy for some considerable time.
Internal conflict on the political-ideological plane was not lacking,
however. In order to secure his own position and that of his young son
Justinian, Constantine decided to deprive his two younger brothers, Hera-
clius and Tiberius, crowned during Constans' reign, of all rights to the
throne and the imperial dignity. This met with immediate opposition from
69
MansiXI, 656; Riedinger, 798.10-11.
70
For the council, see L. Br6hier, in A. Fliche and V. Martin, eds., Histoire de Veglise depuis les
origines jusqu'a nos jours V: Gregoire le grand, les etats barbares et la conquete arabe
(590-757) (Paris 1938), pp. 1831T.; F.X. Murphy and P.Sherwood, Constantinople II et
Constantinople III (Paris 1974), pp. 133ff.; Winkelmann, Die ostlichen Kirchen, pp. 110-12.
See also Stratos, Byzantium in the Seventh Century, vol. IV, pp. 115-31; Herrin, Formation
of Christendom, pp. 275-81.
71
See Mansi, XI201C-D; Riedinger. 10.21-25.
 The East Roman world: the politics of survival 69

the army and the senate, who already in 670 had extracted a confirmation
of his brothers' rights from Constantine; and in 681 troops from the
Anatolikon thema appeared at Chalcedon demanding that Constantine
respect the position and status of the two brothers. 'We believe in the
Trinity', Theophanes reports them to have said, 'and we wish to see three
crowned emperors.' 72 Constantine brooked no opposition. He deprived his
brothers of their titles and shortly afterwards had their noses slit. The
leaders of the Anatolikon mutineers were arrested and executed, and the
troops returned to their bases. Imperial authority had once again been
restored. It has been noted that Constantine's action was similar in its
effects to that of Constans against his brother Theodosius, a move tending
to reinforce the growing tradition of single rule passed on to an eldest son;
so that while sons of emperors are crowned as equals in theory, in practice
power was exercised only by the autokrator himself. This is an important
development, for it sees the gradual ending of the traditional 'college' of
emperors which had been the norm in the later sixth century and under
Heraclius.73

Constantine's last years were marked by a relatively stable situation on

both the northern and eastern fronts. In 684/5 he personally led an
expedition into Cilicia, threatening north Syria, and compelled the caliph
c
Abd al-Malik to pay a tribute of 1,000 nomismata per day, together with a
symbolic horse and a slave, in return for the Byzantines halting their
advance. 74 In Italy the reconciliation with Rome (in spite of the long-term
results of the agreement, which were again to cause dissension between
Rome and Constantinople over their respective ecumenical status),
together with a peace arranged through the papacy with the Lombards,75
brought a situation of relative unrest to an end; while in Africa, in spite of
Arab successes in regularly raiding the Byzantine provinces and, more
importantly, in beginning to outflank them through the large-scale forced
conversion of the Berber population to Islam, the key fortresses and cities of
the exarchate still held firm. Indeed, during Constantine's last years the
Byzantines were able to take advantage of an alliance with the Berbers to
defeat a major Arab force under the great leader cUqba. The Arabs were

72
For the confirmation of 6 7 0 , Dolger, Regesten, n o . 2 3 6 . For the d e m a n d s of the Anatolikon,
see Theophanes, 3 5 2 . 1 5 and see Stratos, Byzantium in the Seventh Century, vol. IV,
pp. 1 3 5 - 4 0 , w h o argues t h a t the brothers were in addition involved in a plot, together
with troops from the Anatolikon thema w h o h a d recently been defeated by the Bulgars, to
m a i n t a i n state monotheletism.
73
See the c o m m e n t of Ostrogorsky, Geschichte, pp. 107f.
74
See Lilie's account, Die byzantinische Reaktion, pp. 100f.; and Stratos, Byzantium in the
Seventh Century, vol. IV, pp 165ff.
75
For a s u m m a r y of the events, Stratos, Byzantium in the Seventh Century, vol. IV, pp. 5 5 - 6 2 .
70 Byzantium in the seventh century

Plate 2.4 Leontius II. Gold solidus

forced to evacuate the territories they had occupied in Tripolitania and
Byzacene and withdraw to the Libyan littoral, where the Pentapolis had
been in Muslim hands since the 640s. And while this situation lasted only
a few years, it demonstrated that the conquest of North Africa would not
be as easy as that of Egypt. On the other hand, it is also clear that the
Byzantines themselves contributed very little to Arab difficulties, with-
drawing to their cities and fortresses and, on the whole, passively awaiting
the outcome. One by one the Berber clans were converted to Islam, the
remaining, isolated Byzantine fortresses and settlements fell, until Car-
thage itself was taken in 697 and then, after a temporary Byzantine
success in retaking it, definitively in 698. 7 6

JUSTINIAN II, 6 8 5 - 6 9 5 AND 7 0 5 - 7 I I

Constantine, who was only thirty-five years old when he died, was suc-
ceeded by his seventeen-year old son, Justinian II, in September 685. He
seems to have been a despotic and arrogant young man, not gifted with his
father's tact or willingness (under certain conditions) to compromise. He
has been compared, unfavourably, with his autocratic grandfather Con-
stans. He was an able ruler, however, and in his first year exploited the
internal situation of the caliphate greatly to his advantage. The threat of
new attacks and a renewed Byzantine offensive persuaded the Caliph
c
Abd al-Malik to confirm the arrangements he had previously made with
Constantine IV. Not only were the tribute payments increased, but he
agreed also to share the income from the island of Cyprus, and that from
Armenia and Iberia, with the Byzantines. Cyprus remained a shared
territory for many years thereafter; while Iberian and Armenian revenues
76
Gabrieli, Muhammad and the Conquests of Islam, pp. 180ff.; Averil Cameron, 'Byzantine
Africa - the Literary Evidence', pp. 59f.; J.Durliat, 'Les Attributions civiles des eveques
byzantins: l'exemple du Diocese d'Afrique (533-709)', JOB 32, 2 (1982), 73-84, see
78(Y. ( = Akten des XVI. Internationalen Byzantinisten-Kongresses II, 2); Diehl, L'Afrique
byzantine, pp. 576-92; M. Brett, 'The Arab conquest and the rise of Islam in N. Africa', in
Cambridge History of Africa, vol. II (Cambridge 1978), pp.490-555, esp. 503-13. See also
Av. Cameron, 'Gelimer's laughter: the case of Byzantine Africa', in EM. Clover, R.S.
Humphreys, eds.. Tradition andlnnovation in Late Antiquity (Madison 1989) 171-90 (repr. in
eadem, Changing Cultures in Early Byzantium [Aldershot 1996] VIQ).
 The East Roman world: the politics of survival 71

were in fact no great loss, since these territories were barely under the
caliph's authority at that time, owing to the civil war. In return, the
Mardaites, who had been a thorn in the flesh of the Muslim authorities in
north Syria and Lebanon, using their mountain fastnesses as a base from
which to plunder and raid the surrounding countryside, were transferred
to western Asia Minor. 77 The Byzantine general Leontius marched against
the forces of Ibn az-Zubair, the common enemy now of both the caliph and
the emperor, in Armenia, Iberia and Caucasian Albania, devastating wide
areas and taking many prisoners. This occurred in 688/9.
The truce in the East, and the relatively favourable position of the
empire, meant that Justinian could turn his attention to the Balkans. In
687/8 troops were transferred from Anatolia to Thrace, and in the follow-
ing year Justinian himself led the expedition against the Slavs and Bulgars,
breaking through to Thessaloniki and subjecting a large number of Slavs
to Byzantine rule. The real situation in the Balkans is reflected in the fact
that the emperor had to fight his way through from Thrace to Macedonia.
Most significant for the empire was the transfer of large numbers of Slavs
to Asia Minor, in particular to Bithynia and Cappadocia, where they seem
to have been eventually drafted into the provincial armies, as a number of
lead seals of the officials who dealt with them demonstrate. 78 Justinian also
undertook the transfer of a section of the population of Cyprus to the area
of Cyzicus, hard-hit by the Arab occupation of 674-8 - as well as of the
Mardaites, mentioned already, to the Peloponnese, parts of the south-west
coastal region of Asia Minor, the island of Cephallenia and the region
around the important port of Nicopolis in Epirus.79
The transfer of the Cypriots seems to have been a breach of the agree-
ment whereby the island was to remain neutral, although it is unclear as
77
Lilie, Die byzantinische Reaktion, pp. 1 0 2 - 8 ; Stratos, Byzantium in the Seventh Century, vol.
V, pp. 19ff.; R.J.H. Jenkins, 'Cyprus between Byzantium and Islam, A.D. 6 8 8 - 9 6 5 ' , in
Studies Presented to DM. Robinson (St Louis 1953), pp. 1 0 0 6 - 1 4 , see 1006ff.; C. Mango,
'Chypre, carrefour du monde byzantin', XVe Congres International d'Etudes Byzantines.
Rapports et co-rapports V, 5 (Athens 1976), pp. 3 - 1 3 , see 4ff.
78
See Theophanes, 3 6 4 and Ostrogorsky, Geschichte, p. 109, notes 2 and 3; Lilie, Die
byzantinische Reaktion, pp. 237ff.; Ditten, 'Zur Bedeutung der Einwanderung der Slawen',
pp. 152f.; Charanis, 'The Transfer of Population', 1 4 3 . See also Ai. Christophilopoulou,
Bv&vTLvrj'Io-TOpCa, vol. II, 6 1 0 - 8 6 7 (Athens 1981), pp. 365f., and esp. H.Kopstein,
'Zum Bedeutungswandel von ax\d(3os/sclavus', BF 7 (1979), 6 7 - 8 8 : the fact that Byzan-
tine seals refer to the Slav settlers as prisoners of war/slaves m a y lie behind the develop-
ment of the equivalent Slav - slave. See chapter 6 below.
79
See Stratos, Byzantium in the Seventh Century, vol. V, pp. 59f.; Lilie, Die byzantinische
Reaktion, p. 105f.; Ditten, 'Zur Bedeutung der Einwanderung der Slawen', p. 156; Ostro-
gorsky, Geschichte, p. 110; Charanis, 'The Transfer of Population', 143-4. The Cypriots in
fact returned to their island shortly after (see Theophanes, 365; Michael Syr., vol. II, 470)
and their settlement, Justinianoupolis (see Mansi XI, 961), was abandoned. Justinian is
also reported to have settled 'Scythians' in east Macedonia in the Strymon region: cf.
Constantine Porphyrogenitus, De Thematibus, 88f.
72 Byzantium in the seventh century

to whether the caliph took it as a serious cause of offence. In 692/3,

however, Justinian decided to break the truce and to attack the Arab forces
in Iraq, newly pacified by cAbd al-Malik, and therefore a weak link in the
Arab defences. cAbd al-Malik had to go on to the offensive, but as a result of
the mass desertion of the recently drafted Slav troops, the Byzantine forces
were soundly defeated. A second Byzantine attack in the following year
was also thrown back. The immediate result was the submission of the
Armenian princes once again to Muslim overlordship and the beginning of
a new series of regular raids into Anatolia, which had barely begun to
recover from the previous forty years of devastation. Justinian's desire for
military glory, and his foolhardiness, did not have by any means a
satisfactory outcome for the empire. 80
According to Theophanes, Justinian had the remaining Slavs and their
families massacred in revenge for the betrayal, although this seems to be a
later and not very reliable accretion to the account. What is known is that
the Arabs settled the deserters in Syria, where they entered Arab service
and proved extremely valuable owing to their knowledge of some of the
localities that formed targets for the yearly Muslim raids. 81 Justinian's
policy of transferring large numbers of Slavs to Asia Minor to repopulate
devastated areas figures large in the historiography of the seventh century.
Its importance in revitalising or in maintaining the economy of consider-
able areas of agriculturally valuable land must certainly have been great.
The policy was in itself not new. Already during the reign of Justinian I
such transfers of population had occurred. Under Maurice, Armenians
were transferred to Thrace. Goths and Vandals had been likewise settled in
Anatolia during the later fourth and fifth centuries; while in the eighth and
ninth centuries the policy was continued by both Constantine V, for
example, and Basil I, among others. 82 Already by the reign of Justinian II
the system of military districts, or themata, under their generals, or strate-
goi, seems to have become established; and there is no reason to doubt that
the resettlement of Slavs in, for example, Bithynia, was designed also to

80
See Lilie, Die byzantinische Reaktion, pp. 109ff.; Stratos, Byzantium in the Seventh Century,
vol. V, pp. 30-9, and see pp. 24-27 on Armenia. For a general survey of the role of
Armenia between Byzantium and Islam in the seventh century see P. Charanis, The
Armenians in the Byzantine empire', BS 22 (1961), 196-240; H. Manandean, 'Les
Invasions arabes en Armenie (notes chronologiques)', B 18 (1946-8), 163-95; and
J.Laurent, VArmenie entre Byzance et Vlslam depuis la conquete arabe jusqu'en 886 (Paris
1919).
81
Theophanes, 366; Lilie, Die byzantinische Reaktion, p. I l l note 22; Ditten, 'Zur Bedeutung
der Einwanderung der Slawen', p. 153.
82
See esp. Ditten, 'Zur Bedeutung der Einwanderung der Slawen,' pp. 155-7, and 'Slawen
im byzantinischen Heer von Justinian I. bis Justinian II.', in H. Kopstein and F. Winkel-
mann, eds., Studien zum 7. Jahrhundert in Byzanz, pp. 7 7 - 9 1 .
 The East Roman world: the politics of survival 73

boost the available manpower for these locally raised forces. I will discuss
this in greater detail in the next chapters.

The Quinisext council of 692

Justinian's religious policy followed on from that of his father with,
however, a very much more pronounced emphasis on the independence
and supremacy of the see of Constantinople, the one question on which
Constantine IV had compromised with Rome. Justinian presented himself
as particularly orthodox, and in placing the motto servus Christi on his
coins, together with a bust of Christ, he emphasised both the source of his
authority and his own subordination to that source.83 In a letter of
February 687 to the pope, the emperor again stressed his function as God's
representative and chosen guardian of orthodoxy;84 and in confirmation
of his role he summoned in 692 the so-called Quinisext or Troullan
council, thus named because it dealt with and confirmed matters treated at
the fifth and sixth ecumenical councils (in 553 and 680/1 respectively)
and was held in the Troullos or domed hall of the imperial palace. In
addition to dealing with matters of dogma and ecclesiastical politics, the
Quinisext dealt also with matters of clerical and lay discipline, problems of
provincial Church and parish administration, and popular practices and
beliefs. It provides in this respect a crucially important insight into provin-
cial life and customs; but its canons also demonstrate the extent to which
Arab raids and the state of almost uninterrupted warfare in the frontier
areas of Anatolia especially had disrupted the patterns of both normal
rural and urban life.85
Perhaps the most significant decisions of the council, however, were
those which dealt with the jurisdiction and traditions of the sees of Rome
and Constantinople and their relationship. The marriage of priests was
permitted, in contrast to the traditions of the Roman Church (canons 3 and
13); the Roman Saturday fast, observed during the Lenten period, was
rejected (canon 55); canon 82 prohibited the representation of Christ as a
lamb - an increasingly popular form in the West - insisting that he be
presented in human form, thereby stressing the greatness of his suffering
and his sacrifice, as God made man, for humankind. But even more
explicitly, canon 36 took up canon 3 of the council of Constantinople of
381 and canon 28 of the council of Chalcedon of 451 in stressing the
83
J.D. Breckenridge, The Numismatic Iconography of Justinian 11 (A.D. 685-95, 705-711)
(New York 1959); cf. also A. Grabar, L'Empereur dans I'art byzantin. Recherches sur Vart
offlciel de Vempire d'Orient (Paris 1936 and London 1971), p. 164.
84 Cf. Mansi, XI. 737f.; Riedinger, 886f.; Dolger, Regesten, no. 254.
85
See, for example, Haldon, 'Some considerations', 9 If. On the Quinisext and its canons, see
Chapter 8, p. 318 and n. 113 below.
74 Byzantium in the seventh century

equality of the sees of Rome and Constantinople in all matters except the
date of their establishment, and having precedence over the sees of Alexan-
dria, Antioch and Jerusalem. 86
The papacy, of course, rejected the canons of the council as not ecumeni-
cal and rejected them in their entirety. Justinian, perhaps thinking of his
grandfather Constans and Pope Martin, ordered the protospatharios Zacha-
rias to Rome, to arrest the pope and return with him to Constantinople. But
the troops of Rome, as well as those of Ravenna, opposed this, and indeed
the imperial officer escaped the Roman mob only through the good offices
of the pope himself.87 Justinian was unable to respond to this opposition
and the humiliation he had suffered; for shortly afterwards, towards the
end of 695, a coup in the city deprived him of his throne. His unpopular
and harsh fiscal policies, put into practice by the general logothete Theodo-
tus and the sacellarius Stephen, had already made him an unpopular ruler.
He seems also to have paid little heed to the will or the authority of the
senate and the leading officers of state, while his policies in general seem to
have won him the hostility of the provinces. In the coup, the Blue deme
(one of the two chief hippodrome supporters' organisations) acclaimed
Leontius, the recently appointed general of the thema of Hellas, as emperor
and, with the assistance of elements of the senate, together with the
connivance of the city and palatine troops, they were able to seize Justinian
and his hated subordinates. Theodotus and Stephen were executed; Justi-
nian had his nose and tongue slit, and was banished to Cherson, where
some forty years earlier his grandfather had sent the hapless Martin. 88

Justinian's successors
The Emperor Leontius ruled for just three years and presided over the final
extinction of Byzantine power in Africa. Little is known of his policies, but
he seems to have been a popular ruler. 89 In 697 he despatched elements of
the fleet - chiefly made up of units of the Kibyrrhaiotai - to retake
86
See L. Brehier, in Fliche and Martin, eds., Histoire de Veglise, V, pp. 194ff.; the canons:
Mansi XI, 9 2 1 - 1 0 0 5 . For a brief s u m m a r y , Winkelmann, Die ostlichen Kirchen,
pp. 1 1 2 - 1 3 . On the purpose of the council, in particular t h e implicit intention of reinfor-
cing Constantinopolitan as opposed to both R o m a n and other practice, see V. Laurent,
'L'CEuvre canonique d u Concile in Trullo ( 6 9 1 - 6 9 2 ) , source primaire du droit de l'eglise
orientale', REB 2 3 (1965), 7 - 4 1 , see 10ff.'
87
See F. Gorres, 'Justinian II. u n d das rbmische Papsttum', BZ 17 (1908), 4 3 2 - 5 4 , see
440ff.; Stratos, Byzantium in the Seventh Century, vol. V, p. 5 3 - 6 ; Guillou, Regionalisme,
pp. 2 0 9 - 1 1 ; Herrin, Formation of Christendom, 2 8 2 - 7 .
88
See Ostrogorsky, Geschichte, pp. 117f.; Stratos, Byzantium in the Seventh Century, vol. V,
pp. 6 6 - 7 4 (although the account is in parts s o m e w h a t eccentric); and Kaegi, Military
Unrest, pp. 186ff.
89
Grierson, DOC, vol. II, p. 6 1 0 ; J.B. Bury, The Later Roman Empire from Arcadius to Irene
(395-800), 2 vols. (London and New York 1889), vol. II, pp. 352f.
 The East Roman world: the politics of survival 75

Plate 2.5 Tiberius III Apsimar. Gold solidus

Plate 2.6 Justinian II (second reign). Gold solidus

Carthage, under the command of the patrikios John. The city and some
neighbouring forts were retaken, but the garrisons placed in them were
soon driven out again in a second Arab attack, and in 698 the city fell a
second time, never to be retaken. While waiting in Crete for reinforce-
ments, and new orders, however, John was deposed by the soldiers of the
Kibyrrhaiot fleet, who proclaimed their own commander, the drouggarios
Apsimar, emperor. The fleet sailed to Constantinople and landed at the port
of Sykai on the Golden Horn. Although the city was ravaged by the plague
at that time, Leontius held out for some months before some of the garrison
units were persuaded to open the gates to soldiers of the besieging forces.
After a short period of plundering and disorder, Apsimar brought his
troops under control; Leontius was mutilated to disqualify him from the
imperial position, and banished to the monastery of Psamathion near the
Xerolophus district of Constantinople. Apsimar altered his name to Tiber-
ius, a move designed to give him some legitimacy and to associate him
with the house of Heraclius.90 It is worth noting that the demes or factions
of the Blues and the Greens seem to have played a role in these political
changes, for Leontius was supported (it would seem) by the Blues, Tiberius
Apsimar by the Greens. But too much has been made of these affiliations
90
On these events see Theophanes, 37O.27sq. and Nicephorus, 39f. Cf. Grierson, DOC, vol.
II, p. 624; Stratos, Byzantium in the Seventh Century, vol. V, pp. 85f.; Kaegi, Military
Unrest, pp. 188f. For Carthage, see Brett, 'The Arab Conquest and the Rise of Islam in N.
Africa', pp. 5O5ff.; Diehl, VAfrique byzantine, pp. 582ff.
76 Byzantium in the seventh century

and involvements and, while it is certain that particular emperors

favoured a particular team and therefore its supporters' club, and that the
hostilities and rivalries that might emerge from such explicit commitments
to the fortunes of a given hippodrome team spilled over into the street and
into the political life of the city and the court, there are few grounds for
thinking that the hippodrome factions represented clear social class divi-
sions, still less religious-doctrinal orientations. Undoubtedly, such alliances
were formed, but these were determined by temporary and fluid conjunc-
tures, rather than by the structure of social relations of production as
such. 91
Tiberius Apsimar reigned from 698 until 705. His reign saw the conti-
nuation of Arab incursions into Asia Minor and a continuous loss of
frontier districts to Muslim forces. No attempt was made to reconquer the
lost North African provinces; in Asia Minor Tiberius' brother Heraclius
was given command of the main thematic cavalry forces and waged a
moderately successful campaign, although he was unable to alter the basic
situation in any fundamental way. Armenia played a significant role
throughout, rebelling in 703 and calling in Arab assistance; but in 704
rebelling again, against the Arabs, calling on Roman support in the
struggle. Tiberius may also have undertaken action in the south Balkans
against various Slav peoples, although little is known of this. 92 He cer-
tainly presided over the resettling in Cyprus of a number of those who had
been removed to the Cyzicus region some years earlier. It was he, too, who
put the sea-walls of Constantinople back into good repair, for they had
apparently been allowed to collapse and decay. 93

JUSTINIAN II AGAIN

In spite of his relatively peaceful internal government and his moderate

successes against the Arabs, Tiberius ruled only seven years. For in 705
the exiled Justinian returned and, with the help of the Bulgar Khan Tervel,
with whom he had reached an agreement, managed to enter the city and
take back his throne. His banishment in Cherson had been cut short when
the authorities there, worried by his plotting, had determined to hand him
back to the emperor. But he was warned of their intentions and fled to the
khagan of the Chazars, who received him with great honour, and whose
daughter he married. Tiberius' envoys demanded that he be handed over,

91
See A. Cameron, Circus Factions: Blues and Greens at Rome and Constantinople (Oxford
1976), pp. 126ff., 297ff., and see 2 6 7 - 8 ; Ostrogorsky, Geschichte, pp. 117f.
92
Lilie, Die byzantinische Reaktion, pp. 112ff.
93
A. Guilland, Etudes de topographie de Constantinople byzantine, 2 vols. (Berlin and Amster-
dam 1969), pp. 263f.
 The East Roman world: the politics of survival 77

and the Chazar khagan, wishing not to endanger the long-standing alli-
ance between the two powers, agreed. But Justinian was once again able to
forestall his enemies, and fled, this time to the Bulgar Khan Tervel.
Accompanied by a considerable Bulgar and Slav army, Justinian arrived in
the autumn of 705 before the land walls of Constantinople. The defences
were too strong for an assault, but Justinian and a few supporters crept in
through one of the ducts of the aqueduct. Tiberius and his supporters,
surprised, fled in panic, and within a few hours Justinian had been able -
with the help of a not inconsiderable faction which favoured his return - to
re-establish himself. It is interesting to recall that the young Leo, later Leo
III, was among Justinian's most enthusiastic supporters. In order to cement
his authority, both Leontius and Tiberius, who had not been able to escape,
were publicly executed.94 Justinian was then able to bring his wife from
Chazaria, together with his young son Tiberius, who was crowned co-
emperor.
In return for his assistance, meanwhile, Tervel received the title of
Caesar and the salutations of the populace of Constantinople. Those who
had opposed Justinian, however, were ruthlessly executed or otherwise
punished, including the patriarch Callinicus, who had crowned Leontius in
695: Justinian had his eyes put out.95 Unfortunately for the empire, these
were only the opening stages of a reign of terror and revenge, in which
Justinian appears to have concentrated most of his resources and attention
in avenging himself on those whom he perceived to be his enemies. But he
was able to restore good relations with the papacy, and in 710/11 Pope
Constantine visited Constantinople, where a compromise arrangement
was reached, although no written accord was drawn up. 96 Justinian has
been generally assumed to be responsible for despatching a punitive
expedition against Ravenna at this time, to avenge the hostile attitude of
the city at the time of his overthrow in 695; in fact, it seems that there had
been some attempt at a coup against Pope Constantine, which involved
Felix, the archbishop of Ravenna and a number of others in both Rome
and Ravenna. Upon hearing that the exarch had met with Constantine at
Naples (where he was en route for Constantinople) late in 710, and had had
those involved in the plot at Rome arrested and punished, the Ravenna
plotters rebelled openly, and were able to kill the exarch upon his return to
the city. Justinian's naval expedition, despatched from the local forces
based in Sicily, was in fact intended to deal with this situation, and it was
94
See Ostrogorsky, Geschichte, p. 119; Stratos, Byzantium in the Seventh Century, vol. V,
pp. 103-29. See the useful account of I. Dujcev, 'Le Triomphe de l'empereur Justinien n en
705', in Bv^dvTiov. 'Aipiepwjxa ordv 'Avdpea N. XrpaTO, 2 vols. (Athens 1986), vol. I,
pp. 8 3 - 9 1 .
95
Nicephorus, 42; Theophanes, 375. See Ostrogorsky, Geschichte, p. 120.
96
See Liber Pontificate, I, 389; cf. Paul the Deacon VI, 3 1 .
78 Byzantium in the seventh century

the misleading and much embroidered version of Agnellus, chronicler of

the Church of Ravenna, which was responsible for the alternative - and, in
the event, barely plausible - account of events. 97
Shortly after this expedition, however, Justinian did send a fleet against
Cherson, the city of his former exile, and had his personal enemies
punished - although the strategic logic behind the expedition was the
reassertion of Byzantine control over the city and its hinterland, which had
come increasingly under Chazar domination. After the departure of the
expedition, however, and for reasons which must remain unclear, the
Chersonites rebelled, and with Chazar help. The expeditionary force
returned on Justinian's order, but was unable to take the city, which with
the aid of its new allies and under the command of the exiled officer
Bardanes - who had been proclaimed emperor - was now well defended.98
There followed a mutiny of the fleet and the army it was transporting,
which also accepted Bardanes as emperor and which sailed back to depose
Justinian. The latter left the city in order to raise the troops of the Opsikion
and Armeniakon districts in his support; but he was deserted by all but his
closest friends, and both he and his son, whom he had left in Constantin-
ople, were killed.99 Bardanes, who now took as his imperial name Philip-
picus, deposed the patriarch Cyrus and appointed in his place John, by
whom he was crowned. 100

THE RISE OF LEO I I I , 717-741

With the death of Justinian II and his son Tiberius, the dynasty of the
Heraclids was finally extinguished, a rather sad conclusion to what has
been seen as the first truly 'Byzantine' dynasty. What followed was further
political confusion and, as the deacon Agathon - a contemporary of the
events in question - described it, a period in which the emperor's authority
counted for nothing, in which the empire had been reduced and humbled,
and tyranny - that is usurpation and violent changes of power - was the
order of the day. 101
In a desperate effort to evoke the glories of the past, to restore imperial
authority and to cement his own position, Philippicus issued an edict soon
97
See Stratos, Byzantium in the Seventh Century, vol. V. pp. 135ff.; C.Head, Justinian II of
Byzantium (Madison 1972) and 'Towards a reinterpretation of the second reign of
Justinian IF, B 4 0 (1970), 14-32; for Agnellus' account, see Liber Pontificalis Ecclesiae
Ravennatis, 368; and cf. Guillou, Regionalisme, pp. 2 1 6 - 1 8 .
98
Bardanes had been exiled in 711 and sent out with the first expedition. See Theophanes,
372ff.; Mansi XII, 192. See also A.A. Vasiliev, The Goths in the Crimea (Cambridge, Mass.
1936), pp.83ff.
99
Kaegi, Military Unrest, pp. 189f.; Stratos, Byzantium in the Seventh Century, vol. V,
pp. 171ff.
101
100 Theophanes, 381. Mansi XII, 192A.
 The East Roman world: the politics of survival 79

Plate 2.7 Philippicus Bardanes. Gold solidus

Plate 2.8 Anastasius II. Gold solidus

after his accession which condemned the acts of the sixth ecumenical
council of 680, which had rejected monotheletism, and officially reintro-
duced the doctrine of the single will. The representation of the sixth council
in the imperial palace, as well as a commemorative inscription on the
Milion gate of the palace, were removed and destroyed. In place of the
inscription, portraits of the emperor and the patriarch Sergius
appeared.102 Philippicus was of Armenian background, and the monophy-
sitism which he may well have found familiar is surely behind this move. It
seems to have been not unpopular among the clergy, including both the
later patriarch Germanus and the theologian and homilist Andrew of
Crete.103 The new policy met with stiff opposition in Rome, of course,
especially in view of the recent entente between Justinian and Pope
Constantine. The latter now returned Philippicus' portrait, which had been
sent to Rome by the new emperor, and rejected his monothelete declaration
of orthodoxy. His name was excluded both from the prayers of the Church
and from the date of documents. And as an added gesture of defiance and

*°2 Mansi, XII, 192D-E; Riedinger, 899.10ft; see A. Grabar, L'lconodasme byzantin: dossier
archeologique (Paris 1967), pp. 48f.
™* Beck, Kirche, pp.474, 500. For Andrew see S.Vailhe, 'Saint Andre de Crete', EO 5
(1902), 378-87, based on the earliest extant life, dating probably to the eighth century.
See Papadopoulos-Kerameus, Analekta V, 169-79; and I.Sevcenko, 'Hagiography of the
Iconoclast Period', 127 and note 105b.
80 Byzantium in the seventh century

self-assertion, the pope had pictures of all six ecumenical councils put up in
St Peter's.104 Whatever the success of Philippicus' new policy among the
clergy, however, he seems not to have commanded much military support.
Throughout the second reign of Justinian, the Balkan provinces had
been at peace, with the exception of occasional raids, a peace due primarily
to Justinian's agreement with Khan Tervel. In Anatolia, in contrast, Arab
raids continued on a yearly basis. In 707/8 the important fortress town of
Tyana fell after a major Byzantine force was defeated. In 711 the frontier
town of Sisium was finally abandoned by its citizens, who could no longer
withstand the constant harassment. At the same time it seems that the
Byzantines were gradually losing their hold in Cilicia, fortress by fortress,
in spite of the occasional counter-attack or the naval expedition against
Damietta in 709. The situation continued to deteriorate under Philippicus.
In 712 Amasia and Misthia fell, along with other forts around Melitene;
and in 713 Antioch in Pisidia was taken. The long-term effects of this
constant raiding was, it has been suggested, effectively to empty the
frontier areas of population, as the local peasantry and townspeople were
either killed, carried off into slavery or driven to seek refuge in areas far
from the conflict zone. 105 In the Balkans, Justinian's defeat and deposition
gave Khan Tervel a pretext for invasion to avenge his friend and ally, and
his forces ravaged Thrace up to the walls of the city itself. When Philippicus
began to organise his troops to oppose these attacks, however, a mutiny
broke out among the Opsikion troops. On 3 June 713 he was deposed and
blinded, and succeeded by the protoasekretis Artemius, a palatine clerical
official, who became emperor with the name Anastasius II.
Anastasius' first act was the restoration of Chalcedonian orthodoxy,
rejection of monotheletism and the rehabilitation of the sixth council. He
was an active emperor who took immediate measures to defend Con-
stantinople against an imminent Arab attack. The walls of the city were
repaired, and a fleet was commissioned, in an effort to attack the Arab
naval forces in their ports and pre-empt the siege. 106 Unfortunately, the
Opsikion division mutinied once more while on Rhodes, where the expedi-
tion was assembling, crossed back to the mainland and, together with the
corps known as the Gothograeci, probably the optimates of an earlier
period,107 acclaimed as emperor an unknown fiscal official named Theo-
dosius. The latter sensibly tried to run away, but was apprehended and
forced to accept the dubious honours bestowed upon him. The provincial

"» Liber Pontificate I, 391.

105
See the list and summary of Lilie, Die byzantinische Reaktion, pp. 116-21, 137ft.
106
Theophanes, 383.1Osq.; Nicephorus, 49.5sq.; MansiXII, 192ff.; Riedinger, 900.12ff. See
Kaegi, Military Unrest, p. 191.
107
Haldon, Byzantine Praetorians, pp. 201f.
 300 miles

1 Exarchate of Ravenna 7 Thema of Thrace

2 Venetia and Istria 8 Thema of Opsikion
3 Duchy of Rome 9 Thema of Thrakesion
4 Duchy of Naples 10 Thema of Anatolikon
5 Duchy of Calabria 11 Thema of Karabisianoi
6 Thema of Hellas 12 Thema of Armeniakon

Man TV The emnire at the accession of Leo III (A.D. 717)

82 Byzantium in the seventh century

Plate 2.9 Theodosius III. Gold solidus

troops then marched to Constantinople which, after some months of

desultory siege warfare, capitulated. Anastasius had meanwhile fled to
Nicaea, but he soon abdicated and retired as a monk to Thessaloniki.108
His successor Theodosius III occupied the throne for just over twelve
months. Almost immediately after news had reached the provinces of his
accession, the strategos of the Anatolikon region, Leo, together with the
strategos of the Armeniakon, Artavasdus, rebelled against the new emperor.
The alliance was strengthened by Artavasdus' marriage to Leo's daughter;
and the conflict took the form of a struggle for precedence between the two
great Anatolian themata of Anatolikon and Armeniakon, on the one hand,
and that of the Opsikion on the other. Theodosius' son and retinue were
captured in Nicomedia in Leo's rapid advance towards the capital; and,
having secured a promise of personal safety for himself and his family,
Theodosius abdicated and took up the monastic life at Ephesus. 109 His
most lasting achievement was probably the treaty he agreed with the
Bulgars in 716, which fixed a formal frontier between the two powers
along a line through northern Thrace, from the Gulf of Burgas across to the
Maritsa between Philippoupolis and Adrianople, a frontier which illus-
trates the extent and power of the khanate of the Bulgars by this time. 110
In March 717 Leo was crowned emperor in the Hagia Sophia by the
patriarch Germanus. Shortly after the second great siege of the capital of
the once mighty East Roman state began. 111
Leo III won his throne largely because he was able to present himself as
the most able general who could deal effectively with the Arab threat to the
empire and especially the capital. For the invasion had already begun
when Leo was finally admitted to the city. His rise to power had started

108
Theophanes, 385.18sq.; Nicephorus, 51.2sq.; Kaegi, Military Unrest, pp. 192f.; G.V.
Sumner, 'Philippicus, Anastasius II and Theodosius III', GRBS 17 (1976), 291ff.
109
See Kaegi, Military Unrest, pp. 192-4.
110
See Obolensky, The Byzantine Commonwealth, pp. 65f.
111
See R. Guilland, 'L'Expedition de Maslama contre Constantinople (717-718)', in Etudes
byzantines (Paris 1959), pp. 109-33, see 114.
 The East Roman world: the politics of survival 83

when, as a young soldier, he had been commissioned as a spatharios

during the second reign of Justinian II. 112 Appointed to the post of
strategos of the Anatolikon by Anastasius II, it was ostensibly in support of
Anastasius that he rose in rebellion against Theodosius III. 113 During his
campaign against Theodosius and his negotiations with the capital, he
had been able to outwit the leaders of the invading Arab forces: he was
able to persuade the Arab commander Suleyman, who invaded from
Cilicia and marched towards Amorion in 716, that Amorion would be
handed over to the invaders. Instead, he succeeded in getting a garrison
into the defenceless fortress town, upon which the invaders had to fall
back again. An Arab fleet was meanwhile ravaging parts of western Asia
Minor; while the main force, under the brother of the caliph (also
Suleyman), Maslama, waited for word of the commander Suleyman. As
the latter began to withdraw, however, Maslama marched into Roman
territory through Cappadocia and, upon discovering Leo's deception,
attacked and took Sardis and Pergamum, wintering in the latter. The fleet
wintered in Cilicia.
In 717 Maslama ordered his fleet up, while he and his forces crossed the
Dardanelles from Abydus, laying waste in the following weeks all of
Thrace, and closely investing the city. The fleet blockaded Constantinople
from the sea. But the capital was well defended and adequately pro-
visioned, chiefly thanks to the measures taken by Anastasius II. With the
help of its terror weapon, Greek fire, and a well-thought-out strategy of
counter-offensives, Leo's forces - helped by the attacks of the Bulgars, the
onset of winter, the cutting-off of Arab supply sources, the outbreak of
disease in the Arab camp and a timely victory over the Arab forces in
Bithynia - were able to defeat the blockade and hold Maslama's forces in
check. In August 718 the new Caliph cUmar II (Suleyman died in 717)
ordered Maslama's withdrawal. The army in Thrace was embarked on his
ships, but Byzantine naval attacks and bad weather resulted in the loss of
many vessels on the return voyage. Arab sources claim 150,000 Muslim
deaths during the campaign, a figure which, while certainly inflated, is
neverthless indicative of the enormity of the disaster in medieval eyes. This
was to be the last Arab attempt to take Constantinople itself, and it is
possible to follow a change in Arab strategy after this time. The caliphate
had finally come to terms with the existence of the Byzantine empire and,
while long-term Byzantine counter-attacks were still a thing of the future,

112
For the contemporary importance of this corps, see Haldon, Byzantine Praetorians,
pp. 182ff.
113
See Theophanes, 386-97 for Leo's early career; and for his 'Caucasian adventure' under
Justinian II, see B. Martin-Hisard, 'La domination byzantine sur le littoral oriental du Pont
Euxin (milieu du VIIe-VIIIe siecles)', Byzantinobulgarica 7 (1981) 141-54.
84 Byzantium in the seventh century

the defeat of this expedition marks an important moment in Byzantine-

Arab relations. 114

LEO III AND IMPERIAL ICONOCLASM

In spite of this major set-back, Arab attacks all along the border and well
into the very deep 'frontier' zone continued throughout Leo's reign. From
720 until his death in 741 there were yearly attacks, in which the border
areas were repeatedly devastated, fortresses and settlements taken, retaken
and lost again, with both sides suffering regular losses. Only in 740 were
the Byzantines able to lure Arab forces on to unfavourable terrain and win
a major victory near Acroinum. 115 Only one of three Arab columns was
actually destroyed, however, and the victory was perhaps more significant
as a shock to Muslim morale and as a heartening triumph for the belea-
guered Byzantines. Arab raids went on until 744, but the civil wars which
led to the overthrow of the Umayyads and the installation of Abbasid rule,
along with the transfer of the centre of power to Baghdad, soon interrupted
the regularity and reduced the devastating nature of the attacks; in
addition to which Leo's son and successor, Constantine V, went over to the
offensive in the 750s, and the nature of the warfare between the two sides
began, slowly but surely, to change both in character and in its effects.116
In the north, meanwhile, Leo cemented the friendly relations enjoyed by
the empire with the Chazars through the marriage of his son Constantine
to a daughter of the Chazar khagan in 733; 1 1 7 while in the Balkans, as a
result of the arrangement of 716, peace reigned. Two rebellions at the
beginning of his reign were quickly crushed. In 717/18 Sergius, the
strategos of Sicily, rebelled before news of the successful outcome of the
siege of Constantinople had reached him and set up a rival emperor by the
name of Tiberius. Upon receiving news of Leo's victory, along with the
imperial letters proclaiming his accession, the soldiers handed the rebels
over to the emperor's envoy, the chartoularios Paul and his troops. 118 In
719 a second coup was foiled. The magistros Nicetas Xylinites, together
with the exiled ex-emperor Anastasius and with assistance from the Bulgar
Khan Tervel, attempted to dethrone Leo. The plotters inside Constantinople

114
For accounts of the siege, see Theophanes, 395-8, 399; Nicephorus, 53--i; and the
discussion of Lilie, Die byzantinische Reaktion, pp. 125-33; E.W. Brooks, 'The campaign of
716-718 from Arabic Sources', JHS 19 (1899), 19-31.
115
See Theophanes, 411; Lilie, Die byzantinische Reaktion, pp. 152-3 with literature and
sources.
116
The best survey of these wars, their nature and their effects is to be found in Lilie, Die
byzantinische Reaktion, pp. 144-55, with discussion at pp. 155-62. For the period from
750, see ibid., pp. 162ff.
117 118
Vasiliev, The Goths in the Crimea, p. 87. Theophanes, 398-9.
 The East Roman world: the politics of survival 85

Plate 2.10 Leo III. Gold solidus

were unable to rally the support of the troops and officers manning the
walls; while Tervel's soldiers seem to have abandoned Anastasius at the
last minute, handing him over to Leo. All the plotters were executed.119
Leo's quick and firm handling of these revolts enabled him to secure his
position and at the same time dissuade others from repeating the mistakes
of Sergius and Anastasius.
Leo's reign is probably best known for two great events: the introduction
of an official imperial policy of iconoclasm and the issue of the Ecloga. The
latter has been traditionally dated to 726; in fact, it appears that it dates to
the end of his reign, to the year 741. 120 It represents in essence a revised
and very much abridged version of the Justinianic codification, with
particular emphasis on family, property and inheritance, and penal law. Its
practical intention was to provide the administrators with a handbook and
reference work, although it hardly replaced the older codifications. But
substantial changes were also introduced, changes which reflect in par-
ticular the powerful influence of canon law, especially with regard to
marriage and the power of the husband over wife and children; there were
also changes in the system and nature of punishment: while capital
punishment and fines dominate the Justinianic codes, the Ecloga, on the
pattern of the Old Testament, introduces a system of corporal mutilations -
slitting of the nose, tongue and so on - which had become increasingly
common in the course of Byzantine history during the seventh century, but
which were quite foreign to Roman legal tradition. Such changes may well
be based on the formal incorporation into the legal framework of the state
of elements of local customary law.121
119
See Theophanes, 400.18sq.; Nicephorus, 55.19sq. and see Kaegi, Military Unrest,
pp.212f.
120
For the older view see V. Grumel, 'La Date de l'Eclogue des Isauriens: l'annee et le jour',
REB 21 (1963), 272-4; D.Simon, 'Zur Ehegesetzgebung der Isaurier', FM I (1976),
40-3; P.E. Pieler, Byzantinische Rechtsliteratur, p. 438 and notes 97ff. For the year 741
see now Ecloga (ed. Burgmann), pp. 10-12 with literature.
121
Ecloga (ed. Burgmann), pp. 4-7; T.E. Gregory, 'The Ekloga of Leo in and the concept of
Philanthropia', Byzantina 7 (1975), 271-5.
86 Byzantium in the seventh century

But in spite of its later partial rejection or condemnation by anti-

iconoclastic polemicists, the Ecloga exercised an important influence on
the development of Byzantine and Eastern medieval legal theory, and
marks an important stage in the history of Byzantine law. 122 It marks
equally the determination of Leo himself and his son Constantine, in whose
joint names it was promulgated, to come to grips with the problems of
Byzantine society at an official level and to address the changes which had
taken place in both social and property relations, as well as in ideas and
morality since the sixth century. The prologue to the work therefore
stresses both the need to preserve the law as the basis of God's will on
earth, of human society, and of imperial competence and success, as well
as the pressing need to make it clearer and more easily understood and
accessible to the judicial authorities. Corruption and bribery were to be
combated, hence the legal representatives of the state's justice are to be
properly salaried. And interestingly, the emperors note that the law has
been frequently difficult of access and comprehension among lawyers and
judges far from Constantinople - a reflection of the difficult conditions, bad
communications, and disruption of urban life during the previous one
hundred or so years. 123
The introduction by Leo of an official policy of iconoclasm has given rise
to a debate among modern historians, as well as among Leo's and Con-
stantine's contemporaries, of a very different nature. The background to the
whole debate lies already in the later sixth and seventh centuries, as the cult
of relics and the associated use of holy images became more and more
prominent, and as both icons and relics were brought more and more cen-
trally into the arena as the focus of popular devotion, both publicly and
privately, as channels of direct access to the heavenly realm. At the same
time, both as a direct result of the collapse of the empire and the massive ter-
ritorial losses of the period up to the 660s - in which the 'failure' of the
emperors and the state were clearly implicated, the more so in view of the
'heretical' policies of the monothelete Emperor Constans and the patriarchs
associated with him and with Heraclius - imperial authority itself came
under pressure as political failure followed political failure. The development
in the 'cult' of icons worked together with the events of the political and eco-
nomic world to encourage a serious questioning of the authority and abili-
ties of individual emperors. Thus the armies of the provinces were encour-
aged to take up arms in favour of one or another candidate for the throne
who it was felt would best answer their needs and those of the state. The
attempt of Philippicus Bardanes - with the blessing of certain leading
Churchmen - to reintroduce monotheletism is an interesting and important
illustration of the perceived need both to reassert imperial and official
122
See the comments of Ostrogorsky, Geschichte, p. 132 with literature.
123
See Ecloga (ed. Burgmann), 'Prooimion', 11.21sq.
 The East Roman world: the politics of survival 87
Church authority, and at the same time unify the disparate religious-ideo-
logical elements within the empire. Monotheletism harked back to the expla-
nations of an earlier epoch. Iconoclasm identified a different source of the
problem.124
Iconoclastic, or anti-image, sentiment was, of course, nothing new in
Christian thought. The debate on the nature of icons had long been part of
Christian doctrinal discussion and argument, although no developed
theory of icons yet existed. Iconoclastic groups may have existed in seventh-
century Anatolia; while the crucial question as to whether icons should be
seen as graven images or as representations through which, rather than to
which, prayer could be directed, had never been formally confronted by the
Church.125
During the 720s, however, an iconoclastic perspective seems to have been
espoused by some provincial churchmen, the best known of whom -
Constantine, the bishop of Nacolea - was later regarded as the founder of
Byzantine iconoclasm and appropriately reviled by the iconophile camp.
Leo III was himself originally from Syria; but whatever contemporary or
later iconophile commentators claimed, the extent to which Leo was
influenced by either Islamic or monophysite thinking (the caliphs cUmar II
(717-20) and Yazid II (720-4) had both issued orders condemning the use
and production of images representing living beings as idolatrous)126 or by
Jewish theology127 must remain unclear. What is probable is that Leo
seems genuinely to have regarded the iconoclastic case as a strong one and
believed that it was 'idolatry' and the sins of the Chosen People - the
Romans - which had brought about their downfall and, as an integral
causal element in that downfall, the decline in, and contempt for imperial
124
See Haldon, 'Some remarks', esp. 1 7 6 - 8 4 and 'Ideology and social change', 161ff.,
where these developments are discussed in greater detail. See also P. Brown, 'A Dark-Age
crisis: aspects of the iconoclastic controversy', EHR 8 8 (1973), 1 - 3 4 ; E.Kitzinger, 'The
cult of images in the age before iconoclasm', DOP 8 (1954), 8 5 - 1 5 0 . For a recent survey
of the origins and cause of official iconoclasm, see Herrin, Formation of Christendom,
pp. 319ff., 3 2 5 - 4 3 . See esp. M.-F. Auzepy, L'evolution de l'attitude face a u miracle a
Byzance (VH e -IX e siecle)', in Miracles, prodiges et merveilles au Moyen Age (Paris 1 9 9 5 ) ,
31-46, and J.F. Haldon, 'The Miracles of Artemios and Contemporary Attitudes: Context
and Significance', in The Miracles of Saint Artemios: Translation, Commentary and Analysis, by
J. Nesbit, V. Crysafulli (Leiden-NewYork-Koln 1 9 9 7 ) . 3 3 - 7 3 .
125
See Kitzinger, 'The cult of images before Iconoclasm', 129flf.; S.Der Nersessian, 'Une
apologie des images du septieme siecle', B 17 ( 1 9 4 4 / 5 ) , 58ff.; N.H. Baynes, 'The icons
before iconoclasm', Harvard Theol. Review 4 4 (1951), 93ff.; Haldon, 'Some remarks',
182, note 6 4 .
126
See.esp. Grabar, L'lconoclasme, pp. 103f.; D.Stein, Der Beginn des byzantinischen Bil-
derstreites und seine Entwicklung bis in die 40er Jahre des 8. Jahrhunderts (Miscellanea
Byzantina Monacensia XXV, Munich 1980), pp. 1 3 9 - 4 1 with literature.
127
See J. Starr, The Jews in the Byzantine Empire, 641-1204 (Athens 1939), esp. on Leo's
persecutions, and A. Sharf, Byzantine Jewry, from Justinian to the fourth Crusade (New York
1971); cf. Theophanes, 401.22. For monophysitism, which seems to have played little or
no role in the affair, see S. Brock, 'Iconoclasm and the monophysites', in A.A.M. Bryer
and J. Herrin, eds., Iconoclasm: Papers given at the ninth Spring Symposium of Byzantine
Studies, University of Birmingham, March 1975 (Birmingham 1977), pp. 53-7.
88 Byzantium in the seventh century

authority which seemed so clear to contemporaries.128 The two were

bound up closely together.
In 726, a dramatic earthquake and volcanic eruption occurred on the
islands of Thera and Therasia, an event which Leo seems to have taken as
a sign of God's anger and as a direct signal for him to take action. As an
immediate result, he appears to have had an icon of Christ on the Chalke
gate of the palace removed and replaced with a cross, a reflection of his
belief that the holy cross and the wood of the cross represented symbols of
salvation for emperor and people, whereas icons, being made by human
hands, were transient symbols of human frailty. But Leo's action was also
a reflection of his personal response to the situation, and it does not begin
any official persecution of icons at this time.
At about this time, the bishop Constantine of Nacolea had been preach-
ing that icons should not be the objects of proskynesis or worship, since this
was idolatrous. As a result of this activity, he had been disciplined by his
superiors and the patriarch Germanus. Leo may have been aware of these
debates and seems eventually to have decided that the iconoclastic viewpoint
- given the historical background to his own reign and those of his prede-
cessors, together with the eruption on the islands - had greater validity. The
public debate, therefore, began only in 72 6. But the context for this debate, and
the reasons for Leo's eventually adopting an iconoclastic position, had a
longer history. After some four years of discussion, in which a split within
the clergy between pro- and anti-icon parties developed, Leo made a first
public move, putting Germanus in 730 in a position where he was forced
either to accept an iconoclastic declaration of faith or to resign. He chose
the latter and was replaced as patriarch by Anastasius, his former syg-
kellos. Contrary to generally received opinion, however, there seems to
have been no official edict issued prohibiting the use of icons. Instead,
Anastasius issued an iconoclastic declaration of faith in which the cross
was presented as the true symbol of Christ's passion and the salvation of
humanity, whereas icons were to be cast out. As these developments took
place, a rebellion of the Helladic troops under their strategos, who set up a
pretender to the throne, had occurred; but it was decisively defeated in a naval
engagement by the imperial fleet near Constantinople in April 727. Again,
contrary to the often accepted view that the rebellion was an iconophile
response to Leo's policies, this was not the case. 129 In the first place, Leo
had as yet made no official pronouncement; in the second place, the
rebellion seems to fit well into the general pattern of coups and revolts
which had plagued the empire. It was the third in Leo's reign itself, and
although it failed, it illustrates the continuing state of internal insecurity

12
« See Nicephorus, 52.2-6; Mansi XII, 192A.
129
See, for example, Ostrogorsky, Geschichte, p. 135.
 The East Roman world: the politics of survival 89

and ideological uncertainty, and the lack of confidence in and respect for
the imperial authority.
It is equally clear that the conflict of interest between the empire and
Rome over the increased rates of taxation for Italy and Sicily had nothing
to do with the iconoclastic policies of Leo from the 730s. The refusal of
Pope Gregory II to pay the taxes was in 723/4; the rebellion of the army of
the exarchate and the cities of Italy shortly after the exarch Paul's failure to
arrest the pope as ordered by Leo took place in 725/6; as did the unsuc-
cessful naval expedition sent by the emperor. Order was restored only with
the arrival of the new exarch Eutychius in 727/8, who formed an alliance
with the Lombard King Liutprand. Indeed, Gregory's general support for
the empire is clear from his opposition to the usurper Petasius, who re-
belled but was defeated by Eutychius shortly after. Iconoclasm played no
part in these events. 130
There is no reason to doubt that Leo's eventual espousal of iconoclasm
represents a genuine religious and theological commitment on his part,
and that of a majority faction within the clergy. But these personal actions
and beliefs must also be set firmly within the context of the political and
ideological developments of the time. Iconoclasm seemed to provide
answers to a number of questions of direct concern to those who perceived
the dangers of a world in which both official Church and imperial auth-
ority had been challenged, at a variety of levels, by the events of the
previous century. While the use of icons, the authority vested in them, and
the consequent tension between two different sources of authority, seemed
to lie at the root of the problem, they were in fact but symptoms of a much
more complex set of developments which had come to a head in the
debates of Leo's reign. 131
13()
See 0. Bertolini, 'Quale fu il vero oggiettivo assegnato da Leone III "Isaurico" all' armata
di Manes, stratego dei Cibyrreoti?', BF 2 (1967), 15-49; Guillou, Regionalisme, p.219f.;
Brown, Gentlemen and Officers, pp. 156, note 24 and p. 180. Cf. Liber Pontiflcalis. i, 403;
Theophanes, 404.
131
The debate on the origins and background to iconoclasm is, as may be imagined,
immense. For its immediate context and development under Leo, however, the excellent
analysis of Stein, Bilderstreit, offers a detailed survey of much of the earlier literature, as
well as a demonstration of the argument outlined here, and upon which I have relied. See
esp. pp. 138ff of Bilderstreit and also H.-G. Beck, 'Die griechische Kirche im Zeitalter des
Ikonoklasmus', in Handbuch der Kirchengeschichte, vol. Ill, part 1 (Freiburg 1966),
pp. 31-66; St. Gero, Byzantine Iconoclasm during the Reign of Leo HI (Louvain 1973) and
Byzantine Iconoclasm during the Reign of Constantine V (Louvain 1977); the collected
articles in Bryer and Herrin, eds., Iconoclasnv, Kitzinger, 'The cult of images before
iconoclasm'; P.Schreiner, 'Legende und Wirklichkeit in der Darstellung des byzantini-
schen Bilderstreites', Saeculum 27 (1976), 165-79; idem, 'Der byzantinische Bilderstreit:
kritische Analyse der zeitgenossischen Meinungen und das Urteil der Nachwelt bis heute',
in Bisanzio, Roma el'Italia nell'Alto Medioevo (Settimane di Studio del Centro Italiano di Studi
sull'Alto Medioevo XXXIV [Spoleto 1988]) 319-427; P. Henry, 'What was the iconoclastic
controversy about?', Church History 45 (1976), 16-31. These all contain surveys of the
literature and fuller bibliographies than can be accommodated here.
90 Byzantium in the seventh century

The results of Anastasius' declarations of faith were soon to be made

clear. In 731 Gregory HI summoned a synod which met at Rome in November.
A general condemnation of the new tendency - there is not enough evidence
to support the notion that it was an official 'policy' - coupled with the threat of
excommunication for those who promoted it, was pronounced, the arguments
being founded upon a collection of Scriptural and patristic texts established
by Gregory. There is no evidence of any condemnation of either the patri-
arch Anastasios or the emperors Leo and Constantine. The conflict of interest
between the two patriarchates of Rome and Constantinople and between
pope and emperor was becoming clearer. And although there is no evidence
of an official condemnation of Constantinople from Jerusalem, the latter
may have produced a condemnation of iconoclastic ideas and of the patriarch
Anastasios, but not the emperor Leo himself. The effects were an increasing
ecclesiastical political isolation of Constantinople. As the gulf between
the Churches widened, so Byzantine political influence and authority
gradually diminished. The transfer of the papal patrimonies of Calabria,
Sicily, east Illyricum and, probably, Crete, from Roman to Constantinopoli-
tan jurisdiction may have occurred shortly thereafter, although a later date,
in the reign of Constantine V, is also possible.132 What is clear is that the
move represents an attempt to secure ecclesiastical and political jurisdiction
over regions which might otherwise become permanently alienated from the

132
The council of 769 in Rome (see Mansi xii, 713-22) used the acts of 731 (which otherwise
do not survive) for the establishment of its own arguments: see the brief account in Herrin,
Formation of Christendom, p. 348 and n. 15. See LP416.5-15. Stein, Bilderstreit, p. 217
n. 98, discusses the other surviving fragments of this synod and shows^at only the most
general condemnation was issued. For the lack of evidence for a condemnation by the patri-
arch John of Jerusalem and the eastern bishops (Theoph., 408.29-31), and the possible
confusion which may have led Theophanes to incorporate this story, see Stein, Bilderstreit,
21 Iff. and M.-F. Auzepy, 'L'Adversus Constantinum Caballinum et Jean de Jerusalem', BS
56 (1995) 323-38 (XTE&AN02. Studia byzantina ac slavica Vladimiro Vavrinek ad annum
sexagesimum quintum dedicata). In respect of the transfer of Calabria, Illyricum, etc., there
are no references to this in the contemporary record; letters of Hadrian I (772-95) and
Nicholas I (858-67) refer to the events; the first reference, in a letter of Pope Hadrian I to
Charles the Great, written between 787 and 794, offers no specific chronology for the
transfer (see MGH Ep. V. Epist. Karol. Aevi iii, 57.5ff.), and L.M. Hartmann, Geschichte
Italiens im Mittelalter, E, 2, pp. 112-14; M.V. Anastos, 'The Transfer of Illyricum, Calabria
and Sicily to the Jurisdiction of the Patriarchate of Constantinople', SBN 9 (1957) 14r-31.
But for a different interpretation, see D.H. Miller, 'Byzantine-Papal relations during the
Pontificate of Paul I: Confirmation and Completion of the Roman Revolution of the Eighth
Century', BZ 68 (19 75) 47-62; and Darrouzes, Notitiae, p. 249 for the statement that these
sees were transferred 'because the Pope was in the hands of the Heathen' (i.e. the Franks).
While this statement also reflected contemporary political concerns in respect of
Byzantine-Frankish relations (see Schreiner, 'Bilderstreit', 376 n. 291) the fact that no
contemporary Byzantine or western source refers to this major event for the 730s is
significant. See V. Grumel, 'L'annexion de l'lllyricum oriental, de la Sicile et de la Calabre au
patriarcat de Constantinople', Recherches des Sciences Religieuses 40 (1951-52) 191-200
(who places the event in the years 752-757).
 The East Roman world: the politics of survival 91
Constantinopolitan Church and the imperial government. On ideological
grounds alone, this would not be permitted.
Leo III died on 18 June 741. 133 His son and successor Constantine soon
found himself fighting a desperate struggle for survival with his brother-in-
law Artavasdus,134 and although the eventual outcome was a complete
victory for Constantine, the civil war demonstrated still that the difficulties
faced by the emperors in Constantinople both ideologically and in terms of
the practical distribution of power to the provincial strategoi were by no
means resolved by Leo's policies, his military successes, or his reaffir-
mation and confirmation of his personal imperial authority. Leo's greatest
achievement lies in the very fact of his success in defeating his rivals and in
maintaining his authority over some twenty-three years, during which the
external and internal problems he had to deal with were hardly less
pressing than those which faced his predecessors. But it should be recalled
in the end that Leo was, after all, a successful usurper. His reign marks
only the tentative beginnings of a reconstruction of imperial authority,
and, through the political struggle which the iconoclastic debate engen-
dered, of a relocation of the imperial position within the wider framework
of the official, imperial ideological system of the Byzantine state.
Leo's reign is significant in one other respect, however. This chapter
began with an account of late Roman political history in the early seventh
century; it ends with an account of the history of the Byzantine state in the
early eighth century: Byzantine, because our map shows a very different
state territorially; because shifts in the ideological world, in the 'symbolic
universe' of the people and culture of this state, had taken place, which lent
to it a vastly different character from that of late Roman culture; and
because, whatever the continuities one may detect, the functional
apparatuses of this state were already very different from those of the later
sixth century. Precise dates at which a social formation or a state can be
said to deserve one descriptive title as opposed to any other are always
liable to criticism. It seems reasonable to conclude this chapter by suggest-
ing simply that it was during the seventh century that the transformation
from late Roman to Byzantine society, culture and institutions took place.
The evidence for this assertion is presented in the following chapters.135

»« Theophanes,413.
134
On the background, development and results of the civil war, see P. Speck, Artabasdos, der
rechtglaubige Vorkampfer der gotthchen Lehren (Poikila Byzantina n, Bonn 1981).
13 5
For recent literature and discussion of the question of the Slav occupation of the southern
Balkan region, see also}. Karayannopoulos, Les Slaves en Macedoine. Lapretendue interrup-
tion des communications entre Constantinople et Thessalonique du 7 erne au 9eme siecle (Centre
d'Etudes du Sud-Est Europeen, no. 25 (Athens 1989)); and Ph. Malingoudis, Oi Slaboi sti
Mesaioniki Ellada (Athens 1988).
 CHAPTER 3

Social relations and the economy: the cities

and the land

The study of the relations between dependent and independent peasant

communities on the one hand, and their landlords - ecclesiastical or
secular, private or state - on the other, has long been the source of much
thought-provoking debate among historians of the early Byzantine period.
The discussion and the variety of often contradictory answers given by
different historians to the questions which the period poses illustrates the
main problems: there are simply not enough written sources for the period
to enable a clear and detailed picture of social relationships in the
provinces (for example) to emerge. We cannot expect the sort of infor-
mation available for the sixth century, therefore, and the sorts of results it
is possible to extract from such material are not forthcoming for the
seventh century. But the problems are not insoluble. It is possible to build
up the basic outlines of the developments of the seventh century from
archaeological, epigraphic, sigillographic and numismatic sources, among
others, which can complement the available literary documentation. By
employing these materials within a coherent framework or set of models, it
is possible to obtain a shadowy picture of seventh-century social and
economic relations, within which the available evidence can play a role
without contradictions and upon which an understanding of develop-
ments in the subsequent period can be reached.
The essential elements in any analysis of seventh-century society are
represented by the relationships between the state and the sources of its
revenue, and by that between the rural population and their means of
production - land, livestock and tools - and those who owned or con-
trolled those means of production. It will be useful to take these
elements separately, to begin with, and in the following chapter I will
look first at the role played by cities in the late Roman and early Byzan-
tine world, both as centres of administration, as well as of economic
activity.

92
 The cities and the land 93

THE BACKGROUND
In the last thirty or so years, debate on the nature and function of the city
in the seventh century and later has been extensive. In essence, two
opposing points of view can be discerned. The first, which represents a
reaction against views held until the 1950s, argues that the seventh
century saw a more or less total collapse of antique urban organisation,
and of social and economic life. The cities of the period up to the sixth
century disappeared as a result of the Arab onslaught and the rapid
ruralisation of the empire; the lack of small denomination bronze and
copper currency from the archaeological record in the second half of the
seventh century confirms the disappearance of market exchange and
illustrates therefore the end of the market function of the towns in the
provinces.1
The opposing view argues that cities did survive physically; that, while
they may have shrunk and often have been confined to their citadels as a
result of constant enemy harassment, they nevertheless retained their role
as centres of market- and exchange-activity, petty commodity production
and administration. The evidence cited in support of this line of reasoning
includes both the clear continuation of occupation on many sites, attested
archaeologically and in literary sources - as well as the occurrence of gold
coins on many sites, regardless of size or importance. 2
Between these two poles, a number of alternatives or modifications have
been suggested. In a series of articles which seemed to strengthen the case
for a total eclipse of urban life after the first decade of the seventh century,
Clive Foss argued that the archaeological evidence was not only fuller, but
also more reliable and explicit, than the relatively sparse literary material;
and that it was the disastrous results of the Persian invasions and partial
occupation of Anatolia from 615 on, which sealed the fate of the cities of
Asia Minor.3 There have been a number of criticisms of this position; in
1
For a basic statement, see A. Kazdan, 'Vizantiiskie goroda v VII-IX vv.\ Sovietskaya
Arkheologiya 21 (1954), 164-88; and for further literature, Haldon, 'Some considerations',
78f.
2
See G. Ostrogorsky, 'Byzantine cities in the early Middle Ages', DO? 13 (1959), 47-66; Sp.
Vryonis, jr., The Decline of Medieval Hellenism in Asia Minor and the Process of Islamization
from the Eleventh through the Fifteenth Century (Berkeley, Los Angeles and London 1971),
p. 7; R.S. Lopez, 'The role of trade in the economic re-adjustment of Byzantium in the
seventh century', DO? 13 (1959), 69-85; see Haldon, 'Some considerations', 78 and 82
with literature.
3
Clive Foss, 'The Persians in Asia Minor and the end of Antiquity', EHR 90 (1975), 7 2 1 ^ 3 ;
'The fall of Sardis in 616 and the value of evidence', JOB 24 (1975), 11-22; Byzantine and
Turkish Sardis (Cambridge, Mass, and London 1976), pp. 275ff.; 'Late antique and Byzan-
tine Ankara', DO? 31 (1977), 29-87; 'Archaeology and the "twenty cities" of Byzantine
Asia', Amer. Journal of Archaeology 81 (1977), 469-86; and Ephesus after Antiquity: A Late
Antique, Byzantine and Turkish City (Cambridge 1979).
94 Byzantium in the seventh century

particular, that it concentrates on cities which may not represent the

typical case and that it fails to situate the cities in question in the context of
their social and historical environment. 4 Most importantly, both for the
case presented in the work of Foss, and in the arguments outlined above,
the long-term development of the city and its changing functional rele-
vance within the structure of late Roman social and economic relations
has not generally been adequately taken into account. The crucial point is,
quite simply, that the Byzantine 'city' was different from its classical
antecedent because it no longer fulfilled the same role, either in the social
formation as a whole or in the administrative apparatus of the state. This
difference would have persisted whether or not there had been any Persian
invasion or Arab attacks; and it is this underlying structural development
which must be understood if we are to grasp the nature of the changes
from the fifth and sixth centuries which affected urban and rural life. Once
this has been understood, the question of whether or not urban sites
continued to be occupied, and the exact causes of their immediate destruc-
tion or shrinkage, become less crucial; and it can be seen from this
perspective, therefore, that much of the debate - that part which argues for
or against the continuity of urban life purely or predominantly on the basis
of the archaeological evidence for destruction caused by hostile military
action - is in fact quite misguided. The question becomes one of recog-
nising the underlying structural tendencies in the evolution of cities in the
context of the late Roman state and its society, and determining the extent
to which the political events referred to in the literary sources and evi-
denced in the archaeological record intensified or otherwise affected that
process. That it was already under way seems to me undeniable.5
4
See in particular the review by W. Brandes, 'Ephesos in byzantinischer Zeit', Klio 64
(1982), 611-22, with a detailed discussion and extensive bibliography; and the comments
of F. Trombley, 'The decline of the seventh-century town: the exception of Euchaita', in
Byzantine Studies in Honor of Milton V. Anastos, ed. Sp. Vryonis, jr. (Malibu 1985),
pp. 65-90, esp. 75ff. For further literature arguing for discontinuity, see Kirsten, 'Die
byzantinische Stadt', 1-35; C. Bouras, 'City and village; urban design and architecture', in
Akten des XVI. Internationalen Byzantinisten-Kongresses, 1 (Vienna 1981), pp. 611-53
( = JOB 31); D. Zakythinos, 'La Grande breche', in Xapi(nr\piov els A.K. Vpkdvdov (3
vols., Athens 1966), vol. Ill, pp. 300-27. For a recent comment on the debate, see J.C.
Russell, 'Transformations in early Byzantine urban life: the contribution and limitations of
archaeological evidence', in 17th International Byzantine Congress, Major Papers (New York
1986), pp. 137-54.
5
See the comments of A.P. Kazdan, in BvCavriva 9 (1977), 478-84; and in particular J.F.
Haldon and H. Kennedy, 'The Arab-Byzantine frontier in the eighth and ninth centuries:
military organisation and society in the borderlands', ZRVI19 (1980), 79-116, esp. 87ff.;
Haldon, 'Some considerations', 77-89 with literature; also E. Frances, 'La Ville byzantine
et la monnaie aux VIP-Vffle siecles', Byzantinobulgarica 2 (1966), 3-14 for a summary of
the debate and full bibliography. For a detailed analysis, see G.L. Kurbatov, Osnovnye
problemy vnutrennego razvitiya vizantiiskogo goroda v IV-VII vv. (Leningrad 1971). For a
fuller survey of the literature and the debate, see Brandes, 'Forschungsbericht'.
 The cities and the land 95

The extent to which demographic factors - decline or stability, for

example, as affected by plague or similar occurrences - can be taken into
account in this schema is very difficult to determine. It has been pointed
out, for instance, that one area - in north Syria - clearly experienced a
rural-based economic and demographic upswing during the sixth century,
a boom which is not reflected in the literary sources at all, only in the
archaeological record. On the other hand, there is clear evidence of
population decline in regions such as Thrace during the second half of the
sixth century which, constantly ravaged and threatened by hostile action,
seems to have suffered greatly. Different regions thus suffered or not
according to often highly localised factors. The important point to make in
this context is that demographic decline is not the cause of urban decline,
even if it does have an effect when it takes place.6
Within this general context, then, it seems entirely reasonable that we
should expect the continued occupation, if on a smaller scale, of some sites,
the abandonment of others, and even, for specific reasons, the preservation
of certain cities within their original limits, rare though this might be. The
city of Euchaita, for example, which has been singled out as such an
exception, may have survived through the seventh century, although its
population and clergy were often forced to seek refuge from Arab raiders in
the citadel or fortified section of the town, which included an acropolis. The
city was the home of St Theodore Tiro, of course, and as such remained an
important cult centre and an attraction for pilgrims. The cycle of miracles
which describes life in the city in the later seventh and eighth centuries
along with the miracle-working acts of the martyr preserves an important
account of the effects of hostile attacks and economic insecurity in the
town at this period; but the presence of the cult itself seems to provide the
main reason for its survival and its relatively populous nature. 7
Such exceptions, in areas both subject to and free from regular attack,
do not disprove the central thesis: namely, that the structural and func-
tional position of the 'city' in the totality of social and economic relation-
ships of the late Roman state was changing and that it was these changes
in function which lie at the heart of any development - whether of decline
or continuity - in the history of the seventh- and early eighth-century city.
In many respects, the crucial changes had already taken place by the
end of the sixth century. As has been noted in chapter 1, the decline of the
curial order, both as a social group with independent economic resources
and as the competent governing element within municipal administra-
6
See Patlagean, Pauvrete economique, pp. 301ff.; and esp. H. Kennedy, 'The last century of
Byzantine Syria: a reinterpretation', BF 10 (1985), 141-83.
7
See the detailed discussion of F. Trombley, The Decline of the Seventh-Century Town (cited in
note 4 above) and Haldon, 'Some considerations', 96.
96 Byzantium in the seventh century

tions, had several consequences. The cities lost their role as crucial fiscal
intermediaries in the extraction by the state of its revenues. Under the
principate, the income of a city had been drawn from several sources. Rent
on city lands, local taxes and dues, customs duties, interest of endowments
(although Jones has estimated that most of the income from this source
was lost during the great inflation of the third century), along with the
voluntary donations of members of the governing body and the citizenry,
all contributed to the municipal treasury. The proportion of these different
sources of income varied from city to city, according to its character:
whether it was chiefly commercial, an ancient foundation and so on. The
first blow to civic economic independence was struck by Constantine I and
Constantius, who confiscated the civic lands and their revenues. Julian
restored them, but Valens and Valentinian confirmed Constantine's action.
The res privata now administered these lands and their income; but after
374 it was decreed that the cities, which had protested that they could not
cope without this revenue, should receive back one third of the income
from their former lands, as well as being made responsible for the manage-
ment of the lands themselves; furthermore they were to get back one third
of their civic taxes, which had been administered in the meantime by the
sacrae largitiones. Eventually, the management of these taxes, together with
the rents from urban sites and buildings, was also returned to the cities.
Some cities were able to increase their lands through purchase or inherit-
ance, of course, and the richer cities, which derived their wealth as much
from taxes and duties on commerce and so on, often suffered less than
those which were dependent entirely on their rural hinterland. But the
latter made up the majority, and it is clear from the fourth-century
legislation that the cities became increasingly unable to cope with their
own maintenance, quite apart from the administration of the revenues of
the state. 8
As a result of this loss of civic economic independence, the curiales no
longer directed the fiscal administration of the state at the municipal level.
Instead, centrally appointed officials - the curator or pater civitatis - repre-
senting in theory locally elected officers, but constituting in practice part of
the vast machinery of the state bureaucracy, took over this role; these
officials were later complemented by the establishment under Anastasius
of vindices in each civitas, responsible for the collection of imperial revenues
in each city, who seem also to have taken over the administration of the
civic revenues.

8
The evidence is fully discussed by Jones, LRE, vol. II, pp. 732ff.; see also J. Durliat, 'Taxes
sur l'entree des marchandises dans la cite de Carales-Cagliari a l'epoque byzantine
(582-602)', DOP 36 (1982), 1-14; and more particularly the discussion of F.Millar,
 The cities and the land 97

The cities thus lost in both economic and social status. That these
developments were already well advanced by the later fifth century is
generally admitted; and the role of the local bishop, who acted and was
recognised as a potential protector of the city and its populace in respect of
both the landed magnates and the state, is illustrative. A law of Zeno,
repeated in the Codex lustinianus, suggests as much. 9 And while the bishop
also represented - as local agent for the estates of the Church in his region
- one, often the foremost, of the powerful landlords and principales of the
municipality, there is no evidence that bishops ever had any formally
constituted civic duties within the framework of the city administration, as
has sometimes been assumed, on a pattern similar to that of the towns of
the early medieval West. 10
The decline of the curial order (the wealthier or luckier members
obtaining exemption from civic duties through membership of the senate,
the poorer increasingly unable to shoulder the burden of both state and
civic liturgies) was halted for a while, although not reversed, by the state
assuming control of the cities' lands and incomes from the middle of the
fourth century. Henceforth, the magistrates and decurions had to adminis-
ter only the affairs of their city. Even in the later sixth and early seventh
centuries there are enough references to the city fathers of certain cities to
make it clear that not all curiales were either poor or indifferent to their
duties. 11 But, as has been pointed out, even contemporaries were aware of

'Empire and city, Augustus to Julian: obligations, excuses and status', JRS 73 (1983),
76-96.
9
See CJ I, 3.35; and for the curator civitatis note 10 below. For the vindices, Jones, LRE, vol.
II, pp. 759 ff. and notes.
10
See Claude, Die byzantinische Stadt im 6. Jhdt, 135 and 157f. Note the cogent arguments of
A. Hohlweg, 'Bischof und Stadtherr im friihen Byzanz', JOB 20 (1971), 51-62;
G. Dagron, 'Le Christianisme dans la ville byzantine', DO? 31 (1977), 3-25, see 19ff. A
law of Anastasius of 505 {CJ I, 55.11 ( = i, 4.19)) repeating a law of 409 for the West,
however, set up a new assembly, including the bishop and clergy, chief landowners and
decurions, which was henceforth to elect the defensor civitatis - supposedly the protector of
the citizens against official oppression - instead of the city council alone. The measure was
designed chiefly to compensate for the fact that the curiales were no longer the wealthiest
or the most powerful in the city, and might easily fall prey to the provincial governor.
Anastasius' law also gave this assembly the duty of electing a corn-buyer in times of need,
as well as the right to elect the curator or pater civitatis, the official appointed originally by
the central government to supervise and regulate civic finances. See Jones, LRE, vol. n,
pp. 726f. and 758f., and esp. J. Durliat, 'Les Attributions civiles des eveques byzantins:
l'exemple du diocese d'Afrique (533-709)', in Akten des XVI. Internationalen Byzantinisten-
Kongresses, II, 2 (Vienna 1982), pp. 73-84 ( = JOB 32); A. Guillou, 'L'Eveque dans la
societe mediterraneenne des VIe-VIIe siecles: un modele', Bibliotheque de VEcole des Chartes
131 (1973), 5-19 (repr. in Guillou, Culture et societe en Italie byzantine (VIe-XIe siecles) II
(London 1978).
11
See, for example, Vita Theod. Syk., 25, 6 and 45, 2-3. For a good discussion of the decline
of the curial class, see de Ste Croix, The Class Struggle in the Ancient Greek World,
pp. 465-74.
98 Byzantium in the seventh century

the decline of the cities and their councils, as a novel of Justinian for the
year 536 implies.12 Civic autonomy had all but disappeared, in fact, by the
later sixth century. The town councils seem, on the whole, to have either
ceased to exist, or to have been bypassed and become ineffectual. Power
resided with the imperial revenue officials, in particular with the vindices -
and with the assembly of bishops, clergy and magnates. But even the
authority of the latter was tempered by the fact that provincial governors
regularly appointed their own representatives to the cities in their
province, representatives who acted, in effect, as city governors, tending to
assume responsibility for civic buildings, walls, water supply, and the like,
and who, together with the vindices of the bureau of the praetorian prefec-
ture, now dominated city administration. The curiales continued to func-
tion, however, if only in their capacity as simple collectors of state and civic
revenues under the supervision of the imperial establishment.13
By the later sixth century, therefore, there is sufficient evidence to
demonstrate that the cities of the empire had lost their fiscal, economic and
political independence to a very large extent to the state; and that the
dominant administrative and legislative body within each city was now
made up of the bishops and clergy, the richer landowners, some of the
centrally appointed officers of the imperial fiscal bureaux, along with the
now relatively unimportant curiales. Of course, richer curiales still were to
be found in many cities; but the general tendency is clear enough, and even
the extraction and administration of local civic revenues and their expen-
diture came under the supervision and sometimes direct control of imperial
officials. In the richer cities - such as Alexandria, Antioch, or Ephesus, for
example, where wealthy merchants or landowners who had invested in
shipping were to be found or in those cities where local magnates retained
an interest in the well-being of their cities - a certain degree of civic
autonomy survived.14 And, of course, this gradual reduction in the rele-
vance of the cities to the fiscal and political administration of the state did
not always have negative effects on local economic activity, whether
small-scale artisan production, services or market exchange. What is
important is the shift in the function of cities, from self-governing,
12
Justinian, Nov. 38, proem.
13
See Jones' comments, LRE, vol. II, p. 759ff.; and see J. Durliat, 'Les Finances municipales
africaines de Constantin aux Aghlabides', in Bulletin archeologique du comite des travaux
historiques et scientifiques, new series, 19 (1983), 377-86; and note also Zacos and
Veglery, nos. 400, 1462, 2890, all datable to the seventh century, of decurions and, in
the last case, the community (to koinon) of the city of Sinope - perhaps a reference to the
municipal council or the municipality itself. Such seals say nothing, of course, about the
functions of their users.
14
But even a city such as Alexandria was financially under the thumb of the vindex, as
Justinian, Edict. XIII, 15 makes clear; and see the remarks of Durliat (see note 13 above),
378-80 with literature.
 The cities and the land 99
economically independent, localfiscaland administrative agencies, acting
for the state as well as on their own behalf, to dependent urban centres
with no real role in the imperial fiscal administrative system and no
autonomous economic existence.15
By the early seventh century, civic autonomy existed in name only. The
state effectively passed the towns by as far as revenue administration was
concerned. They still functioned as administrative centres for imperial
provincial officials, of course, as well as for the Church. As long as there
was no threat to local trade or exchange activity, they will have continued
to serve as market centres or centres of small-scale commodity production.
But while the later Roman and early Byzantine notion of culture and
civilisation were still inseparable from the concept of the city, with all that
(hat entailed in both practical terms (public facilities and services, for
example) and in ideological respects, few urban centres had the income or
the independence of their former selves; none were essential to the actual
functioning of the state.
THE FATE OF URBAN SETTLEMENT AND MUNICIPAL
CIVILISATION IN THE SEVENTH CENTURY
Thus far the term 'city' or 'town' has been used without distinction to
define urban centres of population; but no attempt has been made to define
more precisely what is meant by these terms. And it is this undifferentiated
application of the term 'city' to both late Roman and Byzantine urban
settlements that has led to both misunderstandings and confusion in the
debate.
Since the later nineteenth century, there has been much discussion on
the question of defining what exactly is meant by the term 'city', and a
15
The changes summarised have been analysed in much greater detail by a number of
scholars. See Jones, LRE, vol. II, pp. 726f., 732-4, 737-63 and The Greek City from
Alexander to Justinian, pp. 1481T. and 2671T.; Kirsten, 'Die byzantinische Stadt', 231T.;
W. Liebenam, Stddteverwaltung im romischen Kaiserreiche (Leipzig 1900 and Amsterdam
1967), pp. 476ff.; H. Aubin, 'Vom Absterben antiken Lebens im Friihmittelalter', in
Kulturbruch oder Kulturkontinuitdt von der Antike zum Mittelalter, ed. P.E. Hiibinger (Darm-
stadt 1968), pp. 203-58, see esp. 213ff.; Kurbatov, Osnovnye problemy, pp. 46ff. and
154ff., and 'Razlozenie anticnoi gorodskoi sobstvennosti v Vizantii (IV-VII vv.)\ VV 35
(1973), 19-32 (where he documents the alienation of civic lands to private persons, the
state and the Church); F. Vittinghof, 4Zur Verfassung der spatantiken Stadt1, in Studien zu
den Anfangen des europdischen Stadtwesens (Reichenau 1956). For a detailed discussion of
the earlier literature, both in the Soviet Union and elsewhere, see Frances, La Ville byzan-
tine et la monnaie: and more recently, the excellent discussion in Brandes, Die Stddte
Kleinasiens, pp. 17-22. H.Kennedy, 'The last century of Byzantine Syria: a reinter-
pretation', BF 10 (1985), 141-84, and H. Kennedy and J.H.W.G. Liebeschuetz, 'Antioch
and the villages of northern Syria in the fifth and sixth centuries A.D.: trends and
problems', Nottingham Medieval Studies 32 (1988), 65-90 present valuable surveys of the
pre-Islamic decline in key aspects of urban culture in the sixth and early seventh centuries.
For further literature, see the bibliographical notice at p. 476 below.
100 Byzantium in the seventh century

number of very different ways of using it has been suggested. The issue is
obviously complicated by the fact that modern attempts to define what a
city should mean in social, economic and political terms - which may be
based (as with Max Weber) on analyses of Western medieval cities - may
not always coincide with what either Romans or Byzantines, for example,
considered to be a city or with the character of those settlements which
were, sometimes very loosely, referred to in the sources as civitates or
poleis.16
The term 'city' in the Roman world was essentially a legal administra-
tive definition applied by the Roman state as it expanded to communities
which, as we have seen, were in origin autonomous and responsible to the
state only in respect of the supervision and collection of the state's rev-
enues. These might be either urban communities proper, real centres of
distinct territoria, or they might be groups of villages, attributed jointly
with the administrative responsibility and the constitutional forms of the
classical civitas or polis.17 But behind this lay also a complex web of
cultural meanings which the term evoked for a member of late Roman
society. The city symbolised Roman, or Greco-Roman, civilisation; it
evoked ideas about both local tradition and imperial context, of literacy
and 'letters' in general, of physical space and a certain economic order.
The term 'city' - civitas - thus referred in the first instance to both the
16
The approaches to the problem are many and varied. On the whole, simple definition
based on legal status alone, or the existence of a circuit wall, for example, have proved
unsatisfactory. Weber's approach to the problem was to establish models of urban
communities, which might help to elucidate the historical realities as constructed through
the sources. But the debate about the use, application and validity of these Ideal types'
remains. See G. Korf, 'Der Idealtypus Max Webers und die historisch-gesellschaftlichen
Gesetzmassigkeiten', Deutsche Zeitschrift fur Philosophie 11 (1964), 1328ffM and Max
Weber, 'Die Stadt. Begriff und Kategorien', in Die Stadt des Mittelalters I: Begriff, Entstehung
und Ausbreitung, ed. C. Haase (Darmstadt 1969), pp. 34-59, together with a number of
other valuable contributions in the same volume. See also the article of D. Denecke, 'Der
geographische Stadtbegriff und die raumlich funktionale Betrachtungsweise bei Siedlungs-
typen mit zentraler Bedeutung in Anwendung auf historische Siedlungsepochen', in Vor-
und Fruhformen der europdischen Stadt im Mittelalter, eds. H. Jankuhn, W. Schlesinger and
H. Steuer (Abhandlungen der Akad. der Wissenschaften in Gottingen, phil.-hist. Klasse III,
83 (1973), 1, pp. 33-55); C. Goehrke, 'Die Anfange des mittelalterlichen Stadtewesens in
eurasischer Perspektive', Saeculum 31 (1980), 194-239; M.I. Finley, 'The ancient city:
from Fustel de Coulanges to Max Weber and beyond', Comparative Studies in Society and
History 19 (1977), 305-27; and the comments of G. Warnke, 'Antike Religion und antike
Gesellschaft: wissenschaftshistorische Bemerkungen zu Fustel de Coulanges' La Cite
antique , Klio 68 (1986), 287-304. For a recent detailed discussion, see Brandes, Die Stddte
Kleinasiens, pp. 23-43.
17
See Jones, LRE, vol. II pp. 712ff., and 'The cities of the Roman empire: political, admi-
nistrative and judicial functions', Recueils de la Societejean Bodin VI (1954), 135-73 (repr.
in The Roman Economy, pp. 1-34), Iff., for example; also the useful survey and discus-
sion of J. Koder, 'The urban character of the early Byzantine empire: some reflections on a
settlement geographic approach to the topic', in 17th International Byzantine Congress,
Major Papers, pp. 155-89.
 The cities and the land 101

centre and its dependent territorium; polis was understood in the same way,
a point which hardly needs to be demonstrated. In the second place, the
term refers to a cultural symbol which points to notions about the past and
the present through which individuals could identify themselves. It is
important to recall, therefore, that while a definition of a city or town based
on predominantly economic and social considerations is essential (that is,
a town is a settlement, in which the dominant form of exchange activity
occurs through market transactions, where some elements of petty com-
modity production are present and necessary, and where market-
exchanges constitute a fundamental element for the existence of the
community), 18 it must equally be remembered that many of the 'cities' of
the Roman world were in origin little more than villages with elements of
the local state administration present; and that while the presence of the
latter may also have attracted or stimulated some market-exchange and
small-scale commodity production, this economic activity was in effect a
secondary phenomenon. In times when communications are unhindered
and a stable small-denomination coinage available, such activity will be
favoured; in times of political and economic insecurity, this activity may
well disappear, although the primary administrative function of the settle-
ment may remain. 19
When a late Roman or Byzantine source refers to a polis, therefore, these
considerations must be borne in mind: polis may refer to a thriving centre
of commercial and exchange activity; it may equally refer to an unimpor-
tant provincial settlement which serves merely as an administrative con-
venience and a shelter for the local populace. Each case must be taken, so
far as is possible, on its merits. Indeed, the situation is complicated by the
varied terminology used in the sources. On the whole, modern historians
have added to the problem by reading the term polis uniformly as 'city' or
'town', thus translating the medieval use of the term into a context where
it must necessarily be, at the very least, ambiguous. This is an important
point, since it is clear that the use of the term polis by Byzantine writers did
not always have the meaning we understand by the term. On occasion,
indeed, it may be used merely for effect or to demonstrate a writer's
familiarity with ancient terminology; and it is certainly clear from the use
of terms such as 'fortress' or 'castle' and polis side by side and often
interchangeably among medieval Greek writers that what was meant by
polis often bears no relation to either the ancient or the modern concept

18
See, for example, K. Marx, Pre-Capitalist Economic Formations, ed. E. Hobsbawn (London
1964), pp. 71f. and 77.
19
For the character of such activity, see J.P. Sodini, 'L'Artisanat urbain a l'epoque paleo-
chretienne (IVe-VIIe siecles)', Ktema 4 (1979), 71-119; and the section on Korykos, in
Tinnefeld, Die fruhbyzantinische Gesellschaft, pp. 215ff.
102 Byzantium in the seventh century

associated with the term. 20 Sometimes, a writer deliberately differentiates

between a city and its acropolis or castle, and here it can be argued, as has
been done for Euchaita, for example, that a larger urban settlement is
meant by the polis, not just a fortress.21 On other occasions, the physical
description of a settlement as a kastron is qualified by the remark that the
place in question was also a polis, that is, it possessed the juridical status
and constitutional form - the corporate personality - of a city.22 On the
whole, the juridical distinction between a polis, with its attendant status
and rights, and other settlements, is retained up to the seventh century.
From this time on, however, the term polis begins to lose its technical
meaning, as we shall see. It is thus essential that modern historians
differentiate also between poleis which had an independent economic
existence and a market character, and those which represent primarily
administrative centres or settlements that obtained the title and privileges
of a city for social-cultural - ideological - reasons.

Anatolia and the eastern provinces

Up to the middle of the second decade of the seventh century the majority of
the urban settlements of the eastern provinces had suffered no permanently
damaging attacks on their lands or their territoria, with the exception of the
fortress settlements of the eastern limes most open to attack from the
Persians during the wars of the sixth century. Antioch, for example, which
was destroyed by an earthquake between 526 and 529, was rebuilt with
imperial aid (by this time a normal procedure - even the wealthiest cities
could not raise the resources necessary for such major building pro-
grammes without imperial assistance, an important illustration of their
economic dependence once their lands were no longer entirely under their
control). But apart from occasional Persian successes, the eastern
provinces suffered only minor discomfort at enemy hands.
This is not to say, of course, that the cities and towns of the empire had
not experienced hostile attacks, or suffered considerable destruction, in
previous centuries. In the 250s and 260s, for example, attacks from both
the Goths on the one hand and the Sassanids on the other penetrated as far
as the Pontic coast, Bithynia, Pamphylia, Cappadocia, capturing or sacking
many cities, including Satala, Caesarea in Cappadocia, Comana, Sebastea,
Trebizond, Pergamum, Nicomedea, Nicaea, Prusa, and Apamea. 23 Other

20
See Haldon, 'Some considerations', 9 0 and note 36.
21
See Trombley, 'The decline of the seventh-century t o w n ' .
22
See, for example, Malalas, 3 0 2 , 22sq.
23
For Antioch, see Jones, LRE, vol. I, p. 2 8 3 . For the Persians, D. Magie, Roman Rule in Asia
Minor (Princeton 1950), pp. 7 0 8 a n d 1568f. a n d pp. 1 5 6 6 - 8 for the Goths; see also
 The cities and the land 103
areas of Anatolia suffered at the hands of Isaurian brigands throughout the
fifth and much of the sixth century, as the legislation of the Codex lustinia-
nus and Justinian's novels demonstrate. The Huns similarly raided
eastern Anatolia from the Caucasus region in 515; while the Persian wars
of the sixth century certainly caused widespread devastation in the regions
around the key fortress cities along the Armenian front and in the diocese
of Oriens.24 The construction of walls around the formerly open and
undefended cities of the empire was a phenomenon which began already
in the later second century, however, gathering pace through the troubled
years of the third century. By the fifth and sixth centuries towns or cities
without defences of some sort were a rarity; but the need to maintain them
was a constant drain on limited resources, and itself contributed to the
changed circumstances which forced the municipalities to turn increas-
ingly to local and central imperial sources for financial aid.25
These developments inevitably affected the towns and cities of the areas
in question, at least for the duration of the perceived danger. They thus
contributed also to the overall pattern of decline in urban fortunes over this
period. But it seems clear from both literary and archaeological evidence
that it was the constant and regular devastation of the seventh century
which hastened the end - inevitable anyway in structural terms - of the
towns of both Anatolia and the Balkans.
From 610, following the rapid conquest of much of Syria, successful
Persian attacks on the Anatolian cities of Satala, Nicopolis, Theodosioupo-
lis and Caesarea took place; and although Caesarea was retaken a year
later, the Persians left it in ruins.26 After the failure of the Byzantine
counter-offensive in 613, Tarsus and Melitene were taken; but in the
following years, as the Persians directed their attention toward the con-
quest or consolidation of Syria, Palestine, and Egypt, it was the urban
centres of these regions which suffered.27 From 615/16 Anatolian cities
suffered once again, as the Persians turned their attention to Constantin-
ople. From 615 onwards, Chalcedon, Sardis, Ancyra and a number of
F. Hild and M. Restle, Tabula Imperil Byznatini, II: Kappadokien (Kappadokia, Charsianon,
Sebasteia u. Lykandos) (Vienna 1981), p. 66.
24
For Justinian's legislation, see Jones, LRE, vol. 1, pp. 280ff.; for the Huns, Stein, Bas-
Empire, vol. 2, p. 105; and for the Persian wars, see the summary in chapter one above
and the literature cited in note 23 above.
25
On the walling of cities, see Trombley, 'The decline of the seventh-century town', 76f.;
J.W. Eadie, 'City and countryside in late Roman Pannonia: the Regio Sirmeinsis', in R.L.
Hohlfelder, ed., City, Town and Countryside in the Early Byzantine Era (New York 1982),
pp. 25-42, see 31; T.E. Gregory, 'Fortification and urban design in early Byzantine
Greece', ibid., pp. 43-64; F.E. Wozniak, 'The Justinianic fortification of interior Illyricum',
ibid., pp. 199-209.
26
See Foss, 'The Persians in Asia Minor', 722f.; Kaegi, 'New evidence'; Sebeos, 61ff.; Vita
Theod. Syk., 153, 3sq.: TIB, vol. II, p. 194.
27
See, for example, Sebeos, 65 and 67.
104 Byzantium in the seventh century

other cities were sacked by Persian forces,28 and the devastation of wide
stretches of central and western Anatolia seems to have been repeated in
the following year. Only with Heraclius' counter-attack and successes from
622/3 on did it cease.
The cities of Asia Minor thus received a substantial blow to their
physical structure and to their hinterlands, although hostile action was not
the only cause. A great earthquake in the years 612-16 seems effectively
to have destroyed much of the old city of Ephesus, a disaster from which it
never recovered. The seventh-century settlement was limited to a fortified
area around the old theatre and to a fortified settlement around the
church of St John on the hill of Aya Soliik. As the seventh century
progressed, the latter seems to have become more and more important,
although the theatre area had the harbour at its disposal. While urban
activity certainly did not cease, the earthquake seems to have been the
final blow to an ancient polis and its extensive suburbs. 29
The urban communities of both the Anatolian and the eastern provinces
generally had little time to recover their fortunes before the beginnings of
the Arab attacks on Roman territory and the subsequent conquest of Syria,
Palestine, Egypt and eventually the rest of Roman North Africa. The cities
of these latter areas suffered one of two fates: either they surrendered
unconditionally - generally being left more-or-less unmolested, although
under new political masters - or they resisted and were either taken by
storm or eventually starved into surrender. The penalty for resistance was
the sack of the town, the death or enslaving of much of the population, and
the destruction of all fortifications and defensive structures. 30
From the 640s until the 740s, on the other hand, the towns and cities of
Anatolia, both in those regions which were now to become frontier zones
and in the heart of the empire, were subject to a continuous series of raids,
major and minor attacks and plundering expeditions. These have for the
most part been well documented by historians, and I will detail them only
very briefly here. But the evidence from both literary and archaeological
sources is graphic; and it is quite clear that the massive and constant
insecurity which was a result had far-reaching consequences for both the
rural and urban populations. Communications became uncertain, the
sowing and harvesting of crops, and certainly their consumption, was
frequently impossible, especially in the most exposed zones; market
28
See the account in Foss, 'The Persians in Asia Minor', 7 2 4 with literature; Foss, Sardis,
pp. 53ff. a n d 'Ankara', 62ff.
29
See Foss, Ephesus, pp. 103ff. for a detailed discussion a n d analysis of t h e sources. He dates
the event to 6 1 4 .
30
See the brief a c c o u n t in chapter 2 above; a n d Gabrieli, Muhammad and the Conquests of
Islam, pp. 1 4 3 - 8 0 . The best analysis of the history of u r b a n centres in Anatolia in the
seventh a n d eighth centuries is n o w Brandes, Die Stddte Kleinasiens.
 The cities and the land 105

activity, which depended both upon the safety of local transport at least, as
well as on the existence of secure centres of exchange and the availability
of a suitable medium of exchange, was extensively disrupted and, where it
continued, was limited either to large emporia secured and supported by
the state or to barter and gift-exchange in kind on a highly localised basis
between rural producers and consumers and local administrative or mili-
tary personnel. What is important to recall is the fact that, while these
'external' factors - essentially, the existence of a constant and real military
threat - were without doubt instrumental in the demise of classical civic
life and institutions, they were so only in so far as they dealt the final blow.
There is no evidence to suggest that any of the cities affected by these
developments would have recovered its ancient dynamism had these
attacks not occurred. Indeed, many cities not directly affected, or only very
occasionally affected, by hostile action or its results, nevertheless suffered
ultimately the same fate as the rest. The already existing tendency outlined
above underlies the developments of the seventh century. 31
The following list illustrates clearly enough the activities of Muslim
forces - whether on major expeditions, as, for example, against Con-
stantinople in the years 674-8 or 717-18, or on the yearly spring or
winter raids. 32 Importantly, a larger city often weathered a siege or attack
- unless it was the specific target of an expedition. But the surrounding
countryside was almost invariably devastated, and it was the destruction
of local resources in foodstuffs, livestock and materials, or the impossibility
- indeed the pointlessness - of attempting to maintain local agricultural or
pastoral activity, which must have led to the rapid decline in the popu-
lation of many exposed or frequently attacked cities.33 The larger cities
shrank in size - to a defensible area - and their character changed as their
hinterland became insecure. Only those centres which could be both easily
defended and which had access to the sea, for example, could continue to
thrive - and the best example, exceptional for many reasons, was Con-
stantinople itself.
In the following, an asterisk and a numeral denotes the capture of a city
and the number of times it was taken; otherwise its presence in the list
31
For a similar conclusion, see A. Kazdan and G. Constable, People and Power in Byzantium:
An Introduction to Modern Byzantine Studies (Washington D.C. 1982), p. 57; C. Mango,
Byzantium: The Empire of New Rome (London 1980), pp. 60ff. for a m o r e detailed survey of
the development of u r b a n life from t h e fourth to t h e n i n t h centuries.
32
On these, see Haldon a n d Kennedy, 'The Arab-Byzantine frontier', 1 1 3 ; 'Kudama b.
Dja c far\ 199f.
33
See t h e apposite c o m m e n t s of Lilie, Die byzantinische Reaktion, pp. 190ff. a n d 3O5f., w h o
points o u t that, while Caesarea a n d Euchaita were each taken for a short while in the
opening phases of the warfare, as later w a s Trebizond, Ancyra (654) a n d A m o r i u m (669)
- all quickly re-taken - it is almost entirely t h e smaller a n d less important u r b a n centres
w h i c h were permanently or regularly captured.
 * * * A - < C ! . " . . > - • - - - " ' • Koloh'eia . A . • . • . • . • . " . " .
• Fortresses/defended towns >:: ;:
..J500m•-.-,;•• C'-,. ^J.;.i--T> : : < i J . / ; C : > ' - •
O Towns abandoned during
this period

• . •,. •. ^ Jurandah '

/ (Taranton) /THi^ialyahV
.-'''},<') I \ # (Melitene)" •HisnQalawdhiyah
' HisnZibatrah •

Koron -N,
|..;>(al-Qurrah) r .-- : -' : }'••';• '<-
V-..:'- ^••••••Rhodentonv/>y^,?y--^^^^

AL-JAZIRAH

Manbidj
(Hierapolis)

100 kilometres

Map V The Anatolian frontier region in the seventh and early eighth centuries
 The cities and the land 107

refers to a siege and the accompanying devastation of its hinterland. The

list covers the period from 636 to 740.
Abydus (* 1), Acroinum (* 3), Amasea (* 1), Amastris, Amisus,
Amorium (* 8), Ancyra (* 3), Antioch (Pisidia) (* 4), Antioch (Isauria)
(* 1), Chalcedon (* 4), Chrysoupolis (* 1), Dorylaeum (* 2), Ephesus (* 1),
Euchaita (* 1), Gangra (* 2), Heraclea (* 3), Iconium (* 1), Caesarea (* 4),
Kamacha (* 4), Camuliana (* 1), Koloneia (* 1), Cyzicus (* 1), Melitene
(* 7), Misthia, Mopsuestia, Myra, Neocaesaraea (* 1), Nicaea (* 2),
Nicomedia, Nicopolis (* 1), Pessinus (* 1), Pergamum (* 3), Podandus
(* 1), Sardis, Satala (* 1), Sebastea (* 1), Sebaste (* 1), Sinope, Sisium (* 1),
Smyrna (* 3), Synnada (* 1), Tarenta (* 2), Tarsus (* 3), Trebizond (* 1),
Tyana (* 3).
These towns or fortresses,34 along with many others, were thus the
target of specific attacks on a number of occasions; and they were affected
by the raids which bypassed them on their way elsewhere even more often.
The frontier areas - Armenia II and IV, Cappadocia, Isauria, Pisidia,
Lycaonia - were raided virtually every year for over a century; the regions
behind them - Phrygia, Galatia, Helenopontus; and then Lydia, Bithynia
and Paphlagonia - were traversed or reached almost as often. Frequently
Arab forces wintered deep in Byzantine territory: yearly from 663 to 668,
from 670 to 672, 679 and so on. And it must be realised that there were
probably many raids or targets not explicitly mentioned in any source.
Even taking into account the occasional truce between empire and
caliphate - as in the late 680s - the effects of these raids and attacks on the
economy of Asia Minor must have been considerable. It is difficult to
calculate or quantify, of course, with any degree of accuracy. Comparative
evidence can be adduced; while the archaeological material, where it is
available, can help to flesh out the bare bones of our narrative and other
literary sources.
For comparative material, we might usefully turn to the effect of
Turkmen raids on Byzantine settlements in western Anatolia in the last
forty or so years of the eleventh century, where it has been demonstrated
very clearly how the livelihood of the 'cities' or fortresses and their
dependent rural populations was very rapidly disrupted and then des-
troyed by regular, but very often quite small, raids. The local agricultural
population, where they were not spared in return for supplying the raiders,
was generally forced to abandon its lands, tenaciously though the peasants
tried to hang on to their livelihoods, fleeing first to the nearest fortress
towns and, when these became too obvious a target - or indeed when they
34
The best comprehensive survey, year by year and raid by raid, is that of Lilie, Die
byzantinische Reaktion, who gives full documentation and discusses the chronology of the
attacks.
108 Byzantium in the seventh century

became too overcrowded and unable to support the influx of refugee^ -

further afield. The result was the destruction of the town's resources and,
unless relieved or resupplied by a military expedition, its forced capitu-
lation or capture and sack. 35
The archaeological record suggests a similar pattern for many of the
cities of Anatolia in the seventh century. It demonstrates initially a general
shrinkage of the area covered by the original settlement - a shrinkage
which, it must be emphasised, begins already in the later fifth and sixth
centuries, a result of the structural developments outlined already - and
entailing the abandonment of much of the city and most or all of its
suburban districts; together with the fortification, usually involving the
extensive use of spolia robbed from the older settlement, of a much smaller
area, more often than not on a hill or promontory or some other such
easily defended position. This is certainly the case with Ephesus, Magnesia
on the Maeander, Sardis, Priene, Miletus, Heraclea, Aphrodisias, Laodicea,
Hierapolis, Myra, Tius, ApoUonia, Patara, Side, Ancyra, Nicopolis, Heraclea
(Pontica), Assus, Pergamum, and very many others. 36 Alternatively, some
35
See Vryonis, The Decline of Medieval Hellenism, pp. 144ff.
36
For Ephesus, see Foss, Ephesus, pp. 103ff. with further literature and Brandes, in Klio 6 4
(1982), 6 1 1 - 2 2 ; for Magnesia: Foss, 'The Persians in Asia Minor', 7 4 2 , a n d 'Archaeology
and the " t w e n t y cities'", 4 8 3 ; for Sardis; Foss, Sardis, pp. 53ff. (and see Kazdan, in
Bv£avrivd 9 (1977), 4 7 8 - 8 4 ) ; for Priene: Foss, 'Archaeology a n d the "twenty cities'",
4 7 9 with further literature; for Miletus: Foss, ibid., 4 7 7 - 8 ; for Heraclea: A. Peschlow-
Bindokat, 'Herakleia a m Latmos: vorlaufiger Bericht uber die Arbeiten in den Jahren
1 9 7 4 u n d 1 9 7 5 ' , Archdologischer Anzeiger (1977), 9 0 - 1 0 4 ; W. Miiller-Wiener, 'Mittelal-
terliche Befestigungen im siidlichen Ionien', Istanbuler Mitteilungen 11 (1961), 5 - 1 2 2 , see
14; for Aphrodisias: see K.T. Erim, 'Recent discoveries at Aphrodisias', in Proceedings of the
Tenth International Congress of Classical Archaeologists (3 vols., Ankara 1978), pp. 1077ff.
and Erim in Anatolian Studies 21-4 (1970-3) and 32 (1981) for yearly reports; and
R. Cormack, 'The conversion of Aphrodisias into a Byzantine city', in Abstracts of Papers of
the 5th Annual Byzantine Studies Conference (Washington D.C. 1979), pp. 1 3 - 1 4 ; for
Laodicea: Foss, 'Archaeology and the "twenty cities"', 484, and literature; for Hierapolis:
P. Verzone, 'Le ultime fasi vitali di Hierapolis di Frigia', in Proceedings of the Tenth
International Congress of Classical Archaeologists (3 vols., Ankara 1978), pp. 1057ff.;
Verzone, in RbKll (1971), pp. 1203-23; for Myra: (the centre of the cult of St Nicholas, of
course) see G. Anrich, Hagios Nikolaos. Der heilige Nikolaos in der griechischen Kirche (2
vols., Leipzig and Berlin 1913), vol. II, p. 518; for the excavations, see J. Borchhardt, ed.,
'Myra: eine lykische Metropole in antiker und byzantinischer Zeit', Istanbuler Forschungen
30 (1975); for Tios: see W.W. Wurster, 'Antike Siedlungen in Lykien', Archdologischer
Anzeiger (1976), 23-49, see 30ff.; for ApoUonia: see Wurster, ibid. 38ff.; also R.M.
Harrison, 'Upland settlements in early medieval Lycia', in Actes du Colloque sur la Lycie
antique (Paris 1980), pp. 109-18; for Patara: G.K. Sams, 'Investigations at Patara in
Lycia (1974)', Archaeology 28 (1975), 202-5; for Side: A.M. Mansel, Die Ruinen von Side
(Berlin 1963), pp. 13ff. with literature; for Ankara: Foss, 'Ankara', 30ff. and 62ff.; for
Nicopolis: F. and E. Cumont, Studia Pontica II (Brussels 1906), esp. 304-11; Heraclea
Pontica: W. Hoepfner, Herakleia Pontike - Eregli. Eine baugeschichtliche Untersuchung
(Vienna 1966), esp. pp. 35-48; for Assus: Foss, 'The Persians in Asia Minor', 740f.; J.M.
Cook, The Troad (Oxford 1973), pp. 240ff. and 369; for Pergamum: see Claude, Die
byzantinische Stadt im 6. Jhdt., 23; Foss, 'The Persians in Asia Minor', 742f. and 'Archaeo-
 The cities and the land 109
urban settlements were abandoned more or less completely, the popu-
lation, or part of it, transferring to a nearby and more defensible site. This
was the case with Colonae, for example, one of the best-known instances.
Here, the population simply moved to the nearby fortress of Chonae,
situated on a rocky promontory. Similarly, the site of the antique city of
Faustinoupolis was abandoned for the fortress site of Loulon, some eight
km. to the south; that of Prymnessus for Acroinum; Aurelioupolis in Lydia
(the classical Tmolus) seems to have lost its population to an ancient but
much more defensible site some five km. away; and Perge in Pamphylia
was left for the better-defended fortress town of Syllaeum.37
Many smaller cities in the frontier-zone and their hinterlands were
ultimately permanently abandoned. The extensive excavations at Anemu-
rium in Cilicia show that the town, already in decline in the late third
century following the Persian attack of 260, and after earthquake damage
suffered in the 580s, had been reduced to a very humble settlement by the
second half of the seventh century and may eventually have been aban-
doned, although this remains uncertain.38 Tyana was likewise aban-
doned;39 according to an account of the later eighth century, Euchaita was
saved from abandonment in the 750s only by a timely miracle;40 Sision
logy and the "twenty cities" \ 479ff. This lists only some of the more recent publications.
Further literature can be found in these works, and in Brandes, Die Stddte Kleinasiens,
pp. 81-131, where a detailed analysis of both literary and archaeological evidence can be
found.
37
See Kirsten, 'Die byzantinische Stadt', 29ff.; for Chonai, see Beck, Kirche, p. 171; for
Faustinopolis/Loulon: F. Hild, Das byzantinische Strassensystem in Kappadokien (Vienna
1977), p. 52; TIB, vol. II, pp. 223ff.; for Acroinum: Kirsten, 'Die byzantinische Stadt', 29
and Kirsten, in RE XXIII, pp. 1154ff.; for Aurelioupolis: C. Foss, 'A neighbour of Sardis:
the city of Tmolus and its successors', Classical Antiquity 1 (1982), 178-201; and for
Syllaion-Perge, see V. Ruggieri, S.J., F. Nethercott, 'The metropolitan city of Syllaion and
its churches', JOB 36 (1986), 133-56. For further analysis and literature, Brandes,
Die Stddte Kleinasiens.
38
See the summaries and analysis of J. Russell, 'Anemurium 1976', Anatolian Studies 27
(1977), 25-9; 'Anemurium: the changing face of a Roman city', Archaeology 5 (1980),
31^10; 'Transformations in early Byzantine urban life: the contribution and limitations of
archaeological evidence', in 17th International Byzantine Congress, Major Papers,
pp. 137-54, see 144-9.
39
See Hild, Strassensystem, p. 46; TIB, vol. II, pp. 298f.
40
See Vita et Miracula Theodori, in H. Delehaye, ed., Les Legendes grecques des saints militaires
(Paris 1909), p. 198. 28-31; D. de F. Abrahamse, Hagiographic Sources for Byzantine Cities
(Ann Arbor, Michigan 1967), pp. 347ff.; Trombley, 'The decline of the seventh-century
town', 69. For the date of the attacks described in miracles 4, 6, 7, 9 and 10 (the 750s),
see C. Zuckerman, 'The reign of Constantine V in the miracles of St Theodore the Recruit
(BHG 1764)', REB 46 (1988), 191-210. Kagdan has pointed out that a careful reading of
the text of the miracles shows evidence of little more than a stronghold, and a local
economy based on cattle-raising; in addition, that Euchaita in the eighth-century version
of the Vita of St Theodore is not a city but a ranch or an estate. See A. Kazdan, 'The
flourishing city of Euchaita?', Fourteenth Annual Byzantine Studies Conference, Abstracts of
Papers (Houston, Texas 1988), 4. In contrast, F. Trombley continued to argue for urban
continuity at the economic level. But the examples he cited actually show no more than
110 Byzantium in the seventh century

provides another example of abandonment; 41 and there were undoubtedly

a great number of smaller settlements, of which we know little or nothing,
which suffered a similar fate. Towns well away from the most exposed
areas were also abandoned. Cnidas in Caria seems to have suffered this fate
in the middle of the seventh century; many others like it - with a relatively
weak economy and no other administrative or functional significance -
will have followed the same route to oblivion.42
Those cities which survived as active economic centres did so because of
their particular position. While it was attacked on several occasions,
Nicaea lay far from the frontier; it had been and remained an important
centre of communications and a commercial nodal point. 43 Smyrna, with
its harbour facilities, presents a similar picture, and like Nicaea eventually
recovered from the effects of the seventh and eighth centuries to become
once more a flourishing centre for trade and commercial activity by the
later ninth century. 44 Trebizond, similarly, with its harbour, and lying well
protected behind the Pontic Alps, also survived, although it, too, was not
entirely free from attack. The town was thoroughly sacked by Arab and
Armenian forces in 655; but by the ninth century Arab writers describe it
as a flourishing trading entrepot. 45 Finally, Attalia seems to have pre-
served its character as a harbour and market-town, although little is
known of its history in the seventh and eighth centuries in detail. 46 These
towns, and others like them, owed their physical and their economic
survival to their geographical advantages; it has been plausibly suggested
that many profited also from the influx of refugees from more exposed
areas, an influx which, while it may have strained local resources, may

that sites continued to be occupied where the state and the Church placed their admi-
nistrative establishments. See F. Trombley, 'The Akropolis and lower city of the Byzantine
"Dark Age" town (7th-8th century): the cases of Gortyna, Soloi and Druinopolis', ibid.,
40.
41
See Lilie, Die byzantinische Reaktion, p. 119 and note 50 with sources.
42
See I.C. Love, 'A preliminary report of the excavations at Knidos (1970)', American Journal
of Archaeology 76 (1972), 61-76, and 'A preliminary report of the excavations at Knidos
(1971)', ibid., 393-405; see also Ruggieri and Nethercott (see note 37 above).
43
See J. Soldi, 'Historisch-geographische Studien iiber bithynische Siedlungen', BNJ 1
(1920), 263-337, see 278ff.; and A.M. Schneider and W. Karnapp, Die Stadtmauer von
hnik (Nicaea) (Berlin 1938) ( = Istanbuler Forschungen, 9), esp. p. 42; W.M. Ramsay, The
Historical Geography of Asia Minor (London 1890 and Amsterdam 1962), p. 45.
44
Foss, 'Archaeology and the "twenty cities'", 480ff.; Magie, Roman Rule in Asia Minor,
p. 796.
45
See, for example, P. Lemerle, 'Notes sur les donnees historiques de 1'autobiographic
d'Anania de Shirak', REA, new series, 1 (1964), 195-202; H. Ahrweiler, 'L'Asie mineure
et les invasions arabes', RH 227 (1962), 1-32, see 30 with literature. For its vital role as a
commercial centre, see R.B. Serjeant, 'Material for a history of Islamic textiles up to the
Mongol conquest, Chapter IV, Ars Islamica 10 (1943), 71-104, see 94.
46
See Jameson, in RE S12 (1970), pp. HOff.; E.G. Bean, Turkey's Southern Shore (New York
1968), pp. 41ff.
 The cities and the land 111

also have stimulated local market activity and reinforced the local agri-
cultural labour-force.47
It seems clear from the archaeological and the limited literary evidence
that whether or not sites were abandoned, the long-term decline of the
classical city was completed during the seventh century by the attacks of
the Arabs in Asia Minor and the consequent social and economic effects.
Hostile activity was not the only contributory factor in this process, of
course. A number of cities or towns suffered considerable damage from
earthquakes, often so severe as to permanently end the life of the ancient
settlement - the effect on Ephesus, for example, providing an interesting
illustration;48 and in several cases this may have occurred before the
seventh century. A number of towns affected by such natural disasters
never recovered their former position or wealth: Miletus,49 Aphrodisias,50
Laodicea,51 Nicopolis,52 and Anemurium, referred to already. While
earthquakes did not cause the decline of the ancient cities, of course, they
did constitute an important factor.53 Similarly, the frequent outbreaks of
plague which were a feature of early medieval history affected the popu-
lation of the empire drastically, especially those dwelling in the (relatively)
confined conditions of an urban settlement. The best-known epidemic is
perhaps that which swept across the Mediterranean in the 540s; but
further epidemics affected parts or all of the eastern Mediterranean world
more or less continuously: Constantinople was affected, for example, in
555-6, 560-1, 572-3, 585-6, 592, 598-9, 608-9, and in 618; and it
affected Persia, Syria, Palestine and Egypt at intervals thereafter.54 It
occurred again in Constantinople in 697-8 where it was endemic for a
year or more.55 It was a constant factor in the eastern Mediterranean

47 48
Ahrweiler, 'L'Asie mineure et les invasions arabes'. Foss, Ephesus, p. 103.
49
W. Muller-Wiener et al., 'Milet 1978-1979', Istanbuler Mitteilungen 30 (1980), 23-98,
see 281T.
50 51
Claude. Die byzantinische Stadt im 6. ]hdt, 18 with literature. Ibid., 155.
52
F. and E. Cumont, Studia Pontica II (Brussels 1906), p. 311.
53
See the remarks of Mango, Byzantium: The Empire of New Rome, pp. 68f.
54
See the detailed account with literature in Patlagean, Pauvrete economique, pp. 85ff.;
P. Allen, 'The "Justinianic" plague', B 49 (1979), 5-20; and esp. J.-N. Biraben, Les
Hommes et la peste en France et dans les pays europeens et mediterraneens I. La Peste dans
I'histoire (Paris 1975), esp. pp. 25IT.; and J.-N. Biraben and ]. Le GoiT, 'La Peste dans la
haute moyen age', Annales 2 4 (1969), 1484-1510, see 1485ff.; D. Jacoby, 'La Popula-
tion de Constantinople a l'epoque byzantine: une probleme de demographie urbaine', fi 31
(1961), 81-109; M.W. Dols, 'Plague in early Islamic history', journal of African and
Oriental Studies 9 4 (1974), 3 7 1 - 8 3 and The Black Death in the Middle Ages (Princeton
1977), pp. 13ff. There is an enormous literature on the subject, which can be pursued
through the works listed here. See the discussion in L. A. Conrad, 'Epidemic disease in central
Syria in the late sixth century. Some new insights from the verse of Hassan ibn Thabit',
BMGS18 (1994) 12-58.
55
See Theophanes, 370. 26sq.; Nicephorus, 40. 4sq. and see Teall, 'Grain supply', 101.
112 Byzantium in the seventh century

basin throughout this period. The consequences for the population must
have been drastic. 56
The urban settlements of Asia Minor were already suffering from the
long-term effects of the general shift in the economic relations within the
later Roman empire between the wealthiest magnates of the senatorial
establishment and their cities. And because of the change in the function of
the cities with regard to the state, they experienced a radical upheaval in
their circumstances during the seventh and early eighth centuries as a
result of the combination of factors described above. This change affected
their physical appearance, their extent, as well as their economic and
social function. The nature of the change is well summarised by the
anonymous tenth-century Persian writer of the book The Regions of the
World and by the Arab geographer Ibn Hawkal, both of whom emphasised
the scarcity of cities (in the Muslim sense of the term) at a time when the
empire was economically much stronger and when its urban settlements
had had a century or more to recover from the warfare of the seventh and
eighth centuries. It is significant that Arab writers use the Muslim terms for
castles (qilac) and fortresses (husun) of the Byzantine cities and towns they
describe - significant because the difference between concepts such as
village, town and city is important in Muslim geographical terminology.
Arab writers differentiate carefully between types of settlement when
describing Muslim lands. Clearly, the fact that they refer to many Byzan-
tine cities in this way suggests that they did not regard such settlements as
cities at all (Arabic madinah). They were not to be compared with Muslim
centres such as Baghdad, for example, or Damascus, or even Constantin-
ople itself, centres of commerce, market exchange, administration and so
on. 57
A comment of the chronicler Tabari emphasises the point, for he men-

56
Patlagean, Pauvrete economique, pp. 87ff. h a s demonstrated the effects of the plague of the
5 4 0 s on the u r b a n life of parts of the Eastern empire; see also H. Kennedy, T h e last
century of Byzantine Syria: a ^ i n t e r p r e t a t i o n ' , BF 10 (1985), 1 4 1 - 8 3 , see 181ff.
57
Hudud al-cAlam, The Regions of the World, t r a n s . V. Minorsky (Oxford 1937), p. 156f.: I n
the days of old, cities were n u m e r o u s in Rum, but n o w they h a v e become few. Most of the
districts are prosperous and pleasant, a n d h a v e (each) a n extremely strong fortress, o n
account of the frequency of the raids which the fighters for the faith direct u p o n them. To
each village appertains a castle, w h e r e in time of flight (they m a y take shelter)'; Ibn
Hawqal, Kitdb Surat al-Ard, configuration de la terre, trans. J.H. Kramer and G. Wiet (Beirut
and Paris 1964), p. 194 (text ed. J.H. Kramer (Leiden 1938), p. 200): 'Rich cities are few
in their [the Byzantines'] kingdom and country, despite its situation, size and the length of
their rule. This is because most of it consists of mountains, castles (qilac), fortresses
(husun), cave-dwellings and villages dug out of the rock or buried under the earth.' For
the Muslim terminology, see G. von Grunebaum, 'The structure of the Muslim town', in
Islam: Essays in the Nature and Growth of a Cultural Tradition (London 1961), pp. 141-58,
see 14If. (repr. in Islam and Medieval Hellenism: Social and Cultural Perspectives (London
1976)).
 The cities and the land 113

tions in connection with Ancyra and Amorion that there was nothing in
the land of the Byzantines greater than these two cities. The results of
archaeological investigations have shown that at this time - 838 - neither
was much more than a well-defended and strategically important fortress.
Even allowing for Tabari's probable desire to glorify and to magnify the
deeds of the victorious caliph who took the cities, his comment is telling.
Ancyra shrank to a small citadel within the walls constructed from spolia
robbed from the old city during the reign of Constans II, probably between
656 and 661, a citadel whose walls contain an area of some 350 metres
by 150 metres. 58 Amorium, likewise an important fortress and military
base, and from the later seventh century probably the headquarters of the
thema or military district of the Anatolikon, was also very small. In 716 it
was successfully defended against a major Arab attack by only 800 men, if
the source is to be believed.59 While it may well have lain in a pleasant and
fertile district and been a flourishing administrative centre, and con-
sequently probably attracted some commercial activity, it was hardly a
city in the sense outlined above, certainly not in the eyes of a contempo-
rary Arab. 60
Whether defined in terms of their economic function, their position as
centres of social wealth and investment, or in terms of their constitutional
status, their administrative character and functions, or their role in the
extraction of revenues on behalf of the state, the classical cities of Anatolia
underwent a dramatic transformation in this period. Some were aban-
doned or destroyed; those that survived shrank to insignificance, often
surviving merely as defended villages; others owed their continued exist-
ence - and the existence of a limited degree of commercial activity - to their
function as military and administrative centres, of both Church and state;
yet others to their geographical position in respect of trade-routes and
distance from enemy threat. 61 When cities did recover their economic
well-being, during the later ninth century and after, it was not as revived
or reinvigorated classical or late antique poleis, but as medieval towns,
owing their fortunes to the administrative and military intervention of the

58
See Foss, 'Ankara', 74ff., 78.
59
See Mango, Byzantium: The Empire of New Rome, p. 72; W.E. Kaegi, jr., 'Two studies in the
continuity of late R o m a n and Byzantine military institutions', BF 8 (1982), 8 7 - 1 1 3 and
"The first Arab expedition against Amorium', BMGS 3 (1977), 1 9 - 2 2 .
60
For the region in w h i c h it lay, see Michael Syr., Chron., vol. II, p. 4 4 1 (for the year 647).
61
See Kirsten, 'Die byzantinische Stadt', 2 0 and 28f., for example; the exceptional position
a n d history of the Crimean t o w n of Cherson provides a n illuminating example: see A.I.
R o m a n c u k , 'Die byzantinische Provinzstadt v o m 7. J a h r h u n d e r t bis zur ersten Halfte des
9. J a h r h u n d e r t s (auf Grund v o n Materialien a u s Cherson)', in H. Kopstein, ed., Besonder-
heiten der byzantinischen Feudalentwicklung (BBA L, Berlin 1983), pp. 5 7 - 6 8 with lite-
rature.
114 Byzantium in the seventh century

state, the safety of local and long-distance commerce, and their role as
centres of market-exchange and local society and culture.

The Balkans and Constantinople

While I have concentrated thus far on Asia Minor, the Balkans fared no
better. Only Thessaloniki retained any importance as a centre for trade,
and then only with great difficulty and on a very limited basis. Many older
cities shrank, becoming simply fortresses and/or administrative centres. As
in Asia Minor, many were abandoned or destroyed; only those with access
to the sea and some potential for trading or supplying a local demand in
crafts could hope to survive. The disappearance of the antique pattern of
urban centres took place during the later sixth and seventh centuries, as in
Asia Minor. It is clearly connected with the long-term effects of, in par-
ticular, the Avar and Slav incursions into, and occupation of, much of the
Balkan area south of the Danube. Some cities, of course, survived, and for
the same sorts of general reasons as those outlined for Anatolia.62
Until recently, it was argued by many scholars that the continuity of
names evident for a number of Balkan urban sites was alse evidence of a
continuity of occupation and traditional urban activity and life. But it has
been shown that many classical names also survive in Anatolia, and there
it is no guarantee that continuity in the strict sense was the case. Indeed,
while continuity of name is an element of importance, the mere survival of
a city name in a later form is no evidence that the site in question
continued to house either an urban centre of any sort, still less the market
and economic elements necessary to define it as such.63 The results of an
analysis of literary evidence and archaeological material seem to confirm
the pattern outlined already, a pattern of long-term decline in urban and
municipal fortunes, beginning in the third century, and ending with the
eclipse of urban life - save a few exceptions - during the first half of the
seventh century. Those 'cities' that survived - the harbour town of
Odessus, for example, at the mouth of the Danube, or the Danube settle-
ments of Durostorum and Bononia - did so as closed fortresses, similar in
62
See the c o m m e n t s of B. Baskovic, 'L'Architecture de la basse antiquite et du moyen age
dans les regions centrales des Balkans', in Rapports du XIF Congres Internat. des Etudes
Byzantines VII (Ochrid 1961), pp. 1 5 5 - 6 3 .
63
See, for example, V. Besevliev, 'Zur Kontinuitat der antiken Stadte in Bulgarien', in Neue
Beitrdge zur Geschichte der alten Welt II: Romisches Reich (Berlin 1965), pp. 2 1 1 - 2 1 , who
argues from continuity in toponymy to continuity in economic and urban life. This is a
doubtful procedure; and other evidence has shown that many of these sites were little
more than hamlets during much of the medieval period. See A. Petre, 'Quelques donnees
archeologiques concernant la continuity de la population et de la culture romano-byzanti-
nes dans la Scythie mineure aux VIe et VIIe siecles de notre ere', Dacia new series, 7
(1963), 317-57.
 The cities and the land 115

many ways to the fortress towns of Byzantine Asia Minor, but not in
Roman hands. 64
The exception that proves the rule in this general argument is, of course,
Constantinople. Here, all the archaeological and textual evidence points to
a very considerable decline in population during the later seventh and
eighth centuries. It seems likely that many of the buildings and
monuments of the fifth and sixth centuries fell into disrepair or were
abandoned, and that large areas of the city within the Theodosian walls
were deserted.65 The later plan of the city, as it is known from the tenth
century, with its internal cemeteries, monastic gardens and orchards,
private pastures and estates, suggests the reduction of the urban popu-
lation to well below its sixth-century peak, probably the highest point it
ever reached. 66 Only a very few major public works are mentioned for the
whole of the second half of the seventh and the eighth centuries, and only
two monasteries are recorded as having been founded at that time. 67 The
public works included the building of a defensive wall around the palace
precinct by Justinian II (a valuable indicator also of the political climate
during his reign) and the repairing of the Theodosian circuit after an
64
See the general summary of G. Gomolka, 'Bemerkungen zur Situation der spatantiken
Stadte und Siedlungen in Nordbulgarien und ihrem Weiterleben am Ende des 6. Jahrhun-
derts', in Studien zum 7. Jahrhundert, pp. 3 5 ^ 2 ; also A. Milcev, 'Der Einfluss der Slawen
auf die Feudalisierung von Byzanz im 7. Jahrhundert', ibid., pp. 5 3 - 8 , see 55f. For a
general political-historical survey, see also P. Lemerle, 'Invasions et migrations dans les
Balkans depuis la fin de l'epoque romaine jusqu'au VHP siecle', RH 2 1 1 (1954),
2 6 5 - 3 0 8 ; and Ostrogorsky, 'Byzantine cities in the early middle ages', 107ff., 11 Iff.; R.-J.
Lilie, " T h r a k i e n " und "Thrakesion"', JOB 26 (1977), 7-^t7, see 35ff. For an
instructive parallel, the history of the garrison town of Sirmium is valuable, albeit for an
earlier period: see Eadie, 'City and countryside in late Roman Pannonia' (cited note 25
above), 25ff.
65
The best textual evidence for this comes from a later eighth-century compilation known as
the 'Brief historical notes' {Parastaseis syntomoi chronikai), which repeatedly points out
that a certain monument or building had once existed but was now destroyed or in ruins.
Many other monuments had whole mythologies built up around them, their original
purpose and function being entirely lost. See Th. Preger, Parastaseis syntomoi chronikai, in
Scriptores Originum Constantinopolitanarum (2 vols., Leipzig 1901 and 1907, repr. New
York 1975), vol. I, pp. 19-73. This edition has now been reproduced, accompanied by a
translation into English and a commentary: see Averil Cameron and Judith Herrin,
Constantinople in the Early Eighth Century: the Parastaseis Syntomoi Chronikai (Leiden
1984). In this context, see the comments in the introduction, pp. 3Iff. and 45-53; and
esp. C. Mango, 'Antique statuary and the Byzantine beholder', DOP 17 (1963), 53-75
(repr. in Byzantium and its Image V (London 1984)). For the most concise recent survey of
the decline of the city in the seventh and eighth centuries, see C. Mango, Le Developpement
urbain de Constantinople (IVe-VIIe siecles) (Paris 1985), pp. 51-62.
66
See D. Jacoby, 'La Population de Constantinople a l'epoque byzantine: un probleme de
demographie urbaine', B 31 (1961), 8 1 - 1 0 9 (repr. in Societe et demographie a Byzance et
en Romanie Mine (London 1975)) for the sixth-century population - estimated at about
400,000.
67
See P. Charanis, 'The monk as an element of Byzantine society', DOP 25 (1971), 6 1 - 8 4
see 65f. for the monasteries.
116 Byzantium in the seventh century

earthquake in 740. 68 It has been argued that the fact that the main
aqueduct of the city fell into disuse some time after 626 (when it was
damaged or destroyed during the Avaro-Slav siege of that year) and was
(apparently) only repaired in 766, is good evidence of a very low popu-
lation for the capital throughout this period - it could clearly manage on
the few internal cisterns and springs nearby.69 The point is confirmed by
the testimony of the 'Brief history' of the patriarch Nicephorus, who notes
that after the plague of 747 the city was almost entirely uninhabited; the
dead were disposed of within the Constantinian walls of the city in old
cisterns, ditches, vineyards and orchards.70 The city was struck by plague
in 619 and 698 also;71 and in the eighth century the Emperor Constantine
V had to repopulate the city from Greece and the Aegean islands.72
Incidental evidence reinforces this picture of contraction and aban-
donment of many parts of the previously inhabited area of the city, and the
reduction of commercial and exchange activity. After the sixth century,
the stamping of bricks, hitherto widespread, ceases (although its exact
purpose remains uncertain); while the quality of the locally produced
ceramics seems also to have suffered a decline.73
But in spite of the dramatic fall in its fortunes, Constantinople survived,
primarily because it was both the seat of the emperors, the single source of
social and political power and authority in the empire, and at the same
time extremely well positioned and defended. And exchange- and market-
activity clearly did continue. The fictional Life of St Andrew the Fool,
many parts of which date to the later seventh century, as well as the col-
lection of the miracles of St Artemius, compiled probably in the 660s, both
refer on several occasions to the sale and purchase of vegetables, fruits,
wine and clothing, as well as to the use of small denomination bronze
coins. In addition, we also read of other aspects of city life - taverns,
brothels, street-gangs, beggars, foreign merchants, candle-makers and so
on. Payment in cash is taken for granted.74
From the later eighth century there is some evidence for a revival, a
revival which can be detected also in other urban centres of the empire
68
See Theophanes, 367. 12-14 and 412. 16-20 for Justinian II and Leo. Note that Leo had
to impose a special tax to raise the labour-force.
69
See Mango, Le Developpement urbain de Constantinople, pp. 56f. But for a different view see
Trombley, 'Byzantine "Dark Age" Cities', 43 5f.
70 Nicephorus, 63. 1-64. 12, esp. 64. 10-12; Theophanes, 423. 4-29.
71 Nicephorus, 12. 6-9; Theophanes, 371. 23. 72 Theophanes, 429. 22-5.
73 See J.W. Hayes, Excavations at Sarachane in Istanbul, vol 2: The Pottery (Princeton 1992); see
R.B.K. Stevenson, The Pottery 1936-1937: The Great Palace of the Byzantine Emperors
(Oxford 1947), pp. 33ff., for the pottery; Mango, Le Developpement urbain de Constantinople,
for the bricks.
74 See Vita Andreae Sali, 648D, 656B, 660A, 7O8D (drinking houses); 649A, 652C etc.
(brothels, street theatres); 689C, 713D (purchase/sale of vegetables and fruit); 656B
(purchase of vegetables, payment for entry to baths) and 653C, 656B, 777C for lepta,
follera, miliaresia, tremissis; Miracula S. ArtemiU 32. 26-7; 26. 27, -28. 12.
 The cities and the land 117

from the same time. But the important point is this: that the developments
which could clearly affect Constantinople, the most populous, well
defended and administratively and socially most important city in the
empire, in such a dramatic way, can hardly have been less drastic in their
effects on provincial cities.

CITIES AND MARKET EXCHANGE

It remains to examine briefly the evidence for cities or other settlements as

centres of economic activity, that is, as centres of market-exchange or of
production. The evidence relating to this problem for the seventh and
eighth centuries is slight. Literary texts are, as we have seen, difficult to
interpret, given the regular Byzantine use of the term polis to describe
something which was clearly often simply a fortress. Some texts are more
explicit, however: by the end of the eighth century, according to the
chronicler Theophanes, the fair at Ephesus produced market-taxes of some
100 lbs. of gold for the imperial fisc, a sum which, even if not accurate,
suggests a wealth of market activity, at least on this occasion.75 Fairs, of
course, often connected with specific religious celebrations (in this case the
feast of St John the Theologian) were not always annual; nor do they
necessarily represent the normal state of affairs: in the eleventh century,
for example, John Mauropous, the bishop of Euchaita, comments on the
fact that the great feast of St Theodore, which attracted people 'from every
nation', transformed the city from a 'wasteland' to a populous city with
markets and stoas. 76 Clearly, the wealth and brilliance of the fair did not
represent the day-to-day social and economic reality of life in Euchaita;
and this was probably the case with most other such occasions.77
Other arguments have been adduced to demonstrate the nature and
extent (and usually continuity) of urban economic activity at this time. The
evidence of coins has played an important role in these debates.
There existed in the later Roman and Byzantine worlds in effect two
currencies: a gold coinage designed primarily for the use of the state in
redistributing its revenues and in collecting its income; and a low denomi-
nation copper or bronze coinage, tied at a nominal ratio into the gold
system, issued for the purposes of both revenue and market transactions.
Ostrogorsky, along with a number of other scholars, argued that the
continued presence of gold coins on many Anatolian and Balkan sites

75
Theophanes, 469f.; cf. H. Antoniadis-Blbicou, Recherches sur les douanes a Byzance, V
'oclava', le 'kommerkion' et les commerciaires (Paris 1963), pp. 107f.
76
See P. de Lagarde and J. Bollig, Johannis Euchaitarum Metropolitae quae Supersunt in Cod.
Vaticano Graeco 676 (Berlin 1882), pp. 131-2; note also Hendy, Studies, pp. 141-2.
77
For a list, see Vryonis, The Decline of Medieval Hellenism, pp. 39-41.
118 Byzantium in the seventh century

during the seventh century, while being fewer in quantity than in the sixth
century or in the ninth century, nevertheless demonstrated the continued
existence of a relatively healthy 'monetary' economy and a good level of
commercial activity. While noting at the same time that there was a
dramatic fall-off in finds of the copper coinage from the period in question,
he argued that this was not a significant element.78
In fact, it is crucial. For the presence of gold coins has very little to do
with the existence of a market economy, being a reflection rather of the
needs of the state and its military, administrative and fiscal machinery. A
decline in the number of copper coins, however, assuming that it is not
merely a reflection of their lack of intrinsic value, the whim of collectors, or
accidents of deposit and discovery,79 must surely be ascribed to a reduction
in the number of-coins struck and consequently reflect a general decline in
demand, or the perceived demand. In other words, such coins were
required in smaller numbers (and only in certain places), because the
market-transactions they were designed to facilitate no longer took place,
or took place on a very much smaller scale than had previously been the
case. 80 Coin finds from excavations on the one hand, and coins (both gold
and copper) in collections on the other, tend to bear out the assumed
results for the economic life of the towns and cities of the empire outlined
already: after evidence of much hoarding in or around 615, presumably a
result of the threat of Persian attack, followed by a gradual recovery in the
use and circulation of both copper and gold coins, finds dateable to the last
years of the Emperor Constans II virtually cease. 81 Athens and Corinth in
the Balkans and a whole series of excavated sites in Anatolia (Ephesus,
Sardis, Priene, Ancyra, Assus, Aphrodisias, Anemurium) demonstrate
this feature.82
From the reign of Heraclius through the rest of the seventh century the

78
See Ostrogorsky, 'Byzantine cities in the early middle ages', 50ff.; see, for example, also
Vryonis, The Decline of Medieval Hellenism, p. 7 - they both a r g u e against Kazdan,
Vizantiiskie goroda (an argument repeated also in Derevniya i gorod v Vizantii (IX-X vv.)
(Moscow 1960), pp. 264f.). But see t h e critique of Ostrogrosky by P. Grierson, 'Coinage
and m o n e y in the Byzantine empire 4 9 8 - € . 1 0 9 0 ' , in Moneta e Scambi nell'Alto Medioevo
(Settimane di Studi del Centro Italiano di Studi sull'alto Medioevo VIII, Spoleto 1960),
pp. 4 1 1 - 5 3 , see 445ff.; a n d the detailed account of the literature in Brandes, Forschungs-
bericht, 1 9 2 - 2 0 0 .
79
See the comments of Grierson (see note 78 above).
80
See the survey carried o u t by D.M. Metcalfe, 'How extensive w a s the issue of folles during
the years 7 7 5 - 8 2 0 ? ' , B 3 7 (1967), 2 7 0 - 3 1 0 , for a slightly later period, esp. 272ff., where
similar points are m a d e in respect of the function a n d quantity of low-denomination coins.
81
See, for example, Foss, 'The Persians in Asia Minor', 72Iff.
82
See Foss, Ephesus, pp. 197f., with table; 'The Persians in Asia Minor', 7 3 6 - 4 2 ; a n d esp.
C. Morrison, 'Byzance a u VII e siecle: le temoignage de la numismatique', in 'Aipiepctjia
orov Avdpea N. XrpaTO (2 vols., Athens 1986), vol. I, pp. 1 4 9 - 6 3 , esp. 155ff. and
tables.
 The cities and the land 119

copper coinage was depreciated: the follis' original weight of eleven grams
was reduced in 615/16; it was then further reduced to some five grams;
and while it later stabilised for a while, constant efforts to bolster up the
copper system by various reforms point to its dubious reputation and
unreliability as a means of generalised exchange.83 The gold coinage,
while it retained its integrity, and while it is also found hoarded,84 played
only an accidental role in commercial transactions: where it was available
and relevant to the types of transaction taking place, it might tend to be
drawn in to market exchanges. The copper coinage disappears more or less
completely from the last years of the reign of Constans II, after about 658,
as it became less relevant to the requirements of exchange activity: that is
to say, as market exchanges of a day-to-day variety petered out in all but a
few major emporia - such as Constantinople - or as it found other forms
through which goods could be transferred from one person to another,
such as gift-exchange and barter. The state continued to produce the small
denominations, of course, but their distribution and their use seems to
have been very limited.85 Where finds of copper coins do occur, they are
more often than not to be connected with the presence of soldiers, for it
seems that from the early seventh century both copper and silver had to be
brought into the payment system of the state due to a lack of adequate
supplies of gold.86 And it can be argued that the drastic disappearance of
copper from archaeological sites after the last years of Constans II was due
to the state increasingly financing its forces with produce, equipment and
so forth in kind, rather than through the traditional medium of cash grants
and salaries, and thus making the regular issue of coin for this purpose
unnecessary or, at best, marginal to its fiscal needs. Even the limited
amount of copper found after the reign of Constans, therefore, may not

83
See Hendy. Studies, pp. 498f.; and Sp. Vryonis, jr., 'An Attic hoard of Byzantine gold coins
(668-741) from the Thomas Whittemore collection and the numismatic evidence for the
urban history of Byzantium', ZRV1 8 (1963), 291-300, see 292; P.Charanis, 'The
significance of coins as evidence for the history of Athens and Corinth in the seventh and
eighth centuries', Historia 4 (1955), 163-72; Foss, 'Ankara', 87.
84
See the remarks of C. Morrison, J.N. Barrandon and J. Poivier, 'Nouvelles recherches sur
l'histoire monetaire byzantine: evolution comparee de la monnaie d'or a Constantinople et
dans les provinces d'Afrique et de Sidle', j6B 33 (1983), 267-86, esp. 274f., where
fluctuations in the purity of gold can be related to regional historical contexts and
developments.
85
In the fictional Life of Andrew the Fool, which dates probably to the later seventh century
(see the chapter on the sources, note 20), the regular use of small-denomination copper
coins in day-to-day transactions is taken for granted: see PG CXI, 653C, 656B, 777C.
Compare the results of Metcalfe's analysis with regard to distribution (see note 80 above),
3O5ff. See also C. Morrisson, 'Byzance au VTF siecle: le temoignage de la numismatique', in
Byzantion: Tribute to Andreas K Stratos I (Athens 1986), pp. 149-63.
86
See Hendy, Studies, pp. 640-3 for payment in copper. The question of supplying the
armies in kind will be examined in detail below.
120 Byzantium in the seventh century

necessarily signify the existence of a generalised system of commercial

exchange in urban centres at this time.
Lack of exchange activity through the money medium does not imply
either a lack of exchange activity as such, of course, or the complete
desertion of the sites in question - one of the mistaken inferences drawn by
the first proponents of urban discontinuity.87 Nor does the possibility that
the cessation of finds of copper coins on most excavated sites might be a
reflection of state policy (a response to both changes in patterns of demand
and/or changes in the perceived need for such issues) imply that urban life
continued just as before, but without money. On the contrary, there is
every reason to believe that the fiscal policy of the state itself reflects the
same conditions as those which affected urban culture so drastically, and
which I have described already. 88
Let me summarise. The evidence of texts, numismatics and archaeology
all point uniformly in one direction: the effective disappearance of the late
antique urban economies which had survived up to the reign of Heraclius.
What remained was instead a pattern of defended villages and fortresses,
the strongest of which often came to serve as the administrative and
military centres; and, on the coasts of the Black Sea, the Aegean, the
Adriatic and in south-west Asia Minor, there are a few isolated ports and
emporia. These represented the seeds of the future medieval towns of the
Byzantine world, very different in both social character and urban struc-
ture from their classical antecedents. But in practice, only Constantinople,
the centre of the imperial administration, the seat of the emperors and the
site of the major imperial mints, was able to maintain its identity as a city
in the late ancient sense of the term. But even there, what began to develop
from the middle of the seventh century was a medieval town, not a
classical polis.
There were undoubtedly some exceptions to this general pattern, but
they hardly affect its overall validity. Even Euchaita which, it has been
argued, retained much of its original civic area along with some of the
public buildings which occupied it, was hardly a flourishing polis. The
citadel to which citizens fled during Arab attacks had become synonymous
with the term polis for the hagiographer who describes it in the eighth
century; and while the lower (late antique) town still stretched out within
the Anastasian walls of the early sixth century, there is no evidence to
suggest that the population was particularly large, nor that any degree of
87
See the apt c o m m e n t s of P. Charanis, 'A note o n the Byzantine coin finds in Sardis and
their historical significance', EEBS 3 9 ^ * 0 ( 1 9 7 2 - 3 ) , 1 7 5 - 8 0 a n d 'The significance of
coins as evidence' (cited in note 8 3 above), 164ff.; a n d for gift-exchange, see Haldon,
'Some considerations', 8 4 a n d note 18.
88
See, for example, the c o m m e n t s of J. Russell, 'Transformations in early Byzantine u r b a n
life' (note 38 above), 1 4 2 . For state fiscal policy, see chapter 5 below.
 The cities and the land 121

market exchange was a usual feature of its life. It was essentially a rural
settlement and pilgrimage centre. The comments of its eleventh-century
bishop referred to already - long before it began to suffer under the
Turkmen nomads - places it in its real context. 89
The history of the seventh-century Byzantine town is concerned, there-
fore, not with the question of whether or not life in the classical cities
ended; nor whether the cities themselves were abandoned. It is concerned
rather with the changed conditions in which they found themselves from
the fourth and fifth centuries and after. That many - possibly the majority
- of the 'cities' continued to be occupied (if only as centres of refuge or as
military and administrative bases) is not in doubt. Some were abandoned,
certainly; and the economic and market role of many was also ended
under the changed conditions of the seventh century. On the other hand,
some cities certainly continued to function as both centres of population
and market exchange activity, where their provisioning could be assured.
What is crucial, and what indeed had actually occurred before the physical
destruction of the seventh century, is the change in the function of cities or
towns within late Roman society and economy. They were quite simply no
longer relevant to the state or to the greater part of the ruling elite. Where
they survived, therefore, it was either because they could fulfil a function in
respect of the institutions of Church or state (as an administrative base, for
example) or in respect of genuine economic and social patterns of demand.
The numerous episcopal cities must still have supported a degree of
exchange activity, however limited and closed, for example, in order to
meet the needs of the clergy and ancillary personnel. The same will have
applied to sites where civil or military officials were established. The
occasional coin and the production of local imitations of formerly imported
wares on a number of sites suggest as much. Where such demand was met
in other ways, or ceased, the life of the urban centres ebbed away, too. 90
There remain two categories of evidence which have not been men-
tioned so far, namely the various notitiae of episcopal sees and the lists of
signatories to the ecumenical councils - in our period, those held at
Constantinople in 680 and in 692. Neither source is of any value in telling
us about the actual condition of the cities or sees in question; and of the
two, the notitiae, as has often been pointed out, are of only limited use in
89
See the article of Trombley, 'The decline of the seventh-century town', and note 76 above.
90
See, e.g., J.W. Hayes, 'Problemes de la ceramique des VIIe-IXe siecles a Salamine et a
Chypre', in Salamine de Chypre, Histoire et archeologie (Paris 1980), pp. 375-87, for the
evidence of pottery. For the marked shifts in the organisation of the 'cultural space' of
cities in the sixth and seventh centuries, the dramatic decline of urban culture as
evidenced in artistic production, architecture and public buildings, for example, see
especially the detailed discussion of Miiller-Wiener, 'Von der Polis zum Kastron', esp.
451-62.
122 Byzantium in the seventh century

telling us which cities or sees were still within imperial territory or under
imperial authority.91 They represented rather a theoretical state of affairs,
being often quite anachronistic, particularly for the period with which we
are concerned here, when territory was being lost, or fought over, con-
stantly, and when the political status of a city might change within a very
short period. The conciliar lists of signatories are more useful, since they
represent the signatures of clerics who actually attended the meetings in
question, and they might be useful on that basis in arguing for the
continued occupation of their respective cities and their being still within
imperial territory - although it must be remembered that representatives
from other patriarchates such as Antioch or Jerusalem, clearly outside the
imperial jurisdiction at the time of the sixth and Quinisext councils, were
also present. The use to which these lists has been put has varied. The
question of whether or not there was an exact equivalence of city and see
needs to be borne in mind; while the reliability and completeness of the
documents themselves in their edited form is problematic: as Ostrogorsky
pointed out, there are some 174 signatures to the acts of the sixth council;
yet Theophanes, writing some 130 years later, had information that there
had been 289 bishops present. It has likewise been argued that the lists of
signatories may include absentee bishops whose sees were only nominally
within the empire.92 On the other hand, the archaeological evidence for
the continued occupation of many urban sites, at however limited and
lowly a niveau, does show that such settlements, even when lying in very
exposed areas, were seldom entirely deserted or permanently abandoned.
Part of the reason lies in their actual location, often an ancient site with
access to both routes and amenities, chiefly water; and in the agricultural
resources at their disposal.93 The lists of signatories, while they certainly
91
For a detailed discussion, see H. Gelzer, 'Ungedruckte und ungenugend veroffentlichte
Texte der Notitiae Episcopatuum', Abhandlungen der bayer. Akad. der Wissenschaften
XXI (Munich 1901), pp. 5341T.; and now the edition and commentary of J. Darrouzes,
Notitiae Episcopatuum Ecclesiae Constantinopolitanae (Paris 1981). See also Ostrogorsky,
'Byzantine cities in the early middle ages', 105 and notes.
92
Theophanes, 360. 2 and Ostrogorsky, 'Byzantine cities in the early middle ages', 53 and
notes; a point taken up also by C. Foss, Byzantine Cities of Western Asia Minor (Cambridge
Mass. 1972), pp. 28f. and Lilie,' "Thrakien" und "Thrakesion"', 35; see also J. Darrouzes,
'Listes episcopales du Concile de Nicee (787)', REB 33 (1975), 61. For a counter-critique,
see F. Trombley, 'A note on the See of Jerusalem and the synodal list of the sixth
oecumenical council (680)', B 53 (1983), 632-8; R. Riedinger, 'Die Prasenz- und Sub-
skriptionslisten des VI. okonomischen Konzils (680/1) und der Papyrus Vind. gr. 3',
Abhandlungen der bayerischen Akademie der Wissenschaften, phil.-hist. Kl., n.F. LXXXV
(Munich 1979); and H. Ohme, Das Concilium Quinisextum und seine Bischofsliste. Studien zum
Konstantinopeler Konzil von 692 (Arbeiten zur Kirchengeschichte LVI. Berlin-New York
1990); and for the equivalence of city and see, A.H.M. Jones, Cities of the Eastern Roman
Provinces (Oxford 1937), pp. 519f.
93
The close relationship between continuity of site and occupation, and geographical/
climatic situation has been emphasised by Guillou, La Civilisation byzantine, pp. 19IT., 4Iff.,
followed by Hendy, Studies, pp. 90-100 (Anatolia) and 78-85 (Balkans).
 The cities and the land 123

seem to be incomplete, might support this evidence, and suggest that the
bishops in question were indeed normally resident in their 'cities'. It might
be objected, of course, that conditions in some areas must have led to
abandonment, even if temporary - a point made explicitly in the record of
the Quinisext council.94 And while it has been argued that, since more
bishops attended the Quinisext than attended the council of 680, there
may have been a recovery in the fortunes of cities in parts of Asia Minor as
a result of the truce with the Arabs made in 679, the evidence of Theo-
phanes referred to above casts some doubt on this. In addition, the truce,
which came into force in 680, will just as probably have facilitated the
movement of Churchmen to Constantinople for this council, a point
demonstrated by the fact that news of the empire's Bulgar war reached
Apamaea in Syria II without difficulty.95 But the truce itself only lasted
until 692, and while it certainly made communications between Con-
stantinople and the provinces easier, it was hardly long enough for any
sort of urban economic recovery. The statement regarding clerics who had
left their cities and abandoned their flocks was made in 692 at the
Quinisext, some twelve years after the truce had first taken effect. This
hardly suggests that conditions had improved dramatically. Even if more
bishops did attend the council of 692, this reflects travel conditions only.
The fact that it may imply that bishops were, on the whole, resident in their
sees, says nothing at all about the 'cities' themselves.96
The point made already must be stressed: the crux of the matter is,
surely, that the debate has wrongly assumed an intimate connection
between the question of whether classical urban life continued or died and
that of whether or not its sites were abandoned or deserted. The evidence
we have surveyed makes it clear that both questions are to a large extent
misguided. Classical civic life was already on its death-bed before the
seventh century; what was replacing it was provincial town life of a very
different character, on a very much less wealthy and less physically
extensive basis. Hostile attack and harassment speeded up the former
process and almost smothered the latter. But the organisational needs of
the church alongside local cultural and economic tradition kept many sites
alive, even if chiefly because of the shelter they offered. What survived was
an 'urban' culture of a sort; but it bore little or no relation to the antique
cities on whose sites it evolved, whatever the occasional exception may
suggest.
94
Mansi, XI, 952B-C (canon 18).
95
See Mansi, XI, 617A-B; and note the comments of Trombley, 'A note on the See of
Jerusalem', 636f.
96
The evidence adduced by Hendy, Studies, pp. 76f. (letters, occasional references in narra-
tive and other sources) for provincial bishops normally residing in their sees in peace time,
is hardly relevant to the debate, which is concerned with the specific conditions prevailing
towards the end of the seventh century.
124 Byzantium in the seventh century

Finally, it is worth emphasising the fact that the walled urban settlement
was an important element in Byzantine perceptions of their culture. Towns
or 'cities' which were destroyed were regularly rebuilt;97 new 'cities' were
established, often specifically to provide shelter for a newly immigrant
community under imperial auspices; settlements were given the title of
'city' and the accompanying ecclesiastical hierarchy, when they became
important enough. 98 In other words, the fate of the Byzantine city is not
simply a question of economic resources, market potential, exchange
activity, or the administrative requirements of the state and the Church. It
is intimately connected also with the ideology of the Byzantine world and
its perception of self. Cities served not just as refuges or fortresses, markets
or administrative bases. They constituted an important element of Byzan-
tine self-identity. The continuity of site and settlement from the sixth and
seventh centuries right through to the end of the empire is surely related to
this. 99
97
Eirene rebuilt the t o w n s of Thrace after 7 8 4 (Theophanes, 4 1 7 . 6 - 1 1 ) ; emperors were
regularly involved in the reconstruction of city walls t h r o u g h o u t the empire (cf. Theo-
phanes, 4 8 1 . 9 for Ancyra, Thebasa, Andrasus).
98
E.g. Nicephorus, 6 6 . 1 1 ; Theophanes, 4 2 9 . 2 6 for Constantine V's construction of 'towns'
in Thrace for the Syrians and A r m e n i a n s transported from Anatolia. The city of Gordoser-
bon in Bithynia, w h i c h appears in the list of conciliar signatories for 6 8 0 , w a s probably a
settlement of Serbs from the great population transfer carried out u n d e r Constantine IV.
See Ramsay, Historical Geography, p. 1 9 7 and table. In general, see Ostrogorsky, 'Byzan-
tine cities in the early middle ages', 62f.
99
Hendy, Studies, pp. 90ff. a n d m a p s 2 0 - 3 . For a n important c o m m e n t on the changing
cultural a n d economic function of late antique towns, see n o w also J.-M. Spieser,
'L'Evolution de la ville byzantine de l'epoque paleochretienne a l'iconoclasme', in Hommes
et Richesses dans I'Empire byzantin I: IVe-VIF siecle (Paris 1989), pp. 97-106.
 CHAPTER 4

Social relations and the economy: rural

society

ESTATES AND LANDLORDS

Land, whether exploited by arable farming - cereals, vegetables, fruit and
so forth - or by pastoral farming, constituted the dominant means of
production in the Byzantine world, as indeed it does in all pre-industrial
cultures. The relationship between those who work the land and produce
the wealth from it, and those who own or control the land and the uses to
which it is put, is crucial for an understanding of how late Roman and
early Byzantine society worked. Even more so than in other areas,
however, the limited number of sources which can be usefully employed in
clarifying this aspect of early Byzantine history is a major hindrance. Apart
from the legal texts of the later sixth century, particularly imperial novellae
and apart from the late seventh- or early eighth-century Farmers' Law
(which is still very much debated with regard to its date, its origins and the
extent to which its precepts can be generalised for the whole empire), we
have to rely upon casual references in literary texts, the results of archaeo-
logical work on the economy of the urban settlements - referred to in the
previous section - and upon inferences based on the situation as it is
known in the earlier and later periods, and the possible logic of the changes
which are implied. In this section, I will begin by looking at the situation in
the later sixth century.
As we have seen, the state (in the form of the res privata, the domus
divinae and the fundi pathmoniales)1 along with the Church and the sena-
torial elite formed the most powerful grouping of landowners or landhold-
ers in the empire. The state especially, but other landowners also, sub-let
their lands to tenants who received a rent from the actual cultivators
themselves. The increasing use of emphyteutic leases to both larger land-
1
See for the difference, Karayannopoulos, Finanzwesen, pp. 72-80; JGR I, coll. 1, nov. 1
(p. 2.8-9) (a. 566 = Dolger, Regesten, no. 4); grouped together, however, under Tiberius
Constantine, see/GRI, coll. 1, nov. 12 (p. 2.6sq.) (a. ? = D61ger, Regesten, no. 67). On the issues
dealt with in this chapter see now M. Kaplan, Les hommes et la terre a Byzance du VV
au XT siecle (Byzantina Sorbonensia X) (Paris 1992).

