Christians in al-Andalus

(711-1000)
 CULTURE AND CIVILIZATION
IN THE MIDDLE EAST
Series Editor
Ian R . Netton
Uniuersity if Leeds

This series studies the Middle East through the twin foci of its diverse
cultur es and civilisations. Comprising original monographs as well as
scholarly surveys, it covers topics in the fields of Middle Eastern literature,
archaeology, law, history, philosophy, science, folklore, art, architecture and
language. While there is a plurality of views, the series presents serious
scholarship in a lucid and stimulating fashion .
Christians in al-Andalus
(711-1000)

Ann Christys

I~ ~~o~~~;n~~~up
LONDON AND NEW YORK
 First Published in 2002
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon, OXI4 4RN
270 Madison Ave, New York NY 10016

Transferred to Digital Printing 2007

http://www.roudedge.com

© 2002 Ann Christys

Typeset in Baskerville by LaserScript Ltd, Mitcham, Surrey

All rights reserved. No part of this book may be reprinted or reproduced or utilised
in any form or by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying and recording, or in any information
storage or retri eval system, without permiss ion in writing from the publishers.

British Library Cataloguing in Publication Data

A catalogue record of this book is available from the British Library

Library ofCongress Cataloguing in Publication Data

A catalogue record for this book has been requested

ISBN 0-7007-1564-9

Publisher's Note
The publisher has gone to great lengths to ensure the quality of this reprint
but points out that some imperfections in the original may be apparent
 ...
Contents

Preface and Acknowledgements VI

Abbreviations VIII

List of Rulers x

Maps XII

CHAPTER ONE: lntroduction

CHAPTER lWO: Cordoba and Toledo 14

CHAPTER THREE: News from the east in the eighth-century

chronicles 28

CHAPTER FOUR: The martyrs of Eulogius 52

CHAPTER F1VE: Two more martyrs of Cordoba 80

CHAPTER SlX: Recemund and the Calendar of Cordoba 108

CHAPTER SEVEN: The Arabic translation of Orosius 135

CHAPTER E1GHT: Sara the Goth and her descendants 158

CHAPTER N1NE: Afterword 184

Notes 187
Bibliography 207
Index 226

v
 ---
Preface and Acknowledgments

T his book is a version of my doctoral thesis, which set out to look at the
christian response to the islamic conquest of Hispania. Although there
is little that addressed this question directly, there is a small but important
collection of texts written by or attributed to christians living in al-Andalus
up to the collapse of the Umayyad caliphate. I used these as the basis for a
series of studies of chronicles and calendars, hagiography and histories that
cross the frontiers of Hispanic historiography between christians and
muslims, north and south, Latin and Arabic. Few of these texts are well
known, particularly by medievalists working on areas outside Spain and
Portugal. They have not been considered as a body since Franciso Simonet
wrote his Historia de los Morarabes in the 1870s. Recent developments in the
reading of early medieval authors have opened up the writings of the
christians of al-Andalus to a new interpretation which reveals the variety
and complexity of the christian experience.
During the years when the thesis was 'nearly finished ' I received support
and helpful criticism from many people. I would like to thank the
medievalists of the School of History of the University of Leeds, especially
Wendy Childs, as well as Mayke de Jong, Rosamund McKitterick, Walter
Pohl and their students in Cambridge, Leeds, Utrecht and Vienna, Jinty
Nelson , Barbara Rosenwein and her colleague Zuhair Ghazzal, who
rescued the last copy of the Arabic version of Orosius' Seven Books rifHistory
Against the Pagans from the publisher's basement in Beirut and the
Historians of Medieval Iberia, particularly Roger Wright. I have been
greatly assisted by the staff of the Brotherton library, especially the
InterLibrary Loans office, the Biblioteca Nacional, Madrid and the former
Escuela Espanola de Arqueologia in Rome and its director Javier Arce.
John Wreglesworth read the manuscript with exemplary care and raised
enough questions for another thesis. Richard Hitchcock examined the
thesis and suggested publishing it more or less in its original shape . Peter
Brown nobly read the thesis before breakfast and advised me how to make

vi
 PREFACE AND ACKNOWLEDGMENTS

it easier for a reader to negotiate. I am very grateful to both for their

encouragement.
My debt to Ian Wood goes back to the first few weeks of my time as an
undergraduate in the School of History, where I was a refugee from the
unwarranted certainty of medicine. Ian's approach to the great unknown
of the early Middle Ages, based on imaginative reading of the way early
medieval authors had recorded their world rather than on explanation of
what actually happened, was a revelation. I am grateful to Ian for
humouring my obsession with the history of al-Andalus - almost the only
corner of the early Middle Ages on which he is not himself an authority -
and for giving me so many opportunities to develop as a historian. This
book is dedicated to him .
Finally, although he should perhaps have come first, I should like to
thank Roger Collins for writing EarlY Medieval Spain: Unity in Diversity
400-1000, which I read after my first visit to Spain in 1983 and the rest,
I hope, is history.

vii
 -~-

Abbreviations

AASS: (1634-) Acta Sanctorum quotquot tot orbe coluntur Antwerp Brussels:
Societe des Bollandistes .
Al-Andalus: Revista de las Escuelas de Estudios Arabes de Madrid y Granada
Madrid: CSIC .
Analecta Bollandiana: Brussels: Societe des Bollandistes.
Anuario: Anuario de Estudios Medievales Barcelona: Universidad de Barcelona.
Al-Qantara: Revista de Estudios Arab es de Madrid Instituto «Miguel
Asin Palacios» del CSIC Madrid: CSIC.
BL: British Library manuscript.
BN n.a.lat: Paris Bibliotheque Nationale nouvelles acquisitions latines.
CCSL: (1952-) Corpus Christianorum, Series Latina Turnhout: Brepols.
CSIC: Centro Superior de Investigaciones Cientificas.
CSM: ed. Gil Fermindez,J. (1973) Corpus Scriptorum Murarabicorum 2 vols
Madrid Instituto «Antonio de Nebrija»> .
CHE: Cuadernos de Historia deEspana Buenos Aires: Universidad de BuenosAires.
DHEE: cd. Aldea Vaqu ero, Q, Marin Martin ez, T. and Vives Gatell,
J. (1972- 5) Diccionario de Historia Ecclesidstica de Espana 4 vols, Madrid:
CSIC In stituto Enrique Florez .
EI: new edn Gibb, H.A.R. et al (1960-) Encyclopaedia if Islam Lciden
London: Brill.
The Formation of al-Andalus: Marin, M ., Fierro, M .1. and Samso,
J. cds (1998) The Formation if al-Andalus 2 vols Aldershot: Variorum vol I
Hi story and Society vol 2 Language, Religion , Culture and Sciences.
Conversion and Continuity : Gervers, M . and Bikhazi, RJ. eds (1990)
Conversion and Continuity: Indigenous Christian Communities in Islamic Lands,
Eighth to Eighteenth Centuries Toronto: Pontifical Institute.

viii
 ABBREVIATIONS

The Legacy of Muslint Spain: S.K. jayyusi ed (1992) The Legacy if

Muslim Spain Leiden, New York, Cologne: Brill.
Historians : Lewis, B. and Holt, P.M. eds (1962) Historians ifthe Middle East
Oxford: Oxford University Press.
MBN: Madrid, Biblioteca Nacional.
MGH: Monumenta Germaniae Historica.
AA : (1877-) Auctores Antiquissimi Berlin .
Capit : (1883) Capitularia Regum Francorum.
Concilia: (1906) Concilia Hannover Leipzig.
Epistolae : (1887-) Epistolae Merovingici et Karolini Aevi Berlin.
Poet. lat: (1881-) Poetae Latini Karolini Aevi
SS : (1826-) Scriptores Hannover Leipzig.
SS Rer Ger: (1840-) Scriptores Rerum Germanicarum Hannover.
PL: ed Migne,j.-P. (1841-64) Patrologia Cursus Completus, Series Latina, Paris.
Settirnane : Settimane di studio del centro italiano di studi sull'alto medievo Spoleto.
Medieval Christian Perceptions: Tolanj.Y. ed (1996) Medieval Christian
Perceptions ifIslam:A Book ifEssays New York London: Garland Publishing.

ix
 -
List of Rulers

Urnayyad Enlirs and Caliphs

'Abd al-Rahman I 756- 88

H isham I 788-96
Al-Hakam I 796-822
'Abd al-Rahman II 822-52
Muhammad I 852-86
Al-M undhir 886-8
'Abd Alla h 888-912
'Abd al-Rahma n III 9 12-6 1
AI-l:Iakam II 961-76
H isham II 976-1009, 1010-1 3

Kings of the Asturias

Pelayo 718/22?- 37
Favila 737- 9
Alfonso I 739- 57
Fruela I 757-68
Aurelius 768-74
Silo 774-83
Mauregatus 783-8
Vermudo I 788-9 1
Alfonso II 791- 842
R amiro I 842-50
Ordofio I 850-66
Alfonso III 866-910

x
 LIST OF RULERS

Kings of Leon

Ga rcia 910-14
Ordofio II 914-24
Fruela II 924-5
Alfonso IV 925-30
R amiro II 930-50
O rdofio III 950-5
Sancho I 955- 7, 960- 7
O rdofio IV 957-60
R am iro III 967-84
Vermudo II 982-99
Alfonso V 999- 1027

xi
 A
@
Jliiiill

ATLANTIC
OCEAN OC0
o
Cb
Mediterranean
Sea

_
Muslim
Hlspan la km I SO

Map 1. Christian and Muslim Hispania

 A Roncesvalles'" '-"'-.......... i Pic d'Orty ~
@
aiiiiI& ...\,'""'-"" /"\.A;.p r ..f r>.
~
~e:; Ao." , / ~ Ao.
~ ;
1> / '
A.. .e Pic du M idi
/ ~ ~-~J' ( ~"~ -";
/ '-, S A.:
A ,/
Pamplona . ~ ~ ~«. ~ V1 ..,"'\ A..A..
~ / ." d'O""o:.
-st:l' JIc , Ao..... .. ~
,-,,__-,/
"'
r' . / " " Cal de
'- .... Somport
~
N A v A R E
Burgui /"\ A..

o, km 10
,

• San Juan de la Pena

Map 2. Eulogius's Itinerary in Navarre

 A
@
aiiiiil

GALIC IA
. Santiago

· Silos

R D /lero
~

AA
o1
km so
,
AA
Map 3. North -west Hispania in the Ten th Century
 ---
CHAPTER ONE

lntroduction

'Although formerly Spain abounded plentifully in every liberal art

and in each one those thirsting for the fountain of knowledge were
devoting themselves everywhere to the study of letters, this study
along with the arts vanished entirely when she was inundated by the
forces of the barbarians. And so assailed by necessity both writers
were wanting and the deeds of the Spanish perished in silence'. I

T hese words, written in the twelfth century at the monastery of Santo

Domingo de Silos, south of Burgos, convey the dramatic effect of the
islamic conquest of Hispania on the way her history would be written. The
monk was exaggerating the impact of the conquest for anti-muslim
polemical purposes. The deeds of the christians did not altogether perish in
silence. Yet the events of 711 set christians on one side or the other of a
geographical and ideological frontier between Christianity and Islam
which has dominated the historiography of the peninsula. In the
unconquered northwest, propagandists for the emerging kingdom of the
Asturias practised a rhetoric of resistance to the invaders . 'The Saracens
took over the kingdom of the Goths, which even nowadays they still possess
in part. And the christians have battles with them day and night, and are in
daily conflict, but they cannot take the whole of Spain from them' .2 Thus
wrote a chronicler at the monastery of Albelda in the ninth century. The
author of the Chronicle if Alfonso III, written at about the same time, made
the defeat of Rodrigo, the last Visigothic king, the pre lude to a slow but
inexorable Reconquest of the peninsula, initiated by the Asturian kings
who claimed to be the heirs of the Visigoths. This Reconquest began with
the famous (and perhaps legendary) victory of Pelayo over the Arabs and
their Berber allies at Covadonga. Later christian historians echoed
Asturian triumphalism when writing about the contest between christian
and muslim Spain. Although the word 'Reconquest' was first used in
the twelfth century, and it was only at this period that it began to take on
 INTRODUCTION

the ideology of Crusade, the battle of Covadonga was placed at the head of
a chain of victories which culminated in the fall of the kingdom of Granada
in 1492 and the final liberation of the christians under muslim rule.
The christians of al-Andalus, living on the 'wrong' side of the frontier,
seem to have been consigned to a footnote to the history of al-Andalus. In
the standard histories by Conde," Dozy" and Levi-Provencal," the
indigenous population are incidental to the islamic conquest and the rise
and fall of the Umayyads, Almoravids and Almohads who dominated the
peninsula in turn. Kennedy's recent history of Muslim Spain and Portugal"
devotes three pages to the christian population of al-Andalus in the period
between 711 and the collapse of the Umayyad caliphate at the beginning
of the eleventh century. These works accurately reflect the bias of the
sources. Whilst the christians of al-Andalus were not entirely silenced by
the conquest, they wrote very little. The last chronicle to be written in
Latin dates from 754 . A century later, Eulogius documented the martyr
movement which erupted in Cordoba in the 850s, when christians went to
their deaths for blaspheming against Islam. Very little was written in Latin
in al-Andalus after this. Nearly all the evidence for the history of
al-Andalus is found in works written in Arabic. Historians writing in Arabic
were not interested in the native population and found few reasons to refer
to them. There are accounts of conflicts on the frontiers of al-Andalus
which involved both christians and muslim rebels .' Legal texts dealt with
some of the problems posed by these conflicts, as well as the relations
between muslims, jews and christians and between 'old' muslims and
converts - although most of the legal norms they established did not relate
specifically to al-Andalus." Few christians were mentioned by name. Many
of them adopted Arabic names, and can be identified in the sources only
where a name is qualified by a title such as 'the bishop' or 'the count'.
These men are minor characters in the histories of the Arab governors who
ruled al-Andalus from the conquest to 756, and of the Umayyad dynasty
which succeeded them.
Basing their interpretation on these sources, historians have assumed
that christian society did not survive the impact of Islam? and that the
disappearance of the christians from the historical record must be
explained either by emigration or by conversion to Islam. Emigration will
be one of the themes of chapter five. The meaning of conversion in the
early Middle Ages is obscure. It does not seem to have been a question of
individual conscience, but of social integration. Bulliet defined conversion
to Islam in the following way: 'a convert became a muslim . . . by reciting a
creed, in a language he may not have understood and without necessarily
understanding much of what it implied, learning new ritual practices,

2
 INTRODUCTION

donning Arab dress, and adopting a new Arabic name, often used
alongside his old one . In this way he clearly separated himself from his old
religious community and made himself acceptable to the muslims in his
vicinity' .'? Estimates for the number of christians who took this step vary.
Epalza concluded that most had converted by the end of the eighth
century. I I In the cities which still had bishops, he argued, it was possible to
remain a christian, but the majority had no option but to convert to Islam.
Yet only the sees of Zaragoza, Almeria, Cuenca, Guadix and Siguenza
disappeared entirely. Reilly estimated that christians may still have made
up thirty per cent of the population of the peninsula in the eleventh
century. 12 Bulliet, working from the genealogies of the notables of
al-Andalus, collected in the biographical dictionaries, calculated that the
number of conversions peaked at the beginning of the tenth century; half
the christians in al-Andalus had converted to Islam by the middle of the
tenth century, and eighty per cent by the eleventh.'! He assumed that all
the subjects of the biographical dictionaries were muslims by the time they
were eligible for inclusion, but that some of them were of christian origin .
He dated the conversion of each family by the first appearance in a
genealogy of a name which looked Arabic, although it was not always easy
to distinguish between muslim, Berber and christian names. There are
several problems with this approach. In order to be able to draw general
conclusions from a sample, the data must be accurate, and the sample
representative of the population in general and of adequate size. Bulliet's
study fulfils none of these criteria. He looked at 154 genealogies from five
sources. Nearly all of his subjects were active in Cordoba. Since the
biographical dictionaries included only public figures" - five percent of
the population at most - Bulliet's approach is rather like making
conclusions about the population of Britain in the twentieth century from
a study of W'ho 's W'ho. The genealogies show an accelerating trend towards
conversion in the upper echelons of society and may indicate that
conversion was the christian's main route to such eminence. These
conclusions should not be applied to the population as a whole.
Conversion has always been a major theme in the story of al-Andalus.
In the nineteenth century, Spanish scholars assumed that converts formed
the backbone of Andalusi society. Spanish Arabists of the school of Codera
(1836-1917), Ribera (1858-1934) and Asin (1871-1944), influenced by
contemporary currents of Spanish nationalism, treated al-Andalus as a
special, and of course superior, manifestation of Islam which was
independent of the rest of the islamic world." The last twenty years have
seen an explosion of interest in the history of al-Andalus, particularly
within Spain. A collection of many of the most important articles on this

3
 INTRODUCTION

period, translated into English, has recently been published." Modern

historians have rejected the concept of 'Spanish Islam'. Influenced by
the work of Guichard, 17 many see in the islamic conquest of Hispania the
imposition of oriental tribal patterns of settlement. It is assumed, although
there is little evidence for this, that the indigenous population were
absorbed into this new social structure. Neither in 'Spanish Islam' nor in
tribal al-Andalus is there room for the christians of al-Andalus to play any
role other than that of converts to Islam and clients of the new rulers .
Yet, more than one hundred years ago, Simonet made the 'Mozarabs',
as he called them, the subject of a large volume, of which a substantial part
deals with the first three centuries after the conquest. 18 Historians working
in this field owe an immense debt to him . Simonet wrote with the prolixity
of the nineteenth century but he was not short of material; although
manuscripts are still reappearing, especially in North Africa, most of the
evidence for the christians of al-Andalus had already been discovered .
Simonet's perspective was as deeply rooted in the nineteenth century as his
literary style. First and foremost 'a catholic, above being an Arabist, a
professor, or anything else', 19 he told the story of embattled christians under
alien rule who kept the flame of catholicism burning. Fifty years later,
Isidro de las Cagigas' account of the same period took a nationalist
perspective on the christians' struggle to maintain their identity'" Heirs to
the Asturian chroniclers of the ninth century, both men were working with
the same grand narrative, that Reconquest had freed the peninsula from
the muslim yoke. Similar preconceptions engendered the acrimonious
debate between Sanchez-Albornoz and Castro, as to whether Spain was -
as the former claimed - 'really' a catholic country and part of Europe, or
inescapably set apart by her islamic inheritance." The christian-muslim
frontier in the peninsula was compared with the northwest frontier of the
United States." Sanchez-Albornoz claimed that 'the history of no other
European people has been so decisively modified by a frontier as Castile '."
Combatants on both sides saw the peninsula as unique, a mind set which
Linehan argued is already recognisable in the writings of early medieval
Spaniards." Since the fall of Franco, the secularisation of society,
devolution of political power in Spain and European integration have
robbed this debate of its relevance. Further, the concept of the grand
narrative is losing its power. This alone would justify a re-examination of
the evidence.
It is unfortunate that methodological shifts mean that the christians of
al-Andalus have become, in some ways, less accessible to study in the
twentieth century than they were to Simonet. The opposition between
christianity and Islam remains crucial to Spanish historiography and it has

4
 INTRODUCTION

been given a new twist, which at first sight is purely academic and concerns
the importance of Arabic for the study of al-Andalus." Simonet read both
Arabic and Latin. His History if the Mozarabs deserved its subtitle: 'deduced
from the best and most authentic testimony of christian and muslim
writers'. Few later historians have been able to emulate him . Some were
not deterred by their lack of Arabic. Sanchez-Albornoz wrote many
volumes on this period, far too many to leave time to learn a difficult
language. His reliance on translation involved him in another dispute, this
time with Levi-Provencal, who edited many of the Arabic sources for the
history of al-Andalus, The modern student of al-Andalus cannot afford to
be on either side of this divide. Dependence on translation sometimes
creates additional difficulties for the historian, since translators have been
blamed for compounding the mistakes perpetrated by the editors of the
manuscripts. More important than this, where a work has not been
translated in its entirety, the same passages are quoted, often out of context,
by one historian after another. Some knowledge of Arabic is now the
starting point for work on many aspects of the history of al-Andalus. This
puts most modern scholars into one of two camps and the study of
christians and muslims who lived side-by-side in the early Middle Ages into
different university departments. On one side are the Arabists, who are
sometimes thought by historians to be insufficiently critical of their sources.
Those who are limited to the Latin sources, on the other hand, miss out on
the meat of the evidence. Only recently have historians begun to work
across this frontier, particularly in Spain. Further, the difficulty which
Arabic presents to European scholars in comparison with Latin fosters the
idea that the Arabic histories are alien to twentieth-century minds in a way
that the Latin material is not. This may have more to do with present
mistrust of Islam than with scholars' ability to draw conclusions from
apparently unpromising material; after all, most historians of the early
Middle Ages have learnt to work with hagiography. The divide between
Islam and Christendom is long standing and has been sharpened in recent
years by acts of violence by islamic fundamentalists. It would be a mistake
to extrapolate these feelings back to the early Middle Ages. Although
anti-muslim polemic was written in the Middle East, and to a lesser extent
in Hispania, in this period, enmity between the two religions does not seem
to have crystallised until after the millennium." Yet Crusader ideology has
created a conceptual barrier which loads the interpretation of actual
medieval frontiers." Western historians conceive of Islam as monolithic
and, indeed, unaffected by history, so that observations of modern islamic
communities, particularly those regarded as 'primitive', such as the
Bedouin, are held to apply to their coreligionists of more than a

5
 INTRODUCTION

millennium earlier." Nor is it possible to exonerate all Arabists from the

charge of orientalism, of expecting, and finding, the exotic in their
material. Thus, Hispania can often be represented as having changed
almost overnight from a christian state, heir to the Roman Empire into
something altogether richer and stranger, a 'garden of poets'f? which does
not fit into the narrative of European history"
Some modern historians of this period, most notably Collins, have
rejected Arabic historiography as completely unreliable." It is true that the
Arabic sources for the first centuries after 711 are collections of stories
which are often clearly legendary. Little survives from before the fall of
the caliphate, and much of the textbook history of al-Andalus comes from
the collection of stories assembled by the Moroccan historian al-Maqqari
in the seventeenth century" The value one attaches to such late works
depends to a large extent on the credence one is prepared to give to their
authors' claims to be no more than the transmitters of the exact words of
their predecessors. The influential eastern historian al-Tabari (d.923),
stated the principles of this type of Arabic historiography: 'knowledge
concerning reports of the men of the past and current news about men of
the present is not obtained by one who has not witnessed these men, or
whose lifetime does not reach back to theirs; [knowledge is only obtained]
by the statements of reporters and transmitters, not by rational deduction
or intuitive reference'r'" Yet incompatible accounts of the same events were
sometimes copied into the same manuscript without comment, and stories
appear in different versions in works whose authors claim to be quoting the
same authorities. Faced with conflicting versions of events, al-Maqqari
confessed his inability to determine which of his sources were telling the
truth. There is no surviving charter evidence to validate the narrative.
Al-Tabari's statement is also disingenuous, for judicious editing was
practised." Writers of history were often companions of the rulers and
could not afford to jeopardise their position by impartial scholarship. The
title of a work - TIe Brilliant Qjlalities if the Umayyads (which does not
survive) ~ said to have been written by Qasim ibn Asbagh, the supposed
translator into Arabic of Orosius' Seven Books if History against the Pagans,
gives the flavour of this writing. In this respect, the Arabic histories are not
very different from contemporary Latin chronicles. Medieval Spaniards
certainly seem to have thought this to have been the case. Jimenez de Rada
(archbishop of Toledo 1207-47) used Arabic as well as Latin sources in his
Arab History.35 Recently, historians have begun to evaluate the Arabic
sources, to compare different versions and to rank them in order of
credibility." Valuable evidence for the christians of al-Andalus is scattered
through these texts and it deserves closer examination. Similar attention

6
 INTRODUCTION

should also be paid to the much smaller volume of material written III
Arabic by the christians themselves.
This book is a history of the christians of al-Andalus which places special
emphasis on their own words . It concentrates on a number of texts which
have been attributed to these christians, or their descendants, living
in al-Andalus before the fall of the Umayyad caliphate. It is arranged in
chronological order, as far as this can be determined, but it is not a
narrative. Although the chosen texts all address the impact of the islamic
conquest, however obliquely, they cannot be seen as a concerted literary
response to Islam. They were written under different circumstances and it
is not surprising that they have different perspectives. Each is judged as far
as possible on its own terms, without worrying that the pieces of evidence
may be incommensurable. The result is a series ofbrieflives of a small cast
of characters, which deepen our understanding of the situation of the
christians of al-Andalus.
Most of them lived in Cordoba, the Umayyad capital. A short
introductory chapter contrasts the splendour of Cordoba with the fate of
the Visigothic capital, Toledo. The study proper begins with a number
of Latin sources, some of them familiar, others less well known, and it is
extended by an examination of the translation of Latin sources into Arabic,
and by looking at the work of a muslim historian who wrote about his
christian ancestors. Since the sources cross the divide between Latin and
Arabic, the historian should do the same . The Arabic histories, although
fraught with pitfalls, may be approached with the apparatus of textual
criticism that has been applied to Latin sources for the peninsula, and for
early medieval Europe in general. In order to evaluate a piece of evidence,
it is important to know something about the manuscript from which it
comes, and its textual transmission . How closely do the surviving
manuscripts resemble the original versions of these texts? Is it possible to
reconstruct the thought world of the author? What literary strategies were
available to him or her? How did the audience for the text affect the
rhetoric employed which ' ... is not an element of distortion that the
historian's scalpel should neatly cut away from his evidence, but part of the
reality in question; its content and influence have to be evaluated
separately in each case' .37 It is hoped that this approach will help to
integrate the history of al-Andalus into the main currents of European
history. Textual analysis can be very dry, and the work involved may seem
out of all proportion to the result . As Sanchez-Albornoz remarked at the
end of a paper of 101 pages in which he concluded that Rodrigo, the last
Visigothic king, died in 711, but could not identify the site of his last battle:
'a minuscule fruit for a tree with so many branches'." In other cases, it was

7
 INTRODUCTION

a surprise to discover how much of the textbook history of al-Andalus owes

to the nineteenth-century giants Simonet and Dozy. It is time for a re-
evaluation of their ideas .
I refer to the native population under islamic rule as 'the christians of
al-Andalus'. It is always difficult for the historian to define her terms.
Seeking to avoid burdening the reader with definitions and redefinitions,
one stretches a limited vocabulary to cover a range of meanings which
changed during the period for which they are being used. I have avoid ed as
far as possible the term 'Mozarab' [musta'rib/ Arabicized]. It. is an
anachronism in the first three centuries after the conquest, since it first
appeared in eleventh-century texts. Using 'Mozarab' when referring to the
early islamic period prejudges one of the questions at issue, the degree of
acculturation.P I make the distinction between those who remained
christian and those who converted to Islam only where their religious
affiliations are clear. The use of 'islamic' for the conquest, political
institutions, etc . does not imply that all those involved were muslims.
Al-Andalus refers to that part of the peninsula which was under the control
ofUmayyad Cordoba and its allies, at least in the recollection of the Arabic
historians. It is much less clear how we should refer to the peninsula as a
whole, since 'Spain' and 'Iberia' have unwanted modern connotations.
I have adopted 'Hispania', but used in its widest possible sense to include
the northern fringes of the peninsula as well as the provinces which the
Visigoths and the conquerors of 711 inherited from the Romans.
Each chapter of this study deals with a text or a group of related texts,
beginning with two Latin histories written some fifty years after the
conquest. They have been neglected until recently, when the longer of the
two, known as the Chronicle if 754 or the Mozarabic Chronicle, has received
some attention.t? As the name implies , it seems to have been written in 754
or shortly after, and is thus much earlier than the first accounts of this
period in Arabic. Collins based his history of the Arab Conquest on it, and
demonstrated how valuable a source it is. The Chronicle if 754 gives an
account of the 'ruin of Spain' but it begins with the accession of Heraclius
and incorporates Byzantine history and the islamic conquest of Syria. The
second chronicle, edited by Mommsen under the title Continuatio Byeantia-
Arabica" but now called the Chronicle if 741, is even more eastern in its
focus. Apart from some early chapters about Hispania, the Chronicle if 741
is a bri ef summary of the reigns of the Byzantine emperors and the first
caliphs. There is only one passage about Hispania after 621 and it seems to
have been copied from the Chronicle if 754. Previous commentators on the
two chronicles have passed quickly over the Chronicle if 741, moving with
obvious relief to the later chronicle. Collins dismissed the Chronicle if 741 as

8
 INTRODUCTION

'a confused and confusing work th at fails to int egrate the eleme nts of
which it is compose d in a workmanlike and satisfying way'.42If on e is using
th e eighth-century chro nicles to construc t a narrative of events in
H ispani a, this judgement is clearly valid. Yet ther e is scop e for a more
int er estin g int erpretation of th e way in which both the chroniclers retold
th e history of th e Byzantine Empire and th e Islamic conquests in the east,
which reh abilitates this material as on e aspec t of the authors' response to
events in the peninsula, and thus as part of the story which they were
trying to tell.
For almost a century after the chronicles were writt en there is little
evide nce for christian writing within al-Andalus, The one major exception
is a handful of lett ers written by bishop Elipandus of Toledo, evidence of a
controversy which occupied several of the best theological minds in the
west. In 785, Elipandus wrote a letter condem ning Mig etius, the lead er of a
radi cal sect which seems to have brok en away from the Hi spanic church in
protest at her failure to pr eserve the faith in the face of Islam. " Elipandus
attac ked M igetiu s by ridiculing his heretical views on the nature of the
Trinity and sta ted his own po sition that Christ had adopted his human
natu re. This original att empt to deal with a perennial pr oblem dr ew
Elipandus into disputation first with Beatu s of Lieban a in the Asturias, then
with th e C arolingian church and the pope. The Toledan s wer e trying to
hold onto their spiritua l authority in new political circumsta nces. If they
do cumented their responses to the conquest of 711 in an y more explicit
way, the texts in which they did so have perished .
Again st this background, the execution of forty-eight christians in
Cordoba between 850 and 859 for blasph emy against Islam looks like the
eruption of a volcano pr eviously thought dormant. The Acts of the martyrs
of C ordoba were recorded by Eulogius, who himself was martyred ,
emulating the saints whose passions he had written. Yet this is not an
instan ce of an oppressed people throwing off their chains. It is clear even
from th e writings of Eulogius and of his friend and biographer Alvarus ,
that the Cordoban martyrs were not typical christians, and their sacrifices
wer e condem ne d by the church hierarchy. Mo st of these saints were not
celebrated in their own tim e, and had to wait until the sixteenth century to
be rediscover ed. Eulogius dedicat ed mu ch of his writings to explaining how
the martyrs, who had suffer ed little persecution and performed few
mir acles, and who se pa ssions seemed difficult to write according to the
norms of martyr hagiography, were worthy of being rem emb ered . The
works of Eulogiu s and his circle make up almost two volumes inJuan Gil's
modern edition - a large but pro blem atic volume of material through
which it is difficult to glimp se the social milieu of his age.

9
 INTRODUCTION

Alvarus' complaint that the christians of al-Andalus were losing their

Latin is well-known:
'Alas, the christians do not know their own law, and the Latins pay no
attention to their own tongue, so that in the whole community of
Christ there cannot be found one man in a thousand who can send
letters of greeting properly expressed to his fellow; and there are
found crowds of people without number who can produce (?explain)
learnedly Chaldaic parades of words .. .'+4
Too much has been made of this passage, which is echoed only in the
works of Alvarus himself and of Eulogius. Some of the attention that has
been paid to these two men should be directed towards the reliability of
their testimony. Rereading the works of Eulogius in the light of Alvarus'
Lift if Eulogius and other evidence allows for such a reappraisal.
My fourth chapter presents two Cordoban martyrs who are not nearly
so well known as those documented by Eulogius. Their passions survive in
manuscripts which may date to within a hundred years of the events they
portray, making them in fact the most nearly contemporaneous accounts
we have of christians in Umayyad Cordoba. The Passion ifPelagius tells of a
boy of thirteen who preferred death rather than surrender his virginity to
'Abd al-Rahman 111. 45 The cult of Pelagius, sponsored by a king's daughter,
was adopted almost immediately by christians in the north, which is in
marked contrast to the Asturian dynasty's neglect of the ninth-century
martyrs. Far away in Gandersheim, Hrotswit wrote a poem about Pelagius.
Five years after Pelagius' death, Argentea attained her crown of martyrdom
in Cordoba, having travelled to the capital for that purpose.t" Both
passions, in spite of the hagiographical formulae which they employ, are
about Cordoba in a way that Eulogius' Memorial ifthe saints is not. Looking
at where and how these passions were read gives a new perspective on
Cordoba and the frontier between Christianity and Islam which conflicts
with the accepted view of the relationship between north and south.
In his recent history of Moorish Spain, Richard Fletcher called as one of
his key witnesses 'a Mozarabic christian cleric named Recemund, who
ended his life as bishop of the christian community of Elvira [near present-
day Granada]. Recemund had a successful career as a civil servant under
'Abd al-Rahman III; in 955-6 he had been employed on a diplomatic
mission to the German king Otto I ... in about 960 he commissioned a
work known to scholars as the Calendar if Cordoba'," It is rare to have
enough material to reconstruct the biography of an Andalusi christian.
Fletcher neatly summed up the standard view of Recemund, who has
come to epitomise convivencia : the peaceful coexistence of christians,

10
 INTRODUCTION

mu slims and j ews in the peninsula. Recemund's miSSIOn to Otto is

describ ed in the lift ofJohn of Gorze in Lorraine, who led an embassy to
Cordoba in the 950s .4B Liudprand of Cremona, a prominent memb er of
Otto's court, dedicated a Latin history to Recemund.t? Yet the evidence
which has been used to draw the portrait of Rec emund, found both in the
Latin and the Arabic sources, is a web whose tangles suggest that to link all
this evidence to just one man is to underestim ate the richness of Andalusi
christian society.
Recemund is just the kind of man to have been involved in the
translation into Arabic of the work of on e of Hispania's mo st famous
histori ans , Orosius. The Seven Books if Histories Against the Pagans, writt en in
41 7, is a history of the world which was very popular in the early Middle
Ages. 'Abd al-R ahman III was said to have received a copy of Orosius from
a Byzantine empe ror, and his son al-Hakam II to have commissioned the
translation of the work into Arabic for his famous libr ary. It is possible to
pla ce the tr an slation of Orosius in a wider context. This was not the only
text tran slated from Latin into Arabi c, although the extent of such
tran slation was obscured by the Reconquest and the return to writin g in
Latin. In the twelfth century, a number of Visigothic manuscripts wer e
glossed in Arabic'" and a Latin-Arabic Glossary was compiled to help
christians with their Latin. Christian s may have stopped using Arabic soon
after this. Thus it was easy for later histori an s of medieval Hi spania to give
the impression that the whol e islami c period had been no more tha n a
hiccup in the rise of Latin culture in the peninsula. In this light , the
bilingualism of the Andalusi christians condem ned by Alvarus was skated
over as mer e pragmatism, a temporary accommodation with the current
rulers. However, there is no eviden ce that christians living in al-Andalus
under the Umayyads saw their situation with so much foresight. Wh ether
they liked it or not, it looked as though the muslims were here to stay, and
so was th eir language. The extent of acculturation can be disputed ,
because very few manuscripts survived . Once they could no longer be read ,
they were useless, and vuln erable to charges of heresy. Man y were
destro yed . Ribera found the manuscript of an Arabic grammar in the
University Library in Valen cia, glossed in Catalan by its own er, who noted :
'a s it is writt en in Arabic, I have never found anyone who can read it. I am
afra id it might be the Alcoran of Mahomet '." A few christian Arabic texts
wer e taken to north Africa. A second Latin history translated into Ara bic
was discovered in Qayrawan .V and ther e are Arabi c translations of
scripture and of canon law. Recent studies of this material have shown how
mu ch of their literary herit age the christians of al-Andalus mad e available
in Arabic.f

11
 INTRODUCTION

Although the example of Recemund shows how some christians

prospered without abandoning their faith or their Latin culture, such
acceptance of a new status may have been the lesser road to success for the
indigenous population. For those Spaniards whose ancestors had appeared
in the Visigothic chronicles, their chief hope of remaining near the centre
of power was to convert to Islam and become dependants, or clients, of the
new rulers. It is difficult to say how many noble families made this
transition, since they ar e almost impossible to identify in the sources . Not
all the clients of prominent muslim families were christian converts, since
becoming a client was also the way to advancement for a poor man of
muslim origin. At the same time, very few of the new muslims preserved
the memory of their christian past. It cannot be said whether this was due
to a deliberate denigration of their origins, or merely the absence of a
format in which they might be expressed. There is one important
exception, one scion of a convert family who did not forget his Visigothic
origins. He is the tenth-century scholar Muhammad Abu Bakr, author of a
History if the Conquest if al-Andalus'" which is unique in describing the role of
the indigenous population in the islamic conquest, and thus in being able
to shed some light on the early medieval meaning of conversion and
assimilation. The author was a client of the Qurayshi family, the
descendants of the Prophet, and his given name is impeccably muslim.
Yet he adopted, or was given, the remarkable nickname or title Ibn
al-Qutrya, which seems to mean 'son of the Gothic woman'. The History
implies that this ancestor was a member of the Visigothic royal family
called Sara. The stories in Ibn al-Qjitlya's History about Sara, her uncles
and her descendants show how an accommodation between the Visigoths
and their conquerors may have been negotiated.

Most imperial histories are written by the conquerors, and the fate of the
subjected peoples has to be read from their perspective, which is one of
cultural superiority'? The Roman Empire was posited on the opposition
between 'Romans' and the 'barbarians', which continues to dominate
historiography of the Transformation of the Roman World. Charlemagne's
justification of military expansion with the rhetoric of mission became a
cliche of colonialism. Imperial attitudes to the subject peoples were not
monolithic, but in general, as the 'barbarians' were disparaged, so was
their history. Sometimes the conquerors, by introducing writing to
pre-literate societies, created a new dawn of history. The words of the
vanquished can be heard only faintly through the writings of the settlers
and the vagaries of oral tradition . Later generations of the conquered

12
 INTRODUCTION

peoples would filter their past through the constraints to the imagination
offered by new literary media." Spaniards in the New World authenticated
the historical memory of their subjects in order to make the exaction of
tribute easier. 57 The British in India rewrote history from spurious
genealogies and from both genuine and fabricated Mughal charters in an
attempt to situate the British settlement in an unbroken Indian tradition of
the payment of tribute to conquerors. 58 After independence, the Indian
ruling class found it difficult to escape the colonialist formulation of their
history. It is very difficult in these examples to reconstruct the subject
peoples' view of conquest.
The historiography of the islamic conquest of Hispania is different. The
incomers could not approach the natives from a position of cultural
superiority and the latter did not entirely lose their voice. There is no
evidence for mission ; this is perhaps explained by the fiscal status of the
vanquished, who paid more tax than muslims, but there may have been a
genuine feeling that the christians and jews of Hispania should be
respected as fellow monotheists, 'peoples of the Book' . It may also have had
something to do with the fact that in al-Andalus, as in the Middle East ,
islamic conquerors relied on native administrators'" because the political
institutions of the conquered peoples were more developed than those of
their new masters. The cultural achievements of Visigothic Hispania are
reflected in their influence on the rest of Europe.t" The conquerors, on the
other hand, were a mixture of Arabs and Berbers. Some of them were
muslims but others were adherents of the local religions of North Africa;
there may have been christians as well." Such a group was not in a position
to dominate the intellectual life of the peninsula. Confidence in the
superiority of islamic culture grew as a consequence of Cordoba's contact
with centres of learning in the east. The balance shifted with the adoption
of Arabic for all official transactions, and the patronage of islamic
scholarship in Cordoba. Although christians such as Recemund continued
to playa role, Umayyad Cordoba at its apogee under 'Abd al-Rahman III
and al-Hakam II in the tenth century was an islamic state in the west. After
the collapse of the caliphate, this process went into reverse . The
Reconquest, by restoring the primacy of Latin, ensured that some at least
of the christ ian texts written in Hispania in the islamic period were
preserved. This evidence gives a glimpse of the impact of conquest and
cultural change from the perspective of the subject people .

13
 ._-
CHAPTER TWO

Cordoba and Toledo

T he history of al-Andalus to the fall of the Umayyads is to a larg e extent

the history of Cordoba, and of that city's greatest glory. The Arabic
histori es and th e accounts of travellers to al-Andalus portrayed Cordoba as
on e of the wonders of the world .' Ibn Hawqal, who visited al-Andalus in
948 , wrote:
'There is nothing to equal it in the whole of the Maghreb [North
Africa and al-Andalus] , or even in Upper M esopot ami a, Syria or
Egypt, for th e number of its inhabitants, its extent, the vast area
taken up by markets, its cleanliness, the architecture of the mosques
or th e great number of baths and caravanserais. Several travellers
from th is city who have visited Baghdad, say that it is the size of one
of the quarters of that city. . . . Cordoba is not perhaps equal to half
the size of Baghdad, but is not far off being so. It is a city with a stone
wall, with handsome districts and vast squares."
Cordoba was an important Visigothic centre from 571, when Leovigild
took th e city from the Byzantines and established a bishopric, but it was
overshadowed by Toledo . In the eighth century, this position was reversed.
One of the first islamic governors established his capital in Cordoba soon
after th e conquest, perhaps as early as 717. The Arabic histories focus on
life in the capital. The Umayyads, rulers of al-Andalus from 756, app ear to
hav e conce ntrated their building programmes on Cordoba and its
immediate environs. Above all, they were celebrated by their eulogists
for their contributions to th e great mosque which is the epitome of
Umayyad Spain for modern visitors. Cordoba reach ed the height of its
splendour in the middl e of the tenth century. In 929, 'Abd al-Rahman III
(9 12- 6 1) cast off the nominal allegiance which previous Umayyad emirs
h ad offered to the Abbasids caliphs in Baghdad. H e ordered a
proclamation to be made in the mosque in Cordoba that he should
henc eforward be given the title of caliph, thus asserting spiritual as well as

14
 CORDOBA AND TOLEDO

tempor al power over his subjec ts. Some ten years later he began to build
the palace-city of Madinat al-Zahra', twelve kilometres outside the city.
The splendo ur of court ceremonial at Madinat al-Zahra' was describ ed
both in the Arabic sources," and in the Life ofJohn of Gorze, who went to
Cordoba as an ambassador in the 950 s. M osqu e and palace embodied the
pow er of the caliphate.
It is not known how much of Cordoba the Umayyads developed as the
concrete expression of their dominance, nor, indeed, how big the city was."
The Arabic sources are deceptively precise about size, but frustratingly
vagu e about topography. Cordoba has been estim ated to have as many as a
million inhabitants, and is depicted in a modern historical atlas as being
man y times bigger than contemporary cities in the rest of Europe." This is
the result of taking literally Ibn H awq al's statement that Cordoba was
nearl y half th e size of Baghdad , which begs another question ; the figure of
four million inhabitants often qu oted for Baghdad seems to be derived
from the hyperbole of the Arabic sources. A similar pro cess was at work in
the histories of al-Andalus. Writing in the fourteenth century, Ibn al-Kh atib
said th at tenth-century Cordoba was surrounded by a ditch and a wall of
some twent y-two kilometres long. This implies a city of approxima tely
5,000 hectar es, about eight times the area of the present city. A thirt eenth-
century historian, Ibn Idhari, claim ed that there were thr ee hundred bath s
and three thousand mo sques," but of these only three min arets remain .
The loss of Cordoba's splendour has been blam ed on the civil wars and the
collapse of the caliphate. During this unhappy period, Cordoba seems to
have melted away like the cities of'fable.? Mu ch of what we think we know
about Umayyad Cordoba comes from the collection assembled by
al-Maqqari, Al-Azmeh describ ed this as 'nostalgia at two removes';
rem ember ing a civilisation whi ch has long vanished through th e
vocabulary and sentiments of eastern po etry. In the work of al-Maqqari,
'Cordoba . . . is severally reduced to the general topoi of grandeur, verd ancy
and brillian ce, while, at the same time, illustrating the ruthl ess course of a
cr uel Fare'."
Even where the build ings or their archae ological tra ces remain , their
realities, albeit splendid, do not quit e live up to their literar y reputation .
Almost two thirds of al-Himyari's description of Cordoba is devoted to its
mosque, which 'o bviously function ed inter alia as a symbol '." One of the
longest acco unts of the construc tion of the grea t mosque of Cord oba comes
from the work of Ibn Idhari, who describ ed the additions which each ruler
had orde red to be made." H e claim ed to be qu otin g from earlier
a utho rities, particul arl y Ahmad al-Razi (d .955) whos e origina l work was
lost. Ibn Idhari's history of al-Anda lus is often qu oted , pr ob abl y

15
 CORDOBA AND TOLEDO

because of the seductive amount of detail he included, which is sometimes

more than there was in the sources from whom he claimed to have been
quoting, where these survive. Ibn Idhari's descriptions of the mosque
should be examined in conjunction with other accounts of the building.
Such a study has been carried out for the various descriptions of the palace
of Madinat al-Zahra', and shows that they do not concur with each other
or with the archaeological findings . As 'the tragic circumstances of
Madinat al-Zahra's destruction propelled it from an already fantastic place
of grandeur and magnificence into a truly mythological realm', II the details
of its construction became more fantastic with each retelling.!" The
descriptions of the mosque in Cordoba seem to have been elaborated with
details from descriptions of the mosque in Damascus. The actual building
which remains is undoubtedly impressive, and it seems that each Umayyad
ruler had to be associated with it in the work of their eulogists, if not in
reality.
Archaeologists have tried very hard to find the city of the written
sources, but succeeding generations are gradually, and with great regret,
reducing Cordoba to more modest proportions." In the nineteenth
century, the remains of Roman aqueducts well outside the city were
thought to be the Umayyad walls, rather as many Roman structures in
Spain have been reclassified in folk memory as 'Moorish'. Yet Ibn Hawqal
claimed to have walked round the city walls in an hour. This describes a
city of about the size of late medieval York, and may be nearer the mark.
Traces of a wall running alongside the waterway now known as the Arroyo
del Moro may date to the ninth and tenth centuries, although their
construction is very similar to walls built several centuries later, after the
christian Reconquest. If these walls marked the outer limit of the city in the
caliphal period, they enclose an area of only some seven or eight hectares
to the west of the palace. The palace and mosque, together with the
markets established along the Roman Cardo Maximus, occupied some
2.5 hectares. Thus the Umayyad capital seems to have been based on the
Roman city, although it was a little larger because it was extended towards
the river."
In order to vindicate Cordoba as a great metropolis it is necessary to
include what are always referred to as its 'suburbs'. Ibn Bashkuwal, in a
passage quoted by al-Maqqari, listed twenty-one such suburbs,'! and some
of them are mentioned in the Calendar if Cordoba. The recent archaeology of
Cordoba has been frustrated by rapid redevelopment; little time was
allowed for excavation and large areas of important Roman and islamic
remains have been destroyed. The full extent of Umayyad Cordoba will
probably never be known. Traces of street patterns have been identified

16
 CORDOBA AND TOLEDO

outside the walled city, altho ugh only on e of the se, al-Rusafa, ca n be
identified with on e of the suburbs named by Ibn Bashkuw al. 'Abd
al-Rahrnan I is said to have built a palace here, although its rem ains have
not been found. The rest of the city's hinterland was a rural landscap e of
garden s and orchards, dotted with the palaces of the nobility. Indeed , one
of the subur bs, Secunda, on the other side of the Guadalquivir from the
mo squ e, which later became known by antonomasia as 'the suburb'
[el Arrabal], is elsewhere referred to as a village. 16 Ibn Hawqal's picture of
Cordoba should be quoted in context. It is clear that he too was describing
a compac t city surrounded by a larger area mainly given over to
agriculture, wh ere the nobility had their country palaces:
The ruler of this city, 'Abd al-Rahrnan ibn Muhammad, found ed , to
the west of Cordoba, a city which he called Zahra', on the flank of a
rocky mountain with a flat summit, called Batlash; he brought
markets ther e and had bath s, ca ravansera is, palaces and parks built.
H e invited the people to live ther e and ordered that the following
proclamation should be issued throughout Spain: 'W hoever wished
to build a hous e, choosing a spot next to the sovereign, will receive
400 dirhams '. A flood of people rushed to build ; the buildings
crowde d togeth er and the popularity of this city was such that the
hou ses formed a continuous line between Cordoba and Zahra" . 17
Even if it is taken at face value , this description of ribbon development does
not imply th at the area between Cordoba and Madinat al-Z ahra' was
urbani sed in the commonly accepted sense. Ibn Hawqal went on to
describ e severa l cities in al-Andalus, many of them, like Cordoba, well-
populated and having walls, but all were 'surrounded by a vast rural area'
where the wealth of the city was cultivat ed. " Discussing Damascus in the
later medieval period, Lapidus argued that large villages in the agri cultural
hinterland of a city might be considered part of the metropolitan
conglome ration .'? It is not clear that this argument can be applied to
Cordoba in the tenth century. Although the mosque ma y have been
grander than anything in al-Andalus, the city itself was almost certainly
little lar ger than several other form er Roman cities in Spain. Merid a, for
exa mple, with its spectac ular Roman buildings and the Alcazaba, the
fortress built in the 830s , mu st have been eq ually impressive. Yet Merid a
and other cities wer e rarely menti on ed in the Arabic sources, and their
history in the early islamic period rem ain s obscure. Cordoba attrac ted
ambassado rs and merchants, and oth er visitors hop eful of making their
fortune, not becau se of its size, nor the splendour of its buildings, but
becau se it was the seat of pow er. Amongst those dr awn to Cord ob a, as we

17
 CORDOBA AND TOLEDO

shall see, were christians, some seeking worldly advancement, others

aspiring to martyrdom, who sought their fortune in heaven.

Why did the early governors of al-Andalus chose Cordoba rather than
Toledo? The former Visigothic capital in the centre of the peninsula was a
more obvious base than Cordoba in the south west. According to the
Arabic histories, Tariq ibn Ziyad, the conqueror of Hispania, made for
Toledo soon after his victory over Rodrigo, deputing Mughith al-Rumi to
take Cordoba." The earliest account of this period is probably that found
in the Akhbiir Mqjma<a, a collection of anecdotes about the history of
al-Andalus which may have assembled in the tenth century, although the
surviving manuscript is later.2I Tariq left a garrison in Toledo before
continuing his lightening conquest of much of the peninsula. A year later,
Mnsa ibn Nusayr, the governor of North Africa, arrived in Toledo . The
Chronicle if 754 blamed Musa for Toledo's fall: 'After forcing his way to
Toledo, the royal city, he imposed on the adjacent regions an evil and
fraudulent peace'.22 More than a century later, an Asturian chronicler
emphasised that Toledo's capitulation meant the end of the Goths: 'even
the city of Toledo, victor over all peoples, succumbed, vanquished by the
victories of the Ishmaelites; subjected, it served them'.23 Yet the chronicler
was exaggerating the importance of Toledo's defeat for propaganda
reasons. Only if christian Toledo had been expunged from history could
the Asturians' claim that they were the heirs of the Visigoths be justified. In
this version of events, the Saracens had 'exterminated the Gothic
kingdom'i" Only Pelayo, the victor of Covadonga, who stood in a line
of Gothic succession stretching back to Romulus, was left. In the ninth
century, his descendant, Alfonso II, was to establish in Oviedo 'the entire
order of the Goths just as it had been in Toledo, both in court and
palace'." It was a case of moral as well as political regeneration. A text
preserved in the tenth-century Roda Codex dated the foundation of eight
Galician cities - Lugo, Astorga, Leon, Zamora, Braga, Chaves, Tuy and
'Portugal' ~ to an attack by the founders of these cities, Septemsiderus
and his seven sons, on 'Octavian', king of Toledo. The legend was
illustrated with pictures of Babylon, Nineveh and Toledo; like the biblical
cities, Toledo was brought low by sin.26 When one attempts to read behind
the bias of the Asturian chroniclers and the Arabic historians, Toledo's
status appears to have been more ambivalent. The city was not merely an
islamic satellite of Cordoba. Perhaps the creation of a new capital in
Cordoba is an indication that the invaders were not confident of their
ability to control th e whole of the former Visigothic realm .

18
 CORDOBA AND TOLEDO

It is difficult to know where to place Toledo in relation to the christian-

muslim 'frontier' . In contrast with Cordoba, Toledo appears only
sporadically, in references scattered through the Arabic histories. " usually
in the context of rebellion against Cordoba. The exact details of these
events are uncertain, but the pattern is clear. Toledo was both a centre of
opposition to Cordoba and an important prize for both christians and
muslims . In 740-2, Berber rebels besieged the city.28 The first Umayyad,
'Abd al-Rahman I fought against opposition based in Toledo in 760-1,29
763 -4,30 785 3\ and 788-91. His successor, al-Hakam I defeated his first
Toledan revolt in 796-7 32 and a second in 807 . A century later, this episode
elicited the comment from Ibn al-Quuya that 'the Toledans were such a
rebellious and insubordinate people that they paid no att ention to any
governor, to the point that they never became subjects of any country'i'"
The suppression of the revolt of 807 ended with the so-called 'Toledan
night', when some 5,300 rebels were murdered at a banquet to which
al-Hakam's son had invited them - an episode which was almost certainly
embroidered with details from eastern Arabic and Byzantine sources. " Yet
in spite of al-Hakam's annihilation of his opponents, at least four more
Toledan rebellions were recorded during his reign and that of his successor
'Abd al-Rahman II.
Up to this point, the events in Toledo and its hinterland could be
represented as internal disputes amongst muslims resisting Cordoban
hegemony, since there is little evidence for the involvement of the
indigenous population. Toledan rebellions against Muhammad (852-86)
introduced two new elements - the cooperation between the rebels and the
Asturians, and the rise of the Banu QasI, a family of christian converts to
Islam. In 853/4, the Toledansjoined forces with Ordofio 1.35 The following
year the allies was defeated by Muhammad at the Guazalete in a battle
which was turned into epic by the Arabic historians." Muhammad's
control of Toledo was short-lived, however, and the city was soon under the
control of Mnsa ibn Mnsa of the Banu Qasi:
'A certain man by the name of Musa - a Goth by birth but
deceived by the Muhammadan rite along with all of his people,
whom the Chaldeans call the Banu Qasi - rebelled against the
Cordoban king and took control of many cities partly by the sword
and partly by treachery. First he took Zaragoza, and then from
there Tudela and Huesca, and after that Toledo, where Musa
placed his son by the name of Lupe as prefect. . . . [Musa and Lupe
captured two Frankish commanders and two] 'great rebels from the
Chaldeans . . . on account of these victories, Musa swelled so much

19
 CORDOBA AND TOLEDO

in pride that he demanded that his men call him th e third king of
Spai n' ;"
To ledo was now the locus of mercuri al shifts in allegian ce between the
three main players in the north for which the evidence is patchy and
contradictory. For some years, the Banu Q ast were loyal to Cordoba, but in
882 Muhammad sent a force under H ashim ibn 'Abd al-Azlz against them.
Muhammad ibn Lubb [Lupe] , who may have been a nephew of Miisa ibn
Musa, marched out against Hashim. Then he changed his mind and tried
to persuade Hashim to unite with him against the Asturians, now ruled by
Alfonso III. Earlier hostage-taking by all parties complicated the situation.
H ashim did not want to antagonise Alfonso, who was ho lding his son.
Hashim him self held Ism'ail, the son of Muhammad ibn Lubb, and he sent
his ca ptive and other gifts to Alfonso in return for his son." To ledo escaped
ca pture by Muhammad and the Banii Q ast kept intermittent control of the
city, perhap s until 920 , although the Toledans were unhappy with any
leader for long and in 906, Alfonso III was welcomed to the city for a sho rt
tim e.l? After 93 2, when 'Abd al-R ahman III capture d the city after a lon g
siege, Toledo seem s to have loyal to Cordoba until the civil war period ,
when it became an independent kingdom ruled by the Banii Dhi l-N un.
There is nothing in this narrative that makes Toledo an islami c city in
the same sense as Cordoba. It is not even clear that there was a substantial
mu slim population. Certainly ther e are no surviving bui lding s on the scale
of the great mosque of Cordoba, and most of the evidence for islami c
Tol edo come s from the eleventh century. Ibn H awqal, who wrote so
effusively about tenth-century Cordoba, merely mentioned Toledo as
'among the celebrated an cient cities't" with Jaen and Guadalajara. A
handful of inscriptions refer to bui lding carried out before Alfonso VI's
reconquest of the city in 1085. In 87 1, the bridge across the Tagus was
rebuilt, according to an inscription on the structure which rep laced it in
1259. At least three mosques were completed, of which two have survived.
The small nin e-domed mosque of Bab al-Mardum, now the church of
Cristo de la Luz , was built as a priv at e foundation , perhaps in imitation of
the royal enclosure of th e Cordoban mo sque." The mosque of Tornerias,
closed down between 1498 and 1505 , may date to the eleventh century,"
The pr esent churc h of San Salvador stands on the site of a mo sque which
was eithe r built or restored in 1041. There may onc e have been oth ers, but
Toled o's position on a small, elevated site, and its continuous occupa tion,
mean that archaeology is unlikel y to add very mu ch to these scant rem ain s.
After the reconquest of 1085, Toledo becam e a christian capital once
aga in . The rate of 'reconversion' of the mu slim population seems to have

20
 CORDOBA AND TOLEDO

proceeded at a much faster rate here than elsewhere in the peninsula.P Yet
the existing christians of Toledo may have been slow to assimilate to their
new good fortune. Alfonso VI's archbishop, Bernard de Sedirac, took over
the main congregational mosque, claiming that it had been the basilica of
Mary, wh ere the Visigothic kings wer e crowned." The christians of Toledo
were not convinced . After all, th ey still had their own bishop, the last in a
line that stretched almost unbroken from the Visigothic period. The last
bishop before the Reconquest, Pascal , was consecrated in Leon in 1058.45
They were still using the Hispanic liturgy, although in the north of the
peninsula it had been replaced by the Roman rite." The antiquarianism of
Toledo's 'old christians"? complemented the fake continuity of the Astur ian
Gothic revival in promoting the ecclesiastical pre-eminence of Toledo,
which was confirmed by Urban II in 1088. The fact that the christians of
Toledo preserved the memory of their Visigothic past is the best evidence
for the history of the city in the early islamic period.
At the tim e of the conquest, the Toledan church was defending a
primacy only recently established. Although Toledo had become the
secular capital of Visigothic Hispania under Athanagild and Leovigild in
the second half of the sixth century, it was not recognis ed as the
ecclesiastical centre of the realm until a century later, under Julian,
archbishop of Toledo (68 1- 94). The islamic conquest does not seem to
have deflected the Toledans from their campaign of self-promotion. Soon
after the conquest, Sindered of Toledo (c.704-12) fled to Rome" but a list
of metropolitans in the Codex Emilianense shows th at the episcopal succession
was not permanently disrupted . It was in this context that a short Lift was
written ofIldefonsus, archbishop of Toledo (657- 67),49 who had promoted
Toledo's claims to preeminence. Ildefonsus' On Famous Men concentrated
on his pr ed ecessors in that see, ignoring more eminent saints and scholars
from the peninsula, notably Isidore of Seville. The Lift qf Ildefimsus is
important because it testifies to a continuing ability to compose
hagiography which is related in style and language to similar works from
Visigothic Hispania and from the rest of Europe throughout the early
Middle Ages. It also allows a glimpse of the contemporary concerns which
may underlie its narrative and use of hagiographical formulae. Both the
date and the authorship of the Lift qf Ildifonsus are disputed. The earliest
surviving copy comes from the Codex Emilianensis, and is dated 994 . It is
described as the 'Life and deeds of St. Ildefonsus , metropolitan of Toledo
written by Cixila, bishop of the same city'. Cixila was bishop by c.744 .50
The attribution of the Lift of Ildefonsus to him is not secure; indeed, one copy
attributes it to H elladius, a predecessor of Ildefonsus . Although the text
may be no earlier than its first manuscript." th e author noted that he was

21
 CORDOBA AND TOLEDO

not going to copy episodes which Urban and Evantius had already
described. Since Urban was a melodicus or cantor and Evantius a deacon of
Toledo'" and both died before c.737 , 53 it is possible that the Lift dates from
the second half of the eighth century.
The Lift was written as an introduction to three of Ildefonsus' liturgical
compositions, in honour of Cosmas and Damian, Leocadia of Toledo and
the Virgin Mary, which were originally copied with it. The text is brief, and
in one manuscript, it was supplemented by Julian's short biography of
Ildefonsus from his own On Famous Men.54 The first section of the Lift
describes Ildefonsus' education with Isidore of Seville, his election as abbot
of the monastery of Cosmas and Damian, outside Toledo, and his elevation
to the see of Toledo in succession to Eugenius II (646~57).Julian said that
Ildefonusus had been educated by Eugenius, but linking Isidore with
Ildefonsus connected him, and hence Toledo, with Hispania's greatest
scholar. The second episode in the Lift describes a miracle which took place
on the feast day of Leocadia (9 December) and the third section recounts
the Virgin Mary's appearance to Ildefonsus on her feast day (18 December).
The precise location of Advent Sunday between these two feasts dates the
events to 662.
On the feast day of Leocadia, Ildefonsus, with the help of angels, raised
the lid of the saint's sepulchre, 'which thirty young men could scarcely
move', and took out the perfectly-preserved shroud in which Leocadia's
body had been wrapped. The watching clergy and congregation 'power-
fully sang the song of praise which the same Lord Ildefonsus had lately
composed'. Then Ildefonsus shouted for a knife to cut the shroud, because
Leocadia was still clinging onto it. The watching king Recceswinth 'who
was seen with a proud eye on account of his iniquities noised abroad .. .
offered with tears a small knife which he kept in his treasury' . Ildefonsus
cut the shroud from the saint's grasp and 'he placed the same knife with the
relics in a reliquary of silver, judging it worthy that what had cut holy
things might not again ever cut polluted things'.55 Recceswinth had been
severely criticised at the eighth council of Toledo in 653 for the
depredations which he and his father had made on lands belonging to
the Visigothic nobility and because Recceswinth had succeeded to the
throne without being elected . Perhaps the hagiographer was reminding his
audience of the superiority of spiritual over temporal power. Recceswinth
also witnessed Ildefonsus' second miracle. As Ildefonsus and the clergy
were going into the church to celebrate Ildefonsus' Marian mass, they were
greeted by a host of angels. The clergy 'retracing their footsteps in terror',
Ildefonsus went on alone and saw Mary sitting in the bishop 's seat, 'which
afterwards no bishop dared to sit in except Sisebert, who immediately

22
 CORDOBA AND TOLEDO

losing his see through error, was sent into exile'.56 Mary offered Ildefonsus
'a little gift taken from my son's treasury', which later tradition turned into
a chasuble." The reference to Sisibert (690-3), who was deposed by the
eleventh council of Toledo for plotting against Egica ,58 might also have
reminded readers of Sindered's defection after 712. Writing about
Ildefonsus for an eighth-century audience, the Lift was an attempt to
recall Toledo's glories in less favourable times .
Cixila , the supposed author of the Lift, was commemorated by the
Chronicle qf754 as a 'restorer of churches' P? In a hymn which was included
only in the Toledo hymnary, Cixila was associated with a new church
dedicated to San Tirso.f" This challenges the assumption that christians
under islamic rule did not build churches. Confirmation that the see of
Toledo in particular was promoting church building may perhaps be found
in the church of Santa Maria at Melque, 36 km southw est of Toledo,"
although both th e date of this chur ch and the stylistic influen ces upon it are
controversial. There are no surviving accounts of the building of M elque. It
was at first believed to be Visigothic . There are, however, elements that
appear to copy the decoration of Umayyad palaces such as Khirbat
al-Mafjar near Jericho, built between 739 and 743. It is at least possible
that the church at Melque was built with the help of muslim artisans,
perhaps as early as the late eighth century.
Even after the establishment of an islamic administration in Cordoba,
Toledan clerics acted as though they still had the authority to issue
judgments on the errors of christians elsewhe re in the peninsula. Evantius
(probably the deacon mentioned in the Lift qf Ildifonsus) sent a letter
admonishing the christians of Zaragoza.f who were following jewish
practice in refusing to eat animal blood and the meat of strangled animals,
saying that they were unclean. This was not a new problem. Interpretation
of the strictures found in Leviticus and Deuteronomy, the common
inh eritance of christians and jews, had also exercised the Visigothic church,
especially in relation to the anti-jewish legislation of the seventh centuryf"
In c. 750, Peter, another deacon of Toledo , sent to Seville a treatise on the
dating of Ea ster.64 It do es not survive, but the same Peter may have written
a letter which was copied into a tenth-century manuscript.F' This latter text
is accompanied by a fragment of another letter, dated 764, which could
have been sent by Felix of Cordoba, the recipient of Peter's letter. The
Cordoban was inquiring about the orthodoxy of christians in his city who
had suggested fasting with the jews on the Day of Atonement. Again, the
problem was one of the danger of letting jewish rules influence christian
practice. One might assume that the presence of muslims in Cordoba and
Toledo made the question of defining the boundaries between the thr ee