125
126 Byzantium in the seventh century

owners and to smallholders, and its spread during the sixth century, is
especially important in this connection, as we shall see.2 Along with free
emphyteutic tenants, land was farmed for the most part by dependent
peasants of one category or another: coloni adscripticii and liberi, for
example.3
Until the sixth century, these seem to have constituted the largest group.
The latter were free persons restricted to their holdings for a period of up to
thirty years, after which they were permitted to leave; the former were
bound to their holdings hereditarily. The term colonus designated origi-
nally a free peasant farmer, then a free tenant of a holding, of equal legal
standing to the landlord. From the middle of the third century, it began to
be used not simply of one side in a contractual leasing arrangment, but of a
cultivator-tenant of dependent status. By the middle of the fifth century the
status of colonus was hereditary; and by the sixth century the majority
were regarded as effectively unfree as far as their mobility was concerned,
being classified as 'slaves of their land'. They could be released from their
obligations only by their landlord, and only under certain conditions with
regard to the land itself.4 The term adscripticius refers to the fact that coloni
of this type were entered in the land-tax register, along with their holdings,
under the name of their landlord. The coloni liberi, while free in person, and
free too to make wills and pass on and inherit property, were, from
Anastasius' time, also forbidden to leave their holdings; but they continued
to be entered in the tax-registers under their own names, paying land-taxes
direct to the state, rather than through a landlord.5 But this was in practice
2
See M.V. Levcenko, 'Materiali dlya vnutrennei istorii vostocnoi rimskoi imperii V-VI vv.' in
Vizantiiskii Sbornik pod. red. M.V. Levcenko (Moscow and Leningrad 1945), pp. 12-95, see
p. 64; note that Justinian expressly recommended that such lessees should be relatively
well-off (euporos): see C/I, 2.24/5 (a. 530).
3
In general on coloni, and the various sub-groupings, see Tinnefeld, Die fruhbyzantinische
Gesellschaft, pp. 45-55; Jones, LRE, vol. II, pp. 795-803; for a good general survey see also
Patlagean, Pauvrete economique, pp. 263-340 ('La Terre et la societe'); and A.H.M. Jones,
'The Roman colonate', Past and Present 13 (1958), 1-13 (repr. in The Roman Economy,
pp. 293ff.); P. Collinet, 'La Politique de Justinien a l'egard des colons', SBN 5 (1936); de Ste
Croix, The Class Struggle in the Ancient Greek World, pp. 226-59; M. Kaser, Das romische
Privatrecht, vol. II: Die nachklassischen Entwicklungen, 2nd edn (Munich 1975), pp. 143-9;
D. Eibach, Untersuchungen zum spdtantiken Kolonat in der kaiserlichen Gesetzgebung unter
besonderer Berucksichtigung der Terminologie (Cologne 1977), esp. pp. 47ff. For their status,
see R. Giinther, 'Coloni liberi und coloni originarii: einige Bemerkungen zum spatantiken
Kolonat', Klio 49 (1967), 267-71, and Eibach, Untersuchungen, pp. 132-204, critical of
Jones' traditional perspective.
4
See CTh. V, 17.1 (a. 332): coloni tied to the land; CTh. XIII, 10.3 (a. 357): land not to be sold
without its coloni ( = CJ XI, 48.2); CJ XI, 68.3: status of coloni hereditary; CJ XI, 52.1 (a.
393: the classical definition of a colonus as free in person, but 'slave of the land' on which
they were born; cf. CJ XI, 48.21/1 (a. 530): Justinian asks what differentiates a slave from
an adscripticius; cf. CJ VII, 24.4/1 (a. 531).
5
For adscripticii see CJ XI, 48.22 (a. 531): publici census adscriptio; for the liber colonus, see CJ
XI, 48.19; for the origins of the two categories, see Jones, LRE, vol. n, pp. 797ff.; but more
 Rural society 127

the only real difference between coloni liberi and adscripticii, for it seems
that the latter (contrary to the traditionally held view) could also take out
emphyteutic leases and act as legally independent and free persons. The
fact that their peculium, or personal property, and the rights and duties
attached thereto came under the authority of their landlords - traditionally
taken to be a sure sign of their servile status - has now been shown to be a
factor of their political-juridical position in respect of their landlords and
the estates of which they were a part. For such estates owed revenue to the
state and munera or civic burdens to the local municipality in whose
territory they were situated. From this point of view, the distraint (or the
possibility of distraint) by the landlord of the peculium of the peasants 'tied
to the land' functioned in the same way as the distraint by the res publica,
that is, the city, of the property of a member of the curial class who failed to
fulfil his civic liturgies or duties - munera. It served in effect as a security on
the returns from agricultural production, in which relationship the land-
lord functioned effectively as an agent of the city and thence, ultimately, of
the fisc. It must in addition be remembered that, if the adscripted colonus
had no freedom to leave his holding, neither had the landlord the freedom
to move or to expel him, or to increase the basic rent or tax imposition on
his holding. The relationship was regulated by the state to the mutual
benefit of both parties - in theory if not always in practice - and, of course,
and chiefly, for the securing of the state's revenue. The adscripticius thus
received what the ordinary tenant-farmer or freeholder did not, namely
security of tenure and the protection of state legislation.6
Alongside these two major groupings were free smallholders, often no
different in economic condition than the coloni, with whom they might
share a village and a community. Juridically and economically, of course,
they were of slightly higher status, being both free in person and free to
alienate their own properties; but since they were subject to pressure from
both the state on the one hand, for taxes, and from the more powerful
landlords around them on the other hand, their position was rarely secure.
There is evidence for communities of such freeholders in most regions of
the empire into the early seventh century, 7 and indeed Justin II expressly
forbade the curatores and other officials of the imperial estates throughout
especially D. Eibach, Untersuchungen zwn spdtantiken Kolonat in der kaiserlichen Gesetzgebung
unter besonderer Beriicksichtigung der Terminologie (Cologne 1977), esp. pp. 132-204,
critical of the traditional perspective set out by Jones.
6
See J. Gascou, 'Les grands domaines, la cite et l'etat en Egypte byzantine', TM 9 (1985),
1-90, esp. 22-7; and Goffart, Caput and Colonate, pp. 87ff., for the position of such coloni in
relation to their landlords and the estates to which they were bound.
7
See Jones, 'The Roman colonate', 294f. Note that the sixth-century legislation often treats
such freeholders as subject to the same economic conditions and difficulties as the remai-
ning smallholders - coloni - within the empire. Cf. JGR I, coll. 1, nov. 1 (p. 2.7sq.) (a. 566);
nov. 11 (p. 18.4sq.) (a. 575 = Dolger, Regesten, no. 40); see Jones, LRE, vol. II, pp. 778ff.
128 Byzantium in the seventh century

the empire to lay claim to the nearby villages and their lands, whether
these villages were of freeholders surrounded by civic lands or the estates of
senatorial landowners or of the Church. They were also forbidden to
exercise any form of patronage - patrocinium - over such communities.8
The extent to which this pattern of landownership and land-exploitation
survived into and beyond the seventh century is difficult to say. It has often
been argued that the civil strife under Phocas - followed, first, by the
Persian and then by the Arab onslaught - was responsible both for the
decimation of the senatorial landed elite in the provinces and the reduction
in the numbers of latifundia-style estates in the eastern half of the empire.9
There is, in fact, little evidence to demonstrate that this was the case and,
indeed, little theoretical justification - in terms of how Byzantine society
moved from the situation familiar from the sixth century to that which we
find in the ninth century and later - for assuming this. Partly, the
argument rests on methodological misapprehensions about the possi-
bilities of generalising from the late seventh-century Farmers' Law, which
does not mention such estates. But, as we shall see, there is no reason why
it should; and negative evidence is hardly a suitable foundation for such an
extensive explanatory edifice.10
Of the group of landowners referred to who dominate in the sixth
century, there is no doubt that the Church and the state continued to be
major landowners within the lands remaining to the empire after the
8
JGR I, coll. i, nov. 12 (p. 2O.33sq.). For patrocinium, see Tinnefeld, Die friihbyzantinische
Gesellschaft, pp. 36-44; Jones, LRE, vol. II, pp. 775-8; Brown, The World of Late Antiquity,
pp. 36-7; L. Harmand, Le Patronat sur les collectivites publiques des origines au Bas-Empire
(Paris 1957), pp. 427ff. and 448ff. Patronage functioned quite simply: a community
placed itself under the protection of a powerful person (at first, usually a military officer
with troops at his disposal, sometimes even a whole garrison; later also private persons,
such as a powerful landowner in the district), in order to obtain some protection from state
officials or tax-collectors, or to render assistance in some local feud or conflict with
another landlord. In return, the protector or patron received recompense in cash or kind;
but he could extract greater returns, through the peasants' mortgaging their holdings to
him, for example. Frequently, peasants found it simplest to hand over their land to their
new landlord and receive it back again with security of tenure and tenancy, as his coloni.
In this way, many originally independent and free smallholders became the tenants of
more powerful landowners, including the Church.
9
For example, Ostrogorsky, Geschichte, p. 112. Ostrogorsky's basic argument was that
these developments promoted the more-or-less complete replacement of the traditional
system of latifundia estates and the colonate by a patchwork of free peasant communities,
from among whom the soldeirs of the thema forces were enrolled, a process set in motion
by the Emperor Heraclius. See also Stein, Studien, pp. 157ff. This basic thesis was given a
slightly different nuance, although the general idea remained, by the Soviet historian
M.Ya. Siuziumov, 'Nekotorye problemy istorii vizantii', Voprosy istorii 3 (1959), 98-117,
according to whom the seventh and early eighth centuries witnessed the final transform-
ation from the slave mode of production to a proto-feudal state of development.
10
See for example, F. Winkelmann, 'Zum byzantinischen Staat (Kaiser, Aristokratie, Heer)',
in Byzanz im 7. Jhdt, pp. 161-288, esp. p. 198; Tinnefeld, Die friihbyzantinische Gesell-
schaft, pp. 98-9.
 Rural society 129

territorial losses of the mid-seventh century. The insistence that Church-

men - bishops included - must return to their communities (which they
had abandoned to avoid enemy attacks) in the records of the Quinisext
council of 692 suggests also that the Church was keen not to lose control
over its landed possessions in the more exposed provinces. 11 Bishops, who
represented the landed elite as well as the Church in their sees, were also
crucial to the running of the local administration, as well as to the welfare
of their congregations. It was often the bishop who was responsible in
directing municipal action regarding works of fortification and supplying a
city in times of danger, even extending to the administration of public
granaries and the provisioning of locally based troops. The departure of the
bishop might thus have drastic results for both local fiscal and civil, as well
as military, affairs, quite apart from his role in maintaining morale and
confidence in the state. 12
By the same token, the continued existence of the senate in Constantin-
ople, the power and the authority it exercised on occasion during the
seventh century, and its reinforcement through newcomers from the
imperial bureaucracy and administrative apparatus, meant that large
landed properties in all probability survived. The actual composition of the
senate, and the degree of continuity from the sixth century in terms of
families and specific estates, remains impossible to assess. From a detailed
analysis of members of the senatorial and administrative elite during the
seventh century, however, it seems that the older senatorial magnate
families gradually lose in power to the rising service aristocracy of the
Constantinopolitan establishment - in which the choice of individual
emperors in promoting those they favoured or judged worthy of a par-
ticular office or honour was crucial. 13 At the same time, however, there is
no reason to doubt that a large proportion of the old senatorial elite were
assimilated to the new service elite of the second half of the seventh
century; and whatever the origins of the military and civil officials of the
period, personal wealth could only be secured beyond one generation
through the acquisition of land. Landed property remained an essential
11
See Mansi XI, 952B-C (Quinisext council, canon 18).
12
Dagron, 'Le Christianisme dans la ville byzantine', 20ff.; Durliat, 'Les Attributions civiles
des eveques' 77-8, notes 32-4 and literature (for the sixth century; but the institutional
continuity is not to be doubted for the following period). For bishops responsible for
fortifications, see Gregory I, Ep. IX, 121.
13
See esp. the surveys of Winkelmann, Zwn byzantinischen Staat, pp. 185-219 and Quellen-
studien zur herrschenden Klasse von Byzanz im 8. und 9. Jahrhundert {BBA LIV, Berlin 1987),
pp. 143ff.; and the articles of R. Guilland on 'Les Patrices byzantins' in To/mos els
Mvr\yiT]v K.I. Afidi^rov (Athens 1960), pp. 11-24 (the period 602-68); in Hellenika 23
(1970), 287-98 (the period 668-717) and in B 40 (1970), 317-60 (the period
717-829). The last two are repr. in Guilland, Titres et fonctions de Vempire byzantin VIII
and IX (London 1976).
130 Byzantium in the seventh century

element in securing one's future and also in cementing one's position

within the establishment. Marriage into older and established landed
families on the one hand, and the purchase of land - from other land-
owners, including both Church and state in the form of perpetual leases,14
or from impoverished or threatened peasant communities - must have
made the continued existence of large-scale landed property a necessary
element in seventh- and eighth-century society. It should also be recalled
that it is the wealthier landowners or landholders who are able to survive
natural or man-made disasters, for they have greater resources and
reserves and hence greater economic flexibility. The importance of patron-
age, for example, and the effects of a drought or blight on the poorer
farmers is enough to demonstrate this. In addition, it is clear from the later
evidence that enemy raids had the most devastating effects on small-
holders, not on the large estate owners. The tenth-century semi-official
fiscal treatise dealing with the desertion of properties and the abondon-
ment of smallholdings names enemy attacks as one of the chief causes of
this. It was a development from which wealthier and more powerful
landowners could always take advantage.
Military personnel in particular - the thematic officers, for example -
will have been able to acquire land with relative ease, buying out impover-
ished peasants, arranging emphyteutic leases with absentee landowners,
imposing their 'protection' on local rural communities, simple extortion
through the use of their soldiers and, of course, taking over the property,
where it existed, of their soldiers themselves. Military officers represented
the only group which was in a position to maintain its interests under the
sorts of conditions prevailing in areas regularly threatened by hostile
action. They had not only the resources (the authority of state power) but
the armed force necessary to impose their will on any opposition, and also
to protect the area in which they were established. The 'protection' offered
by military personnel must often have been welcomed, but as in an earlier
period, patrocinium tended ultimately to favour the patron at the expense of
the client.
Considerable areas of south and central Asia Minor, for example, must
have been effectively lost to the control of those who had extensive lands
there, whether lay or Church, at times. The Quinisext council, referred to
already, explicitly mentions the abandonment of many congregations by
the clergy due to fear of the Arab raids. This seems to include bishops also,
among the most important landlords, albeit acting on behalf of the Church,
14
The Church, for example, was permitted to sell off land both to the fisc and to private
persons under specific circumstances - as when it had debts to make good. See Justinian,
Nov. 46, proem; 1; 2; 6.2; 120.4; the Church could also sell land in order to raise funds for
the release of prisoners from the enemy. See Justinian, Nov. 65 (a. 538).
 Rural society 131

in any district. If bishops had left, it is unlikely that many lay landowners
would have stayed to be killed or captured. But the abandonment of their
lands made possible both the imposition of authority from a new source
and the casting-offof the ties between landlord and tenants' obligations on
the part of the peasantry. I will discuss this and related developments
below. 15
One of the most frequently cited texts in this debate is the ninth-century
Life of St Philaretus the Merciful, a Paphlagonian landowner of some
means who gave up his considerable wealth in charitable acts and who
was compelled to surrender properties he could no longer adequately
exploit, to the advantage of his neighbours, both richer landowners and
the villagers of the area. 16 Even taking into account the probable exagger-
ations in the number of animals and properties Philaretus is supposed to
have possessed (designed to emphasise the extent of his generosity and
piety), it is clear that Philaretus stood for a landowner of considerable
wealth, presumably a character not unfamiliar to those who read or heard
the hagiographer's account. His livestock included 600 bullocks, 100 pairs
of oxen, 80 pack-horses and mules/donkeys, 800 horses, 12,000 sheep;
while he owned 48 large parcels of land distributed over a wide area,
including also land in the areas of Pontus and Galatia. He was born,
according to the Life, in 702; and it is reasonably clear from the description
that his estates were for the most part inherited from his family.17 His
neighbours included other landowners, 18 as well as smallholders; and it
seems equally logical to infer that the history of such estates reached back
well into the seventh century. Philaretus himself was regarded by the small
farmers and peasants around him as a powerful landowner; while it is
clear from the rest of the narrative that the most powerful and wealthy
elements of the ruling class, both in the provinces and in Constantinople,
regarded him as their equal. 19
15
See Haldon, 'Some considerations', 97f., and 'Some remarks on the background to the
iconoclast controversy', BS 38 (1977), 161-84, see 174f. For the relevant canons of the
Quinisext, see Mansi XI, 945B-D (canon 8), 951B-C (canon 18), 960C-E (canon 37),
961A-C (canon 39). For the tenth-century evidence, see the Fiscal Treatise (ed. Dolger),
p. 116.2-3; 118.42-119.1; and cf. 119.6-7.
16
M.-H. Fourmy and M. Leroy, 'La Vie de S. Philarete', B 9 (1934), 85-170; and see also
L. Brehier, in B 1 (1924), 177-90.
17
See J.W. Nesbitt, 'The Life of St Philaretos (702-92) and its significance for Byzantine
agriculture', Greek Orthodox Theol. Review 14 (1969), 150ff.; H. Evert-Kappesowa, 'Une
grande propriete fonciere du VIIIe siecle a Byzance', BS 24 (1963), 32-40.
18
VitaPhilareti, 117.
19
See H. Kopstein, 'Zu den Agrarverhaltnissen', in Byzanz im 7. Jahrhundert, pp. 1-72, see
63f., although historians are divided as to whether Philaretus was a landed magnate or
merely the son of a wealthy peasant family which made good. See, for example, M. Loos,
'Quelques remarques sur les communautes rurales et la grande propriete terrienne a
Byzance, VIF-XIe siecles', BS 39 (1978), 3-18, see 9ff.
132 Byzantium in the seventh century

That large private estates existed, therefore, side by side with the estates
of state and Church in the late seventh and eighth centuries is clear. There
seems no reason to doubt that they had continued to exist, and to come into
being, throughout the seventh century. But Philaretus' own background
and that of his family remains unknown, and a question mark must remain
over the issue of the degree of physical continuity of such estates.
The text of the Vita Philareti also refers to the large estates of the Church
as something that is taken for granted in provincial life; and it refers to
villages of independent peasants, who own and farm their own holdings
and who pay taxes directly to the state. 20 And this brings us to the next
question, namely, what was the extent of such village communities within
the empire? And what is the significance of the so-called Farmers' Law for
this question?

VILLAGES AND TENANTS

The Farmers' Law is usually thought to date from the last twenty years or
so of the seventh century or the first two decades of the eighth century. It
appears to be a privately sponsored compilation, although this does not
exclude the possibility of its having been consulted by officials with judicial
authority, and it was designed to regulate property-relations and tenure
within an agricultural-pastoral community - a village - the majority of
whose occupants seem to have been independent smallholders, but among
whom were to be found various categories of lessee and tenant, as well as
domestic slaves and wage-labourers or journeymen. Its legal foundation is
at least Justinianic, and some aspects may well be pre-Justinianic. The date,
however, does remain very uncertain, and it may well be a compilation of a
later period. In this respect, arguments based upon it must be treated
carefully, although it is possible to argue that the relationships described
therein can be applied with hindsight to the period with which we are
concerned.
Now it has been argued by several historians that the Farmers' Law is
good evidence for (a) the dominance of such free village communities in the
Byzantine world at this time (and at the expense of the earlier dominance of
large senatorial latifundia); and (b) the strong influence of Slav social
institutions, in particular the obscina or commune, on Byzantine peasant
society, a result of the settlement by the state of large numbers of Slavs from
the Balkans in Asia Minor. 21
20
Vita Philareti, 117; for the Church estates, see ibid., 15.26; 157.5 (Philaretus' grandsons
handed their lands over to a convent).
21
For the Farmers' Law (nomos georgikos) see W. Ashburner, 'The Farmers' Law', JHS 30
(1910), 85-108; 32 (1912), 68-95 (text, translation and commentary). There is a
 Rural society 133
The first argument is based on the existence and widespread dissemi-
nation during the medieval period of the Farmers' Law itself and on
evidence from the ninth and tenth centuries which suggests that commu-
nities of peasant proprietors subject directly to the fisc formed a central
element in the rural population of the empire.22 The second argument is
based on the known introduction of Slav colonists into the empire and on a
failure to consider adequately other elements which might have contri-
buted to this apparent increase in peasant proprietorship which are jnter-
nal to late Roman and Byzantine social and economic relations.23
In fact, as has been shown, the growth in the numbers of such commu-
nities, their increased importance to the economy of the state and the
related changes in the basic system of taxation - from the late Roman
capitatio/iugatio to the middle Byzantine land- and hearth-tax - can all be
explained within the context of developments internal to the dynamic of
relations of production in the late Roman social formation, although
external factors - warfare on the one hand, immigration on the other -
certainly played an important role.
We have seen that those who worked the land can be divided into a
number of categories, both from the legal and from the economic point of
view. But the proportion of free tenants paying both taxes and rents, of
coloni liberi who were effectively treated as free tenants after they had
cultivated their holdings for thirty or more years, and of small-scale
wide-ranging discussion. For the best recent survey of the problems and the literature, see
Kopstein, 'Zu den Agrarverhaltnissen', 40ff.; P. Lemerle, 'Esquisse pour un histoire
agraire de Byzance: les sources et les problemes', RH 219 (1958), 32-74, 254-84; 220
(1958), 42-94; see 219 (1958), 32-74; and the revised, English, translation: The
Agrarian History of Byzantium from the Origins to the Twelfth Century. Sources and Problems
(Galway 1979), pp. 32ff. For an attempt to date part of the contents of the Farmers' Law
to sources of the pre-Justinianic period, see N. Svoronos, 'Notes sur l'origine et la date du
Code Rural', TM 8 (1982), 487-500. Svoronos' analysis suggests that many features of
the code or law predate the codification of Justinian I and are drawn from texts of an
earlier period which were available to the compilers of the later seventh century. But this
has been heavily criticised and shown to be based on dubious methodological suppositi-
ons. See L. Burgmann, '1st der Nomos Georgikos vorjustinianisch?', Rechtshistorisches
Journal 1 (1982), 36-39.
For the arguments over Slavisation, see for example E.E. Lipsic, Byzanz und die Slawen.
Beitrdge zur byzantinischen Geschichte des 6. -9. Jahrhunderts (Weimar 1951), pp. 30ff. with
literature; but see the detailed discussion of Ostrogorsky, Geschichte, pp. 113-14 and note
3. A more recent review of the problem is presented in R. Poptodorov, 'Sledi ot slavjans-
koto pravo v'v Vizantijskoto', Izvestija Crkov Noistoric. i Arkhiv. Institut, Soflja II (1984),
126-37, which argues that both the Ecloga and the Farmers' law show traces of a
barbarisation and Slavisation of classical Roman law; although, as with the earlier
arguments, the methodological grounds upon which such conclusions are based are not
very satisfactory.
22
For text, tradition and dissemination of the Farmers' Law, see also the detailed comments
of Ostrogorsky, Geschichte, p. 75 note 10; and J. Karayannopoulos, 'Entstehung und
Bedeutung des Nomos Georgikos', BZ 51 (1958), 357-73.
23
See Ostrogorsky, Geschichte, pp. 113f. and note 3.
134 Byzantium in the seventh century

peasant emphyteutees, paying a fixed and possibly very low rent (and often
holding their leases, which were transferable, in perpetuity and hereditari-
ly) 24 seems to have become increasingly important during this period, as
the relevant imperial legislation demonstrates. Just as significantly, the
later Byzantine term for a dependent peasant, paroikos, is to be found in this
sixth-century legislation,25 referring, however, to a lessee of land, a locator
(or colonus, in one of its original meanings). This points the way to the
gradual assimilation of all three categories of tenant smallholder into a
single body of tenants, paying a rent to their landlord and tax to the state,
bound to their properties according to late Roman law or to their lease
(although, as we have seen, emphyteutic lessees could cede or sell the land
- that is, the lease - to a third party, and were often regarded, at least in
respect of non-Church lands, as the possessor and not simply the locator, or
tenant).
It is important in this connection to note that the Ecloga of Leo III and
Constantine V places a great deal of emphasis on this particular form of
contractual relationship, which again seems to point to its importance at
this period.26
In the first half of the seventh century, the Life of Theodore of Sykeon
portrays a western Anatolian society of free peasant smallholders and
communities, in which the community clearly plays a role as a corporate
body; the Life refers likewise to peasant farmers fleeing from the tyranny of
their landlords 27 and to the existence of large estates belonging to the
Church of the nearby city of Anastasioupolis.28 The late seventh-century
Farmers' Law similarly refers to the village community - TJ TOV x^ptov
- and to the communal lands belonging to the village as a

24
See above; and P. Lemerle, The Agrarian History of Byzantium from the Origins to the Twelfth
Century. Sources and Problems (Galway 1979), p. 2 5 .
25
See Justinian, Nov. 7, proem, 1 (a. 535); Nov. 1 2 0 . 1 proem (a. 544); C/I, 2.24. proem; I,
3 4 . 1 . The term occurs earlier for coloni in a variety of documents, see A. Deleage, La
Capitation au Bas-Empire (Paris 1945), pp. 182ff.
26
See in particular the c o m m e n t s of M. Kaplan, 'L'Exploitation p a y s a n n e byzantine entre
l'antiquite et le m o y e n age (VI e -VIII e siecles): affirmation d ' u n e structure economique et
sociale', in V. Vavfinek, ed., From Late Antiquity to Early Byzantium (Prague 1985),
pp. 1 0 1 - 5 , see 102f., and 'Remarques sur la place de l'exploitation p a y s a n n e dans
Teconomie rurale', in Akten des XVI. International Byzantinisten-Kongresses, U, 2 (Vienna
1982), pp. 1 0 5 - 1 4 ( = JOB 32), see p. 107f. with references. Kopstein, 'Zu den Agrarver-
haltnissen', p. 6 6 , argues t h a t paroikos relates only to ecclesiastical lands a n d their leases.
In fact, as Kaplan suggests, the term begins with this general meaning, but h a d by the
early n i n t h century b e g u n to acquire the m e a n i n g later attributed to it (a dependent
peasant) as a result of the rapid spread of this system of leasing land, a n d its replacement
of the colonate. See Ecloga (ed. B u r g m a n n ) , 1 2 . 1 - 6 .
2
7 Vita Theod. Syk., 147.49sq. (fleeing colonus); 98.1sq.; 114.1sq.; 115.2sq. etc. (free pea-
sants); 1 4 3 . 1 (the village community).
28
Ibid., 76.1sq.; and cf. 3 4 . 6 - 7 for the lands of the Church of Helioupolis in Bithynia.
 Rural society 135
whole.29 There seems little reason to doubt that the Farmers' Law, com-
piled on the basis of extracts from the Codex lustinianus and the customary
law of part of the empire and designed to regulate the institutions of the
independent peasant village, represents a relatively ancient tradition both
socially and economically.30 There is no reason to doubt either that such
communities existed in an unbroken tradition throughout the seventh
century; for although the point is still debated, it seems that the Farmers'
Law presents nothing in its regulations or assumptions that is not already
present in the late Roman village community.31 But the terminology has
undergone a certain evolution: the term georgos (7ea>p70s), which could
refer to both coloni adscripticii or liberi, or yet again afreesmallholder or a
lessee (locator, e|jup\n"evTT|s) in the fifth and sixth centuries, occurs in the
Farmers' Law to describe the members of the village. Since the latter
clearly hold land, over which they have sole jurisdiction and which they
are free to leave, the term can refer to the last three categories only.32
The Farmers' Law describes the relationships which might come before a
court within the village community; it makes it clear that social differenti-
ation was a normal element of the community: wealthier and poorer
smallholders, and hired labourers, are mentioned, for example. It describes
an economy of cereal, fruit and vine cultivation, of sheep and cattle raising;
it also frequently refers to the sub-tenanting of holdings among the vil-
lagers, and to the fact that some villagers might also be the tenants of
larger landowners - individuals or institutions - from outside the village.33
29
See in particular on this question the old but still very valuable work of A.P. Rudakov,
Ocerki Vizantiiskii kul'turipo dannim greceskoi agiografii (Moscow 1917 and London 1970),
pp. 174ff. For the village community, see the Farmers' Law, art. 81.
30
See the literature in notes 117 and 118 above; and St. Maslev, 'Die soziale Struktur der
byzantinischen Landgemeinde nach dem Nomos Georgikos', in F. Winkelmann and
H. Kopstein, eds., Studien zum 7. Jahrhundert, pp. 10-22, see 10-12.
31
See the summary of the debate in Maslev, 'Die soziale Struktur', and Svoronos, Notes sur
Vorigine et la date du Code Rural, esp. the final comments on p. 500.
32
See the evidence summarised by Kopstein, 'Zu den Agrarverhaltnissen', pp. 41f. For the
wide application of georgos - originally limited just to coloni - during the later sixth
century, see P. Lemerle, The Agrarian History of Byzantium from the Origins to the Twelfth
Century (see note 21 above), pp. 20ff.
33
See the Farmers' Law, esp. art. 9 and 10 for the tenant {mortites) of the (larger) landowner
or landholder. In general on the content of the Farmers' Law, see Kopstein, 'Zu den
Agrarverhaltnissen', pp. 4 1 - 8 and 49-53; Maslev, 'Die soziale Struktur' (note 30 above);
and H. Kopstein, 'Zu einigen Aspekten der Agrarverhaltnisse im 7. Jahrhundert (nach den
juristischen Quellen)', in Studien zum 7. Jhdt., pp. 23-34, see 29fT. The morti arrangement,
which involved a rent of only ten per cent on the income from the land, seems to represent
in practice, if not in name, the earlier emphyteutic lease, designed to enable the landowner
to keep land under cultivation by offering a relatively attractive arrangment with tenants.
As we have seen, such leases were increasingly employed through the sixth century, see
below, and note that the Ecloga knows only emphyteutic and ordinary leases. The term
morti does not occur, and it is tempting to assume that the morti arrangement was the
equivalent of the emphyteusis.
136 Byzantium in the seventh century
It refers to the common land of the village; and it refers to the practice of
bringing abandoned land back into cultivation and dividing it among
(some) members of the community. The common grazing of livestock on
the fields after the harvest was also a feature - normal in many peasant
societies, of course.34
The relationship between such a community and the state is referred to
only obliquely: when a peasant is unable to gain a livelihoodfromhis land,
whether he has rented a part of his holding to another (being unable to
farm it himself) in return for a portion (usually a half) of the resulting yield;
or whether he has rented all of his holding, taking up work for others on a
waged basis; or whether he hasfled,unable to pay his taxes and his debts,
then the burden of taxation was transferred to those who could maintain
his plot. If no one was in a position to do this, the whole community was
made responsible for the taxes owed, although possibly - as in the ninth
and tenth centuries - a reduction in the total owing was made in order to
avoid the impoverishment of the remaining members of the village. If, after
a certain (unspecified) time, the land was still unoccupied, it was - along
with other untenanted holdings - formally confiscated by the state and
redistributed to members of the village collectivity.35
I will deal with this question in more detail in chapter 5. But there is no
explicit evidence in the Farmers' Law that communalfiscalsolidarity was
firmly established.
The Farmers' Law therefore seems to represent a village community not
too different from those known from Anatolia, Syria and other regions of
the empire in the sixth century. The differences which have been noted
provide no real objections to continuity of basic structure and social
organisation. For example, there is evidence in the late Roman metrokomiai
(or extended villages) of Syria and Egypt for a regular redivision of
holdings, an absence of family property and less emphasis on the rights of
individuals to occupy and cultivate abandoned village land. In theory -
although the legislation itself hints at the reality being rather different -
such community land could not be held by an outsider. The village
community of the Farmers' Law, in contrast, has both heritable family
For social differentiation in this sort of community, note the opposition between the
'landowners of the village' and the ordinary Inhabitants' (tenants, leaseholders etc.) of
the village in the Vita Theod. Syk., 116; 118; for the leading men' of the village, see also
ibid., 114; and Vita Philareti, 1 3 7 . 2 3 . Cf. Loos, 'Les Communautes rurales', 9 with
references; Patlagean, Pauvrete economique, p. 2 6 5 .
34
For example, P. Stirling, 'A Turkish village', in T. Shanin, ed., Peasants and Peasant
Societies (Harmondsworth 1971), pp. 3 7 - 4 8 , see 40ff.; D. Thorner, 'Peasant economy as
a category in economic history', ibid., pp. 2 0 2 - 1 8 ; R.H. Hilton, Bond Men Made Free
(London 1973), see pp. 71f.
35
See Kopstein, 'Zu den Agrarverhaltnissen', p. 53 note 1; Lemerle, Agrarian History,
pp. 41-5.
 Rural society 137
property and outsiders in the village; it expressly permits the cultivation of
abandoned land by neighbours; and it envisages no mass redivision of
holdings. But the late Roman evidence is specific to Egypt and Syria, while
the Farmers' Law represents a geographically and chronologically very
different context. And the evidence of the Life of Theodore of Sykeon does
suggest an Anatolian village economy and structure very similar to that
described in the Farmers' Law, but for a period one hundred years earlier.
In discussing the nature and evolution of rural communities, people who
argue for and against continuity must also bear traditional regional
variations and cultural differences in mind, as these examples suggest.36
The juridical character of the village community seems to have evolved
also during the period from the sixth century to the later seventh or early
eighth, and these are developments which are important for our under-
standing of the role played by village communities in the social relations of
the seventh-century Byzantine world. Central to this question is the nature
of the relationship between the state, the cities and the countryside.
This change in juridical status can be detected in the term used to
describe and to define the village. In the late Roman period, certainly by
the sixth century, two terms of equivalent value - kome and chorion, the
former the traditional word for a village of small proprietors, the latter for
an inhabited holding or part of an estate within a singlefiscalunit - were
used to describe village communities.37 The reasons for the semantic
equivalence of the two, as has been demonstrated, lie in the fact that, as the
development of the colonate and smallholding came to dominate on large
36
See especially Kazdan, Derevniya i gorod, pp. 3 Iff., for the metrokomiai. Lemerle, Agrarian
History, pp. 7f. believes that the metrokomia with its collective liability for taxes was
already in existence in the fifth century as the standard form of non-dependent village
throughout the empire. M. Kaplan, 'Les Villageois aux premiers siecles byzantins (VI e -X e
siecle): une societe homogene?' BS 4 3 (1982), 2 0 2 - 1 7 (see 206f. and note 31) argues in
contrast that it was a limited phenomenon occurring in some eastern districts only.
Partly, the debate revolves around the technical meaning of the terms 6\L6XI\V(TOV and
6JJL68OV\OV in respect of the allocation of deserted lands (epibole). But as Jones, LRE, vol.
II, p. 8 1 5 note 105 points out, the terms refer to fiscal census units, not necessarily
communities in the collective sense, thus supporting Kaplan's contention.
As far as concerns the area to which the Farmers' Law was originally intended to apply,
it should be noted that pastoral activity occupies more than fifty per cent of the text's
interests. Regulations concerning livestock, especially cattle, outnumber those dealing
with agricultural activity. The Anatolian plateau, and especially the central and eastern
sections, present themselves as obvious candidates. But this remains still in the realms of
hypothesis.
37
Original sense of kome: Libanius, Discours sur les patronages, ed. L. Harmand (Paris 1955),
15 (cap. 4) and 17 (cap. 11). For a comment, see G. Dagron, 'Entre village et cite: la
bourgade rurale des IV e -VII e siecles en Orient', Koivavia 3 (1979), 2 9 - 5 2 . For the
original sense of chorion: Digest X, 1.4/5 (168); and for a comment, M. Kaplan, Les
Proprietes de la Couronne et de VEglise dans Vempire byzantin (V -VIe siecles) (Paris 1 9 7 6 ) -
e

for kome = vicus, chorion = fundus. See also Kopstein, 4Zu den Agrarverhaltnissen',
pp. 56ff.
138 Byzantium in the seventh century
estates (as opposed to the dominance of agricultural slavery), the small-
holders and tenants themselves usually dwelt in village communities on
their lands. As the social and juridical differences between free peasant
smallholders in komai on the one hand and dependent peasants on the
chorion or fundus of their landlord on the other, were gradually ironed out,
so the technical difference between the two terms vanished. Chorion
eventually predominates, both because it reflects an equivalent semantic
use of the term, but also because it bore afiscalsignificance, representing
as it did in origin a singlefiscalunit, the property of a landlord occupied by
his tenants.38 As the cities ceased to function as the administrative centres
for the fiscal assessment of their regions, this role was transferred to the
smaller, but more immediately relevant, unit of the village. Chorion, with
its original fiscal significance, as well as its meaning of village, replaced
kome. But the development was a gradual one. Even in the Farmers' Law
the term chorion, while referring also to the village, still bears a wider
significance.
These changes in terminology are important, however. For while it
seems clear that the village community described in the Farmers' Law is in
itself not new, it is clear that such villages now represented a more
important element in the totality of the relations of production, especially
with respect to the state and its revenue-raising apparatus. The compi-
lation of the Farmers' Law and its wide dissemination themselves reinforce
this view: both in Byzantine society in general and in the eyes of the state
(and the officials who exercised judicial power in the provinces) the village
community was, or was becoming, an economic and social element of
much greater relative importance than had hitherto been the case.
Such communities existed throughout the later Roman world, and
indeed the nucleated settlement was the usual form of settlement outside
the cities, whatever the legal status of its inhabitants.39 The qualitative
change in their role does not necessarily imply that there were more of
them. What it does suggest is that the village, and its individual landhold-
ers, were becoming the key element in the state's administration of its
revenue-collection, in contrast to the later Roman period, when it had been
38
This development has been admirably summarised by Kaplan, 'Les Villageois aux prem-
iers siecles byzantins', 2O3ff.; see also Patlagean, Pauvrete economique, pp. 241f.
39
For the settlement-pattern and the nucleated village, see Patlagean, Pauvrete economique,
pp. 241ff.; Dagron, 'La bourgade rurale'; M. Kaplan, 'Quelques remarques sur les paysages
agraires byzantins (VIe siecle - milieu XT siecle)', Revue du Nord 62 (1980), 156ff. On
social differentiation, see note 33 above; for an example of a village of tenants (locatores,
emphyteutai, coloni liberi - albeit not specified so exactly) see Vita Theod. Syk., 34, 76, 162
(villages on the estates of a landowner from Ankara). See esp. N. Svoronos, 'Sur quelques
formes de la vie rurale a Byzance, petite et grande exploitation', Annales 11 (1956),
325-35 (repr. in Etudes sur Vorganisation interieure, la societe et Veconomie de VEmpire
Byzantin II (London 1973).
 Rural society 139
the landlord and the city who had been the chief intermediaries. The
evidence from the later period, when the village or chorion as afiscalunit
and as a juridical entity played the key role, demonstrates the results of this
evolution.40 But the terminological changes evidenced by the increasing
use of chorion for village and then fiscal community, which begins in the
sixth century and proceeds throughout the seventh, show that the social,
economic and administrative shifts which this development represents
were already under way.
The village community of the Farmers' Law is thus a traditional element
of the rural economy. Indeed, the degree of social differentiation and
dependence presented in the Farmers' Law suggests that such communi-
ties already had a long history. The important question, however, is how
widespread such communities were in the Byzantine world. In the
Farmers' Law, it is clear that the dominant element seems to be the
independent smallholders who either own their land or rent it on a more or
less permanent basis - the last point is hypothetical, but is supported by the
context which we have described for the period. As we have also seen,
Philaretus possessed a number of holdings which will have been farmed by
his own tenants or by hired labourers. From later sources it is clear that
such holdings will normally have been part of other village communities
which had come under the sway of the landlord in question - the Farmers'
Law already presents some evidence of this process, which was itself
nothing new. According to the Vita Philareti, both the local big land-
owners and some of the villagers themselves were keen to occupy and
cultivate the lands which Philaretus himself was unable to keep up, again
a procedure implicit in the provisions of the Farmers' Law.41 Together, the
Farmers' Law and the Life of Philaretus show that both independent
peasant village communities and large estates with interests in the sur-
rounding villages - as well as, presumably, villages of a greater dependent
status - were an integral part of the rural economy of later seventh- and
eighth-century Byzantine society.
40
For taxation in the tenth century (assessment, application, extraction), see F. Dolger,
Beitrdge zur Geschichte der byzantinischen Finanzverwaltung besonders des 10. und 11.
Jahrhunderts (Byzantinisches Archiv IX, Leipzig 1927 and Hildesheim 1960): G. Ostro-
gorsky, 'Die landliche Steuergemeinde des byzantinischen Reiches im X. Jahrhundert',
Vierteljahresschrift fur Sozial- und Wirtschaftsgeschichte 20 (1927), 1-108; Ch.M. Brand,
'Two Byzantine treatises on taxation', Traditio 25 (1969), 35-60; J. Karayannopoulos,
'Fragmente aus dem Vademecum eines byzantinischen Finanzbeamten', in Polychronion.
Festschrift Franz Dolger zum 75. Geburtstag (Heidelberg 1966), pp. 317-33; N. Svoronos,
'Recherches sur le cadastre byzantin et la fiscalite aux XP-XIIe siecles: le cadastre de
Thebes', BCH 83 (1959), 1-166 (repr. in Etudes III).
41
For proasteion, see Lemerle, Agrarian History, and the Fiscal Treatise (ed. Dolger, Beitrdge);
see esp. M. Kaplan, 'Les Villageois', esp. 214, for the social differentiation in the Vita
Philareti.
140 Byzantium in the seventh century
The increased importance of such independent communities - as impli-
citly evidenced by the Farmers' Law and as is made clear from later
evidence (in the tenth century it is taken for granted that they represent the
central element in the state'sfiscaloperations) - may suggest also that the
number of such communities increased during this period. Some of the
possible causes of such an increase have been argued at length: the
abandonment by landlords of their country estates and the consequent
assertion by the peasants of their independence;42 the immigration of large
numbers of Slav settlers with their community structure and organi-
sation;43 and the growing independence of peasant smallholders with
perpetual, or long-term, heritable leases, paying low and fixed rents to
landlords who may often have been permanently resident away from their
estates and properties.44
There is little reason for doubting that the Slav immigrants, having been
settled by the state, constituted communities which held and cultivated
land to maintain themselves. Justinian II's explicit intention was to
promote the availability of soldiers from this source, a point on which I will
say more in chapter 6.45 But it seems unlikely that these transfers of
population affected the social structure of the rest of the empire in such a
way as to bring about administrative changes in the fiscal system and at
the same time promote a sudden development of free peasant communities.
Two fundamental causes seem to be operating. On the one hand, a
development which we have already noted, the change in emphasis in the
mode of exploitation of large properties: the gradual weakening of the
adscripted colonate in favour of long-term and often heritable leases on
state, Church and private lands tended to reduce the need for estate-
owners to supervise either directly or indirectly the production process on
their lands. As long as rents were collected, interference was unnecessary;
and anyway the producers were responsible under the terms of the leases
for the state taxes. This reduced both the administrative costs and the fiscal
obligations of landowners to the state. It removed the need for bailiffs,
except as rent-collectors. But it also weakened the landlords' direct control
over their property, while reinforcing the relationship between producers
and state. The juridical differences in status of the various types of producer
- coloni, leaseholders, smallholders and peasant proprietors - will thus
42
See Kopstein, 'Zu den Agrarverhaltnissen', p. 59, basing her argument on the fact that
evidence from the canons of the Quinisext in 692 for clerics fleeing their cities might also
apply to secular landowners.
43
Kopstein ibid, and literature; Charanis, 'Ethnic changes', 42f. and 'Transfer of population',
143.
44
See above, pp. 134f.
45
See Theophanes, 347f. (A.D. 658), 364 (A.D. 689); see Charanis, 'Transfer of population',
143f.
 Rural society 141
have become less significant to the fiscal requirements of the state, less
relevant to the process of production and less relevant to the landowners
themselves. The long-term cumulative result must surely have been a very
considerable increase in the numbers of communities subject directly to the
fisc, albeit made up of persons of very varied legal status. Most importantly,
the great majority will have possessed freedom of movement and have
been, in effect, possessores, even if not owners, of their holdings, whatever
their original condition. The peasants of the Farmers' Law, as well as those
in the Life of Theodore of Sykeon and those referred to in the Life of
Philaretus, must not be taken to represent a uniform body of free peasant
proprietors. They represent farmers with freedom of movement, freedom to
transfer their land or to transmit it to their heirs. All of these provisions
were possible within the terms of the normal emphyteutic lease, as well as
straightforward proprietorship.46
On the other hand, we are probably dealing also with a change in
emphasis - reflected in the sources from the Farmers' Law on - in
government fiscal policy. The cities could no longer - and had been for
some time unable to - cope with the problems of takingfiscalresponsibility
for their territoria. State officials had taken over the supervision and
administration of these tasks since thefifthand sixth centuries. As we have
also seen, landlords - estate owners - no longer played such a crucial role
in this context either. The state had, as a result, to concentrate on the level
at which wealth-production actually took place; in other words, on the
land and the communities which farmed it. The centres of production were
the villages, and these now replace the towns or cities in the fiscal
administrative structure of the early Byzantine state. In a much more
significant way than before, the village comes to occupy a central position
in the society and the administration of the empire from the later seventh
century onwards.

TAXATION AND THE LAND

The basic taxes had been assessed in the sixth century on the capitatio/
iugatio equation, according to which land was taxed relative to its produc-
tivity when under cultivation (or otherwise exploited), and the agricultural
producers (or animals grazing on pasture land) were taxed only in associ-
ation with productive land.47 Supposedly, the last reference to this system

46
See the remarks of M. Kaplan, 'Remarques sur la place de l'exploitation paysanne', 106f.
with literature, and the comments on the rise of emphyteutic leasing in the sixth century
above and in chapter one infra.
47
See chapter 1.
142 Byzantium in the seventh century

occurs for Sicily during the first years of the first reign of Justinian II; 48
between then and the reign of Nicephorus I (802-11) a major change took
place, when tax was no longer raised on the basis of the combined
assessment of capitation and iugation, but by separate assessments, the
kapnikon, or hearth-tax; and the synone, or land-tax. The former was, in
effect, a tax on the household-property of the adult members of a house-
hold; the latter was a tax on productive land.
It has further been argued that a change took place at this time in the
methods by which the state tried to maintain the continued productivity of
the land. Abandoned or otherwise uncultivated holdings in the later
Roman period had, from the fourth century at least, been transferred to the
neighbouring landowners or landholders, who were made responsible for
the taxed normally pertaining to it. This system, referred to as the epibole
(ton aporori), or adiectio sterilium, had by the later ninth century been
replaced by a different system, in which the state intervened directly to
exempt abandoned properties until the owner or his/her heirs could bring
it back into cultivation again. If this had not happened within a period of
thirty years, then the state simply took it out of its original register and
fiscal district and gave it to a new owner.
Now, in spite of the arguments of several scholars, the old system seems
still to be in operation in the Farmers' Law, although not explicitly referred
to as epibole. Thus, while there was a change in terminology, the new
system does appear only from the second half of the ninth century. 49
It has generally been assumed that the abandonment of the older system
of assessment based on the capitatio/iugatio formula - which reflected the
needs of the state to ensure that land was cultivated in order to secure its
revenue - demonstrates that the lack of manpower which gave rise to it no
longer existed; in other words, that manpower was no longer a problem.
And it has been argued further that this is because of the growth in the
number and extent of communities of free peasants, and in particular the

48
See E. Stein, 'Vom Altertum zum Mittelalter. Zur Geschichte der byzantinischen
Finanzverwaltung', Vierteljahresschrift fur Sozial- und Wirtschaftsgeschichte 21 (1928),
158-70, see 150 and 152; G. Ostrogorsky, 'Das Steuersystem im byzantinischen Alter-
tum und Mittelalter', B 6 (1931), 229-40, see 237; the references are in the accounts of
the Lives of Popes John V and Conon (Liber Pontiflcalis, I 366.8-10; 368.19-369.2).
49
See Haldon, Recruitment and Conscription, pp. 52f.; Lemerle, 'Esquisse', RH 219, 60f.,
263f.; Agrarian History, pp. 46f.; and esp. J. Karayannopoulos, 'Die kollektive Steuerver-
antwortung in der friihbyzantinischen Zeit', Vierteljahresschrift fur Sozial- und Wirtschafts-
geschichte 43 (1956), 289-322; Kopstein, 'Zu den Agrarverhaltnissen', p. 47 and note 4.
For the first reference to the kapnikon, see Theophanes, 486.29-487.1; and Theophanes
cont. 54.4 for the rate at which it was raised (2 miliaresia per household). In general, see
Ostrogorsky, 'Die landliche Steuergemeinde', 5Iff.; Dolger, 'Beitrage', 52 (but with a
different view on its origins from that of Ostrogorsky).
 Rural society 143
drafting in of large numbers of Slavs, which made such a legislative tying
of manpower to the land unnecessary.50
But this line of argument is weak: neither the Farmers' Law, nor the
supposed thematic reforms of the Emperor Heraclius, nor the appearance
of new districts in the provinces, nor indeed the much later evidence for
smallholding on a widespread basis can actually prove this point.51
There can be no doubt that the system of capitatio/iugatio does reflect the
interests of the state and itsfiscaladministration in ensuring that land was
cultivated and that revenues could be appropriated from the maximum
possible number of cultivators exploiting the maximum amount of produc-
tive land.52
But the Byzantine state of the seventh century was just as interested in
this question, indeed, more so, given the loss of some two-thirds of the total
revenue sources of the empire in a period of some sixty or so years.53 Even
assuming that the situation with regard to labour power had improved
(which in the conditions of the seventh century can hardly be certain), it
seems inherently unlikely that the state would voluntarily adjust its system
of revenue calculation and assessment. Surely some other factors were at
work here, which made it necessary for the state, in order to maintain or
intensify the extraction of revenues from a much smaller territory, to
reform its operation. A hypothetical increase in peasant freeholders and a
hypothetical decline of large estates is hardly sufficient, and would in any
case make no difference to the total income of the state derived from this
source.
In fact, the conditions of the later and middle seventh century provide
some clues to the real nature of the change. In thefirstinstance, as we have
seen, vast tracts of land stretching deep into Anatolia were made more or
less permanently insecure for urban life. They will hardly have been any
more conducive to rural and agricultural exploitation. The populations of
the 'cities', whether they remained or whether they moved orfledto safer
areas, were themselves almost entirely agricultural; the rural populations
of the same areas will have been subject to equally or more insecure
conditions, and they can hardly have remained or been able to extract a
livelihood from their regularly pillaged and plundered lands and villages.
50
See Ostrogorsky, 'Das Steuersystem', passim.
51
See J. Karayannoppulos, 'Die vermeintliche Reformtatigkeit des Kaisers Herakleios', JOBG
1 0 (1961), 5 3 - 7 ; Kopstein, 'Zu den Agrarverhaltnissen', 59; W.E. Kaegi, jr., 'Some
reconsiderations on the themes: seventh-ninth centuries', JOBG XVI (1967), 3 9 - 5 3 .
Lemerle, Agrarian History, pp. 48ff., places great emphasis on the importation of Slav
populations in relieving the labour-shortage. But see below.
52
See, for example, the comments of Jones, 'Capitatio and iugatio', 29If.; and esp. Goffart,
'CapuV and Colonate, pp. 47ff.
53
See Hendy's estimate, Studies, p. 6 2 0 .
144 Byzantium in the seventh century

Several sources attest to the abandonment of such areas by the clergy

(higher and lower), by the agricultural population and by town-
dwellers.54 Later sources detail the effects of enemy activity in a way which
only reinforces this impression.55 The seventh-century Apocalypse of
Pseudo-Methodius, written in fact during and after the devastation of the
Anatolian provinces but purporting to tell the future, gives a graphic
account of the devastation of the eastern provinces of the empire, the
capture, massacre or flight of the population and the abandonment of cities
and villages. We must, of course, make allowances for the intentions and
the rhetoric of the text; but even so, it must have had a recognisable basis
in reality and therefore in the popular experience in order to achieve its
aims.56
Given this basic context, it is difficult to believe that the seventh century
really did witness a dramatic rise in the available labour force and a
consequently less urgent need for the state to concern itself with the
relations between land and labour. On the contrary, the flight of the rural
population from many areas will have meant a considerable reduction in
revenues; it must certainly have adversely affected the demography of the
empire, for the effects of both warfare and endemic outbreaks of plague can
hardly have done otherwise. The reinforcement of the population through
the transfer of Slav peoples may well have helped to re-establish a certain
demographic equilibrium; it certainly represents an enormous effort on the
part of the state and hints at the magnitude of the problem. But it clearly
also points to the dramatic consequences for the population of parts of
Anatolia of the events of the second half of the seventh century. It is,
moreover, to be stressed that these Slav immigrants were settled not just in
districts which we might reasonably assume had been badly affected by
warfare; but in areas also in the heart of the empire's territory. Thus there
are seals referring to the Slav prisoners of war for the districts of Cappado-
54
See above, the canons of the Quinisext, the miracles of St Theodore at Euchaita, the
archaeological and narrative evidence for the abandonment of urban sites, the catalogue
of Arab raids itself, year by year. Note also M. Kaplan, 'L'Economie paysanne dans
TEmpire Byzantin du V e au Xe siecle', Kilo 68 (1986), 198-232, see 221-2, who surveys
the disastrous effects of such hostile activity on peasant agricultural production; and cf.
the Fiscal Treatise (ed. Dolger, Beitrdge), 116.2-3, 119.1. Note especially canon 95 of the
Quinisext, which notes that the ecclesiastical authorities were faced by a number of
problems as a result of the large number of refugees coming out of Galatia (Mansi, XI,
984B-E). But see also Haldon, The Miracles of Artemios and Contemporary Attitudes',
p. 34 and n. 8 for some problems with this text.
55
The Vita Philareti (115) notes that Philaretus lost much of his livestock to Arab raiders; the
Fiscal Treatise details enemy raids as the cause of insecurity and the abandonment of land
by even the better-off peasants - see 116.2-3, 119.1.
56
Ps.-Methodlus, Apokalypse, see XI, 9-17 (the text makes explicit mention of the abandon-
ment of cities, the enslavement of the population, the destruction of crops, as well as the
decline in the population of the affected districts); Haldon, 'Ideology and social change',
168 note 74 with literature.
 Rural society 145

cia I and II, for Phrygia Salutaria, Caria and Lydia; but also for Bithynia, in
the Opsikion district neighbouring Constantinople.57 Areas such as Bithy-
nia clearly suffered greatly during the Arab siege operations from 674 to
678, and the long-term effects of this warfare must have been apparent for
many years. But Arab raids penetrated deep into Anatolia on many other
occasions and, while frontier regions and areas where the Arab forces
wintered or were regularly present will have been most affected, it would
seem that nowhere was entirely free from harassment and economic and
demographic disruption.58
The generally unquestioned assumption that there was a demographic
increase in Anatolia at this time, therefore, seems to be fundamentally
flawed, based on a set of hypotheses that have little foundation in the
sources for the period, and which indeed are in clear contradiction to the
logic of the situation which the sources do portray. Ostrogorsky's carefully
developed thesis is based, in effect, on three elements: the destruction or
disappearance of large estates; the supposed establishment under Heraclius
of soldiers' lands; and the assumption that the Farmers' Law represents the
generality of rural communities throughout the empire. None of these is, in
fact, more than a hypothesis; the first two are extremely improbable, and
there is certainly no evidence of any substance in their favour. The
Farmers' Law, as I have suggested, does represent an increase in the
importance of such communities and may represent also an increase in
their numbers. But I do not think that this is enough to assume the
veritable social and demographic revolution for which Ostrogorsky argu-
ed. 59 On the contrary, the evidence seems to point the other way; and
while the transfer of Slavs in such large numbers must have improved or
stabilised the situation, the fiscal response of the state, in eventually
severing the link between land and labour for tax purposes, actually points
in the opposite direction to Ostrogorsky and those who have followed his

57
See Zacos and Veglery, vol. I, pt. 1, pp. 190f. table 33 (dated to the year 694/5).
58
The best synoptic account of this warfare is to be found in Lilie, Die byzantinische Reaktion,
see pp. 76(T.; also Haldon, 'Ideology and social change', 184.
59
See Ostrogorsky, Geschichte, pp. HOfT.; Guillou, 'Transformation des structures', 77, also
argues for increased agricultural productivity during the seventh and eighth centuries
and into the early ninth. Over such a long period (during the later part of which the
empire's lands were a good deal more secure than in the earlier part) he may be correct -
but his argument rests partly upon the assumption that the state could only intensify its
demand for revenues if productivity had increased, an assumption which is dubious.
While it is true that a peasant subsistence economy has only a limited potential for
quantitative extension before it collapses in on itself (see W. Kula, An Economic Theory of
the Feudal System: Towards a Model of the Polish Economy, 1500-1800 (London 1976) see
pp. 28(T. and 1651T. for comparative data), the desperate situation of the Byzantine state in
the later seventh and early eighth centuries must have made increased demand on limited
resources inevitable. But it will have been the form that this demand took that was crucial
to its potential success.
146 Byzantium in the seventh century

reasoning. Shortage of manpower can traditionally be met by landlords

and/or tax-collecting institutions in one of two ways (dependent also, of
course, on which options were available as cultural and political possi-
bilities). The first is to tie the labour force to the land, a process which
involves close supervision by the state, a stable population, the co-
operation of the landowners and landlords, and - as far as possible - stable
political conditions. This was the route taken by the Roman state during
the fourth century, a process which continued into the fifth and sixth
centuries with decreasing momentum. It was a pattern followed also in
Eastern Europe, for example, from the fifteenth century, as population
decline became apparent, and as the flight of peasants to hitherto unex-
ploited lands outside seigneurial political control threatened to damage the
economic interests of the feudal class of landlords. 60
The second way is to attract labour power through offering or accepting
land on conditions and terms relatively advantageous to the cultivator:
low rents, for example, or a low rate of taxation, heritable tenures, freedom
of movement, freedom to alienate the property or the lease and so on. This
pattern was partly followed in England for a period after the devastation
caused by the Black Death. 61 It seems also to have been followed in the
East Roman world during the sixth century and after, and the popularity
and spread of advantageous emphyteutic leases, or similar contractual
forms, during this period certainly seems to represent already the
awareness of landlords, institutional or private, of the advantages such
leases had in retaining their agricultural labour force, and in bringing back
uncultivated land into productive use. 62
Now, in the seventh-century context described above, is it really likely
that the state 'relaxed' its fiscal policy because sources of labour were
plentiful? It seems most improbable. In fact, it must have been faced with a
major difficulty in terms of the effectiveness of its legislation, for it was
clearly simply impossible to enforce regulations which compelled coloni, for
example, to remain on their holdings or to impose fiscal solidarity on
communities of smallholders which may well have resulted in mass aban-
donment of the land, under the conditions we have described for many
areas. Parts of Anatolia, and indeed of the Balkans (or those districts
remaining in Byzantine hands), were at times full of refugees of one sort or
60
See, for example, M. Dobb, Studies in the Development of Capitalism (London 1967),
pp. 53ff.; R.H. Hilton and R.E.F. Smith, The Enserfment of the Russian Peasantry (Cambridge
1968), pp. 1-2 7; and for a summary of developments throughout Eastern Europe, Ander-
son, Passages, pp. 252ff.
61
See G. Duby, Rural Economy and Country Life in the Medieval West (London 1968),
pp. 72ff.; and see M.M. Postan, 'The fifteenth century', Economic History Review 9
(1938-9), 160-7.
62
See chapter 1, and pp. 134f. above.
 Rural society 147

another, seeking safety, or work, or both. Every Arab raid set elements of
the population in motion, even if flight was only to a local fastness; many
of the larger and safer urban centres in the west and north-west of Asia
Minor must have contained far more people - who needed food and shelter
- than they could adequately cope with. And even if this did not affect all
areas at the same time and in the same degree, it represented a more or less
continuous state of affairs from the 650s and 660s well into thefirsthalf of
the eighth century. Not only coloni, whether liberi or adscripticii, but
agricultural producers with freedom of movement, will have been
affected.63 The net result was that the close connection forged over the
preceding centuries between a settled population and its land was, in many
regions/broken. Land was abandoned, even if only temporarily, and this
meant an immediate reduction in resources and, in particular, provisions
for the personnel of the state - the army especially. Faced with the
impossibility of enforcing the older legislation, and the need to find some
alternative way of raising revenues, as well as of restoring abandoned land
to productive use, here was reason enough for remodelling the traditional
system of assessment, even if only selectively at first.
But a more compelling reason exists, and one which helps to locate the
change in the middle years of the seventh century.
As we shall see in chapter 5, there are good grounds for believing that
the greater part of the state's revenue had, from the late 650s if not already
in the 640s, been assessed in kind, primarily in order to maintain and
support the armies or themata which, from the early 640s, began to be
cantonned across Anatolia. In itself, the procedure was not new, a part of
the regular assessment of many areas having always been collected in
produce for the provisioning of the army. But a result of the generalisation
of this practice was that the land-tax itself came to be referred to by the
term traditionally applied to a compulsory purchase or levy of provisions
and materials for the state, especially the army. The term in question is
synone or, in Latin, coemptio. The only context for such a shift, from a cash
assessment and collection to one carried out mostly in kind, can reason-
ably be shown to be the period with which we are concerned, the 640s and
650s. That the change had occurred already by the later seventh or early
eighth century is nicely confirmed by the Farmers' Law, where the regular
land-taxes on a property are referred to as the extraordina of the public fisc
(T& e£rpa6p8iva TOV 8ir]|Aoaun> \670v). 6 4 Now the term extraordinaria
had referred exclusively to exceptional levies above the usual assessment
and could in no way have applied in the sixth century, for example, to
63
See in particular the 95th canon of the Quinisext already referred to (Mansi XI, 984B-E),
regarding the refugee problem in Asia Minor.
64
Farmers' Law art. 19 (JGR II, p. 66).
148 Byzantium in the seventh century

regular taxes. The synone or coemptio, on the other hand, was just such an
extraordinary levy and would be perfectly well described in this way. 65 As
the hitherto limited practice of collecting a portion of the state revenues in
kind was thus applied throughout the empire to the land-tax, so the terms
which had described extraordinary levies in kind seem to have been
generalised and applied to the ordinary assessment on land. The common
element was collection in kind.
There is indeed some evidence to suggest that a reorganisation of the
land-tax assessment had taken place by the 660s. In about 667, an
imperial iussio or command was issued, ordering the drawing-up of tax-
rolls for the populations of Calabria, Sicily, Sardinia and Africa. The same
iussio ordered also the registration of units of assessment - capita - and the
raising of a ship-tax from the landowners of the said provinces.66 The
order coincides, of course, with the presence of the Emperor Constans II in
Sicily and reflects accordingly the needs of the court, army and fleet at that
time. In 681 another imperial iussio was issued, by which the number of
capita, or units of assessment, and the rate of collection of the coemptio,
along with other yearly assessments, was reduced for the population of the
papal patrimonial lands in Sicily and Calabria.67 These two texts together
suggest a significant development. In thefirstplace, the drawing-up of new
tax-registers (diagrafa) implies a major reassessment or reorganisation of
fiscal liabilities. In the second place, the coemptio - synone - is treated
explicitly as a regular, yearly assessment, a character which it had never

65
See the detailed discussion in J.F. Haldon, 'Synone: Re-Considering a Problematic Term of
Middle Byzantine Fiscal Administration', BMGS 18 (1994) 116-53 (repr. in idem, State,
Army and Society in Byzantium [Aldershot 1995] VIE).
"Tiber Pontlflcails, I, 344.2; Ddlger, Regesten, no. 234.
67
Liber Pontificalis, I, 366.8; Ddlger, Regesten, no. 250. Note also Liber Pontificalis, I,
368.19-369.2, according to which an imperial command to Pope Conon in about 687
reduced the taxes of the papal patrimony in Bruttium and Lucania by 200 annonacapita
per annum. The term used strongly suggests a unit of assessment upon which the
calculation of the tax - annonalsynone - was based. By the 730s the system was well
established. An edict of Leo III of 731 for Sicily and Calabria orders the increase of the
taxed by thirty-three per cent and on the papal patrimonial lands the entering of adult
male taxpayers and their heirs on the tax-registers. See Theophanes, 410.8; <p6povs
xecpaXixovs; Ddlger, Regesten, no. 300. Whether this latter procedure can be taken as
evidence for the existence of the kapnikon or its ancestor - since male heads of households
only are to be registered - is uncertain. For the letter of Theodore the Studite (see p. 149
below), see PGIC, 929-33, esp. 932B, and the detailed discussion on the whole question
of distributive and contributive tax-assessment in N. Oikonomides, 'De l'impdt de distri-
bution a l'impdt de quotite a propos du premier cadastre byzantin (7 e - 9 e siecle)', ZRVl 26
(19fc7), 9-19 - the first really to address this crucial issue. Of course, miliaresia were first
issued under Leo HI so that, if the kapnikon or its ancestral form were already in existence
as a cash levy, it would have been raised in either copper folleis or silver hexagrams. But
neither the copper nor the silver coinages were particularly stable in the later seventh
century, and this must have presented a number of problems. Possibly, then, it is to Leo in
that the introduction of the kapnikon must be ascribed. For the coinage, see Hendy, Studies,
pp. 494-6 and 498-501.
 Rural society 149

possessed in the late Roman period. The occurrence of coemptio and capita
together, the former representing the actual tax, the latter the units of
assessment, raised regularly, is good evidence for the existence by this
time of the new procedure. The date of the two orders, applying to the
West, would tend to support the date and the reasons suggested for its
probable introduction in the East in the 640s or 650s. It also suggests
the way in which the later kapnikon was to evolve, as a separate assess-
ment on heads of families in fixed dwellings, calculated differently from
the capita/synone system, but clearly fixed to the notion of residence, as
the name implies.
The synone element of the later kapnikon/synone pair can thus be taken
back with very great probability to the 640s or 650s. There is no reason to
assume that the kapnikon element - whether under that name or not - was
not originated at the same time. The older assessment was, after all, a
land-tax which combined both agricultural land and livestock and other
property in a single calculation; the new procedure may simply have
meant the separate assessment of each. Given the preconditions described
already, and the need to raise a very large part of its revenue in kind, it
must have made good fiscal and administrative sense to add to the original
assessment, based on both capita and iuga and now raised in kind, a second
assessment. For while the collection of the land-tax in kind will have
reduced some of the state's difficulties, it was also much less flexible than a
collection based in cash; and the overall effect must have been a reduction
in the amount of actual wealth thus expropriated. The introduction of a
'new' tax, raised on households - which is what the term kapnikon clearly
implies - may have been intended to compensate for this by introducing a
more flexible (for the state, at any rate) element into the equation, to be
raised in cash.
A letter of Theodore the Studite, dated to the year 801, offers some
valuable evidence in this context. Referring to the fact that the Empress
Eirene had substantially lightened the burden of many tax-payers, it
remarks in particular that the need for the poor to engage in extra (paid in
cash?) work, not to free themselves from poverty, but to be able to pay the
tax-collectors the levy 'which cannot be combined', is now over. This levy,
it states, has existed for a long time; and it has been suggested that this is a
reference to the kapnikon, a flat-rate hearth-tax, levied at a rate of two
miliaresia per household, which cannot be acquitted in kind or in any other
way except in coin. Like any flat-rate levy, therefore, it probably hit the
poorest members of the community hardest. As we have seen, the kapnikon
is first explicitly mentioned for the reign of Nicephorus I, although the
context implies that it was already then old and well established. It seems
not unlikely that it, too, should be taken back into the seventh century, to
150 Byzantium in the seventh century

the time when the state was obliged to split its major tax-assessment on
land and persons into two portions.
A second point in this connection concerns the nature of the assessment
itself. A recent study has pointed out that, whereas the late Roman
assessment was distributive, that is to say, the state fixed its total revenue
requirements (as far as it was able) in advance, and then distributed the
demand across the different units of assessment, the middle Byzantine
system was contributive, that is, taxes were assessed on the basis of the
ability of individual units of assessment to pay. The assessment could vary,
therefore, according to the wealth of the tax-payers as well as the demands
of the state. The split between the assessment on land and the fixed rate on
households which underlies the difference between synone and kapnikon
may well represent the first step in this shift. Oikonomides has collected
evidence which suggests strongly that the weight placed by the state on
oaths in respect of statements of property-value and wealth in the later
seventh century and up until the early ninth century (again a situation
changed by Eirene's reforms mentioned in the letter of Theodore the
Studite and evidenced also in a novel ascribed to her reign) is a sign that
such a system was possibly already in operation. A distributive system did
not require such accurate information, since demands were issued accord-
ing to state requirements only, not the amount of taxable wealth available.
It is precisely the conditions of the seventh century which we have
described - population movement, insecurity of agriculture and rural
production in general in many areas - which may have forced the state to
fall back upon a contributive assessment, secured on the basis of central
assessments of the wealth available from specific regions and specific
tax-payers. Censuses in this context will have been unavoidable.
The loosening of the direct tie between tax-liability and land with labour
will have had several consequences. In the first place, it had the advantage
for the state that refugee populations (which could hardly be punished as
the old laws demanded) were no longer in breach of imperial legislation - a
trivial point in some respects, but important with regard to the authority of
the government. Secondly, it represented an effort to regain some at least
of the lost' tax-payers - who, according to the traditional system of
assessment, were only liable if they cultivated land - by making the
'hearth-tax' or kapnikon (in its ancestral form) independent and separate
from the possession of land. Households or heads of households were
henceforth liable to tax regardless of their relationship to the land. This
alteration, while it must have been difficult to carry through, necessitating
many detailed censuses throughout the empire, will also certainly have
increased the tax-revenue, since it taxed the inhabitants of cities and
non-agricultural communities, too, unless they received specific exemp-
 Rural society 151
tion.68 Thirdly, the change in the system of assessment must have recog-
nised and at the same time thereby promoted greater mobility, since -
although the state still required the maximum amount of land under culti-
vation - it was no longer possible rigidly to tie the agricultural population
to the land, nor was it in many regions at all practicable. In recognising the
change and in adjusting its methods of assessment, the state also averted
the possibility that tied coloni on state lands or over-taxed freeholders would
abandon their holdings and take up advantageous leases on Church or
private estates.69 Fourthly, the conditions which gave rise to these changes
in fiscal practice must have promoted the frequent and widespread transfer
of liability for taxes and/or the cultivation of the land - the old epibole - as
individuals and sometimes groups of peasants gave in to the pressure to
escape hostile action or the effects of bad harvests, for example, both before
and after the changes. The consequence was more regular supervision on
the part of thefiscover the affairs of the rural communities and, if the sixth-
century pattern is anything to go by, increased hardship on those landhold-
ers and communities who had to shoulder the extra burden.70
Finally, and given the options for dealing with demographic decline dis-
cussed already, it is highly probable - and the existence of the Farmers'
Lawl, amongst the texts, is good evidence for this - that the state both reass-
essed the rate of taxation downwards and drew substantial numbers of
refugees to its own estates (which were extensive) on attractive terms.
Indeed, the considerable remission of taxes during the last years of the sixth
centiiry suggests that the state had come to recognise the potential increase
in thi total area of productive land and consequently of revenue which
migh^ thus be achieved. Perhaps this reflects the success of emphyteutic
lessor^ in obtaining both labour for their land and in maintaining their own
incomes.71 A lessening of the tax-burden permitted the investment by
peasarit cultivators of a greater proportion of their resources in agricultural
production, with a consequent increase in absolute terms of output and,
indirectly, of population.72
68
Note that censuses were carried out in the West in the 6 6 0 s and in 7 3 0 - not exactly
frequently, but illustrative of their importance to the fiscal administration of the state. See
notes 6 6 - 7 above.
69
The notion that such a reform also promoted mobility was first expounded by N.A.
Constantinescu, 'Reforme sociale ou reforme fiscale?', Bulletin de VAcade'mie Roumaine,
section historique, 11 (1924), 94ff., but was rejected by Ostrogorsky, Geschichte, p. 115
note 3.
70
See chapter 1, p. 2 9 and Procopius, Historia Arcana, XXHI, 1 5 - 1 6 . Cf. Kaplan, 'L'Econo-
mie paysanne', 2 2 9 (for the tenth century) and 2 0 2 (general).
71
Tax evasion in the sixth century: Justinian, Nov. 1 4 7 (a. 5 5 3 for Oriens and Illyricum);
JGR I, coll. 1, nov. 1 (Justin II, a. 566); nov. 11 (Tiberius Constantine, a. 575) (Dolger,
Regesten, nos. 4 and 40).
72
See, for example, Fiscal Treatise (ed. Dolger, Beitrdge), 1 1 5 . 2 8 - 3 3 and the discussion in
Kaplan, L'Economiepaysanne: a reduction in tax/rent burden can result in the expansion of
152 Byzantium in the seventh century

All these changes amount to a series of fairly drastic organisational

alterations in the late Roman system of tax-assessment and collection.
Exactly where and how they were first implemented must remain unclear
for the time being.
There is no reason to doubt that these changes in tax-assessment
together with the conditions of economic disruption and dislocation which
brought them about, can also have contributed towards an increase, at
least in some areas, of independent peasant smallholders. Large estates,
while their owners may have been able to take refuge, and whether
ecclesiastical, imperial or private, must have suffered under the same
conditions and the consequent shortage of labour; they will have been only
too happy to maintain production on the basis of less oppressive contract-
ual arrangements, long-term, short-term or perpetual. Theflightof much
of the rural population of the most exposed areas will have benefited safer
regions and will also inevitably have promoted the gradual disappearance
of the remaining legislative ties between land and peasants. Many peasants
of relatively humble status must have escaped their former condition
entirely in this way. The importation by the state of new elements to
repopulate the countryside, whatever the scale on which it was done, will
also have reinforced this tendency,73 and the need for the state to recruit its
soldiers from among the ordinary population of the provinces, and for the
soldiers to provide at least a part of their own equipment during the later
seventh century, is good evidence that by this time there existed a suffi-
ciently numerous and prosperous source of volunteers or conscripts from
among the rural population on which it could draw.74
The result of this brief survey, therefore, is that there does seem to have
taken place an expansion of independent smallholding. This expansion
was both a result of the changes in the relationship between agricultural
producers and landlords, and to a degree also of the reorganisation of the
system of fiscal assessment by the state, in turn brought about by the
the basic unit of production both in land and in demographic terms - in a growth,
therefore, of productive capacity in the quantitative sense. The disadvantages are herita-
ble partition of the family holdings and ultimately a reduction in the size and the viability
of such units - which become as a consequence more vulnerable in adverse condition
(drought, pestilence, warfare, bad harvests, etc.). It is nevertheless clear, as the Fiscal
Treatise demonstrates, that the effects of a reduction in the overall burden of tax and rent
were recognised by contemporary authorities.
73
The question of where and h o w these settlers were accommodated, as well as that of the
fiscal system of the state in general, will be dealt with in chapters 5 and 6. But note that
the picture was further complicated by the movement of refugees from both inside and
outside the empire. For the year 6 8 6 , for example, Theophanes (364.3^£) notes that a
famine in Syria forced many people from that region to flee to 'Romania', an event
recorded also in a contemporary Syriac chronicle. See G.J. Reinink, 'Pseudo-Methodius
und die Legende vom romischen Endkaiser', 9 5 and note 6 2 .
74
See Haldon, Recruitment and Conscription, pp. 72f.
 Rural society 153
dislocation of the warfare of the second half of the seventh century. The
conditions promoted by war and constant insecurity cast a large number of
the agricultural population adrift, as refugees - either from the tax-collec-
tor, the rent-collector and bailiff, or from the Arabs. They stimulated an
increase in demographic mobility. The relocation of refugee populations to
several different areas of the empire, in particular to a number of the
Anatolian provinces, and - given the endemic shortage of labour power -
under much improved conditions, completes the picture. The state, faced
with a fait accompli, could only conform to the changed situation and
adjust, to try to maximise the potential of the new conditions.

THE LANDED WEALTHY AND THE NEW ARISTOCRACY

The changes which I have outlined so far clearly did constitute a shift in the
relations of production in the countryside. While labour power was scarce,
cultivable land deserted and the state able to impose a uniform policy on the
rural population with great difficulty - the latter an effect of the divergent
interests of the older landowning element of the ruling class and the new,
particularly the military, service meritocracy - the condition of the
peasantry with regard to the assertion of its rights of tenure, and commu-
nal solidarity in respect of the fisc, had improved. But clearly, a number of
factors militated against the disappearance of large estates and of social
equilibrium in peasant communities.
In thefirstplace, as we have seen, areas where a strong military presence
was required will have been easy prey to the landed property ambitions of
officers who, in return for protection from various threats - whether from
the enemy or from government officials, for example - will have been able
to gain power over smallholders and peasant communities and indeed over
the property of both resident and absentee landlords without much diffi-
culty. As in an earlier period, it would not have proved difficult to turn the
'protection' into a rent, until the patronage was effectively converted into
one form of possession or another through lease, donation or sale.75
Social differentiation within peasant communities, the impoverishment
of some smallholders to the advantage of others (clearly present in the
Farmers' Law) and in particular the subdivision of holdings among heirs,
leading to holdings which might eventually become too small to be
economically viable, these factors must all have facilitated the encroach-
ment of the wealthier peasant on the land of the poorer, and of the big
estate owner on the lands of peasant communities in general.76 Together
75
See, for example, Haldon, 'Some considerations', 95-8; Jones, LRE, vol. II, pp. 776f.
76
For the effects of these factors in the late Roman period, up to the later sixth century, see
Jones, LRE, vol. II, pp. 773-81.
154 Byzantium in the seventh century

with poor harvests, seasonal or climaticfluctuation,natural disasters and

similar perennially present elements, and in addition the demands of the
fisc, of the local provincial or military administration and other depart-
ments of the state administration, such factors must always have left
peasant farmers very vulnerable to fairly dramatic changes in fortune and
sudden household economic failure.77 The abandonment of holdings and
the attribution of their tax-dues to their neighbours was clearly a normal
occurrence for the compilers of the Farmers' Law. The speed with which
Philaretus' community set about taking over his unworked holdings (quite
apart from the greed of the larger landlords in simply taking Philaretus'
property), together, presumably, with the taxes that went with those
holdings, demonstrates several aspects of such a situation: both the
process, described in the Farmers' Law, whereby unwanted land might be
alienated (temporarily in the first instance) to a third party within the
community who was responsible also for its taxes; and it demonstrates the
ways in which a landholder or owner could lose his lands (although, in the
case of Philaretus, a lack of prudence rather than of resources was to
blame). In Philaretus' community, or in his neighbourhood, the needy
peasant farmer was not uncommon - witness the number of animals or
amount of other property that was given away.78 On the other hand, the
powerful local landowner, with estates spread over several regions or
provinces was also present. Church and state continued to represent
substantial landowning interests. And there is no reason to doubt that
peasants of colonus status, bound to their land by private contract, con-
tinued to farm such estates, sharing villages with free tenants, emphyteutic
leaseholders and so on.79 In areas relatively free from enemy harassment,
the well-established bond between land and tenant, whatever the legisla-
tive intervention of the state may have done to relax or sever the connec-
tion, probably remained, if only because the tenant farmer attained
thereby a degree of economic security for himself and his family - as well
as the (dubious) protection of a landlord. The advantages and dis-
77
Compare the effects of the severe winter and bad harvests of the years 927/8 and 938/9,
described in the novel of Romanus I which dealt with the results (Dolger, Regesten,
no. 628; text: JGR I, pp. 2O5ff.). See Ostrogorsky, Geschichte, pp. 227ff.; R. Morris, 'The
powerful and the poor in tenth-century Byzantium: law and reality', Past and Present 73
(1976), 3-27.
78
The best summary of the provisions of the Farmers' Law is to be found in Kopstein, 'Zu den
Agrarverhaltnissen', pp. 4 0 - 5 3 ; for Philaretus, see ibid., pp. 6 1 - 2 (summary of text);
Kaplan, 'Les Villageois', 214f.
79
The tenants of Church and monastic lands are referred to in the early ninth century as
paroikoi, which may mean simply tenant or bear a more specific and technical meaning
similar to that of colonus liber of the legislation of Anastasius and Justinian. See Theo-
phanes, 4 8 6 . 3 0 . For the technical meanings of paroikos, see Lemerle, Agrarian History,
pp. 1 7 9 - 8 2 .
 Rural society 155

advantages were, from a theoretical standpoint, evenly balanced. Indeed,

it must be emphasised that while the conditions of the seventh century do
seem to have resulted in a number of important and substantive changes,
not all areas within the empire, nor all of the rural population, will
necessarily have been affected in precisely the same way. Considerable
room for regional variation should be allowed within the general explana-
tory framework.
And so, while the overall effect of the economic and social dislocation of
the second half of the seventh century and of the opening years of the
eighth century seems to have been to promote an increase in peasant
mobility and in the numbers of peasant smallholders subject directly to the
fisc, with all the contingent results for the state and its fiscal apparatus,
already the age-old disadvantages which accompanied the subsistence
economy of the peasantry were leading to a new phase of polarisation
between estate owners and peasant communities, on the one hand, and
within such communities on the other. The evidence of the Life of Philare-
tus, for the later eighth century, and of the measures taken by Nicephorus I
with regard to the tax-burden of the impoverished peasantry, as well as
those of Eirene in respect of taxes which affected the poorest members of
society in the provinces and in the capital, make this much clear. 80
The origins of the provincial landed elite, the 'aristocracy' of the ninth
and tenth centuries and after, are to be sought for the most part in the
same social and economic dislocation and turmoil as the increased mobi-
lity and numbers of the peasant freeholders. Whereas the latter were
spread geographically throughout the empire (at least, there is nothing to
contradict this assumption), the later evidence for the location of the more
powerful provincial families suggests that it was precisely in areas of
relatively low population density, and which formed the wide band of
territory constituting the frontier zone, that this section of the ruling class
had its origins.
It has already been demonstrated by several scholars that the dominant
elements of the Byzantine ruling class in the period from the ninth to the
eleventh centuries originated in the eastern themata of Anatolikon, Kappa-
dokia, Paphlagonia, Charsianon and Boukellarion.81 Out of the forty-seven
80
Theophanes, 486.23-6; see Haldon, Recruitment and Conscription, p. 50 note 87. For
Eirene's reforms, see the letter of Theodore the Studite of 801 (PG IC, 929-33) with note
67 above.
81
See especially Sp. Vryonis, jr., 'Byzantium: the social basis of decline', GRBS 2 (1959),
159-78, see 161f.; Vryonis 'The internal history of Byzantium during the "Time of
Troubles" (1057-1081)' (Diss., Harvard 1956), pp. 173ff. and 390f.; and The Decline of
Medieval Hellenism, pp. 24f and list. See also A.P. Kazdan, Socialniy sostav gospodstvuius-
cego klassa Vizantii XI-XII vv. (Moscow 1974), pp. 195, 204 and lists. More recently
Winkelmann has shown how the development of nomina gentilia is, initially, to be found
among these provincial families. The degree to which this reflects the bias of the sources
156 Byzantium in the seventh century

family names located by the sources either explicitly or by implication

(ownership or large-scale estates), thirty derive from the districts of Anatoli-
kon, Kappadokia and Paphlagonia. Similarly, Kazdan's analysis of the
origins of the eleventh- and twelfth-century civil elite shows that, whereas
the great majority of the known members of the military elite came from
central or eastern Anatolia and the Caucasus, with smaller percentages
from Macedonia, Bulgaria and outside the empire, the majority of the civil
elite hailed from Constantinople or its immediate hinterland, coastal
Greece and coastal Anatolia, with the Aegean region. 82 The close connec-
tion between the homelands of the military elite, and the non-urban and
plateau zone of Asia Minor, with its predominantly pastoral economy, and
the concomitant identification of the greater part of the civilian elite with
the urbanised or more densely populated regions of the empire, becomes
immediately apparent. 83 Now it is significant that not only was the
military elite drawn for the most part from areas of a chiefly pastoral
economic character where towns had usually been limited to those dis-
tricts in river valleys, for example, which could support them through a
limited amount of arable exploitation;84 they also represented those areas
which, from the seventh to the later ninth century, formed the frontier, as
mentioned above. 85 Their association with a pastoral economy was
ancient, 86 and some were the site of large-scale imperial stud-farms in the
later Roman period. 87 The same type of economic activity was still carried
on in the tenth and eleventh centuries, as the Life of the magnate-turned-
monk Michael Maleinus implies: the property he gave away on taking holy
orders consisted chiefly of herds of cattle and flocks of sheep. 88 Cedrenus
notes that the estates of Eustathius Maleinus in the themes of Boukellarion
and Charsianon consisted mostly of large ranches; 89 and already in the
fourth century (and therefore probably earlier) some senatorial magnates

and their interest in the political-military conflicts of the empire, on the one hand, and a
more-or-less accurate picture, on the other, remains uncertain. But the general context of
the growth of such families would suggest that they do illustrate an actual tendency. See
Winkelmann, Quellenstudien, passim and pp. 2 1 If.
82
Kazdan, ibid. Clearly, a complete catalogue is not possible; but the bias of the sample
offered by the sources is suggestive and is supported by other evidence discussed below.
83
See Haldon, 'Some considerations', 95ff., esp. 9 7 - 8 ; Kazdan, ibid.; Hendy, Studies,
pp.lOOff.
84
The best survey is that of Hendy, Studies, pp. 9 0 - 1 0 0 .
85
See Lilie, Die byzantinische Reaktion, map at p. 336; and cf. Hendy, Studies, map 2 5 .
86
See Strabo, Geographica, XII, 2.7ff.
87
Jones, LRE, vol. II, pp. 7 6 7 - 9 ; E. Kirsten, art. Cappadocia, in Reallexikon fur Antike und
Christentum II (1954), cols. 869f.
88
Vie de S. Michel Maleinos (ed. L. Petit, in ROC VII (1902), 5 4 3 - 6 0 3 ) , see 557.31sq.
89
Cedrenus, vol. II. 4 4 8 . 9 - 1 6 ; Scylitzes, 3 4 0 .
 Rural society 157
lived not in cities, but on their fortified ranch-villas.90 Just as significantly,
there were areas in which the activities of local magnates in appropriating
both imperial and private lands, farms and herds of horses had attracted
the attention of the Emperor Justinian. As has been pointed out, the
districts primarily involved were the Cappadocias, Helenopontus, Paphla-
gonia, Phrygia Pacatiana, Galatia I, Pisidia and Lycaonia, districts which
were later to be subsumed within the themes of Anatolikon (Phrygia
Pacatiana, Pisidia, Lycaonia, part of Galatia), Kappadokia (Cappadocia I
and part of II) and Charsianon (Cappadocia II and part of Galatia I).91
It seems clear, therefore, that the regions of central and south-eastern
Anatolia from which the later provincial military magnates appear for the
most part to have been drawn coincided with areas which were both
geographically and climatically unsuited to arable exploitation, favouring
instead a pastoral economy. They were areas which had always supported
such an economy, and in which the extended 'ranch' had formed the
traditional unit of exploitation. They were, furthermore, areas of relatively
low population density and of few urban centres, as Hendy has most
recently pointed out.
The fact that the later provincial magnates originated in large part in
these districts is surely no accident. It seems to be a direct result of the
coincidence of warfare, patterns of settlement and patterns of economic
exploitation. And in the context of the dislocation of the seventh century it
is worth noting in passing that unlike arable farmers, pastoralists and
ranchers - given reasonable warning - may often be less vulnerable,
requiring as they do much less manpower and being able to bring herds
and flocks more quickly to safety.92 It is significant that the later Byzantine
aristocracy of the tenth and eleventh centuries placed more emphasis on
their moveable wealth - whether in the form of livestock or jewellery or
similar items - than on land itself, an illuminating indicator of the relative
ideological values attributed to these two forms of investment in social
wealth.93 Contrary to the assumptions of some historians, therefore, a
90
Cf. B. Treucker, Politische und sozialgeschichtliche Studien zu den Basilius-Briefen (Frankfurt
and Munich 1961), p. 15.
91
The arguments are well summarised by Hendy, Studies, pp. 103-4. For Justinian's
legislation, see Justinian, Nov. 28 (Helenopontus), 29 (Paphlagonia), 24 (Pisidia), 25
(Lycaonia), 30 (Cappadocia), 8 (Galatia and Phrygia Pacatiana). For these reforms and
their background, see Jones, LRE, vol. I, pp. 280ff.
92
In the tenth-century treatise on guerrilla strategy, De Velitatione Bellica, it is predomi-
nantly livestock that the author envisages being taken to safety before raiders arrive, with
the implication that the economy of the regions in question was predominantly pastoral
or transhumant in character. See G. Dagron, Le Traite sur la guerilla (De Velitatione) de
Vempereur Nicephore Phocas (963-969) (Text G. Dagron and H. Mihaescu; transl. and
comm. G. Dagron; Appendix J.-C. Cheynet) (Paris 1986), XII.8-9; XX.59.
93
See the analysis of G.G. Litavrin, 'Otnositel'nye razmeny i sostav imuscestva provincial'-
noi vizantiiskoi aristokratii vo vtoroi polovine XI v.' Vizantiiskie Ocerki (Moscow 1971),
158 Byzantium in the seventh century
context of ongoing warfare and military conflict may well have been more,
rather than less, conducive to the appropriation of lands and the evolution
of large-scale landed property in the provinces during the seventh and
eighth centuries, at least where a pastoral economy dominated.94
The implications of this concordance are considerable. In thefirstplace,
it seems that the dominant position won over the period up to the tenth
century in these eastern and central Anatolian lands by this class of
military magnates was not necessarily replicated in the more densely
populated regions with a more strongly arable tradition. It has been
demonstrated that the newly reconquered territories in south-eastern and
southern Asia Minor and in northern Syria, taken in the period from the
middle of the ninth to the middle of the eleventh century, were almost
without exception keptfirmlywithin state hands.95 By the same token, the
military elite seem never to have established themselves asfirmlyeither in
the European provinces or in the western and northern plains and littoral
of Asia Minor, areas where the heavily agricultural and densely settled
nature of the land made such a dominance much less straightforward.96
This does not mean that large estates, relatively speaking, farmed by
tenants of varying degrees of dependent status, did not develop. But it is
clear that they did not develop on the same scale as in the East,
undoubtedly a result of both lack of opportunity on the part of the local
military and state establishment: western Asia Minor, for example, was
never subject to the degree of population collapse and economic dislo-
cation that typified the history of the eastern regions during the seventh
and early eighth centuries. It also suggests that the geography of owner-
ship in the predominantly agricultural zones, with a much greater parcelli-
sation and subdivision of the land among both peasantry and estate-
owning families, prevented the rapid consolidation of large tracts of land
into estates. This was clearly the case in the sixth century and before. In
western Asia Minor, for example, in the district of Magnesia on the River
pp. 1 6 4 - 8 ; and the comment of Kazdan and Constable, People and Power in Byzantium,
p. 5 1 .
94
See Winkelmann, Quellenstudien, esp. pp. 28f., for example. But comparative evidence
from the later period of Byzantine history - from the twelfth century in Asia Minor, along
the Byzantine-Turkish frontier region, and from the middle of the thirteenth century,
along the border regions between the empire of Nicaea and the Turks - suggests very
strongly that warfare and economic insecurity are not incompatible with a pastoral
economy and the growth of estates. See the comments of M. Angold, A Byzantine
Government in Exile: Government and Society under the Laskarids of Nicaea (1204-1261)
(Oxford 1975), p. 1 0 1 ; and also Vryonis, Decline of Medieval Hellenism, pp. 1 4 5 - 5 5 . See
also C. Cahen, Pre-Ottoman Turkey: A General Survey of the Material and Spiritual Culture and
History c. 1071-1330 (London 1968), pp. 143ff.
95
See N. Oikonomides, Les Listes de preseance byzantines des IXe-Xe siecles (Paris 1972),
pp. 355f. and 363; and Hendy, Studies, p. 104.
96
Hendy, Studies, pp. 85ff. (Balkans) and 132ff. (general discussion).
 Rural society 159
Maeander, and at Tralles, land-holdings and farms, whether on larger or
smaller estates, or pertaining to village freeholders or to cities, were
extremely fragmented in the early fourth century, as fragments from
surviving census records illustrate; there is no reason to suppose that the
picture changed through the later Roman period and into the seventh
century. It is interesting also that this region later formed the southern
section of the Thrakesion thema, an area for which the late tenth-century
evidence might imply a very similar pattern of property-holding.97 The
tenth-century evidence adduced by Hendy demonstrates that the state was
more easily able to defend its interests in this region, among others, in
maintaining peasant taxpayers independent of large landowners; and
although it is a clear implication of the legislation that wealthy landlords
already existed in the districts dealt with in the novels of Constantine VII
and Romanus II (chiefly the Thrakesion thema), they were clearly not
landlords on the scale of the magnates of central and eastern Anatolia.98
There seems little doubt, therefore, that the rise to prominence of the
military magnates of Anatolia which becomes apparent in the sources
during the ninth century and after is predicated upon the events and
developments of the second half of the seventh century. Whether through
their power and authority alone, or whether also through legal and
marital assimilation with the pre-existing civilian (senatorial) nobility,
they became the dominant landowners in the East and represented a
powerful faction at court. But their pre-eminence should not blind us to the
existence of other wealthy landowners in other regions of the empire, nor
to the fact that the power and wealth of the latter, where it was not
inherited, was also, in origin, due to their position in the imperial estab-
lishment, their closeness to the palace and to the emperor's circle, and the
97
See esp. A.H.M. Jones, 'Census records of the later Roman empire', JRS 43 (1953), 49-64
(repr. in The Roman Economy, pp. 228-56).
98
See/GR I, coll. 3., nov. 6 (pp. 214-17, a. 947, Constantine VII; Dolger, Regesten, no. 656);
nov. 16 (pp. 243-4, a. 962, Romanus II; Dolger, Regesten, no. 690). Compare Basil n's
novel of 996 dealing with the vast estates of Eustathius Maleinus, JGR I, coll. 3, nov. 29
(pp. 264-5). For the landed elite of the period, see S. Stavrakas, The Byzantine Provincial
Elite: A Study in Social Relationships during the Ninth and Tenth Centuries (Ann Arbor 1978);
J.-F. Vannier, Families Byzantines: les Argyroi (IXe-XIIe siecles) (Paris 1975), esp. pp. 16ff.;
W. Seibt, Die Skleroi: eineprosopographisch-sigillographische Studie (Vienna 1976); I. Djuric,
4
La Famille des Phocas', ZRVI17 (1976), 195-291 (Fr. resume 293-6); also Morris, The
powerful and the poor', 16; and D. Papachryssanthou, in TM 3 (1968), 309-23 (for the
Monomachoi); E. Honigmann, 'Un itineraire arabe a travers le Pont', A1PH0S 4 (1936),
268ff.; H. Gregoire and N. Adontz, 'Nicephore au col roide', B 8 (1933), 2O3ff.; and esp.
M. Kaplan, 'Les grands proprietaires de Cappadoce (VIe-XIe siecles)', in Le aree omogenee
della civiltd rupestre nellambito dell'impero bizantino: la Cappadocia (Galatina 1981),
pp. 143-8. See further H. Ditten, 'Prominente Slawen und Bulgaren in byzantinischen
Diensten (Ende des 7. bis Anfang des 10. Jhdts.)', in Studien zum 8. und 9. Jahrhundert in
Byzanz, eds. H. Kopstein and F. Winkelmann {BBA LI, Berlin 1983), pp. 95-119, see
lOOff.
160 Byzantium in the seventh century
various networks of clientship and patronage which invariably develop as
corollaries to a court and its apparatuses. However much influenced by the
topoi of the hagiographical genre, the Life of Philaretus does make it clear
that large-scale property accumulation or maintenance could take place or
continue outside the state and its institutions. For the same period as this
Life, other hagiographies similarly refer to the wealth of provincial land-
owners. The origins of such wealth, of course, must remain unknown -
whether acquired by earlier members of the family while in imperial
service, granted as a reward for such service, purchased gradually or
obtained fraudulently, all are possible. But while a degree of continuity
from before the seventh century may be present, it would seem on the basis
of the admittedly very limited evidence not to have been very great."

THE SENATORIAL ARISTOCRACY AFTER THE SIXTH CENTURY:

SURVIVAL AND ASSIMILATION

The fate of the older landowning elite is uncertain, as we have seen, and it
is equally impossible to relate its fortunes to particular regions of the
empire or to specific patterns of landholding and their survival, or their
decline from the sixth century. On the other hand, something can be said
about the changes in the composition of the ruling class of the seventh-
century world, from which certain inferences may be drawn.
The chief landowners of the late Roman period had usually been
members of the senatorial order, which had since the fourth century
undergone an enormous expansion, chiefly a result of the Diocletianic
reorganisation, through which senators were initially excluded from all
but a few minor civilian posts. The military and most civilian posts of the
state were thenceforth supposed to be occupied by members of the eques-
trian order, membership of which depended upon office, granted by the
emperor. The senate itself was a mostly hereditary body, although new
members were regularly adlected by the emperor.100 As a result of Diocle-
tian's reforms, the senate lost much of its former political power, although
as a body it still represented the old aristocracy of land and office in the
West, and always had great social prestige.
This situation began to change during the fourth century, partly
because Constantine I began to employ members of the senatorial order in
the administrative machinery of the state on a much wider basis,
effectively ignoring the restrictions imposed by Diocletian. Through large
numbers of grants of senatorial status to outsiders the senate expanded;
99
See the remarks of Winkelmann, Quellenstudien, pp. 29-31, with sources and literature,
for the ninth century.
100
For the procedure, see Jones, LRE, vol. II, pp. 53Of.
 Rural society 161
while both Constantine and his sons increasingly appointed senators to
posts formerly reserved for members of the equestrian order. Many posts
now came to be reserved for senators only, thus excluding the equestrian
order. The senate in the East, of course, had neither the wealth nor the
prestige of that in Rome; but it experienced similar changes and, while in
the 350s there had been only some three hundred members, by the 380s
there were some two thousand.101 The reasons for the increase are not
hard to find: as an ever greater number of posts were opened up to or
reserved for senators, more and more outsiders were appointed to them
and received senatorial status; since the occupancy of such offices was
limited, the regular intake of newcomers was considerable. By the same
token, an ever greater number of palatine officials received senatorial
rank on retirement; while the emperors also made a great number of
grants of honorary senatorial status. The long-term result was that sena-
torial rank became increasingly devalued, so that, as Jones pointed out,
decurions who had held the high-priesthood of their city were also
admitted.102
The result of the rapid expansion of the senatorial order was the com-
plete devaluation and ultimate disappearance of the equestrian order,
membership of which had depended on office.103 Diocletian's reorgani-
sation had improved the position of this order at the expense of the senate.
Its membership increased greatly, but as the fourth century wore on it
became more and more inflated and lost ever more in status and value to
the newly expanding senatorial order, until it had effectively disappeared
- only some officials in certain palatine bureaux retained the grades of the
equestrian order.104
The senatorial order, as it expanded, was also regraded internally. Pre-
viously, all senators had been clarissimi; from the time of Valentinian I a
revised system was introduced, in which, while senators by birth and
newcomers admitted by a grant of the title, who held no office, still ranked
as clarissimi, they were preceded in rank by those of illustris and spectabilis
grade, the award of these titles being attached to the holding of imperial
posts. The exceptions to the general rule were provided by the titles of
consul and ex-consul, who had precedence over all senators - and by the
revived title of patricius, reintroduced under Constantine I, awarded to

101
Themistius, Oratio 34, 456.
102
The process is well summarised by Jones, LRE, vol. II, 527; see also Tinnefeld, Die
fruhbyzantinische Gesellschaft, pp. 66-71.
103 The order was divided into a number of grades: egregius, centenarius, ducenarius and
perfectissimus (with eminentissimus for praetorian prefects) according to salary and post.
See Jones, LRE, vol. n, p. 525.
104
Jones, LRE, vol. n, pp. 525-7; Tinnefeld, Die fruhbyzantinische Gesellschaft, p. 71.
162 Byzantium in the seventh century

individuals as a personal distinction. 105 This alteration was important,

especially for the later development of the system of precedence, since its
effects were to change the aristocracy of the state from one of birth to one
of office. The son of a senator of illustris rank was still hereditarily a
member of the senatorial order, of course, but he received only the generic
title of clarissimus.106 Higher status was attained by holding office or by an
imperial grant. During the fifth century, the senatorial order became so
enlarged that a tendency to increase the status and privileges of the highest
grade, the illustres, at the expense of those of spectabilis and clarissimus
rank, is clear. The latter, particularly if they were also curiales, were
encouraged or indeed compelled to reside in their provinces; under
Marcian, they were excused from being nominated for or holding the
praetorship and praetorian games in Constantinople, which was their only
remaining link to Constantinople and the senate body. By the time of
Justinian, only illustres could belong to the senate proper and bear the title
senator. 107 The title of clarissimus was still hereditary, and all sons of
senators inherited it. Spectabilis was merely an honorific dignity, bearing
only very few privileges and attached to a number of lowlier functions.
Membership of the senate itself, which had thus once again become a
relatively small body, was attained by imperial award of an office carrying
the rank of illustris, whether active or honorary. This does not mean that
the formal rules were always adhered to, since senators could still, often
with success, petition the emperor for the title oiillustris for their sons. The
result was that the senate of the sixth century was still made up of a
mixture of both hereditary members and those appointed by the emperor.
The system was openly pluralistic.
The order of ranks and precedence was complicated still further by
distinctions within the highest grade of illustres. Honorary offices, with
illustrious rank attached, were granted, but with equivalent status to
active offices, as opposed to the lower-ranking, purely honorary posts
without such a title. The former were referred to as illustres inter agentes
and were described as titular holders of actual offices (vacantes) in oppo-
sition to the honorarii who held the title as an honorific only. 108 Within the
grade of illustres new, higher grades were established, those of magnificus
and gloriosus (also as gloriosissimus), which in effect replace illustris in
105
See CTh. VI, 6.1 (a. 382) for consuls; and Zosimus, vol. II, 4 0 for the title ofpatricius. The
Emperor Zeno had insisted upon the rank of consul or the office of praetorian praefect in
order for the title of patricius to be bestowed (see CJ XII, 3.3 proem), but Justinian
withdrew this ruling in 537 (Justinian, Nov. 62, 2.5). See Stein, Bas-Empire, vol. II,
p. 430.
106 1Q
Cf. CJ Xn, 1.11 7 See Digesta I, 9.12 c.l and Jones, LRE, vol. H, p. 529.
108
See Justinian, Nov. 62, proem and 1; CJ XII, 16.1 (a. 415 = CTh. VI, 23.1). Note Stein,
Bas-Empire, vol. II, p. 429; Jones, LRE, vol. n, p. 535.
 Rural society 163

respect of most of the higher-ranking posts to which it was attached; while

Justinian awarded the title of illustris to a number of provincial officials and
even curiales who had previously borne only the ranks of spectabilis or
clarissimus.109 The reasons lie primarily in the distinctions between the
illustres inter agentes and active illustres, or gloriosi, on the one hand, and
the lower-ranking illustres, on the other. 110 The complexity of the system
as a whole is demonstrated for an earlier period in a law of Theodosius
II, 111 according to which the order of precedence is: (i) those who actually
held offices with illustrious rank; (ii) those who received a titular office
(vacans) while at court; (iii) those who had received such an office in their
absence; (iv) those who received an honorary office at court; (v) those who
received such an office in their absence. Members of the first group - the
active office-holders - always ranked above the rest; but in groups (ii) to (v)
position depended on the rank of the titular or honorary office held: thus
an honorary office-holder, the position of whose office in the hierarchy of
posts was higher than that of a titular office-holder, would have prece-
dence. 112
The expression 'senatorial aristocracy' should be used with care, there-
fore. It refers to both the hereditary group of those who held the title of
clarissimus, which by Justinian's time must certainly have numbered
several thousand, as well as to those lower-ranking illustres in the
provinces. Many of these would have been landowners of small or mid-
dling status, whose property will have been dispersed among several
villages and cities, mixed in with the properties of both local curiales and of
much larger landowners. Some of those referred to above in the Maeander
region will have been typical. But the term refers also to those of magnificus
or gloriosus rank, holding actual titular or honorary offices of state
(whether or not they resided at Constantinople or had permission to live in
their province), men of much greater wealth and social standing. 'Sena-
torial aristocracy' is a term which, for the sixth and early seventh century
at least, is perhaps better used of this group alone, since in relation to

109
Justinian, Nov. 70 (a. 538).
110
See P.Koch, Die byzantinischen Beamtentitel (Jena 1903), pp. 4 Iff. and 70f.; Stein,
Bas-Empire, vol. II, pp. 4 2 9 - 3 2 with literature.
111
CTh. VI, 22.8 (a. 425).
112
All forms of service in the imperial establishment were covered by the general term
mihtia; but a distinction was drawn in practice between its specific sense of service as a
member of a particular department or unit in the army; and service in one of the higher
posts, military or administrative, described also as dignitates, honores or administrationes.
The former were filled by issuing probatoriae or certificates of appointment; the latter by
an imperial codicil and at the emperor's pleasure. Even active positions in this group were
regarded as dignities in the general sense, hence the great demand for titular and
honorary appointments to such positions, and they were generally referred to as dignities
rather than 'posts'.
164 Byzantium in the seventh century

wealth, power, the patronage of the court and the administration of the
government, both in the civil and the military spheres, they constituted the
ruling elite - the governing elite within the ruling class - of the empire.
By the late sixth century, and in spite of the reduction in the size of the
senate proper which had taken place, the granting of titular and honorary
senatorial offices had once again increased its numbers. Justinian, as
mentioned already, decreed that only patricii, ex-consuls, illustres and
illustres inter agentes could sit on the senate; he also decreed that all
members of the consistory (usually all senatorial offices) should be present
during senate meetings - and that all senators should participate at
judicial sittings of the consistory. 113 The active illustres, consisting of those
generally referred to as gloriosi or gloriosissimi, formed the supreme group
of senatorial dignities in the civil and military establishment. And while
their rank could on occasion, and upon application to the emperor, be
conferred upon their sons, such positions were not hereditary and
depended to a very great extent upon the emperor. Of course, once a
senatorial family had established both contacts to the palace and a basis in
landed wealth, it was able to further the interests of its offspring and
promote their careers, so that offices and titles could eventually be received
by successive generations as their services and suitability were brought to
the attention of the emperor or the appropriate officials. But such a system
left a great deal of leeway for the emperor or his advisers to promote
outsiders, and it left the way open for persons of quite humble rank to rise
to positions of considerable power. 114
In spite of some alterations, it is evident that the system of ranking

113
See Justinian, Nov. 62, 1 and Stein, Bas-Empire, vol. n, p. 432 and literature. Justinian
also decreed (Nov. 62, 2, a. 537) that patricii were henceforth to have precedence over
those with consular titles and that precedence among this group was to be determined on
their having held one of the variety of consular offices (whether titular, honorary or
whatever). According to this novel, the prefect of the city (praefectus urbis) was now to
have first place, followed by the 'other' patricii, then the consuls and consularii, and then
the praetorian prefects, magistri militum, and other illustres. The consistory (consistorium)
included leading members of the senate and the civil and military establishment appoin-
ted ad hominem: the magister officiorum, quaestor, comes sacrarum largitionum, comes rei
privatae, praetorian prefect (when present), comes domesticorum, comes excubitorum,
magister militum praesentalis, along with other former and titular holders of offices, as well
as those who were not formally members of the senate but whom the emperor wished to
consult. See Jones, LRE, vol. I, pp. 333ff. By the early seventh century, the sacellarius
seems also to have played a key role: he is referred to as the highest-ranking of the
senators present at the interrogation of Maximus Confessor (see PG XC, 88C; 101A-B;
113A-B). A description preserved in the tenth-century Book of Ceremonies (De Cerimo-
niis), 628.10-14 for the year 638 describes a senatorial procession of the endoxotatoi
patrikioi, along with those of apo hypaton rank and others of senatorial grade down to
simple illustres.
114
See Jones' commentary, LRE, vol. II, pp. 5 5 Iff.; and for evidence of continuity, especially
in the West, see Arnheim, Senatorial Aristocracy, pp. 103ff. and 155ff.
 Rural society 165
senatorial offices does not change dramatically through much of the
seventh century. The titles of patrikios, hypatos (consul) and apo hypaton
(ex-consul) continue to be awarded by the emperor or claimed by office-
holders of the appropriate rank. The development is complicated, how-
ever, by the increase in importance and status of titles conferring member-
ship of a number of palatine ordines, both civil and military, which are
used together with the traditional titles of rank as well as with titles of
offices. Concomitantly, the value of several titles and their status seems to
decline gradually during the seventh century. Most obvious is the title of
stratelates or magister militum, originally designating one of the leading
military commands, already by the reign of Justinian awarded to relatively
humble commanders of contingents, both on an active, and a titular and
honorary basis.115 During the seventh century, the title oistratigos came
to be applied to the actual commanders of the provincial forces - the
themata - and this further depressed the value of stratelates, which now
signified no active duties or competence. Like stratelates, the general
descriptive terms for members of the senatorial establishment - gloriosus/
gloriosissimus, or endoxotatos in Greek - continue in use, to define the
members of the Constantinopolitan senate as determined by Justinian in
novel 62. 116 But the fact that titles and the system of precedence
demonstrate a great deal of continuity tells us little, if anything, about
continuity in the senatorial class itself - whether defined in the narrower
sense or in the more general meaning of all the clarissimi and spectabiles
as well.
A hint may be supplied by looking at the names of those mentioned in
the various sources who occupy positions of power, and by looking at the
relationship between names, titles and offices.
To begin with, even a superficial prosopographical survey of the period
shows a marked increase in the number of non-Latin and non-Greek
names among palatine officials, imperial advisers, and especially among
military personnel. Charanis has shown how central a role was played by
Armenians at this time, for example, who already from the later sixth
century seem to have become increasingly important. By the end of the
reign of Heraclius, and during the reigns of his son Heraclius Constantine
and grandson Constans II, members of Armenian noble families occupy
key positions in both the central and the provincial military administration

115
For example, the magister militum vacans Martinus, in 53 5; and the three magistri militum
vacantes under the command of Belisarius in Sicily in the same year. See Stein, Bas-
Empire, vol. II, pp. 323-4 and note 1; p. 430 and note 3.
116
For a more detailed analysis of titulature, see chapter 10 below.
166 Byzantium in the seventh century
of the empire, often endowed with titles of senatorial rank.117 But persons
of Iranian, Slav and Germanic origins were also to be found.118
At the same time, while there seems little reason to doubt that many of
the leading officers and persons mentioned in the sources in this connec-
tion did hold senatorial dignities, whether or not this was made explicit,
the position and role of the senate changes over the period under examin-
ation here. Traditionally, the senate had played, since the fifth century, a
central role in the selection and ratification of a new emperor - and during
the events surrounding a succession in general. It has been shown that this
was the case during the fifth and sixth centuries, as well as in the events
surrounding the election of Heraclius and his immediate successors.119
And while, according to Justinian's novel 62, the senate had by that time
entrusted its executive functions to the emperor,120 it is clear that the
senate as a body, made up in part at least of active officials of state of
considerable power and wealth, along with other members of the landed
wealthy elite of the empire, continued to exercise a very great de facto
authority. Those who were included among the group of leading advisers
to the emperor had formed a cabinet within the older consistory and were
referred to as the proceres sacri palatii. They represented the real power at
the court. The senate and the consistory themselves - which from Justi-
nian's time were effectively equivalent bodies with a shared membership -
had few opportunities formally to intervene. Only at times of crisis, such as
when a problem over the succession arose, for example, were they able to
function as an independent body, usually to represent, as well as their own
117
See P. Charanis, 'The Armenians in the Byzantine empire', BS 22 (1961), 196-240, esp.
2O5f. and 'Ethnic changes', 32ff. See also W. Seibt, Die byzantinischen Bleisiegel in
Osterreich I, Kaiserhof (Vienna 1978), p. 165; Winkelmann, Quellenstudien, pp. 2O3ff.
and literature at p. 230 note 849. But note the more reserved position of S. Gero,
'Armenians in Byzantium: some reconsiderations', Journal of Armenian Studies 2 (1985),
13-26.
118
See Winkelmann, Quellenstudien, pp. 199ff. (mostly eighth century or later); H. Ditten,
Prominente Slawen und Bulgaren, pp.lOOff.; R. Guilland, 'Les Patrices byzantins de la
premiere moitie du VIIe siecle', Topcx; eis fiurjfirjv K. I. A/naPTOv (Athens 1960),
pp. 11-24 (repr. in R. Guilland, Recherches sur les institutions byzantines (2 vols., Berlin
and Amsterdam 1967), pp. 162-9) and 'Patrices de Constantin IV a Theodose QT,
Hellenika 23 (1970), 2 8 7 - 9 8 (repr. in Titres etfonctions VIII).
119
See the detailed studies of Ch. Diehl, 'Le Senat et le peuple byzantin aux VIIe et VHT
siecles', B 1 (1924), 201-13; H.-G. Beck, Senat und Volk von Konstantinopel Probleme der
byzantinischen Verfassungsgeschichte, in SBB (1966), pp. 1-75 (repr. in Ideen und Realitd-
ten in Byzanz XII (London 1972).
120
Justinian, Nov. 62, proem: 'Antiquissimis temporibus Romani senatus auctoritas tanto
vigore potestatis effulsit, ut eius gubernatione domi forisque habita iugo Romano omnis
mundus subiceretur, non solum ad ortus solis et occasus, sed etiam in utrumque latus
orbis terrae Romana dicione propagata ... Postea vero quam ad maiestatem imperato-
riam ius populi Romani et senatus felicitate reipublicae translatum est, evenit ut ii, quos
ipsi elegerint et adminstrationibus praeposuerint, omnia facerent quae vox imperialis eis
iniunxisset...'
 Rural society 167
interests, those of the population of Constantinople and the ruling adminis-
trative circles.121 The occasions on which senatorial intervention was
important during the sixth century are illustrative: the selection and
acclamation of a successor to Anastasius in 518; during the Nika riot in
532; during the reign of Maurice, when the senate formally advised the
emperor to treat with the khagan of the Avars; and before and during the
revolution which led to the downfall of Phocas.122 In addition, of course,
the senate was present and played a formal role at all accessions and
acclamations; but its power as a body was, on the whole, less important
than the power exercised by its members as individuals within the
government.
In the seventh century this general pattern is repeated. The senate
appears to have played a formal role as one of the three constitutive
elements in ^very imperial election (senate, people and army): in the
accession of Phocas,123 of Heraclius,124 of Heraclius Constantine - in the
last case the Armenian chronicler Sebeos emphasises the fact that Hera-
clius placed his son's future in the care of the senate, a fact which may
suggest the reason for the central role played by this body in the events that
followed, and hints also at the differences between the various factions
surrounding Heraclius on the one hand, and his wife and niece Martina on
the other.125 In the confused events following the death of Heraclius
Constantine, the deposition of Martina and Heracleonas, the rebellion of
Valentinus and the accession of Constans II as sole ruler, the senate again
plays a crucial role and is, according to the words of the young Constans,
reported by Theophanes, the adviser to the emperor and the protector of the
people. Sebeos also notes the central role of the senate and lays stress on the
fact that it was the latter body which worked against the plans of the
general Valentinus to have himself confirmed as co-emperor.126 Heraclius
seems to have placed a great deal of value on the senate and its support,
both in the proceedings against Priscus, the son-in-law of Phocas (and a
member of the senate himself) and in his dealings with the Avars.127 The
121
See Jones, LRE, vol. I, p. 3 2 5 .
122
See the brief summary of Tinnefeld, Die fruhbyzantinische Gesellschaft, pp. 77ff. with
references and literature; Jones, LRE, vol. I, pp. 267f. and 27If.; Stein, Bas-Empire, vol. n,
pp. 219ff. and 449ff.; and Diehl, 'Le Senat et le peuple byzantin' (cited in note 1 1 9
above). For the embassy to the khagan, see Theophylact Simocatta VII, 15.8; Theo-
phanes, 279.2Osq.
123 124
Cf. Theophylact Simocatta VIII, 10.2 and 3.15. Nicephorus, 5 . 1 4 - 1 6 .
125
Sebeos, 67, 8 0 . Sebeos' reliability is, of course, open to question, writing as he was well
away from the scenes he reported.
126
Theophanes, 331.3sq., 3 4 1 . 2 4 - 7 , 342.15sq.; Sebeos, 105.
127
Nicephorus, 6 . 9 - 1 1 ; see A. Pernice, L'imperatore Eraclio, Saggio di storia bizantina (Flo-
rence 1905), pp. 48f.; Stratos, Byzantium in the Seventh Century, vol. I, p. 100; and Chron.
Pasch. 706.17sq.
168 Byzantium in the seventh century

interrogations of both Pope Martin and of Maximus Confessor were carried

out before the senate, and it was senators who continued the examinations
afterwards - the patrician and sacellarius Troilus and the patrician Epi-
phanius played a major role as intermediaries. 128 Some continuity of
family and position is suggested in the account of the trial of Maximus,
where Theodore, the son of John the candidatus and brother-in-law of
Platon the patricius is brought as a witness against Maximus. 129 Sergius,
the epi tes trapezes, is also a leading member of the senatorial opposition.
Similarly, the patrician Troilus may be the father of Andreas, the member
of the retinue of Constans II responsible for his death in 668; while the
patricius Justinian, the father of the later patriarch Germanus, was also
involved in the same plot. 130 During the reign of Constantine IV the senate
seems to have supported the emperor in opposing the wishes of the soldiers
of the Anatolikon district that the emperor's brothers should be equal
co-emperors with him; 131 and again under Justinian II the senate sup-
ported his preparations against Cherson. 132 The senate seems also to have
supported the attempt of Philippicus Bardanes to reintroduce monothele-
tism; but it was equally involved, although to what extent is difficult to say,
in the deposition of Philippicus and the accession of Anastasius II. 133 The
senate, or members of it, may well have been behind the rebellion of
Leontius against Justinian II in 695; and it was the senate, together with
the city troops, which in all probability called upon Theodosius in to
abdicate in favour of Leo III in 717. 1 3 4
As a formally constituted body, therefore, made up of the chief palatine
officials, both civil and military, the senate seems to have continued to play
the role which it had in the sixth century throughout the seventh and well
into the eighth century. This is hardly surprising, since it included all the
most powerful advisers and officials of the imperial administration in

128
See PG XC 88C; 113A-B: the sakellarios Troilus ranked highest a m o n g the senators. For
possible seals, see Seibt, Bleisiegel, n o . 1 3 2 , a seal of Troilus patrikios; a n d Zacos a n d
Veglery, no. 3 0 6 1 , seal of Troilus, apo hypaton. See also PG XC 101A-B (Troilus and
Epiphanius, both patrikioi). Other senators of consular r a n k n a m e d include Paul and
Theodosius (PG XC, 96D; a n d see 109D, 113A-B, 137B).
129
PG XC 113C. See also note 1 3 9 below.
130
PG XC, 120B. For Andreas, the son of Troilus, see Seibt, Bleisiegel at no. 132; Theo-
phanes, 3 5 1 . 2 9 ; a n d P. Peeters, 'Une vie grecque du Pape S. Martin', AB 51 (1933), 2 5 9
and note 1. For the patricius Justinian, see Guilland, Recherches, vol. 2, p. 1 6 6 .
131
Theophanes, 3 5 2 . 1 9 s q . See Diehl, 'Le Senat et le peuple', 2 0 8 ; also Michael Syr. XI, 13
(vol. II 455f.).
132
Nicephorus, 4 4 . 1 6 - 2 0 .
133
Nicephorus, 4 8 . 1 2 - 1 4 ; Zonaras XIV, 2 6 ; A g a t h o n D i a c , 1 9 2 . See Beck, Senat und Volk,
p. 3 1 .
134
For Theodosius III see Theophanes, 390.20sq.; Nicephorus, 52.15sq. On these events,
see Beck, Senat und Volk, pp. 3 Iff.
 Rural society 169

Constantinople. 135 And that some degree of continuity of family and

personnel existed is suggested by the few examples mentioned just now. 136
But the increasing number of newcomers to the ranks of the senate as
imperial officials in the reigns of Justinian II and his successors especially
suggests that by this time the relative proportion of 'old' senatorial families
to 'new' members was shifting in favour of the latter. As I have said,
membership of the clarissimate had been hereditary; that of the higher
grades - spectabiles, illustres and gloriosi - more limited, although evidence
for imperial concessions of a father's rank and title to his son upon formal
request exists, at least for the sixth century, as Jones has shown. 137 Even
without this element, however, a relatively high degree of continuity might
be expected, given the wealth, power and contacts of the senatorial
establishment in Constantinople and its richer associates, certainly up until
the 650s, as our evidence might suggest.
The names of those with the title patrikios collected by Guilland for the
years up to 668 compare and contrast interestingly with those for the
following period up to 717. Of forty named patrikioi known from 602 to
668, only three have non-Greek or non-Latin names, two of which are
Persian and one Armenian. For the period to 717, however, thirty-four
named patrikioi are known, of whom three have Armenian or Caucasian
names, and eleven are given epithets - such as Myakios, Rizokopos,
Bouraphos, Pitzigaudes and so on. Guilland's list and figures can be
supplemented from the seals now available, and they do not appreciably
alter this picture. This evidence suggests the arrival of 'outsiders' on the
scene, whose advancement is a result of imperial favour and promotion; as
well as the need to differentiate among them by the use of such descriptive
terms. It suggests, in effect, the increasing importance of men from outside
the older establishment and the older circle - however large it may have
been - of senatorial families.138 Of course, it provides only the crudest of
135
All those with the title of patricius will have been senators; in addition, the leading
palatine officers were members (see note 1 1 3 above), and by the early eighth century
some of their staff as well. There is no reason to doubt, for example, that the various
officials involved in the interrogation of Maximus were also senators (see, for example,
PG XC, 88C: the sakellarios; and 120B: the e m TTJS Tpaire^Ti? TT\<$ $a(ri\ixr\s, and note
2 2 6 above); while the officials present at the sixth council in 6 8 0 (Mansi XI 209ff.:
master of offices, count of the Opsikion, military logothete, curator of the imperial House
of Hormisdas, the lieutenant of the count of the excubitors, quaestor, dioiketes of the
Eastern provinces, domestic of the imperial table and five others with no named
functions) will all have been endoxotatoi and senators. In the proem to the Ecloga of 741
the quaestor and his staff are counted as endoxotatoi, the former being a patrikios, the
latter merely hypatoi (see Ecloga, ed. Burgmann, 104f., 1 6 2 . 4 0 - 2 ) .
136
See Haldon, Byzantine Praetorians, pp. 156f and 158f.
" 7 Jones, LRE, vol. II, pp. 5 2 9 - 3 0 .
138
The lists are to be found in the articles of Guilland cited in note 118 above; and see the
comments of Winkelmann, Zum byzantinischen Staat, pp. 204ff., and 'Probleme der
170 Byzantium in the seventh century

guides: many newcomers will certainly have borne traditional Greek or

Latin names, while some of the epithets - such as strouthos, for example,
the sparrow - may simply have been popular nicknames. But the com-
parison is suggestive, in spite of the methodological uncertainty. 139 And it
is important to bear in mind also that this development seems to represent
only the opening stage of the genesis and formation of a new ruling class,
marked during the later eighth and ninth centuries in particular by the
appearance of some 150 family names - nomina gentilia - and the clear
appearance in the sources of an elite whose social and political power is
based upon both imperial service, as well as clan property in land. In this
gesellschaftlichen Entwicklung im 8. und 9. Jahrhundert', in 17 th International Byzantine
Congress, Major Papers, pp. 577-90, see 58If. and note 34. The evidence of the names
used on seals, which represents a much wider variety of offices, is also telling: there is an
increase in the number of names of non-Greek and non-Latin origin, through the seventh
and into the eighth century, which, while hardly conclusive, is evidence of their
introduction from a variety of sources (particularly Armenian) into the Greek cultural
milieu. Names such as Vasakios (Seibt, Bleisiegel, no. 1335; Zacos and Veglery, nos. 721,
1089, 1672), Vaanes (Zacos and Veglery, nos. 1086, 1671, 2569), Arazat (Zacos and
Veglery, no. 59OA), Arsakios (Zacos and Veglery, no. 591), Mouselios/Mousilios (Zacos
and Veglery, nos. 662, 946, 947), Arsaphios (Zacos and Veglery, nos 1115, 1420),
Rostom (Zacos and Veglery, nos. 2343, 3046), which illustrate the degree of Armenian
influence - as well as those such as Baianos (Zacos and Veglery, no. 594), Tourganes
(Seibt, Bleisiegel, no. 199) or Oulid (Zacos and Veglery, nos. 1566, 1567, 3042), of
Turkic and Arabic origin respectively - demonstrate the penetration of the early Byzan-
tine establishment, both military and civil, by 'new' elements on a considerable scale, and
in stark contrast with the sixth century, when such elements - whether Germanic, Turkic
or Iranian - were limited for the most part to the military. See also Winkelmann,
Quellenstudien, pp. 146ff. on nick-names which became nomina gentilia, and pp. 197ff. on
names of non-Greek origin.
139
Equally tantalising is the appearance of seals in the later seventh and eighth centuries of
individuals, sometimes with their title (usually hypatos, apo hypaton, patrikios, but inclu-
ding also those of kandidatos and spatharios) along with an expression of a familial
relationship: son of . . . (e.g., Zacos and Veglery, nos. 363, 388, 370, 531, 782, 879,
1827, 2229, 2336, 2641, 3040, 3117, and possibly nos. 2126, 2060, and 1460).
Might these represent the efforts of older senatorial families to emphasise their lineage? Or
of persons whose fathers were newly established (and thus in their time well known),
among the elite of the state establishmentPThe question has been addressed by J. Nesbitt,
'Double names in early Byzantine lead seals', DOP 31 (1977), 111-21, see 109ff. and by
Winkelmann, Quellenstudien, pp. 144-6, with further examples from the eighth century,
from literary sources. On the whole, such references would seem to back the second
rather than the first suggestion, depending as they do on the person or persons named on
the seal being known to the reader. Of course, as Winkelmann's detailed treatment has
demonstrated, the generalisation from the specific 'son of to 'of the line/family of (which
is what the phrase 6 x a r a TOV/TOVS most probably is meant to convey, cf. Quellenstu-
dien, pp. 15 If.) is evidence for the establishment of specific families with an awareness of
their lineage identity. But the great majority of the known examples seem to represent
families evolving in the eighth century and after - there is no real evidence for continuity
here. Interesting is the possibility that a seal of John, kandidatos, son of Hadrian (dated c.
600-50) may have belonged to the John, kandidatos, nicknamed Chilas, whose son
Theodore, the brother-in-law of the patrikios Plato, was a witness during the interroga-
tion of Maximus (PG XC, 113C). For the seal (the reading of which is, however,
problematic) see Zacos and Veglery, no. 370.
 Rural society 171
respect, the later seventh century seems to mark a real transformation,
with the subsumption of the older establishment into the nascent service
aristocracy of the middle Byzantine period.140
The sum of this evidence suggests, therefore, that the older senatorial
establishment, in the narrower sense as defined above, survived, in spite of
some confiscation during the reign of Justinian after the Nika riot and in
spite of the supposed mass execution of many senators during the reign of
Phocas, more or less intact into the seventh century.141 It continued to
dominate the administration of the state as a class; but, as the dramatic
changes of the second half of the seventh century began to take their toll, it
became increasingly populated by newcomers whose arrival served to
strengthen its essential character as an elite of service rather than of birth
or cultural tradition, and to weaken the cultural-political hegemony of the
older aristocracy. That some families may have survived is quite probable,
of course, and an element of continuity in landed property and family
ideology is not to be excluded, especially in areas least affected by the
physical disruption of the times. Such a family may have been that of
Botaneiates, a name which occurs as early as the sixth century, although
the family (if it is the same, and not a coincidence) is best known from the
eleventh century.142 It is unfortunately impossible to generalise either in
favour of or against continuity from this single example; it is difficult to
know whether the limited sources reflect the historical situation or their
own limitations.
In terms of the social relations of production in the empire as a whole,
therefore, the seventh century marks the emergence of a new factor within
the ruling class, formed from among the members of the provincial and
Constantinopolitan service elite. It also witnessed the beginnings of a
process of assimilation between the two groups dominating the social and
political life of the state, the older 'senatorial' establishment and the newer
'service' elite. I will discuss some of the implications of this in chapter 10
below. Large estates continued to exist, probably on a very similar scale to
that of the sixth century; but the growth in long-term leases on estates at
the expense of the older colonate, together with increased mobility among
the rural populations of the empire, a change in the system of tax-assess-
ment and collection, and the increased availability of land and the arrival
140
In this respect, Winkelmann's survey in Quellenstudien of the names for the eighth and
ninth centuries is extremely important, and it illustrates a caesura in the nomenclature of
the elite between the late Roman and middle Byzantine periods very clearly. See esp.
pp. 143ff.: 'Symptome und Tendenzen der Konsolidierung einer Magnatenschicht'.
141
For confiscations after 5 3 2 , see Procopius, Historia Arcana XII, 12f.; and for the events of
the reign of Phocas, see above, and Tinnefeld, Die fruhbyzantinische Gesellschaft, pp. 9 7 - 9 .
142
See G. Buckler, 'A sixth-century Botaneiates?' B 6 (1931), 4 0 5 - 1 0 , and Winkelmann,
Quellenstudien, pp. 158 and 1 8 1 .
172 Byzantium in the seventh century

of imported labour power, all promoted the importance and the numbers of
communities of independent peasant smallholders subject directly to the
fisc. By the early eighth century the sum total of all these changes and
developments adds up to a very different set of relations of production
within the Byzantine world of Asia Minor and the Balkans.
 CHAPTER 5

The state and its apparatus: fiscal

administration

THE LATE ROMAN SYSTEM

In chapters 3 and 4, I outlined the chief developments in the relations
between cities and their hinterlands on the one hand, and between land-
owners and the rural agricultural population of the empire on the other.
These developments provide the backdrop to the subject of this chapter and
the next, that is, the administrative structures of the Byzantine state, its
mode of revenue extraction and distribution, and its military machinery.
For the fiscal and military establishments form the two central and inter-
related elements of the state administrative apparatuses which secure both
its physical defence and its political and institutional reproduction. Con-
tingent upon these are the mode of civil provincial administration and the
conduct of foreign policy, and it is to these areas that I shall turn in this
chapter.
Let me begin, therefore, with the machinery of revenue extraction. The
changes that took place in the organisation of this aspect of state admin-
istration during the seventh century were considerable; and the best way
to illustrate this, and then to attempt to demonstrate the causes of change
and their process, will be to sketch out briefly the systems of the later
Roman period (up to about 600) and of the Byzantine period (of about
800-900), respectively.
During the sixth century the extraction and administration of revenues
had been organised under the auspices of three departments: the praeto-
rian prefectures, the comitiva sacrarum largitionum and the res private. The
res privata was responsible for the administration and collection of rents
from state lands; it included all lands which lapsed to the state, as well as
lands which had been inherited from the previous rulers of districts as they
had been incorporated into the empire. Properties bequeathed to the
emperors, the family property of those who had occupied the imperial
throne, former public and civic lands, estates confiscated for one reason or

173
174 Byzantium in the seventh century

another, all these lands were administered by this department, whose staff,
therefore, was considerable. Often, large tracts of whole provinces - as, for
example, much of Cappadocia I or Bithynia - belonged to the imperial
domain. The administration of this property was headed by the comes rei
privatae. His bureau (officium) was divided into a number of subdepart-
ments or scrinia, each responsible for specific functions: grants of lands,
rents, leases and so on. In each diocese and province were provincial staffs
to collect and supervise rents, and to administer the various consolidated
or dispersed estates. The revenues drawn from this source were employed
in the first instance in the maintenance of the imperial household, being
dispensed probably through its treasury, the scrinium largitionum privata-
rum. But the emperor could also draw upon the res private as a source of
reserves for purposes of state or the granting of largesses.
Anastasius formalised this arrangement by setting up a new depart-
ment, separate from the privata, called the patrimonium, at the disposal of
the emperor and subordinate to the department of the sacrae largitiones (see
below) - the privata continuing to be used as a source of state and fiscal
income. Already in the later fourth century, the estates in Cappadocia (the
domus divinaper Cappadociam) had been transferred directly to the charge of
the praepositus sacri cubiculi, the palatine official directly responsible for the
imperial household and its administration. Under Justinian, further
marking-off occurred, with the result that the res privata was divided by
566 into some five branches: the original res privata, the patrimonium, the
domus divina per Cappadociam, the domus dominicae (two sections, each
under curatores, administering estates whose revenues were allocated
apparently to the personal expenses of the rulers) and the patrimonium
Italiae under its comes (following the reconquest of Italy: it seems to have
been constituted of land confiscated from the Ostrogothic nobility and
rulers, as well as of older imperial lands). 1
1
For a full and detailed treatment, see Jones, LRE, vol. I, pp. 412-27; Stein, Bas-Empire, vol.
I, pp. 174 and 341; vol. II, p. 67 and note 1, pp. 423ff., 472-3 and 748ff.; note also J.B.
Bury, The Imperial Administrative System in the Ninth Century, With a Revised Text of the
Kletorologion of Philotheos (British Academy Suppl. Papers I, London 1911), p. 79, who
assumes the curator mentioned in a novel of Justin II {JGR I, coll. 1, nov. 1; a. 566 = Dolger,
Regesten, no. 4) to be an independent official in charge of both the Cappadocian lands and
the domus dominicae, a forerunner of the later megas kourator (cf. Oikonomides, Preseance,
p. 318). According to a novel of Tiberius Constantine {JGR, I, coll. 1, nov. 12; a. 578-82 =
Dolger, Regesten, no. 67) the domus divinae included also the patrimonium for practical
administrative purposes; and although several curatores are referred to in this novel, there
is every reason to suppose that the system was again centralised during the seventh
century - possibly before the Arab wars, possibly as a result of them. There may in any
case, as Bury supposes, have been several curatores under the disposition of a (megas)
kourator at Constantinople, perhaps that in charge of the estates of Hormisdas. See Jones,
LRE, vol. I, p. 426; Stein, Studien, p. 98 and note 7: the curator of the estates of Hormisdas
was clearly one of the leading officials in charge of imperial properties. See also the
 Fiscal administration 175

The department of the sacrae largitiones administered bullion from gold

and silver mines, as well as the mines themselves, together with the mints,
the workshops where parade armour was decorated with precious metals,
the production and collection of clothing (for historical reasons) and the
state dye-works, During the fourth century the armour-decorating work-
shops were transferred to the supervision of the magister officiorum. The
count of the sacred largesses (comes sacrarum largitionum) also issued the
donatives in silver and gold which the soldiers received. As with the res
privata there was a central bureau with provincial staffs. The central
offlcium consisted of ten departments or scrinia, responsible for general
clerical matters, revenues, accounts, messengers, military donatives,
bullion, diocesan reserves of gold, and mints; as well as the imperial
wardrobe, silver and silver coin, palatine silversmiths, copper, and (origi-
nally) arms and armour decorators. Each diocese had a representative
from the department, and there were a number of regional treasuries
where gold, silver and finished clothing could be stored. The mints were
supervised from these points. In addition, the sacrae largitiones had repre-
sentatives in the cities, responsible for administering the revenues drawn
from confiscated civic lands; it also had its own transport service (like the
res privata); and it was responsible for the controllers of foreign trade, the
comites commerciorum. The income of the largitiones derived from the
surviving cash taxes in gold and silver, customs and import duties, the
donations made by cities and the senate on the occasion of an imperial
accession (the aurum coronarium and aurum oblaticium) and quinquennial
celebrations, as well as the commutations received in lieu of recruits, the
aurum tironicum, and of military horses, each rated at about 25-30 gold
solidi. A cash assessment on land may also have provided some funds,
although the rate of assessment is unknown, as is the extent of application
of such an assessment.2
The mints were probably the most important element in the administra-
tive machinery of the largitiones. They were run by procuratores and staffed
by monetarii, who constituted a hereditary group from the fourth century,
often quite wealthy, although by origin they were mostly imperial slaves.
Following the reconquests of Justinian's reign, there were mints at Con-
discussion of D. Feissel, 'Magnus, Megas et les curateurs des "maisons divines" de Justin II a
Heraclius*. TM 9 (1985), 465-76, esp. 474f.; and M.Kaplan, 'Quelques aspects des
"maisons divines" du VIe au IXe siecle', in Melanges N. Svoronos (Paris 1986), pp. 70-96.
For the scrinium largitionum privatarum, see Notitia Dig., occ. XII, 4. Bury surmises {Admi-
nistrative System, pp. 81-2) that part of the income from this source also went into the
emperor's private treasury, the sacellum. See below.
2
See Jones, LRE, vol. I, pp. 427-34; Stein, Bas-Empire, vol. II, pp.426f., 761f. and 766-9;
J.P.C. Kent, in E. Cruikshank-Dodd, Byzantine Silver Stamps (Dumbarton Oaks Studies VII
Washington D.C. 1961), pp. 3 5ff.; and the contribution of R. Delmaire, in Hommes et richesses
1.265-77.
176 Byzantium in the seventh century

stantinople, Thessaloniki, Carthage and Ravenna, the last two acquired by

reconquest, all producing gold coin; and at Nicomedia, Cyzicus, Antioch,
Alexandria (and temporarily also at Rome, Salona, Carthagena, and
Cherson) producing copper. During the reign of Maurice, a mint was set up
in Sicily, at Catania; and during that of Heraclius, temporary mints were
established at Constantia in Cyprus and at Seleucia in Isauria (later at
Isaura itself) (from 582/3, and in 615 to 618 respectively).3 The mints
themselves were organised at the diocesan level, and although this had not
been the original intention, since several of the mints actually pre-dated
the reforms of Diocletian and Constantine through which the dioceses were
established, the result in the sixth century was that the major dioceses had
each one mint for copper; while the minting of gold was limited to
prefectural capitals or their equivalents, that is, Constantinople, Ravenna
and Carthage.
The bullion for the coinage came from levies in kind where the land was
ore-bearing, from mining and panning, from confiscated plate (not always
inconsiderable amounts) and, chiefly, through the recycling of gold coin
via the fiscal machinery of the state back to the largitiones. From Valens
and Valentinian onwards gold coin was melted down in the provinces
before being shipped back to the capital, or the imperial comitatus, in
bullion form for re-minting. The export of gold was strictly prohibited, as
was the clipping of any coin, in particular gold. By these means the state
tried to recover the gold it issued in coin, although inevitably it was not
entirely successful. But as Jones notes, the frequent re-minting of the gold
must have constituted a crucial factor in the maintenance of the weight
and purity of the solidus.4
In spite of their sizeable incomes and their functions, both the sacrae
largitiones and the res privata were over-shadowed by the praetorian
prefectures. The prefecture comprised a specific territorial area, subdivided
into dioceses, under vicarii, and then into provinces under local governors
with a variety of titles (moderatores, rectores and so on), each comprising a
group of civitates. The office of the prefecture was responsible for the corn
supply for the capital(s) and certain other cities, as well as for the costs
incurred in transporting this by the guild of state shippers or navicularii; it
was responsible for the rations or ration allowances, in kind or in cash, for
the army, civil service, and related departments of state, as well as the
3
Jones, LRE, vol. I, pp. 374 and 437ff.; A.R. Bellinger, Catalogue of the Byzantine Coins in the
Dumbarton Oaks Collection and in the Whittemore Collection, vol. I, Anastasius I to Maurice,
491-602 (Washington D.C. 1966), pp.64f.; M.F. Hendy, 'On the administrative basis of
the Byzantine coinage, c. 400-900, and the reforms of Heraclius', Birmingham University
Historical Journal, 12/2 (1970), 129-54, see 140ff.; Ph. Grierson, DOC II, 1, pp.209 and
327-30.
4
Jones, LRE, vol. I, p. 43 5f.
 Fiscal administration 177

fodder allowance of officers and troopers in cavalry units and their civilian
counterparts. It was also responsible for the maintenance and operation of
the public post; for that of the state arms workers along with their
materials (although not for the arms factories or fabricae, which came
under the authority of the magister officiorum); and for the public works of
the provinces, levying labour and materials from local communities for
roads, bridges, granaries, post stations and so on. From Diocletian's time
the rate of assessment in kind (later commuted into cash) and the estimates
needed to meet this wide range of needs had been systematised into a single
general levy, raised each year, referred to as the indictio. The rate might
vary from year to year, however, according to need, and the assessment
was at least in theory not a fixed tariff. As the greater part of this
indictional levy was commuted to gold during the last years of the fifth
century, in the East, so the prefectures were able to build up reserves to
meet exigencies and thus, to a degree, make the maintenance of a single
established rate possible.
Until Justinian's reign, there were two prefectures in the eastern part of
the empire, that of Illyricum and that of Oriens, the former based at
Thessaloniki, the latter at Constantinople. The size and wealth of the latter
gave it a pre-eminent position. During the reign of Justinian, the prefec-
tures of Italy (based at Ravenna) and Africa (based at Carthage) were
re-established (in 537 and 534 respectively), and as we have seen, the
quaestura exercitus was also set up. Map VI illustrates the diocesan and
provincial composition of the empire c. 565. 5
Since it is of some importance for an understanding of the development
and origins of the later provincial and central administration, it is worth
pausing for a moment to examine in greater detail the structure of the
prefecture of the East. Each diocese was represented by a scrinium
(although Egypt, which was otherwise an independent diocesan district
under the Praefectus Augustalis, was included in the scrinium for Oriens, to
which it pertained); there was further a scrinium for Constantinople, for
public works, for arms (the assessment and collection of materials for the
state factories) and other departments for the purchase of corn for Con-
stantinople and other cities; for the assessment and collection of the
annonae (rations) for the troops (referred to as the stratiotikon); and there
was also a prefectural treasury or area, divided into two sections on
account of the size and complexity of the fiscal operations of the Eastern
prefecture, the general and the special bank (genike trapeza, idike trapeza).
The special bank was responsible for the assessment of the collatio lustralis,
the tax on merchant and craft sales introduced by Constantine I, but

5
See Jones, LRE, vol. I, pp. 370-2 and 448f. For map VI see p. 228 below.
178 Byzantium in the seventh century

abolished by Anastasius. The other prefectures had a similar, if less

ramified, bureaucracy. The diocesan scrinia were headed each by two
numerarii, assisted by an adiutor and a chartularius or accountant; the staff
consisted of tractatores or trakteutai, each responsible for the accounts of
one province, assisted in turn by a clerical member of staff who dealt with
correspondence. Supervisory staff were despatched to each province to
advise on the collection of the indictional assessment and to help with the
extraction of arrears, the latter referred to as compulsores or expelleutai.
The calculations for the indiction depended upon returns submitted by
the competent state officials: those responsible for the military, civil and
administrative staffs of all branches of the state apparatus. While allow-
ances were fixed, yearly requirements often varied according to circum-
stances, so that the indiction was not always adequate. In this case,
superindictiones on the whole prefecture, or on a single diocese or province,
could be imposed to raise the necessary shortfall in supplies or cash. In
addition corvees, or munera sordida (such as providing bread for the troops
quartered in the district, supplying stone or timber, craftsmen and labour
for public works, billetting and providing animals for officials of the state
and so forth), were also levied, although certain privileged groups were
exempted from such duties, such as senators, for example. In short, the
assessment of tax was based on the principle of a distributive allocation:
the state calculated its needs first, and then proceeded to assess tax-payers
at a rate which would, as far as could be predicted, ensure the meeting of
those targets. 6
The actual assessment of the indiction was based upon the capitatio/
iugatio calculation, described in chapter 4, although the rates varied from
province to province with regard to both the size and the number of units
of land and labour power in question. Thus the total needs for each
province would be calculated in accordance with the number of capita and
iuga, together with a variety of other factors regarding the economic
situation of the territory in question, the amount of arrears that might be
owed and so forth. The prefecture then circulated the assessment to the
diocesan vicarii and provincial governors. From then on the municipal
officers appointed by the town councils were responsible for the issue of tax-
demands to tax-payers, and for their collection by the curial susceptores/
hypodektai and procurators/epimeletai. The vindex appointed from the time
of Anastasius to each city was effectively a collection-manager and tax-
farmer, who continued to employ the curial collectors, but who was
responsible directly to the prefecture. While unpopular, they were appar-
ently very efficient, and this system was retained through Justinian's
6
Jones, LRE, vol. I, p. 449 and pp. 450-3. For the difference between distributive and
contributive modes of assessment, see Oikonomides, 'L'Impot de distribution', 9-11.
 Fiscal administration 179

reign. 7 But the tendency of provincial governors to appoint their own

officials to govern cities on the one hand, a development which, although
frowned upon by the emperors, led to the eventual disappearance of the
town councils as governing bodies, and on the other hand, of the vindices to
take over the administration of the city revenues, as well as the imperial
taxes, meant that by the time of Justinian, and certainly by the end of the
sixth century, the cities had suffered an almost complete loss of civic
autonomy in both local administrative and fiscal matters - even though
the curial order continued to exist, and its members were still called upon
to collect the taxes and to bear personal liability for their return. Parallel to
this, of course, the state also delegated the collection of taxes to estate
owners, including the Church, a practice which saved it (or the local
provincial governor) the cost of organising and supporting the fiscal
officials who would otherwise have been responsible for this task. 8
The greater part of these taxes had, until the later fifth and early sixth
centuries, been collected in kind and then distributed by local officials of
the prefecture to the various points of consumption - chiefly military units
and centres of the civil administrative establishment. From this time on,
however, the bulk of the assessment was commuted into gold, although at
rates which varied over the prefectures, usually according to local con-
ditions and costs. From the reign of Justinian there is some evidence that a
standardisation was taking place, and it seems likely that by the end of the
sixth century this process had been completed.9 Military supplies could still
be demanded in kind, of course, and the coemptio, or compulsory purchase
of corn, was a regular imposition in some areas. Part of the land-tax seems
also still to have been raised in kind for distribution to the army. 10
As will by now be clear, the three departments responsible for revenue
collection were, at various points, closely related. The income of the sacrae
largitiones was normally collected by the susceptores and similar officials
appointed by the cities to collect the prefectural taxes; the rents of the lands
of the res privata were often collected by officials of the provincial gover-
nors, supervised by officials deputed from Constantinople.11 At the same
time, the diocesan level of prefectural administration seems certainly by
Justinian's reign to have become less important and relevant to the civil,
and to some extent the fiscal, administration of the state, and indeed may
have ceased to function as an effective intermediate instance of govern-
ment. The territorial prefecture with its constituent provinces made up the
7
Ibid., pp. 453-8.
8
Ibid., vol. II, pp. 759f.; and chapter 3 above, pp. 95-9. For the procurators in the later
sixth and early seventh centuries, and the delegation of fiscal obligations, see Guillou,
'Transformations des structures', 72f.
9 10
Jones, LRE, vol. I, pp. 458ff. Ibid., vol. II, pp.
11
Ibid., vol. I, pp. 414 and 434.
180 Byzantium in the seventh century

key elements, with only a few exceptions where the diocesan level was
retained for specific purposes. 12

THE MIDDLE BYZANTINE SYSTEM

This system of revenue extraction and administration seems to have
continued to operate uninterrupted until the early seventh century. From
the reigns of Phocas and Heraclius, however, a number of changes seem to
have taken place and, together with the events of the seventh century and
the drastic loss of revenues and territory, the whole system underwent a
transformation. I have already noted that a shift occurred in the mode of
tax-assessment in this period, from the late Roman distributive to the
middle Byzantine contributive pattern. I will look next at the chief elements
of the state's fiscal establishment in the ninth century before attempting to
trace more closely the course and the nature of the transformation. As in
the case of the later Roman administration, that represented in the sources
for the ninth century and later has been examined by a number of
scholars, and I shall refer to the results of their analyses for what follows.
In contrast to the tripartite division of competences between prefecture,
sacrae largitiones and res privata, the fiscal administration of the ninth
century is represented by a number of departments, or sekreta,13 of more or
less equal status, subject directly to the emperor. These comprise the
genikon (logothesion), the eidikon (or idikon) (logothesion), the stratiotikon
(logothesion), the first and last coming under the authority of a logothetes,
the idikon under an official known as the epi tou eidikou, or simply the
eidikos; and the sakellion and vestiarion, each under a chartoularios. Accord-
ing to the Kletorologion of Philotheus, from the year 899, and in which
these bureaux are detailed, the sakellarios was the general supervisor of all
these departments, presumably having achieved this distinction by virtue
of having been originally in charge of the emperor's own treasury, which
was left as an independent bureau under the authority of his assistant, the
chartoularios.14 The origins of these departments are not difficult to
discern: the large officia of the prefecture, of the sacrum cubiculum (imperial
bedchamber, which meant in effect the imperial household), as well as of
the magister officiorum and the res privata and sacrae largitiones, seem to
have been divided into groups of scrinia, which, under the officials immedi-
ately subordinate to the former departmental heads, became independent
sekreta. The genikon, idikon (or eidikon) and stratiotikon derived from the

12 Ibid., pp. 374f.

13
For the origins of the term, see Bury, Administrative System, pp. 83f.
14 See T.Usp., 47.12, 49.22, 51.1, 3, 7, 8, 53.21, etc.; Kiel Phil 113.23-115.4, 12-20,
121.3-26, 123.7-10; Dolger, Beitrdge, pp. 16ff. and 27.
 Fiscal administration 181

general, special and military departments of the praetorian prefecture of

the East;15 the sakellion from the old sacrum cubiculum subdepartment of the
sacellum;16 and the vestiarion from the department of the sacrum vestiarium
within the sacrae largitiones.17 In a similar fashion, the department of the
public post, originally under the praetorian prefect and supervised by the
magister officiorum, had become an independent bureau under the logothetes
tou dromou;18 while the old praepositus gregum under the comes rei privatae
became independent as the logothetes ton agelon.19 A number of other
departments developed similarly out of older and larger bureaux. But from
the fiscal-administrative standpoint, the changes also meant a concentra-
tion and centralisation in Constantinople, since the bureaux in question
had their own officials in each provincial circumscription, or thema, during
the ninth century, and there was only one level of administration, as
opposed to the civic, provincial and diocesan levels of the later Roman
period. The department of the sakellion under its chartoularios directed local
thematic civil and fiscal administration through its protonotarioU and liaised
with the local officials of the genikon;20 the department of the stratiotikon
likewise directed local military administrative matters, the assessment of
supplies and provisions, and so on, through its thematic representatives,
the chartoularioi ton thematon;21 while the bureau of the quaestor in
Constantinople, whose tribunal formed the penultimate court of appeal
before that of the emperor, supervised the thematic kritaU or judges. 22

15
See the original analysis of Stein, Studien, pp. 149ff.; also Bury, Administrative System,
pp. 86, 9 0 and 98; and emending Bury's identification, Oikonomides, Preseance,
pp. 3 1 3 - 1 4 and 316 with literature; Hendy, Studies, p.412; V.Laurent, Le Corpus des
sceaux de Vempire byzantin, vol. II: VAdministration centrale (Paris 1981), pp. 1 2 9 - 3 0 and
3O3ff.
16
Hendy, Studies, p. 412; cf. Jones, LRE, vol. II, pp. 567f. on the sacrum cubiculum; Oikono-
mides, Preseance, pp. 314f.; Laurent, Corpus, vol. II, pp. 383ff.
17
Bury, Administrative System, pp.95f.; Jones, LRE, vol. I, p. 428. This department should
not be confused with the oikeiakon vasilikon vestiarion under the epi tou oikeiakou vestiariou
or protovestiarios, descended from the palatine comitiva sacrae vestis of the sacrum cubicu-
lum. See Oikonomides, Preseance, p. 305 with literature; Bury, Administrative System,
pp.95f. and 125; Hendy, Studies, pp. 4 1 2 - 1 3 ; Seibt, Bleisiegel, p. 184.
18
Jones, LRE, vol. I, p. 369, vol. II, pp. 832ff. for the cursus publicus, its administration and
supply by the praetorian prefecture, and the inspectorate exercised by the master of
offices. See Bury, Administrative System, pp. 9 1 - 2 ; Oikonomides, Preseance, p. 3 1 1 ; D.A.
Miller, 'The logothete of the drome in the middle Byzantine period', B 36 (1966), 4 3 8 - 7 0 ;
Hendy, Studies, p. 608 and note 240.
19
Jones, LRE, vol. I, p . 4 1 4 ; Bury, Administrative System, p. I l l ; Oikonomides, Preseance,
p. 338; Laurent, Corpus, vol. II, pp. 2 8 9 - 9 9 .
20
Klet. Phil. 121.6; Oikonomides, Preseance, pp. 314f. and literature; H. Ahrweiler, 'Recher-
ches sur radministration de l'empire byzantin aux IX e -XI e siecles', ECU 84 (1960),
1-109, see 4 3 .
21
Klet. Phil. 15.15; Oikonomides, Preseance, p. 314; Ahrweiler 'Recherches', 4 3 ; Dolger,
Beitrage, pp. 2 If.
22
See Ahrweiler, 'Recherches', 67ff.; 4 3 - 4 .
182 Byzantium in the seventh century

The function and responsibilities of the various fiscal sekreta of this period
are reasonably clear, although the sources of revenue for departments
such as the eidikon, for example, are difficult to determine. The genikon,
through its officials deputed to the provinces - epoptai, exisotai and dioiketai
(only the last group of whom seem actually to have been based in the
themata) - was responsible for the calculation, assessment and collection of
the chief public taxes, the demosia, primarily the land-tax and the kapnikon,
or hearth-tax. 23 The stratiotikon was responsible for calculating the regular
and expeditionary requirements of the troops, who were paid through it,
and through the thematic protonotarios and chartoularios from the geni-
kon.24 The later eidikon, in its original form of the idike trapeza of the
praetorian prefecture of the East, had probably functioned as a central and
provincial clearing-house for assessments in kind - ores, weapons,
clothing and so forth - before passing on the materials to the relevant
departments: iron, for example, to the state arms-factories under the
praetorian prefects, later supervised by the magister officiorum;25 clothing
(assessed and collected by the department of the largitiones, although the
prefecture played a key intermediary role), passed on at provincial and
diocesan level to the actuaries of the military units. 26 In the same way, the
sacrae largitiones had stored and sold off bullion excess to the state's
requirements. 27 The later eidikon seems to have retained these functions (as
the regular use of both the forms idikon, special, and eidikon, dealing in
items in kind, suggests) 28 and certainly controlled the imperial arms
storehouses and was responsible, via the protonotarioi and strategoi of the
themata, for the production and distribution of weapons and other military
equipment. 29 The public vestiarion, once an element within the largitiones,
had inherited that department's function with regard to mints and bullion,
as well as, apparently, the stores for the imperial fleet and iron for other
items of military hardware. 30 The sakellion exercised a general fiscal

23
See esp. Dolger, Beitrdge, pp. 14ff. and 47; Laurent, Corpus, vol. n, pp. 1 2 9 - 3 0 .
24
See Haldon, Byzantine Praetorians, pp. 3 1 4 - 1 8 with literature.
25
The scrinium armorum of the prefecture w a s responsible for this function: see Lydus, De
Magistratibus III, 4 - 5 .
26
See CJ XII, 1 - 4 (three constitutions addressed to the praetorian prefect, one to the comes
sacrarum largitionum).
27
See Cruikshank-Dodd, Silver Stamps, pp. 23ff.
28
See Stein, Studien, pp. 1 4 9 - 5 0 ; Dolger, Beitrdge, pp. 1 9 - 2 0 and 3 5 - 9 ; Laurent, Corpus,
vol. II, pp. 1 9 1 - 3 ; and esp. Hendy, Studies, pp. 6 2 8 - 9 .
29
See Haldon, Byzantine Praetorians, pp. 3 1 6 - 1 7 , 3 2 0 - 2 with sources and literature.
30
See Kiel Phil 1 2 1 . 1 5 - 2 6 ; Oikonomides, Preseance, p. 316; Bury, Administrative System,
pp. 9 5 - 7 ; Hendy, Studies, p. 4 1 2 ; Seibt, Bleisiegel pp. 1 8 4 - 5 ; Laurent, Corpus, vol. II,
pp. 353ff. The vestiarion may also have continued to operate weaving establishments or
supervise the production of cloth, as the existence of magistri linteae vestis in the old
offlcium of the comes sacrarum largitionum might suggest: cf. Notitia Dig., Or. XIII, 14; C/XI,
8.14; although the majority of such establishments now came under the eidikon.
 Fiscal administration 183

supervision in the themata through the protonotaries; but it seems also to

have supervised some of the imperial charitable institutions, which it
presumably supported. 31 In addition to these bureaux, the imperial private
treasury, or koiton, the private vestiarion and the imperial estates under the
(great) curator (megas kourator) also provided funds in cash and materials
from a variety of sources: the last-named from imperial estates throughout
the empire and representing the surviving elements of the former res
privata minus the estates of the praepositi gregum (now under the logothetes
ton agelon);32 the koiton under the parakoimomenos (originally the chief
koitonites), which was itself originally a department of the sacrum cubiculum
under the praepositus sacri cubiculi;33 and the related oikeiakon vasilikon
vestiarion, or imperial private wardrobe, under the protovestiarios, received
resources allocated from the genikon, as well as the income from tribute
and some income from the imperial estates in addition. Both the last-
named departments had treasuries, from which donatives and largesses
might be issued; and the private vestiarion was also a storehouse for
precious silks and other items of imperial or ceremonial clothing and
vestments, jewels and plate. 34

THE PROCESS OF CHANGE

The later system of fiscal and economic management, while clearly derived
directly from that known in the sixth century, has obviously undergone a
series of changes - changes which produced a less hierarchical and very
much more centralised set of institutions. It is the evolution from one to the
other that concerns us here.
In the first place, the central role of the sakellion and the sakellarios
deserves attention. The subdepartment of the sacellum within the sacrum
cubiculum had acted as a personal treasury for the emperors from the later
fifth century, possibly earlier, although the first reference to a sacellarius is
for the reign of Zeno. 35 Thereafter the proximity to the emperor and
31
Klet Phil. 1 2 1 . 3 - 1 4 ; Oikonomides, Preseance, pp. 314f. Its income derived in the first
instance from the res privata - the properties later under the megas kourator.
32
Klet. Phil. 123.11-20; Oikonomides, Preseance, p. 318; Bury, Administrative System,
pp. 1 0 0 - 3 ; cf. Jones, LRE, vol. I, pp. 4 2 5 - 7 ; the megas kourator seems to have been a new
creation of Basil I, associated with the estates of the Mangana; although a single curator
had remained in charge of imperial estates (excluding those of Cappadocia) since the later
sixth century. See Kaplan, 'Maisons divines', 83ff.
33
Bury, Administrative System, pp. 1 2 4 - 5 ; Oikonomides, Preseance, p. 305.
34
For the protovestiarios, see note 17 above. The koiton w a s embraced within the oikeiakon
vasilikon vestiarion, although it constituted a n independent section. For tribute, see DAI (De
Administrando Imperio) 50.53. See also Hendy, Studies, p. 2 2 7 and note 4 1 ; and Dolger,
Beitrdge, p. 2 5 note 3.
35
John of Antioch, frg. 214.4. Sacellarius was a title applied to a number of officials serving
in the capacity of treasurer to a particular department or individual, in the Church, for
184 Byzantium in the seventh century

consequent importance of the imperial sacellarius meant that officials

entrusted with this post occupy a more significant position in the imperial
administration than the position alone might warrant. Ad hominem
appointment of particular persons not formally connected with them to
military commands or other such specialised tasks was a common feature
of imperial administrative practice; and the sacellarius was on several
occasions entrusted with such missions. Narses was Justinian's sacellarius
between 530 and 538, and went on to high office;36 the sacellarius
Rusticus was entrusted by Justinian with the delivery of money for the
army in Lazica;37 the former sacellarius of Phocas was killed in 610; 38 the
sacellarius Theodore Trithyrius is referred to during the reign of Heraclius
by later chroniclers as leading imperial troops (in 635) in Syria;39 while
the sacellarius Philagrius was entrusted by the Emperor Heraclius Con-
stantine with the protection of his son, the future Constans II, as well as
with considerable funds in cash. 40 Sacellarii were thus becoming increas-
ingly important during the later sixth and the first part of the seventh
century; and indeed during the reign of Constans II, the sacellarius, referred
to as the leading dignitary of the palace, led the interrogation of Maximus
Confessor and appears to have been one of the emperor's most trusted
political aids. 41
Not only was the sacellarius, therefore, one of the leading officials in the
imperial personal retinue, he and his department seem to have been
increasingly involved from the time of Justinian in the fiscal affairs of the
state: the delivery of pay to the troops, which seems to have been regarded
as part of his regular functions,42 or (with a similar supervisory mission in
respect of military pay and recruitment) actually commanding military

example, or in the army: the sacellarius of the general Peter, magister militwn in Numidia
(see PG XC, 112A, 113A) is mentioned during the interrogation of Maximus Confessor in
the mid-seventh century. See Bury, Administrative System, pp. 80f.
36
Procopius, De Bello Persico, I, 15.31 (a. 530) and De Bello Gothico, H, 13.16 (a. 538).
37
Agathias III, 2; see Bury, Administrative System, p. 85; Stein, Studien, p. 146.
38
Chron. Pasch. 701; Nicephorus, 5.6.
39
Theophanes, 337.23, 338.3; Nicephorus, 23.12; see Lilie, Die byzantinische Reaktion,
p. 42; Stratos, Byzantium in the Seventh Century, vol. II, pp. 64ff.
40
See Nicephorus, 28.12; John of Nikiu, 192. See Laurent, Corpus, vol. II, no. 740, a seal of
Philagrius, koubikoularios and sakellarios. Philagrius is referred to by Nicephorus as 6 TWV
PacriXixwv XPT||JUXT(OV Ta|xta<5, as is Theodore Trithyrius at Nicephorus, 23.12, Leontius,
at ibid., 5.6, and Stephen, the sakellarios of Justinian II at ibid., 37.13, all of whom are
otherwise referred to by the specific title of sakellarios (Theophanes, 337.23; 338.3;
367.15; Chron. Pasch. 701), as we have seen. The seal provides final confirmation of the
views of Bury (Administrative System, pp. 84-5) and Stein {Studien, p. 146) that the phrase
6 TWV pa(Ti\ixd>v xpT| ixctTwv T a f i a s is equivalent to sakellarios.
41
PGXC, 8 8 C 8 9 A , 113B.
42
See notes 37 and 4 0 above; note Hendy, Studies, pp. 4 1 0 - 1 1 , who points out that Pope
Gregory claimed to be acting as imperial sacellarius in paying the expenses of the local
troops in 595. See Greg. I, epist. V, 39 {MGH, Ep. I, p. 328).
 Fiscal administration 185

forces.43 The importance of the sacellarius in the central fiscal administra-

tion had clearly become very great; and by the reign of Justinian II, if not
well before, the sacellarius had, together with the general logothete, become
one of the chief fiscal officers of the state. 44 In the process, his department
seems also to have taken on responsibility for the administration of its
chiefs duties, probably including also the supervision of the other fiscal
bureaux. It is interesting that the sacellarius (sakellarios), as a confidant of
the emperor as well as an official of state, was regarded as the chief of a
trustworthy and more easily controlled department; that at least is the
impression gained from the proem to the Ecloga (741), where the emperors
note that the quaestor and the judiciary are henceforth to be paid direct
from the sakellion, to avoid the possibility of bribery. 45
The role of the sakellarios during the seventh century suggests in the first
instance, therefore, a more direct intervention in a closer control of the
state finances by the emperors, the tool of this intervention being the
sakellion, that is, the relevant department of the sacrum cubiculum. This
impression is reinforced in other respects, as other sections of this depart-
ment came to prominence: the increasing importance of the corps of
spatharioi through the middle and later seventh century, both as soldiers in
the emperor's presence and as imperial agents in a variety of duties; the
increased use and higher worth attached to the dignity of spatharios; and
the increased importance and similarly increased bestowal of the title
cubicularius (koubikoularios) over the same period.46 Both the spatharioi,
and the sakellarios and his staff, constituted originally subdepartments of
43 44
See note 39 above. Theophanes, 367.15sq., 22-30; cf. 369.26-30.
45
Ecloga (ed. Burgmann), proem. 166.103-9.
46
Haldon, Byzantine Praetorians, pp. 182ff., where the textual and sigillographic evidence is
discussed. The spatharii praesentales in actu increasingly seem to replace the older corps of
the excubitores, which appears to become a parade-ground force only. See ibid., pp. 161ff.
See also Guilland, Recherches, vol. I, pp. 275ff. for a prosopographical list of some
important cubicularii. A sample from the sigillographic collection of Zacos and Veglery is
illustrative: of some 42 officials bearing the title cubicularius for the period c. 500 to 900, 7
belong to the period before c. 650, 21 to the period from c 600 to c. 750, and 10 to the
eighth century; of the approximately 256 or so officials bearing the title spatharius or
imperial spatharius, 3 belong to the period up to c. 600, whereas there are some 26 for the
period c. 600 to 750, and the remaining 227 belong to the late eighth or ninth centuries.
Other collections, such as that of Schlumberger, of Pancenko and of Laurent, bear out the
same pattern. The spread of seals reflects both the incidence of finds and collections, of
course, as well as the historical fact that seals became increasingly popular, and their
inscriptions increasingly detailed, from the seventh to the tenth century. But the results of
this breakdown of the date and the numbers of seals bearing the titles cubicularius and
spatharius is borne out also in the literary sources, and it does point to a dramatic increase
in the role and importance of the imperial cubiculum and the titles to which it gave rise.
Whether or not the greater part of the seals represents titles rather than offices (and from
the later seventh century this seems to be the case), it is significant that it is titles
connected with the cubiculum that demonstrate this increase in use, and they reflect
therefore the corresponding increase in the importance of this palatine department.
186 Byzantium in the seventh century

the sacrum cubiculum, a connection which is clearly reflected in the seals of

sakellarioi of the period, the majority of whom bear also the dignity of
koubikoularios.47 Similarly, as we have seen, the (later) important depart-
ment of the private imperial vestiarion (oikeiakon vasilikon vestiarion) was
descended from the comitiva sacrae vestis of the sacrum cubiculum. The
importance of the personnel of the sacred bedchamber and its various
subdepartments in the central government, therefore, seems to have
increased dramatically at this time, an impression reinforced also by the
fact that the official in charge of the imperial table, the epi tes trapezes, had
also become an officer of considerable importance by the middle of the
seventh century: Sergios Eukratas, who held this position (epi tes trapezes
tes vasilikes), was centrally involved, along with the sakellarios, in the
interrogation of Maximus Confessor; the domestikos tes vasilikes trapezes,
Leontius, was one of the leading officials of state present at the sessions of
the sixth ecumenical council in 680/1. 4 8 The post of epi tes trapezes was
descended from that of the castrensis of the sacred table, with his staff, who
formed a section of the department of the castrensis sacri palatii, also within
the sacrum cubiculum.49
This process of centralisation and supervision through the departments
of the imperial bedchamber and household was well under way by the
second half of the seventh century, but seems to have begun during the
reign of Heraclius, if not earlier. 50 As has been shown, the civil wars which
followed the accession of Heraclius, and the exigencies of the long war with
Persia, resulted in a number of changes in imperial administrative practice,
both in fiscal administration as well as in other fields - for example,
military organisation - which mark the beginnings of this tendency.
Particularly significant is the organisation of mints, which underwent
considerable restructuring during the period 627-30. The mints of the
exarchates of Ravenna and Carthage, along with that of Alexandria,
remained in operation, but those of Thessaloniki, Nicomedia, Cyzicus,
Cyprus, Antioch and Catania in Sicily were closed down. The effects of this

47
See, for example, Zacos and Veglery, nos. 747: seal of Antiochus, koubikoularios, vasilikos
chartoularios and sakellarios (7th cent.); 911: seal of Leontius, koubikoularios, chartoularios
and sakellarios (7th cent.); 932: seal of Mauritius, koubikoularios, imperial chartoularios,
and sakellarios (650-750); 1365: seal of Philagrius, koubikoularios and sakellarios (7th
cent., and cf. 747, above; and 750: seal of Antiochus Philagrius, c. 550-650); 1678: seal
of anon., koubikoularios and sakellarios (7th cent.).
48
PG, XC, 120B; Mansi XI, 209C.
49
For the castrensis sacri palatii, see Notitia Dig., Or., xvii; and for a xaorpT|cn,o<; T-qs fteias
TpcnTe^Tis, see Vita Danielis StyL, 25; for the same official under Justinian: Jones, LRE, vol.
2, pp. 567-8 note 7.
50
The autonomy and the momentum of organisation demonstrated by the establishment of
the imperial estates and the patrimonium in the later sixth century, independently of other
institutions, has already been noted by Kaplan, 'Maisons divines', esp. 92.
 Fiscal administration 187

development were to terminate effectively the role of the sacrae largitiones in

the dioceses, with a consequent further reduction in the relevance of the
diocesan level of administration for the praetorian prefectures.51 As we
have seen, the diocesan element had already lost much of its administrative
significance by the time of Justinian. This change in the administration of
mints must have removed one of the last remaining supports to its con-
tinued existence. From c. 630 there was only one mint for the prefectures of
Illyricum and Oriens, that at the capital, Constantinople, with the exception
of that at Alexandria. The new system constitutes, therefore, a fairly radical
break with the past. Gold continued, as before, to be minted at the prefectu-
ral level of administration, at Constantinople, and at Carthage and
Ravenna for the two exarchates. The diocesan level seems to disappear
from the record entirely, and the production of copper coins for these areas
from then on depended upon the prefectural mints as well. The Constanti-
nopolitan mint now produced copper in fixed quantities for particular desti-
nations, sometimes minting the coins specifically for the destination in
question. 52 As Hendy has noted, this is a reversal of the traditional late
Roman procedure. And it implies several important consequences: in the
first place, a clear centralisation of the minting, and the distribution, of
coin, both gold and, for the first time, copper; and in the second place, a
reorganistion and possibly the disappearance of the sacrae largitiones, and
the assumption of their functions and remaining administrative personnel
by another department. The last explicit mention of a comes sacrarum larg-
itionum is to Athanasius, komes largitionon, for the year 605. 5 3 A logothete
is referred to for the year 626, and while claims for his being a logothetes tou
stratiotikou seem improbable, 54 the reference to the title is significant.
Already in 602/3 Constantine Lardys is described as ex-praetorian prefect,
logothete and curator of the imperial estates of Hormisdas. 55 Since logo-

51
This development, and the numismatic evidence for it, has been fully analysed by Hendy,
'Administrative basis'. Justinian had effectively abolished the separate vicariates (diocesan
governorships) of Asiana, Pontica and Oriens by combining them with the provincial
governorships of Phrygia Pacatiana, Galatia I and Syria I respectively; those of Pontica
and Asiana were effectively restored after 5 4 8 . See Jones, LRE, vol. 1, pp. 280ff, 2 9 4 .
52
Hendy, 'Administrative basis', 1 4 7 - 5 2 , and Studies, pp. 4 1 7 - 2 0 . The temporary mints at
Jerusalem and Alexandretta, at Constantia in Cyprus, at Seleucia and Isaura in Isauria,
are to be explained in terms of the exigencies of the warfare of the period 6 0 8 - 2 6 / 7
(Hendy, Studies, pp.415f.).
53
Chron. Pasch., 6 9 6 (cf. Theophanes, 2 9 7 . 2 0 w h o mentions his execution in 609).
54
Chron. Pasch. 7 2 1 : 0eo86crio<s 6 ev8o£oTaTO<; iraTptxios x a t \oyoftsTX)<$. For his identity as
a military logothete, see R.Guilland, 'Etudes sur l'histoire administrative de l'empire
byzantin: les logothetes', REB 2 9 (1971), see 2 5 - 6 and note 9; N. Oikonomides, 'Les
premiers mentions des themes dans le chronique de Theophane', ZRVI16 (1975), 1-8,
see 6 note 2 3 ; Hendy, 'Administrative basis', 154; Studies, p. 4 1 3 . Against this, see
Haldon, Recruitment and Conscription, p. 34, note 4 3 with literature.
55
See Chron. Pasch. 6 9 4 . 8 .
188 Byzantium in the seventh century

thetes is a Greek equivalent of the Latin technical title numerarius (or

scriniarius), the chief of a scrinium, the important position and high rank
attributed by the sources to these officials at this time is significant.56 It
suggests that they were already persons of high standing and that they
may already be heads of independent bureaux. Constantine Lardys was an
ex-praetorian prefect, and both he and Theodosius were gloriosissimi, both
therefore belonging to the topmost group of officers of state. The older
scriniarii, while they had been important officials in their departments, had
not attained this elevated status. And there are other high-ranking logo-
thetai, connected with the customs and trading depots or apothekai, also
known from the reigns of Maurice and Phocas. 57
We have, therefore, at least the possibility that by the reign of Heraclius,
from 610, the fiscal departments of the praetorian prefecture were becom-
ing, or had already become, independent bureaux supervised by high-
ranking officials. Such a development would immediately involve also a
reassessment of the relationship between the largitiones and the special,
general and military departments of the praetorian prefects (who certainly
continued to operate, at least in a general capacity, until 629 if not
beyond), 58 and the eventual result seems to have involved the organi-
sation of the mints on a single, centralised level and the absorption of
responsibility and authority for this operation into the single bureau of the
sacrum vestiarium within the largitiones which, with its much reduced staff
(no longer needing regional or urban representatives or staffs, merely the
central staffs of the mints themselves, and the officials in charge of receiv-
ing the bullion for coining) could now handle all the demands made upon
it. The local affairs of the largitiones had in any case been handled by
officials of the prefecture or by officials seconded from Constantinople.59
The later establishment of the general logothesion seems to reflect this fact,
for its staff includes chartoularioi ton arklon, sometimes called chartoularioi
of the exb arklaU that is, chartularies of the provincial treasuries formerly
under the largitiones.60
The prefectures had each had, of course, a central fisc (two for Oriens, as
we have seen, the special and the general), but they had never had local
56
Stein, Studien, pp. 1 4 8 - 5 0 , contra Bury, Administrative System, p. 8 3 . These were the
numerarii of the scrinia - see Jones, LRE, vol. I, p p . 4 5 0 a n d 5 8 9 . See, for example,
Justinian, Edict XIII, 10. proem; 1 1 . 1 ; 13 for scriniarii = members of scrinia, including the
chiefs also.
57
See Jones, LRE, vol. II, p. 589; and for other commerciarii see Zacos and Veglery, p. 2 1 4 ,
nos. 3 a n d 6, with earlier literature; ^Tecpavos, ev8o£6T(XTOs ctiro ixa^Ccrrpwv, {teios
Xo^ofteTKis x a i xojifxepxiapios Tvpov; and liaavvj]^, ev5o£6T(XTO<; CLTTO VTTCLTWV
iraTpCxios, X ^ o f t e r r i s x a i f3aai\ix6<; (probably xo(X|xepxtdpios).
58
JGR I, coll. 1, nov. 2 5 , 2 (a. 6 2 9 = Dolger, Regesten, n o . 199).
59 See above, a n d Justinian, Edict XIII, 11.2; 3; XIII, 2 0 .
60 For example, Klet. Phil, 1 1 3 . 2 9 ; De Cer., 6 9 4 . 1 8 .
 Fiscal administration 189

treasuries or reserves, such as those operated for the largitiones at diocesan

level by the praepositi or comites thesaurorum.61 It seems highly likely, as
Bury long ago suggested,62 that the later chartularies of the arklai (Lat.
arcae) were in fact these diocesan officials of the largitiones, absorbed into
the bureau of the general bank of the prefecture of the East - that is, into
that section of the prefecture's activities handling public and general taxes.
The praepositi thesaurorum had, after all, received the income from minor
taxes on land, for example, which were owed to the sacrae largitiones, and it
was to these local depots that such revenues were delivered. In the same
way, gold, silver and other goods for which the largitiones was responsible
were also delivered there (usually by officials of the prefecture), before
being passed on to the capital. Similarly, the commercia, and the diocesan
comites commerciorum, were also absorbed into the bureau of the general
bank. 63 It seems probable, then, that this element of the organisation of the
largitiones was absorbed in its entirety into the structure of the general
bank of the prefecture, a development which will also have enhanced the
status and power of the latter bureau. While the mints continued to be
administered separately at a later date, within the bureau of the public
vestiarion, the vestigial remnant of the largitiones (the name alone reminds
of its origins, and the loss of its wider fiscal and administrative func-
tions), 64 they received recirculated gold coin and bullion now entirely
through the medium of the general logothesion and the eidikon.65
The reorganisation of the mints in the period 627-30, thus, must have
involved also the transfer of their regional depots, along with full responsi-
bility for the original cash revenues of the largitiones, to the relevant
branch of the praetorian prefecture, chiefly the genike trapeza under its
numerarii. The greatly increased importance of the sacellarius at the same
period, it may be guessed, was in some way connected with this
rearrangement; and it seems not unreasonable to suggest that, whatever
vestigial authority the praetorian prefect retained, the relative indepen-
dence of at least two, and possibly three, of his former departments - the
general, special and military banks - goes back to the reign of Heraclius
61
For the operation of their regional chests, see Jones, LRE, vol. I, pp. 4 2 8 - 9 .
62
Bury, Administrative System, p. 87. These officers are clearly not to be connected with the
prefectural (i.e. central) arcae as such, pace Stein, Studien, p. 150.
63
Jones, LRE, vol. I, pp. 4 2 8 - 3 3 . For the comitiva sacrarum largitionum, see Notitia Dig, Or.
XIII; and see below.
64
Its function appears to have been twofold: the minting of coin (distribution via the other
fiscal bureaux) and the storage and provision of naval equipment - although the origins of
this aspect of its role is unclear.
65
See, for example, Dolger, Beitrdge, pp. 19ff. and 35ff.; Laurent, Corpus, vol. H, pp. 191ff.;
but the komes tes lamias (X6|AT|S TT|<; Aa|iias) has nothing to do with mines, being in fact in
charge of the Constantinopolitan granaries. See J.F. Haldon, 'Comes horreorum - komes
tes Lamias?', BMGS 10 (1986), 2 0 3 - 9 .
190 Byzantium in the seventh century

and that the supervisory authority of the sacellarius and his department is
similarly to be dated to this time, if not slightly earlier. In consequence, the
rationalisation of the remaining functions of the largitiones into the 'new'
and independent bureau of the vestiarion, likewise under the general
supervision of the sacellarius, must also date to approximately the same
time. 66 While this scheme is, to a degree, hypothetical, the implications
follow directly from Heraclius' proven reorganisation of the system of
minting quite logically, although the later sources, such as the so-called
Taktikon Uspenskij and the Kletorologion of Philotheus, both from the
ninth century, present an already evolved structure which may no longer
exactly represent the original form of these new departments. But the
mention of two logothetai, for the years 602/3 and 626, demonstrates that
the Greek title for the head of the chief fiscal bureau was already current. 67
Heraclius' reorganisation of the mints, therefore, may reasonably be
taken, at least partially, to reflect a wider process of fiscal administrative
reorganisation, a process which seems already to have commenced before
his reign, however, and which this reorganisation may have completed.
The causes of this process of centralisation are more difficult to locate
and can on the whole only be guessed at. The precarious economic posi-
tion of the empire in the last years of the sixth century certainly provides a
context. 68 But inherent structural tendencies within the organisation of
the fiscal-civil administration may equally well have been working them-
selves out. 69 The result seems to have been that the greater autonomy of
66
It is significant that the traditional system of control-stamps on silver plate associated with
the largitiones begins to break down at about this time, a point emphasised by Hendy,
Studies, p. 4 1 3 . See Cruikshank-Dodd, Silver Stamps, pp. 31 and 45f. It is also significant
that the single zygostates in the bureau of the sakellion in the ninth century and later (see
T.Usp., 61.12; Kiel Phil, 121.8; 153.29; Oikonomides, Preseance, p. 315 with sources
and literature; Hendy, Studies, pp. 317f.), responsible probably for controlling the purity
and weight of the coinage, appears on a seal of the seventh century (Zacos and Veglery,
no. 2803: John, skribon and imperial zygostates). Zygostatis is here clearly an office, and
the epithet Imperial' is therefore all the more significant. Whether the seal dates to the
period with which we are concerned here or to the later part of the seventh century, it is
evidence already that the functions of the earlier municipal zygostatai under the supervi-
sion of the praetorian prefecture (see CTh. XII, 7.2 = C/X, 73.2, emended; and cf. Justinian,
Edict XI) had probably been centralised within the bureau of the sakellion at an early stage;
given the mint reforms discussed by Hendy for the years 62 7-30, the latter date is very
probably the moment at which such a centralisation might have occurred. For the system
of control-stamps, and especially the gradual transfer of their administration to the city
prefect of Constantinople, as the department of the largitiones was progressively fragmented,
see D. Feissel, 'Le Prefet de Constantinople, les poids-etalons et l'estampillage de l'argenterie
au VIe et au VIP siecle', Revue Numismatique 28 (1986), 119-42.
67
Note also the reference to a certain George, chartoularios of the sacred logothesion,
mentioned in a story from the last years of the reign of Heraclius or the first years of
Constans II: see Miracula S. Artemii 25.29.
68
See chapter 1; and the summary of Jones, LRE, vol. I, pp. 2 9 8 - 3 1 5 , esp. 3O5ff.
69
The development of two separate reserves within the Oriental prefecture, for example,
which appears in the last decades of the fifth century as a response to the fiscal
 Fiscal administration 191

the two banks of the oriental prefecture on the one hand, and the increas-
ing intervention in, and supervision of, fiscal affairs by the sacellarius, the
emperor's personal treasurer, on the other hand - an effect of the lack of
cash resources and the multiplicity of demands on the limited revenue
available - coincided. The political-military collapse of the reign of Hera-
clius before 622-6 may have encouraged attempts to reorganise the state's
fiscal administration, to recognise some of the changes which had become
effective in the interim, and to continue and extend the process of central
supervision and control in respect of revenues, resources and expenditure.
As a corollary of Heraclius' reorganisation of the mints, it is therefore likely
that the departments of the general, special and military logothesia, under
the general supervision of the praetorian prefect and the sacellarius,
together with those of the vestiarion and the sakellion, appeared in a form
which was recognisably that of the Byzantine rather than of the later
Roman state. The concomitant disappearance of the largitiones, and, with
them, of the final significant prop of the diocesan level of provincial
administration, together with the weakening of the power of the praeto-
rian prefect, will have had important implications for the later develop-
ment of the provincial civil administration, as well as the pattern of the
palatine establishment.
The lack of direct evidence concerning these developments has, of
course, turned this problem, and the related question of the origins of the
military provinces or themata, into the major vexata quaestio of this period
of Byzantine history. On the whole, it is now generally accepted that the
processes of both civil and fiscal, as well as military, reform or change were
gradual, beginning in the later sixth century, and concluded only during
the second half of the eighth century at the earliest.70 Apart from the
administrative exigencies of the times, is illustrative of such tendencies; there is no reason
to suppose that the institutional framework of the state's fiscal administration did not
continue to develop, that is, that it was not a fixed, static block of organisational
relationships. For the fifth century, see Stein, Studien, p. 149 and Bas-Empire, vol. I, p. 221
and note 6.
70
The literature, as is frequently pointed out, is considerable. For the most recent general
surveys, see Lilie, Die byzantinische Reaktion, pp.287ff. and 'Die zweihundertjahrige
Reform: zu den Anfangen der Themenorganisation im 7. und 8. Jahrhundert', BS 45
(1984), 27-39 and 190-201, see 27ff.; Haldon, Recruitment and Conscription, pp.20ff.;
Hendy, Studies, pp. 409flf. and 62 Iff. Of the older literature, the work of J. Karayannopou-
los, 'fiber die vermeintliche Reformtatigkeit des Kaisers Herakleios', JOBG 10 (1961),
53-72, and Die Entstehung der byzantinischen Themenverfassung (Byzantinisches Archiv X,
Munich 1959), esp. pp. 5ff. and 55-8; and Karayannopoulos, 'Contribution au probleme
des themes byzantins', L'Hellenisme contemporain 10 (1956), 458-78; and cf. A.Pertusi,
'La Formation des themes byzantins', in Berichte zum XI. International Byzantinisten-
Kongress (Munich 1958), vol. I, pp. 1^40; and 'Nuova ipotesi sull' origine dei temi
bizantini', Aevum 28 (1954), 126-50, are all still important. See also J.Toynbee, Constan-
tine Porphyrogenitus and his World (London 1973), pp. 134ff. and 224ff.; and Oikonomi-
des, 'Les Premiers mentions'; Kaegi, 'Two studies'.
192 Byzantium in the seventh century

references to the logothetai, and to the sakellarios and his important role
during the reign of Constans II already mentioned, the first references to
the leading officers of the developed establishment occur for the year
680, when certain high officials accompanied the emperor at the ses-
sions of the sixth ecumenical council. These are named in order as
follows: Nicetas, most glorious ex-consul, patricius and master of the
imperial offices; Theodore, most glorious ex-consul and patricius, comes of
the imperial Opsikion and deputy general of Thrace; Sergius, most glori-
ous ex-consul and patricius; Paul, also most glorious ex-consul and
patricius; Julian, most glorious ex-consul, patricius and logothete of the
military treasury; Constantine, most glorious ex-consul and curator of
the imperial estate of Hormisdas; Anastasius, most glorious ex-consul,
patricius and second-in-command to the comes of the imperial excubitores;
John, most glorious ex-consul, patricius and quaestor; Polyeuctes, most
glorious ex-consul; Thomas, also most glorious ex-consul; Paul, most
glorious ex-consul and director of the eastern provinces; Peter, most
glorious ex-consul; Leontius, most glorious ex-consul and domestic of the
imperial table.71
Of the older establishment, the master of offices appears to have still
exercised important functions and counted among the leading officials;
and indeed he continued to exercise some of these functions well into the
eighth century.72 Likewise, the curator of the imperial estates of Hormisdas
seems still to be functioning and may by this stage be the equivalent of the
later megas kouratort in charge of other imperial properties formerly
attached to the domus divina and the res privata.73 The quaestor was also
present, and although his functions were altered in some respects, his
duties and authority remained much the same in the later period as in the

71 Mansi, XI, 209A-C; Riedinger, 14.19ff.

72
See the studies of Boak, The Master, pp. 491T., and Bury, Administrative System, pp. 29f.; for
the master in the eighth century, when he seems still to have exercised authority over the
scholae, until the reforms of the guard units undertaken by Constantine V, see Haldon,
Byzantine Praetorians, pp. 145-50 and 229ff.; and F. Winkelmann, Byzantinische Rang-
und Amterstruktur im 8. und 9. Jahrhundert (BBA LIII, Berlin 1985), p. 31.
73
See above, note 1. The fate of the imperial estates of the res privata, the patrimonium and
the domus divinae is unclear. The stud-farms and other livestock-farming estates came
under the logothetes ton agelon, descended from the older praepositus gregum (see above,
note 19); the remaining properties may well all have been administered by the later
(megas) kourator, and the income derived therefrom seems to have gone into the koiton, the
successor to the sacellum as the emperor's private chest, and the imperial private vestia-
rion. See Hendy, Studies, p. 199 note 235; Stein, Studien, pp. 168-85. By the early eighth
century, seals of officials in charge of imperial estates demonstrate the new bureau's
development. Cf. Zacos and Veglery, no. 3221, seal of anon., hypatos and chartoulahos ton
vasilikon ktematon. See Bury, Administrative System p. 102; Oikonomides, Preseance,
p. 318.
 Fiscal administration 193

sixth century. 74 He may well already have taken over by this time the
scrinia libellorum and epistolarum originally under the disposition of the
magister officiorum, since the Greek term antigrapheis represents the Latin
magistri scriniorum and, as Bury noted, the antigrapheis are associated with
the quaestor in the proem to the Ecloga.75
Leaving aside for the moment the military officials named, the remaining
palatine officials there represent either new offices or give their titles only,
but not their posts. The military logothete is represented, the head of the old
bureau of military finance in the officium of the praetorian prefect, as we
have seen, but from his high rank by now evidently an important
independent official. His presence would argue indirectly also for the
independence by this time of both the general and the special banks of the
prefecture;76 while the presence of the domestikos of the imperial table
suggests that the subdepartment of the sacred table within the old sacrum
cubiculum was by this time also an independent bureau. 77 If this conclusion
is accepted, the further implication is that the sacrum cubiculum had by now
devolved into its constituent parts, an inference which the clear pre-
eminence of the sakellarios in the middle of the seventh century would tend
to support. 78 Thus the koiton under the chief parakoimomenos,79 the
imperial private vestiarion under the protovestiarios (the old comitiva sacrae
vestis under its comes), as well as the sacellum and the imperial table, were
now all independent services in the palace.
References to the various officials of the central sekreta begin to occur in
the literary sources for the later seventh century. Thus Theodotus, the
monk, genikos logothetes during the reign of Justinian II, is mentioned for
the years 694/5, for whom there exists also a seal; 80 a seal of Paul, apo
hypaton and genikos logothetes may date to the same period; while seals of

74
See the detailed discussion of Bury, Administrative System, pp. 73ff.
75
Bury, Administrative System, pp 7 5 - 6 ; Ecloga (ed. B u r g m a n n ) , proem. 1 0 3 - 4 . See Kiel
Phil 1 1 5 . 5 - 1 1 ; Oikonomides, Preseance, pp. 32 If. with literature; see also below, o n the
epi ton deeseon.
76
A dated seal for t h e years 6 5 9 - 6 8 of Stephen, apo hypaton, patrikios a n d stratiotikos
logothetes, emphasises the independence of this b u r e a u at a relatively early stage. See
Zacos a n d Veglery, n o . 1 4 4 ; a n d see p . 1 4 5 , table 1.
77
See above, note 4 9 ; Bury, Administrative System, pp. 125f.; Oikonomides, Preseance,
pp. 3O5f.
78
See note 1 7 above with literature.
79
Bury, Administrative System, pp. 124f.; Oikonomides, Preseance, p. 3 0 5 .
80
Theophanes, 3 6 7 . 2 2 - 4 ; 3 6 9 . 2 7 ; cf. Zacos a n d Veglery, n o . 1 0 6 4 A , seal of Theodotus,
m o n k a n d genikos logothetes. The description genikos logothetes is used by Theophanes of
the chief finance officer of the Caliph c Abd al-Malik - Sergius, the son of M a n s u r - for the
year 6 9 2 . But it is difficult to k n o w exactly w h a t value should be placed o n its use here. It
m a y at least suggest t h a t Theophanes' source used the term a n d t h u s suggest t h a t it w a s
already c u r r e n t a m o n g Greek-speakers for such a n official. Cf. Theophanes, 3 6 5 . 2 4 .
194 Byzantium in the seventh century

(another) Paul, genikos logothetes, and of Michael, hypatos and genikos

logothetes, also belong to the second half of the seventh century. 81 Subord-
inate officials of the genikon are also evidenced by seals for this period.82
Similarly, seals of officials for the imperial koitonion and from the public
vestiarion for the early eighth century suggest that this branch of the older
cubiculum in its new form, as well as the public vestarion, were now well
established;83 while seals of officials described as epano ton deeseon of the
later seventh and early eighth centuries again demonstrate that this
sekreton, formerly the scrinium of the magister memoriae under the dis-
position of the magister officiorum, was by now of independent status and
probably - as in the ninth and tenth centuries - under the authority of the
quaestor. 84

THE FATE OF THE PRAETORIAN PREFECTURE: THE

'SUPERVISORS OF THE PROVINCES'
The evidence of these literary and sigillographic sources, while by no
means definitive, does suggest that the basic framework of the fiscal
administrative and palatine administrative systems known from docu-
ments such as the Taktikon Uspenskij and the Kletorologion of Philotheus,
of the ninth century, were already well established by the year 700; and
the earlier material, in particular the fact of Heraclius' reform of minting
practice and the inferences which must logically follow from it, suggests

81
See Zacos and Veglery, nos. 961, 3162, 2903; N.P. Lihacev, 'Datirovannye vizantiiskie
pecan", Izvestiya Rossiiskoi Akademii Istorii Materid'noi Kul'tury 3 (1924), 153-224, see
180, no. 11; V.Laurent, 'Bulletin de sigillographie byzantine, I', B 5 (1929/30),
571-654, see 6 0 7 note 3: seal of Cyriacus, apo hypaton and genikos logothetes (dated
696/7).
82
For example, Zacos and Veglery, no. 1231, anon., hypatos and chartoularios of the genikon
logothesion.
83
Zacos and Veglery, no. 1093, seal of anon., chartoularios tou vasilikou koitoniou and
chartoularios tou vasilikou vestiariou. See also nos. 1409 and 1714, of Andrew, chartoula-
rios tou vasilikou vestiariou. All these seem to date to the early part of the eighth century. It
is unclear whether no. 1093 belongs to an official of the koiton who functioned also within
the public vestiarion, or whether the private vestiarion is meant. The latter might seem
more likely, in view of the origins of both departments within the sacrum cubiculum,
although the later private vestiarion had no chartoularios in the ninth century and after
(see Oikonomides, Preseance, p. 305). The public vestiarion, on the other hand, was later in
the charge of a chartoularios, and this official, and certainly the Andrew of the seals 1409
and 1714, may well represent this bureau. See Oikonomides, Preseance, p. 316 and
references.
84
See V. Laurent, Les Sceaux byzantins du medailler Vatican (Medagliere della Biblioteca
Vaticana I, Citta del Vaticano 1962), no. 8; Zacos and Veglery, no. 2466; Bury,
Administrative System, pp. 77f.; Oikonomides, Preseance, p. 322. For a further comment on
the fate of the sacrae largitiones, see R. Delmaire, 'Le Declin des largesses sacrees', in
Hommes et Richesses dans VEmpire byzantin I: IVe-VIle siecle (Paris 1989), pp. 2 6 5 - 7 8 .
 Fiscal administration 195

that this pattern was probably already in existence much earlier, and by
the end of his reign. But this in itself is no argument for a major
transformation of either the principles or the practice of civil or military
administration - at least not at this stage.
While the various bureaux of the prefecture dealing with financial affairs
became increasingly independent, the prefecture seems still to have oper-
ated into the last years of the reign of Heraclius. In 629 the praetorian
prefect of the East is referred to in a novel, issued at Jerusalem, on the
juridical situation of the clergy, addressed to the patriarch Sergius.85
Although no more explicit references to this official occur, other evidence
suggests that the prefectures continued for a while to be the main civil
administrative subdivisions within the surviving territories of the state. In
the first place, the praetorian prefect of Illyricum continued to exist until the
second half of the seventh century, although probably with only nominal
uthority over much of his former prefecture, which was now mostly
outside Byzantine imperial control. 86 In the second place, an early ninth-
century ceremony, preserved in the tenth-century Book of Ceremonies of
Constantine VII, lists the praetorian prefect and the quaestor (along with
the eparchs of the themata) together. 87 While it is certainly true that only
the title may have survived, as in many similar cases 88 (that of stratelates,
for example, or anthypatos or magistros), one might have expected it to have
been generalised as a title in the system of precedence, just like these latter,
which it was not, however; and it seems reasonable to suppose that, like the
offices of quaestor and prefect of the themata which it accompanies, it, too,
referred to an office, not a dignity alone. It may well be, of course, that the
praetorian prefect exercised a nominal supervision only by this time.
Already in the later sixth century he was referred to by the function for
which his department was best known: TTJV T)7e|xoviav TWV cpopcov rqs
ecoas ... 6v eirapxov TTpaiToapicov euo'daaiv ovojxd^eiv T?a>|iaioi,89
that is, for the collection and assessment of revenues. And whereas, by the
middle of the seventh century, the bureaux of the general, special and
military logothesia had taken on an independent existence, this does not
preclude the prefect's having still exercised overall authority.
85
JGR I, coll. 1, nov. 2 5 , 2 (a. 6 2 9 = Dolger, Regesten, no. 199).
86
See E. Stein, 'Ein Kapitel v o m persischen u n d v o m byzantinischen Staate', BNJI (1920),
5 0 - 8 9 , see 8 3 ; Ch. Diehl, 'L'Origine du regime des themes d a n s l'empire byzantin', in
Etudes Byzantines (Paris 1905), pp. 2 7 6 - 9 2 , see p. 2 9 0 ; H. Gelzer, Die Genesis der byzanti-
nischen Themenverfassung, A b h a n d l u n g e n der konigl. sachsischen Gesellschaft der Wissen-
schaften, phil.-hist. Klasse (Leipzig 1 8 9 9 and A m s t e r d a m 1966), see pp. 35f.; Bury, ERE,
pp. 2 2 3 a n d 2 3 4 note 1; Lilie, 'Die zweihundertjahrige Reform', 3 5 .
87
See Stein, 'Ein Kapitel', 71ff.; most recently, Kaegi, 'Two studies', 98ff. Cf. De Cer. 6 1 . 2 5 ;
a n d see below.
88
See Lilie's objection, 'Die zweihundertjahrige Reform', 3 5 .
89
Theophylact Simocatta, VIII 9.6.
196 Byzantium in the seventh century

The praetorian prefect had charge also, of course, of the local civil and
judicial administration, and the provincial governors and their staffs; and
the late Roman provincial geography continues to exist well into the eighth
century, as the evidence of lead seals shows quite clearly. The provinces of
the sixth century occur on seals of the later seventh and eighth centuries to
the extent that, of those mentioned in the Justinianic legislation and whose
territory was still within the empire in the later seventh and eighth
centuries, twenty-six occur on seals of kommerkiarioi or apothekai (on which
see below) for the years c. 654-c 720. 90 While the traditional provinces
continued to exist, therefore, their local administrative apparatuses prob-
ably continued to function as far as was possible.
In this respect, the seals of a large number of provincial dioiketai are
significant. In the ninth and tenth centuries, the dioiketai were those officials
responsible for supervising the collection of taxes, and they were respon-
sible to the general logothete. They represent the earlier provincial trakteu-
tai of the diocesan scrinia, responsible to the central clerks and chartularii of
the prefectural bureau. 91 As such, it is likely that they were also closely
connected with the chartularies of the provincial arcae taken over from the
largitiones, although evidence for these officials occurs only from the ninth
century. 92 But evidence for the continued existence of the dioiketai occurs
through the seventh and eighth centuries and into the ninth century in an
unbroken tradition.
From the second decade of the seventh century, general supervisors seem
to have been appointed over a whole group of provinces, possibly over the
whole of the Eastern prefecture. A seal of Theodore, megaloprepestatos
illoustrios and dioiketes of all (province names or designation illegible),
dating to the years 614 to 631 demonstrates that this practice was known
during the period of fiscal reorganisation under Heraclius;93 while the
mention of Paul, the endoxotatos apo hypaton and dioiketes of the eastern

90
For the Justinianic provinces, see Jones, LRE, map VI. Including the islands, Cyprus and
Honorias (which Justinian joined with Paphlagonia, but which seemed afterwards still to
have been treated as an independent province), there were 2 7 such provinces. For the
seals, see Zacos and Veglery, pp. 146ff. and tables 3ff. See below for seals of dioiketai
bearing provincial names. There are seals for 2 3 Anatolian provinces, plus seals for the
various isles making up the Justinianic insulae, and Cyprus. Armenia I and II are the only
provinces not found.
91
See, e.g., Theophanes, 3 6 7 . 2 7 for 6 9 4 ; and Bury, Administrative System, p. 89; Oikonomi-
des, Preseance, p. 3 1 3 . See also Dolger, Beitrdge, pp. 70ff.; N. Svoronos, 'Recherches sur le
cadastre byzantin et la fiscalite aux XP-XII e siecles: le cadastre de Thebes', BCH 83
(1959), 1 - 1 6 6 , see 56f. For the earlier trakteutai, see, for example, Vita Eutychii 68; and
Justinian, Edict XIII, 9 - 1 2 ; 27; 2 8 . In the Vita Eutychii, they are described as TOIJS TT^V

92
See Oikonomides Pre'seance, p. 3 1 3 .
93
Zacos and Veglery, no. 1 3 1 .
 Fiscal administration 197

provinces in 680 at the sixth ecumenical council; the seals of George and
Theodore, each apo hypaton and dioiketes of the eparchiai (or provinces,
unspecified), dated c. 650-700; of Marinus, apo eparchdn and dioiketes of the
eparchiai, dated 650-700; of Paul, hypatos and dioiketes of the Anatolikoi,
dated c. 700; of Leontius, patrikios and dioiketes of the eparchiai, dated c.
650-700; and of Stephen, apo hypaton and dioiketes of the eparchiai, or
patrikios and dioiketes of the eparchiai, dated c. 650-700, demonstrates the
continuation of the same practice.94 Throughout the same period, officials
issued seals bearing the title dioiketes alone, without further elaboration,
and these may represent the subordinate provincial officials placed under
the authority of these general dioiketaL95
From the first half of the eighth century, a shift in the pattern becomes
evident, as ordinary dioiketai or supervisors begin to issue seals with the
name(s) of their provinces also. Thus there are seals for dioiketai of Thrace,
Hellas, Sicily, Euboea, Seleucia, Cyprus, Lydia, Bithynia, Galatia.96 But
what is interesting is that seals for those fiscal officials bearing the general
title 'of the provinces' are limited almost without exception to the second
half of the seventh century and the first years of the eighth.97

w See Mansi XI. 209B; Riedinger, 14.32; Zacos and Veglery, nos. 821, 1031,1178, 2290,
2897,1008 and 1014, respectively.
95
See Zacos and Veglery, nos. 1464, 1527, 1528 and 1439; K. Konstantopoulos.
Bu(avTiotxdt iio\vp86povAAot TOV ev 'Atitivott? 'E-dvixoO Notuo*iiaTixoi> Movaetov
(Athens 1917), nos. 325a, b, g; 586; Zacos and Veglery, nos. 749, 1724, 1991, 2069,
2018, 3109, 885, 956, 2352, 3189, 3192, 1698, 1866, 2120, 2158, 2399, 1951.
1847, 1917, 2302, 2297, 2531 and 2139; V.Laurent. Documentsde sigillographie byzan-
tine. La collection C. Orghldan (Paris 1952), nos. 249 and 251; G. Schlumberger, Sigillogra-
phie de Vempire byzantin (Paris 1884 and Turin 1963), pp. 497-8, no. 13; p. 499, nos. 19
and 20; p. 536, no. 4; J.Ebersolt, Musses imperiaux ottomans. Catalogues des sceaux
byzantins (Paris 1922), nos. 362 and 364 and 'Sceaux byzantins du Mus6e de Constanti-
nople', RN, 4th ser., 18 (1914). 207-43 and 377-409. see 361 and 363; W.de Gray
Birch, Catalogue of Seals in the Department of Manuscripts in the British Museum V (London
1898), pp. 1-106, nos. 17, 615 and 17, 617. These date to the period c. 650 to c 900, the
majority from the eighth century. For a fuller and more detailed list, see Winkelmann,
Rang- und Amterstruktur, p. 134.
96 See Zacos and Veglery, nos. 1044, 2114, 2081, 2082, 2078, 2079, 2019, 2020, 1895,
2183, 2426. 1628. 1642 and 3189 all dating to the period c 700-850. The giving of
more explicit detail may have reflected a personal fashion, as well, of course. Thus the
Theodore, dioiketes, of Zacos and Veglery, no. 3189. seems to be the same as the Theodore.
Zacos and Veglery. no. 2426, who was also dioiketes of Lydia. See Zacos and Veglery,
p. 1778. There are for the same period a considerable number of seals of dioiketai which
give the name of a town rather than a province. For a list, see Winkelmann, Rang- und
Amterstruktur, pp. 134f.
97
That of Sergius, hypatos and dioiketes of the eparchiai (Zacos and Veglery, no, 487), dated to
thefirsthalf of the eighth century is the latest I have been able tofindextant. From its style
and lettering, however, it might equally belong in the years around c. 700. Similar
considerations apply to Schlumberger, Sigillographie, pp.499f., seal of Sergius, dioiketes of
the eparchiai; and to Konstantopoulos, Molybdoboulla, no. 326a, seal of Theophylactus,
hypatos and dioiketes of the eparchiai (in the first and last cases partially confirmed by the
rank, hypatos).
198 Byzantium in the seventh century

Three further points deserve our attention. First, the use of what are
clearly thematic circumscriptions begins to increase on seals of kommerkia-
rioi and kommerkia (customs officials and depots - but see below) from the
early eighth century, suggesting that the military districts which had come
to overlie the older civil provinces were becoming increasingly important
from the general administrative point of view. 98 Second, a slight change in
the formula expressing the titles of two dioiketai of the eparchiai suggests a
shift from a prefectural to a thematic emphasis. The dioiketes at the sixth
council formulated his title, as we have seen, thus: Paul, endoxotatos ago
hypaton, kai dioiketes ton anatolikon eparchion." In contrast, the seal of Paul,
hypatos kai dioiketes ton Anatolikon, dated to the turn of the century,100
seems clearly to name the thema of the AnatolikoL The first Paul seems still
to bear the hallmarks of the oriental prefecture - compare the usual title of
the prefecture itself:-Td dvaToXixa irpai/rcopia, in legislative documents.101
He is thus in charge of thefiscaldioikesis of the provinces of the prefecture of
the East. The second Paul, however, was in charge of a smaller area,
namely the provinces within the military district of the Anatolikon army.
The change in emphasis, if such it is, must reflect a recognition of a state of
affairs in which the importance of new and developing provincial circum-
scriptions outweighed the relevance of the older establishment. Provinces -
eparchiai - were being seen from the standpoint of which thema they
belonged to, a point which the seals of the kommerkiarioi, apothekai and
kommerkia referred to already makes quite explicit. As the new thematic
districts became more closely defined, and as officials of the central bureaux
- the chartularies of the genikon in particular - were able to assume
responsibility for provincial fiscal administration under the local gover-
nors, so the need to appoint officials to oversee large groups of provinces
(perhaps a result of the obsolescence of the diocesan level of administration)
within, or over all of, the prefecture, will likewise have diminished.102
98
For example, Zacos and Veglery, no. 222, dated to the years 717-18, of anon., genikoi
kommerkiarioi of the apotheke of Colonea and all the eparchiai of Armeniakon', no. 155, with
table 18/2 (p. 164), seals of the Armeniakon redated by Seibt, in BS 36 (1975), 209 to the
early eighth century; no. 263, dated 745/6, of the kommerkia of the eparchiai of the
imperial God-guarded Opsikiow, and cf. nos. 242, dated 732/3, for the Anatolikom no. 261,
dated 741/2, for Thrakesion; nos. 258 and 259. dated 730-41, for Thrace; and so on.
99
MansiXI. 209B; Riedinger, 14.32.
100
Zacos and Veglery, no. 2290. Lilie, Thrakien und Thrakesion', 12 would identify the
two Pauls as one and the same, with an assumed promotion from apo hypaton to hypatos
in the interim. See Winkelmann, Rang- und Amterstruktur, p. 35. But in view of the
frequency of the name, this must remain hypothetical.
101
E.g., JGR I, coll. 1, nov. 2 5 , 2 (a. 629 = Dolger, Regesten, no. 199): TOIS evSogordTOis TCOV
dvorroXixwv lepu>v irpatTtopicov.
102
A late seventh-century seal of an anon., hypatos and chartoularios tou genikou (Zacos and
Veglery, no. 1231) shows that this bureau already had its officials responsible for
provincial fiscal administration.
 Fiscal administration 199

Finally, the fact that dioiketai responsible for 'the provinces' in a general
and non-specific sense seem no longer to issue seals after the first years of
the eighth century is extremely suggestive. It implies that officials with such
a wide jurisdiction throughout the empire were no longer necessary to the
fiscal administration. The fact that, with one dubious (and Western)
exception, all the known seals of dioiketai of named provinces are dated to
the eighth century and later, makes the conclusion that the two develop-
ments are somehow connected inescapable. 103
The significance of these general dioiketai must now be apparent. The first
of whom we have any mention is the Theodore whose seal, referred to
already, was issued some time between 614 and 631, a period of fiscal and
military crisis for the state. Thereafter, there are seals of six dioiketai 'of the
provinces' datable to the period c. 650-750, and there is mention of a fifth,
for the eparchiai of the East, in 680. Finally, for the years around 700, there
is a seal of a dioiketes of (the provinces of) the Anatolikon. All these officials,
with two exceptions, are very high-ranking persons, with the titles of
hypatos, apo hypaton or patrikios, testifying to their importance. 104 It is
surely no coincidence that the period in which these dioiketai functioned
begins at the time of substantial fiscal reforms under Heraclius and ends,
both as the themata begin to occur in literary sources and on seals as
administrative entities and as individual provincial dioiketai start to
(re)appear on issues of seals.
The solution to the problem of the general dioiketai is, therefore, an
administrative and institutional one. If, as seems probable, the genike
trapeza, or genikon, had become more or less independent during the reign of
Heraclius, whether already before his mint reform or, as seems more likely,
as a result of the ramifications of that reform, then the close connection
between this bank and the diocesan scrinia,105 which actually dealt with
the calculation and collection of the public taxes - a connection hitherto
assured by the presence of all these bureaux within the same officium - will
have been severed. Yet it is clear that in the developed genikon logothesion of
the ninth century, the descendants of the officials of the former diocesan
103
The exception is the seal of Theodore, dioiketes of Hellas, dated to the period 6 5 0 - 7 5 0 .
But it, too, m a y well be of eighth-century date. See Zacos a n d Veglery, no. 1 6 2 8 .
104 x h e t w o exceptions are the seals of Sergius (Schlumberger, Sig., p. 4 9 9 ) , described simply
as dioiketes ton eparchion (which m a y suggest a n earlier r a t h e r t h a n a later date - the
regular use of titles a n d r a n k increases during the seventh a n d especially the eighth
century. See W i n k e l m a n n , Rang- und Amterstruktur, p. 132); a n d of Marinus, apo epar-
chbn a n d dioiketes ton eparchion (Zacos a n d Veglery, n o . 1178). Apo eparchon seems to
h a v e been a relatively h u m b l e r a n k during the seventh century (see Zacos a n d Veglery,
no. 142; W i n k e l m a n n , Rang- und Amterstruktur, pp. 40f.).
105
Diocesan describes, of course, the competence of these scrinia. They were physically
located at Constantinople. For the theion logothesion at the end of Heraclius' reign, see
note 6 7 above.
200 Byzantium in the seventh century

scrinia, and their several provincial controllers, assessors and collectors of

taxes, were all within the general logothesion itself.106 At some stage,
therefore, the department of the general bank of the prefecture must have
absorbed or otherwise been given administrative authority over these
scrinia. Whether the diocesan scrinia themselves survived is difficult to say;
but originally each scrinium had (usually) one tractator or supervisor of
taxes for each of its provinces. The appointment within the general bank of
a single supervising dioiketes to oversee the operations of all the scrinia
(which will eventually, if it did not happen through administrative fiat,
have in practice brought the diocesan and provincial scrinia into the
purview of the general bank) would be a logical step, both to maintain
communication between the scrinia and the bank itself, and to co-ordinate
the extraction of revenue throughout the prefecture. Once the themata had
emerged as territorial administrative districts - in the last years of the
seventh century - that is, as groups of provinces and once the functions of
the old diocesan scrinia with regard to their constituent provinces had been
fully assumed by the enlarged department of the genikon - which could now
supervise the activities of the epoptai and dioiketai directly - then the general
supervisors, the dioiketai ton eparchion, became redundant. As we have
noted, themata are clearly seen as consisting of groups of provinces by the
early eighth century. At about this time, seals of general dioiketai cease to be
issued; and seals of named provincial dioiketai begin to appear. 107 While the
internal chronology of the process outlined must remain hazy, therefore,
the limits to the transformation are set by the process of fiscal administra-
tive reform and centralisation on the one hand, which seems to have begun
in the last years of the sixth century and reached a definitive stage with
Heraclius' reorganisation of the mint system in 627-30; and on the other
hand, by the appearance of thematic provincial groupings, the cessation of
references to general dioiketai and the concomitant appearance of indi-
vidual provincial tax officials in the early eighth century.
106
Kiel Phil, 113.26-115.4. Of the staff of the logothesion, the chartoularioi megaloi tou
sekretou represented the older chartularii scriniorum; the chartoularioi ton arklon (also
called exb chartoularioi: see Oikonomides, Preseance, p. 313) represented the former
praepositi thesaurorum of the comitiva sacrarum largitionum; the epoptai and dioiketai
represent the former censitores and inspectores, and tractatores for each province, responsi-
ble to the diocesan scrinium within which their province was located. The koummerkiarioi
(sic), by this time simply customs and trade-regulation officials, are descended from the
comites commerciariorum of the sacrae largitiones; while the komes tes Lamias represents
the old prefectural comes horreorum. On the latter, see Haldon, 'Comes horreorum -
Komes tes Lamias?'; for the other officials, see Notitia Dig., or. XIII (comes sacr. largit.) and
Jones, LRE, vol. I, pp. 449-51, 455.
107
Of course, the sigillographic material itself cannot form an entirely solid basis for
argument, since there may be seals which do not fit this pattern. But the seals known at
the present time do support the argument; a r 1 statistically the sample does provide a
sound basis for generalisation.
 Fiscal administration 201

THE FATE OF THE PRAETORIAN PREFECTURE: EPARCHS AND

PROCONSULS
Whether the praetorian prefect still had a role to play is, in view of the
almost total lack of evidence in this regard, impossible to say with certainty.
That the post continued to be filled, however, possibly as a civil administra-
tive figurehead, is suggested by the existence of two seals of the period
650-700, of Marinus, OLTTO eirdpxwv xai SLOIXTJTTJS TCOV euapxuov, and
of the same Marinus, now promoted, diro vrraTtov xai errapxos TCOV .. . 1 0 8
It seems highly likely that the latter title should be completed rrpaiTcopicov;
although a thematic title, such as Anatolikon or Armeniakon is a possibility.
If the first suggestion is correct, then the praetorian prefect was still in
existence in the later seventh century and, furthermore, probably still
exercised a general authority over the departments formerly under his
direct control. 109 That the office survived until the early ninth century is
clear from its inclusion in a ceremony dating to the years 809-43,
although the status of the title is open to question. 110 Like that of
magister officiorum, the post of praetorian prefect of the East may
have retained some vaguely defined supervisory capacity, as has already

108
See Zacos and Veglery, nos. 1 1 7 8 a n d 1 1 7 9 . A further three seals of Marinus, as apo
hypaton kai dioiketes ton eparchion exist. See Zacos a n d Veglery, note to no. 1 1 7 8
(Konstantopoulos, Molybdoboulla, no. 58; N.P. Lihacev, Istoriceskoe znacenie italogreceskoi
ikonopisi Izobrazeniya Bogomateri (St Petersburg 1911), 1 1 7 (figs. 2 5 9 a n d 2 6 0 ) .
109
If the second suggestion is preferred, then Marinus m a y well h a v e been one of the
thematic prefects responsible for the provisioning of the provincial military at this time.
See Kaegi, 'Two studies', 104ff., esp. 107ff.
110
The history of the debate a r o u n d this passage, first discussed in detail by Stein, 'Ein
Kapitel', 7 0 - 8 2 , h a s been briefly detailed by Kaegi, 'Two studies', 99ff. Stein based his
date for the passage (which actually describes a ceremony for the feast of Easter, not for
Pentecost as in the manuscript), o n the presence of w h a t he s a w as a series of archaic
Latin terms a n d titles. He concluded from this, a n d from the fact t h a t a dioiketes ton
anatolikon eparchion, b u t not a praetorian prefect, appears in 6 8 0 at the sessions of the
sixth ecumenical council, t h a t the praetorian prefect h a d therefore ceased to exist by 6 8 0 .
He concluded further t h a t the passage m u s t itself date to the later seventh century, but
before 6 8 0 . The importance of the passage in the De Cerimoniis for Stein lay in the fact
t h a t both the praetorian prefect a n d the 'eparchs of the themes' occur a m o n g the list of
dignitaries presented to the emperor. The passage in question reads: p^Xov TeTapTOv • TOV
iJirapxov TWV TrpaiTwpieav, TOV x o i a i o r w p a , dvfruTrdTOu<s TWV fteixaTcov x a i eirdpxovs
{DeCer. 6 1 . 1 5 - 1 6 ) .
Two sets of observations need to be made. In the first place, it m u s t n o w be clear that,
unless the passage h a s been very heavily interpolated (which would render it useless for
Stein's a r g u m e n t also), it m u s t date from after 8 0 9 , since the domestikos of the Hikanatoi is
listed also {ibid. 6 1 . 1 7 , cf. also 67.23), a unit which w a s first established in t h a t year (see
Haldon, Byzantine Praetorians, pp. 245f.). Quite a p a r t from this, the presence of tagmatic
officers of the scholai, a n d of the vigla, the former established as a tagma only u n d e r
Constantine V, the latter established probably by Eirene, makes it clear t h a t the passage is
at the earliest of the early n i n t h century (Haldon, Byzantine Praetorians, pp. 228ff. a n d
236ff.; note also Guilland, Recherches, vol. n , p. 70). In general style, a n d in respect of the
202 Byzantium in the seventh century

been suggested, perhaps over the provisioning and supplying of troops

in the provinces: the thematic eparchs (see below), who do seem to
have fulfilled this function until their replacement during the first half of
the ninth century by thematic protonotarioi appointed from the sacellium,
certainly came under the authority of the praetorian prefect.111 And it
may well be that the praetorian prefect who survives in this early ninth-
century ceremony, equivalent in rank to the quaestor - with whom he was
grouped - did retain a general authority over the administration of
military supplies, through his subordinates, the anthypatoi and eparchai of
the themata, exercised through a much-reduced sekreton at Constantinople,
with a local bureau in each thema.
Now it has been suggested that the thematic protonotarioi, who begin to
appear in the literary sources from the second quarter of the ninth century,
and on seals shortly before this, were originally the local representatives of
the office of these thematic eparchs who, when the praetorian prefecture
and the dependent thematic eparchate were finally superseded, inherited
the functions of the thematic eparch within the administration of the
strategos.112 But the protonotarioi were members of the sakellion at a slightly
later date, 113 and it is probable that this reflects a long tradition, since it

palatine titles which occur, it could well be contemporary with the Taktikon Uspenskij of
842/3. The text can tell us little or nothing about the later seventh-century or indeed the
early eighth-century establishment at all.
In the second place, where Stein assumed that the title anthypatos, which occurs in
respect of two groups of dignitaries {De Cer. 61.13: TTcn-pixious TOIS xai dv-fhrn-cmns;
61.16. dv^hmdTOvs TWVfteiAcmovxai eirdpxovs) signified, at least in the second case,
two distinct functions, it seems more likely that one group is meant, the eparchs of the
themata who were also anthypatoi. Note De Cer. 67.17: eirapxio-a*; ftejxaTixds
avfruTTaTLaas (the presentation of the wives of the respective officials on the same
occasion): and cf. T.Usp. 51.25: oi dv-fhrrraTOi xai eirapxoi T<OV -deixdTwv. The
eparchs and anthypatoi clearly represent one group, and Stein's whole discussion of the
question of the 'survival' of proconsuls therefore needs considerable revision. Guilland's
attempt (Recherches, vol. II, p. 69 note 15) to emend the text of De Cer. 67.17, which he
clearly thought problematic, in order to fit in with Stein's view, which he shared, by
inverting the order so that it would read dvftvrraTioro-aq, fteixomxas, eirapxuras, must be
regarded with some suspicion. Correctly understood, such an arbitrary intervention in
the text is quite unnecessary. It is worth noting that two seals dated to the first half of the
ninth century, of Eustathius, imperial spatharokandidatos and anthypatos of the Anatolikoi
(Zacos and Veglery, no. 1901); and of John, imperial spatharios and anthypatos of the
Anatolikoi (Zacos and Veglery, no. 2049) do provide evidence for thematic officials
bearing the title anthypatos, suggesting that there may well have been thematic anthypa-
toi acting as civil governors, even if by the time of the Taktikon Uspenskij they had been
combined in practice with the position of thematic eparchs. Stein's argument is, there-
fore, partially vindicated.
111
Kaegi, 'Two studies', 106f., and De Cer. 61.15-16.
112
Kaegi, 'Two studies', 109ff., partly following and emending Stein, 'Ein Kapitel', 79ff. See
especially Winkelmann, Rang- und Amterstruktur, p. 142 with the evidence from both
seals and literary sources cited there. See also Haldon, Byzantine Praetorians, pp. 215f.
113
Klet. Phil. 121.6; cf. Oikonomides, Preseance, p. 315.
 Fiscal administration 203

was the sakellarios, with his bureau, who in the late sixth and seventh
centuries seems to have been responsible for the general fiscal supervision
of state departments. More probably, the establishment of the themata as
territorial administrative units from the early eighth century meant an
increasing intervention in and supervision of the activities of the thematic
eparchs by the sakellarios, and the consequent increasing irrelevance of the
officials of the old prefecture in favour of those of the sakellion.
Two further arguments would favour this general hypothesis. In the first
place, it has plausibly been argued that the eparchs of the themata, who
occur in the ninth-century protocol in the Book of Ceremonies, and in the
Taktikon Uspenskij (dated to 842/3), are the descendants of those ad hoc
praetorian prefects appointed during the sixth and early seventh centuries
to supervise the arrangements for supplying troops in transit. 114 The fact
that they bear the title of anthypatos, however, deserves more attention.
Stein believed that the title represented a separate office and argued that
the anthypatoi of the themata were the civil governors, and that the eparchs
were the governors of the provinces within the themata.115 This line of
reasoning then proceeds to argue that, since the ceremony is to be dated to
the seventh century (which is, as we have seen, incorrect), the mention of
these two groups of officials is evidence for a thematic civil administration
by 680 at the latest, and therefore for the existence of the themata as both
civil and military administrative units by this date. The argument is based
on several misapprehensions and is now generally rejected.116 But Stein
was not entirely on the wrong track. For the fact that the eparchs are also
anthypatoi, that anthypatos is not yet a title or rank, even in the Taktikon
Uspenskij of 842/3, and that, crucially, thematic anthypatoi are evidenced,
for the Anatolikon at least, on seals of the early ninth century, is impor-
tant. 117 Given the continued existence of a praetorian prefect in the early
ninth century - whatever his exact functions - it is not entirely improbable
that the posts of the former ad hoc prefects who had been responsible for
military provisioning, and those of the leading civil governors of the
provinces within the thema for which they were responsible, were eventu-
ally amalgamated. Both groups were at the disposition of the praetorian
prefect; both would have fulfilled functions which were in practice very
closely related. The fusion of the two, with the result that thematic prefects

114
Kaegi, 'Two studies', 103ff.
115
Stein, 'Ein Kapitel', 71f. Anthypatos had been the title of several provincial governors in
the sixth century and earlier, although the majority of such governors bore titles such as
praeses, moderator, or consularis. See Jones, LRE, vol. Ill pp. 386-9, table, col. 10.
116
See the detailed critique by Karayannopoulos, Entstehung, pp. 55ff.
117
See Oikonomides, Preseance, pp.287 and 294 with literature; Winkelmann, Rang- und
Amterstruktur, pp. 35f. and above, note 110, for the seals.
204 Byzantium in the seventh century

were also the civil governors of the bureau of the praetorian prefect in
Constantinople, would not have been illogical.
In the second place, it is surely significant that, at approximately the
time after which the thematic eparchs and anthypatoi are last mentioned
(that is, in 842/3), both thematic protonotarioi and thematic strategoi
bearing the title anthypatos appear in the sources. 118 It is tempting to see in
these two developments a related phenomenon, namely, the phasing out
(or abolition) of the skeletal residue of the old prefecture, with its thematic
representatives (anthypatoi and eparchs combined), and the transfer of the
civil governorships to the thematic strategoi - hence the use of the older
functional title as a rank granted to the governor of a thema; and the
concomitant establishment of representatives of the sakellion within the
themata, the protonotarioi, to take up the role of the older eparchs in respect
of provisioning and supplying the army, liaising with the central sekreta
(the genikon, eidikon and, to a degree, the stratiotikon), as the eparchs would
have had to do before them. What was involved, therefore, was the
removal of an intermediate administrative instance, that of the thematic
proconsular eparchs; and the establishment of direct surveillance from the
sakellion. That the protonotarioi replaced the eparchs/proconsuls func-
tionally, but not in terms of power and rank, is evident from the fact that,
whereas strategoi, and eventually other functionaries also, come generally
to hold the rank of anthypatos, the protonotarioi in the sakellion remain fairly
humble officials: mostly of the rank of hypatos, spatharios, kandidatos and

118
The first mention of the title anthypatos as a rank seems to be that in Theophanes cont,
108.1 (cf. Guilland, Recherches, vol. II, p. 71; Oikonomides, Preseance, p. 294; Bury,
Administrative System, p. 28) for the reign of Theophilus. In the Taktikon Uspenskij,
however, firmly dated to 842/3 (Oikonomides, Preseance, p. 45) the term seems still to be
a function, see T.Usp., 51.25; with the possible exception of the entry 6 iraTpixios xai
AvduTraTOq which, it has been suggested, represents the position created for Alexios
Mousele by Theophilus referred to in Theophanes cont. above (see T.Usp., 49.1 and Bury,
Administrative System, p. 28. For its still being a functional rank, see Winkelmann, Rang-
und Amterstruktur, p. 35). The dating of a number of seals of the early ninth century
bearing the title anthypatos, on the other hand, suggests that it had already towards the
end of Eirene's reign a titular value - see the seals of Gregory Mousoulakios, who appears
to have held the ranks of anthypatos, patrikios, while also being komes of the Opsikion (see
Haldon, Byzantine Praetorians, p. 360 and note 1100; Zacos and Veglery, no. 3113A;
and Seibt, Bleisiegel, no. 158). But Gregory may be, like Alexios Mousele after him, an
exception. Other seals bearing the title are all to be placed in the middle decades of the
ninth century or later: see the list and commentary of Winkelmann, Rang- und Amter-
struktur, pp. 35-6. By the later ninth century, in contrast, and certainly by the time the
Kletorologion was compiled, all the strategoi of themata, as well as members of the central
sekreta holding the higher posts, could be anthypatoi. See Guilland, Recherches, vol. II,
pp. 71ff. for a prosopographical list.
 Fiscal administration 205

the like. 119 Seals of protonotarioi are extant in considerable numbers, seals
of anthypatoi are relatively few. Two, with specific thematic titles (of the
Anatolikon), have been mentioned. Others cover the period from the
seventh to the early ninth centuries. 120 Whatever the explanation for the
imbalance, there is nevertheless a clear continuity from the sixth century
until the middle of the first half of the ninth century in the use of the title
anthypatos, often in conjunction with other ranks, to denote an office, and
not simply a dignity.
However uncertain the picture, therefore, there is good reason for
believing that the praetorian prefect continued to exist until some time in
the period of 809 to 842; and that his bureau, which was responsible for
elements of the provincial civil administration and the supplying and
provisioning of provincial troops was represented by anthypatoi and
eparchs, who in the final stages seem to have been amalgamated as a
single post in the themata. This amalgamation (with one such joint post for
each thema?) is the more likely in view of the fact that both officials were
responsible for exactly the same territorial circumscriptions. The general
hypothesis is supported by the survival of the older provincial governors,
or praetors, until the time of the Taktikon Uspenskij in 842/3, and their
subsequent emergence as the leading judicial officials in the themata. Once
more, a key element of the traditional civil and fiscal administrative
apparatus continued to play a role into the ninth century; and it seems
unlikely that all these titles represent no more than a fossilised echo of a
past system. Within the framework I have outlined, they played an active,
functional role in the state establishment.
A final point is worth making here. The boundaries of the original large
themata in Asia Minor corresponded, as we shall see, fairly closely, and in
most cases exactly, with older provincial boundaries. There was no agree-
ment with diocesan boundaries, however, since the themata represented
areas which could support a certain number of soldiers, as well as a degree
of strategic planning. This will be demonstrated below. Each thema,
however, did consist of a group of older provinces, and the question arises

119
For example, Zacos and Veglery, nos. 3118, 2496, 3214, 1727, 2324 and many more.
See the remarks of Winkelman, Rang- und Amterstruktur, p. 142, with lists at pp. 120 and
122-31 where some 65 seals of protonotarioi are listed. By the ninth century, these
formerly prestigious titles had been greatly reduced in status.
120
See note 110 above. The other seals are: Zacos and Veglery, no. 2881, of John,
anthypatos (550-650); no. 775, Konstantinos, anthypatos (550-650), no. 1085,
Tryphon, Stratelatis and anthypatos (late seventh century); Schlumberger, Sig., 438,4;
David, anthypatos (eighth-ninth century); Konstantopoulos, Molybdoboulla, no. 295:
John, illoustrios and anthypatos (seventh to eighth century); cf. also Schlumberger, Sig.,
p. 438,2, seal of Andreas Botaneiates, imperial spatharios and anthypatos.
206 Byzantium in the seventh century

as to why the titles of anthypatos on the one hand (representing civil and
fiscal administration) and praetor on the other (representing judicial
administration), which in Justinian's time had equivalent status (both were
spectabiles), came to have different statuses by the early ninth century and
how the division of labour came to be attached to the two titles in this way.
Other titles, such as moderator or comes, were also available, but were not
retained. The only reasonable answer would suggest that some deliberate
administrative/legislative decision was taken at a certain point in the later
seventh or early eighth century; and that, perhaps for traditional reasons of
association, the two terms were thus assigned, the paramount civil
governorships being described as anthypatoi, the judicial functions being
awarded to the praetors. But the Taktikon Uspenskij would not contradict
the notion that, until this time at least, the term praetors of the themata
referred to the civil governors of the individual provinces within each
thema, while the anthypatoi and eparchai were responsible for the whole
thema, including all its constituent provincial subdivisions.121
To summarise briefly the results of the foregoing analysis. The evidence
points to a major restructuring of the fiscal administration of the state
during the reign of Heraclius. While the reform and reorganisation of the
system of mints constitutes the major single change, the logical con-
sequence of this centralisation of administration for the provincial organi-
sation of the sacrae largitiones, together with the evidence for the increas-
ingly important role of the sakellarios, and the appearance of high-ranking
logothetai during the reigns of both Phocas and Heraclius, seems to be that
the sekreta of the genikon, eidikon and stratiotikon, together with the sakellion
and the vestiarion, already existed in their essential form by the end of
Heraclius' reign. Apart from the reorganisation of the mints, however,
which itself may well have been a response to pre-existing tendencies,
exacerbated by the civil war and the war with the Persians, these changes
- essentially a reversal of the pattern which had hitherto operated -
evolved as logical and necessary consequences of each successive adminis-
trative alteration in the older establishment and follow quite consistently
from one another. They also conform to the possibilities presented by the
available evidence and represent the only coherent explanation for that
evidence.
At the same time, shifts in the pattern of provincial fiscal administration
occurred. In order to ensure a coherent policy, and to fill the gap left by the
disappearance of the diocesan tier, general supervisors were appointed to
oversee provincial tax-assessment and collection; and it is presumably not

121
See T.Usp., 53.3 (and note Oikonomides, Preseance, p. 344, who makes a similar sug-
gestion).
 Fiscal administration 207

a coincidence that the first such official of whom we learn, Theodore, was
appointed during the reign of Heraclius, and possibly at the same time as
the reorganisation of the mints took place. By the early eighth century,
however, the establishment of the provincial field armies in Asia Minor and
the fixing of their boundaries lent to these new districts an administrative
identity (the equivalent of a group of provinces) which provided an
intermediate level of administration. They replaced, in effect, an adminis-
trative level similar to that represented by the older dioceses. In con-
sequence, fiscal supervisors could be appointed to these districts, the
themata, with their subordinate provinces, and the general (prefectural)
supervisors were superseded. Similarly, while the praetorian prefect
remained in existence, the civil provinces within the themata formed new
groupings, placed under a senior civil official, the proconsul, again equiv-
alent in many respects to the vicarii of the older dioceses; while the ad hoc
prefects appointed from the sixth century to deal with troops in transit
were made permanent as a result of the permanent presence in what had
been predominantly civil provinces of the field armies. By the later eighth
or early ninth centuries the functions of these two 'thematic' civil officials
had come to be formally assimilated; while the gradual evolution of the
various fiscal bureaux at Constantinople meant that their role was increas-
ingly redundant. Eventually - some time after 842/3 - they were abol-
ished, their function being assumed by the thematic protonotarioi of the
sacellium, their high status being transferred to the thematic governors or
strategoi, who were henceforth formally endowed with both civil and
military authority.
This is only a part of the story, of course, and it is necessary now to turn
to the military administration of the state, and to the vexed question of the
military provinces themselves, the themata.
 CHAPTER 6

The state and its apparatuses: military

administration

THE THEMATA
As with most other aspects of the state's administrative machinery, its
structures of military organisation also evolved along lines which resulted
in a system of very different appearance in the eighth and ninth centuries
from that which operated in the sixth and earlier seventh centuries. A
number of processes, which form in practice an evolutionary whole, need
to be clarified or explained. All, however, depend to a greater or lesser
extent on one key problem: the origins and development of the so-called
themata or military provinces and the phenomenona associated with them
- the mode of recruiting, equipping and supplying the soldiers, the
methods of paying them, and the fate of the civil provincial government.
The differences between the late Roman army of the sixth century and
that of the Byzantine state of the ninth century are not difficult to perceive.
The late Roman system was characterised by a fundamental division
between the civil and military spheres of administration: civil officials had
no authority over the military, and vice versa. The armies were divided
into two essential groups: those of the mobile field forces, technically
referred to as comitatenses (although in their turn made up of a number of
distinct types of unit, differentiated originally by their posting, their mode
and source of recruitment and so on), and the permanent frontier or
garrison units, described and known as limitanei (although again made up
from units of widely differing origins).1 But while in theory remaining
mobile, many units of the comitatenses came to be based for long periods in
1
For detailed descriptions and analyses of the late Roman army, see Jones, LRE, vol. II,
pp. 607-86; D. Hoffmann, Das spdtromische Bewegungsheer und die Notitia Dignitatuum
(Epigr. Studien VII, 1, Dusseldorf and Cologne 1969); R. Grosse, Romische Militdrgeschichte
von Gallienus bis zum Beginn der byzantinischen Themenverfassung (Berlin 1920);
Th. Mommsen, 'Das romische Militarwesen seit Diocletian', Hermes 24 (1889), 195-279;
A. Miiller, 'Das Heer Iustinians nach Prokop und Agathias', Philologus 71 (1912), 101-38;
G. Ravegnani, Soldati di Bisanzio in eta Giustinianea (Materiali e Ricerche, Nuova Serie VI
Rome 1988).

208
 Military administration 209

one area or garrison town, the soldiers putting down local roots, and even
- as some examples from Africa, Italy, Egypt or Palestine demonstrate -
taking up other occupations.2 For the limitanei, recruitment was assured
through hereditary conscription, and membership of these units was
regarded as a privilege, in view of the favourable tax-status of soldiers who,
while subject to the regular land-and-head tax, were exempted from all
extraordinary burdens. It was further assured through various forms of
voluntary recruitment for the regular units of the comitatenses.3 Weapons,
uniforms and mounts were provided by the state, which also issued cash
allowances for their purchase, through the commissariat, and from civil-
ian suppliers or - as in the case of weapons and some items of clothing -
from the state factories or workshops. 4
The system of command at the beginning of Justinian's reign was, in
most respects, unchanged from that of the fifth century. The limitanei along
the frontiers were divided into a number of independent commands,
consisting of nine duces (of Palestine, Arabia, Phoenice, Syria, Euphraten-
sis, Osrhoene, Mesopotamia, Pontus, Armenia) on the eastern frontier, a
comes rei militaris in Egypt, three duces, of the Thebaid, Libya and the
Pentapolis in North Africa; and five duces, of Scythia, Dacia, Moesia I and II
and Pannonia II on the Danube. The field armies were similarly grouped
into divisions, in this case five, two of them in and near Constantinople (in
Thrace and in north-west Anatolia) under the two magistri militum
praesentales, one in Thrace, one in Illyricum and one in the East. In
addition, military officers entitled comites rei militaris, who had authority
over civil governors also, were appointed to provinces where banditry was
a problem, notably Isauria, Pamphylia, Pisidia and Lycaonia. While the
master of offices exercised a general inspectorate over the limitanei, both
they and the field troops were subject directly to the authority of their
respective magistri militum. Justinian introduced several changes. The zone
of the magister militum per Orientem was divided into two, and a new
magister militum per Armeniam took over the northern section; new com-
mands for south-eastern Spain, for Italy (magister militum per Italiam) and
for Africa (magister militum per Africam) were established after the recon-
quest, the latter circumscription including also the isles of Sardinia and
Corsica. A combined military civil command of the Long Walls (established
by Anastasius under two vicarii, one civil and one military) under a praetor
2
Italy: Brown, Gentlemen and Officers, pp. lOlff.; Egypt: Jones, LRE, vol. II, pp. 662ff.;
Syria/Palestine: Patlagean, Pauvrete economique, pp. 255ff. and 313-15; Durliat, 'Les
Finances municipals', 379 and 'Les grands proprietaires et l'etat byzantin', 527f.
3
For sources and literature, see Haldon, Recruitment and Conscription, pp. 20-8 and Byzan-
tine Praetorians, pp. 103ff.
4
See Jones, LRE, vol. II, pp. 671 and 834-7; Haldon, Byzantine Praetorians, pp. 114f. with
literature.
210 Byzantium in the seventh century

of Thrace, was also established. The number of commands of limitanei was

increased along the eastern front to deal with Persian incursions; and in
the Armenian zone the two duces of Armenia and Pontus were replaced by
five commands along a more advanced frontier. Similar commands were
established along the African frontier, in Tripolitania, Byzacium, Numidia
and Mauretania, Sardinia and in northern Italy. In general, Justinian main-
tained the division between military and civil spheres of authority; but in
Asia Minor he combined civil and military authority to deal with brigand-
age (in Isauria, Pisidia and Lycaonia); in Egypt the position of Praefectus
Augustalis and dux were combined; and the dux of the Thebaid was given
civil administrative authority to deal with raids from the desert tribes.
Apart from these units and their dispositions, the emperors also had the
palatine units of the scholae (by the sixth century a parade-force only) and
the excubitores, created by Leo I, who replaced the scholae as the elite
imperial bodyguard. In addition, the palatine corps of protectores et dom-
estici, both horse and foot, under their comites (later a single comes) formed
another parade unit.5 Justinian's main strategic innovation was the estab-
lishment of the quaestura exercitus, comprising Caria and the Aegean isles,
together with the Danubian provinces of Scythia and Moesia II. The aim
was to secure supplies and a sound base for the frontier units which could
thus avoid further impoverishing an already devastated region.6
After Justinian's reign, several administrative changes were made, and
certain organisational reforms within the structure of the field armies took
place. Most of these seem to have occurred in the reign of Maurice. Of the
administrative changes, the establishment of the two exarchates of
Ravenna and Carthage were responses to the arrival of the Lombards on
the one hand, and to the constant raiding of the Berbers along the African
limes on the other. This move was partially the formal recognition of what
had hitherto been de facto the case, namely the greater importance of the
magister militum in each of the two prefectures as a result of the prevailing
political and military situation. Already both Solomon and Germanus had
combined the offices of magister militum with that of praetorian prefect in
Africa; and although the praetorian prefect continued to exist as an
independent office separate from the exarch, the latter was officially
endowed with supreme authority.7 It has recently been demonstrated in
addition that the mints of both Carthage and Ravenna began to produce
(for the first time in the former case) a regular issue of dated solidU those of
Carthage dated indictionally, those of Ravenna by imperial regnal years.
5
See Jones, LRE, vol. II, pp. 655-7: Haldon, Byzantine Praetorians, pp. 119-41.
6
Jones, LRE, vol. I, p. 280. Cf. Justinian, Nov. 41 (a. 536) and Lydus, De Magistrates II,
28-9 (pp. 83-5).
7
See Jones, LRE, vol. II, pp. 655-6; Brown, Gentlemen and Officers, pp. 46ff.; Guillou, Riqio-
nalisme, pp. 150ff. For the older literature, see Diehl, Etudes byzantines, pp. 157-84;
L'Afrique byzantine, pp. 484ff.
 Military administration 211

This occurred between December 581 and August 582. Some time
between August 582 and August 583 a new mint, producing copper and
possibly also gold, was established at Catania in Sicily, which had since
537 been under a separate jurisdiction jointly administered by the quaestor
sacri palatii and the comes patrimonii per Italiam. The first reference to an
exarch for Italy is for 584; the first for Africa, 591. 8 There is little doubt
that this reorganisation of minting and the new issues which resulted
reflected also the new statute of the affected prefectures, and these develop-
ments seem to have coincided with and reflect the establishment of the
exarchates, with their greater political, military and fiscal independence.9
In the tactical organisation of the armies, reforms also took place, mostly
during the reigns of Tiberius Constantine and Maurice. The numerous
units of foederati, effectively by this time units of mixed - Roman and
barbarian - regulars in the comitatenses, came to be brigaded together in a
single corps. Similarly, the private bucellarii of the leading officers in the
praesental forces were also brigaded together as an elite force and were
incorporated into the state establishment, receiving pay and provisions on
the same basis as the regular troops. Both developments seem to have
occurred in the context of the wars in the Balkans in the 580s and 590s;
and at the same period an elite 'foreign legion' of Germanic cavalry, the
optimates, seems to have been established, made up chiefly of Gothic and
Lombard mercenary soldiers at first, although later recruited from within
the empire also. These changes were accompanied by reforms in the
command structure of the armies, which represented the culmination of a
long-drawn-out development in which the older internal structure of
regular units dating back to the time of Diocletian and Constantine was
replaced or modified by a system which more closely reflected the func-
tional needs and the language of the late-sixth-century army.10 Finally,
there is some evidence that the old division between the two magistri
militum praesentales and their armies ceased to have any practical sig-
nificance, so that there came to be usually only one praesental force under
a supreme commander, normally operating in Europe. This development
was to have important consequences in Heraclius' reign and after.11
The system which is described in the sources of the ninth century is very
different. The magistri militum, with their field armies, and the limitanei
under their duces are no longer to be found. Instead, the territories
8
See Pelagius II, Ep. I, 703-5 (PL LXXII); Greg. I, Ep. I, 59.
9
For the most recent treatment, see Hendy, Studies, pp. 406ff., with sources and literature.
10
On these developments see Haldon, Byzantine Praetorians, pp. 96ff. and 107ff; and Recruit-
ment and Conscription, pp. 31-2.
11
Byzantine Praetorians, pp. 176ff. See also idem. Administrative continuities and structural
transformations in East Roman military organisation c. 5 80-640', in L'Armee romaine et les
barbares du 4e au 7e siecle, eds. F. Vallet, M. Kazanski (Paris 1993)45-51 (repr. in State, Army
and Society in Byzantium, V).
212 Byzantium in the seventh century

remaining in the empire (or regained since the seventh century) were
divided up into a number of military provinces, or themata, each governed
by a strategos, or general, who - at least in terms of general authority -
held both civil and military powers; although the civilian officials and their
spheres of action within each thema were closely supervised by their
respective chief bureaux at Constantinople. The themata varied in size, but
were subdivided for military purposes into tourmai, or divisions, each
consisting of a number of drouggoi, or brigades, the latter in turn made up
of several banda, or regiments. By the ninth century, the chief territorial
division within the thema was the bandon, also called a topoteresia; but it is
clear that the older provinces of the late Roman period survived well into
the eighth century and possibly into the early ninth. The strategoi were
subject directly to the emperor, by whom they were nominated. They
commanded mixed forces of infantry and cavalry, made up of volunteers
and conscripted soldiers, the latter enlisted according to an hereditary
personal obligation. The frontier districts were organised into 'passes', or
kleisourarchiai, each under a kleisourarch.12
The thematic administration, which by the middle of the ninth century
clearly included also the civil government of the areas encompassed by
each thema, was controlled from Constantinople by the various fiscal
administrative sekreta - the genikon, eidikon and stratiotikon, supervised by
the sakellion and the sakellarios - through their thematic representatives:
the protonotarios for the sakellion; the chartoularioi of the provincial treasu-
ries for the genikon, along with their subordinate fiscal officials - dioiketai,
epoptai, exisotai and so on - and the thematic chartoularioi of the stratioti-
kon. The latter were attached directly to the staff of the strategos, who had
his own corps of guards, and a staff of clerical officials, in addition to the
military officers of his command. 13
While the number of themata had multiplied during the eighth century
and especially the ninth century, partly through the subdividing of older
and larger themata, partly through the acquisition of new (reconquered)
territories, there had been originally only four such regions: the Opsikion,
the Anatolikon, the Armeniakon and the Thrakesion; together with a fifth
division, the fleet of the Karabisianoi, which included the islands of the
Aegean and part of the south-west coast of Asia Minor. It is now generally
agreed that the Thrakesion army, which represented the older division of
the magister militum per Thracias, had been established in the area later
known as the thema of the same name some time during the middle years
12
For a descriptive survey of the themata, see CMH, vol. IV, part 2, pp. 35ff.; Toynbee,
Constantine Porphyrogenitus, pp. 134ff. and 224ff. For further literature on their origins
see chapter 5, note 70 above.
13
For a summary, see Oikonomides, Prese'ance, pp. 34Iff. with literature.
 Military administration 213

of the seventh century and that it is as old as the other original themata.14
During the reign of Justinian II these five divisions existed alongside the
forces of the exarchates of Italy and Africa; and, as has long been recog-
nised, they were descended respectively from the armies of the magistri
militum praesentales, per Orientem, per Armeniam, per Thracias, as we have
already noted, and from the forces under the quaestor exercitus.15
The first reference to these divisions together occurs in the iussio of the
Emperor Justinian II, sent to the pope in 687 in confirmation of the acts of
the sixth ecumenical council of 680. As witnesses to the orthodoxy of the
emperor and his subjects, the emperor lists the following:
deinceps militantes incolas sancti palatii, nee non et ex collegiis popularibus et ab
excubitoribus, insuper etiam quosdam de Christo dilectis exercitibus, tarn ab a Deo
conservando Obsequio, quamque ab Orientali, Thraciano, similiter et ab Arme-
niano, etiam ab exercitu Italiano, deinde ex Caravisiensis et Septensiensis seu de
Sardinia atque de Africa exercitu.16
By this date therefore the basic elements of the later system of thematic
armies appears to have been established. It is the process of this estab-
lishment, the origins of the armies concerned, and the transformation from
late Roman institutions and practices to which I shall now turn.
Two questions present themselves: When did these forces occupy Anato-
lia? And how were they recruited, organised and supported? Two related
questions follow: What was the relationship between these forces and the
civilian population, and between their commanders and the civil admin-
istration? And what was the relationship between the establishment of
these forces in Asia Minor and the later system of military land-holding
familiar from the legislation of the tenth century? The first and third
questions are probably the most straightforward, and I will deal with them
first.
Thefirstexplicit reference to the Opsikion (Latin Obsequium) division is for
680, when the komes of the Opsikion is listed among the officers accom-
panying the emperor during the sessions of the sixth ecumenical council;
although there is good reason to think that the Opsikion army existed

14
See the now out-of-date but still important work of Gelzer, Themenverfassung, pp. 9f. and
19ff.; and also Ostrogorsky, Geschichte, pp. 8Of., with map and literature; Lilie, Die
byzantinische Reaktion, pp. 2871T.; Haldon, Byzantine Praetorians, p. 165. For the first
references to the districts and armies of the themata in the sources, see Pertusi, De
Thetnatibus, comm. pp. 104-11, 114-20, 124-30 and 149ff. For the Thrakesion thema,
Lilie, '"Thrakien" und "Thrakesion"', 22-7.
15
See Pertusi, De Thematlbus, comm. pp. 105f. and 149; Haldon, Byzantine Praetorians, p.
165 with literature; H. Antoniadis-Bibicou, Etudes d'histoire maritime a Byzance, a propos du
Theme des Caravisiens (Paris 1966), esp. pp. 63ff.; Toynbee, Constantine Porphyrogenitus,
pp. 235f.
16 MansiXI, 737f.; Riedinger. 886.20-25.
214 Byzantium in the seventh century

already in the 640s and before;17 in addition, it is almost certain that the
Armenian noble Mzez Gnouni - the Mizizios of the Greek sources - was
komes of the Opsikion in the 660s, and up to his death after the assassi-
nation of Constans II in 668. 1 8 Certainly by the late 680s this army was
seen as occupying a defined region, as later sources testify, although the
evidence of seals of kommerkiarioi show that the traditional provincial
names continued to be employed by contemporaries as before.19 The
Armeniakon and Anatolikon themata appear in references for the years 667
and 669 respectively;20 the Thrakesion thema is that referred to as the
Thracianus exercitus in the iussio of 687 referred to above;21 while the
Caravisiani - Karabisianoi - appear in the same document, and at approxi-
mately the same time in an account contained in the collection of the
miracles of St Demetrius at Thessalo'niki.22
The word thema is itself a problem. In the older literature, and in
particular in the work of Ostrogorsky, it was argued that it bore a specific
technical meaning, namely a definite region, occupied by a military corps
commanded by a strategos, who was the governor-general of his region
and in whose hands both civil and military authority were united.23 This
view has been challenged from a number of different perspectives, but most
importantly, the earliest reference to the word themat in the ninth-century
Chronography of Theophanes Confessor, has generally been regarded as
an anachronism, contemporary with the compiler of the chronicle, not
with the events described. The case cannot actually be conclusively proved
in either direction. In the last analysis, the most that can safely be said
about the term in these earliest occurrences is that it refers merely to
armies: it is clear that in mentions of themata from the later seventh century
this is how the word is used.24 That it was later applied to the districts
where these forces were based is not in doubt. But this does not seem to
happen until the later seventh century (according to the ninth-century
historical sources) or the beginning of the eighth century (according to the

17
MansiXI.209;Riedinger. 14.20-21; Haldon, Byzantine Praetorians, pp. 144f. and 175-80.
18
Haldon, Byzantine Praetorians, pp. 179f.
19
See Theophanes, 364; Nicephorus, 36. 20. For the seals, Zacos and Veglery, no. 186
(dated 694/5) of George, apo hypaton, (for) the Slav prisoners of the eparchia of Bithynia;
no. 187 (dated 694/5) of George, apo hypaton, of the apotheke of the Slav prisoners in
Phrygia Saiutaria; no 243 (dated 731/2). of the imperial kommerkia of Bithynia, Salutaria
and Pakatiane; and so on. See Zacos and Veglery, tables.
2" Theophanes. 348. 29; 352. 14-23.
21
MansiXI, 737; Riedinger. 886.23. SeeLilie,' "Thrakien" und Thrakesion" \ 22ff.
22
See the references assembled in Pertusi, De Thematibus, p. 149.
23
See the classic summary of this position in Ostrogorsky, Geschichte, pp. 80ff., with the
older literature up to 1960.
24
For summaries of the debate and a critique of the Ostrogorsky thesis, see Lilie, Die
byzantinische Reaktion, pp. 287ff.; Haldon, 'Some remarks', 167, and Recruitment and
Conscription, pp. 30ff.
 Military administration 215

contemporary documentation).25 Note that the traditional title of the

regional military commanders - magisth militum - was still in use in 662,
in the district of Lazica in the Caucasus.26
The word thema, then, seems to be a term introduced during the
seventh century - possibly right at the beginning, but by no means cer-
tainly - to mean an army or division. From its use in later sources, on
contemporary seals and from the list of armies in the iussio of Justinian II
of 687 it is clear that it was applied to the forces of the various magisth
militum, each of whose corps was described as a thema. At some point
during the seventh century, these forces occupied areas of Asia Minor on
what was to become a permanent basis, each covering a number of
provinces; so that by the early eighth century, the civil provinces could be
counted on occasion in 'thematic* groups, that is, according to which
army occupied them. From this time, therefore, the term thema, while
retaining its original sense of army, could refer also to a distinct geo-
graphical area, and fiscal administrators could be referred to officially as
supervising the provinces of such-and-such a thema.27

THE OCCUPATION OF ANATOLIA

The process by which these armies arrived and were established in Asia
Minor seems now generally agreed: the armies of the magister militum per
Thracias were at some point transferred to Anatolia, either under Hera-

25
Theophanes, 3 6 4 , 1 4 - 1 6 : e i s ta TOV Ov|itxiou . . . pipTj; Nicephorus, 36. 20: eis TT\V TOO
OilaxCov Ae-yonivnv x<*>P«v- On the value of these later references and the problems of
interpretation which accompany them, see the comments of Lilie, 'Die zweithundertjahr-
ige Reform', 3 0 . For the first references in contemporary documents to themata, see the
seals noted in chapter 5, note 9 8 above for the years 7 1 7 / 1 8 and later: and that of Paul,
dioiketes of the Anatolikon, dated to c. 7 0 0 (Zacos and Veglery, no. 2 2 9 0 ) .
26
See Devreesse, Hypomnesticum, 70. 1-2; it occurs also - as CTTparriAdTns ev TOIS
dvcnroXixois jtepeaiv - in the (probably late seventh-century) fictional Life of St Andrew
the Fool: PG CXI, 632A.
27
The origins and etymology of the word remain unclear: see the literature cited in Haldon,
Recruitment and Conscription, p. 31 and notes 3 5 and 36. More recently it has been sug-
gested that the term is an early seventh-century loan word, from the Chazar turkic word
tiimen, meaning a division of 1 0 , 0 0 0 soldiers, although I do not find this particularly
convincing. See J. D. Howard-Johnston, 'Thema', in Maistor. Classical, Byzantine and Ren-
aissance Studies for Robert Browning, ed. A. Moffatt (Byzantina Australiensia V, Canberra
1 9 8 4 ) , pp. 1 8 9 - 9 7 . More convincingly, see J. Koder, 'Zur Bedeutungsentwicklung des
byzantinischen Terminus Thema', JOB 4 0 ( 1 9 9 0 ) 1 5 5 - 6 5 , suggesting that the term may
already have been employed with the meaning 'designated region' before the middle of the
seventh century. It has also been argued that the themata are to be identified with a
number of districts which appear in the earliest Arab accounts of the conquest of Syria
and which clearly predate the actual conquest of these areas. These districts are referred to
as Junds (Arab. Jund, pi. Ajnad), and it has been argued that they actually represent the
military zones of a Heraclian thematic reorganisation. See I. Shahid, 'Heraclius and the
theme system: new light from the Arabic', B 6 7 (1987), 3 9 1 ^ * 0 3 ; and I. Shahid,
•Heraclius and the Theme System: Further Observations', B 6 9 (1989), 2 0 8 ^ 3 . In fact,
216 Byzantium in the seventh century

clius or shortly thereafter, to oppose the Arabs; those of the magistri

militum per Orientem and per Armeniam were withdrawn in the late 630s as
the Romans were forced to deal with the effects of the defeat at the Yarmuk
in 636. The regions already occupied by the praesental forces, the
Obsequium/Opsikion, from the later sixth century, formed the later district of
the same name. When exactly the Thracian forces were established in Asia
Minor is unclear, but it was probably at the same time as the troops of the
magister militum per Orientem withdrew into Asia Minor, since we hear of a
magister militum per Thracias commanding an army on this front and being
sent with his troops to reinforce Egypt in 638. 2 8 Thrace seems to have
come already under the occasional protection of the praesental forces,
although it clearly had troops of its own. The more pressing threat in the
East, together with the lack of resources in Thrace itself to maintain such
an army, must lie behind the decision to billet at least a large proportion of
the troops in western Asia Minor. The forces which remained in and were
maintained in Thrace were established as a thema, together with its
dependent region, only after 680, or so it would seem, as the threat from
the Bulgars became serious.29
The process of withdrawal and of the distribution of the troops in their
new garrisons and camps was in all probability completed by the
mid-640s: already one Arabic source refers to a military unit called
Armeniakon for this period, suggesting that the forces of the magister
militum per Armeniam were by this date established in their new districts;
while it has been pointed out that Arab attacks on Amorium in 644 and
646 may reflect the new headquarters of the magister militum per Orientem,
whose forces had by this time also been withdrawn.30 The praesental
forces of the Opsikion seem to have been active in north-west Anatolia at
about the same time, under an officer who may have been their komes.31
As has been shown elsewhere, the term Obsequium/Opsikion applied to this
they probably represent the regions under the command of the duces of Phoenice
Libanensis, Arabia and the three Palestines: see J.F. Haldon, 'Seventh-century continuities:
the Ajndd and the "Thematic Myth" \ in States, Resources and Armies (Studies in Late
Antiquity and Early Islam 1, IE), ed. Averil Cameron (Princeton 1995), 379-423. On all
these developments and the evolution of the debate on the origins of middle Byzantine
military structures, see idem, 'Military service, military lands, and the status of soldiers: cur-
rent problems and interpretations', DOP 47(1993) 1-67 (repr. in State, Army and Society in
Byzantium, VII).
28
Nicephorus, 24, 19. See Haldon, Byzantine Praetorians, p. 173 note 349.
29
For the Opsikion and Thrace - Haldon, Byzantine Praetorians, pp. 194f., and note 442 for
Thracian troops in the 650s; Lllie,' "Thrakien" und "Thrakesion"', 28-35.
30
See W. E. Kaegi, jr., 'Al-Baladhuri and the Armenlak theme*, B 38 (1968), 273-7. and
The first Arab expedition against Amorium1, BMGS 3 (1977), 19-22; Haldon, Recruit-
ment and Conscription, p. 33 and note 40; Lilie, 'Die zweihundertjahrige Reform', 36. The
removal of Armenia from Byzantine control in 652-5 must have meant the establishment
of the forces of the magister militum per Armeniam in their new districts by this date at the
latest. See Stratos, Byzantium in the Seventh Century, vol. II, pp. 73f. and 108fT.
*> Haldon, Byzantine Praetorians, pp. 19Iff.
 Military administration 217

army derives not from a description of thefieldarmies themselves, but from

the title of the officer who, during the reign of Heraclius, was appointed to
command them on the emperor's behalf. This officer was the commander
of the palatine corps of domestici the comes domesticorum, also called the
comes Obsequii.32 Finally, the command of the Caravisiani (KapaPtaictvoi /
r\ arpornnia TCOV Kapdpcov), constituted by the central and southern parts
of the old quaestura exercitus (i.e. the Aegean islands and Cyprus) and
probably based at Samos, must have attained its final form by about 654, if
not slightly earlier, with the Arab conquest of Cyprus in that year.33
The fate of the limitanei in the Eastern and Armenian commands is
unknown. Since, according to Procopius, Justinian had 'deprived them of
the name of soldier', their status seems to have undergone some changes -
possibly they were no longer granted the usual immunities and privileges
of military status. At any rate, they formed for the most part a settled
garrison force and participated in regular campaigns infrequently. It is
quite possible that the great majority - certainly those of the frontier
districts overrun most rapidly by the Arabs - simply stayed in their
settlements and on their lands, where their security, and their families,
were to be found. In the northernmost districts of the eastern frontier, units
of limitanei may well have survived intact at first; but the changed situ-
ation, their loss of status, and the permanent establishment of the field
troops around them, must have quickly led to their assimilation to the
ordinary civil population and the loss of any vestigial militia or police
functions.34
By the mid-650s, therefore, it seems probable that all the original
themata - armies - had taken up their new positions. It has long been clear
that these new quarters consisted in essence of groups of older civil
provinces, so much so that later subdivisions of themata tended to follow
the lines of late Roman provincial boundaries:35 and where older provin-
cial boundaries are not respected, there are usually specific reasons. For
example, the splitting of the province of Phrygia Salutaris between the
themata of the Opsikion and the Anatolikon reflected the fact that the
palatine units of scholae and domestici protectores deputati (that is to say,
32
Jfocf., pp. 174ff.
33
Stratos, Byzantium in the Seventh Century, vol. HI, pp. 39ff. and 44f. This command was
later broken up, probably under Leo III, into its constituent drungariates (of Kibyrrhaio-
ton, Samos and the Aigaion Pelagos); but there is little doubt that it represents the Aegean
section of the quaestura, which was also referred to in a novel of Tiberius Constantine (JGR
I, coll. 1, nov. 1 1 , a. 5 7 5 = Dolger, Regesten, no. 4 0 ; issued during the reign of Justin II)
as the otpxT) T<OV VTICTWV (xoti T<OV e m Xxvdux? TE x a i Mucxiaq orpaTUimx&v Tot^ixcmov).
See Antoniadis-Biblcou, Histoire maritime, pp. 6 3 - 8 9 . On Cyprus, see W. E. Kaegi, 'The
disputed Muslim negotiations with Cyprus in 6 4 9 ' , in Fourteenth Annual Byzantine Studies
Conference, Abstracts of Papers (Houston, Texas 1988), 5 - 6 with literature and sources.
34
For Procopius' account, see Historia arcana XXIV, 13; and Jones, LRE, vol. II, pp. 6 8 4 ,
649fT. and 66If.
35
See the parallel chart in Gelzer, Themenverfassung, pp. 1 2 7 - 3 0 .
218 Byzantium in the seventh century

those based outside Constantinople) were in part based in its northerly

section (the cities of Cotyaeum and Dorylaeum), which seems to have
represented also the territory occupied by the forces of the magister militum
praesentalis II and therefore the later Opsikion army. When the forces of the
magister militum per Orientem arrived, or rather, when the state allotted
new bases and quarters to the troops withdrawn from Syria, only the
southern part of the province was occupied, and it thus fell to them. 36 The
fact that provincial boundaries were generally observed, however, and
that provinces continued to be named explicity or referred to indirectly on
seals until well into the eighth century is significant, for it shows that,
when the field armies were established in Asia Minor, they were estab-
lished at a time when the older civil administration was still fully
operational, and by officials who were aware of the boundaries of their
provinces and their jurisdiction. In the first instance, therefore, the field
forces were merely based throughout the districts into which they had
been withdrawn, probably on exactly the same basis as field troops
anywhere in the empire at the time. (See maps VI and VII.)
In contrast, the diocesan boundaries of Asiana, Pontica and Oriens were
entirely ignored, suggestive of both the insignificance of this - effectively -
defunct level of prefectural administration (the more so after the closure of
the diocesan staffs of the largitiones) and also the way in which the troops
withdrew into Anatolia: both the magister militum per Armeniam and the
magister militum per Orientem simply withdrew into the provinces closest to
their last field of action and were then distributed over a number of
provinces able to support and maintain them. As we have seen, the
praesental troops of the second magister militum in praesenti already occu-
pied the regions later called the districts of the Opsikion which they had
shared with some palatine units. The allocation of provinces between the
Oriental and Thracian troops in the west of Anatolia probably reflected
both the economic wealth of the provinces over which the troops were
billeted and the total numbers in each corps, a point to which I shall return
in a moment.
36
A point elaborated by Hendy, Studies, pp. 62 3f. The pattern is not entirely preordained,
however. There is evidence for units of scholarii in the provinces of Phrygia Salutaris
(Dorylaeum, Cotyaeum), Bithynia (Nicomedea, Cion, Prousa) and Hellespontus (Cyzicus)
- see Theophanes, 236. 16sq.; Vita Theodore Syk., 156. 68; 159. 9-11, 38; for domestici
protectores in Galatia I (Ancyra, Anastasioupolis) - see Vita Theodore Syk. 45. 3; 76.3;
Procopius, Historia Arcana XXIV, 25. These areas were all to become part of the Opsikion
district. But there is also evidence for scholarii and domestici in Cappadocia (Justinian, Nov.
30, 7.2 (a. 536)); and for domestici in Galatia Salutaris (Vita Theodore Syk. 101. 5, at
Pessinus) as well as scholarii (Justinian, Edict VIII, 3 (a. 548). These areas later formed part
of the Anatolikon and Armeniakon districts. Scholarii were for a time also based in Thrace:
see Theophanes, 236. 16sq., and cf. Hoffmann, Bewegungsheer, vol. I, p. 298; Haldon,
Byzantine Praetorians, pp. 128 and 135 arid notes.
 Military administration 219

The actual date of the withdrawal and the stationing of the Armenian
and Oriental forces seems to have been after the battle of Yarmuk in 636,
in which the joint forces of both field divisions were heavily defeated.
Heraclius withdrew with part of the Oriental force into southern Anatolia,
leaving some units to garrison the cities of Palestine and north Syria that
were still in Roman hands. The forces of the magister militum per Armeniam
withdrew to their bases in Mesopotamia and Armenia. The remaining
Oriental forces were withdrawn from Egypt under treaty in 641 after the
fall of Alexandria. 37 Shortly after 636, Heraclius issued a general order to
the Roman forces in the East, to maintain themselves in their quarters and
bases and to defend their positions against the Arabs, but not to go over to
the offensive or to risk open battle. 38 This is a significant development, for
it clearly represents a first step in the process of permanently establishing
troops in Asia Minor on the one hand, but implies on the other no more
than that the troops should remain (in the districts under the magister
militum per Armeniam, at least) in their traditional quarters and garrison
towns; or (in the case of the troops of the magister militum per Orientem) in
the regions which had been (temporarily) assigned to them in southern
and eastern Asia Minor. By the same order, the troops in the cities and
fortresses still in imperial hands in Palestine and Mesopotamia were also to
adopt a purely defensive strategy.
From the later evidence of the tenth-century compilation De Thematibus
('On the themata'), compiled by the Emperor Constantine VII, it is possible
to reconstruct, approximately, the areas occupied by these forces. The
Opsikion division was already based in the provinces of Hellespontus,
Bithynia, Galatia I, part of Phrygia Salutaris, and Honorias; 39 the Anatoli-
kon forces occupied eventually the southern part of Phrygia Salutaris,
together with Phrygia Pacatiana, Galatia Salutaris, Lycaonia, part of
Isauria, Pamphylia, Pisidia and Lycia;40 those of the Thracian contingent
the provinces of Asia, Lydia, Caria and a small part of Phrygia Paca-
tiana; 41 and those of the Armeniakon the provinces of Cappadocia I and II,
37
See Baladhuri, 207, 210 and 253; Michael Syr., vol. II, 424; and cf. Lilie, Die byzantinische
Reaktion, p. 42 note 5 and pp. 300-1; and for Egypt, pp. 48-51 with literature and
sources. See also Stratos, Byzantium in the Seventh Century, vol. n, pp. 88ff.
38
See Dolger, Regesten, no. 210 with sources. The order was issued probably in 637. See
also the discussion of W. E. Kaegi, jr., 'The frontier: barrier or bridge?', 17th International
Byzantine Congress, Major Papers (New York 1986), pp. 279-303; cf. Baladhuri (as in note
37 above).
39
See Pertusi, De Thematibus, 128,132 and 135; Jones, Cities of the Eastern Roman Provinces,
app. iv, tables viii, xii, xiv, xix, xx.
40
Pertusi, De Thematibus, 114f.; Jones, Cities of the Eastern Roman Provinces, app. iv, tables xi,
xii, xiii, xv, xvi, xvii, xviii, xxx.
41
Pertusi, De Thematibus, 125; Jones, Cities of the Eastern Roman Provinces, app. iv, tables vii,
ix, x, xi: only the cities of Ierapolis, Laodicea and Colossae/Chonae from Phrygia Paca-
tiana - see De Thematibus, HI, 36-7; and Pertusi, ibid., 150.
220 Byzantium in the seventh century

Armenia I-IV, Paphlagonia and Helenopontus.42 The naval division of the

islands seems at this stage not to have had any mainland territory,
although it is difficult to state with certainty whether Caria belonged
entirely to the region settled by the Thracianus exercitus/Thrakesianoi. It
may be that the littoral of the province, or certain cities and ports along it,
such as Corycus on the Gulf of Attalia, or Kibyra (Cibyra), from which the
later naval thema of the Kibyrrhaiotai takes its name, did form part of this
command. 43
It is possible that the two Cilicias, or parts of them, were originally
intended to be included within the districts covered by the Oriental forces;
but although Byzantines and Arabs continued to fight over these
provinces, it seems clear that it was Heraclius who decided to withdraw to
the Taurus mountains and to establish a new defensive line there, which
was, in the event, to become the new frontier. Both Arabic and Syriac
sources report that Heraclius evacuated the troops and the civil population
alike from Cilicia, and indeed that he established a kleisoura, or several
kleisourai there, on analogy with those existing already along the north-
eastern frontier.44 These were garrisoned outposts designed to control
access to the mountain passes into Roman territory and had been a normal
feature of the Armenian frontier zone for many years. The establishment of
such garrisons in the Taurus at this time - 638-40 - illustrates the fact
that the Byzantines clearly recognised the danger which now threatened
the provinces of Asia Minor. 45 It seems most likely, therefore, that the
forces of the three magistri militum of Thrace, Oriens and Armenia were
already in the areas which they were to occupy according to later seventh-
century sources by about 640.