23
 CORDOBA AND TOLEDO

religions more acute, but there is no reference to Islam , and no evidence

within th e texts them selves th at Evantius, Peter and Felix were doing any
mo re th an re-exploring problems which ha d troubled the H ispanic churc h
before the conq uest.
It is mainly for its involvem en t in another theological controversy,
Adoptionism, that the eighth-century Toledan church is rememb ered .
Adoptionism is the subject of a recent mon ograph." altho ugh this
addresses th e theological rather than the political aspec ts of the controversy
and its impli cat ion s for the churc h in al-Andalus have yet to be examined
in detail. Like the writings of Evantius and Peter of Toledo and Felix of
Co rdo ba , it was a reworking of an old probl em. The debate develop ed
Visigothic formulation s of Trinitari an doctrin e which had bee n derived
from th e thou gh t of Augustine by such autho rities as Isidore and Julian of
To ledo . Eu genius II of Toledo had compose d a work on the Trinity, now
lost, which may have been used in formulatin g the creed of the eleventh
council of Toled o (675). Ildefonsus composed a christological work, also
lost, which Julian listed in his encomium of the bishop. The debate was
reawakened and, perhaps ina dvert entl y thrown ope n to a wider field, by
Elipandus (754-after 800), who succee ded Cixila in the see of Toledo . The
controversy arose from a letter which Elipandus sent to a cer tain Migetius,
sometime in the 780s, in respon se to a sho rt work which Migetius had sent
him ." M igetiu s seems to have been the lead er of a schisma tic sect. H is
followers may have succee ded in estab lishing themselves near Cabra, thir ty
miles southeast of Co rdoba - if they ca n be identified with the 'Cassianists'
condem ned at a council held in Cordo ba in 839. 68 Migetius' work does not
surv ive, and his erro rs have to be recon stru cted from Elipand us'
overwro ught con demna tion of them. H e accused Migetius of being too
rigorous in his demands for clerical purity and of refusing to eat with
pa gans , but the brunt of his attack was dir ected at Mi getius' Trinitarian
beliefs. As Elip andus represented th em , they are way outside the
main stream of Hi spanic T rinitarian theology. Migetius seems to have
been pr eachin g that the three person s of the Trini ty had taken corpo rea l
for m in the persons of David, C hr ist and Paul. Nevertheless, Cavadini
argued that Mig etius' beliefs m ust be taken serio usly as a contribution to
the debate abo ut the natu re of the Trini ty. Elipandus' statement of his own
position was a defen ce of a one-pe rson christology against M igetius'
for m ulation of the problem . Elipa nd us outlined the argu ment that J esus,
considered solely from th e point of view of his human nature, was an
'adoptive' son of Cod. Yet the essence of the 'a doptionist' position which he
describ ed in the letter to Migetius do es not represe nt Elipandus' beliefs. It
was qualifi ed in several places by the phrase 'as you say'; Elipandus qu oted

24
 CORDOBA AND TOLEDO

from Migetiu s in order to discredit him ." A surviving tract known as

the Symbol qf the ftith qf Elipandus70 seems to be a more formal rebuttal
of Mig etiu s' theology which Elipandus may have composed later. In spite
of the language in which Elipandus expressed himself, his was an intelligent
defen ce of current Hispanic orthodoxy."
Nevertheless, Elipandus attracted condemnation within the penin sula as
well as outside. He was in dispute with Ascaricus;" who may have been
an Asturian bishop, and Theudila of Seville, who was mentioned in
connec tion with a restatement of Elipandus' arguments almost a century
lat er by Alvarus of Cordoba." Elipandus' bitterest critic was Beatus of
Lieb an a. Beatus may have been a monk in the remote monastery of San
Torribio de Liebana in the Asturias, although the Lift qf Beatus included in a
martyrology from Astorga in northern Spain 74 is unreliable.P H e was
probably th e author of a Commentary on the Apocalypse of which several
manuscripts survive, notable for their illustrations rather than for the
originality of Beatu s' text. Hi s substantial tract Against Elipandus" was
writt en c.785 with Heterius, bishop of Burgo de Osma on a confluence of
the river Duero. The debate centred, at least in Elipandus' view, not so
mu ch on theological niceties, as on the authority of Toledo . In a Letter to
Felix of Torribio of 785, lat er copied into Beatus' Against Elipandus,
Elipandus was scathing about Beatu s' credentials:
'For never has it been heard that the Libanese instru cted the
Toled an s. It is known to the peopl e of the entire world th at from
the very beginning of the faith , this see has been luminous with holy
teachings, and that nothing schismatic whatsoever ha s gon e forth
from it'. 77
In fact, in the context of th e wider debate about Adoptionism, Beatus and
Elipandus' views were remarkably similar. Both could cite Visigothi c
authoriti es who had used the word adoptivus, although they were unable to
agree on what it meant. It was the int ervention of the Carolingian Church
which gave the controversy new momentum.
The first mention of Adoptionism outside the peninsula was a lett er sent
by pop e Hadrian I to the Hi spani c bishops , in which he warned them
again st Mig etiu s, but also condem ned Elipandus' position in terms which
make it clear that he did not understand it. The letter was the last of thr ee
which H adrian sent between 785 and 791 concern ing a pri est nam ed
Egila,78 who had been dispatched on a pr eachin g mission to the penin sula
c.780 by Wilcharius of Sen s, where he had fallen und er the influence of
Migetiu s. The Papacy bracketed Elipandus' unorthodoxy with Egila's
defection as another exam ples of the Hi spani c Church's long-standing

25
 CORDOBA AND TOLEDO

resistan ce to Roman jurisdiction. I? At about the sam e time, Felix of U rgel

introduced the debate to the Carolingians by pr eaching Elipandus'
do ctrine in his see of Urgel, which was part of the Carolingian realm after
789. In 792 , Felix was summoned to Regensburg to defend him self before
Charlemagne and Alcuin. Bullough suggested that Alcuin's rise at the
Carolingian court was at least partly attributable to the lett ers he wrote
against Adoptionism to Elipandus and the Hispanic bishops. " If so, it is
hardly surprising that Alcuin remained a zealous opponent of Adoptionism.
H e reworked the letter to Elipandus into a book designed for wider
readership and asked Charlemagne's permission to refute Felix of Urgel."
Thus began a long and civilised debate between Felix and Alcuin which
resulted in the composition of Seven Books Against Felix if Urgel. Adoptionism
was furth er condemned at church councils held at Frankfurt in 795 and
Aachen in 799.
Faced with such criticism from the Franks , the Hi spanic clergy closed
ranks.82 Elip andus and his colleagues sent a number of lett ers to
Ch arlem agne after the condemnation at Frankfurt. Yet, if the debate
continued in the peninsula after the 780s, it has left little trace. Elipandus
him self did not take pa rt in th e lat er stages of the controversy between
Alcuin and Felix, although Beatus received a lett er from Alcuin in 797 or
798 . By this stage, the debat e had becom e irrelevant to the Hispanic
church, which 'existed in an environme nt which ultim ately was protective
of olde r ecclesiastical traditions again st the sort of cultur al hegemony that
Charlemagne's renaissan ce of learning . . . entailed' i'" It was a qu estion of
on e tradition against another. Felix knew it had been customary for
theologians from the peninsula to refer to Christ as 'adopted ' and Alcuin
menti on ed a eucharistic prayer writt en by Ildefonsus which spoke of 'the
passion of the adoptive man'. Alcuin and Hadrian, how ever, reformulated
th e debate in their own terms. By the end of the eighth century, the two
sides of the Pyrenees existed in different christological worlds.
None of the protagonists of the controversy saw it as being a respons e to
the invasion of 711. Felix may have writt en a Disputation with a Saracen which
Alcuin inquired about but seem s not to have read." Yet the pr esen ce of
mu slims in the peninsula, which is often assumed to have enge ndered
the controversy, was not mentioned in any of the surviving texts from the
Adoptionist debate. To ask about the relationship between Adoptionism
and the islami c conquest may be the wrong qu estion. Christian s, both in
the east and in the peninsula saw the conquest as a military disaster. They
did not yet identify Islam as a new religion. P In his letter to Aethelred of
Mer cia (746- 7), Bonifac e had used the threat of the Saracens to urge the
backsliding English to rep ent of their sins: 'So it has befallen othe r peoples

26
 CORDOBA AND TOLEDO

of Spain, Provence and Burgundy. . . . Almighty God allowed the penalties

of such crimes to destroy them, first by letting them lose the knowledge of
God and then by loosing the attacks of the Saracens upon thern'i'" There
was no suggestion that the Saracens came to bring a new kind of God. The
times still allowed the Church to pursue old controversies rather than
forcing her to formulate new ones .
Much less is known about the Toledan church after Elipandus.
Wistremirus of Toledo presided over a council in Cordoba in 839 and
received Eulogius, the hagiographer of the martyrs of Cordoba, who may
have been chosen to succeed him c.858, although he did not take up the
appointment." The list of the metropolitans of Toledo in the Codex
Emilianense 88 ends with John (d.926). although a bishop of Toledo named
'Ubayd 'Abd Allah, later bishop of Seville, was an advisor to al-Hakam 11. 89
After this, the record of episcopal activity falls silent until the consecration
of Pascal in 1058. Yet it is clear that the Toledan church, which was still an
important player in the peninsula in the eighth century, preserved some at
least of its Visigothic inheritance through the years of political turmoil.
Toledo was a city with a sizeable christian population which the Umayyads
could rarely control for long. It may be in this milieu that the eighth-
century chronicles were written.

27
 CHAPTER THREE

News from the east in the

eighth-century chronicles

'He [al-Samh, governor of al-Andalus 718-21] took possession of

Gallia and Narbonense through the leader of the army M azlema
and harassed th e leaders of the Franks with frequ ent battl es. And
assembling his forces, the aforeme ntioned leader reached Toulouse,
surrounding it with a siege and trying to overcome it with slings and
other machin es. But the Franks, informed of these events, gathered
togeth er under th eir commander Eudo. Thus gathered, they reached
Toulouse. At Toulouse, th e battl e lines of both sides engaged in
serious fighting. They killed Zema, [al-Samh] the lead er of the
Saracen army and some of his troops and they pursued the
rem aining part of the army in flight';'

T he history of H ~sp ani a b e~een the arrival of ~riq ib~ Ziyad in 711
and the establishment of Abd al-Rahman I III 756 IS obscure and
controversial. Arabi c historians elaborated their accounts of the conquest
with stories of the rivalry between Tariq and Mus a ibn N u ~a yr, and of the
miraculous discoveri es th at they made in Toledo. Yet the outlin e which can
be extracted from these stories is corroborated by the Chronicle if 754 and
by contemporary chronicles written north of the Pyren ees. Within four
years, islam ic forces dominated the southeast of the peninsula and
adva nced as far north as Zaragoza and the Picos de Europa. Led by a
series of governors, they pu shed on over the Pyrene es. Al-Hurr (717- 8)
attacked the Visigothic kingdom based on Narbonne. His successor,
al-Samh, took over Narbonne, and carried out raids on Aquit aine and
Provence. According to the Frankish histori es, Charles Martel brought this
expansion to an end with his victory over 'Abd al-Rahman al-Ghafi:kI,
somewhere betw een Tours and Poitiers, in 733 or 734.2 In the Hispanic
sources, both in Latin and in Arabic, this battle assumed less importance.
The Chronicle if 754 not ed that the Franks 'despicably put up their swords,
saving themselves to fight another day, for night had fallen during the

28
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

battle' and the Saracen army was allowed to escape ." Both the Chronicle if
741 and the Chronicle if 754 placed more emphasis on an earlier islamic
setback, the defeat of al-Samh before Toulouse. The version of this defeat
recounted by the Chronicle if 741 which is particularly problematic, as we
shall see, introduces this chapter. This episod e was recounted as part of the
history of Hispania and Byzantium which the two chronicles told in very
different ways. Since the Chronicle if 741 is mainly an account of the
struggle between Christianity and Islam in the east, this chapter
conc entrates on the eastern material in both chronicles. What relevance
did the decline of Byzantium have to Hispanic christians who had recently
suffered their own islamic conquest?
Chronicles used to have a poor reputation. They were thought to be the
work of minor clerics copying mechanically from their equally ill-educated
predecessors and th en adding a few trifles of their own to bring the work up
to date. Chronicles were mined for snippets of information, but otherwise
disparaged. Recently, the use of chronicles as a historical resource has
become more sophisticated . It is not enough simply to mark a chronicle out
of ten for factual accuracy and list its sources. Comparisons between two or
more accounts of the sam e period show that chroniclers chose what to write
and th e way that events and their protagonists were emphasised or played
down , making polemical use both of long past and of recent history" The
nearest one can get to extracting historical fact from such compositions is
to find a near contemporary with a different perspective on the same
material ." Chronicle writing can be flexible, like any other historical genre ,
so that 'the style and form of each work, however short, is determined by
th e author's current literary strategy'." In retrospect, it is sometimes easier to
decide what the chronicler was trying to say than what really happened.
This chapter is an attempt to determine how eighth-century readers
might have understood the Chronicle if 741 and the Chronicle if 754. It
developed out of recent work on the fifth-century chronicles in Gaul. Few
of the texts from which the Gallic chroniclers were selecting their material
have survived, making it more difficult to unravel the chroniclers' purposes.
In the case of the two eighth-century Hispanic chronicles, we can compare
two chronicles from the same generation with each other and with at least
one of th eir common sources. It is possible to get some idea of what the
authors included and left out and the potential for understanding why they
did so seems high . The chronicles must be located as firmly as possible
within a historiographical tradition, and in time and place. Once we know
what their authors might have read and how they might be expec ted to
have written, it is possible to interpret their words . This will involve a
number of detours and some guesswork.

29
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

The authors of the eighth-century chronicles inherited the Visigothic

tradition of writing history in the form of a universal chronicle," based on
the work of Eusebius, whose Chronicle and Ecclesiastical History they could
have read in Latin translation. This was history with a message, the
triumph of the empire of God, which for Eusebius was co-extensive with
the Roman empire. In Hispania, this idea was taken up by Orosius and
Prudentius, early in the fifth century. Eusebius' influence pervaded
historical writing even after the fall of the Roman empire in the west to
the barbarians, along with a powerful but controversial apocalyptic strain .
Augustine proposed that the City of God was set apart from earthly powers
and denied the possibility that God's purpose might be accomplished on
earth but western historians preferred to give the role previously played by
Rome to the empire's barbarian successors. In Visigothic Hispania it was
not easy to envisage the barbarian kingdom as God's chosen instrument,
since the persistence of the Arian heresy made the Visigothic kings suspect
in the eyes of the catholic clerics who wrote the chronicles. It was only after
the conversion of Reccared to catholicism in 589 that John of Biclar and
Isidore were able to write convincing providential history in the peninsula.
They could also link the history of Hispania with the survival of the Roman
empire in the east.
John of Biclar studied in Constantinople and it was there that he began
to write his Chronicle, a history whose introduction spells out the author's
mission to continue the work of Eusebius into his own days." John of
Biclar 's text is most clearly Byzantine in using imperial regnal dating but he
also included many episodes from Byzantine history, although he did not
take a pro-Byzantine attitude to events in Hispania but reported the defeats
suffered by the Byzantines at the hands of the Visigoths. The Chronicle ends
in 592 but one version was extended to 602 by an unknown hand with the
addition of information about the emperor Maurice and Gregory the
Great." In the next generation, Isidore summarised the history of the world
from Adam to 615 in his Chronicle and his History if the Goths, Sueves and
Vandals improved onJohn of Biclar's version of events in Hispania, ending
with the triumph of the Gothic monarchy, the expulsion of the Byzantines,
and Reccared as the new Constantine. Hispania was firmly integrated into
a universal history of the christian empire.
The dramatic events of 711 could be envisaged as part of the same story.
Both the Chronicle if 741 and the Chronicle if 754 are continuations ofJohn
of Biclar and incorporate passages from Isidore 's History if the Goths. Both
texts survived in copies made in the christian north of the peninsula. There
are four surviving manuscripts of the Chronicle if 741, one from the
thirteenth century and three from the sixteenth. They were all recensions

30
 NEWS FROM THE EAST IN THE EIGHTH - CENTURY CHRONICLES

of the same exemplar, now lost. T he best description of this exemplar is

that ofJuan Bauti sta Perez (1537- 97), a canon of Toledo and later bishop
of Segorbe , who collected many Hispani c medieval texts. H e fou nd the
chro nicle in a codex present ed to the library of Philip II at El Escorial in
1578 by J orge de Beteta y Cardenas, a noblem an from Soria; the codex
was destroyed in the fire of 1691. 10 Perez called the Chronicle of 741 a
'co ntin uation of J ohn of Biclar by an unknown author from the Gothi c
volume which is in San Lor en zo [El Escori al] with Eusebius and Victor ...
I believe that it is by Isidore Pacensis', demon strating that the habit of
calling unid entified authors of historical texts 'Isidore' persisted at least
until the sixteenth century. H e describ ed the manu script as 'ex tremely old'
bu t was not able to date it, alth ough five manu scrip ts donated by J orge de
Beteta that survived the fire have all bee n dated to the ninth or tenth
centuries. Apart from the C hro nicle of 741, the codex contained a number
of imp ortant historical texts from the Hi span o- Roman and Visigothic
periods - the chronicles of J erom e (to 378), Prosper (to 455), Victor of
Tunnona (to 567) and J ohn of Biclar, Isidore's Chronicle and History if the
Goths, Sueves and Vandals, Julia n of Toledo's Against Wamba (c.675), the
Laterculus regum visigothorum (a list of the Visigothic kings to Witiza, written
before Witiza's death in 710), and the lift if San M illan (c.63 I ). The latest
text to be copied into th e codex was the version of the Chronicle ifAlfonso III
known as the ad Sebastianum, which ends with the death of O rdofio II
(850-66). This dates the whole collection no earlier than the late ninth
century and suggests that the codex was copied as well as preserved in the
north.
The Chronicle if 754 survived in two appa rently independ ent versions.
The earliest is incomplete and consists of six folios, now divided between
M adrid and London, I I which seem to be slightly earlier th an J orge de
Beteta's surviving manuscripts, perha ps mid-ninth century.12 Anoth er
version was copied into a thirteenth-century codex now in Madrid.'! The
M adrid codex contains a large number of histories, from Eusebiu s to the
chronicle of Ricardo Pictaviense continued to 1244. The list of texts
composed before the eighth century is very similar to those accompanying
the Chronicle if 741, with the addition of a chro nogra phical comme ntary by
QuintusJulius Hilarianus, the Gallic Chronicle (to 5 11), a Chronicle if Cartagena
(to 525), the On the Birth and Death ifthe Fathers by Isidore and his On the lives
ifFamous Men, and works with the same title by Gennadiu s, Ildefonsus and
J erome. The Chronicle if 754 was also copied into at least two other mixed
codices of histories. One, from Alcobaca, disappeared somet ime after
1622,14 bu t there is a fourteen th-century manu script from Navarre now in
Pari s."

31
 NEWS FROM THE EAST IN THE EIGHTH -CENTURY CHRONICLES

Both the Chronicle if 741 and the Chronicleif 754 were copied into codices
which told the history of the peninsula, sometimes by the continuation of
an earlier chronicle, but often by a process of accumulation of texts which
seems to be more random . Both wer e read in northern Hi spania within a
century of their composition. For th e Chronicle if 741 it is possible to tra ce
this process back to the date of its inclusion in Jorge de Beteta's codex,
perhaps in the ninth century, and the six loose folios of the Chronicle if 754
show that this chronicle was being copied at about the same time. It was
during this period that the earliest Asturian chronicles were composed as
propaganda for the emergent christian kingdom, although there is no
indication that the Asturian annalists had read th e eighth-century
chronicles. The Asturians wrote the history of the islamic conquest only
as it had affected their own small corne r of the peninsula. Yet the inclusion
of the Chronicle if 741 and the Chronicle if 754 in northern historiographical
compe ndia is an indication that history which, as we shall see, seems to
have been writt en by christian clerics und er islamic rule was con sidered
sufficiently orthodox to be copied in a christian kingdom.
The vagari es of manuscript survival have left only on e appar entl y
com plete version of each of the two chronicles. The int erpretation of a sole
surviving manuscript of a text is difficult because it is impos sible to be sure
that this text is either th e original version, or even the product of on e
author at on e ascertainable date. The plain style of many chronicles means
that later altera tions more easily blend in with original material than is the
case in other genres. The com pilation of new versions of the past by linking
one or more chronicles was very popular in the Middle Ages. We can see
this process in action in the surviving manuscripts of the Chronicle if
Hydatius, written in Galicia in the middle of the fifth century!" A ninth-
century manuscript of this work has survived almost intact. Lat er copies
from Hispania belong to two separate traditions. Both were interpolated
with err oneous dates and material from other, unknown sources. Had the
ninth-century version not survived, it would have been even more difficult
to know what Hydatius actually wrot e. Readers must also be aware that in
som e cases, of which Hydatius' chronicle is a good example, editors from
the sixtee nth century onwards '? sometim es pr eferred their own dating and
arrangeme nt of the material to what they found in the manuscripts. This
analysis of the Chronicle if 741 and the Chronicle if 754 depends on the
assumption that the modern editions represent coherent pieces of work
dating from the middle of the eighth century.
The two authors compiled their chronicles from the same, or very
sim ilar sources, as we shall see. Since the use of this material in the Chronicle
if754 is the more straightforward of the two, it is easier to begin by looking

32
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

at this text in more detail before conside ring the very different syn thes is
which th e autho r of th e Chronicle of 74 J achieved.

The Chronicle of 754

The author of th e Chronicle of 754 seems to have known a contin ua tion of

John of Biclar's Chronicle with an epilogue 'up to th e present year whi ch
is [Hispanic Er a] 680' [642].18 H e used Isidore's History of the Goths
extensively, and also knew Isid ore's Chronicle, the Erymologies, a christological
treati se by Braulio of Zar agoza , Julian of To led o's Book on the Proving of
the Sixth Age and th e Acts of th e ecclesiastical councils of Toled o." Hi s
extensive knowledge of his Visigothic literary heritage is evidence th at
education in Latin was not perman ently disrupted by th e conq uest. The
question of th e collapse or co n tin uity of Visigothic institution s is
controversial, but th e survival of mu ch of th e framework of governme nt
also bears witness to th e du rability of pre-islamic str uctures.i" The autho r
m ay have been writing in Toledo, because he seems to have known th e city
well, altho ugh Lop ez put forward a case for the south-east corner of the
peninsula formerly held by the Byzan tines." The chro nicler wrote at least
one othe r histori cal work, an 'e pitome ' on the civil wa rs between the
gove rnors of al-Andalus in th e ea rly 740s, which does not survive.F
The Chronicle of 754 begin s with th e access ion of H eracliu s in Era 649
[61 I : th e accep ted dat e is 6 10] and, like J ohn of Biclar 's Chronicle, it is
str uc ture d by Byzantine imperi al accessions. Although the autho r included
m aterial on th e Byzan tin es and on the islamic conquest of Syria and th e
calip hal succe ssion, his conce r n was not with th ese events as such. H e used
them to establish a secure chro no logica l fram ework. Most of the entries
about events in the east ar e bri ef, and wh ere th e chronicler included a
more detailed episode, he did this to make a point, as we shall sec. On this
fram ework he hung a history of th e Visigothic kings, using his eas tern
m at eri al to turn th e histo ry of Hispani a int o a universal ch ro nicle. Apart
fro m a few discrep an cies, he was successful in tying in Byzantine impe rial
regnal yea rs with Hispani c Era dates, th e regn al years of th e calip hs, th e
years of th e H egira, and 'ann us mundi' d ating whi ch began with
th e creation of th e world in 5200 BC .23 For mu ch of the chronicle, th e
three histori es of th e Byzantine, Visigothic and islamic empires run in
par allel, with one cha pter given to eac h in turn. Where th e writer was sho rt
of information he merely gave a list of da tes: 'At this tim e, in the Eras 736,
737 and 738 [698, 699 and 700] , the first, seco nd, and third yea rs of his
reign [Leontius] and th e seventy-ni nt h, eightieth and eighty -first years of

33
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

the Arabs, 'Abd al-Malik completed the thirteenth, fourteenth and fifteenth
years of his rule'.24 For the period after the islamic conquest, the chronicler
concentrated exclusively on the history of al-Andalus. This section is more
detailed, and this is what makes the Chronicle if 754 the best account of this
period.
The chronicler was particularly interested in ecclesiastical matters.
Much of his emphasis, together with his vocabulary, came from Isidore."
The main concerns of Isidore's Chronicle and his On Famous Men were the
Arian and anti-Chalcedonian heresies . The Chronicle if 754 showed Isidore
himself dealing with a heretical Syrian 'Acephalite' bishop at the Council
of Seville.r" The chronicler picked out one church council for each reign
up to 711.27 This religious focus, in the context of an obsession with
chronology, is the key to the chronicle. A sense of calamity caused by the
fall ofVisigothic Hispania and the End of the World which it may portend
is heightened by reports of famines, eclipses and other natural disasters .
Reports on the defeats of the Byzantines by the Arabs served to reinforce
the message that Hispanic christians had brought disaster upon
themselves. This was particularly clear in the chronicler's account of the
career of Heraclius. The emperor's pride sealed his downfall even at the
moment of his triumph over the Persians: 'When Chosroes' kingdom was
finally destroyed and had surrendered to imperial dominion, the people
did honour not to God, but to Heraclius, and he, accepting this with
pride, returned to Constantinople'." Heraclius' nemesis was not long in
coming. Forewarned in a dream 'that he would be ravaged mercilessly by
rats from the desert', Heraclius saw armies from Arabia overrun Syria ,
Mesopotamia and Egypt. The chronicler wrote the whole of Heraclius'
reign to 631 as a continuous narrative, even though it forced him to
backtrack in order to resume the history of Hispania, distorting the
chronological framework of the chronicle. His manipulation of chronology
underlined the parallel between events in Byzantium and Hispania. He
showed how the peninsula flourished under a succession of Visigothic
kings, starting with the pious and victorious Sisebut (612-21), until the
kingdom was destroyed by the ambition of the usurper Rodrigo, who
'rebelliously seized the kingdom at the instigation of the senate'. At the
Transductine mountains, Rodrigo faced the invading armies of 'Arabs and
Moors ... and in that battle the entire army of the Goths, which had come
to him fraudulently and in rivalry out of ambition for the kingship, fled
and he was slain. Thus Roderic [Rodrigo] wretchedly lost not only his rule
but his homeland, his rivals also being killed . . .'29 and Hispania was
ruined:

34
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

'W ho can recount such perils? Wh o can enumerate such grievous

disaster? Even if every limb were tran sform ed into a tongu e, it would
be beyond human nature to exp ress the ruin of Hisp ania and its
many and great evils. But let me summarise everything on one brief
page. Leaving aside all of the innumerable disasters from the time of
Adam up to the pres ent which this cru el, unclean world ha s brought
to countless regions and cities - that which, historically, the city of
Troy sustained when it fell; th at which J erusalem suffered, as foretold
by the eloquence of th e prophets; that which Babylon bore,
according to the eloq uenc e of th e scriptures; that which Rom e went
through, martyrially graced with the nobility of the apostles - all this
and more Hispania, onc e so delightful and now rendered so
miserable, endur ed as mu ch to its honour as to its disgrace'r' "
The chronicler's picture of his country's tribulations may be exaggerated,
but his view of history locat ed her changed fortun es firmly within the
christian historiographical tradition .

The Chronicle of 741

At first reading, the Chronicle if 741 does not address the same issues. The
author did not con cern himself with chronological exactitude, as we shall
see. H e said nothing about religious matt ers, the islamic conquest of
Hispania, or signs and portent s of the End of the World. He quoted his
sour ces selectively, perhaps even more so than the author of the Chronicle if
754. Although the author probably knew John of Biclar's Chronicle in the
version which had been extended to 602, he began with the death of
Reccared in 60 I. H e listed some but not all of the succeeding Visigothic
kings to Sisebut (6 12- 2 1), dated according to the Hispanic Era, with
noti ces summarised from the History if the Goths. The rest of the chronicle is
a series of brief accounts of the reigns of the Byzantine emperors from
Phocas (602- 10) to Leo III (7 19- 4 1). Although the first imp eria l reigns are
dat ed, the last date of any sort is the accession of Constantine, son of
H eraclius in Era 678 [640; the accep ted date is 641] . Mo st curious of all,
the chronicler ignored the history of Hispania after 621 except for the
account of al-Samh's attack on Toulouse a century later.
Although the Chronicle if 741 functions as a continuation ofJo hn ofBiclar,
it is mu ch more difficult to read as provid ential history than the Chronicle if
754.John of Biclar had focused on Reccared as the ruler responsible for the
conversion of the Goths to catholic orthodoxy. Reccared gained the throne

35
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

witho ut opposition," founded churches and monasteries and restored

ecclesiastical pr operty.' " H e summo ned a council at Toledo in 590, and
helped to destroy the Arian heresv" For J ohn of Biclar, the failure of attac ks
on Reccar ed an d his faith ." his divinely-aid ed triumphs against the Franks,"
and the peace and harmony which he established in Hi spani a were reflected
in events in the east, wh ere 'the empero r of the Persian s received the faith of
Christ and establish ed peace with the empe ror M aurice'r" Isidore
elabo rated this view of Reccar ed , portr aying him as an ideal empero r
who was 'pious and outstanding in peace . .. gloriously elevatin g the ...
people by the victory of the faith . For in the beginning, Reccar ed adopted
th e catholic faith, recalling all the peoples of the entire Gothic nation to the
observance of the correct faith '." Isidore described Reccar ed 's vital role in a
syno d which confirmed the suppression of Arianism, and his generosity
towards the church. Of Reccar ed 's victory over the Franks, Isidore wrote:
'No victory of the Goths in Hi spania was greater or even comparable. Many
tho usands of the enemy were laid low or captured and the remaining part of
th e army, beyond hop e, turned in flight with the Goths following after them
until they were cut down with in the bou ndaries of their own kingdom '.
Any H ispan ic chronicler who had read J ohn of Biclar and Isidore might
be expec ted to have included these tales of Recca red's ortho dox y and piety,
which Go d had rewarded with victory in ba ttle. As he was relying on the
History if the Goths as his source for events in Hi spania up to 62 1, the author
of the Chronicle if 741 could have followed J ohn of Biclar's pan egyric of the
king with Isidore's. Yet his use of Isidore is odd. Co mpa re the Chronicle if
74l's trea tme nt of the successors to Reccar ed with the corresponding
passage from Isidore. The phrases which the chronicler copied from
Isidore are in italics.
Chronicle if 741 3: 'Era DCXLI Vuittericus regnum, quod a Liuvan e
tyrannice invaserat, sibi vendicat annis VII; nam quia gladio operatusj uit,
gladio periit. Mors quippe innocentis Liuvani s filii Reccaredi inulta in illo
non j uit; inter epulas enim prandii a suis est interfictus'.38
Historia Gothorum 58: 'Aera DCXLI anno imp erii M au rici XX
extincto Liuvan e Wittericus regnum, quod vivente illo invaserat, vindicat
annis VII, vir quidem stre nuus in armorum arte, sed tam en expers
victoriae, namque adversus militem Rom an orum pr oelium saepe
molitu s nihil satis glori ae prater quod milites quisdam Sagontia per
du ces obtinuit. Hi e in vita plur ima inlicita fecit, in mor te cuida m quia
gladio operatusj uerat, gladioperiit. Mors quippe innocentis inulta in illo non.fuit:
inter epulas enim prandii coniuratione quorundam est intefectus. Co rpus
eius viliter est exportatum atque sepultum'; "

36
 NEWS FROM THE EAST IN THE EIGHTH-C ENTURY CHRONICLES

The chro nicler did not include everything th at Isidore said, but he stuck to
most of th e sense , and copied some oflsido re's exac t word s. Yet he had j ust
dismissed Reccared in one sentence : ' Reccared d ied after reigning for
fifteen years'i '" Although th is section deals with events before the islam ic
conq uest, it is worth looking at th e chronicler's use of the History ofthe Goths
in some detail, because it help s to explain how he used his sources for th e
lat er period .
Only th e first half ofIsidore's entry for Gundemar (6 10- 12) was copied
into th e Chronicle qf741 , without Byzantine imperi al dating: 'Era 648: After
Witteric, Gundemar was set in autho rity over th e kingdom for two yea rs',"
wh er ea s the History of the Goths continue d: 'He devastat ed th e Basqu es
during one expedition and besieged th e army of th e Roman s on ano the r.
H e died a natural death in Toled ot? The chro nicler took the same
approach in his entry on Sisebut (612-2 1), omitting Isidor e's praise of
Sisebut as a man of letters, his force d conversion of the j ews and his
milit ary victo ries. The Chronicle of 754, in contras t, summarised Isidor e's
account of Sisebut in th e first entry in this chronicle to m ention a
Visigothic king: 'He was famous for his m ilitary example and victories.
Di spatching an army, he brou gh t th e rebellious Astu rians under his
dominion. Through his gene rals, he overca me the Ru ccon es, . .. he had the
goo d fortune to tr iumph twice over the Rom an s and to subject certa in of
the ir cities to himself in battl e' .43 Sisebut was succeede d by his son
Reccared , a child who survived his fathe r only very briefly and was left out
of the Chronicle qf741. Then came Suinthila (621- 3 1) wh ose reign was the
culm ina tion of one version of th e History qf the Goths. Unfortunately, it is
impossible to know which text of Isidore th e chronicler was using. M an y
different rece nsions of th e History qf the Goths survive. They divid e int o a
sho rt version ending with th e reign of Sisebut and a lon g version to
Suinthila;" If, as Collins argue d, only th e longer version sho uld be
attributed to Isidore, he was writin g th e 'fina l chapter of wh at [he)
co nce ived of as a history of th e rise of th e Goths at th e expe nse of Rome';"
'After he had asce nde d to th e summ it of royal dignity, he [SuinthilaJ
waged wa r and obtain ed th e rem aining cities which th e Rom an
a rmy held in Hispania and, with amazing fortune, triumphed even
more gloriously than had the ot he r kings. H e was the first to obta in
the mon ar chy of th e entire kingdom of H ispani a north of th e straits
which had not been ac hieved by any previou s ruler ' r'"
This version of th e History qf the Goths ends: 'S ubjected, th e Rom an soldier
serves th e Goths, who m he sees being served by m any peoples and by
H ispani a itself"." Once again, the Chronicle qf754 echoe d Isidore's report.t"

31
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

Yet th e Chronicle of 741 made no referen ce at all to Suinthila. Lacunae of

this sort ar e often blamed on the chro nicler's possession of a defective copy
of his source, as though manuscripts becam e as motheat en within a few
years of their writing as they and their successors have become in later
centuries. Yet it is more likely tha t the compiler of the Chronicle of741 had
read an intact copy of the History of the Goths but chose not to include
Suinthila in his list ofVisigothi c kings. It may be argued that he und erstood
Isidore's line on Gothic history and recognised it as inapplicable to his own
times.
Since th e Visigothi c empire had itself been overthrown, th e compiler of
th e Chronicle of 741 may have thought it better to consign its glories to
oblivion. Yet the re was at least one altern ative approach the chronicler
could have taken - the stra tegy adopted by the compiler of the Chronicle of
754, who describ ed the Visigothic triumphs but went on to show the
migh ty cast from their thron e as punishment for their sins. The writer of
the Chronicle of 741 took a different line. Mu ch of the material tha t he
omitted to qu ote from Isidore dealt with the victor ies of the Visigoths over
the Byzantines, beginn ing with Reccared . The chro nicler was consistent in
his use of material from this section of the text. This resulted in the
omission of Suinthila, the king most me mo rable for his victories agai nst
New Rom e. The place which Suinthila should have occupied in the text is
taken by Byzantium's struggles agai nst the Persians and the islami c
conquest of Syria. By writin g a continuation ofJohn of Biclar, the compiler
of the Chronicle of741 mad e his own contribution to Visigothic history but
he took a completely original line on Byzantine history, both in the east and
in the peninsula. H e avoided any disparagem ent of the Byzantine empire
which might be implied by reference to Byzantine defeats in Hispani a. Yet
he did not use Byzantine history merely as a chro nological peg on which to
han g his story, or as another example of the sinful receiving their just
desserts. H e implied that Hispanic christians no longer had any interest
in the empire's defeat, since Hispani a and Byzantium were now allies in
the struggle to reverse the islami c con quests.

Eastern sources for the eighth-century chronicles

It would be fascina ting to pursue the chro nicler's polemic throu gh the rest
of the chro nicle by analysing his selective use of a text or texts on
Byzantium and the islam ic conquest of Syria. U nfortuna tely, this is
impossible. The historio graphy of this period in the east is obscure and
controversial." The sources are fragm entary, late and contra dictory. To

38
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONlCLES

make matters even more difficult, few historians of Hispania will be

familiar with all the languages - Greek, Syriac and Arabic are the most
common, but there are others - used by the authors of the texts whose
subject matter overlaps with that of the Hispanic chronicles. In the last few
years much of this material has been translated and put into context.50
Although no-one has yet discovered what the eighth-century chroniclers
read, one can say something about the sort of text they had at their
disposal.
Chronicle writing in the east crossed linguistic and sectarian barriers.
Several fragments of chronicles referring to the islamic conquest of Syria
survive from th e seventh and early eighth centuries." The earliest of these
is a note written soon after the battle of Yarmuk (636) in a codex
containing the Gospels of Matthew and Mark.t" Another is a chronicle
composed by a pro -Chalcedonian Maronite christian in 660s; what
remains of this text is in a nineteenth-century transcription beginning with
the civil wars between 'All and Mu'awiya (655-7).53 In 692, James of
Edessa compiled a continuation of Eusebius' Chronicle laid out in three
columns according to Persian, Byzantine and islamic chronology, with brief
historical notices which have become displaced in the surviving tenth- or
eleventh-century copy." Individual chroniclers were multilingual and may
have written in more than one language.P Conrad believes that the
widespread use of Arabic vocabulary in Syriac and Greek chronicles from
the ninth century and after and their detailed narration of muslim history,
means that these events must have been written in Arabic in the mid-
seventh century, 56 although the first surviving text known to have been
translated from Arabic is dated 724.
By the 740s, chronicles mixing Byzantine and islamic history had
probably entered a common Byzantine--Svriac tradition ." In the mid
eighth century, a Melchite monk in Palestine translated into Greek a
Monophysite Syriac chronicle ending in 746, written soon after that date
by an otherwise unknown John, son of Samuel, as a continuation of
another chronicle, now lost, which was composed between 724 and 31.
This Greek translation was taken to Constantinople in 813 after the
dissolution of the Syrian monasteries. It was read by Theophanes, whose
Chronicle 58 of c.8l3 is the one of most important witnesses to the gradual
accumulation of Syriac texts by Byzantine chroniclers. A completely
unrelated source which came out of the same tradition is the Chronicle of
John of Nikiu, a Coptic bishop born at about the time of the islamic
invasions of Egypt, who wrote in Greek and whose work now survives in a
seventeenth-century Ethiopic translation." Eastern christian chronicles
varied in content and emphasis because christian attitudes to the islamic

39
 NEWS FROM THE EAST IN THE EIGHTH -CENTURY CHRONICLES

invad ers had more to do with curre nt christological controversies th an with

actual events. In the ea rly yea rs of th e conquests, th e incom ers were
conside red to be pagans. In 634, bishop Sop hronius ofJ erusalem describ ed
th em as 'beastly and barbar ou s ... filled with every diab olical savagery'i'"
Yet th ey were not perceived as a religiou s threat until as late as th e eleventh
cen turv'" On th e con trary, N estori an christians welcomed th e end of
Byzantine per secution and th eir patriar ch Isho 'yahb (d.65 9) praised th e
new ruler s for th eir tolerance of christians.f
Links betw een Hispania and Byzantium may hav e been m aintain ed
after th e conquest of 711 , giving Hispanic christians access to Byzantine
histori es writte n in Greek. H owever, th ere is little evide nce th at such ties
survived in Hispania, in co ntras t to th e situa tion in Rome and Fran cia."
No r ca n the eastern mater ial in th e two chronicles be attrib uted to any
known Byzantine text , simply becau se non e surv ive which are ea rlier than
th e ninth cen tury, and analysis of th e sources used and the gaps in one of
th ese later histori es, th e Short History of Nikep ho rus (d.828),64suggests eithe r
th at such histori es were never writte n, or th at they never becam e well
known . It would be tedi ou s and unprofitabl e to m ake a line-b y-line
comparison of all the surviving texts with th e Chronicle if 741 and the
Chronicle if 754. They all dem on strate commo n features with th e H ispani c
texts, but th eir simila rities are outwe ighe d by th eir differ en ces. Dubler
compared th e histories of T heophanes and N ikep horus with the Chronicle if
741 6, and found th at th ey are sim ilar in outline, but not in detail. Noldeke
postulated a M onophysite text written in Syria as the commo n source for
the eighth-century H ispan ic and th e ninth-century Byzantine chron iclers."
The exten t of the problem ca n be illustrat ed by one of the chro nicles
mention ed above, th e 'Maronite' chro nicle." Its author conce n tra ted
on the few Byzantine victori es worthy of mention in order to enco urage his
read ers in th e hop e that th e empire might be restored . So did th e compiler
of the Chronicle if 741, as we shall see. The 'Maronite' also discussed th e
war s between 'All and Mu'awiya which led to th e first and most important
schism of Islam, over th e succession to Muhammad . The H ispani c
chro niclers, however, took a different line, which was com mo n to most of
th e othe r Syriac and Arabic histori an s, om itting an y referen ce to 'All, and
referring only in passing to th e civil wars : 'Prese n tly M oabia obtained the
kingd om , ruling for twe nty-five yea rs. Alth ough he fou ght civil wa rs for five
yea rs, in truth the Ishmaelites were obe d ient to him for twen ty years and
he bro ugh t his reign to a successfu l conclusion'i '" Even a brief comparison
between the Chronicle if 74 1, the Chronicle if 75 4 and th eir pu tative easte rn
mod els shows that it is impossible to ide nti fy which chro nicles from the east
were read in Hispani a in the eight h cen tury.

40
 NEWS FROM THE EAST IN THE EIGH TH-CENTURY CHRONICLES

Lop ez and Collins postulated a number of different sources for the

eastern material in the Chronicle 0] 754. Collins thought there mu st have
been both a Byzantine chronicle source and a 'work on the rulers of the
Arab s, that in structure modelled itself closely on Isidore's Historia
Gothorurn' , which he lab elled the Historia Amburn, supposedly writt en in
the reign of Hisham by a christian with a knowledge of Islam, and Syrian
and Umayyad sympathies.t" Lopez suggested a similar text as the common
source for the history of the caliphs in the Chronicle 0] 741 and the Chronicle 0]
754 70 and another Arabic work , available in a Latin translation , as the
source for the entries in the Chronicle 0] 754 dealing with events in Hispania
between 711 and 740.71 Modern historiography of Hispania in the
medi eval period is full of 'lost chronicles' ; Sanchez-Albornoz was
particularly fond of postulating such links between texts otherwise difficult
to connec t. From the evidence of the two surviving chronicles alon e, it is
possible to simplify the argument a little.
The eastern material used by the com pilers of the Chronicle 0] 741 and
the Chronicle 0] 75 4 is closely related. The latter chronicle could have been
dependent on th e former, although Lop ez thought not. 72 Hi s argument is
two-fold. Firstly, the chronologies are slightly different. The Chronicle 0]741
places Yazld's attack on Constantinople in the reign of Constantine IV
(668- 85), wh ereas the Chronicle 0] 754 correc tly dates it to the reign of his
fath er C ons tan s (64 1-68).73 We sho uld not perhaps be too concerned
about errors in chronology. A Syriac writer of the eighth century
recognised the problem when compiling his own chronicle, and added
these consoling remarks:
'If anyon e who reads this is of a mind to despise it, he should reflect
that, just as affairs and doings of various kinds do not occur in one
place alone, . .. so it is here also. If he has consulted a history which
does not match this one, he should realise that not even the authors
of form er times agree with on e another... . It causes no injury to the
discerning and the god-fearing if a date is on e or two years out either
way... .' 74

Dating events in the seventh century is a problem even for mod ern
histori an s who ar e able to compa re many different texts. The 'mista kes' in
the Chronicle 0] 741, how ever, may not mean that the chronicler read a
different source from that used by the compiler of the Chronicle0]754. Dates
may have been alter ed for em phas is. For example, the dat e for the
access ion of Ph ocas (602- 10) is given as Er a 642 [604] but this entry was
placed after the entry on Witteric (603- 10), who was murdered after
seizing the throne from Liuva. Perhap s the chronicler wanted to emphasise

41
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

the blood y fate of all tyra nts. Ph ocas' usurpation was punished when 'the
Persians, leaving their own lands, were successful again st th e Rom ans.
They subjected Syria , Arabia and Egypt, expelling the Romans', and
Ph ocas himself was later mu rdered." Other instan ces of the chro nicler's
manipulation of his material will be considere d later. Lopez' second reason
for thinking that the two chro nicles are not closely relat ed was the
observa tion that there ar e a few passages in the Chronicle if 754 which do
not appear in the Chronicle if 741. The Chronicle if 754 lists the years reigned
by Constantine, son of Heraclius and Constantine IV and notes the trib al
or igin of Abu Bakr. These differences in content, however, are min or in
compa rison with the great overlap in conte nt and vocabul ary between the
two texts. Like th e two chroniclers' use of the History if the Goths, their
selection of the news from th e east dem onstrates two different approaches
to similar, perhaps identi cal material.
As a compre hensive analysis of the eastern history copied into the two
chro nicles would take too long, the following selection must serve to make
th e point. The Chronicle if 741 begins a little earlier than the Chronicle if 754,
with the usur pa tion of Phocas. This is followed by a passage which is
almos t identi cal to the opening of the Chronicle if 754, on Heraclius'
rebellion aga inst Ph ocas:
Chronicle if 741: 6: 'Era DCXL VIIII Romanotum L VI Eraclius imperio
coronatur, qui rebellionem adversus Focam ex Africa moliens ob Flaoiae
nobilissimae virginis illi apud Africam desponsatae atque iussu Focaeprincipis de
Libyaefinibus Constantinopolim deportatae, tali causa praedictu s princeps
corre ptus, armatis atque aduna tis totius occidentis viribus, navale
proelium contra rem pulicam cum mille et amplius navibus peragit,
Nicitamque magistrum militi ae Rom anorum ad gregati terrestris
exercitus ducem facit, tali sub pacto invicem dljinito, ut quiquis illorum
primus Constantinopoli adventaret, illi totius administrati o contra deretur
imperii. Igitur Eraclius Africa degrediens ocius regiam urb em
navigando pervenit. Qyem aliquanto obsistentem bello adgressus cst.
Sicque Bizan tii Focam captam Eraclio offerunt iugulandurri '."
Chronicle if 754: 1: 'Era DCXL VIIII Romanorum L VII Eraclius imperio
coronatur. Regnat annis XXX peractis a principio mundi a nnis
VDCCCXXXVIII. Hi s ob amo rem Ptamae nobilissime virginis, illi apud
Africam ante sumpto imp erio desponsatae et iussu Foce principis ex Libie
finibus Constantinopolim deportate, rebellionem adversus Focam cum Nichita
magistro militie moliens contra rem publicam consilio definito, Eraclius
eq uoreo , Nichita terrestris exercitu adunato tali invecem dljiniunt pacto, ut
quisquis eotum primus Constantinopolim adventaret, in loca corona tus digne

42
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

frueretur imperio. Sed Eraclius ab Africa navali ascendens collegio

ad regiam usque ocius pervenit navigando. Q!tem aliquantem obsistentem
in bello Focam Bizantio cap tum flammigero ferunt gladio . Qui mox
eum perspicit iugulatum, ilico imperio sublimatur'.
The two chroniclers continued to take the same line on Heraclius' Persian
wars , often using identical words, but their treatment of the islamic
conquest of Syria was different. They made the same mistake in dating the
islamic invasions to the seventh year of Heraclius' reign c.618 , perhaps
the result of a misreading of VII for XII by their source . Both attributed
the Saracens' success to guile rather than military prowess." Yet there is
hardly any overlap between the vocabularies of the two accounts of the first
Saracen incursions. The author of the Chronicle qf 754 introduced
disparaging references to Muhammad that are absent from the Chronicle
qf 741, a difference in emphasis which becomes clearer after reading the
next chapter of the Chronicle of 741 : 'Assembling a very large army,
the Saracens invaded Syria, Arabia and Mesopotamia; they were ruled by
Mahmet [Muhammad], scion of the most noble of their tribes , a most
prudent man about whom many stories would be told'." It is not surprising
that this passage did not appear in the Chronicle qf 754, since it was difficult
to harmonise with the author's view that the invasion of Hispania by the
'Saracens' was a disaster. Similarly, the Chronicle of 754 omitted to mention
the death of Muhammad, whilst the Chronicle qf 741 noted that, after his
death, Muhammad was venerated as a prophet." There are several other
disparities between the Chronicle of 741 and the Chronicle qf 754. Briefly, the
two chronicles give different accounts of the accession of Constans, and
the Chronicle qf 741 has passages on the reign of Marwan, 'Abd al-Malik
and Leo III, and on the victories of Maslama ibn 'Abd al-Malik in Western
Anatolia which the compiler of the Chronicle qf 754 omitted or shortened.
Some of these passages will be discussed later. Overall the two accounts are
so similar that they appear to have been derived from the same source,
The news from the east in the Chronicle qf 741 stops before that in the
Chronicle of 754. Since the author of the Chronicle qf 741 noted that Leo
(717- 4 1) reigned for twenty-four years, but did not name his successor, it
has becn assumed that he stopped writing in 741. Of the reign of Leo he
said only : 'T he sixty-seventh emperor Leo , expert in the military arts, as
the Saracens were hurrying to storm the royal city (Constantinople),
received the royal sceptre for twenty-four years , with the acclamation of
the senate' i'" In the previous entry, the chronicler had implied that Leo
raised the siege of Constantinople imposed by Maslama ibn 'Abd al-Malik
in 717, but he did not cite his other victories, including that at Akroinos in

43
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

Western Anatolia. It is possible that he did not know of these , since the
Chronicle if 754 does not note them either. Yet the later chronicler knew of
at least one contemporary event, the victory gained by Hisham against
the Byzantines in 723.8 1 Similarly, the author of the Chronicle if741 seems to
have known more about the caliphs than he wished to mention. He listed
most of the successors to Muhammad, ending with the accession of al-
Walld (743-4). The brief entry about aI-Walid links him with his
predecessor Hisham without giving details about Hisham's reign, which
the Chronicle if754 treated at some length. 82 Collins postulated that the first
version of the chronicle, or its source, stopped in the reign of Hisham, and
that the final entry on al-Walld must have been added later." However, the
Chronicle if 754 continued to incorporate eastern material for the period
after 741 which is very similar to what had gone before. It is possible,
although it seems unlikely, that the later chronicler had two eastern sources
which were so similar that he was able to switch to the second when the
first gave out, just at about the time when the author of the Chronicle if 741
stopped writing. But, unless the surviving manuscript is incomplete, it may
be argued that the compiler of the Chronicle if 741 deliberately put down his
pen with Leo's raising of the seige of Constantinople because he had come
to the end of his story.
The news from the east in the Chronicle if 741 was carefully selected to
tell this story. Comparison with the Chronicle if 754 helps to strengthen
some general impressions about his selection criteria. The compiler of the
Chronicle if 741 chose from the material on Heraclius mainly his successes
against the Persians, although it is odd that neither of the two chroniclers
knew about, or chose to include, Heraclius' recapture ofJerusalem and the
restoration of the True Cross. In his account of the islamic campaigns in
Syria , the compiler of the Chronicle if 741 mentioned Heraclius only in
connection with measures taken to defend the empire and attributed the
Byzantine defeats to Theodore's failure to take his brother Heraclius'
advice not to campaign against the Saracens. This is a topos which the
Hispanic chronicler shared with several eastern historians of the islamic
conquests, who avoided blaming Heraclius for the Byzantine defeats, even
though his support of the Monophysites made him theologically suspect."
The islamic armies were able to take over large areas of the Byzantine
empire, it is implied, because the Byzantines suffered a series of
usurpations; the chronicler had already emphasised his disapproval of this
road to the throne in his treatment of Phocas and Witteric. Not until the
succession of Leo were the islamic armies defeated, when Leo raised
the siege of Constantinople. Here the story ends, apparently with more
which the chronicler could have said, but chose to leave out.

44
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

The defeat of al-Sanll;1 at Toulouse

The story of Byzantium told by the Chronicle if 741 may be read as a long
prologue to his account of the defeat of al-Samh at Toulouse, an islamic
defeat which was closer to the chronicler's home. Any discussion of the
Chronicle if 741 must face up to the problems posed by this passage, which is
placed anachronistically as the penultimate entry in the chronicle. Here the
chronicler returned to events in the peninsula, which he had passed over
for a century after the reign of Sisebut (612-1). The passage seems to have
been copied from the Chronicle if 754:
Chronicle if 741 : 42: 'Galliam quoque Narbonensum per ducem exercitus
Mazlema nomine suam facit gentemque Francorum jrequentibus bellis
stimulat. Atque incongruenti virtute iam dictus dux exercitus Tolosam usque
pervenit, eamque obsidione cingens.fundis et diversis generum machinis expugnare
conavit. Francorum gentes talio de nuntio certae apud ducem ipsius gentis
Eudonem nomine congregatur. Sicque collecti Tolosam usque perveniunt.
Apud Tolosam utraque exercitus acies graui dimicatione corifligunt. Zema ducem
exercitus Sarracenorum cum parte exercitus sui occidunt, reliquum exercitum per
.fugam lapsum sequuntur' .85
Chronicle if 754: 57: 'Tunc in Occidentis partibus multa illi
preliando proveniunt prospera; atque per Zama nomine tres minus
paululum annos in Spania ducatum habente ulteriorem vel
citeriorem Iberiam proprio stilo ad vectigalia inferenda describit;
preda et manu<bi>alia vel quidquid illud est, quod olim predauiliter
indiuisum retemtabat in Spania gens omnis Arabiae sorte sociis
diuidendo partem ex omni re mobiIi et inmobili fisco adsociat.
Postremo Narbonensem Galliam suam facu gentemque Francorum jrequentibus
bellis stimulat et seditas Saracenorum in predictum Narbonensem
oppidum ad presidia tuenda decenter conlocat. Atque inconcurrenti
virtute iam dictus dux Tolosam usque preliando pervenit, eamque obsidione
cingens.fundis et diversis generum machinis expugnare conavit. Francorum gentes
talio de nuntio certae apud ducem ipsius gentis Eudonem nomine congregantur.
Ubi dum apud Tolosam utrique exercitus acies graui dimicatione corifligunt.
Zama ducem exercitus Sarracenorum cum parte multitudinis congregata
occidunt, reliquum exercitum perfuga elabsum secuntur. Quorum Abdorra-
man suscipit principatum uno per mense, donee a principalia iussa
veniret Ambiza eorum rector . .. 86
Collins argued that the passage on al-Samh's defeat in the Chronicle if 741
could not have been copied by the author of the chronicle but must have
been added later." This is possible, but it does not affect the incongruity of

45
 NEWS FROM THE EAST IN THE EIGHTH- CENTURY CHRONICLES

this passage. The entry, however, may be intenti onally misplaced . The
passage has other strange features. Taken togeth er, they suggest that
the chronicler was deliberately drawing parallels between the history of the
islamic conquests of th e Byzantine empire and of Hi spani a.
The bald stateme nt in the Chronicle qf 741 shows the Arabs, defeated,
retreatin g befor e the chr istian forces. Readin g on in the related entry from
th e Chronicle qf 754, this impression soon van ishes. The invad ers were not
under threat of expulsion. On the con trary, they were stabilising their
control of al-Anda lus. The defeat at Toulou se did not hinder the peaceful
succession of mu slim governors. The Chronicle of 754 report ed that Eudo,
the victor of Toulouse, was him self defeated ." The compiler of the Chronicle
qf 741 ame nde d as well as shortened the passage by brin ging into the story
a certa in M azlem a who does not feature in the Chronicle qf754 at this point.
Now, th is might be explained as a slip on the copyist's part, writing
M azlem a instead of Zema. It may, however, be a deliberate reference to
M aslam a ibn 'Abd ai-M alik, the victor of several campa igns against
Byzantium. T he two chronicles have slightly different acco unts of
M aslam a's trium phs in Asia Min or, his failed attack on Co nstantinople
and his defeat of Yazld ibn M uhallab 'in the Babylonian fields', which
occurred during the reign of Yazid II (720 - 4).89 The Chronicle qf741, but
not th e Chronicle qf 754, went on to say tha t M aslam a 'triumphed against
Rom ani a' in the West throu gh the lead ers of the army. The phrase in
occiduis denique partibus presum ably refers to Western Ana tolia, but the
chro nicler was perh aps being deliberately vague about what he meant by
'the west' in order to prepare the rea der for his next entry, the participation
of M aslam a in the defeat at Toulouse. Although the events at Toulouse
should be placed before M aslam a's defeat of Yazid ibn Muhallab, the
chro nology is plau sible, even if the involvemen t of Maslama is less so. The
Chronicle qf 754 dated the defeat at Toulouse to 719 but this may wron g
becau se th e chro nicler also said that al-Samh had 'held power in Hispani a
for a little less than three years', which put s the attac k on Toulouse closer to
72 1. It is imp ossible to acco unt for all M aslam a's movem ent s during this
period and the datin g of those campa igns which were noted by the eastern
chro niclers is, of course, uncert ain . Mas lama besieged Co nstantino ple in
7 16- I 7 and next appea rs in the Arab ic histories as the governor of Iraq at
th e time of the struggle for the succession to 'Umar I in 720 .90 T here is no
evide nce in any of the easte rn histories tha t Maslam a was active in
Hi spani a. By introducing M aslam a into an acco unt of the failed attack on
Toulouse, the autho r of the Chronicle qf 741, prompte d perhaps by the
similarity between the nam es Zema [al-Samh] and Maslama , may have
been using M aslama, a famous mu slim general and bro ther to the caliphs

46
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

Sulayman and Yazid II, to draw attention to parallels between the events
before Toulouse and Constantinople. In this tightly-edited version of recent
history, the chronicler seems to be claiming that the conquerors could be
beaten, even one of the greatest among them. Perhaps the writer was trying
to do no more than encourage his fellow christians with the hope that,
since the vulnerability of the conquerors had been demonstrated in
Hispania just as it had before Constantinople, the invaders might
eventually be expelled. He may have been going further than this. Before
putting forward an even more speculative hypothesis, it is necessary to
make another excursus, returning to Byzantine and Syriac texts, but this
time to look not at temporal history but at stories of the Last Days.