THE ARMIES IN ANATOLIA: MAINTENANCE AND PROVISIONS

The early relationship of these field armies with the civilian government
and population is clearly a more difficult problem. What can be said at the
outset, and which has already been demonstrated, is that the late Roman
provinces continued to exist. Their fiscal administration was still super-
42
Pertusi, J)e Thematibus, 118; Jones, Cities of the Eastern Roman Provinces, app. iv, tables
xxii, xxiii, xxiv, xxv, xxvi, xxvii and xxxii. Paphlagonia has until recently been taken to
belong originally to the Opsikion - see Pertusi, De Thematibus, 134 with references, 136f.
But see W. Treadgold, 'Notes on the numbers and organisation of the ninth-century
Byzantine army', GRBS 2 1 (1980), 2 6 9 - 8 8 , see 2 8 6 - 7 ; and Hendy, Studies, p. 649 with
note 414.
43
See Pertusi, De Thematibus, 149-50; Jones, Cities of the Eastern Roman Provinces, p. 517;
Antoniadis-Bibicou, Histoire maritime, p. 85; and the remarks of Hendy, Studies, p. 652.
44
Baladhuri, 2 1 0 and 2 5 3 ; Michael Syr., vol. II, 424. For the kleisoura, see Michael Syr.,
vol. II, 4 2 2 - 3 ; and the discussion of Lilie, Die byzantinische Reaktion, p. 303 note 35.
45
For the older kleisourai, see Karayannopoulos, Entstehung, p. 52 with note 1.
 Military administration 221
vised and organised from Constantinople, under the general supervision of
a dioiketes of the provinces; and, as the ninth-century evidence - together
with some contemporary material - suggests, they continued to have local
governors under the authority of the praetorian prefect. In the early years
after the arrival of these troops in Asia Minor, it is likely that they were
billeted in the traditional manner on the population of the towns or rural
areas they garrisoned, or they were maintained in garrison fortresses.46 Of
course, this will have been a considerable burden on the populations of the
areas where the troops were based and will probably have been regarded
as a short-term measure - there is no evidence to suggest at this stage that
the government did not intend to strike back as soon as it had consolidated
its position. On the other hand, the size of the areas occupied by the
different divisions does suggest, as we shall see, that the burden on the
civilian population had been taken into account and that it was spread as
equitably as was possible in the circumstances. In addition, it is quite clear
that fiscal administrative changes were also introduced in order to cope
with this new burden - in particular a shift towards full-scale maintenance
of the troops through regular levies in kind, extraordinary or exceptional
levies which eventually became regularised. I will discuss this below.
In these respects, there is no reason to assume otherwise than that the
traditional and standard rules pertaining to the military in their civilian
provinces will have applied, both in regard to legal matters and juridical
authority governing civil-military relations, and to the provisioning of the
soldiers.47
A good deal is known about supplying and provisioning troops in the
fifth and sixth centuries. Rations and fodder were issued either in kind, or
were commuted for cash, in which case the regimental commissary officers
purchased supplies from local producers or landowners. According to the
system as set out by Anastasius, a proportion of the land-tax was assessed
in kind. Units received warrants or requisition orders (delegatoriae) from the
office of the praetorian prefect, which the quartermaster of a unit could
present to a local civil governor or his representative. The latter then issued
orders to selected villages or estate owners to supply specified provisions to
the quartermaster (actuarius); and in return, the actuarius provided, a
receipt, which entitled the supplier to deduct the appropriate quantities or
values from their next tax-assessment. Where commutation was the norm,
the regimental actuaries purchased the requisite supplies according to a
schedule of fixed prices established at each indictional assessment by the
46
For the system in the sixth century and before, see Jones, LRE, vol. II, pp. 630ff. and esp.
CJ XII, 37.19f. See also Teall, 'Grain supply', 113f.
47
For the legal situation, see Jones, LRE, vol. II, pp. 63 If.; Haldon, Byzantine Praetorians,
pp. 304fT. and notes 914-24.
222 Byzantium in the seventh century

prefecture.48 Troops on active service, however, were usually issued

rations in kind. The same applied to troops in transit, such as expedition-
ary detachments. In such cases, the route of march having been trans-
mitted in advance to the governors of those provinces through which the
troops were to pass, the local civil governors would be responsible for
collecting the required supplies and depositing them at appropriate points
- in cities or on estates - along the way. The actuaries of the regiments
drew the rations in kind and issued receipts to the tax-payers, who could
then have the respective amount deducted from their next assessment.
Sometimes the amount requisitioned exceeded the normal assessment, in
which case the province or the prefecture would repay the tax-payers in
cash or credit them appropriately.49 During the fifth and sixth centuries,
an ad hoc praetorian prefect was often appointed specifically to organise the
requisitioning and delivery of supplies; and in Justinian's reign, the tempo-
rary prefect on the eastern front became permanent, a result of the
constant warfare. The practice of appointing such prefects seems to have
continued through the seventh century. 50
By the later sixth century, it is clear that garrison troops and those in
winter quarters were usually paid in cash - a regulation of the Strategikon
of Maurice, compiled probably in the 580s or 590s, stipulates that officers
must ensure that the districts in which the forces winter can provide
adequate supplies - in both foodstuffs, as well as clothing and mounts - for
the soldiers to purchase; 51 but the rule for units on active service seems still
to have been to supply provisions in kind. 52
Now it is obvious that the events of the 630s and 640s must have
dramatically affected this traditional establishment in Asia Minor, if only
with respect to its ability to cope with the sudden arrival, however well
planned and executed, of most of the field forces from the eastern front. As
has been emphasised before, the effects of the loss of the greater and richer
parts of the Eastern prefecture outside Asia Minor by the mid-640s (Egypt
was lost by 642) on the revenue of the state were considerable. Something
over half of the total revenue of the state from the Balkans and the eastern
prefecture has recently been estimated to have been lost. 53 And whether
this is entirely accurate or not - something which, given the available
figures for resource potential on the one hand, and actual exploitation on

48
For a detailed description, see Jones, LRE, vol. II, p. 672, with notes 150 and 151; Haldon,
Byzantine Praetorians, p. 113 and note 111.
49 Jones, LRE, vol. II, p. 673; see CJXU, 37.1-3, 5 - 1 1 , 14-19; 38.1-2; Justinian, Nov. 130,
1-8; cf. J. Maspero, Organisation militaire de VEgypte byzantine (Paris 1912), pp. 109ff.
50
Jones, LRE, vol. II, pp. 673-4; Kaegi, 'Two studies', 103ff. with sources.
51
Maurice, Strategikon I, 2. 16. See Haldon, Byzantine Praetorians, p. 113.
52 53
Kaegi, 'Two studies', 106f. Hendy, Studies, p. 620.
 Military administration 223
the other, is impossible to arrive at - this estimate provides a reasonable
idea of the implications of such losses.
The result of such developments must have presented the government
with almost impossible problems. Not only was the maintenance of the
army traditionally one of the most expensive items, if not the most
expensive item, in the state budget;54 the state was now faced with the
difficulty of supplying and maintaining a disproportionately large number
of troops in a much-reduced empire, both in territorial and in fiscal terms;
and, in addition, it was faced with the effects of constant and debilitating
warfare on its eastern frontiers and of economic and therefore fiscal
disruption. Fortunately, recent work on some aspects of this problem
makes it possible to obtain a clearer picture of how the state managed to
respond to what was, in effect, a major financial and resource crisis.
In the first place, the ad hoc praetorian prefects seem to have continued
to exist and to play a role up until the early ninth century. So much is
suggested by their appearance in both the Taktikon Uspenskij of 842/3
and in a ceremony of roughly the same period. Since it is also reasonably
certain that they were replaced by the thematic protonotarioi55 and since
the method of calculating the necessary provisions for troops in transit and
for those based locally, and of reducing the tax-assessment of those who
supplied the troops accordingly, is known to have been almost the same in
the ninth and tenth centuries as in the sixth century continuity of practice
in this respect is assured.56 There seems no reason to doubt that units of
soldiers on the move were provisioned in the same way in the seventh
century as they had been in the sixth and were to be in the ninth and tenth
centuries: the local military commanders sent in an assessment of require-
ments to the central bureaux; the latter then calculated the taxes in kind
accordingly; and the provisions were deposited by the relevant prefects
along the route. Tax-payers received receipts, which they could claim
against the next assessment.
The longer-term maintenance of large numbers of soldiers, however,
presented a more intractable problem, since the resources which had
traditionally been available to pay the troops their cash rhogai were simply
no longer available in sufficient quantity.
Up to this time, soldiers had been supported from two main sources: the
54
Jones, LRE, vol.1, p p . 4 5 8 and 4 6 3 ; Hendy, Studies, pp. 1 5 8 - 9 and 172; note the
anonymous sixth-century treatise Peri Stratigias (ed. and trans. G. Dennis, Three Byzantine
Military Treatises (CFHB XXV, ser. Wash., Washington D.C. 1985), 1 - 1 3 5 , see 2. 1 8 - 2 1 :
The financial system was set up to take care of matters of public importance that arise on
occasion, such as the building of ships and walls. But it is principally concerned with
paying the soldiers. Each year most of the public revenues are spent for this purpose.
55
See above, pp. 2 0 1 - ^ .
56
For a detailed account, see Haldon, Byzantine Praetorians, pp. 314ff.
224 Byzantium in the seventh century
issue of annonae and capitus, in kind or commuted into a cash equivalent
(according to afixedtariff); and the cash donatives issued on the occasion
of an imperial accession and every fifth year thereafter, made up of five
solidi and a pound in silver in the first case and five solidi only on the
quinquennial occasions. In addition, new recruits received weapons and
clothing through the commissariat, purchased with a cash allowance
issued for the purpose, arms coming from the state weapons and arms
factories, clothing from state weaving establishments or through levies in
kind raised as tax.57 During the reign of Maurice, attempts were made to
reform the system, in the first instance to divide the regular commuted
annonae into three parts, issuing clothing and weapons directly; and in the
second attempting to reduce the sum of the annonae by twenty-five per
cent.58 Both seem to have misfired, but both represent the efforts of the
government to reduce its cash expenditure and move towards a direct-
issue system such as had operated before the laterfifthcentury, at least in
respect of clothing and equipment.59 This shortage of cash reserves for
military expenditure was, then, already a feature of the last years of the
sixth century. During Heraclius' reign, it clearly became critical. Not only
was the emperor forced to borrow and turn into coin the gold plate of the
Church;60 it is equally clear that the troops, while still paid in cash, had to
be content with copper rather than gold - in itself a procedure that was not
new, but which illustrates the dramatic lack of gold revenues in the
conditions prevailing in the period after 610. So much is clear from the fact
that Heraclius had a statue in Constantinople melted down and turned into
coin to pay troops in the Pontus; and the fact that the temporary mints set
up to produce coin for the army during the Persian wars up to 627
produced predominantly copper.61 While Maurice probably failed to effect
the twenty-five per cent reduction in cash emoluments which he attempted
57
See Haldon, Byzantine Praetorians, pp. 1 1 3 - 1 5 with notes 1 1 3 - 2 3 , and Recruitment and
Conscription, p. 6 9 and note 123; Hendy, Studies, pp. 1 7 5 - 8 and 646ff. Procopius,
Historia Arcana, XXIV, 2 7 - 9 , claims that the quinquennial donative was allowed to lapse
by Justinian, but it has been argued that this is unlikely. Jones, LRE, vol. n, p. 6 7 0 and
vol. I, pp. 2 8 4 - 5 , argued that it may have been incorporated at the rate of 1 solidus per
annum into the commuted yearly annona.The accessional donatives survived until the
seventh century, certainly, and there is no reason for suggesting that the quinquennial
donatives for the field armies did not do so either. See Hendy, Studies, pp. 646ff.
58
See Haldon, Byzantine Praetorians, pp. 113f.; Recruitment and Conscription, p. 6 9 note 123.
59
For the original system, Jones, LRE, vol. II, pp. 624f., and see 670f.
60
See Theophanes, 3 0 2 - 3 .
61
For the melting of the statue, Kaegi, 'Two studies' 90ff. (but see Haldon, Byzantine
Praetorians, pp. 6 2 7 - 8 ) ; Hendy, Studies, pp. 4 1 5 - 1 7 (mints at Alexandretta, Constantia
(Cyprus) and Jerusalem during the revolt of Heraclius; and at Seleucia (Isaura) and
Constantia during the Persian war). Soldiers in the sixth century, and from the ninth
century, were usually paid in gold (although not exclusively) - this was a fundamental
element in the redistributive fiscal operations of the state. See Haldon, 'Some considerati-
ons', 80ff.
 Military administration 225

to introduce, Heraclius, faced by a much more drastic situation, actually

did cut military and civil rhogai by fifty per cent, probably in 615; and,
furthermore, calculated the payment of salaries largely in the newly
introduced silver hexagram. 62 Finally, in 641 a donative probably issued
to celebrate the accession of the young Constans II (or also the coronation
of Heracleonas) was issued, consisting - according to a much later source -
of only three solidi, that is, only one third of the traditional donative on
such occasions. 63
All of this points to a dramatic reduction in the gold reserves of the state
and the available coined gold in circulation. If, in 641, the state was able to
put together only one-third of the donative usually issued on the occasion
of an imperial accession, while at the same time trying to finance roughly
the same number of field troops as it had possessed before the loss of most
of its Eastern territories, the finances must already have been near
breaking-point. It is from this time on, therefore - the early 640s - that
major shifts in the methods of paying and financing the armies are to be
sought.
The most pressing problem, in the first instance, may have been to
reduce the overall burden on the fisc. One of the solutions to this problem
seems to have been to pay the various provincial forces on a rotational
basis, rather than all at once, at least in so far as the payment of donatives
was concerned. The system of paying the troops their rhogai on a four-
yearly rotational cycle, described by Constantine VII in the tenth century
as 'the old system', 64 has been plausibly shown to be tied in with the older
quinquennial donatives (collected quadrennially, in the last year of each
quinquennium). Since the rota presented by Constantine is clearly a
developed form, it is highly likely that the original cycle concerned the four
original themata in Asia Minor: the Anatolikon, Armeniakon, Thrakesion and
Opsikion; and it is also highly likely that it should be taken back to the
seventh century, perhaps as early as the 640s and 650s. 65 This rotational
system must have applied to donatives only, however, enabling the state to
spread the burden of cash payments over a longer period and to extract the
necessary revenue year by year to cover the cost; for while the next stage in
reducing costs might well have been to cut down or cut out altogether the
commuted annonae and capitus, the payment of reduced cash annonae on
such a rotational basis cannot have been a practical solution, since such a
cycle would run the risk on a regular basis of depriving the troops of their
62
Chron. Pasch. 706. See Hendy, Studies, pp. 4 9 4 - 5 .
63
See Cedrenus I, 753 and Stratos, Byzantium in the Seventh Century, vol. II, pp. 187 and
217-18.
64 De Cer. 493-4.
65
The argument has been fully developed in Hendy, Studies, pp. 6 4 7 - 5 1 .
226 Byzantium in the seventh century

livelihood before it was completed. Cutting down the numbers of troops

may also have been considered, but there were obvious limits - in terms of
the effectiveness of defence - to this alternative.
On the other hand, a reversion to the payment of the field forces largely
or entirely in kind would have gone part of the way towards solving the
problem. For, however the troops were quartered or housed, they still
needed to be fed, clothed and supplied with weapons, horses and so on. The
permanent establishment of an ad hoc praetorian prefect for each thema (or
army), who supervised the calculation, extraction and delivery of the
military annonae and capitus in kind, seems to represent part of the solution
finally adopted by the state. As we have seen, such a permanent post had
already existed for the Oriental forces in the sixth century. Similar posts
may have been created for the other field armies, especially the Armenian
division and the forces in Thrace, where providing adequate supplies had
for long been a problem. That there was such an official for each thema is
reasonably clear from the ninth-century material referred to already. And
it seems a reasonable hypothesis that the withdrawal of the troops into
Anatolia marks the point at which this system is most likely to have come
into existence as a permanent arrangement. 66 Possible corroboration for
this development comes from a reference in the 660s to the annonae of the
scholae, parade-troops who, as far as is known, would normally have
always been paid in gold, in the sixth century at least; and to a manceps, or
baker, in charge of the bread-ration of these palatine soldiers.67
Corroboration also comes from the numismatic evidence in Asia Minor
(which, as Hendy has shown, contrasts for specific institutional reasons
with the Balkans) where finds of copper coins die out almost entirely
during the last years of Constans II. Up to 658, it has been pointed out that
the copper coinage is represented by small but regular quantities on a
number of excavated sites in Asia Minor: Aphrodisias, Anemurium,
Ancyra, Sardis, Priene, Ephesus and Pergamum. From 658 to 668,
however, the copper coins peter out and effectively disappear from Anato-
lian sites until the ninth century. 68 A number of the groups of copper have
been interpreted as coming from a military context, which would explain
the physical concentration - on sites which might be associated with the
presence of soldiers as a garrison, or in the construction of defences. The
disappearance of copper coins from Anatolian sites has, however, usually
been taken to reflect the removal of the military units stationed there and
the effective abandonment of the sites in question. In fact, neither of these is
necessarily a corollary, as Hendy has also demonstrated. Instead, it has
been suggested that the disappearance of copper - which during the
66 67
See above, pp. 201^1,223. See Haldon, Byzantine Praetorians, p. 125 and note 164.
68
The phenomenon is fully documented in Hendy, Studies, pp. 640ff.
 Military administration 227

seventh and early eighth centuries does seem to have been employed to
pay the troops on a number of occasions - signifies more probably the
abandonment of a system of cash payments to soldiers.69 If this were the
case, then a reasonably precise date can be ascribed to the implementation
of these changes, that is to say, the last years of the reign of Constans II.
This is, of course, some twenty years after the arrival of the field armies in
Asia Minor. But the discrepancy need be more apparent than real -
payments in copper, possibly at a reduced rate (in comparison with the
sixth century), as well as occasional payments in silver and gold, were still
made, 70 while the troops might already have been supported by supplies
and equipment issued for the most part in kind.
That the extent to which the land could support the armies which
occupied it was taken into account from the beginning is clear from the
actual size of the districts which were allotted to each thema or army. As
maps VI and VII will show, the forces occupying the most fertile zone
received by far the smallest district in which to billet its troops and from
which to draw its resources. In contrast, the Armenian forces, which fell
back into eastern Anatolia, occupied the least fertile area made up of some
of the most arid sections of the central plateau, and in consequence -
although there is every reason to suppose that the magister militum per
Armeniam commanded forces significantly less numerous than those of the
magistri militum per Thracias or per Orientem - seem to have been allotted a
much wider resource-area from which to support themselves. The district
of the Opsikion is not part of this pattern, in so far as it existed in effect - as
the area in which the praesental and some palatine forces had always been
stationed - before the 640s and the withdrawal from the East. The
Anatolikon, on the other hand, occupied the area between the Thrakesion,
Opsikion and Armeniakon armies, and once again its extent, contrasted with
both the Thrakesion and the Armeniakon, suggests the nature of the land
from which it was intended it should support itself. While this line of
reasoning is hardly conclusive by itself, it is very suggestive, and it is a
further indication that the withdrawal of 637 and after was carefully
organised and co-ordinated, and planned with a view to the problems of
supplying and maintaining the armies in the future. 71
There seems, therefore, to be good circumstantial evidence that the
withdrawal into Asia Minor was, if not planned greatly in advance,
nevertheless planned carefully; and that the provinces into which the
69
Hendy, ibid., with sources and older literature.
70
The cash issued to the troops in 641 was, according to Cedrenus, in gold; that put aside by
Heraclius Constantine and entrusted to the sacellarius Philagrius was similarly (appa-
rently) in the form of gold coins. See note 63 above: and Nicephorus, 28. llsq.; John of
Nikiu, 192. See Haldon, Byzantine Praetorians, pp. 191f. with note 427.
71
See excursus to chapter 6, pp. 2 5 1 - 3 below.
 300mOes

Map VI Justinianic prefectures and provinces c. A.D. 565

 Military administration 229
armies withdrew were intended to be able to support them on the basis of
requisitions of supplies and provisions in kind (the close relationship
between size of army and size of districts occupied would be otherwise
pointless). From the last years of Constans II the numismatic evidence
suggests that payments in cash, even though in copper, were more or less
entirely cut out, at least as a regular form of salary equivalent in (nominal)
value to the annonae and capitus of the previous years. In other words, the
state had come to rely entirely upon a system of supporting the armies
through the levying of all requirements in kind. This implies that the
greater part of the tax-burden also took the form of payment in kind.
Similarly, the origins of the quadrennial cycle of payments for donatives
also seem to lie in this period, although exactly when it might have been
introduced is impossible to say.
There is one further piece of evidence to support this line of reasoning. As
we have seen, up to the end of the sixth century at least, the armies were
supported both by cash, with which goods, provisions and such like could
be purchased, and by provisions in kind, for expeditionary or moving
forces. Such provisions were part of the regular tax-assessment, and if

KEY

(a) Prefecture of Italy 27 Palaestinal 56 Phrygia Pacatiana

1 AlpesCottiae 28 Arabia 57 Caria (in quaestura exercitus)
2 Aemilia 29 Palaestinall 58 Lycia
3 Venetia (with Histria) 30 Phoenice 59 Lydia
4 Liguria 31 Theodorias 60 Bithynia
5 Flaminia 32 Cyprus (in quaestura 61 Hellespontus
6 Tuscia et Umbria exercitus) 62 Asia
7 Picenum 33 Phoenice Iibanensis 63 Insulae {Aegean Isles - in
8 Samnium 34 Syrian quaestura exercitus)
9 Campania 35 Syria I 64 Creta
10 Apulia et Calabria 36 Euphratensis 65 Europa
11 Lucania et Bruttium 37 Osrhoene 66 Bosporus
12 Sicilia (under quaestor sacri 38 Mesopotamia 67 Haemimontus
palatii) 39 Armenia III 68 Rhodope
40 Armenia IV 69 Scythia (in quaestura
(b) Prefecture of Africa 41 Armenia I exercitus)
13 Corsica 42 Armenia II 70 Moesia II (in quaestura
14 Sardinia 43 Helenopontus exercitus)
15 Numidia 44 Cappadocial 71 Thracia
16 Zeugitania 45 Cilicia II 72 Macedonia I
17 Byzacena 46 Cilicia I 73 Thessalia
18 Tripolitania 47 CappadociaH 74 Achaea
48 Lycaonia 75 Epirus vetus
(c) Prefecture of Oriens 49 Isauria 76 Epirus nova
19 Libya Pentapolis 50 Pamphylia 77 Macedonian
20 Libya Inferior 51 Pisidia 78 Dacia Mediterranea
21 Arcadia 52 Galatia Salutaris 79 Dardania
22 Thebais Inferior 53 Galatia I 80 Praevalitana
23 Augustamnica II 54 Paphlagonia 81 Dacia ripensis
24 Aegyptus I and II 55 Phrygia Salutaris 82 Moesia I
25 Augustamnica I 83 Dalmatia
26 Palaestinam
 A/y Thrakesion Armeniakon

Map VII The Anatolian themata and the late Roman provinces c. A.D. 660
 Military administration 231

extra supplies were needed, what was taken was set against the tax-assess-
ment for the following years so that, in theory at least, the tax-payers were
not unfairly exploited. The compulsory purchase of provisions - coemptio
or synone - in addition to, or instead of this procedure, was, except in very
special cases, strictly forbidden.72 Only one area was exempted from this
rule, namely Thrace where, as the relevant law of Anastasius states,
because the revenue from the land-tax in gold was insufficient, due to
disruption, population decline, and the devastations of the barbarians, the
coemptio was the only reliable way of providing for the troops there. 73
By the early ninth century, if not indeed much earlier, however, the term
synone referred quite simply to the basic land-tax, raised by the state
together with the hearth-tax or kapnikon. The synone, by the eleventh
century at least, could be commuted into cash. 74 But in the early ninth
century, without doubt, it was a regular yearly assessment which could be
collected in kind. 75
The question inevitably arises, by what process did a sixth-century tech-
nical term for a compulsory purchase of military provisions, applied only
in exceptional circumstances, come to mean the standard land-tax, also
collected in kind? The answer can only lie in the context outlined above, in
which the state, faced by a drastic shortage of cash, was forced to return to
a system of provisioning all of its troops in kind. The basic land-tax hitherto
raised both in cash and in kind was now raised more regularly and on a
widespread basis in kind alone; and the term henceforth applied to it was
that already available describing precisely such a process, the synone,76
72
Jones, LRE, vol. I, pp. 2 3 5 , 4 6 0 and note 120.
73
C/x, 27.1 (a. 491) x, 27.2/5-10 (a. 491-505).
74
See Ostrogorsky, 'Steuergemeinde', 49ff. and 60f., 'Steuersystem', 232, and Geschichte,
p. 55 note 2. Contrast Dolger, Beitrdge, pp. 5Iff. and 78, who sees the synone still as a
simple extra levy in kind. That synone = land-tax is clear from a number of sources, for
example, De Cer. 6 9 5 . 6-14, where certain households are to be exempt from all regular
fiscal impositions, especially the kapnikon and the synone. This unusual exemption,
granted for a specific purpose, was clearly intended to free such households from all the
state taxes, including those not usually listed in such grants.
75
See Haldon, Byzantine Praetorians, pp. 314f. and notes 9 5 0 and 9 5 1 (letters of the
metropolitan Ignatius of Nicaea). Theophanes (443. 1 8 - 2 2 ) and Nicephorus (76, 5-14)
refer to the raising of taxes in gold, which suggests that the state was by this time in a
more favourable situation and could choose according to its needs whether to collect in
gold or in kind. But whether this refers to the land-tax, the hearth-tax or both remains
unclear.
76
The system for supplying troops in transit in the fifth and sixth centuries detailed above, in
which military supplies are taken from the tax assessment of each province in kind, is
almost exactly the same in the ninth and tenth centuries, in which the thematic protonota-
rios supplies the troops from the synone in kind. The similarities are so obvious that it is
clear once more that synone must be the Byzantine equivalent of the late Roman basic
land-tax. By the later seventh century, this form of assessment seems to have been well
established, for the Farmers' Law refers to the regular land-tax as the 'extraordinary tax of
the public fisc' (T& e^Tpaop8iva TOV 8TI (XOCTLOD \6yov). Extraordinaria had meant originally
232 Byzantium in the seventh century

The only context in which such a dramatic shift in the form of revenue
collection is likely to have taken place - based as it is implicitly upon the
assumption of a lack of, or shortage of, cash - is that of the seventh
century; and the chief reason was, as we have seen, to support the armies
and other state bureaux in the provinces. It is quite possible that the
original arrangement actually involved both the usual tax-assessment, as
well as a generalised supplementary coemptio for the newly-arrived armies.
The gold or copper thus expended by the state in the process of the
compulsory purchases would be recouped through the usual tax-payments
(the Anastasian regulations on coemptio had envisaged the tax-payer being
recompensed in gold coin).77 But a proper coemptio as widespread as this
would have demanded reserves of gold and coin which, as we have seen,
the state does not seem to have possessed. As the state found itself unable
to recompense tax-payers for the produce it needed, yet continued to
demand the necessary provisions, the term will have become generalised
and applied to the ordinary land-tax assessment also.

THE ROLE OF THE 'APOTHEKE'

The final cessation of cash payments and allowances (for weapons and for
clothing, for example) will have involved changes in the method of arming
and equipping the troops, of course. In the sixth century, and probably up
to the reign of Heraclius, soldiers received their initial uniform and issue of
arms directly along with mounts for the cavalry troopers. Weapons and
uniforms were delivered through the prefecture to the units, whose actua-
ries were responsible for submitting claims for what was needed, direcly
from the state manufactories. Serving soldiers received a portion of their
cash pay to cover the costs of purchasing new equipment or remounts; and
it was the responsibility of the officers in each unit to check the soldiers'
equipment and to order new material.78 Part of this arrangement, of
course, involved the issue of cash to the soldiers, which could be used to
purchase material from the state, but which might also be spent unwisely
on non-military items. Hence the unpopularity of Maurice's attempted
reforms in the 590s. With the cessation of cash allowances altogether this

exceptional levies in addition to the regular taxes. Here in its Greek variant, extraordina, it
clearly means the regular tax itself, presumably undergoing a similar terminological
transformation to that of synone, which was itself originally classed among the extraordi-
naria. Note that, in spite of the fact that a large number of soldiers in the forces of the
exarchate of Ravenna did possess land, the local military authorities still provided their
supplies and rations, as a compulsory purchase (coemptio) of grain in 686 demonstrates:
Liber Pontificalis I, 366; see Brown, Gentlemen and Officers, pp. 87-8.
77
See note 73 above, and Haldon, 'Synone', for the detailed development of this argument.
78
See Haldon, Byzantine Praetorians, pp. 113-15 and Recruitment and Conscription, p. 69.
 Military administration 233

clearly became impossible. And here another element in the pattern seems
to fit into place.
A number of historians have noted that it is during this period that dated
seals of genikoi kommerkiarioi suddenly become much more common, seals
on which an association between an imperial apotheke and a region or
regions of the empire is apparent, and from which it is clear that one official
often had jurisdiction over a number of widely dispersed areas. On the
whole, this phenomenon has been explained in terms of imperial control
over trading in luxury or other goods, since it is clear that the apothekai
were in origin state depots in which private merchants might also have an
interest. 79 More recently, it has been shown that these apothekai were very
probably storehouses and also emporia for surplus produce from state fac-
tories, chiefly luxury goods such as silks, gold- and silverware, dyed cloths
and so on. 80 It has further been argued that the division within the prefec-
ture of the East between its general and special banks reflects this activity;
for while the actual production of the items in question came under the
supervisory authority of either the magister offlciorum (dyeing and weaving)
or the comes sacrarum largitionum (coin and plate), the raw materials were
intially provided through the prefecture and were returned for sale or redis-
tribution to a branch of the prefecture. The special bank seems to have ful-
filled this function, and the later confusion in its name, between eidikon and
idikon - 'things' and 'special' - may reflect this. The production of such
goods certainly seems to have been controlled by the later eidikon, and it
would seem that the newly independent special bank of the prefecture
gained control of these aspects of state manufacturing in the 630s or 640s.
The fact that officials in charge of the silk-producing establishments, for
example, could also be genikoi kommerkiarioi reinforces this impression.81
But the sudden increased importance of genikoi kommerkiarioi holding
high rank - the first clear example is dated to the years 654-9 8 2 - can have
79
See, e.g., Antoniadis-Bibicou, Douanes, p p . 164ff.; R. S. Lopez, T h e role of trade in the
economic re-adjustment of Byzantium in the seventh century', DOP 13 (1959), 6 7 - 8 5 ,
see 73ff.; Lilie, Die byzantinische Reaktion, p. 2 8 2 ; a n d Zacos and Veglery, vol. I, pp. 135f.
For the apothekai as imperial depots, see Zacos and Veglery, no. 130bis (Tyre, late sixth
century); Antoniadis-Bibicou, Douanes, p. 1 5 9 ; Hendy, Studies, p. 2 4 6 (Alexandria).
80
Hendy, Studies, p. 6 2 8 . T h a t the fabricae were involved is, however, unlikely, pace Hendy,
ibid. The production, transport a n d possession of w e a p o n s w a s strictly monopolised by the
state a n d its personnel, i.e. the soldiers. Surplus w e a p o n s were stored in the imperial a n d
municipal arsenals, a n d it is improbable t h a t they were involved in open commercial
exchanges. See Justinian, Nov. 8 5 (a. 539).
81
For a detailed treatment, see Hendy, Studies, pp. 6 2 8 - 3 0 ; and see above, pp. 186ff., on the
results of the fiscal reorganisation of Heraclius' reign. For seals of officials w h o were in
alternate years both genikos kommerkiarios and archontes tou vlattiou (silk-weaving
establishment), see Zacos a n d Veglery, tables 6/2, 8, 11 etc.
82
Zacos a n d Veglery, n o . 1 3 6 , seal of Theodore, hypatos and genikos kommerkiarios of the
apotheke of Galatia.
234 Byzantium in the seventh century

had little to do with the sale of luxury goods at a time when the empire was
in such political and fiscal straits. In fact, the kommerkiarioi and the
apothekai which they administered have been plausibly shown by Hendy to
be connected with some aspects of the supplying of the imperial forces. On
a number of occasions, dated seals of kommerkiarioi and apothekai for
specific regions can be related to specific military undertakings connected
with those regions. 83 In view of this connection, and the original role of the
apothekai as stores for the produce of imperial factories, it has been argued
that the main function of the apotheke was to provide the troops with the
necessary equipment for their expedition.84 The apotheke is thus seen as a
derivative institution or system, rather than an actual storehouse -
although these were certainly involved. The apotheke of a province or
group of provinces thus involved a process, the transport from the point of
manufacture of certain goods, and their distribution to the soldiers of a
given region for a specific purpose or campaign, a process supervised and
directed by a general kommerkiarios. This connection is borne out by
numerous other examples - campaigns in 687/8 in eastern Asia Minor are
accompanied by the appearance of seals of the apotheke in Cilicia, Cappado-
cia and Armenia; the campaign in Thrace in 689/90 is accompanied by
the appearance of the apotheke in Constantinople and Helenopontus, and in
the Cyclades and Crete; the naval expedition to recover Carthage in 696/7
is similarly accompanied by the appearance of the apotheke in the Cyclades
and in Sicily. Examples can be multiplied, and while not every such seal
can be so closely tied in with a specific example of an expedition in the
literary sources, the connection is too strong to be a coincidence.85

83
The clearest example is that for Justinian ITs campaign against the Arabs in 692/3 and
694/5. Four seals, dated to the 8th indiction, that is, to 694/5, exist for a certain George,
apo hypaton, all connected with the Slav prisoners settled in Asia Minor by Justinian in
688/9 and, respectively, with the apotheke of the provinces of Caria, Asia and Lycia, of
Bithynia, of Phrygia Salutaris, and of the two Cappadocias. Theophanes reports Justi-
nian's forced transfer of the Slavs to Asia Minor, specifically to the Opsikion district (i.e.
Bithynia and Phrygia Salutaris) in 688/9, and two unsuccessful campaigns in the region
of Sebastopolis, i.e. in Armenia, in 692/3 and 694/5. See Theophanes, 364. 11-366, 23.
Whether or not Theophanes' dates are entirely correct, it is clear that the transfer of Slavs
to Asia Minor, the establishment of a corps of Slav soldiers by Justinian from among the
captive populations, the campaigns of 6 9 2 - 5 , and the apotheke of the regions in which the
Slavs were settled, are closely connected. For the seals see Zacos and Veglery, nos. 2764,
186, 187 and 188, and table 33 (for Asia, Caria and Lycia, Rhodes and the Cherson-
nesos); and see table 18/2: the seal of George, apo hypaton and genikos kommerkiarios of the
(apotheke of the) Armeniakoi, read by Mordtmann and followed by others (most recently
Hendy, Studies, p. 631 note 340) for 694/5, is in fact for one of the regions of Armenia,
not the district of the Armeniakoi. But it is connected with the first four seals mentioned
above. See W. Seibt, review of Zacos and Veglery, in BS 36 (1975), 209.
84
Hendy, Studies, pp. 6 3 1 - 3 .
85
For a detailed account, with both sigillographic and literary evidence, see Hendy, Studies,
pp. 6 5 4 - 6 0 and 6 6 7 - 9 .
 Military administration 235

This explanation, which represents a major advance in our understand-

ing of the function of the kommerkiarioi and the apotheke in the seventh
century, has been challenged, however. 86 Several objections have been
raised: first, no text refers explicitly to a connection between the supply of
weapons and the kommerkiarioi or apotheke; second, the chronology of the
events of 688 to 694/5, which constitutes a central element in the
demonstration of the validity of the argument outlined above, has been
forced - in fact, the seals of the kommerkiarioi and of the Slav prisoners of
the eighth indiction, i.e. 694/5, post-date the campaigns for which,
according to Hendy, they were mobilised; third, the kommerkiarioi and the
apothekai were connected specifically with silk, its production and sale - the
emperor's effigy on these seals was, like that on imperial coins, a validation
of the purity of the product thus marked. According to this argument, the
Slav prisoners and the seals pertaining to them can be more readily
explained in terms of a mass distribution and sale of the rebellious Slavs in
694/5 as slaves in various provinces of the empire, a contract farmed out
to the kommerkiarios George, who used the apotheke system already in
existence to fulfil it. The argument is underpinned with the assumption
that the demand for silk in the seventh century was considerable enough to
warrant a major investment and interest in its production; that this
investment took place in the seventh century - from about 629 in Cyprus,
in the period 641 and after in North Africa, and from about 654 in
Anatolia - and that it was during this time that the silk industry 'grew and
became one of the pillars of the empire's state economy'; and also that this
growth took place chiefly in Asia Minor until the 730s and after, when it
was transferred to Thrace and the south Balkan area.
Now there are a number of serious objections to this theory as it stands.
There is no reason to doubt that the late sixth- and early seventh-century
kommerkiarios may have been a contract-farmer who held a monopoly
over the purchase and resale of silk within the empire; but there is no
reason to doubt either that - as Oikonomides in fact admits 87 - these
official contractors also dealt in goods other than silk alone and had charge
of the apothekai in which they were deposited, a point already made by
Hendy. 88 There are no grounds for assuming that this did not continue to
be the case throughout the seventh century and the eighth century. The
fact that seals bearing the emperor's effigy, and used to validate quality
(and origin - in imperial workshops or imperial-controlled or supervised
centres of production), are employed, is no argument that the goods in
question involved silk alone: the products of the imperial arms factories
86
N. Oikonomides, 4Silk trade and production in Byzantium from the sixth to the ninth
century: the seals of kommerkiarioi', DO? 40 (1986), 33-53.
87 88
Ibid., 34-5 and 38-9. Hendy, Studies, pp. 627ff.
236 Byzantium in the seventh century

had also been a state monopoly (like silk in the later sixth century), and
such seals might equally serve to confirm the origin and legality of
consignments of weapons produced in the state workshops or elsewhere
under state licence. If Oikonomides is correct as far as the Slav prisoners
are concerned - that is, that the seals in question relate to a mass sale of
Slavs through the kommerkiarios - this single example is alone sufficient to
refute the notion that kommerkiarioi dealt only in silk, the more so since on
one of the seals the apotheke is specifically mentioned: if the kommerkiarios
in question were bidding for the contract to sell slaves in his private
capacity then it seems unlikely that he would put the apotheke on his seal.
In fact, it seems quite logical to assume that the emperor's effigy on the
seals is simply a validation of the formal and imperially sanctioned nature
of the transaction (whatever its exact nature), and that it need have
nothing at all to do with silk - a point also tacitly conceded, although
Oikonomides wishes to argue that the example of the sale of slaves was
probably an exception.
There are a number of other points to be considered, however. Even if
Hendy is wrong as regards the sale of the Slav prisoners, the general
hypothesis remains strong. The coincidence between the dated seals of
kommerkiarioi and apothekaU and known military expeditions, is really too
marked to be written off as mere chance. In the second place, the argument
relating kommerkiarioi to silk and to the importance of silk for the imperial
economy in the seventh and eighth centuries depends upon a number of
dubious assumptions about the nature of sericulture. Crucial to Oikono-
mides' argument is the assumption that mulberry trees - which require a
moderate climate - could flourish throughout Anatolia (except in Galatia I,
for which no seal survives), in provinces as diverse as Asia or Caria, on the
one hand, which have a relatively high rainfall and standard Mediter-
ranean climatic conditions, and Lycaonia, the Cappadocias and the
provinces of Armenia on the other, which have extreme seasonal variation
between winter and summer, and very much lower rainfall. The mulberry
tree (Morus alba) needs deep, well-drained and fairly rich loams to prosper,
with adequate supplies of moisture. Eastern and central Anatolia can offer
these conditions, when at all, on an extremely limited basis. On these
grounds alone, therefore, the long-term production of silk in eastern Asia
Minor, supposedly represented by the seals of kommerkiarioi and apothekai
for these provinces, is very doubtful. It is admitted, however, that the
'silk-producing areas' were forced to move westwards as a result of the
effects of Arab raids and attacks - but only from the 730s and later. 89

89
Oikonomides, 'silk trade', 44. In modern Turkey the very small silk industry is limited to
the region around Bursa, in the north-west of Anatolia, and Antakya (Antioch on the
 Military administration 237

But Asia Minor, especially the central and eastern regions, was already
subject to regular and devastating enemy action from the 650s if not a
decade earlier. The evidence surveyed already (see chapter 3 above) is
overwhelming. The stable political conditions for such an investment
simply did not exist at this time. Even if one were to concede that isolated
groves of mulberry trees could survive in certain secluded valleys, and that
silk production could continue uninterrupted, this could hardly have
provided the major source of silk - and imperial revenue - at this time. And
indeed, one might also ask, where was the market for such large-scale
production as is envisaged? Given the restrictive and redistributive nature
of the state economy at this period, such sales can have brought in very
little in net terms to the imperial fisc. It could be pointed out, of course, that
silk - like gold - formed a major element in Byzantine diplomacy: used to
buy off foreign powers, as gifts or bribes, as a form of payment and so on.
Silk may well have been used in this way, and indeed in the ninth and
tenth centuries military officers and others received silk vestments as a
form of remuneration. But if silk were really on a par with gold, and
indeed, almost replacing gold, why is it never referred to in this sense for
this period in the sources? Gold and horses, and sometimes slaves, are
mentioned as the usual forms of such gifts or tribute; precious silks, when
they do occur, clearly occupy a relatively minor role. And the contention
that the connection between the department of the blattion (the imperial
silk-workshop) and some of the kommerkiarioi is a proof of this notion, is
hardly a convincing argument - the fact that kommerkiarioi were (also)
responsible for silk production is not to be doubted.
In the third place, Oikonomides has argued that the westward move-
ment of the 'silk-producing areas' in the eighth century illustrates a flight
away from the war-zone - thus implicitly countering the idea that the
kommerkiarioi and the apothekai were connected with supplying the mili-
tary. Two observations are relevant here. First, it is nowhere argued that
the kommerkiarioi had to do only with the supplying of the troops with
equipment - merely that this became one of their functions. Second, this
function did not last much longer than the crisis which called it into being;
and indeed, the movement westwards (that is to say, the appearance of
fewer seals for kommerkiarioi and apothekai for the Anatolian provinces,
and a corresponding increase in the number of such seals for the Balkan
districts) actually conforms to a real shift in the military strategy of the
empire, as the Balkans became a more significant - and more contested -
element in its political environment. Again, the close connection between
known military undertakings and the seals of kommerkiarioi and apothekai
Orontes) in the far south, both areas of Mediterranean climatic type. See J. C. Dewdney,
Turkey (London 1971), pp. 135-6 and 182.
238 Byzantium in the seventh century

is compelling, and it is a relationship not challenged by the arguments put

forward by Oikonomides.90
While there is no reason to doubt, therefore, that the kommerkiarioi and
the kommerkia or apothekai were also connected with the sale and pro-
duction of silk, there is every reason to reject the hypothesis that they were
exclusively to do with this activity. The historical context, quite apart from
the nature and demands of seventh- and eighth-century state fiscality,
make it inherently unlikely. The evidence in favour depends in certain
crucial ways upon arguments which can be shown to be false; while the
evidence against it is enough to flaw it in fundamental ways. In respect of
the sale of Slav prisoners in 694/5, Oikonomides may well be right. But
this neither supports nor detracts from his argumentation in any crucial
way.

To return to the main argument once more. What seems to have hap-
pened, therefore, is that the state, having cut down to a bare minimum (the
donative) the cash paid out to the armies, had to find an alternative way of
supplying and equipping its soldiers. Supplies could be raised in kind, as we
have seen. Hitherto, the troops had received initial grants of equipment
free, and cash allowances thereafter for remounts, replacement weapons
and so on. With the cessation of cash annonae and grants for such
equipment, and the return to supplying the armies in kind, a system had to
be found whereby weapons and equipment could also be supplied and
distributed. The extension of the role of the apotheke, which represented the
central and provincial storehouses of the special bank of the prefecture -
now the eidikon/idikon logothesion - offered an obvious solution.
But Hendy has also argued that the equipment involved - including
weapons - was sold to the troops, partly on the grounds that the kommer-
kiarioi have generally been assumed to be tax-collectors, and that the
pattern demonstrated by the seals - appointment by the year, dated by
indiction, often alternating with similar appointments for different regions
or for different spheres of imperial manufacturing - suggests that the
kommerkiarioi were tax-farmers. Further, it has been taken for granted that
the material to be distributed to the soldiers was the product of imperial
workshops: the ergodosia in Constantinople and elsewhere, and the
imperial arms-factories. But here we meet a number of problems.91
In the first place, while the state weaving- and cloth-factories had
90
See Hendy, Studies, pp. 654ff. See also A. Dunn, The Kommerkiarios, the Apotheke, the
Dromos, the Vardarios, and The West', BMGS 17 (1993) 3-24, at 11; and D. Jacoby, 'Silk in
Western Byzantium before the Fourth Crusade', BZ 84/85 (1991/1992) 452-500, see454
n.7.
91
See esp. J. W. Nesbitt, 'Double names on early Byzantine lead seals', DOP 31 (1977),
111-21, see esp. 115-17, and note 20; Hendy, Studies, p. 626 and note 310, and pp. 633
and 636.
 Military administration 239

originally supplied a considerable proportion of military clothing, they had

never entirely covered the demand, which had been met by private
producers whose products were handed over to the state as part of their
tax-assessment.92 Even after the commutation of much of this tax into gold
- in the laterfifthcentury - the soldiers were still equipped only partially by
the state and were expected to purchase replacement items with their cash
allowances.93 As a result of the territorial losses suffered by the empire
during the seventh century, however, the state must have found it impos-
sible to cater for all but a very small proportion of these requirements. The
majority of the weaving and clothing establishments of the later Roman
state now lay outside the empire's control: only Heraclea in Thrace,
Cyzicus in Hellespontus and Caesarea in Cappadocia, together with the
workshops in Constantinople itself, remained; and of these, both Heraclea
and Caesarea were in extremely insecure areas, subject to regular devas-
tation. It is unlikely that their establishments remained in operation into
the later 640s. 94
Similar considerations applied to weapons, which had been a state
monopoly. But by the 640s only five of thefifteenEastern fabricae remained
within imperial territory - at Nicomedia, Sardis, Adrianople, Thessaloniki
and Caesarea in Cappadocia, together with the workshop which produced
parade-armour in Constantinople.95 Of these, those at Adrianople and
Caesarea may well have ceased production due to their exposed position;
while those of Thessaloniki and Sardis - if the latter survived at all - must
have had production dramatically curtailed after the reign of Heraclius, if
not before.
Now it is clear that the few establishments left to the state can hardly
have catered for the requirements offieldarmies whose numbers, at a most
conservative estimate, must still have attained some 80,000 soldiers in
Europe, Africa, and Asia Minor in the 650s. The only option left to the state
was to promote private production - whether or not this was supervised by
government officials - of clothing and other items of equipment: boots,
leatherwork, saddlery and so on, as well as weapons; and the latter
constituted, of course, a radical break with the traditional system. What
the apotheke system and the kommerkiarioi represent, in fact, is not the
transport of materials from Constantinople, for example, to the districts
92
Jones, LRE, vol. I, pp. 433f.
93
See Jones, LRE, vol. II, pp. 625 and 670-1; Haldon, Byzantine Praetorians, p. 113 and
notes 113 and 114.
94
For the workshops see Jones, IRE, vol. II, p. 836. Those at Tyre, Scythopolis, and in
Phoenicia and Cyprus (together with others about which nothing is known, since no
complete list (unlike for the West) exists) were lost to the empire by the 650s.
95
There had been arms factories specialising in different types of production at Damascus,
Antioch, Edessa, Marcianopolis, Naissus, Ratiaria, Irenopolis (Cilicia) and Horreum
Margi. See Notitia Dig., Or. XI, 18-39; and Jones, LRE, vol. II, pp. 834ff.
240 Byzantium in the seventh century

where they were needed, and the consequent sale of such material to the
soldiers, but a much more fundamental shift.
If, as has been surmised, but need not necessarily have been the case, the
kommerkiarioi are contract-farmers, dealing in materials (and it is worth noting
that a number of their seals bear the impression of sacking on one side, imply-
ing their attachment to the actual goods)96 and administering the require-
ments for specific military undertakings, then the accumulated evidence sug-
gests the only viable option left to the state: they were contractors who were
actually in charge of levying the equipment - weapons, clothing and so on -
from local producers and craftsmen in the areas in which the troops were based.
What must have operated was, in effect, a compulsory levy in kind, raised
either in the form of an advance against future tax-assessments, or as a
compulsory purchase at a low fixed price (although the emphasis placed by
the state on avoiding cash transactions may argue against this) or as a
variant of the simple munera extraordinaria or aggareia. The crucial point for
the state was to minimise its own outgoings and maximise the extraction
of resources from the population. And here the later system operated in the
ninth and tenth centuries becomes relevant, for it is exactly the same
procedure - albeit now under the supervision of the thematic prdtonotarioi
and strategoi - that is to be found. In the account of the preparations for the
expedition to Crete of 911 in the Book of Ceremonies of Constantine VII,97
the method of raising arms for the thematic forces on a mass scale is
abundantly clear: each local commander, or appropriate official, was
required to ensure the production of a certain number of weapons (shields,
arrows, lances and so on), which can only have been raised by compulsory
levy from the local population.98 This seems to have been the standard
system from the ninth century on, and there is no reason for assuming that
it is not the developed form of the ad hoc system introduced in the 640s and
after to cope with the crisis which the state had to face in the production of
weapons and clothing for its soldiers.
The kommerkiarioi therefore, were contracted to arrange the production
and delivery, and possibly the distribution of weapons and equipment from
local craftsmen and producers to the soldiers in a given region or regions,
as well as from the state's own factories and warehouses. Their contract
will have lain in the state's offering them some form of return on excess
production, which could be stored in the state apothekai or armamenta; and
in the kommerkiarioi undertaking to cover any shortfall out of their own
resources. There was nothing at all new in this principle, except its

96
Nesbitt, 'Double names', 115-17; Hendy, Studies, p. 626 note 310.
9
7 De Cer., 657. 12sq.
98
For a detailed analysis, see Haldon, Byzantine Praetorians, pp. 319-22.
 Military administration 241

application to the production of military equipment." And the reasons for

the kommerkiarioi being commissioned in this sphere are quite straight-
forward: they had traditionally supervised the taxes and duties on the sale
of finished goods produced by state workshops and disposed of through the
special bank of the prefecture. The special bank, in its independent guise as
the eidikon/idikon logothesion, continued to supervise these activities,
gaining in addition control over the actual production establishments
themselves, including - as the later evidence makes fairly clear - the
production of weapons and other items of military equipment. 100 It was a
natural extension of the role of the kommerkiarioU as production began to
be contracted out by force of circumstances to provincial workshops and
craftsmen, that they should also supervise this aspect of the activities of the
eidikon. And that this was the case is strongly suggested by the fact that on
at least one occasion between 659 and 668 the general kommerkiarios of a
provincial apotheke was also stratiotikos logothetes, the official in charge of
military finance and the assessments for military provisions. 101
But such transactions could in no way have involved the sale of
weapons and equipment to the troops. If the state had stopped paying the
regular cash allowances to its soldiers, it could hardly have expected them
to pay for their equipment with cash. Instead, the state returned to the
system of the fourth century: supplies were raised in kind from the local
population as part of the regular tax-assessment, supervised by the local
prefects (of the themata or armies); equipment and clothing was now raised
similarly, by levies on local producers, supervised by the genikoi kommerkia-
rioU and for specific undertakings. The actual distribution of the material
probably continued to be administered as before, regimental actuaries
collecting and issuing it at the regular musters or adnoumia, as described in
the Strategikon of the Emperor Maurice of the late sixth century, and in
later hagiographies and military treatises. 102

99
For the classic example of tax-farming in the late Roman state, see Jones, LRE, vol. I,
p. 4 5 7 .
100
See the evidence cited in Haldon, Byzantine Praetorians, pp. 320-2 and notes. The fact
that the kommerkiarioi were in origin purely civilian officials is not a serious objection to
this development. In the first place, there were neither juridical nor institutional hindran-
ces to it; in the second place, the appointment of civil officials to military commands, for
example, where the emperor placed particular trust in the individual concerned, was
quite usual - cf. the example of the deacon John, genikos logothetes, appointed to
command the expedition assembling on Rhodes in 714 (see Theophanes, 385. 5sq. and
the discussion of Hendy, Studies, pp. 657f.). In the crisis faced by the state at this period,
such measures are only too easily explained.
101
Zacos and Veglery, no. 144, seal of Stephen, apo hypaton, patrikios, stratiotikos logothetes
and genikos kommerkiarios of the apotheke of... (dated 659-68).
102
See Maurice, Strategikon I, 2.11; and cf. Theophylact Simocatta VI, 6.4, VII, 7.1, VIII,
12.12; Haldon, Byzantine Praetorians, p. 104 and note 6 3 . For the later adnoumion, see
242 Byzantium in the seventh century

By the same token, no loosening of state control over the production of

weapons and, more significantly, the ownership of weapons need be
implied (although this may have been an unavoidable result). The strict
regulations on the shipment and possession of weapons in Justinian's
novel 85 were retained in later legislation; and while the state therefore
gave up its monopoly on production, it still intended to limit the possession
of arms to its soldiers alone. 103
There seems little doubt that by the time of the Ecloga of Leo and
Constantine in 741 provincial soldiers were assumed to own the basic
elements of their equipment - weapons, armour and, where relevant,
horse. They were responsible for its maintenance in good condition and -
certainly by the later eighth century - for the replacement of items, such as
horses, which had been lost in the interim.104 An eighth-century legal text
also makes it clear that private persons were able, but not obliged, to
contribute to the cost of equipping a soldier, thus emphasising the personal
ownership of such material.105 Equally, while the legal texts state that the
soldiers did receive some cash emoluments, referred to as rhogai, they also
received annonae, that is to say, provisions in kind. This suggests that the
term rhoga had already come to be used of the donative (and, indeed, of any
cash payment) - even though, as we have seen, this was issued probably
on a four-yearly rotational basis by this time - and that the rations in kind
were the main support of the armies.106 The fact that the same texts refer
to the annonae of civil officials makes the situation clear - the state had had
to revert almost entirely to a system of payment in kind, in the provinces at
least.
What appears to have been happening, therefore, was that the state

Leo, Tactica, VI. 15; Anon. Vdri, 48 (28.3sq. Dennis); Haldon, Recruitment and Cons-
cription, p. 63; and on the issue of weapons, see Haidon, Praetorians, pp. 114-15.
103
See Justinian, Nov. 85 and cf. Basilica LVII, 9.1; Procheiros nomos XXXIX. That private
production may have taken place - particularly in specialist production such as bows, or
mail and lamellar armour - is suggested by the existence of a memorial stone of a
bowmaker from the Attaleia region. See Gregoire, Recueils, no. 308 ( = C1G 9239), of the
seventh or eighth century.
104
See Haldon, Recruitment and Conscription, pp. 67-72 for a detailed treatment and analy-
sis of the relevant sources. The key text is Ecloga XVI, 2.
105
See especially the (probably) mid-eighth-century text of a legal decision attributed by its
editor to Leo III and Constantine V, in which a soldier's father-in-law is referred to as
contributing to the expense of equipping the former: ed. D. Simon, 'Byzantinische
Hausgemeinschaftsvertrage', in Beitrdge zur europdischen Rechtsgeschichte und zum gelten-
den Zivilrecht. Festgabe fur ]. Sontis (Munich 1977), pp. 91-128, see 94; see also Burg-
mann, Ecloga, 135-6.1 have discussed this, and related texts, elsewhere: see J. F. Haldon,
'Military service, military lands, and the status of soldiers', 2Off.
106
See Ecloga XVI, 4, where rogai are carefully defined as coming respectively from the
emperor's hand (i.e. donatives or similar payments from the emperor's own treasury);
whereas annonai and synetheiai came from the demosion, that is to say, the public fisc.
 Military administration 243

issued soldiers with their basic equipment, as in the fourth century, which
the soldiers were expected to look after and present in good order at
musters and parades (adnoumia). This was produced partly in imperial
workshops, but mostly under contract or by levy on the provincial popu-
lations. Whatever the initial legal status of such equipment may have been,
it is clear that by the time of the Ecloga text, and certainly by the time of the
writing of the Life of St Philaretus, the weapons and other material,
including the soldier's horse, were regarded as his possessions, for which
he was responsible. And while the state seems to have issued the basic
requirements in the first instance, more expensive weaponry and armour
was clearly available from private producers, or possibly from certain state
workshops. This much is clear from the second legal text of the reign of Leo
III and Constantine, referred to already. It is also evident from hagiographi-
cal sources, and from Theophanes' account of the second evil deed of the
Emperor Nicephorus I that soldiers were expected to pay for and maintain
their own equipment by the early ninth century. 107 The cumulative
pay-scales, recorded by ninth-century Arab geographers, which may
reflect the seventh-century situation or an evolution from it no doubt
contributed to soldiers' ability to cover such additional expenditure, and
may indeed represent this initial state issue when service was first begun
and the increasing obligation of the soldiers as time passed to cater for their
own equipment and mounts. But even in the ninth and tenth centuries, it
seems that the state still equipped some soldiers and levied large amounts
of military materiel in addition to expecting thematic soldiers to provide for
themselves. 108
What this means in practical terms is that two complementary systems
operated side by side. Troops raised for specific campaigns were armed and
equipped by the state - this is presumably the chief function of the eidikon
and its apothekai in our period, and of the eidikon and the thematic
protonotarioi in the ninth and tenth centuries. On such occasions, troops
already available were mobilised, too, and were expected to turn out with
their weapons, mounts and so on. Failure to do so had severe con-
sequences, as the evident fear of the soldier Mousoulios in the Life of St
Philaretus, as well as other, later examples, amply demonstrates. 109 But
107
See Vita S. Philareti, 1 2 5 , 34sq. and see Haldon, Recruitment and Conscription, pp. 5 0 - 1
and note 58 and pp. 58ff. For the reforms of the Emperor Nicephorus I, by which
impoverished soldiers were to be equipped by their fellow villagers, Theophanes, 4 8 6 .
2 3 - 6 . For the Arabic source, see lbn Khurradadhbm, Kitab al-Masalik w'al-Mamdlik, in
M.-J. de Goeje, Bibliotheca Geographorum Araborum VI (Leiden 1889), pp. 7 6 - 8 5 , see
p. 8 4 .
108
See De Cer. 6 5 7 . 12sq., 6 5 8 . 4sq.
109
For Mousoulios, see note 1 0 7 above; and cf. the similar case of a horseless soldier
reported in the Life of Eustratius {Vita Eustratii, 377. 3 - 6 ; see Haldon, Recruitment and
Conscription, p. 51 note 88). Oikonomides has argued plausibly that the four-yearly cycle
244 Byzantium in the seventh century

the state can hardly have expected soldiers whom it did not pay in cash to
buy their equipment. On the other hand, those who could afford it would
undoubtedly provide themselves with the best weapons and armour. And
as the state, keen to minimise its own outgoings, and indeed the burden on
local populations, insisted on the soldiers' responsibility for their weapons
and mounts, so it would have encouraged private investment in such
equipment, with the inevitable result that possession on the one hand, and
legal ownership on the other, were assimilated.

THE PROBLEM OF MILITARY LANDS

We can now turn to another aspect of the possible sources of support for
the soldier, that is, the question of the military lands and the degree to
which the state actually promoted a policy of providing soldiers with land
from which they could support themselves and their duties.
It has been argued that, on the assumption that the state sold weapons
and military equipment to the soldiers through the apotheke system, and
had also withdrawn regular cash payments to them except for the quad-
rennial/quinquennial donatives, it must have had to provide them with an
alternative form of income; for how else could the soldiers have paid for the
equipment available for purchase through the kommerkiarioi? Given the
types of resources available, the only viable solution in such a situation, it
is argued, would be to give soldiers land, which they and their families
could cultivate, hence providing an income for their support.110
But there are a number offlawsin this argument also. In thefirstplace,
even if some soldiers and their families were given grants of land and the
time required to bring it under cultivation, or at the least to achieve the
minimum requirements in terms of autarky - let alone the production of
the considerable surpluses necessary to support a soldier, purchase his
equipment and maintain a sustained level of productivity - even if this
were the case, it is inherently improbable that all soldiers could be sup-
ported in this way. Some would simply lack the requisite technical skill and
knowledge; others might have no family adequate to farming the land.
Quite apart from this, of course, is the fact that most campaigns took place
at precisely the times of the year when agricultural production demanded

as it is described in the tenth century represents the state's rationalisation of a system

whereby soldiers received cash (and other) emoluments due to them usually on the
occasion of a campaign. See N. Oikonomides, 'Middle Byzantine Provincial Recruits:
Salary and Armament', in Gonimos. Neoplatonic and Byzantine Studies presented to Leendert
G. Westerink at 75, eds. J. Duffy and J. Peradotto (Buffalo, N.Y. 1988), pp. 121-36, see
125-7.
110
The argument is set out in detail in Hendy, Studies, pp. 634ff.
 Military administration 245

the most attention and labour, in particular during harvesting.111 Equally,

if- as has been argued112 - soldiers were originally given lands on imperial
estates, complete with tenants to carry out the agricultural labour for them,
it is very difficult to explain how such large holdings came to be reduced to
the state of poverty implicit in the case of the soldier Mousoulios from the
Life of Philaretus - for surely such an argument must presuppose a degree
of state protection of such holdings against alienation and impoverishment
(as was the case, for example, from the first half of the tenth century). The
provision of soldiers with land, therefore, can at best have been a very slow
and partial procedure.
In the second place, it is clear from the available evidence for the eighth,
ninth and tenth centuries that there existed prior to the legislation of the
mid-tenth century no legislation to bind soldiers, land and military service
together; and if the state had given land to troops directly in return for
military service, there would surely have remained some echo of this in the
legislation or in some narrative source, somewhere. But the sources remain
either quite silent on this point, or they positively assume otherwise.
Indeed, neither of the two early eighth-century legal texts thus far referred
to suggest any obligation upon soldiers' families to support them; and they
actually make it quite clear that soldiers could leave their households and
lands at liberty. This can only have been the case if the state, in principle at
least, was still assumed to maintain its troops, if not also with afirstissue of
equipment and weapons, then certainly with regular annonae or provisions
(and later cash rhogai). What the sources do make clear is that military
service in the provincial, that is to say thematic, armies was, for many of
the soldiers if not for all of them (for some continued always to be recruited
on a mercenary basis and for the duration of a specific campaign)113 a
11
* Soldiers were officially prohibited from engaging in farming for others. For the legislation
and its relevance, see Haldon, Recruitment and Conscription, pp. 48-9 and note 83; 72-3
and note 128; Ahrweiler, 'Recherches', 8f. For an illustrative example, prohibiting
soldiers from engaging in agriculture and trade, see Leges militares (ed. Korzensky), 80, 2.
As Lilie, 'Die zweihundertjahrige Reform', 194, has noted, the continued official emphasis
on this ruling would tend to exclude an official promotion of the personal involvement of
soldiers in such activities. On the other hand, the efficacy and purpose, as well as the date,
of the 'Military Laws' remain unclear; while such prohibitions may have been intended to
apply only to full-time salaried troops. See the remarks of Oikonomides, 'Middle Byzan-
tine Provincial Recruits', 133 note 36.
112
W.T. Treadgold, 'The military lands and the imperial estates in the middle Byzantine
empire1, in Okeanos. Essays presented to I. Sevcenko {Harvard Ukrainian Studies, VII, Cam-
bridge, Mass. 1983), pp. 619-31.
113
See Haldon, Recruitment and Conscription, p. 79 note 145, and Byzantine Praetorians,
pp. 216-20. This is undoubtedly also the implication of the comments of Theodore the
Studite in a letter of the year 801 (ed. G. Fatouros, Theodori Studitae Epistulae (CFHB XXXI
1/2. Berlin 1992), ep. 7.61-63) to the effect that widows of soldiers needno longer'lament
bitterly at the piteous and inhuman demands made on their dead husbands' account'.The
text suggests that the state continued to demand the equivalent of the husband's military
246 Byzantium in the seventh century

personal and hereditary obligation. Only in the tenth century was this
obligation transferred to the land. 114
It seems, therefore, unlikely that a great many soldiers were able to take
up such an option on land, even if it were available. Yet all soldiers had
been deprived of the greater part of their regular cash income and their
cash allowances for weapons, clothing and other equipment. The notion
that the apotheke was in the business of selling weapons to such soldiers
thus becomes even more unlikely. Soldiers in this position will have been
simply unable to purchase weapons on this basis. And the only conclusion
must be that the state, as I have surmised, did not sell weapons and
equipment; it issued them directly, exploiting both urban and provincial
craftsmen and skills, and thereby saving its own resources. The apotheke
system was a convenient and pre-existing means of widening and systema-
tising the basis of production and distribution of military requirements.
The fact that it is just such a system which operated in the ninth and tenth
centuries (although no longer through the kommerkiarioi) lends strength to
the argument.
An otherwise apparently powerful argument for the state's attribution of
land to soldiers for their maintenance is thus crucially weakened. The state
could clearly support and maintain its armies on a traditional basis of
levies of provisions in kind, as it had always done for military expedition-
ary forces anyway, and as it had done during the fourth and much of the
fifth centuries. It may well have attempted to lighten the burden by
granting leave to soldiers who had families that could support them during
the winter, for example, thus having to provision those soldiers who
remained in camp or on active service only. There is some evidence, again
from the Ecloga text (xvi, 1) that this was indeed the case. 115 As the
themata became firmly established, so the probability that soldiers have
families or relatives in the garrison area increases, and thus the possibilities
for the state to transfer the leave-time support of the troops to the soldiers
themselves will also have increased. Given the difficult situation, there is
no reason to doubt that the state exploited such possibilities to the full.
None of this means that soldiers could not or did not hold or own land;
service in some other form, possibly cash or support for a replacement - the threat of the
withdrawal of the special exempt status of soldiers' families in respect of certain fiscal and
other dues will probably also have played a role.
114
The question and the sources are analysed in detail in Haldon, Recruitment and Cons-
cription, pp. 41-65. The clear evidence for a personal and hereditary military service in
the ninth and tenth centuries, but one which was in no way formally and legally
associated with land (until the legislation of Romanus I and Constantine VII) invalidates
any line of reasoning which argues for a Heraclian origin for this relationship, of course.
The argumentation of both Hendy and Treadgold is in this respect flawed. For a similar
critique of Treadgold, see R.-J. Lille, 'Die byzantinischen Staatsfinanzen im 8./9. Jahr-
hundert und die orpcmamxa xrniiotTa', BS 47 (1987), 49-55, see 50-1.
115
See the apposite comments of Lilie, 'Die zweihundertjahrige Reform*, 194 and note 77.
 Military administration 247

nor that military service did not come to be supported by and associated
with such holdings. Soldiers were clearly expected to be at least partially
self-supporting by the later eighth, and certainly by the ninth century;116
and some of the divergent views of historians who have examined this
question can be partially reconciled in this respect.
For there is also no doubt that the state did give land to individuals, from
whom it then expected to recruit soldiers. In the 590s, the Emperor
Maurice is reported to have decreed the raising of 30,000 cavalry soldiers
in Thrace, to which end he ordered the settling by forced transfer of the
same number of Armenian families to Thrace. While the figures may no
doubt be considerably inflated, the principle is clear.117 Similarly, it is clear
that the Slavs, whom Justinian II transferred to Asia Minor in 688/9
together with their families, were used as a source of recruits. Again, while
the figure of 30,000 is no doubt inflated, along with that of the 20,000
who supposedly deserted (Syriac sources give for the latter a figure of a
mere 7,000), 118 the principle seems to be clear: to draft in new populations
(and not simply individuals, but whole communities and families) who, in
the first case mentioned above certainly, and in the second case almost as
certainly, were given land, and from among whom recruits could be
conscripted. But it is important to note that these groups were ethnic bands
and, if the Slavs are anything to go by, they were organised as such, under
their own leaders, on a similar basis to that of late Roman foederati, or
possibly the laeti (who had less independence as distinct groups), within
the Western empire in the fourth and fifth centuries. They were intended
by Justinian, just like the earlier federates, to operate as an independent
corps which might fight in conjunction with Byzantine forces. The system
they represent, therefore, is not a novelty,119 but neither is it a generalised
means of recruitment including the granting of land to soldiers, for which

116
Haldon, Recruitment and Conscription, pp. 72ff. and the sources cited there. It is clear from
the Italian evidence from the exarchate of Ravenna that after about 640 the majority of
the garrison troops were, to a greater or lesser degree, able to support themselves;
although vestigial cash payments such as donativa were still made. See Brown, Gentlemen
and Officers, pp. 87f. For Africa, see Durliat, Les grands proprietaires et Vetat byzantin, 527f.
with sources and literature.
117
See Sebeos, 54f.; and for the sixth-century examples, Charanis, 'Ethnic changes', 29f.
and 32f.; and 'Transfer of population', 142.
118
See in the last instance Hendy, Studies, pp. 63 Iff.; and Stratos, Byzantium in the Seventh
Century, vol. V, pp. 34fT. Groups of Slavs had also been settled in Asia Minor by Constans
II, as their reported desertion to the Arabs in 665 suggests. For these episodes, see
Charanis, 'Ethnic changes', 42f., 'Transfer of population', 143, and 'The Slavic element
in Byzantine Asia Minor in the 13th century', B 18 (1946-8), 69-83, see 70f.
119
A point recognised by Hendy, Studies, p. 637; the settlement of Slavs to repopulate the
land was not confined to the Byzantine empire only: in the eighth century the Duke of
Istria also settled Slavs on deserted lands. See Brown, Gentlemen and Officers, p. 88 note
14; Guillou, Rtgionalisme, pp. 304 and 306 (Plea of Rizana).
248 Byzantium in the seventh century

there is no evidence at all. Indeed, the fact that these Slav 'prisoners' were
probably given land on a 'federate' basis makes more likely the probability
that, as has been suggested, they were settled on imperial estates. 120 For,
of course, part of the purpose of introducing such populations as is
sometimes made quite explicit in the sources, was the revitalisation of the
local population and the bringing back into cultivation of abandoned or
deserted lands.
There is no reason to doubt, therefore, that groups of barbarians such as
the Slavs were settled in Asia Minor, under both Constans II and Justi-
nian II, in order to provide auxiliary military support for the hard-pressed
Byzantine corps and in order to reoccupy deserted agricultural lands. The
terms on which the land was received are unknown, but if it was on the
basis of a fixed relationship between land and soldiers, one would again
expect to find echoes of the practice in the later sources. Much more likely -
since it has been shown that the late Roman imperial estates in Cappadocia
seem to disappear some time before the twelfth century and represent an
area associated on the seals of kommerkiarioi with the captured Slavs 121 -
is the probability that large tracts of imperial lands were granted to the
new 'federates', perhaps on an emphyteutic basis: that is to say, the basis
for their settlement was the provision of a corps of soldiers, but this was not
tied directly to individuals and specific properties or holdings. This hypo-
thesis is indirectly supported by the fact that under Justinian I authority
over the Cappadocian estates, as well as those of the diocese of Pontica,
was granted to the proconsul of Cappadocia, who was at the same time
(and like several other Anatolian provincial governors) given military
authority. The local administration was therefore already in a position to
facilitate the establishment of new settlers on imperial lands. 122 Initially,
the land may have been freed from all fiscal dues for three years - as was
the practice with captive 'Saracens' in the tenth century and probably
before - in order to permit them to reap an adequate return on their initial
labour. 123 And emphyteutic leases will have meant ultimately that the
120
Hendy, Studies, p. 6 3 7f.; a point m a d e also, a l t h o u g h for all soldiers, by W.T. Treadgold,
The Byzantine State Finances in the Eighth and Ninth Centuries (New York 1 9 8 2 ) and T h e
military lands a n d the imperial estates' (cited in note 1 1 2 above).
121
Hendy, Studies, pp. 637f., with pp. 1 0 4 - 6 a n d 1 3 3 - 5 a n d m a p 2 9 ; Kaplan, 'Maisons
divines', 8 5 .
122
See Kaplan, 'Maisons divines', 8 5 ; Justinian, Nov. 30, 1.6 (a. 536); a n d Jones, LRE, vol. I,
pp. 280ff.
123
See De Cer. 6 9 4 . 2 2 - 6 9 5 . 14. For a brief s u m m a r y of the principles u p o n which land w a s
allotted to federates a n d to the b a r b a r i a n s in the West in the fifth century, see Jones, LRE,
vol. I, pp. 2 4 9 - 5 3 . Note t h a t the land t h u s granted remained subject to the basic
land-tax, w h e t h e r it w a s granted in perpetuity or not. Perpetual emphyteutic leases of
imperial lands, of the sort already c o m m o n in the East in the later sixth a n d seventh
centuries, would h a v e provided the East R o m a n state with a m e a n s of both bringing land
back into cultivation, recovering lost tax revenue, a n d a p e r m a n e n t source of recruits.
 Military administration 249

land so granted will eventually have come to be regarded as much as the

property of those settled on it as that of the state. The disappearance of the
Cappadocian estates might thus be explained.
There is therefore good evidence that the state could and did settle
groups of immigrants on the land and in return demand military service
from the first generation at least of such 'federates'. But such a system, as
we have seen, can hardly have applied to the regular soldiery. Instead, the
probability is that these continued at first to be garrisoned in camps, or be
barracked or billeted on the population - supported by direct levies of
provisions from the local provincial population. Where soldiers had their
families in the region, they may have been encouraged to support them-
selves to a degree, certainly when on leave in winter quarters - so much is
clear from the early eighth-century legal texts referred to. Gradually - I
have argued this elsewhere - and just as in the earlier cases of Africa,
Egypt, Syria and, differently, in Italy, as the armies became more per-
manently established across their cantonments, the soldiers will have
inevitably become almost entirely recruited from local sources, as the first
generation put down roots, intermarried with the provincials, obtained
land or interests in other property and so on. Indeed, members of the army
of the exarchate in Italy seem regularly to have taken out emphyteutic
leases on Church properties, encouraged both by the Church, as a means of
maintaining a steady income, and by the state, as a means of ensuring
adequate support for the soldiers. The result in the East seems to have been
that during the later seventh and the eighth centuries the local military
command and the state were able to rely increasingly on the soldiers
supporting their service at least partially out of their own or their families'
pockets - especially in view of the fact that the property of soldiers and
their immediate dependents had traditionally been, and continued to be,
exempt from all but the basic public taxes. 124 In this respect, military
service entailed privileged status for the soldier's household and family.
This development implies no formal obligation, of course, but the legal
stipulations of the Ecloga and the decision of Leo and Constantine dis-
cussed above accurately reflect a situation in which private support for
military service was becoming usual. The only formal obligation was in
respect of the hereditary nature of military service, probably reintroduced

124
Haldon, Recruitment and Conscription, pp. 74ff. For the privileged position of soldiers' land
(the possession of which, of course, did not necessarily imply their actual physical
involvement in its exploitation) see ibid., p. 54 and note 94, p. 73 and note 129, p. 60
and note 104; Brown Gentlemen and Officers, pp. 87f. for a similar situation in the
exarchate of Ravenna. For emphyteutic leases to imperial troops, see ibid., p. 177 and
esp. pp. 105f.
250 Byzantium in the seventh century

under Heraclius, possibly later, but certainly during the seventh cen-
tury. 125
The state thus had at least one major, traditional and above all well-tried
option open to it, through which it could continue to support and supply its
armies. The granting of land, while it certainly occurred, was on the other
hand neither the only available means of supporting soldiers, nor indeed a
particularly effective one, given the difficult conditions of the time; nor
again is there evidence for it, except indirectly - but undeniably - for the
transplanted Slavs during the reigns of Constans II and Justinian II. Here,
however, special circumstances prevailed, and the institutional conditions
on which the latter were established approximates to those of the fifth-
century foederati and laetl rather than to those of the recruitment and
maintenance of the regular soldiers in the imperial forces. For the latter,
the state returned to a system similar to that which had operated in the
fourth and fifth centuries.
One final observation needs to be made. AH the developments outlined
above seem to apply throughout the lands of the empire in the seventh and
early eighth centuries (with the possible exception of the two exarchates),
except for the methods of paying expeditionary forces in the Balkans. In
Asia Minor, as we have seen, the evidence suggests that the state found it
necessary to support and equip its troops almost entirely in kind, with the
exception of donatives and other occasional cash payments. No doubt cash
- gold - was still found with which to attract new recruits for specific
campaigns, for example; but in general, cash was not the rule. On the other
hand, and as Hendy has pointed out, excavations from Athens, where
relatively high concentrations of copper coins of Philippicus and Leo III
were found, all minted in Constantinople and associated both in time and
in location with the presence of the military on specific campaigns or
actions, suggest that cash still played a role in some aspects of the military
administration of these regions, in contrast to Asia Minor. 126 It may well
be a reflection of local conditions and the dearth of available resources in
the Balkan lands still held by the empire at this time, which meant that the
state found it easier to pay a part of the soldiers' salary in cash - rather
than provide the supplies in kind - and to use coin as the medium of

125
Haldon, Recruitment and Conscription, pp. 36ff. Against this, see Lilie, 'Die zweihundert-
jahrige Reform', 1 9 3 note 6 9 . That hereditary service w a s introduced as late as the reign
of Leo IV, however, seems improbable - ibid., 199f. The fact t h a t the hereditary n a t u r e of
military service is not referred to in either Ecloga XVI, 2, or the decision attributed to Leo
a n d Constantine (see Lilie, 'Die byzantinischen Staatsfinanzen' (cited in note 1 1 4 above),
51) is n o evidence t h a t hereditary service w a s n o t applied at that time - such a n argu-
m e n t e silentio is hardly persuasive, t h e less so in view of the fact t h a t neither text is in the
least w a y concerned with this aspect.
126
See Hendy, Studies, pp. 659ff., with sources a n d literature.
 Military administration 251
exchange with local traders and producers. It may also reflect imperial
administrative tradition in these districts.127 But while this evidence does
show that the state could and did employ coin on occasion, it does not
appreciably alter the general picture drawn in this chapter, one in which
the reversion to both the appropriation and redistribution of revenues in
kind plays a major role.
The seventh century thus witnessed a major transformation in the fiscal
and military administrative machinery of the state, in which the late
Roman institutions inherited from the preceding period were recast to cope
with a radically changed set of circumstances. It is important not to
underestimate the degree of planning and foresight which this process
demonstrates. For while the state was certainly responding to pressures
which could hardly have been foreseen, it did not react blindly to circum-
stances. Both in the shape offiscalas well as of military reorganization and
readjustment, the late Roman and early Byzantine state demonstrates a
remarkable functional coherence, an aspect which has perhaps been
rather underplayed in recent debates.

EXCURSUS

The unequal allocation of territory to the different army corps, or themata, is more
apparent when compared with the size of the field armies in question. These are
difficult to estimate exactly, but the figures which can be obtained from the Notitia
Dignitatuum for the East, which dates to about 420, where they can be corroborated
by figures from the reign of Justinian or afterwards, suggest that the main field
armies of the empire in the mid-sixth century were maintained at approximately
the same strength as in the early fifth century. According to Notitia Dig., Or. V-IX,
the armies will have numbered very approximately as follows:

127
See Hendy, Studies, pp. 418ff., 65Iff and 662 for a detailed discussion and analysis.
Hendy has also convincingly argued that the later (ninth- and tenth-century) differences
in methods of paying Eastern as opposed to Western thematic strategoi and troops reflects
the original late Roman pattern: the themata which occupied the territories of the
praetorian prefecture of the East were paid from Constantinople on a four-yearly rotation
which, as we have seen, was based on the old quinquennial issue of donatives. Their
strategoi were paid annually also from Constantinopolitan funds. In contrast, the Balkan
themata, which developed on the territory of the old praetorian prefecture of Illyricum,
were maintained on a different basis, their soldiers and strategoi being paid from local
funds, reflecting the late Roman source of cash, through the prefectural headquarters at
Thessaloniki. Such continuities can also be shown to exist in the case of the naval themata
which developed via the command of the Karavisianoi from the 'pseudo-prefecture' of the
quaestura exercitus, with its bank (arca/trapeza) in Constantinople. The later strategoi of
these themata were paid also from Constantinople in the ninth and tenth centuries.
Likewise, the later thematic commander of the district of Mesopotamia was paid from the
proceeds of the local kommerkion, just as the late Roman dux of the same command had
received his salary via the local commerciarius. See Hendy, Studies, pp. 650-4.
252 Byzantium in the seventh century

magister militum praesentalis I -20,500

praesentalis II -22,500
per Orientem -20,000
per Thracias -23,500
per Illyhcum -17,500
total -104,000.
Calculations depend, of course, on the strength of the various units, which is also
difficult to assess - see the figures offered by Jones, LRE, vol. II, pp. 682f. and
vol. Ill, p. 379, table xv. Figures for the sixth century suggest comparable
numbers: in 499 an army of 15,000 was available to oppose a Bulgar attack in
Illyricum (see Marcellinus comes, s.a. 499; Procopius, De Bello Gothico III, 29.3); in
530, after the new magister militum per Armeniam had been established, Belisarius
commanded a reinforced Eastern army of 25,000; and of 20,000 in 531; while in
543 a total of some 30,000 men is recorded on the Persian front (Procopius, De
Bello Persico 1,13.23,18.5; II, 24.16). Procopius notes that a force of some 12,000
men operated in Lazica in 554 (De Bello Gothico IV, 13.8), and this must represent
the greater part of the corps under the magister militum per Armeniam. Thefigureof
30,000 for 543 probably represents most of the combined field troops of both
Eastern and Armenian commands. In Italy, the field armies numbered variously
from 12,000 to 18,000 between 542 and 554 (Procopius, De Bello Gothico I,
5.2-4, III, 3.4; and Agathias II, 4); in Africa thefieldforces seem to have numbered
about 15,000 (Procopius, De Bello Vandalico I, 11.2, 11-12, 19). Agathias also
suggests, perhaps over-pessimistically, a grand total for Justinian's armies of about
150,000 in about 560 (Agathias V, 13). The size of the field armies suggested by
this evidence (approximately 12,000 to 20,000, depending upon the theatre)
seems to represent the norm at this period. See R. MacMullen, 'How big was the
Roman imperial army?', Klio 62 (1980), 451-60; and H. Castritius, 'Die Wehr-
verfassung im 5. Jahrhundert n. Chr. und der Untergang des romischen Reiches',
Jahrbuch des Deutschen Evangelischen Instituts fur Altertumswissenschaft des Heiligen
Landes 2 (1990). Also on numbers, and the relationship between size of army,
civilian population and resource base, see J.-M. Carrie, X'Esercito: trasformazioni
Funzionali ed economie localf, in Societd Romana ed Impero Tardoantico. Instituzioni,
Ceti, Economie, ed. A. Giardina (Rome 1986), pp. 449-88, 760-71, with further
literature. The figures can also be revised downwards, to produce considerably small-
er units and overall sizes: see the useful discussion in M. Whittow, The Making of
Orthodox Byzantium, 600-1025 (Basingstoke-London 1996), pp. 181-93 and, on
the figures offered by Jones, loc.cit, see R. Duncan-Jones, Structure and Scale in the
Roman Economy (Cambridge 1990).
These figures suggest that in the period c, 500 to c. 550 at least, thefieldforces of
Illyricum and Oriens numbered approximately the same as in the 420s; that the
forces of the magister militum per Armeniam, which covered a much less extended
area than those of the Eastern forces from which they had been separated,
numbered about 12,000-15,000; that those of Italy and Africa numbered about
15,000 each; and that those of the praesental and Thracian forces numbered also
about the same as in 420. The forces of the mid-sixth century must therefore have
numbered as follows:
 Military administration 253

magister militum per lllyhcum -17,500 „ including troops of the

Thracias -23,500 J quaestura
Orientem -25,000
Armeniam -15,000
praesentalisI -20,500
praesentalisII -22,500
Africam -15,000
Italiam -15,000
total-154,000
This figure - granted its very approximate nature, the fact that troops were also
based in Spain, and the estimate of Agathias - seems to be reasonably representa-
tive. There must have been some changes in the period from Justinian to Heraclius,
of course, and the total was probably reduced. The command of the magister
militum per lllyhcum, some of whose units (and perhaps some from the army of
Thrace as well) will have been separated to form the army of the Justinianic quaestura,
had been effectively lost by the 630s and 640s - its vestiges may well have been
incorporated into the praesental forces. The two praesental corps themselves seem
to have been amalgamated (the maximum praesental force envisaged in the
so-called Strategikon of Maurice is of 24,000 men: see Haldon, Byzantine Praeto-
rians, pp. 96ff. and 546 note 782; and C. M. Mazzucchi, 4Le KATArPA4>AI dello
Strategicon di Maurizio e lo schieramento di battaglia dell'esercito romano nel
VI/VII secolo', Aevum 55 (1981), 111-38, see 1251T.), and it is very likely that units
were transferred from both to the Eastern, African and Italian armies. But given
these changes, there is no reason to doubt that the Thracian, oriental and
Armenian corps continued to be maintained at roughly their Justinianic strengths,
that is, of some 20,000-25,000 for the first two, and some 12,000-15,000 for the
last. Thefiguresare, as we have said, hypothetical; but they do illustrate what must
have been the proportional numerical relationship between the three armies. Thus
regardless of whether or not the higher hypothetical totals for the armies arrived at by
Jones and the smaller numbers supposed by Whittow actually pertained, the propor-
tional difference between them remains more or less constant. The losses incurred in
the war with the Persians, and with the Arabs, the probable desertion or disbandment
of many units, particularly those which had been based in a given locality for many
years and on a more or less permanent basis, will have meant ultimately the with-
drawal of forces probably smaller than those previously enlisted. But the figures are
indicative of the fact that the smallest division seems to have been allocated the most
extensive region as its resource area in Anatolia.
 CHAPTER 7

Society, state and law

STATE AND LEGISLATION

The study of Byzantine jurisprudence in the seventh and eighth centuries,
of judicial procedures, the operation of the courts, the appointment and
background of the judiciary and notarial staff, these are all areas which, on
account of the sparsity of adequate sources, are particularly difficult to
elucidate. Of the imperial legislation of the period, the novels of Justinian's
immediate successors1 and of Heraclius,2 the Ecloga promulgated in 741, 3
together with the anonymous treatises or law books' known as the
Farmers' Law 4 and the Rhodian Sea Law,5 with the so-called 'mutiny act',
or treatise on military discipline and punishment, 6 are all that we have.
And the date of the last three texts - all very derivative and drawn from
Justinianic or pre-Justinianic material - is still very much debated. Only
with the help of casual references in other sources is it possible to say
anything about judicial practice, the application of the law and the
availability of justice. In this section, I shall make some attempt to see how
1
See/GUI, Coll. I, 1-21.
2
JGR I, Coll. I, 22-5; new edn by J. Konidaris, 'Die Novellen des Kaisers Herakleios', FM V
(1980), 33-106 (text: 62-95).
3
Ecloga. Das Gesetzbuch Leons III. und Konstantinos' V, ed. L. Burgmann (Forschungen zur
byzantinischen Rechtsgeschichte X, Frankfurt a. M. 1983). For the date, see ibid.,
pp. 10-12; and 0. Kresten, 'Datierungsprobleme'. On the legal judgement appended in
some Mss. to the Ecloga (as chapter 19), but which represents an independent legislative
act, see D. Simon, Byzantinische Hausgemeinschaftsvertrdge, pp. 91-100.
4
See the literature cited in chapter 3, notes 117-18; and Pieler, 'Rechtsliteratur', p. 440 and
notes 115ff.
5
W. Ashburner, Ndfiog'Podtcop Navrixog. The Rhodian Sea Law (Oxford 1909 and Aalen
1976) (and in JGR II, 91-103). See Pieler, Rechtsliteratur, pp. 441f. For the two latter texts
as 'law-books' rather than as formal legislation, see the literature in Pieler, 'Rechtslitera-
tur', p. 433 and notes 30 and 31.
6
W. Ashburner, 'The Byzantine mutiny act', JHS 46 (1926), 80-109 (and in JGR U, 75-9);
E. Korzensky, 'Leges poenales militares e codice Laurentiano LXXV, Egypetemes Philologiae
Kozlony (Budapest 1930), 155-63, 215-18 (in JGR II, 80-9). For literature and discussion,
see Pieler, 'Rechtsliteratur', p. 444, notes 147-50.

254
 Society, state and law 255

these aspects of the traditional and sophisticated system of jurisprudence

and judicial practice fared during the period in question.7
The sixth century, and in particular the reign of Justinian, had been a
Golden Age of late Roman legislative productivity and codification. This is,
with hindsight, largely a reflection of the fact that the fifth and sixth
centuries were the only period in Byzantine history when a professional-
ised legal apparatus existed, and in which a formal legal training and
career-structure determined the nature of jurisprudence and the admin-
istration and activities of the court system.8 The Codex Iustinianus, the
Institutes and the Digest, along with the collection of novellae consti-
tutiones of Justinian's reign (together with several novels of Justin II and
Tiberius Constantine and three prefectural edicts) established some time
after 575, constitute a milestone in late Roman and early Byzantine legal
history.9 Thereafter, and in spite of the possible loss of some legislation
from the following century and a half, the amount of legislation promul-
gated in the traditional form of the novel dies to a bare trickle. Four novels
and the Ekthesis of Heraciius, the Typos of Constans II, the Edict of 681 of
Constantine IV,10 the Ecloga and a judicial ruling probably from the reign
of Leo III and Constantine V, and the three treatises referred to above,

7
For the history of jurisprudence and legal literature in this period and later, see the older
survey of Zacharia, Geschichte, pp. 5ff.. esp. 11-15; L. Wenger, Die Quellen des romischen
Rechts (Vienna 1953); B. Sinogowitz, Studien zum Strafrecht der Ekloge (Upay. rfj? Axct8.
'A$i)v<hv XXI, Athens 1956); also P.I. Zepos, 'Die byzantinische Jurisprudent zwischen
Justinian und den Basiliken', in Berichte zum XL Internationalen Byzantinisten-Kongress V,l
(Munich 1958), esp. pp. 7-13; Pieler, 'Rechtsliteratur*, pp. 429ff. For some general
remarks on the relationship between law and state, albeit in an extremely formalised way,
see P.E. Pieler, 'Verfassung und Rechtsgrundlagen des byzantinischen Staates', Akten des
XVI Internationalen Byzantinisten-Kongresses, 1,1 (Vienna 1981), pp. 213-31 ( = JOB
31,1 (1981)). It must be stressed that the difficulties facing the historian in respect of
law. justice and legal administration in this period are particularly great. The lack of
sources makes definite conclusions about many aspects impossible, and hypotheses
hazardous. I have tried to interpret this limited material through the context - cultural,
social and ideological - already elaborated; and it is this context, therefore, which informs
the suggested evolution outlined here.
8
See the remarks of D. Simon, 'NoiiOTpipovfievot/, in Saturn Roberto Feenstra Sexagissimum
Quintum Annum Aetatis Complenti ab Alumnis Collegis Amicis Oblata, eds. J.A. Ankum, J.E.
Spruit and F.B.J. Wubbe (Fribourg 1985), 273-83.
9
See esp. P. Noailles, Les Collections des novelles de Vempereur Justinien. La collection grecque
des 168 novelles (Paris 1914). The Justinianic codex is most easily consulted in the edition
of Th. Mommsen, P. Kruger, R. Scholl and W. Kroll, Corpus luris Civilis (3 vols., Berlin
1892-95 and 1945-63), I: Institutiones. ed. P. Kruger and Digesta, ed. Th. Mommsen; II:
Codex Iustinianus, ed. P. Kruger; III: Novellae Constitutions, eds. R. Scholl and W. Kroll. For
the most recent general survey, see Pieler, 'Rechtsliteratur', pp. 407-19.
10
Ekthesis: Mansi X, 991B-997A; Lateran, 156.20-162.13 (Dolger, Regesten, no. 211);
Typos: Mansi X, 1029C-1032D; Lateran, 208.15-210.15 (Dolger, Regesten, no.225); edict
of Constantine IV: Mansi XL 697A-712D; Riedinger, 832.1-856.6 (Dolger, Regesten, no. 245).
256 Byzantium in the seventh century

dating to the later seventh or first half of the eighth century, are all that
survive, and they probably represent in fact most of what was enacted. 11
This lack of imperial legislative activity is significant. It reflects at the
very least a change in the methods employed by the emperors in enforcing
and publicising their policies. For we have already seen that a whole series
of major administrative transformations took place, especially from the
later years of the reign of Heraclius on, and it seems inherently improbable
that all the novels of the emperors from Constans II to Leo IE have been
lost. The later tradition itself makes this unlikely, for there would surely,
somewhere, in the codification of the Macedonian period, for example, be
some reference to this legal and promulgatory activity.
It is, in fact, far more likely that no such activity took place, that is, that
no novels (for example) were produced, because the matters of administra-
tive organisation, civil law and Church affairs traditionally handled in this
way were now dealt with quite differently. This change must have set in
during the reign of Heraclius, for we would surely expect something of the
legislative activity of this emperor - whose fiscal reorganisation, for
example, introduced a whole series of major changes in the relevant areas
of the state administrative machinery - to have survived if it had existed.
But from Heraclius on, we have only a small group of novels dealing with
matters of Church discipline and clerical organisation. 12 The seventh
century, indeed, is a period of dramatic, often rapid, and certainly con-

11
See the comments of Pieler, 'Rechtsliteratur', p. 434. Legislation which has survived only
in accounts of other contemporary or later sources does not change this picture dramati-
cally. The greater part is concerned with matters of ecclesiastical jurisdiction and/or fiscal
policy. See Dolger, Regesten, nos. 174, 176, 182, 197, 205, 206, 227, 234, 237, 246
and 258, for example.
12
Of the legislative activity of Justinian - the Novellae Constitutiones and related texts - some
50% deal with matters of administrative, fiscal and military organisation, problems of
provincial government and so on. Some 30% deal with civil and criminal law, and the
remaining 20% are concerned with Church property and related issues. See M.-Th. Fogen,
'Gesetz und Gesetzgebung in Byzanz. Versuch einer Strukturanalyse', lus Commune.
Zeitschriftfiir europdische Rechtsgeschichte 14 (1987), 137-58, see esp. 140ff., and N. van
der Wai, Manuale Novellarum. Apergu systematique du contenu des Novelles de Justinien
(Groningen and Amsterdam 1964). For the novels after Maurice, see the remarks of P.J.
Zepos, 'Die byzantinische Jurisprudenz zwischen Justinian und den Basiliken', 7. There are
no novels after Heraclius until the reign of Eirene, and then of Leo V. Those of the latter, as
well as those of Heraclius, are concerned exclusively with Church matters (Heraclius),
divorce, or the question of the validity (and moral acceptability) of oaths. See L. Burg-
mann, 'Die Novellen der Kaiserin Eirene', FM IV (1981), 1-36; and Kresten,
'Datierungsprobleme'. It may justifiably be objected that the failure of legislation to appear
in later sources or to survive in an extremely limited manuscript tradition is no guarantee
that it did not exist - one of the known novels of Irene being a case in point (see
Burgmann, 'Die Novellen der Kaiserin Irene'). On the other hand, while this may be true of
one or two such items, a regular series of promulgations is most unlikely to disappear so
completely, either from the later jurisprudential tradition or from the known manuscript
tradition.
 Society, state and law 257

tinuous change, both in the administrative competence of various depart-

ments of state administration, in military organisation, in economic and
fiscal structures, and in the conditions within which these organs had to
function. Nowhere is this reflected or mentioned in a legislative text -
although the passing reference of the Emperor Constantine IV to the effect
that he and his armies had been so preoccupied with matters of civil and
military importance that he had been unable to consider the question of
summoning an ecumenical synod until 680 (to consider the pressing
problem of monotheletism) is eloquent.13 The Ecloga of Leo III and Con-
stantine V deals exclusively with questions of property relations, marriage,
divorce and inheritance, contracts, loans, leases and punishment.
Nowhere does it touch explicitly upon the question of either a specific
context for the legislation or the administrative framework within which
the law is held to operate (although again, the introduction to the Ecloga
speaks volumes about the background conditions with which the emper-
ors were confronted in issuing the legislation).14
An initial conclusion from this data must be that emperors simply did
not legislate in the same way as they had done previously. In other words,
the novel was effectively abandoned as a way of intervening in matters of
either judicial or civil-political interest. Two questions follow: what
medium replaced the novel as a means of expressing and enforcing the
emperor's will, on the one hand, and of affecting and modifying the
operation and activity of the state apparatus, on the other? And what are
the reasons for this sudden break? The answer to the second question, I
believe, goes some of the way to providing an answer to the first.
In thefirstplace, we must begin by recalling that the word 'law' depends
for its strength very much on the context in which it is employed. There
may, to begin with, be various forms of legislative activity which had the
strength of 'law' in the modern sense - i.e. universally valid, regulatory
and normative statements - but which are not referred to as 'norms' in
medieval texts. Secondly, there may be a considerable difference between
the promulgation of laws and their actual effectiveness, the more so if the
law in question was not widely or sufficiently known. The latter is gen-
erally recognised to have been a major problem in the reign of Justinian
and after; and it must be borne in mind that 'laws' may be ineffective both
because they remained insufficiently known or understood, as well as
because they were tacitly or explicitly ignored (that is, because they
represented a theoretical or desired state of affairs too far removed from the
reality of the times and impractical properly to enforce).15
13 14
MansiXL201C-D;Riedinger,10.21ff. Pieler, Rechtsliteratur, pp. 438^*0.
15
See especially F. Wieacker, 'Zur Efiektivitat des Gesetzesrechts in der spiiteren Antike', in
Festschrift fur Hermann Heimpel turn 70. Geburtstag (Gottingen 1972), pp. 546-66 (repr.
258 Byzantium in the seventh century
In the second place, the failure of laws, in pragmatic terms, their lack of
effectiveness, does not mean that they remained without function.16 On
the contrary, a corpus of laws or legislative acts which enunciate a (more
or less) consistent world view, a moral system, or whatever, regardless of
its practical relevance in day-to-day terms, nevertheless constitutes a
normative system and, where regularly invoked, acts as a symbolic refer-
ent, invoking also key aspects of the values of the culture in which it
exists.17 In Byzantine terms, of course, the invocation of the Justinianic
legislation and codification played a crucial role throughout the history of
the empire, recalling the commonality of orthodox Christianity, Roman
civilisation, and the role and paramount position of the emperors. Just as
significant was the figure of the Emperor Justinian himself in the later
tradition.18 The invocation of the Justinianic corpus in the Ecloga and in
the 'Farmers' Law', quite apart from the codification of the ninth century
and later,19 is evidence enough of this. And it has been already demon-
strated that the legislative activities of the Emperor Leo VI, for example,
were grounded in the assumption that, whatever the changes which had
actually taken place in the interim, it was still the legislation of Justinian
which provided both the framework within which society had to be
interpreted and understood and the law applied, and which determined the
interpretation of Byzantine efforts to re-establish the Roman oikoumene as it
had once been, as a valid enterprise.20
It is just this symbolic aspect which is, I would argue, crucial to the
period with which we are concerned. The law, whether or not its detailed
prescriptions and demands were understood or applied, symbolised the

in Ausgewdhlte Schriften I (Frankfurt a. M., 1983), pp. 222^*0). For the later seventh and
early eighth centuries this was clearly a major consideration. In the prooemium to the
Ecloga (line 43) it is remarked that the members of the commission responsible for
producing this new selection needed to collect the necessary texts together and that the
laws are, for the most part, intellectually inaccessible. Whether even an imperial archive
existed which possessed all the relevant material is a debatable point. See Pieler, 'Rechts-
literatur', p. 430; L. Burgmann, review of Pieler, in Rechtshistorisches Journal 1 (1982), 14
and n. 3.
16
See Fogen, 'Gesetz und Gesetzgebung', 147.
17
See P. Noll, 'Symbolische Gesetzgebung', Zeitschrift fur schweizerisches Recht 100 (1981),
347-64, and G. Lanata, Legislazione e natura nelle Novelle giustinianee (Naples 1984).
18
G. Prinzing, 'Das Bild Iustinians I. in der Uberlieferung der Byzantiner vom 7. bis 15.
Jahrhundert', FM VII (1986), 1-99.
19
See A. Schminck, Studien zu mittelbyzantinischen Rechtsbuchern (Forschungen zur byzanti-
nischen Rechtsgeschichte XIII, Frankfurt a. M. 1986), esp. p. 80 and note 136; pp. 103
and 107 (for Leo VI); and Ecloga, tit. lsq.; Fanners' Law, tit. 1 (ed. Ashburner, 85; JGR n,
63). See Pieler, 'Rechtsliteratur', pp. 440-1 and 449f. Note that the Justinianic motif
occurs in other contexts - in the reign of Constantine IV, for example, whose son was so
named (hardly an accident). Constantine was acclaimed in the final session of the sixth
ecumenical council as the 'new Justinian': see Mansi XI, 656B.
20
Fogen, 'Gesetz und Gesetzgebung', 148f.; Pieler, 'Rechtsliteratur', pp. 449f.
 Society, state and law 259
Roman state and everything that accompanied that notion. It symbolised
the power of the emperor and his relationship to his subjects, and it acted
as a symbol of his role and the tasks which he had to fulfil. It existed as the
theoretical backdrop, as it were, to the practical ideology of the state and to
the political-cultural beliefs and assumptions upon which people based
their understanding and explanation of the world as they perceived it. like
the soteriological theology of orthodox Christianity, which provided one
aspect of the theoretical understanding of the world, the law as an abstract
but systematised structure represented another facet of that understand-
ing, at least as far as it was available to the literate elements of society. And
in this respect, of course, new legislation was not perceived as a need.
What was required in the era after Justinian, and especially from Heraclius
on, was conformity (assumed or real) to the norms set out by the system, or
the reassertion and re-establishment of such norms, in so far as they were
understood, of course (a major consideration in itself).
Seen from this perspective, the legal 'system' became less a practical
instrument for intervening in the world of men in order to modify relation-
ships or individual behaviour, but more a set of theories which represented
a desired (if recognisably not always attainable) state of affairs. Emperors
needed to issue no new legislation, therefore, but rather to establish (or to
re-establish) the conditions within which the traditional system would
once again conform to actual practice. Imperial action was thus not
directed at emending laws to conform to reality, but rather at emending
reality to conform to the inherited legal-moral apparatus.21 The struggle to
maintain imperial authority and imperial intervention (and insistence
upon the right to intervene) in matters of dogma can be readily understood
in this light.
What appears to have happened between the later years of the sixth and
the middle of the seventh century, therefore, was in effect the exaltation of
an interventionist, regulatory legal system, which concerned both
administrative-functional aspects of the state's existence, as well as legal-
ethical practices of its subjects, into an abstract and idealised 'world', in
which the latter aspect attained a pre-eminence at the expense of the
former. So much is clear from the concerns and preoccupations of Byzan-
tines in the period from the end of the reign of Heraclius.22 For the
underlying reasons for this development are clearly connected with the
21
This can be seen already in the second novel of Heraclius, A.D. 6 1 7 , in which the explicit
intention is to lend an older piece of Justinianic legislation new authority and practical
relevance. See Konidaris, Die Novellen des Kaisers Herakleios, 74.22sq. and commentary,
1 0 0 - 2 . It is even more clearly expressed in the Typos of Constans II issued in 6 4 8 , which
is specifically intended to re-establish the situation which prevailed before the conflict over
the wills and energies of Christ developed: see Lateran, 208.19sq.
22
See Haldon, Ideology and social change; and see chapters 9 to 11 below.
260 Byzantium in the seventh century

parallel developments in late Roman cultural attitudes already discussed.

Indeed, the deliberately fostered gulf between emperor and court on the
one hand, and the ordinary population on the other, emphasised by ritual
and public ceremonial, must have been a major stimulus to this shift in
attitude to the law and imperial legislation. Promoted coincidentally by the
activity of the antecessores and scholastikoi (discussed below), and testified
to by the collections of novels made during the period up to the end of the
reign of Heraclius,23 the Justinianic corpus became the measure against
which all alternatives had to be assessed, a process that was furthered by
the intervention of its originator himself, incorporated into the prooemia of
the novels. For with some exceptions, where Justinian had to withdraw or
emend his own legislation, the novels were intended to be relevant to all
subjects of the empire, and for all times. The ban on interpretations of
commentaries to the Digest reinforces the message. 24 Those novels which
concern radical administrative measures form an exception, but even here
the underlying assumption is that these 'specific' measures are short-term
and should lead to the restoration of the 'healthy' situation, which had
been originally intended. 25
A system was thus erected which came to represent for the generations
that followed, and particularly in the cultural context as I have described
it, a perfect world, to which one could refer and direct one's energies but
which, in practical terms, and in the terms of any perceived or day-to-day
reality, was quite unreal. It was a world both emperors and subjects
aspired to return to and that provided an invisible, taken-for-granted
permanence against which experience and actual developments and situ-
ations were to be measured. Changing the framework laid down by
Justinian was tantamount to tampering with the world as it really was,
hidden, of course, behind the facade of the temporary set-backs of his-
23
See Pieler, 'Rechtsliteratur', p. 4 3 1 a n d note 2 0 , a n d 436f.; a n d note, for example,
L. B u r g m a n n a n d Sp. Troianos, 'Appendix Eclogae', FM III (1979), 2 4 - 1 2 5 .
24
See especially t h e discussion of Fogen, 'Gesetz u n d Gesetzgebung', 140ff. For t h e prohibi-
tion o n writing interpretations of commentaries into legal manuscripts, see H.J. Schel-
tema, 'Das Kommentarverbot Justinians', Tijdschrift voor Rechtsgeschiedenis 4 4 (1976),
3 0 7 - 3 1 , w h o shows t h a t t h e legislation in question does n o t place a b a n o n commen-
taries to t h e Digest as such, as h a s usually been thought. See, for example, A. Berger, 'The
Emperor Justinian's b a n u p o n Commentaries to the Digest', Quarterly Bulletin of the Polish
Institute of Arts and Sciences in America 3 (1945), 6 5 6 - 9 6 .
25
Cf. Justinian, Nov. 1 4 5 (a. 553), proem, 16sqq.: Toi<s d e l TTapeixTrCirrovcrL rr\v
TTpocnixovaav ExdoTOTe ftepcnreiav evpurxovTes, eireiS&v T& TTJS xpc^a<5 TrapeXftri,
TTCLXIV rfj*; irpoTepas 7iv6|Ae / da Td£e(o<;, |xexpi |xovov TCU TTETTOVTIXOTOS TT|V laTpetav
iorcovTes ('Since we find for all irregular situations the appropriate cure, so we return to
the original state of things when the need is over, applying the cure only to the cause of
the problem'.) Compare with the sentiments expressed in the proem to novel 29, on the
restoration of the older administrative structure (A.D. 535).
In this connection, see in particular the important discussion of M. Maas, 'Roman
history and Christian ideology in Justinianic reform legislation', DO? 40 (1986), 17-31.
 Society, state and law 261

torical change. Legislation that was enacted, therefore, represented

'temporary' responses to 'temporary' situations, the long-term intention
being to restore the world of the Justinianic era. And already for the period
beginning with Justin II, it has been suggested that the considerable decline
in imperial legislative activity which sets in at this time reflects this
unwillingness to tamper with Justinian's great edifice.26
By the reign of Constans II, as I will argue,27 the situation had
changed sufficiently, in terms of people's appreciation of what was happen-
ing and of their immediate political-ideological priorities and interests, that
the moral-ethical aspect of their world and its meaning bore far more
significance than problems of administration. Imperial 'legislation' of the
period is concerned almost exclusively with such matters - problems of
belief, of everyday observance, the avoidance of sin and the common
responsibility for incurring the wrath of God, of the relationship between
the earthly authority of the emperor and the Church and that of heaven -
these are the dominant motifs of the times.
In ecclesiastical terms, the proceedings of the Quinisext, and in secular
law, the Ecloga - itself explicitly based upon the Justinianic system - are
key moments.28 The major instances of imperial legislative acts for the
period are, almost exclusively, similarly preoccupied with such matters -
the Ekthesis of Heraclius, the Typos of Constans II, the Edict of 681 of
Constantine IV, all mentioned above. The prefatory sacrae and the speeches
before the sixth ecumenical council in 680 and the Quinisext prioritise,
along with the difficulties faced by the emperor on the military front, just
these issues, although at a higher level of abstraction.29
The administrative and functional legislation of Justinian and his
immediate successors, however, finds no such echo. The last examples of
such legislation concern the introduction of the silver hexagram in 615
and the abolition of the public bread ration in 618. 3 0 For the rest of the
seventh century and much of the eighth, imperial interest in matters of
administration and, especially, of fiscal affairs is expressed through the
issue of imperial 'commands': iussiones, keleuseis or prostagmata. But even
these are relatively few in number: only five are explicitly referred to by the
literary sources up to the reign of Leo III, although there must have been
considerably more.31 The arbitrary nature of such references, of course, is
no real guide to the frequency or infrequency of issues of such commands.
26
See Zepos. 'Die byzantinische Jurisprudenz zwischen Justinian und den Basiliken', 7.
27
See especially chapter 9 below.
28
See Ecloga, 160-1.1-6 with comment at p. 4; Pieler, 'Rechtsliteratur', pp. 438-9.
29
See Mansi XI, 201C-D; Riedinger, 10-12 {sacra to patriarch George); 195B-D; Riedinger,
2-10 (letter of the emperor to Donus, received by Agatho).
30
Dolger, Regesten, nos. 167, 173 and 174.
31
Dolger, Regesten, nos. 234, 249, 250. 255 and 256.
262 Byzantium in the seventh century

But the lack of legislative activity in the form of novels on the part of the
emperors, and the fact that such matters seem henceforth to be dealt with
by the issue of iussiones to the relevant parties, is suggestive of the nature of
the change. The legal administrative framework of the Justinianic era now
became the 'norm', whether it actually reflected the real situation or not.
Legislative enactments intended permanently to emend this framework
were therefore, and in principle, extraordinary. The novels of Heraclius
provide an example, albeit on a very limited scale. 32 The emperors could
modify the arrangements of their illustrious forebears by introducing
'temporary' measures, especially where these concerned fiscal policies and
the appropriation of resources. Thus, even a major work of codification
such as the Ecloga was compiled with the specific intention of making the
Justinianic legislation more readily available and comprehensible. The
modifications it introduced, especially in respect of divorce and marriage
law, were clearly perceived as an attempt to strengthen the force of the older
legislation and to reaffirm the already existing values of a Christian
society.33
The various iussiones or keleuseis which do receive a mention in the
sources are thus concerned with a range of matters, from fiscal policy to
questions of imperial policy in respect of the Church, for example. But I
suggest that the drastic and far-reaching changes which the administra-
tive and military apparatuses of the state underwent at this time were
perceived as (temporary) deviations from a structure that was still imma-
nent in the legislative framework of the Justinianic era; and because these
changes would ultimately be made good, there was no need for permanent
legislation. Temporary modifications, carried out through specific com-
mands, were all that was required. Most of these commands - and there
must have been many hundreds of them - receive no mention in the
sources because the sources are, by their very nature, not interested in
such matters. Occasionally, as with the 'ten evil measures' of the Emperor
Nicephorus I in the early ninth century, or the fiscal measures that affected
the papal lands in southern Italy, they do attract the attention of a
chronicler or commentator for specific ideological reasons. 34 But such
examples are few.
A context thus developed in which legislative activity of the sort pursued
32
See Konidaris, 'Die Novellen des Kaisers Herakleios', 64.39sq., where Heraclius notes that
Justinian's arrangements are no longer in force due to the passage of time. See Konidaris'
commentary, 94ff.
33
Pieler, 'Rechtsliteratur', pp. 4 3 0 - 1 and 438ff.
34
See Theophanes, 4 8 6 - 7 , and Ostrogorsky, Geschichte, pp. 152ff.; Dolger, Regesten,
nos. 2 3 4 and 2 5 0 . Dolger (Regesten, nos. 372ff.) followed by Pieler, 'Rechtsliteratur',
p. 4 3 4 , assumes that many of Nicephorus' measures were in the form of novels. But there
is no evidence for this at all.
 Society, state and law 263

by Justinian and his immediate successors became ideologically redun-

dant. It was not only a question of a shift in the perception of the function
and relevance of such legislation, however. The political-military environ-
ment was itself unfavourable to the promulgation of formal legislation
such as Justinian had produced. In the first place, the question of the
dissemination and the assimilation of legislative enactments, referred to
already, had to be faced; and in the context of much of the reign of
Heraclius, when the empire was to a greater or lesser extent harassed or
occupied by enemies, when communications were vulnerable, and when
the emperor himself was fully committed to managing the very survival of
the state, the difficulties will have been so much greater. If this problem
presented difficulties to the legislators of the (relatively) peaceful sixth
century, how much more so will it have affected the situation in the middle
and later seventh centuries? It is surely no accident that the four surviving
novels of Heraclius all deal with the affairs of the Constantinopolitan
Church, and of these only one addresses the question of the clergy in the
provinces and their relationship with the civil and military authorities.
In the second place, the implementation of a legislative device such as a
novel presupposed a degree of stability and continuity of both economic
and especially of administrative institutions within the state. It assumed
also the ability of the relevant imperial officials in the provinces, for
example, to devote their attention to the matter in hand, unhindered by
other major disturbances. Again, during much of Heraclius' reign, and in
the remaining parts of the empire from the 640s on, this situation did not
exist for many areas over considerable periods of time. Legislation of the
Justinianic type was for the most part both too unwieldy and too slow to
take effect, as well as being impossible uniformly to put into operation. It is
not without significance that Justinian II overcame the problem in part by
commanding the presence of the court, the representatives of the popu-
lation of Constantinople and the leading military officers of capital and
provinces in order to witness and sign their agreement to the acts of the
sixth ecumenical council, which were read out to them in 687. By this
means he could be sure that the key representatives of the state and the
civil and military population of the empire were made aware of the formal
acceptance of the decisions of the sixth council and the end of any
monothelete policies.35
In the context of the rapid changes forced upon the state administrative
and military machinery in the second half of the seventh century
especially, therefore, the traditional form of interventionist imperial legisla-
tion must have become quite impracticable. In order to respond to the

35
See Mansi XI, 737-8; Riedinger, 886-887.
264 Byzantium in the seventh century

multifarious demands of distant provinical officials, of the armies and their

needs, of the state fisc and the central administration, it must have become
much more efficient and effective to respond to each specific situation as
and when it arose, a pattern of legislating which was, of course, already to
hand. And the evidence that is available supports this hypothesis. No
general legislation is recorded, but a large number of individually tailored
imperial commands, edicts, letters to local officials, both military and civil,
and so on do occur in the sources. The contrast could not be more obvious,
and the total absence of the traditional legislative forms in this period is
compelling. The vast array of changes - in fiscal administration, military
organisation, the structure of the palatine bureaux - are nowhere dealt
with in the traditional form. Legislative acts concerning them are entirely
lacking, as are references in other sources to such acts. The changes
themselves appear usually only obliquely, sometimes in more detail than
others, in narrative sources and, indirectly, through the evidence of seals,
coins and inscriptions.
The lack of legislative activity on the part of the emperors in the seventh
and much of the eighth centuries thus becomes more easily compre-
hended. The effects of a shift in emphasis in the perception of the Justinianic
codification and legislation, further affected by the radically changed
economic and political environment of the seventh century, produced a
very different view of the role and significance of imperial legislation. This
view is reflected in practice in the Ecloga. It is a view in which the novel of
the sixth century, as a convenient means of responding to and modifying a
given situation, as well as a symbol of imperial authority, lost the former
function to the advantage of the latter. In turn, the legislative framework of
the Justinianic age became fossilised and hypostatised, representing an
ideal, but lost (or at the least, 'hidden'), world which must be restored. If we
put to one side the two novels of Irene and that of Leo V, the novel is only
revived as an effective instrument of imperial authority in the tenth
century. 36

THE ADMINISTRATION OF JUSTICE

This major change in emphasis seems also to reflect, and may also be
partially responsible for, what is generally recognised to have been a
considerable decline in legal scholarship during the seventh century,
although there was a clear distinction between imperial legislation on the
one hand, and the activity of lawyers, jurists and legal scholars in general
on the other. In the mid-sixth century a specialist legal education was
36
Bearing in mind also the particular function of, and intentions behind, the novels of Leo
VI: see Fogen, 'Gesetz und Gesetzgebung', 148ff.
 Society, state and law 265

available in several cities of the empire and represented one of the most
reliable routes to rapid promotion in the imperial civil administration.
Professors of law were respected socially, and a career in the profession
was greatly sought after.37 The most famous law school was in Beirut,
which flourished until the devastation of the city by an earthquake in 551.
The curriculum established by the professors of law, or antecessores, lasted
from five to six years, and it is chiefly on the basis of the paraphrases of and
detailed notes on the Justinianic legislation established in their teaching
that later curricula were based. 38 For whatever reason, the antecessores
seem to have ceased their activity in the mid-560s, perhaps a result of the
breaking off of imperial financial support after Justinian's death; thereafter
legal education seems to have been conducted by the scholastikoi, practis-
ing lawyers or barristers. The principal difference lay in the method of
teaching. The antecessores based themselves on the original texts - usually
Latin - adding a loose translation and a series of exegetical notes for
clarification and interpretation. In contrast, the scholastikoi based them-
selves on summaries of the original, or paraphrases, together with the
Greek commentaries and text of the original legislation. In spite of the ban
on certain types of exegetical commentary laid down by Justinian in the
Digest, commentaries, interpretations and paraphrases of the Justinianic
material continued to be produced, intended both as teaching materials
and as exegetical texts for magistrates and judges, their form determined
principally by the fact that Greek, rather than Latin, had by now become
the dominant language of legal and administrative practice. Many of these
commentaries, such as the paraphrases of the antecessores Stephanus and
Theophilus, were in any case exempt from the ban, counting as Indices'
rather than as interpretations in the true sense. 39 A number of handbooks
of a practical nature for actual use by lawyers were also produced, dealing
with a variety of specific themes, although not all of these have survived or
are known. 40
From the reign of Phocas, however, the legislative activity of the emper-
ors declined almost to nothing, and this seems to have been paralleled by a
similar reduction in the activity of lawyers, commentators and inter-
preters. From the evidence of the Appendix Eclogae, a late eighth- or early
ninth-century compilation based on the Ecloga of Leo III and Constantine
3
7 See P. Petit, Les Etudiants de Libanius (Paris 1957), esp. pp. 166ff.
38
See HJ. Scheltema, 'Byzantine law', in CMH, vol. II, part 2, pp. 55-77, see 55-60.
39
HJ. Scheltema, UEnseignement de droit des antecesseurs (Leiden 1970), esp. pp. 61^1.
40
See the summary with literature of Pieler, Rechtsliteratur, pp. 434-8; and esp. Wenger,
Quellen, pp. 682-92; Zepos, Die byzantinische Jurisprudenz zwischen Justinian und den
Basiliken, 8f. Note also HJ. Scheltema, 'Korreferat zu P. Zepos, 'Die byzantinische Juris-
prudenz zwischen Justinian und den Basiliken", in Berichte zum XL Internationalen
Byzantinisten-Kongress, pp. 35^41, see 37-8.
266 Byzantium in the seventh century

V,41 some evidence of the legalistic activity of the seventh century can be
culled. Thus, references to excerpts from the Justinianic material, as well as
to specific treatises, such as the Poinalion (based probably on the work of
the Justinianic antecessores), occur, suggesting that a degree of juristic
activity continued, in Constantinople at least, in this period, albeit limited
chiefly to tracts on canon law or on punishment. 42 On the whole,
however, the situation in the seventh century does not seem to have been
favourable, either to the teaching of law or to an interest in its literary
tradition. The situation up to the end of the reign of Heraclius may, in itself,
have presented no major obstacles; although the failing fortunes of the
provincial civitates and the lack of resources with which the state could
finance the Constantinopolitan law school must have adversely affected
the study of law. 43 Thereafter, with the loss of the Eastern provinces, and
the devastation of much of the Balkans and Asia Minor, the study of law in
the provinces must have been drastically affected.44
The prooemium of the Ecloga paints a gloomy picture. According to the
emperors, the available legal literature (of which there must have been a
reasonable amount in private possession at least in the capital) 45 was
barely understood, if it was accessible at all, outside the capital. The
purpose of the Ecloga and the commission which was constituted to
compile it was, in the first instance, to collect the older texts and codifi-
cations, organise the material and select from it what was deemed relevant
to the situation hinted at, and to compile this material in readily under-
stood language into a concise reference-book. While the Ecloga did intro-
duce a number of novelties, especially in respect of its dependence on
canon law, it was based largely on pre-existing, that is, Justinianic, for the
most part, precepts. The extent to which elements of local or regional
customary law were also intended to be represented is difficult to say. But it
is highly probable that the maintenance of the Justinianic system in the
provinces was partial and heavily influenced by local custom. The fact that
the older legal handbooks were neither easily available nor understood
implies as much; and together with the nature and the explicit purpose of
the Ecloga, this seems more than likely. Even in the sixth century, the
population (and the officials) in frontier districts occasionally adopted the

41
New edn by Burgmann, Troianos. See also Burgmann, Ecloga, pp. 134-5.
42
See Pieler, 'Rechtsliteratur', p. 437; Burgmann, Ecloga, p. 2; Zepos, 'Die byzantinische
Jurisprudenz zwischen Justinian und den Basiliken', 24ff.
43
See Stein, Studien, pp. 3f. While the academy in Constantinople may have survived into
the second half of the seventh century, there is no evidence that legal studies continued to
be pursued there. See Scheltema, L'Enseignement de droit des antecesseurs, p. 63; and Pieler,
'Rechtsliteratur', 429.
44
Note also C.E. Zacharia von Lingenthal, V Ilpoxeipos NO/J,OS (Heidelberg 1837), XHI.
45
See the remarks of Burgmann, Ecloga, pp. 2 - 3 .
 Society, state and law 267

legal practices of neighbouring districts outside the empire; the situation

must have been very much more difficult to control at times during the
seventh century.46 Particularly interesting is the intriguingly brief refer-
ence in the prooemium to the corruption of those entrusted with judicial
functions, the influence of the powerful and the wealthy as compared with
the poor. These are, of course, not unusual topoi in imperial legislation. But
in the context of the prooemium, there is no reason to doubt either that they
were meant to reflect a real situation.47 Partly to ensure that the law was
practised impartially, the emperors also undertook to secure the regular
salary (from the sakellion) of the quaestor and the antigrapheis and all
others appointed by the state to supervise and promote the execution of
justice.48
The prooemium of the Ecloga, therefore, and the internal arrangement
and contents of the compilation itself, are valuable pointers to the social
and cultural situation of the Byzantine state in the early eighth century;
and there is no reason to assume other than that they reflect a situation
which had prevailed for several decades already - the fact that the
emperors express the hope that the general applicability of imperial legisla-
tion might once again be re-established is indicative of their perception of
the situation. The disappearance of the late antique cities, the economic
and social disruption of the wars with the Arabs and the exigencies of state
administrative and military reorganisation had produced a situation in
which imperial legislation and Roman (Justinianic) jurisprudence were
confined to those areas over which the imperial government had direct
and constant supervision - effectively, Constantinople and its environs. No
doubt the situation was, in theory at least, also controlled by imperial
bureaucrats and officers in the provinces, but in practice, as the prooemium
to the Ecloga implies, the judiciary was corrupt and venal and, to judge
from the initial impetus behind the compilation, unable to understand or
apply the traditional legislation in a competent or effective manner.
46
Detailed analyses in Burgmann, Ecloga, pp. 3-19; Pieler, 'Rechtsliteratur', 438-40, as
regards local traditions and customs. Note that Justinian also incorporated regulations for
the application of regional customary law into the corpus of Roman law in the codex: see
VIII, 52: 'quae sit longa consuetudo'; cf. Digesta, I, 3. For the adoption of 'foreign' law
(regarding marriage), see Justinian, Nov. 154, 1; and the novel of Justin II of 566, JGR I,
Coll. I, 3. On the ways in which Byzantine judges handled customary law, the existence of
which was, of necessity, clearly recognised in the imperial legislation, see D. Simon,
'Balsamon zum Gewohnheitsrecht', in Xxo\ia. Studia ad Criticam Interpretationemque
Textuwn Graecorum et ad Historiam luris Graeco-Romani Pertinentia Viro Doctissimo D.
Holwerda Oblata, eds. W.J. Aerts, J.H.A. Lokin, S.L. Radt and N. van der Wai (Groningen
1985), pp. 119-33.
47
Ecloga, prooemium, 52sq. Note in particular the hope, expressed in lines 87-95, that the
changes which the Ecloga would inaugurate would also make possible the rehabilitation
of the general applicability and acceptance of the imperial legislation.
48
Ibid., prooemium, 102-9.
268 Byzantium in the seventh century
What might loosely be termed the 'companion' texts to the Ecloga - the
Farmers' Law and the military code in particular (texts probably produced
at about the same period) - tend implicitly to bear out this conclusion. It
remains unclear as to whether the Farmers' Law actually constituted a
legally valid promulgation, or - like the military code, excerpted from the
so-called Strategikon of the Emperor Maurice of the later sixth century,
and from the Digest and the Codex Iustinianus - merely a legal handbook
consisting of extracts drawn mostly from older sources.49 The Farmers'
Law in particular applied to a clearly provincial context and was expressed
in an informal, albeit precise style closely paralleled by that of the Ecloga.50
Again, while based on extracts from the Corpus Iuris, an element of
customary law is evident which illustrates the fact that the monolithic
Justinianic codification could no longer have been regarded as either
accessible or easily understood in this provincial context, whatever its
ideological import may have been.51
The evidence of the surviving legal texts, therefore, both legislation and
handbooks, suggests that until some time in the middle of the seventh
century the traditional legislative framework and its administrative basis
continued to exist, although under rapidly changing circumstances. The
extent to which the activity of the scholastikoi was maintained is impossible
to say, although the references to the legal literature in contemporary or
later texts suggests that from the reign of Phocas it was very restricted and
- given the situation in the provinces after the 630s - probably limited
almost entirely to Constantinople. This is also borne out by the remarks on
the subject in the prooemium to the Ecloga. That scholastikoi continued to
function after the middle of the century is suggested by canon 71 of the
Quinisext council, which refers to their students as those who learn the
civil laws', forbidding them to follow traditional (pre-Christian) custom
such as going to the theatre or horse-races, wearing gowns and so on. The
canon says nothing specifically of the location of this activity, of course,
but Constantinople is almost certainly to be understood - the reference to
theatres and horse-racing suggests as much. But these lawyers and their
students appear in no contemporary legal text or literature.52 After the
640s, and as a result of the withering away of municipal cultural life, the
49
See Pieler, 'Rechtsliteratur', pp. 442-4 with literature; but see D. Simon, 'Provinzialrecht
und Volksrecht', FMI (1976), 102-16, see 106.
50
Pieler, 'Rechtsliteratur', p. 441.
51
For a useful definition of 'popular' law - that is, as regional interpretations of imperial
legislation, as opposed to something of quite different origins - see the remarks of Simon,
'Provinzialrecht und Volksrecht', 115-16.
52
Mansi XI, 976A; but note that one manuscript tradition of the Ecloga mentions scholasti-
koi in the compilation of the text. Its reliability remains unclear. See Ecloga, prooemium,
42, and pp. 106f.
 Society, state and law 269
exigencies of warfare and defence, and the sharp contrast between provin-
cial and Constantinopolitan life, it is highly likely that officials with judicial
responsibilities had little or no legal training and were thrown back on
their own resources for the resolution of judicial matters. The misinter-
pretation of late Roman law and the introduction of local and traditional
modes of resolving certain types of case or conflict must have been an
unavoidable consequence, and this is, in effect, the situation hinted at in
the prooemium to the Ecloga.
As far as concerns the administration of justice at this period, there is
virtually no evidence apart from one or two mentions in hagiographical
literature pertaining to Constantinople.
In the sixth century, and probably well into the reign of Heraclius, the
lowest courts had been those of the defensores civitatis in each city, who
could try minor civil suits and could also hear and then refer minor
criminal cases. The next level were the courts of the provincial governors,
the ordinarii iudices; and above them, acting as courts of appeal, the courts
of the diocesan vicarii, or spectabiles iudicii. The courts of the bishops were
also empowered to act in civil cases parallel to the courts of the first
instance, providing (after the reigns of Arcadius and Honorius) both
parties involved had agreed to transfer the case from the secular to the
ecclesiastical authority. Final appeal rested with the praetorian prefects,
although a supplication could be made to the emperor if it was felt that an
unjust decision had been reached (unjust, of course, very much determined
by the context of the case in question and by the resources of the opposing
parties!). Justinian himself introduced a number of reforms in an attempt to
eliminate unnecessary appeals coming from the provinces to Constantin-
ople; in Asia Minor especially, he increased the number of governors of
spectabilis rank to bolster the intermediate courts of appeal, permitting
some governors (those of Armenia I, Cappadocia, Palestine and Armenia
III) to hear appeals from their own, as well as the neighbouring, province.
In addition, the prefect of Constantinople heard appeals from certain
provinces, as well as being responsible for the administration of justice in
Constantinople and environs; and the provincial appeals of the quaestor
exercitus were split on account of its peculiar geography: appeals from
Scythia and Moesia were heard by the quaestor himself; those from Caria,
Rhodes and Cyprus by his representative in Constantinople, who sat with
the quaestor sacri palatii.53
This straightforward system of regular courts was complicated by the
existence of a wide range of special courts which had jurisdiction over the
53
The best general survey of judicial administration and the law in practice is still that of
Jones, LRE, vol. I, pp. 479ff.; see also Stein, Bas-Empire, vol. H, pp. 467ff.; Brehier, Institu-
tions, pp. 219-24.
270 Byzantium in the seventh century

affairs of members of the various fiscal, administrative and military depart-

ments of the state, including the army. The relationship between the
jurisdiction of such courts and the regular courts was very complex, and it
was regulated by equally detailed and complex legislation. In general, the
jurisdiction of special courts over their staff and dependents tended to
prosper at the expense of the ordinary courts, partly because the state
usually found it advantageous to privilege the activities of its own officials,
partly because of the inherent institutional momentum, fuelled by the
self-interest of the officials in question, which promoted such a privileged
status. The system was further complicated by the existence of the privilege
of praescriptio fori, according to which certain types of case were to be
heard in specific courts only, and by which certain classes of persons could
claim the jurisdiction of their own special courts, rather than that under
which they would normally have come according either to domicile or (in
criminal cases) the place of the crime. 54
Many administrative bureaux possessed their own courts, chiefly the
fiscal departments of the largitiones and the res private, and those of military
commanders and civil governors. The complexities and confusion which
this proliferation of jurisdictions caused promoted in consequence an
extension of the use of praescriptio fori wherever this right was granted, as a
sure way of circumventing the maze of conflicting bureaucratic interests.
The system inevitably favoured the privileged and the wealthy; and while
the emperors made some effort to simplify the procedures involved and to
disentangle the complex web of jurisdictional competences - including a
limitation on the application of prescription of forum (later rescinded as
impracticable!) - the system in its essentials seems to have survived into
the seventh century. 55
Since the majority of judges in the usual understanding of the term were
attributed with their function by virtue of their position as governor,
administrator or military officer, or as one of the heads of the great
administrative departments at Constantinople, it is not surprising that
most of them had little or no legal training. To ensure a degree of legal
expertise, therefore, all magistrates with judicial duties had one or more
assessors or advisers, selected usually by the officials in question them-
selves, but salaried by the state. These advisers were usually selected from
barristers, and a career as assessor was one way of approaching a provin-
cial governorship or magistracy. In Justinian's reign a panel of twelve

54
Jones, LRE, vol. I, pp. 484-6.
55
Jones, LRE, vol. I, pp. 4 8 7 - 9 4 . For restrictions on the use of prescription of forum, see C]
III, 25.1 (a. 439); but note III, 23.2 (a. 440). For a graphic account of the way in which
the administration of justice inevitably favoured the privileged, see Priscus of Panium
Fragmenta (in FHG IV, 71-110), frg. 8 (86ff.).
 Society, state and law 271

trained lawyers and judges (the radices pedanei) was re-established (it was
originally instituted under Zeno) upon whom the emperor or leading
officials could call to act as delegate judges. But in general, provincial
judges remained relatively unlearned in the law - the majority purchased
their office or obtained it through similar means - and were easily over-
awed by powerful local magnates or state officials of higher rank than
themselves.
The imperial court and legal bureaux in Constantinople (as opposed to
the courts of the urban prefect and other special jurisdictional courts) were
administered by the quaestor sacri palatii, instituted by Constantine I, and
the magistri of the scrinia memoriae, epistularum and libellorum (responsible
respectively for imperial prescripts, references from judges to the emperor
and the preparation of trials). The quaestor and the praetorian prefect of
the East were delegated by the emperor to represent him in a great number
of suits, but he also exercised his personal jurisdiction. In such cases, he sat
together with the members of the consistory; and under Justinian, it
became the rule that the full senate should be involved, in other words,
that for important trials every meeting of the consistory should be regarded
as a meeting of the senate. 56
Barristers were attached strictly to the bar of the judge in whose court
they were enrolled. They were trained in the law schools in Constantinople
and Beirut, and could transfer out of their courts only by becoming an
assessor to a magistrate. Higher office could be achieved by enrolment in
one of the higher courts, as opposed to those of the provincial or municipal
level. The law seems to have been a popular profession, bringing social
status and possibly wealth and an official dignity as well. Many of the
higher state officers of the fifth and sixth centuries rose through the bar
and into the state administrative apparatus through service in Constantin-
ople or in the court of a high-ranking official. By the sixth century,
candidates for the bar had to produce certificates and evidence of their
training and competence; but at the same time, as a strict limit on the
numbers of barristers attached to each bar was enforced, retired lawyers
began to claim priority for entry to the profession on behalf of their sons. 57
This was sanctioned by the state, but the effect was to turn the profession
into a closed and almost hereditary caste. While a rigorous education in
the law and in rhetoric (an indispensable element of an advocate's train-
56
Jones, LRE, vol. I, pp. 4 9 9 - 5 0 7 . For the ruling on silentium and conventus of the consistory
and the senate respectively, see Justinian, Nov. 6 2 , 1 (a. 537). See also Stein, Bas-Empire,
vol. II, pp. 469f.; and esp. pp. 71-A with notes. For the panel of judges, see Justinian, Nov.
82, proem and 1; and cf. CJII, 7.25 (a. 519).
57
For certificates and qualifications, see CJ II, 7 . 1 1 / 2 (a. 460); II, 7. 2 2 / 4 and 2 4 / 4 (a. 517).
For quotas and priority, C/II, 7. 11 (460); II, 7. 2 2 / 5 (505); II, 7. 2 4 / 5 (517); H, 7. 2 6 and
proem (524).
2 72 Byzantium in the seventh century

ing) remained a sine qua non, this development was not in itself damaging
to the profession.58 But in the conditions of the seventh century, its effect
may well have been to promote a rapid decline in the number of barristers
and in their standards of learning and competence.
As well as the lawyers, notaries were also an important, if lower-
ranking, element in the legal profession. Notaries were very common, a
fact which probably reflects the reasonable living that was to be made from
the profession, since there was always a demand for the drawing up of
wills, contracts, conveyances and so on. Notaries themselves and their
assistants had to be registered, presumably to ensure that they possessed
the requisite skills and knowledge. In Constantinople, according to novel
44 of Justinian, they were organised in offices or stations, where their
assistants and apprentices worked. 59
The extent to which this legal profession, and the jurisdictional system
which supported it, survived beyond the middle of the seventh century is
difficult to say. There is no evidence at all for the continued education of
barristers after the early seventh century, although it is probable that a
small number continued to train in Constantinople and to fill posts there. 60
Evidence for the continued activities of notaries is equally lacking. The
prooemium to the Ecloga implies that provincial judges were either entirely
ignorant of the Justinianic law, or unable to interpret it; but in doing so, it
suggests that, as before, the system of provincial courts continued to
function in the hands of local governors. This was certainly still the case in
the reign of Heraclius, as a trial of a priest in Cyprus in 629, presided over
by the local archon, or governor, demonstrates. 61 The Ecloga statement
might also suggest that assessors were no longer available to assist the
provincial governors, implying that the teaching of law had been very
much reduced (and confined perhaps only to Constantinople).62 On the
other hand, there is good evidence, albeit for the ninth century and after,
that the numerous special courts with jurisdiction over their staff (and
their dependents) continued to exist, in other words, that the principles
58
Jones, LRE, vol. I, pp. 514f.
59
Jones, LRE, vol. I, p. 5 1 5 . For the fees of notaries, see the table in the Edict of Diocletian,
VII, 4 1 (they were paid by the line). See S. Lauffer, ed., Diokletians Preisedikt. Texte und
K o m m e n t a r e 5 (Berlin 1971), p. 1 2 0 .
60
See B u r g m a n n , in Ecloga, pp. 106f., however, for a possible reference to scholastikoi in
741.
61
Ecloga, prooemium, 5 2 - 9 5 . For the trial in 6 2 9 , see F. Nau, 'Le Texte grec des recits utiles
a l'ame d'Anastase (le Sinaite)', OC 3 (1903), 5 6 - 9 0 , see 69f.
62
The later symponos attached to the u r b a n prefect appears to be descended from the late
R o m a n assessor: see Bury's remarks, Administrative System, pp. 7 0 - 1 ; Oikonomides,
Preseance, p. 3 2 0 . But this is n o g u a r a n t e e t h a t assessors were still attached to provincial
governors. The consiliarius (assessor) w h o w a s purportedly involved in negotiations
between Maximus Confessor and imperial officials in 6 5 6 m a y h a v e been such a n official;
but the context offers n o certainty for such a n identification: PG XC, 169B.
 Society, state and law 273

upon which the late Roman jurisdictional system had been based sur-
vived.63 Whether or not prescription of forum continued to exist is difficult
to say. 64
Until at least the reigns of Heraclius and Constans II there is good reason
for supposing that the system familiar from the sixth century continued to
function, however. The collection of miracles of St Artemius, put together
in the later seventh century and recording events dating from the reign of
Heraclius to the 660s, refers to a number of officials familiar from the
Justinianic period. The commentariensis of the prefect of the city, the
secretarius, a subadiuva and the standard judicial procedures of the earlier
period all occur. 65 In the same collection, reference is made to a patrikios
and senator who was also one of the twelve judicial commissioners
established in Constantinople. He is referred to in this text as a fteios
8iKaorf|s, the correct title. The narrator situates the story in the reign of
Heraclius, and again it is clear that the system of the Justinianic period was
still in operation. 66 In the reign of Constans II, the trial of Maximus the
Confessor took place before the full senate and consistory, as was tradi-
tional. 67 The so-called Hypomnestikon, or commemoratio, written by Theo-
dore Spoudaios shortly after the death of Maximus in exile in Lazica in
662, refers also to the judicial officer of the prefect of the city, in this
instance called protosecretarius.68 It can reasonably be assumed that nota-
ries continued to be employed for much the same purposes throughout this
period, since they appear again in the ninth-century Book of the Eparch,
and a degree of continuity seems likely; although their organisation and
the extent of their legal knowledge remain unknown. In the Kletorologion of
63
See Peira LI, 29; and note Basilika VI, 1 for sixth-century legislation retained in later
codifications. Cf. also Procheiros Nomos IV, 1 1 , 18 and 24; DAI, 5 1 . 5 4 - 6 5 . For the special
jurisdiction as applied to soldiers, see Haldon, Byzantine Praetorians, pp. 304ff. The Ecloga
also makes it clear that courts with differing jurisdictional competences continued to exist:
see XIV, 7 (line 656sq.).
64
See Haldon, Byzantine Praetorians, pp. 306f.
65
Miracula S. Artemii, 22.26-23.27', probably of the later years of Heraclius. The commen-
tariensis was an important official on the judicial side of all offlcia of leading state officers,
civil and military, such as the praetorian prefect, the urban prefect, the magister offlciorum
and the magistri militum. There was usually only one such official attached to each
officium; his superior was the primiscrinius, or subadiuva, of the judicial section, and above
the latter came the cornicularius and the head of the section, the princeps. See CJ I, 27.
1/24 (a. 534); John Lydus, De Magistratibus II, 16; III, 8. For the duties of the commentarie-
nsis, who was responsible for the custody of prisoners, criminal proceedings and related
matters; John Lydus, De Magistratibus HI, 16 and 17; CTh., IX, 4 0 . 5 (a. 364), VIII, 15. 5
(a. 368) and IX, 3. 5 (a. 371). See also Jones, LRE, vol. II, p. 587. The princeps in this text
seems to be represented by the secretarius, who would normally hear the case himself. In
this example, the prefect happens to be present and intervenes personally (23.1sq.). Note
the fee paid to the secretarius (8 hexagrams) and the commentariensis (3 hexagrams): ibid.,
23.23^.
66 67
Miracula S. Artemii, 1 7 . 1 0 - 1 2 . PG CX, 88C and 109C.
68
Devreesse, Hypomnesticum, 68.26-9.
274 Byzantium in the seventh century

Philotheus of 899 the nomikoi (or tabellarioi) are still under the authority of
the urban prefect.69 The judicial side of the urban prefect's bureau seems
also to have survived, under a different nomenclature, in a developed form:
the later logothetes tou praitoriou is almost certainly to be identified with the
older princeps, or secretarius, of the prefect's judicial department. 70 Simi-
larly, the bureau of the quaestor can be followed - with less difficulty, since
the post was an important one and is mentioned more frequently in the
sources - from this period through to the tenth century. 71
While the evidence is rather thin, therefore, it is possible to see a degree
of structural continuity in the capital at least, in respect of the principles
upon which justice was organised, administered and dispensed. The evi-
dence of the Ecloga suggests that the study of the law and knowledge of the
Justinianic and pre-Justinianic corpus were maintained in Constantinople,
however difficult the circumstances - although it is worth noting that in
the prooemium the legal texts from which the commission is to draw its
material have first to be sought out before they can be collected together
and excerpted; a fact which suggests that, if the law did continue to be
studied (by scholastikoi, for example) it was on an individualised and
private basis. No single library of legal texts seems, even in the palace, to
have existed at that time. 72
The pre-eminent position of the praetorian prefecture must have suf-
fered, however, as the administrative changes which have already been
discussed took effect, resulting in a diminution of the importance of the
prefect's court. In its stead, the court of the urban prefect seems to have
become the first court in the capital, alongside the imperial court and that
of the quaestor - note that it is the prefect of Constantinople who is
responsible for both Pope Martin and Maximus the Confessor after their
arrest and trial, as well as for the brothers Theodore and Euprepius, who
had also opposed imperial monotheletism. 73
69
Eparchikon Biblion I (Trept T a P o v M a p i w v ) , 1 3 , 1 5 , 16. See also Klet. Phil. 1 1 3 . 1 8 ; A. Dain,
in REB 16 (1958), 166ff.; Noailles a n d Dain, Les Novelles de Leon VI le sage, p. 3 7 7 ; Bury,
Administrative System, p. 72.
70
See note 6 5 above. Bury believed t h a t the protokagkellarios represented the older princeps;
but this leaves t h e n n o r o o m for the origin of the logothetes tou praitoriou. The protokagkel-
larios is m o r e probably to be associated with the late R o m a n primiscrinius or subadiuva. See
Jones, LRE, vol. II, p. 5 8 7 .
71
See Oikonomides, Preseance, pp. 32If.; Bury, Administrative System, pp. 7 3 - 7 . Note the
important position of the quaestor as one of the imperial officials accompanying Constan-
tine IV at the sessions of the sixth ecumenical council (see, e.g., Mansi XI, 209B) and in the
prooemium to the Ecloga, w h e r e he a n d his antigrapheis (the older magistri sacrorum
scriniorum) are entrusted with the n e w codified selection a n d with the honest dispensation
of justice. See Ecloga, pr., 40sq. and 102sq.
72
Ecloga, prooemium, 40sq.
73
Mansi X, 8 5 7 B a n d E; PG CX, 104B a n d 172A. Cf. Devreesse, Hypomnesticum, 6 8 . 2 6 - 9 ;
7 2 . 2 6 - 7 3 . 2 ; 7 6 . 2 3 - 7 7 . 1 . For Theodore and Euprepios, see ibid., 7 1 . 1 7 - 1 8 .
 Society, state and law 275

In the provinces, things seem from the 640s to have been very much less
structured, that is, from the time when the Eastern armies are withdrawn
into Asia Minor and Arab attacks begin in earnest. Civil governors (archon-
tes) and judges (dikastai) are referred to in the Ecloga, although it is unclear
as to whether the latter are based in the provinces also; and the civil
governors were still clearly credited with the judicial authority they had
formerly held. 74 Courts are mentioned in a matter-of-fact way, too, which
suggests that the compilers of the Ecloga assumed the continued existence
of the provincial courts, both ordinary and appeal courts, within the
empire. 75 The fate of the civil courts of first instance, those of the defensores
civitatis, is unknown; but it is highly likely that, in view of the fate of the
municipalities and of civic culture generally during the seventh and eighth
centuries, they fell into abeyance as they became less and less relevant,
their functions being subsumed during the later sixth and seventh cen-
turies by the courts of the bishops. 76 Similarly, the nature of the relation-
ship between the episcopal courts, the courts of provincial governors and
the generals of the territorial themata (as these became fixed during the
later seventh and eighth centuries) remains obscure. There were clearly no
thematic kritai, who appear much later, at this time. That pertaining
between the first two presumably followed on from the earlier system with
little change (except in respect of the knowledge and understanding of the
law, and the degree to which local traditions and customs were used as a
prism through which to interpret the Justinianic corpus); and it may be
that the officials whom we have already encountered, the anthypatoi and
eparchs, responsible for the civil government of these provinces, retained
the judicial functions of the praetorian prefecture, representing up to the
end of the seventh century a higher, prefectural, court of appeal. 77
Initially, the strategoi, who represented the older magistri militum under a
different name, must likewise have exercised the jurisdiction over their
troops and officiales familiar from the sixth century. But as the themata
became territorially permanent, so the strategos seems to have gained a
general authority over the civil government of the provinces within his
thema. Thus the civil governors of the provinces, while they survived, will
have continued to exercise their functions under the authority of the
strategos, but within the jurisdiction also of the old prefecture which, as we
have seen in chapter 5, continued to function.
74
Ecloga VIII, 1. 6 (490); VHI, 3 (503); XVII, 5 (787); XVII, 21 (827) (archontes); Vffl, 3
(504); XIV, 1 (638); XIV, 4 (645); XIV, 7 (656, 6 6 0 etc.) (dikastai).
75
Ecloga IV, 4 (345) and XIV, 7 (658).
76
See Hohlweg, 'Bischof und Stadtherr', 5 Iff.; and esp. K.L. Noethlichs, 'Materialien zum
Bischofsbild aus den spatantiken Rechtsquellen', Jahrbuchfur Antike und Christentum 16
(1973), 2 8 - 5 9 .
77
See chapter 5, above.
276 Byzantium in the seventh century

Now it is important to remember that the posts of proconsul, or anthypa-

tos, and of the permanent, but originally ad hoc, praetorian prefect, or
prefect of the themata, came to be amalgamated at some point during the
later eighth or early ninth centuries; in other words, that the chief provin-
cial governorships and the permanent prefectural officials responsible for
military supplies were combined, continuing to fall within the jurisdic-
tional competence of what was by then a supervisory prefecture in Con-
stantinople, but coming under the general authority of their thematic
strategoi. The title of praetor, however, also survives into the ninth century,
for the praetors of the themata are mentioned in the Taktikon Uspenskij,
although lower in the hierarchy than the anthypatoi and eparchai.78 It
seems quite probable that, until the general reorganisation which seems to
have occurred shortly after 843 (the date at which the Taktikon Uspenskij
was compiled), the praetors of the themata represented the ordinary civil
provincial governors, with judicial as well as fiscal administrative func-
tions, just as the anthypatoi and eparchai represented the general (and
prefectural) civil and fiscal supervision of local administration. All will
have come under the remnants of the prefectural office in Constantinople,
as well as under the general authority of each specific thematic strategos.
And just as the fiscal administration seems to have been reorganised under
the protonotarioi, so the judicial administration of the themata seems to
have been reorganised - perhaps at the same time - under the krites (still
called praetor at times). 79 If this was the case - and the evidence does not
contradict such a conclusion - then it can be assumed that the organi-
sation and administration of justice in the provinces did carry on along
very much the same lines as in the late Roman period, with the exception
that the diocesan level of fiscal/judicial administration was replaced by the
thematic level - under the strategos, together with the anthypatos, the
permanent civil administrator of the groups oieparchiai that made up each
territorial thema - below which came the provincial praitores. As I have
argued in chapter 5, this picture remains hypothetical; but it does provide
a more adequate explanatory framework for the appearance of thematic
praetors, proconsuls and similar officials as late as the early ninth century.

THE FUNCTION OF THE LAW

The role played by the judicial system, and more particularly by the law at
a symbolic level, in seventh-century society and ideology was central in
the maintenance of a sense of Roman tradition and cultural identity. It was
78 r. Usp., 5 3 . 3 .
79
For the function a n d competence of the later krites, see Ahrweiler, 'Recherches', 67ff.;
Oikonomides, Preseance, pp. 323f.; a n d for krites-praitor, see Leo VI, Tactica, 7 0 5 .
 Society, state and law 277

also a crucial element in the maintenance of the Roman state itself, for it
was through the law that provincial governors and officials could invoke
and legitimate their authority, which flowed from the emperor and Con-
stantinople. That emperors realised this is apparent from the prooemium to
the Ecloga, where Leo and Constantine are concerned not simply with the
clarification of the traditional jurisprudence, but its relevance and applica-
tion to the provinces as well as to the capital city. The law embodied
implicitly the assumption that the highest judge on earth was the emperor,
who was himself chosen and protected by God; and the application of the
law meant at the same time accepting and furthering the political ideology
of the Roman state and all that this meant for contemporaries. Invoking
the juristic and legal tradition, therefore, whether in Constantinople or in
the furthest province, meant invoking also the political ideology of the
orthodox Roman state. 80 The fact that, during the period with which we
are concerned, the degree of ignorance or incompetence of local provincial
governors in legal matters was such that the emperors felt eventually
obliged to admit this state of affairs publicly and undertake action to
correct it, does not in itself mean that the ideological strength of provincial
government and the law was any weaker. As we have seen, 'the law' had a
symbolic, an evocational value in itself. Whether it was applied according
to the precepts and on the basis of the precedents of the Justinianic
legislation, or on the basis of provincial interpretations which may have
owed more to customary and traditional social practice, therefore, its force
when described as the law of the Roman state was just as powerful.81
Quite apart from this consideration, however, another important aspect
of judicial practice must be borne in mind. Modern legal systems, however
they developed historically, are founded upon the assumption of normative
prescription, precedents and interpretations, established by a judiciary
whose members share the same technical education. More importantly,
the judiciary strives to assert a normative framework within which civil
and criminal cases can be handled and within which the evidence of
witnesses, for example, can be processed on a consistent basis. Byzantine
judges did not work in this way. On the contrary, attempts to make sense
of the contradictions within Byzantine legal compilations (between appar-.
ently conflicting norms, for example) and especially between the formal
legislation and the actual practice and decisions of judges, where these are
known, have generally failed because of such assumptions. The enormous
complexity of the legislative literature available to Byzantine judges, the
corruption, the poor judicial training of judges, social and political vested
80
Ecloga, prooemium, 9 - 3 1 .
81
See esp. F. Dolger, 'Rom in der Gedankenwelt der Byzantiner', in Byzanz und die europdi-
sche Staatenwelt (Darmstadt 1964), pp. 70ff., esp. 75-6.
278 Byzantium in the seventh century

interests - all these have been invoked as grounds for the apparent
mismatch between legal theory and practice. 82 But as has now been
shown, these conclusions were flawed because based upon entirely false
assumptions. Byzantine judges worked within a widely respected norma-
tive judicial framework only at the most general level. They did not order
and interpret their case material within a pre-existing normative system
according to which the correct interpretation and the solution to a given
problem could be read off. Their activity was seen instead as determined by
moral-ethical considerations within a Christian framework, drawing upon
the accepted principles of an orthodox culture and the accepted 'common-
sense' understanding of the society as a whole. 83 Judges were selected
according to their general literacy, moral standing, and according to their
experience within the Church or state apparatus and appropriate adminis-
trative establishment. Judges were not, therefore, expected to fulfil their
obligations through applying the law, in the modern sense. On the con-
trary, the law they applied was the morality of the society - this replaced
the normative legal framework - interpreted through the prism of the
inherited legislation on an ad hoc basis according to the needs of the
particular case, the knowledge of the tradition of the judge in question, and
the prevailing moral climate, as well as the personal feelings of the
judge(s).84 Ultimately, therefore, formal justice in the Byzantine world
depended upon the moral preconceptions and ethical disposition of each
individual judge, whose decision would be grounded in a selection of topoi
drawn from the available legal literature. The decision would be legally
grounded in so far as all legislation was, ultimately, ascribed to the
emperors, who thus bore also the responsibility for its misapplication or
misinterpretation. Challenges to a legal decision were thus based on an
alternative and contradictory set of topoi, assembled to demonstrate the
fallibility of the first decision. The success of an appeal, however, rested
once more upon the personal morality and knowledge of the judges before
whom the appeal was heard.
That some generally applicable normative legislation did serve the
function of structuring social relations, of course, is apparent - marriage
law, for example, to name one of many such areas. But this still left
enormous scope for the moral personality of the judiciary. Even in the
imperial court, it was clearly ultimately the moral universe of the emperor
and his advisers which gave a particular slant or nuance to a decision,

82
See Pieler, 'Rechtsliteratur', pp. 346f.
83
The best discussion of this whole question is D. Simon, Rechtsfindung am byzantinischen
Reichsgericht (Frankfurt a. M. 1973), see pp. 18ff.
84
Pieler, 'Rechtsliteratur', p. 3 4 7 .
 Society, state and law 2 79
rather than a consciously regulated intention to arrive at a decision which
accorded in fact with the pre-given socially structuring framework.85
The breakdown of the traditional legalistic framework of the Roman
state in the seventh century, at least in respect of the education of the
judiciary and the accessibility of legal literature - while it may have
introduced a number of qualitative changes into the administration of
justice within the empire - does not seem to have had any dramatic effect
upon the survival of the state or indeed upon the availability of 'justice' to
the ordinary population. The ideological force of the concept of Roman law
will hardly have been affected, for Roman law did not symbolise an
abstract notion of justice alone, but rather the Roman state and the world
order which that concept evoked.86 'Justice' and the law were by no means
the same thing, of course, as is indeed the case in every state; and, as in the
fifth-century account of Priscus of Panium, so in the later period, it was
status, wealth and influence which usually ensured 'justice', rather than
the neutral objectivity, theoretical or actual, of the law and the judiciary.
The almost ritualistic complaints of emperors of corruption or bias in the
judiciary, repeated in the prooemium to the Ecloga, speak for themselves.87
Emperors who conscientiously interested themselves in justice were
marked out in the popular imagination, as the legend of the emperor
Theophilus demonstrates.88 The personal element in law-giving and the
application of justice is unmistakable, whether we are speaking of an
emperor or an ordinary governor or judge.89 None of this is to say, of
course, that there were not also judges who, within the cultural limitations
imposed by their world, did not strive for an 'objective' and as fair an
assessment of the cases they heard as was possible. But those cultural
limitations, rather than the individual, were the determining factor.
Roman-Byzantine law, therefore, and the legal-administrative appara-
tus which maintained it, plays a crucially ideological role in the Byzantine
world of the seventh and eighth centuries and after. The law was invoked
as a symbol of Romanitas, of continuity and Roman tradition, indeed as
confirmation of Roman orthodoxy and the role of the new Chosen People.
The maintenance of a judicial apparatus was essential for this to happen;
and the fact that Roman-Byzantine law experiences both a renaissance, of
85
Ibid., pp. 3 4 8 - 5 1 ; Simon, Rechtsfindung, pp. 1 5 - 3 2 .
86
Although this statement does not reflect the intention and the ideology of lawgiving and
justice - see, for example, Hunger, Prooemium, pp. 184f. Even in the sixth century,
however, it is very doubtful that the judiciary ever applied, or were able to apply, the
prescriptions of the codes systematically. See Scheltema, in CMH, vol. IV, part 2, pp. 7If.
87
Ecloga, prooemium, 5 2 - 6 8 . Cf. Jones, LRE, vol. I, pp. 516ff.
88
Ch. Diehl, 'La Legende de l'empereur Theophile', Seminarium Kondakovianum 4 (1931),
33ff.
89
Cf. Cecaumeni Strategicon, 8.1 lsq., and note the comments of Brehier, Institutions,
pp. 2 2 4 - 5 .
280 Byzantium in the seventh century
sorts, and a new vitality outside the Byzantine world from the ninth
century onwards, demonstrates not only the fact that the early Byzantine
state did manage, against remarkable odds, to restructure and reassess its
resources and their potential; it demonstrates above all the potency of the
law as a symbol of Roman power and claims to universal sovereignty
within the imperial ideology.
 CHAPTER 8

The imperial Church and the politics of

authority

THE CHURCH AS INSTITUTION

The Christian Church constituted one of the most powerful economic and
ideological institutions of the late Roman world, a role it continued to play
in the barbarian successor kingdoms and in the Byzantine empire
throughout its history. While there is no doubt that Christianity was not
universally accepted throughout the lands of the later Roman state -
enclaves of traditional, local non-Christian beliefs may have survived well
into the seventh and eighth centuries, for example - it was certainly by the
early seventh century the majority faith, in which traditional beliefs were
subsumed and through which older ways of seeing the world came to be
expressed. In theory, it represented a universalist soteriological belief
system, with a highly developed and sophisticated theological-philosophi-
cal arsenal at its disposal. In practice, while its formal debates, its day-to-
day teachings, and the political theory implicit in it, were represented by its
clergy and by the literate minority within society, as the single and correct
form of belief, it included also a plurality of ways of intepreting the world
inherited from the cultural traditions amongst which it grew up and
matured and upon which it was eventually superimposed. This syncretism
existed throughout the empire and throughout its history, although the
form of its expression and where it was expressed varied greatly. And while
the formal, sanctioned directions of belief expressed by members of the
orthodox establishment at any given moment (and depending upon which
form of 'orthodoxy' prevailed) tend to monopolise the theological and
political debates of the time, the underlying tensions out of which they
were refined must never be forgotten. This is particularly true in the
seventh century, for it was the period in which the contradictions which
eventually led to the iconoclastic policies of emperors like Leo III became
apparent, in which the possibility of 'pagan' survivals and heterodox belief
had to be confronted and in which theologians came into direct conflict

281
282 Byzantium in the seventh century

with the state and the emperor. And it was the latter conflict which
perhaps more than anything else in these years points to the real nature of
the political crisis within the Byzantine world.
Already in the fourth century, the crux of the matter had been summed
up by two Churchmen: in his debate with the North African Donatists,
Optatus of Milevis stated 'non enim respublica est in ecclesia, sed ecclesia in
re publica, id est in imperio Romano'; while Ambrosius of Milan pointed
out to the young Valentinian III: Imperator enim intra ecclesiam, non
supra ecclesiam est.' 1 These two statements represent two potentially quite
contradictory points of view, and they represent a tension between Church
and state, emperor and patriarch, which subsisted throughout Byzantine
history and was effectively resolved in the East only with the replacement
of the Byzantine emperor by an Ottoman sultan. In the seventh century,
they were to be invoked in a conflict which was, itself, to appear as
symbolic of these two poles of opinion for later generations.
Let us stress two key points at the beginning. First, it should never be
forgotten that the Byzantine Church, an organisational and doctrinal part
of a much wider Christian whole, is part and parcel of the history of the
Byzantine state and of Byzantine society and culture. Second, that its
development, and its relations with the state, are dynamic: there is no
static parallelism between emperor and patriarch, repeated in essence,
albeit concealed by the nuances of the conjunctural politics of a given
moment, throughout the course of Byzantine history. On the contrary, the
Church and its clergy need to be seen in the perspective of their own
society, responding and changing - or not - to the exigencies of temporary
crises or permanent changes in their circumstances. The Church repre-
sents, therefore, a fundamental element in the history of Byzantine culture
as a whole.
Perhaps the single most important determining feature of the Byzantine
Church was its close political-ideological relationship with the position of
emperor. These two institutions - priesthood and secular ruler - had been
tied theoretically by the development of an imperial, Christian ideological
system with its roots in the Roman and Hellenistic past, from the early
fourth century. In its most idealised form, it was expressed as a relationship
of mutual interdependence, and yet with a duty on the part of the secular
authority - the emperor - to defend 'correct belief and to protect the
interests of the spiritual authority. The Emperor Justinian I gives a vivid
statement of this position in a novel of the year 535:

1
Opatatus Milevitanus, Contra Parmenianum Donatistam, in CSEL, vol. XXVI (1893), iii, 3;
Ambrosius, Sermo de Basilicis Tradendis, or Contra Auxentium, in PL. vol. XVI, 875-1286,
see 1007-18.
 The imperial Church and the politics of authority 283

Mankind has been endowed by God from the mercy of heaven with two great
blessings, those of priest and emperor. The former ministers to matters divine; the
latter is set above, and shows diligence in, matters human. Both proceed from one
and the same source, and both adorn the life of Man. For this reason nothing lies
closer to the emperor's heart than honour and respect for the priestly office, while
the latter is bound to pray constantly for the emperor. For when this duty is carried
out without blemish and in true Godliness, and when in return the imperial power
accepts the application of its secular authority justly and morally, then there exists
harmony, and all mankind is blessed, and benefits.2
The passage continues with an expression of the conviction that the
health of the state can be assured only if the traditions of orthodox belief
are faithfully and correctly handed down and followed, a tradition
bestowed by the Apostles of Christ, and protected and preserved obviously
by the Fathers of the Church.
Clearly, this expression of the harmony of the secular and the spiritual
spheres was a Utopian statement. But it mirrors, although in a highly
idealised form, the aims of imperial policy with regard to both orthodoxy in
general and the fate of the empire, and to the ecclesiastical institutions and
personnel of the Byzantine world in particular. And it is demonstrated in
reality by the role of the emperor in the convening of synods and, more
particularly, in the way in which the decisions of such ecclesiastical
meetings were taken up in imperial legislation, where they received the
added force of being backed by the secular authority and its judicial
apparatus. From the time of Constantine I, the emperors had been involved
in both the politics and the theology of the Christian Church, and imperial
laws dealing with such matters ensured that, by the sixth century, emperor
and Church, state and Christianity, were inextricably bound together.
Justinian I presents in many respects the completion of this evolution, for it
was he who first set about formulating clearly the extent of, and limits to,
imperial authority in both secular and religious affairs. As God represented
the only source of law, he argued, so the emperor, who was chosen and
appointed by God to represent Him on earth, was the ultimate source of
law in the earthly sphere. 3 The exact extent of this imperial authority,

Justinian, Nov. 6, proem.