The Last Em.peror and the defeat of Islam

The Church's official line on the end of the world, that it was merely
symbolic, failed to engage with popular concerns. Texts on the Apocalypse
circulated and influenced the writing of history, particularly in the east."
Seeking to put the islamic invasions into the context of universal history,
Syriac authors incorporated the Book if Daniel's picture of the succession of
world empires into the Eusebian model. Sybilline prophecies dating from
before the islamic conquests influenced the author of the text known as
Pseudo-Methodius, which was probably written in Syriac in the
mid-seventh century in Singana in northern Mesopotamia.F It was
translated both into Greek and into Latin sometime before 800, was
widely disseminated and became the most influential apocalyptic text. The
author may have been influenced by jewish Messianic hopes. He also knew
the legend that Alexander the Great would return in the Last Days, which
passed into the written record in Syriac shortly after Heraclius' victory over
the Persians. Pseudo-Methodius' text also has much in common, in
conception, if not in detail, with the World Chronicle (c.686) of John bar
Penkaye from the monastery of Bezirkes Qardu, not far from Singana
where Pseudo-Methodius' text may have been written. Both authors saw
the islamic conquest as part of a divine plan to punish the Byzantines for
their religious deviations. A similar view was expressed by the author of a
chronicle written in Armenian soon after the conquest" and it would later
find widespread acceptance through the anti-muslim polemic of Timothy,
patriarch of Baghdad (d.823).94John bar Penkaye expressed the idea clearly:
'When the kingdom of the Persians came to an end , in the days of
their king Khosro, the kingdom of the children of Hagar [the

47
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

'Saracens'] at once gained control over more or less the whole world,
for they took the whole kingdom of the Persians, overthrowing all
their warriors who prided themselves in the arts of war.
We should not think of the advent [of the children of Hagar] as
something ordinary, but as du e to divine working. Before calling
them, [God] had prepared them beforehand to hold christians in
honour; thus they also had a special commandment from God
concerning our monastic situation, that they should hold it in
honour. Now when these people came at God's command, and took
over as it were both kingdoms, not without any war or battle, but in a
menial fashion , such as when a brand is rescued out of the fire; not
using weapons of war or human means, God put victory into their
hands, in such a way that the words written concerning them might
be fulfilled namely, 'One man chased a thousand and two men
routed ten thousand' (Deuteronomy 32: 30).. . . Only a short period
passed before the entire world was handed over to the Arabs.
Only half the Byzantine empire was left to them'.95
The author of the Alexander legend had warned his readers that the
Persians could not save them. Pseudo-Methodius told his audience that
their help lay with Byzantium. He listed the historic victories in which
the Romans [Byzantines] had overcome various opponents, ending with
the repulsion of the Avar siege of Constantinople in 626. Relying heavily
on Psalm 68: 31, Pseudo-Methodius predicted that a Byzantine emperor
would hand over the empire to God in the Last Days after the destruction
of his enemies. This prediction was abbreviated and interpolated into
works belonging to a popular genre known as the Visions qf Daniel, of which
many very different versions survive. As the return of the Byzantine empire
to its former glories looked increasingly unlikely, the authors of the Visions
qf Daniel abandoned Pseudo-Methodius' pro-Byzantine stance and
concentrated on his descriptions of the islamic invasions in apocalyptic
terms. In many of these texts, the Last Emperor, often unnamed, suddenly
appeared at a specific moment when oppression was at its height. In other
texts, only the prediction of a final victory over the enemy indicated the
authors' debt to the apocalyptic tradition.l" Yet the search for a Byzantine
Last Emperor continued. Some of the Visions ofDaniel used the metaphor of
a lion. During the reigns of Leo III and his son Constantine V, oracles
about victories of a lion's whelp were in circulation.
A version of the story of the Last Emperor passed into Arabic
historiography.97The following prophecy was attributed to Muhammad in
the Book qf civil wars of Nu'aym ibn Hammud (d.843):

48
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

'It is in the hands of the fifth of the family of Heraclius (al-Hiraqi) that
the wars (malahim) will take place . Heraclius may ru le, then after him
his son Constans (Qusta) son of Heraclius; then his son Constantine
(Qustan{fn) son of Constans; then his son Justinian (Ustanilln) son of
Constantine, then the kingship of the Romans moved from the
family of the Heraclius to Leo (Liylln) and his son after him . But
th e kingship will return to the fifth of the family, at whose hands the
wars will take placc'.?"
Several variants of the prophecy were copied into the same manuscript,
which is a collection of quotations from Syrian and Egyptian scholars .
Most of these include a corrupt regnal list, in which the fifth of the family of
Heraclius is named Tibiris. [Leo's son was Constantine]. The idea that
the Last Emperor was called Tiberius is interesting. It is unlikely to have
anything to do with the historical Tiberius, who was made regent with
Justinian and murdered with him, while still a small child, in 711, although
it wou ld have reminded readers of the Tiberius who was emperor at the
time of the crucifixion. In another passage copi ed into the same
manuscript, a man calling himself Tiberius appears to Maslama ibn
'Abd al-Malik during the 716-17 campaign against Constantinople.
Maslama himself was cited as the source of this story. Tiberius was
denounced as false by Abu Muslim al-Rumi, a christian convert to Islam ,
who prophesied that the real Tiberius would lead a huge Byzantine army
into a last battle in which the muslims eventually, with only a few soldiers
remaining, would defeat the Byzantines. Neither Tiberius appears in other
Arabic sources, but Theophanes included a related story in his Chronicle:
'Annus Mundi 6229 [737]: In this year, Sulayrnan, son of Hisham
took many captives in Asia among them a certain native of Pergam en
who claimed to be Tiberius, son ofJustinian. In order to honour his
own son and to frighten the emperors, Hisham dispatched this man
to Jerusalem with the appropriate imperial honours, namely a guard
of soldiers with banners and sceptres , and decreed that he should
tour all of Syria with great pomp so that all should see him and be
amazed' i''"
The meaning of these stories is obscure. They have parallels in stories
about Theodosius, the son of the emperor Maurice, who was believed to
have been murdered by the usurper Phocas, whose reappearance was the
theme of several legends . Tiberius' links with the siege of Constantinople
may be particularly significant. The siege fell at a chronological turning
point, on the eve of the first century of the Hegira, which may have

49
 NEWS FROM THE EAST IN THE EIGHTH-CENTURY CHRONICLES

increased its resonance for both muslims and christians.l'" It may have
given rise to many variants on the Last Emperor legends, including the one
which was remembered in connection with Maslama ibn 'Abd aI-Malik,
the man who failed to bring down the Byzantine Empire.
Apocalyptic texts are even more difficult to interpret than the sources
for Byzantine and islamic history discussed earlier. The picture is confused
by garbled transmission, by the loss of intermediary texts, and by the
concordance and discordance amongst the Arabic, Syriac and Byzantine
sources. There is no evidence that any of these texts reached Hispania.
Nevertheless, the vocabulary of apocalypse was one which early medieval
Hispanic authors were able to exploit. Apocalyptic writing in Hispania
began with Hydatius, who may have been spurred on to produce his
chronicle by his belief that the world would end in 482 . 101 Hydatius was
concerned with signs and portents of the End of the World . Isidore had
inadvertently increased millenarial tension by turning Augustine's six ages
of history into real time,102 although he emphasised the biblical injunctions
against calculating the years remaining in his Sixth, and last, Age of the
world.l'" Julian of Toledo attempted to bypass the Last Days by
recalculating the end of sixth millennium to a date which had already
passed unnoticed,'?' but this ruse did not put an end to millenarial
speculation. Beatus of Liebana composed a Commentary on the Apocalypse and
in 883, the Asturian author of the Prophetic Chronicle was to use Biblical
calculations to predict that the Arabs would be expelled from the peninsula
in the following year. 105 The islamic conquest of Hispania featured in
apocalyptic texts from outside the peninsula. One surviving, corrupt
manuscript, called the 'Vision of Daniel concerning the Last Time and
concerning the End of the World', probably written in Constantinople in
867 or 869, includes a fragment referring to the conquest of Hispania and
the threat to Aquitaine.l'" The author of this text named Leo III as the Last
Emperor. If the stories of the Last Emperor had reached Hispania in the
mid-eighth century, they would have found a receptive audience.
Each of the eighth-century Hispanic chronicles can be read as a version
of the Last Days as they were being acted out in the peninsula. The
message of the Chronicle if 754 was that the followers of Muhammad had
been divinely ordained to punish God's people for their sins. The story
began with the conquest of Syria, Egypt and Mesopotamia and ended with
the 'ruin of Spain'. The chronicler evoked the famines and signs in the sky
characteristic of apocalyptic writing, and ended his chronicle with a
discussion of the age of the world , which must be calculated in order to
know 'the fullness of time' .107 The author of the Chronicle if 741 did not
write history in this way but he too was using his eastern source to tell a

50
 NEWS FROM THE EAST IN THE EI GHT H- CENTURY CHRONICLES

story abo ut H ispani a. By omission and the rearranging of events from

eastern history, he created a new, pro-Byzantine version of history which
fed into the history of his own times. By painting a favourable portrait of
H eraclius and sho wing that it was not Heraclius' sins but the usurpation s of
his successo rs that wer e punished by the expa nsion of the islami c empire
and by ending with Leo 's defeat of M aslam a ibn 'Abd ai-Malik before
Consta ntino ple, and M aslam a's p articip ation in th e defeat of th e
conq uerors at Toulouse, he mad e a coded referenc e to the Last Emperor
and the Last Days in which the invad ers would finally be overcome.

51
 CHAPTER FOUR

-~-

The martyrs of Eulogius

'The pr eservation of our church is no thanks to this infidel people,

into who se hands Spain fell, for our sins, after the ruin of the Gothic
kingdom - which for a long tim e, mo st fortunatel y, was strong in
veneration of the christian faith, flourish ed in the worthiness of the
vener able priests and shone with the construc tion of wond erful
basilicas - but for the sake of her Redeemer and for her sake was
continually worthy to be defended by the company of the church,
who said: ':-\s the lily among thorns, so is my love among the
daughters" , and again: "in the midst of a crooked and perverse
nation, among whom ye shin e as lights in the world"'.'

1 n this passage Eulogius seems to be taking up th e story of the Andalusi

christians , suffer ing under muslim rule, from the Chronicle of 754. For
perhaps the first tim e since the conquest the christians responded violently
to their subj ugation to Islam . Between 850 and 857, forty-eight christians
wer e executed in Cordoba for apostasy from Islam or for blasphemy. The
first to die was Perfectus, who was denounced before a judge becaus e he
had expressed his condem nation of Islam to a group of muslims." Isaac,
who died in 851 , was the first to go to his death voluntarily.3 H e had retired
to the monastery of Tabanos, outsid e Cordoba after a successful career as a
secretary in the Cordoban administration but returned to the city to make
a stand against Islam. Fluent in Arabi c, he was able to harange the judge in
his own language. Isaa c's death prompted a wave of imitators. Seven men
and women met their end over the next four da ys and five more followed
them before the end of the year. M any of the martyrs wer e the children of
mixed marriages between christians and muslims and wer e experiencing
th e tensions of conflicting loyalties. Flora was the daughter of a christian
mother and a muslim fath er. After her fath er died , she was brought up as a
christian - in secret, becaus e, under islamic law, if a christian wom an
married a muslim, all her childre n had to be brought up as muslim s.

52
 THE MARTYRS OF EULOGIUS

Eventually the strain proved too great for Flora to bear and she ran away
from home. Her brother sought her return, and denounced her to the
authorities. Flora was imprisoned but offered the chance of release if she
would return to the faith of her father, but she denounced Islam before
the judge, in the company of Maria, the child of a christian father and a
muslim mother who followed her brother to voluntary martyrdom." Mixed
marriages do not provide the motives of all the christians who courted
death in the next six years . They came from monasteries and from the laity.
Many were Cordobans, but the movement swept up christians from
outside the capital - and, in the case of George, from the monastery of
Mar Sabas nearJerusalem - in a wave of self-immolation. Eulogius himself
met the same fate and his death in March 857 may have brought the
movement to an end.
Even through the words of their champion Eulogius it is clear that the
martyrs were a group of extremists whose actions were at best an
embarrassment to the church in Cordoba. The church's response to the
first round of protests was to round up and imprison a number of priests ,
including Eulogius, in an attempt to damp down enthusiasm for
martyrdom. Christians were protected under islamic law and although
there were restrictions on displays of christian ritual and symbols, religious
allegiance was not generally a matter of public concern. The church had
been able to conduct its affairs without harassment from the islamic
authorities, holding councils in Cordoba. Even the idea that blasphemy
against Islam was automatically punishable by death may be questioned.
This explanatory detail was added to a version of the passions of George,
Aurelius and Nathalia which Usuard of St-Germain took to Paris : 'The
Saracens think that only those who leave their sect and turn to the christian
faith , and those who utter blasphemies against their Legislator, deserve
death'." Yet, although the Qur'an ruled against blasphemy, it is far from
clear how often the death penalty was imposed. According to islamic law,
the perpetrator could not be punished if he was insane, and attempts were
to be made to persuade the blasphemer to withdraw his accusations. A
ninth-century Andalusi legal scholar, al-'Utbi, whose work can be partially
reconstructed from the commentary which the grandfather of the famous
Ibn Rushd [Averroes] wrote on it in the twelfth century, gives a very
moderate judgement on this subject." Moreover, the vocabulary of
blasphemy could take on different meanings at different times and places .'
Without the corroboration of Arabic sources for this period, we have only
Eulogius' word that the martyrs faced certain death. Eulogius set out to
defend the martyrs from the charge of bad faith . He prefaced his Memorial
of the Saints with a book justifying the actions of the martyrs and returned to
53
 THE MARTYRS OF EULOGIUS

the same theme in the Apologetic Book if the Martyrs. What need was there for
christians to seek martyrdom? What persecution had these martyrs faced,
and what miracles had they performed? The brief details which Eulogius
offered about his subjects were buried under the weight ofjustification and
apologetics.
The writings of Eulogius and his friend Alvarus of Cordoba make up
the bulk of the evidence for the christians of al-Andalus, and they have
been extensively studied. The mountain of publications on the martyrs of
Cordoba under which the student of this period is submerged illustrates
one of the pitfalls of the patchy evidence for the early Middle Ages - the
danger of accepting a snapshot as representative of the whole. Although
the period of the martyrdoms is less than ten years, Eulogius' writings have
been used describe the situation of the christians in al-Andalus during the
whole of the Umayyad period. From here, historians have taken the short
step to using this material to confirm their own interpretations of the past."
Spanish historians to the middle of the twentieth century, following
Simonet's lead , emphasised religious oppression, to which Andalusi
christians responded with a continuous tradition of heroic resistance .
Some saw in the martyr movement the first stirrings of nationalism. Later
writers on the martyrs have concentrated on social tensions . What is rarely
emphasised is that the story was written almost entirely by one man.
Eulogius' contemporaries shared his spiritual concerns - and deadly
writing style - but with the exception of Alvarus, in his biography of
Eulogius, they did not comment on the martyrs.
In early medieval Hispania, Eulogius' tomes seem to have remained
unread. Perfectus and Emila were included in the Calendar if Cordoba, but
none of the other martyrs were commemorated in their homeland. Some
of them were listed in the martyrology compiled by Usuard, who obtained
the relics of George, Aurelius and Nathalia for his community of
St-Germain in Paris ." Apart from this, the Acts of the Cordoban martyrs
survived only in the copy of Eulogius' writings which may have been taken
to Oviedo with Eulogius ' relics by Dulcidius, who went on a mission to
Cordoba in 883. 10 The Inquisitor General Pedro Ponce de Leon
rediscovered Eulogius' works in the sixteenth century. The manuscript
itself was lost and only the copy made by Ambrosio de Morales at about
this time saved the martyrs of Cordoba from oblivion. Wolf acknowledged
the problems raised by an uncorroborated source, but shied away from the
implication that we cannot be sure whether to believe what we read. He
explored the difficulties Eulogius faced in writing conventional hagiography.
This inclined him to believe Eulogius simply because his material appeared
so intractable: 'We can trust Eulogius at least to identify the martyrs and to

54
 THE MARTYRS OF EULOGIUS

inform us of the circumstances surrounding their deaths' . 11 Coope's

sociological approach to christian-muslim relations in Cordoba depended
on her acceptance of Eulogius' account: 'Writing for neighbours and
contemporaries, they wer e free to exaggerate, embellish and reinterpret,
but not to make up events out of whole cloth'. 12 Yet, before we can draw
any conclusions from Eulogius' picture of Cordoba in the 850s, it is
necessary to reach a judgement on the value of his testimony.
Eulogius is a particularly interesting witness to the past because we
know as much if not more about the biographer as about his subjects. The
two principal sources for Eulogius' life ar e Alvaru s' Lift if Eulogius'? and
the letter Eulogius wrote to bishop Wiliesindus of Pamplona on his return
from a journey to Navarre.!" Apart from Eulogius ' own writings , the only
other source for his life is a brief reference to his meeting in Cordoba with
Usuard of St-Germain.'! It is thought that early medieval writers were so
hidebound by literary convention that th e personalities of the individuals
they described became buried in a mass of topoi. and did not reemerge until
'the discovery of the individual' in the twelfth century. 16 It would be rash to
conclude that Alvarus' biography gives us a psychological profile of the
man. Nor can we be certain that Eulogius ' visit to Navarre was one of
the key events in his life. This would be making the same mistake with the
eviden ce for Eulogius as has been made with the martyrs in general , by
making a whole from these parts which may be the chance survivors from a
larger collection of material and which are subject to distortion of emphasis
and perhaps of fact. Yet, in his writings on the martyrs, Eulogius returned
many times to themes which ar e pres ent in the Lift and in the letter to
Wiliesindus. There is much to be gained from looking at Eulogius'
biography as presented in these documents, to see what influence it had on
his account of the martyrs of Cordoba.

The Life of Eulogius

Alvarus wrote his Lift if Eulogius from the perspective of a lifelong friend :
'Setting out to write the passion of the blessed and most learned
martyr Eulogius, I thought that before describing the glorious
combat of his end , I should include an ord ered narrative of his life.
I am doing this, in the first plac e, so that the reader might know who
and what he was, and then to demonstrate clearl y that he deserved
the palm of victory. Trusting in the help of Our Lord and Rede emer,
at the beginning of this work I declar e that I am not writing about

55
 THE MARTYRS OF EULOGIUS

doubtful matters that I have been told, but what I myself have seen
and felt. Through the grace of God we were united from the first
flowering of youth with the single tie of sweetness, of affection, and of
love of the Scriptures. We lived, if not in the same station in life
[Alvarus was a layman] , guided by the yoke of all subjects with equal
attachment. He, adorned with the priestly office and carried on the
wings of virtue, flew higher than I, stained by the filth of luxury and
concupiscence, who still walk the earth. Thus I have set myself to
narrate events, not uncertain and learned from the lips of others, but
things done in my presence and which I have known'."
Eulogius was born into 'a Cordoban senatorial family' .18 His parents
dedicated him at an early age to the church of San Zoilus in Cordoba,
where he soon became noted for his scholarship and piety. He met Alvarus
when both were students of the abbot Speraindeo. The two pupils were
united by their delight in theological disputation. Eulogius' devotion to
scholarship was undeviating. When he was imprisoned by Reccafred
bishop of Seville, 'he paid more attention to his prayers and reading than
to his chains' . 19 Here he wrote the Martyr Document to inspire his fellow
captives Flora and Maria on their path to martyrdom, at the same time
making himself master of Latin metre 'which, until this moment, was
unknown by the Hispanic scholars and which he explained to us after his
release' r'" Eulogius took his revenge on Reccafred by refusing to say mass,
appealing to patristic authority. He suggested to an unnamed deacon that
he read before the bishop and assembled ecclesiastics a letter of Epiphanius
of Salamis (367-404), a jewish farmer who converted to christianity and
became a scourge of heretics." Epiphanius had allied himself with Jerome
against John ofJerusalem in a dispute over the doctrines of Origen and in
his letter he defended the right of two priests ofJerusalem to refuse to say
mass in protest over the stance taken by their bishop.
'Eulogius understood this with great happiness and recognized the
opportunity that God had given him. He sighed deeply from the
depths of his heart as if he had received a great wound. Looking first
of all at me , he turned to the bishop and said : "If the lamps and
pillars of the Church acted in this way, what should we do who are
afflicted by the weight of sin? Therefore understand, my lord, that it
is not lawful for me to say mass" . And remaining of the same mind
he kept to this decision during the time of Reccafred' r"
At this point in his narrative, Alvarus introduced Eulogius ' journey to
Navarre, although Eulogius must have made the journey before

56
 THE MARTYRS OF EULOGIUS

Reccafred 's ac tions against him , since he wro te the letter of thank s to his
hosts in the north from pri son - unless he suffered more than one episode
of imprisonment. The journey to Navarre is on e of the most pu zzling
episodes in Eulogiu s' life. It has been mu ch discussed, withou t scho lars
being able to reach an y agreeme nt abo ut Eulogiu s' real purpose. The
consensus view is that it shou ld be dated 848- 9. By Novem ber 85 1
Eul ogius was back in Cordoba and writing about it in a letter of thanks to
Wiliesindu s of Pamplona, one of his hosts. One aspect of the visit to
N avarre which may be em phasised is the discrepancy between Alvarus'
and Eulogius' versions of it. According to the Lift, Eulogius' purpose when
he went north was spiritua l: 'He was not satisfied with visits to the
mon asteries of his own land. Thanks to his brothers who at this time were
und ert akin g a journey to the land of the Frank s, he set off and, arriving
at Pamplon a, stoppe d in th e mon astery of San Zachar ias' F' Eulogius' lett er
to Wiliesind us has a different and muc h more circumsta ntial acco un t; we
must remem ber that the letter was written for public consumption and tha t
Alvaru s cer ta inly read it. 2{ Eulogius said that he set off in searc h of two of
his brothers, who m 'the cruel fortun e of the century had up rooted from
the ir native soil . .. to lead them almost to th e fronti ers of Gaul, in exile
with king Lou is of Bavaria'r'" H eading for Catalonia, Eulogius was forced
to turn back becau se of the warfare in the regio n between count William
of Bar celon a and Cha rles the Bald. H e tried to cross the Pyrenees at
Ron cesvalles, but aga in he failed to reach Fran cia, this time because of
hostilities involving a rebel called Sancho Sa nchez . So he resigned himself
to a visit to Pamplon a.
Yet the impression of an un fortunate man wanderin g in the foothills of
the Pyren ees which Eulogius gave mu st be false. It is possible that he even
had a map." Eulogius may have owned th e surviving copy of the Itinerarium
provinciarum Anto[ni]ni Augusti, based on a third-century origin al, which
ca rr ies the po stscript 'Eulogi M ementote Pecatori'. It was added to a codex
which Ambrosio de Morales saw in Ovi edo in 1572, and could have been
taken the re with Eulogius' writings and relics." In his letter to Wiliesindu s,
Eulogius nam ed the abbo ts who had been hospitable to him in Navarre;
the ir mon asteries have been located , some with more confide nce than
others, by assuming tha t Eul ogius' visited the mon asteries in the order in
which he listed th em , and by a logical route. The monasteries clustered in a
sma ll area of the foothills of the Pyrenees betwee n the Pic d' Orty and Pic
du M idi d' Ossa u." Eulogius said tha t he went directly from Pamplona to
Leire, where Fortu n was the abbot. Eulogius is the sole source for the
mon aster y of San Salvador de Leire in the ninth century. T he first
surviv ing cha rte rs date from ea rly in the following century, when it was the

57
 THE MARTYRS OF EULOGIUS

principal monastery of the diocese of Pamplona. Pamplona was linked to

Leire by a major route which followed the lower course of the Irati into the
valley of the Aragon. He next visited abbot Scemenus at Igal, which might,
therefore, have been situated in the valley of the next tributary of the
Aragon, the Salazar. By the same argument, Urdaspal, where Dadilanis
was abbot, might have been in the next valley, the Ronca!' Madoz sited it
near the present village of Burgui, on the grounds that, according to the
cartulary of Leire, this village was attached to Urdaspa!. By the eleventh
century, some or all of the monasteries in the valleys of the Roncal and the
Urdaspal were dependent on Leire ." In the next valley of the Anso lay
the monastery of San Martin de Cillas. After 858 it may have been
attached to San Juan de la Pefia, although the charter which gives this
information is suspect. 30 Sanjuan de la Pefia itself, on the main route to the
north from Zaragoza via the Col de Somport, is perhaps an odd omission
from Eulogius' itinerary, although it did not become famous until the early
eleventh century. The abbot of San Juan de la Pefia in the mid ninth
century was Athilius, which is the name Eulogius gave to the abbot of San
Martin de Cillas, and it is possible that Eulogius visited both of these
monasteries. Serasa may be the monastery of San Zacharias which
Eulogius mentioned in his letter as being ruled by abbot Odarius: 'a man of
supreme sanctity and eminent science'. It was obviously an important
establishment, which Eulogius said had more than a hundred monks;
Alvarus said one hundred and fifty. 3 \ Eulogius' description of its situation is
very imprecise, but if he visited the other monasteries in the order listed in
the letter to Wiliesindus, it makes sense to have San Zacharias in the next
valley towards the east, that of the river Sabardan, where lies Hecho, site of
the monastery of San Pedro de Silesa. This was certainly an important
monastery from the tenth century, and may be Eulogius' Serasa. From here
Eulogius returned to Zaragoza, and his route back to Cordoba took him
via Siguenza, Complutum and Toledo. The itinerary gives the impression
of an intensive tour of a small number of neighbouring monasteries, rather
than an improvised alternative to a foiled visit to Francia.
Alvarus did not record the problems which had beset Eulogius on his
journey north, nor did he give details of his itinerary in Navarre, referring
the interested reader to Eulogius' letter to Wiliesindus . The significance of
the journey for Alvarus lay in the number of books Eulogius saw 'hidden
away and almost unknown to many'. Eulogius brought back copies of
several works, including Augustine's Ciry if God, Virgil, Horace and
Aldhelm. The Cordoban church's need of these famous texts has been
taken as an indication of its cultural decline . However, it is unlikely that all
these important works were unknown in the capita!. A compilation made

58
 THE MARTYRS OF EULOGlUS

by Alvarusf contains amongst other texts the works of Smaragdus of St-

Mihiel and three homilies of Bede. The prose and poetry of Eulogius and
his contemporaries reveal a sound knowledge of Carolingian models , in
advance of scholars in the christian north." As I shall argue below,
Eulogius' acquisition of a Lift if Muhammad may have been of more
significance for his circle of Cordo ban christians than any number of works
of the Fathers of the Church. Alvarus implied that Eulogius got what he
went for. His journey was commended as being of great benefit both to
Eulogius and to the Church in general.
The biographer was, however, skating over a setback experienced by
Eulogius which may be connected with the journey to Navarre. Alvarus
said that Eulogius had been elected unanimously to succeed Wistremirus as
bishop of Toledo. Epalza argued that Eulogius went north to be
conse crated as bishop in Pamplona in order to be eligible to serve in
Toledo becaus e there were not enough bishops in Cordoba who were
willing to perform the office." But, said Alvarus , 'Divine Providence, which
had reserved martyrdom for him, put obstacles in his way'.35 If Toledo
could not have Eulogius, Alvarus added, the post would be left open. This
may indeed have happened, for there is no record of an alternative
successor to Wistremirus. There is no indication of the reason why
Eulogius was chosen, nor why he failed to take up the appointment but
th ere may be some connection betwe en Eulogius' relations with Toledo
and the discrepancy of purpose which Eulogius and Alvarus report for the
journey to Navarre. Fontaine developed the episode of the two brothers in
exile, suggesting that they were conspiring against Cordoba, and that
Eulogius went north on their behalf." There is not enough evidenc e to
substantiate this hypothesis. Alvarus , making light of the disappointment ,
said that 'although intrigue denied him the episcopal rank, it could not
deny him the honour of the order', which he gained through martyrdom.
As Alvarus added pointedly: 'All saints are bishops, but not all bishops are
saints'.
It is striking that Alvarus described his friend's life as a series of such
false starts and disappointments. Eulogius had not been a model student. "
The two friends' youthful efforts at exegesis of the Scriptures were
destroyed 'lest they pass to post erity' perhaps because they were
unorthodox; the word used is 'inadibilia' [unapproachable]. Eulogius
appeared unsettled. Attempting to atone for the sins of his youth , he
planned a pilgrimage to Rome, but was restrained from departing by his
fellows. He 'frequently scurried off to the sacred flocks of the monasteries,
but lest he be thought to disdain his own office, he always went back to his
clerical duti es, in which he would persist for some time .. . only to return to

59
 THE MARTYRS OF EULOGIUS

his monastery'.38 This echo of Gregory the Great may be the only example
ofa tapas in Alvarus' account of Eulogius' life before his passion. Otherwise,
the vacillations seem a genuine reflection of a troubled man. In spite of
'such great virtues, he went sadly and anxiously along life's way' .
Eulogius seems at first to have resisted being carried away by the
enthusiasm for martyrdom. Although he encouraged Flora and Maria to
remain steadfast in the face of muslim hostility, he made it clear that he had
no wish to follow them. On the contrary, he implored them to work a
miracle for him and obtain his release from prison. Nevertheless, Alvarus
cleared Eulogius of any accusation of weakness: 'When he came to know all
those who were advancing towards their deaths, encouraging all, venerating
and interring their bodies, he was so inflamed by the fire of martyrdom that
he seemed to be the inspirer of the martyrdoms in those days '.39 Perhaps
Alvarus protested too much. Apart from his writings, Eulogius' only
contribution to the early phase of martyr activity was his disobedience
towards Reccafred. Moreover, this action was censured by Eulogius' own
bishop, presumably the man with whom he had been imprisoned, who
ought to have been sympathetic to Eulogius' stand. The bishop threatened
the rebellious priest with excommunication if he continued in his refusal to
say mass.t" Eulogius may indeed have been excommunicated, at least
temporarily, since Alvarus reported that 'although they had ordained him
priest, he dispensed with this marvellous usage, in spite of which he did not
wish to resume the state which he had left'. It seems an unpromising start
for a saint.
Alvarus devoted the second half of the Life to an account of Eulogius '
passion. Eulogius had given religious instruction to a muslim convert to
Christianity named Leocritia. At first, Leocritia was a secret christian, but
she felt compelled to state her beliefs openly, provoking the wrath of her
parents. Leocritia escaped from their custody and went into hiding with
some christians. Her parents appealed to an unnamed judge, who ordered
her arrest. When the soldiers discovered Leocritia, Eulogius was with her.
Both were brought before the judge, condemned to death, and executed.
In a long closing passage, Alvarus asked his beloved friend to intercede for
him in heaven. Yet even in the account of his passion there are
prevarications which are at odds with the way Eulogius himself wrote up
the passions of the other martyrs. Eulogius' crown of martyrdom was
unintended, 'a judgement not expected, but at the same time caused by his
free choic e'." He had been in hiding from the muslim authorities:
'Learning of the deplorable plan [to persecute the christians] we fled,
we departed, we wandered, we hid and having changed our clothes

60
 THE MARTYRS OF EULOGIUS

we made our way in timid flight through the nocturnal silence. We

were frightened by falling leaves, we frequently changed our place of
residence, we searched for safer places, and we constantly trembled,
fearing death by the sword . . . perhaps we fled martyrdom not
because we feared death, which comes when it will, but because we
were unworthy for martyrdom, which is given to some but not to all.
Those who have been and are being martyred were predestined from
the very beginning.'?
Eulogius did not seem to believe that he had transgressed against Islam in
teaching Leocritia. He tried to turn away the 'odious fury, savage aspect
and impatience' of the judge with a soft answer: 'My lord, we are obliged
to preach and it is part of our faith to extend the light of the faith to those
who ask it, and not deny it to anyone who hastens along the holy paths of
life'.43 It was only when the judge ordered Eulogius' death that the saint
spoke out boldly against Islam for the first time . The scene reads like a
recreation of similar episodes in the Acts of the Roman martyrs. From this
point on, Alvarus seems to be coasting through some of the hagiographical
formulae. The elaboration of this section of the Lift may owe more to
literary exigency - the need to give the same weight to the passion as to the
life - than to historical accuracy. Eulogius was whipped, and offered
blandishments. One of the emir's advisers, moved by the fame of Eulogius'
scholarship, promised to obtain his release if he would modify his ardour
just a little; the judge would allow him to continue to practise his faith
unobtrusively, but Eulogius, all hesitation now gone (one might almost say
miraculously) ordered the judge to sharpen his sword. After the execution,
a dove hovered above Eulogius' corpse, which the muslims' best efforts
could not destroy, and his body was saved for burial at San Zoilus .
In spite of this ending, the Lift if Eulogius leaves the reader with the
impression of a saint whose route to martyrdom was far from conventional.
Eulogius' many volumes of writings on the martyrs of Cordoba should be
read in the light of the reservations evoked by his Lift. Whatever the
circumstances of the martyrs' protest, it seems that either the way in which
Eulogius wrote about them or the man himself were so unorthodox that his
works could not be read by his own and subsequent generations. It would
be a mammoth task to re-examine every detail of Eulogius' works from this
angle , and one would still be working largely in the vacuum created by
Eulogius' monopoly of the evidence. There are two aspects of Eulogius'
writing which can be interpreted in the light of material from other
sources. They both relate to the journey to Navarre. The first is the Lift if
Muhammad which Eulogius found at Leire." The second is the passion of

61
 THE MARTYRS OF EULOGIUS

Nunilo and Alodia, the only saints to be included in the Memorial if the
Saints" who were not martyred in Cordoba. Nunilo and Alodia died at
Huesca, near Zaragoza, and their cult was celebrated at Leire. These
documents can be read both as an illustration of the way Eulogius used his
material and as a test of his veracity.