I have necessarily simplified a much more complex political-theological ideology. See
Haldon, 'Ideology and social change', 156f. and literature; and the survey of J.-M.
Sansterre, 'Eusebe de Cesaree et la naissance de la theorie "cesaropapiste"', B 42 (1972),
131-95 and 532-94. For Justinian's legislative statements, see esp. 0. Mazal, Die Prooi-
mien der byzantinischen Patriarchenurkunden (Vienna 1974), pp. 146ff., and note
W. Ensslin, 'Der Kaiser in der Spatantike', Historische Zeitschrift 177 (1954), 449-68, see
463ff.; see also Beck, Kirche, pp. 3 6-7, and esp. K. Aland, 'Kaiser und Kirche von Konstan-
tin bis Byzanz', in Kirchengeschichtliche Entwurfe (Gutersloh 1960), pp. 257-79 (repr. in Die
Kirche angesichts der konstantinischen Wende, pp. 42-86, cited here); with the contribution
of G.L. Kurbatov, in Z.V. Udal'cova, ed., Kul'tura Vizantii.
284 Byzantium in the seventh century

however, was an ill-defined area, which was to lie at the focal point of
Church-state relations during the seventh century.
The close ties, both ideologically and institutionally, between Church
and state, can be demonstrated in other spheres. Most notably, the role of
bishops in the civil administration, especially in respect of municipal
government and the regulation of revenue collection and distribution. As
we have seen, bishops played a central role in this area, and not just as
Churchmen - they were rather, by virtue of their position, regarded also as
part of a single establishment which was divided into two mutually
overlapping spheres, the spiritual and the material. There was nothing
anomalous from this standpoint.4
By the same token, the increasing liturgification of imperial ceremonial
and especially of the coronation5 - a real break with the antique traditions
of the Roman past - emphasised the officially sanctioned corporate identity
of the secular and the spiritual spheres, their interdependence and their
reciprocal influence. Increasingly, the Church within the East Roman
empire became the East Roman imperial Church - the two were initially by
no means the same. Church and state establishments worked together, and
together they represented the formally sanctioned institutions of govern-
ment. Faced with this monolithic concentration of authority, it is not
surprising that oppositional tendencies were represented through a rejec-
tion of these poles of authority and a search for alternative routes of access
to God and His spiritual authority from those of the emperor, the secular
Church and the power of the state and its apparatuses.6
The unresolved tensions in the Justinianic formulation were expressed in
practice through the weakest link in the chain, the relationship between
imperial and patriarchal authority in day-to-day terms. The 'harmony' of
Justinian's vision depended more or less entirely on the strength of per-
4
See chapter 3. A particularly striking example is that of the African Church after the
reconquest, where Justinian clearly intended to involve the Church very closely with the
rebuilding of an imperial administration and the reincorporation of North African society,
ideologically and politically, into the general framework of the empire. See esp. CJ I, 27 and
Justinian, Nov. 37; with the remarks of R. Markus, 'Reflections on religious dissent in North
Africa in the Byzantine Period', Studies in Church History in (1%6), 140-9.
5
Phocas' coronation in 602 marked the first such occasion actually in a church; coronati-
ons in 610 and 638 took place in the palatine chapel of St Stephen; and in 641 (and
thereafter), coronation occurred in Hagia Sophia. The development illustrates the general
tendency of late sixth- and early seventh-century cultural evolution, and the increasing
'exclusivism' of Christian imperial society; and it contrasts starkly with the militarised,
secular and still pre-Christian-dominated coronations of the previous rulers, even when the
patriarch had himself been directly involved. See Van Dieten, Geschichte der Patriarchen,
pp. 2f.; and esp. Treitinger, Die ostromische Kaiser- und Reichsidee, pp. 7ff.; W. Ensslin, Zur
Frage der ersten Kaiserkronung durch den Patriarchen und zur Bedeutung dieses Aktes im
Wahlzeremoniell (Wiirzburg 1948); the remarks of N. Svoronos, in REB 9 (1951), 125-9;
Brehier, Institutions, pp. 8-11 and 12f.
6
See Haldon, 'Ideology and social change', 148f. and 161ff.
 The imperial Church and the politics of authority 285
sonality of emperor and patriarch, together with the political and religious
situation of the moment, for no formal and clearly demarcated division of
competence was ever achieved. Emperors and clergy determined the limits
of their power, therefore, in accordance with the exigencies of the moment,
and with the abilities, strengths and weaknesses of those persons in power
with whom they had to deal, and in the context of the factional interests of
different power-groupings within the ruling and governing class.
Such tensions were most clearly evident when the state, in the person of
the emperor, intervened directly in matters of ecclesiastical jurisdiction or
of theology and dogma. This is not to say that they were not intellectually
qualified to do so - on the contrary, many emperors were competent
theologians in their own right. But they inevitably represented for the
Church more than a merely neutral interest. The degree to which the
church could tolerate such intervention depended, as we have seen, on the
context in which it took place, and on its perceived repercussions. Emper-
ors often found it easy to bully Churchmen into conformity, especially in
Constantinople, where patriarchs could be threatened, or removed and
replaced, when they failed to conform to the imperial line; although
actually resolving the questions at issue was not always so easily achieved.
The most coherent statement of the different competences of Church and
state, and one which was to be invoked frequently by churchmen in later
centuries, was that set out by Maximus Confessor during his trial and in his
correspondence. Maximus' argument was directed at one particular aspect
of the Church-state relationship, however. It was directed at the theoreti-
cal grounds upon which the emperors had attempted to justify their rights
of intervention, and of directing debate, in matters of dogma and the
nature of orthodoxy, grounds which - in the reign of Constans II at least -
were presented by the imperial party as elements within Christian theology
in the narrower sense. The central notion, the identity of the secular ruler
as both emperor and priest, represented a position strongly refuted by
Maximus, who queried likewise the emperor's claim to be the sole source
conferring authority on synods and councils - a claim which was founded
upon the model of Constantine I, but which was actually bound up with
the refusal of the state to recognise the Lateran synod in 649.7 While these
arguments were henceforth to be reproduced in similar confrontations in
later years, it was in the West, and particularly in the context of the
political authority of the papacy during the later eighth century and
afterwards that its importance for the development of the Church became
most apparent. For the Byzantine world in the seventh century, it was a
combination of perceived threats to imperial authority in the changed
7
Haldon, 'Ideology and social change', 173ff. with literature; Winkelmann, Die ostlichen
Kirchen, pp. 133ff.; Aland, 'Kaiser und Kirche' (cited note 3, chapter 8 above), 65-7.
286 Byzantium in the seventh century
political, military and social climate of the times, together with the lack of
any clear demarcation of spheres of influence and authority between
Church and state which lay at the root of the further development of both
the Byzantine Church and of the state itself.

Monophysites and dyophysites

The Constantinopolitan patriarchate, of course, was only one offivesuch
archiepiscopal sees: Rome, Alexandria, Antioch and Jerusalem each had
their apostolic traditions, and claims to equivalence and pre-eminence
played an important role in the history of their relations with one another.
But the patriarchate of Constantinople, by virtue of the loss of Alexandria,
Jerusalem and Antioch to the Arabs in the 630s and 640s, took on a
double significance for the development of Christianity after this time. On
the one hand, it continued to vie with Rome for ecumenical status and
pre-eminence, claims first expressed in the second council of Constantin-
ople held in 381, and always strongly rejected by Rome.8 On the other
hand, it became the patriarchate of the capital city of the empire, which
implicitly gave it an enhanced status in the Christian world generally, but
which most importantly made its final transformation into an imperial
Church inevitable. The disagreements with Rome thus received a very
different nuance, for Rome remained politically and militarily - in practice
if not in theory, and in spite of the few dramatic imperial interventions - an
independent authority. Divisions within the Byzantine Church, therefore,
which might otherwise have been resolved internally, could still be
referred to Rome; and the result was, predictably enough, a further
heightening of the built-in structural tensions within the Byzantine world
between state and Church, and between factions within the Church itself.
The loss of the Eastern provinces was, in the event, in many respects the
final solution to the intractable problems faced by the emperors, from the
time of the council of Chalcedon in 451 and afterwards, of imposing
orthodoxy (as it was there defined) on the increasingly recalcitrant and
resentful monophysite populations in Syria, Egypt and in parts of Armenia.
The establishment during the sixth century of independent monophysite
Churches in Syria and in Egypt provided a focus for anti-Chalcedonian
sentiment, and a framework through which the monophysite population
of town and countryside could resist the efforts of Church and state to
impose the Chalcedonian creed. From 543 the monophysite Jacob Bara-
daeus, as monophysite archbishop of Syria, pursued by the imperial
8
See especially canon 3 of the council of 381 (Mansi HI, 557f.) where Constantinople was
placed second in the hierarchy after Rome, as the 'New Rome', and before both Alexandria
and Antioch.
 The imperial Church and the politics of authority 287
authorities, established a monophysite 'shadow' Church. During the
period of Persian occupation it developed apace, so that by the time of
Heraclius'firstefforts at reconciliation (the synod of Hierapolis, 631) it was
strong enough to reject his overtures, even though its leaders seem initially
to have accepted a mononergite compromise formula.9 In Egypt matters
were complicated by competing factions within the monophysite camp.
While local cultural identity, language and traditions gave Egypt its own
very different and self-aware history, it was hardly surprising that the
rejection by the council of Chalcedon of the Alexandrian - monophysite -
position increased this feeling of difference, and that monophysitism took
on also thenceforth the role of vehicle for local, regional - rather than
separatist - expressions of independence. The varying fortunes of the
monophysite Church depended, as in Syria, on the strength and determi-
nation of the imperial authorities and their success in repressing anti-
Chalcedonian opposition. Many Syrians, indeed, took refuge in Egypt from
just such repressive measures, and in 575 the Jacobite exile community in
Egypt elected a Syrian archimandrite, Theodore, as anti-patriarch of Alex-
andria. The Coptic monophysites reacted by electing their own patriarch to
succeed Peter IV (557-78), although the Chalcedonian patriarch John II
was, of course, still in office. The matter was yet further complicated by the
fact that a theological difference of opinion developed between the Syrian
and the Egyptian monophysite communities as a result of the theology of
the new monophysite patriarch of Alexandria, Damian (578-605).
This schism between Syrian and Egyptian monophysites was healed
only in 616, when leading elements of the Egyptian landed elite and
monophysite Church realised that unity was in their best interests. Once
again, the Persian occupation of Egypt (from 619) could only favour the
monophysite cause, since the Chalcedonian - Melkite - clergy, as repre-
sentatives at the same time of the imperial Church, were forced to leave.
Even after the departure of the Persians the Chalcedonian Church, under
the imperial candidate for the see, Cyrus, a convinced supporter of the
compromise formulae of the patriarch Sergius and the emperor, failed to
recover the ground that had by then been lost. While the monophysite
clergy, as before, was forced out of the city into the countryside - where it
9
Van Dieten, Geschichte der Patriarchen, pp. 30ff. For the history and development of the
Syrian monophysite - Jacobite - Church, see especially R. Devreesse, Le Patriarcat d'Antio-
che depuis la paix de Veglise jusqu'd la conquete arabe (Paris 1945); D. Bundy, 'Jac°b
Baradaeus', Le Museon 91 (1978), 45-86; E. Honigmann, Eveques et eveches monophysites
d'Asie anterieure au Vle siecle (CSCO CXXVH, subsid. 17, Louvain 1951); W.A. Wigram, The
Separation of the Monophysites (London 1923). For a general account of political-theological
developments, see Beck, Kirche, pp. 288-91; Winkelmann, Die ostlichen Kirchen,
pp. 121-5, with sources and literature. On the regional and local feelings of independence
in the monophysite areas, see Guillou, Regionalisme, pp. 236ff., and see above, chapter 2.
288 Byzantium in the seventh century
took refuge in the monasteries and sketes of the hinterland - by the
imperial authorities, the Melkite Church remained for the most part dis-
tanced from the mass of the Coptic population. Only under archbishops of
great ability and diplomacy did the 'establishment' Church meet with any
success - the popularity of the Chalcedonian John III (610-19) in later
monophysite sources is indicative. With the arrival of the Arabs, the efforts
of the imperial Church to reassert its position finally ended.10
There is some evidence that the Chalcedonian clergy - including the
patriarchs - did continue to reside in their sees in the areas taken by Islam
from the 640s, attending the sixth ecumenical council and the Quinisext
council, although thefirstunequivocal evidence suggests that it was finally
only in 742 that the Chalcedonian patriarch of Antioch was allowed to
take iip his position, although the skeleton of a Melkite Church subsisted
throughout the intervening years; and it was only in 706 and 744 that
Chalcedonian patriarchs were elected once again in Jerusalem and Alexan-
dria respectively. Again, the Chalcedonian Church subsisted, electing
bishops and clergy and, when possible, sending representatives to the
general councils, although its history is difficult to discern.11
In what follows, I shall frequently refer to the 'Chalcedonian' or dyophy-
site theology which had become the imperial orthodoxy by the later sixth
century, in opposition to the monophysite theology of the Eastern
provinces. This must not be taken to imply that there were not monophy-
sites throughout the empire - merely that the majority tradition in the East
was, by the later fifth century, a monophysite tradition. In addition, we
should also speak more accurately of neo-Chalcedonian orthodoxy: from
10
Most recent general survey: Winkelmann, Die ostlichen Kirchen, pp. 114-21; also E.R.
Hardy, Christian Egypt (New York 1952); D.C. Miiller, 'Die koptische Kirche zwischen
Chalkedon und dem Arabereinmarsch', Zeitschrift fur Kirchengeschichte 75 (1964),
271-308; R. Remondon, 'L'Eglise dans la societe egyptienne a l'epoque byzantine',
Chronique d'Egypte 47, 93 (1972), 254ff.; M. Roncaglia, Histoire de Veglise copte (Beirut
1966); P. Verghese, Koptisches Christentum (Stuttgart 1973); and for the relations
between Greeks and Copts, the role of the Alexandrian patriarchate and the intervention
of the state, F. Winkelmann, 'Die Stellung Agyptens im ostromisch-byzantinischen Reich',
Graeco-Coptica. Griechen und Kopten im byzantinischen Agypten 48 (1984), 11-35, and
'Agypten und Byzanz vor den arabischen Eroberungen', BS 40 (1979), 161-82. For
Damianus and his followers, referred to as Damianites, as Angelitai and Tetraditai, see
I. Rochow, 'Zu einigen oppositionellen religiosen Stromungen', in Byzanz im 7. Jahrhun-
dert, pp. 225-88, see 264.
11
See the account of L. Brehier and R. Aigrain, in Fliche and Martin, Histoire de Veglise, vol.
V, pp. 479-84; Beck, Kirche, pp. 93-8; but for good circumstantial evidence that the
patriarchs of Antioch and Jerusalem were in their sees, at least in the period of truce
between empire and caliphate from 678 to 692, see F.R. Trombley, 'A note on the see of
Jerusalem and the synodal list of the sixth oecumenical council (680-681)', B 53 (1983),
632-8. Note also Brehier, Institutions, pp. 456-60; and especially H. Kennedy, 'The
Melkite Church from the Islamic conquest to the Crusades: continuity and adaptation in
the Byzantine legacy', Seventeenth International Byzantine Congress, Major Papers (New
York 1986), pp. 325^3.
 The imperial Church and the politics of authority 289
the late fifth century a division within the Chalcedonian camp developed.
This was concerned chiefly with the position within dyophysite theology of
the Antiochene tradition. The strict Chalcedonians wished to emphasise
the twin aspects of Christ's two natures, and to retain the rich and
sophisticated dyophysite tradition of the school of Antioch; the neo-Chalce-
donians, in contrast, were concerned to incorporate the tradition of Cyril of
Alexandria into Chalcedonian theology, and consequently to exclude its
antithesis, the Antiochene theology. The debate was eventually resolved,
as a result of direct imperial intervention through the affair of the Three
Chapters, in favour of the second direction. By the end of Justinian's reign,
neo-Chalcedonian theology dominated the dyophysite position.12

Wealth and administration

As we have said, the Church was a powerful economic force in the late
Roman and early Byzantine world, not simply in respect of its own power
and wealth, but also in the context of the economy of the state as a whole;
for the Church also absorbed and consumed surplus wealth on a large
scale, and thus directly affected the distribution of resources across the rest
of society. It has been estimated that the resources consumed by the
Church in the sixth century to support its charitable institutions, its clergy
and episcopate, its buildings and its public ceremonial, surpassed that of
the state and its administrative establishment (excluding the army) in both
wealth and manpower.13
The wealth of the Church lay for the most part, of course, in land,
although considerable amounts existed also in the form of precious plate
and - last but not least - in buildings themselves. From the time of
Constantine I the emperors had been the greatest benefactors of the
Church. Property was obtained through private grants and gifts, as well as
by donations from the state. It was protected by rulings on the inalienabi-
lity of Church property - established by Leo I and extended to the whole
empire by Justinian I - and by exemptions from certain forms of state tax or
imposition.14 Churches could receive property, in the form of willed lega-
cies and from intestate estates, and from the time of Theodosius II also from
individual members of the clergy who had no relatives entitled to inherit.
Gifts of money or produce became a regular feature of the income of the
12
For a brief summary, see Winkelmann, Die ostlichen Kirchen, pp. 58-62; Beck, Kirche,
pp. 372-84; G. Every, The Byzantine Patriarchate (451-1204) (London 1947), pp. 61ff.
13
Already Valentinian III had legislated to limit the flow of decurions and others into the
clergy; Justinian had limited the number of clergy attached to the Hagia Sophia to 525, a
figure raised by Heraclius in a novel of 612 to 600. See Winkelmann, Die ostlichen Kirchen,
pp. 137-8; and esp. Jones, LRE, vol. II. pp. 933-4.
14
See CJ I, 2.14 (a. 470) for Leo I; and Justinian, Nov. 7 (a. 535).
290 Byzantium in the seventh century

Church and its personnel, and Justinian even had to legislate to prevent the
clergy from demanding such voluntary donations as an automatic right.
The Church owned property in cities, too, and extracted wealth through
rent and through the sub-letting of estates on an emphyteutic basis. By the
time of Justinian, the Church was unable to manage all its property
adequately, and the rapid expansion of the use of emphyteutic leases was
the chief way in which this was made good. In practice, of course, the use
of perpetual emphyteutic leases meant ultimately the effective loss of the
estates or lands in question to secular landlords; but in spite of Justinian's
legislation to protect the notion of the absolute inalienability of Church
property, this type of lease continued to be the most effective way for the
Church to continue to extract at least a limited revenue from its properties.
But the fact of its widespread adoption by the Church throughout the
empire is indirectly good evidence for the considerable extent of the
property of the Church by this time. 15
The seventh century was a century of devastation, demographic dislo-
cation and decline, and of economic disruption for that part of the empire
which remained after the initial expansion of Muslim power and the loss of
the Eastern provinces. The extent to which the Church suffered in this,
along with other landowners - both private citizens and the state - is
difficult to say. But it has been plausibly suggested that the efforts of the
ecclesiastical authorities to make sure that the clergy did not abandon their
communities and sees - expressed most clearly in canon 18 of the Qui-
nisext council 16 - were intended to ensure that Church property continued
to be administered and, as far as was possible in the circumstances,
exploited economically.17
The history of the Church of Ravenna provides interesting parallels.
Although conditions were indeed more favourable to agriculture and
economic activity in Italy in the seventh century than they may have been
across much of Anatolia and the Balkans, the Church - unlike the secular
landowning elite, which seems to have taken refuge in Constantinople and
similar urban centres - was able to hold on to and even extend its landed
wealth; so that it emerges from the 'dark ages' as an even more substantial
15
On the economic organisation of the Church, and its property, see the general remarks in
CMH, vol. IV, part 2, pp. 118ff.; Kopstein, Zu den Agrarverhdltnissen, pp. 18-22; Brehier,
Institutions, pp. 518ff.; and esp. E. Wipszycka, Les Resources et les activites economiques des
eglises en Egypte du IVe au VIIF siecles (Brussels 1972); G.R. Marks, 'The Church of
Alexandria and the city's economic life in the sixth century', Speculum 28 (1953),
349-62; 0. Grashof, 'Die Gesetze der romischen Kaiser uber die Verwaltung und Veraus-
serung des kirchlichen Vermogens', Archiv fur katholisches Kirchenrecht 36 (1876),
193-203; E.F. Bruck, 'Kirchlich-soziales Erbrecht in Byzanz', in Studi in Onore di S.
Riccobono (3 vols. Rome and Palermo 1933ff.), vol. Ill, pp. 377-423; Beck, Kirche,
pp.65-7; Jones, LRE, vol. Ill, pp. 894-90.
16 17
Mansi, XI, 952B-C (canon 18). See Kopstein, Zu den Agrarverhdltnissen, p. 65.
 The imperial Church and the politics of authority 291

landowner than before.18 It is this institutional tenacity in maintaining

and reinforcing its hold on landed wealth which lies behind the above-
mentioned canon of the Quinisext and which explains why the Church,
unlike individual private landowners, was able to conserve its position
where the latter were, more often than not, unable to do so with the same
degree of success. What is clear is that, after the dislocation of the seventh
and eighth centuries, the Church emerges still as the greatest institutional
landowner next to the state.
One of the most significant aspects of Church and monastic finances, of
course, was their charitable foundations. Since the beginning, the Chris-
tian communities had devoted a part of their income to charitable pur-
poses. From the early fourth century the revenues from Church endow-
ments could also be employed for such purposes, and the care of the sick,
the elderly, of orphans and widows, along with the poor in general,
became an important part of the pastoral work of the Church. As the
available wealth grew, so orphanages, hospitals, almshouses and so on
were endowed and built, maintained from the rents of Church lands,
sometimes endowed also by a wealthy lay person. Such institutions often
had their own lands, from which they could extract rents; others were
maintained by monastic foundations, either as separate institutions or as
part of the monastery itself. Still others were built and maintained by the
state, and the imperial hostels in Constantinople provide illustration of
this. 19 Under Justinian, care for the needy in society and a tutelary concern
for society in general - both spiritual and physical well-being - were
officially recognised, indeed promoted, aspects of the activities of the
Church. 20 Such activity lent the Church both moral authority and political
power - the eleemosynary activities of John the Merciful, patriarch of
Alexandria from 610 to 619, for example, won the respect of monophysite
as well as Chalcedonian communities and commentators, and indeed of
the local officials of the state. 21
The organisation and administration of the Church followed the pattern
of late Roman administrative institutions. Below the patriarchs were
metropolitans, autocephalous archbishops and bishops. The metropolitan
was the senior bishop in each province, appointed by the patriarch.
Bishops were elected by the provincial synods, although until the reign of
18
See Guillou, Regionalisme, esp. pp. 181-7 and 232f.
19
See Jones, LRE, vol. II, p. 901; and H.R. Hagemann, 'Die rechtliche Stellung der christli-
chen Wohltatigkeitsanstalten in der ostlichen Reichshalfte', Revue Int. des Droits ant, 3
ser., 3 (1956), 265-83; Brehier, Institutions, pp.524-6; and esp. the account in D. Savra-
mis, Zur Soziologie des byzantinischen Monchtums (Leiden and Cologne 1962), pp. 25-38
with sources and further literature.
20
See CJ I, 2.23 (a. 530) and Justinian, Nov. 65 (a. 538).
21
See Vita Joh. Eleemosyn., 8.16sq.
292 Byzantium in the seventh century

Justinian, the ordinary clergy and the citizens of the diocese also partici-
pated in the nomination of candidates for the election. Justinian curtailed
these rights, and they were later restricted to the provincial bishops alone,
although popular intervention - acclamations, for example, in support of
one or other of the candidates - still took place. The bishop was the chief
ecclesiastical authority in his diocese, and both clergy and monasteries
were under his control. He was responsible for the maintenance of ortho-
doxy, the seeking out and destruction of heresy, and the imposition and
application of the canon law of the Church. He was also the chief manager
of Church properties, although in practice this was delegated to local
managers and stewards - oikonomoi - and he supervised the distribution of
Church revenues to his clergy and to Church foundations such as
almshouses and orphanages. In respect of the clergy in his diocese he also
presided over the ecclesiastical court and arbitrated at cases between
laypersons and clergymen. From the reign of Heraclius (from 629) the
privileges of the clergy in such legal proceedings were explicitly safe-
guarded, to prevent their being unjustly treated by the civil or military
authorities. But one area in which conflict continued to arise - until the
eleventh century, at least - was that of matrimonial law, where the civil
law of the state encoded in the Codex lustinianus and later collections, based
as it was on traditional Roman law, was sometimes in conflict with canon
law. Matrimonial cases were dealt with by the civil authorities, therefore,
since civil marriages could be dissolved by the agreement of both parties,
the legal question of the redistribution of property presenting the most
difficult problems. According to the Church, in contrast, marriage was
indissoluble (except in the case of adultery), and the conflicts which ensued
- for example, the question of the fourth marriage of the Emperor Leo VI -
were ultimately resolved only when the state ceded complete jurisdiction in
such matters to the Church, with a ruling of the Emperor Alexius I in
1084. 22
The bishop, of course, played a role not merely in the ecclesiastical
organisation of his diocese, but in its economic and social life, too. From
the sixth century, bishops were among the most important figures in
municipal government, being entitled to sit also in civil courts and carry
out civil-administrative functions. Bishops were, certainly at this time,
drawn from among the social and economic elite of their cities or dioceses,
22
See Dolger, Regesten, no. 1116. For the organisation of the Church and the relative
positions of the patriarch, bishops, upper and lower clergy, rural and urban parishes, see
esp. the survey in Beck, Kirche, pp. 67ff. and 79-86; CMH vol. IV, part 2, pp. 106-18;
Jones, LRE, vol. II, pp. 874ff.; Brehier, Institutions, pp. 477ff.; and G. Dagron, 'Le Christia-
nisme dans la ville byzantine', DOP 31 (1977), 3-25 (repr. in La Romanite chretienne en
Orient (London 1984)), see pp. 19ff. On the question of matrimony and divorce, see Beck,
Kirche, pp. 86-90.
 The imperial Church and the politics of authority 293

and it is unlikely that this picture changed dramatically during the seventh
century. 23 As such, they formed also an important element of the ruling
class of late Roman society, in which they occupied a central position,
acting as mediators of both the ideological interests of Church and state
and the economic interests of the Church as a great landowner. 24

The monasteries
Just as significant an element in the late Roman and early Byzantine world
was the monastic community. During the period from the early fourth to
the later sixth centuries, monasticism - confined originally to Egypt, Syria
and parts of Palestine - experienced a dramatic expansion. 25 In spite of
strenuous efforts on the part of the secular Church and the state to
establish some form of control and authority over both the monastic
communities themselves and their tendency to expand which was an
important element in the decline of the urban economy and culture of the
late Roman period, monasteries represented a source of independence and
anti-authoritarianism to the regular establishment, and more particularly
to unacceptable or novel departures in imperial policy with regard to the
Church and dogma. In 451 the council of Chalcedon reached a number of
decisions in respect of monastic property, the position of monks and
monastic communities within the Church as an institutional body, deci-
sions which were taken up and expanded by the Emperor Justinian I. The
continued independence of such communities, however, and their tacit
rejection of many such limitations on their size, their activities and their
sources of recruitment is evidenced in the canons of the Quinisext, where
the Church once again attempted to enforce some degree of generally
recognised conformity.26 Individual hermits and anchorites, 'holy men' of
23
See the survey of A. Guillou, 'L'Eveque d a n s la societe mediterraneenne des VI e -VIP
siecles: u n modele', Bibliotheque de Vecole des Chartes 1 3 1 (1973), 5 - 1 9 (repr. in Culture et
societe en Italie byzantine (VIe-XIe siecles) II (London 1978).
24
See chapter 3; a n d Jones, LRE, vol. II, pp. 923ff.; Brown, Gentlemen and Officers, p. 176.
Note also the c o m m e n t s and literature cited by F. Winkelmann, 'Kirche u n d Gesellschaft
in Byzanz v o m ende des 6. bis zum Beginn des 8. J a h r h u n d e r t s ' , Klio 59 (1977), 4 7 7 - 8 9 ,
see 4 8 Iff. For a comprehensive general survey, see Beck, Kirche, pp. 1 2 0 - 4 0 .
25
See CMH, vol. IV, part 2, pp. 161ff. and 167ff.; and especially the discussion in Mango,
Byzantium: the Empire of New Rome, pp. 1 0 5 - 2 4 . On the types of monastic community, see
also Beck, Kirche, pp. 120ff. with literature. On the expansion of monasticism, see Jones,
LRE, vol. II, pp. 930ff.; Brehier, Institutions, pp. 529ff.
26
For Chalcedon a n d Justinian, see Beck, Kirche, p. 126; a n d esp. B. Granic, 'Die rechtliche
Stellung u n d Organisation der griechischen Kloster n a c h dem justinianischen Recht', BZ
2 9 (1929), 6 - 3 4 a n d 'Die privatrechtliche Stellung der griechischen Monche im V. u n d
VI. J a h r h u n d e r t ' , BZ 3 0 (1930), 6 6 9 - 7 6 ; G. Dagron, 'Les Moines et la ville; le m o n a c h i s m e
a Constantinople j u s q u ' a u concile de Chalcedoine (451)', TM 4 (1970), 2 2 9 - 7 6 , for a
detailed analysis of the role of m o n k s in ecclesiastical a n d imperial politics in the fourth
and fifth centuries. For the Quinisext, see c a n o n s 4 0 to 4 9 : Mansi XI, 939ff.
294 Byzantium in the seventh century

the type familiar from the late Roman period, remained even more difficult
to control and continued to exercise considerable influence and authority
among the ordinary rural population. 27
Like the secular Church, monasteries could also own property, granted
to them in the same ways. The efforts of the Church to retain some control
over monastic foundations gave to bishops a pre-eminent role. Not only did
bishops have to give their assent to the building of any new monastic
establishment; monasteries were also obliged to pay them a regular episco-
pal tax, or canonicum, and to commemorate the local bishop in the liturgy.
The bishop also supervised the election of the abbot of a monastery and
consecrated him. Needless to say, these regulations were not always
strictly observed, and conflicts between monasteries and the episcopate
were not infrequent.28 Monasteries were endowed by private persons,
from the richest to the poorest, for the salvation of their souls - for which
the monks waged a continuous battle in prayer - and by grants of property
and wealth from the imperial government. The extent of monastic property
is difficult to assess, although it must from early on have been considerable,
to judge from the number of monasteries which existed in some regions of
the empire: there were, over several centuries, some 300 establishments in
Constantinople alone, endowed either with land or other property from
which the monks could be supported, or given cash grants and provisions
on a yearly basis from their patrons and/or founders. Monastic centres
flourished in the Aegean area, in western Asia Minor, in Cappadocia, in the
Pontus, in parts of the Balkans, to name only those areas still within the
empire after the seventh century. 29
Monasticism, beginning as an Egyptian phenomenon, reflects the degree
of popular piety among the ordinary populations of the late Roman and
Byzantine world, and its often turbulent history provides an accurate

27
See the comments of H.-G. Beck, Das byzantinische Jahrtausend (Munich 1978), esp.
pp. 214ff. The council of Chalcedon (canons 4 and 24) had ordered that all those fol-
lowing the monastic life should have a fixed abode, and that they should confine their
activities to fasting and prayer. They should not intervene in politics or in the wider affairs
of the Church. In his fifth novel, Justinian ordered that even anchorites and recluses
should live in their own cells, but within the bounds of the monastery. But such rulings
clearly had little effect at this early time and afterwards, until the later ninth century at
least. For Justinian's legislation, see CJ I, 3.43 and 46; V, 3.7 and 133. For monks and
'holy men' in the seventh century, see below.
28
See Mango, Byzantium: The Empire of New Rome, pp. 108ff. and 120ff.
29
See especially P. Charanis, 'The monastic properties and the state in the Byzantine
empire', DO? 4 (1948), 51-118, and 'The monk as an element of Byzantine society', DO?
25 (1971), 61-84; Savramis, Zur Soziologie des byzantinischen Monchtums, esp. pp. 39ff.
and 45-52; Brehier, Institutions, pp. 553ff.; Beck, Das byzantinische Jahrtausend,
pp. 214-15. See also the brief survey of F. Trombley, 'Monastic foundations in sixth-
century Anatolia and their role in the social and economic life of the countryside', Greek
Orthodox Theol. Review 30 (1985), 45-59.
 The imperial Church and the politics of authority 295
measure of the intensity of feeling aroused within the Byzantine Christian
world over key issues - the Christological debates of thefifthto the seventh
centuries, iconoclasm, and so on. It is notable that both Maximus Con-
fessor and Sophronius had spent their early years in monastic communi-
ties or as anchorites; and it is important to remember that it was just such
people - men and women both - who demonstrated their piety and
nearness to God by their ability to endure physical and emotional degra-
dation and humiliation, gaining thereby a more direct and better felt
access to the true source of the holy and who also fulfilled the half-pious,
half-superstitious needs of ordinary people of all social strata in their
day-to-day difficulties and personal problems, constituting thereby an
alternative source of spiritual authority which was implicitly a challenge
to the formally endowed authority of the secular clergy and the estab-
lishment Church.30
On the other hand, calculations of the size of monastic communities and
the numbers of monks within the empire have often been absurdly exag-
gerated, and on very little evidence, to give the impression that the late
Roman and Byzantine state was at times almost overrun with such
monastic establishments and that vast revenues were lost to the fisc as a
result of wide-ranging exemptions and other privileges granted by pious
rulers to calculating abbots and monastic patrons. In fact, while the
number of monks may at times have been considerable, and while the
amount of monastic property may at times have attracted the attention of
the state - one thinks of the policies of the Emperor Nicephorus II in the
960s - there was never the numerical superabundance of monks which
this tendency implies. On the contrary, very many people adopted the
monastic life only in their last years and after a secular career; others left
the monastic life for the secular world once more; while, as at least one
scholar has stressed, the greater the number of monks, the greater their
variety and the more diverse and dilute their effects.31
The extent to which monastic property suffered during the seventh
century is impossible to assess. One estimate has suggested that the basic
pattern, at least as regards Constantinople and the less exposed areas of
Asia Minor, remained essentially the same.32 No doubt monasteries in
exposed districts were abandoned, but their property - especially in land -
could hardly be destroyed, and since it remained inalienable, it may have
been taken over by the local diocese, where this survived administratively,
30
See Haldon, 'Ideology and social change', 1 8 4 with literature; and P. Brown, 'The rise and
function of the holy man in late Antiquity', JRS 6 1 (1971), 80ff. and 'A Dark-Age crisis:
aspects of the iconoclastic controversy', EHR 3 4 6 (1973), 1 - 3 4 , see 23ff.
31
Charanis, 'The monk as an element of Byzantine society', 6 9 .
32
See Beck, Das byzantinische Jahrtausend, pp. 2 0 7 - 1 2 ; and in general on monasticism in its
social, cultural and institutional effects, pp. 2 1 2 - 3 1 .
296 Byzantium in the seventh century

or they were reclaimed at a later date. Monks do play an important role,

even as refugees - the problems they caused for the North African Chris-
tians after the Arab conquest of Egypt, when large numbers of monophy-
site monks and nuns arrived, to the dismay of the local clergy, are well
known. 33 Monks figure prominently in the events connected with the
monothelete debate from the mid-63Os on; 34 they seem often from the
earliest times - a reflection (in theory, if not always in practice) of their
more rigorous life and greater piety, which gave them a greater spiritual
authority - to have acted as 'pressure-groups' on the secular clergy and
the episcopate, partly, of course, through the greater respect afforded them
by the ordinary people of their society and their ability consequently to
influence the latter in one direction or another. Monks represented in the
seventh century, as often before and in later years, the hardline and
uncompromising element in the Christian community, standing up against
imperial Interference' in matters of the spirit. They play a central role in
the cultural-ideological politics of seventh-century Byzantium; and from
the early eighth century they begin to supply the Church with an increas-
ing number of leading clergymen, both patriarchs and bishops.35

Through its administrative and economic activities and organisation,

through its clergy, its ecclesiastical courts, its role of missionary, preacher
and defender of orthodoxy, the Church in its widest sense, both monastic
and secular, was intimately bound up with every aspect of later Roman
and Byzantine society and culture. While it is to a degree inevitable that it
should be studied and examined as a separate institution or set of institu-
tions, within the framework of Byzantine society, it should nevertheless not
be forgotten that it was itself a structured and structuring element of that
society, in the history and development of which it played a fundamental
role.
The relationship between imperial Church and imperial state was not,
therefore, a fixed one. Social and political developments affected the struc-
ture of this relationship as they affected all other aspects of early Byzantine
society. The political ideology of Church and state incorporated ancient
traditions - Hellenistic, Roman, oriental and early Christian - each of
which was invoked at any given time and in a particular situation in a
particular way. The Church represented the implementation of a formal
33
See below; a n d Van Dieten, Geschichte der Patriarchen, pp. 67ff.
34
Both Sophronius a n d Maximus h a d begun their careers as monks; a large n u m b e r of their
followers likewise.
35
See esp. L. Brehier, 'Le Recrutement des patriarches de Constantinople pendant la periode
byzantine', Actes du Vle Congres Internationale des Etudes Byzantines, vol. I (Paris 1956),
pp. 221-7. The first patriarch from monastic circles was Cyrus (705-11) - see Van
Dieten, Geschichte der Patriarchen, pp. 16Iff.
 The imperial Church and the politics of authority 297

and sophisticated theology, yet in a social and cultural context - a

symbolic universe - in which different elements clearly had differential
effects and in which heterodoxy and the cultural assumptions of a variety
of pre-Christian cultural traditions played a central role. Christianity
represented not just the formal political theology of the Greco-Roman state,
it represented also a regional, local and personal-individual system of
values, with its appropriate logic and explanatory efficacy.
Conflicts within Christianity, and the continued existence of varying
degrees of explicitly non-Christian belief- 'paganism' - give the impression
of an uneasy pluralism until the middle of the sixth century. From Justi-
nian's reign, neo-Chalcedonian universalism and a single orthodoxy start
to become the keynote values of early Byzantine society and the state. The
marginalisation and exclusion of non-conforming groups becomes
increasingly apparent under Justinian's successors. The unity of state and
Church, the future salvation of the oikoumene, depends upon their partner-
ship in orthodoxy. To be different was dangerous. The long evolution of the
imperial Church which began in the fourth century reaches its final stages
in the reign of Justinian; thereafter, state and Church are inseparable. As
the patriarch Antony IV puts it in his letter to Prince Basil I of Moscow in
the last decade of the fourteenth century:
My son, you are wrong in saying 'We have a Church, but not an emperor.' It is not
possible for Christians to have a Church and not an empire. Church and empire
have a great unity and community; nor is it possible for them to be separated one
from the other.36
Much had changed in the intervening centuries, of course. But the
political ideology which this statement represents was the result of the
developments of the period from the fourth to the sixth centuries. The
seventh century confirmed and finalised this evolution.

CHURCH AND STATE C. 6 I O - 7 1 7 : AUTHORITY VS.

LEGITIMATION

The origins of the monothelete conflict

Under the influence of his wife, the Empress Sophia, Justin II began his
reign with an attempt to bridge yet again the gulf between Chalcedonian
and monophysite Churches. Sophia, a niece of the Empress Theodora, had
been brought up in the tradition of Severus of Antioch, the great monophy-

36
The text is in F. Miklosich and J. Miiiler, Acta et Diplomata Graeci Medii Aevi Sacra et Profana
I and II {Acta Patriarchatus Constantinopolitani) (Vienna 1860-2), vol. II, pp. 188-92; Engl.
trans. E. Barker, Social and Political Thought in Byzantium (Oxford 1957), pp. 194-6.
298 Byzantium in the seventh century

site theologian of the first half of the sixth century. 37 Her influence over
Justin became clear when he recalled and restored all the bishops exiled by
Justinian I, and when in 566/7 he issued a document of unification, or
Henotikon, in which the ecclesiastical policies of Justinian - with the
important exception of the policies invoked and imposed during the Three
Chapters controversy - were more or less entirely abandoned and in which
a return to the position at the end of the reign of Anastasius (491-518)
was evident. Unfortunately for Justin's initial efforts, this attempt was a
failure, for the monophysite bishops who met at Kallinikon in 567 to
debate and ratify the policy under the auspices of the imperial representa-
tive John in fact rejected it, since it did not manage actually to condemn
dyophysitism, that is, the creed of Chalcedon.38 In 571 a second Henotikon
was promulgated in which Justin modified his position very slightly,
stressing the fact that, however the question of the single or double natures
of Christ (as man and God) might be resolved, He still occupied a single
person and being, both God and man in one. In the same document, Justin
also repeats and extends the basic position expressed in Justinian I's edict
on correct belief, issued in 551. In practice, therefore, whatever his
intentions or his hopes, the official position would be seen by the monophy-
sites as one based fundamentally on Chalcedonian principles, in which two
natures could still be admitted. 39 The result was, perhaps, predictable, and
certainly typical: monophysite rejection of the principle of the Henotikon,
followed by intensified state persecution. The majority of the monophysite
bishops were forced to put their names to the document; those who refused
were exiled.40 The issue, therefore, remained unresolved, and the indepen-
dent Jacobite Church organised by Jacob Baradaeus continued to devel-
op. 41 Justin's short reign, which began with an attempt at compromise
and unification, ended in persecution and repression.
Tiberius II and Maurice appear to have acted much more pragmatically,
indeed the monophysite Church during Tiberius' reign seems to have been
able to consolidate its organisational foundation, and monophysite sources

37
See esp. W. de Vries, 'Die Eschatologie des Severus von Antiochien', OCP 23 (1957),
354-80; J. Lebon, Le Monophysisme Severien (Louvain 1909).
38
See Michael Syr., vol. II, 289ff.; and Averil Cameron, 'The early religious policies of Justin
IT, Studies in Church History XIII (1976), 65ff. For Zeno's Henotikon, see Every, The
Byzantine Patriarchate, pp. 50f.; Winkelmann, Die ostlichen Kirchen, pp. 97f.
39
See the long account in Evagrius, 199-200; and see, for the Justinianic position,
E. Schwartz, 'Drei dogmatische Schriften Justinians', Abhandlungen d. bayer. Akad. d.
Wiss., phil.-hist. Klasse, new series XVm (Munich 1939). For a brief survey of Justinian's
religious policies, see Every, The Byzantine Patriarchate, pp. 57-68.
40
See Michael Syr., vol. II, 295ff.; John of Nikiu, 94; and especially John of Ephesus, 1,19sq.;
II, lsq.; II, 9sq.
41
For the most useful general account, see W.H.C. Frend, The Rise of the Monophysite
Movement (Cambridge 1972); and Winkelmann, Die ostlichen Kirchen, pp. 122ff.
 The imperial Church and the politics of authority 299
speak well of the emperor, who put an end, temporarily, to the persecu-
tions.42 But under Maurice, the policies of repression were once more
applied, although their effect was again merely to drive the monophysite
Church underground, but not to reduce its influence nor to damage its
organisation.43 And under Phocas the situation seems to have remained
very much the same.
In contrast to the last years of the reign of Maurice, Phocas once more
established friendly relations with the papacy and Rome. In the 590s the
disagreement between the patriarch at Constantinople and the Pope in
Rome over the former's use of the title 'ecumenical' resurfaced. Since
Maurice took no action in the matter, his inactivity was assumed by Rome
to signal his approval of the patriarch's position, and relations were
considerably soured as a result. In contrast, Phocas recognised Roman
authority over the whole Church, and in 607 he issued a decree in which
he expressly forbade the Constantinopolitan use of the title ecumenical,
recognising instead the supremacy of Rome.44 Again, this question was to
become one of the distinguishing features of later Byzantine ecclesiastical
history in its relations with Rome and with the West in general.
The central question for the state, however, remained that of the schism
between dyophysite Chalcedonians and the monophysites of the Eastern
provinces. During Heraclius' wars with the Persians, the question natur-
ally enough was relegated to a secondary position. Imperial Church and
state worked together to restore the empire and to re-establish orthodoxy
in the civilised world. The enormous loans made by the Church to finance
Heraclius' armies, the central role of the patriarch Sergius during Hera-
clius' absence in the East, and especially during the great siege of 62 6,45 all
point to the close collaboration between the spiritual and political estab-
lishments of the Byzantine world. But in the provinces temporarily occu-
42
Michael Syr., vol. II, 310.
43
Although the repression was not as severe as it had formerly been - see Stratos, Byzantium
in the Seventh Century, vol. I, p. 13. See R. Paret, in REB 15 (1957), 42ff.; and note the
remarks of Michael Syr., vol. II, 372. In general on Church-state relations in this period,
see I. Rochow, 'Die Heidenprozesse unter Tiberios II. und Maurikios', in Studien zum 7.
Jahrhundert, pp. 120-30; Winkelmann, 'Kirche und Gesellschaft' 477ff.
44
For Phocas' policies in respect of the monophysites, see John of Nikiu, chapter 104, for
example; and Brehier and Aigrain, pp. 71-2. On the Rome-Constantinople debate and the
question of ecclesiastical supremacy, see Ostrogorsky, Geschichte, p. 71; V. Laurent, 'Le
Titre de patriarche cecumenique et la signature patriarcale', REB 6 (1948), 5-26; Beck,
Kirche, pp. 63f. with literature; Brehier and Aigrain, pp. 64ff. and 7 0 - 1 . For a brief
summary of Byzantine-papal relations during the seventh century, see F. Dvornik, Byzanz
und der romische Primat (Stuttgart 1966 = repr. of Byzance et la primaute romaine (Paris,
1964)), pp. 95-108; and Herrin, Formation of Christendom, pp.206ff., 213-15, 2 5 0 - 6 7
and 274ff. For Phocas, ibid., pp. 180f.
45
See the account in Stratos, Byzantium in the Seventh Century, vol. I, pp. 126f.; Brehier and
Aigrain, p. 86. For the siege, see F. Barisic, 'Le Siege de Constantinople par les Avares et les
Slaves en 626', B 24 (1954), 371-95; Van Dieten, Xkschichte der Patriarchen, pp. 12-21.
300 Byzantium in the seventh century

pied by the Persians, things were very different. For here, the removal of
Byzantine authority had permitted the open establishment and strengthen-
ing of the hitherto illegal monophysite Church. Whether or not Persian
rule was regarded favourably, it certainly encouraged the self-confidence
of the monophysite clergy and people; and upon the completion of the
reconquest, it became evident that throughout the Eastern provinces -
Egypt, Syria, Palestine and Armenia in particular - a potential or actual
hostility to, and a resentment of, Constantinopolitan rule now constituted
a real danger, both to the unity of the empire politically and ideologically,
and to the authority of the emperor and of the Chalcedonian Church. 46
In an effort to promote a reconciliation, the patriarch Sergius, the
emperor and Cyrus, Chalcedonian patriarch of Alexandria from 631,
adopted the 'doctrine of the single energy', known as monenergism. The
real originator of this theological solution had been a certain Theodore,
bishop of Pharan, towards the end of the sixth century, and the principal
aim of his theology was to overcome the contradiction between the
monophysite and dyophysite perspectives by pointing to the unity of effect
of the natures of Christ, that is, the single energy which emanates from the
logos. At the same time, he emphasised also the single will which was an
inevitable corollary of the single energy. 47 The debate over one or two
energies was thus in origin a purely theological issue reaching back into
the sixth century arising out of the Christological debate around the
question of the natures of Christ and expressed in the sophisticated vocabu-
lary of Christian thinkers. And it was the position outlined in the writings
of Theodore of Pharan that Sergius, Cyrus and Heraclius now tried to build
upon, but with only limited success. Even though Heraclius himself led the
discussion and negotiations, the monophysites could not be persuaded that
this was not still essentially a dyophysite position; and in part they were
correct, for Theodore of Pharan had grounded his theology in a dyophysite
46
For the opening stages of the development of an imperial compromise solution, see the
summary and literature in Stratos, Byzantium in the Seventh Century, vol. I, pp. 283-304;
Van Dieten, Geschichte der Patriarchen, pp. 24ff. For Persian policy in the monophysite
provinces, see Brehier and Aigrain, pp. 88-90 and 11 If. Heraclius was clearly already
well aware of the nature of the problem: his Edict on Faith of 610, promulgated shortly
after" his accession, was firmly dyophysite, but was couched in terms intended to be
acceptable to a monophysite reader, including a formulation of Cyril of Alexandria. See
Michael Syr., vol. II, 4 0 2 - 3 ; Brehier and Aigrain, pp. 86-7. The 'monophysite danger'
should not, of course, be exaggerated: see chapter 2, and J. Moorhead, The Monophysite
response to the Arab invasions', B 51 (1981), 579-91.
47
See the summary of the complex argument in W. Elert, Der Ausgang der altkirchlichen
Christologie (Berlin 1957), pp. 2O3ff. For the role of Sergius, see Brehier and Aigrain,
pp. 112ff.; and the agreement of 633, ibid., pp. 117-18; Van Dieten, Geschichte der
Patriarchen, pp. 2 4 - 3 1 ; Herrin, Formation of Christendom, pp. 2 0 6 - 1 1 ; and for the sources
for and beginnings of the debate, see Winkelmann, Die Quellen, esp. nos. 1-11, and
pp. 55f.
 The imperial Church and the politics of authority 301

framework, and his argument inevitably allowed room for the possibility
that there might be two natures. In spite of an agreement reached between
Cyrus and the so-called Theodosiani, a monophysite sect in Egypt, formally
announced in a document of union in June 633, no real resolution of the
problem was reached. Pope Honorius, in a carefully expressed letter, made
his agreement with the discussions for unity known; but the patriarch
Sophronius of Jerusalem rejected the attempt out of hand, and his oppo-
sition - from afirmlydyophysite position - meant in effect the failure of the
whole exercise, for it was announced in a local synod, held in 634 shortly
after his accession to the patriarchal throne. The failure of thisfirstattempt
was made all the more apparent by its rejection from the Chalcedonian
side. The rejection from Sophronius produced an edict in late 634 or early
635 from the emperor, supported by Cyrus and Sergius, to the effect that
the question of the number of energies should no longer be debated at all,
since union had now been achieved and since such debates would serve
only to weaken the empire further. But the monophysites seem simply to
have ignored the efforts of the emperor and his advisers. They were, far
from being reunited with the Chalcedonian, now in the position of
observers of a new schism within the ranks of the latter. And as first Syria
(636), then Palestine (638) and finally Egypt (642) fell to the Arabs, the
attempt lost much of its relevance for them, both from the theological and
the political points of view. 48
For the second time in his reign Heraclius now saw the Eastern
provinces of the empire lost to an invader. In 638, perceiving that the
monenergite doctrine had failed, he made a second attempt to attract the
loyalty and support of the monophysites, but without stirring up Chalcedo-
nian opposition, he hoped, and issued the Ekthesis, in which a monothelete
doctrine was proclaimed, a doctrine which emphasised the single will of
God, but which left the question of nature and energy to one side. Once
again the debate on the question of the number of energies was strictly
prohibited.49 The Ekthesis was posted in the narthex of Haghia Sophia in

48
Pope Honorius: Brehier and Aigrain, pp. 121-3; andMansiXI, 537-44; Riedinger, 548-58;
Sophronius' opposition: Brehier and Aigrain, pp. 118,120-1 and 12 3-4; and his synodal letter,
Mansi XI, 831-53. See also Van Dieten, Geschichte der Patriarchen, pp. 32ff. For the edict of
634-5 and its background, stimulated by a letter of the Pope to Sophronius, suggesting
that the latter permit Honorius to handle the discussion personally with Sergius, see
Dolger, Regesten, no. 205; Van Dieten, Geschichte der Patriarchen, pp. 39ff.; Brehier and
Aigrain, pp. 123-4.
49
Dolger, Regesten, no. 211; V. Grumel, Les Regestes des Actes du patharcat de Constantinople I:
Les Actes des patriarches, vol. I (Paris 1972); vol. II (Chalcedon 1936); vol. II (Chalcedon
and Bucarest 1947), vol. I, no. 292; Brehier and Aigrain, pp. 131-2; Van Dieten,
Geschichte der Patriarchen, pp.47ff. For the text, see Lateran, 156.20-162.13 (Mansi X,
991-8); and for general historical context, Stratos, Byzantium in the Seventh Century, vol.
II, pp. 141-9; and Van Dieten, Geschichte der Patriarchen, pp. 179-218 and 219-232.
302 Byzantium in the seventh century

Constantinople, and copies were sent to the patriarch Cyrus in Alexandria,

and to the exarch of Ravenna for the successor to Pope Honorius, Sever-
inus. In Constantinople Sergius convoked a council to approve the docu-
ment and the statement of faith, but then he died in December 638. His
successor Pyrrhus called a new council after his accession in January 639.
All the attending bishops ratified the document. The new patriarch of
Jerusalem, the monothelete Sergius of Joppa, also ratified it; while the
Chalcedonian patriarch of Antioch, Macedonius, appointed shortly before
by Sergius, stayed in Constantinople rather than occupy his see, and
likewise ratified the Ekthesis.50
In Rome, meanwhile, Pope Honorius had died in October 638, and his
successor Severinus seems to have come under direct pressure from the
exarch to ratify the Ekthesis. In addition, his apocrisiaries were held up in
Constantinople until they agreed to accept the document on the pope's
behalf. This they refused to do, although they eventually agreed to per-
suade Severinus himself to accept it. The new pope was finally confirmed in
his office by Constantinople in April of 640, but died shortly after in August
of the same year. He never ratified the Ekthesis. His successor, John IV,
seems to have been confirmed in his election relatively quickly, but
immediately adopted a hostile position, calling a synod in Rome which
rejected and condemned monotheletism and the Ekthesis.51 Shortly before
his death in April 641 Heraclius realised that his policies had failed to
achieve their aims, indeed that they had created new divisions within the
Church. In a letter to John IV he attributed the original authorship of the
Ekthesis to Sergius and conceded also that it had brought further trouble to
the Christian community. 52 His death did not end the matter, however,
and it was in the reign of Constans II in particular that the divisions within
Church and empire became critical. But monotheletism was itself not
simply an ineffective and short-lived imperial compromise. Indeed, it seems
from the later evidence - notably the adherence to monotheletism of the
patriarch Makarios of Antioch and his supporters at the sixth ecumenical

50
See Brehier and Aigrain, pp. 132f.; Van Dieten, Geschichte der Patriarchen, pp. 47f. and
5 8 - 6 3 ; Stratos, Byzantium in the Seventh Century, vol. II, pp. 143f.; Grumel, Regestes, vol.
I, nos. 2 9 3 , 2 9 5 and 2 9 8 .
51
Brehier and Aigrain, pp. 133f.; Van Dieten, Geschichte der Patriarchen, pp. 6 2 - 3 ; Dolger,
Regesten, no. 214; Stratos, Byzantium in the Seventh Century, vol. n, pp. 146-7; Ostro-
gorsky, Geschichte, pp. 9 0 - 1 . The split in the Chalcedonian community was ultimately
founded upon much deeper cultural differences between East and West, of course, than
these ecclesiastical-political disagreements suggest. See the discussion of P.Lemerle,
'L'Orthodoxie byzantine et l'ecumenisme medievale: les origines du "Schisma" des
eglises', Bulletin de VAssociation Guillaume Bude, 4,2 (1965), 2 2 8 - 4 6 (repr. in Essais sur le
monde byzantin VIII (London 1980). See also Herrin, Formation of Christendom,
pp. 213-15.
52
See Dolger, Regesten, n o . 2 1 5 .
 The imperial Church and the politics of authority 303

council of 680/1 - that the monothelete doctrine was adopted, and

probably on a widespread basis, in the neo-Chalcedonian communities of
Syria and Palestine, where it appears to have taken firm root. The Syriac
Life of Maximus Confessor in particular suggests that he and his followers
did not find the massive support in the East which their own propaganda
at the height of the controversy with the emperor and afterwards might
suggest. On the contrary, it was only in Africa and in Rome that Maximus
was really able to muster vocal and committed opposition to monothele-
tism. The explanation for this support, and the stubbornness of the Chalce-
donian Syrian communities in adhering to the doctrine, must lie partly in
the fact that from the 640s they were effectively cut off from the political
body of the empire: one clear way of denoting their identity, for themselves
and especially for their monophysite neighbours, was by sticking firmly to
the official policy of the state to which they had once belonged - and might
still belong again. 53
While Heraclius' policies regarding the Church are dominated naturally
enough by the great question of the relationship between Chalcedonian
and monophysite Churches, the emperor concerned himself also with the
less public matters of ecclesiastical administration, the organisation and
regulation of monastic communities, and the Church hierarchy. Decrees
issued in 612 and in 619 to regulate the numbers of the clergy in positions
in the Great Church in the capital and in the Church of the Blachernai and
others; and ordinances in 629 granting juridical privileges to the clergy,
and in 638, on matters concerning both the Constantinopolitan and
provincial clergy, are illustrative of this activity.54 He was also very much
concerned with the question of the Jews in the empire, particularly in view
of the frequent collaboration of Jewish communities with the Persians.
While initially granting an amnesty for their actions (issued at Tiberias in
630), he seems thereafter to have hardened his views, forbidding Jews to
53
For Makarios, see below. For the Syriac Life of Maximus, see S. Brock, 'An Early Syriac Life
of Maximus the Confessor', AB 9 1 (1973), 2 9 9 - 3 4 6 , dated to the later years of the
seventh century; and the comments of Winkelmann, Die Quellen, 5 5 2 and 5 5 7 - 9 on the
propaganda effort of Maximus and his followers.
54
See Dolger, Regesten, nos. 165, 175, 199, 2 1 2 and 2 1 3 ; Grumel, Regestes, vol. I, nos. 2 7 8
bis, 2 7 9 bis (with vol. Ill, 196); Van Dieten, Geschichte der Patriarchen, pp. 3 - 5 , 10 and
2 2 - 3 for discussion. Heraclius' genikos nomos of 6 2 9 is particularly significant, for it
marks a new stage in the privileges of the clergy and in Church-state relations. According
to the decree, no layperson, including military officials, had henceforth the right to
imprison any clerical person. Clergy w h o travelled to Constantinople to have their pleas
heard now had the right to choose which court they wished to appear before - that of the
praetorian prefect, that of the patriarch or of a specially delegated imperial judge. Lay
courts no longer had any jurisdiction to alter the decisions of the Church courts; but for
clerics w h o lay charges against non-clerics, the traditional and standard rule - that the
plaintiff must accept the decision of the court under whose authority the defendant comes,
w a s maintained. See Beck, Kirche, p. 8 2 .
304 Byzantium in the seventh century

live within three miles of Jerusalem after the return of the True Cross.55
There had on occasion been forced baptisms of Jews also, but in 634 he
issued a general decree ordering the baptism of all Jews in the empire.
Needless to say, such measures had little hope of even a limited success,
and the loss of Syria and then Palestine to the Arabs by 638 made the
decree irrelevant for these areas. But forced baptisms did occur in other
parts of the empire, as the story of Jacob, 'the recently baptised', testifies.56
If anything, of course, the decree made the Jewish communities in Pal-
estine even more hostile to Roman rule and will have ensured, if not
co-operation, then at least neutrality in the struggle against the Muslim
invaders.
One further event of significance deserves mention, if only because of the
consequences for the state after Heraclius' death and the attitude of later
Byzantine writers to him: his marriage to his niece Martina. Heraclius' first
wife, Eudocia, died on 13 August in the year 612 and was laid to rest in the
Church of the Holy Apostles.57 Shortly afterwards, the emperor decided to
marry his niece, Martina. The Church clearly saw this marriage as con-
trary to canon law, and the patriarch Sergius tried to convince the
emperor that his actions would also make him very unpopular. Heraclius
replied politely, we are told, that Sergius was correct and had done his
duty; but that he could leave the matter in the emperor's hands
thenceforth. The marriage went ahead, indeed with the patriarch's bless-
ing; 58 but although the temporary disagreement between Heraclius and
Sergius was quickly forgotten in the troubles that followed - the emperor
suffered a major military defeat in 613 at Persian hands; and in 614
Jerusalem fell and the Holy Cross was carried off to Persia - it remained a
point of contention among both clergy and population, especially in
Constantinople, where Martina was apparently very unpopular. 59

The crisis of authority: Constans II vs. Maximus Confessor

Heraclius was succeeded by his eldest son Heraclius Constantine and by his
first son with Martina, known as Heracleonas, who were to rule jointly
according to his will. 60 In effect, Heraclius Constantine, usually referred to

55
Dolger, Regesten, nos. 1 9 6 and 1 9 7 .
56
Dolger, Regesten, no. 2 0 6 ; and the account of Brehier and Aigrain, pp. 110f.; see also
Beck, Kirche, pp. 3 3 2 - 3 and note 1.
57
See Chronicon Paschale, 702.19sq.; Ostrogorsky, Geschichte, p. 9 3 .
58
See Pernice, L'imperatore Eraclio, pp. 54f.; a n d Nicephorus, 14.11sq.; Theophanes,
3 0 . 2 5 - 8 . For Sergius' baptism of the first child of the marriage, Constantine, in 6 1 3 , see
Theophanes, 3 0 . 6 - 7 . See Grumel, Regestes, vol. I, no. 2 8 4 .
59
Nicephorus, 2 7 ; and see Ostrogorsky, Geschichte, p. 9 3 .
60
Dolger, Regesten, no. 2 1 6 .
 The imperial Church and the politics of authority 305

simply as Constantine, ruled alone at first, although Martina attempted to

promote her own position and that of her son. Her unpopularity, however,
and the fact that the patriarch Pyrrhus was accused of illicitly appropriat-
ing state funds for Martina's cause, together with the support of the senate
and the army for the 'true' line of Heraclius and - in addition - Con-
stantine's apparently anti-monothelete position 61 meant that she had to be
content with a background role. Pyrrhus lost temporarily the influential
position he had held over the throne. But Constantine died after only a few
months, and the young Heracleonas, supported by his mother and the
patriarch, became sole ruler. 62 The rule of Martina as regent, backed by
Pyrrhus, did not last long. One of her first acts was to rescind an order of
Constantine to drive out, or to convert, the monophysite refugees in Africa
(although her order was never carried out - the exarch Gregory, realising
the potential unrest this would cause among the orthodox Africans,
declared the decree a falsification and ignored it). 63 Martina's lack of
support in the senate and her unpopularity in the city, as well as the
opposition of the commander of the troops in Anatolia, Valentinus, forced
her and Pyrrhus to arrange for the crowning of the son of Heraclius
Constantine, Constantine, or Constans, as he is normally known, as
co-emperor. This compromise did not hold, however, and as rumours
spread that Martina and Pyrrhus were plotting with the commander of one
of the Anatolian field armies - possibly of the Armenian forces - to march
against Valentinus and assist in the deposition of Constans, the senate
ordered Martina and her offspring to be arrested, mutilated and banished.
Pyrrhus managed to flee to Carthage, from where he was eventually
recalled to office.64
After the turbulent events of 640-1 the young Constans II commenced
sole rule on 9 November 641. Within a few months the patriarchal throne
had been filled by the newly elected Paul II, who remained in office until
653. He was a monothelete and, although the young emperor and the
senate accepted the letter of John IV to Heraclius Constantine and recog-

61
See the text of Pope John VFs letter to Constantine, Mansi X, 682-6, and Van Dieten,
Geschichte der Patriarchen, pp. 63-4 and note 22; and the strongly anti-monophysite order
of the emperor to the bishops and exarch of Carthage in response to their request to deal
with the refugee nuns spreading monophysite ideas in the exarchate. See Dolger, Regesten,
no. 222; Brehier and Aigrain, pp. 160ff.; Van Dieten, Geschichte der Patriarchen, pp. 67ff.;
Stratos, Byzantium in the Seventh Century, vol. II, pp. 176-85; Ostrogorsky, Geschichte,
p. 94.
62
See Van Dieten, Geschichte der Patriarchen, pp. 6 3 - 8 ; Brehier and Aigrain, pp. 143-^i.
63
Van Dieten, Geschichte der Patriarchen, pp. 6 8 - 9 and note 37; Stratos, Byzantium in the
Seventh Century, vol. Ill, pp. 59ff.
64
See Van Dieten, Geschichte der Patriarchen, pp. 7 0 - 5 for a summary; and chapter 2 above,
pp. 49ff.
306 Byzantium in the seventh century

nised the Chalcedonian creed,65 his appointment suggests the existence of

powerful monothelete elements in the senate and at court, who were the
effective government during the early years of Constans' minority. The
general Valentinus, with a strong detachment of troops, had also estab-
lished himself in Constantinople, however, and seems to have attempted to
make himself officially regent or co-emperor. His attempt failed, chiefly as a
result of the efforts of the patriarch Paul, and he was killed when the
populace rioted against his use of soldiers brought into the city from the
field army.
The rift between Rome and Constantinople was not overcome, however,
by the imperial recognition of two natures and two wills, for the Pope
Theodore insisted also on a formal and canonical deposition of the patri-
arch Pyrrhus 66 and on the removal of the Ekthesis from the narthex of
Haghia Sophia. Initially, there was no response to these demands, perhaps
a sign of uncertainty in the factions clustered around the young emperor.
The debate did not rest, however. In 643 the archbishop Sergius of Cyprus
wrote to Pope Theodore affirming his orthodoxy and that of the island;
while in 645 the famous disputation between the monk Maximus and the
ex-patriarch Pyrrhus took place - organised by the exarch Gregory himself
in Carthage - in which Pyrrhus conceded the debate, travelled to Rome to
confess his sins and received the pope's forgiveness and blessing. Maximus
travelled with him. 67
Maximus is perhaps one of the most important individuals about whom
events revolve at this period. He was a confirmed Chalcedonian and
anti-monothelete, a keen-witted theologian and a brilliant debater. Like his
older contemporary Sophronius, also a fierce opponent of monotheletism
from the Chalcedonian position, Maximus spent his early years as a monk
in Palestine, where he came into contact with such men as John Moschus,
compiler of one of the most important collections of edifying tales and
extracts from the Lives of the desert holy men and monks, the so-called
Spiritual Meadow. As refugees from Islam, both Maximus and Sophronius
were active in the struggle against monophysitism and against the 'new'
heresy, monotheletism. It was Maximus who wrote around to the ortho-
dox communities of the Mediterranean world enlisting their support and
promoting their opposition to the imperial policy; it was Maximus who
became the effective leader of the African Church after his defeat of Pyrrhus

65
Dolger, Regesten, n o . 2 2 1 ; Brehier a n d Aigrain, pp. 162f.; Van Dieten, Geschichte der
Patriarchen, pp. 76f.
66
Van Dieten, Geschichte der Patriarchen, pp. 8 0 - 2 .
67
Ibid., pp. 8 2 - 7 ; Cameron, 'Byzantine Africa', 56ff.
 The imperial Church and the politics of authority 307

in the debate of 645; and it was Maximus who was the guiding hand
behind the convoking of the Lateran synod of 649. 6 8
As a result of the defeat of Pyrrhus, the African Church held a number of
local synods in the following year, all of which condemned monotheletism.
The bishops of Byzacena, Numidia and Mauretania directed two letters, via
Rome, to Constantinople, one to the patriarch and one to the emperor,
appealing to the latter to invoke his imperial authority and to compel Paul
to return to orthodoxy. 69 The degree of suspicion of the African Church in
Constantinople - clearly expressed in letters of Victor, bishop of Carthage,
and of other African bishops to Rome - was, of course, greatly exacerbated
by the exarch Gregory's rebellion, which began soon after Maximus'
victory over Pyrrhus. 70 The logic of Gregory's action is not entirely clear,
but it is clearly connected with the results for the African Church and
population of the defeat of Pyrrhus, the reaffirmation of orthodoxy, and the
conflicting options open to him: either to follow his Constantinopolitan
orders and enforce imperial policy - that would have entailed a massive
persecution and political repression, which might well have seemed an
impossible task - or to maintain his position and authority (and his popu-
larity), throw in his lot with the orthodox sentiments of the greater part of
the population and all the episcopate and, by refusing to follow orders, call
down upon himself the wrath of the imperial government. Gregory decided
evidently on the latter course, pre-empting the court's response by formally
stating his rejection of the rule of the Emperor Constans. 71
Gregory's death in battle with the Arabs in 647 brought a dramatic
change in the situation. Pyrrhus - who had been anathematised by Paul
following his rejection of monotheletism and who had perhaps hoped for a
patriarchal throne in Carthage beside the pretender Gregory72 - now
changed his position once more, returned to monotheletism and claimed
that his 'conversion' had been obtained under duress. Pope Theodore
68
See Cameron, 'Byzantine Africa', 56f.; Haldon, Ideology and social change', 1 7 3 and note
8 5 ; Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 60f. For the Lateran council, see
the extensive works of R. Riedinger, listed in Winkelmann, Die Quellen, 5 3 8 . On Maximus
himself, see esp. J.M. Garrigues, Maxime le confesseur. La charite, avenir divin de Yhomme
(Paris 1976), esp. pp. 35-75 for a detailed biography; and the papers in F. Heinzer and
Chr. Schonborn, Maximus Confessor. Actes du Symposium sur Maxime le Confesseur.
Fribourg 2 - 5 Sept. 1980 (Fribourg 1 9 8 2 ) .
69
See Vita Maximi Confessoris, 84B; Theophanes, 3 3 7 . 8 - 1 0 ; Cameron, 'Byzantine Africa',
57; Stratos, Byzantium in the Seventh Century, vol. in, pp. 6 1 - 2 .
70
See the literature a n d discussion in Van Dieten, Geschichte der Patriarchen, pp. 8 5 - 7 . It is
uncertain w h e t h e r t h e letters were sent, however, a l t h o u g h the pope, Theodore, certainly
used t h e m in his o w n letter to the patriarch Paul in 6 4 7 . See below.
71
Cf. Cameron, 'Byzantine Africa', 57f.; Stratos, Byzantium in the Seventh Century, vol. IE,
pp. 6 2 - 7 ; Ostrogorsky, Geschichte, pp. 98f.
72
Van Dieten, Geschichte der Patriarchen, pp. 87; 8 4 - 5 .
308 Byzantium in the seventh century

immediately anathematised him and wrote a strongly worded letter to the

patriarch Paul in which he intimated that the political loyalty of the West
and Africa depended directly on the imperial government and the Con-
stantinoplitan patriarchate returning to a dyothelete/dyophysite position.
The consequence for the East of accepting this ultimatum, of course, would
be - apart from the political defeat for the emperor - a de facto recognition
of Rome's precedence in the ecclesiastical hierarchy and in matters of
dogma. Paul replied in a gentler tone, affirming quite explicitly his and the
imperial government's adherence to monotheletism and the formulation of
the patriarch Sergius, invoking also the Pope Honorius' claimed adherence
to this doctrine. 73 The pope answered quite simply by formally pronounc-
ing the deposition of the patriarch Paul. 74
This unilateral act immediately changed the tone of the disagreement.
Up until this point, the debate had been restricted to matters of faith and
confined to the Church. The imperial government had anyway been unable
to respond effectively to Gregory's rebellion owing to its difficult situation
in the East, and the problems of government by a council of advisers and
the still minor Constans. Gregory's death had ended that particular threat,
and the state seems to have been willing to remain silent while Paul carried
on the debate - although the clear division within the Church, and
especially the alienation of Africa, must have been very worrying. The
pope's letter of 647 and his deposition of Paul changed all this. The first
result was the issue in 648 of the Typos, clearly designed less to resolve the
theological problem than to quell the uproar which this had effected and to
restore order within the Christian community - for there were other, yet
more pressing dangers to be faced.75 Paul may well have been closely
involved in its composition, but it was issued in the emperor's name and
thereby took the debate on to a different level. Already in his letter of 647
the Pope had, by implication, threatened the state; his deposition of Paul
was now a direct challenge to imperial authority. The emperor decreed
that those who refused to comply with the demands of the Typos were to be
severely punished: bishops would be removed from office, monks from
their sanctuaries; civil and military personnel would lose their office and
73
Paul's letter: Lateran, 1 9 6 . 1 6 - 2 0 4 . 4 . Theodore's letter h a s n o t survived, but its tone a n d
content c a n be re-established from Paul's reply. Note that Paul spent some time with the
papal emissaries to Constantinople trying to reach some sort of compromise solution,
without success, before m a k i n g his position clear in his letter. See Lateran, 198.22sq.;
Liber Pontiflcalis I, 3 3 3 ; cf. Grumel, Regestes, vol. I, n o . 3 0 0 .
74
Lateran, 1 8 . 8 - 1 9 (Pope Martin's account at t h e Lateran council in 6 4 9 ) .
75
See Dolger, Regesten, n o . 2 2 5 ; text: Lateran, 2 0 8 . 3 - 2 1 0 . 1 5 ; see Haldon, 'Ideology a n d
social change', 173f., a n d Van Dieten, Geschichte der Patriarchen, pp. 92ff.; Ostrogorsky,
Geschichte, p. 99; E. Caspar, 'Die Lateransynode von 649', Zeitschrift fur Kirchengeschichte
51 (1932), 75-137; Herrin, Formation of Christendom, pp. 217-9 and 250-5. See also the
references to the work of Riedinger in note 68, chapter 8, above.
 The imperial Church and the politics of authority 309

their titles, senators and others would lose their lands, yet others would be
chastised with corporal punishment and with imprisonment. And since the
Typos forbade all debate, ordering simply that the orthodox should observe
the canons of the first five ecumenical councils - thus leaving the question
of the number of wills and energies open - the Ekthesis was also removed,
thus complying with one of the earlier demands of the Roman primate. 76
The Roman response, suggested and encouraged by Maximus, was not
long in coming. 77 Pope Theodore commissioned the fiercely orthodox
Stephen of Dor, spiritual comrade of Sophronius of Jerusalem, to extract a
confession of faith from the clergy of Palestine. Those who refused to
co-operate or remained monothelete were to be deposed. Such action was,
of course, a direct interference in the affairs of the eastern patriarchates
and a direct challenge to imperial policy, and resulted partly from the
appeal of the monothelete Sergius of Joppa, who had succeeded Sophro-
nius as patriarch of Jerusalem, together with the bishops whom he had
appointed, to the patriarch Paul for confirmation of their appointments. 78
The imperial authorities responded by depriving the papal emissaries of
their priestly titles and position, punishing them physically and exiling
them from Constantinople.79 The new Pope Martin - elected on 5 July
649, but without imperial confirmation - held a synod in the Lateran
palace, a synod which was in fact organised by and dominated by
Maximus and Stephen of Dor and their associates. The originators of
monotheletism - Theodore of Pharan, Cyrus of Alexandria, the patriarchs
Sergius, Pyrrhus and Paul, together with the Ekthesis of Heraclius and the
Typos of Constans, were all anathematised. The acts of the council, again
written up by Maximus and his associates, were then sent to Constans,
together with a letter urging him once more to compel his patriarch to
return to orthodoxy. The emperor himself, along with his grandfather
Heraclius, was carefully excluded from the anathematisations and the
condemnations, almost certainly a political gesture rather than a reflection
of real sentiment. 80
The events that followed are well known. The new exarch of Ravenna,
76
See the s u m m a r y of Van Dieten, Geschichte der Patriarchen, pp. 9 3 - 4 ; Stratos, Byzantium in
the Seventh Century, vol. Ill, pp. 95ff.
77
It is not clear from the sources w h e t h e r Pope Theodore or his successor Martin received
the Typos. Theodore died o n 13 May 6 4 9 , and since Martin's accession w a s never
confirmed by the emperor, it is probable t h a t the d o c u m e n t arrived in time for Theodore to
see it. See Van Dieten, Geschichte der Patriarchen, pp. 9 2 - 3 and note 7 3 .
78
Lateran, 46.1sq. For Sergius of Joppa, see Van Dieten, Geschichte der Patriarchen, p. 50.
79
Cf. Lateran, 1 8 . 8 - 3 1 ; Liber Pontificalis, 3 3 6 ; Hypomnesticum, 70 (Greek version) a n d 196A
(Latin version).
80
For the text, see Riedinger, Lateran. See Van Dieten, Geschichte der Patriarchen, pp. 97ff.;
Haldon, 'Ideology and social change', 1 7 4 a n d note 8 8 ; Stratos, Byzantium in the Seventh
Century, vol. Ill, pp. 9 8 - 1 0 4 ; Ostrogorsky, Geschichte, p. 9 9 .
310 Byzantium in the seventh century

Olympius, had already been ordered to arrest the pope, who had ascended
the papal throne without imperial ratification, and to have the Typos read
out in all the churches of Italy. Instead, Olympius came to an arrangement
with Martin and ruled independently, ignoring his orders, just as Gregory
had done in Africa. Only with the death of Olympius in 652 was the
emperor able, through the new exarch, Theodore Calliopas, to rectify the
situation. 81 Martin was arrested by Calliopas with contingents of the
Ravenna army and taken to Constantinople where he was tried, initially
condemned to death for high treason, then to exile. He died in Cherson on
16 September 655. Maximus was also arrested and taken to Constantin-
ople. His trial, banishment, second trial or interrogation, followed by his
mutilation and exile - he died eventually in Lazica in 662 8 2 - accurately
reflects the anger at his obstinate and intellectually sharp defence of his
position and also the main issue at stake: imperial authority. Indeed, it is
interesting to observe how, throughout the debate, and from the earliest
days of the conflict in 645/6 through to the final execution of Maximus'
punishment, the government and Constans were eager to come to a
peaceful compromise, an arrangement that would involve nothing more
than the end of the debate, rather than any admission of fault on the part of
either Pope Theodore or his successor Martin, or of Maximus. In the last
resort, and with the secular demands of state politics firmly in view, the
defeat of the anti-monotheletes was, perhaps, a foregone conclusion. The
emperor had brutally reasserted imperial rights to be directly and centrally
involved in both matters of faith, in the calling of synods and in the
ratification of higher ecclesiastical appointments. This success, however,
was bought at the cost of the alienation, even if only temporary, of Africa,
which remained stubbornly anti-monothelete and, as its later history
illustrates, unable or unwilling to offer any dynamic opposition - political
or cultural - to the arrival of Islam. 83

81
Ostrogorsky, Geschichte, pp. 99-100; Haldon, 'Ideology and social change', 182 and note
123; Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 105-11. There is some
numismatic evidence to suggest that Martin did approve and bless Olympius' rebellion: a
silver coin struck in the mint of Rome and dated to the years 651-2 by those who have
examined it bears an effigy which is not that of Constans II and is probably to be identified
with Olympius. See M.D. O'Hara, 'A find of Byzantine silver from the mint of Rome for the
period A.D. 641-752', Swiss Numismatic Review 65 (1985), see no. 7, type 3, and
'Numismatic evidence for the treason of Pope Martin (A.D. 649-654)', in Fourteenth
Annual Byzantine Studies Conference, Abstracts of Papers (Houston, Texas 1988), 53.
82
See Van Dieten, Geschichte der Patriarchen, p. 1 0 1 ; Ostrogorsky, Geschichte, p. 100; Brehier
and Aigrain, pp. 170ff.; Stratos, Byzantium in the Seventh Century, vol. HI, pp. 1 1 2 - 2 5 .
83
See Brehier a n d Aigrain, ibid.; Van Dieten, Geschichte der Patriarchen, pp. lOlf. and 107ff.;
Haldon, Ideology a n d social change', 1 7 3 - 7 ; Cameron, 'Byzantine Africa', 57f.; o n the
question of synods a n d their legitimacy, see also, in addition to the literature cited by these
authorities, V. Peri, 'I concili ecumenici come struttura portante della gerarchia ecclesia-
stica', in 17th International Byzantine Congress, Major Papers, pp. 59-81. For a summary of
 The imperial Church and the politics of authority 311

Before Maximus was exiled, however, the emperor had effectively won
the battle. On 27 December 653, the patriarch Paul died and, after a lively
debate as to his suitability, in which the emperor even called the captive
Pope Martin as a character witness, Pyrrhus was elected for the second
time to the patriarchal throne, although he died shortly after - in June of
654 - having achieved very little.84 His successor, Peter, presided over the
provisional reconciliation of Rome and Constantinople. Martin was in exile
and, under pressure from the exarch, a new pope was elected in Rome,
Eugenius I, whose position was confirmed in 655. The patriarch Peter
presented the papal emissaries with a compromise formula, a formula
which in essence resolved nothing, using an ambiguous terminology
which left both the traditionalist and monothelete positions open. The
apocrisiaries accepted it without demur. 85 The Typos remained in force,
although its ban on discussion now became less significant; but in spite of
these developments, Maximus still refused to admit that the emperor had
any jurisdiction in the debate. Indeed, his disciples were able to inform the
Roman Church of events in Constantinople, with the result that the
patriarch Peter's synodika - his formal announcement of appointment and
request for recognition - were rejected. Pope Eugenius was clearly under
local pressure to conform with this turn of events. 86 On Eugenius' death,
however, his successor Vitalian (657-72) recognised Peter and sent his
own declaration of faith, in which he adopted as open and neutral a
position as possible. Relations seemed to be restored, in spite of the ongoing
difficulties with Maximus and his supporters. 87 And in 658, Constans
formally renewed the privileges of the Roman Church in recognition of
Vitalian's co-operative and reconciliatory stance. 88
Maximus continued to maintain his position, however, in spite of these
developments in the situation and, although from the emperor's point of
view the situation was now under control, Maximus still constituted a
vocal and dangerous source of opposition to imperial authority. In 662,
finally, he and his disciples were brought back from exile in Thrace to the
capital, where they were mutilated and exiled to distant Lazica. The great

the conflict between Rome and Constantinople, see Herrin, Formation of Christendom,
pp. 2 5 5 - 9 .
84
See Theophanes, 351.23-4; Nicephorus, Chron., 118.17; Grumel, Regestes, vol. I,
no. 194; Van Dieten, Geschichte der Patriarchen, pp. 1 0 4 - 5 .
85
Van Dieten, Geschichte der Patriarchen, pp. 106ff.
86
Van Dieten, Geschichte der Patriarchen, pp. 107, note 6 and pp. 108-9; Stratos, Byzantium
in the Seventh Century, vol. Ill, pp. 1 2 6 - 8 . Cf. Grumel, Regestes, vol. I, no. 305; Liber
Pontificalisl, 341.
87
Stratos, Byzantium in the Seventh Century, vol. Ill, pp. 126-8; Van Dieten, Geschichte der
Patriarchen, pp. 112-14; cf. Mansi XI, 200D, for Vitalian's name being added to the
Diptychs at Constantinople.
88
Dolger, Regesten, no. 229.
312 Byzantium in the seventh century

opponent of the emperor died in the same year. 89 His second 'trial' and
exile, followed by his death, marked the end of the 'public' controversy
over imperial and official Church monotheletism in the Eastern empire.
The debate and its political ramifications illustrate the struggle conducted
at an ideological level over the locus of authority - both in secular and
theological respects. But, although this conflict took centre stage, it was by
no means the only instance during the reign of Constans where the
question of the nature and extent of imperial authority was raised.
Armenia, which had been predominantly monophysite since about 500,
had been officially united with the imperial Church since an agreement of
union made in 571; and in 591, as Maurice received further districts from
the grateful Chosroes II, a Chalcedonian Catholicate was established in the
Byzantine region of the country, paralleled on Persian territory by a
monophysite Catholicate. In 633 at the synod of Theodosioupolis
(Erzerum), Heraclius was able to win the Armenian Church over to his
monenergite formula. But there was powerful opposition to this develop-
ment. In 648/9, the patriarch Paul II and the emperor tried to reaffirm the
union with the Armenian Church, partly inspired by the complaints of the
non-Armenian soldiers in the Armenian districts of the empire that the
local monophysite population and its leaders excluded them from commu-
nion and treated them as heretics. Constans and Paul ordered the formal
union of the Churches, and the acceptance by the Armenians of both the
Tome of Leo and the doctrine of Chalcedon.90 In reply, the Armenian
Church returned a clear statement of its monophysite faith at the synod of
Dvin. Constans was only able to assert his authority by marching with a
large army into Armenian territory and forcing acceptance of his policy, in
653. Needless to say, such an agreement remained superficial in the
extreme, and monophysitism continued to be the creed of the Armenian
Church. 91
Constans' treatment of Maximus and his followers, however, had poli-
tical repercussions for the emperor and his policies. It seems clear from the
sources that he was not a popular ruler in Constantinople, especially after
his brother Theodosius was compelled to enter the clergy; and it is not
unlikely that his decision to transfer the seat of government to the West -
Rome or Sicily - was at least in part stimulated by this atmosphere. 92
While in the West, the most important feature of his relations with the
89
See Brehier a n d Aigrain, pp. 1 7 3 - 5 ; Van Dieten, Geschichte der Patriarchen, p. 1 1 4 .
90
Dolger, Regesten, n o . 2 2 7 . For the background, see Frend, The Rise of the Monophysite
Movement, pp. 3 0 8 - 1 5 ; R. Grousset, Histoire de VArmenie des origines a 1071 (Paris 1947),
pp. 234ff. a n d 2 9 8 - 3 0 2 ; Brehier a n d Aigrain, pp. 116f.
91
See Sebeos, 112ff. a n d 134ff.; Van Dieten, Geschichte der Patriarchen, pp. 100f.; Brehier
and Aigrain, pp. 1 5 7 - 6 0 .
92
Brehier a n d Aigrain, pp. 1 7 5 a n d 178f.; Van Dieten, Geschichte der Patriarchen, p. 1 1 5 .
 The imperial Church and the politics of authority 313

Church was his grant of autocephaly to the Church of Ravenna. This

appears to have been primarily a result of the arguments and influence of
the archbishop of Ravenna, Maurus, who will no doubt have pointed out
the difficult and anomalous position of his Church, a suffragan of Rome
and whose position in the hierarchy was obviously not commensurate
with its much higher status in the civil and military establishment. The
exarch Gregory supported Maurus' petition, possibly - as the document
itself makes clear - because the Church of Ravenna had made payments in
cash to the exarch - whether on a personal basis or officially for the
support of the military-administrative needs of the exarchate remains
unknown. 93 It is certainly clear that the imperial fisc benefited from the
taxation raised from the Sicilian patrimony of Ravenna; and since it seems
clear that from the sixth century the state had used the Church of Ravenna
as its fiscal agent in its Italian territories, the reward of autocephaly was a
measure of Constans' gratitude and the reliance of the state upon it. 94

The sixth ecumenical council and the end of monotheletism

Constans' reign ended with his assassination in Sicily in 668. The abortive
coup of the Armenian commander of the Opsikion forces, Mzez Gnouni
(Mizizios in the Greek sources) was speedily followed by the establishment
of a firm rule by Constantine IV, who also initiated the process of recon-
ciliation with Rome. Constantine seems to have tried genuinely to adopt a
neutral stance in the debate, concerned as he clearly was with the political,
military and economic situation of the state. From his accession in 668,
however, political and military problems had overshadowed such matters,
and the great siege of Constantinople by the Arabs in the years 674-8
prevented any attempts to conduct a meaningful and consequential
ecclesiastical policy. Only in respect of the Church of Ravenna does the
emperor seem to have taken any decisions, and in two decrees of 671 and
677 the emperor regulated the rights of ordination of the archbishop of
Ravenna and freed clerics from harbour- and customs-dues. He also
limited the time that the archbishop of Ravenna should spend in Rome at
his consecration - the latter already suggesting the first move in the
subordination of Ravenna to Rome which was ordered in 682. 9 5
Once the threat from the Arabs was past, however, and a truce had been

93
See Dolger, Regesten, n o . 2 3 3 ; a n d esp. T.S. Brown, 'The Church of R a v e n n a a n d the
imperial administration in the seventh century', EHR 9 4 (1979), 1 - 2 8 , see 12ff. a n d
16ff.; Guillou, Re'gionalisme, pp. 167ff. a n d 2 0 6 - 7 .
94
Brown, 'The Church of R a v e n n a ' , 1 7 - 2 0 .
95
Dolger, Regesten, nos. 2 3 7 a n d 2 3 8 ; Brown, 'The Church of Ravenna', 20ff.; Guillou,
Regionalisme, pp. 177f.
314 Byzantium in the seventh century

arranged (in 678), 96 Constantine turned to the question of the relationship

between Rome and Constantinople. Although the names of neither of the
Popes Adeodatus (672-6) or Donus (676-8), the successors of Vitalian,
had been entered in the Diptychs at Constantinople, relations seem never-
theless to have been less hostile than during thefirsthalf of Constans' reign.
Hostility between Ravenna and Rome persisted until the deaths of the two
main protagonists, Pope Vitalian himself (in 672) and archbishop Maurus
(in 673), when the tension relaxed somewhat. The monothelete patriarch
Theodore, however, who was elected in August 677, together with the
patriarch of Antioch, Makarios (also a firm monothelete), requested that
the emperor permit the erasure of the name of Pope Vitalian also from the
Diptychs.97 Constantine was not prepared to condone this step, however,
the more so since, as he states in his letter to the pope, Vitalian had
supported him against the usurper Mzez in 668. In this letter, addressed to
Donus in 678, but delivered to the latter's successor Agathon, the emperor
asked the pope to convene local councils to debate the key question of the
relationship of the one or two wills and energies to the pronouncements of
the Fathers of the Church and the five ecumenical councils. He represented
the debate as one over terminological differences between two equally
orthodox schools of thought. And while he intimated that an ecumenical
council would be desirable, it was important first to debate the matter
thoroughly, and he requested that the pope send representatives from
Rome and the Western Churches to Constantinople to discuss the whole
matter thoroughly.98 Interestingly, Constantine admitted that an imperial
edict would not itself resolve either the problem or end the debate, a
marked change from the point of view of his predecessors, from Zeno on up
to Constans II.99
The pope took some time to consider this approach; and in doing so he
wrote to all the metropolitans of the Western Church, asking them to
convoke local councils and transmit their conclusions to him. Some sent
their responses directly to the emperor.100 But by Easter of the year 680,
the Western Church had sent its replies to Rome and, following a local
meeting in Rome itself, the pope sent two letters to Constantinople, one to
the emperor and another, shorter letter, containing a formal profession of

96
Dolger, Regesten, no. 239.
97
Brehier and Aigrain, pp. 183f.: Van Dieten, Geschichte der Patriarchen, p. 126.
98
Text of the letter: Mansi XI, 195-201; Riedinger, 2-10; Dolger, Regesten, no. 242; general
background: Stratus, Byzantium in the Seventh Century, vol. IV, pp. 57ff. and 119ff.; and for
background and summary of events leading up to the sixth ecumenical council, see Herrin,
Formation of Christendom, pp. 2 74ff.
99
Mansi, XI, 200C; Riedinger, 8.20-22. « » Brehier and Aigrain, pp. 184f.
 The imperial Church and the politics of authority 315

faith and a rejection of the heresy of monotheletism.101 In the interim, the

patriarch Theodore had managed to have Vitalian's name removed from
the Diptychs, partly because the long delay in Rome's reply may have
suggested to Constantine a rejection of his proposals. But in November 679
Constantine replaced Theodore, who was proving an embarrassment to
his attempts at reconciliation, and enthroned the more amenable patriarch
George, a Syrian by birth.102
In September of 680, therefore, the Western delegation arrived with the
pope's letter in Constantinople; and on 10 December the emperor ordered
the patriarch George to convoke a council and to invite also the patriarch
of Antioch, Makarios, and his representatives, along with the representa-
tives of the orthodox (Chalcedonian) patriarch of Alexandria.103 The
council met for its opening session in the imperial hall within the palace
known as the Troullos, or cupola, on 7 November 680. The emperor
himself opened the deliberations and personally presided over the first
eleven and the last sessions of the eighteen which took place up to 16
September 681. 1 0 4 The results of the council were the abandonment of an
imperial policy of monotheletism. Almost immediately, in the first session,
the Roman delegation levelled accusations against the patriarchs Sergius,
Pyrrhus, Paul and Peter, against Cyrus of Alexandria and Theodore of
Pharan. Only Makarios of Antioch and his supporter, the monk Stephen,
who seem to have genuinely represented the feelings of the neo-Chalcedo-
nian communities of Syria and Palestine, defended their monothelete
position, although with little success. In the ninth session, they were
condemned also; and in the thirteenth session the supposed founders of the
heresy were anathematised.105
In the final session, at which the emperor was once again present, the
condemned patriarchs, along with Makarios of Antioch and Stephen, were
formally condemned by name. Copies of the acts of the council were sent to
all patriarchates, that to Rome accompanied by a letter in the name of the
council and the emperor confirming the orthodoxy of the Roman see. But
Agathon had died while the council was still meeting, and when news

>°> Mansi, XI, 234-86; Riedinger, 52-120 (to the emperor); and 286-315; Riedinger,
122-160.See Brehier and Aigrain, p. 185; Van Dieten. Geschichte der Patriarchen, pp. 132-4.
">2 See Theophanes, 354.24; 355.1-4; and Brehier and Aigrain. p. 185; Van Dieten,
Geschichte der Patriarchen, p. 129.
103
Dolger, Regesten, no. 244. The meeting was not originally understood as an ecumenical
council, but adopted this title during its first session: see Brehier and Aigrain, p. 186 and
note 7.
104
Br6hier and Aigrain, pp. 187-90; Van Dieten, Geschichte der Patriarchen, pp. 134-42;
Stratos, Byzantium in the Seventh Century, vol. IV pp. 1231T.; Ostrogorsky, Geschichte,
pp. 106f.
105
Mansi, XL 213A-B, 216C, 383C and 553D (Riedinger, 20.20-22.6; 24.7ff.; 270.4ff.;
578.12ff.). For the confessio of Makarios, see ibid., 349-59; Riedinger, 216.11-230.26.
316 Byzantium in the seventh century

arrived that Leo II had succeeded him, the letter was readdressed. Follow-
ing tradition, an imperial edict was then drawn up and exhibited in the
Hagia Sophia, in which the emperor confirmed the decisions reached
during the council and threatened punishment on those who refused to
conform.106 Similarly, a iussio was issued to all the dioceses of the empire
informing them in turn of the decisions of the council.
The sixth council thus sounded the death-knell for imperial monothele-
tism in the eastern empire, although it was to be briefly and unsuccessfully
revived again under Philippicus Bardanes (711-13). It also marked the
reconciliation of Eastern and Western Churches, although at the expense
of Constantinopolitan claims to equality with Rome - the condemnation of
the erring patriarchs by name meant as much. It signified in addition,
however, the end of the unified Christian world, for the monophysite
communities of the lost provinces were not even considered in the council's
deliberations. This reflects primarily, of course, the fact that these mono-
physite communities were outside the empire. The questions which the
sixth council was assembled to resolve were all questions which concerned
the Christian communities still under Christian political authority. But the
sixth council marks a break with the past, and at the same time the
beginnings of a more emphatically introverted political-theological culture
and ideological consensus. The East was no longer as important as it had
been. 107 And the monophysites themselves regarded Constantine IV as
having been bought by the papacy and as having abandoned his own
convictions. Whether this later comment has any value is difficult to say,
but there must be little doubt that the monophysite communities in the lost
Eastern provinces can have had little sympathy with their Western
brethren after this time. 108
Constantine continued his policy of reconciliation with Rome at the
expense of Ravenna, too. In 681 he ordered a reduction in the rate of
assessment of the basic taxes on the papal patrimonial lands in both Sicily
and Calabria;109 and in 682/3 he placed the see of Ravenna once more

"* Brehier and Aigrain, pp. 189-90; Van Dieten, Geschichte der Patriarchen, pp. 141-3. For
the letter to the pope, see Mansi XI, 683-8; Riedinger, 830.4-5 and the comments of Van
Dieten, Geschichte der Patriarchen, p. 143 and note 53; see also Dolger, Regesten, no. 247;
Grumel, Regestes. vol. I, no. 312. For the imperial edict, Dolger, Regesten, no. 245; and for the
council and its significance, see Herrin, Theformationof Christendom, pp. 2 77-80, who stresses
in particular the innovative procedures adopted to verify the texts read to the meetings, a
major development in the intellectual and political history of the Church.
107
See Winkelmann's comment, Die ostlichen Kirchen, pp. 111-12.
108
See Michael Syr., vol. II, 447-8 and 457. For the later history of these communities, see
Br6hier and Aigrain, pp. 4791T.
109
Dolger, Regesten, no. 250.
 The imperial Church and the politics of authority 317

under papal jurisdiction.110 On his death in 685 there was once again
religious unity across the Christian world, although the continued exist-
ence of strong monophysite communities, as well as monothelete commu-
nities in the lands under Muslim political control, remained an important
factor in the political and ecclesiastical make-up of the cultures of the east
Mediterranean world.

The Quinisext council and the divergence of East from West

Constantine's son, Justinian II, clearly saw himself also as a true defender
of orthodoxy. So much is clear in his letter to the pope of 687, in which he
writes of his adherence to the decisions of the sixth council of 680/1, a
letter drawn up after a general assembly of representatives of the pope, the
patriarch, along with the metropolitans and bishops present in Constantin-
ople, the demes, the palatine parade units and the provincial field-armies -
Opsikion, Anatolikon, Armeniakon, Thrakesion, Italy, Sicily, Africa and the
imperial fleet - which was held to confirm the acts of the council.111 He
saw himself, too, as occupying the traditional position of the emperor in
respect of the Church - it was his duty to convoke and organise general
councils and to cater for the legislative activity of the Church; he was the
defender of dogma and the faith; and he made this position clear by his
actions in not inviting papal representatives, for example, to attend the
Quinisext council of 692. They were present, but not by the emperor's
invitation.112
The Quinisext council met primarily to impress upon the oikoumene
Justinian's own role, and to deal with a wide range of disciplinary and
related matters. It did not concern itself directly with the major theological
issues which had concerned the fifth and sixth general councils, the acts of
which the council of 692 was called upon to confirm and to elaborate,
hence its popular title, penthekte, or quinisextum. Instead, and as the
emperor's opening statement makes clear, it was concerned with such
matters as the evidence for paganism in the provincial communities. The
canons of the council make this concern with Church discipline apparent.
Opening with a general statement of orthodoxy and affirming the decisions
of all six ecumenical councils, the first canon of the Quinisext repeats the
anathemata pronounced in the sixth council - in particular that on Pope
Honorius. The second canon recognises all the apostolic canons, but goes

110
Dolger. Regesten, no. 251; Brown, 'The Church of Ravenna' above, 22f.; Guillou, Rtgio-
nalisme, pp. 207-8; Herrin, Formation of Christendom, pp. 280-2.
i » TextrMansiXI, 737-8;Riedinger,886f.SeeD61ger,K6^sten,no.254;BrehierandAigrain,
pp. 192-3; VanDieten, GeschichtederPatriarchen, pp. 146ff.
112
Cf. Liber Pontiflcalis, I 372ff.; Stratos, Byzantium in the Seventh Century, vol. V, pp. 45flf.
318 Byzantium in the seventh century

on to regard the disciplinary canons of the ecumenical councils and the

local synods of the Eastern Church only (with the exception of those of the
council of Carthage) as valid. The remaining one hundred canons of the
Quinisext deal exclusively with questions of clerical discipline and the
regulation of the Christian communities, and they make clear the fact that,
in spite of the numerous divergences between Eastern and Western
Churches, this council assumed that it was acting for the whole Christian
community - that it was ecumenical - and that it could, therefore, impose
'Byzantine* practice on the West, on the monophysite communities of the
East and upon Armenia. Roman practice, indeed, was openly criticised in
the acts and canons themselves.113 Quite apart from this, canon 36
asserted the equivalence of Rome and Constantinople in the ecclesiastical
hierarchy,114 a ruling which demonstrates quite clearly the difference
between the policies of Justinian II and those of his father. The latter
worked for reconciliation and, as we have seen, was regarded in some
circles as having betrayed the interests of the Eastern Church to Rome.
Justinian, partly to reassert his authority on the Byzantine Church, partly,
no doubt, out of a genuine conviction, defended the interests of Constantin-
ople and thus, inevitably, came into conflict with Rome.115
The Roman response can hardly have come as a surprise. Pope Sergius,
who succeeded in 687 after the death of Conon, repudiated the signatures
of the papal legates, when he received his own copy of the canons, and
refused to sign his own name. He protested that a number of the canons
directly contradicted established Roman practice and traditions, in some
cases even in the East. He refused likewise to recognise the claims of the
Quinisext for the apostolicity of a number of canons from earlier synods;
and he objected especially to canons 3 and 13 relative to the marriage of
the clergy (which fixed less severe penalities for clerics involved in relation-
ships within the prohibited degrees of affinity and which ordered clerics
who were married to remain with their wives if they were of the rank of
deacon or priest - conflicting with Roman practice, by which priests had to
renounce their wives before entering the diaconate).
113
For the canons of the Quinisext, see Mansi, XI, 921-1006. The acts of the council have
not survived, only the concluding address to the emperor, the signatures of those present,
and the canons themselves. For discussion, see Winkelmann, Die ostlichen Kirchen,
p. 113; Van Dieten, Geschichte der Patriarchen, pp. 153-4; Brehier and Aigrain,
pp. 195-6, 474f. and 485-8; and esp. V. Laurent, 'L'ceuvre canonique du concile in
TfuUo (691-692), sourceprimairedu droltdel'eglise orientate', REB 23 (1965), 7-41. For
the most modern analysis, see Ohme, Das Concilium Quitiisextum und seine Biscfwfsliste.
114
Mansi, XI, 959.
115
For a detailed treatment of subsequent events and Justinian's relationship with the
papacy, see F. Gb'rres, 'Justinian II. und das romische Papsttum', BZ 17 (1908), 440-50;
also J.D. Breckenridge, 'Evidence for the nature of the relations between Pope John VII
and the Byzantine emperor Justinian II', BZ 65 (1972), 364-74; Stratos, Byzantium in
 The imperial Church and the politics of authority 319

Justinian responded in what was almost predictable fashion. The magis-

tros Sergius was ordered to Italy, where he arrested immediately the bishop
John of Porto and the clerk Boniface, two of the pope's close advisers, and
had them sent off to Constantinople. Shortly thereafter, the protospatharios
Zacharias was despatched with orders to arrest the pope himself.116 But
the situation was now very different from that of the 650s. Local sentiment
and tradition, feelings of separateness and difference from the Eastern parts
of the empire, together with a less pressing need for imperial military
support (a peace had been concluded with the Lombards in about 680),
meant that even the Ravenna troops were no longer willing to act unques-
tioningly on behalf of distant emperors. The Roman troops and those of
Ravenna turned on Zacharias, and he escaped only with the aid of the
pope. 117
Justinian's deposition in 695 and his replacement by Leontius (695-8)
and then Tiberius II Apsimar (698-705) do not seem appreciably to have
altered this situation, and it was only during Justinian's second reign
(705-11) that a reconciliation of sorts took place. In spite of his emphasis
on the pre-eminence of Constantinople and his anti-Roman perspective in
regard to Church discipline and practice, Justinian II seems also to have
been fully aware of the wider context of his policies. In two iussiones of
about 687, for example, that is at the outset of his reign, he ordered the
repayment of monies extracted in lieu of unpaid tax from the Roman
Church in Sicily and elsewhere, and the reduction of the tax-assessment on
papal properties in Bruttium and Lucania, orders which follow on from the
policies of his father's reign. 118 The papal response to the canons of the
Quinisext changed this, of course, but the original direction of the first
years of his reign are clear enough.
Justinian's second reign was marked by a distinct effort at reconciliation.
In his first year he directed a request to the pope, that the latter might
convene an apostolic Roman synod to discuss the decisions and canons of
the Quinisext and decide which were acceptable and which not to the
Roman Church. 119 The pope, John VII, died in 707 without taking advan-
tage of this opportunity, and he was succeeded by Constantine, whom
Justinian asked to pay a formal visit to Constantinople. In 710, the new
pope, accompanied by a considerable retinue of clergy, set out, stopping en

the Seventh Century, vol. V, pp. 48-53; Ostrogorsky, Geschichte, p. 116; Herrin, Formation
of Christendom, pp. 284-7; and see Grumel, Regestes, vol. I, no. 317.
116
Dolger, Regesten, no. 259; Stratos, Byzantium in the Seventh Century, vol. V, pp. 53-6;
Ostrogorsky, Geschichte, pp. 116-17.
117
Brehier and Aigrain, p. 197; Brown, 'The Church of Ravenna' 25; Van Dieten, Geschichte
der Patriarchen, pp. 153-5; Ostrogorsky, Geschichte, pp. 116-17; Guillou, Regionalisme,
pp. 209-11.
118 119
Dolger, Regesten, nos. 255 and 256. Dolger, Regesten, no. 264.
320 Byzantium in the seventh century

route at Naples, Palermo, Reggio and Otranto (where the winter was
spent), before proceeding to Chios and on to Constantinople. He met the
emperor at Nicomedia in 711, and it seems that a compromise formula
was worked out. Later evidence suggests that some fifty of the 102 canons
of the Quinisext were formally accepted by the Western delegation. After
state receptions and a mass, at which the pope celebrated the communion,
he left in October 711 for Rome once more. 120 Shortly after, Justinian was
deposed and killed by the rebels under Philippicus Bardanes.
While his second reign was marked by an eccentricity which has earned
him a certain notoriety among historians, 121 the policies he pursued,
especially in the period of his first reign from 685 to 695 demonstrate an
emphasis on the assertion of his authority as emperor not over the details
of dogma, but over the general welfare and direction of the Church,
reinforced by the emphasis in the imperial ideology on his position as pious
defender of the faith. 122 His relations with the Church, as far as the limited
evidence suggests, were those of a protector and patron - the edict of 688
in favour of the church of St Demetrius in Thessaloniki is a good illustra-
tion. 123 His relations with the patriarch Callinicus, in contrast, were
stormy, primarily on the grounds of Callinicus' efforts to hold the emperor
back from his cruel persecutions of the last years of his first reign. But
Justinian was to let nothing stand in his way, and when the patriarch
attempted to change his mind with regard to the demolition of a church,
which was to make way for a new, imperial construction, Justinian was
unmoved. 124
Justinian's deposition, and the short rule of the Armenian Philippicus
Bardanes, brought with it a major shift, albeit of short duration, in imperial
ecclesiastical policy. The new emperor announced his rejection of the acts
of the sixth council even before entering the city and ordered both the
removal of the image of that council from the palace precincts (before he
would enter) and the restoration to the Diptychs of the patriarch Sergius
and the other Churchmen condemned for their espousal and promotion of

120
Brehier and Aigrain, pp. 1 9 8 - 2 0 0 ; Van Dieten, Geschichte der Patriarchen, pp. 161ff. See
Dolger, Regesten, nos. 2 6 6 - 9 (note 269, for October 711, by which the emperor renews
all the privileges of the Roman Church). For the fifty acceptable canons, see the
ninth-century account of Anastasius Bibliothecarius, in Mansi XII, 982; and for the
background and sequence of events, Stratos, Byzantium in the Seventh Century, vol. V,
pp. 1 3 1 - 5 ; Herrin, Formation of Christendom, pp. 2 8 7 - 9 .
121
Although attempts have been made to redeem his reputation: see C. Head, 'Towards a
reinterpretation of the second reign of Justinian II: 7 0 5 - 7 1 1 ' , B 4 0 (1970), 1 4 - 3 2 .
122
J.D. Breckenridge, The Numismatic Iconography of Justinian II (A.D. 685-695, 705-711)
(New York 1959), esp. pp. 92f. For the general context, see Haldon, Ideology and social
change', 189; also Van Dieten, Geschichte der Patriarchen, p. 146 and note 2.
123
Dolger, Regesten, no. 2 5 8 .
124
Van Dieten, Geschichte der Patriarchen, pp. 157f.; see Theophanes, 367.22sq.
 The imperial Church and the politics of authority 321

monotheletism. His actions are explained by the deacon Agathon, who

later recopied the acts of the sixth council destroyed on Philippicus' orders,
in terms of his probable monophysite background, but more especially in
terms of his having been a pupil of the abbot Stephen of Antioch who,
together with Makarios, had been the last defenders of monothelete doc-
trine at the sixth council.125
The patriarch Cyrus was deposed, probably because, like a number of
others, he refused to sign Philippicus' tomos condemning the sixth council.
He was replaced with a man more to Philippicus' liking, the deacon and
chartulary John, who became John VI; although the orthodox estab-
lishment seems to have been able to have an original and much less
acceptable imperial candidate dropped.126 The new patriarch, along with
other prominent Churchmen, including Andreas of Crete and Germanus of
Cyzicus, later patriarch, faced with imperial pressure to conform, had
little choice but to go along with the revived monothelete policy. The
extent to which they may have seriously regarded it as a valid doctrine
remains unknown;127 but the context of the later seventh century provides
a good reason for this last effort (as it turned out) to arrive at a compromise
formula in the Christological debate.
In the first place, relations with Armenia were extremely difficult, due
both to the constant strife within Armenia between different factions,
pro-Byzantine and pro-Arab, and to the monophysite creed of the majority
of the population. In addition, there was the frequent intervention of the
Caliphate on one side or the other. The political and military importance of
the country was recognised by both empire and caliphate, but religious
differences meant that there existed a constant tension between the Byzan-
tine and Armenian populations and their leaders. It is possible that Philip-
picus' monothelete policy reflects a desire to overcome these conflicts and
represents a last effort at union. An edict of Justinian II for 711 ordered all
Armenians within the empire to recognise the authority of the patriarch of
Constantinople, partly a response to the recent desertion to the caliphate of
the pro-Byzantine faction of the Armenian nobility: and the Caliphate took
advantage of the situation by granting Armenia its autonomy and relig-
ious freedom, restoring the exiled nobles and placing Smbat Bagratuni,
from a pro-Byzantine clan, at the head of the Armenian forces. Henceforth,
125
See Mansi XII, 192C-E; Riedinger. 899.1Off.; and the account In Van Dieten, Geschichte der
Patriarchen, pp. 163f.
126
Van Dieten, Geschichte der Patriarchen, pp. 166f.
127
Br&iier and Aigraln, pp. 206f.; Van Dieten, Geschichte der Patriarchen, pp. 167-71. See
S. Vailhe, 'Saint Andre de Crete', EO 5 (1902), 378-87. Andreas seems later, like
Germanus, to have repented of his decision: see PG LXXXXVII1437-44, a short poem in
which he suggests as much; and the discussion of this by H. Heisenberg, 'Ein jambisches
Gedicht des Andreas von Kreta', BZ 10 (1901), 505-14.
322 Byzantium in the seventh century

Armenia remained under Arab suzerainty, and although contacts with the
empire were never severed, the synod of Mantzikert of 719 - which
ordered the expulsion of all Chalcedonians from Armenia - meant the end
of any Byzantine pretensions in the region. Philippicus' monotheletism,
given his Armenian, and possibly monophysite, background, seems to
have been a forlorn attempt to bring the Armenians back into the political
orbit of the empire. 128
In the second place, Philippicus may have hoped that a return to the
policies of Constans II would both enhance his own authority within the
empire and regain divine support - the defeats of the previous years, the
final loss of North Africa after 698, the increasing power of the Bulgars, all
had a profound effect on contemporaries; and this policy may well have
been seen as one potential way of reversing the process. 129 In the event, of
course, it was a failure. The papacy predictably refused to condone the shift,
rejecting both the imperial edict and Philippicus' portrait, and declaring
him a heretic. At the news of the destruction of the icon of the sixth council,
the clergy and population of Rome paraded all six conciliar images in
affirmation of their orthodoxy. 130 The Bulgars, on the pretext of avenging
their erstwhile ally Justinian, invaded Thrace and met with little or no
opposition. Even within the army Philippicus seems to have had little real
support, although whether as a result of his innovation in religious policy
or his failures as emperor generally is unclear, and in 713 he was deposed
by officers and soldiers of the Opsikion army based in Thrace. 131 His
successor, the imperial secretary Artemius, took the imperial name Anas-
tasius II and immediately reversed Philippicus' policies. The image of the
sixth council was restored to its position; the pope was apprised of the new
emperor's orthodoxy and his adherence to the sixth council's decisions;
while the patriarch John VI now sent an explanatory letter to the pope,
detailing his reasons for going along with the previous emperor's religious
policies. John died in the summer of 715 and was succeeded by Germanus
of Cyzicus who, having repudiated monotheletism, was supported by the
papal legate in the election to the patriarchal throne. 132
Anastasius himself did not survive John by more than a few weeks.
Within a few months a mutiny of naval and land units compromised his

128
Justinian's edict: Dolger, Regesten, no. 2 7 2 (wrongly ascribed to Philippicus); J. Laurent,
L'Armenie entre Byzance et I'lslam depuis la conquete arabe jusqu'en 886 (Paris 1919),
pp. 2 0 2 - 6 ; Brehier a n d Aigrain, pp. 2 0 3 - 4 ; Grousset, Histoire de VArmenie, pp. 3 0 7 - 1 5 .
129
See Haldon, Ideology a n d social c h a n g e ' , 1 8 6 - 9 .
130
Brehier a n d Aigrain, p. 2 0 7 ; Ostrogorsky, Geschichte, pp. 1 2 7 - 8 ; Van Dieten, Geschichte
der Patriarchen, pp. 1 7 0 - 1 .
131
See chapter 2, above.
132
Brehier a n d Aigrain, p. 2 0 8 ; Van Dieten, Geschichte der Patriarchen, pp. 172f.; Dolger,
Regesten, no. 2 7 3 .
 The imperial Church and the politics of authority 323

position, and he was compelled to abdicate. Theodosius, his successor, was

able to continue the preparations which had been set in motion by
Anastasius for the imminent Arab siege of the city; but he, too, was soon
compelled to vacate the throne for the general of the Anatolikon thema, Leo.
And it was Leo who opened the next chapter in Byzantine ecclesiastical
and theological history. But the iconoclastic controversy is the subject of
another book. 133
As in the case of secular politics, the seventh century opened with a
unity, a Christian Church which represented a unified oikoumene - what-
ever the realities of the debate between Chalcedonian and monophysite
communities, such differences were thought to be surmountable and
subject to discussion and agreement. By the end of the seventh century, the
Eastern monophysites were no longer part of the empire, and no longer
merited much consideration, although the orthodox communities of Egypt
and Syria contined to play a role in the affairs of the Church as a whole.
More important, however, the split between Roman and Constantinopoli-
tan politics and practice had finally come out into the open, and in a way
which could no longer be papered over. The great split in the Christian
world, which was to play such a crucial role in defining the cultural
appearance of East and West, had taken place, and was to prove irre-
vocable.
133
See chapter 2, above.
 CHAPTER 9

Religion and belief

THE CHURCH AND CHRISTIANITY! NEO-CHALCEDONIAN

UNIVERSALISM AND HETERODOX PLURALISM
Religion and belief are not, of course, synonymous. In this chapter, I want
to look at the 'public' and the 'official' aspects of seventh-century religious
and ideological history, rather than at the individual psychology of belief
in the early medieval world of the Eastern Roman empire. But since the
latter subject is, in several ways, the prerequisite for an understanding of
the world as it was perceived and understood by people of this period, a few
words on the assumptions which inform my approach are in order.
Although several scholars have approached this problem, of the Byzan-
tine mentalite, at a general level, their work has on the whole been
descriptive rather than analytic. That is to say, Byzantinists have generally
tended towards a presentation of the data from the sources on the forms of
belief and its appearance - the specific ideas which together form the
imperial ideology, or orthodoxy, or attitudes to the everyday. In the
process, the question of the actual generation and reproduction of ideas in
their social-psychological context tends to be forgotten. More particularly,
the question of the relationship between what people think (as this is
presented at different levels in our sources, through histories, hagiogra-
phies, letters and so on), about both themselves and about their per-
ceptions of the world and how they act upon that world, how those
perceptions inform the possibilities open to them in respect of their social
existence has been neglected. There is a dialectical relationship here, of
course, between these two forms of human social praxis, and it is through
the examination of that dialectical relationship - by relating contexts to
effects - that the historian can best approach the whole problem of causal
relationships.
The key question is, quite simply: what enabled Byzantines in the period
with which we are concerned to think and act as they did? I have

324
 Religion and belief 325

elaborated an approach to this problem elsewhere in terms of the inter-

action between a number of levels of cognition and practice which play a
role in constituting both human beings as social beings and the world as
perceived by them.
Briefly, these levels or 'moments' can be described as follows. First, the
concept of symbolic universe, which refers to the totality of cultural
knowledge and practice in a society, within which and through which
everyday life is carried on. Within the symbolic universe and drawn from a
number of its elements, specific ideological discourses are generated,
chiefly a result of the contradictory nature of the social relations of
production (that is to say, through the existence of antagonisms between
different social groups in respect of their relationships to the means of
production and distribution of wealth, to the sources of power and so on;
antagonisms which need to be made sense of). Ideological discourses
usually serve to mask such contradictions in social relations by first
explaining and then legitimating them within a wider framework - that of
the symbolic universe. Out of this ideological consciousness and practice
specific discourses are then refined, from which they take their logic and
vocabulary, and which legitimate through a series of symbolic statements
the relationships between groups of individuals on one side, and between
such groups and their world on the other. These are what are commonly
termed political ideologies, and it is in this sense that the notion of an
imperial ideology is employed.
As I have said, the relationship between consciousness and practice
should be construed as dialectical and mutually constituting; an instru-
mental relationship through which individuals receive their sense of self
and subjectivity, and which provides them with the conceptual apparatus
with which they can express what they 'know' about the world, but that
also limits what they can know, and how, to the culturally possible.
Contingently, the symbolic universe, within which and through which this
instrumental relationship works to constitute human individuals, is itself a
product of, and is constantly reproduced by, social practice: by the activi-
ties of individuals in their social and physical context which serve to
reproduce both themselves and the social relations of production, the roles
and institutions of their society. It needs to be said that in reality, that is, in
the expression of people's experience, consciousness and practice are
hardly distinguishable; but an analytic distinction is necessary if this model
of cognition and practice is to have any heuristic value.
A useful way into these relationships is to employ the concept of
personal and group narrative as an element which generates the construc-
tion of social reality and a sense of self. I have argued elsewhere that
narratives are important because they are crucial to the link between
326 Byzantium in the seventh century

thinking and doing. For narratives are ways of organising experience in

time, they reconstruct experience, whether first- or second-hand, and
they therefore act as patterns for future action. But since they are
reconstructions, they also necessarily involve evaluation and thus the
potential for change. In particular, a change in the order or the relation
of elements within a certain narrative involves a change in evaluation,
and hence a change in perception of the relationship between self (or
group) and the world. Narratives are the key structuring elements in
ideology.
Now this is a useful approach to the question of the relationship between
thinking and doing on the one hand, and to that of historical change on
the other. For the latter is manifestly a result of human action (or reaction).
Narratives serve as basic elements in describing and prescribing social
practice. But if narratives, and the realities they purport to describe, no
longer match the world as it is perceived or experienced, then changes in
one or the other can follow: either the narratives are suitably altered in
order to make them fit the new perceptions, or the world is acted upon in
an effort to make it correspond more closely with the narratives. In the
later sixth and seventh centuries it is precisely such efforts at both re-
evaluation and the re-establishment of a desired world-order which our
sources represent.1

1
For useful descriptive surveys of the forms of orthodox Byzantine beliefs, see Beck, Das
byzantinische Jahrtausend, pp. 257-89, 'Die Byzantiner und ihr Jenseits', SBB (Munich
1979), Heft 6, and 'Orthodoxie und Alltag', in Bv^dvnov. A<piep<o/j,a arrov AvSpea N.
XrpdTO (2 vols., Athens 1986), vol. II, pp. 329-46. See also A.Kazdan and G. Constable,
People and Power in Byzantium (Washington D.C. 1982), esp. pp. 76ff.; and the useful
general survey by P. Kawerau, Das Christentum im ostromisch-byzantinischen Reich bis zur
osmanisch-tiirkischen Eroberung Konstantinopels, CSCO, vol. 441, Subsidia 64 (Ostkirchenge-
schichte, II) (Louvain 1982). See also C. Mango, Byzantium: The Empire of New Rome, esp.
pp. 151-217.
I have outlined my own approach in greater detail in Ideology and social change', esp.
145-6 and 150-3, with literature. On the relationship between cognition and practice, see
in particular R. Bhaskar, 'Emergence, explanation and emancipation', in Explaining Human
Behavior: Consciousness, Human Action and Social Structure, ed. P.F. Secord (Beverly Hills,
Calif. 1982), 275-310, esp. 278-88; D.-H. Ruben, Marxism and Materialism: A Study in
Marxist Theory of Knowledge (Brighton 1979), pp. 95ff.; T.W. Goff, Marx and Mead: Contri-
bution to a Sociology of Knowledge (London 1980). On the construction of roles and social
realities, see A. Schiitz, Der sinnhafte Aufbau der sozialen Welt (Vienna 1960); P. Berger and
Th. Luckmann, The Social Construction of Reality (Harmondsworth 1967), pp. 65-70 and
llOff.; and on narrative, see esp. H. Garfinkel, Studies in Ethnomethodology (Englewood
Cliffs, N.J. 1967) (together with the important critique of the work of Schiitz and an
appreciation of Garfinkel by J. Heritage, Garfinkel and Ethnomethodology (Cambridge and
Oxford 1984)); D.Sperber, Rethinking Symbolism (Cambridge 1975), pp. 85-149; Goff,
Marx and Mead, pp. 112ff.; and W. Labov and J. Waletzky, 'Narrative analysis: oral versions
of personal experience', in Essays on the Verbal and Visual Arts, ed. J. Holm (Seattle 1967),
pp. 12-44.
 Religion and belief 327

Pagan survivals?
The degree to which Christianity had been adopted by the populations of
the territories remaining to the empire in the second half of the seventh
century is still far from clear. While it had been the official religion of the
Roman state since the time of Theodosius I, and while, by the end of the
sixth century at least, non-Christians are treated by both Chalcedonian
and monophysite writers alike as either marginal or backward, it is
nevertheless clear that considerable numbers of people continued to
observe traditional and pre-Christian cult practices, whether this occurred
in a thinly disguised but nevertheless Christian form, in which non-
Christian rituals and practices received a Christian veneer, or whether it
occurred in an overtly pagan form. To some extent, the question is bound
up with the survival of the native - non-Greek - languages of the East
Roman world, particularly in Asia Minor, where it has sometimes been
assumed that the survival of languages such as Phrygian or Galatian was
accompanied also by the survival of pre-Christian cultural traditions and
beliefs.2 Now it is certainly true that non-Greek languages may have
survived in Asia Minor well into the seventh century; 3 and they existed, of
course, in the Middle Eastern provinces of Syria and Egypt, for example.
But while linguistic differences certainly facilitated a degree of local par-
ticularism, and - in the cases of Syria, Egypt and North Africa - were on
occasion combined with religious differences or disagreements to function
as vehicles for sentiments and actions hostile to the state or to the
Chalcedonian establishment, there is no reason for assuming automatic-
ally that the survival of pre-Hellenistic languages meant also the survival
of pre-Christian religions.4
The universality of Christianity was the proclaimed reality of the
seventh-century Byzantine world. It was necessarily accompanied, in
theory at least, by the universality of a single orthodoxy, which from the
council of Chalcedon of 451 was represented in the so-called Chalcedonian
creed of the two natures - and from the middle of the reign of Justinian I by
the neo-Chalcedonian theology based on this and the theological system of
Cyril of Alexandria. The only serious challenge to this, within the frame-
2
See especially K. Holl, 'Das Fortleben der Volkssprachen in Kleinasien in nachchristlicher
Zeit', Hermes 43 (1908), 240-54; criticised by Sp. Vryonis, jr., Decline of Medieval Hellenism,
pp.45ff. and 59f.; and in Bv&vTivd 1 (1969), 214f.
3
See esp. P. Charanis, 'Ethnic changes', 23ff., although vitiated by an acceptance of Holl's
methodological assumptions. On the survival of Galatian in the first half of the seventh
century, see Timotheus Constantinopolitanus, De Receptione Hereticorum {PG LXXXVI
11-68), 13B-16A; and note J. Gouillard, 'L'Heiesie dans l'empire byzantin des origines au
XIIe siecle', TM 1 (1965), 299-324, see 304, on this text.
4
So Charanis, 'Cultural diversity and the breakdown of Byzantine power in Asia Minor',
DOP 29 (1975), 9-10.
328 Byzantium in the seventh century

work of Christianity, was monophysitism, which eventually lost the battle

within the East Roman state when the provinces that professed it were lost
to the Arabs. But while Christianity in its neo-Chalcedonian or its mono-
physite form was represented as universal, and was certainly taken by the
establishment of Church and state to be so, non-Christian practices and
traditions seem to have survived in a number of places. There is a
difference, of course, between the subsumption of pagan traditions and
practices, and sometimes cult observances, within a Christian framework
and the survival of pagan cults as independent active forms of religious
devotion and expression. The former represented a process which was, for
the great mass of the population of the ancient and early medieval world,
an inevitable feature of the transfer of devotion from one - syncretic and
pluralistic - set of religious narratives and symbolic systems, to the
explicitly anti-pluralistic soteriology of Christianity. In the West, the
papacy explicitly permitted the incorporation of Christianised pagan prac-
tices or calendar events as a means of promoting the conversion of the
barbarians. 5 In the East, such practices and traditions constituted an
unacceptable challenge to Christian universality, and an intolerable threat
to its claim to mediate between man and God. They had to be dealt with
accordingly. In the context of the seventh century, the opposition between
these two poles appeared even more starkly.
Pagan cults and practices were thus from the mid-fourth century pros-
cribed. As the imperial Church evolved, along with the increasing identifi-
cation of the state with the Christian oikoumene, so the legal condemnation
and persecution of heathens and 'pagans' increased in intensity, and those
who did not openly proclaim their Christianity were progressively
excluded from the apparatus of the state. The practice of 'magic', of taking
haruspices, and related observances were strictly forbidden. Pagan temples
and sites of worship were destroyed, sometimes being deliberately replaced
with Christian buildings. Officials found guilty of participating in or con-
doning pagan rituals were dismissed, as were Christians who were found
to have lapsed into heathen ways. 6 It was, of course, impossible to
eradicate such ancient traditions in a matter of a few years or even over
two centuries, forming as they did central elements in people's day-to-day
experience and practice and in their narratives of that experience and of
5
The policies of the early Church in Kent provide a good example: see Brehier and Aigrain,
p. 286.
6
For a general survey of the state's response to the traditional religions and cults, see K.L.
Noethlichs, Die gesetzgeberischen Massnahmen der christlichen Kaiser des 4. Jahrhunderts gegen
Hdretiker, Heiden und Juden (Cologne 1971). For the closing or destruction of temples,
proscription of cult practices and 'magic', see CTh. XVI, 10.4 (a. 354) and XVI, 10.16 (a.
399), with IX, 16.1 and 2. For dismissal of officials and the formal declaration of
Christianity, see CTh. XVI, 7.5 (a. 391).
 Religion and belief 329

their place in the world. On the other hand, Christianity was itself a
dynamic social and ideological force, and the repressive measures of the
state, while they certainly promoted the interests of the Church, were not
themselves alone responsible for the long-term development and success of
the new soteriology.
'Pagan' beliefs and practices did, therefore, continue to exist within the
late Roman and Byzantine world well into the seventh century at least,
and in certain more isolated areas probably longer. It is possible to dis-
tinguish two forms of 'survival', however. On the one hand, we have the
conscious efforts of members of the social elite and the literate class of the
state to maintain, somewhat self-consciously, an antique way of life and
system of beliefs - these are chiefly members of the senatorial elite, of the
academies with their ancient traditions, and of the educated urban
bureaucracy. This form certainly disappears during the later sixth and
early seventh centuries. On the other hand, there are the unreflective
values and traditional practices of rural populations, particularly in the
more remote areas of the empire, or in regions where local cults, for
example, retained their social relevance and vigour. 'Pagan' is in this con-
text a misnomer - indeed, the continued use of the term by historians today
reflects a somewhat uncritical, and very unfortunate, because quite mis-
leading adoption of a medieval term of reproach and condemnation.
In the former category belong the Academy of Athens, for example,
closed by Justinian in 529, and the numerous members of the senatorial
and governing elite, many of them purportedly Christians, indicted and
tried for paganism during his reign. 7 Indeed, Justinian carried out a series
of purges, re-enacting and reinforcing the anti-pagan legislation of his
imperial predecessors.8 Cult centres such as Baalbek/Hierapolis or
7
See A. Cameron, 'The last days of the Academy of Athens', Proc. Cambridge Philol. Soc. 195
(1969), 7-29, and Malalas, 451. See also P.Lemerle, Le premier humanisme byzantin. Notes
et remarques sur enseignement et culture a Byzance des origines au Xe siecle (Paris 1971),
pp. 69f.; H. Blumenthal, '529 and after: what happened to the Academy?', B 41 (1978),
369-85; and for Justinian's persecution of senators and bureaucrats, see Jones, LRE, vol. I,
pp. 285ff. In 529 he ordered the baptism of all pagans and non-Christians under threat of
sequestration of property and dismissal from office: CJ I, 11.1-10. See also the survey of
J. Irmscher, 'Heidnische Kontinuitat im justinianischen Staat', in Seventeenth International
Byzantine Congress, Major Papers, pp. 17-30.
8
Jones, LRE, ibid.; Stein, Bas-Empire, vol. II, pp. 369fT.; note J. Constantelos, 'Paganism and
the state in the age of Justinian', Catholic History Review 50 (1964), 372-80. This affected
naturally enough many other aspects of traditional elite culture. Classical learning,
inevitably associated in the Christian mind with the pagan past, tended to take an
increasingly background role. Classical patterns of presentation and style retreat before the
forms and modes of scriptural and patristic writing. While this does not necessarily mean
that classical learning was driven out - as has sometimes been argued - it does suggest that
it lost its formerly privileged position. Knowledge of the classics, of Attic style, and of
classical mythology still retained a crucial significance and symbolised an important
element in the self-image of the Byzantine ruling elite. But it had henceforth to be presented
330 Byzantium in the seventh century

Carrhae/Harran attracted a great deal of imperial attention, of course, and

during the sixth century there were a total of seven purges there and in the
surrounding districts.9 But less well-known centres were also purged, and
the legislation of the sixth century demonstrates clearly enough that
'paganism' was still regarded as a living force in many parts of the empire.
The cult centres at Heliopolis and Harran were still active at the end of the
reign of Maurice; the latter occurs in the early Islamic tradition as a well-
known pagan centre which flourished well into the Islamic period. The
tenacity of such ancient religions makes it clear that legislation and per-
secution alone were rarely enough to destroy such practices and the beliefs
that promoted them. 10 On the other hand, it is clear from the response of
the urban population of cities such as Antioch, who was outraged by the
continued practice of pagan communities, that Christianity was the major-
ity religion and that pagan centres were intensely localised phenomena.11
'Pagan' practices, as defined by contemporaries, existed on a more
widespread - a more popular - basis in parts of Asia Minor, however. In
542 Justinian commissioned the monophysite John of Ephesus to carry out
a large-scale missionary expedition in western Asia Minor, as a result of
which tens of thousands, according to contemporary sources, were
brought into the Church. His missionary activity continued throughout
the reign of Justinian and into that of Justin II, chiefly in the provinces of
Caria, Asia, Lydia and Phrygia.12 Similar persecutions and missions took
place in other parts of the empire, notably Syria and Egypt, although
aimed usually at specific cult centres,13 while in the Western provinces
isolated elements of pagan cults continued to exist, although the evidence
is sparse. According to a tenth-century source, the (non-Slav) inhabitants
within an entirely Christian and Christianised context. See Averil Cameron, Agathias
(Oxford 1970), esp. pp. 89ff.; H.-G. Beck, 'Bildung und Theologie im fruhmittelalterlichen
Byzanz', in Polychronion. Festschrift F.Dolger (Heidelberg 1966), pp. 72-81; Lemerle,
Premier humanisme, pp. 85-7; Cameron, 'Images of authority', 4 and 27f.; and 'New and
old in Christian literature', in Seventeenth International Byzantine Congress, Major Papers,
pp. 45-58.
9
Jones, LRE, vol. II, p. 943; also Rochow, Die Heidenprozesse unter den Kaisern Tiberius II
Konstantinos und Maurikios, pp. 120ff. See Evagrius, V, 18 (212-14); John of Ephesus,
114ff.
10
See Rochow, Die Heidenprozesse, pp. 126ff.; and Cambridge History of Islam, vol. II: The
Further Islamic Lands. Islamic Society and Civilisation, eds. P.M. Holt, A.K.S. Lambton and
B. Lewis (Cambridge 1970), pp. 786-7, on the Sabians (§abi'a) of Harran.
11
Rochow, Die Heidenprozesse, pp. 128f. See also Herrin, Formation of Christendom, pp. 184f.,
for examples of high-ranking pagans in Edessa and Antioch including, according to John
of Ephesus and Evagrius, the patriarch Gregory of Antioch. See in general F.R. Trombley,
Hellenic Religion and Christianization c. 3 70-529 (Religions in the Graeco-Roman World CXV.
Leiden 1993).
12
See Jones, LRE, vol. I, pp. 285f. and Rochow, Die Heidenprozesse, pp. 121ff. with sources
and literature.
13
See Jones, LRE, vol. II, pp. 942-3 (the centre at Philae on the Nile, and the centre and
temple of Jupiter Ammon in the Augila oasis). See also Stein, Bas-Empire, vol. II, pp. 300-1
and 372.
 Religion and belief 331

of the Mani in the Peloponnese were Christianised only during the reign of
Basil I, 14 while local cults in Thrace and Macedonia may also have
persisted. Here, however, the arrival of Slav peoples may well have
overwhelmed the old tradition, and it was the missionary activity of the
eighth and, especially, the ninth centuries which reintroduced Christianity
on a widespread basis. Paganism in Africa - excluding the case of Egypt -
seems on the whole not to have been a significant problem, and to have
been anyway confined to the Berber population of the hinterland.15 But it
is clear that, right to the end of the sixth century, pagan cults of one sort or
another continued to exist in many parts of the empire and, in certain
cases, to enjoy considerable local support; and that a wide range of
pre-Christian practices continued to make up an important element of the
day-to-day belief patterns of a great number of the ordinary population of
the empire.16
It is not surprising, therefore, that evidence for pagan survivals, for
'magic' and 'superstition' is still to be found in seventh-century sources;
and indeed, the conditions brought about by the Arab incursions and the
consequent disruption of life in the provinces as well as in major urban
centres - with the interruption in the activities of the local clergy and the
episcopate - may have permitted many ancient traditions, which had
survived in one form or another as long as this, to experience a revival of
sorts.17 The arrival of new populations - notably Slavs - either as immi-
grants into the Balkans or as the result of forced transfers to depopulated or
underpopulated areas of Asia Minor, must also have affected the areas in
which they were settled. Some older, non-Christian religious and linguistic

14
DAI 50.71-6; see D.A.Zakythinos, Le Despotat grec de More'e, vie et institutions, revised edn
Chr. Maltezou (London 1965), pp. 6-14 and 381f.
15
See for a good general survey Rochow, 4Zu einigen oppositionellen religiosen Stromun-
gen\ 245-50. The slow assimilation and Christianising of the South Balkan Slavs had
already begun in the middle of the seventh century. The leader of the Rouchinai in 674,
Perbund, knew Greek, and this may reflect a Hellenisation of some leading elements
among the immigrant clans. Whether this involved Christianisation is difficult to say. See
Lemerle, Les plus anciens recueils, vol. II, pp. 113-14; P.Lemerle, 'La Chronique
improprement dite de Monemvasie. Le contexte historique et legendaire', REB 21 (1963).
5-49, esp. 18fT. Note for the following period also F. Dvornik, Les Slaves, Byzance et Rome
au lXe siecle (Paris 1926), pp.235f. For Africa, see Jones, LRE, vol. II, p. 942; and see
Cameron, Byzantine Africa, p. 40.
16
See the apposite remarks of Jones, LRE, vol. II, p. 941.
17
This is, at least, the implication of canons 8 and 18: the former attempts to re-establish the
practice of calling yearly synods for each ecclesiastical province, abandoned due to enemy
raids; the latter orders the clergy who had abandoned theirflocksfor the same reasons (or
claimed to have done so) to return to their parishes. See H.J. Magoulias, 'The lives of
Byzantine saints as a source of data for the history of magic in the sixth and seventh
centuries A.D.: sorcery, relics and icons', B 37 (1967), 228-69. More recently, F.R.
Trombley, The Survival of Paganism in the Byzantine Empire during the pre-lconoclastic period
(540-727) (Ann Arbor 1981) is useful; but on the context and significance of these
'survivals', see below.
332 Byzantium in the seventh century

traditions may well have been dealt a fatal blow in this way; more
importantly, the supposed universality of Christian belief within the empire
must have been seriously jeopardised by these migrations, for such new
arrivals can have been only superficially, if at all, Christianised. The
regions in which John of Ephesus carried out his missions seem still in the
later sixth century to have been populated by many non-Christians.
During the trials of large numbers of 'pagans' under Tiberius Constantine
in 579 and the following years, the province of Asia was picked out as a
particular source of paganism. 18 It is highly likely that the traditional cult
practices of such areas continued to receive popular support, in the
countryside at least, and they may very well have been given a new lease
of life by the destruction of urban centres - which had, on the whole,
always been Christianised early and more completely than the surround-
ing country 19 - and by the consequent removal, or at least the weakening,
of Christian supervision.
The probability is that both indigenous and imported non-Christian cults
and beliefs proliferated in many regions of Asia Minor, and certainly in the
Balkans, during the seventh century, especially in those regions which the
clergy had abandoned. 20 But the evidence remains patchy. The canons of
the Quinisext which, as we have seen, set out specifically to regulate the
clergy and the behaviour and practices of the subjects of the empire, are
directed at Christians and tell us little, and then only indirectly, about
non-Christians. What they do illustrate, however, is the degree to which
the population in different regions of the empire (although again, the
canons rarely specify which regions) had subsumed many features of their
traditional religious practices and beliefs into Christianity. As mentioned
already, this was actually encouraged on occasion in the Western Church
as a means of promoting conversion. It may well have been, implicitly, the
case in the East, too, on occasion.
What the evidence does illustrate, importantly, is the difficulty faced by
any commentator, modern or medieval, in disentangling the complex
elements which go to make up any given set of beliefs. Christianity also had
its demons and evil spirits, its exorcisms and its rites of purification or
damnation, many of which were in effect no different from those of pagan
belief, at least at the descriptive level. Members of late Roman and early
Byzantine society, from the highest and most privileged to the lowest and
18
See John of Ephesus III, 33sq.; and Rochow, Die Heidenprozesse, pp. 12Off.
!9 See Jones, LRE, vol. II, pp. 942f.
20
The cult of Cybele, for example, which received great support from the local population of
Caria and which was condemned by John of Ephesus in the sixth century, was still in
existence in the mountains of the same region in the eighth century, as reported by
Cosmas of Jerusalem - see Cosmas Hierosolymitanus, Scholia in Gregorii Nazianzeni
Carmina {PG XXXVIII), 502. See also note 10, chapter 9 above.
 Religion and belief 333

most oppressed, accepted such ideas and practices as perfectly common-

sensical, as part and parcel of day-to-day reality and practice. In Byzantine
terms, the difference between magic and miracle is, quite simply, a func-
tional one: the former serves, and is inspired by, the devil; the latter by God.
The wondrous qualities of icons, relics, saints' tombs, pieces of the cross,
talismans of various sorts, even the power of words recited or chanted at
the correct moment in the appropriate form, this is the 'magic' of the
Christian symbolic universe. The artifacts are different, but the principles,
much older than the Church, are inscribed within the narratives of daily
life for all the members of this early medieval world. And it is when these
elements take on a broader and more public role - when they threaten the
institutionalised and consecrated authority of the Church - that they also
pose a threat to, and attract the (unfavourable) attention of the estab-
lishment. 21 What the canons of the Quinisext clearly reflect is a view
among Churchmen that such practices were taking too significant a role
within the otherwise generally Christian framework of life.22
The practices themselves reveal a great deal about rural belief and
attitudes in the provinces; but what is important about them is that, while
they were certainly condemned by the council as 'Hellenic' or pagan, they
21
These points are made in greater detail and more graphically in the discussion of
H.J. Magoulias, 'The Lives of Byzantine Saints'; see also the comments of Beck, Das
byzantinische Jahrtausend, esp. pp. 264-9; A. Momigliano, 'Popular religious beliefs and
the late Roman historians', Studies in Church History VIII (1971), 1-18; P.R.L. Brown,
'Sorcery, demons and the rise of Christianity: from late Antiquity into the Middle Ages', in
Witchcraft Confessions and Accusations (Cambridge 1970), pp. 17-45; P.P.Joannou,
Demonologie populaire - demonologie critique au XIe siecle (Wiesbaden 1971); C. Mango,
'Discontinuity with the classical past in Byzantium', in Byzantium and the Classical
Tradition, eds. M. Mullett and R. Scott (Birmingham 1981) pp. 48-58 (repr. in Byzantium
and its Image: History and Culture of the Byzantine Empire and its Heritage (London 1 9 8 4
third essay), see pp. 5 5 - 7 ; and esp. Mango, 'Antique statuary and the Byzantine behol-
der', DOP 17 (1963), 5 5 - 7 5 , see 5 9 - 6 4 . For the excellent survey of the form and
functions of talismans in this context see G. Vikan, 'Art, medicine and magic in early
Byzantium', DOP 3 8 (1984), 6 5 - 8 4 ; and for the functions of icons in this context,
C. Belting - Ihm, art. 'Heiligenbild', in Reallexikon fur Antike und Christentum, fasc. 1 0 5
(Stuttgart 1987), 8 9 - 9 5 . The debate impinges, of course, on the whole question of the
nature of medieval religious culture, of the nature of different levels of practice and
understanding, and of different discourses within a shared symbolic universe. See P.R.L.
Brown, 'The saint as exemplar in late Antiquity', Representations 1 (1983), 1-25; and the
survey article of J. Van Engen, 'The Christian Middle Ages as an historiographical
problem', American Historical Review 9 1 (1986), 5 1 9 - 5 2 . For the same problem in the
context of a very different society, see I. Sorlin, 'Femmes et sorciers. Notes sur la perma-
nence des rituels paiens en Russie, XI e -XIV e siecle', TM 8 (1981), 4 5 9 - 7 5 .
22
Compare also the response of the state and the Church to the activities of a priest in the
village of Triachia on Cyprus, whose trial is reported by Anastasius of Sinai to have taken
place in 6 3 8 : the priest had secretly led various magical rites and w a s accused also of
corrupting the holy vessels of the Church liturgy. The prefect of the island himself presided
over the trial, at which the accused was found guilty and condemned to death by burning.
See Beck, Kirche, p. 4 6 4 ; for the text, see F. Nau, 'Le Texte grec des recits utiles a Tame
d'Anastase (le Sinaite)', OC 3 (1903), 5 6 - 9 0 , see 69f.
334 Byzantium in the seventh century

do not in fact represent paganism, that is, non-Christian religions or cults

in operation. There is an important difference between the two - as I have
already tried to stress - and it would, therefore, be a mistake to argue from
the canons of the Quinisext that there was a 'revival of paganism' or
anything of the sort at this time. The canons of the Quinisext represent the
anxiety of Church and state, and their attitudes both to their own control
over the patterns of belief of the populations of the provinces, and to the
ways in which their authority was received among the ordinary people of
the empire.
The practices which were condemned and forbidden had on the whole,
then, nothing to do with paganism - they represented merely the deeply
rooted and age-old traditions of local rural or urban culture, a way of life
which, without being the least bit un-Christian, could hold on to those
features of its pre-Christian traditions which helped it to regulate and
understand the vagaries of daily existence, of the seasons and of natural
and man-made disasters, or which helped protect against, or ward off,
harmful events. The list is revealing. Canon 65 condemns the practice of
building fires in front of houses and workshops at the time of the new
moon, and of leaping over suchfires'according to the old tradition'; canon
61 condemns soothsayers and diviners, fortune-tellers and similar persons,
including those who go about with trained animals such as bears and trick
the simpler folk; canon 50 proscribes gaming with dice, and canon 62
forbade a whole series of 'hellenic' practices: celebrating the feast of the
Brumalia, the Vota, the Calends, and the dances on thefirstday of March;
it also forbade the invocation of the god Dionysus at the grape-crushing.
Similarly, it proscribed the use of comic and satyric masks at such popular
festivals, and in particular the practice of masquerading as members of the
opposite sex. Canon 94 prohibits the use of 'Hellenic' oaths and curses, and
so on.23 All of these practices were clearly part of the traditional seasonal
and yearly rituals of a predominantly Christian population; and while the
Church may have had some successes in stamping them out, the survival
of feasts such as the Brumalia - which was still celebrated at the imperial
court in the mid-eighth century24 - throws considerable doubt on the
efficacy of the rulings. Indeed, the commentary to the canons of the
twelfth-century canonist Balsamon shows quite clearly that very many of
these practices continued unaffected well into the late Byzantine period,
justified on the grounds of 'ancient tradition'.25 The ninth-century Life of
23
Canon 65: Mansi, XI, 973A-B; canon 61: 969E-972A; canon 50: 968A-B; canon 62:
972A-C; canon 94: 984B.
24
See Vita Stephani iun., 1169B-C.
25
Rhalles-Potles, Syntagma, vol. II, see the commentary to canon 24 (p. 359); canon 51
(pp. 424f.); canon 60 (pp. 440f.); canon 61 (pp. 442-7); canon 62 (pp. 448-52). See also
D. Simon, 'Balsamon zum Gewohnheitsrecht', in Xxokia- Studia ad Criticam Interpretatio-
 Religion and belief 335
St Anthony the Younger records that the saint cured a woman's infertility
by writing an incantation on a strip of parchment torn from a bible, a prac-
tice explicitly forbidden in canon 68 of the Quinisext;26 in the early ninth
century St Ioannicius was famed for his powers over wild animals - on one
occasion at a feast he issued commands to a trained bear, again a practice
expressly forbidden in canon 61. 27 The same canon is aware of the dangers
of such practices for the ordinary population - as is Theodore of Studion
over a century later;28 yet only a few years later, the hagiographer of
Michael of Synnada could write that his saint worked miracles in order to
instil fear and respect into the uneducated.29 The plain fact is that these
practices were widespread, ancient and deep-rooted; legislation alone was
unable to alter this; and they had been thoroughly integrated into the
pattern of daily life.
Many other traditions, including the use of masks, the masquerading in
the clothing of the opposite sex, the lighting of fires and so on, have
remained part and parcel of peasant culture in the Balkans and Anatolia to
the present day.30 They have been thoroughly absorbed into a Christian
framework of beliefs and a Christian calendar. Other practices reflect, simi-
larly, folk-traditions which existed independent of any specific religion or
cult, but which were clearly regarded for convenience as 'pagan' by the
authorities. Canon 60 condemns the practice of imitating a trance-like state
and imposes exorcism on the practitioners who thus permit evil spirits to
take hold of them.31 Mimes, popular theatre, pantomime and wild-animal
hunts were similarly proscribed.32 Yet the satirical theatre, certainly, con-
tinued throughout the Byzantine period, particularly in the cities, and dealt
with popular and contemporary themes - political affairs, the character
and mores of foreigners such as Armenians or Arabs, even aspects of New
Testament history and, perhaps inevitably, the Church and its clergy.33
nemque Textuum Graecorum et ad Historiam luris Graeco-Romani Pertinentia Viro Doctissimo
D. Holwerda Oblata, eds. W.J. Aerts, J.H.A. Lokin, S.L. Radt and N. van der Wai (Groningen
1985), pp. 1 1 9 - 3 3 , for Balsamon's understanding of 'customary law' and its relationship
to the civil law.
26
F. Halkin, 'Saint Antoine le Jeune et Petronas le vainqueur des Arabes en 863', AB 6 2
(1944), 1 8 7 - 2 2 5 , see 195; Mansi XI, 973C-D.
27 Vita S. loannicii, in AS, Nov. H71, Vita a Petro, 3 8 4 - 4 3 4 ; see 3 9 0 B - C 4 1 2 C - 4 1 3 B .
28 S. Theodori Studitae Epistolarum Lib., I {PGIC, 9 0 4 - 1 1 1 6 ) , ex. XIX (965D-968B).
29
The Life is to be found in C. Doukakis, Meyas ovvagapiorrjs irdvTayv T(OP ayitov (12 vols.,
Athens 1 8 8 9 - 9 6 ) , Maii, 4 1 1 - 2 2 , see 4 1 7 ; see BHG3, 2 2 7 4 x .
30
See, for example, J.du Boulay, Portrait of a Greek Mountain Village (Oxford 1974), esp.
pp. 60ff., and on local taboos and myths, such as fortune-telling, pp. 64ff. See also M.F.
Herzfeld, The Poetics of Manhood. Contest and Identity in a Cretan Mountain Village (Prince-
ton, N.J. 1985), esp. pp. 238ff.
31 32
Mansi XI, 969D. Canon 51: Mansi XI, 968B.
33
See F.Tinnefeld, 'Zum profanen Mimos in Byzanz nach dem Verdikt des Trullanums
(691)', Bv£av7iva 6 (1974), 325ff.; and esp. Ja.N. Ljubarskij, 'Der Kaiser als Mime: zum
Problem der Gestalt des byzantinischen Kaisers Michael m \ JOB 37 (1987), 39-50;
336 Byzantium in the seventh century

The effectiveness of the canons needs to be seen in this context, therefore;

and it should be remembered that most of the prohibitions taken up by the
Quinisext were, in fact, reiterations of earlier prohibitions going right back
to the fourth century, measures which had clearly had only very limited
success.34
The great majority of the practices condemned in the canons of the
Quinisext, therefore, are 'pagan' primarily because the ecclesiastical estab-
lishment was bound to identify them as non-Christian and, on a variety of
grounds, morally questionable, if not downright dangerous. But their
existence both before and after the council need not in itself cause us to
doubt that Christianity was by this time the dominant belief system in the
early Byzantine world. Pockets of 'real' paganism - actual living cults, as
opposed to occasional festivities and so on - undoubtedly did survive, even
if we know next to nothing about them (the cult of Cybele in Caria was
probably not an isolated phenomenon). 35 But the practices condemned by
the Quinisext throw more light upon the traditional beliefs and super-
stitions of ordinary people who were also Christians, than upon paganism.
And in this respect, it is perhaps time that historians abandoned the
terminology of their medieval sources and viewed these practices for what
they were (and are): essential elements of the social reproductive practice
of the ordinary rural and urban populations of the empire. The anti-pagan
legislation of the sixth century and before is not repeated in later codifi-
cations; 36 the Quinisext canons themselves make no mention of pagans as
such, only 'Hellenic', which is to say, pagan, practices. Later Church
legislation concentrates on the prevalent heresies of the time, but

see in particular 4Iff. and 4 4 - 6 . Note also the comments of Zonaras, the twelfth-century
canonist and jurist, in Rhalles-Potles, Syntagma, vol. Ii, pp. 424ff. Those bishops who
attended the Quinisext and w h o were familiar with life in the capital, will have been very
well aware of the continued popularity of street theatres and related entertainments. If the
fictional Life of Andrew the Fool is to be placed in the later seventh century, as Mango has
reasonably argued (see introduction, above), there is plenty of evidence for this: see Vita
Andreae Sali 648D, 652C and so on.
34
Cf. Jones, LRE, vol. II, pp. 977f.
35
The attitudes and fears of contemporaries and later commentators is well summed up in
the gruesome story of the sacrifice of an unborn baby by the Byzantine population during
the siege of Pergamon in 716. For, as has n o w been shown, the actual events were taken
from an older apocalyptic tradition by an anonymous writer contemporary with the siege
of Constantinople in 7 1 7 - 1 8 , and later incorporated into the historiographical tradition
of Theophanes and Nicephorus. They did not actually take place; but they symbolised the
anxieties of the time both with regard to the ancient pagan rites and to the barbarians
w h o threatened the empire. See W. Brandes, 'Apokalyptisches in Pergamon', BS 4 7
(1987), 1 - 1 1 .
36
Only one piece of Justinianic anti-pagan legislation was taken up in the later ninth-
century Basilika, cf. C/1,11.9, a law forbidding the testamentary transmission of property
to persons or institutions which support pagan cults and beliefs. Why this pronouncement
alone was taken up is not immediately apparent. Cf. Bas., I, 1.14 and II, 3.7.
 Religion and belief 337

paganism is not mentioned at all, except in the commentaries to the

canons. Where pagan cults did survive, therefore, they must have done so
in relative isolation; but it is unlikely that they survived long after the
seventh century.
It seems quite reasonable to suppose that the conditions of the second
half of the seventh century had exacerbated the problems of the Church in
respect of the maintenance of Christian standards of morality and practice.
The absence over several months or even years of the local clergy in areas
most exposed to Arab or Slav activity can hardly have promoted a
reinforcement of Christian mores. 37 On the other hand, the traditions,
practices and beliefs which the canons of the Quinisext condemn are likely
to have been an integral part of everyday life, in one form or another
according to local cultural traditions, throughout much of the empire in
normal times, and it was the conditions of the period in question which
drew the attention of the Church to such matters, as much as the existence
of the practices themselves. It is important to remember that there exists
within any symbolic universe a whole series of registers of beliefs and
practices, from the 'pure' and formalised theology and religious-philoso-
phical logic of thinkers and intellectuals - theologians - right down to the
day-to-day beliefs and attitudes of the uneducated and illiterate rural and
urban populations. Such beliefs and ideas, and the practices which accom-
panied them, may well employ the vocabulary and the imagery of the
formal system within which they are subsumed, but may themselves be
both older, and fulfil a different functional role in the society as a whole,
than the formal system. In the case of medieval Christianity, and in spite of
the formal claims of Eastern orthodox political theology, there always
existed a pluralism of beliefs, or registers of belief, of this sort. The con-
tinued existence of 'Hellenic' traditions is illustration enough. And it is at
root not a question of the 'survival of paganism'.

Heretics and Jews

More important for the state, politically and also ideologically, were the
various heretical groups which existed within the empire, important
because they posed a more immediate and coherent challenge to neo-
Chalcedonian orthodoxy, and thereby to the authority of the emperors. 38
37
See V. Laurent, 'L'CEuvre canonique du concile in Trullo (691-2), source primaire du droit
de l'eglise orientate', REB 23 (1965), 10ff.; see also F.Trombley, T h e council in Trullo
(691-2): a study of the canons relating to paganism, heresy and the invasions', Comitatus
9 ( 1 9 7 8 ) , 1-18, see 1 1 - 1 3 .
38
For a good discussion of the origins of the term 'heresy' in Christian thought, see
F. W i n k e l m a n n , 'Einige Aspekte der Entwicklung der Begriffe Haresie u n d Schisma in der
Spatantike', Koivuvia 6 (1982), 8 9 - 1 0 9 ; a n d see the useful discussion of E. Patlagean,
338 Byzantium in the seventh century

But the most dangerous of these, monophysitism, became ever less rele-
vant during the course of the seventh century. True, the imperial govern-
ment and Church probably never wrote off entirely the possibility of
reconquering the lost Eastern provinces in which monophysitism was the
dominant creed. But by the 640s imperial attempts to find and impose a
compromise - in the event, monotheletism - reflected rather the per-
ceptions of those in power that the imperial authority was endangered,
than the real desire to win over the populations of Syria and Egypt.
Armenia remained a problem, of course, but again, a distant imperial
power, even with the occasional use of force, could not seriously hope to
compel a monophysite population to accept imperial and Chalcedonian
authority. 39 And in the event, the alienation of the non-Byzantine Arme-
nian nobility and the much cleverer politics of the caliphate lost this
region, too. No doubt monophysitism remained among elements of the
population of Asia Minor - according to John of Ephesus, Cilicia, Isauria,
Asia and Cappadocia all possessed strong and flourishing monophysite
communities. Other well-known monophysites, such as Jacob Baradaeus,
had also been active in converting pagans to monophysitism in Asia
Minor. Jacob travelled in Cappadocia, Cilicia, Isauria, Pamphylia, Lycao-
nia, Lycia, Phrygia, Caria, Asia, as well as on Chios, Rhodes and Lesbos.
Monophysite bishops were consecrated for many of the cities of these
regions. 40 And it is most unlikely that these communities did not survive
well into the seventh century and beyond. But the effects of the Arab
invasions and attacks and, in particular, the loss of the monophysite
provinces, cut such communities off from the wider world which had
nourished them hitherto. They can only have survived - where they were
not destroyed or fragmented through hostile military activity - in relative
isolation, and the pressing insistence of the state on orthodoxy within its
own apparatus, at least for the vast mass of the bureaucracy, cannot have
contributed to the maintenance of their ideological integrity. At any rate,
while monophysite sentiment continued to exist within the empire, the
canons of the Quinisext suggest that it played only a minor role. Only two
canons, 81 and 82, reflect a clearly anti-monophysite tendency, the first
'Byzance, le barbare, l'heretique et la loi universelle', in Ni Juif ni Grec. Entretiens sur le
racisme, ed. L.Poliakov (Paris 1978), pp. 8 1 - 9 0 ( = Structure sociale, famille, chretiented
Byzance (XV London 1981)).
39
Note canons 32, 33 and 99 of the Quinisext, which attempt to bring the Armenian
Church into line in respect of certain matters of liturgical practice. See also the comments
of Balsamon: Rhalles-Potles, Syntagma, vol. II, pp. 373-81 and 543-4.
40
See Frend, The Rise of the Monophysite Movement, pp. 285ff.; and John of Ephesus I, 39, II,
4, HI, 36f., IV, 19 and V, 6; and especially John of Ephesus, Lives of the Eastern Saints {PO
XVIII), 535f.; (PO XIX), 154ff. Monophysite communities existed also in Constantinople
and on Cyprus - see Van Dieten, Geschichte der Patriarchen, pp. 28-9; and John of Ephesus
ffl, 15-16.
 Religion and belief 339

prohibiting the addition of the phrase 'who was crucified for us* to the
creed, a theopaschite formulation which reflected a monophysite theol-
ogy; the second prohibiting the depiction of Christ as a lamb, a practice
which, it was argued, jeopardised the assumption of two natures, human
and divine, of Christ.41 Canon 95 also implies the continued existence of a
variety of monophysite sects and tendencies, although the list must be treated
with caution. In listing the process whereby heretics who wish to join or
rejoin the true faith might be admitted to the Church, the canon notes that
for certain groups, Severus of Antioch and the archimandrite Eutyches, two
of the best-known monophysite polemicists, were to be first condemned
(along with the patriarch Nestorius) before they could be accepted.42 It is prob-
able that various minor sects, which still held views that might be charac-
terised as either monophysite or Nestorian, were intended here. On the
whole, however, monophysites do not seem to be a serious concern, and
indeed the fact that the canon stipulates the requirements for admission may
well suggest that such monophysites as remained within the empire were
being gradually assimilated into the Chalcedonian fold.
The hostile activity of the Arabs may well have played a central role
here, for the devastation they brought about seems to have dislodged large
numbers of the rural populace in many areas, forcing them to seek refuge
in less exposed districts. Interestingly, canon 95 notes that numbers of
heretics - including Eunomians, Sabellians and Montanists - had left the
region of Galatia and sought baptism; and it is highly likely that this
movement was a result of the economic and physical dislocation caused by
raiding Muslim forces. More importantly, it forced hitherto isolated rural
communities, who had probably been able to practise their form of Chris-
tianity with little interference, into the 'outside' world of neo-Chalcedonian
orthodoxy, where the differences and divergences were more apparent,
attracting the attention of the authorities. This general phenomenon must
41
See Mansi XI, 977D-E and 977E-980B.
42
Mansi XI, 984B-E. For Eutyches and Severus, see Winkelmann, Die ostlichen Kirchen,
pp. 44 and 81ff.f and 50 and 54ff. For Nestorius, ibid., pp. 36f. The text of canon 95 is itself
taken mostly from earlier sources, chiefly the seventh canon of the council of 381 in
Constantinople, together with other elements such as canon 19 of the council of Nicaea
(325): see Hefele-Leclercq II, 1; Mansi II, 676f.; and Rochow, Zu einigen oppositionellen
religiosen Stromungen, p. 266. Although monophysitism continued to be a target of
theological polemic well into the seventh century (along with Arianism and other
Christological heresies), as can be seen from a survey of the relevant literature in CPG III,
this reflects the fact that it was seen as part of the monothelete debate, as well as the fact
that the most important of these writings came from outside the empire, where the debate
between Chalcedonian and monophysite Christians remained a live issue - the Hodegos of
Anastasius of Sinai is perhaps the best example of this. See K.-H. Uthemann, Anastasii
Sinaitae Viae Dux (Corpus Christianorum, series Graeca, vol. VIII) (Turnhout 1981); and
see CPG III, nos. 7685 (Euboulus of Lystra), 7745, 7756-7, 7771 (?) (Anastasius of
Sinai), 7798 (anon.).
340 Byzantium in the seventh century

surely have applied to other non-Chalcedonian or heretical groups. On at

least one occasion, and possibly more often, we know that considerable
numbers of refugees from a famine in Syria entered the empire (in 686,
according to Theophanes), and this was yet another source of potential
difference or disruption which the Church had to handle.43
It seems not unreasonable to suppose, therefore, that it was in fact the
disruption of the seventh century which produced a greater uniformity of
belief within Byzantine Asia Minor.
Other heresies and schismatic sects existed across the empire, of course.
In North Africa, Donatism - which had come to represent local cultural-
political feeling as well as religious dogmatic differences of opinion - seems
to have experienced a very limited revival during the later years of the sixth
century, if the letters of Pope Gregory are to be given credence; although
their numbers seem to have been small, and their influence limited.44 But
whether the people in question are really Donatists, or whether the term is
simply used of non-Chalcedonian heretics generally, is uncertain. Even so,
a reference to Donatists in North Africa in a letter of Pope Gregory II for the
year 722 suggests that it may have survived, albeit in isolation and on a
very limited scale, throughout the period with which we are dealing.45
Arianism likewise survived on a limited scale, chiefly among Germanic
mercenaries in the imperial armies in the later sixth and early seventh
centuries. John of Ephesus reports the presence of an Arian church for the
German community in Constantinople, and the popular hostility to it. 46
But this was an isolated incident; and while Arianism may have persisted in
the West, particularly in North Africa after the defeat of the Vandals, and in
Italy among the Lombards until 680, 4 7 it seems to have disappeared from
43
Mansi ibid., repeatedfromthe 7th canon of the council of 3 81: see Rhalles, Potles, Syntagma,
n, 529-3If.; for the Syrians, see Theophanes, 364.3-4; and the seventh-century Syriac
chronicle of Johannan bar Penkaye - see G.J. Reinink, 'Pseudo-Methodius und die Legende
vom romischen Endkaiser', 95 and note 62.
44
See W.H.C. Frend, The Donatist Church: A Movement of Protest in Roman North Africa
(Oxford 1952), pp. 308-11; Diehl, L'Afhque byzantine, vol. II, pp. 5O8ff.; Br&iier and
Aigrain, pp. 217-18; Goubert, Byzance avant I'Islam, vol. II, 2, pp.229ff.; and B.H.
Warmington, The North African Provinces from Diocletian to the Vandal Conquest (Cambridge
1954), pp. 76-102. R.A. Markus, 'The imperial administration and the Church in
Byzantine Africa', Church History 36 (1967), 18 argues that Donatism in its proper sense
no longer existed at this time.
45
See PL LXXXIX, 502; and Frend, The Donatist Church, p. 313; Brehier and Aigrain, p. 228.
note 3. The debate is complex and depends also on the question of Gregory's use of the
term for his own ecclesiastical-political ends. See esp. R.A. Markus, 'Christianity and
dissent in Roman North Africa: changing perspectives in recent work', Studies in Church
History IX (1972), 21-36; P.R.L. Brown, 'Religious dissent in the later Roman empire: the
case of North Africa', History 46 (1961), 83-101; and the summary with literature in
Averil Cameron, 'Byzantine Africa', 49f.
46
John of Ephesus III, 13 and 26, and V, 16.
4
? See W.E. Kaegi, jr., 'Arianism and the Byzantine army in Africa, 533-546', Traditio 21
(1965), 23-53; and for Italy, K.D. Schmidt, Die Bekehrung der Ostgermanen zum Christen-
 Religion and belief 341
within the empire, including its Spanish territories (held until c. 625: the
final conversion of the Visigothic Church to Chalcedonian orthodoxy took
place at the council of Toledo in 589)48 by the time of the accession of
Heraclius.
The Montanist heresy, which had its centre in Phrygia, may have
lingered on into the seventh and early eighth centuries, although the
evidence is inconclusive: it is difficult to ascertain whether a source such as
Theophanes means the real Montanists or merely uses the term as a
convenient and well-known element in a list.49 Later legislation - notably
the condemnation of Montanists and Manichaeans in the Ecloga - seems
to rely on the Justinianic codification for its inspiration, and the term is
used once more as a convenient 'catch-all', as may also be the case with
the 95th canon of the Quinisext. Montanism certainly seems to have
existed into the late sixth century, however, as John of Ephesus, who led a
major drive against their cult centres and converted many to orthodoxy
(i.e. monophysitism), reports in detail. But his missionary activities,
together with thefiercepersecutions under Tiberius Constantine, may well
have dealt the fatal blow to the Montanist communities.50
Similar considerations apply to the Messalians, sometimes called Mar-
cianists, both terms used in the sixth and seventh centuries for defamatory
purposes, rather than signifying actual practices or beliefs. Messalianism
originated as an ascetic heresy in Syria during the fourth century; but by
the later sixth century the term seems to have lost its real significance.51
And the fact that it occurs in the scholia of Maximus Confessor to the
turn (2 vols. Gottingen 1939), vol. I, pp. 387ff.; S.C. Fanning, 'Lombard Arianism
reconsidered', Speculum 56 (1981), 2 4 1 - 5 8 .
48
See E.A. Thompson, The Goths in Spain (Oxford 1969), pp. 320ff.; P.Goubert, Influences
byzantines sur l'Espagne wisigothique (554-711)', REB 2 (1944), 5-78, see 4 3 - 9 ; and
the council of Toledo: Mansi IX, 985f.
49
They were heavily persecuted by Justinian; see Stein, Bas-Empire, vol. II, pp. 374f.; P. de
Labriolle, La Crise montaniste (Paris 1913), pp. 532ff. For their origins in the second
century, see esp. S. Runciman, The Medieval Manichee. A Study of the Christian Dualist
Heresy (Cambridge 1947), pp. 18ff.; F.E. Vokes, 'The opposition to Montanism from
Church and state in the Christian empire', Studia Patristica IV, 2, ed. F.L. Cross (Berlin
1961), pp. 518-26; W. Schepelern, Der Montanismus und die phrygischen Kulte (Tubingen
1929). For the story of Theophanes (according to which an edict of Leo III in 722 ordered
the forced baptism of all Montanists, and as a result of which the adherents to the heresy
locked themselves in their churches and set them on fire, rather than be converted), see
Theophanes, 4 0 1 . But see also A. Sharf, 'The Jews, the Montanists and the emperor Leo
III', BZ 59 (1966), 37-46, see 43; and Procopius, Historia Arcana XI, 23; Stein, Bas-
Empire, vol. II, p. 375, which may be Theophanes' original source. Runciman, The
Medieval Manichee, p. 18, accepts the story, as do Brehier and Aigrain, p. 439.
50
See Ecloga XVII, 52; and J. Gouillard, 'L'Heresie', 308-9. For John of Ephesus, see Michael
Syr., vol. II, 269ff.; John of Ephesus III, 13, 20, 2 1 , 26 and 32. In general, see Sharf, 'The
Jews, the Montanists and the emperor Leo III', esp. 38f.
51
See the summary and the literature in Rochow, 'Zu einigen oppositionellen religiosen
Stromungen', 2 7 4 - 8 .
342 Byzantium in the seventh century

Pseudo-Dionysius Areopagite - 'Lampetianoi, that is, Messalianoi or Adel-

phianoi, which is to say, Markianistai' - is no evidence for the existence of
Messalian heretics or communities. 52 Part of the difficulty in identifying
such groups lies in the fact that orthodox critics of aberrant beliefs and
heresies tended to use the descriptive terms of earlier writers and of an
earlier period in their lists and classifications of heretics and heretical
beliefs, and it is difficult to know to what extent these lists represent a
contemporary situation, or merely reflect an antiquarian and condemna-
tory topos. Older terms were often applied, with little or no theological
justification, to quite new groups of heretics. The term 'Messalian' was
thus used also of Bogomils in the later period, while terms such as
Montanist or Arian occur similarly alongside other names of actual, living,
heresies from the seventh century on. 53 Thus references to the Meletians,
Paulianists, Marcionists, Sabbatians, Novatians, Quartodecimans (Tessa-
rakaidekatitai), Eunomians, Sabellians and many other sects must be
treated cautiously. 54 In the early seventh-century handbook of the presby-
ter Timothy of Constantinople, De receptione haereticorum, and in the 95th
canon of the Quinisext, such sects are mentioned without any real evi-
dence that they actually survived. What both texts represent is a catalogue
of traditional heresies, subdivided into three categories according to their
degree of difference from orthodox Christianity and the mode in which
their adherents are to be readmitted to the Church. 55 Similarly, in the
anti-heretical compilation of Anastasius of Sinai and in a (probably)
seventh-century text on the Messalians by the monk and presbyter George,
lists of heretics also occur, and it is clear that in part, at least, the polemical
tradition and the intention of the authors of such works have determined
the composition of the lists themselves.56
52
PG, IV, 169. Messalianism seems to have died out by the seventh century - see Rochow,
'Zu einigen oppositionellen religidsen Stromungen', 2 7 8 and note 1.
53
D. Obolensky, The Bogomils (Cambridge 1948), pp.240f.; Runciman, The Medieval
Manichee, pp. 63ff.; M. Loos, Dualist Heresy in the Middle Ages (Prague 1974), pp. 72 and
33Off.; Rochow, 'Zu einigen oppositionellen religiosen Stromungen', 2 7 4 note 1.
54
For the origins and history of these sects, see Dictionnaire de theologie catholique, eds.
A. Vacant, E. Mangenot and E. Amann (Paris 1930), vol. DC, 2, 2 0 0 9 - 3 2 (Marcionism);
1 4 6 4 - 7 8 (Macedonianism); XIII, 2, 1445ff. (Quartodecimans); XIV, 1, 4 3 0 - 1 (Sabbatia-
nism); T. Gregory, 'Novatianism. A rigorist sect in the Christian Roman empire', Byzantine
Studies 2 (1975), 1-18; also Lexikonfiir Theologie und Kirche, eds. J. Hofer and K. Rahner,
2nd edn (Freiburg i. Br., 1 9 5 7 - 6 8 ) , vol. VI, 1313f. (Macedonianism); VII, 257f. (Mele-
tians); IX, 192 (Sabbatianism); VIII, 213f. (Paulianism); see also CMH, vol. IV, part 2,
pp. 1 9 0 - 2 ; Beck, Kirche pp. 3 4 9 - 5 0 (Messalianism); also Jones, LRE, vol. II, pp. 9 5 0 - 6 .
" See PG LXXXVI, 1 1 - 6 8 and Beck, Kirche, pp. 4 0 1 - 2 (for Timothy of Constantinople);
Mansi XI, 984fr-E.
56
See PG LXXXIX, 2 6 8 , and Beck, Kirche, pp. 442f. and 4 4 6 ; and for George, see M. Richard,
in EEBX 2 5 (1956), 3 5 0 - 6 2 . See esp. Gouillard, 'L'Heresie', 3 0 2 and 3 0 4 with note 37;
Beck, Kirche, p. 4 4 7 . Note the comments of Rochow, 'Zu einigen oppositionellen religiosen
Stromungen', 2 7 7 .
 Religion and belief 343

While it is quite probable that much of the rural population of Asia

Minor - where these heresies had mainly had their roots in the period from
the second to the fifth centuries 57 - did express their beliefs through ideas
and liturgical practices which might have seemed heretical or aberrant to
the strictly orthodox observer (and to the official Church position), and
while such groups may even have referred to themselves by such sectarian
names, it seems unlikely that any organised movements or sect Churches
existed. Many may have seen their differences from the orthodox estab-
lishment as a symbol of their own cultural traditions - and even at times of
their opposition to aspects of state rule as vehicles for the expression of
their cultural and social solidarity.58 The occurrence of these terms in later
commentators, therefore, especially when used in connection with a
known heretical movement or unorthodox activity, must be viewed with
especial suspicion.59 On the whole, with the exception of the Marcionist
sect, which may have contributed to the development of the Paulician
Church, the great majority of these sects seems to have vanished by the
later sixth century. 60 The use of their names in later sources reflects either
the age-old practice of using old names for new heresies where a connec-
tion, spurious or real, could be found; or their application to regional
customary usages for which the observer may have felt little sympathy. 61
If the great majority of the older heresies had disappeared in all but
name by the later sixth or early seventh centuries, two groups seem to
have their origins in this period. In the first, that of the Paulicians, the
question of dualism in its Manichaean forms was encountered once more.
The confrontation with dualism begins much earlier, of course, in the early
Christian centuries: Manichaeism, with its systematised and sophisticated
theology, gave dualism a new strength and provided later dualist heresies
with a valuable arsenal of arguments. 62
The beginnings of Paulicianism are shrouded in obscurity, and the
question of when and where the sect was first established remains a subject
57
See W.M. Calder, 'The epigraphy of the Anatolian heresies', Anatolian Studies presented to
Sir William Ramsay (Manchester 1923), pp. 5 9 - 6 1 ; H. Ahrweiler, 'L'Asie mineure et les
invasions arabes', 3-5; and for further literature, see Rochow, 'Zu einigen oppositionellen
religiosen Stromungen', 2 7 8 - 8 2 .
58
See Ahrweiler, 'L'Asie mineure et les invasions arabes', 5; Kopstein, 'Zu den Agrarverhalt-
nissen', pp. 3 3 - 4 on the bandit activities of the mountain population of Pisidia and
Lycaonia in the sixth century.
59
Thus the use of the term 'Nestorian' of the iconoclasts in the Life of Peter of Atroa, 22.9
probably has nothing to do with Nestorians as such, pace Vryonis, in BV£OIVTIV6L 1 (1969),
215. See esp. Gouillard, 'L'Heresie', 302ff.
60
See P. Charanis, 'Cultural diversity and the breakdown of Byzantine power in Asia Minor',
who revises his earlier arguments in this respect. See Gouillard, 'L'Heresie', 304 and note 36.
61
Gouillard, 'L'Heresie', 3 0 7 - 1 2 .
62
For dualism, and more particularly Manichaeism, see the literature at Beck, Kirche, p. 335
and commentary and Loos, Dualist Heresy, pp. 2 Iff.
344 Byzantium in the seventh century

of debate. It was a dualist and egalitarian sect, which based its beliefs on an
interpretation of the New Testament, regarding the Old Testament as
irrelevant. Its rejection of baptism and of any formal ecclesiastical hier-
archy brought it into conflict with the established Church and thus with
the state; and in the second half of the seventh century it was heavily
persecuted - the later evidence, at least, suggests that it already had at this
time a considerable popular following in the still limited areas of the
provinces Armenia I and II where it is first encountered. It was, of course,
to become a major threat to imperial territorial and administrative integ-
rity in the ninth century; but at this period although persecutions took
place under Constantine IV and Justinian II, it remained a minor
problem.63
The second heretical sect whose origins have been traced to the later
seventh or early eighth centuries is that of the Athigganoi. They are first
mentioned in the writings on heresy usually ascribed to the patriarch
Germanus (715-30) between 727 and 733. 64 The nature of their beliefs is
unclear, and they have been connected with both the Novatians, Montanists
and Paulicians, and with the Jews. They were associated in Byzantine times
with Phrygia and Lycaonia especially, and the Montanist connection is sug-
gested by the fact that they were sometimes referred to as 'Phrygians', a term
applied also to the older sect.65 But they seem to have been few in number and
relatively insignificant until the middle and later eighth century. From our
point of view, their significance, as that of the Paulicians, lies in their
63
The literature on the Paulician movement and its beliefs is vast. For the most recent
survey, which includes also the older secondary literature, see P. Lemerle, 'L'Histoire des
Pauliciens d'Asie Mineure d'apres les sources grecques', TM 5 (1973), 1-144; N.Gar-
soian, 'Les Sources grecques pour l'histoire des Pauliciens d'Asie Mineure: texte critique et
traductlon', TM 4 (1970), 1-227; and M.Loos, 'Deux publications fondamentales sur le
paulicianisme d'Asie Mineure', BS 35 (1974), 189-209, and Dualist Heresy, pp. 32-40;
N.G. Garsoian, The Paulician Heresy. A Study of the Origins and Development of Paulicianism
in Armenia and the Eastern Provinces of the Byzantine Empire (The Hague and Paris 1967);
D. Obolensky, The Bogomils, pp. 28-58; Runciman, The Medieval Manlchee, pp. 26ff. The
major Greek source is the report of Peter of Sicily, compiled in the ninth century on the
basis of his own observations, and accounts of their origins from the Paulicians them-
selves. See 'Pierre de Sidle, histoire des Pauliciens'. ed. and trans. N. Garsoian, in Les
Sources grecques, pp. 6-67, and esp. C. Ludwig, 'Wer hat was in welcher Absicht wie
beschrieben?', in Varia U (Poikila Byzantina VI. Bonn 1987), pp. 149-227. Peter places the
beginnings of Paulicianism in the middle years of the seventh century, Armenian sources in
the last years of the seventh and first years of the eighth century. Some historians have
argued for a mid-sixth-century origin; many have connected the heresy with the
Marcionists. See Loos, Dualist Heresy, pp. 30-5; Obolensky, The Bogomils, pp. 45ff.: also
Garsoian, The Paulician Heresy, pp. 13 Iff. (sixth-century origin). Against the Marcionist con-
nection: H. Soderberg, La Religion des Cathares (Uppsala 1949), p. 120. See also Beck, Kirche,
pp. 335-6 with literature.
<* SeeGouillard, X'H6r6sie', 306-7 and 310; Stein, Bilderstreit, p. 262f.; the text, DeHaeresibus
etSynodis,\siaPGXCVm, 39-88; see 85.
65
See especially J. Starr, 'An Eastern Christian sect: the Athinganoi', Harvard Theological
Review 29 (1936), 93-106; Loos. Dualist Heresy, p. 61; Gouillard, 'L'HerSsie', 307-12.
 Religion and belief 345

possible connection with some of the earlier sects, which may have been
reshaped during the seventh century, or may indeed have attracted new
followers and adherents as the remnants of the majority of the older sects
were dispersed or died out. It is at least likely, therefore, that while the
Paulician and athigganoi movements developed their own independent
and dynamic traditions, they were rooted in, and owed their beginnings to,
the conditions which signalled the end of most of the other Christian sects
of Anatolia.
The one group of believers who presented a real problem for both
Christian theologians and the Christian state throughout the period dealt
with here, and indeed throughout the history of the state itself, a group
which, in spite of more or less constant persecution, occupying always a
subordinate position within Eastern Christian society and culture, does
survive and is able even to prosper, is represented, of course, by the Jews.
Judaism presented a number of difficulties to Christian thinkers.
But it also presented difficulties for the state and for society at large, for
the structure of Jewish beliefs and kinship gave Jewish communities a
stubborn resilience which was able to weather the fiercest storms of
persecution, forced baptism and so on. Up to the sixth century, the Jews
had been tolerated, with only minor and occasional persecutions directed
specifically against them. But they, along with certain other groups within
Roman society, had fewer rights than orthodox subjects.66 From Justi-
nian's reign, however, begins a long period of persecution: Justinian
himself deprived Jews, and Samaritans, of their few remaining rights
within the state with regard to public office or state service in general.
Under Justin they had already been deprived of the right to make wills and
to receive inheritances, being also debarred from carrying out any legal
act, such as being a witness in a law court. Justin and Justinian both began
to apply the laws against heretics and pagans to Jews and Samaritans as
well; and although forcible baptism had been applied to Jewish communi-
ties in the fifth century on occasion, such measures were enacted increas-
ingly from Justinian's reign on in the East, and from the later sixth century
66
The growing anti-Semitism of Christianity even produced imperial legislation to protect
Jewish property; but during the later fourth and fifth centuries, increasingly, legislation to
exclude them from state service was introduced. See Jones, LRE, vol. II, pp. 944-8. The
best detailed analysis of the relationship between Jews and the rest of late Roman society
and culture is the excellent article of L. Cracco Ruggini, 'Pagani, Ebrei e Cristiani: odio
sociologico e odio teologico nel mondo antico', in Gil Ebrei nell'alto Medioevo (Settimane di
Studio del Centra Italiano di Studi sull'alto Medioevo XXVI, Spoleto 1980), vol. I,
pp. 13-101, esp. 90ff. For a survey of anti-Jewish polemic, see A.L. Williams, Aduersus
ludaeos. A Bird's-Eye View of Christian Apologiae until the Renaissance (Cambridge 1935);
and P. Browe, 4Die Judengesetzgebung Justinlans', Analecta Gregorlana 8 (1935) 109-46;
R.-M. Seyberlich, Die Judenpolitik Kaiser Justinians I (Byzantinische Beltrage. Berlin 1964),
pp. 73-80; S. Baron, A Social and Religious History of the Jews III (New York 1957),
pp. 4-15.
346 Byzantium in the seventh century

in the West also. Under Tiberius Constantine, Maurice and Heraclius,

similar persecutions took place, and the Jews were increasingly reduced to
the position of a very marginalised social and cultural element within a
predominantly Christian society.67
Under Heraclius, a series of forced baptisms of Jews culminated in 634 in
the first general edict to compel all Jews within the empire to accept
baptism.68 While it seems to have met with very little success, and to have
been only half-heartedly imposed, it is nevertheless symptomatic of a
change in the situation of Jews. The days when Jews were accorded a
privileged status, enshrined in the legal codifications of the emperors -
including Justinian's - and based on both Roman legal theory and Chris-
tian notions of the Jews as a living testimony to the Christian interpretation
of the Old and New Testaments, were gone. Even though Jews lost their
'civil rights' under Justin and Justinian, they were still, in general, set apart
from pagans and heretics alike. Judaism was both permitted and protected.
By the time of Heraclius, this situation had begun to change; and although
there now existed a constant tension between the traditional approach on
the one hand, enshrined in the Codex Theodosianus and the Codex lustinia-
nus, and on the other the exclusivism of emperors like Heraclius, or Leo III
(who is reported to have issued a similar edict to that of Heraclius), the
attempts to force the conversion of Jews to Christianity, repeated sporadi-
cally throughout Byzantine history, were on the whole a failure; the older
Christian tradition, which set Jews apart from other marginal groups,
retained its hold on Eastern Christian attitudes and beliefs.69
Accusations of Judaism began to be used increasingly from the later
sixth century to denigrate heretical tendencies, or those suspected of
deviating from the proper orthodox path - notably in the canons of the
Quinisext, in the introductory declaration of which the assembly addressed
the emperor and noted the survival of 'pagan and Jewish perversity'.70 But
already the increasing persecution of Jews within the empire in the sixth

67
See Jones, LRE, vol. II, pp. 948-50; for the recent literature, J. Starr, The Jews in the
Byzantine Empire, 641-1204 (Athens 1939); A. Sharf, Byzantine Jewry from Justinian to
the Fourth Crusade (New York 1971), esp. pp. 191T.; Yannopoulos, La Societe profane,
pp. 243-51; Beck, Kirche, pp. 332-3 with further literature.
68
See Dolger, Regesten, nos. 196, 197 and 206. The interesting contemporary Doctrina
Jacobi nuper Baptizati records the events of the reigns of Phocas and Heraclius relevant to
this policy, from the point of view of a supposedly genuine convert. See Beck, Kirche,
p. 447; and R. Devreesse, 'La Fin inedite d'une lettre de S. Maxime: un bapt&ne force de
Juifs et de Samaritains a Carthage en 632'. Revue des sciences religieuses 17 (1937).
25-35; see CPG III, 7699 (Epistula 8).
69
See especially Sharf, Byzantine Jewry, pp. 19ff. and 53ff. On Leo's edict, ibid., pp. 61ff.
70
Mansi XI, 933E. Note also canon 11 (ibid., 945E warning against the use of unleavened
bread, mixing with Jews at the baths, dealing with Jewish doctors); canon 33 (957D) and
canon 99 (98 5E) (against a variety of liturgical or related practices, especially in Armenia,
which smacked of strong Jewish influence).
 Religion and belief 347

century - possibly, if not probably, an imperial response to popular

pressure - had encouraged Jews to re-examine their position and their role
within the state. Their resentment expressed itself in terms of civil disturb-
ances, and in particular in the collaboration of some Jewish communities
with the Persians in the early seventh century.71 The consequent Christian
outrage merely exacerbated a vicious circle. By the time of the Quinisext,
Jews were explicitly as bad as pagans and heretics in terms of the danger
they posed to the state, and especially of the challenge they presented to the
universalist claims of neo-Chalcedonian imperial orthodoxy. Anti-Jewish
tracts now became a standard element of Christian polemic.72 Regular
persecution of Jews becomes a commonplace in Byzantine political history;
and the regular use of the term 'Jew' to demean or insult becomes a topos
of some genres of Byzantine literature.73 Judaism itself becomes synony-
mous with a drift away from orthodoxy and with potential political-ideolo-
gical subversion.74 Whether there were, in fact, any substantial communi-
ties of Jews in the territories remaining to the empire after about 650 is
71
Jewish and especially Samaritan revolts: see the summary in Kopstein, 'Zu den Agrarver-
haltnissen', pp. 36-7; S. Winkler, 'Die Samariter in den Jahren 529/30', Klio 43 (1965),
435-57. For the Jewish response to the Persians and Arabs, see Sharf, Byzantine Jewry,
pp. 4 8 - 5 7 with literature, and analysis of the results for later Christian reprisals and state
repression. On anti-Jewish polemic, see G. Dagron, V. Deroche, 'Jmfc e t Chretiens dans
rOrientduVITsiecle', TM11 (1991), 17-273; V. Deroche/La poiemique anti-Judai'que au
VP et au VHe siecle. Un memento inedit, les Kephalaia', TM11 (1991), 275-311; G. Dagron,
'Juda&er', TM11 (1991), 359-80.
72
Of the surviving polemical literature usually ascribed to the sixth century, there is only
one anti-Jewish tract, itself in the form of a dialogue between a Persian magus, Christians,
pagans and Jews at the court of the Sassanid Great King. See Beck. Kirche, p. 381. Another
'Debate with the Jew Herban* dates also to the middle or later sixth century (Beck, Kirche,
pp. 386 and 407); whereas from the seventh century there are, in contrast, at least three
major polemical treatises: those of Stephen of Bostra (ed. G. Mercati, in Theologische
Quartalschrift 77 (1895), 663-8; repr. in Opere minori I (Studi e Testi 76, Rome 1937),
pp. 202-6): of Hieronymus of Jerusalem (cf. P. BatifTol, 'Jerome de Jerusalem d'apres un
document inedit', Revue des questions historiques 39 (1886), 248-55; and PG XL,
845-66); and the so-called Tpifaraux xorrd 'IovSaCwv ev Aa^daxcp (ed. G. Bardy, in PO
XV (Paris 1927), pp. 189ff.). as well as a number of minor polemics, the Doctrina lacobi
nuper Baptizati, and hostile references in the hagiographical tradition. See Beck, Kirche,
pp. 447ff. The contrast is very clear; and see the survey of M. Waegemann, 'Les Traites
adversus ludaeos: aspects des relations judeo-chretiennes dans le monde grec', B 56
(1986), 2 9 5 - 3 1 3 . See Averil Cameron, 'Byzantines and Jews: some recent work on early
Byzantium', BMGS 20 (1996) 2 4 9 - 7 4 .
73
See Beck, Kirche, pp. 443 and 447-8. Note the use of the term to emphasise the superiority
of Christian beliefs to Judaism (and implicitly to other doctrines) in a polemical context in
the later seventh-century miracles of St Artemius (see 48.26sq., 63.13sq.); and note also
the position of the Jew as instrument of the Devil typical in middle Byzantine literature,
and exemplified in the later-seventh- or early-eighth-century story about the Church
administrator in Adana (Cilicia II), who is deceived into making a pact with the Devil
through a Jewish intermediary. See L. Radermacher, Griechische Quellen zur Faustsage
(Vienna and Leipzig 1927) ( = Sitzungsber. d. Akad. d. Wiss. in Wien, phil.-hist. Kl. CCVI,
4). pp. 164ff.
74
See Sharf. Byzantine Jewry, passim, and 'Byzantine Jewry in the seventh century', BZ 48
(1955), 103-15.
348 Byzantium in the seventh century

difficult to say. Individuals and their families there must certainly have
been; but (and ignoring the numerous literary topoi referred to) there is no
evidence for organised communities of Jews in the empire until a much
later date. Jews, nevertheless, became scapegoats for Christian apologists,
as Byzantine society became more and more exclusivist and introverted. I
will return to this below.

CHURCH, STATE AND SOCIETY: POLITICAL IDEOLOGY AND

SYMBOLIC UNIVERSE

Introversion and the failure of mission

Byzantine society became during the later sixth and seventh centuries
increasingly inward-looking and exclusivist. But this increasing uniform-
ity of appearance and consensus, and the accompanying demand or
expectation of conformity represented in many of the sources should not
mislead us into thinking that the cultural practices and beliefs of the
populations of the empire actually became uniform and less differentiated.
On the contrary, formal adherence to, and public recognition of, all the
key elements of political and religious orthodoxy does not mean that in
day-to-day terms observances, assumptions and explanations did not
continue to vary very greatly from region to region and from community
to community - geographical isolation and the strength or weakness of
local cultural identities will have played an essential role. 75 But those who
did not clearly commit themselves to the orthodox oikoumene, which meant
also the imperial state, were perceived as being against it. Persecutions of
pagans and heretics, especially during the reign of Tiberius Constantine
and those of his successors, became more severe as the Christian exclus-
iveness which promoted them became an ever more central element in
Byzantine perceptions of their world and themselves as the Chosen People.
The attitude is summed up by the question of the prefect of the city of
Carthage to the Jewish community there in 632, two years before Hera-
clius' general edict: 'Are you all servants of the emperor? If you are, then
you must accept baptism.' 76
Hostility to non-conforming groups is apparent throughout the later
sixth century, of course, and in many respects is merely a natural progres-
sion from the political theology of exclusiveness which becomes more
obvious after the loss of the West and from the reign of Justinian in
particular. But as the empire shrinks, and as East Roman orthodox culture
75
I have described this phenomenon elsewhere as a form of 'ideological re-orientation'. See
Haldon, 'Ideology and social change', 161ff.
76
Doctrina lacobi nuper Baptizati, 1.18sq.
 Religion and belief 349

begins to reassess the narratives of its place in the world, so this hostility
becomes more pronounced. And it is not confined simply to elements of the
social and economic elite. On the contrary, and for social reasons as well as
for political-theological reasons, the popular hostility to the (mostly)
upper-class persons accused, rightly or wrongly, of adherence to pagan
beliefs and practices during the trials and persecutions of the years 579-82
and after, is an indicator of the direction of the feelings and the attitudes of
ordinary people in cities such as Constantinople, Antioch, Edessa, Baalbek,
Harran and others. What is important about the events is the fact that,
whereas persecutions varying in intensity, imperial legislation and large-
scale missionary activity - that of John of Ephesus is known best - had
occurred before, they had been directed by the state and had rarely found
more than a passive, if accepting, response among the ordinary orthodox
populations. 77 Now popular riots ensued when the penalties imposed upon
the accused were thought to be too lenient. 78 John of Ephesus and Evagrius
ascribe this to the intense piety of the Christians in the empire; and while
there is no reason to doubt the latter, it seems clear that this response is
something quite new. Once again, the notions of exclusiveness and of
religious solidarity, of marking off clear boundaries between true Chris-
tians and Romans, as opposed to pagans, heretics and 'outsiders' in
general, is evident. The hostility of the population of Constantinople to the
Arian church of the Germanic mercenary troops in the early seventh
century is in the same mould. 79 And while it is no doubt possible to find or
to assume more immediate reasons for these popular reactions, it was the
changed ideological atmosphere which determined the particular form of
the reaction, and not the immediate context.
This tendency to exclude those defined as marginal to the orthodox
Christian community, which becomes increasingly evident in the actions
and attitudes of ordinary people, and especially of the urban populations of
the empire, is emphasised by the political shrinkage of the middle of the
seventh century. Following the loss of Jerusalem, Antioch and Alexandria,
only Constantinople remains to serve as the focus of Church and state
hierarchies, as the source of power and authority, privilege and prestige.
At the same time, there is a marked absence of any effort to continue the

77
See I. Engelhardt, Mission und Politik in Byzanz. Ein Beitrag zur Strukturanalyse byzantini-
scher Mission zur Zeit Justins und Justinians (MBM XIX, Munich 1974), esp. pp. 12ff. for
John of Ephesus; also pp. 128ff. and 1 7 8 - 8 6 . For general remarks, H.-G. Beck, 'Christliche
Mission und politische Propaganda im byzantinischen Reich', in Settimane di Studio del
Centro Italiano di Studi sull'alto Medioevo XIV (Spoleto 1967), pp. 6 4 9 - 7 4 (repr. in Ideen
und Realitdten in Byzanz IV (London 1972)).
78
See the account in Rochow, Die Heidenprozesse, pp. 124ff. Sources: John of Ephesus in,
2 7 - 3 4 ; Evagrius V, 18.
79
See Haldon, Ideology and social change', 162 and note 59.
350 Byzantium in the seventh century

policies of the reign of Justinian to proselytise among the pagans or the

unorthodox. Instead, political fiat followed by repression seems to become
the order of the day. The initial attempts of Sergius and Heraclius to win
public support for their monenergite and monothelete doctrines through
debate at local synods marks the only significant gesture towards recon-
ciliation and persuasion. Once failed, the effort was not repeated; and as
the seventh century wears on, even the monophysite communities come to
be branded as a dangerous and subversive element, effectively ignored by
the sixth council as irrelevant to the interests of Church, state and
oikoumene; and condemned in passing in the canons of the Quinisext as
simply another subversive tendency.
It is difficult to know whether any missionary activity was carried on.
There is virtually no evidence for it. It is often assumed, for example, that
the Slav immigrants to Anatolia must have been converted before their
settlement.80 But the empire's immediate needs were for agricultural and
military manpower; and in the two or three years over which the Justi-
nianic transfer of 689 was organised and carried out, it seems unlikely that
anything more than the most superficial proselytisation could have
occurred. Once in Asia Minor, of course, it is probable that conformity with
orthodox Christian practices became unavoidable, as the state and Church
could hardly ignore the existence - and possible influence - of large
numbers of non-Christians in the heart of the empire. Such groups, which,
as we have seen, were under the general supervision of officials appointed
from the capital, will undoubtedly have received ecclesiastical attention,
and it is quite likely that new bishoprics were established specifically for
them. The bishop of Gordoserbon in Bithynia (in the Opsikion district) was
present at the Quinisext in 691/2 (although not at the sixth council in
680), 81 and the bishopric is referred to in the seventh-century Notitia of

80
See Brehier and Aigrain, pp. 200-1, where the Christianisation of the Slavs in the Balkans
is assumed to have been under way by the time of the establishment of the Bulgars in the
680s.
81
Mansi XI, 996E. See B. Grafenauer, Die ethnische Gliederung und geschichtliche Rolle der
westlichen Sudslawen im Mittelalter (Ljubljana 1966), p. 19. But doubts have been expres-
sed as to the Slavic origin of the name - see P. Charanis, 'The Slavic element in Byzantine
Asia Minor in the thirteenth century', B 18 (1946-48), 78 and note 1 (repr. in Studies on
the Demography of the Byzantine Empire VII (London 1972)); but these are not very
convincing: whether or not the name Serb has a Slavic root, the point is that the people in
question used this name themselves, and Gordoserba may well reflect the establishment of
an episcopal see for them in Bithynia. See Obolensky, Byzantine Commonwealth, pp. 59f.;
and Lilie, 'Kaiser Herakleios und die Ansiedlung der Serben', 26ff. That such a procedure
was not unusual at the time is clear from the canons of the Quinisext itself, where canon
38 (Mansi XI, 960E-961A) notes that newly founded or re-founded 'cities' should be
subject to the traditional civil, fiscal and ecclesiastical authorities. Note also canon 39
{ibid., 961A-C) on the establishment of the 'city' Nea Ioustinianopolis in Hellespontus for
the refuge congregation of Constantia in Cyprus.
 Religion and belief 351
Pseudo-Epiphanius.82 It is tempting to conclude that this reflects the
activity of Church and state in the conversion and integration of the new
settlers. The imposition of Christianity on such immigrant populations
must anyway have been regarded as a duty of the emperor, and the fact
that there is virtually no explicit evidence for this does not alter that. The
exclusion of the monophysites from the Christian oikoumene, implicit in the
acts of the sixth council and in the canons of the Quinisext, demonstrates
that the state can in no way have tolerated such overt heterodoxy. That
the canons of the Quinisext wish to deal with pagan and Jewish tendencies,
but make no reference to actual communities of pagans (for example), is
surely suggestive of the fact that there were no such obvious targets to
name - it is unlikely that all reference to such a dangerous element within
the Christian community would have passed over in such silence, if it had
actually existed.
We may reasonably conclude, therefore, that in name at least the
population of the empire was by this time assumed to be entirely Christian,
an assumption which certainly underlies the canons of the Quinisext; and
that the chief problem was perceived to lie in the continued use of
cult-practices, superstitions and so on, which for the contemporary Church
were dangerous leftovers of pre-Christian tradition and had to be excised.
The fact that the empire, nevertheless, retained its multiethnic and polyglot
character - Greek-, Slav- and Armenian-speakers dominated, but other
ethnic/linguistic groups counted themselves as 'Romans', too - gave Chris-
tianity and orthodoxy a particular importance, of course; and the rele-
vance of notions of 'conformity' and 'belonging' at this time is particularly
apparent, as we shall see.
If missionary activity continued at all in this period, it did so at a very
low-key level. Specific groups - such as the Slav immigrants - may have
been compelled by the power of the state to convert, even if at first only
superficially. But massive undertakings of the sort carried out by John of
Ephesus and others in the mid-sixth century do not appear in the sources
at all, and it is tempting to conclude that they simply did not occur because
it was felt that there was no need, or rather, no perceived need, for them.
For the cultural introversion which typifies Byzantine society from the
later sixth to the middle of the eighth centuries (approximately) meant also
a narrowing of horizons, cultural and ideological, as well as, more obvi-
ously, geopolitical. The shrinking of the empire to a rump of the Justinianic
state brought with it, naturally enough, a loss of interest in what had now
become very distant affairs - the question of the Christian communities in
82
In Gelzer, Ungedruckte und ungenugend veroffentlichte Texte der Notitiae Episcopatuum, p. 538
number 187. In general, see also Ditten, 'Zur Bedeutung der Einwanderung der Slawen',
pp.l53ff.
352 Byzantium in the seventh century

south Arabia or Ethiopia, for example, must have seemed very distant from
the concerns of the government of the mid-seventh century. On the other
hand, conversion for diplomatic-political reasons remained on the agenda;
for Christianity marked out the Romans and their world empire clearly
from their barbrian foes.83
Under Heraclius, attempts were made to Christianise the Serbs and the
Croats Invited' to settle in the western Balkans. While these efforts were
not successful in the long term, short-term diplomatic success did follow.84
The conversion of the Onogur ruler Organa with his son Kovrat and other
followers at Constantinople in 619 similarly served diplomatic ends and
belonged to a long and respected Roman political-diplomatic tradition. It
was apparently - according to the limited and much later tradition
incorporated into the History of the patriarch Nicephorus in the early ninth
century - quite successful. But only the leadership of these Bulgaric clans
seems to have accepted Christianity, so that their demise meant not only
the end of their new religion, but also the hitherto friendly relations
maintained with the Byzantine court. 85 In the same way, it has been
argued that a serious plan to achieve the conversion to Christianity of
Sassanid Persia was evolved and pursued during the 620s and 630s, a
plan which was, in the event, brought to nothing by the Arab invasions. 86
Such missionary efforts may well have been stimulated by the eschatalogi-
cal attitudes which became apparent at the end of the sixth century and in
the reign of Heraclius. 87
Thereafter, and until the later eighth and early ninth centuries, neither
state nor Church seems to have promoted or supported in any significant
way - that is, which merited mention in the sources - a deliberate policy of
missionary activity and proselytisation either within or without the
empire. The conversion of the Chazar wife of Justinian II provides an
example of one way in which non-Christians were converted; but it is an
isolated and, in general terms, an insignificant example. 88 No doubt
individuals did preach Christianity in neighbouring lands or along the
83
Compare the speech attributed by Theophanes (307.3sq.) to Heraclius, delivered to the
soldiers before battle: the Christian Romans are placed clearly on the side of God's struggle
with evil, against the pagan and blasphemous barbarians.
84
See the summary in Obolensky, Byzantine Commonwealth, pp. 59ff.; and Ditten, 'Zur
Bedeutung der Einwanderung der Slawen', p.. 1 2 9 with literature; Lilie, 'Kaiser Her-
akleios und die Ansiedlung der Serben', 26ff. with sources and literature.
85
Summarised in Obolensky, Byzantine Commonwealth, pp. 6 2 - 3 .
86
See C. Mango, 'Deux etudes sur Byzance et la Perse sassanide, II: Heraclius, Sahrvaraz et la
vraie croix', TM 9 (1985), 1 0 5 - 1 8 ; note the earlier study by I. Shahid, 'The Iranian factor
in Byzantium during the reign of Heraclius', DO? 2 6 (1972), 2 9 5 - 3 2 0 .
87
See Mango's final remark, ibid., 117, with reference to the prophecy of Chosroes II
reported in Theophylact Simocatta V, 1 5 . 1 - 1 1 ; and compare the argument in the text of
the Doctrina lacobi nuper Baptizati.
88
Theophanes, 372.3Osq.; Ostrogorsky, Geschichte, p. 119.
 Religion and belief 353

border regions, when the opportunity presented itself; but these seem to
have had little effect. As far as the sources reflect the actual situation in the
Slav-occupied Balkans and the Peloponnese, in the early ninth century, for
example, the population was still predominantly pagan.
Cultural introversion thus came to mean a preoccupation with the
attempt to force the 'real' world of the time to fit the Ideal' world or model
of the political theology of state and Church. The only response the
imperial government and the Church seemed able to offer to opposition
within its frontiers, and within the Christian world at large, where
straightforward argument or command proved insufficient, was military
force and violence. And the imperial government - and Church - seem to
have been far too involved with problems of internal political authority
and the distribution of power, together with the determination to stamp
out all heterodox belief and practice which accompanied this concern, to
be able to commit the effort and the resources necessary to large-scale
missionary activity. In ideological terms, such activity must have seemed
simply ineffective, if it was considered seriously at all. Perhaps it was also
the pressure placed on the Christian polity by the militant and aggressive
force of Islam which emphasised this tendency - conversion by the sword
seemed, at least according to this example, to be both quicker and more
effective. The traditional missionary activity of the Church must have
appeared in comparison simply not effective enough to make the effort
worthwhile. For, more importantly, the military success of Islam was
taken commonly by Byzantine society to be a sign of God's wrath against
the Chosen People for the sins they had committed. It is not surprising,
therefore, that the longer-term tendencies evident from the later sixth
century were focused and intensified in the second half of the seventh
century around a whole group of symbolic referents which bore directly or
indirectly on this theme within the framework of the traditional imperial
ideology: imperial authority, the universality of orthodoxy within the
empire, imperial responsibility for both ensuring and maintaining ortho-
doxy, and for perceived failure to do so. In this context, the overriding
concern was for internal order and correct belief - a prerequisite, after all,
for any attempt to expand and recover the areas wrested from imperial
control. 89 Only then could the question of the wider Christian world be
considered - and this meant, first of all, relations with the West and in
particular with the papacy. Only when all these problems had been
resolved could the question of the conversion of the barbarians outside the
empire seriously be considered.
Given these concerns and the different routes by which they could be
89
See Haldon, Ideology and social change', esp. 165-7 and 175-7 with sources and
literature for a summary of some aspects of this development.
354 Byzantium in the seventh century
achieved - represented by the state and its political ideology on the one
hand and by elements within the Church, especially the traditional oppo-
sition to state intervention in (or attempts to direct) matters of theology
and interpretation, on the other hand - it is not surprising that the conflicts
which arose, over monenergism and monotheletism, over imperial auth-
ority and what amounted to the theory of imperial Infallibility' and, later,
over icons, did occur. Neither is it surprising that the intellectual - as
opposed to the political-military - confrontation with Islam did not even
begin until the middle of the eighth century. Islam, even according to John
of Damascus still a Christian heresy, was not the cause of Byzantine
defeats; it was the Christian community itself which had brought the forces
of Islam down upon it, and the situation could be recovered, therefore, only
by first putting the Christian house in order. The strengths or weaknesses
of Islam, the existence of pagans in the Balkans - such questions were
insignificant compared with the quest for orthodox purity, correct belief
and the maintenance of the order and harmony derived from, and imi-
tating, that reigning in heaven.90
These are some of the key motifs of seventh-century political-theological
debate; and it is the attempts of different social/political groups within the
empire to come to terms with the radically changed social, economic and
political conditions of the imperium Romanum after the 640s which must
figure at the centre of any modern attempts to understand and explain the
political and ideological history of the Byzantine world in this period. For
members of the Byzantine cultural-ideological world - the 'symbolic uni-
verse' of east Mediterranean imperial orthodox culture - the crux of the
matter was the imbalance between the way the world was supposed to be
and the perceived and experienced realities of the times. What the events of
the seventh century actually present is the harsh juxtapositioning of
realities against theoretical representations; or, put another way, it
demonstrates the effects on people's understanding and explanation of the
90
It was primarily the iconoclastic controversy, in particular the supposed Islamic influence
on the first iconoclasts, especially Leo HI, which stimulated this interest and the need to
confront Islamic doctrine. Thefirstinformed treatise against Islam is by John of Damascus,
being the second part of his Fount of Knowledge (11^7*1 7va>o-eo)s): see PG XCIV, 677-780
(De Haeresibus); see D. Sathas, John of Damascus on Islam. The 'Heresy of the Ishmaelites'
(Leiden 1972). Later polemics - by Theodore Abu Qurra, for example, a pupil of John of
Damascus - elaborate the basic argument which consists of a chapter-by-chapter refuta-
tion of the Koran, intended to demonstrate on the basis of patristic writings and of the
scriptures the contradictions and inconsistencies of Islamic thought. See Beck, Kirche,
pp.337ff., 478 and 488f.; A.Th. Khoury, Polemique byzantine contre Vlslam (VIW-XIW
siecle) (Leiden 1972); also E. Trapp, 'Manuel EL Palaiologos, Dialoge mit einem 'Terser'",
Wiener byzantinistische Studien II (Vienna 1966), see pp. 11-95 with literature and a
general survey of the beginnings and development of Byzantine anti-Muslim polemic.
Note also E.M. Jeffreys, 'The image of the Arabs in Byzantine literature', Seventeenth
International Byzantine Congress, Major Papers, pp. 305-23, see pp. 316ff.
 Religion and belief 355
world and their own experience, both as individuals and as members of
different social and institutional groupings, of the changes which occurred
in their environment, which the traditional explanatory models available
to them could not handle.91

Ideology vs. experience, or: old narratives for new

Already in the later sixth century the evident mismatch between the
grandiloquent claims of the Justinianic world empire, and the fact of its
actual structural weakness, had led to a reassessment of a number of key
elements within the imperial ideology.
This does not mean to say that there had not been substantial criticisms
of the emperor in, say, the time of Justinian I; nor that there had been no
anxieties about the nature of imperial rule, the direction and effects of
imperial policies and so on. Indeed, a number of writers voiced doubts, in
different ways, about various facets of Justinian's rule. But such doubts
were limited, on the whole (or so it would seem), to a relatively limited
number of literati in Constantinople or other major cities. Opposition to
Justinian's policies, and especially to the effects of hisfiscaladministration,
high taxation, and the maladministration which led to hardship for the
armies, were voiced in the Nika riot, for example, or in the various
mutinies and rebellions which affected the army and some provinces. But
such opposition was primarily concerned with immediate material con-
ditions, as I have argued elsewhere, and was crushed quickly and
effectively. The divisions it implied were papered over through the more
obvious political and military successes of the reign, and the presentation -
in an idealised and codified form (through the Code and the Novels) - of an
orderly, imperially ruled and God-protected empire. The failure of the
Justianianic reconquista, the collapse of Byzantine power in Italy with the
arrival of the Lombards, the increasingly heavy raidsfromboth Turkic and
Germanic, as well as Slav, peoples in the Balkans, the costly warfare with
the Sassanian state in the east - along with the consequent lack of
resources, shortage of troops to defend the newly extended frontiers,
increased oppression of the producing population by the state's fiscal
apparatus - all these developments seem to have stimulated a move away
from reliance upon, and trust in, the imperial cult and the symbols of
imperial authority, and the hierarchy of the established imperial Church -
features of late Roman society and its symbolic universe that evoked the
relationships between God and mankind and provided a straightforward
framework of authority within which the world could be understood.
91
What follows has been argued at greater length and in more detail in Haldon, Ideology
and social change'.
356 Byzantium in the seventh century

Instead, attention was drawn to less material and less fallible symbols of
heavenly authority and, more importantly, of heavenly intercession. Local
communities, in particular urban communities, overcame, at certain levels
of social experience and group identity, the social and class differentiation
within their society through appeal to a single divine intermediary and
intercessor - a local saint or a figure from the divine hierarchy of angels,
even (in the case of Constantinople) the Virgin herself. And this new
emphasis was brought out and given expression in the cult of relics of many
saints and in the icons through which the powerless human could approach
directly - and not through the formal networks of the Church - a patron or
protector.92 At the same time, the peculiar holiness of the ascetic and hermit,
who had confronted the devil and his hordes directly and not been overwhelmed,
provided a similar attraction. A feature of both urban and rural society in the
late Roman period, the significance of the 'holy man' lay in the function he
came to fulfil as both social intermediary and intermediary between God
and man. Importantly, however, neither icons nor holy men were part of
any official framework of authority consecrated through the Church. On
the contrary, they were implicitly blessed directly by God and the Holy
Spirit, as the icons 'not made by human hand1 demonstrate all too clearly.
And as such they were both immensely attractive to people in search of
answers and support, comfort and affirmation of their lives as they led
them; and at the same time potentially subversive and anomic in respect of
the 'establishment'.93 They could undermine confidence in the Church,
challenge its representatives in their interpretations and explanations,
mislead or corrupt those who followed them. But this was not (yet) the
issue at stake. Indeed, the popular shift of attention was quickly taken up
92
Haldon, 'Ideology and social change', 161f.; and Cameron, 'Images of authority'; Nelson,
'Symbols in context'; E. Kitzinger, 'The cult of images in the age before iconoclasm', DO? 8
(1954), 85-150; Brown, 'Dark-age crisis'; for the cult of the Virgin, see Cameron,
Theotokos'; and for the cult of relics, see Auzepy, 'devolution de l'attitude face au miracle
a Byzance'. For the sixth-century situation and the latent contradictions, see Cameron,
Procopius, esp. pp. 242-60 (Trocopius and sixth-century political thought').
93
See Kitzinger, 'Images before iconoclasm'; and esp. Brown, 'Holy man'. For the icons 'not
made by human hand* - acheiropoieta - see Kitzinger, 'Images before iconoclasm',
112-15. Note also, however, the critical remarks of H.W. Drijvers, 'Hellenistic and
oriental origins', in The Byzantine Saint, ed. S. Hackel (London 1981) ( = Studies supple-
mentary to Sobornost, V), pp. 25-33, who modifies Brown's original thesis in a number of
aspects, noting that while the particular socio-ideological function of holy men evidenced
in the Syrian tradition for that region was determined by local social and cultural
structures, the holy man as such was not a purely Syrian phenomenon, having also a
general relevance in late Antique society, the significance of which was structured and
nuanced by local conditions. This comes out clearly in the miracle-collections, particularly
those of St Artemius and St Therapon, and in the accounts of wonders worked by both
holy men and monks, and icons or holy amulets - the stories attributed to Anastasius of
Sinai, for example, or those in the Life of Andrew the Fool. See chapter 11 with notes 53
and 55.
 Religion and belief 357

Plate 9.1 Seventh-century icon (from a triptych) of St Theodore Tiro

358 Byzantium in the seventh century

Plate 9.2 Sixth- to seventh-century icon of St Peter

 Religion and belief 359

Plate 9.3 Seventh- to eighth-century icon of St Athanasius and St Basil

360 Byzantium in the seventh century

by the establishment of Church and state, and served to refocus the

divergent tendencies within the symbolic universe of the late sixth-century
Roman world around the Christ-loving emperors protected by God at
Constantinople, and the ideological framework of Christian imperial
theory. 94
This displacement of ideological attention was, therefore, a result of the
failure of the leading symbolic elements within the established political
ideology of the state and the imperial Church to evoke adequately and to
answer the challenge to Roman imperial universality and authority
thrown down by the events of the post-Justinianic years. It had shaken and
loosened the identity of Christians with their world order which was well
structured and very much taken for granted. The clear divide between
Roman and barbarian, Christian and non-Christian became less well
defined;95 and the response - diffused at first throughout the social for-
mation, taken up and directed by the imperial authorities later - was the
phenomena which I have described. The attempts, both deliberately
fostered by the state and spontaneously generated within late Roman
society, to reaffirm and restate the boundaries and structures which were
thus shaken are clear enough in the formal ritualisation of imperial
ceremonial, the distinction drawn between the sacred and the profane,
holy and secular, and the emphasis placed upon exclusiveness, but -
within the political bounds of the empire at least - universality of belief: the
persecutions of the reigns of Tiberius and Maurice, and the popular echo
they found, is one good illustration. Another, perhaps even more vivid, is

94
See Cameron, I m a g e s of authority', and esp. P J . Alexander, 'The strength of empire and
capital as seen t h r o u g h Byzantine eyes', Speculum 3 7 (1962), 3 3 9 - 5 7 , see 3 4 5 . It must be
emphasised t h a t the Constantinopolitan element in the rhetoric of imperial ideology w a s
itself nothing new. From the time of Constantine's transfer of the capital from Rome,
efforts h a d been m a d e to clothe imperial tradition in a n e w Christian garb, in terms of the
'new Rome' a n d the symbolism of renewal (see Alexander, 'The strength of empire and
capital', 348ff.). But it w a s eminently suited to the ideological needs of the emperors of the
later sixth century, w h o took it up a n d developed it in a m u c h m o r e explicit form - along
with the symbolism of its divine protectress, the Virgin - a n d thereby determined the
future course of a crucial component of Byzantine imperial ideology. See the literature
a n d discussion in Cameron, 'Theotokos', passim, Alexander, 'The strength of empire a n d
capital, 341ff., and esp. E. Fenster, Laudes Constantinopolitanae (Miscellanea Byzantina
Monacensia IX, Munich 1968), pp. 20ff. See also the references in note 1 2 2 , chapter 9,
below.
95
For the importance of such categories a n d the boundaries they represented, see esp.
F. Dolger, 'Bulgarisches Cartum u n d byzantinisches Kaisertum', Bull de I'lnstitut archeolo-
gique Bulgare 9 (1935), 5 7 - 6 8 ( = Byzanz und die europdische Staatenwelt. Ausgewdhlte
Vortrdge und Aufsdtze (Ettal 1953 and Darmstadt 1964), pp. 140-58), see 58f., and 'Rom
in der Gedankenwelt der Byzantiner', Zeitschriftfur Kirchengeschichte 56 (1937), 1 ^ 2 ( =
Byzanz und die europdische Staatenwelt, pp. 7 0 - 1 1 5 ) ; note also P. Brown, Religion and
Society in the Age ofSt Augustine, pp. 4 6 - 7 3 , see 55.
 Religion and belief 361

that expressed in the concern to draw the boundaries between 'real' and
'false' holy men - in the 'questions and answers' attributed to Anastasius
of Sinai and, more dramatically, in the hagiographical tradition - in the
Life of Andrew the Holy Fool and, thereafter, in the narrative literary
tradition of the Byzantine world. 96
The reafflrmation of the late sixth century, therefore, represents an
attempt to explain change as perceived by members of the East Roman
cultural world from their own particular position and sense of place within
their society. It was thus a relatively diffuse set of developments. It is
expressed most obviously in the focusing of attention on imperial cere-
monial and, more specifically, in the centralising of a variety of elements
within the framework of the imperial ideology around the figure of the
emperor, the ruler appointed by God and the symbolism of the divinely
bestowed imperial authority. An increase in the ceremonial and ritual
aspect of court life - both public and private - had been an element of the
imperial cult since the fourth century, of course; but the later sixth century
saw a distinct quickening of pace. 97 But, crucially, this refocusing now also
involved other elements which had hitherto been less central - heavenly
guardians and guarantors of imperial authority, elements which had
represented also the tendency to redirect attention away from God's agent
on earth (and the vast established institutional framework through which
imperial power was exercised) back to God Himself.98 The saints and
martyrs and other divine intercessors about whom civic and local saints'
cults proliferated played a double role. While they publicly evoked the
heavenly sources of imperial authority, they served at the same time to
represent a more immediate and tangible source of divine authority and
grace. Power in this respect, rather than being centralised, was in fact
diffused and diversified at a multiplicity of local levels. Evidence for such
cults and for the devotion they attracted increases dramatically at this
time; so that while the central authority struggled at the formal, official
and public level of Constantinopolitan and provincial ceremonial observ-
ance to centre attention on the emperor and his relation with God, it was
this very emphasis on the divine source of the emperor's authority which
contributed to and promoted the power of local saints and martyrs; for it

96
See Anastasius of Sinai, Quaestiones et Responsiones {PG, LXXXIX, 329-824), qu.
62(648-52); Vita Andreae Sali, 776C-784A. See Brown, 'Sorcery, demons and the rise of
Christianity', 17ff. On the 'Question and Answer' literature, see chapter 11.
97
See A. Alfoldi, 'Die Ausgestaltung des monarchischen Zeremoniells am romischen Kaiser-
hofe', Mitteilungen des Deutschen Archdologischen Instituts, romische Abt, 49 (1931),
1-118, see esp. 35ff., 63 and 100.
98
See esp. Cameron, 'Images of Authority', 1 5 - 2 1 ; Theotokos', 120ff.
362 Byzantium in the seventh century

was the latter who represented a locally involved, committed and -

crucially - an easily accessible source of mediation and intercession."
The 'new' elements which appear within the imperial political ideology
were new only in so far as they were brought to the fore to occupy a
position that had hitherto been otherwise filled. Indeed, the response to the
situation of the late sixth century in general was latent in the symbolic
universe of the East Roman world, and more particularly within the
framework of the imperial ideology itself, which was refined out of a
complex of elements within that symbolic universe. The parallel between
the Chosen People of the Old Testament and the Christian Romans, the
notion of the emperor's position as divinely ordained, the notion of political
and military success and social harmony being dependent upon right belief
and practice, and the notion that God can - and does - punish the Chosen
People when they stray from the path of righteousness; these and many
other elements drawn from the traditional Roman imperial cult and from
Hellenistic political theory, the Judaeo-Christian tradition as well, made up
the complex whole of the Christian imperial ideology. They were articu-
lated together in a unity, in which modern commentators have seen three
main components: the religious-political rhetoric which tied the fate of
Constantinople, the New Rome, to that of the Christian oikoumene and to
the future as foretold in the Old Testament; the theory of imperial rule
based on the concept of renewal, that is, of the renewal of an original state
of affairs; and the theory of the God-ordained nature of the Christian
Roman empire. 100
Now these varied elements were welded together both at the level of
99
See Kitzinger, 'Images before iconoclasm', esp. 95ff. The realignment of imperial foreign
policy under Justin II (an attempt to regain the prestige it was felt had been lost in the later
years of Justinian's reign and to reaffirm strong imperial leadership) and the imperially
sponsored persecutions under Justin, Tiberius II and Maurice illustrate the efforts of the
emperors and the central government to put themselves back into the centre of the stage:
see Haldon, Ideology and social change', 167 and note 73. Similarly, the Emperor
Heraclius was placed even more prominently at the forefront and was presented as a
parallel and indeed fulfilment of the Old Testament tradition, as divinely appointed ruler
in the image of David. See Cameron, Images of authority', 21 and note 93; S. Spain
Alexander, 'Heraclius, Byzantine imperial ideology and the David plates', Speculum 52
1977), 217-37, see 225 and 232-3.
IOO p o r t n e notion of the Chosen People, see F. Dvornik, Early Christian and Byzantine Political
Philosophy (2 vols., Washington D.C. 1966), vol. II, pp. 797 and 823; Cameron, 'Images
of authority', 21-2; Haldon, Ideology and social change', 166 and note 68. For the
tripartite division of the imperial ideology, see esp. Alexander, 'The strength of empire
and capital', 340ff. For some general discussion of key aspects of this ideological system,
see Treitinger Kaiser- und Reichsidee, pp. 158ff. and 220ff.; Winkelmann, 'Zum byzantini-
schen Staat', pp. 171^1; H. Hunger, 'Ideologic und systemstabilisierung im byzantini-
schen Staat', AHASH 27 (1979), 263-72; and the literature cited in Haldon, 'Ideology
and social change', 156 note 37. See, for the older literature, the essays collected in
H. Hunger, ed., Das byzantinische Herrscherbild (Darmstadt 1975). All these elements
were present in the sixth century, for example - see Cameron, Procopius, pp. 256-7.
 Religion and belief 363

formally expressed statements of political-religious ideology - in the prooe-

mia to imperial legislation, for example; or in secular or religious texts
where the nature of the emperor's rule was set out for a wider readership
or audience - where they were most coherently expressed; and at the level
of ordinary, day-to-day rationalisations of personal and group experi-
ence. 101 And as far as possible, the leading elements represented and gave
meaning to the perceived realities of that experience. Once circumstances
had changed sufficiently for this no longer to be the case, that is, once
contradictions between the world and man's place in it according to the
generally accepted views held in society, and the world as it now came to
be perceived, became apparent, then of course both individuals and groups
within society would, sooner or later, have to confront the fact that the
explanations available to them for what was happening to their world -
however broadly or narrowly it was perceived - did not work. The result
was, in the later sixth century, an attempt to rearrange the chief elements
of the imperial ideology, bringing hitherto less central aspects or symbols
to the fore; and an attempt on the part of the ordinary population of the
empire - across all social classes, of course - to understand that experience
of the world through different channels. The developments of the period, in
particular the swing towards reliance upon sources of 'direct' and localised
access to God, evident especially in seventh-century texts, represents this
effort at reassessment. 102
The reaffirmation of the traditional values and framework of the Chris-
tian Roman world, which took place through this reassessment of narra-
tive elements both in people's day-to-day lives and in the formal political

101
See, for example, H. Hunger, Prooimion. Elemente der byzantinischen Kaiseridee in den
Arengen der Urkunden (Vienna 1964); and cf. an eighth-century view, in the romance of
Barlaam and Joasaph, trans, in E. Barker, Social and Political Thought in Byzantium
(Oxford 1957), esp. pp. 82f.; a n d see Beck, Kirche, pp.482f. For this type of literature -
the 'Fiirstenspiegel' - see Hunger, Profane Literatur, vol. I, pp. 15 7ff.
102
A useful w a y to look at such changes is t h r o u g h the medium of personal and group
narrative, t h a t is, accounts of events or situations in the sources which, whatever their
conscious intention, contain implicit assumptions about the w a y s in w h i c h the world
w a s perceived a n d h o w the individual perceived him or herself. It is t h r o u g h narrative -
t h a t is, t h r o u g h accounts of experience - t h a t the construction of a personal social
identity a n d a reality is m a d e possible. I h a v e discussed this a p p r o a c h at greater length in
Ideology a n d social change', esp. 15Iff., w h e r e further literature will be found. A further
pointer to the direction of c h a n g e c a n be seen in the hagiographical and other writings of
the seventh century in particular, in which the direct personal relationship between God
(or his representative, the saint, m a r t y r or similar figure) a n d individual is both taken for
granted and a sine qua non for the construction and understanding of the texts them-
selves. The accounts of the miracles of Artemius a n d of Therapon, for example, compiled
during the later seventh a n d early eighth centuries, provide classic examples; and note
the c o m m e n t s attributed to Anastasius of Sinai in Quaestio 55 (PG LXXXIX, 6 1 7 A -
620B) - but also in other questions - w h e r e this relationship a n d its m u t u a l n a t u r e is
discussed a n d stressed.
364 Byzantium in the seventh century

ideology of the state, seemed to be vindicated by the successes of the

Emperor Heraclius in his wars with the Persians. It also explained the
disasters of the reign of Phocas and of the early years of Heraclius
himself.103
The rise of Islam, and the Arab conquests, marked a new, and in the
event much more real, threat to imperial stability and tp the very existence
of the empire itself. It dislocated Byzantine society much more funda-
mentally and dramatically, and at all levels - political, economic, and in
terms of beliefs and ideas about the world. Reasons for the onslaught, and
for its successes, were not difficult to find within the existing framework of
belief, in particular the assumption that the Chosen People had sinned in
some way and were in consequence suffering their chastisement, a motif
well established in Christian political theology.104 Monotheletism, the last,
unsuccessful attempt of the neo-Chalcedonian imperial establishment to
win over the monophysite East - offered reason enough for many in both
West and East for the continuing success of the Arabs, and of other
barbarians in the Balkans. But the very fact of opposition to imperial policy
- even though justified by Maximus Confessor on strictly theological
grounds - was, in the context outlined above, in which imperial authority
itself was at stake, a direct challenge that could not be ignored. Imperial
insistence on acceptance of its policies, therefore, whether that involved
accepting monotheletism itself, or merely agreeing not to discuss the
nature of imperial authority, becomes quite understandable. All the
emperor wanted was an acknowledgement - through acceptance of the
Typos of 648 - that he remained the fount of all authority on earth,
because he was God's direct choice for this role. 105 And the relevance of
monotheletism to the question of reuniting the monophysites in the East in
communion with the dyophysites in the West, whatever the hopes that
may have been cherished about their political recovery, slips into the
background.
The conscious and unconscious efforts to rethink the elements of the

103
See especially P. Lemerle, 'Quelques remarques sur le regne d'Heraclius', Studi Medievali I
(1960), 3471T. (repr. in Le Monde de Byzance: histoire et institutions III (London 1978));
S. Spain Alexander, 'Heraclius, Byzantine imperial ideology and the David plates';
Cameron, 'Images of authority', 2 Iff.; Shahid, 'The Iranian factor', 3O3f., and 'Heraclius,
mends ev Xpi(rr<i> pa<xi\ei)<5f, DO? 34/5 (1980/1), 225-37.
Jo* See S. Sophronii... Epistola Synodica ad Sergium Patr. CP (PG LXXXVII, 3, 3148-200),
3197D; note also Anastasius of Sinai, Quaestio 17 (PG LXXXIX, 484A-500A), regarding
the evils which befall the Romans; and cf. Quaestio 114 (765C-768A), on the reasons for
pestilences and similar misfortunes. Note also Lateran, 40.28sqq.: Sophronius of Jerusa-
lem's reason for not attending.
105
For Maximus' position, see Ep. X (PG XCI, 449-53), 542D; and compare his other
statements on his position during his various interrogations: PG XC, 116-17, 145C-D,
161D-164A and 164D-165A. See Haldon, 'Ideology and social change'. 173-6 for a
more detailed discussion.
 Religion and belief 365

imperial ideology and the traditional explanatory models of day-to-day

realities thus received a further check. The framework which had been
carefully constructed still depended for its cohesion and its explanatory
power on the position of the emperor. Any dislocation of this structure
would shake the whole system, and an imbalance between theory and
'reality', between divine and imperial authority, would once more appear.
In the event, the disasters of the 630s, 640s and 650s, the massive
restructuring of resource-allocation and distribution, and the damaging
conflicts between the central government and some of its citizens on the
issue of monotheletism, the Typos, and the unpopular trials and punish-
ment of Maximus and Martin and their confederates, had precisely this
effect. The tendency to reassess the relationship between the various
elements which made up the social narratives of the subjects of the empire,
represented as a transfer of spiritual and ideological allegiance away from
the earth-bound to the less vulnerable heavenly sources of authority,
picked up speed, facilitated by the fact that the heavenly intercessors had
by now been thoroughly integrated into the fabric of the imperial ideology
itself, and to a degree at the expense of the emperor's own position.106
The emperors found themselves in an even more difficult position. On
the one hand, the cult which the court and Church maintained and
fostered formally and explicitly recognised the centrality of heavenly
authority and the emperor's absolute dependence upon it for his own
authority; on the other hand, it was those very aspects of the imperial
ideology which laid emphasis upon divine support and mediation which
threatened and even bypassed imperial political power. While the Chris-
tian state, represented and symbolised by the emperor, attempted to
reinforce its position by discrediting or damaging any claims to share or to
circumvent the authority vested in it by God (illustrated, for example, by the
fact that Constans II issued copper coins with the reverse inscription ananeo-
sis and en touto nika),107 other sections of Byzantine society likewise attempted
to reinforce their own position and theframeworkof their beliefs (the order
and structure of their narratives of the world), by clinging to those symbols
which appeared to represent a less vulnerable and fallible source of authority;
or, in the case of those who were able to do so, by opposing changes - and the
imperial government - when such changes seemed to pose a threat to the
traditional values, structures and framework of Christian society and the
efficacy of its symbols.
Thus the state argued with the articulate and literate opposition pre-
106
See especially Kitzinger, 'Images before iconodasm', 126C; Cameron, 'Images of autho-
rity', 6ff., and Theotokos', 100. Note also Haldon, 'Some remarks', 176f.
107
Compare the nature of the charges levelled against Maximus and Martin, which make
this concern absolutely clear: PG XC, 112A-D and 113C; Mansi X, 855D-E, 856A; and
also 850D-E. See Haldon, 'Ideology and social change', 173ff. Constans'coins: Grierson,
DOCn, pt 1,101-2; pt. 2,406,408f.
366 Byzantium in the seventh century

sented by the Western and African clergy, and particularly those grouped
around Pope Martin, Maximus Confessor and the organisers of the Lateran
council of 649, and carried on a public debate over the source of authority
within the empire, with all the public political implications this involved.
But ordinary attitudes among the majority of the population - whether
urban or rural, or members of the state apparatus or the army or whatever
- had already shifted sufficiently to make the debate over the Typos or
monotheletism itself irrelevant. In effect, the values represented by the
more personal symbols of the icon, or the holy patron, in their roles as
accessible mediators and intercessors, seem already to have superseded
those vested explicitly in the figure of the emperor as the symbol of God's
mediation between Himself and mankind. New narratives were already
available, and the result was a parallelism or plurality of sources of
authority on a scale not hitherto encountered.
Soldiers who visited Maximus during his imprisonment were more
interested in ascertaining whether or not he had slandered the Virgin, than
whether he rejected or challenged imperial authority. Indeed, the fact that
the imperial government could find no better way of arousing hostility to
Maximus than attributing this slur to his name is sufficient and revealing
evidence of the direction in which private and public devotion was mov-
ing. 108
In 692, the Quinisext had to confront one element of the problem when,
in canons 41 and 42, it sought to regulate entry into the monastic life and,
more especially, to control the movements and influence of itinerant holy
men. It suggests that by the 690s, there had been a considerable increase
both in the numbers and importance of such individuals, clearly not
subject to any ecclesiastical discipline, a result, we may assume, of both the
disruption of the Church administration in the provinces and the dislo-
cation of provincial populations. Refugees, as we have seen, were clearly a
problem for the ecclesiastical administration, and presumably for the civil
administration, too. 109 In such conditions, the importance of these hermits
and itinerant preachers must have risen considerably as the needs of the
ordinary rural population for interpretation and explanation increased.
Some of the questions supposedly addressed to Anastasius of Sinai reflect
this preoccupation, concerned as they are with the existence of 'false
prophets' and 'miracle workers', and with the question of whether or not it
is possible to obtain the services of one holy man to undo the work of
another. Similarly, a warning tale about a woman who took advice from a
'sorcerer' in the mistaken belief that he was a holy man, which occurs in
los SeePG.XC, 168C-169B.
109
See Mansi XI, 964A-C and 964D. A number of refugees from Syria who arrived in 686
(see Theophanes, 364.3^1) must have added to the confusion.
 Religion and belief 367

the fictional Life of Andrew the Fool (later seventh century) deals with the
same point. Such questions and stories illustrate the concerns of this time
of uncertainty - as do queries such as why there were so many lepers,
cripples, epileptics and others among the Romans, in contrast to other,
non-Christian peoples; or whether prophesying and fortune-telling
through lachnisterion (chance searching through pages of the Old and New
Testaments) were permitted. 110
The traditional ideological framework offered little comfort. But the
apocalyptic prophecies, which represented a long popular tradition and
responded most clearly to the needs felt by the uneducated mass of the
population in the provinces especially, had become the bread-and-butter of
such preachers and itinerant holy men. 111 And it is not surprising that it is
at just this time that the greatest and most widely disseminated of the
medieval compilatory apocalypses, that of Pseudo-Methodius of Patara,
was produced. 112 What is particularly significant is that apocalyptic
writings of this sort represented an alternative ideology of the future: the
formal political ideology permitted an abstract and distanced statement of
the relationship between empire and God; the eschatology of writings like
the apocalypse of Pseudo-Methodius - which, following an increasingly
important trend within the apocalyptic-eschatalogical tradition, identified
the historical Christian Roman empire with the last of the four world
empires of the vision of Daniel - in contrast, predicted an ultimate and

110
See Quaestiones 20 (PG, LXXXIX, 517C-532B), 62 (648A-652D), 94 (732B-733C) and
108 (761A-B); and Vita Andreae Sali, 777C-781A. Compare canon 61 of the Quinisext
(Mansi XI, 970E-972A), which condemns all forms of fortune-telling, palm-reading,
prediction, as well as the use or sale of amulets and so on. These practices were certainly
not peculiar to the seventh century; but the fact that the Quinisext felt obliged to deal
with them, along with a wide range of problems of a similar nature, in such an explicit
and directed manner, is eloquent in the context in question.
111
See the comments of G. Podskalsky, Byzantinische Reichseschatologie. Die Periodisierung
der Weltgeschichte in den vier Grossreichen (Daniel 2 und 7) und dem tausendjdhrigen
Friedensreiche (Apok. 20) (Munich 1972), pp. 70-1, on the increasingly polemical and
compilatory character of the apocalyptic tradition from the late fifth century on, and the
gradual reduction of its original technical exegetical aspect. But it should be recalled that
such writings were also part of a much older, and still very lively, Jewish tradition, which
also experienced a revival at this time: see S.W. Baron, A Social and Religious History of the
Jews V (8 vols.): Religious Controls and Dissensions, 2nd edn (New York 1957),
pp. 138-50.
112
See Podskalsky, Byzantinische Reichseschatologie, pp. 53ff.; Haldon, Ideology and social
change', 168 and note 74; P.J. Alexander, 'Medieval apocalypses as historical sources',
American Historical Review 73 (1968), 997-1018, see 998ff. (repr. in Religious and
Political History and Thought in the Byzantine Empire XIII (London 1978); and especially
G.J. Reinink, 'Pseudo-Methodius und die Legende von romischen Endkaiser', 82ff., who
argues that it was originally intended as a sermon or homily: see also 'Ismael, der
Wildesel in der Wiiste. Zur Typologie der Apokalypse des Pseudo-Methodios', BZ 75
(1982), 3 3 6 ^ 4 , esp. 338 and note 14.
368 Byzantium in the seventh century

assured victory over the enemy, in this case, the Arabs.113 The Pseudo-
Methodius apocalypse, originally written in Syriac, but very soon available
in Greek and Latin, circulated probably in the 690s, and it was clearly
intended to be directly relevant to the experiences of a confused and
battered Byzantine population, as well as to the Christians outside the
empire. The great popularity of the text bears out the observation that this
type of literature flourished most at the times when people most needed it,
that is, in times of social, political and economic upheaval. The apocalypse
contained in the Life of Andrew the Fool, which dates to the same period,
was likewise intended quite explicitly to generate hope and optimism about
the future.114 In the early part of the century, the Persian and early Arab
attacks had similarly generated such apocalyptic compositions, notably
the apocalypse of Pseudo-Ephraim and the Syrian Christian legend of
Alexander.115 Particularly telling is the short treatise composed by the
monk Theophanius in about 710, which attempts to calculate the date of
the end of the world using a cabbalistic process of attributing numerical
values to the names of the key figures in the Old Testament. The work was
compiled towards the end of the period with which we are concerned, at a
time when popular fears about the end of the world, the threat from Islam
and the problems of imperial authority were all central issues. It is an
interesting reflection of the scepticism with which some popular

113
Podskalsky, Byzantinische Reichseschatologie, pp. 54-5 and 72ff.
114
Alexander, 'Medieval apocalypses as historical sources', 1005f. and 1002; J. Wortley,
'The literature of catastrophe', Byzantine Studies 4 (1977), 1-17. For the Life of Andrew
the Fool, see C. Mango, 'The Life of St. Andrew the Fool reconsidered', Rivista di Studi
Bizantini e Slavi 2 ( = Miscellanea A. Pertusi II) (Bologna 1982), 297-313 (repr. in
Byzantium and its Image VIII). The apocalypse known as the Exegesis of Ps-Daniel, dated
to the years 716-17, is discussed in the same article (see 310-13) and fulfilled a similar
function. On the Pseudo-Methodius Apocalypse in particular, see G. Reinink, 'Pseudo-
Methodius: A Concept of History in Response to the Rise of Islam', in Averil Cameron and
L.I. Conrad, eds., The Early Medieval Near East: Problems in the Literary Source Material
(Princeton 1990), where its general ecumenical relevance is discussed. Note also, on
another Syriac Apocalypse of the same period, H.J.W. Drijvers, 'The Gospel of the Twelve
Apostles: A Syriac Apocalypse from the Early Islamic Period', ibid.
115
See W. Brandes, 'Apokalyptisches in Pergamon', BS 47 (1987), 1-11, see 5-7 with
literature, and 'Die apokalyptische Literatur', in Quellen zur Geschichte des friihen
Byzanz (BBA, LV Berlin 1989). See also Th. Frenz, 'Textkritische Untersuchungen zu
"Pseudo-Methodios": das Verhaltnis der griechischen zur altesten lateinischen Fassung',
BZ 80 (1987), 50-8. Note also the interesting discussion of H. Suermann, Die geschichts-
theologische Reaktion auf die einfallenden Muslime in der edessenischen Apokalyptik des 7.
Jahrhunderts (Europaische Hochschulschriften, Reihe XXII: Theologie, 256. Frankfurt, a.
M. Bonn and New York 1985); and G. Reinink, 'Die Entstehung der syrischen Alexander-
legende als politisch-religiose Propagandaschrift fur Herakleios' Kirchenpolitik', in After
Chalcedon. Studies in Theology and Church History Offered to Professor A. van Rooey
(Orientalia Lovanensia Analecta, 18. Leuven 1985), pp. 263-81. For a convenient
survey of the effects of the Arab conquests on the Syrian monophysite Church and its
later tradition, see S. Ashbrook-Harvey, 'Remembering pain: Syriac historiography and
the separtion of the Churches', B 58 (1988), 295-308, esp. 304f.
 Religion and belief 369

mythology about the date of the last judgement, and the fate of the empire
in the immediate future, may have been handled, that Theophanius
produced the date of A.D. 880 (in our reckoning) for the end of the
world. 116
The activities of the itinerant holy men, the inability of the established
Church to maintain ecclesiastical discipline and, in particular, the already
established fact that the location of what was holy or sacred in Eastern
Christian culture was both more widely dispersed and more ambiguous
than in the West - and, therefore, more directly accessible, more open to
immediate personal experience - all these elements contributed to a new
atmosphere. Especially indicative is the proliferation of private chapels
and, indeed, the fast-growing custom of celebrating the liturgy in private
homes, something quite absent in the West, where the Church retained a
much stricter control over such activities.117 It was this new atmosphere
which generated both an understanding of what was happening to the
traditional framework of society, as well as the action which was felt to be
necessary to preserve it intact, by allowing a more openly critical position
with regard to imperial authority and the individual emperor's interpreta-
tion of that authority. It is worth adding, perhaps, that the attempts of the
Church to control the hermits and preachers were, on the whole, unsuc-
cessful: the itinerant, wonder-working holy man or monk remains a classic
of Byzantine hagiography in the ninth and tenth centuries. The popularity
and importance of such figures is borne out by the number of references to
holy men competing for their audience and following - there must have
been many more such figures than the hagiographical tradition alone
would suggest. Just as significant is the effort of the Church to assert its
authority over the scriptural and textual tradition upon which dogma and
the canons were founded. Canon 9 of the Quinisext is particularly strongly
worded: the clergy and bishops should interpret Scripture strictly in
accordance with the Fathers of the Church, and not improvise - once
more, the establishment of authority and the signalling of the existence of

1"> See Beck, Kirche, p. 473. The text is discussed by A. Dobschutz,' Coislinianus 296', BZ 12
(1903), 534-67, see 549ff.
117
See especially P.R.L. Brown, 'Eastern and Western Christendom in late Antiquity: a
parting of the ways', Studies in Church History XIII (1976), 1-24, see llf.; Nelson,
'Symbols in context, 11 If. and 115f. For private chapels and the liturgy, see canons 31
and 59 of the Quinisext (Mansi XI, 956E and 969C); and the discussion of T. Mathews,
'"Private" liturgy in Byzantine architecture', Cahiers Arche'ologiques 30 (1982), 125-38.
Whether there was indeed a proliferation of 'domestic' icons during the seventh century, as
has been maintained (e.g. J. Herrin, 'Women and the faith in icons in early Christianity', in
Culture, Ideology and Politics, ed. R. Samuel and G. Stedman-Jones (London 1982), pp.
56-83, see 66ff.) has been seriously questioned by P. Speck, 'Wunderheilige undBilder: Zur
Frage des Beginns der Bilderverehrung', in Varia m (Poikila Byzantina XI. Bonn 1991)
163-247; idem, 'Das Teufelsschlofi. Bilderverehrung bei Anastasios Sinaites?', in Varia V
(Poikila Byzantina Xm. Bonn 1994) 295ff.
370 Byzantium in the seventh century

clear boundaries are the chief concerns. And these concerns are echoed
also in the Questions and Answers of Anastasius of Sinai. Indeed, the effort
made at this time to establish an authoritative set of texts from which
dogma could be adduced, clear in the proceedings of the sixth ecumenical
council and in the canons of the Quinisext, is an important affirmation of
this preoccupation.118
The political and ideological history of the second half of the seventh
century illustrates the extent of the changes. Political coups, military
rebellions and attentats all had their immediate, conjunctural causes. But
that they could occur, and that such actions could be envisaged as
ideologically acceptable, shows that the ground rules within which indi-
viduals and groups situated themselves socially had been drastically
revised. The later seventh century contrasts vividly with the sixth century,
a period of relative internal stability, political expansion and ideological
security, at least up to the 550s and 560s. In such a context, the imperial
ideology left no room and made no provision for direct challenges to
imperial authority. Conversely, however, the very different situation of the
later seventh century and the realignment of elements within the symbolic
universe and especially within the imperial ideology itself meant that such
a challenge could appear both justifiable and worthwhile.
Justinian II's reformed coinage offers a clear example of the desperate
efforts made by the emperors to keep up with these shifts in people's world
view: the Quinisext had ordained that Christ should no longer be repre-
sented as a lamb, but in his human form, thereby the better to stress his
incarnation (canon 82). Justinian's reformed coinage moved the emperor's
bust from the obverse to the reverse of the gold coins, introducing a bust of
Christ to the obverse. The intention must have been to emphasise the
ordinance of the Quinisext; but it served also to stress the nature of Christ's
role as mediator and intercessor between heaven and earth. And, given the
ideological context we have described, it must also have served to empha-
sise the emperor's particular role in this relationship. The Hodegos of
Anastasius of Sinai, written during the years from c. 640 to c. 680 and
incorporating his own later additions, argues strongly for the value of
certain representations from the life of Christ, in particular the crucifixion,
as didactic means of refuting heresy. Together with canons 73, 82 and
100 of the Quinisext, which all deal with images, this may be evidence of a
tendency at this time to recruit icons into the service of the Church as one
more weapon in the battle against heterodoxy and the struggle to main-
tain the boundaries between orthodoxy and the 'outside' world. Once
118
See the useful brief summary of K. Ringrose, Saints, Holy Men and Byzantine Society, 726
to 843 (Ann Arbor 1976), pp. 83ff. For the Quinisext, see Mansi XI, 952; and for
Anastasius, see qu. 117.
 Religion and belief 371
more, it would seem to represent an attempt on the part of the central
authority to stress and reinforce the divine nature of their authority on
earth. Justinian's successors, in reverting to the traditional type, seem to
have sought to cement their authority by reversing the process and
returning to the traditional, and legitimating, form.119

The search for order: the case of the soldiers

In this changed situation, and on the basis of this very different interpreta-
tion of the function and position of imperial authority, individuals and
groups were able to take action directly and to intervene in imperial
politics in order to protect what they perceived to be their interests, or the
interests of the state; or indeed to re-establish a stability and security which
had been lost.
One group in particular deserves our attention, since their activities at
this time are highlighted in the sources, and hence serve to elucidate some
of these developments, developments which were central to the contempo-
rary perception of the world. For as the seventh century wore on, the
provincial soldiery came increasingly to be drawn from, and to represent,
at least implicitly and in a partially refracted form, the attitudes and
sentiments of the provincial and rural populations. Their actions represent
at the same time a generalised loss of faith in the traditional symbols of
authority and the increasing ineffectiveness of traditional legitimating
narratives; and, in however distorted a way, and in spite of the very
different consequences which followed, they represent an attempt to
restore an older, and irretrievably lost, pattern of relationships of auth-
ority. Officers and soldiers played the central role in all the political
upheavals of the second half of the seventh century. Unlike the period up
to, approximately, the last years of Heraclius, pay and conditions of service
do not seem to have fired the discontent. On the contrary, ideological
motives lie clearly at the root of a number of such demonstrations - the
support for Constans II among the troops of Valentinus in 641-2, the
rebellion of Gregory, the exarch of Africa, in 647 and of Olympius in Italy
in 649-52, the demonstration of the Anatolikon troops at Chrysoupolis in
681, the coups of 695, 698, 711 and of the years up to 717 - they all were
119
See Haldon, 'Some remarks', 164-6; and for Justinian's coinage, see chapter 2, note 83
above; and A. Grabar, L'Iconoclasme byzantin: dossier archeologique (Paris 1967), esp.
pp. 36ff. and 77-80 and plates. For Anastasius and the icons see Anna D. Kartsonis,
Anastasis: The Making of an Image (Princeton 1986), pp. 59ff., and 'The Hodegos of
Anastasius Sinaites and Seventh-century pictorial polemics', in XVI. International
Byzantinistenkongress. Resumes der Kurzbeitrdge (Vienna 1981), 10.3. For a general
comment on the search for authoritative tradition, see C. Head, Justinian II of Byzantium
(Madison, Wisconsin 1972), p. 61.
372 Byzantium in the seventh century

made possible by the ideological climate and assumptions of the times, as

well as the fact of the political and military, and - indeed - the economic,
instability of the times. 120
Of course, soldiers may not always have been the direct instigators in
these events; they may sometimes have been merely tools in the hands of
politically more astute and self-interested officers; they may also have
taken action for purely 'selfish' motives, having become aware of their
political relevance and their strength. But in all these cases they were
voicing more often than not the attitudes of the provinces and reflected in
turn a genuine concern with the politics of the imperial government, given
expression through the realignment of the key elements in the symbolic
narratives through which the whole fabric of beliefs, intuitions and prac-
tices of Byzantine society was represented. The emperor and the imperial
ideology stood at the centre of this complex, and it is not surprising,
therefore, that they stood also at the centre of political conflict.
Soldiers became important in this period for a number of reasons. In the
first place, the army represented a set of institutions which generated its
own rules of conduct, identity, role-playing and its own language. It
represented, in short, a specific set of practices, a discourse, within the
larger framework of the social formation. It represented also one of the few
organised contexts in which large numbers of individuals, especially those
drawn from the dispersed rural communities of the provinces, came
together. And in the context of the seventh century, this becomes especially
significant. While the urban-centred society of the period up to the early
seventh century was able to channel its social and cultural resources into
and through the cities, the result of the long decline of corporate municipal
life and of the economic-administrative independence of the cities, which
had set in during the third century, meant that the cities lost this role. 121
They ceased to be self-governing administrative units responsible for both
their own and the state's revenues. By the seventh century, the majority
were rapidly becoming local centres of state provincial administration
only, their single remaining function being as a collecting point for the
extraction and forwarding of the state's taxes and other revenues. The
cities themselves had no economic role in this process. As corporate bodies,
they had lost effective control over their lands and rents, which now went
to the state or to private landlords. The net result was the dissolution of the
formal administrative and fiscal ties which had hitherto bound the city to
its territorium; and the loss of the historic function of the cities within the

120
See chapter 2, above; and Haldon, Ideology and social change', 177-89 with sources
and literature.
121
See chapter 3 above; and see also W. Goffart, 'Zosimus, the first historian of Rome's fall',
American Historical Review 76 (1971), 4 1 2 - 4 1 , see 4 2 5 - 6 and notes 6 3 - 6 .
 Religion and belief 373

structure of the Roman state, as the basic and essential component of

provincial fiscal administration. Cities may well have housed locally
appointed fiscal and administrative officials, selected from the traditional
sources; but it was centrally appointed, salaried bureaucrats who con-
trolled and directed such appointments, supervised their activities and took
charge of the revenue-collecting operations. In these circumstances, a
transfer of the social and economic wealth previously invested in the
provincial cities to the capital and the state apparatus is not surprising. The
city had lost not only its economic and corporate personality, it had lost
thereby its social and cultural relevance and attractiveness. Provincial
notables and others whose family careers and importance had hitherto
been firmly entrenched in provincial municipal culture now shifted their
attention to Constantinople, where they vied for position, titles and
honours within the ambit of the imperial court and the bureaucracy, and
within the establishment of the Church. 122
Given this development, cities could no longer function adequately as
channels of acceses to authority at a local level, nor to the capital and the
governing elite there. The corporate nature of provincial municipal life was
replaced by a kin-based solidarity of family groupings, by the individualism
of the competition for power and wealth within the imperial bureaucracy
and the Church. And it was this 'blocking off of routes of access to
authority which disadvantaged the ordinary provincial, especially rural,
populations. The wealthy could overcome the physical and cultural bar-
riers; simple artisans and peasant cultivators could not.
The effect of this phenomenon was to push the provincial soldiery into
the limelight. This group became, in practical terms, the only representa-
tives of provincial opinion and attitudes which was in a position to make
its views known; it had also the institutional organisation and hence the
power to take action to make these views known. And while it is clear that
such views must always, to some extent, have reflected the particular
interests of a special - and in many ways privileged - group within society,
they must also have represented the sentiments of their own social class
and culture, their own loyalties and the loyalties of their region and so on,
an association which can only have gained in strength as recruitment and

122
See chapter 3 above; and note especially H. Hunger, Reich der neuen Mitte, on the process
whereby Constantinople was transformed from the city of the imperial government, one
among several equally important and prestigious cities in the sixth century, to the city,
during the seventh century, a process completed by the loss of Alexandria, Antioch,
Jerusalem and other large centres to the Arabs. But the process was initiated under
Justinian, and taken up and promoted by his immediate successors. Cf. Cameron,
Theotokos'; and H.-G. Beck, 'Konstantinopel, das neue Rom', Gymnasium 71 (1964),
166-74; W. Hammer, 'The concept of the new or second Rome in the Middle Ages',
Speculum 19 (1944), 50-62.
374 Byzantium in the seventh century

service came to be identified with particular districts. Provincial fears and

uncertainties, provincial attitudes to the government or the local administra-
tion, anxiety about and frustration at the failures of the government to deal
effectively with the various barbarians besetting the empire and wrecking
the rhythm of economic and social life; all of these feelings could now be
expressed by the provincials through only one channel - the local army.
The attitudes and actions of soldiers thus take on a new significance.
They serve to a degree, and however muted and refracted, as a barometer
of the views and beliefs of provincial society. This explains also their
'incorporation' by the state into the framework of legitimacy, even if only
nominally - both Constantine IV and Justinian II attempted to integrate the
armies as an element of the state and its authority into some of their formal
statements.123 While military unrest or opposition to the government was
obviously determined in its form and content by the specific context of the
times, the public activities of soldiers can suggest a great deal about the
responses of the ordinary people of the times, about whom we hear
otherwise virtually nothing. It was because the existence of traditional
society as it was understood seemed to be threatened and because the
stable framework through which experience could be reconstructed via
personal and social narrative was shaken, that soldiers could take the sorts
of action that they did in the middle of the seventh century and afterwards.
They were simply responding to their understanding of events, and their
position both socially and institutionally gave them the power to intervene
directly to restore the situation, as they perceived it, to a former state, a
state of order, certainty and political and social harmony.
Taken as a whole, therefore, the period from the reign of Justin II up to
that of Leo III presents a fascinating story, of a society and a culture in the
process of redefining itself. 'Redefining' was, of course, not the conscious
intention - reaffirmation of the traditional was what was intended - but
this was nevertheless the end result, viewed from our vantage point. Every
society or culture generates sets of legitimating ideas and theories,
explanatory and descriptive mechanisms which are, in effect, the social-
functional forms of personal and shared narrative reconstructions of
experience, through which each group within the society is able to 'justify'
or legitimate its practice and explain its position in terms of its physical and
social environment. Such theories and explanations are themselves deter-
mined in their content by the subjective evaluation of the experiences of
the institutional group or social class concerned, and, by the functions they
fulfilled, given expression in the terms of what was culturally available,
through the symbolic universe. They represent various aspects of the main
ideological currents prevailing within a culture and, less immediately, the
123
SeeMansi, XI, 201C; Riedinger, 10.23f., and 737B-738A; Riedinger, 886.20-25.
 Religion and belief 37S
material conditions of existence of the sections of society which maintain
them; and they tend to be expressed through one set of terms or another
within the dominant political ideology.
Now, when the stability of a social formation, or the position of one or
more sections within it, is felt to be under threat, it becomes necessary to
reaffirm the validity, the authority, and indeed the relevance of its own
values and the narrative representation of those values. The history of
Byzantine society in our period represents just such a situation. This is not
to imply that Byzantine society - or indeed any socio-cultural formation -
was a monolith. On the contrary, different sections reacted differently and
at different times to their own subjective interpretations of what was going
on, and how they should respond. The situation was stabilised at first
through the development of social and political-ideological narratives
which absorbed and tried to compensate for the effects of the apparent
contradictions within the symbolic universe: an increased private and
public, official reliance upon icons, for example, as more accessible media-
tors and intercessors between God and man; imperial ceremonial designed
to augment and bolster the emperor's authority by emphasising its divine
source, and stressing the emperor's piety and devoutness. This develop-
ment was reinforced by Heraclius' successes against the Persians. But the
successes of Islam introduced new strains upon the system, while the
long-term structural changes in society - the fate of the cities, for example,
with all that that entailed - increased those strains to breaking point. The
attempts at religious unification made by Sergius and Heraclius, the
attempt to impose absolute imperial authority under Constans II, these
represent two approaches to the problems as they were perceived. The
'intervention' of soldiers and officers reflects the position of the provincial
armies and the attitudes of the provincial populations themselves to these
developments. It indicates also the growing inadequacy of the traditional
legitimating narratives. Only with the iconoclastic controversy and its
outcome would new sets of such legitimating theories be fully worked out.
But it must also be emphasised that iconoclasm, as an imperial policy, was
not simply a product of the 'crisis' generated by the advance of Islam in the
East and the establishment of a new, threatening and rival cultural and
political force, a point of view maintained by some historians.124 It was
much more complex than that. As we have seen, both long-term internal
social and ideological tendencies, as well as the effects on Byzantine state
and society of military, fiscal and political crises, were combined in an
intricate pattern in the creation of the conditions which made imperial
iconoclasm first of all thinkable and, secondly, practicable.'

124
Most recently Herrin, The Formation of Christendom, p. 343.
 CHAPTER 1 0

Forms of social and cultural organisation:

infrastructures and hierarchies

As we have seen in the foregoing account, late Roman society was marked
by a distinct introversion and introspectiveness from the late sixth century
on. While this tendency is represented at the level of official, public
consciousness - where it is most easily detected - as a form of ideological
reorientation, as an attempt to reaffirm the traditions and values of the
past, its roots lie in the material conditions of existence and the experiences
and perceptions of people at all levels of society, from the lowliest tied
peasant to the emperors themselves, and in the ways these people were
able to give meaning to and come to terms with these experiences.
The interpretation placed upon these perceptions and experiences both
promoted, and in its turn was promoted by, changes which also affected
people's relationships to one another, and the concepts and vocabulary
available to describe and explain these relationships. In terms of their
response to the wider world, we have seen how some of these ideas were
marked out. But they had effects upon what might be termed the social
infrastructure, too, the relationships of individuals and groups through
marriage and kinship, patronage and clientship and, most especially, in
the context of seventh-century Byzantium, through and within the appar-
atuses and the hierarchy of the state.

INDIVIDUAL, FAMILY AND THE STATE

One of the most significant developments of the later Roman period was in
the area of kinship structures, 1 and the ways in which Christian marriage,
as a formal system of gift-exchange, property-transmission and social
1
For some introductory comments on the problem, see J. Thirsk, 'The family', Past and
Present 27 (1964), 116-22, the essays in S.C. Humphreys, The Family, Women and Death:
Comparative Studies (London 1983), and in R.N. Anshen, ed., The Family: Its Function and
Destiny (New York 1959), and esp. P. Laslett, 'Family and household as work group and
kin group: areas of traditional Europe compared', in R. Wall et al., eds., Family Forms in
Historic Europe (Cambridge 1983), along with the other essays in that volume.

376
 Infrastructures and hierarchies 377

alliances, came to play an increasingly central role in both the organi-

sation of social relations and in the self-image of the society, as represented
at least through the writings of Churchmen, hagiographers and jurists.
The recognised forms of liaison which were inscribed in Roman law were
modified in a number of ways. The possibility of divorce came to be
increasingly restricted,2 and by the eighth century the Church was devel-
oping a specific betrothal service.3 While concubinage (the regular union
between a free man and a slave or freedwoman) was tolerated by the
Church (as was simple cohabitation if consented to by the parents), it was
gradually assimilated to regular marriage, as the Church, during the sixth
and seventh centuries, gained greater success in obtaining recognition
from the state for its principles.4 But even in the eleventh century, such
relationships were still acceptable within the terms of the civil law; while
the Ecloga clearly gives full recognition to a union which takes place by
consent alone, made known either through a Church ceremony or
through civil witnesses - this was especially so when poverty prevented
the exchange of marriage gifts and the drawing up of a formal, written
contract. Only in the reign of Leo VI (886-912) did the benediction of the
Church become an absolute condition for the validity of a marriage. 5 The
original distinction between the written and publicly witnessed marriage
agreement of those belonging broadly to the estate of honestiores, and the
non-binding arrangements (in terms of the law, of course, rather than in
terms of the social bonds that followed) of the poorer in society, however,
while it did not disappear, was gradually blurred over as both types of
arrangement were brought within the supervision of the Church and the
2
Divorce by mutual consent remained, however: L} V, 17.8; 9; Justinian, Nov. 22 (a. 536).
Justinian's novel 117 (a. 542) had stipulated a series of specific grounds, but Justin II
withdrew these regulations in 566. Thereafter, novel 117 seems to have been in abeyance
until the eighth century. See especially the survey of L. Bressau, II divorzio nelle chiese
orientali (Bologna 1976), pp. 21ff., esp. 22-3 and C.E. Zacharia von Lingenthal, Geschichte
des griechisch-romischen Rechts, 3rd edn (Berlin 1892 and 1955), pp. 76-81.
3
See CTh. Ill, 5.2 (a. 319), 6 (a. 336); CJ VI, 61.5 (a. 473); and especially E. Herman, 'Die
Schliessung des Verlobnisses im Recht Justinians und der spateren byzantinischen Gesetz-
gebung', Analecta Gregoriana 8 (1935), 79-107; K. Ritzer, Formen, Riten und religioses
Brauchtum der Eheschliessung in den christlichen Kirchen des ersten Jahrtausends (Munich
1961), pp. 77tt. and 143ff.
4
See Digesta 25, 7.3 for the original Roman definition; and for the harsh measures of
Constantine I against the institution, see CTh. IV, 6.2f.; CJ V, 26.1. Anastasius I and
Justinian I both ameliorated these restrictive measures, on the grounds of the hardship
caused to the offspring of such unions: Justinian, Nov. 89, see proem. Although permitted to
unmarried, widowed and divorced men and women in the Ecloga, it was regarded
effectively as an unwritten marriage contract, and treated thus. See Ecloga II, 6.
5
See the cases in the Peira, for example, XLIX, 25 (/GKIV, 204-5); and see Ecloga II, 2 and 6.
Note also the valuable discussion of A.E. Laiou, 'Consensus Facit Nuptias - et Non. Pope
Nicholas I's Responsa to the Bulgarians as a source for Byzantine marriage customs',
Rechtshistorisches Journal 4 (1985), 189-201. For the legislation of Leo VI, see esp. novel
89.
378 Byzantium in the seventh century

state; although again, the first clear evidence of this process is from the
ninth century. Similarly, by the ninth century betrothal and marriage
were seen, at least by the Church, as more or less equally binding: so much
is implied in the replies of Pope Nicholas I to the questions on Christian
practice addressed to him by the Bulgar Tsar Boris-Michael in 866; and
this is the effect of novels 74 and 109 issued by Leo VI.6 At the same time,
the process of prohibiting marriage within an ever-widening range of
cognates - beyond first, and later second, cousins - had been completed, in
theory, by the later seventh century. Such prohibitions are embodied in the
canons of the Quinisext council of 692 and are firmly laid down in the
Ecloga - although it is probable that they finally obtained general and
widespread acceptance and enforcement only in the eleventh century. 7
The Ecloga took up once more, although in a revised form, the heart of the
Justinianic legislation on divorce, which had been abrogated by Justin II,
and there is a very marked tendency to reinforce the permanence of the
marital contract and to limit the possibilities for divorce. By the eighth
century, divorce could only be obtained (again at least in theory) with
some difficulty, after a hearing, a radical change from the traditional
Roman practice. With its emphasis on family-law, especially the dis-
appearance of the patria potestas of the family head over his offspring once
they had attained the age of majority, the Ecloga marks an important stage
in the consolidation of the orthodox nuclear family of the middle ages.8
The stimulus to this legislative activity on the part of both Church and
state, it has been suggested, is to be located in the reaction to a very
different tendency in many parts of the empire, namely the reassertion
from the later third century of close ties of consanguinity in marriage
arrangements, and in particular the emphasis on cross-cousin marriage
(that is, between sons of sisters and daughters of brothers of the same

6
See Justinian, Nov. 94 (a. 538), 4-5; and compare with Ecloga II, 6. See H.J. Wolff, The
background of the post-classical legislation on illegitimacy', Seminar 3 (1945), 2 1 ^ 5
(repr. in Opuscula Dispersa (Amsterdam 1974), pp. 135-59). For the ninth-century evi-
dence, see, for example, Epanagbge XVI, 1.
7
See the summary in Beck, Kirche, pp. 8 7ff. with literature.
8
See canon 54 (Mansi XI, 968D-E); Ecloga II, 2. For some general comments, see Sp. Troia-
nos, "H ixeTdpacrri dm) TO pwumxd OTO Bv£avTivo 8ixaio\ in Seventeenth International
Byzantine Congress, Major Papers, pp. 211-35; and D. Simon, 'Zur Ehegesetzgebung der
Isaurier', Fontes Minores I (1976), 16-43, see 30-42 (although the novel in question has
now been shown to date from the reign of Leo V, the general argument made in this article
remains valid). For the redating, see 0. Kresten, 'Datierungsprobleme "Isaurischer" Ehe-
rechtsnovellen. I. Coll. 126', Fontes Minores IV (1979), 37-106, see 49-53. For the Roman
background, see the survey of Beryl Rawson, 'The Roman family', in The Family in Ancient
Rome: New Perspectives (London and Sydney 1986), pp. 1-57; and esp. W.K. Lacey, 'Patria
Potestas', ibid., pp. 1 2 1 ^ 4 . For a summary of the Byzantine developments, see
A. Schminck, art. 'Ehebruch', in Lexikon des Mittelalters HI (1986), 1660, and Zacharia,
Geschichte, pp. 55-83.
 Infrastructures and hierarchies 3 79

parental group) and the consequent importance of uncle-nephew relation-

ships. It has been argued that such consanguinous marriages, designed to
maintain the continuity of lineage-property, were the norm in the period of
the principate.9 But, in fact, the elite strata of Roman society in the first
three centuries A.D. and, as far as can be determined, much of provincial
society, placed no more emphasis on this type of relationship than on the
alternative exogamous forms.10 A tendency to endogamy can be detected
in the West, under the influence of Germanic settlers and their cultural
traditions; while in parts of the East, especially Syria and Palestine, the
revival of such endogamous structures has also been taken to mark the
strengthening of village-based as opposed to urban-based forms of prop-
erty acquisition and transmission. But whereas in the West, the estab-
lishment of the extended kinship-group or clan is firmly attested during the
period from the sixth century and after, it is clear that in the East, this
'revival' (if the evidence for it has been correctly interpreted, a point about
which there is still some debate) was strongly disapproved of by the state
and frowned upon by the Church. 11 The reasons, I suggest, lie to a degree
at least in the increasing cultural introversion within the Eastern Christian
imperial lands from the sixth century, a phenomenon discussed at greater
length in chapter 11, below; and in an increasing desecularisation of ideas
about society and the relationship between God and man. In addition, the
Constantinian legislation of the fourth century on sexual relations and
marriage - even though later considerably revised and relaxed under
Justinian - had condemned and proscribed many features of the traditional
Roman marriage law and concubinage, effectively forcing changes in
patterns of marriage behaviour among the social elite which, in spite of
later ameliorations, had permanent results. 12
All these pressures, together with the legal embodiment of the Church's
assumption that parents were responsible for ensuring that their children
9
See Patlagean, Pauvrete economique, pp. 118-28.
10
See Brent D. Shaw and Richard P. Sailer, 'Close-kin marriage in Roman society?' Man,
new series 19 (1984), 432-44; note also K. Hopkins and G. Burton, 'Ambition and
withdrawal: the senatorial aristocracy under the emperors', in Death and Revewal. Social
Studies in Roman History, ed. K. Hopkins (2 vols., Cambridge 1983), vol. II, pp. 120-200.
11
See, for example, Justinian, Nov. 22, the 48 paragraphs of which deal with a whole range
of matters related to the question of marriage. See Patlagean, Pauvrete economique,
pp. 118ff. For the West, compare the comments of D. Bullough, 'Early medieval social
groupings: the terminology of kinship', Past and Present 45 (1969), 9-18. See also
K. Schmid, 'fiber das Verhaltnis von Person und Gemeinschaft im friiheren Mittelalter',
Fruhmittelalterliche Studien 1 (1967), 2 2 5 ^ 9 , the contributions in T. Reuter, ed., The
Medieval Nobility (Amsterdam 1979), and C.J. Wickham, Early Medieval Italy (London
1981), pp. 115ff.
12
See CTh. VIII, 16.1 (a. 321); and the commentary of HJ. Wolff, 'Doctrinal trends in
post-classical Roman marriage law', Zeitschrift der Savigny-Stiftung, romanist. Abt, 67
(1950), 261-319.
380 Byzantium in the seventh century

were betrothed and married, strengthened the nuclear family unit at the
expense of both older, more diffused forms of kinship group, and of simple
concubinage and cohabitation. In effect, the Church tried to extend its
supervision over all areas of sexual relations by confining and limiting
them to a specific form of marital relationship, accompanied by a series of
concomitant parental rights and duties, property-regulating stipulations
and associated institutions such as God-parenthood. Another result was
the legal and the moral privileging of legitimate as opposed to illegitimate
children in respect of inheritance.
Parallel to these developments there evolved also the institution of
spiritual parenthood, between godparents and godchildren. Even these
relationships were assimilated within the general system of prohibited
degrees of affinity, so that the two systems constituted a mutually reinforc-
ing whole. Indeed, in Christian thought the spiritual relationship was
regarded as much superior to the merely fleshly kinship bond itself.13
These changes did not occur evenly or all at once. Their evolution begins
already in the later third and early fourth centuries, if not before, in Roman
civil law; and still in the seventh century the stable system of later centuries
had not been attained. In the later ninth century, Leo VI had still to
legislate on relationships of spiritual and adoptive parenthood in order to
deal with continued failures to observe the relevant canons of the Quinisext
council, a point which demonstrates nicely the often considerable gap
between legislative theory and social practice. A similar divide no doubt
existed in respect of many other of the developments outlined above. But
the popular response to Heraclius' marriage to his niece Martina shows
that by this time at least the idea of conjugal and sexual ties between such
close kin was regarded with considerable distaste. One must recall, of
course, that figures greatly in the public eye attract more attention, and are
often expected to behave more closely in accordance with the codes
stipulated by the moral universe, than the rest of society. But the story is at
least indicative.14
« See Ecloga II, 2, and XVII, 25 and 26; cf. canon 53 of the Quinisext (Mansi XI, 968C). For
the Justinianic legislation (less restrictive) see CJV, 4.26/2 (a. 530); and compare with the
ultimate results of these developments in terms of the attitudes to marriage, kinship, and
'blood' relationships in modern Greek rural society, as surveyed in J. Du Boulay, 'The
blood: symbolic relationships between descent, marriage, incest prohibitions and spiritual
kinship in Greece', Man, new series, 19 (1984), 533-56, with literature. In general on
godparenthood, see E. Patlagean, 'Christianisation et parentes rituelles: le domaine de
Byzance', Annales E.S.C., 33 (1978), 625-36 (repr. in Structure sociale, no. XII); and, for a
recent survey of later developments, Ruth Macrides, 'The Byzantine godfather', BMGS 11
(1987), 139-62. See also the essay by A.A. Cekalova, in Udal'cova, ed., Kul'tura Vizantii.
14
See chapter 8 above, p. 304; and see Leo VI, Nov. 24 (in P. Noailles and A. Dain, Les
Novelles de Leon VI le sage: texte et traduction (Paris 1944), pp. 92-5). Note also the
comments of the twelfth-century canonist Balsamon on the relevant canon, 53, of the
Quinisext (Rhalles-Potles, Syntagma, vol. II, pp. 430f.). Whether the patriarch Sergius was
 Infrastructures and hierarchies 381

In addition, a very important change in attitudes towards the family

seems to have taken place during the fourth century, most clearly repre-
sented in the startling expansion of monasticism and the evolution of the
concept of an alternative, celibate life devoted to God. The specifically
Christian tendency to reject marriage and the reproduction of the family
became significant enough to make celibacy a reasonable social alter-
native for many, although the motivation for such a choice varied greatly
from individual to individual - poverty, tax-burdens and debts, the
onerous duties of the curial class, or at least the poorer members thereof, as
well as less concrete religious grounds - all played a role. By the sixth
century, the choice was clear and open to all, and it retains henceforth a
central position in late Roman and Byzantine thinking about the family,
sexuality and 'society': marriage, family and the reproduction of the social
ties and the economic relationships which accompanied them on the one
hand; rejection of these relationships, with a clear choice of celibacy in
contrast, on the other. The stark opposition between the two modes of
living reinforced the identity and the perceived role of the family unit. 15
The Christianisation of the institution of marriage and the system of
kinship-relations revolving around the family was thus a gradual and
many-sided process. A crucial impetus was lent to the development by
Justinian's order that the canons of the Church should henceforth have the
validity of state law - although there still remained substantial areas of
disagreement and even conflict between the secular and the ecclesiastical
legislation. The process was further consolidated by the appearance of the
first collection of nomocanones in the later sixth century, in which both
secular and ecclesiastical regulations were assimilated in a common
corpus. The Ecloga of the early eighth century is based in these respects
even more clearly on the canon law of the Church and is usually taken to
mark an important moment in the process of change. 16
None of this is to suggest that the nuclear family unit had not always
been at the basis of most kinship structures which existed in the east
Mediterranean world. Indeed, there is a problem in reconciling the gen-

also involved in this is unclear: see P. Speck, 'Die Interpretation des Bellum Avaricum und
der Kater MexXeia/ire', in Poikila Byzantina VI: Varia II (Bonn and Berlin 1987),
pp. 374-5.
15
See Patlagean, Pauvrete economique, pp. 1 2 8 ^ 5 ; and note the review by L. Couloubarit-
sis, in Revue beige de philosophie et d'histoire 60 (1982), 374-82, esp. 379f.
16
See Justinian, Nov. VI, 1.8 (a. 535); for the nomocanons, see Beck, Kirche, pp. 145ff.; and
see A. Schminck, art. 'Ehe\ in Lexikon des Mittelalters III (1986), 1641^4.
For the different trajectories of Eastern and Western Christianity in these respects, see
J. Goody, The Development of the Family and Marriage in Europe (Cambridge 1983),
pp. 103ff.; but note the critique of C.J. Wickham, in Journal of Peasant Studies 14 (1986),
129-34; and D. Herlihy, Medieval Households (Cambridge, Mass, and London 1985),
pp. 10ff.; cf. also A. Esmein, he Manage en droit canonique (Paris 1929-35).
382 Byzantium in the seventh century

eralised use of the term familia in the sources, to refer to the wider
'extended' family network, with that for the actual practical, functioning
family unit, which was clearly the 'nuclear' unit. For the latter group no
real equivalent term existed; and it has been convincingly shown that,
until the wider network began to lose its more extended and ramified
elements, this remained the case. This evolution was part and parcel of the
social and cultural changes of the period from the third century on, of
course, and it is only when they are well under way that familia comes
actually to represent the reality of the elementary family group. The
practical disappearance after the fifth century of clan names, and the
reappearance of nomina gentilia only during the ninth century on any
substantial scale in the Byzantine world - a development which corres-
ponds more or less exactly with the rise of the new aristocracy of the
middle Byzantine period - is significant. The intervening centuries were
marked by the radical loosening of such wider ties and the absence of any
integrative kinship structure higher than the elementary family. The
almost complete lack of family names on seals or in the literary sources -
excluding nicknames or ethnonyms - during the later sixth, seventh and
earlier eighth centuries is in this respect particularly relevant. 17
By the end of the seventh century, therefore, Christian marriage seems to
have become, in legislative terms at least, the preferred form of social-
sexual organisation; while the elementary, or nuclear, family was rapidly
becoming the only significant unit of any social relevance. The legislation
itself marks only specific moments in the evolution and generalisation of
the social institutions themselves, of course, and its clear-cut formulations
can by no means be taken to mean that other forms of familial structure
did not continue to exist for some considerable time. There remained still
substantial areas of conflict between Church and state - over the question
of divorce, for example, and especially of remarriage, and over the question
of whether civil or ecclesiastical courts should have the final competence
over matters pertaining to marriage and divorce; but the attention paid to,
and emphasis placed upon, the elementary family unit, with its limited
lateral extensions, and the parallel institutions of spiritual parenthood,
demonstrates that by the later seventh and early eighth centuries these
were well on the way to becoming the key terms within which the social
relationships of Byzantine society were understood.

17
See the excellent analysis of Brent D. Shaw, 'The family in late Antiquity: the experience of
Augustine', Past and Present 115 (May 1987), 3-51, with extensive literature; and the
comments of B. Rawson, The Roman Family, pp. 7ff. For the Byzantine period, see F. Win-
kelmann, 'Probleme der gesellschaftlichen Entwicklung im 8./9. Jahrhundert', in Seven-
teenth International Byzantine Congress, Major Papers, pp. 577-90, see 58Iff., and Quellen-
studien, pp. 143ff.
 Infrastructures and hierarchies 383

These changes have been examined at length elsewhere,18 and we need

not dwell on them further here. What has, on the whole, not received
adequate attention is their long-term role within the structure of late
Roman and early Byzantine society. For the substitution of the limited
nuclear family unit for the extended kinship group from the fourth to the
seventh century meant in the first instance a potential fragmentation of
kinship solidarity and property transmission. While in the latter case this
will hardly have affected the greater part of the peasant population, it
certainly affected the property-owning elite throughout the empire. And
kin-groups must have been affected by this compartmentalisation. The
Church now replaced the extended kinship-community. Indeed, it can be
plausibly argued that the institution of godparenthood was an effective
social-communal substitute for wider kinship ties, if not in respect of
property, then certainly in respect of social and family solidarity.19
The close-knit, autonomous elementary family unit was the result of all
these changes - already in the ninth century it contrasted with the basic
family unit in the West 20 - but it seems unlikely that Christian ideology
alone, even in conjunction with the various late Roman factors referred to,
brought about the qualitative transformation which has been described. In
the West, the Church was engaged for a far longer period in trying to
enforce canon law evenly and throughout the area under the jurisdiction
of Rome. 21 Likewise, the legislation itself reflects the theoretical demands of
18
On all these developments, see Jones, LRE, vol. II, pp. 972ff.; and esp. B. Cohen, 'Betrothal
in Jewish and Roman Law', Proceedings of the American Association for Jewish Research 18
(1949), 67-135; J. Gaudemet, 'Les Transformations de la vie familiale au Bas-Empire et
l'influence du Christianisme', Romanitas 4/5 (1962), 58-85, and L'Eglise dans Vempire
romain (IVe-Ve siecles) (Paris 1957), pp. 515ff.; E. Herman, 'Die Schliessung des Verlobnis-
ses' (see note 3, chapter 10, above), and 'De Benedictione Nuptiali quid Statuerit Ius
Byzantinum sive Ecclesiasticum sive Civile', OCP 4 (1938), 189-234; H. Hunger, 'Chri-
stliches und Nichtchristliches im byzantinischen Eherecht', Osterr. Archivfiir Kirchenrecht
18 (1967), 305-25 (repr. in Byzantinische Grundlagenforschung XI (London 1973));
J. Gaudemet, La Formation du droit seculier et du droit de Veglise aux IVe et Ve siecles (Paris
1957); A. Hage, Les Empechements de manage en droit canonique oriental (Beirut 1954); E.F.
Bruck, 'Kirchlich-soziales Erbrecht in Byzanz. Johannes Chrysostomus und die mazedoni-
schen Kaiser', Studi in onore di S. Riccobono (3 vols., Rome and Palermo 1933), vol. in,
pp. 377-423; A. Christophilopoulos, Xx^osig yovecov xal TSXVCJV xara rd Bv^avnvov
bixaiov (Athens 1946); Patlagean, Pauvrete economique, esp. pp. 114-28, and 'Christiani-
sation et parentes rituelles' (see note 13, chapter 10, above). Note also the remarks of
Kazdan and Constable, People and Power in Byzantium, pp. 32ff., and the old, but still
valuable general survey of Zacharia, Geschichte, pp. 71-6.
19
See Macrides, 'The Byzantine Godfather' (see note 13, chapter 10, above), esp. 155ff.
20
See K. Ritzer, Eormen, Riten und religioses Brauchtum der Eheschliessung in den christlichen
Kirchen des ersten Jahrtausends (Miinster 1961), p. 104.
21
See Goody, The Development of the Family and Marriage in Europe, passim; and see also
D. Herlihy, 'Family solidarity in Medieval Italian history', in Economy, Society and Govern-
ment in Medieval Italy: Essays in Honour of Robert C. Reynolds (Kent, Ohio 1970),
pp. 177-84. For the results of these developments, see the comments of A.P. Kazdan,
'Small social groupings (microstructures) in Byzantine society', Akten des XVI. Int. Byzanti-
384 Byzantium in the seventh century

the Church or the state, as is clear from the canons of the Quinisextum, for
example, where canons 53 and 54 assume a continuing gulf between
theory and practice. Traditional relationships between cousins, for
example, seem to have survived.22 What other factors, therefore, contri-
buted to the transformation?
In the first instance, we may recall the general situation of early
Byzantine culture. The introversion described in an earlier chapter affected
the whole of society, albeit in different ways in different contexts. 23 It is
surely no accident that the dominance of the elementary family unit offers
parallels to this development. It represents a turning-away from the wider
context of the kinship group, a system of social relations in which the units
that make up the group exchange property - including wives - on the
basis of a self-reproducing, horizontal reciprocity. Instead, the isolated
nuclear unit, which depends absolutely on the wider institutions of society
for its legitimation and its perpetuation (and less as a group than as a
community of individuals) takes its place. Under the powerful influence of
the Church, which stressed the reliance of human beings upon heavenly
authority, rather than upon the community and its ties, the tendency can
only have been encouraged. The imperial ideology itself reflected this
emphasis on heavenly authority, of course; the harsh penalties imposed in
the Ecloga serve to underline this, the stress on conformity and orthodoxy
across the whole society, unity of belief and culture, and the determination
to eradicate any form of oppositional thinking or practice, are significant
features.24 The extent to which formal statements in legal and canonical
writings reflect, or fail to reflect, the wide diversity of local and regional
practice, of course, cannot be gauged. But the ideological message and
intention is clear enough. 25 It is perhaps also useful to remember that the

nisten-Kongresses, II/2, pp. 3-11 ( = JOB 32, 2 (1982)); but with the reservations
expressed by Laiou, (see note 5, chapter 10, above), 198-9 and note 32.
22 23
Mansi XI, 968C. See chapter 9, pp. 348ff.
24
Cf. Ecloga XVII, esp. 2 3 - 7 and 30ff. It should be remembered that the various punishments
involving corporal mutilation listed in the Ecloga and embodied in the Farmers' Law, for
example, were not new: from late Roman times (and indeed earlier) physical mutilation
had been practised, intended as a physical demonstration of the authority vested in the
state and its institutions, and the marginalisation and exclusion, both literally and
metaphorically, of offenders. On the 'philanthropic' element behind the nature of punish-
ment in the Ecloga, see T.E. Gregory, 'The Ecloga of Leo HI and the concept of Philanthrop-
ia', BvfcvTLvd 7 (1975), 267-87, with the remarks of D. Simon, in BZ 69 (1976), 665.
For a differently nuanced view, see E. Patlagean, 'Byzance et le blason penal du corps', in
Du chdtiment dans la cite: supplices corporels et peine de mort dans le monde antique (Rome
1984), pp. 4 0 5 - 2 6 , who - in my view quite rightly - attempts to relate the types of
punishment to a symbolic system in which the body and its parts were related through
metaphor to both power and sexuality. A more traditional approach is that of Sp.
Troianos, 'Bemerkungen zum Strafrecht der Ecloga', in A(piep<ofjLa ordv NCxo Xfioptivo
(2 vols., Rethymnon 1986), vol. I, pp. 9 7 - 1 1 2 .
25
See Hunger, 'Christliches und Nichtchristliches', 3O5f. and 324f. (cited note 18, chapter
10, above).
 Infrastructures and hierarchies 385

concept of apartness and privacy seems to have received a considerable

stimulus at just this time. A number of historians have noted in particular
the great contrast between the open, public lifestyle of the classical polis,
with its style of domestic architecture, which seems to last until the sixth
century, and the closed, private, almost secretive lifestyle of the Byzantine
town-dweller from the seventh century on. This aspect of Byzantine life is
especially evident in Byzantine literature of the period, which speaks of,
and often emphasises, the importance of privacy, and the closed-off,
compartmentalised organisation of domestic architecture. 26
In the second place, the physical and cultural disruption which the
Byzantine world experienced from the middle of the seventh century, and
earlier in the West, must also have had effects on the population in respect
of its social-institutional organisation and coherence. Many communities
were dislodged or destroyed, as we have seen. 27 The traditional ties of
dependence will in many regions have suffered, and - given the circum-
stances - have become irrelevant to the immediate demands of individual
families or communities. In such conditions, reliance upon the state and
the Church, both materially as well as spiritually, for leadership and
defence, arid upon heavenly intercessors, or earthly mediators with access
to the heavenly realm, for explanations and reassurance, must have
increased; and indeed, we know that in the latter case this did occur.
In the third place, the presence of large numbers of soldiers in the
Anatolian provinces, mostly for the first time, must have been significant.
The permanent garrisoning of substantial forces in provinces which had
hitherto had - with some exceptions - little or no military presence, and
the logistical problems alone that this brought with it, must have affected
local society in many ways. 28 The increasing emphasis on localised
recruitment, for example, and the status and privileges which soldiers
brought with them, made them as individuals an attractive proposition to
a family with daughters of marriageable age or younger. The socio-sexual
intervention of the first and second generations of these newcomers will
have affected the demography of the provinces where they were stationed
from the 640s and 650s, and consequently the traditional bonds between
and within communities in respect of patterns of marriage and kinship. We
have already seen that soldiers appear to identify themselves with provin-
cial and local issues in the later years of the seventh century; and there is
no reason to doubt that this local identity is rooted in their incorporation or
absorption into local society. Given this influx of alien groups into the
26
See Kazdan and Constable, People and Power in Byzantium, p. 50 and Mango, Byzantium:
the Empire of New Rome, p. 83. Note also P. Brown, The Body and Society. Men, Women and
Sexual Renunciation in Early Christianity (London and Boston 1988), pp. 320-1.
27 28
See chapter 3 above. See chapter 6, pp. 220ff.
386 Byzantium in the seventh century

social and demographic structures of the different regions of Asia Minor,

and given the fact that the armies were clearly billeted or quartered on a
relatively widespread basis, the effect must have been to introduce a whole
new set of variables into the social make-up of the districts concerned. And
this will undoubtedly have affected patterns of marriage within local
communities, along with the patterns of distribution of wealth both in land
and in movable property.
The reinforcement of local populations from outside in this way - even
though the numbers in absolute terms may have been relatively limited -
must, therefore, have affected the demographic balance in m a n y villages,
modifying the customary patterns of conjugal selection. It will have also
directly affected the structure of kinship relations and contributed thereby
to further strengthening the nuclear family unit and to breaking down
endogamous marriage practices. Soldiers' status and privileges, both juri-
dically and fiscally, will have been a major factor.
The last point must remain, of course, to a large degree, hypothetical.
But it is an hypothesis based firmly in the actual context, and in respect of
the known developments of the period it possesses a coherency and a logic
that cannot be denied. There is some evidence to support it, of course; for
the lateral fragmentation of lineage and the centrality of the nuclear family
have been emphasised also in respect of the village community. The
so-called Farmers' Law exemplifies the point, for it deals very clearly with a
community of separate, independent households, each with (or without)
its own property in land and livestock, each carefully guarding its rights vis
a vis its neighbours, and each physically marking itself off from its neigh-
bours with fences, walls, ditches or other forms of boundary. 29 The key
feature of Byzantine rural society from the time of the Farmers' Law and
after was the possession by each family unit - each head of household, in
effect - of its own holding, whether owned or leased on a long-term
basis. 30 Common rights took the form of collective interests in specific and
often seasonal activities, such as the pasturing of animals on uncultivated
fields, or making hay; or in village land held to belong to the community; 31
or in respect of the fisc, as the common liability for fiscal assessments,
especially on abandoned or uncultivated land. 32 Village solidarity appears
as a function of specific activities and fiscal obligations, therefore, as the
totality of socially distinct yet interconnected family units; and as the
public expression - religious festivals and activities, for example, or on the
29
See Farmers' Law, articles 50, 5 1 , 58 a n d 6 6 .
30
See Kazdan, 'Small social groupings' (cited note 2 1 , chapter 10, above), pp. 4 - 5 a n d A.E.
Laiou-Thomadakis, Peasant Society in the Late Byzantine Empire: A Social and Demographic
Study (Princeton 1 9 7 7 ) , esp. pp. 72ff.
31
See Farmers' Law, articles 23ff., 30ff. a n d 8 1 - 8 2 .
32
Farmers' Law, articles 18 a n d 1 9 .
 Infrastructures and hierarchies 387
occasion of marriages or funerals - of a common identity of interests in
respect of the outside world.33 But the economic independence of the
households, and the structured competition within the village, the possi-
bility of both the fragmentation of property and the amassing of wealth,
illustrate the separateness and the potential internal structural oppositions
latent within the community.34 The medieval village society depicted in
the Farmers' Law, with its constant emphasis on neighbours' rights, or in
the lives of saints such as Theodore of Sykeon or Philaretus the Merciful
(set in the early seventh and later eighth centuries respectively) was a
community of independent and often competing economic units. Village
solidarity through kinship - even where marriage was not unusual within
a single community - was not a significant structuring element, at least in
respect of enhancing mutual solidarities and promoting interdependencies.
In these respects, of course, medieval village society was not dissimilar to
that of contemporary or recent Greek and Aegean rural culture.35
These considerations bring us to a fourth, and much wider, point:
namely, the centrality of the state as a source of power and authority, a
determinant of social status, and a route to personal advancement.