The Life of MulJam:mad

The Lift if Muhammad which Eulogius copied into the Apologetic Book if the
Martyrs is a brief, scurrilous biography of the prophet which begins with the
death of his parents and his upbringing by 'a certain widow' whom he later
married. The devil sent to Muhammad a vulture claiming to be the angel
Gabriel, who ordered Muhammad to become a prophet. After defeating
the armies of Byzantium, Muhammad and his supporters established
themselves in Damascus, where he ruled for ten years and married the
divorced wife of one of his followers, having changed the law to allow him
to do so. Muhammad predicted that after his death he would rise from the
dead after three days, but his body decomposed and was eaten by dogs. For
all the crudeness of its polemic, the Lift if Muhammad reveals an extensive
knowledge oflslam and contains many echoes of the Qur'an." The text is
either Hispanic in origin, or was given a local flavour by its introduction:
'In that time Bishop Isidore of Seville excelled in the catholic doctrine, and
Sisebut held the throne in Toledo. The church of the blessed Euphrasius
was built over his tomb in the town of Ildai [Andujar, near Cordoba].
Furthermore in Toledo the church of the blessed Leocadia was enlarged
with a high roof of wonderful workmanship by order of the aforementioned
king'."?
Eulogius gave the impression that he had rescued the text from
obscurity. In his introduction he noted: 'when I found myself in the past in
the town of Pamplona and detained at the monastery of Leire, leafing
through all the manuscripts there, incompletely known, which were worthy
of reading, I suddenly discovered part of a certain work containing this
anonymous little history concerning the impious prophet'." Perhaps
Eulogius had never read anything similar in Cordoba. Yet this was not the
only extant Lift if Muhammad. A version very similar to that copied by
Eulogius was preserved in four manuscripts from northern Hispania.r?
The earliest of these are the Albelda codex'? of 975 and the Codex
Emilianense of 994 5 1 which is almost identical. In the Codex of Roda,
a collection of texts of Navarrese and Oviedan origin dating from
the eleventh century," the Lift if Muhammad was transcribed between the

62
 THE MARTYRS OF EULOGIUS

Asturi an Prophetic Chronicle of 88 3 and other texts dating from before the
tenth century. The fourth manuscript is a late twelfth-century copy of
the Roda codex, probabl y from C astile. Co mparison with the first two
surv iving manuscripts shows that the Rod a text represents a different line
of transmission , even though the second half of the codex was copied at
San Millan. The 'correc tions' which Morales made in his edition of
Eul ogius' text make it impossible to draw conclusions about the date of the
manuscript from the orthography of the published version but the close
links between all the other manuscripts make a Nav arrese origin for
Eul ogius' acq uisition credible. The text was almost certainly well known , at
least in the christian north.
The Rod a codex also has the only surviving copy of a text with the
cur ious title Tultusceptru de libro domni Metobii.53 It is the story of a monk who
corrupted a naive boy with an ad ultera ted version of christian dogm a. This
legend was well known in the east in several variants. The monk was called
Sergi us, Bahira or Nes torius and th e boy was some times identified as
Muhammad, It may have been ada pte d for a Hi spani c audience by giving
the bearer of the corrupt message the nam e O sius, a reminder of the
bishop O sius of Co rdoba who was still notori ou s in Hi spania for his
apostasy at Sir mium in 357. The Tultusceptru was written on the blank verso
of the folio concluding the Chronicle if Alfonso III, in a Visigothic script
which appears to come from the Rioja, perhap s betweeen 1030 and
1060 .54 The text is full of incon gru encies and lacunae, and seems to have
been copied from a hand which the scribe found difficult to read, altho ugh
th e rep roduction of Arabic phrases is accurate, showing that he knew both
the language and religion of Islam , and was aware that the mu slims
rej ected Christianity as being a cor ruption of the message of J esus.
Di az believed that both the Life if Muhammad and the Tultusceptru
originat ed in al-Andalus and wer e taken to the Asturias in the ninth
century'" Ther e is on e possible piece of corroborating evidence for this
hyp othesis. A very bri ef Lif t if Muhammad appe ars as the penultimate
paragraph of a lett er which J ohn, bishop of Seville sent to Alvaru s. It is
conserve d in a manuscript in the Cathe dral Arc hive of Co rdoba with
Alvarus' other letters and his poetry," The two Lives ar e similar, bu tJ ohn of
Seville's version includes two details which are missing from Eulogius';
Muhammad was show n occupying the mind of a camel, and after his death
his suppo rters claim ed miracles for him. The manuscript of Alvaru s' lett ers
is wr itten in a single hand, that of 'Sisuertus pr esbiter ', perhap s ea rly in the
eleven th cen tury.57It has not survived intact and the letter ofJ ohn of Seville
is incomplete and seems to have bee n copied in the wrong order.J ohn said
of his Lif t if Muhammad: 'We send you this adnotatio on the heretic M ammet

63
 THE MARTYRS OF EULOGIUS

at the end of this letter'. This sentence is immediately followed by a

grammatical discussion . Then comes the life of Muhammad, which appears
inadvertently to have been sandwiched between the grammatical excursus
and a request to Alvarus to send him a treatise on Latin metre composed
or edited by Eulogius.t" After the text of the Life, several lines were
subsequently erased, perhaps because a later reader recognised that the
manuscript was now in a complete muddle. Most of the letter is a
theological discussion with no direct anti-muslim polemical content.
Madoz dated all the letters in the manuscript between 849 -51, on the
grounds that after the beginning of the martyr movement there was no
time for theological discussion, but there is no reason to accept this date.
The Life appears to have been despatched in response to a previous
requcstr'" the phrase: 'Direximus vobis illam adnotationem/" implies that
the writer is referring to a subject which the correspondents have already
discussed. John and Alvarus may have been engaged in the project
recommended by Eulogius in the Apologetic Book qf the Martyrs, that the
learned should collect polemical material against Islam."! It is impossible to
say whether John had read Eulogius' version as well as his own. He may
have written to Alvarus to find out how the two compared.
It is not clear how the Life qf Muhammad fits in with other anti-muslim
polemic. In the east the tradition was already nearly a century old .62 The
monastery of Mar Sabas outside Jerusalem, home of the Cordoban martyr
George, was associated with one of the earliest polemicists against Islam,
John ofDamascus (c.652-c.750).63In the final section ofhis work OnHeresies,
John of Damascus treated Islam as a deviant christian sect, rather than as a
new religion. Muhammad, 'having conversed supposedly with an Arian
monk, devised his own heresy'. The term Christianorum conventiculus for Islam
in the Life qf Muhammad shows the same argument being used in Hispania.
Eulogius and Alvarus were pupils of Speraindeo, said to be the author of a
work of polemic against Islam which does not survive." The existence of
several versions of the Life of Muhammad, together with the Tultusceptru
suggests that such anti-muslim polemic was being read throughout the
peninsula. Perfectus, the first martyr, raised the subject of Muhammad's
adultery with Zaynab - although we have only Eulogius ' testimony to this.65
The prior existence of a polemical tradition in Hispania denies Eulogius the
'suddenness' of his discovery of the Life qf Muhammad at Leire.
In the Apologetic Book of the Martyrs, Eulogius set out to warn his fellow-
christians against the seductions of Islam:
[With regard to] 'the catholic, whoever he is, who wishes to instruct
himself in the folly of the error of Mahomet, of the delirium of his

64
 THE MARTYRS OF EULOGIUS

pr eaching, of the pr ecepts of his new kind of impi ety, it is clear that
during his enquiry he sho uld turn away from the followers of this
sect; for believing them selves in possession of a sacred tru st they are
not content to preach th e dogma of th eir prophet amo ngst
them selves, but proclaim it to whoeve r wishes to hear. For they find
many thin gs in the works of certain of our writers, who armed with
the zealous ardo ur of God , are inflam ed , with their prudent pen ,
against this shameless cha rla tan. We our selves, in our work entitled
Memorial cf the Saints will stigmatise, refute and criticise in part the
erro rs of this sect'. 66
Eulogius bor rowed his terms of abuse - 'he resia rch', 'AntiC hrist' and 'false
prophet' - from the anti-Aria n controversy, and presented Islam as a
christian hcresy.'" Yet there are rela tively few passages in the Memorial of the
Saints which tr eat directly with the confronta tion with Islam . In a work
based on the apocalyptic writings of the Book ofDaniel, Alvaru s was mu ch
mor e strident in his anti-m uslim polemi c. It is difficult to believe that these
words were written for immediate reading, unless Alvaru s too was seeking
death:
'T hey are swollen in pride, haugh ty with swollen hearts, lan guid
in th e enjoyme nt of ca rnal ac ts, gour ma nds in ea ting, usurpers in
the seizing of things and gree dy in th e pillage of the poor,
gras ping with out any feelings, liars witho ut sha me, false witho ut
discriminat ion , impuden t with no mod esty of mind , crue l without
mer cy, usurping witho ut j ustice, witho ut hon our or truth, knowing
neither beni gn ant affection nor th e feelin g for godliness, followin g
modes and fad s, foppi sh, sly, cra fty, and bes mirche d with the dr egs
of all evils, not moder ately so but mainl y so, deriding humility as
madness, spurn ing chas tity as some thing dirty, detractin g virginity
as ru st or mildew, trad ing upon the virtues of th e soul and the
vice of th e bod y, sho wing th eir own mor als in the ir dress and
ac tions' i'"

The polemi cal line followed by Alvarus and Eulogius had too mu ch in
commo n with the eastern christian approach to Islam to be a completely
ind ep endent development. Alth ough ther e are no surv iving Hi spani c
copies of eastern polemi c from the eight h cent ury, eastern polemical
writings in the form of dialogues between individual mu slims and
christians, such as thos e between the patri ar ch Timothy I of Baghdad
(775- 85) and the caliph al-Mahdi survived in later manuscripts from
al-Anda lus.F' C hristians in al-Anda lus may also have read martyr

65
 THE MARTYRS OF EULOGlUS

hagiography from the east which was relevant to the situation in which
they found themselves. The Passion of Anthony of Damascus describes the
martyr's death at the hands of Harun al-Rashid (786-809) at Raqqa on the
Euphrates." Like several of the Cordoban martyrs, Anthony had been
denounced before a judge by his parents after he apostatised to
Christianity. The ninth-century Passion of Rabi ibn Qays ibn Yazid
al-GhassInI tells of a young christian who set out on a pilgrimage to
Jerusalem, but fell in with a band of muslims and renounced his faith .
Thirteen years later he repented, returned to his faith and became a monk
at the monastery of Mar Sabas and then at Mount Sinai , where the earliest
copy of his Passion was preserved. Rabf ibn Qays made a public profession
of his apostasy from Islam, for which he eventually paid the ultimate
penalty. 71 Another such text is the Passion of Peter of Capitolias in
Transjordan." Peter was a married priest who, out of concern for his own
salvation, divested himself of some of the world's temptations by sending
his wife and children away to monasteries. Peter became obsessed with the
idea of martyrdom, and courted death by blaspheming against Islam. He
appeared before the district commander, who ordered that Peter's sanity be
established and both the judge and the caliph, named as al-WalId
(705-15), tried to persuade him not to persist in his error, but without
success. There are obvious parallels between Eulogius' accounts of the
passions of the martyrs of Cordoba and the way in which Peter insisted on
martyrdom against the instincts of the muslim authorities, although these
parallels may be derived from common Late Antique models rather than
demonstrating that Eulogius knew this text .
It is unlikely that the Cordoban martyr movement was a reaction to
contemporary christian-muslim tensions in the Abbasid caliphate, although
there are many points of comparison between east and west. It was a time
when muslims were beginning to resent their dependence on christians
active in administration and the professions. They fear ed that the
distinctions between muslims and christians were becoming blurred,
reducing the status of muslims." The caliph al-Mutawwakil (847-61), who
imposed strict orthodoxy on his muslim subjects, reintroduced the rules
governing the behaviour of christians which were attributed to 'Umar I
(717- 20).74 Christians were to dress in distinctive yellow clothing and
wooden images were to be hung outside their houses to differentiate them
from those of muslims. All churches of recent construction were to be
demolished. The christian response to these restrictions was the propagation
ofpolemic against Islam. It was read not by muslims, but by fellow christians ,
and was intended to put some distance between the two religions. Perhaps
the same concerns affected christian-muslim relationships in al-Andalus .

66
 THE MARTYRS OF EULOGIUS

Eulogius' diatribe against the persecutions of the emir Muhammad

(852-86), also recalls the legislation of 'V mar:
'At that moment, as the cruel conspiracy of the prince against God's
flock redoubled in violence and overwhelmed the christians every-
where, and thus not all, in a general ruin, had thrown themselves
into their religion, as was believed, he gave the order to demolish the
recently-constructed churches and to throw down all the new
ornaments which shone in the ancient basilicas, and which a
maldroit art had added to them in the times of the Arabs . The satrap
of darkness having seized this occasion, even the pinnacles of the
temples which had been erected in a time of peace by the enthusiasm
and industry of our fathers were pulled down , whose foundation
went back almost three hundred years'."
Although historians have assumed that Muhammad ordered the destruction
of the monastery of Tabanos, the home of the several of the martyrs, in
853/6 Eulogius did not in fact identify any churches which were
destroyed." There is little independent evidence for such destruction,
although the discrepancies between the monasteries and churches which
were noted in the Calendar if Cordoba and those mentioned by Eulogius
suggests that some may have disappeared in the ninth century. Neither
Eulogius nor Alvarus referred to the sufferings of christians in the east.
Furthermore, the dates do not fit. Al-Mutawwakil promulgated his first
decree against the christians in 850, but persecution in Baghdad did not
start until two years later, by which time more than twenty of the
Cordoban martyrs had already died. Polemic and resistance to Islam in
east and west developed as parallel responses to similar situations.
It is clear that Eulogius gained a good grounding in anti-muslim
polemic whilst still a young man, before his journey to Navarre. He was not
dependent on his discovery of the Lift if Muhammad at Leire. Eulogius'
reference to this text, if not deliberately misleading, has been misinter-
preted by modern scholars. The text was known but it may not have been
widely read. Why did Eulogius decide to make use of the Lift? Tensions
between christians and muslims may have been rising before Eulogius
departed on his journey to the north. There have been many differing
opinions on the degree of Eulogius' responsibility for encouraging the
martyrs. One fact is not in dispute. His chief response to the crisis was to
write. Perhaps, on finding the Lift in the library in Leire, and in discussion
with his colleagues in Navarre, Eulogius was struck by the text's potential
value in a new context. Polemic against Islam helped Eulogius to refute the
criticism of the martyrs that was coming from more orthodox circles. What

67
 THE MARTYRS OF EULOGIUS

ca me 'suddenly' to Eulogius was the idea of using the Life, not aga inst
Muhammad , which was obvious, but in defen ce of his version of the
christian life. For, said Eulogius, 'those who assert that these soldiers
[martyrs] of our own day were killed by men who worship God and have a
law, are distinguished by no prudence .. . becau se if such a cult or law is
said to be valid, ind eed the strengt h of our christian religion mu st
necessarily be impaired' .78

Nunilo and Alodia

Eulogius may have found a second source of inspiration in Leire. A brief

acco unt of the passion of the sisters N unilo and Alodia, whose cult was
celebrated at Leire, is included in the second book of the Memorial if the
Samtsl? In this volume, Eul ogius wro te biographies of varyi ng length s of
twenty-nin e martyr s which, acco rding to his pr eface, compleme nte d wha t
he had writt en in the first book and were laid out in cha pters for the
purpose of rnedi tation. i" The pr eservation of the mem ory of N un ilo an d
Alodi a was qui te differ en t from that of the other martyrs whom Eulogius
descr ibed. They were comme mo ra ted in a Passionary from the monastery of
Sa n Ped ro de Cardefia, near Burgos." All the surviving Latin liturgical
calenda rs from the penin sula includ ed them except for the Calendar if
Cordoba/" Ther e is also an acco unt of the tran slation of their relics to
Leire,8:l and the saints are menti on ed in donation s to that mon astery and
to other foundation s in the north." The connec tion between Eulogius
and the cult at Leir e is, how ever, no t as obvious as on e might suppose .
Firstly, neith er Eulogius nor Alvaru s said tha t Eulogius learned abo ut these
two saints wh en he wa s in Leire. Secondly, problems in datin g the death
and tran slation of the two martyrs make their relationship with Eulogius'
journey to Navarre impos sible to pin down . These problems have been
considered before, but it is worth attempting ano the r solution becau se if
Eulogius learned abo ut N unilo and Alodia from the monks at Leire, the
way th at he wrote abo ut them in the Memorial ifthe Saints ra ises interesting
q uestions which suggest a rein terpretation of his writings on the martyrs as
a who le.
Ther e are man y similarities between Eulogius' acco unt of the saints and
th e Passion from Cardefia, but Eulogius's version is mu ch shorter, conta ins
less circumstantial detail and differs in significant respects. According to
Eul ogius, N unilo and Alodia were the dau ghters of a mu slim father and a
christian mother. When their father died, their mother married a second
mu slim who did not allow them to pr actise the christian faith to which they

68
 THE MARTYRS OF EULOGlUS

had been devoted since infan cy. Their stepfathe r had them brought befor e
the pr efect [praeficto urbis] , who tried to get them to apostatise by tempting
them with riches and threatening torture and death by the sword . The
sisters made a speech of defiance, a convention of the passions writt en
to comme mora te the Roman martyrs. The prefect handed them over to
muslim wom en relatives who tried in vain to get them to accept Islam .
Nin e days later, still obdurate in their beliefs, they were decapitated in the
public square. Their bodies were taken under guard to a remote plac e and
buried so deep that the christians would not be able to find them and profit
from their relics. But , by signs and miracles which Eulogius did not record,
their worth be came known both to the christians and to the infidels.
In th e Cardefia version, the two girls were brought up as christi an s after
their mu slim fath er abandoned them , and it was only after their moth er
died that their muslim relative s tried to get them to recant their
Christianity. This text is mu ch more explanatory than Eulogius ' version.
The sect of their father was describ ed not as a faith but as a grea t
deception, and this was elaborated in a paragraph of polemic, describing
how the Saracen s had been polyth eists until Muhammad was deluded by
Lucifer in th e guise of the Angel Gabriel into believing that he was the
great est of all the prophets. The Passion continued; a memb er of the sisters'
family denounced them before Ghalaf, the proconsul 'whi ch in the
Chaldean tongue is called emir'. H ere the girls made their speec h of
defian ce, but the proconsul answ ered them gentl y, urging them to convert.
Wh en they refused , he took pity on their youth and returned them to th eir
family. They were next denounced before 'king' Zumahel, whom the 'king
of Hispania' had sent to Huesca as his deputy. The king sent them away,
this tim e not to their home but to another muslim household, where they
resisted blandishments for several days . For a third time they were
denounced, this time before a form er pri est who had apostatised to Islam .
He too did not want to condemn them to destruction and told them to go
to live with the christians in the mountains. This lenient treatment of the
sisters recalls al-Walld's relu ctance to make a martyr of Peter of Capitolias.
Fin ally, Nunilo was executed, but even at this late stage her youn ger sister
was urged to change her mind, but she refused, and was condem ned to
death. The sisters' bodies were exp osed to be ea ten by scavengers. Their
legs wer e tied together and they wer e thrown into a ditch, where they
rem ain ed under the protection of two vultures. At last, a christian was
allowed to wrap them in linen and bu ry them . The mo st significant
differ en ces between this version and the Memorial ifthe Saints are in this final
section. It was th e C ardefia account which gain ed cur rency in the north; a
sho rter version found its way into the Acts of the cathedral of Huesca .P

69
 THE MARTYRS OF EULOGIUS

Eul ogiu s said th at th e events occurred 'in urb e Bosca ad oppidum

Barbitanum'. The Carde fia Passion firmly links th e m ar tyrs with Navarrer"
'pass io beati ssimarum virgi num N unilonis atque Alodi a martires Xpi
que passe sunt in civita te osche nse sub preside Zumahel , . . contigit
in regione Spanie Sarracen orum [rex] Abderra haman precepi t , ..
Acc idit autem in territorio berb etano iuxta antiquissimum locum qui
dicitur Castro Vigeti in Villa Ab osch a , esse duas germanas , . ,87

Ther e are severa l possible int erpretati on s for 'Bosca'/'civitate osche nse'
including th e Huesca near Granada, bu t they do not account for the othe r
top on ym s. LOpez, who wrote extens ively on the origins of Nu nilo and
Alodia, adopted th e th esis th at th ey were born in Ad ahuesca ['Abosch a'] ,
shut up in the castle of Alquezar ['Castro Vigeti'] , bo th in the region of
Barbastro ['t erritorio Berbetan o'] and th at th ey were martyred at Huesca
['Osca']. All th ese places lie in a small area northeast of Zaragoza. This
identificat ion has m et with general suppo rt. There are othe r indi cation s to
link th is text with th is part of the peninsula , whe re th e po pulation were
mu ch less Arabicized th an in Cordoba. The ruler of ' Osca' was ign orant of
the lan gu age of the christia ns of the zone and the sain ts had to talk to him
via an int erpreter. This is not a feature of th e passion s in th e Memorial ifthe
Saints, where Eulogius noted of several mar tyrs th at th ey knew Arabic.
T he translation of the bodies of Nun ilo and Alodia to Leir e strengt hens
the link be twee n thi s cult and th e north. Only one version of the Translation
survives. It was edited by Pellicer de Salas in 1668 'fro m an anc ient codex'
of th e Acts of saints from C ardefia. The text m ay have origina ted in Leire,
since ano the r, later, transcrip tion was copied from th e Breviary if Leire.
C ertainly, th e Translation does not see m to have been writte n to
compleme n t th e Cardefia Passion, and indeed , what th e Translation says
abo ut th e disposal of th e bodi es is much closer to Eulo gius' acco unt than
to Carde fia's:

'The tyr ant . . , first threw th eir bo dies to th e wild birds, to deny th em
the hon our of a ch ristian bu rial, and so th at th eir m em ory would be
erased from th e ea rth, he threw them int o a deep pit , wh ich he filled
with a great pile of stones so th at they would lie the re in ob livion for
ever,.88

The atte mp t at obsc urity was unsuccessful, since twenty- nine years later
qu een Oneca, prompt ed by the fame of the saints' names 'throughout th e
length an d breadth of Hispani a', inquired where the ir rem ains had been
b urie d. At the same time, the mo nks of Leire, under th eir abbot Fortu n,
pu t forwa rd th eir cla im to th e bodi es, which was based on the don ation s of

70
 THE MARTYRS OF EULOGlUS

land and goods which the monastery claimed to have received from the
sisters' family. The rest of the account of the translation contains many of
the topoi of the genre. Similarities with such texts as the Translation oflsidore
led Gil to believe that it was composed in the eleventh century or the
beginning of the twelfth. However, it has much in common with ninth-
century texts written in Francia, which used as a model Einhard's
Translation ifMarcellinus andPeter. 89 As so often in these texts, the discoverer
of the relics was a shadowy figure - a certain man called Auriatus ~ who
was roused by a dream to go to Huesca, where, 'with divine help' he found
the bodies 'hidden in a deep pit', from which he moved them to an
unnamed place before he returned to give first notice of the discovery to
abbot Fortun at Leire . The abbot or the queen sent him back to Huesca
with companions and gifts to conduct the difficult business of obtaining the
relics, a difficulty perhaps inserted in obedience to convention, but left
unclear. The relics were placed in danger of misidentification, another
feature of the genre, when one of Auriatus' companions 'of the tribe of the
blessed virgins' expressed doubts about them. Auriatus promised that these
doubts would be overcome and took the doubter to the place where he had
hidden the martyrs' bodies. The men prayed by the light of the stars, then
dug up the relics, which were authenticated by their sweet smell. They took
the relics back to Leire, where they were received with great rejoicing by
the abbot in the company of the king Enneco (liiigo) and bishop
Wiliesindus, together with many churchmen and other faithful. Ifiigo and
Wiliesindus were credited with the translation of Nunilo and Alodia to
Lcire in other sources, as we shall see. Although it features these two
authentic figures, there seems little doubt that the account of the discovery
is a pious forgery written to authenticate the Leire relics.
Examination of a number of documents from Leire, almost all of them
probably falsified, gives some clues to the possible date of composition of
the Translations" The first is a charter recording a royal donation to Leire
and to the saints Nunilo and Alodia by Inigo and bishop Gulgesindo
[Wiliesindus, presumably the recipient of Eulogius' letter], dated 842. The
second describes a donation by Inigo's son Garcia and bishop Eximio
to Leire dated 880, and the third, with the same date as the second, is a
compilation of the first two, giving much more precise details on the extent
of the bequests, the lay and episcopal witnesses to the charter, and the dire
punishment which would descend on anyone who contravened the terms
of the donation. A fourth charter dated 901, which may be genuine, notes
a donation by Garcia's son Forum. It is the charter of 918 recording a
donation of Sanchez, son of Garcia, which may be a clue to the forgeries .
Comparison of these documents with a contemporary collection, equally

71
 THE MARTYRS OF EULOGIUS

suspec t, from Sanjuan de la Pefia shows th at th ere are two matters at issue.
There were , as might be expected, disagr eem ents between th e two
foundati on s conce rning th e owne rship of land. Leire was making a
ret ro spective claim to land origina lly gran ted to San Juan de la Pefia,
Di screp an cies betw een th e royal nam es listed in th e charters and the
genealogy con taine d in th e Cod ex of Rod a?! suggests that even mor e
important matters wer e at issue. D eliber ate obscur antism was used in th e
Leir e documents in describing th e roy al succession in Navarre. The most
likely explanation for this is th e usurpati on of Navarre by San ch o Garces I
(905- 25). San cho seems delib er ately to have confused his lineage in orde r
to link himself with the dynasty of Ifiigo Arista whi ch held N avarre during
th e previou s cen tury. The autho rs of th e Leir e charters also cla ime d for
Sanc ho a connec tion with th e Jimen ez dynasty, who m th e charters from
San Juan de la Pefia identifi ed as th e legitimate rulers of Navarre, by
changing Inigo Arista to Inigo Jimen ez. The attempt at deception was
successful in at least one sense, in th at subseq uent historian s have had great
difficulty in unravelling th e pro sopogr aphical tan gle thus created. F To
confuse m atters even further, one tradition m ain tain ed that the last king of
the line of Ifiigo Arista, called Fortun Garces, becam e a monk at Leire . It is
difficult to follow the tr ail of tho ught of th e tenth-cen tury scribes th rough
th e wr iting or rewriting of th e charters and of the tran slation ofNunilo and
Alod ia , but it appears th at th ey were used to bolster San ch o Garces' cla im
to legitimacy by inventing links between th e king and th e early pat rons of
th e cult at Leire. The mon astery was, in its turn, rewarded by becoming
th e pantheon of th e kings of Pamplon a.l"
Sa ncho Gar ces I furthered his claims to N avarre by dedicating
m on asteri es to Nunilo and Alodia in th e lands which he took from th e
muslims in th e Rioj a. The first such foundati on was a conven t dedicated
to the two m artyrs near N ajera , which San ch o cap tured in 925.94 D evotion
to N unilo and Alodia was well establishe d in the Rioj a, and was ado pted at
Los H ornajos, betw een Castro viejo and Bezar es, a few kilom etres south-
eas t of N ajera, as lat e as th e fifteen th century. The associa tion of N unilo
and Alodi a with Leir e cont in ued . In 1682 th e mon astery was granted th e
right to celeb ra te the dat e of th e tran slation , 18 April, with th e mi nor rite
'o n wh ich day th eir bodi es were tr an slated , in th e tim e of the Sarace ns, to
this m on astery ' i'" Nor was th e assoc iat ion be twee n th e m ar tyrs and Huesca
forgotten. After Ped ro I of Aragon took Huesca in 1096 he gave one of the
city's m osques to Leire , thus emp hasising the link between th e site of the ir
passion an d th eir final resting-place. A ch urch was erected in th e squa re
whe re they were executed. These references to the saints clea rly link them
with northern Hi spania, and with th e politics of Navarre .

72
 THE MARTYRS OF EULOGIUS

When and where did Eulogius hear about the martyrs of Huesca? The
answer depends on dating the martyrdom and the translation of the relics
to Lcire . Commentators on the martyrdom of Nunilo and Alodia have
assumed that it was inspired by events in Cordoba and the year of their
martyrdom is usually quoted as 85l because this seems to be the date given
by Eulogius . In fact he specified the day, the eleventh Kalends of
November (22 October), and indicated the year simply 'as above' ['aera qua
supra']. The date of the death of the first martyr whose passion Eulogius
recounted in the second book of the Memorial if the Saints was given both an
Incarnation and a Hispanic Era date, and the second martyr Isaac 's death
was dated 'aero' 889 (85l). Of the remaining martyrs of Book 2, in only two
instances is the year given in full, the others appearing as 'aera qua supra'.
No date is specified between chapter two, on Isaac and chapter seven, on
Nunilo and Alodia. Although 'aero' here probably means 'year', Eulogius
may have used 'aero' in its Isidorean sense to mean 'an era or epoch from
which time is reckoned'. 96 In those entries where the days of the week are
mentioned, they are consistent with the martyrdoms having occurred in
85l. It would be easy to make too much of this point, since only the day on
which the saint's feast was to be celebrated was important, and the year of
his or her death had no significance . But, when Eulogius used the
expression 'aero qua supra' for many of the other martyrs included in Books
2 and 3 of the Memorial if the Saints, and in the letter to Wiliesindus he may
not have meant that all the deaths took place in the same year.
A date of 85l for the martyrdom of Nunilo and Alodia seems at first
glance to be confirmed by the text of the Passion:" It begins: 'Factum est
igitur in anno incarnationis Domini DCCC quinquagesimo primo'.
Lopez argued that this was not referring to the passion, but was a note by
the scribe: 'This work was undertaken in 85l AD ', since to continue the
sentence 'it occurred in the region of Hispania of the Saracens 'Abd
al-Rahman ordered .. .' ('contigit in regione Spanie Sarracenorum
Abderrahaman precepit') makes little grammatical sense. It is also
possible that the copyist confused Era and Incarnation dating, because it
is not clear which dating was in force in the north in the ninth and tenth
centuries. Attempts to date the martyrdom are further thwarted by
discrepancies between the date and day of the week given in the sources .
All have 2l October except Eulogius, who has 22 October. This may be a
simple move by Eulogius to avoid the feast of St. Ursula on 2l October;
in the Breviary if Leire, it is St. Ursula and her sisters who have been
moved . Eulogius did not give the day of the week, although he had done
so for all the previous martyrs in the second book of the Memorial if the
Saints. The Passionary if Cardeiia says that the passion occurred on a

73
 THE MARTYRS OF EULOGIUS

Thursday. This rules out 851, unless, as Gil does, one 'corrects' quintafiria
to quarta."
The date 851 makes some sense in conjunction with the Translation,
which says that the relics were taken to Leire twenty-nine years after the
passion and begins 'Therefore in the year 880 ', It is not clear, however, if
the scribe meant Era 880 or 880 AD. Further investigation reveals that the
dating in the Translation is completely scrambled. The opening sentence
reads 'Therefore in the year 880, after the Resurrection and Ascension of
our Lord, in the part of Hispania ruled by Abderrahamen . . .' and goes on
to refer to the persecutions of this monarch and to date the deaths of
Nunilo and Alodia in connection with these events . This seems to mean
that it is the martyrdom itself which is being dated 880, and not 880 AD,
but 880 years after the Resurrection, that is, sometime after 911. This date
can be made to fit if the emir was 'Abd al-Rahman III (912-29), but surely
'Abd al-Rahman II (822-52) is meant. The reference to post-Ascension
dating sounds like an acknowledgement that some thirty years have gone
astray, without the author's being clear where. This sentence looks like an
attempt by a later hand to turn a Hispanic Era date to an Incarnation date.
The finger of suspicion points towards Pellicer, who favoured 880 AD,
although he did not admit to having altered the date when he copyied the
manuscript. Pellicer recognised the inconsistencies in the names of the
kings mentioned in documents from Leire. He could not believe that Inigo
Arista had been present at the translation of the relics to Leire , and
invented another Inigo , grandson of the first, which he felt provided a
solution to the problem. Gil believed the confusion was caused by reading
the false charter of 842, which described Inigo Arista and bishop
Wiliesindus as making their donation shortly after the saints' relics arrived
in Leire.?" Yet however spurious the charter, the reasons for its production
do not rule against a tradition that the translation occurred when Inigo was
king and Wiliesindus bishop of Pamplona. These citations are surely the
reason to credit an earlier date. This was a Leire text, and although it was
easy to change the dates, it would have been much less so to introduce
people thirty years too late . The Translation clearly places the events in the
time of abbot Fortun, when Wilesindus was the bishop of Pamplona and
king Inigo and queen Oneca reigned. The chronology of this period is
unreliable, depending on the work of Ibn Hayyan, who gave the date of
Inigo Arista's death as 851 ~ 2. 100 If Fortun and Wiliesindus are the same
men whom Eulogius visited, it is unlikely that they were both still alive in
880. If there is any credence in the clues to dating offered by the Translation,
it is in favour of Era 880 [842], rather than 880 AD as the date of the
translation, with the martyrdom perhaps in 813. This seems to be

74
 THE MARTYRS OF EULOGIUS

confirmed by the Breviary if Leite; which gives XIV Kalends May, Era 880
as the date of the translation, albeit in a thirteenth- or fourteenth-century
manuscript; the Breviary also preserved the tradition that the sisters perished
twenty-nine years earlier. 101
The people featured in the Passion do not really help in dating the events
it describes. The emir of Huesca who sentenced Nunilo and Alodia to
death was named as Zumael, but the ruler of Huesca from c.850 was Musa
ibn Musa of the Banu QasI. Zumael could be Musa's son Isma'tl, who,
after Musa's death in 862, took 'Barbitaniya' as his personal fief, and who,
it is argued, already ruled in this region and acted for his father in dealing
with N unilo and Alodia. The emir of the Passion could equally well be a
predecessor of Musa. Unfortunately, the rulers of Huesca in the first half of
the ninth century are unknown, although it seems from frequent references
to 'Barbitanrya' in the Arabic sources that the area between Huesca and
Lerida was under muslim control during the Umayyad pcriod.P? The
Banu QasI were sometimes loyal to the Umayyads, and it is possible that
'Zumael' was a Cordoban appointee, as the Passion claimed. Ghalaf, the
proconsul before whom the sisters were first denounced, could be Ghalaf
ibn Rashid, to whom the geographer al-'UdrI (d. 1085) from Almeria
attributed an implausibly long reign of 70 years, from 802, over two castles
of Barbitanrya, Barbastro and Alquezar - both mentioned in the Passion of
Nunilo and Alodia. In c.813 Ghalaf tried to extend his rule to Huesca.l'"
This favours an earlier rather than a later date for the martyrdom.
Most modern historians have preferred Eulogius' date of 851 for the
martyrdom simply because neither Eulogius nor Alvarus mentioned
Nunilo and Alodia in connection with the journey to Navarre. Even those
writers who favoured an early date for the martyrdom have been forced to
postulate that the saints perished before Eulogius' visit to Leire, but that the
relics were not translated until after this date; the exact dates selected vary
but the sleight-of-hand is the same. It seems odd that Eulogius did not refer
to them in his letter to Wiliesindus of Pamplona, either with reference to
his journey to Navarre, or - if the deaths did occur in October 851 - in his
list of martyrs brought up to the date of writing the Icttcr (15 Nov 851).
The date of the letter is given credence by Eulogius' remark that he was
entrusting it to Galindo, who went to Cordoba in 844, but may have
returned to Navarre in 851 or 852 at the death of his father Inigo Arista.
Eulogius said that he heard about Nunilo and Alodia from Venerio of
Complutum, whom he visited on his way home from Navarre, although he
used the phrase 'refetente Venerio' which could also mean that Venerio had
sent him a letter about it. Eulogius spent fivc days in Complutum, then
returned to Toledo, where he spent 'many days' with Wistremirus.l'" Thus

75
 THE MARTYRS OF EULOGIUS

it is just possible that in the twenty-four days which ar e supposed

to separate the dates of the martyrdom and the writing of the letter to
Wiliesindus , Eulogius mad e his way back from Navarre, heard about the
new martyrs, but chos e not to menti on them in his letter. On the other
hand, an earlier date for the martyrd om suits the facts just as well. This
doe s not seem a matter that can ever be resolved, but perhaps it does not
really matter where and when Eulogius heard about the sisters. It is clear
that his information came from clerics in the north and it is very likely that
when Eulogius returned to Cordoba the martyrs Nunilo and Alodia were
already dead, perhaps long dead, and he had heard of their passion and
perhap s of their translation to Leire. Like the acquisition of the Life if
Muhammad, or the idea which it gave him of promoting polemic, the Passion
ofNunilo and Alodia looks like som ething that Eulogius brought back from
the north.
In the Memorial if the Saints, the chapter devoted to Nunilo and Alodia is
immediately followed by a long acco unt of the pa ssions of Flora and
Maria. l'" The Martyr Document which Eulogius wrote for them was
modelled on the treatises which bishop s from the time of the Rom an
persecuti on s composed to enco urage potenti al martyrs. Eulogiu s, acco rding
to his Life, 'tenaciously fortified the virgins for martyrdom and tau ght
them , by means of letters and words, to disdain death' l'" (although it seems
rath er unlikely th at two wom en , on e of them brought up as a mu slim,
would have been sufficientl y ed ucated to understand Eulogiu s' convoluted
Latin). All three wer e togeth er in pri son when Eulogius wrot e to
Wiliesindus on 15 November, and the wom en perished nin e da ys later.
The parallels between Flora's passion and that of Nunilo and Alodia in
Eulogius' accounts of them ar e striking. Flora, the daughter of a christian
mother and a muslim fath er, was brought up as a christian after her fath er
died , and denounced to the authorities by her brother when neith er blows
and threats nor sweet words would induce her to adopt Islam. Interrogated
by the judge, Flora declared her int ention of being martyred for her faith .
Enraged , the judge ordered her to be beaten. She was returned half dead
to her br other, who was told to bring her back only if he failed to convert
her. Flora soon returned to prison , wher e she was joined , not by her sister
- who sympa thised with Flora but did not pay the price of martyrd om -
but by Maria, propelled towards martyrdom by the death of her brother
Walab on sus. Both responded willingly to Eulogius' exhor tations to j oin
their 'brothe rs' Perfectus, Isaac, San ctius, Petrus, Walabonsus, Sabinianus,
Wistremundus, Habentius,]erimi as, Sisen andus, Paulu s and Theod omirus
- all the martyrs comme mora ted in the Memorialif the Saints up to the death
of Flora and Maria with the exception of N unilo and Alodia - and were

76
 THE MARTYRS OF EULOGIUS

put to death.'!" Eulogius could have used the example of Nunilo and
Alodia to encourage Flora and Maria, but he did not do so. Yet he wrote
the two passions in a very similar way. Whatever kernel of truth there may
be in the Passions of these female martyrs is difficult to extract from the
parallels which Eulogius imposed upon them.
Eulogius seems to have finished the second book of the Memorial of the
Saints in 853 or 854 . 108 Before recounting the passion ofNunilo and Alodia,
he thought that a note of explanation was necessary, justification for adding
to the list of saints whose passions he had recorded in the first book of the
Memorial qf the Saints and the first six chapters of the second:
'I had thought at this point that 1 was at the end of the second book,
at the end of the saints running to this contest [martyrdom], nor did
1 think that anyone else was about to be taken to prison after so many
crises of our church and such fierce struggles. Truly, because divine
presage, hitherto thus disposed, increased by the profession of such
people the number of such saints, therefore presently, inflaming
many hearts with the manifold ardour of their predecessors, it
propelled them to that contest, and to the same vow and greatness,
inspiring other men, women and children, and armed them to the
exercise of their contest. ... Whose names, ages and deeds of war are
set out in their place and time . First therefore, the victories of the
saints are laboriously written down, by reckoning and progress of
the months, in the order of their acts, brought together just as
the deeds in other regions will have been visible. But because there is
one confession which crowned both [this could refer either the saints
in Cordoba plus the saints in other regions, or to Nunilo and Alodia
alone] and because it was this same time which sent both to heaven
through beastly impiety, 1 think it not absurd, being capable of
writing down the squadrons of the saints that were joined to them in
the assembly of those elected together, so that those of equal
magnificence of praise might be raised by us on earth who are
recorded in one codex in heaven. .. . Therefore, reported by the
saintly man and venerable father Venerio, bishop of Complutum, we
learned . . . ' 109
Eulogius does not say that the movement was confined to Cordoba; it is
not he but modern historians who have obscured Nunilo and Alodia's
distant origins .
Eulogius' passions become increasingly suspect the more one looks at
them. It is little wonder that he had to write such lengthy apologetics for
them. Not only did he include martyrs from other parts of the peninsula,

77
 THE MARTYRS OF EULOGIUS

but also saints who died before the 850s and people who may not have
been martyrs at all. Another passion which Eulogius added to the second
book of the Memorial qf the Saints is that of Aurea, who was denounced to the
muslim authorities as an apostate after having lived quietly for thirty years
in the monastery of Cuteclara. Her muslim family were supposedly
unaware of this defection. When they found out, they came all the way
from Seville to denounce her. This has been taken to represent a sudden
increase in tension between the two communities causing crises in families
of mixed religious affinities.'!" The passage may be open to another
interpretation. According to Eulogius, Aurea was the sister of Joannes
and Adulphus, two martyrs of the 820s whose passion had been
commemorated in a work, now lost, by Eulogius' teacher Speraindeo.'!'
Nevertheless, Eulogius described Aurea, somewhat implausibly, as still
being a young virgin at the time of her death. Eulogius may have wanted to
include Joannes and Adulphus in his martyrology but they were too well
known; their feast is listed in the Calendar qf Cordoba. 112 They could not
contribute to Eulogius' picture of a crisis of the 850s. Aurea, on the other
hand, was not known and it is possible that she is Eulogius' invention. By
linking the bona fide saints with the martyr Aurea he legitimised her
passion and the movement he was trying to defend. Eulogius may also have
included men and women with no claims to sanctity. Although he was
caught up in the martyr movement and met many of the participants, some
passions have scarcely any detail. The chapter devoted to Sanctius, a
christian captive, perhaps a slave, from Albi in Francia, is a single sentence
long. Some of these briefest of passions could be little more than lists of
christians who died in Cordoba in a variety of circumstances about which
Eulogius could not afford to be too clear.
Eulogius was inspired by his journey to Navarre and his reading of the
Life qf Muhammad to write in defence of martyrdom. Some deaths had
already occurred in Cordoba, but in order to make his collection of saints
more inspiring, Eulogius added Nunilo and Alodia celebrated at Leire, and
perhaps invented or reworked material from other sources. Only in the
case ofNunilo and Alodia do enough independent details survive to expose
this process. Eulogius may have been assembling the great and the not so
good to bolster the witness of an unknown number of Cordoban saints
whose sacrifices had not been accepted by the leaders of the church. He
was firing a volley of saints as a response to the accommodation which the
christians of Cordoba had made with their heretical rulers. In addition,
given that there are so many indications of Eulogius' unorthodoxy in the
Life qf Eulogius, he may have been writing partly to justify himself against
the attacks of his ecclesiastical superiors. He may also have been motivated

78
 THE MARTYRS OF EULOGIUS

by a pious horror that these martyrs would otherwise be forgotten. Indeed,

the danger was very real. Popular support for the martyrs of Cordoba
disappeared - if indeed it had ever existed ~ leaving Eulogius as their sole
chronicler. There may have been few christians who were willing to read
him. Eulogius wrote to Alvarus:
'This work [the Memorial if the Saints] was almost finished when an
insane decision on the part of the authorities landed me in prison
. . . I thought that it would end up dispersed all over the place. But
having been preserved at that time by the Lord, now, with his help,
amidst the anxieties of prison life, it has not only been completed
but delivered to you, whom the Lord chose to see it before anyone
else .. .' " 3
Alvarus spent several years under excommunication after unexpectedly
recovering from a serious illness during which he had received the final
sacrament of penance. This gave him an ambiguous status in the eyes of
the Church and his attempts to have this judgement revoked led him into
a bitter dispute with the bishop Saul. 114 He was not in a position to ensure a
wider audience for Eulogius' writings. Further, there are several aspects
of Eulogius' ouput which make it possible that the works were not intended
to be read immediately. It is clear from the way he described the locations
of the monasteries around Cordoba, which a local audience would have
already known, that his work was not written for his contemporaries in
Cordoba. And only an audience of people who did not know recent history
could have accepted Eulogius' additions to the tally of martyrs.
It is not easy to find a context in which the works of Eulogius could have
been read. Modern historians have perhaps proved Eulogius' most
receptive audience. Poring over the Memorial if the Saints, they have been
willing victims of Eulogius' strategems, arguing over the significance of
various gnats, but swallowing the camel whole . It is time for a radical
reappraisal of the martyrs of Cordoba. Eulogius made the martyr
movement seem more important than it was and distorted our picture of
christiarr-muslim relations in the ninth century. If we cannot trust his
account of the martyrs of Cordoba, how much less should we accept them
as being representative of the christians of al-Andalus as a whole .

79
 ---
CHAPTER F1VE

Two more martyrs of Cordoba

'In the western part of the world there glowed an ornament bright,
A city famous in lore, proud of its new might at war
It throve under the reign of colonists from Spain.
Cordoba was its name; wealthy it was, and of fame;
Well known for its pleasures and for its splendid treasures.
Held, too, in great esteem , for the seven-forked stream
Of learning. Also in the fore for its great triumphs at war.
Once this famous town to Christ in faith was bound
And gave its sons to God, cleansed in the baptismal font
But sudd enly a martial force changed the well-established cours e
And laws of holy faith, by spreading through the state
Errors of false dogma, harming th e faithful folk.
For the faithless tribe of unrestrained Saracens
Fell upon the stout people of the town .. .' 1

A ndalusi christians continued to seek martyrdom , although now they had

no Eulogius to commemorate them. All the later martyrs died in
Cordoba. Scholars have dismissed them as no more than a footnote to the
protest of the 850s, assuming that they were a response to the earlier
deaths. Yet there are no obvious links. Aimoin ofSt-Germain, the author of
an account of the translation of the relics of George , Aurelius and Nathalia
[Eulogius called her Sabigotho] to Paris, noted that Mancio, the envoy sent
by Charles the Bald to investigate these martyrs, witnessed the executions
of two sisters." After this, no martyrs were recorded until the tenth century.
When the impuls e to martyrdom was reawakened, it was as a series of
individual gestures. The evidence for these martyrs is sparse and difficult to
interpret, and must be gathered from a variety of sources, both Latin and
Arabic.
Som etime during the reign of 'Abd Allah (888- 9 12), a woman called
Dhabha denied the divinity of Allah , and called Muhammad a false

80
 TWO MORE MARTYRS OF CORDOBA

pr ophet." Dhabha appeared be fore the j udge Ahmad ibn Muhammad ibn
Ziyad al-La khmi (d .924),4 who decreed that she should be burned at the
stake. This bri ef refer ence to Dhabha comes from a legal text comprising a
collection of judgem ents compiled some fifty years later and it is not clear
whet her Dhabha suffered the pr escrib ed punishment, or whethe r this is
just a textbook exa mple. A few years later, ano ther Cordoban judge, Aslam
ibn 'Abd al-Aziz, had to deal with a would-be martyr. Perh ap s this was not
a commo n occurrence. Al-Khushani, in his biography of the judge,
exp lained that 'the nonsen se or igno ra nce of the christians led them to
attribute grea t merit to this action of offering them selves to death' , for
which the re was no justification in Scripture." This christian did not
achieve his goal. The judge had him flogged to pro ve that the body which
the christian had describ ed as mer e semblance would suffer the pain of
exec ution in an all-too-real way; the purpose of the anec do te was to
illustrat e the judge's sagac ity. Apart from these two instances, christian
martyrdo m did not attract th e atte ntion of muslim autho rs.
In his history of The Mozarabs, Cagigas named th ree mart yrs from the
920s, Euge nia, Abu Nasr an d Maria," gathering togeth er meagre
fragm ents of evide nce. In the sixteenth cent ury Ambrosio de M orales
saw a badly-da maged sto ne in the Marmo lejos district of Cordoba, which
seems to have bee n inscribed with the dat e 26 March Era 96 1 [923] and
acrosti c hexam eters rea ding 'Eugenia martyr'. " Abu Nas r was a rebel
agai nst the Umayyads who was cru cified at the gates of Co rdoba," Writing
three centuries after the event, Ibn Idh ari did not say that Abu Nasr was a
christia n, but Simo net thought he was 'probably a chr istian' , no doubt
beca use of the mode of execution, altho ugh the Arabic histories often refer
to dea th by cru cifixion as the penalty for treason, irr espective of the
religiou s affiliation of th e traitor. Cagigas did not cite his sources and I have
been un able to discover M ari a. M orales also seems to be source of the story
of ano ther martyr, a certain Domi ngo Sarracino who was capture d with
othe r christians at the battle of Simancas and taken to C ord ob a, where he
was imprison ed , refused to abj ure his faith, and died in 980. Hi s wife
followed him to Co rdoba, where she died in 982, according to a stone
which M orales saw in the wall of a hou se next to the church of Acisclo and
Victoria." The inscripti on , in Morales' tra nscr ipt ion , does not substantiate
his story and, since the battle of Sima nca s took place c.939, it seems rather
improbable. Morales, Simo net and Cagigas were all stretching the
evidence as far as it wou ld go in an attempt to prove that the christians
of al-Anda lus continued to oppose the idea of peaceful co-existence with
Islam. Since the evidence is exiguo us, it is difficult to discuss these mar tyrs
furt her.

81
 TWO MORE MARTYRS OF CORDOBA

There are, however, two tenth-century Cordoban martyrs who are

much better documented. Three copies of the Passion of Pelagius and a single
copy of the Passion qf Argentea survive in manuscripts from monasteries in a
small area south of Burgos in northern Hispania. The passions may have
been written in response to actual martyrdoms, although it is difficult to
disentangle the circumstances of the martyrs' deaths from the conventions
imposed by the models on which the passions were based. A careful
reading of the passions uncovers something of the perspective from which
they were written and tells us a little about the city where the martyrs met
their deaths. In attempting such a reading, it is difficult to avoid becoming
sucked into the mire of assumptions and generalisations about the
relationship between christians and muslims and between the christians of
north and south which colour all discussions of this period. Not all these
assumptions can be overcome, but some of them will be questioned in the
course of the analysis of the Passion ofPelagius and the Passion of Argentea
which follows. I begin by investigating the date and authorship of the
passions , by looking at the context in which they survived .

The Hispanic Passionary

All but one of the earliest copies of the Passion qf Pelagius were discovered in
the monasteries of Silos and Cardefia, Santo Domingo de Silos was
probably a Visigothic foundation but it is documented only from the
period of its refoundation by Fernan Gonzalez, named in the first extant
charter from Silos dated 954 . 10 A rich collection of early medieval
manuscripts survives from this monastery, but not all of them originated
there. Silos did not achieve prominence until the death of Domingo Manso
in 1041, when the monastery became the focus of his cult, and the
repository for other monastic archives . San Pedro de Cardefia may have
been founded in the ninth century; the first extant charter is dated 899. 11
All but one of the codices containing the Passion qf Pelagius are passionaries
- collections of the Passions of martyrs and a few lives of confessors, usually
arranged in calendar order, which were read aloud during the celebration
of th e saint's feast day. The individual entries in the Silos and Cardefia
passion aries follow a common formula, but although these collections may
have been assembled about the same time and not far from each other,
they are not identical. The differences between the manuscripts help to
date them. Before embarking on an analysis of the manuscripts, however, it
must be pointed out that they have been placed in different chronological
order by recent authorities. Fabrega, who completed an edition of the

82
 TWO MORE MARTYRS OF CORDOBA

Hispanic Passionary based on two manuscripts from Cardefia, ranked the

manus cripts according to the number of saints they contain, arguing that
passions gradually accumulated and that each scrib e had read most or all
of the earlier collections. In contrast, the dates which Diaz assigned to the
manuscripts dep end to a large extent on his expert opinion of the script .!"
The following brief introduction to the different versions of the passion aries
illustrates some of th e problems of working from printed texts which reflect
the opinions of their editors.
Two manuscripts of the Cardefia passionary survive. The earlier of the
two!" remained in Cardefia until 1864, when it went to the British
Mus eum. This passionary may repre sent the earliest version , becaus e it has
the fewest saints. The Passion if Pelagius was not included, although the
manuscript seems to have been copied after Pelagius' death." A marginal
note names the scrib e as the pri est Endura. 15 This may be the same Endura
who copied Cassiodorus' Commentary on the Psalms in 949 and Isidore's
Etymologies in 954 16 and witnessed charters of donations to Cardefia in 950 ,
966 and 969.17 Of th e manuscript's 269 folia, all but the last ten ar e in the
same hand. After the 'Explicit' ('Here ends th e first part of the passionary'),
and written in another hand, is an appendix. It consists of the sole
surviving copy of the Passion ifArgentea, the passions of Cyriacus and Paula,
two African martyrs of the fourth century, and the Invention if <'pilus, a
Cordoban martyr from the sam e period. Many versions of the passion of
Cyriacus and Paula (sometimes Paul) exist, but the Cardefia passion is very
similar to a hymn in a tenth-century manuscript from Toledo;" which
suggests that this version was widely disseminated throughout th e
peninsula. Zoilus' cult was also well known, and continued to be
celebrated in Cordoba. 19 Thus, although the additions might be Cordoban
in origin, and the palaeography is said to be characteristic of manuscripts
written in al-Andalus.P these texts could have come to Cardefia from
almost anywhere in the peninsula.
The so-called Passionary if Valdeavellano, discovered at Silos, may also be
tenth-century. This manuscript doe s not include the Passion if Pelagius even
though it was offered to the monastery of Valdeavellano, dedicated to
Pclagius, in 992 21 - although the manuscript may have been copied some
time before this. The provenance of the collection is not known, nor is the
date when it was taken from Valdeavellano to Silos. The manuscript, which
is now in two parts. " is incomplete, and the copyist complained that he was
working from an incomplete exemplar. As far as one can deduce from its
lacunae, this passionary comprised th e majority of the passions of the first
Passionary if Gardena, plus another ten. There are other, less important
differenc es between the two passionaries. The Passionary if Gardena, like the

83
 TWO MORE MARTYRS OF CORDOBA

Roman martyrology, celebrates the feast of Sebastian on 21 January;

Valdeavellano has 20 January. No direct dependence between the two can
be demonstrated.
Perhaps the earliest manuscript of the Passion qf Pelagius comes from an
incomplete passionary which forms part of a miscellaneous codex now in
Madrid.P Diaz dated the Passion qf Pelagius to the beginning of the eleventh
century, although the codex was put together much later." The passions
included in this fragment suggest that the collection as a whole was larger
than that of Valdcavellano. There is nothing in this manuscript to link it
with Silos or to any other monastery, although Diaz had no doubt that
it came from the Burgos area. A second passionary which includes the
Passion qf Pelagius may also have been copied early in the eleventh century. "
It is known as the Passionary qf Silos on the grounds that the scribe John
whose name appears on folio 48 is the 'Ioannes presbyter' who copied
another Silos codex." The saints included in this passionary are very
similar to those included in three calendars from Silos" and in the first
Passionary of Cardeiia." Another Silos manuscripti" contains the Passion of
Pelagius copied in the liturgical context of the office and mass for the feast of
Pclagius. This codex is a compendium of texts for nuns, including the Life
of the Frankish saint Seculina of Troclara, the Rule of Leander of Seville
and works on virginity by Jerome. Most of the manuscript may have been
written by one Vuilfurus, who signed folio 47v., but the hand of the passion
is different, and the text appears as a separate section of the codex from
which it was clearly meant to be detached and used . There is no indication
when it was added to the codex.
The second Passionary qf Gardena seems to have been copied to
complement the first, since it contains all the passions which are in the
Passionary qf Silos but were not included in the first Passionary qf Gardena. It
also has nine passions which are not in the Passionary of Silos, including that
of Nunilo and Alodia. These 'extra' passions were added in no particular
order, suggesting a gradual and random accumulation of saints as they
became known. Perhaps, under different circumstances, the acquisition of
so much new material would have prompted a recopying of the whole
collection in calendar order. Fabrega dated this second collection to the
end of the eleventh century. However, Diaz pointed out that the passions
in the second Passionary cf Gardena were numbered to follow on from
the first, minus its appendix." Thus the two manuscripts may be nearly
contemporary. Yet Diaz rejected the date given in a marginal note to folio
30 of the second Passionary qf Gardena ~ annus praesens Era 1020 [982] ~
arguing that the script of the manuscript makes it later, and that the gloss
must have been copied from an earlier manuscript.

84
 TWO MORE MARTYRS OF CORDOBA

The passionaries may tentatively be listed in the following order:

Passionary if Cardeiia I: BM Add .25600
Passionary if Valdeavellano : Madrid BN 494 and Paris n.a.l. 2180
Madrid BN 822 - [includes the Passion if Pelagius]
Passionary if Silos: Paris BN n.a.l. 2179 - [includes the Passion if Pelagius]
Passionary if Cardeiia II : Escorial b-I -4 - [includes the Passion if Pelagius]
appendix to BM Add .25600 - [includes the Passion if A rgentea]
The passion aries can be dated with no more certainty than to say that they
were probably compiled before the suppression of the Hispanic liturgy in
1080, after which time only the passions of saints whose cults were
celebrated in each monastery may have been preserved there . The Passion
if Pelagius continued to be copied in the thirteenth century, at least in Tuy
and Oporto."
Fabrega based his edition of the Hispanic Passionary on the two
manuscripts from Cardefia as though the other passion aries were mere
variants. Reference to the resulting work as the Hispanic Passionary,
however, produces an oversimplified view of the dissemination of
hagiographical texts in the peninsula. In addition to the codices already
mentioned, some eighty manuscripts and fragments survive, and there
must have been many more, judging from the number of times the terms
Passionum, Liber Passiones and Passionarium appear in library catalogues.F
Many of the manuscripts come from Catalonia, especially from Vich and
Montserrat. The different permutations of saints suggest a tradition that
was in constant evolution until the suppression of the Hispanic liturgy.
How do the manuscripts from Silos and Cardefia fit into this development?
The nucleus of the Hispanic passion aries is made up of the saints whom
Prudentius commemorated in verse. Some of the passions based on
Prudentius continued to be copied until the eleventh century, although the
passion of Eulalia of Merida seems to have been lost to the tradition which
fed Cardefia and Silos, since the text which they incorporate is in a later,
less classical Latin. The late sixth or early seventh century saw the
development of a composite Hispanic martyr text, to which the memories
of several martyrs were adapted. It was based on the itinerary of Dacian,
whose persecution of Hispanic christians started, in literary terms at least,
in the fourth-century passion of Vincent, martyred in Valencia c.304.
Dacian was held responsible for the deaths of Felix in Gerona, Cucufat and
Eulalia in Barcelona, Leocadia in Toledo, Vincent, Sabina and Cristeta in
Avila and the Innumerable Martyrs of Zaragoza, all of whose passions
share a common model which may be derived from the fifth-century
passion of Saturninus of Toulouse. Indeed, the model itself created the

85
 TWO MORE MARTYRS OF CORDOBA

three martyrs of Avila, who seem to be a complete fantasy. The second

canon of the Fourth Council of Toledo, held in 633, established uniformity
of the liturgy and encouraged the restoration of cults and the writing and
rewriting of passions, some of which were included in the works of Valerius
of Bierzo . Hispanic texts travelled to Francia in the eighth century;
they were copied throughout Septimania and Aquitaine, and the saints
they commemorated were included in the anonymous Mar!Jrology if Lyon,
written before 806 .33
Several passions may have been written or introduced into the
peninsula after the islamic conquest." In order to identify these texts,
Fabrega made several assumptions. The first is the idea that linguistic
analysis of a text will yield its approximate date. The further the text's
grammar has travelled down the road from classical to colloquial forms ,
the later it is likely to be, although the decline into vulgar Latin was not
linear. Secondly, Fabrega argued that if a passion had not been copied
north of the Alps at a given date, it had not yet been composed. Since the
author of the Martyrology if Lyon did not list the feast of Ascisclus and
Victoria, or that of Facundo and Primitivo, their passions must be ninth-
century or later." Yet he did not include Eulalia of Barcelona, nor other
saints whose passions his Hispanic informants would have known. The
compilers of martyrologies were selective about which saints to add to their
canon and the presence or absence of a particular saint in such texts
cannot be used as proof of the date of the dissemination of his or her cult.
Thirdly, Fabrega identified phrases in several of the passions which
he thought indicated that they had been written after 711.36 The epilogue
to the passion of Emeterius and Celedonius refers to a wandering 'getulus',
which Fabrega interpreted as a reference to the armies of the islamic
conquest. The Gaetuli were of North African origin, but although an
Asturian chronicler used the term on one occasion as a synonym for
Saracen," the author of the passion could have meant Vandals rather than
Arabs or Berbers. The penultimate passage of the passion of the three
Cordoban saints Faustus, Januarius and Marcialis begins 'vas, clarissime,
nolite credere', a speech which Fabrega interprets as being addressed to the
beleaguered christians of al-Andalus, although it is clearly addressed to the
three martyrs. Thus it is not clear which saints were added to passionaries
in the peninsula after the eighth century.
The general similarities between the Silos and Cardefia manuscripts,
and the absence of contemporary sources with which to compare them,
create the impression that they mark the penultimate stage of evolution of
a unified Hispanic passionary. This impression is misleading. The
transmission of other liturgical texts such as collections of prayers and

86
 TWO MORE MARTYRS OF CORDOBA

hymns suggests that the dissemination of the cult of the saints was not
uniform throughout the peninsula. These collections are difficult to date.
The earliest seems to be an collection of prayers known as the Oracional qf
Tarragona,38 which had left the peninsula for Italy by the early eighth
century. It includes prayers dedicated to saints not mentioned elsewhere
in the peninsula, such as Cecilia, Hippolitus and Ciprian, but omits the
Innumerable Martyrs of Zaragoza, Eulalia of Barcelona and others. The
author of a prayer to Eugenia appears ignorant of the details of her passion
as it appears in the passionaries from Silos and Cardefia. The Oracional of
Silos, which may date from the ninth century, includes many of the saints
mentioned in the Tarragona collection but not in the Silos passionary, but
omits many saints whose passions were well known in the peninsula.l?
There seem to have been two traditions of the transmission of liturgical
texts , one in the north ~ witnessed by manuscripts from Tarragona,
Carcassonne, Sanjuan de la Pefia, San Millan, Silos, Cardefia and Leon -
and one in the south. Manuscripts originating from Toledo show that the
two traditions coexisted in the former Visigothic capital.t" The earlier
the cult of a saint in the peninsula, as attested by inscriptions, the more
likely are the details of his passion to be common knowledge. Comparison
between the two manuscripts which preserve the majority of the hymns
copied or composed in the peninsula before the twelfth century, one from
Silos, the other from Toledo," shows that hymns continued to be added
after 711, but the resulting collections are very different. Some eight or ten
hymns, including the Ave maris stella from St-Gall, were introduced from
outside the peninsula, but only three of these were included in the Toledo
manuscript. There are other differences between the Silos and Toledo
traditions. A hymn to Nunilo and Alodia copied into the Silos manuscript
is not found in the Toledo hymnary'" and a hymn to Eugenia which
survives in two northern collections is quite different from that in the
Toledo manuscript. Since the number of surviving liturgical manuscripts
from the peninsula is so small , it is difficult to draw definite conclusions, but
it is likely that Hispanic traditions were several and various.
Thus the passionaries from Silos and Cardefia may not be
representative of the tradition as a whole . Further, the differences betw een
the Silos and Cardefia passionaries might be explained by different
circumstances of composition, their similarities by their geographical
proximity. The fact that the Passion of Pelagius may have been copied in Silos
before it reached Cardefia, and that the Passion qf Argentea was not copied
into any extant Silos text and was a late addition to the Passionary ofCardeiia;
has marginal utility for determining the provenance of the two passions , or
for dating their composition, and does not rule out the possibility that there

87
 TWO MORE MARTYRS OF CORDOBA

were othe r copies, or even var iant version s of th ese passions in other places.
Any clues to the origin and significance of these texts mu st be sought in the
texts them selves and in the places where we know that th ey were read.