HIERARCHIES, SOLIDARITY AND STRATIFICATION

As we have seen, the political and economic irrelevance and, in many
cases, the virtual physical disappearance, of a great number of provincial
cities within the area still under effective imperial control after the 650s,
left a vacuum in provincial society, a vacuum that was partiallyfilled,in
terms of local modes of access to the capital, by soldiers. In particular, the
army represented a vocal source of provincial opinion - refracted though it
may have been through the localised and self-interested concerns of the
soldiery and their leaders - and in many ways, given the disappearance of
the independent municipalities of the previous era, the only source which
had any authority. The army was, of course, only one aspect of the
institutions of the state. But there were other aspects which were equally
important. The demise of the corporate municipal culture which had
survived - albeit in a much debased form - into the early seventh century,
buttressed, it is true, by the state itself, meant not only the end of local
representation through the medium of the cities, in the structure of state
and administration at large; it meant the transfer of provincial interest to
33
Compare the common front presented by the inhabitants of Amnia to the imperial visit to
Philaretus' house: Vita Philareti, 1 3 7 . 2 3 - 6 .
34
The varied forms of independent exploitation of resources mentioned in the Farmers' Law
(e.g. art. 81) and the social differences noted in, for example, the Life of Philaretus, testify
to this. See Kopstein, 'Zu den Agrarverhaltnissen', pp. 5 0 - 3 .
35
See J. Du Boulay, Portrait of a Greek Mountain Village (London 1974), for example.
388 Byzantium in the seventh century

the centre of power, wealth, authority and status, to Constantinople. Only

state and Church now offered any prospect of advancement, of social
improvement, and yet at the same time held out the promise of stability;
and the state, with its complex system of ranks and titles, represented in
addition an alternative social structure, a bureaucratic, hierarchical
society dependent upon the emperor - God's appointee - with its own
rewards and its own values. Even as a simple soldier, one received an
enhanced social status, with tangible benefits in respect of fiscal exemp-
tions, and with the possibility of promotion, or attachment to the personal
retinue or guard of an officer, as powerful inducements offering potentially
greater rewards. Various classes of clerk and bureaucrat received similar
benefits. It is reasonable to suppose, therefore, that in a society where
lateral solidarity through kinship had begun to fragment, and where
access to power at the local, provincial level through the municipalities
had been blocked, horizontal and intra-provincial social mobility and
stability was replaced by the competitive, vertical mobility of individuals
operating within the apparatuses of state and Church.
That the enhanced development and prominence of the elementary
family unit during the later seventh century is not merely hypothetical is
borne out by other forms of evidence: in the first instance, the effective
disappearance of the old senatorial-municipal elite, and its subsumption
within and replacement by a new, meritocratic elite of provincials of
widely varied social and cultural backgrounds. Armenians in particular
have been singled out, but other ethnic groups are also represented. The
crucial point is that there are a great number of individuals of high rank
who are clearly 'new men'. Their presence is itself suggestive of the open,
competitive and individualistic nature of the upper levels of Byzantine
society.36
In the second place, the dominant position of the system of titles and
precedence dependent upon the emperor and court during the seventh and
eighth centuries and after is crucial. Social and economic status in the sixth
century and before had been marked out by a variety of attributes: family
name, although less important than during the principate, was an
element; occupation of an imperial position and the possession of an
appropriate title, active or honorific, another; origin, and position in the
local community - the municipality - constituted a third element; and
wealth, together with membership of the senate and the possession of the
appropriate cultural capital, a fourth. These could be, and were, combined
in a variety of ways. But they constituted together a rather loose, open and
pluralistic system - wealth and family background, along with the appro-

36
The classic survey of Charanis, 'Ethnic changes', illustrates the point well.
 Infrastructures and hierarchies 389

priate classical education were as relevant as the office or title held by an

individual, and the two were not always or necessarily interdependent.37
Membership of the wider senatorial order, one's role in the local commu-
nity, an imperial sinecure or title, they all bestowed status and social
respect - although from the fourth century the imperial bureaucracy and
its system of precedence had become increasingly important. The relative
social and economic stability of the late Roman, east Mediterranean world,
however, along with its social and political institutions, in particular the
city and the ideology of municipal civilisation and local patriotism -
however weakened they may have become - still contributed to the
pluralism of status which is reflected in the sources until the early seventh
century. 38 By the fifth century, of course, the senate, and membership of
the senatorial order, represented the ultimate source of status, and escape
from curial burdens by the acquisition of senatorial rank pointed to a
dangerous drain of wealth and manpower from the curial order. But the
senatorial order was itself broad and even in the later sixth century
represented only superficially a unified social elite: its members were still
drawn from a wide variety of social backgrounds and attained senatorial
rank by a multiplicity of routes.
With the upheavals and the territorial losses of the Persian and then the
Arab attacks, however, and the concomitant shifts in ideological perspec-
tives which had slowly taken form during the later sixth century, late
Roman and early Byzantine society came to adopt a much more defensive
posture. While the senate itself, sitting in Constantinople, retained its
importance until the middle of the seventh century - until at least the later
years of the reign of Constans II - representing as it did an assembly of the
leading officers of the court, the senatorial order in the wider sense seemed
to lose in status and eventually to disappear during the seventh and eighth
centuries. Of course, by the middle of the sixth century, if not before, the
role of the emperor and of the court had become crucial in the selection of
the leading officials and title-holders in the state establishment, and there-
fore of the senate. 39 But given the educationally and economically privi-
leged position of the landowning social elite of the empire, the majority of
such appointees tended in any case to be drawn from their ranks, and this

37
For later R o m a n systems of status-recognition, see Jones, LRE, vol. II, pp. 543ff.; K.M.
Hopkins, 'Elite mobility in the R o m a n empire', Past and Present 32 (1965), 1 2 - 2 6 , and
'Social mobility in the later R o m a n empire: the evidence of Ausonius', Classical Quarterly,
n e w series, 1 1 (1961), 2 9 - 4 8 ; R. MacMullen, 'Social mobility a n d the Theodosian Code',
JRS 54 (1964), 4 9 - 5 3 .
38
See Jones, LRE, vol. II, pp. 546ff. a n d 7 3 7 - 5 7 ; a n d chapter 4 above.
39
Jones, LRE, vol. I, pp. 3 8 7 - 9 0 , a n d vol. II, pp. 55Iff.; note also Cameron, 'Images of
authority', 27f.
390 Byzantium in the seventh century

has often given a false sense of the political hegemony of the senatorial elite
and those aspiring to membership of it.
In fact, the crucial position of the emperor in the whole edifice becomes
really clear only during the seventh century, for it is then that the
economic and cultural dominance of the old broad senatorial estab-
lishment, with its roots ultimately in the landowning, but still municipal,
elite of the provinces, finally crumbles under the various internal and
external pressures that have been discussed already. Its survival into the
middle decades of the seventh century is suggested by what little prosopo-
graphical evidence there is; beyond the 650s and 660s, however, it seems
clearly to have been unable to provide the staff and the leaders of the state's
apparatus that it had done hitherto. Increasingly, 'outsiders' of all kinds,
linguistically, ethnically and culturally, appear in important positions in
the administrative apparatus of the state, military and civil.40 This
phenomenon is accompanied by a marked increase in the prominence and
relative value of what are in later sources referred to as Imperial' as
opposed to 'senatorial' titles of rank, a distinction found in treatises on
titulature and precedence from the ninth century on, but which has its
roots in the late Roman period, more particularly in the developments of
the seventh century.
The original distinction is clear enough in the sixth century. As we have
seen, those who belonged hereditarily to the senatorial order bore the title
of clarissimus, and this is what was passed on by higher-ranking persons -
of illustris rank, for example - to their sons. By Justinian's reign, only those
who also held the titles of patricius, ex-consul, illustris and illustris inter
agentes could sit on the senate proper; and of these, it was mostly the active
illustres, who bore the epithet gloriosus or gloriosissimus (Gr. endoxotatos)
who represented the real power and who also occupied the higher posts,
civil and military, in the state.
These epithets all signified membership of the wider senatorial order.
They were not equated with specific positions or functions, referred to as
dignities, dignitates. Apart from titles awarded by the emperors in excep-
tional circumstances, such as nobilissimus, curopalates, Caesar (the latter
regularly employed during the sixth century to signify the emperor's
intended successor) and more commonly patricius, dignities were in the
first instance derived from offices, which could be bestowed on an active
basis, a titular basis (including the cingulum, or mark of office) or on an
honorary basis (involving the title only). It is likely that initially at least a
titular office was bestowed on persons who had actually held the office in
question. But by the sixth century the titles of consul, prefect and magister
40
See chapter 4, p. 153-72.
 Infrastructures and hierarchies 391

militum were conferred regularly on persons who held or had held no

related office. At the same time, the highest offices were automatically
associated with senatorial status. 41
The lower palatine titles represented a different route of access to
privilege and - potentially - senatorial status. They were drawn from
active, titular and honorary office-holders, at court or in absentia, of posts
in the various palatine scholae: the spathariU candidate scribones, stratores,
silentiarii, cubicularii, even the imperial bath-attendants, balnitores, and
palatine doctors, archiatroi.
The hallmark of the late Roman system, therefore, was the fixed associ-
ation of titles - dignities - with offices. The few exceptions have already
been mentioned. During the seventh century, however, this relationship
was severed. The original association of the later Byzantine dignities was
still recalled, so that the Kletorologion of Philotheus, for example, along
with many other sources, notes the division of titles into senatorial and
imperial.42 But the senatorial order itself ceased to exist, as the dis-
appearance of the epithets which denoted it illustrates - endoxotatos,
gloriosus, illustris, clarissimus and related grades do not survive into the
second half of the eighth century. 43 At the same time, a distinction
41 42
See chapter 4. See Klet.PhiL, 87.32-3; Oikonomides, Preseance, pp. 295-6.
43
The last datable reference to gloriosissimus/evbo^oiaiof; I can find: Ecloga, pr. 40-1 and
103; see P. Koch, Die byzantinischen Beamtentitel (Jena 1903), pp. 69ff. The title clarissi-
mus had become more or less meaningless by the seventh century, although it continued
to have some application and value in the exarchate of Ravenna: see Brown, Gentlemen
and Officers, p. 133 with note 11. Illustris disappears from use during the seventh century.
For examples, see Maximus Confessor, Ep. XIII {PG XCI, 5O9B) and Ep. XLIV (644D) - note
that in the latter case Maximus uses the appropriate epithet for an illustris -
|i,E7aXoTrpe'irecrraTO<; - which suggests that the illustrate still retained a degree of social
relevance and status; see also Martini Papae, Ep. II (Mansi X, 825A), probably to the same
Peter as in letter XIII of Maximus. The letter is datable to 646. See also Miraculi S. Demetrii,
161.7 (Lemerle); Miracula S. Artemii, 42.7 - where again the illoustrioi are taken for
granted as an ordo distinguished by their clothing, signifying in the last example the
members of the Constantinopolitan senate. For seals, see Schlumberger, Sigillographie,
pp. 518-19; cf. Zacos and Veglery, index V, p. 1884, s.t. IXAotKXTpios. Some 35 persons
are listed bearing this title, some with administrative functions, such as trakteutes of the
islands (914A, cf. 909A) or dioiketes (131), all dating to the sixth and seventh centuries.
See also Laurent, Orghidan, 273, and Vatican, 52; and note Schlumberger, Sig., p. 518,
no. 4, for a seventh-century seal of an illoustrios and hypatos, possibly implying that
illoustrios was still associated with senatorial rank; and Konstantopoulos, Molybdoboulla,
no. 295, for a seal of John, illoustrios and anthypatos (seventh to eighth century). But by
the later seventh century a seal of Theodore, illoustrios and basilikos chartoularios (Laurent,
Orghidan, 273) suggests a relatively low value for the title. After this time, the title
disappears. The epithets endoxotatos and peribleptos are revived in the later ninth century
and applied to magistroi and patrikioi, but bear no relation to a senatorial ordo. See, for
example, Laurent, Orghidan, 186 and 187 (tenth century), and Vatican 51 (eleventh
century), and p. 37 with notes 2 and 3 for commentary. Many other epithets used in the
later Roman period for senators occur also in middle Byzantine texts; but these appear to
have been used with little or no technical precision. See the list in Yannopoulos, La societe
profane, p. 75 note 458.
392 Byzantium in the seventh century

between active posts and the older order of dignities appears, so that the
posts known in the later period as d£ica 8td \6yov (meaning that they
were awarded by the emperor's word), actual functional positions, while
they could be held on a titular basis (chrpaTcas), were held in addition to a
title from the older order of dignities; the latter represent positions in the
hierarchy only. The development reflects a major change from and elabo-
ration upon the late Roman system, according to which any office among
higher grades could be awarded as a dignity of one sort or another and
was recognised as such even when it was at the same time an active
appointment. 44 The reasons for the change are not difficult to see. The
great government departments of the late Roman state were for the most
part fragmented during the seventh century, and a whole new range of
leading posts came to the fore. The majority of the older functional posts
became obsolete and disappeared entirely; only those dignities which were
already forming a distinctive order of titles and honorific ranks, alongside
the newly created or newly prominent functional posts of the seventh
century Byzantine state, survived.
The textual and sigillographic material for the period illustrates the
development. In the first place, a reference in the Typos of Constans II of
648, followed by an edict of Constantine IV in 680, orders the punishment
of laypersons who do not conform to the imperial order by depriving them
of their axial zone or strateia.45 Here are the three Latin terms of dignitas,
cingulum and militia, and it seems reasonable to assume that they represent
the traditional late Roman distinction between, respectively, the possession
of an honorary or actual dignitas; the conferment of an office (whether in
actu or in vacante) and straightforward service in a branch of the imperial
civil or military establishment.46 The distinction between axia and strateia
is more clearly expressed in the Ecloga, where a difference between dignity
and function is clear; although the former may well signify also the
assumption of an accompanying high post. 47 But the Byzantines them-
selves used the terms interchangeably, which leaves the modern commen-
tator, unsure of the exact context or the intentions of the writer, with a
number of difficulties.48
What is clear, however, is the formation and existence by the later

44
See chapter 4. For the later system, see the discussion of Oikonomides, Preseance,
pp. 28Iff. and Bury, Administrative System, pp. 2If. The difference was expressed in the
ways the dignities were awarded: known as d£iai 8i& (Spapeiwv, they were bestowed
through insignia. The sixty or so functioning offices were awarded directly by the
emperor, hence their being referred to as d£iai 8id \6yov. See Oikonomides, Preseance,
pp. 2 8 1 - 2 , and Bury, Administrative System, pp. 20ff. and 36ff.
4 46
* Mansi X, 1O32D and XI, 712D. See CJ XII, 8.2 (a. 440-1).
47
EclogaXTV, 1 (Burgmann, 214.636).
48
Oikonomides, Preseance, pp. 2 8 1 - 3 ; Yannopoulos, La Societe profane, pp. 30ff.
 Infrastructures and hierarchies 393

seventh century of a definite number of imperially awarded titles whose

origins lie in the dignities of the late Roman period and in the granting of
titular membership of various late Roman palatine ordines. The seals of the
seventh and early eighth centuries suggest the following titles or combin-
ations of titles, in a roughly descending order of value: (1) hypatos, (2) apo
hypaton patrikios, (3) apo hypaton, (4) patrikios, (5) protospathariosf (6)
spatharios, (7) stratelates, (8) skribon, (9) balnitor, (10) apo eparchbn, (11)
kandidatos and (12) silentiarios. Other titles, such as those of vestitor or
mandator, were also awarded. Of those listed here, (l)-(4) represent consu-
lar and senatorial dignities or combinations thereof, of the highest order in
late Roman times, and all accompanied by the senatorial epithet gloriosissi-
mus; while (7), (10) and (12) represent titular senatorial dignities of the
late Roman period. The rest represent titular membership of the appro-
priate palatine ordo. All these titles are found in combination with a variety
of functional posts; although the first four seem to represent the leading
group and are generally attached to functions of importance. 49 The title of
protospatharios appears late in the seventh century and represents in the
first instance a post, becoming by the early eighth century a title as well. 50
Similarly, the title of spatharokandidatos represents the combination of two
titles and appears in the later eighth century, although officers bearing
both titles - one probably representing a function - appear in the early
seventh century. 51
The hallmark of the Byzantine system, therefore, was the clear distinc-
tion between the chief palatine offices on the one hand and the orders of
dignities on the other. But the majority of dignities were themselves drawn
from what continued to be active palatine scholae - such as those of the
spatharii and mandatores, for example - so that the title spatharios in a text
or on a seal by itself might signify either a dignity or membership of the
actual palatine schola of the same name. Similarly, active posts could be
held in vacante, but did not represent stages in the order of dignities itself.
Careful analysis of the ways in which these titles are combined, either
together or with functional posts, has shown that the traditional notion of
a clear vertical hierarchy, based on assumptions taken from much later
49
Winkelmann, Rang- und Amterstruktur, pp. 31-7. For the senatorial titles and their
origins, see esp. Bury, Administrative System, pp. 23ff.t and Stein, Bas-Empire, vol. II,
p. 430 and note 3. The establishment of a privileged group of silentiarii with senatorial
dignity, and the consequent association of the schola silentiariorum with the senatorial
privileges lies in the time of Theodosius n. See CTh. VI, 23.4. Vestitor was likewise a
senatorial rank, whose origins lie in similar circumstances.
50
See Haldon, Byzantine Praetorians, p. 184 and note 394; Winkelmann, Rang- und Amter-
struktur, pp. 46f.
51
Haldon, Byzantine Praetorians, pp. 156 and 188, and note 417 with literature; but see also
Winkelmann, Rang- und Amterstruktur, p. 39, who argues for the appearance of the rank
spatharokandidatos in the early eighth century.
394 Byzantium in the seventh century

documents which attempt formally to set out the whole system (such as
the Kletorologion of Philotheus of the year 899), simply does not work at
this period, or indeed through much of the eighth century. 52 There does
appear to be an association between the titles of hypatos, apo hypaton and
patricius with the higher posts. But equally, persons with quite low-ranking
titles occupy powerful functional posts. At the same time, the order of
precedence among the different groups of titles seems to vary: it is unclear,
for example, whether hypatos or patrikios had precedence in the second half
of the seventh century; although patrikios seems to rise at the expense of
hypatos, which during the eighth century could also be combined with the
ranks or titles oi spatharios, silentiarios and vestitor.53 Indeed, the multipli-
city of combinations of lower titles suggests a roughly horizontal system of
alternative titles, awarded according to very approximate spheres of com-
petence. Even here, however, no exact relationship between a given title
and, say, military or civil posts seems to hold: and it has been pointed out
that in the eighth and early ninth centuries at least both posts and titles
must have been awarded on the basis of individual competence or patron-
age, rather than according to any formalised systematisation of functions
actually exercised. The same will undoubtedly have applied during the
seventh century. 54 One or two examples serve to illustrate the point. The
deacon of the Hagia Sophia, one John, a confidant of the Emperor Anasta-
sius II, was made genikos logothetes and was also given command of an
expedition to intercept an Arab fleet in 715. In a similar fashion the Abbot
Theodotus, a confidant of Justinian II, was made genikos logothetes. There
are many other examples of the practice. 55
There appears, however, at least initially, to have been an attempt to
differentiate between those who held a title only and those who held a title
which signified an actual function (such as spatharios, for example)
through the addition of the epithet 'imperial' to the title of certain officials,
presumably those in active service and in praesenti. It is noticeable that this
relates always to the titles of persons who are associated with a palatine
ordo or schola (and hence originally signifying also presence at court):
spatharioi, kandidatoi, mandatores and so forth; but never to titles awarded
on an individual basis: magistros, patrikios, hypatos, stratelates and so on, a
difference which seems to confirm the point. 56

52
See the important c o m m e n t s of W i n k e l m a n n , Rang- und Amterstruktur, pp. 1 9 - 2 8 .
53 ibid., p. 4 1 .
54 ibid., pp. 4 5 - 6 1 a n d 1 3 8 .
55
See Zacos a n d Veglery, no. 2 0 0 7 ; Laurent, Corpus, vol. II, no. 2 7 8 with sources; and
Theophanes, 367.22sq.
56
See Haldon, Byzantine Praetorians, pp. 1 8 3 - 4 , a n d Haldon's review of Winkelmann, Rang-
und Amterstruktur, in BS 47 (1986), 2 2 9 - 3 2 , see 2 3 2 .
 Infrastructures and hierarchies 395

THE NEW POWER ELITE

But what is particularly important is that the titles - including those which
ought properly to be classed as 'senatorial': hypatos, apo hypaton, apo
eparchon, stratelates, silentiarios and vestitor - now seem to be part of a
single system dependent upon the emperor's pleasure, which can be
combined with other titles according to the appointments currently or
formerly held. 57 Indeed, the titles listed above seem now to be the only
marks of membership of the senatorial 'order' itself, since the generic
epithets ofillustris and magniflcus seem to drop out of using during the first
part of the seventh century. Gloriosissimus is retained, as we have seen, in
its Greek form of endoxotatos, but only until the 740s. The fact that the first
two - hypatos, apo hypaton - were, at least until the eighth century, of high
status supports the contention that the senate as a body in Constantinople
retained a degree of prestige; but equally, the titles of eparchon and strate-
lates both lose in value. The titles of apo hypaton and hypatos had rated the
epithet gloriosus and ranked at the top of the scale according to novel 62 of
the Emperor Justinian; the titular ranks of apo eparchon and magister
militum were classed among the second grade of illustres, as magnifici.58
The devaluation of the latter titles during the second half of the seventh
century and the first part of the eighth is important, because it reflects the
firm establishment of the title stratelates, for example, as a titular dignity;
its replacement in functional and practical respects by the office and title of
strategos; and its concomitant loss of status and value as it becomes
progressively divorced from the active, functional establishment of the
state. 59 Similarly, apo eparchon loses in value over the same period; and
both titles disappear after 899. The loss of status in the case of the title
ex-prefect may reflect the demise of the praetorian prefecture and the
splitting up of its civil-administrative functions, a development outlined in
chapter 5. The sigillographic evidence makes it clear that the 'senatorial'
titles were now part of a single system of hierarchy and were awarded on
just the same basis as the palatine or Imperial' titles. Senatorial rank no
longer existed, for no senatorial order existed. Instead, a number of
senatorial titles survived, titles which presumably also conferred mem-
bership of the ceremonial body of the senate which played such an
important part in imperial and state ritual. 60
57
Winkelmann, Rang- und Amterstruktur, pp. 45ff.
58
Justinian, Nov. 62, 2 (a. 537); for the equivalence of magister militum a n d prefect see
Justinian, Nov. 70 (a. 538).
59
By the middle of the n i n t h century, it h a d come to occupy the lowest place in the
hierarchy: see W i n k e l m a n n , Rang- und Amterstruktur, p. 3 9 .
60
See Oikonomides, Preseance, p. 2 9 5 a n d references. The history of these titles in the
exarchate of R a v e n n a is, as might be expected, very similar. See the analysis of Brown,
396 Byzantium in the seventh century

What this implies, of course, is that the older senatorial order had
withered away. 61 All 'senators' were henceforth imperially sponsored, as
the incorporation of 'senatorial' titles into an imperial system of prece-
dence demonstrates. And this implies that the economic power and cul-
tural authority of the socio-economic groups from which the illustres had
been drawn had been fragmented or destroyed. In effect, while the senate
may still have had some authority in state affairs, by nature of its physical
context - in the capital city of the empire - and its traditions, it no longer
represented any sort of economic interest, a class of landowners whose
existence, however broadly defined and loosely composed in the late
Roman period, nevertheless reflected the dominance of a landowning
aristocracy of privilege, sharing in a common cultural heritage.
Additionally, the older senatorial elite had also based its position on the
tenure of high civil office, the provincial magistracies and governorships,
and so forth, and the majority of these seem either to have disappeared as
they became irrelevant to the changed situation or were very greatly
reduced in status, from about the middle of the seventh century. 62 From
this time, the senate was increasingly an assembly of imperially sponsored
title-holders, whose only real functions were to act on ceremonial occa-
sions as a symbol of Roman imperial tradition. Obviously, the senate could
still serve as a focus of opposition to, or a source of counsel and support for,
an emperor, simply because it included many of the leading officers of the
court - all those with the ranks of hypatos, apo hypaton and patrikios, for
example. This can clearly be seen at the beginning of the reign of Constans
II, at the beginning of this evolution, and in the last years of the seventh
century. 63
At first, as we have seen, the highest senatorial titles, those of hypatos
and apo hypaton, retained their status. But even they begin during the
eighth century to lose ground to Imperial' titles, and the reasons for this
must surely be sought in the development sketched out above. As the
Gentlemen and Officers, pp. 136ff., with a critique of E. Stein, 'La Disparition du Senat de
Rome a lafindu VIe siecle', Bull de la Classe des Lettres de VAcad. de Belgique 23 (1937),
365-90 (repr. in Opera minora selecta (Amsterdam 1968), pp. 359-84).
61
While the situation in Italy was by no means the same, the process of the withering-away
of the older senatorial elite offers a number of important parallels: see Brown, Gentlemen
and Officers, pp. 21-37. Continuity, on the other hand, can be much more clearly
observed in the early medieval West, as shown in the analysis of A. Demandt, 'Der
spatromische Militaradel', Chiron 10 (1980), 601-36.
62
See Arnheim, Senatorial Aristocracy, esp. pp. 155-71 for the senate as a social-economic
elite in the later Roman period.
63
See Beck, Senat und Volk, pp. 31f., 42ff. and 49, and the remarks at 57f. Since the leading
state offices were usually accompanied by the rank of patrikios, apo hypaton, hypatos and so
on, all such officials, whether active or titular, will have been of senatorial status. Even in
the later period this was the assumption: see, for example, De Cer., 15.6-7; 174.11-12;
290.16sq.
 Infrastructures and hierarchies 397

senate came increasingly to represent an assembly of title-holders

appointed at the emperor's pleasure, and especially as the hereditary
element - the clarissimate - lost its social relevance during the seventh
century, so its institutional significance and its social-political attrac-
tiveness must have dwindled. Those anxious for advancement may no
longer have sought senatorial titles, of course; while the emperors them-
selves may have increasingly disregarded these titles, preferring instead to
promote their favourites or others who had been brought to their attention
by appointing them to titular membership of the palatine ordines or by
granting them the highest imperial titles, such as that of patrikios. This does
not seem to have been a particularly long drawn out process. The lesser
senatorial ranks lost ground before the middle of the eighth century; apo
hypaton disappears during the second half of the eighth century; 64 the title
of hypatos itself is widely held during the eighth century, and the tendency
to award it to an ever-widening range of officials and others might again
reflect the process of devaluation in operation. 65 The devaluation of the
senatorial titles also suggests the increasing dominance of the military
administrative element within the state, as has been suggested else-
where. 66 At any rate, the incorporation of senatorial titles into a single
hierarchy of status, whatever the actual complexities of its operation and
in spite of the survival of the different traditions, precedents and privileges
associated with them, parallels and reflects the disappearance of the late
Roman senatorial establishment and the hereditary clarissimate, and the
concomitant rise of a purely imperially sponsored 'meritocratic' elite
whose status and authority were secured, initially at least, only through
imperial patronage. The relative pluralism of the late Roman period
disappears, to be replaced by an authoritarian system of precedence and
advancement rooted in state service alone. The contradiction inherent in
the whole lies in the fact that this rather more rigid and conformist system
seems to have made social mobility no less common or difficult; it merely
channelled it through different routes of access, and demanded different
abilities and social qualifications from those aspiring to advancement.
The effects of this centralisation of access to authority, status and
precedence around the emperor and the court, in a more radical and more
pronounced way than had ever been the case in the later Roman period, is
clear by the later seventh century. The titles and dignities held by those
selected to carry out diplomatic or other imperial missions reflect this
imperial monopoly. Without exception, these persons bear titles such as
64
See Winkelmann, Rang- und Amterstruktur, pp. 32f., 34f. and 36-7.
65
Ibid., p. 41 and the list at pp. 48ff. Note that the Liber Pontiflcalis (I, 351) associates those
of consular rank (ypati) with the senate and patricii in 681, but not in 710 (I, 390).
66
Winkelmann, Rang- und Amterstruktur, p. 51.
398 Byzantium in the seventh century

spatharios, silentiarios and koubikoularios, or higher grades still, or they

occupy posts which will have merited such titles. 67 Whether these titles
signified membership of an active corps based at court, or the honorary
title alone, is in this context not important. The point is that all forms of
social and political advancement now depended on association with the
imperial court and its system of precedence. It might be objected, of course,
that this was also the case in the fourth, fifth and sixth centuries. But in
practice, there existed always a greater openness of opportunity, a greater
pluralism and less rigorously centralised system of advancement in the
earlier period. The attention and favour of the emperors had always been
crucial. But the traditional senatorial elite establishment of landed wealth
and state office had been able to maintain itself by the assimilation of new
elements into its culture and traditions, and thereby retain its position as
the social and economic elite, the 'ruling class' of the later Roman period.
Nobility of birth and ancestry was as relevant in the late Roman world as
any other qualification for membership of the senatorial order. 68 The
changed economic and political circumstances of the years after the middle
of the seventh century especially destroyed this capacity for social and
political renewal. The senate as an institution and as a social elite suffered
as it became entirely dependent upon the emperors - the court - for its
personnel and for the status thereby attached to membership of it. Over the
same period, local centres of wealth and power lost in prestige and in
relevance to Constantinople and the court, which thus came to represent
the only source of access to social advancement. Notions of lineage and
nobility of ancestry became, for a while at least, secondary issues; where
the social origins of an individual from a well-to-do background are
described in the contemporary or near-contemporary sources, then terms
such as 'well-born', 'of noble/ancient lineage', 'of high rank' or 'high-born'
seem to denote the wealth and status of a family at the time, rather than its
real or supposed lineage. And the fact that a very wide range of terms is
used, in both hagiographical and in other sources, demonstrates that no
single and uniformly accepted formula for 'noble' or 'aristocratic' lineage
was actually in common use. The use of the description 'son of on seals
and in the literary sources reflects most probably the assertion of a
new-found status rather than any long-term continuity of lineage. 69 The
67
See, for example, Haldon, Byzantine Praetorians, pp. 185ff. (spatharioi); Guilland, Recher-
ches, vol. I, pp. 2 7 7 - 9 ; Diehl, L'Exarchat de Ravenne, p. 173 note 2 (koubikoularioi); see
Laurent, Corpus, vol. II, s.t. koubikoularios, silentiarios, chartoularios, sakellarios; and
Haldon, Byzantine Praetorians, p. 188 note 415; Brown, Gentlemen and Officers,
pp. 1 3 4 - 5 .
68
See Jones, LRE, vol. II, pp. 523ff.; Arnheim, Senatorial Aristocracy, pp. 103ff.
69
See the useful list of Yannopoulos, La Societe profane, pp. 14ff. In general, see the articles of
R. Guilland, 'La Noblesse de race a Byzance', BS 9 (1948), 3 0 7 - 1 4 (repr. in Recherches,
 Infrastructures and hierarchies 399

popular idea of noble ancestry persisted, of course, and appears frequently

in hagiographical literature, for example, as a means of expressing
approval of the character of the individual in question, especially in the
period up to the middle of the seventh century and from the later eighth
century onwards. A very similar development has been noted for Italy and
especially the exarchate in the same period.70 But it was only through
association with the state and its apparatuses, or the Church, that social
advancement and preferment could be assured, the more so since the state
now intervened more directly than ever before in the affairs of the
provinces. And it is worth noting that it is in the period from the 640s to
the middle of the eighth century that this association predominates, after
which - as concepts of nobility and lineage begin to become relevant once
more to the state and begin to be applied with more precision and care -
both associations coexist.71

While it would be wrong to take either of the two developments outlined

above as alone responsible for the change in the character of late Roman
society which we have observed, together they do add up to a very real
transformation of the ways in which that society perceived itself (as seen,
for example, through the canons of the Quinisext or through the Ecloga)
and the ways in which it functioned. Under the all-pervasive influence of
Christian ideas, reinforced by imperial legislation and stimulated by the
radicalising effects on social practice of the social and economic changes of
the seventh century, the late Roman or early Byzantine idea of the family
and its practical relationship in the world underwent a major revision.
Society was perceived as an agglomeration of individual primary family
units occupying a definite space with regard to one another and to God,
rather than as a corporate body of individuals united within, first, the

vol. I, pp. 15-22); 'La Noblesse byzantine a la haute epoque', 'E\\j)vixa, Trapdprruxa 4:
IJpocrcpopd els XrikTruiva IT. Kvpiaxibr]v (Thessaloniki 1953), pp. 2 5 5 - 6 6 (repr. in
Recherches, vol. I, pp. 2 3 - 3 1 ) . For the epithet 'son of in its various forms, see Nesbitt,
'Double n a m e s o n early Byzantine lead seals', 109ff.
70
E.g. Vita Theod. Stud., 1 1 6 B (ev 7evei . . . Xa^-rrpcov); Vita Blasii Amor., 659E (TWV
eTrioT|(jLa)v); Vita Evaristi, 3 0 0 . 2 0 - 1 (eTrupaveoTctTOi x a i Trepi8o£oi); Vita Nicephori
Medic, 4 0 5 , c a p . 5.9 ftevovs TrepipXeTTTOv); Vita Euthymii lun., 1 6 . 2 1 (einraTpi5ai);
Vita loannis Psichaitae, 1 8 . 2 2 (eirupaveis); Vita Alypii styl., 1 6 1 . 1 1 (irepicpavEis); a n d so
on. For the exarchate of R a v e n n a , see Brown, Gentlemen and Officers, pp. 166ff.
71
For example, Anastasius of Sinai, Quaestiones, 504A; Vita loannis Eleemosyn., 77.18;
Miracula S. Artemii, 14.22; Vita Greg. Agrigent., 633B; Vita Antonii lun., 1 9 7 . 2 8
(dpxovTes); Miracula S. Artemii, 11.21-2; 44.22 (ev$o£oi TO\3 TraXaTiov); and the list at
Yannopoulos, La Societe profane, p. 15 notes 32-50. For the reassertion of these concepts
in the ninth century, see E. Patlagean, 'Les Debuts d'une aristocratie byzantine et le
temoignage de rhistoriographie: systeme des noms et liens de parente aux IXe-Xe siecles',
in The Byzantine Aristocracy, IX to XIU centuries (BAR, Int. Ser., 2 2 1 , Oxford 1984),
pp. 23-43. See also Winkelmann, Quellenstudien, pp. 143ff.
400 Byzantium in the seventh century

municipality, and second, the wider polity. Personal loyalty to God and to
one's family eventually came to be expressed as priorities at the very least
equivalent to, when not actually superior to, loyalty to the state or to a
notion of 'society*. The development is clearly expressed at a later date - in
the vernacular tradition of the akritic cycle, for example, or in the eleventh-
century Strategikon of Kekaumenos72 - but its origins lie in the trans-
formations of the later sixth and seventh centuries.
Individuals were now perceived of as the responsible - culpable -
elements in God's universe and in society, in a way which had not been the
case before. The new emphasis, as has been pointed out, is most apparent
in the way in which the body and the individual become the centre of
attention in the penal law of the later seventh century and the eighth
century, especially and most explicitly in the Ecloga.73 Indeed, by a series
of metaphors drawn from the Christian and the Judaic traditions, a
body-symbolism was evolved which, while it had come into being already
in the Roman period, was first formally elaborated in the Ecloga. Nasal
mutilation was thus associated with sexual offences, while the putting out
of eyes was related to crimes of sacrilege and, by extension, to treason. The
two were, of course, associated, for power and sexuality, and the meta-
phors by which these were represented in symbolic discourse, are univer-
sally related.74
Emphasis on the body as the individual, and on public and dramatically
obvious physical punishment as retribution, as evidence of the nature of
the crime and as a demonstration of imperial authority, is only one side of
the coin, however. For at the same time, the linking of notions of corporal
mutilation with both Christian morality and imperial philanthropy and the
very concept of the individual's direct responsibility to God for his or her
sins 75 flew in the face of the traditional Roman ideal of the family headed
and guided by the just paterfamilias. It was a set of attitudes which could
have flourished only at the cost of both the concept and the structure of the
traditional family, however these may have been shaped; and, as we have
seen, one of the key developments over this period was the reduction in the
authority of the head of the family and a strengthening of the moral
72
For some valuable comments on the representation of relationships of honour, shame and
patronage In these texts, see P. Magdallno, 'Honour among Romaioi: the framework of
social values In the world of Dlgenes Akrltes and Kekaumenos', BMGS 13 (1989)
1 8 3 - 2 1 8 . On the developments outlined here, see Haldon, The miracles of Artemios and
contemporary attitudes'.
73
See E. Patlagean, Byzance et le blason pinal du corps, 405ff. (cited note 24, chapter 10,
above).
74
SeeiWd.,407.
75
These elements are all fundamental to the narrative in the miracles of Artemius and
Therapon, and many others; and they are echoed in many other texts with increasing
emphasis from the seventh century on. See, for example, Anastasius of Sinai, Quaestlones,
55 (617A-620B, esp. 617&-C).
 Infrastructures and hierarchies 401

personality and therefore individual responsibility of the separate members

of the family.
The traditional Roman notion of the state as an enlarged or 'super
familia' was given a Christian guise and fitted without difficulty into this
new framework. The emperor was still the paterfamilias of his family, but
now with the clearly delineated, but less absolute, authority of the Chris-
tian head of the household as set out in the canons of the Quinisext or in
the provisions of the Ecloga.76 The subjects of the emperors remained their
'children'.77 But whereas late Roman society had provided a series of
intermediate levels of social and community organisation, within each of
which the individual could find an identity, the society which had evolved
by the later seventh and early eighth centuries was a society of individuals
deemed directly responsible for their actions before God. The Church
proclaimed (and legislated for) the elementary nuclear family; and the
collapse of the older order under the strain imposed by the change in its
circumstances brought on by the events of the seventh century could only
have promoted its rapid evolution and consolidation. The individual, and
the concomitant open and competitive social individualism, the career-
building on a necessarily 'meritocratic' basis in response to the urgent
needs of the state and its apparatus, were the result. The symbolism
embodied in mutilation, the firm establishment of the nuclear family unit,
and the foregrounding of the relationship between individual and God as
both a private and public one, are thus all closely interconnected, albeit in
ways which may not at first be apparent.
None of this is to say, of course, that the nuclear or elementary family
was itself a novelty or that loyalty to one's kin and family was something
new in late Roman times. As we have said, the former had been a feature of
Mediterranean culture for centuries, although its extent and its importance
varied from region to region and from culture to culture. What I am
suggesting here is that the imposition of uniformity and consensus from
above - by state and by Church - and the striving for a common identity,
along with both social and political survival which was a feature of daily
existence in the lands which remained 'Roman', took its toll on the
traditions and the practices necessary at a structural level for the social
reproduction of late Roman culture, a culture which still remained, even in
the later sixth century, relatively open, both in its attitude to the outside
world and in its attitudes to physical space and to institutions. Certainly,
signs of considerable change were already apparent from the end of the
reign of Justinian and during the reigns of his immediate successors. But
the physical shrinking of the Roman world promoted the changes already

76
On the patria potestas, see the literature in note 8, chapter 10, above.
77
See the evidence assembled in Hunger, Prooimion, passim.
402 Byzantium in the seventh century

implicit in the structure of late Roman social organisation and culture, and
also emphasised the isolation of the East Roman world and the imperial
Church. It is hardly surprising that the motif of the Chosen People found a
particular echo in the surviving literature of the time. 78
78
The concomitant representation of the emperor as a new David illustrates and emphasises
the attitudes of the times: see Cameron, Images of authority', 21-2; see also S. MacCor-
mack, in B 52 (1982), 287-309, see 295ff.; Dvornik, Early Christian and Byzantine
Political Philosophy, vol. II, pp. 797 and 823.
 CHAPTER I I

Forms of representation: language,

literature and the icon

The literature and the art of the period we are concerned with represents
one aspect of the social and cultural whole of the late antique and early
medieval east Mediterranean civilisation which is the object of our
enquiry. In itself it is, of course, a vastfieldwhich has been studied under a
number of subcategories, and I shall not attempt a descriptive account of
them all here.
My concern is rather with the ways in which these forms of represen-
tation functioned during the course of the seventh century in a way
which has rendered them peculiarly difficult of access to later commen-
tators. They were vehicles for the self-representation of the culture
which produced them. They symbolised and transmitted, for different
elements of the population, according to their means of access to them,
fragments of an ideological system and a cultural universe. And at the
same time, they constituted the shape of that cultural and symbolic
universe, which we can observe both through these forms and through
other sources.
These are normal and fundamental functions of artistic and literary
production. But in the east Mediterranean world of the later sixth century
and after, symbols that were culturally available evoked and were focused
around specific aspects of that symbolic universe in a way which contrasts
with that of the preceding years, a focusing which is part and parcel of the
attempts at re-evaluation and reaffirmation discussed already in chapter 9.
This is not to suggest that this process, at least in its main elements, was in
any way one which was consciously undertaken or executed. In one of its
most obvious - and fundamental - features, it clearly was not: by the later
sixth century, the linguistic Hellenisation of the state and its administrative
apparatuses, and all that depended upon or was associated with the state
(the senatorial elite or the Church, for example), was well advanced, a
process that was effectively completed during the years of the early Arab

403
404 Byzantium in the seventh century

conquests.1 The old world of Latin West and Greek East, sharing a unified
political culture, was a thing of the past, in spite of the continued existence
of Byzantine possessions in North Africa until the late seventh century and
in Italy until after the middle of the eighth century. 2
Latin left its mark, of course, most clearly in legal terminology and
literature, where it remained the basis of the technical jurisprudential
vocabulary; and also in the field of mechanics and especially military
equipment and organisation, as well as in numerous words and terms for
items of everyday use, both in domestic and agricultural terms. But the
language of state and diplomacy was Greek, and already by the middle of
the century letters and other documents to and from Constantinople to the
West were accompanied by translations. 3
By the end of the reign of Phocas, the Slav occupation of much of the
Balkans, and the existence of the Avar dominion, had severed any regular
and direct land link between the Greek-speaking southern Balkans and the
Latin-speaking provinces around the Danube. Most importantly, the lan-

1
For general surveys, see H. Zilliacus, Zum Kampf der Weltsprachen im ostromischen Reich
(Helsingfors 1935); L. Hahn, 'Zum Sprachenkampf im romischen Reich bis auf die Zeit
Justinians', Philologus, Suppl. 10 (1907), 675-798, and 'Zum Gebrauch der lateinischen
Sprache in Konstantinopel', in Festgabe fur Martin von Schanz (Wurzburg 1912),
pp. 173-83; note also the review by F. Dolger of Zilliacus, Zum Kampf der Weltsprachen, in
BZ 36 (1936), 108-17. More recently, H. Mihaescu has studied the relationship between
Greek, Latin and Daco-Roman or Slav cultural-linguistic zones in the late Roman and early
Byzantine world. See especially the summary article, with more up-to-date literature, in
'Die Lage der zwei Weltsprachen (Griechisch und Latein) im byzantinischen Reich des 7.
Jahrhunderts als Merkmal einer Zeitwende', in Studien zum 7. Jahrhundert, pp. 95-100. On
the background to the changes of this period, see G. Dagron, 'Aux origines de la civilisation
byzantine: langue de culture et langue d'etat', RH 241 (1969), 23-56; see esp. 36ff.
2
For North Africa, see Cameron, 'Byzantine Africa', and for Italy, Brown, Gentlemen and
Officers, pp. 65ff. and esp. H. Steinacker, 'Die romische Kirche und die griechischen Sprach-
kenntnisse des Fruhmittelalters', Mitteilungen des Institutsfiir osterreichische Geschichtsforsch-
ung 62 (1954), 28-66. Southern Italy and Sicily had since the sixth century B.C. had
notable Greek elements, of course, and this seems to have been strengthened during the
seventh century A.D., at least in the far south and in Sicily, by refugees from the
Peloponnese. See A. Guillou, 'Italie meridionale byzantine ou Byzantins en Italie meridi-
onale', B 44 (1974), 152-90 (repr. in Culture et societe en Italie byzantine XV (London
1978); C. Mango, 'La Culture grecque et l'Occident au VIIIe siecle', in Settimane di Studio del
Centro Italiano di Studi sulV alto Medioevo XX, 2 (Spoleto 1973), pp. 683-719; P. Charanis,
'On the question of the Hellenization of Sicily and southern Italy in the Middle Ages',
American Historical Review 52 (1946/7), 74-87.
3
Note the comments of Pope Gregory I {Gregorii 1 Papae Registrum Epistolarum, eds. P. Ewald
and L.M. Hartmann, in MGH, Epp. 1 and 2 VII, 27 (2nd edn, Berlin 1957), p. 474: 'hodie in
Constantinopolitana civitate, qui de latino in graeco dictata bene transferant non sunt.' For
Latin in military language, see Mihaescu, Die Lage der zwei Weltsprachen, 99 and note 24,
and 'Les Termes de commandement latins dans le Strategicon de Maurice', Revue de
Linguistique 14 (1969), 261-72. In the field of law, Latin clearly had a considerable
influence; but already by the later sixth century the greater part of imperial legislation
promulgated from Constantinople was in Greek. See L. Wenger, Die Quellen des romischen
Rechts (Vienna 1953), esp. p. 660.
 Language, literature and the icon 405

guage of the orthodox Church in the East was always firmly Greek, and it
had indeed been predominantly Greek-speaking thinkers and theologians
who had been pre-eminent in the religious controversies and who had set
the pace in theological debate in the centuries since the peace of the
Church. It is perhaps symptomatic that the great leaders of the religious
opposition to monotheletism in North Africa and Italy were - with the
exception of Pope Martin - Greek-speakers and that the key debates in the
controversy were carried out in Greek, even in North Africa - the famous
debate between Maximus and Pyrrhus, for example, in Carthage in the
year 645. 4 It is this accelerating process of linguistic Hellenisation which
provides the backdrop to the developments described in what follows.
Two phenomena in particular deserve our attention; and although
neither is limited exclusively to the seventh century, it is then that the
context described in the foregoing chapters of this book gives them a
particular weight and significance. It should be made clear from the outset
that these elements played themselves an equal part in painting the picture
of the seventh century that we can construct from our sources, they were
integral aspects of a continuing dialectical process. The two phenomena in
question are: the increased centrality in both ideological and artistic-
representational terms of icons and the decline and near disappearance of
secular literature. They are not linked directly, but they are part of the
same pattern of cultural changes which I have attempted to describe. And
they constitute a crucial pointer to the character of seventh-century
culture and belief.

ICONS, EVOCATIONS AND EXPECTATIONS

The increasing importance of relics and icons, as both channels of access to
their archetypes and as sources of miracles and divinely inspired power in their
own right has been discussed and analysed in detail by several scholars.5
All have noted that it is particularly from the later sixth century - the
immediately post-Justinianic period - that the sources begin to mention the
existence, the use of and the power of icons more and more frequently. As
we have seen in chapter 9, this was associated causally with an attempt
4
See PG XCI, 287-354. See Van Dieten, Geschichte der Patriarchen, pp. 571T., and Guarrigues,
Maxime le confesseur, pp. 5If. Of course, North Africa remains in some respects a special
case, upon which a great deal of work remains to be done; but the example is illustrative of
the developments described.
5
See esp. Kitzinger, 'Images before iconoclasm', 95-128; A. Grabar, L'Iconoclasme byzantin
(Paris 1957), pp. 21ff.; Brown, 'A Dark-age crisis'; Cameron, 'Images of authority', 18ff.;
Haldon, 'Ideology and social change', 16Iff.; G. Dagron, 'Le Culte des images dans le
monde byzantin', in Histoire vecu du peuple chretien, ed. T. Delumeau, vol. I (Toulouse
1979), pp. 133-60 (repr. in La Romanite chretien en Orient. Heritages et mutations XI
(London 1984)); Herrin, Formation of Christendom, pp. 3O7ff.
406 Byzantium in the seventh century

within the culture of the East Mediterranean in general to reaffirm the

narratives of social practice by seeking a more immediate mode of access to
God and the heavenly realm, and by the imperial power in particular to
reassert its paramount position by stressing the divine origins of its auth-
ority. But as first relics, and then icons became part and parcel of daily life, so the
context within which they originally became so important became irrelevant,
and their centrality to the experience of the everyday unquestioned. It was, of
course, this very situation and the narratives that derived from it which
made the imperial iconoclasm of the eighth century possible.6 But it was
also the all-present icon - whether static or portative, large or small, public
or private - which lends Byzantine Christianity from the seventh century
on one of its most prominent and enduring characteristics. It was icons
which focused people's attention on local saints, and on local traditions
and observances, beliefs and practices. Icons reminded the congregations
of churches of their history, and with it the traditions of Christianity and -
if distantly - of empire and imperial rule, of identity with a wider Christian
community and so on. Icons worked didactically and performed functions
similar to those of the holy man or hermit familiar from the late Roman
period.7 There was one crucial difference, of course: icons outnumbered
holy men by far. Icons became a commonplace in a way that the holy man
- at least in the normal run of events - never could. Icons were available to
all - whether in monumental form in churches or in the humbler context
of the household. Pope Gregory the Great justified their use in a much-
quoted passage from a letter to the bishop of Marseille: 'For what writing
presents to readers, the picture presents to the unlearned, who behold; for
in it even those who are ignorant see what they should follow; in it, the
unlettered read. Thus is a picture, especially to the barbarians, instead of
reading.'8 Thus, icons were attributed with a functional ideological aspect,
they could be seen as the literature of the illiterate in early Byzantine
society; and although this argument for their role - while it was revived
and elaborated during iconoclasm - had become a less significant justi-
fication by the later sixth century, as we shall see, pictorial representation
was clearly regarded by the later seventh century as an important alter-

6
See my comments in 'Some remarks', esp. 176ff.
7
See the comments of R. Cormack, Writing in Gold (London 1985), pp. 75ff.; and H.-G. Beck,
'Von der Fragwurdigkeit der Ikone', SBB 7 (1975), pp. 1-44; and G. Lange, Bild und Wort.
Die katechetischen Funktionen der Bilder in der griechischen Theologie des sechsten bis neunten
Jahrhunderts (Wurzburg 1969); J. Gouillard, 'Contemplation et imagerie sacree dans le
christianisme byzantin', Annuaire de la Ve section de I'Ecole Pratique des Hautes Etudes 86
(1977-8), 29-50, see esp. 37ff. (repr. in La Vie religieuse a Byzance II (London 1981)). Note
also the useful discussion of L. Ryden, 'The role of the icon in Byzantine piety1, in Religious
Symbols and their Functions, ed. A. Biezais (Uppsala 1979), pp. 41-52.
s PL LXXVII, 1128C.
 Language, literature and the icon 407
native element to texts in the elaboration of Christian dogma - Anastasius
of Sinai, for example, argues strongly in its favour.9
The seventh century marks an important stage in the evolution of the
icon, and in particular of the modes of representation which came to be
associated with it. It has been suggested that of the modes available to
artists in the later sixth century, the dominant tradition (which Kitzinger
terms 'Hellenistic', but which others prefer to call Illusionist')10 seems
increasingly at this time to give way to an alternative tradition, referred to
as the 'abstract' style. The latter is characterised by a linear, two-dimen-
sional mode of representation, with passive, motionless figures, contrast-
ing with the naturalistic, three-dimensional representation of the 'Hellenis-
tic' or Illusionist' style.11 Both systems continue to exist side by side
throughout the period with which we are concerned and through the
iconoclastic era; and their employment was by no means even and regular
in all areas of the empire at the same time.12 A clear-cut distinction
between separate examples of both styles is sometimes difficult to detect,
indeed, since many works contain elements from both. For example,
formal portraits of Justinian and Theodora in the church of San Vitale in
Ravenna are set in a clearly three-dimensional physical context within a
building, perhaps a part of the palace. Yet the figures themselves, repre-
sented as approaching the observer, are portrayed in a way that suggests
9
See Kitzinger, Images before iconoclasm', 136ff. The argument was given new strength in
the context of iconoclasm by John of Damascus. See Contra imaginum calumniatores
orationes tres, in B. Kotter, ed., Die Schriften des Johannes von Damaskos III (Berlin 1975), I,
17. But the 'educational' element was never forgotten: the canons of the Quinisext include
an ordinance ordering the destruction of corrupting or misleading pictures. See Mansi XI,
98 5D; in his Rodegos, Anastasius of Sinai recommends pictorial representations for
didactic purposes to refute heresy. See especially Anna D. Kartsonis, Anastasis: The Making
of an Image (Princeton 1986), especially pp. 40-67, for a detailed analysis.
10
E. Kitzinger, 'Byzantine art in the period between Justinian and iconoclasm', in Berichte
zum XL Internationalen Byzantinisten-Kongress IV 1 (Munich 1958), pp. 1-50. For the use
of 'illusionist', see R. Cormack, 'The arts during the age of iconoclasm', in Iconoclasm, eds.
Bryer and Herrin, pp. 35-44, see pp. 41f.; D.H. Wright, 'The shape of the seventh century
in Byzantine art', in First Annual Byzantine Studies Conference, Abstracts of Papers (Cleve-
land, Ohio 1975), pp. 9-28; and esp. J. Onians, 'Abstraction and imagination in late
Antiquity', Art History 3 (1980), 1-23. 'Illusionist' is also used, along with 'Hellenistic',
by D. Talbot Rice, The Appreciation of Byzantine Art (Oxford 1972); and G. Mathew,
Byzantine Aesthetics (London 1963).
11
The 'abstract' style had evolved gradually from a variety of elements, as Kitzinger has
suggested, 'Byzantine art in the period between Justinian and iconoclasm', 16ff., esp. 28, a
function of both traditional Hellenistic and oriental modes of figural representation. The
roots of the Hellenistic style lie, as the name suggests, in the development of classical
Hellenistic styles of the period from the second century B.C. to the second century A.D. The
shift applies to all figural and representational art, of course, whether three- or two-
dimensional.
12
See the valuable and detailed survey by Kitzinger, Byzantine Art in the Making,
pp. 99-126; and 'Byzantine art in the period between Justinian and iconoclasm', 30ff.;
and Cormack, 'The arts during the age of iconoclasm,' 42-3.
Plate 11.1 The Empress Theodora and attendants, San Vitale, Ravenna
 Language, literature and the icon 409

to a modern western eye a lack of motion, and great gravity. Dynamism is

present in the composition as a whole, but it is not located in the actual
figures. (See plate 11.1) This combination of modes does not mean an
artistic eclecticism, and certainly not an unconscious application of such
elements by the artists who produced the works - icons, frescoes, mosaics -
in question. Rather, and as Kitzinger has suggested, the artists applied dif-
ferent modes with the deliberate intention of thus signifying or suggesting
different levels of symbolic reference.13 So, in the case of the Ravenna
mosaics, the imperial figures can be understood to gain in stature, both
physically and symbolically, from their mode of portrayal. In icons of
saints and holy figures, the dominant trend is to represent in abstract,
linear form, the key figures, and this does seem in part to be a reflection of
the way in which the icons were meant to be conceived of and employed.
To illustrate this point we must, of course, refer to the compositions
themselves. Perhaps the clearest contrast between the two traditions in a
single context comes from the church of St Demetrius in Thessaloniki,
where the 'Hellenistic' or Illusionist' style of the later sixth and early
seventh centuries is represented by a number of mosaics of the period
before the fire of the 620s which led to its refurbishing and redecoration
shortly thereafter;14 and where the 'abstract' style of the phase of rebuild-
ing, dated to about the middle of the century, can be compared directly.
(See plates 11.2(a) and (b).) Elsewhere, and within a single composition,
the use of these two modes, along with other elements to emphasise
different degrees of symbolic evocation and the different roles of the figures
portrayed with regard to the beholder can be found in some of the
well-known icons in the monastery of St Catherine on Mt. Sinai. The icon
of the Virgin and child, for example, who are flanked on both sides by two
saints, and with two angels behind, provides a good example of the use of
space, movement (or lack of it) and frontality within a single composition.
The saints look sternly and protectively out at the onlooker, the angels are
portrayed in a quite different, illusionist and naturalistic style, gazing up at
the hand of God and heaven. (See plate 11.3)
These shifts in 'style' and in the choice of representational technique is
significant. It has been suggested on the one hand that the popularity of
icons was in part stimulated in the sixth century by their illusionist style
13
Kitzinger, 'Byzantine art in the period between Justinian and iconoclasm', 47-8, although
he sees a more explicit evocation of levels of spiritual being. See also J. Trilling, 'Late
antique and sub-antique, or the "decline of form" reconsidered', DO? 41 (1987), 469-76.
14
For the date of the fire and the mosaics, see Lemerle, 'La Composition et la chronologie des
deux premiers livres des Miracula S. Demetrii', BZ 46 (1953), 349-61, see 356; Kitzinger,
'Byzantine art in the period between Justinian and iconoclasm', 26 and notes 100 and
101; and see Cormack, Writing in Gold, p. 83, and 'The mosaic decoration of St Demetrios,
Thessaloniki: a re-examination in the light of the drawings of W.S. George', Annual of the
British School at Athens 64 (1969), 17-52.
 H

Plate 11.2(a) Late sixth- to seventh-century mosaic decoration from St Demetrius, Thessaloniki (water-colour W.S. George)
 Language, literature and the icon 411

Plate 11.2(b) St Demetrius with patrons and benefactors. Mosaic of post-

reconstruction period, second quarter of seventh century
412 Byzantium in the seventh century

and their lifelike and almost humanly accessible quality. This may have
been the case; although it assumes without argument a questionable
essentialism of perception - that what we might today feel to be 'more
approachable' held the same symbolic and emotive values for a medieval
observer. I will return to this problem shortly. For it is clear, on the other
hand, that accessibility and receptiveness could equally well be expressed
in the 'abstract' form. It is less the style itself which promotes or inhibits
accessibility than the context in which the figures are represented, and in
which 'style' and 'form' are attributed with specific meaning, and
symbolic-evocational functions. As we will see, it is important to remember
that different cultures also perceive differently - from each other as well as
from us. And it seems a reasonable surmise that it was the linear, hieratic
and abstract mode of representation which came to meet best the demands
of the producer and the beholder. It seems to have responded to and
reflected the attitudes of the late sixth and seventh centuries, which sought
an affirmation of the heavenly, and therefore unsullied, unassailable,
status and authority of the prototypes which the icon encapsulated.
Imperial parallels, on coins of the later seventh century, for example,
would appear to bear this out, given the need to express and reinforce
imperial authority at this time. We may reasonably conclude that auth-
ority and status were represented iconographically by an abstract, hieratic
style, more effectively than by the alternatives. 15
Both in theological argument (elaborated during the sixth and seventh
centuries) 16 and in the common perception, icons came to be understood
as being transcendentally related to their prototypes; they were sources,
therefore, of the holy and the sacred. Icons seem to have responded, both in
their physical dissemination and their easy availability, on the one hand,
and in the style in which they were executed, on the other hand, to the
need for a more immediate mode of access to God or his representatives. At
first, the Illusionist' or 'Hellenistic' mode dominated. But I would argue
that the adoption of an abstract mode - a style in which, as we have said,
15
For a detailed description and analysis of this icon, see K. Weitzmann, The Monastery of St.
Catherine at Mt. Sinai, the Icons, I: From the sixth to the tenth century (Princeton 197£),
pp. 19-21. For the notion that icons were rendered more approachable through the us£ of
an illusionist mode of representation, see J. Herrin, 'Women and the faith in icons in early
Christianity', in Culture, Ideology, Politics: Essays for Eric Hobsbawm, eds. R. Samuel and
G. Stedman-Jones (London 1982), pp. 56-83; and for imperial models of solemnity,
authority and power, and the ways in which these were thought to be best represented,
see the detailed fourth-century description of the Emperor Constantius by Julian: Oratio I
(in Julian, Letters and Works, ed. and trans. W.C. Wright (London and Canmbridge, Mass.
1954), p. 96); and see note 19, chapter 11 below
16
Kitzinger, Images before iconoclasm', 139-46. Note in particular qu. 39 of the Ps.-
Athanasius Quaestiones ad Antiochum ducem {PG XXVIII, 597-708) datable possibly to the
630s, which is a brief justification of the use of holy images, and in which their miraculous
attributes are taken quite for granted - see 621A-D.
 Language, literature and the icon 413

Plate 11.3 Icon of the Virgin and child (detail) between St Theodore and St George,
with flanking angels, monastery of St Catherine, Sinai
414 Byzantium in the seventh century

Plate 11.4(a) Solidus of Justinian I (obverse and reverse)

Plate 11.4(b) Solidus of Constans II with Constantine, his son (obverse and reverse)

the key figures were represented as passive and motionless, attentive and
ready to be approached by the supplicant - must have been also a
functional response to a qualitative change in the nature of this demand;
more particularly, to the need for both accessibility, differentiation from
the earthly world, and undeniable spiritual authority - something which
the naturalistic portraits of the Hellenistic tradition were perhaps less able
to communicate. For the Illusionist' mode, at least when individual or
small groups of figures were concerned, was less approachable, if only
because it depended upon a dynamism and narrative involvement within
the composition itself, which necessarily restricted the observer to that role
alone, excluding him or her from actually approaching and being received
by the object of devotion. 17
The process by which painters of icons and mosaicists arrived at this
particular mode of expression is a more difficult problem of course, and one
which I cannot do justice to here. What is clear is that, by the later sixth
17
See E. Kitzinger, 'On some icons of the seventh century', in Late Classical and Medieval
Studies in Honor of Albert Mathias Friend, jr. (Princeton 1955), pp. 132ff. For the lack of
personal accessibility and involvement inherent in the three-dimensional, dynamic art of
the classical period, see T.J. Boatswain, Images of uncertainty: some thoughts on the
meaning of form in the art of late Antiquity', BMGS 12 (1988). Note also the comments of
J.-M. Spieser, 'Image et culture: de l'iconoclasme a la renaissance macedonienne', in
G. Siebert, ed., Methodologie iconographique (Strasbourg 1981), 96f. and esp. Onians (see
note 10 chapter 11 above).
 Language, literature and the icon 415

century, Constantinople had become the chief centre of artistic inspiration

in the eastern Mediterranean basin, exercising an influence over artistic
production in both Italy, for example, and in the Eastern provinces. The
building activities of the reign of Justinian alone seem to have contributed
a great deal to the spread of a broadly Constantinopolitan style and
provided a further unifying element in the developing art of late antique
Christianity. The common symbolic and representational elements in
mosaics, frescoes, sculpture and metalwork, for example, which had devel-
oped from the fourth century, are apparent in the Christian art of the sixth
century. The shared artistic heritage is expressed in both the production of
clay ampullae from the holy places, for example, brought back by pilgrims,
and in the commissioning by the wealthy of silverware in the style of
Constantinople. As the sixth century drew on, so this Constantinopolitan
influence became more and more important - not to the exclusion of
purely provincial styles, but nevertheless as a leader in style and form. The
presence of the court was, of course, the crucial factor in this. 18
It is not impossible that a preference for a more abstract style for
devotional and authoritative figures, such as the Virgin and the saints, for
example, was given added stimulus by the adoption of such a style for
imperial figures - it has been argued that, as a general rule (although
there are exceptions) the imperial coinage up to Justin II exemplifies a
broadly Hellenistic or illusionist style for the figures of the emperor, but
that thereafter a more distinctly linear, two-dimensional style dominates.
Religious icons clearly did owe a great deal to the portrayals of secular
figures of authority such as emperors, as has been pointed out by several
art historians. And although there are specific exceptions to this general
tendency, as we have noted - notably the coinage of Constantine IV, for
example, which seems to reflect a 'Justinianic revival' based on the styles of
the mid-sixth century 19 - there is no doubt that it was in the end just this
18
Kitzinger, 'Byzantine art in the period between Justinian and iconoclasm', 33ff., esp. 39,
and Byzantine Art in the Making, esp. pp. 113ff.; J. Hubert, J. Porcher and W.F. Volbach,
Europe in the Dark Ages (London 1969), pp. 245f.; K. Weitzmann,' "Loca sancta" and the
representational art of Palestine', DOP 28 (1974), 31-55; A.P. Kazdan and A. Cutler,
'Continuity and discontinuity in Byzantine history', B 52 (1982), 429-78.
19
For the connection between imperial and religious iconography, see K. Weitzmann, with
M. Chatzidakis, S. Radojcic and K. Miatev, Icons from South Eastern Europe and Sinai
(London 1968), p.x; Kitzinger, 'Images before iconoclasm', 12Iff.; 'Byzantine art in the
period between Justinian and iconoclasm', 20; and esp. Graber, L'Empereur, in particular
pp. 189ff., with pp. 196-243. For abstraction and linearism on the coin-portraits of
emperors in the period after Justinian and up to Constantine IV, see esp. M. Restle, Kunst
und byzantinische Miinzpragung von Justinian I. bis zum Bilderstreit (Athens 1964), esp.
pp. 65ff.; and Kitzinger, Byzantine Art in the Making, pp. 102ff., and 'Byzantine art in the
period between Justinian and iconoclasm', 120 (with fig. 20); P.D. Whitting, Byzantine
Coins (London 1973), pp. 106-60, esp. p. 150. For the 'Justinianic' revival under Con-
stantine IV, see Kitzinger, Byzantine Art in the Making, pp. 12 If.
416 Byzantium in the seventh century

Plate 11.5 (a) Solidus of Justin II (obverse and reverse)

Plate 11.5(b) Solidus of Maurice (obverse)

Plate 11.5(c) Solidus of Phocas (obverse)

Plate 11.5(d) Solidus of Heraclius and Heraclius Constantine (obverse)

 Language, literature and the icon 417

Plate 11.5(e) Solidus of Constantine IV (obverse and reverse)

Plate 11.5(f) Solidus of Justinian II (obverse)

Plate 11.5(g) Solidus of Anastasius II (obverse)

Plate 11.5(h) Solidus of Philippicus Bardanes (obverse and reverse)

418 Byzantium in the seventh century

abstract and hieratic style, even if tempered by many illusionist elements,

which was adopted by the artists of the post-iconoclastic era for devotional
portraits. 20 (See plates 11.5(a)-(h).)
Of course, there are difficulties in assessing the aesthetic significance of a
particular style or mode of execution of a work of art for contemporary
beholders, both in respect of the possible variety of different situation-
bound perceptions (social class and status, for example, may play a role in
the evocational process which a specific image sets in train) and in terms of
the function of a representational or figural aesthetic as understood by
contemporaries. The effect a particular iconographic style has on the
modern interpreter, and the language we might use to describe those
effects, might be quite inappropriate when applied to the culture which
produced the artifact. And we are confronted here, of course, with the old
problem of whether or not we can posit some universal aesthetic principle;
or at least the possibility of long-term cultural-aesthetic continuities within
a specific cultural-ideological continuum, in terms of both geography and
space. In this particular case, the assumption of a common Greco-Roman
and Christian 'heritage' plays a significant role.
There are a number of arguments here, of course, not the least of which is
the practical and functional difference in respect of the production of works
of art: modern, individualist art-production (and consumption) operates in
a very different context from that of the medieval religious art, whatever
the individual elements contributed by the actual artist or craftsman. The
question of intentionality and the ambiguity of aesthetic codes - and,
therefore, the number of potential interpretations - in modern cultures
makes the notion that meaning is immanent, rather than constructed in
the process of consumption, dubious. There may be a (culturally or
individually) 'preferred' understanding on the part of the artist or onlooker,
but this cannot be exclusive.21 In medieval religious art, on the other hand,
painters or craftsmen clearly worked within a specific and sometimes very
tightly controlled framework or set of instructions, both as far as the
demands of the patron of the work and the inherited modes of represen-
tation available were concerned. Individual interpretation was not thereby
excluded, but it was very limited. And within this limited interpretational
field, it is clear that certain meanings or readings were both intended and
20
See, for example, K. Weitzmann, 'The study of Byzantine book illumination, past and
present', in The Place of Book Illumination in Byzantine Art, ed. G. Vikan (Princeton 1975),
p. 160, see 3-4.
21
For a general survey of the concept of 'art' in modern culture, see Janet Wolff, Aesthetics
and the Sociology of Art (London 1983), and The Social Production of Art (London 1981);
and for the question of aesthetic value-attribution, R. Ingarden, 'Artistic and aesthetic
values', in H. Osborne, ed., Aesthetics (Oxford 1972), pp. 39-54. See also R. Wollheim, Art
and its Objects (Harmondsworth 1968).
 Language, literature and the icon 419

understood. The crucial problem for the modern interpreter is to grasp the
modalities of style which represent shifts in the interpretational patterns,
first of all; and then - much more difficult - to relate these changes in style
to the functional demands placed upon artistic representation (whether
those demands were explicit or - as is more usual - implicit). This assumes,
of course, that shifts in modes of representation also tie in with, affect and
reflect, shifts in the perceived or the required significance of a represen-
tation. The real problem is to see to what emotional and psychological
needs a change in aesthetic actually responds, or may itself stimulate. And
it is at just this point that the question of universals must be addressed.
Now, it seems to be a reasonable assumption that certain elements of our
aesthetic responses are universal - if only because all human creative and
artistic activity is carried on by the same biological being. Just as there
appear to be phylogenetically determined capacities to generate linguistic
structures or to employ different modes of cognition - symbolic, semantic
and so on - so the probability of some phylogenetically determined faculty
to respond to certain types of form - size, depth and so on - and colour,
which structures the possibilities of production and interpretation, seems
unavoidable. Sentiments or moments in human experience which remain
universal and constant - such as love, hate, fear, death, birth, ageing and
so on - whatever the cultural variation in their mode of expression and
representation, are common to all. And, if - using socially inflected
conventions - an artist, or a whole style of representation, is able to evoke
such basic elements of experience, this serves at least as part of the
explanation for the ability of some works to transcend cultural barriers. 22
The problem remains to recognise exactly what each different culture
understands by a specific style and how its effects were given expression.
If we accept the possibility of both a deep structure of aesthetic responses
and a given cultural-aesthetic continuum, then we can justifiably apply
some of our criteria to artifacts not directly of our own culture, but of
cultures which have a specific and determinate historical affinity with ours.
If we reject these premises, of course, then we must seek some direct
statement from the culture itself on the aesthetic qualities concerned, and
the role they were held to play. 23
In the case of the orthodox, Byzantine world we are fortunate to have
some comments from members of the culture itself; and so, while I would
certainly argue for both a historical-cultural affinity as well as a generative

22
See in particular the work of Sebastian Timpanaro, On Materialism (London 1975); and
Raymond Williams, 'Problems of materialism', New Left Review 109 (May-June, 1978),
3-17, see 10.
23
See in this respect R. Wollheim, 'Aesthetics, anthropology and style', in M. Greenhalgh
and V. Megaw, eds., Art in Society (Ithaca and London 1978), esp. pp. 5ff.
420 Byzantium in the seventh century

and phylogenetically determined aesthetic faculty, this need not be the only
basis for any assumptions we might make. Such statements confirm the
validity of our interpretation of the different modes of representation; in
particular, the difference between, and different aesthetic-psychological
functions of, the Illusionist' and the 'abstract' styles described.24 Later
Byzantine writers stress particularly the centrality of 'order' and
motionlessness, both in invoking the accessibility of the figures portrayed
and in emphasising their other-worldliness and authority, themes which
have been discussed and stressed by other scholars. 25 Of course, such
conscious formulations were also subject to long-term historical evolution
and represent possibly only a certain stage in the development of Byzantine
understanding and explanations of their figurative art. But it is at least
clear that such art was always open to interpretation and explanation and
that there existed at different times differently nuanced but generally avail-
able sets of notions about the function of images. 26 In the sixth century,
Nilus Scolasticus remarked that the image of an angel immediately evoked
the heavenly sphere; Agathias comments that the image imprints itself
within the beholder.27 What these, and many other comments, suggest is
the evocational, which is to say the symbolic, power of images.
Some scholars have argued that there is,no (conscious) symbolism in
Byzantine art, and that, on the contrary, it was intended to be explicit and
realistic. But this seems to miss the point at issue. The problem revolves not
around whether or not Byzantine commentators or theologians imbued
religious art with a conscious symbolism, but rather whether the notion of
'reality' in figural art was broadened or redefined and whether it became
more nuanced, according to the nature and explicit function of the work in
question. 28 In other words, and as I have argued above, the shifts in mode,
and the choice of a particular mode, may well represent explicit decisions
and intentions on the part of the craftsman or the patron; but this in itself
reflects the availability to the culture of a choice in mode of representation
at a tacit or subconscious level, that is, at a level at which emotional and
evocational stimuli operate, and as a function of iconographic intention. 29
24
See especially Henry Maguire, 'Truth and convention in Byzantine descriptions of works
of art', DOP 2 8 (1974), 1 1 3 - 4 0 ; and the essays of Byckov, Popova and Komec (chapters
15, 16 a n d 17) in Udal'cova, ed., KuVtura Vizantii.
25
For example, Kazdan a n d Constable, People and Power in Byzantium, pp. 104f.
26
See Mathews, Byzantine Aesthetics, for example, whose survey of the literary sources
makes this a b u n d a n t l y clear, whether one accepts his conclusions or not.
27
Anthologia Palatina I, 33 and 34.
28
Mango, Byzantium: the Empire of New Rome, p. 2 6 4 , for example, argues against any
symbolism. But see H. Maguire, Earth and Ocean: The Terrestrial World in Early Byzantine
Art (London, Pa. 1987), pp. 5 - 1 5 , 8 1 - 4 .
29
See, for example, the comments of Th. Adorno, Aesthetic Theory, eds. G. Adorno and
R. Tiedemann, Engl. trans. C. Lenhardt (London and New York 1984), esp. pp. 8 - 9 .
 Language, literature and the icon 421

Representation changes, therefore, in response to a need to fulfil the

functional requirements placed upon it.
What we are talking about, of course, is symbolism as a mechanism of
evocation and metaphor. Whatever theologians and Churchmen may have
claimed, all works of art are endowed with a latent or potential symbolic
value, whether this is understood, made explicit, desired or not. For this is
the way in which the human cognitive system functions.30 The symbolic
value in question may or may not be part of a conscious theological or
other intellectual system, of course - symbols operate at a variety of levels,
from the most personal and individual experience to the most general social
level. The crucial point is that a particular mode of representation will have
on each beholder both a personal specific and a general social, evocational
effect. A gradual shift in the particular mode, therefore, or the use of a
combination of modes in a particular way, whatever the intentions of the
artist might be, cannot but evoke a shift in response of some sort from the
onlooker. Since systems of figural representation do not develop by chance
or in an anarchic manner - in a social vacuum, as it were - the gradual
dominance of a particular style or combination of styles, in a particular
context, must also reflect in some way the expectations of the onlooker. It
operates at both an explicit and a symbolic level.31
Symbolic systems, and the narratives with which they are bound up,
depend, of course, on the experience of individuals and groups within the
culture for their effectiveness and the specificity of the evocations they
generate. Whether Byzantines endowed their icons with explicit symbolic
values is, therefore, not such an important question. 32 What is important is
to understand that a figural representation inevitably 'evoked' in the mind
of the onlooker, just as it also portrayed in a more or less explicit manner, a
particular individual or event or whatever; and that changes in style reflect
30
See esp. R. Bhaskar, 'Emergence, explanation, emancipation', in P.F. Secord, ed., Explain-
ing Human Behaviour: Consciousness, Human Action and Social Structure (Beverly Hills, Ca.
1982), pp. 2 7 5 - 3 1 0 ; T.W. Goff, Marx and Mead, Contribution to a Sociology of Knowledge
(London 1980), pp. 86ff. See Haldon, Ideology and social change', 150ff.
31
The best demonstration of how symbolic systems operate, and also the best challenge in
this context to a semiological approach to symbolism, is D. Sperber, Rethinking Symbolism
(Cambridge 1975). On the dialectic between perception and representation, see the
comments of V. Burgin, 'Diderot, Barthes, Vertigo1, in The End of Art Theory. Criticism and
Post-Modernity (London 1986), pp. 1 1 2 - 3 9 ; a n d 'Seeing sense', ibid., pp. 5 1 - 7 0 .
32
In fact, of course, Byzantine C h u r c h m e n did endow t h e m with express symbolic functions:
c a n o n 8 2 of the Quinisext makes this quite clear w h e n it orders the representation of
Christ in h u m a n form, r a t h e r t h a n in t h a t of a lamb, in order to proclaim a n d emphasise
the incarnation of the word in Christ a n d to evoke the doctrine of salvation. See Mansi XI,
9 7 7 E - 9 8 0 B . This is, precisely, a conscious symbolism, in the sense used here. The fact
t h a t the medieval Church a b a n d o n e d the totemic symbolism of lamb a n d fish, for
example, h a s nothing to do with a relinquishing of symbolic functions inevitably bound
up with t h e production a n d use of visual representations such as icons or frescoes in
churches.
422 Byzantium in the seventh century

deeper social-psychological trends. Such trends must, therefore, be one

element in the wider set of social and cultural transformations which late
Roman and early Byzantine society underwent at this time.
The modal changes which I have described seem to fit in also with (and
presumably are causally linked with) a change in the perception of the
effects of an icon or similar figural portrayal, which can be shown to have
occurred over the period from the later sixth century, approximately, to the
ninth century. The central element in this change involves the transfer of
emotional weight from the representation itself to the onlooker. This
impression is gained from examining the composition and form of figural
representations, the Hellenistic or illusionist mode depending upon an
inwardly directed and narrativistic involvement within the frame of the
composition; the abstract mode invoking attentiveness, accessibility, the
direct involvement of the onlooker with the main subject of the com-
position, and, potentially, the intervention of the portrayed figure into the
world of the onlooker. In the former, the figures inhabit their own world; in
the latter, they look out and touch the world of the onlooker.
Later Byzantine writers lay greater stress on the physical pain endured in
scenes of martyrdom, for example, than do earlier writers, and they
emphasise the emotion and pathos expected from the onlooker.33 Tears
and grief - greatly in contrast to the earlier period - were assumed, if not
demanded, from a true Christian. The ninth-century account of Ignatius
the Deacon, in the Life of patriarch Tarasius, of a martyrdom cycle,
assumes just this response; 34 the council of 787 clearly accepted as
fundamental the principle that the sanctity of an icon was guaranteed if it
moved the onlooker to tears; 35 and, perhaps most significantly - since it
clearly relates to a general attitude to, and set of assumptions about,
representative art, whether two- or three-dimensional - the anonymous
writers of the eighth-century Brief Historical Notes (Parastaseis syntomoi
chronikai) are clearly interested in the effects antique monuments and
statuary have on the onlooker, paying little or no attention to their actual

33
See, for example, C. Mango, The Art of the Byzantine Empire. Sources and Documents
(Englewood Cliffs, NJ. 1972), p. 38 (description of the martyrdom of St Euphemia, late
fourth century); and esp. H. Maguire, Art and Eloquence in Byzantium (Princeton 1981),
pp. 34ff., and 'Truth and Convention' (cited, note 2 4 , chapter 1 1 , above), 22f. The
development has been discussed at length, for the ninth century, by Leslie Brubaker,
'Byzantine art in the ninth century: theory, practice and culture', BMGS 13 (1989),
2 3 - 9 4 and 'Perception and conception: art, theory and culture in ninth-century Byzan-
tium', Word and Image 5 (1989), 1 9 - 3 2 . The importance of tears as a purifying and
cleansing element is made apparent in the writings of Anastasius of Sinai. In qu. 105 (PG
LXXXIX, 757C4-76OA8) and at several points in his Oratio de Sacra Synaxi (PG LXXXIX,
8 3 2 D - 8 3 3 A , 837A) their importance for true repentance and understanding is made
very clear.
34 35
See Vita Tarasii, 4 1 4 . 1 8 s q . Mansi XIII, 9, 11 and 32.
 Language, literature and the icon 423

appearance. 36 For the twelfth-century Nikolaos Mesarites, describing the

Church of the Holy Apostles in Constantinople, the figures represented for
him and for fellow onlookers the reality of their prototypes; they could be
smelled and seen to move, they evoked an instant emotional response. 37
These descriptions of the effects a work of art had on an onlooker tell us
little or nothing about the mode of representation of course. And one must
be careful not to assume that the response expected from an onlooker was
constant from the fourth or fifth centuries through to the twelfth: this was
clearly not the case. But that a change in the perception of effects did take
place, and began at about the same time as the increasing importance of an
abstract figural mode becomes apparent, and that this change was well
developed by the middle of the eighth century, does seem significant.
The change is implicit in the enormous increase in the use of icons,
particularly the intensely personal and private nature of their use, which
can be dated from the later sixth century. The argument has returned to its
starting-point. 38 The change is quite clear from the expectations which
people clearly came to have of icons and the powers which were ascribed to
them. Images intervened directly in the day-to-day lives of ordinary people.
According to a (possibly interpolated) passage in the Spiritual Meadow of
John Moschus, an image of Jesus appeared before a crowd at Antioch
dressed in clothes previously given to a beggar, thus acting out a story from
the New Testament. 39 The pilgrim Arculf, who visited the Holy Land in the
670s, relates a number of tales he heard about an image of St George at
Lydda - it made promises to those who prayed before it, was immune to
attack, worked miracles; 40 other icons bled when attacked or even fought
back at their opponents. The famous image of Christ at Edessa was later
credited with beginning a fire which destroyed the siege-ramps of the
Persians in 544. In the Life of Theodore of Sykeon, images are able directly
to effect cures, while in the miracles of Artemius, a collection of which was
put together in Constantinople in the 660s, wax taken from a holy icon is
an effective cure. 41 Images gave aid in battles and sieges; they provided
assistance in other contexts as well; and they were clearly venerated and
36
See edition by Cameron and Herrin, introduction, p. 53.
37
G. Downey, ed., 'Nicholas Mesarites, Description of the Church of the Holy Apostles at
Constantinople', in Trans. American Philos. Society 47 (1957), see ch. 26, 6
38
See the texts for this period collected in Mango, The Art of the Byzantine Empire, pp. 133-9;
and Kitzinger, 'Images before iconoclasm', 95ff.
39
See Th. Nissen, 'Unbekannte Erzahlungen aus dem Pratum Spirituale', BZ 38 (1938),
351ff., see 367 no. 12.
40
Arculf, Relatio de Locis Sanctis III, 4 (ed. T. Tobler, Itinera et Descriptiones Terrae Sanctae I
(Geneva 1877), 195ff.)
41
See the long list of such wonders, all dating to the later sixth and seventh centuries, given
by Kitzinger, 'Images before iconoclasm', 101-9; and excerpts in Mango, The Art of the
Byzantine Empire, pp. 133-9; and see Miracula S. Artemii, 16-17; Vita Theod. Syk., 8.5-10.
424 Byzantium in the seventh century

respected, both in public and in private. Supplicants placed candles before

them, lamps were lit and maintained before them, curtains or hangings
were placed before or around them, to the extent that the latter, too, gained
a certain power through association; and incense was burned in their
presence.42 In short, there takes place a marked intensification of interest
in the role and effectivity attributed to icons by the sources of the later sixth
and seventh centuries, in comparison with the preceding period. Again,
this development parallels the changes in the modes of representation
which were employed for images at the same period. The most likely
explanation for this shift in mode is, quite simply, therefore, that it repre-
sented in ways which can no longer be entirely accessible to us the
evocational and interventionist power of images, conjuring up through
the frontal, unavoidable gaze of the figure or figures represented in the
image an immediacy, authority and, above all, an externality which an
illusionist mode - a mode which, as we have seen, evoked rather an
internalised narrative - could not attain so effectively. It did not exclude or
inhibit the use of the illusionist mode, of course, both in images and in
other contexts; and we should take into account the fact that frontality and
directness of gaze might be represented in a more naturalistic or Hellenistic
mode, and still achieve the desired result. But the abstract mode does seem
to indicate what we can associate most clearly both visually (aesthetically)
and through the textual evidence of the period in question with the
intensification and elaboration of the use of icons, and the enormous
emotional power vested in them.43 By the end of the seventh century, the
evidence suggests that icons had begun to be an absolutely unremarkable
feature of Byzantine piety and of daily life, both public and private. As we
have seen, canon 82 of the Quinisext specifically ordained the represen-
tation of Christ in human form, rather than as a lamb, for didactic
purposes - to make clear the incarnation of the word in Christ, to affirm the
dogma of Chalcedonian Christology; canon 100 ordained that art which
might corrupt or confuse be destroyed, again stressing the key role it was
seen to play both as an educational and as a spiritual mediator. Such texts
confirm the suggestion that there existed a general tendency in this period
to enlist images formally in the service of the Church - more particularly,
the imperial Church.44 The icon was, in several ways, a true 'sign of the
times'.

42
Kitzinger, 'Images before iconoclasm', 96-100; and see V. Nunn, 'The encheirion as
adjunct to the icon in the middle Byzantine period*, BMGS 10 (1986), 73-102.
43
For a similar point, although from a different standpoint, J. Onians, 'Abstraction and
imagination in late Antiquity', Art History 3 (1980), 1-23.
44
Mansi XI, 977E-980B and 985D; see Kitzinger, Images before Iconoclasm', 120f.; Anna
Kartsonis, Anastasis: The Making of an Image (Princeton 1986), p. 59.
 Language, literature and the icon 425

LITERATURE, PIETY AND THE END OF ANTIQUITY

The second of the two phenomena noted above, namely the more or less
complete disappearance after the first quarter of the seventh century of a
secular literature, has invoked much comment and discussion. The obvious
explanation - that the conditions of that century, and indeed of much of the
following century, did not provide an environment at all conducive to such
activity - is clearly correct. But how does this very general explanation
relate to the specific developments in question? After the late-sixth- or
early-seventh-century works of Theophylact Simocatta, for example, or the
anonymous Paschal Chronicle, or that of John of Antioch, there is a lacuna
of almost two centuries until the next surviving historical work. Similarly,
the seventh and much of the eighth century provide no examples of
geographical, philosophical or philological literature; ithere is no historical
poetry after George of Pisidia, and only a trickle of legal literature and
secular rhetoric. Apart from supposed lost histories of a certain Trajan
patrikios of the seventh century and of the so-called megas chronographos, or
'great chronographer', of the eighth century, the letters of a few Church-
men and the powerful, the very small amount of legal literature mentioned
already and the surviving documents of state and Church as institutions,
the literary output of the seventh century appears to have been almost
entirely theological in nature, or at the least concerned with matters of
dogma, devotion, various aspects of liturgical practice, problems of day-to-
day piety and observance, and so on. 45
Now the fact that theological and related literary genres maintain a
written existence, whereas the forms of secular literary activity suffer a
dramatic decline, is clearly a reflection of the difficulties of the seventh
century and must in the first instance be a reflection of a change in the
45
The relevant data with detailed surveys can be found in Hunger, Profane Literatur, vols. I
and II. See, too, the chapter of A.A. Averincev on late Roman literature in Udal'cova, ed.,
Kul'tura Vizantii, who remarks also on the break in genre and continuity which sets in
with the work of George of Pisidia and John Moschus. For Trajan patrikios see Hunger,
Profane Literatur, vol. I, pp. 337 and 345; D.Serruys, 'Recherches sur l'epitome', BZ 16
(1907), 1-52; A. Markopoulos, 'A la recherche des textes perdus. L'historiographie
byzantine de la haute 6poque jusqu'au VIIIe siecle. Etude preliminaire', in From Late
Antiquity to Early Byzantium, ed. V. Vavfinek (Prague 1985), pp. 203-7. The cultural
break which the seventh century represents affected secular letter-writing also. See
M. Mullett, 'The classical tradition in the Byzantine letter', in Byzantium and the Classical
Tradition, pp. 75-93, see p. 86. Note also the survey of E. Jeffreys, 'The image of the Arabs
in Byzantine literature', Seventeenth International Byzantine Congress, Major papers (New
York 1986), pp. 305-23, see pp. 312-14. For a good survey of the nature of late Roman
historical writing and the problem of the break in the tradition, see L.M. Whitby, 'Greek
historical writing after Procopius: variety and vitality', in Averil Cameron, and L. Conrad,
eds., The Early Medieval Near East: Problems in the Literacy Source Material (Princeton 1990);
also J.D.C. Frendo, 'History and Panegyric in the Age of Heraclius: the Literary Background
to the Compositon of the Histories of Theophylact Simocatta', DOP 42 (1988), 143-56.
426 Byzantium in the seventh century
conditions within which the literary culture associated with late Roman
society had existed andflourished.Two elements in particular seem to be
relevant: the decline during the later sixth, but especially during the
seventh, century of traditional municipal society and culture; and the
qualitative change in the nature and constitution of the ruling elite. The
two are intimately bound up together, as I have tried to show in chapters 3
and 4 above. And it is clear that the demise of the urban civilisation within
which late Roman literary culture hadflourishedhad a dramatic effect. A
real shift took place in the relationships between social class and literacy,
and more especially between the traditional sources of late Roman culture
and education, and the developing imperial elite of the court and the state
bureaucracy. The latter - for which there is evidencefromthe middle of the
seventh century - while it will certainly have incorporated established
families and representatives of the old senatorial elite, was recruited more
directly through imperial service and received its rewards through a more
centralised and autocratic system of precedence and hierarchy. It was
drawn from a variety of ethnic and cultural sources, and selection was
based more than previously, it would appear, upon practical administra-
tive, political and military ability, along with the patronage and influence
of those already In place'. It was this new, meritocratic elite which began to
dominate both provincial and central administrative apparatuses, along
with the sources of wealth and the means of controlling its distribution. The
older, leisured cultural elite quite simply lost its social prestige and its
economic standing. There is no reason to assume, of course, that the new
elite did not acknowledge the value of the old culture and its traditions, nor
indeed that it did not itself aspire to participate in that culture - so much is
clear from the later revival of interest in 'classical' learning and models of
literacy, and the enormous power of the notion of 'the past' in Byzantine
thinking. But the dominant tendency does seem to have been at least
apathetic in our period with regard to these traditions.
The lack of literature in the fields mentioned above - which cannot
possibly be a reflection of some supposed (and mysteriously selective)
failure of the literary output of the seventh and early eighth centuries to
survive - confirms the hypothesis; the fact that theological literature does
survive, and in considerable volume, makes it certain. For the Church
maintained its traditional administrative organisation and, to a great
extent, within the empire, its sources of revenue. It needed to be able to
educate its clergy, and it needed literate and cultured men for its highest
offices.46 But even here, it was topical questions of the day, the study of the
writings of the Church fathers and of the general councils, along with
46
Beck, Kirche, pp. 67-74 and 79ff.
 Language, literature and the icon 427
scripture and exegesis that provided the mainfieldsof concern. Interest in
the pre-Constantinian, much less the pre-Christian, culture of the past was
a rarity.47
The nature of hagiographical writing from the eighth century underlines
the character of the shift. Not only does the hagiography of the later eighth
century and after deal with a very different type of saint or holy man from
that familiar from the period up to the Arab conquests (as we might expect,
given the nature of the social and cultural changes already discussed); it
represents a different style of hagiographical narrative, in which the
connection with the forms of late antique literary types has been severed.
The seventh century did not by any means see the end of hagiography, but
it did witness a major qualitative change in its composition and presen-
tation.48 Just as significantly, the ethos which underlay the collection and
preservation through recopying of classical, patristic and historical texts in
the ninth and tenth centuries especially reinforces the nature of the break.
For the fear of losing forever the literature of the past (meaning, in effect, of
the period up to the mid-seventh century) seems to be a central motif.49
By the same token, it is illustrative that the traditional secular system of
education seems to have disappeared almost entirely - with one or two
individual exceptions - and even in Constantinople was reduced to
insignificance. Whatever the value of the remark of the ninth-century
chroniclers that after 726 learning in the Roman world was extinguished,
their comments certainly reflect a general situation.50 Only an elementary
form of primary schooling seems to have carried on, if we exclude the
private tuition that must have been available to the wealthier; and it is
interesting that the regular use of the Psalter as the basic reading primer in
both Constantinople and the provinces probably begins in the years about
the middle of the seventh century.51 As we have seen, there does seem to
47
Beck, Kirche, esp. pp. 430-73 on the period of the monenergite and monothelete debates,
for a representative survey.
48
See the discussion of L. Ryden, 'New forms of hagiography: saints and heroes', Seventeenth
International Byzantine Congress, Major Papers, pp. 537-54; and note W. Lackner, 'Die
Gestalt des Heiligen in der byzantinischen Hagiographie des 9. und 10. Jahrhunderts',
ibid., pp. 523-36.
49
As has been emphasised by H. Hunger, 'The reconstruction and perception of the past in
literature', Seventeenth International Byzantine Congress, Major Papers, pp. 507-22, esp.
519.
50
Theophanes, 405.10sq., and especially the story recorded in Georg. Monachi Chronicon,
742.2-22.
51
See Mango, Byzantium: The Empire of New Rome, pp. 136f., for a brief survey of the
evidence; and for a more detailed picture, Anne Moffatt, 'Schooling in the iconoclast
centuries', in Iconoclasm, eds. Bryer and Herrin, pp. 85-92. See also Lemerle, Le premier
humanisme byzantin, pp. 98-103. See, for example, Vita Andreae Cretensis, ch. 3, where
Andreas receives an education in ta peza grammata, and later in ta hypselotera mathemata.
That a traditional education of sorts continued to be available, to some, however, albeit
subordinated to the demands of theology and 'orthodoxy' (whether from a Chalcedonian
428 Byzantium in the seventh century

have been a limited legal training available at Constantinople in the last

years of the seventh century, as canon 71 of the Quinisext suggests. But
there is no evidence for any other form of higher education. The use of the
Psalter reinforces the impression gained from the types of literary activity
that did flourish at this time that the Church and clergy came to play a
much more central role in basic and more advanced education than
hitherto, as well as in the transmission of literacy and a literary culture in
general, especially in respect of what was read. In effect, a cycle of
developments was set in train that could have had only one result: a
decline, even if temporary, of interest in, and transmission of, the secular
literature of the sixth century and before, contrasting with the emphasis
that fell upon the reading and production of theological literature. The
more pivotal position of the clergy in the maintenance of literacy and the
selection of reading matter was to have important consequences for a later
period. And it marks also a decisive break with the literary and cultural
pluralism of the late antique world.
What I am suggesting, therefore, is quite straightforward. The old
culture had been maintained within a specific ideological and social-
economic framework, that of the municipal economies and society of the
late Roman world and the senatorial elite which dominated central and
provincial culture and politics. During the seventh century this economic
and ideological framework was shattered. In its place there developed
gradually from the second half of the seventh century a new, and quite
different, social and political elite drawn from a wider variety of social and
cultural backgrounds, and founded within a centralised imperial estab-
lishment which came to be the sole effective source of wealth, power and
privilege. This new elite had very much less interest in and commitment to
the traditional literary culture and its values, concerning itself more
immediately with cementing its own power and position by and through
the imperial administrative apparatus which it served. Needless to say, this
tendency represents the sum of a host of individual aspirations and careers
- a conscious set of discourses and narratives which provided a common
political ideology for this group had hardly begun to develop at this stage.
Slowly, as this new elite became entrenched and as the structures of the
medieval Byzantine state began to emerge from the difficulties of the
seventh and eighth centuries, it began once more to cultivate its interest in
a supposedly classical past that was, as several historians have shown,
more late Roman than pre-Christian. Both the very limited degree of
literacy in the Byzantine world (in contrast with popular views on the
or a monophysite point of view) is clear from the case of the Armenian Ananias of Sirak,
whose 'biography' illustrates the point well. See J.-P. Mane, 'Quadrivium et cursus
d'etudes au VIIe siecle en Armenie et dans le monde Byzantin', TM 10 (1987), 159-206.
 Language, literature and the icon 429

subject), and the methods and subject-matter of Byzantine literati in the

ninth century and after - in other words, the very nature of the break in the
secular literary culture at this time - testify clearly to this aspect, a point
borne out by the motivations underlying the so-called renaissance of the
later ninth and tenth centuries. 52
These two developments, in the use of icons on the one hand and in the
nature of literary activity on the other, can now be considered together. For
in reflecting on the forms of supposedly 'popular' literary activity -
hagiography and miracle-collections especially - and the expansion of
icon-use over the same period, the real nature of the changes in late Roman
and early Byzantine culture can be most clearly seen. The crucial import-
ance for the inhabitants of this threatened world of belonging, and being
seen to belong, to the Chosen People, of being a meaningful part of God's
plan, is the message constantly reiterated in the public and private culture
of the second half of the seventh century: the icon, whether of Christ, the
Virgin, a saint or an archangel, reminds of and invokes a Christian
orthodox and local tradition. 53 The church sermon, and more especially
the hagiography, the collection of apophthegmata, or sayings of the Church
Fathers, the accounts of the doings of holy men and monks, or the
collection of miracles, read out in a church or in some other community
gathering (and it must not be forgotten that a parallel oral tradition which
has not always left clear traces had as great an influence), fulfilled the same
function. The very process of teaching to read and write were now almost
entirely under the guidance of the clergy. And these forms were, in a sense,
classless. They were not directed at any specific social group, but at all the
faithful; and the sources make it clear that their effectiveness was indeed
universal. 54 The popularity of hagiographical and miracle literature, of the
52
See esp. R.D. Scott, 'The classical tradition in Byzantine historiography', in Byzantium and
the Classical Tradition, pp. 6 1 - 7 4 ; a n d C. Mango, 'Discontinuity with the classical past in
Byzantium', ibid., pp. 4 8 - 5 7 , see 52ff.; see also the literature in notes 4 8 a n d 4 9 ; and note
the discussion of P. Lemerle, 'Eleves et professeurs: Constantinople a u X e siecle', CRAl
(1976), 576-87.
53
Dagron, Le Quite des images dans le monde byzantin, esp. pp. 138ff. and 152ff. with
literature; Ryden, The role of the icon (cited note 7, chapter 11 above), esp. 45ff. and 5 1 .
The importance of icons to the ordinary household, as well as their vulnerability to the
agents of the devil, c a n be seen in the story of the w o m a n w h o unwittingly goes to a
magician or false holy m a n for help, recounted in the Life of Andrew the Fool (see PG CXI,
7 7 6 C - 7 8 4 A ) . The magician is able to drive out the grace of God t h r o u g h magical spells
uttered before a lighted oil lamp, w h i c h results eventually in the icon being defiled and
covered in excrement. Only t h r o u g h the 'white magic' of the holy fool are the icons
purified. Compare these stories with the context discussed by G. Vikan, 'Art, medicine and
magic in early Byzantium', DOP 38 (1984), 6 5 - 8 6 ; and C. Belting-Ihm, art. 'Heiligenbild',
in Reallexikon-fur Antike und Christentum, fasc. 1 0 5 (Stuttgart 1987), 6 6 - 9 6 , especially
8 9 - 9 5 o n icons as media of healing a n d miraculous intervention.
54
See Beck's r e m a r k in Das byzantinische Jahrtausend, pp. 1 4 7 - 5 2 a n d Kirche, pp. 2 6 7 - 7 5
(on hagiography); also H. Delehaye, Les Recueils antiques des miracles des saints (Brussels
430 Byzantium in the seventh century
erotapokriseis, or collections of 'questions and answers', selected from the
sayings of famous hermits, holy men and monks, the icons in churches and
in homes, all expressed a common cultural focus and a unity of belief, a
common fund of narrative accounts about the nature of the world.55 More
importantly, perhaps, it demonstrated the quest for common motifs,
common interests, shared explanations of the world and what was
happening, a quest which reached its goal first in local solidarities and
identities, but which evoked and symbolised also a wider Christian polity
and ideological community. The cultural pluralism and relative openness
of literary forms, which still typified the period before the Persian and
especially the Arab invasions (and in spite of the increasingly successful
screening out of classical motifs which became a feature of post-Justinianic
culture), was past. Religious writings which had been tolerated, but which
had come to be regarded by the later seventh century as dangerous or

1925) (originally in AB 43 (1925), 5-85 and 303-25); note esp. E. Patlagean, 'Ancienne
hagiographie byzantine et histoire sociale', Annales ESC 1 (1968), 106-26 (repr. in
Structure sociale V) who stresses precisely this point in relation to hagiography, collections
of miracles, and so forth. See also Patiagean, 'Discours ecrit, discours parle. Niveaux de
culture a Byzance aux VIIIe-IXe siecles', Annales ESC 2 (1969), 264-78 (repr. in Structure
sociale VI).
55
See Beck, Kirche, pp.430ff., esp. 437 and 444; G. Dagron, 'Le Saint, le savant, l'astrolo-
gue. Etude de themes hagiographiques a travers quelques recueils de "Questions et
reponses" des Ve-VIIe siecles', in Hagiographie, cultures et socie'te (IVe-VUe siecles). Etudes
Augustiniennes (Paris 1981), pp. 143-55 (repr. in La Romanite chretienne IV); A. Garzya,
'Visages de l'hellenisme dans le monde byzantin (IVe-XIIe siecles)', B 55 (1985), 463-82.
Cf. H. Domes, art. 'Erotapokriseis', in Reallexikon fur Antike und Christentum 6,
pp. 347-70. See also G. Bardy, 'La Litterature patristique des Quaestiones et Responsiones
sur l'Ecriture sainte', Revue Biblique 41 (1932), 210-36, 341-69 and 515-37; ibid., 42
(1933), 328-52. Two collections of 'Questions and Answers' in particular deserve our
attention: the Quaestiones ad Antiochum ducem of Ps.-Athanasius (in PG XXVIII, 597-708)
and those attributed to Anastasius of Sinai (in PG LXXXIX, 312-824; see CPG III, 7746).
The former collection, compiled in the eighth or ninth century, clearly draws on the latter,
itself a product of the last years of the seventh century. The Ps.-Athanasius incorporates
what may be an earlier text, compiled at about the time of the forced baptisms of Jews
(632-4) undertaken by Heraclius, as the presence of a long argument clearly designed to
persuade Jews of the justice of the Christian case might suggest (PG XXVIII, 684C-700C;
see CPG IE, 7795). See in particular the literature cited at CPG IE, 7746; and J.F. Haldon,
'The works of Anastasius of Sinai: a key source for the history of seventh-century east
Mediterranean society and belief, in Averil Cameron and L. Conrad, eds., The Early
Medieval Near East: Problems in the Literary Source Material (Princeton 1990).
Just as important a source, both for the attitudes and beliefs of ordinary people, as well
as for the cultural and social history of the period, are the collections of narrationes or
stories about the doings and sayings of holy men and monks of the time. Perhaps the best
known are those contained in the Spiritual Meadow of John Moschus, a collection of the
early seventh century, including both earlier and contemporary eye-witness reports (CPG
HI, 7367 for editions and literature). But those attributed to the monk or abbot Anasta-
sius, who may be Anastasius of Sinai, are equally important for the seventh century,
shedding light in particular on the Christian populations of the areas conquered by Islam.
See CPG III, 7758; and Haldon, 'The works of Anastasius of Sinai'.
 Language, literature and the icon 431

subversive, were banned. Canon 63 of the Quinisext council, for example,

ordered the destruction of the apocryphal Acts of the Apostles.56
The disappearance of secular literature and the concentration of both
rural and especially urban culture (for it is from the latter context that the
greater part of the evidence comes) on these devotional forms of a
consciously collective identity, an awareness made all the more powerful
through the emphasis on the individual's relationship and responsibilities
within the wider society to the Almighty, marks a qualitative cultural
transformation. Late antique civilisation had given way to that of the early
medieval world.
The focusing on a (broadly defined) theological and religious literary
production marks also a change in the centre of attention of literate
Byzantine society. For the most popular forms - in particular the hagiogra-
phical genre and the 'questions and answers' - are at the same time
representatives of differing and sometimes contradictory positions, deeply
concerned with the immediate, the 'here and now', everyday world and the
place of the individual within that world. The hagiographical tradition and
the collection of miracles associated with relics, icons and saints stresses
direct divine intervention in the affairs of humankind. But an alternative
tradition, represented, for example, by some late seventh-century 'ques-
tions and answers' among those attributed to Anastasius of Sinai, stresses
the role of physiological and natural elements in the history and environ-
ment of human society. It avoids the determinism of ancient astrology,
relating the natural world and the human (social) world to God's plan. It
thus attempts to reconcile both the contemporary Christian world view
with an older, secular (and pre-Christian) tradition.
This alternative can be traced back into the literature of the sixth
century, for example. John Lydus, John Philoponus and Procopius, in their
very different works, present versions of both Aristotelian material causal
explanation (for earthquakes, for example) and of a purely scriptural
explanation, related to the sins of mankind and the will of God. But the
latter tradition is clearly the more widespread, especially among the
illiterate mass of the population (or so we may, I think justifiably, surmise),
and dominates, certainly by the later seventh century. 57
56
For c a n o n 6 3 of the Quinisext, see Mansi XI, 972D. For comments o n the literary
production a n d the variety of genres of the sixth century see Cameron, I m a g e s of
authority', 4; a n d esp. Averil a n d Alan Cameron, 'Christianity a n d tradition in the
historiography of the late empire', Classical Quarterly 14 (1964), 3 1 6 - 2 8 ; Averil
Cameron, 'The scepticism of Procopius', Historia 15 (1966), 4 6 6 - 8 2 ; S.S. Averincev, in
Udal'cova, ed., Kul'tura Vizantii (ch.6) a n d Garzya, Visages de Vhellenisme. Both of the last
two detect a qualitative shift in the writings of Moschus a n d George of Pisidia, as noted
above, chapter 11 note 4 5 .
57
The n a t u r e of the debate h a s been summarised a n d analysed by Dagron, 'Le Saint, le
savant, l'astrologue' (cited note 5 5 , chapter 1 1 , above), esp. 1 4 4 - 6 . For the sixth century,
432 Byzantium in the seventh century

The extent of the debate between these two opposing interpretational

poles, and its effect upon its audience or readership, remains unclear. But
that it took place at all is important, for it demonstrates the liveliness of
these forms of literature and consequently - since both are known from
their wide dissemination in manuscript form to have been listened to or
read by a large circle - the centrality of this sort of discussion to the
understanding, the piety and the interest and concern of ordinary people of
all social groups. Questions on the nature of earthly rulers, the satisfactory
identity of a true holy man or saint, the nature and origins of miracles, the
cause of the different states of public health between the Christians and
their barbarian and pagan neighbours, the origins, power and value of
wise men and 'magicians', talismans and incantations, and many other
topics - they all reflect the common concerns and anxieties of this genre
and of hagiography.58 Particularly to the point are questions on how to
differentiate between Satan's trials and God's just punishments, why
sinners appear often to be spared the wrath of God (a poignant query in
such a bureaucratic and socially stratified culture), whether or not the
incursions of the barbarians represent God's punishment, or why Satan has
caused more schisms and heresies among the Christians than among other
peoples.59
This material reinforces once more the impression of a society pas-
sionately interested in reassuring itself about the nature of its relationship
with the divine world, an interest that was of as much relevance to the
cultural and spiritual survival of individuals, as was that of the physical
defence of the empire to the career interests of members of the political-
administrative elite. The answer to the question of how one differentiates
between true and false holy men - an old question, of course, and not
see Cameron, Procopius, pp. 257-9. The alternatives are represented equally, although
less self-consciously, in the later Ps.-Athanasian Quaestiones ad Antiochum Ducem; and it
appears most clearly in relation to medical matters. For Christianity, which saw itself as a
healing faith in particular, inevitably had difficulty in reconciling the Hellenistic and
Roman medical tradition with its own principles and dogmas. Evidence of a rather loose
compromise is clear in both the Ps.-Athanasian and the earlier Anastasian collections. For
a survey, from the medical perspective, see V. Nutton, 'From Galen to Alexander: Aspects
of medicine and medical practice in late antiquity*, DO? 38 (1984), 1-14, see 5-9; and
especially O.Temkin, 'Byzantine medicine: tradition and empiricism', DO? 16 (1962),
9 7 - 1 1 5 (repr. in 0 . Temkin, The Double Face of Janus and Other Essays on the History of
Medicine (Baltimore 1977, pp. 202-22)); D.W. Amundsen, 'Medicine and faith in early
Christianity', Bulletin of the History of Medicine 56 (1982), 326-50. On the 'Question and
Answer' literature, see J.F. Haldon, 'The Works of Anastasius of Sinai', cited at note 55,
chapter 11 above.
58
On all these questions, see Haldon, The works of Anastasios of Sinai', esp, 129-47; idem,
'The miracles of Artemios and contemporary attitudes'; and Auzepy, 'L'evolution de l'atti-
tude face au miracle a Byzance'. For literary and archaeological evidence for these concerns,
see G. Vikan, Art, medicine and magic in early Byzantium', DO? 38 (1984), 6 5 - 8 6 .
59
Quaestiones 9 (PG LXXXIX, 409-32), 10 (432-6), 17 (484-500) and 118 (769-72).
 Language, literature and the icon 433

confined just to the seventh century - takes on a new importance in this

period, illustrating the nature of ordinary hopes and fears.60
The questions in the collection of 'questions and answers' ascribed to
Anastasius of Sinai, the popularity of apocalyptic literature, 61 the collec-
tions of miracles and hagiographies, which increasingly through the
seventh century elaborate on the theme of the relationship between God
and man, the divine realm and the earthly empire, all point to a greater
intensity in the search for security, for clear-cut and solid answers and
explanations to the questions which most reflected the concerns of day-to-
day existence in this changing and hence - in many ways - threatening
world. As we have seen, the process of change is reflected also in the mode
of figural representation on icons. The increasingly hieratic, distanced and
spiritualised figures of this period represent a world above which is orderly,
dignified and holy, presenting implicitly a stark contrast to the drama-
laden and disorderly chaos of the earthly world. And all the shifts and
tendencies I have noted so far seem to aim to defeat this disorderliness, and
lack of harmony: the developments in imperial ceremonial, in popular
literary forms and in the interests and priorities they reflect, even in
attitudes towards the legal inheritance of the Justinianic 'golden age',
people's activities seem directed towards limiting the damage, as it were,
wrought by the sins of the Chosen People on the one hand, and by the
satanic hordes who had invaded the empire, on the other. 62
The results of this brief survey can be summarised as follows. The period
from the middle decades of the seventh century demonstrates the rapid
collapse of the traditional forms of secular literary culture, a phenomenon
which accompanied the disappearance of the old educated social elite and
the conditions which maintained it. Municipal culture died with the cities.
In its place occurred a levelling out of cultural niveaux, in which a new
ruling elite of administrative officialdom, many of whom can have been
only marginally acquainted with the older literary culture, comes to share
a set of common cultural beliefs and values with the greater part of the

60
See Dagron, 'Le Saint, le savant, l'astrologue', 146ff. and chapter 9, above, pp. 364ff.
Compare the story of the magician and the icon recorded in the Life of Andrew the Fool
(note 53, chapter 11, above); Vikan, 'Art, medicine and magic', 68-70.
61
The so-called Apocalypse of pseudo-Methodius seems to have had a very wide audience;
that appended to the Life of Andrew the Fool occupied a similar place in the tradition. Both
represented and stressed the powerful (and predestined) bond between the Chosen People
and the divine plan, and placed much emphasis on the temporary nature of the set-backs
suffered by the empire in the second half of the seventh century. See the discussion in
chapter 9 above, pp. 367ff.; and Mango, 'The Life of Andrew the Fool reconsidered',
305-8.
62
See E. Kitzinger, The Art of Byzantium and the Medieval West (Bloomington and London
1976), pp. 200ff.; a n d the r e m a r k s of N. Baynes, 'The Hellenistic civilisation and East
Rome', in Byzantine Studies and Other Essays (Oxford 1960), pp. 5f.
434 Byzantium in the seventh century

rural and urban population of the empire, cultural patterns determined by

the stubborn anti-pluralism of the Church and the increasingly dominant
role played by the clergy - and by religious literature - in the maintenance
of literacy.
Of course, this is a generalised picture: individuals at court and in the
provinces still continued to be acquainted with and to foster the literature of
the past; and there can be no doubt that the average state official or
member of the imperial senate was different in his cultural horizons and
aspirations from the average peasant - even though these aspirations oper-
ated within the shared general framework just described. Further differenti-
ation between the populations of the now limited number of real urban
centres and that of the rural hinterland of the empire was inevitable.
Indeed, the population of Constantinople as a whole will have had a very
different view of things - bearing in mind the fact that the perspectives
expressed through hagiography, apocalyptic writings, collections of 'ques-
tions and answers' of miracles and so on, while shared by the whole culture
at a general level, differed in respect of local and regional outlook and inter-
ests. The maintenance of a literary tradition, with its classical pretensions,
rhetorical artifice and stylistic conservatism is not to be doubted. The theo-
logical, homiletic and hagiographical traditions testify to this, quite apart
from the surviving secular texts.63 In addition, it must be remembered that
the functional importance of style in different contexts was crucial - the
level of style, as has been shown, was associated with specific genres,64 and
this certainly ensured a crucial functional framework and stimulus for the
continued cultivation of the literary, especially the rhetorical, registers.
Exegesis demanded very different stylistic techniques from, for example,
panegyric, and these different styles were already firmly embedded in the
literary tradition, religious and secular, by the fifth century.65
63
Compare, for example, the style of the e n c o m i u m o n the miracles of the m a r t y r Therapon,
compiled in the late seventh or early eighth century (ed. L. Deubner, De Incubatione Capita
Quattuor (Leipzig 1900), 120ff.) with t h a t of the miracles of Artemius; or the classical
allusions a n d references in the homilies of Andreas of Crete - compiled in a m o r e
accessible style to reach a wider audience, but nevertheless full of traditional rhetorical
motifs. See Th. Nissen, 'Diatribe u n d consolatio in einer christlichen Predigt des achten
J a h r h u n d e r t s ' , Philologus 9 2 (1937), 1 7 7 - 9 8 a n d 3 8 2 - 5 . On the importance of rhetoric,
see esp. H. Hunger, 'The classical tradition in Byzantine literature: the importance of
rhetoric', in Byzantium and the Classical Tradition, pp. 3 5 - 4 7 ; a n d for the rhetorical link
between a r t a n d literature, Maguire, Art and Eloquence, esp. pp. 2 2 - 5 2 .
64
See I. Sevcenko, 'Levels of style in Byzantine prose', Akten des XVI. Internationalen
Byzantinisten-Kongresses, I, 1 (Vienna 1981) ( = JOB 3 1 , 1). For a valuable comparative
analysis of these stylistic trends, see M.B. C u n n i n g h a m , 'Andreas of Crete's Homilies
on Lazarus a n d Palm Sunday: a critical edition and c o m m e n t a r y ' (Unpubl. PhD., Univ. of
Birmingham, Centre for Byzantine Studies a n d Modern Greek, 1 9 8 3 ) .
65
See in particular T.E. Ameringer, The Stylistic Influence of the Second Sophistic on the
Panegyrical Sermons of St. John Chrysostom: A Study in Greek Rhetoric (Washington D.C.
1921); R.R. Ruether, Gregory of Nazianzus, Rhetor and Philosopher (Oxford 1969); Sev-
 Language, literature and the icon 435
But the elite literary culture of late antique municipal society had passed
away for ever, and with it the pluralistic and multi-faceted world which it
represented. The icon and the Psalter are the hallmarks of the later
seventh-century world, a world which, while now more compact and
tightly controlled than before, was inhabited by individuals searching for
unity and a common identity through the symbols available to them - the
orthodox faith, the icon and the emperor.
cenko, 'Additional remarks to the report on levels of style', Akten des XVI. Internationalen
Byzantinisten-Kongresses II, 1 (Vienna 1982), 220-38 ( = JOB 32, 1), with examples.
 CHAPTER 1 2

Conclusion: the transformation of a culture

PIETY AND SECURITY! THE POLES OF AUTHORITY

The history of East Roman or Byzantine culture in the later sixth and
seventh centuries can be most easily represented in terms of two dominant
motifs: the increasing introversion of orthodox culture and the quest for
security. By the former, I mean the generalised concentration in the
thinking of the society as a whole on the personal relationship of indi-
viduals to God, the identity of orthodox thinking with the survival of the
Roman oikoumene, and the exclusion of all marginal or heterodox groups
from consideration, in a much more emphatic way than had been the case
before the last years of Justinian. By the latter, I mean the manifest collapse
of confidence in the traditional symbols of earthly supremacy, in particular
the institutions of the imperial establishment, and the search for ways in
which an imagined older order - of stability and confidence - could be
recovered, which would fulfil the desire both to conform to the spirit and
the letter of Chalcedonian orthodoxy, and at the same time to reassert the
political dominance of East Rome and the symbolic universe it represented.
The first is evident in the literary texts of the period; the second is implicit in
the assumptions, actions and responses of individuals and groups within
Byzantine society in the period from the 640s and beyond.
These two motifs were, of course, inseparable, since the maintenance of
orthodoxy and the exclusion of heterodoxy automatically - it was believed
- brought with it, ultimately, political supremacy, at least until the events
immediately preceding the Last Judgement. In this respect, it seems to me
that some of the apocalyptic literature of the times conceals also - some-
times rather obviously - elements of optimism which belie the belief in a
supposedly imminent Day of Judgement. The Apocalypse of Pseudo-
Methodius especially - understandably extremely popular and widely
disseminated throughout the orthodox world in the Middle Ages - looks
forward to the defeat of the Muslims and the reconquest of all the lost lands

436
 The transformation of a culture 437

of the empire before the Second Coming, and thus - given the bleak
outlook in the second half of the seventh century - implicitly provides what
must have been a rather lengthy breathing-space before the end of the
world could be expected.1
It is in this context that we must try to interpret and understand the
evolution of seventh-century social, political and cultural history, and not
just within the empire. Similar concerns and anxieties affected also the
monophysite and neo-Chalcedonian communities in the Near and Middle
East, too. It is a world in which the public pluralism of the late antique past
had been eradicated, because it no longer represented a comfortable mode
of understanding and acting within (and upon) the world as it was
perceived. Security, and the assurance of doing the 'right thing', could be
found in a uniformity of belief, both imposed from above and at the same
time desired by the 'ordinary' people. Yet the contradictions between the
theory of the world represented by Church and state at the public level, and
the rapidly changing circumstances in which people found themselves, at
different times and in different places, could not be papered over so easily
by the imposition of a rationale from above. Individuals continued to ask
questions and to seek answers, whatever their social and cultural position.
Men such as Maximus Confessor represented the anxieties and uncertain-
ties of many less literate and articulate than themselves. A common motif
remained 'orthodoxy' - but how to attain it, how even to understand it?
These were questions to which neither the official Church nor the Byzan-
tine state were able always to find answers that could be translated
immediately into action, although not for want of trying. And even
Maximus' theology remains just that: a theology, intellectually rigorous,
but politically no more nor less valid than the theories of imperial absolu-
tism against which he pitted his talents.
Within the framework of orthodox piety, therefore, and the exclusivist,
anti-pluralistic political ideology of the imperial state, recourse to the
resolution of problems through an intensely individualistic or personalised
devotion was predictable; for the umbrella of official belief could not handle
the multiplicity and variety of personal needs, nor could it provide more
than the most generalised response to the detailed questions of people
anxious about the very foundations of their day-to-day lives.2 The daily
worries which afflicted everyone were ever-present, of course; and it was
1
Ps.-Methodius, Apokalypse XIII, 7ff. Even more optimistic is the short chronological treatise
of the monk Theophanius, compiled in about 710, in which the calculation of the end of the
world, based on a cabbalistic-numerical equation, falls in the year 880. See E. Dobschiitz,
'Coislinianus 296', BZ 12 (1903), 534-67, see 549ff.f and Beck, Kirche, p. 473. See
chapter 9 and note 116 above.
2
Some of these aspects are described by N.H. Baynes, in Byzantine Studies and Other Essays
(Oxford 1960), p. 5.
438 Byzantium in the seventh century

partly as a result of their effectiveness in providing answers and advice, as

well as alternative models for repentance and the absolution of sins, that
the hermits and holy men of the later Roman world had become so
numerous and popular. And in the shrinking world of the second half of
the seventh century, such issues took on a greater significance: not only
did they affect one's personal life and salvation; but individuals were
regarded as being jointly responsible for the salvation and maintenance of
orthodoxy, the safety of the Chosen People and the restoration of the
Christian oikoumene. The sorts of questions which occur in the 'questions
and answers' attributed to Anastasius of Sinai, and in particular the
tension which they express between divine will and foresight, on the one
hand, and human self-determination and responsibility on the other
(although understood within a divine plan in which this limited freedom
was inscribed), bears a new significance in this context.
It would be quite misleading, of course, to suggest that the sorts of
conditions which I have described above and the attitudes, beliefs and
practices which seem to me typical of late sixth- and seventh-century
society, in both Constantinople and the provinces, came into being over-
night or developed uniformly across the empire in clearly detectable 'acts'
on the historical stage. On the contrary, the evidence is relatively sparse
and relates to only limited areas of the empire, or to a limited context at a
specific moment. It represents a brief glimpse into the lives of individuals or
anonymous groups of people for which we have to provide both the
interpretational framework and the specific rationale or logic. In this
respect, any conclusions we may draw are bound to be incomplete, flawed
and, in some cases, still quite hypothetical or tendentious. On the other
hand, the evidence from all the different sources - literary, archaeological,
sigillographic and so on - when considered as a whole and within a single
cultural context, does bear out the interpretation offered here. It is impor-
tant to remember that we are dealing with several generations of lives - a
short period for the historian, perhaps, but each life represents the growth
of an individual personality and psychology constituted within the cultural
context into which it is born, with all the aspirations, desires and relation-
ships of a social being. The changes which we observe, detected through
different forms of evidence in different ways, affected people differentially
and qualitatively over a longer or shorter period in all these respects. We
observe what we take to be the results of change, but it is not always easy
to discern the causal relationships which generate change in the first place,
nor to determine the point at which quantitative evolves into qualitative
transformation.
It is useful to bear these points in mind when looking at a period of such
complexity as the seventh century. For there is no doubt that many of the
 The transformation of a culture 439
practices which we can detect or observe - for example, in the accounts of
miraculous cures and the wonder-working power of icons, or implicit in
the questions and answers which we have referred to - are, individually,
not new, nor are they surprising in the cultural milieu of early medieval
Christianity in the east Mediterranean basin. And it also seems to be clear
that we are dealing with a period which represents only the resolution of
many much more long-term developments; indeed, it is perhaps only the
later eighth century which sees the completion of these processes of
transformation. But the specific context of the seventh century, in which
the cultural traditions we have mentioned were carried on, does give them
a particular significance and importance, both in respect of their moti-
vation and in respect of the results which they were intended or thought to
have, on the one hand, and which they may actually have had, on the
other.
It is, therefore, precisely this 'whole', this context, which makes it
possible to understand individuals and their actions, whether singly or in
groups, consciously or unconsciously, in the historical past. The seventh
century is clearly a qualitatively very different era from the sixth, or it
becomes so as it progresses; and the beginnings of this cumulative qualita-
tive evolution, which affected different aspects of different elements of East
Roman culture and perceptions at different times, can be situated, as we
have seen, already in the last years of Justinian. What the evidence
represents, therefore, are the efforts of different elements and groups within
Byzantine culture to come to terms with what was perceived as a changing
situation over several generations. The central concern was the reaffir-
mation of the validity and relevance of the values and perceptions of the
world which had been affected by the dislocations we have described - in
other words, the reassertion of the narrative representations of the society
as a whole and in terms of the various social groupings which made it up.
The growth of social and political explanations of the world which
seemed to make sense of the changes of the later sixth and early seventh
centuries is evidence of their initial success. Increased public and private
reliance upon icons, for example, as more direct channels of mediation and
access between God and man, imperial ceremonial emphasis upon the
emperor as devout servant of God and the conjunctural effects of Heraclius'
successful 'crusade' against the Persians, they all served to stabilise the
situation, to reaffirm an interpretation of the world in which the imperial
state and the Chosen People would ultimately triumph. But shifts in
perception, leading in turn themselves to shifts in social praxis, are them-
selves constitutive of change. And the new situation thus attained was
very different from that of the sixth century, both as regards the position of
the empire militarily and politically with respect to neighbouring powers
440 Byzantium in the seventh century

(especially in the West), and in terms of the ideological issues taken for
granted which now dominated.
The trauma of the Arab attacks and victories, the loss of the East, the
devastation of much of Anatolia and the great siege of the years 674-8, as
well as the loss of effective imperial control in the greater part of the
Balkans - together with the cumulative results of the withering away of
urban centres as meaningful social, economic and political elements
within both state and society - all had a drastic effect, throwing once more
the legitimating theories of the imperial political ideology into question,
opening up a gulf between the 'reality' of actual events and the traditional
narrative representation of the world. The emperors (rather than the
imperial position as such) were now also to feel the effects of the con-
sequent search for a reordering of the elements of the available legitimat-
ing theories, and their response was predictable. The monenergite and
monothelete policies of Heraclius and Sergius, but especially of Constans II,
signalled one attempt to redress the balance and to try to regain the
stability which had been regained for a short time by the 630s. And while
the attempt failed, the question of the maintenance of (imperial) authority
became increasingly crucial in the ever-more uncertain situation which
developed during the 650s and beyond, as is clear from the ways in which
the imperial government handled the cases of Pope Martin and Maximus
Confessor and their adherents.
As we have seen, one of the most significant aspects of later seventh-
century history is the increasing Interference' of officers and soldiers from
the provincial armies in the internal politics of the state. This activity - less
Interference' than 'participation', for soldiers had traditionally long had a
role in certain aspects of the imperial succession, for example - reflects the
position of the soldiers in provincial society, and given the increasingly
localised recruitment-patterns which take hold from the 650s, the atti-
tudes also, however refracted, of the provincial populations. The actions of
soldiers at this time can be shown to represent a general loss of faith in the
efficacy of the traditional symbols of authority, and hence the growing
ineffectiveness of traditional legitimating narratives. Military unrest was
obviously determined in form and content by its specific context.3 It
expressed itself in terms of the situation as it existed, and it was precisely
because the traditional, stable framework through which the world could
be made sense of began to break down, and was perceived to be under
threat, that soldiers took the sorts of action which are described in the
sources.
The activities of soldiers, which constituted a direct threat to imperial

3
Haldon, Ideology and social change', 178ff.
 The transformation of a culture 441

authority, and the activities of the emperors themselves, seem to confirm

this interpretation. Whether usurpers or not, emperors struggled through
various means to reinforce and to assert their authority, both through the
traditional ideology and by emphasising all those aspects of the systems of
legitimation which were central to Byzantine ideas about the world - the
role of the emperor as God's vicegerent on earth, servant and implementer
of his will, guardian of orthodoxy and so on. Constans II's reign is
particularly striking in this respect. But equally, Constantine IV did not
hesitate to deal in an exemplary manner with the soldiers who wished him
to crown his two brothers as co-emperors in 681, 4 at the same time
attempting clearly to integrate the armies into the framework of his
authority, just as Justinian II did a few years later.5 Constantine's conven-
ing of the sixth ecumenical council and the confirmation of the doctrine of
Chalcedon similarly served to reassert imperial authority and prestige;
Justinian II, with the Quinisext, reasserted even more forcefully the position
of the Eastern emperor, both with regard to the Byzantine Church and more
especially the papacy.6 His coinage stresses both his role as victorious
emperor over his enemies and the enemies of orthodoxy, as well as his
position as the defender of orthodoxy and the humble servant of God. The
failed attempt of Philippicus Bardanes to re-establish a monothelete policy
on a permanent basis likewise represented an attempt to reassert imperial
authority and the search for a set of narrative evaluations of the past which
would provide explanations and solutions to the problems of the earthly
rulers and provide a practice acceptable to God which would bring an end
to the chastisement of the Chosen People at the hands of their enemies.
The activities of soldiers, when seen in ihis context, become also much
easier to understand. Their intervention seems to have very little to do with
grievances over matters such as pay, equipment or service conditions
(although these undoubtedly existed), in clear contrast with the years up to
the middle of the reign of Heraclius, all the more surprising in view of the
desperate situation of the state at this time.7 From the early 640s there are
some eleven recorded instances of successful military rebellion - excluding
the deposition of Justinian II in 695 - up to and including the accession of
Leo III.8 All these examples demonstrate either an immediate political

4
J.B. Bury, The Later Roman Empire, from Arcadius to Irene (395-800) (2 vols., London and
New York 1889), vol. 2, pp. 308-9; Winkelmann, 'Zum byzantinischen Staaf, p. 217;
Stratos, Byzantium in the Seventh Century, vol. IV, pp. 135-40.
5 MansiXI, 201C; Riedinger, 10.2Iff. and 737 (Riedinger, 10.2Iff. and 886.2-24).
6
See Mansi XI, 201C-D for Constantine IV's opening remarks in the sacra to the patriarch
George; and for the Quinisext, see Winkelmann, Die ostlichen Kirchen. In general, see
chapters 2 and 8 above.
7 See Haldon, 'Ideology and social change', 178f.; Kaegi, Military Unrest, pp. HOff.
8
See the summary in Haldon, 'Ideology and social change', 182ff. with literature.
442 Byzantium in the seventh century

objective or commitment of the soldiers or officers or a general attitude of

hostility to the Constantinopolitan government or the emperor himself,
which could be exploited for or against a particular party. The accession of
Leo III, generally looked at by modern historians as a watershed in these
trends, did not in fact end this activity. Indeed, the rebellion of Sergius, the
commander in Sicily, in 718, and the plot of Anastasius II, exiled in
Thessaloniki, in the same year, emphasise the fact that Leo's position was
no more secure at this stage than that of any other usurper.
What is clear is that officers and soldiers were central players in all the
coups and attempted coups of this period. It would be an oversimplifi-
cation, of course, to suggest that the same motives, nuanced in the same
ways, were present in every case. Soldiers may well have been the tools of
more particularly aware and/or self-interested officers at times (although
the latter must have been able to exploit an already given frame of mind
which did not prohibit such actions); while on at least one occasion - the
overthrow of Anastasius II in 715 - the troops of the Opsikion seem to have
acted quite arbitrarily. But this in itself seems to indicate both the soldiers'
awareness of their central position and the power they could consequently
invoke, as well as a possible provincial hostility to Constantinople and its
privileges (as perceived, perhaps, from the provinces).9 Soldiers and officers
were involved in the politics of government at this time because they were
directly committed to reasserting a stable framework within which the
world could be understood.
The quest for security referred to at the beginning of this chapter,
therefore, was carried on at several levels and produced a range of conflicts
between the representatives of different interpretations of how the
'problem' - likewise, perceived differently from each of several perspectives
- should be tackled. We are confronted in the seventh-century Byzantine
world with a society and culture full of contradictions. On the one hand,
the theoretical (but in many ways also quite real) unity of an orthodox
culture in which individuals sought to maintain a common front against
outsiders and heresy, through the attainment of piety and the observance
of the commands of the Church. The degree of actual uniformity between
regions, and within them, whatever the exhortations or formal demands of
the Church and canon law, was probably minimal, given both the physical
disruption of the times, the degree of awareness of the problems and the
degree of literacy of the priesthood. But a consensual uniformity - a desire
to conform - did exist.
On the other hand, the conflicts which took place, as those with the
power to pursue a particular solution took up their case, show how fragile

9
See my comments in chapter 9: The search for order: the case of the soldiers, above.
 The transformation of a culture 443

this consensus might be and how variously the perspectives of different

social and economic groups affected their interpretation. It is no accident
that provincial soldiers concentrate their hostility, and their interest, on
the capital and on the emperor, for this was ideologically the only sensible
way of approaching the problem as it was perceived. While orthodox piety
and conformity to the canons of the Church represented the values in
accordance with which the Chosen People could recover their sense of
direction and destroy their enemies, as well as marking out the role of
emperor and Church in society at large, it also provided the justification for
alternative interpretations of the world, interpretations which were
informed by and generated within a very different context from that of the
imperial court.

CONTRADICTION AND REGENERATION: THE DYNAMIC OF THE

BYZANTINE SOCIAL FORMATION

Why did the Byzantine empire not succumb to the various forces, internal
and external, which during the seventh century threatened to destroy it?
The question has often exercised the minds of historians. Some have seen
its survival as mere accident, the failure of its foes adequately to organise
their efforts at conquest or the result of unavoidable internal divisions
within the caliphate. Others have seen the impregnable position of Con-
stantinople, the queen of cities, as the key; yet others have regarded the
strength of orthodox Christianity and the cultural bonds it forged as a
crucial factor; while some historians have seen the well-structured and
flexible administrative, fiscal and military apparatuses of the state as the
foundation of its survival. 10 All of these - although I should wish to modify
each statement in different ways - played a role, of that there can be little
doubt. But to look for single causes, or indeed prime movers, is to misun-
derstand the very nature of historical change. For in many ways the late
Roman state did not survive, at least not in the sense that protagonists of a
'continuity' approach to the problem would have us believe. The physical
space - albeit much reduced - the geography and climate (with natural
and usually very gradual shifts) remain much the same. But late Roman
urban culture vanishes entirely, along with much of the cultural baggage
it carried with it. Instead, new systems of thought develop, new
10
For the most recent attempt, see G. Huxley, Why Did the Byzantine Empire not Fall to the
Arabs? (Athens, Gennadius Library, 1986), pp. 1-14; and the resume of the argument in
BS 68 (1987), 267; and see the article of G. Weiss, 'Antike und Byzanz: die Kontinuitat
der Gesellschaftsstruktur', Historische Zeitschrift 224 (1977), 529-60. For a useful survey
of the literature on the question of continuity and discontinuity, see the discussion of
V. Vavfinek, 'The Eastern Roman empire or early Byzantium? A society in transition', in
From Late Antiquity to Early Byzantium, ed. V. Vavfinek (Prague 1985), pp. 9-20.
444 Byzantium in the seventh century

approaches to art and representation are refined, new administrative

structures are evolved. Power relationships within the ruling elite also
change - the old senatorial establishment, with much of the literary
culture associated with it, disappears, to be replaced by a very different
elite, of different social, cultural and often ethnic origins. Those aspects of
the traditional elite culture that did survive came to play a different role in
the ideological world of this new class, although there is no reason to
doubt that this new medieval elite included elements of the older estab-
lishment. Continuity of language, at least as regards Greek and Armenian,
is clear; while the native languages of Asia Minor were by the early
seventh century mostly extinct. Continuity of tradition, in terms of
economic organisation at community level, of the rituals accompanying
birth, marriage, death, of popular forms of music and song, these can all be
found, or assumed, to a degree. But even here, the developments of the
seventh century brought about considerable changes, both in respect of
the organisation of property and property relations (and hence the
exploitation of arable and pastoral land) and in the Christianisation of the
rituals and meaning of social and biological relationships.
Whatever continuities and survivals one might point to, therefore, a
very different culture emerges on the surviving remains of late Antiquity.
Of course, it is not hard to find a vast array of single elements which
continued to exist, whether in terms of social and cultural praxis, adminis-
trative organisation, titles and ideas about order and hierarchy, literary
motifs and stylistic systems, and so on. But this is hardly surprising, given
the fact that the entire population of the late Roman world, along with its
culture, was neither wiped out nor enslaved and transported far away. No
one would seriously claim that eighteenth-century English culture and
society were the same as that of the fourteenth century, in spite of the
obvious elements of continuity. The point is that continuity in some form
can always be discovered where one social and cultural formation devel-
ops qualitatively enough to be transformed into something which is clearly
different. But when the basic structures which determine how that social
formation works and reproduces itself can be seen to have been funda-
mentally changed, then we can hardly talk of a real continuity, but only, at
the most, of the survival of traditions and forms - survivals which may
often be determined by the levels of technology available, for example,
rather than by any preference for doing things in a particular way; or by
ideological 'hangovers' from earlier times, self-conscious notions about
how things 'ought' to be done, none of which affect the structural dyna-
mics of the society.
An East Roman or Byzantine state 'survived' the seventh century,
therefore, for all the reasons listed at the beginning of this section. But it
 The transformation of a culture 445

was, as I hope I have been able to show in this book, a very different state
in its workings from that of the late Roman period. The question now
arises, how we set about characterising the society and state formation
which developed, and within what sort of explanatory framework we can
clarify the nature of the transformations which took place in order to
understand how that society functioned at a structural level.
The first, and probably the most obvious point, concerns the fate of the
late Roman cities.11 The evidence which we have reviewed suggests quite
clearly that the urban centres of the late ancient world were almost totally
eclipsed during the period with which we are concerned. The classical city
had come during the Roman period to occupy a central role both in the
social and economic structure of Mediterranean society and in the
administrative machinery of the Roman state, as focus of market-exchange
activity, of regional agricultural activity and, especially, as tax-collectors.
Where cities in the Mediterranean sense did not exist, the Roman state
created them, either establishing entirely new foundations, or amalgamat-
ing earlier settlements and providing them with the corporate identity and
legal personality of a civitas or polis. These are terms which, we must
remember, bore a specific administrative and fiscal significance in the late
Roman world and did not necessarily imply a major urban centre. The
long-drawn-out shift in the economic and social function of these cities
from the third century on, the final suppression of their economic indepen-
dence during the later fifth and early sixth centuries, and their physical
and social devastation during the seventh century, had consequently very
considerable implications for the fiscal and civil administrative structure of
the late Roman state. Equally, it marked the end of classical city life and
culture, with its urbanocentric economy. One way of looking at these
changes is to see in them a decline in urban civilisation. A more productive
approach is to see them as a gradual change in the function and relevance
of cities to the needs of the late Roman state and the society which
supported it. However we approach the problem, of course, the results
were the same for late Roman culture and, in particular, for the state.
Centralisation of tax-collection and the replacement of local centres of
power by Constantinople, with the resultant attraction of social and
cultural life to the capital, were part of the history of the late Roman world
which the events of the seventh century brought to completion. The
Byzantine state which emerged into the eighth century was effectively an
empire of one major urban centre, together with a few fortunate provincial
urban centres which were able to survive as emporia and ports.
This development had several implications. In the first place, it meant

11
For a fuller survey of the points made here, see Haldon, 'Some considerations', 78ff.
446 Byzantium in the seventh century
that the state on the one hand, and large landowners (private and
institutional) on the other, inherited the wealth, or at least the territories,
which had previously been exploited by the municipalities. But it also
meant that, on the whole, the wealth extracted from such regions (chiefly
in the form of surplus agricultural and pastoral produce) tended to flow
towards Constantinople, for this was where power was increasingly con-
centrated. It was to Constantinople that the provincial wealthy would send
their sons to be educated or to gain an entree to Church or state service; it
was there that the emperor and court were resident, from where titles and
privileges, offices and emoluments were bestowed, and from where the
imperial administrative machinery, as well as that of the Church, was
directed. The state centred at Constantinople inherited the social role of the
cities, in so far as the absence of cities meant that, with the exception of one
(state) institution, the armies, there was no longer a formal and
institutional intermediary between the mass of the provincial population
and the imperial government. It is this lack which goes part of the way to
explaining the significance of the armies and their leaders - a point we
have already examined.
The second point relates to the economic structure of late Roman and
early Byzantine society. The city had never been a centre of industrial
production. Commercial and industrial centres existed, of course, but they
were relatively few in number and were in any case more or less entirely
dependent upon their hinterland for basic supplies of foodstuffs, means of
transport and, more often than not, for the extra labour needed for major
building programmes. The cost of overland transport was, for most cities,
prohibitive, which meant that those situated away from the coast or from
easy access to a port were unable to exist on a non-agricultural basis. Ports
were in a different position, of course, and the major cities of the late
Roman world were almost all on navigable rivers or had good coastal
harbours. The exceptions can generally be explained in terms of state
intervention - the state was the only agency which could maintain the
resources sufficient for this activity - or local and usually short-lived
economic booms. The greater part of the state's income came from taxes
on land and property of one sort or another. Commerce brought in a very
limited income, although it was probably greater in the eastern part of the
empire than in the western. Until the middle of the seventh century the
empire was made up, in effect, of an agglomeration of city-states and
subsistence economies; with the final demise of the cities, this essential
local subsistence character did not change, although the local centres were
eradicated. The agricultural population of the empire, whatever the differ-
ent social and juridical categories into which it was divided, had to rely
upon its own resources, depending upon local market-exchange through
 The transformation of a culture 447

the medium of the lower-denomination bronze currency or, in the absence

of the latter, upon direct exchange between producers, for those items
which they could not produce themselves. Since overland transport was so
expensive and slow, the provinces of the empire had of necessity to be
economically self-supporting. While wealthier landowners could invest
surpluses from their estates in luxuries, the extent of this investment was
limited again by their proximity to trade routes or large entrepots, and by
the state of market-exchange and the availability of cash with which to
buy. Since it was the local wealthy - both curiales and independent
landowners, members of the senatorial elite - who provided much of the
wealth to invest in their municipalities, the shift in their attention away to
the capital must have clearly had drastic implications for the physical
maintenance and the social and cultural life of their cities. Of course, in
times of relative political stability, market-exchange was able to flourish in
the most remote regions of the empire, too; and even after the cities had
lost their fiscal-administrative functions, the larger ports and adminis-
trative-economic centres continued to flourish under state patronage. But
such activity generated only a very small proportion of the surplus wealth
expropriated by the state as tax or by landowners as rent. What was the
role of the state in this wider structure of economic relationships?
The only institution capable of maintaining large-scale overland trans-
port activities was, of course, the imperial government, whose complex
system of posting- and baggage-stations was intended to further the
movement of supplies and other necessary requirements for the army - by
far the single most expensive item in the imperial 'budget' - the civil
administration and related institutions. The transport system was itself
supported through extraordinary levies and impositions in services and in
kind, placed upon the agricultural population of the empire.
Just as this transport system was evolved to serve the needs of the state,
so the imperial coinage fulfilled a similar function, and the role of the gold
coinage is the linchpin of the system. Coin was issued in thefirstinstance to
serve the requirements of the administration and the army. Naturally
enough, it also gravitated towards commercial centres, but this was not a
deliberate policy in its issue. It was a tool and a symbol of the state and,
more specifically, of the emperor, and it was also expected that gold coins
issued for the state's purposes would eventually be returned through the
medium of tax. In the period before a stable relationship between gold and
bronze currencies, in the latefifthcentury, the lack of an adequate medium
had forced the state - during the fourth and first part of the fifth century
especially - to fall back upon taxation and the maintenance of its own
apparatus, in kind, for which a vast and complex system of tariff equiv-
alents had been set up. The establishment of a reliable gold currency and a
448 Byzantium in the seventh century

related bronze coinage made much of this redundant, and it also made
easier market-exchange on a much more widespread basis. The fact that
the state demanded the payment of taxes in gold implies the existence of
some sort of market sphere in which the agricultural produce of either
peasant smallholders or large estates could be exchanged for gold coin. But
this market was itself stimulated to a large extent by state demands, since it
was chiefly the army and the state administration who consumed agri-
cultural surpluses on a grand scale, and they received their salaries -
usually - in gold. Limited amounts of gold might be absorbed through
commercial activity in larger cities or taken out by elite trade or hoarding.
But it was intended that, through taxation, as much as possible would be
recovered. To what extent gold actually changed hands, for example,
between peasants, or even big landowners, and the tax-collectors, is very
difficult to establish. Much of the agricultural produce may have been
collected directly from individual farmers or the storehouses of larger
landowners by the prefectural officials responsible for supplying the army.
Receipts against their tax-assessment will have been issued in return.
Likewise, the payments for the military supplies may have been taken out
of the soldiers' salaries before these were issued, being then handed over
directly by the military paymasters or optiones to the prefecture, the cost of
annonae and capitus having been established in advance. These procedures
are known to have applied quite normally under certain conditions, and
there is no reason to suppose that they did not apply in these circum-
stances, too. Such a process permitted the state to maximise the extraction
of surpluses for its apparatus, while at the same time ensuring that as great
a proportion as possible of its gold issue was returned directly as tax. At the
same time, the tax-officials practised a system of unequal exchange, in
which taxes demanded in gold were rounded up for convenience and
change given in the bronze currency. Large-scale tax-farming - either to
wealthy landowners or to other private contractors - also encouraged the
maximisation of the return, of gold, although it undoubtedly meant in
addition the existence of widespread profiteering.
The gold and bronze currencies served different functions. Gold was
intended to promote the operations of the state and the extraction of
surpluses in the form of tax. The bronze and occasional silver issues (after
the reign of Anastasius) were intended for local exchange activities and
day-to-day commerce. The bronze coinage in particular was a nominal
money of account used at a relatively low level of transaction and was
dependent for its acceptability as a means of exchange on the stability of
the gold, to which it was tied. As we have seen, the lack of any evidence in
the period from c. 660 to c. 800 for issues of bronze on any substantial
scale and its use in market contexts strongly suggests that no such activity
 The transformation of a culture 449

took place, except in the highly localised context of major centres such as
Constantinople - by its very nature an exception. In contrast, the con-
tinued issue of a stable gold coinage demonstrates the role of the state in
regulating the distribution of surplus wealth according to its own require-
ments. Of course, in certain circumstances the issue of gold coin may not
have been sufficient to meet the exigencies of the moment, and so state
officials may have had to accept bronze coins as payment - such was the
case in the reign of Heraclius, for example, and under Constans II in Italy,
where the scarcity of resources reached crisis proportions. 12
These considerations are valid both for the later Roman period and for
that from the ninth century on. What evidence there is makes it clear that
they apply equally to the seventh and eighth centuries. While the munici-
palities of the late Roman world vanish, the essential relationship between
state and resources continued to be mediated on the same principles. The
disappearance of the cities, the centralisation of fiscal control, the shift in
production relations which followed the expansion of emphyteutic leasing
and the demise of the senatorial landowning elite, all these factors,
together with the dramatic loss of revenue which resulted from the Islamic
conquests of the second half of the seventh century, meant that the
structural details of the state's extraction and redistribution of resources
were changed, sometimes rapidly, sometimes more gradually. The slow
dissolution of the prefectural instance of fiscal and civil administration
provides a good example of the latter; the rapid transformation of the
central officia in the 620s and 630s one of the former. In terms of
continuity, therefore, the principles of the state's mode of surplus appro-
priation remain much the same: a surplus redistribution system based on a
subsistence peasant economy, a system facilitated by a gold coinage which
served both to promote the extraction of tax and the maximisation of the
state's returns. On the other hand, the structures which evolved to fulfil
these functions changed very considerably after the later sixth century,
and there is no doubt that the seventh century marks a major break with
the late Roman system.
More than anything, however, the seventh century marks in many
respects the reassertion of imperial authority and the power of the state
over its resources. All the factors we have examined in the course of this
book point to this conclusion. The decline of the senatorial elite, the
shrinkage of the empire territorially, the centralisation of fiscal administra-
tion, the disappearance of the cities as intermediate levels of government,
the rise of a new military and administrative bureaucracy dependent much
more directly upon the emperor than hitherto - all are crucial to this shift.

12
See the comments of Hendy, Studies, pp. 228ff. and 415-16.
450 Byzantium in the seventh century

Along with the developments in ideology in the period after Justinian I and
the urgent need for the state to intervene much more directly in provincial
administrative matters in order adequately to respond to the political and
financial demands of the period after about 640, these factors resulted in a
much stronger and more centralised state apparatus and a society ideo-
logically much more homogeneous, and in some respects more dependent
on the emperor, than hitherto. The contrast between the loose hegemony
of the sixth century and the authoritarian centralisation of the later
seventh century is clear. Of course, government control was considerably
tempered by the difficulties of communications and transport, and so we
should not assume a great deal of 'efficiency' in the contemporary sense of
the term, for either period. But the dominant and leading element in
seventh- and eighth-century society was the imperial court, which had no
longer to compete for wealth or authority with a large and wealthy, and
potentially - economically and in many respects culturally - independent
ruling class, nor with centres of urban civilisation outside Constantinople
itself.
All these changes, both in structure and in emphasis, within late Roman
social relations and ideology, are part of a process through which Byzan-
tine society - that is, the medieval, orthodox culture of the east Mediter-
ranean - comes into being. In many ways, the seventh century is the end,
rather than the beginning, of many of the qualitative transformations I
have described.13 But, in a wider context, how are we to define this social
formation?
Here, inevitably, we must address the question of concepts and theories
- often dismissed by historians as an irrelevant exercise involving the
setting-up of descriptive boxes which serve no functionally relevant
purpose to the real task of research and analysis. Sometimes, it is true,
historians and social scientists have tended to use general theories as a
substitute for real explanation, fitting in the evidence and their description
of it to pre-ordained explanatory models which are often in themselves
inadequately thought through or far too general to be of any value in
understanding the causal relationships which bring about historical
change. I have touched upon this subject in the introduction, where I have
also suggested the importance and heuristic value of such theories. For
13
Here I would part company with Patlagean, Pauvretf economique, who regards the seventh
century as only one stage in a longer process, culminating in the eleventh century; and I
would argue a radically different historical development from that set out by Weiss (see
note 10, chapter 12, above). Of recent commentators, both Mango, Byzantium: The Empire
of New Rome, and D. Zakythinos, 'La grande breche dans la tradition historique de
l'Hellenisme du septieme au neuvieme siecle', in Charisterion eis Anastasion K. Orlandon, 3
vols. (Athens 1966), vol. Ill, pp. 300-27, argue for a transformation in, or completed by,
the seventh century. My own view is that the seventh century marks a crucial turning
point in an evolution that was completed by the early ninth century.
 The transformation of a culture 451

Marxist historians, the problem of the adequacy of the analytical cate-

gories at their disposal is particularly pressing, since historical materialism
claims above all to be able to provide both a general theory of social and
historical change and the heuristic categories which facilitate a detailed
analysis of historical phenomena within a social formation. And for any
historian, adequate categories are essential if we are to be able to focus on
real causal relationships.14
The two fundamental categories for a historical materialist analysis are
those of mode of production and social formation or, more generally,
'society*. In the Byzantine and late Roman case, therefore, we are faced
with the specific problem of whether or not a transition took place over the
period from the later third century to the later eighth and ninth centuries,
and at what level: from one mode of production to another - from the
'ancient' to what is traditionally referred to as the'feudal' mode but which I
prefer to call the 'tributary' mode? - orfromone social formation to another:
what elements within the social formation were representative of these
modes, which of them dominated, and how were they articulated to
produce a specific and recognisable late Roman or Byzantine society?
Let us review, briefly, the two categories mentioned above, since these
provide the framework within which any discussion of the generative
structures which causally explain qualitative change must take place. A
mode of production consists in 'the relations of production in their totality',
that is to say, the forces of production (land, water-power, draught animals
and so on) and the relations of production (chiefly between those who
operate the forces of production, and those who effectively control those
forces, the produce which results from their exploitation and the distri-
bution of that produce). This totality constitutes in the widest sense the
economic structure of society, the framework within which political,
cultural and ideological structures are embedded and with which they
'correspond', that is, within which their fields of effect and potential paths
of development are delineated. A mode of production is thus an abstrac-
tion, a purely theoretical and heuristic construct designed to provide a
model of how certain types of economic and social organisation of human
beings actually work. And within this construct, what is known as the
mode of surplus appropriation (that is to say, the mechanisms by which
one social class is able to exploit the labour of another in order to extract
surplus wealth for its own consumption or some other purpose) which
dominates is a crucial feature in differentiating between modes of pro-
14
See J.F. Haldon, '"Jargon" vs. "the facts"? Byzantine history-writing and contemporary
debates'. BMGS 9 (1984-5), 95-132; F. Favory, 'Validite des concepts marxistes pour
une theorie des societes de l'antiquite. Le modele imperial romain', Klio 63 (1981),
313-30; and the remarks in Haldon, 'Some considerations', 99ff.
452 Byzantium in the seventh century

duction. It is this, above all, which enables a distinction to be made

between the relations of production in so-called feudal, or ancient, or
capitalist societies.15
A social formation, in contrast, is a particular historical articulation of
these relationships under specific conditions, an articulation which deter-
mines the effectivity of the different elements; that is to say, the dialectical
relationship between these elements and the sort of result this might
produce - in a structural, rather than a particular, sense. 16
Now it has usually been argued that the Roman state consisted of a
number of locally distinct social formations in which one variant or
another of the ancient mode of production dominated. The ancient mode is
characterised, in its initial form, as an economic and social system in
which a (city-based) citizen body controlled a surrounding agricultural
hinterland as private persons, but with collective rights in their ownership
of public land and co-operating together in their economic and political
activities as a community. The state represents this body politic, 'exploit-
ing' the citizens only in so far as it is responsible for the collection of taxes
and other customary obligations. Taxation represents, in the first instance,
merely that part of the surplus contributed by the producer-citizen body on
a contractual basis towards the maintenance and reinforcement of the
political and legal apparatus which secures them in the possession of their
means of production and social reproduction.
Of course, in historical terms, such 'abstract' systems were greatly
coloured by older tribal and clan divisions and organisation, the sexual
division of labour and so on. And as the division of labour increased (that
is, as an increasing differentiation of economic and craft, as well as
intellectual, activities evolved - a process usually accompanying improve-
ments in the mode of exploitation of the means of production, either
qualitatively or quantitatively), so the 'community' of citizens becomes
more and more fragmented. Conquest and expansion might then bring in
supplies of cheap slave labour, so that the egalitarian, peasant-agricultura-
list citizens are displaced both economically and socially, resulting in
limited areas in a 'slave mode of production', in which large estates or
plantations dominate the productive sector, replacing small-scale freehold-
15
See K. Marx, Pre-Capitalist Economic Formations, ed. Eric Hobsbawm (London 1964),
intro., pp. 38ff.; and text, pp. 67ff.
16
See K. Marx, A Contribution to the Critique of Political Economy, ed. M. Dobb (Moscow
1970), pp. 20ff. (Dobb's introduction); and for modern discussions of the concepts and
their relevance, see G. Cohen, Karl Marx's Theory of History: A Defence (Oxford 1978),
pp. 28ff.; and more especially the much-criticised B. Hindess and P.Q. Hirst, Pre-Capitalist
Modes of Production (London 1975), esp. pp. 1-20; P.Q. Hirst, 'The uniqueness of the
West', Economy and Society 4 (1975), 446-75 (review of P. Anderson, Passages from
Antiquity to Feudalism (London 1974)); A. Foster-Carter, 'The modes of production contro-
versy', New Left Review 107 (Jan.-Feb., 1978), 47-77.
 The transformation of a culture 453

ers or tenant-farmers entirely. At the same time, the subordination of

citizen to citizen and the generation of objective antagonisms between
different groups in respect of their relationships to the means of production
and distribution of wealth means that the state becomes the legislative and
executive organ of the ruling class of citizens, which can henceforth use it
to maintain and promote the extraction of tax and their own social-
economic position. The early history of the city of Rome and the republic
provide a good illustration of such an evolution. In the case of Rome, of
course, its dramatic expansion and the gradual imposition of its own
administrative-institutional and tax-raising traditions (much influenced,
naturally, by the methods of the conquered states or peoples) resulted in
the qualitative assimilation of a number of different social formations
within this ancient mode to the Roman model - although, inevitably, a
wide range of regional particularities and idiosyncrasies survived.17 As I
will suggest below, there are problems with this way of understanding the
ancient mode.
The tributary mode of production operates on different principles/and it is
important to us because, in the effort to understand how and why the
change from ancient to medieval social, cultural and economic forms took
place in the Roman and Byzantine world, it has been held by some
historians that Byzantine society was, in contrast to Roman society, feudal
(in the wider usage described above) - although the datefromwhich this can
be said to be the case is also debated: the seventh, eleventh and the twelfth or
thirteenth centuries have all been put forward on different grounds.18
Tributary relations of production are characterised by a mode of surplus
appropriation dependent upon the exercise of non-economic coercion over
producers in possession of their own means of production; in other words,
17
See S. Cook, 'Beyond the Formen: towards a revised Marxist theory of pre-capitalist
formations and the transition to capitalism', Journal of Peasant Studies 4 (1976-7),
360-89; and see also Hindess and Hirst, Pre-Capitalist Modes of Production, pp. 18f. and
79ff. For the debate on slavery and the 'slave mode of production', see 0. Patterson, 'On
slavery and slave formations', New Left Review 117 (Sept.-Oct., 1979), 31-67; E. Wood,
'Marxism and Ancient Greece', History Workshop Journal 11 (Spring 1981), 3-22; and
J.-P. Vernant, 'La Lutte des classes', in My the et societe en Grece ancienne (Paris 1974),
pp. 11-29.
18
The debate has taken place for the most part in the Soviet and Eastern European literature.
For representative works see V. Hrochova\ 'La Place de Byzance dans la typologie du
feodalisme Europeen', in V. Vavfinek, ed., Beitrdge zur byzantinischen Geschichte im 9.-11.
Jahrhundert (Prague 1978), pp. 31-45; G.G. Litavrin, Vizantiiskoe obScestvo i gosudarstvo v
X-XI vv. (Moscow 1977); Z.V. Udal'cova and K.A. Osipova, Tipologiceskie osobennosti
feodalisma v Vizantii', in Problemy social'noi struktury i ideologii srednevekovogo obscestva I
(Leningrad 1974), pp. 4-28. For a Western comment, see Anderson, Passages from
Antiquity to Feudalism, pp. 190ff., and Haldon, 'Some considerations', 99ff. with literature.
For some theoretical observations in respect of the current debate within Western
Marxism, see the brief comments of G. McLennan, 'Marxist theory and historical research:
between the hard and soft options', Science and Society 50 (1986), 85-95.
454 Byzantium in the seventh century

the payment of rent or tax by peasant producers on the basis of customary

or other obligations or rights, backed up ultimately by the exercise of
physical force. This emphasis on non-economic pressure is to contrast with
capitalist relations of production, which are economic in so far as the
labourer is free, but possesses in effect only his labour-power, not his
means of production and subsistence - the need to work, to earn a living by
selling that labour-power as a commodity on the labour market is a result
of purely economic pressures.
'Feudalism' is often summarised in terms of the dominance of landlord-
tenant relations, in which the economic and social position of the landlord
depends upon rent appropriated from the productive labour of tenants and
upon the landlord's ability effectively to control the means of economic and
social reproduction of those tenants. But the extraction of tribute or taxes
by a state power is, at the level of the mode of surplus appropriation, no
different; and in respect of this, crucial, differentiating factor, most pre-
capitalist social formations not dominated by either slave-exploitation or
by the communal or kinship mode can be said to be dominated by 'feudal*
relations of production, in this very general sense. But since this usage has
frequently led to misunderstanding, I prefer to use the term 'tributary' to
define the general category, and to restrict 'feudal' to those social formations
which were dominant in the medieval West from the ninth to the fifteenth
centuries. It can be objected, of course, that such a broad application of the term
deprives it of any specific analytical and descriptive value. But this is easily
countered when it is remembered that tributary relations of production
describe a general mode of surplus appropriation - the specific forms that
general mode takes in different historical contexts is for the historian to
ascertain. The general category helps in establishing certain criteria for that
analysis. And, anyway, the alternative is to abandon all forms of general theory
of pre-capitalist economic systems or to provide a specific mode to match
every particular historical articulation - every different society - of the
forces and relations of production. The latter really does take us in to the
realm of category-collecting for its own sake!
The maintenance of a particular form of exploitative relations of pro-
duction, of course, depends itself upon features which are in themselves not
strictly part of the economic equation - the legal, juridical and political
conditions necessary to the reproduction of a particular social-political
formation. Serfdom alone, for example, if it coexists with other forms of
exploitation, may not be enough to justify categorising a society as 'feudal'-
a social formation is dominated by one mode of production or another
according to whichever mode of surplus appropriation is dominant and
determines the main features of the society as a whole. But this dominance
 The transformation of a culture 455

does not need to be exclusive - elements from other modes of production

may also be present, and it is quite usual for societies to contain elements
from at least two modes, often mutually antagonistic, the possibilities
being determined by conjunctural as well as technical and environmental
factors. Importantly, too, the dominant mode will be reflected also in the
institutional and economic machinery of the state, for example. Where
different relations of production are gaining ascendancy, the state - which
is a product of an older order - will generally adopt an oppositional stance
as the representative of an established set of power relations; but, where
the old order is unable to resist the new, it will be undermined or radically
transformed.19 This process, of course, involves also a process of class
struggle, since every change in the relations of production is precisely a
change in the relationship of one socio-economic class to another.20
Looked at from this point of view, it must be clear that the ancient mode
as traditionally understood is by no means dominant in the Mediterranean
world after the end of the fifth century B.C. For already, class antagonisms
and the evolution of clear levels of social and economic subordination
within the city-states of Greater Greece and Italy signify a shift in the
relations of production. While the dominance of slavery at times and in
limited areas varies the pattern (in the Greek city-states of the fifth century
B.C., for example, or in Roman Italy during the period from the second
century B.C. to the second century A.D.), it seems to me that the dominant
mode of production in the ancient world was, in the strictly political-
19
Discussion of the elements of the feudal mode are, as might be expected, numerous: see
Marx, Pre-Capitalist Economic Formations, ed. Hobsbawm, pp. 125ff. For some useful
surveys, see Anderson, Passages from Antiquity to Feudalism-, Hindess and Hirst, Pre-Capita-
list Modes of Production; and G. Bois, Crise dufeodalisme (Paris 1976), pp. 351ff.
20
Class is used by Marxist historians as more than just a description of the status of a given
social group. Class refers to the relations pertaining between those who own or control the
means of production and distribution of social wealth, and those who do not - usually the
producers themselves. It is in this sense that the term 'class struggle' is to be understood -
not to signify constant, or even occasional, open conflict between social groups occupying
antagonistic positions in respect of the means of production; but rather the fact of the very
existence of such a structural antagonism between social groups. This may express itself
in terms of violence, but this then depends on the context and the conjuncture - it may
equally express itself within the framework of a legal system. It is the actual relationships
of a social class to the means of production which determines its objective economic
interests, interests which may in themselves not necessarily be apparent to the members of
that class, but represented instead in different forms through ideology. By the same token,
we cannot expect to find 'class consciousness' in the modern sense of the term, that is, an
awareness of the economic position of a social group in relation to other social groups. But
this does not affect the validity of the concept of class as both a heuristic device and as a
real description of embedded economic relationships. Awareness of social position, of
course, does occur - represented in terms of status groups, wealth, power, birth and
ancestry, and so on. See my comments in 'A touch of class', Rechtshistorisches Journal 7
(1988). 37-50.
456 Byzantium in the seventh century

economy sense defined above, that of the tributary mode.21

What has confused the issue for medievalists and historians of the late
ancient world has been the nature and structure of the late Roman state.
For it is also apparent that, in the light of the evidence which we have now
reviewed, the dominance of the state and taxation, the imposition by the
state of its surplus-extracting machinery and the employment of those
surpluses in the maintenance and extension of state supervision of the
surplus wealth produced within its lands (however inefficient this may
appear to modern eyes) are key elements - both in the idea we have of the
nature of East Roman society and in the reasons for its survival.
Unlike the situation prevailing in the late Roman West, the contra-
diction between the state and its interests, on the one hand, and the
wealthy senatorial elite who dominated its civil apparatus, on the other,
never reached crisis proportions. In the West, the growth of senatorial
landed property stimulated a polarisation of economic and ideological
interests. Increased wealth meant a quest for the avoidance of tax-liability,
so that eventually the interests of the ruling elite as landowners out-
weighed their public interests and duties as tax-payers and supporters of
the state. For a variety of reasons, partly connected with the force and
effect of the Germanic invasions, partly with the greater sparsity of large
cities and the more central position of large-scale estate exploitation in the
relations of production, partly also with the extension of relations of
patronage and subordination between landlords and tenants on a wide-
spread basis, the extraction of rent by private landlords as opposed to that
of tax by the state came to dominate production-relations - and therefore
surplus appropriation - as a whole. The result was that taxation and the
hegemony of the state were weakened to the point where they were no
longer able to maintain themselves against the interests of the powerful
private landlord or magnate class.22
The East presents a vivid contrast to this picture. There, during the
fourth, fifth and sixth centuries, and as a result of a much more diversified
set of production-relations, a far greater proportion of independent small-
holders and tenants, a less dominant senatorial elite and a much greater
number of towns and commercial centres, the state was able to keep the
contradiction between its own interests and those of the ruling class in
check, although it was certainly present. But during the sixth and seventh
centuries, as we have seen, the dominance of this class of landowners is
21
See J.F. Haldon, 'The feudalism debate once more: the case of Byzantium', Journal of
Peasant Studies 17/1 (1989), ( 5-39; and especially the detailed discussion in my The State and
the Tributary Mode of Production (London 1993).
22
For a more detailed analysis, see CJ. Wickham, 'The other transition: from the ancient
world to feudalism', Past and Present 103 (May 1984), 3-36.
 The transformation of a culture 457

broken. The state remains as the sole focus of economic and political
power, and although the Church, as the major single landowner after the
state, must be seen as a potential rival, the interests of Church and state
were ideologically so closely bound together (in the long term: ruptures of
the sort typified by the monothelete controversy, for example, did not affect
the institutional, property-owning Church in its economic relations with
the state) that this was not an important source of conflict at this period.
Landlord-tenant relations and the privatised extraction of surpluses - as
rent - o n a widespread basis represented the major element in the relations
of production, of course. The state was itself a major landowner, along
with the Church. But the evidence suggests that the proportion of wealth
extracted by the state through tax was at least as structurally significant as
that collected as rent; and the continued existence of a powerful, central-
ised state, with an elaborate bureaucracy and professional (or at least
permanent) army, maintained by the extraction of tax, necessarily limited
the degree to which private rent could be extended. It certainly made any
challenge to the dominance of the state at this period impossible.
The taxation policy of the state dominated the economy of the empire,
promoting some developments, while hindering or preventing others. In
particular, it meant the constant involvement in and attention of the state
upon rural relations of production and the possibilities for surplus extrac-
tion. So much is clear from the incidence of state legislation concerned with
safeguarding the sources of its revenue: it is surely no accident that the
major collections of state legislation dealing with land and the production
of wealth which control of land brought with it come from the sixth and
tenth centuries, two periods when major shifts in rural relations of pro-
duction were taking place (excluding, of course, the difficulties associated
with the lack of legislation in the later seventh century on these matters,
discussed in chapter 7 above) and when a potential or actual threat to the
tax-base was perceived.
Only when the state began to lose control over the means of production
and the rate of surplus appropriation (that is to say, when private rent
began to outstrip tax) and hence to obtain the income necessary to its
existence - in other words, only when the exaction of tribute or private rent
by the landowning class rather than by the state becomes dominant - can we
speak of a real shift within the balance of the tributary relations of production
in Byzantine society.23 But even at this stage, the entrenched ideology of
23
See Litavrin, Vizantiiskoe obsdestvo i gosudarstvo, pp. 289f.; A.P. Kazdan, Social'nii sostav,
pp. 253ff. Note also G. Ostrogorsky, 'Observations on the aristocracy in Byzantium', DOP
25 (1971), 12ff. The first clear, albeit tentative, institutional steps towards this process
can be seen in the eleventh century, with the development of the system of pronoia grants
and the temporary award of fiscal revenues by the state to private persons in return for
(military) service. But again, this is only one element in a complex picture. For the classic
458 Byzantium in the seventh century

the state, which found expression in the role of the emperor and the
existence of a centralised state bureaucracy, held the continued expansion
of privatised tributary relations - feudal relations in a sense akin to those
pertaining in the medieval West - back. For the Byzantine aristocracy was
committed to this ideology and political theory even after they had ceased to
represent their objective economic class interests. Like the state which it
represented, and yet for whose resources it competed, the aristocracy was
ideologically divided between serving the state and thus not promoting (or
actually damaging) its own interests as a group of independent land-
owners and landlords, on the one hand, and opposing the state - the
emperor - in order to protect and enhance its privileges and its power-base.
This compromise existed in its clearest form after the eleventh century. But
while the economic contradictions were gradually resolved in favour of the
magnates, the political contradiction was less apparent, a result of the
close ties between imperial political ideology, orthodoxy and the Church,
and the fact that this political system was focused upon the emperor, not
on a concept of 'the state' as such. The political and ideological support of
the elite, and the residual political authority of the emperors (and therefore
of the state) made possible the survival of the state as a parasitic political
form doomed to economic and therefore military and political extinction.
The seventh century thus saw the establishment of conditions which
made possible the survival of one form of the ancient state. For the
senatorial elite had been weakened sufficiently for the state to ignore it as a
class, while the new service elite (from which the middle Byzantine aristoc-
racy develops by the ninth century) was still in its infancy, dependent
entirely upon the state for its existence. Until the eleventh century at least,
the transformed structures of the ancient state provided the dominant struc-
tures of the Byzantine social formation. It is clear from our survey that a
fundamental transformation took place in the late Roman world, a transfor-
mation that made possible, and prolonged, the survival of the Roman state
in the East.

study of this development, see G. Ostrogorsky, Pour Vhistoire de la feodalite byzantine

(Brussels 1954). The question has been extensively discussed in the Soviet literature: see
the references in chapter 1, note 36.
 Addendum: Further observations on the
question of the late ancient city
(see Chapter 3)

This issue continues to receive substantial attention, but many problems

remain to be resolved or even addressed. Further archaeological work has
produced results which confirm the general pattern outlined here, but with
more differentiation between settlement types and functions. It has recently,
and quite correctly, been stressed that the evolution of the types of settle-
ment most commonly associated with the term kastron is in fact rather more
complex than has often been assumed. Indeed, the evidence suggests that
one should avoid overemphasising the contrast between the late ancient
polis and the middle Byzantine kastron: rather, of the large number of settled
sites which are differentiated from undefended rural settlements in form,
function and situation, only a small proportion bore the official or unofficial
characteristics of a polis; whereas a large number were characterised
already in the fourth and fifth centuries, and especially in the sixth century,
by features normally associated archaeologically and topographically with
the later Byzantine kastron. The examples upon which this conclusion is
based come from the Balkan context; but they are as valid, from the sixth and
seventh centuries, for Asia Minor. See A. Dunn, 'The transformation from
polis to kastron in the Balkans (III—VII cc): general and regional perspec-
tives', Byzantine andModern Greek Studies 18 (1994), pp. 60-80; J.F. Haldon,
'Quelques remarques sur l'economie byzantine de 600 a 1100. Esquisse
comparative', in R. Francovich, G. Noye, La Storia delVAlto Medioevo Italiano
(VI-X secolo) alia Luce dell'Archaeologia (Rome 1994) 71-84, esp. 75-6;
and T.E. Gregory, 'Kastro and diateichisma as responses to early Byzantine
frontier collapse', B 62 (1992) 235-53. In particular, it should be clear that
the transformations which occurred did not, except in a relatively small
number of cases, involve a universal abandonment of formerly urban sites
(poleis) in favour of hilltop fortified sites (kastra). Rather, it involved a change
in the way populations were distributed between such sites, and how they
were occupied. At the same time, it is possible to argue that many cities
did indeed continue until well into the first half of the seventh century to
459
460 Addendum

flourish as centres of local provincial society, but that the mode of socio-cultural
investment had changed. Churches rather than civic, secular public build-
ings seem to have attracted investment, for example: see J.-M. Spieser, 'Les
villes en Grece du IIP au VIP siecle', in Villes etpeuplement dans Tilly ricumpro-
tobyzantin (Rome 19 84) pp. 315-3 8. While certain major cities did decline -
as a result both of warfare and natural calamities (Antioch, for example, or
Apamaea: see J.C. Baity, Apamee au VP siecle. Temoignages archeologiques
de la richesse d'une ville', in Hommes et richesses I, pp. 79-96) - there is
plenty of incidental evidence for the continuity of provincial urban life (see
M. Whittow, 'Ruling the Late Roman and early Byzantine City: a Continuous
History', Past and Present 129 (Nov. 1990),pp. 3-29). What changed was the
emphasis on civic and corporate 'monumentality', a point well illustrated in
the work of Spieser. The important question of the extent to which cities may
have continued to exercise control over their territoria is taken up in F.R.
Trombley, 'Byzantine "Dark Ages" cities in comparative context', in To
EWTJVLKOV. Studies in Honor of Spews Vryonis, jr. I: Hellenic Antiquity and
Byzantium, eds. J.S. Langdon et al. (New York 1993) 429-49. But the pos-
sibilities for clarifying the situation in the sixth and seventh centuries are still
severely restricted by the patchy and limited archaeological data, in particu-
lar the still very partial and fragmentary ceramic record.
The archaeological material, albeit sparse, can nevertheless be used to
suggest a resolution to one aspect of the problem. In some Byzantine texts,
mostly hagiographical, there occur descriptions of 'cities' which can be
interpreted to mean that there remained a population inhabiting the 'lower'
town. As we have seen, it has been argued that either this means that the
whole ancient city area continued to be occupied; or that the text(s) in ques-
tion consists of topoi and that only a citadel is actually meant (see p. 109
above, and note 40). The preliminary results of excavations at Amorion and
several other sites show that while the very small fortress-citadel continued
to be defended and occupied, discreet areas within the late Roman walls also
continued to be inhabited, often centred around a church. In Amorion there
were at least two and probably three such areas (C. Lightfoot, in Anatolian
Studies 44 (1994) 105ff.). It seems probable that what these findings repre-
sent are small but distinct communities whose inhabitants regarded them-
selves (in one sense, that of domicile, quite legitimately) as 'citizens' of the
city within whose walls their settlement was located, and that the kastron,
which retained the name of the ancient polis, provided a refuge in case of
attack (whether or not it was permanently occupied, still less permanently
garrisoned). Many of the poleis of the seventh to ninth centuries thus may
have survived as such because their inhabitants, living effectively in distinct
'villages' within the area delineated by the walls, saw themselves as belong-
ing to the polis itself, rather than to a village, which may well have been
 On the question of the late ancient city 461

referred to by the name of its church or its older suburban quarter. Cf. the
example of Ephesos, which served as a refuge for the local rural population,
as a fortress and military administrative centre, but also retained its role as a
market town. Survey and excavation suggest that it was divided into three
small, distinct and separate occupied areas, including the citadel (Foss,
Ephesus After Antiquity, pp. 106-13); Sardis similarly shrank to a small forti-
fied acropolis, and one or more separate occupied areas within the circum-
ference of the original late ancient walls (Foss, Byzantine and Turkish Sardis,
pp. 5 5-61); Miletos was reduced to some 2 5 per cent of its original area, and
divided into two defended complexes (W. Miiller-Wiener, 'Das Theaterkastell
von Milet', Istanbuler Mitteilungen 17 (1967), pp. 279-90; C. Foss,
Archaeology and the "Twenty Cities" of Byzantine Asia', 469-86, at 477f.);
Didyma, close by Miletos, was reduced to a small defended structure based
around a converted pagan temple and an associated but unfortified settle-
ment nearby (Foss, Archaeology and the "Twenty Cities" of Byzantine Asia',
479 with literature). Examples can be multiplied. See the survey of Brandes,
Stddte, pp. 82-111,132ff. with further literature and sources.
For the wider perspective, see the essays in J. Rich, ed., The City in Late
Antiquity (London 1992) and in N. Christie, S.T. Loseby, eds., Towns in
Transition. Urban Evolution in Late Antiquity and the Early Middle Ages
(Aldershot 1996); and in Trade and Exchange in the Late Antique and Early
Islamic Near East (Studies in Late Antiquity and Early Islam 1, V), eds. L.A.
Conrad, G.R.D. King (Princeton forthcoming).
 Bibliography

Primary sources
Agathias ed. R. Keydell, Agathiae Myrinaei Historiarum Libri V (CFHB 2, Berlin
1967)
Agathon diaconus (account of events in 713), in Mansi XII, 189-96
Ambrosius (of Milan), Sermo de Basilicis Tradendis vel contra Auxentium, in PL XVI,
875-1286
Anastasius Monachus ed. F. Nau, 'Le texte grec des recits utiles & Tame d'Anas-
tase (le Sinaite)', OC 3 (1903), 56-90; see also OC 2 (1902), 58-89
Anastasius of Sinai lnterrogationes et Responsiones, in PG LXXXIX, 311-824
Sermo adversus Monotheletas, in PG LXXXIX, 1152-80; new edn by K.-H.
Uthemann, Anastasii Sinaitae Opera. Sermones duo in Constitutionem Hominis
secundum Imaginem Dei necnon Opuscula adversus Monotheletas (Corpus Christia-
norum Series Graeca XII. Brepols and Turnhout, 1985), pp. 55-83
Anonymi Chronicon ad Annum Domini 1234 Pertines, trans J.-B. Chabot (part 1), in
CSCO scriptores Syri, 56; trans. A. Abouna (part 2), in CSCO scriptores Syri, 154
Anonymi Peri strategias, ed. and trans. G.T. Dennis, in Three Byzantine Military
Treatises (CFHB XXV. Washington, D.C., 1985), pp. 1-135
Anonymus Vdri Incerti Scriptoris Byzantini saec. X. Liber de re Militari, ed. R. Vari
(Leipzig, 1901); new edn and trans. G.T. Dennis, Three Byzantine Military
Treatises (CFHB XXV. Washington, D.C. 1985), pp 247-327
Anthologia Palatina Anthologia Graeca, I-IV, ed. and trans. H. Beckby (Munich
1957-8, second edn 1970)
Appendix Eclogae eds. L. Burgmann and Sp. Troianos, 'Appendix Eclogae,' PM III
(1979). 24-125
Arculf, Relatio de locis sanctis, ed. T. Tobler, in Itinera et Descriptiones Terrae Sanctae I
(Geneva 1877), pp. 195ff.
Baladhuri Al-Baladhurf, Kitdb futuh al-Buldan. The Origins of the Islamic State,
trans. P.K. Hitti (London 1916 and Beirut 1966)
Basilica eds. H.-J. Scheltema, N. Van Der Wai and D. Holwerda, Basilicorum Libri
LX. A I-VII (Textus librorum MJX); B I-VIII (Scholia in libros I-LX, 16)
(Groningen, Jakarta and The Hague 1953-83)
BeSevliev, V. Spdtgriechische und spdtlateinische Inschriften aus Bulgarien (Berlin
1964)
Cedrenus Georgii Cedreni Compendium Historiarum, ed. I. Bekker 2 vols. (CSHB,
Bonn 1838-9)
462
 Bibliography 463

Chronicon Anonymi ad Annum Domini 813 Pertinens, ed. and trans. E.W. Brooks, in
CSCO scriptores Syri, ser. 3., vol. IV: Chronica Minora, 3, 1, 185-96
Chronicon Maroniticum, ed. E.W. Brooks, trans. J.-B. Chabot, in CSCO scriptores
Syri, ser. 3, vol. IV: Chronica Minora, 2, 3, 35-57
Chronicon Miscellaneum ad Annum Domini 724 Pertinens, ed. E.W. Brooks, trans.
J.-B. Chabot, in CSCO scriptores Syri, ser. 3, vol. IV: Chronica Minora, 2, 4,
63-119
Chronicon Miscellaneum ad Annum Domini 846 Pertinens, ed. E.W. Brooks, trans.
J.-B. Chabot, in CSCO scriptores Syri, ser. 3, vol. IV: Chronica Minora, 2, 5,
123-80
Chronicon Paschale, ed. L. Dindorf (CSHB, Bonn 1832)
Codex lustinianus in CICII
Codex Theodosianus Theodosiani Libri XVI cum Constitutionibus Sirmondianis, eds.
Th. Mommsen, P. Meyer et al. (Berlin 1905)
Concilium universale Constantinopolitanum tertium, ed. R. Riedinger, 2 vols. (Acta
Conciliorum Oecumenicorum H/2.1-2. Berlin 1990/1992)
Constans II, Typos in Lateran 208. 15-210. 15; Mansi X, 1029C-1032D
Constantine IV, Edict in Mansi XI, 697A-712D
Corpus lnscriptionum Graecarum, ed. A. Bockh (vols. I and II) and I. Franz (vols.
Ill-ff.) (Berlin 1828-77).
Corpus luris Civilis I, lnstitutiones, ed. P. Kruger; Digesta, ed. Th. Mommsen; II,
Codex lustinianus, ed. P. Kruger; III, Novellae, eds. R. Scholl and W. Kroll
(Berlin 1892-5; repr. Berlin 1945-63)
Corpus Scriptorum Ecclesiasticorum Latinorum (Vienna 1866-)
Cosmas Hierosolymitanus, Scholia in Gregorii Nazianzeni Carmina, in PG XXXVIII,
341-679
Cumont, F. 4Les Inscriptions chretiennes df Asie Mineure', in Melanges d'archeologie
etd'histoire 15 (1895)
De Administrando Imperio Constantine Porphyrogenitus, De Administrando lmperio I:
Greek Text, ed. Gy. Moravcsik, Engl. trans. R.J.H. Jenkins. New revised edn
(CFHBI = Dumbarton Oaks Texts I, Washington, D.C. 1967); II, Commentary,
ed. R.J.H. Jenkins (London 1962)
De Cerimoniis Constantini Porphyrogeniti De Cerimoniis Aulae Byzantinae Libri Duo,
ed. J.J. Reiske (CSHB, Bonn 1829-30)
De Thematibus Costantino Porfirogenito, De Thematibus, ed. A. Pertusi (Studi e Testi
160, Citta del Vaticano 1952)
Devreesse, Hypomnesticum R. Devreese, 'Le Texte grec de l'hypomnesticum de
Theodore Spoudee', AB 53 (1935), 66-80 (BHG32261)
DOC, vol. II Ph. Grierson, Catalogue of the Byzantine Coins in the Dumbarton Oaks
Collection and in the Whittemore Collection, vol. II: Phocas to Theodosius HI,
602-717 (Washington D.C. 1968)
Doctrina lacobi nuper Baptizati, ed. N. Bonwetsch, Abhandlungen der koniglichen
Gesellschaft der Wissenschaften zu Gottingen, phil.-hist. Klasse, XII, 3 (Berlin
1910)
Ecloga Ecloga. Das Gesetzbuch Leons 111 und Konstantinos' V., ed. L. Burgmann
(Forschungen zur byzantinischen Rechtsgeschichte X, Frankfurt am Main
1983)
Epanagoge ed. C.E. Zacharia von Lingenthal, in Collectio Librorum luris Graeco-
Romani Eclogam Leonis et Constantini, Epanagogen Basilii, Leonis et Alexandri
Continens (Leipzig 1852), 53-217; repr. in/GK II, 229-368
464 Bibliography

Eparchikon Biblion I. Duj6ev, Td 'Etrapxixdv BipKCov - The Book of the Prefect -

Le Livre du pHfet. Text, translation and commentary (London 1970)
Exc. de lnsid. Excerpta Historica lussu Imp. Constantini Porphyrogeniti Confecta III:
Excerpta de lnsidiis, ed. C. de Boor (Berlin 1905)
Exc. de Leg. Excerpta Historica lussu Imp. Constantini Porphyrogeniti Confecta I:
Excerpta de Legationibus, 2 parts, ed. C. de Boor (Berlin 1903)
Farmers' Law W. Ashburner, 'The Farmers' Law*, JHS 30 (1910), 85-108; JHS
32 (1912), 68-95.
Fiscal Treatise ed. F. Dolger, in Beitrdge zur Geschichte der byzantinischen Finanzver-
waltung besonders des 10. und 11. ]ahrhunderts (Byzantinisches Archiv IX
Munich 1927 and Hildesheim 1960)
Georgii Monachi Chronicon, 2. vols., ed. C. de Boor (Leipzig 1904)
George of Pisidia A. Pertusi, ed., Georgio di Pisidia, Poemi 1. Panegirici Epici (Studia
Patristica et Byzantina 7, Ettal 1959): Expeditio Persica, pp. 84-136; In Bonum
Patricium, pp. 163-70; Bellum Avaricum, pp. 176-200; Heraclias, pp. 240-61
Ghevond G. Chahnazarian, Ghevond: histoire des guerres et des conquetes des Arabes
en Armenie (Paris 1856)
Gregoire, H.t Recueils des inscriptions grecques chrttiennes d'Asie Mineure (Paris
1922)
Gregorii I Papae Registrum Epistolarum, eds. P. Ewald and L.M. Hartmann, in MGH,
Ep. 1 and 2 (second edn, Berlin 1957)
Grumel, V., Les Regestes des actes du patriarcat de Constantinople; I: Les Actes des
patriarches (Paris 1972); II (Chalcedon 1936); III (Chalcedon and Bucharest
1947)
Heraclius, Ekthesis in Mansi X, 991B-997A; Lateran 156, 20-162.13
Hierocles E. Honigmann, ed., Le Synekdemos d'Hierokles et Vopuscule geographique
de Georges de Chypre (Corpus Bruxellense Historiae Byzantinae I, Brussels
1939)
Hudud al-cAlam, The Regions of the World, trans. V. Minorsky (Oxford 1937)
Ibn Hawqal, Kitab Surat al-ard, Configuration de la terre, trans. J.H. Kramer, G. Wiet
(Beirut and Paris 1964) (text, ed. J. Kramer, (Leiden 1938))
Inscriptions grecques et Mines de la Syrie, eds. L. Jalabert, R. Mouterde et al. (Paris
1929-70)
Inscriptions grecques et Mines de la Syrie, ed. W.H. Waddington (Paris 1870 and
Rome 1969) (= Inscriptions grecques et Mines recueillies en Grece et en Asie
Mineure, III, 1/2 (Paris 1870 and 1876))
Isidore of Seville Chronicon Isidori luniohs, in MGH {AA) XI, 2, 424-81
Isidori lunioris Episcopi Hispalensis Historia Gothorum Wandalorum Sueborum ad an.
DCXXIV, in MGH (AA) XI, 2, 267-303
Continuations Isidorianae Byzantina Arabica et Hispana, in MGH (AA) XI, 2,
334-68
lus Graeco-romanum, eds. J. and P. Zepos, 8 vols. (Athens 1931 and Aalen 1962)
John of Antioch loannis Antiocheni Fragmenta, in FHG V, 27-38; Exc. de lnsid.,
58-150
John of Biclar loannis Abbatis Monasterii Biclarensis Chronicon, in MGH (AA) XI, 2,
211-20
John of Damascus Contra Imaginum Calumniatores Orationes tres, in B. Kotter, ed.,
Die Schriften des Johannes von Damaskos III (Berlin 1975)
Fount of Knowledge Ui\yx] Tvwaecos, part 2: De Haeresibus, in PG XCIV,
677-780
 Bibliography 465

John of Ephesus The Third Part of the Ecclesiastical History of John of Ephesus, ed.
and trans. R. Payne-Smith (Oxford 1860)
Lives of the Eastern Saints, ed. and trans. E.W. Brooks, in PO 17-19 (Paris
1923-5)
John of Nikiu The Chronicle of John, Bishop ofNikiu, ed. and trans. R.H. Charles
(London 1916)
Julian, Oratio I in Julian, Letters and Works, ed. and trans. W.C. Wright (London
and Cambridge, Mass. 1954)
Justinian, Digesta in CICI
Justinian, Edict. Edicta, in CIC III, 759-95
Justinian, Nov. Novellae, in CIC III
Kekaumenos Soveti i rasskazi Kekavmena: socinenie vizantijskogo polkovodtsa XI
veka, ed., trans, and comm. G.G. Litavrin (Moscow 1972); older edn B. Wassi-
liewsky and V. Jernstedt, Cecaumeni Strategicon et incerti scriptoris de offlciis
regiis libellus (St Petersburg 1896 and Amsterdam 1965)
Kiel Phil. Kletorologion of Philotheos, in Oikonomides, Preseance, pp. 81-235
Konidares, J., 'Die Novellen des Kaisers Herakleios', FM V (1980), 33-106 (text
62-95)
Konstantopoulos, K., Bv(avTiaxa fjLo\v(386(3ovA\a TOV ev 'A'd'nvals 'E-dvixov
Nofiiafianxov MovaeCov (Athens, 1917) (= JIAN 5 (1902), 149-64,
189-228; 6 (1903), 49-88, 333-64; 7 (1904), 161-76, 255-310; 8
(1905), 53-102, 195-222; 9 (1906), 61-146; 10 (1907), 47-112
John Mauropous P. de Lagarde and J. Bohlig, eds, Johannis Euchaitarum Metropoli-
tae quae Supersunt in Cod. Vaticano Graeco 676 (Abhandlungen der Gottinger
Gesellschaft der Wissenschaften 28, Berlin 1882)
Lateran Concilium Lateranense a. 649 Celebratum, ed. R. Riedinger (Acta concilio-
rum oecumenicorum, ser. 2, vol. I, Berlin 1984)
Laurent, V., 'Bulletin de sigillographie byzantine I', B 5 (1929-30), 571-654
Le Corpus des sceaux de Vempire byzantin, vol. II: VAdministration centrale (Paris
1981)
Documents de sigillographie. La collection C. Orghidan (Bibliotheque byzantine,
Documents, I, Paris 1952)
Les Sceaux byzantins du mtdailler Vatican (Medagliere della Biblioteca Vaticana I,
Citta del Vaticano 1962)
Leges Militares ed. W. Ashburner, 'The Byzantine mutiny act', JHS 46 (1926),
80-109 (repr. in JGR (Zepos) II, 75-9) (earlier redaction); ed. E. Korzensky,
'Leges poenales militares e codice Laurentiano LXXV, Egypetemes Philologiae
Kozlony (Budapest 1930), 155-63, 215-18 (repr. in JGR (Zepos) II, 80-9)
(later redaction)
Leo VI, Nov. P. Noailles and A. Dain, Les Novelles de Lion VI le sage: texte et
traduction (Paris 1944)
Leo, Tactica Leonis imperatoris tactica, in PG CVII 672-1120; ed. R. Vari, Leonis
Imperatoris Tactica I (proem., const. I-XI); II (const. XII-XIII, XIV, 1-38) (Sylloge
Tacticorum Graecorum III, Budapest 1917-22)
Liber Pontiflcalis, ed. L. Duchesne, 2 vols. (Paris 1884-92)
Lydus loannis Lydi De Maglstratibus Populi Romani Libri Tres, ed. R. Wiinsch
(Leipzig 1903)
Malalas ed. L. Dindorf, loannis Malalae Chronographia (CSHB, Bonn 1831); Engl.
translation: The Chronicle of John Malalas. A Translation, by E. Jeffreys,
M. Jeffreys, R. Scott et al. (Byzantina Australiensia IV, Melbourne 1986)
466 Bibliography

Mansi J.D. Mansi, ed., Sacrorum Conciliorum Nova et Amplissima Collectio

(Florence, 1759-1927).
Martin, Pope S. Martini Papae Commemoratio, in Mansi X, 853-61; and PL 129,
591-600
Epistolae, in Mansi X, 682-853
Encyclical letter, in Mansi X, 1169-84
Maurice, Strategikon Das Strategikon des Maurikios, ed. G.T. Dennis, trans. E. Gam-
illscheg (CFHB XVII, Vienna 1981)
Maximus Confessor Maximi Confessoris Relatio Motionis, in PG XC, 109-29
Gesta in Primo Eius Exsilio, in PG XC, 135-72
Disputatio S. Maximi cum Pyrrho, in Mansi X, 709-60
Maximi Confessoris Epistolae, in PG XCI, 364-649
Menander Protector Menandri Protectoris Fragmenta, in Exc. de Leg. I, 170-221;
II, 442-77
Michael Syr. La Chronique de Michel le Syrien, Patriarche Jacobite d'Antioche (4
vols., Paris 1899, 1901, 1905, 1924)
Miklosich, F. and Muller, ]., Acta et Diplomats Graeca Medii Aevi Sacra et Profana I, II
(Acta Patriarchatus Constantinopolitani) (Vienna 1860-2)
Miracula S. Artemii, in A. Papadopoulos-Kerameus, Varia Graeca Sacra (St Peters-
burg 1909), 1-75 (BHG3173)
Miracula S. Demetrii P. Lemerle, ed., Les plus anciens recueils des miracles de
S. Demetrius I: Le Texte (Paris 1979); II: Commentaire (Paris 1981)
Miracula Therapontis in L. Deubner, De Incubatione Capita Quattuor (Leipzig 1900)
Monumenta Asiae Minoris Antiqua, eds. W.M. Calder, J. Keil et al. (Manchester
1928-62)
Nicephorus Breviarium, in Nicephori Archiepiscopi Constantinopolitani Opuscula His-
toricat ed. C. de Boor (Leipzig 1880), 1-77
Nicephorus, Chron. Chronographikon syntomon, in Nicephori Archiepiscopi Con-
stantinopolitani Opuscula Historicat ed. C. de Boor (Leipzig 1880), 79-135
Nicephorus, Patriarch of Constantinople. Short History, ed. and trans. C. Mango
(Washington D.C., 1990)
Notitia Dignitatuum Utriusque Imperil ed. 0. Seeck (Leipzig 1876)
Notitiae Episcopatuum Ecclesiae Constantinopolitanae, ed. J. Darrouzes (Paris 1981)
Notitia Ps.-Epiphanii ed. H. Gelzer, Ungedruckte und ungenugend verqffentlichte Texte
der Notitiae Episcopatuum. Bin Beitrag zur byzantinischen Kirchen- und Ver-
waltungs-geschichte (Abhandlungen der bayerischen Akademie der Wissens-
chaften, phil.-hist. Klasse, XXI 3, Munich 1900)
Optatus Milevitanus, Contra Parmenianum Donatistam, in CSEL XXVI (1893)
Parastaseis syntomoi chronikai, ed. Th. Preger, in Scriptores Originum Constantinopoli-
tanarum (2 vols., Leipzig 1901,1907; repr. New York 1975), vol. I, pp. 19-73
see Averil Cameron and Judith Herrin, Constantinople in the Eighth Century
(Leiden 1984)
Paul the Deacon Pauli Diaconi Historia Langobardorurn, ed. G. Waitz, in MGH
(Scriptores Rerum Langobard. et Ital. Saec. VI-IX), 45-187
Peira Ileipa sive Practica ex Actis Eustathii Romani, in JGRIV, 1-260
Pelagius II Papa Epistolae et Decreta, in PL LXXII, 703-60
Peter of Atroa, Life La Vie merveilleuse de S. Pierre d'Atroa, ed. V. Laurent (Subsidia
Hagiographica 29, Brussels 1956) (BHG32364)
Popescu, E., Inscriptiile Grecesti si Latine din Secolele IV-XIll descoperite in Romania
(Bucharest 1976)
 Bibliography 467

Priscus of Panium, Fragments in FHGIV, 71-110

Procheiros Nomos 0 Ilpdxeipos Ndfxog. Imperatorum Basilii, Constantini et Leonis
Prochiron, ed. C.E. Zacharia von Lingenthal (Heidelberg 1837) (partially repr.
in JGRU, 107-228)
Procopius Procopii Caesariensis Opera Omnia, ed. J. Haury (3 vols., Leipzig
1905-13; revised edn with corrections and additions by G. Wirth, 4 vols.,
Leipzig 1962-4)
Procopius, De Bello Gothico. in Procopii Caesariensis Opera Omnia II
De Bello Persico, in Procopii Caesariensis Opera Omnia I
De Bello Vandalico, in Procopii Caesariensis Opera Omnia I
Historia Arcana (Secret History), in Procopii Caesariensis Opera Omnia III
Ps.-Athanasius, Quaestiones ad Antiochum ducem, in PG XXVIII, 597-708
Pseudo-Denys de Tell-Mahrt, Chronique, ed. J.B. Chabot (Paris 1895)
Ps.-Methodius, Apokalypse A. Lolos, ed., Die Apokalypse des Ps.-Methodios (Beitrage
zur klassischen Philologie LXXXIII, Meisenheim am Glan 1976)
Radermacher, L. Griechische Quellen zur Faustsage (Sitzungsberichte der Akademie
der Wissenschaften in Wien, phil.-hist. Klasse CCVI, 4, Vienna and Leipzig
1927)
Regesta Pontificum Romanorum ab Condita Ecclesia ad Annum MCXCVlll eds. P. Jaffe,
revised by P. Ewald, F. Kaltenbrunner, S. Lowenfeld et al. (2 vols., Leipzig
1885 and 1888)
Rhalles, K. and Potles, M. Syntagma (Zvvrayfia T&V tieCav xai ieptiv xavdvtov)
(4 vols., Athens 1852-9)
Rhodian Sea Law ed. W. Ashburner, Ndtws Poduav Navrixds. The Rhodian
Sea-Law (Oxford 1909 and Aalen 1976) (repr. in JGR II, 91-103)
Schreiner, P., ed., Die byzantinischen Kleinchroniken (2 vols., CFHB XII, 1 and 2,
Vienna, 1975 and 1977)
Scylitzes loannis Scylitzae Synopsis Historiarum, ed.). Thurn (CFHB V, Berlin and
New York 1973)
Sebeos F. Macler, trans., Sebeos, Histoire d'Heraclius (Paris 1904)
Seibt, W., Die byzantinischen Bleisiegel in Osterreich I: Kalserhof (Vienna 1978)
Strabo, Geographies ed. H.L. Jones (8 vols., London and New York 1917-32)
Supplementum Epigraphicum Graecum, eds. J.J.E. Hondius and A.G. Woodhead
(Leiden 1927-71, 1976-7)
Taktikon Uspenskij ed. N. Oikonomides, in Preseance, pp. 47-63
Tabari ed. M.J. De Goeje, Annales quos Scripsit Abu Djafar Mohammed Ibn Djarir al
Tabari cum Aliis (3 vols., Leiden 1879)
Themistius Orationest ed. W. Dindorf (Leipzig 1832)
Theodore of Studion S. Theodori Studitae Epistolarum Lib. I, in PG XCIX,
904-1116
Theodori Studitae Epistulae, ed. G. Fatouros (CFHB XXXI1/2. Berlin 1992)
Theodore Sygkellos, Tlepi rqs r&v d&ewp Bappdpwv xai Ilepawv... xivrjcrews xai
dvaxo>prj(re(o<;, ed. L. Sternbach, in Analecta Avarica (Rozprawy Akademii
Umiejetnosci Wydzial Filologisczny, ser. 2, vol. XV, Cracow 1900), pp. 2-24 ( =
L. Sternbach, Studia Philologica in Georgium Pisidam (Cracow, 1900),
pp. 297-334)
Theophanes Theophanis Chronographia, ed. C. de Boor (2 vols., Leipzig 1883-5)
Theophanes cont. Theophanes continuatus, Ioannes Caminiata, Symeon Magis-
ter, Georgius Monachus continujatus, ed. I. Bekker (CSHB, Bonn 1825),
pp. 1-481.
468 Bibliography

Theophylact Simocatta Theophylacti Simocattae Historic ed. C. de Boor (Leipzig

1887; ed. and corr. P. Wirth, Stuttgart 1972)
Timotheus Constantinopolitanus, De Receptione Haereticorum, in PG LXXXVI,
11-68
Taktikon Uspenskij in Oikonomides, Prisiance, pp. 43-63
Vie de S. Michel Maleinos, ed. L Petit, in ROC VII (1902), 543-603 (BHG31295)
Vita Alypii Stylita ed. H. Delehaye, in Les Saints stylites (Subsidia Hagiographica,
XIV, Brussels, 1923), 170-87 (BHG31295)
Vita Andreae Cretensis ed. A. Papadopoulos-Kerameus, in 'Avd\exra
'lepoaoAvfunxris Iraxvkoyia^ V (St Petersburg 1891), 169-80, 422-4
(BHG3113)
Vita Andreae Sali in PG CXI, 621-888 (BHG3117)
Vita Antonii lunioris ed. A. Papadopoulos-Kerameus, in Pravoslav. Palestinskii
Sbornik, 19/3 (57) (St Petersburg 1907), 186-216; suppl. F. Halkin, in AB 62
(1944), 187-225 (BHG3142)
Vita Blasii Amoriensis in AS Nov. IV, 657-69 (BHG3278)
Vita Danielis Stylitae ed. H. Delehaye, in AB 32 (1913), 121-214; repr. in H.
Delehaye, Les Saints stylites (Subsidia Hagiographica XIV, Brussels 1923),
1-94
Vita Eustratii ed. A. Papadopoulos-Kerameus, in 'Avdkexra 'IepoaoXv/junxrjs
iraxvXoyCas, TV (St Petersburg 1891-8), 367-400 (BHG3645)
Vita Euthymii lunioris ed. L. Petit, 'Vie et office de S. Euthyme le jeune', ROC VIII
(1903), 155-205; 503-6 (repr. L. Clugnet in Bibl. Hag. Or., 5 (1904),
14-51) (BHG3655)
Vita Eutychii Archiepiscopi Constantinopolitani, in PG LXXXVI/2,
2273-390 (BHG3657)
Vita Evaresti ed. Ch. van de Vorst, 'La Vie de S. Evariste, higoumene a Constantin-
ople', AB 41 (1923), 288-325 (BHG32153)
Vita Gregorii Agrigentini in PG, XCII, 549-716 {BHG3707\ see Beck, Kirche,
pp. 466f.)
Vita S. loannidi (aPetro) in ASNov. II/l, 3 8 4 ^ 3 4 (BHG3936)
Vita Johannis Eleemosyn. ed. H. Gelzer, Leontios' von Neapolis Leben des heiligen
Johannes des barmherzigen Erzbischofs von Alexandrien (Sammlung ausgewahlter
kirchen- und dogmengeschichtlicher Quellenschriften V Freiburg and Leipzig
1893), 1-103 (BHG3886)
Vita loannis Psichaita ed. P. Van den Ven, in Le Muston, new series 3 (1902),
103-25 (BHG3896)
Vita Maximi Confessoris in PG XC, 68-109 (BHG31234); Syriac Life, ed.
S. Brock, 'An early Syriac Life of Maximus the Confessor', AB 91 (1973),
299-346
Vita Michaelis Synnadensis, ed. C. Doukakis, in Meyas ovvagapurrrj? vdvrojv ra>v
ayC<av (12 vols., Athens, 1889-96), Mail, 411-22 (BHG32274x)
Vita Nicephori Medicii ed. F. Halkin, 'La Vie de S. Nicgphore, fondateur du Medi-
cion en Bithynie (mourut 813)', AB 78 (1960), 396-430 (BHG32297)
Vita Philareti eds. M.H. Fourmy and M. Leroy, 'La Vie de S. Philarete', B 9 (1934),
85-170 (BHG31511z)
Vita Stephani lunioris in PGC, 1069-186 (BHG31666)
Vita Tarasii ed. J.A. Heikel, 'Ignatii Diaconi Vita Tarasii Archiepiscopi Constanti-
nopolitani', Acta Societatis Scientiarum Fennicae 17 (1889), 395-423
(BHG31698)
 Bibliography 469

Vita Theodori Syceotis ed. A. Festugiere, Vie de Theodore de Syke'on (2 vols., Subsidia
Hagiographica IIL, Brussels 1970) (BHG31748)
Vita S. Theodori Studitae ed. A. Mai, in NPB VI, 291-363 (BHG31754)
Vita et Mracula Theodori (tironis) ed. H. Delehaye, in Les Ugendes grecques des
saints militaires (Paris 1909), 183-201 (BHG31764)
Zacos, G. and Veglery, A., Byzantine Lead Seals, vol. I, parts 1-3 (Basel 1972)
Zonaras loannis Zonarae Epitomae Historiarum Libri Xlll usque ad XVHl, ed. Th.
Buttner-Wobst (CSHB, Bonn 1897)
Zos. (Zosimus) Zosimi Comitis et Exadvocati Fisd Historia Nova, ed. L. Mendelssohn
(Leipzig 1887 and Hildesheim 1963)

(B) SELECTED SECONDARY LITERATURE

Abrahamse, D. de F. Hagiographic Sources for Byzantine Cities (Ann Arbor, Michigan
1967)
Ahrweiler, H. 'L'Asie mineure et les invasions arabes', RH 227 (1962), 1-32
'Recherches sur l'administration de l'empire byzantin aux IXe-XIe siecles', BCH
84 (1960), 1-109
Alexander, P.J. 'Medieval apocalypses as historical sources', American Historical
Review 23 (1968), 997-1018 (repr. in P.J. Alexander, Religious and Political
History and Thought in the Byzantine Empire XIII (London 1978))
'The strength of empire and capital as seen through Byzantine eyes', Speculum 37
(1962), 339-57
Anderson, P. Passages from Antiquity to Feudalism (London 1974)
Antoniadis-Bibicou, H. Etudes d'histoire maritime a Byzance, a propos du Theme des
Caravisiens (Paris 1966)
Recherches sur les douanes d Byzance, V 'octava', le 'kommerkion'et les commerciaires
(Paris 1963)
Arnheim, M.T.W. The Senatorial Aristocracy in the Later Roman Empire (Oxford
1972)
Auz6py, M.-F. 'L'6volution de 1'attitude face au miracle a Byzance (VIP-IXe siecle)', in
Miracles, prodiges et merveUles au Moyen Age (Paris 1995), pp. 31-46
Avenarius, A. Die Awaren in Europa (Bratislava 1974)
Baity, J.C. Apamee au VIe siecle. Temoignages archeologiques de la richesse d'une
ville', in Hommes et richesses I, pp. 79-96
Barker, E. Social and Political Thought in Byzantium (Oxford 1957)
Beck, H.-G. Kirche und theologische Literatur im byzantinischen Reich (Handbuch
der Altertumswissenschaft XII, 2.1 = Byzantinisches Handbuch 2.1, Munich
1959)
Senat und Volk von Konstantinopel. Probleme der byzantinischen Verfassungsge-
schichte (Sitzungsberichte der bayerischen Akad. d. Wissenschaften, phil.-hist.
Kl. (1966), Heft 6, 1-75). Repr. in H.-G. Beck, Ideen und Realitdten in Byzanz
XII (London 1972)
Bertolini, 0. 'Riflessi politic! dell'controversie religiose con Bisanzio nelle vicende del
sec. VIII in Italia', in Caraterri del secolo Vlll in Occidente (Settimane di Studio
del Centro Italiano di Studi su 11'alto Medioevo V, Spoleto 1958), pp.
733-89
Boak, A.E.R. The Master of Offices in the Later Roman and Byzantine Empire (New
York 1919), in A.E.R. Boak and J.E. Dunlop, Two Studies in Late Roman and
Byzantine Administration (New York 1924)
470 Bibliography

Brandes, W. 'Die byzantinische Stadt Kleinasiens im 7. und 8. Jahrhundert - ein

Forschungsbericht', Klio 70 (1988), 176-208
Die Stadte Kleinasiens im 7. und 8. Jahrhundert (BBA LVI Berlin 1989)
Brandes, W., Winkelmann, E, eds., QueUen zur Geschichte desfruhen Byzanz (BBA LV.
Berlin, 1990).
Brett, M. 'The Arab conquest and the rise of Islam', in The Cambridge History of
Africa, vol. II (Cambridge 1978), pp. 490-555
Breckenridge, J.D. The Numismatic Iconography of Justinian II (A.D. 655-95 and
705-711) (New York 1959)
Br6hier, L. in A. Fliche and V. Martin, eds., Histoire de Vtglise depuis les origines
jusqu'd nos fours V: Grigoire le grand, les itats barbares et la conquete arabe
(590-757) (Paris 1938)
Brooks, E.W. 'Byzantines and Arabs in the times of the early Abbasids', EHR 15
(1900), 728-47
'The Arabs in Asia Minor (641-750) from Arabic sources', JHS 18 (1898),
182-208
Brown, P. The Body and Society. Men, Women and Sexual Renunciation in Early
Christianity (London and Boston 1988)
'A Dark-Age crisis: aspects of the iconoclastic controversy', EHR 88 (1973),
1-34
Religion and Society in the Age ofSt Augustine (London 1972)
'Sorcery, demons and the rise of Christianity: from late Antiquity into the Middle
Ages', in Witchcraft Confessions and Accusations (Cambridge 1970), pp. 17-45
'The rise and function of the holy man in late Antiquity', JRS 61 (1971), 80-101
The World of Late Antiquity (London 1971)
Brown, T.S. Gentlemen and Officers. Imperial Administration and Aristocratic Power in
Byzantine Italy, A.D. 554-800 (Rome 1984)
'The Church of Ravenna and the imperial administration in the seventh century',
EHR 94 (1979), 1-28
'The interplay between Roman and Byzantine traditions and local sentiment in
the Exarchate of Ravenna', in Bisanzio, Roma e IItalia nellalto Medioevo (Setti-
mane di Studio del Centro Italiano di Studi sulFalto Medioevo XXIV, Spoleto
1988), pp. 127-60
Bury, J.B. A History of the Eastern Roman Empire from the Fall of Irene to the Accession
of Basil I (802-67) London 1912)
The Imperial Administrative System in the Ninth Century, with a revised text of the
Kletorologion ofPhilotheos (British Academy Suppl. Papers, I, London 1911)
Byzanz im 7. Jahrhundert: see Winkelmann et al.
Cambridge Medieval history, vol. IV, parts 1 and 2, ed. J. Hussey (Cambridge
1966-7)
Cameron, Averil 'Byzantine Africa - the literary evidence', in University of Michi-
gan Excavations at Carthage VII (1982), pp. 29-62
'Images of authority: elites and icons in late sixth-century Byzantium', PP 84
(1979), 3-35
Procopius and the Sixth Century (London 1985)
'The Theotokos in sixth-century Constantinople: a city finds its symbol', JThS 29
(1978), 79-108
'Byzantines and Jews: some recent work on early Byzantium', BMGS 20 (1996)
249-74
The Mediterranean Worldin Late Antiquity AD 395-600 (London 1993)
Changing Cultures in Early Byzantium (Aldershot 1996)
 Bibliography 471

Cameron, Averil, Conrad, L, eds., The Byzantine andEarly Islamic Near Eastl: Problems
in the Literary Source Material (Studies in Late Antiquity and Early Islam 1,1.
Princeton 1992)
Charanis, P. 'Ethnic changes in the Byzantine Empire in the seventh century', DOP
13 (1959), 23-44
'Observations on the demography of the Byzantine empire', in Xlll International
Congress of Byzantine Studies, Main Papers XIV (Oxford 1966), pp. 1-19
Studies on the Demography of the Byzantine Empire (London 1972)
'The monk as an element of Byzantine society', DOP 25 (1971), 61-84
'The transfer of population as a policy in the Byzantine Empire', Comparative
Studies in Society and History 3 (The Hague 1961), 140-54
Christie, N., Loseby, S.T., eds., Towns in Transition. Urban Evolution in Late Antiquity and
the Early Middle Ages (Aldershot 1996)
Claude, D. Die byzantinische Stadt im 6. Jahrhundert (Byzantinisches Archiv, XIII,
Munich 1969)
Conrad, L.I. 'Theophanes and the Arabic Historical Tradition: Some Indications of
Intercultural Transmission', BF15 (1990), 1-44
'Epidemic disease in central Syria in the late sixth century. Some new insights from
the verse of Hassan ibnThabit', BMGS18 (1994) 12-58
Cormack, R. 'The arts during the age of iconoclasm', in Iconoclasm, eds. A.A.M.
Bryer and J. Herrin (Birmingham 1977), pp. 35-44
Writing in Gold (London 1985)
Cruikshank-Dodd, E. Byzantine Silver Stamps (Dumbarton Oaks Studies VII, Wash-
ington D.C. 1961)
Dagron, G. 'Entre village et cite: la bourgade rurale des IVe-VHe stecles en Orient',
Koinonia 3 (1979), 29-52 (repr. in G. Dagron, La Romanite chretienne en Orient
HI)
La Romanite Chrtiienne en Orient. Heritages et mutations (London 1984)
'Le Saint, le savant, l'astrologue. Etude de themes hagiographiques a travers
quelques recueils de "Questions et reponses" des Ve-VIIe siecles', in Hagiogra-
phie, cultures et societe (1V-V1T siecles). Etudes Augustiniennes (Paris 1981),
pp. 143-55 (repr. in G. Dagron, La Romanite chretienne en Orient IV)
Dagron, G. 'Juda&er', TM11 (1991) 359-380
Dagron, G, Deroche, V. 'Juifs et Chretiens dans l'Orient du VIP siecle', TM 11 (1991)
17-273
Deroche, V. 'La polemique anti-Judai'que au VIe et au VHe siecle. Un memento inedit,
\esKephalaia\TM 11 (1991) 275-311
Diehl, Ch. L'Afrique byzantine: histoire de la domination byzantine en Afrique
(533-709) (2 vols., Paris 1896)
Etudes byzantines (Paris 1905)
'Le Senat et !e peuple byzantin aux VIIC et VIIIe socles', B 1 (1924), 201-13
Ditten, H. Ethnische Verschiebungen zwischen der Balkanhalbinsel und Kleinasien vom
Endedes6. biszurzweitenHdlftedes9.Jahrhunderts(BBAUX.Berhn 1993)
Ditten, H. 'Zur Bedeutung der Eihwanderung der Slawen', in Byzanz im 7. Jahrhun-
dert, eds. F. Winkelmann et al., pp. 73-160
Dolger, F. Beitrdge zur Geschichte der byzantinischen Finanzverwaltung bes. des 10.
und 11. Jahrhunderts (Byzantinisches Archiv IX, Munich 1927 and Hildesheim
1960)
Regesten der Kaiserurkunden des ostromischen Reiches von 565-1025, parts 1-3
(Munich and Berlin 1924, 1925 and 1932)
Duncan-Jones, R. Structure and Scale in the Roman Economy (Cambridge 1990)
472 Bibliography

Dunn, A. 'The transformationfrompolis to kastron in the Balkans (III-Vn cc): general

and regional perspectives', BMGS18 (1994) 60-80
'The Kommerkiarios, the Apotheke, the Dromos, the Vardarios, and The West', BMGS
17(1993)3-24
Durliat, J. Les Dtdicaces d'ouvrages de defense dans I'Afrique byzantine (Collection de
l'ecole francaise de Rome, Rome 1981)
'Les Finances municipales africaines de Constantin aux Aghlabides', in Bulletin
archeologique du comite des travaux historiques et scientiflques, new series, 19
(1983), 377-86
'Les Grands proprietaires africains et l'&at byzantin', Cahiers de Tunisie XXIX
(1981), 517-31
De la ville antique a la ville byzantine. Leprobleme des subsistences (Collection de l'ecole
francaise de Rome 136. Rome 1990)
Dvornik, F. Early Christian and Byzantine Political Philosophy, 2 vols. (Washington
D.C. 1966)
Ebersolt, J. Musees impfriaux ottomans. Catalogues des sceaux byzantins (Paris 1922)
'Sceaux byzantins du Musee de Constantinople', RN, 4 e ser., 18 (1914), 207-43,
377-409
Fogen, M.-Th. 'Gesetz und Gesetzgebung in Byzanz. Versuch einer Strukturana-
lyse', in lus Commune. Zeitschrift fur europdische Rechtsgeschichte 14 (1987),
137-58
Foss, C. 'Archaeology and the "Twenty Cities" of Byzantine Asia', American Journal
of Archaeology 81 (1977), 469-86
Byzantine and Turkish Sardis (Cambridge, Mass, and London 1976)
Ephesus after Antiquity: A Late Antique, Byzantine and Turkish City (Cambridge
1979)
'Late antique and Byzantine Ankara', DOP 31 (1977), 29-87
'The Persians in Asia Minor and the end of antiquity', EHR 90 (1975), 721-43
Frend, W.H.C. The Donatist Church: A Movement of Protest in Roman North Africa
(Oxford 1952)
The Rise of the Monophysite Movement (Cambridge 1972)
Gabrieli, F. Muhammad and the Conquests of Islam (London 1968)
Garsoian, N. 'Les Sources grecques pour l'histoire des Pauliciens d'Asie Mineure:
texte critique et traduction', TM 4 (1970), 1-227
Gelzer, H. Die Genesis der byzantinischen Themenverfassung (Abhandlungen der
koniglichen sachsischen Gesellschaft der Wissenschaften, phil-hist. Klasse,
Leipzig 1897 and Amsterdam 1966)
Goflart, W. 'Caput' and Colonate: Towards a History of Late Roman Taxation
(Phoenix suppl. vols., 12, Toronto, 1974)
Goodchild, R. 'Byzantines, Berbers and Arabs in seventh-century Libya', Antiquity
41 (1967), 114-24
Goubert, P. Byzance avant Vlslam. Byzance et VOrient sous les successeurs de Justinien
I: L'Empereur Maurice; II: Byzance et les Francs; II2: Rome, Byzance et Carthage
(Paris 1951, 1956 and 1965)
Gouillard, J. 'L'Her&ie dans l'empire byzantin des origines au XIIC siecle*. TM 1
(1965), 299-324
Grabar, A. VEmpereur dans Vart byzantin. Recherches sur Vart officiel de Vempire
d'Orient (Paris 1936 and London 1971)
Gray Birch, W. de. Catalogue of Seals in the Department of Manuscripts in the British
Museum V: Byzantine Empire (London 1898)
 Bibliography 473

Great Britain, Naval Intelligence Division, Geographical Handbook series. Greece, 3 vols.
(London 1944-5)
Naval Intelligence Division, Geographical Handbook series. Turkey, 2 vols. (London
1942-3)
Gregory, T.E. 'Kastro and diateichisma as responses to early Byzantine frontier col-
lapse', B62 (1992)235-53
Grousset, R. Histoire de VArmMe des origines a 1071 (Paris 1947)
Grumel, V. 'Recherches sur l'histoire du Monothelisme', EO 27 (1928), 6-16,
257-77; 28 (1929), 272-83; 29 (1930), 16-28
Guilland, R. 'Les Patrices byzantins de la premiere moitie du VIIe siecle', Tdfios eis
livrjiirivK.l. A^iavrov (Athens 1960), pp. 11-24 (repr. in R. Guilland, Recher-
ches sur les institutions byzantins, pp. 162-9)
'Patrices de Constantin IV a Theodose III', 'E\\t)vixd 23 (1970), 287-98
(repr. in R. Guilland, Titres etfonctions VIII)
Recherches sur les institutions by2antins, 2 vols. (Berlin and Amsterdam
1967)
Titres etfonctions de Vempire byzantin (London 1976)
Guillou, A. La Civilisation byzantine (Paris 1974)
Rdgionalisme et indtpendance dans Vempire byzantin au VIF siecle: Vexemple de
Vexarchat et de la pentapole d'ltalie (Istituto Storico Italiano per il Medio Evo,
Studi Storici, LXXV-LXXVI, Rome 1969)
'Transformation des structures socio-economiques dans le monde byzantin du
VIe au VIHe siecle', ZRVI19 (1980), 71-8
Haldon, J.F. Byzantine Praetorians: An Administrative, Institutional and Social Survey
of the Opsikion and Tagmata, c. 580-900 (Poikila Byzantina III, Bonn and Berlin
1984)
'Comes horreorum - Komes tes Lamias?', BMGS 10 (1986), 203-9
Ideology and social change in the seventh century: military discontent as a
barometer', Klio 68 (1986), 139-90
'"Jargon" vs. "the Facts"? Byzantine history-writing and contemporary
debates', BMGS 9 (1984-5), 95-132
Recruitment and Conscription in the Byzantine Army c. 550-950: A Study of the
Origins of the Stratiotika Ktemata (Sitzungberichte der osterreichischen Akade-
mie der Wissenshaften, phil.-hist. Klasse CCCLVII, Vienna 1979)
'Some considerations on Byzantine society and economy in the seventh century',
BF 10 (1985), 75-112
'Some Remarks on the background to the iconoclast controversy', BS 38 (1977),
161-84
'The Writings of Anastasius of Sinai: a key source for seventh-century East
Mediterranean history', in: Cameron, Conrad, eds., The Byzantine and Early
Islamic Near East I, pp. 107-47
The State and the TributaryTMode of Production (London 1993)
'Quelques remarques sur reconomie byzantine de 600 a 1100. Esquisse compara-
tive', in: R. Francovich, G. Noye, La Storia dell'Alto Medioevo Italiano (Vl-Xsecolo)
alia Luce deWArchaeologia (Rome 1994) 71-84
'Synone: Re-Considering a Problematic Term of Middle Byzantine Fiscal
Administration', Byzantine and Modern Greek Studies 18 (1994) 116-53 (repr. in
State, Army and Society in Byzantium VJH)
State, Army and Society in Byzantium (Aldershot 1995)
'The Miracles of Artemios and Contemporary Attitudes: Context and Significance',
474 Bibliography

in The Miracles of Saint Artemios: Translation, Commentary and Analysis, by J.

Nesbitt, V. Crysafulli (Leiden-New York-Kdln 1997), 33-73.
Administrative continuities and structural transformations in East Roman military
organisation c.580-640', \nVArmeeromaineetlesbarbaresdu4eau 7esiecle,eds.
F. Vallet, M. Kazanski (Paris 1993) 45-51 (repr. in State, Army and Society in
Byzantium V)
'Seventh-century continuities: the Ajndd and the "Thematic Myth"', in The
Byzantine and Early Islamic Near East YE: States, Resources and Armies, ed. Averil
Cameron (Studies in Late Antiquity and Early Islam 1. m. Princeton 1995),
379-423
'Military service, military lands, and the status of soldiers: current problems and
interpretations', DOP 47 (1993) 1-67 (repr. in State, Army and Society in
ByzantiumVE)
Haldon, J.F. and H. Kennedy, 'The Arab-Byzantine frontier in the eighth and ninth
centuries: military organisation and society in the borderlands', ZRVl (1980),
79-116
Hayes, J.W. Excavations at Sarachane in Istanbul, vol. 2: The Pottery (Princeton 1992)
Hendy, M.F. 'On the administrative basis of the Byzantine coinage, c. 400-900,
and the reforms of Heraclius', Birmingham University Historical Journal 12/2
(1970), 129-54
Studies in the Byzantine Monetary Economy, 300-1450 (Cambridge 1985)
Hermann, E. 'Die Schliessung des Verlobnisses im Recht Justinians und der
spateren Gesetzgebung', Analecta Gregoriana 8 (1935), 79-107
Herrin, J. The Formation of Christendom (Princeton 1987)
Hild, F. Das Byzantinische Strassensystem in Kappadokien (Denkschriften der osterrei-
chischen Akademie der Wissenschaften CXXXI, Vienna 1977)
Hindess B. and P.Q. Hirst, Pre-Capitalist Modes of Production (London 1975)
Hoffmann, D. Das spdtromische Bewegungsheer und die Notitia Dignitatum (Epigraphi-
sche Studien, VII/1, Dusseldorf and Cologne 1969)
Hommes et richesses dans VEmpire byzantin 1: We-VUe siecle (Paris 1989)
Hunger, H. 'Christliches und Nichtchristliches im byzantinischen Eherecht', Oster-
reichisches Archivfur Kirchenrecht XVIII (1967), 305-25 (repr. in H. Hunger,
Byzantinische Grundlagenforschung XI (London 1973))
Hunger, H. ed., Das byzantinische Herrscherbild (Wege der Forschung CCCXXXXI,
Darmstadt 1975)
Die hochsprachliche profane Literatur der Byzantiner (2 vols., Handbuch der
Altertumswissenschaft, XII, 5.1 and 2 = Byzantinisches Handbuch, 5.1 and 2,
Munich 1978)
Reich der neuen Mitte. Der christliche Geist der byantinischen Kultur (Vienna, Graz
and Cologne 1965)
Jones, A.H.M. 'Capitatio and iugatio', JRS 47 (1957), 88-94 (repr. in A.H.M. Jones,
The Roman Economy, pp. 280-92)
The Greek City from Alexander to Justinian (Oxford 1940)
The Later Roman Empire 284-602: A Social, Economic and Administrative Survey (3
vols., Oxford 1964)
'The Roman Colonate', PP 13 (1958), 1-13 (repr. in A.H.M. Jones, The Roman
Economy, pp. 293-307
The Roman Economy: Studies in Ancient Economic and Administrative History, ed.
P.A. Brunt (Oxford 1974)
Kaegi, W.E. jr., Byzantine Military Unrest 471-843: An Interpretation (Amsterdam
1981)
 Bibliography 475

'Two studies in the continuity of late Roman and eariy Byzantine military
institutions1, BF 8 (1982), 87-113
Kaplan, M. 'L'Economie paysanne dans l'empire byzantin du Ve au Xe siecle', Klio
68 (1986), 198-232
'Les grands proprietaires de Cappadoce (VIe-XIe siecles)', in Le aree omogeneee delta
Civilti Rupestre nelVambito delVimpero Bizantino: la Cappadocia (Galatina 1981)
Les hommes et la terre a Byzance du WY au XT siecle (Byzantina Sorbonensia 10. Paris
1992)
Les Proprietes de la couronne et de Veglise dans l'empire byzantin (V-VV siecles)
(Paris 1976)
'Les Villageois aux premiers siecles byzantins (VP-Xe siecles): une societe homo-
gene?' BS 43 (1982), 202-17
'Quelques aspects des maisons divines du VIC au EXe siecle', 'Acptepoy^ia <rrdv Nixo
Zpopwm vol. I (Rethymnon 1986), pp. 70-96
'Remarques sur la place de l'exploitation paysanne dans l'economie rurale', in
Akten des XVI. Internal Byzantinisten Kongresses, II/2 (Vienna 1982),
pp. 105-14 ( = JOB 32/2)
Karayannopoulos, J. Das Finanzwesen desfriihbyzantinischen Staates (Sudosteiropai-
sche Arbeiten, Munich 1958)
Die Entstehung der byzantinischen Themenordnung (Byzantinisches Archiv X,
Munich 1959)
'Fragments aus dem Vademecum eines byzantinischen Finanzbeamten', in Poly-
chronion. Festschrift Franz Dolger zum 75. Geburtstag (Heidelberg 1966),
pp. 317-33
llrjyal rqs Bv(avrivi}s 'laropCag (Thessaloniki 1978)
Karayannopoulos, J. and G. Weiss, Quellenkunde zur Geschichte von Byzanz
(324-1453) (2 vols., Wiesbaden 1982)
Kartsonis, A. Anastasis: The Making of an Image (Princeton 1986)
Kazdan, A. 'Vizantijskie goroda v VII-IX vv.', Sovietskaya Arkheologija 21 (1954),
164-88
Kazdan, A. and G. Constable, People and Power in Byzantium: An Introduction to
Modern Byzantine Studies (Washington D.C. 1982)
Kirsten, E. 'Die byzantinische Stadt', in Berichte zum XL Internationalen Byzanti-
nisten-Kongress (Munich 1958), V, 3, pp. 1-48
Kitzinger, E. Byzantine Art in the Making. Main Lines of Stylistic Development in
Mediterranean Art, 3rd-7th Century (London 1977)
'Byzantine art in the period between Justinian and iconoclasm', in Berichte zum
XL Internationalen Byzantinisten-Kongress (Munich 1958), IV, 1, pp. 1-50
'The cult of images in the age before iconoclasm', DOP 8 (1954), 85-150
Koder, J. Der Lebensraum der Byzantiner (Vienna, Graz and Cologne 1984)
'Zur Bedeutungsentwicklung des byzantinischen Terminus Thema', JOB 40
(1990)155-65
Kopstein, H., ed. Besonderheiten der byzantinischen Feudalentwicklung (BBA L, Berlin
1983)
'Zu den Agrarverhaltnissen', in Byzanz im 7. Jahruhundert, eds. F. Winkelmann
et al., pp. 1-72
Konstantopoulos, K. Bv(avnaxd fio\vp86/3ov\\a rov ev Atirivafe 'Edvixov
Nofiia-fianxov MovaeCov (Athens 1917)
Kresten, 0. 'Datierungsprobleme "Isaurischer" Eherechtsnovellen. I. Coll. 126', FM
IV (1979), 37-106
Kulakovskij, J. Istoria VizantiU vol. Ill (Kiev 1915 and London 1973)
476 Bibliography

Kurbatov, G.L. Osnovnje problemy vnutrennego razvitija vizantijskogo goroda v IV-VH

vv. (Leningrad 1971)
Laurent, V. 'L'CEuvre canonique du concile in Trulio (691-2), source primaire du
droit de l'eglise orientale', REB 23 (1965), 7-41
Lemerle, P. 'Esquisse pour une histoire agraire de Byzance: les sources et ies prob-
lfcmes', RH 219 (1958), 32-74,254-84; RH 220 (1958), 42-94. (See now the
revised version in English, published as The Agrarian History of Byzantium from
the Origins to the Twelfth Century: The Sources and the Problems (Galway 1979).)
Invasions et migrations dans les Balkans depuis la fin de l'6poque romaine
jusqu'au VIIIe sifcde', RH 211 (1954), 265-308
Le Premier humanisme byzantin. Notes et remarques sur enseignement et culture a
Byzance des origines au X* siecle (Paris 1971)
Les Plus anciens recueils (See Miracula S. Demetrii in the bibliography of primary
sources.)
Lilie, R.-J. Die byzantinische Reaktion auf die Ausbreitung der Araber (Miscellanea
Byzantina Monacensia XXII, Munich 1976)
'Die zweihundertjahrige Reform: zu den Anfangen der Themenorganisation im 7.
und 8. Jahrhunderf, BS 45 (1984), 27-39, 190-201
'Kaiser Herakleios und die Andiedlung der Serben', Sudost-Forschungen 44
(1985), 17-43
' "Thrakien" und "Thrakesion". Zur byzantinischen Provinzorganisation am
Ende des 7. Jhdts.\ JOB 26 (1977), 7-47
Litavrin, G.G. Vizantijskoe oMcestva i gosudarstvo v X-XI veka (Moscow 1977)
Loos, M. Dualist Heresy in the Middle Ages (Prague 1974)
'Quelques remarques sur les communautes rurales et la grande propriete ter-
rienne a Byzance, VIIe-XIe steles', BS 39 (1978), 3-18
Louggis, T.K. AOXLJJLIO yia rr\v xoivwixij egeAigT] arq Si&pxeict rwv keyofxevoov
oxorew&v caa)va)v (Athens 1985)
Ludwig, C. 'Wer hat was in welcher Absicht wie beschrieben?' in Varia II (Poikila
Byzantina VI. Bonn 1987), pp. 149-227
'Kaiser Herakleios, Georgios Pisides und die Perserkriege', in Varia JH (Poikila
Byzantina XI. Bonn 1991), pp. 73-128
Maas, M. John Lydus and the Roman Past (London 1992)
Macrides, R. 'The Byzantine godfather', BMGS 11 (1987), 139-62
Magie, D. Roman Rule in Asia Minor (Princeton 1950)
Magoulias, H.J. 'The lives of Byzantine saints as a source of data for the history of
magic in the sixth and seventh centuries A.D.: sourcery, relics and icons', B 37
(1967), 228-69
Maguire, H. 'Truth and convention in Byzantine descriptions of works of art', DOP
28(1974), 113-iO
Earth and Ocean: The Terrestrial World in Early Byzantine Art (London, Pa. 1987)
Malingoudis, Ph. Oi Slaboi sti Mesaioniki Ellada (Athens 1988)
Mango, C. 'Antique statuary and the Byzantine beholder', DOP 17 (1963), 53-75
(repr. in C. Mango, Byzantium and its Image V)
'The Breviarium of the Patriarch Nicephorus', in Byzantion: Tribute to Andreas N.
Stratos (Athens 1986), H, pp. 545-8
Byzantium: The Empire of New Rome (London 1980)
Byzantium and its Image (London 1984)
Le Developpement urbain de Constantinople (IV-VW siecles) (Paris 1985)
The Art of the Byzantine Empire. Sources and Documents (Englewood Cliffs. N.J.
1972)
 Bibliography

'The Life of St Andrew the Fool reconsidered', Rivista di Studi Bizantini e Slavi II
( = Miscellanea A. Pertusi II, Bologna 1982), pp. 297-313 (repr. in C. Mango,
Byzantium and its Image VIII)
'Who wrote the Chronicle of Theophanes?', ZRVl 18 (1978), 576-87 (repr. in
C. Mango, Byzantium and its Image XI)
Martin-Hisard, B. 'La domination byzantine sur le littoral oriental du Pont Euxin
(milieu du VIP-VET siecles)', Byzantinobulgarica 7 (1981) 141-54
Marx, K. Pre-Capitalist Economic Formations, ed. E. Hobsbawm (London 1964)
Mathew, G. Byzantine Aesthetics (London 1963)
Momigliano, A. ed. The Conflict between Paganism and Christianity in the Fourth
Century (Oxford 1963)
Moravcsik, Gy. Byzantinoturdca, vol. I (third edn, Berlin 1983)
Morrisson, C. 'Byzance au VIF siecle: le temoignage de la numismatique', in
Byzantion: Tribute to Andreas N. Stratos (Athens 1986), I, pp. 149-63
Muller-Wiener, W. 'Von der Polis zum Kastron', Gymnasium 93 (1986), 435-75
Murphy, F.X. and P. Sherwood, Constantinople II et Constantinople 111 (Paris 1974)
Nelson, J. 'Symbols in context: rulers' inauguration rituals in Byzantium and the
West in the early Middle Ages', Studies in Church History XIII (1976), 97-119
Nesbitt, J.W. 'Double names on early Byzantine lead seals', DOP31 (1977), 111-21
Noth, A. (with Lawrence Conrad), The Early Arabic Historical Tradition. A Source-
Critical Study (Studies in Late Antiquity and Early Islam 3. Princeton 1994)
Obolensky, D. The Byzantine Commonwealth. Eastern Europe 500-1453 (London
1971)
Ohme, H. Das Concilium Qunisextum und seine Bischofsliste. Studien zum
Konstantinopeler Konzil von 692 (Arbeiten zur Kirchengeschichte LIV. Berlin-
New York 1990)
Oikonomides, N. Les Listes de prfoeance byzantines des IX*-^ siecles (Paris 1972)
'De l'impdt de distribution a l'impot de quotite a propos du premier cadastre
byzantin (7 e -9 e siecles)', ZRVl 26 (1987), 9-19
'Les Premiers mentions des themes dans le chronique de Theophane', ZRVl 16
(1975), 1-8
'Middle Byzantine provincial recruits: salary and armament' in Gonimos, Neopla-
tonic and Byzantine Studies presented to Leendert G. Westerink at 75, eds.
J. Duffy and J. Peradotto (Buffalo, N.Y. 1988)
'Silk trade and production in Byzantium from the.sixth to the ninth century: the
seals of kommerkiarioi', DOP 40 (1986), 33-53
Olster, D. The Politics of Usurpation in the Seventh century: rhetoric and revolution in
Byzantium (Amsterdam 1993)
Onians, J. 'Abstraction and imagination in late Antiquity', Art History 3 (1980), 1-23
Ostrogorsky, G. 'Byzantine cities in the early middle ages', DOP 13 (1959), 47-66
'Das Steuersystem im byzantinischen Altertum und Mittelalter', B 6 (1931),
229-40
'Die Chronologie des Theophanes im 7. und 8. Jahrhundert', BNJ 7 (1930), 1-56
'Die landliche Steuergemeinde des byzantinischen Reiches im X. Jahrhundert',
V5W 20 (1927), 1-108
Geschichte des byzantinischen Staates (Handbuch der Altertumswissenschaft, XII, 1.2
= Byzantinisches Handbuch, 1, 2, Munich 1963)
Patlagean, E. 'Byzance et le blason penal du corps', in Du chdtiment dans la cite:
supplices corporels et peine de mort dans le monde antique (Rome 1984),
pp. 405-26
478 Bibliography

'Christianisation et parents rituelles: le domaine de Byzance', Annales E.S.C. 33

(1978), 625-36 (repr. in E. Patlagean, Structure sociale XII)
Pauvrete' economique et pauvrete' sociale d Byzance, 4e-7e siecles (Paris 1977)
Structure sociale, famille, chr&tientia Byzance (London 1981)
Pernice, A. L'imperatore Eraclio (Florence 1905)
Pieler, P.E. 'Byzantinische Rechtsliteratur', in Hunger, Profane Literatur, vol. II,
pp. 343-480
Piganiol, A. L'Empire Chre'tien, I (325-93). Histoire romaine, vol. IV, part 2 (Paris
1947)
Podskalsky, G. Byzantinische Reichseschatologie. Die Periodisierung der Weltgeschichte
in den vier Grossreichen (Daniel 2 u. 7) und dem tausendjahrigen Friedensreiche
(Apok. 20) (Munich 1972)
Pringle, D. The Defence of Byzantine North Africa from Justinian to the Arab Conquest
(British Archaeological Reports, Int. Series 99, Oxford 1981)
Rawson, B. 'The Roman family', in The Family in Ancient Rome: New Perspectives
(London and Sydney 1986), pp. 1-57
Reinink, G.J. 'Pseudo-Methodius und die Legende vom romischen Endkaiser', in The
Use and Abuse of Eschatology in the Middle Ages, eds. W. Verbeke, D. Verhelst
and A. Welkenhuysen (Leuven 1988), pp. 82-111
Rich, J., ed., The City in Late Antiquity (London 1992)
Riedinger, R. 'Die Lateransynode von 649 und Maximos der Bekenner', in
Maximus Confessor. Actes du Symposium pour Maxime le confesseur (Fribourg,
2-5 Sept. 1980), eds. F. Heinzer and Chr. Schonborn (Fribourg 1982),
pp. 111-21 (= Paradosis 27)
Rochow, I. 'Die Heidenprozesse unter den Kaisern Tiberius I. Konstantinos und
Maurikios', in Studien zum 7. Jhdt, eds. F. Winkelmann et al, pp. 120-30
Byzanzim 8.JahrhunderttinderSichtdesTheophanes (BBALVtL. Berlin 1991)
'Die monenergetischen und monotheletischen Streitigkeiten in der Sicht des
Chronisten Theophanes'. Klio 63 (1981), 669-81
'Zu einigen oppositionellen religiosen Stromungen', in Byzanz im 7. Jahrhundert,
eds. F. Winkelmann et al., pp. 225-88.
Ruhrbach, G. ed. Die Kirche angesichts der konstantinischen Wende (Darmstadt 1976)
Runciman, S. The Medieval Manichee. A Study of the Christian Dualist Heresy
(Cambridge 1947)
Ryden, L. 'The Life of St Basil the Younger and the Life of St. Andreas Salos', in
Okeanos. Essays Presented to lhor Sevcenko on his Sixtieth Birthday by his
Colleagues and Students ( = Harvard Ukrainian Studies, 7 1983 (Cambridge,
Mass. 1983) pp. 568-86
Sathas, D. John of Damascus on Islam. The 'Heresy' of the Ishmaelites (Leiden 1972)
Schlumberger, G. Sigillographie de Yempire byzantin (Paris 1884 and Turin 1963)
Schminck, A. art. 'Ehebruch', in Lexikon des Mittelalters III (1986), 1660
art. 'Ehe' in Lexikon des Mittelalters III (1986), 1641-4
Schreiner, P. 'Der byzantinische BUderstreit: kritische Analyse der zeitgenossischen
Meinungen und das Urteil der Nachwelt bis heute', in Bisanzio, Roma e Vltalia
neWMto Medioevo (Settimane di Studio del Centro Italiano di Studi sull'Alto
Medioevo XXXIV. Spoleto 1988) 319-427
Seibt, W. Die byzantinischen Bleisiegel in Osterreich I: Kaiserhof (Vienna 1978)
Sevcenko, I. 'Hagiography of the iconoclast period, in Iconoclasm. Papers of the ninth
Spring Symposium of Byzantine Studies, University of Birmingham (March
1975), eds. A.A.M. Bryer and J. Herrin (Birmingham 1977), pp. 113-31
 Bibliography 479

Sharf, A. Byzantine Jewry from Justinian to the Fourth Crusade (New York 1971)
'The Jews, the Montanists and the Emperor Leo III', BZ 59 (1966), 37-46
Studien zum 7. Jahrhundert: see Winkelmann et al.
Simon, D. 'Byzantinische Hausgemeinschaftsvertrage', in Beitrdge zur europdischen
Rechtsgeschichte und zum geltenden Zivilrecht. Festgabe fur J. Sontis (Munich
1977), pp. 91-128
Trovinzialrecht und Volksrecht', FMI (1976), 102-116
Rechtsfindung am byzantinischen Reichsgericht (Frankfurt a. M. 1973)
'Zur Ehegesetzgebung der Isaurier', FM I (1976), 16-43
Spain Alexander, S. 'Heradius, Byzantine imperial ideology and the David plates',
Speculum 52 (1977), 217-37
Speck, P. 'Wunderheilige und Bilder: Zur Frage des Beginns der Bilderverehrung', in
Varia m (PoMa Byzantina XI. Bonn 1991) 163-247
'Das Teufelsschlofi. Bilderverehrung bei Anastasios Sinaites?', in Varia V (Poikila
Byzantina Xm, Bonn 1994) 295-309
'Die Interpretation des Bellum Avaricum und der Kater MexXejiire', in Varia II
(Poikila Byzantina VI. Bonn 1987), pp. 371-402
Das geteilte Dossier. Beobachtungen zu den Nachrichten uber die Regierung des Kaisers
Herakleios und die seiner Sohne bei Theophanes und Nikephoros (Poikila Byzantina
K. Bonn 1988)
Ich bin's nicht. Kaiser Konstantin ist es gewesen. Die Legenden vom Enflufi des Teufels,
desjuden und des Moslem auf den Ikonokslamus (Poikila Byzantina X. Bonn 1990)
'Der "zweite" Theophanes. Eine These zur Chronographie des Theophanes', in
Varia V (Poikila Byzantina XIII. Bonn 1994), pp. 433-83
Spieser, J.-M. 'Les villes en Grdce du &T au VST siecle', in VOles et peuplement dans
rUlyricumprotobyzantin (Rome 1984), pp. 315-38
Starr, J. The Jews in the Byzantine Empire, 641-1204 (Athens 1939)
Ste Croix, G.E.M. de. The Class Struggle in the Ancient Greek World (Oxford 1981)
Stein, E. 'Ein Kapitel vom persischen und vom byzantinischen Staate', BNJ 1
(1920), 50-89
Histoire du Bas-Empire, 2 vols. (I: Paris and Bruges 1939 and Amsterdam 1968;
II: Paris and Bruges 1949 and Amsterdam 1968)
Studien zur Geschichte des byzantinischen Reiches vornehmlich unter den Kaisern
Justinus U und Tiberius Konstantinus (Stuttgart 1919)
Stratos, A.N. Byzantium in the Seventh Century (Eng. trans.), I: 602-34 (Amsterdam
1968); II: 634-41 (Amsterdam 1972); III: 642-68 (Amsterdam 1975); IV:
668-85 (Amsterdam 1978); V: 686-711 (Amsterdam 1980)
Svoronos, N. 'Notes sur l'origine et la date du Code Rural', TM 8 (1982), 487-500
'Recherches sur le cadastre byzantin et la fiscalite aux XIe-XIIe siecles: le
cadastre de Thebes', BCH 83 (1959), 1-166 (repr. in N. Svoronos, Etudes sur
Vorganisation interieure, la societe et Veconomie de Vempire byzantin III (London
1973)
Teall, J. 'The Byzantine agricultural tradition', DO? 25 (1971), 34-59
'The grain-supply of the Byzantine empire', DO? 13 (1959), 87-139
TIB Tabula Imperil Byzantini: vol. I, J. Koder, F. Hild, Hellas und Thessalia (Vienna
1976); vol. II, F. Hild, M. Restle, Kappadokien (Kappadokia, Charsianon, Sebastia
und Lykandos) (Vienna 1981); vol. Ill, P. Soustal (with J. Koder), Nikopohs und
Kephallenia (Vienna 1981); vol. IV, K. Belke with M. Restle, Galatia und Lykao-
nien (Vienna 1984) (= Denkschriften der osterr. Akad. der Wissenschaften
CXXV, CIL, CL, CLXXII)
480 Bibliography

Tinnefeld, F.H. Diefruhbyzantinische Gesellschaft. Struktur - Gegensatze - Spannungen

(Munich 1977)
Toynbee, A. Constantine Porphyrogenitus and his World (London 1973)
Treadgold, W.T. The military iands and the imperial estates in the Middle Byzan-
tine Empire', in Okeanos. Essays Presented to Ihor Sevcenko on his Sixtieth
Birthday by his Colleagues and Students ( = Harvard Ukrainian Studies VII
(1983)) (Cambridge, Mass. 1983), pp. 619-31
Treitinger, 0. Die ostromische Kaiser- und Reichsidee nach ihrer Gestaltung im
hoflschen Zeremoniell (Jena 1938)
Trombley, F. 'The decline of the seventh-century town: the exception of Euchaita',
in Byzantine Studies in Honor of Milton V. Anastos, ed. Sp. Vryonis, jr. (Malibu
1985), pp. 65-90
Trombley, F.R. Hellenic Religion and Christianization c. 370-529 (Religions in the
Graeco-Roman World CXV. Leiden 1993)
Udal'cova, Z.V. ed., Kuttura Vizantii, lV-pervaja Vll v. (Moscow 1984)
Udal'cova Z.V. and K.A. Osipova, 'TipologiSeskie osobennosti feodalizma v Vizantii',
in Problemy social'noj struktury i ideologii srednevekovogo obscestva I (Leningrad
1974), pp. 4-28
Van Dieten, J.-L. Geschichte der Patriarchen von Sergios I bis Johannes VI. (610-715)
(Amsterdam 1972)
Vryonis, Sp., jr., The Decline of Medieval Hellenism in Asia Minor and the Process of
Islamization from the Eleventh through the Fifteenth Century (Berkeley, Los
Angeles and London 1971)
Weber, Max 'Die Stadt. Begriff und Kategorien', in Die Stadt des Mittelalters I: Begriff,
Entstehung und Ausbreitung, ed. C. Haase (Darmstadt 1969)
Wenger, L. Die Quellen des romischen Rechts (Vienna 1953)
Whitby,L.M. 'The Great Chronographer and Theophanes', BMGS 8 (1982-83) 1-20
Whittow, M. 'Ruling the Late Roman and early Byzantine City: a Continuous History',
PP129 (Nov. 1990) 3-29
The Making of Orthodox Byzantium, 600-1025 (Basingstoke-London 1996)
Wickham, CJ. Early Medieval Italy. Central Government and Local Society, 400-1000
(London 1981)
'The other transition: from the ancient world to feudalism', PP 103 (1984),
3-36
Winkelmann, F. 'Agypten und Byzanz vor der arabischen Eroberung', BS 40
(1979), 161-82
Byzantinische Rang- und Amterstruktur im 8. und 9. Jahrhundert (BBA LIII, Berlin
1985)
Die ostlichen Kirchen in der Epoche der christologischen Auseinandersetzung (5.-7.
Jahrhundert) (Kirchengeschichte in Einzeldarstellungen 1/6, Berlin 1980)
'Die Quellen zur Erforschung des monenergetisch- monotheletischen Streites',
Klio 69 (1987), 515-59
ed., Quellen zur Geschichte desfruhen Byzanz (BBA LV Berlin 1989)
Quellenstudien zur herrschenden Klasse von Byzanz im 8. und 9. Jahrhundert (BBA
LIV, Berlin, 1987)
'Zum byzantinischen Staat (Kaiser, Aristokratie, Heer)' in Byzanz im 7. Jahrhun-
dert, eds. F. Winkelmann et al., pp. 161-288
Winkelmann, F., and H. Kopstein, eds., Studien zum 7. Jahrhundert in Byzanz.
Probleme der Herausbildung des Feudalismus (BBA XLVII, Berlin 1976)
 Bibliography 481

Winkelmann, F. H. Kopstein, H. Ditten and I. Rochow, Byzanz im 7. Jahrhundert.

Untersuchungen zur Herausblldung des Feudalismus (BBA XLVIII, Berlin 1978)
Yannopoulos, P. La Sodtti profane de Vempire byzantin des VIF, VIW et IX* siecles
(Universite de Louvain, Recueil de Travaux d'Histoire et Philologie, 6C s6r.f
fasc.6) (Louvain 1975)
Zacharia von Lingenthal, C.E. Geschichte des griechisch-romischen Rechts (third edn,
Berlin 1892 and 1955)
Zepos, P.I. 'Die byzantinische Jurisprudenz zwischen Justinian und den Basiliken',
in Berichte zum XL Internationalen Byzantinisten-Kongress (Munich 1958), V, 1,
pp. 1-27
 Index

Abbasid rule, 84 Anastasius of Sinai, xxvfT.. 342,371,

c
Abd al-Malik, 69f., 72, 193 431ff.,438
Abydus, 83, 107 Anastasius (Sygkellos), 88, 90
Academy (of Athens), 329 Anatollkon (district, thema). 69ff., 197ff.,
Achaemenids, 20 212ff.
acheiropoieta, 356 Anchialus, 67
Acroinum, 84, 107, 109 Ancyra, 103, 107, 108, 113, 118, 218,
Acts of the Apostles, 431 226
actuarius, 221 Andreas (son of Troilus), 168
Adelphianoi, 342 Andreas of Crete, 79, 321, 434
adiectio sterilium {see also epibole), 29, 142 Anemurium, 109, 111, 118, 226
adiutor, 178 annona(e), 28,177, 224-6, 229, 238, 242,
adnoumia, 241, 243 245, 448
Adrianople, 82. 239 Antakya, 236
Adriatic, 21, 120 antecessores, 260, 265, 266
Aegean, 10,12, 116, 120, 156, 210, 217, anthypatoi, 195, 202-6. 275f., 391
294, 387 antigrapheis, 193, 267, 274
Africa, 4f., 10f., 15, 20, 34, 36, 41, 57, Antioch.xxiii
60, 69f.. 74. 76. 177. 209f., 235, Antioch (Isauria), 107
239, 249, 252, 282, 305, 3O7f., 310, Antioch (Pisidia), 80, 107
366 Antiochene tradition, 289
Agathias, 420 Anthony IV (Patriarch 1389-90), 297
Agathon (deacon), 78, 321 Apamea. 34, 102f.
Agnellus, 78 Aphrodisias, 108, 111, 118, 226
Alexander (the Great), 20, 368 apo eparchon, 393, 395
Alexandretta, 224 apo hypatdn, 165, 193ft, 199, 393-7
Alexandria, 38f., 48, 54, 287f., 291, 302, Apocalypse (of Ps.-Methodius), 144, 433,
315,327 436
Alexios Mousele. 204 apocalyptic literature. 36711., 433, 436f.
c
Ali, 55 Apollonia, 108
Amaseia, 80. 107 apophthegmata, 429
Amastris, 107 Apostles of Christ, 283
Ambrosius of Milan, 282 apothekai, 188(1.
Amisus. 107 Appendix Eclogae, 265
Amorium, 55, 8 3 , 1 0 7 , 1 1 3 , 216 Apsimar (Tiberius, drouggarios), 75
ampullae, 415 Arabic chronicles, xxiiif.
c
Amr, 54f. arcae/arklai, 177, 189, 196, 251
Anastasian walls, 120 archiatroi, 391
Anastasioupolis, 134, 218 archon (Us), 272, 275
Anastasius (ex-consul), 192 Arculf (pilgrim), 423
482
 Index 483
Areopagite (Ps.-Dionysius), 342 Boris-Michael (Tsar), 378
Arian(s), and Arianism, 339f., 342, 349 Bosphorus, 43
armamenta, 240 Boukellarion, 155f.
Armenia, and Armenians, 62f., 76, 209, Bouraphos, 169
213f., 216. 219f., 252, 286. 300. 305. Brief History (of Patriarch Nicephorus).
312, 321f., 338, 351 xxi-xxu,116,352
Armeniakon (district, thema), 78ff., 1971T., brigandage, 31
212ff. Britannia, 39
army: late Roman, 208-11; Byzantine, Brumalia, 334
21 Iff.: equipment, 224, 232ff., 238ff.: bucellarii, 211
pay/remuneration, 223(T.; Bulgaria, 156
provisions/supplies, 22Off., 227f., Bulgaro-Slav state, 67
229-32; in politics, 371ff. Bursa, 236
Artavasdus, 82, 90 Byzacena, 70, 307
Artemius (protoasektretis), 80 Byzacium, 210
Asia (province), 13, 39, 234, 236, 330,
332^338 Caesar, 51, 54, 77, 390
Asiana, 187, 218 Caesarea, 42, 55, 102f., 107, 239
Asparuch (Khan of Bulgars), 47, 67 Cain, 59
Assus. 108.118 Calabria. 90. 148, 316
Athens, 118,250 Calends, 334
Athigganoi, 344f. Callinicus (Patriarch), 77, 320
Attalia, 110 Callinicus (mechanic), 64
Attic (style), 329 Camuliana, 107
Augila (oasis), 330 candidate 391
Aurelioupolis/Tmolus, 109 canon law, 304, 318, 334ff., 378ff.
Avars, xxv, 3 2 , 3 7 . 4 4 - 8 canonicum, 294
Aya Soluk, 104 capita/capitus, 148f., 224-6, 229, 448
axial, 392 capitatio/iugatio, 28f.. 133,141-3. 178
Cappadocia, 13. 42, 55, 71, 83, 144f.,
Baalbek/Hlerapolis, 329, 349 174, 219, 229, 236, 239, 249
Bacaudae, 31 caput, 28f.
Baghdad, 84, 112 Caravisiani (Karabisianoi), 214, 217
Baladhurf, xxiv Caria. 110,219,229
balnitores, 391. 393 Carrhae/Harran, 330
Balsamon (canonist), 334 Carthage, 5, 35, 42, 56, 75, 210f., 234,
banda/bandon. 212 306f.
Bar Hebraeus, xxiii Carthagena, 176
Basil I of Moscow, 297 castrensis, 186
basilikoi, 391 Catania, 176, 186, 211
Belisarius, 16, 21,165, 252 Catholicate, 312
Benevento, 60 Caucasus (Albania), 71
Benjamin (monophysite patriarch), 54 Caucasus (region), 13, 20, 46, 58, 103,
Berber tribes. 21, 34. 36. 69f., 210, 331 156,169,215
Bessarabia, 67 Cephallenia, 71
Bithynia, llff., 197, 219 Chalcedon, 52, 69, 73. 103, 107, 286f..
Blachernai, 303 293, 298
Black Death, 146 Chalke gate, 88
Black Sea, 10, 20f., 210 Charsianon, 155-7
blattion, 237 chartouhrios/chartularius, 84, 178, 180-2,
Blues (deme), 74f. 188, 192, 194, 196, 200, 212
Bogomils, 342 Chazars, 67, 76ff., 84, 352
Bononia, 114 Chazaria, 77
Boniface (papal clerk). 319 Cherson, 58, 74, 76, 78, 168, 176, 310
Book of Ceremonies, 195. 203, 240 Chersonnesos (Aegean), 234
BookoftheEparch, 273 Chios. 320. 338
484 Index
Chonae, 109 Constantia (Cyprus), 55, 224, 350
chorion, 137ff. Constantine (son of Heraclius), 46
Chosroes I, 20, 46 Constantine (bishop), 87f.
ChosroesII, 35, 312, 352 Constantine (ex-consul), 192
Chronica maiora, xxiii Constantine Lardys, 187f.
Chronicle of George the Sygkellos, xxi Constantinople (sieges of), 45f., 63f., 82ff.
Chronicle of John of Nikiu, xxiii consularis, 203
Chronicle of Michael the Syrian, xxiif. conventus, 271
Chronicle of Monemvasia, 44 Copts, xxiif., 43, 54,287f.
Chronography of Theophanes the Corinth, 60, 118
Confessor, xxif., 214 cornicularius, 273
Chrysoupolis, 107, 370 Corpus luris Civilis, 16, 19, 25, 268
Church, xxff.; in Armenia, 312, 321f.; Corsica, 209
imperial, 28Iff.; relations with state. Corycus, 220
281-6; and wealth, 289f.; Cos, 55, 63
administration, 291-3; and monasteries, Cotyaeum, 218
293f.; schisms. 301f., 306ff., 318f.; of councils (of Church, lists of signatories),
Ravenna, 78, 313, 316f. 121ff.
Church buildings, xxvi courts: ecclesiastical, 303; secular, 269IT.,
Church of the Holy Apostles, 61, 304, 292. 303
423 Crete, 55. 75, 90. 234. 240
Church of the Holy Sepulchre, 43 Crimea, 10
Church of the Holy Wisdom, 16 Croats, 47, 352
Church of St Peter, 42 Ctesiphon, 43
Cilicia, 39. 43, 69. 80. 83. 109. 220. 234. cubicularius/koubikoularios, 60, 185, 191
338. 348 cubiculum, 194
cingulum, 390, 392 curator, 96, 127
cities: evolution, function, decline of, 445f.; curiales, 96ff., 162f., 447
definition of, 99ff.; fate in seventh curopalates, 390
century, 102-17; and market exchange, Cybele, 336
104f., 117ff.,445ff.;and Cyclades, 234
plagues/earthquakes, 11 If. Cyprus, 55ff.
civitas, 96, 100, 176, 266,445 Cyril of Alexandria, 289, 300, 327
clarissimus, 161ff., 165, 390f. Cyrus (Patriarch of Alexandria), 48f., 54f.,
climate (of imperial lands), lOfT. 300ff., 309, 315
Cnidas, 110 Cyrus (Patriarch of Constantinople,
Codex Iustintanus, 25, 97, 103, 135, 255, 705-11), 78, 296, 321
268, 292, 346 Cyzicus, 63f., 71, 76, 107, 176, 186, 218,
Codex Theodosianus, 346 239
coemptio (see also synone), 147f., 179, 23 If.
coinage: function of, 117-20, 447IT.; Dacia, 209
reform, 370 Dalmatia, 66
collatlo lustralis, 177 Damascus, 1, 42, 112, 216, 239
collegia, 28 Damian (Patriarch of Alexandria), 287
Colonae, 109 Damietta, 80
Colossus (of Rhodes), 55 Daniel (vision of), 367
Comana, 102 Danube, 1, 21, 32, 43
comes, 54ff. Danubian forces, 12, 41
comitatenses, 28, 208f., 211 Dardanelles, 83
comltatus, 176 'Dark Ages', 1
comitiva, 173, 186. 193 David Tiberius, 5If.
commemoratio, 273 Day of Judgement, 436
commentarlensis, 273 defensores civitatis, 97, 269, 275
commercialcommerciarlus, 189, 251 delegatoriae, 221
compulsores (expelleutai), 178 demosia, 182
consiliarius, 272 Denis of Tell-Mahre, xxiii
 Index 485

De Thematibus, 219 256, 262, 272f., 287, 300ff.f 346, 348,

diagrafa, 148 352, 371, 375, 380,440,449; Heraclius
Dniester, 67 Constantine, 51-3,165,167,184, 304f.;
Digest. 25, 255. 260. 265. 268 Honorius, 269; Julian. 96; Justin I. 17.
dignitas, 390, 392 31, 35, 345; Justin II, 31f., 127, 255,
dikastai, 275 261, 330, 362, 374, 378, 415; Justinian
dioceses (boundaries), 218 1,15ff., 35f., 177,179, 185.187, 209f.,
dioiketai, 182, 196-201, 212, 221, 391 254, 265, 283f., 289ff., 327, 329f., 345,
Dionysus, 334 377f., 395, 407, 439; Justinian II, 56ff.,
diptychs, 314, 320 71-4, 76-80, 83, 168, 193, 213, 250,
Dobrudja, 67 263, 317-22,370. 374, 394,441; Leo I,
domesticus, 186,193, 202, 216f. 210,289; Leo m. It, 18,82,84-91,
domusdivinae, 125, 174, 192 168,242,277, 323, 346, 374,441f.;
Donatists/Donatism, 282, 340 Leo IV, 250; Leo V, 256, 264; Leo VI,
donativa, 247 258, 292, 377f., 380; Leontius, 74f., 77,
Dorylaeum, 107, 218 168, 319; Marcian, 162; Maurice, 31,
drouggoi, 212 35ff., 39. 211, 224, 268, 298f., 330,
ducatus, 215 346, 360; Nicephorus I, 142, 149, 154,
duces, 209ff., 251 243, 262; Nicephorus 0, 295;
Dukeoflstria, 247 Philippicus Bardanes, 78f.; Phocas, 31,
Durostorum, 114 35f., 41ff., 167,185, 265. 268, 299,
Dvin (synod of), 312 404; Romanus I, 246; Romanus II, 159;
dyophysite (Chalcedonian), 286f., 308 Theodosius II, 163, 289; Theodosius HI,
dyothelete, 308 80-3, 168; Theophilus. 204, 279:
Tiberius Apsimar, 75-7, 319; Tiberius
earthquakes. 111 (usurper), 84; Tiberius Constantine, 1,
Ecloga, 85ff., 254ff. 31, 36, 211, 255, 298, 332, 341, 346,
Edessa, 59, 239, 349, 423 348, 360, 362; Tiberius (son of Justinian
edicta. 22 H), 77f.; Valens, 96, 176; Valentinian,
Edict (of 681), 255, 261 96, 161, 176; Valentinian HI, 282; Zeno,
Edict on Faith (of 610), 300 97, 183, 270, 314
education, 427ff.; legal, 265, 268f., 271f., empresses: Eirene, 149f., 155, 202, 204,
274, 428f. 256; Eudocia, 304; Martina, 5If., 54,
Egypt, 1, 10f., 31. 35, 41, 46, 50, 55, 70. 167, 304f., 380; Sophia, 297; Theodora,
177, 209f., 216, 229, 249, 286f., 301 16f., 31, 297. 407
eidikon/etdikos, 180,182, 189, 204, 206, emphyteusis, 135
212, 233, 238, 240, 243 endoxotatos, 165, 391
Ekthesls (of 638), 49, 56f., 255, 261, 301f., endoxotatoi patrikloi, 164
306, 309 England, 146
Emesa, 216 epano ton demon, 194
emperors: Alexius I, 292; Anastasius I, eparchtai, 197ff., 202, 206, 276. 395
96(T.; Anastasius II, 80, 82-5,168, Ephesus, 82, 98,104, 107f., I l l , 117f.,
322f., 394, 442; Arcadius, 269; Basil I, 226
72,331;ConstansH,xxif., 52ff., 67,113. eptbole, 137,142,151
119.168,226f., 255f., 273, 304ff., 371, Eplphanius, 168
389, 392;ConstantineI,96,160,176f., Epirus, 71
211,270,283,289; Constantine IV, epistolarum (scrinium), 193, 271
63-70, 313ff.. 374, 392; Constantine epi Us trapezis, 186
V, 72, 84f., 90f., 116,134,192,202, epoptai, 182, 200, 212
242,255; Constantine VH, 47,159,195, ergodosia, 238
219,225,240,246; Constantius, 96. Erzerum (see Theodosloupolls)
412; Diocletian, 22,160f., 176f., 211; eschatology/eschatological literature, 367f.
Hadrian. 25;Heracleonas. 51. 53f.. 167, Ethiopia, 352
225, 304f.;Heraclius, xxv, 37,41ff.. 75, Euboea, 197
118,166f., 186,188,190f., 194f., 196, Euchaita, 95, 107, 109, 117, 120
199f., 206f., 2 1 7 , 2 1 9 , 2 2 4 , 2 5 0 , 2 5 4 , Eunomians, 339, 342
486 Index
Euphratensis, 209 Great Bulgaria, 66
Euphrates, 1,10, 25 Great Church (sec also Church of the Holy
Euprepius, 274 Wisdom/Hagia Sophia). 56
Eustathius Maleinus, 156 Greek fire, 64, 83
Eutyches (archimandrite), 339 Green deme, 75
Eutychius (exarch), 89 Gregory (exarch), 57f., 60f.. 305-8. 310,
Evagrius, 349 313, 369
exarchates, 35-6 Gregory Mousoulakios, 204
excubitores, 54, 192. 210 GulfofAttalia, 220
exisotai, 182, 212 Gulf of Burgas, 82
extraordina/extraordinaria, 147
Hagia Sophia (see also Great Church/Church
fabricae, 77, 233, 239 of the Holy Wisdom), 49, 52, 57, 82,
familia, 382, 401 284, 301, 306, 316. 394
family: structure and evolution, 376flf.; hagiography, xxi, 427, 4291T.
legal status - betrothal, marriage, Harran, 349
divorce, 377f. haruspices, 328
Farmers' Law, 132ff., 254, 386f. Helenopontus, 107,157, 220, 234
Paustinoupolis, 109 Heliopolis. 330
Felix (Bishop of Ravenna), 77 Hellas, 39, 74,197
feudalism/feudal mode of production. 45 Iff. Hellespont^, 218f.. 239, 350
Fiscal Treatise, 144 Henotikon, 298
foederati, 211, 247, 250 Heraclea. 45,107f., 239
follis, 119 Heraclea (Pontica), 108
Francia, 39 Heraciius (the elder), 41
fundi (patrimoniales), 125,138 Heraclius (brother of Tiberius Apsimar),
76
Galatia. 107ft Heraclius (son of Constans II), 59, 68
Gangra, 107 heresies/heretics, 337ff.
genikon, 18011. Hierapolis, 108, 287
genikos logothetes, 193f.. 200, 394 Hiereia, 46
genikos nomos, 303 Hikanatoi, 202
geography (of the empire), 9iT. Hispania, 39
George (apo hypaton), 197, 234 historiography, 425
George (kommerkiarios), 235 Hodegos (of Anastasius of Sinai). 370
George I (Patriarch), 68, 315 Holy Land, 423
George (presbyter), 342 holy men, 356flf., 369
George (of Pisidia), xxv, 425,431 honestiores, VI7
George (Sygkellos), xxi honorarii, 162
georgos, 135 Honorias. 196, 219
Gepids, 32 Hormisdas, 187, 192
German/Germanic, 17, 32, 166, 211, 349, Horreum Margi, 239
355. 379. 456 hwniliores, 27
Germanus (magister militum), 210 Hungary (plain of), 64
Germanus (of Cyzicus, Patriarch of Huns, 103
Constantinople), 79, 82. 88, 168, 32 If., hypatos, 165, 192, 199, 205, 391, 393-7
344 Hypomnestikon, 273
Ghassanids. 20
Ghevond, xxiu Iberia, 70f.
Gibraltar, 15 Ibn az-Zubair, 71
gloriosus/gloriosissimus, 162-5, 169, 188, IbnHawkal, 112
390, 391, 393, 395 Iconium, 107
Golden Horn. 75 iconoclasm, 84, 87. 323
Gordoserbon. 124, 350 icons: role and function of, 356ff., 366,
Gothograeci, 80 405ff., 423f.: style and mode of
Goths, xix, 21, 72, 102,211 expression, 407ff.
 Index 487

ideology: imperial, 19f., 37-9, 348; and Kamacha, 107

symbolic universe, 324ff.. 3481T. kandidatoi, 170, 205, 393f.
idike trapeza, 182 kapnikon, 142, 149f, 182. 231
idikon, 180,182. 233 Kappadokia, 155ff.
Ignatius the Deacon, 422 Karabisianoi, 212
illustris/llloustrlos, 1611T. kastron, 102.459-61
Illyricum, 12, 21, 9 0 . 1 5 1 . 1 7 7 , 1 8 7 , 195, Kavadh/Siroes. 46
209, 252 keleuseis, 261f.
impehum Romanum, 354 Kibyra/Cibyra. 220
indictio, 177 Kibyrrhaiotai. 74f.. 220
Institutions, 25, 255 kinship. 376ff.
Ionian Sea, 10 kleisoura, 212. 220
Iraq, 172 K/etorotoflion (of Philotheos), 180. 190.
Irenopolis, 239 194, 273, 391, 394
Isauria, 43, 103, 107. 176, 209f., 219, koitdn, 183, 192
224, 338 koitdnion, 197
Isidore of Seville, xxiii Koloneia, 107
Islam: rise and conquests of, 49ff.; kome, 137f., 204
invasions and economic effects, 143ff.; kommerkia/kommerkiarioi, 196ff.
iudices pedanei, 271 komes, 213f., 216
iuga, 28f., 149 koubikoularioi, 186, 398
iussio, 148, 213ff.. 261f., 316, 319 Kovrat, 47, 67, 352
kritai, 181, 275f.
Jacob Baradaeus. 286. 298. 338 Kuban. 47
Jacob (Jew of Palestine), 39
Jacob ('recently baptised'), 304 lachnistinon, 367
Jacobite(s), 287, 298 laeti, 247, 250
Jerusalem, 38ff. Lampetianoi, 342
Jews/Judaism, 87IT., 345IT. Laodicea. 108. I l l
JohnofAntioch, 425 largitiones, 182. 188-91, 196, 218, 270
John Athalarich, 51 Last Judgement, 436
John of Biclar, xxiii Lateran (council), xxiv. 57.285,307.
John the Cappadocian, 16 309,366
John II (Chalcedonian Patriarch of latifundia, 132
Alexandria), 287 Latin, xxii, 368,404
John III (Chalcedonian Patriarch of law: function and symbolism of, 258ff.,
Alexandria), 288. 291 276ff.; legal texts, 254fT.
John VI (Patriarch of Constantinople), 77', Lebanon, 66. 71
32 If. legislation (imperial), 254-64
John (ex-consul), 192 Leontius (general), 71
John of Damascus, 354 Leontius (domestikos tes vasilikes trapezes),
John (deacon of Hagia Sophia), 394 186
John of Ephesus, 330, 332. 338. 341. 349, Leontius (ex-consul), 192
351 ' Leontius (dioiketis of the eparchiai), 197
John Lydus, 1 5 , 2 4 , 4 3 1 Lesbos, 338
John Mauropous, 117 libellorum (scrlnium), 271
John Moschus. 306, 423. 43Of. Uber Pontificate, xxiii
John (patrikios), 75 Libya. 70.209
John of Porto (bishop). 319 life of Andrew the Fool xxvii. 116.336,
John Philoponus, 431 356. 361. 367f.. 429.433
Julian (ex-consul). 192 Life of Anthony the Younger, 335
Jupiter Ammon (temple of), 330 Life of the Patriarch Tarasius, 422
justice (administration of), 264ff. Life of Philaretos the Merciful, 131.141.
Justinian (patrldus), 168 155, 160, 243, 245, 387
Life of Theodore of Sykeon, 134.137.141.
Kallinikon. 298 387, 423
488 Index
Liguria, 34 Melkite (Church.) 287f.
limes, 210 mentality (Byzantine), 324ff.
limitanei, 28, 208f., 211. 216f. Mesopotamia, 50, 209, 219
literature (theological and secular), 425ff. Messalians, 34If.
Liutprand (Lombard king), 89 metrokomia, 136
locator, 134f. Michael (hypatos), 194
logotheston. 180.188f.. 191, 195, 200 Michael of Synnada, 335
logothetis, 181, 183, 185, 187f, 190, Miletus. 108, 111
192f., 206, 274 miliaresia, 149
long walls (of Constantinople), 45, 209 Milion gate, 79
Loulon, 109 military lands, 244-50
Lucania. 148. 319 militia, 163, 392
Lycaonia, 31.107.157. 209f.. 219, 236, missionary activity, 35OfT.
338, 344 Misthia, 80,107
Lycia, 219, 234. 338 mode of production, 45Iff.
Lydda, 423 moderator, 203, 206
Lydia, 13, 107. 109. 145, 197, 219, 330 Moesia, 67, 209f., 269
monasteries, 293-6
Macedonia, 71,156,256, 331 monenergism, 300f.
Macedonius (Patriarch of Antioch), 302 monophysites, 24, 31f., 286ff., 3O5f., 321,
Maeander, 108, 159 328f., 364, 440
magic (and superstition), 331, 333 monotheletes, 56fT., 67f., 86f., 2971T.,
magister memoriae, 194 314ff., 32If., 364f.,440
magister militum, 43, 165, 209ff., 215ff., Montanists, 339, 341f., 344
227. 252. 39Of. Moorish troops, 41
magister officiorum, 181, 194, 201, 233 Mopsuestia, 107
magistri (of palatine scrinia), 271 morte, 135
magistros, 84, 195 Mount Sinai (icons), xxvu, 409
Magnesia. 108. 158 Mousoulios. 243, 245
magniflcus, 162f.. 395 Mucawiya. 55, 63, 64
Makarios (Patriarch of Antioch), 56, 302, Muhammad, 55
314f.. 321 Mulberry {morus alba), 236
manceps, 226 mtrnera, 127, 178, 240
mandatar, 393f. Myakios, 169
Mani, 331 Myra, 107f.
Manichaeans, 341, 343 Mzez Gnouni (Mizizios), 61, 213, 313f.
Mantzikert (synod), 322
Marcionists, 342f. Nacolea, 87
Marcianoupolis, 239 Naissus, 239
Mardaites, 66, 71 Naples, 60, 77, 320
Marinus (son of Martina), 52 Narses, 16, 184
Marinus (apo eparchdn), 197, 201 navicularii, 176
Maritsa, 82 Nea Justinianopolis, 350
Markianistai, 342 Neocaesarea, 107
Marseille, 406 Nestorian (theology), 339
Maslama, 83 Nestorius, 339
Mauretania, 210, 307 'new Mercian', 'new Justinian' (of
Maurus (Archbishop of Ravenna), 313f. Constantine IV), 68
Maximus Confessor, xxv, 56ff., 285,306f., New Testament, 335, 344, 346, 367, 423
309ff., 365f. Nicaea, 82, 102, 107,110, 158
Medina, 55 Nicetas (cousin of Heraclius), 41, 43
megaloprepestatos, 196 Nicetas (ex-consul), 192
Megas chronographos, 425 Nicetas Xylinites, 84
megas kouratdr, 183, 192f. Nicomedia, 82, 102,107, 176, 186, 218,
Meletians, 342 239, 320
Melitene, 80. 103. 107 Nicopolis, 71, 103, 107f., I l l
 Index 489

Nika riot, 15,17, 22, 167, 171, 354 pater civltatls, 96

Nikiu, xxiii Patras, 60
Nikolaos Mesarites, 423 patria potestas, 378
Nile, 330 patrlcius/patrtkios, 47, 75, 161f., 168ff.,
Nilus Scholastlcus, 420 192, 39Of., 393f., 396f.
Nineveh, 46 patrlmonium, 174,186, 192
nobilissimus, 390 patrocinium, 128, 130
nomlkol/tabellarloi, 274 Paul (apohypaton), 193
nomlna gentllia, 155,170, 382 Paul (endoxotatos apo hypaton), 196, 198
nomismata, 69 Paul (ex-consul), 192
nomocanones, 381 Paul (chartoularios), 84
NotitiaofPs.-Epiphanius, 351 Paul (exarch), 89
Notitia Dignitatum, 251 Paul (genikos logothetes), 194
Novatians 342, 344 Paul (hypatos), 197f.
novels (novellae), xxiv, 22,25,255ff., 261ff. Paul (Patriarch of Constantinople), 57, 68,
numerarii, 178, 188f. 305-9, 31 If., 315
Numidia, 210, 307 Paulianists, 342
Paulician Church, 343f.
Odessus, 114 peculium, 127
officiates, 27S Peloponnese, 44f., 48, 66, 71, 331, 353
offldum, 174f., 180, 193, 200, 273, 449 Pentapolis, 54, 70, 209
oikeiakon (vasilikon vestiarion), 183 penthekte (quinisextwn), 317
oikonomoi, 292 Perbund, 331
oikoumene, 258ff. Pergamum, 83, 102. 107f., 226, 336
old Great Bulgaria, 47 Perge, 109
Old Testament, 85, 344, 346, 362, periblebtos, 391
367f. Persians (see also Sassanid), xxv, 32,34f.
Olymplus (Exarch of Ravenna), 57f., 61, Persian war (of Heraclius), 42f., 45f.
310, 370 Pessinus, 107
Onogur Huns/Bulgars, 47, 66, 352 Petasius. 89
Opsikion, 54, 212ff., 218f., 225, 234 Peter (ex-consul), 192
OptatusofMilevis, 282 Peter (Exarch), 311
optimates, 80,211 Peter (Patriarch of Constantinople), 315
optiones, 448 Peter IV (monophysite Patriarch of
ordinarii iudices, 269 Alexandria), 287
ordlnes, 165, 393 Philae, 330
Organa, 47, 352 Philagrius, 184
Oriens, 103,157, 170,187f., 218, 220, Philippoupolis, 82
252 Philon, 23
Orontes, 237 Phoenice, 209
Osrhoene, 209 Phoenicia, 36, 239
Ostrogoths, 15,19, 2 If., 174 Phrygia, 13, 217ff., 229, 330, 338
c Pisidia, 31, 107, 157. 209f., 219
Othman, 55
Otranto, 60, 320 Pitzigaudes, 169
Ottoman Sultan, 282 plague (effects of), 111, 144
Platon ipatricius), 168
paganism (survivals), 327ff. Pliska, 67
Palermo, 320 Plotinus, 23
Palestine, 39ff. Podandus, 107
Pamphylia, 102, 109, 209, 219, 338 Poinalion, 266
Pannonia Secunda, 32, 209 polls (see also cities). 1001T., 113.116, 120.
Paphlagonia, 11. 107, 131, 155ff., 220 385,445,459-61
parakoimomenos, 183, 193 Polyeuctes (ex-consul), 192
paroikos, 136. 154 Pontic, 102, 110, 187, 218
Paschal Chronicle, xviii, 425 Pontus, 11, 13, 31, 59, 131, 209f., 224,
Patara, 108 294
490 Index
popes: Adeodatus, 314; Agathon, 314f.; Quinisext council. 73. 317f., 332ff., 338,
Conon, 148, 318; Constantine, 77, 79, 351, 366, 384, 421, 431
319; Donus, 314; Eugenius. 311;
Ratiaria, 239
Gregory I, 42, 406; Gregory II, 89, 340;
Ravenna, 35, 58, 74, 77, 309, 313f., 316,
Gregory in, 90; Honorius, 49, 68, 301f.,
319
308. 317; John IV, 302, 305; John VII.
Reggio, 320
319; Leofl, 316; Martin, xxv, S7L 305,
renovatio imperil, 17, 19, 26
309ft, 365£; Nicholas 1,378; Sergius,
reconquests (of Justinian I), 16ff., 20f.
318; Severinus, 302; Theodore, 306f.,
relics, cult of, 86,405£
3O9f.; Vitalian. 60, 311, 3 l 4
res private, 96ff.
population transfers, 63, 71f., 145
respublica, 127
possessor, 134,141
resources (agricultural and mineral),
potentlores, 27
10-13
power elite, 395ff.
Rhodes, 52, 55, 63, 80, 234. 269. 338
Praefectus Augustalis, 177, 210
Rhodian Sea Law, 254
praepositus, 174.181. 183. 189, 192
rhogai, 223. 225, 242, 245
praescrlptiofori, 270
Rizokopos, 169
praeses, 203
Romanitas, 279
praetorian prefects, 195-207, 223
Rome, 17. 49. 56, 176. 286, 314
praitores, 276
Rouchinai, 331
precedence (system of). 388ff.
Rum, 64
Priene, 108,118,226
Russian steppes, 45
primiscrinius, 273f.
Rusticus (sacellarius), 184
princeps, 273f.
principales, 97 Sabbatians, 342
Priscus, 167 Sabellians. 339, 342
Priscus of Panium, 279 sacellariusIsakellariosIsakellion, 74, 18 Off.,
proceres sacri palatii, 166 192ff., 296f.
Procopius, 15, 21, 29, 217, 252, 431 Sahara, 15
procuratores/epimeletai, 175, 178 Sahin, 46
prooemium (of Ecloga). 266-9. 272. 274 Sahrbaraz, 46
proskynesis, 88 saints: Artemius, 116, 273, 347, 356, 363,
prostagmata, 261 400, 423. 434; Catherine. 409;
protectores et domestic!, 210 Demetrius. 39, 45. 214, 320, 409;
protokagkellarios, 274 George, 423; Ioannicius, 335; John the
protonotarioi, 18 Iff. Theologian. 117; Philaretus. 132. 138.
protosecretarius, 273 154; Stephen (chapel of), 41, 284;
protospatharioas, 74. 319. 393 Theodore (Tiro), 95,117; Therapon,
protovestiarios, 183, 193 356,363,400
provinces (late Roman), 217ff., 227ff. Salona, 176
Prusa, 102, 218 Samaritans, 345
Prymnessus, 109 Samo, 46
Psalter, 427, 435 Samos, 217
Psamathion (monastery), 75 San Vitale (Ravenna), 407
Pseudo-Dionysius (Areopagite), 23. 342 Saracens, 248
Pseudo-Ephraim, 368 Sardinia, 148, 209f.
Pseudo-Methodius of Patara, 367f. Sardis, 83, 103, 107f., 118, 226, 239
Pyrrhus (Patriarch of Constantinople). 49, Sassanid(s) (see also Persians). 20. 45. 50.
57, 3O5ff., 315 102. 347, 352, 355
Satala. 102f.. 107
quaestor, 185, 211, 213, 269, 271. 274 Scamares, 31
quaestura exerdtus, 12, 177, 210, 217, Schemarion, 58
251, 269 scholai, 202, 210, 217f., 226. 391, 393
Quartodecimans/Tessarakaidekatitai, scholastikoi, 260, 265, 268, 274
342 Scotia (i.e. Ireland). 39
Questions and Answers, 370, 43Iff., 438 Scotland, 15
 Index 491
scribones, 190, 391, 393 Soviet historians, 27
scrinla, 174, 1771, 180, 188, 193, 196, spatharius/spatharocandidatus, 83.185,
200, 270 202, 205, 391, 393f., 398
Scripture, 369 Spain, xxiii, 21,34,209
Scythians, 71, 209f., 269 spectabihs, 161ff.. 165, 169, 206, 269
Scythopolis, 239 Spiritual Meadow (of John Moschus), 306,
Sea of Azov, 47 423, 430
Sebaste, 107 state: structure and form of, 14f., 22-4;
Sebastea, 102, 107 fiscal apparatus, 125.173-201; ideology
Sebastopolis, 234 and authority. 449flf.; and taxation,
Sebeos, xxiii, 167 452ff.; in the west. 456; in the east.
Second Coming, 437 4561T.
secretarius, 273f. Stephanus {antecessor), 265
Secret History, 16 Stephen (sacettarius), 74
sekreta, 180, 182, 193. 202. 204. 206. Stephen of Antioch (abbot), 321
212 Stephen of Dor, 309, 315
Seleucia, 176, 197.224 Strategikon (of Kekaumenos), 400
senatorial 'class', 129f., 160-72, 389ff., Strategikon (of Maurice), 222, 241, 252.
395ff., 444 268
Serbs, 47, 350. 352 strategos, 62, 72, 82f., 88, 202, 204, 212
Sergius (Archbishop of Cyprus), 306 strateia, 392
Sergius (epi tes trapezes), 168 stratelates. 165,195, 393ff.
Sergius (magistros), 319 stratiotikon, 180ff., 204, 206, 212, 241
Sergius (son of Mansur), 193 stratores, 391
Sergius ofjoppa, 302. 309 Strymon region, 71
Sergius (Patriarch of Constantinople), 4If., Suania, 20
48f., 57, 300ff., 309,315, 320 subadiuva, 273
Sergius (strategos), 84f., 442 Suevi, xxiii
Sergius (ex-consul), 192 Sufetula, 57
servus Christi, 72 Suleyman (Caliph), 83
Severi, 67 Suleyman (commander), 83
Severus of Antioch, 297. 339 superindictiones, 178
Short Chronicle (of the Patriarch susceptores/hypodektai, 178f.
Nicephorus), xxii Sykai. 75
Sicily, 21, 59f., 142, 148, 211, 229, 313 Syllaeum, 109
Side, 108 symbolic universe, 24f., 325f.. 348. 403
silentiarii, 391, 393ff., 398 symbolism/symbolic systems, 420f.
silentium, 271 Synnada, 107
silk production, 235-8 synodika, 311
Sinope, 98, 107 synone, 142, 147-50, 231
Sirmium, 32, 44 Syracuse, 60
Siroes {see Kavadh) Syria. 1, 20, 28, 31, 35f., 42f., 46, 49f.,
Sisium, 80,107, 109 56, 69, 209, 249, 286f., 301, 303f..
Sklavinlai, 44, 56 379
Slavs, 21, 32, 35, 37, 42ff., 52. 56, 66f., Syriac (Life of Maximus Confessor), 303
144f., 234ff., 247f.
Smbat Bagratuni, 321 Tabari,xxiv,112f.
Smyrna, 63,107, 110 tabellarioi (see nomikoi)
social formation (and mode of production), tagma, 202
451ff. Taktikon Uspenskij, 190, 194, 202-6, 223,
society (chief characteristics and structure 276
of), 13ff.. 26-31.125ff.. 153-60 Tarenta, 107
soldiers (in political context), 371ff., 44Iff. Tarsus, 43, 103, 107
Solomon (magister milltum), 210 Taurus, 13, 220
Sophronius (Patriarch of Jerusalem), 49f., taxes/taxation (see also capitatio/iugatio,
295f., 301, 306, 309 coemptio, synone), 28IT. 36, 141-52
492 Index
terrltoria (municipal), lOOff., 141 topoteresta, 212
Tervei (Khan of Bulgars), 76f., 80, 84f. tourmai, 212
Thebaid. 209f. trade, llf.
thelema, 49 Tralles, 159
themata/theme system, 35, 202(T., 208f., Trajan (patrikios), 425
212ff. tractores/trakteutai, 178,196, 200, 391
Theodore (ajw /ii/pafon), 197. 199 Triachia (Cyprus). 333
Theodore (llloustrlos), 196 Tribonian, 16
Theodore (ex-consul), 192 tributary mode of production. 451,
Theodore (commander), 46 453f.. 456-58
Theodore (Bishop of Pharan), 300. 309, Trinity, 69
315 Tripolis/Tripolitania, 54. 70.210
Theodore Calliopes (exarch), 310 Troilus [sacellarius), 168
Theodore (anti-Patriarch of Alexandria), Troullos (imperial palace), 73, 315
287 True Cross, 43, 46, 304
Theodore Spoudaios, 273 Turkey, 236
Theodore (son of John the candidates), 168 Turkic Reoples/Turks, 21,158, 355
Theodore the Studite, 149f., 335 TUrkmen, 107,121
Theodore Trithyrius, 184 Tyana, 80, 107, 109
Theodore (the Sygkellos), xxv Tyre, 10, 239
Theodore (Patriarch of Constantinople), 68, Typos (of Constans II), 57f., 67, 255, 261,
314f. 3O9f., 311, 364f.
Theodosian walls, 115
Theodosiani, 49, 301 c
Umar H (Caliph), 83. 87
Theodosius (brother of Constans II), 59, 69, c
Uqba (general), 69
312
Theodotus (general, logothete and abbot),
Valentinus, 54, 61, 3O5f., 370
74. 193. 394 Vandals, xxiii, 15,19-21, 72,340
Theodosioupolis/Erzerum, 103, 312 Venetia, 32
Theophanes Confessor, xxi£, 56,69,122,
vestiarion, 180,182f., 186,190, 193f.,
214,243,340 206
Theophanius (monk), 368f., 437 vestitor, 3931T.
Theophilus (antecessor), 265 vicarii, 178,207,209, 269
Theophylact Simocatta, 425 Victor (Bishop of Carthage), 307
Theophylactus (hypatos), 198 vigla, 202
Thera/Therasia, 88 village, 1321T.
Thessaloniki, 21, 35, 45, 60, 66, 71, 176f., vindtces, 96L, 178f.
214. 442 Virgin (Theotokos), 38, 356f., 409, 415,
Thomas (ex-consul), 192 429
Thrace, 10f.. 21, 48, 71, 124, 209, 213ff.,
Visigoths, 15, 21. 34, 341
218, 220, 229, 234 Vota, 334
Thracianus (exercitus)/
Thrakesianoi/Thrakesion, 159, 212, 214,
220,225,317 Wallachia, 67
Three Chapters controversy, 24, 289, 298
Tiberias, 303 Xerolophus (in Constantinople), 75
Tiberius (son of Constans II), 59, 68
Timothy of Constantinople, 342 Yarmuk (battle), 50, 216, 219
titles (senatorial/imperial see precedence) Yazid I (Caliph), 64
This, 108 Yazid H (Caliph), 87
Tmolus (see Aurelioupolis)
Toledo. 341 Zacharias (protospatharios), 74. 319
Tome of Leo, 312 zone (cingulum), 392
tomos (of Philippicus), 321 Zwangswirtschaft, 28
topoi (legal/legislative). 267. 278 zygostates, 190