The Passion of Pelagius

The prelude to the Passion if Pelagius was a Umayyad campa ign aga inst the
north. The christian army was defeated . Some were put to flight, others
wer e carried off to imprisonment in C ord ob a. Among the captives was a
bishop nam ed Ermogius. Exh au sted by the privation s of imprisonment, he
asked that his nephew Pelagius, then aged ten , might be sent as hostage in
his place. Pelagius is said to have rejo iced at the oppo rt unity to escape from
the many temptat ion s of life as a free man. In prison , he becam e famo us
for his piety: 'he was chas te, sober, peaceful, prudent, attentive to prayer,
assiduo us in reading, not forgetting the pr ecepts of the Lord , promoter of
im proving conversa tion, taking no pa rt in evil, not easily disposed to levity.
Refutin g all her esy, displaying the beauty of his body as mu ch as that of his
soul, he prepare d himself for th e dou ble crown of virginity and
martyrdom' .+3
After th ree and a half years of imprisonment, news of Pelagius' beauty
came to the ears of the king, who asked tha t Pclagius be brought to him.
The king offered Pelagius a number of inducements to renounce his faith:
'The king spo ke to him imme diately: "Boy, I will ra ise you up to the
hon ours of a high office, if you are willing to deny Christ and say that
our pr ophet is tru e. Don 't you see how great and how man y are
my realm s? M oreover, I shall give you a grea t deal of gold and silver,
fine clothing and costly baubles. You will also take whichever of these
young knight s you should choose to serve you according to your
tastes. I will give you compa nions to live with , horses to ride, and
luxu ries to savour. Then , I will release from pr ison whomever you
choose. If you wish, I will bring members of your famil y here and
confer grea t hon ours up on the m'";"
Pclagius scorned all th ese temptations. The next passage mu st also be
qu oted , as it has becom e the focus of rece nt interpretation s of the passion :
'Then, whe n the king tried to caress him playfully, holy Pclagius said,
"Get bac k, you dog! Do you thi nk that I am effeminate like
yourse lves?" Pelagius ripp ed off the robes in which he had been
dressed and mad e himself like a bold athlete in the are na, choosing

88
 lWO MORE MARTYRS OF CORDOBA

to die honourabl y for Chr ist rather than to live sha me fully with the
devil and to be defiled by his vices. The king, thinking tha t he could
per suad e him, instructed his atte ndants to sed uce him with
pandering speeches, so th at he might apostatise and submit to his
royal vanities. But Pelagiu s, with the help of God, stood stro ng and
remaine d undaunted , pr ofessing only Christ and saying th at he
would always obey his commands alone' .
The king becam e angry. H e ordered that Pelagius be strun g up in pin cers.
When he saw that the boy rem ained steadfast, he called for furth er tor tures
to be inflicted . Pelagius called on Go d as he was cut limb from limb. At last
he expired and his body was th rown into the river so that the faith ful
christians would not be able to find and bu ry him . It was eventually
recovered ; the head was buried in the ceme tery of San Ciprian , his body in
that of Sa. Ge nesius. The text ends: 'T he most holy Pelagius, at roughly the
age of thirteen and a half years , suffered martyrdom in the city of Co rdoba,
as it is said, during the reign of Abd al-Rahma nn [sic], certainly on a
Sunday, at the tenth hour, the twenty-sixth ofJ une, in the Era 964 [926] . . .'
The second Passionary of Cardeiia, in a gloss of the epigra ph to the Passion
qfPelagius, nam es the author as the pri est R aguel. Although this man uscript
is unlikely to be the earliest copy of the passion, as we have seen, Diaz
argue d that this m arginal gloss - Raguel presbiter doctor fuii huius passionis
cordobensis - was, like the dat e of 982, copied from an ea rlier manuscript.
As the Passion said that the relics were buri ed in Co rdo ba, R aguel mu st
have bee n writing before 96 7, when the relics of Pelagius were tran slat ed to
Leon.i 5 This is the only copy of the Passion to menti on R agucl, an d Diaz'
hypoth esis ca nno t be confirme d or refuted. The ph rase has been
int erpreted as meaning th at R aguel was the author of the Passion and
came from Cordoba. Ther e is no evide nce within the text that it was
writt en by an eyewitness, and the gloss is ambiguous - firstly beca use
R aguel could have been either the author or the copyist of the Passion, and
secondly, becau se cordobensis could refer either to R aguel , or to the place
whe re the ma rtyr died ." Although this am bigu ity was pointed out by
Flor ez in the eightee nth cent ury," subsequent commenta tors on the
passion have maintain ed that R aguel was a christian or a conver t to
C hr istianity from C ordoba and that the Passion qf Pelagius was written either
in th at city, or perhap s in the north, whence R aguel had fled from mu slim
persecution.
T he geographical terms em ployed by the hagiographer may help to
revea l his origins. The story of Pelagius bega n whe n the 'enemies [of the
christians] from th e who le of Hi spani a moved aga inst G allecia'. Altho ugh

89
 TWO MORE MARTYRS OF CORDOBA

'G allecia' is usually tra nslat ed as thou gh it is eq uivalent to pr esent-day

Galicia, th is ter m is ambiguou s. T he hagiograp her wen t on to say tha t 'if
Gallecia were entirely overthrown, these outsiders would have wielded
power over the faithful'." The campaign in which Er mogius was captured
is taken to be the first successful ca mpa ign which 'Abd al-Rahman III led
agai nst the north, in 920. 49 Accordi ng to Ibn H ayyan , 'Abd al-Rahrnan
captured Sa n Esteban de Gormaz and other towns in the U ppe r Duer o
before defeating th e combined Leonese and Basqu e ar mies at Val de
Junquer a, between Pamplon a and Estella. Ibn Hayyan 's chro no logy is not
always reliabl e50 but a dat e of 920 fits reason abl y well with the chronology
of the passion , if we allow that Er mogi us might have spent a couple of
years in pr ison befor e arranging to be exchanged for Pelagius. The
Chronicle ofSampiro'" men tion ed ano the r bishop , Dulcidius, captured in the
same campaign.F but since the compiler followed this imm ediately with
the martyrd om of Pclagius, this passage may not be inde pendent of the
Passion qf Pelagius. T he next cam paign against the nor th mention ed in the
Ara bic sourc es is th e cam pa ign of923 or 924 dur ing which Pamplon a was
sac ked.P Both these ca mpaigns wer e dir ected agai nst the northeast, not
aga inst G alicia. The hagiographer concluded the Passion with a prayer to
th e saint, th at he sho uld 'defe nd and cherish witho ut rest the churc h
which you see hon ours you with prayers and offerings, so that it will have
you, raised in Galicia but glorified th rough martyr's blood in Co rdo ba as
an advocate before G od '. C hristian writers in the north used 'H ispania'
for al-Andal us, but confined ' Gallecia' for roughly th e area now known as
Galicia. Writers in Ara bic, on th e other hand, used JilliqIya' for th e
who le of the christian north. 54 This usuage seems to be what the
hagiographer had in mind, and points to an Anda lusian origin for the
text .
In order for either Raguel or his text to have originated in Cordoba,
ther e mu st have been christians in the capital capable of composing such a
work. Yet it is often said that by the tenth century, Latin learning was
confined to th e north. Severa l of the Cordoban martyrs knew Arabic.
A cent ury later, it is argue d, the christians of al-Anda lus spoke Rom an ce
an d Arabic, but wrote only in Arabic. Wright holds Eulogius an d his circle
themselves respon sible for the loss of Latin, arguing that they develop ed a
style that was so convoluted tha t everyone else stoppe d writing Latin in
despair.55 When C ixila went from Toledo to found the mon astery of
Cos m as and Damian in Abellar, near Leon , in 905, he took with him a
number of Latin texts by Visigothic autho rs, listed in his don ation to the
mon astery dated 927.56 This has been taken to mean tha t they were no
lon ger of any use in the south. It is far from clea r that this was the case.

90
 TWO MORE MARTYRS OF CORDOBA

Latin works may still have been composed and copied in al-Andalus.
One illuminated copy of the Bible survives from al-Andalus. It was given to
the cathedral of Seville in 988, and was probably copied a generation
earlier.57 The manuscript, 'made at a very capable scriptorium' has several
islamic motifs; the neck of a bird illustrating the capital at the head of the
Book qf Daniel is inscribed with an Arabic legend. Another Bible, from
Cordoba, given by a muslim to Alfonso II, is listed in his donations to
Oviedo of 908. 58 The language of the Passion cf Pelagius shows a command
of hagiographical formulae and could hardly have been written where
Latin was in terminal decline. Furthermore, a cleric with little or no Latin
would have been of no use to a new patron in the north. Either Raguel was
not a Cordoban, or, as seems more likely, the Passion qf Pelagius could have
been composed either in the north or in al-Andalus.
Just as the origin of Raguel and the Passion of Pelagius have proved
impossible to pin down, so the wider question of christian emigration is
more opaque than some scholars have assumed. The subject of emigration
is beset by onomastic problems. Men with Arabic names appear as
witnesses more than three thousand times in Leonese charters, with a peak
incidence in the tenth century. In only two cases is the witness' religion
mentioned and both were converts to Christianity. The others may have
been christians, jews or even muslims ." Oliver, in a study of the etymology
of the name 'Saracen' concluded that some of the witnesses to charters
with Arabic names were converts to Christianity from families of eastern or
North African origin.60 Even some of those with obviously christian names
gave muslim names to their children, which suggests either that they were
not trying to hide their Andalusi origins , or a degree of 'onomastic
indifference'i'" It cannot be assumed that all these men were christians
from al-Andalus, fleeing from islamic persecution.
Similar problems afflict discussions of Asturian art and architecture.
Scholars have been so convinced of the existence of a profound cultural
divide between the christian kingdoms and al-Andalus that the borrowing
of islamic styles in the north can be explained only as the effect of christian
emigration from al-Andalus. The striking differences between the
architecture of the church of San Salvador de Valdedios of 893 and San
Miguel de la Escalada, dedicated in 913, have led to the first being called
'Asturian', the second 'Mozarab', the result of new techniques coming
north with Cordoban monks.F This interpretation of the evidence is
controversial. Although some aspects of the construction of the church at
Escalada may have been influenced by islamic techniques, the church
might also be seen a return to a Visigothic style, and its closest parallel may
be Sanjuan de Banos, founded by Recceswinth in the seventh century'" As

91
 TWO MORE MARTYRS OF CORDOBA

we saw in chapter two, the church of Santa Maria de Melque, outside

Toledo, seems to combine Visigothic revival with islamic decoration. It is
difficult to determine the ideological content, if any, of christian use of
islamic art. 64 The pictorial reform which frontier monasteries witnessed in
the tenth century may have been inspired by immigrants from al-Andalus.
Yet these manuscripts rarely have a specifically anti-muslim content. When
the Book qfDaniel was added to copies of Beatus' Commentary on the Apocalypse,
Muhammad was not added to the list of AntiChrists.65 Christian artists
sometimes used islamic motifs, such as the rider who appears twice in the
Girona Beatus of 975, probably produced at Tabara, south of Leon/" The
artist may have employed the figure of the rider to represent Herod, the
persecutor of christians. Such borrowings are rare, and confined to a small
area, and islamic motifs could also be used in an unpolemical way.67 The
difficulty in dating and ascribing models to the art and architecture of both
north and south suggests that educated men throughout the peninsula -
christians, but probably muslims and jews as well ~ remained in contact
during this period.
There were two periods during which christians may have left
al-Andalus in significant numbers. The first was the late eighth century,
when the principal destination of the emigrants was Francia. The Hispani
who settled in Septimania and the Hispanic marches were presented in
Carolingian documents as the victims of persecution. A privilege granted
to them referred to 'the iniquitous oppression and most cruel yoke which
the Saracens, so hostile to the christians, have imposed upon them'; the
Spaniards had 'abandoned their own lands which belong to them by
hereditary right, have fled from Spain and have been taken to Septimania
to live there' .68 Yet a document of812 which names 42 Hispani shows that
some of them, rather than being exiles, may have been native to the area,
and two of these men were using the Arabic names Zoleiman and Zate. 69
The best known emigrant to Fran cia is Theodulf of Orleans, who may
have come from Zaragoza. His call to Charlemagne to take up again the
fight against Cordoba seems to encapsulate the plight of the exile." Yet
Theodulf prospered in his adversity. Like his Hispanic contemporaries
Benedict of Aniane and Agobard of Lyon, he played a prominent role in
Carolingian religious reform and monastic expansion. These men may
have been attracted to a centre of christian patronage rather than
propelled by religious persecution. Elsewhere in Europe, only the
occasional find of a Visigothic manuscript which may have left the
peninsula after 711 testifies to those whom Perez de Urbel, in a detailed
study of medieval Hispanic monasticism, called 'refugee monks from
Spain'."

92
 TWO MOR E MARTYRS OF CORDOBA

The Kin gdom s of the Asturias an d lat er Leo n may, like C ha rlemagne's
court , have attracted emigrants from al-Anda lus, During the reign of
Fru ela (757-67), the mon astery at Samos was restored by monks from the
sou th. Bishop Odoario, said to have been exiled to Africa, retu rn ed to
Lugo, where he repaired the walls and reorganised the diocese - altho ugh
this story comes from a series of forged charters which have Odoario
leaving the peninsula in 711 and returning after the victori es of Alfonso I,
some 100 years later," When Ordofio I (850- 66) restored Leon, Astorga,
Tu y and Am aya , 'he filled them partly with peopl e from his own kingdom
and partl y with th ose coming up out of Spain 'r' " The majority of emig ra nts
to the nor th recorded in the Asturi an charters, inscriptions and chronicles
mad e th e j ourney in the tenth century. Alfon so III invited Sebast ian,
bishop of Ercavica to becom e bishop of Orense." In addition to C ixila,
abbot of th e mon astery of Cosmas and D am ian in Abellar, monks from the
south founded or refounded the mon asteries of San Cebrian de Mazote,
Vallado lid, 75 Sa n M artin de Castaneda," and Sa n Migu el de la Escalada. "
The dedication inscrip tion at Escalad a rea ds: 'T his place, of old dedicated
in hon our of the archa ngel Michael and built with a little bu ilding, after
falling int o pieces, lay lon g in rui n un til Abbot Alfon so, coming with his
brethren from C ordoba, his fatherland , built up the ru ined house in the
time of th e powerful and sere ne prince Alfonso"." Anda lusi churc hmen
moved to the kingdom of the Asturias, j ust as their compa triots had
pr eviou sly gone to Fran cia, becau se they saw it as a land of opportunity.
The key to chr istian emigra tion from al-Andalus was the expansion
southwards of the kingdom of th e Asturias and Leo n at the end of the nin th
and th rou gh out the tenth cent ury. Most of those emigra nts from the south
whom we can identify in th e sources were involved in the great wave of
mon astic settleme nt of this period. The Asturi an chroniclers por trayed
Alfon so III as moving back into the 'no-man's land' which Alfon so I had
crea ted in the eighth century by laying waste the ar ea which they called
the 'Gothic fields' , thus creating a barri er between al-Anda lus and the
beleaguer ed christians of the north. This 'co rdon sanitaire' was, for
Sa nchez-Albo rnoz and his disciples, the key to Spa in's development ," a
par ap et from behind which the chr istians fought the 'ba ttles with [the
Saracens] day and night ' described in the Chronicle if Albelda.8o Here the
Reconqu est began . Yet the appa rent emptiness of the Duero regio n may
be, at least in part, an artefac t produced by the lack of any surviving
cha rters dating from before the late ninth century. The conti nu ity of place
nam es from Roman times th rou gh to the present day argues aga inst their
abando nme nt. It is likely that mon asteries such as San Millan continued
to exist from the Visigothic period , altho ugh they were not in a position to

93
 TWO MORE MARTYRS OF CORDOBA

attract wealthy donors who desired their gifts to be commemorated in

charters. The survival of the local law-codes of Sepulveda'" is further
evidence of continuity of settlement. Few excavations have been carried out
in the region, but there is so far little trace of depopulation and
repopulation of the Duero.P except perhaps for a small area near the river
Lirnia." The vocabulary of 'repopulation', of places being described as
'desertus et incultus' in charters from the north-west of the peninsula, was
used to justify Asturian territorial claims to the region. Alfonso III granted
landholding rights to men prepared to take control of the Duero in his
name. In turn, these men gave land to monasteries. The majority of
foundation or refoundation charters from these monasteries do not
mention any men from al-Andalus, and their signatories may have been
Asturians. It is likely that those clerics who came north to join them would
have been attracted by the carrot of newly-endowed monasteries rather
than driven by the stick of persecution. They would have been educated
men, capable of participating in the cultural life of the christian kingdoms,
including the writing of hagiography. One of them could have written the
Passion qf Pelagius, either in Cordoba or in an Asturian monastery.

The cult of Pelagius

Unlike the martyrs of the 850s, Pelagius was honoured in his own country.
The Latin version of the Calendar qf Cordoba, for June 26, notes that 'This is
the feast of Pelagius, and his tomb is in the church of Tarsil' r" There is no
known dedication of a church to Pelagius at Tarsil , which lies in the
Campifia, the countryside around Cordoba, but there was a basilica
dedicated to Genesius, where Pelagius' body was said to have been buried,
and this may be the church to which the Calendar refers. The Calendar qf
Cordoba is in general fairly well informed about Cordoban saints, but as we
shall see in the next chapter, the surviving Latin version may not have been
compiled in Cordoba, and cannot be relied upon as testifying to the
strength of any cult associated with that city. It is interesting that the
Calendar refers four times to the church of San Cyprien without mentioning
that Pelagius' head was buried there. In spite of this Cordoban connection,
details within the Passion qf Pelagius and the subsequent development of the
cult of Pelagius link the text very firmly with the north.
The Passion of Pelagius should be read through northern christian eyes
and in the context of a growth of female spirituality. As we have seen, on e
of the codices containing the Passion of Pelagius is a collection which may
have been put together for a community of nuns at Valdeavellano who

94
 TWO MORE MARTYRS OF CORDOBA

celebrated his cult. M an y of the new mon astic founda tions were for
women . Several texts on virginity and lives of virgin saints were copied at
this period. About 9 12, Leodegundia copied a collection of texts includ ing
the Rules of Benedict, Isidore, Lea nder and Fructuosus, letters from j erome
and th e Lives of Consta ntina and M elani a." Another manuscript,
containing the Lives of seven female saints.i" was copied by a scribe
nam ed j ohn in 954. One of these is the Lif t qf Helia, which is probably
fictional and may have been writt en in Hi spania. It draws on j erome's
writings on virginity, and has scenes reminiscent of the Acts of Paul and
Thecla." a very popular text which circulated in the penin sula from the
fourth century and was copied int o the Passionary ofSilos and the second
Passionary qfCardefia.John's collection also includes Ildefonsus' Onthe virginiry
qfMary, j erome's Against ]ovian and Against Helvidius and Braulio's Lift ofSan
Millan, added in a different hand , which strengt hen the collection as a
manual of fem ale spiritua lity, with an emphasis on virginity.
It was Pelagius' virginity which made his story so popul ar with wom en.
T he fact that he was a male rather th an a fem ale expose d to sexual
temptation sho uld not be overemphasised. The Passion qf Pelagius carefully
spec ified, in three plac es, that Pelagius was thirteen and a half years old at
the tim e of his martyrdom. Read ers would have recognised tha t Pelagius
was on the brink of pubert y, defined by Isidore as the end of the fourteent h
year.88 j ordan ana lysed the Passion qfPelagius in terms of the develo pment of
th e theology of homosexu ality," arguing that it is impossible 'to disentan gle
th e rete llings of the Passion qf Pelagius from the amb ivalent rela tions of
Iberian Christianity to th e same-sex love it is tho ught was pr eached and
pr actised by Islam '. j ordan 's rea ding has as mu ch to do with modern ant i-
mu slim polemi c as with ea rly medieval ideas. The sexual imagery used by
autho rs of hagiography and other writings on virginity is difficult for
modern read ers to underst and, as is the correspo ndence between christian
clerics; similar probl em s affect our reading of Arabic poetry?" It is not clear
wha t the overheated lan guage of this idealisation of same -sex relationships
had to do with ac tua l sexual activity. Visigothic law codes had legislated
agai nst hom osexu al pr actices," as had autho rities as disparate as the
Theodosia n Code and Irish penitentials. Anti-muslim po lemic, which
accused mu slims in genera l, and Muhammad in particular, of sexual
licence, did not specify hom osexuality. W hen, in respon se to the king's
adva nces, Pelagius stripped off the fine cloth es in which the court iers had
dr essed him he (or rath er his ha giogr apher) was employing a topes -
Pelagiu s was taking off the clothes which represented falsehood in
preparation for the contest leading to a martyr's crown, rather than
'playing with same -sex desire'.92 In fact, the Passion qf Pelagius is rem arkably

95
 TWO MORE MARTYRS OF CORDOBA

free of anti-m uslim polemi c. The story of Pelagius is reminiscent of that of

Paulillus, recounted in th e fifth-century Chronicle if Prosper. Gaiseric, the king
of the Vandals, had condem ne d to death four men who m he had ea rlier
favoured , becau se they refused to adopt Arianism:
'A boy called Paulillu s, the brother of Eutychianus and Paschasius
[two of the condem ne d men] , who had been somewha t favoured by
the king because of his beauty and intelligenc e, when no thr eats
could dissuad e him from his profession and love of the catholic faith,
was beat en for a long while and condem ned to the lowest servitude .
Yet, as it appears, he was not killed in case his youthful age might be
glorified, throu gh the defeat of the impi ou s king's savagery'.93
If the Passion if Pelagius is a lesson on the sin of Sod om , it is in the wider
sense used by writers such as H incmar of Rheims,?' whe re sexual
perversion s of all kind s go hand in hand with extravagance and pride. As
J erome had explained: 'the Sod omi tic sin is pride, bloatedn ess, ab unda nce
of all things, leisure and delicacies'." Pelagius resisted the material
temptations offered by the king as well as his sexua l adva nces: 'E nric hed
bo th by his virginity and by the crown of his suffering, he bore a do ub le
victory aga inst the enemy - scorn ing riches, and not giving way to vices' i?"
We will see Arge ntea achieving a similar victory.
It was as a text on virgini ty that th e Passion of Peiagius was retold in verse
by H rotsvit, a ca no ness in the Ottoni an imp erial abbey of Gandersheim .
H rotsvit finished her last work before 973.97 The poem was copied into the
earliest surv iving collection of her works, the eleventh-cen tury Emmeram-
Munich Co dex. It is one of eight legends in verse which also includ e the
stori es of the virgin M ary and of Agnes who, having refused to mar ry a
pagan , was shut up in a brothel but rem ain ed uncorrupted until her
martyrdom. In the prologue to the legends, Hrotswit declared:
'I found th e mat erial I have used in thi s book in vari ou s ancient
works by authors of reputati on , with th e exception of the story of
th e martyrdom of St. Pelagius, which has been her e told in verse .
The details of thi s were supplied to me by an inh ab itant of the
town whe re the saint was put to death . This truthful stranger
assure d me that he had not only seen Pelagius, who m he descr ibed
as th e most beautiful of men , face to face, but had been a witness to
his end '."
H rotsvit's informa nt could be Recemu nd of Cordoba , whose visit to the
Ottonian court in the lat e 950s is described in the next cha pter. Yet
H rot svit's acco un t, which is melodr am atic even for hagiography, differs in

96
 TWO MORE MARTYRS OF CORDOBA

many details from th e Passion if Pelagius, and shows very little kn owledg e of
Islam. After th e fall of the peninsula to th e pagan king, Hrotsvit explaine d,
on e cond ition alone was m ad e for pea ce, 'that no dweller of th e aforesaid
city should presume to blasphem e the gold en idol's name, whom thi s
prince adored, or else, it was so willed , this man was promptly to be
killed' i'" Peace ensued, although th er e were some martyrs, until th e
accession of 'Abd al-Rahrnan, a degen erate and harsh persecutor of th e
christians. Pelagius, the only son of a duke captured during th e emir's
campaign against Galicia, languished in Cordoba because his fath er could
not rais e th e money to ransom him. When the king tried to embrace the
beautiful youth, Pelagius struck him on th e face, drawing blo od. Enraged,
the king ca tap ulted Pelagius' body over th e city walls, but he remained
miraculously unharmed until his head was cut off. Pelagius' relics
performed so m an y miracles that his head was submitted to an ordeal
by fire in order to prove that the boy saint was ind eed responsible for th em.
These details owed more to conte mpo rary mores in northern Europe '?"
th an to Hrotsvit's informant. It is also possible that th e story of Pelagius
had already ac quired legendary embe llishme n ts during th e half century
after his death.
It m ay have been on th e initiative of a woman that th e relics of Pelagiu s
wer e translated to the Asturias. Sever al claims to Pelagiu s' relics were being
made at about the same tim e. It is difficult to weigh up th eir merits, as this
dep ends on th e int erpretation of ch ronicle and charter evid en ce whi ch has
still to be ad equately evaluated. The Chronicle if Sampiro says that in 960,
Sancho th e Fat and his sister Elvir a , daughter of Ramiro II and regent for
her young nephew Ramiro III (967- 84), sent an embassy to Cordoba led
by bishop Velasco to obtain th e relics of Pelagius for Leon. lUI Elvira was the
abbess of San Salvador in Leon, 1U2 wh ere the relics arrived in 967. The
most reliable version of the Chronicle if Sampiro is probably that extant in th e
anonymous Historia Silense, composed at the beginning of th e twelfth
century,1 03 although where th e Historia Silense overlaps with th e Chronicle if
AlfonsoIII th eir details do not always conc ur. Pelayo of Ovi edo, a notorious
forger, int erpolated his own version of th e Chronicle if Sampiro into a work
called th e Chroniclefrom the beginning if the world to 1170 AD, wh ere he seems
to be att ributing it to bishop Sampiro of Astorga (fl.c. 1035 -40).104 Another
Sampiro was a royal notary in Leon in th e late tenth or early eleventh
century. In 99 5, th e relic s of Pelagius may ha ve been mo ved from Leon to
Ovied o becau se of th e threat to Leon of attack by al-M ansur, 105 the def acto
rul er of Cordoba. They were hou sed in the churc h of John the Baptist,
which was attach ed to a convent rul ed by Elvira 's sister Teresa. In 1063,
wh en Fernando I and Sanch a dedi cated a new church to Isidore on th e

97
 TWO MORE MARTYRS OF CORDOBA

site, they donated a reliquary for J oh n the Bapti st and Pelagius, which is
still in the church treasury.'?"
The competing claim s of Leon and O viedo to th e relics of Pelagius have
more to do with political rivalry within north ern Hispani a than with the
actual fate of Pelagius' bon es. The pr om otion of the cult of Pelagius can be
compa red with th e cult of six Anglo-Saxo n saints, all young kings or
princes, whose status as inn ocent martyrs was used to bolster the royal
minsters where th ey were bu ried.' :" As in the Passion if Pelagius, the English
hagiographers emphasised the saints' physical and mental purity. The
Chronicle if Sampiro recorded th at , towards the end of R ami ro Ill's reign, a
revolt in Gal icia elevated a rival king Vermudo II , the son of Ordono. l'"
The autho r of th e Chronicle took a fairly neutral line on the revolt, recording
that Ram iro and Vermudo met in battle without settling their claims, and
that Ram iro died shortly afterwa rds. The chro nicler did, however, place
the passage on Sancho's quest for the relics of Pelagius immediately after
mentioning an ea rlier revolt in Ga licia, and before recount ing Sancho's
atte mpt to subdue Galicia, which suggests that he saw the acquisition of the
relics as part of Leon 's claim to hegem ony over Galicia. Later, Pelayo of
O viedo expanded the whole episode of the revolt and blamed the success
of al-Mansur's campaigns aga inst the north on the sins of Vermudo. T his,
like all Pelayo's additions to his sources, was pure propagand a for Oviedo.
Rami ro Ill 's moth er was Teresa, the abbess of the monastery in Ov iedo
which claimed Pelagiu s' relics. The same mon astery had sheltered
Velasqui ta, Vermudo's first wife, whom he had repudiated. Even worse
perh aps in bishop Pelayo's eyes was the fact that Vermudo had driven an
earlier bisho p of Ovi edo from his see. 109 Vermudo's claim to legitim acy in
th e eyes of later histori ans was based on his enforceme nt of the Visigothi c
law codes. In contrast, R am iro's orthodoxy rested to some extent on the
association between his family and the relics of Pelagius. It should also be
noted that at least one oth er monastery was making a claim to Pelagius'
relics which may antedate that of Leon and Oviedo. San Pedro de
Valerani ca was founded c.930 near th e river Arlanza . The prologue of a
codex of Smaragdus copied by Floren cio, active at Valeranica between 940
and 965, lists Pelagius amo ng the saints whose relics were pr eserved in the
monastery.'!" It is clear tha t man y royal and noble foundations were
anxious to celebra te the cult of this saint.
It was in Galic ia, in the narrowest geographical sense of this ter m, that
the stro ngest claim to Pelagius was mad e. There are now more than
40 pari shes in Ga licia which preserve the memory of Pelagius, altho ugh
man y of these may be fairly modern. III In the earliest copies of the Passion
if Pelagius, and in a Silos Sacramentary of the eleventh cent ury,'!" the see
98
 TWO MORE MARTYRS OF CORDOBA

which Pelagius' uncle Ermogius occupi ed is not named. The version of the
office for Pelagius inserted into Vuilfurus ' codex of texts for nuns'! " gives
Ermogius' see as Tu y and this manuscript seems to have been retouched at
a later date in order to mention Tuy,' B It is possible that Pelagius was not a
native of that city, but that Tu y began to assert its claim to Pelagius at about
the time when his cult was being promoted in Leon and Oviedo . During
the second half of the tenth century, the diocese ofTuy was under threat of
dissolution, perhaps because of Viking attacks. I IS In 1000 a certain Viliulfo,
bishop of Tu y, witnessed a charter. Perhaps he was the Viulfurus whose
nam e appears on the codex containing the Passion cf Pelagius altered to
mention Tuy. 116 It is not clear when the clergy of Tuy began to assert their
claim to Pelagius. A ma ss in th e saint's honour was edited by Prudencio de
Sandoval, bishop ofTuy (1608- 12), in a work on the city and its bishop s. II7
The attribution of the mass to 'a mozarab of about the year 930' may be
Prudencio 's. H e not ed that there were still three churches dedicated to
Pelagius in Galicia. The ma ss takes a standard form and ends with a very
bri ef account of Pelagius ' passion. The church of Tu y claimed Pelagius as
'their martyr', 118 saying that he had been born in the western part of
Galicia where his family held lands , but they do not seem to have possessed
his relics, saying only that 'Tu y received this our patron, your martyr who
suffered in his bod y in Cordoba'; '!" that is, Pelagius was their patron saint
even though they ma y not have received his bod y. In fact, Tu y ma y not
have had any relics of Pelagius until the millennial celebrations of his
death, in 1925. 120 If Pelagius had once been the patron saint of the city of
Tuy, he stopped being so some time after 1251, when the Dominican
missionary Pedro Gonzalez Telmo was adopted; at least Tuy had Telmo's
relics. 121 Pelagius was most closely associated with the mon astery of Alveos,
which preserv ed the tradition that Pelagius had been born there. 122 This
association was linked to a tradition that Ermogius had retired to a
monastery which he found ed in honour of Pelagius, overcom e by the fate
of his nephew. There may not be any historical basis for this tradition. A
letter from an Ermogius 'confessor' from Celanova, dated 951, links this
man with the monastery of Cabrugia Rivolimia, but the monastery itself
was dedicated to St. Christopher. It is difficult to say how this Ermogius is
related to the bishop of the Passion ofPelagius.
There ar e, however, tantalising snippets of information about the
episcopate in the northwest which seem to tie in with the story of Pelagius.
The tenth- and early eleventh-century bishop s came from the nobility, and
seem to have passed on their office from uncle to nephew - however this
relationship was defined. A charter dated 976 from San Miguel in Lemos ,
which may have been drawn up early in the eleventh century, named the

99
 TWO MORE MARTYRS OF CORDOBA

donor as the brother of bishop Vimara of Tuy, nephew of Ermogius, who

himself was the nephew of bishop Nausto.!" Some of these bishops became
involved in the struggles between Ramiro III and Vermudo II. One of the
exiles at Celanova was a bishop named Pelagius expelled from the see of
Iria by Vermudo in 985 . 124 Since so many of these details are found in
forged charters, it is difficult to evaluate them. The saint and his uncle may
indeed have been natives of Tuy, and perhaps Pelagius had been destined
to succeed his uncle as bishop. Yet it is equally possible that the diocese of
Tuy based their claim to the cult of Pelagius on their reading of the Passion,
and the coincidence between the name of one of their former bishops and
the Ermogius of the Passion if Pe/agius.
The cult of Pelagius was one aspect of a struggle towards self-definition
for the christian north. The dominance of historiography by the Kingdoms
of the Asturias and then of Leon and the rhetoric of Reconquest, with its
requirement for christian unity in the face of the enemy, obscured the
continuing struggles between Galicia and Leon for mastery in the north, in
which competing claims to the cult of Pelagius played a role. Pelagius may
also have been revered as a symbol of the struggle between the christian
kingdoms and al-Andalus. In spite of its literary conventions, there is a
contemporary feel to the story which, although the Passion does not use the
language of anti -muslim polemic, draws its significance from the fact that
Pelagius suffered martyrdom in Cordoba. Pelagius' hagiographer may
have been presenting him as the saviour of his people when he wrote: 'For
the nation which he left, he possessed the paradise he desired'!" -
although this is a topes. The story of Pelagius reminded educated readers of
the Pelagius/Pelayo of the Chronicle ifAlfonso III, the hero of Covadonga. 126
Pelayo was also sent to Cordoba, as an envoy or, in another version, as a
hostage, by Munnuza, a companion of Tariq ibn Ziyad, the conqueror of
al-Andalus, so that Munnuza could marry Pelayo's sister. The writer makes
it clear that this was more significant than just a personal affront to Pelayo:
'When Pelayo returned, he by no means consented to it. Since he had
already been thinking about the salvation of the church, he hastened to
bring this about'. Tariq's soldiers were sent to apprehend Pelayo and take
him back to Cordoba but Pelayo escaped to the mountains, where he was
exposed to worldly temptations similar to those offered to the child martyr.
Oppa, the bishop of Toledo and son of king Witiza, the penultimate
Visigothic king of Hispania taunted him: ' ''If'', said Oppa, "when the
entire army of the Goths was assembled, it was unable to sustain the attack
of the Ishmaelites, how much better will you be able to defend yourself on
this mountain top? To me it seems difficult. Rather, heed my warning and
recall your soul from this decision, so that you may take advantage of many

100
 TWO MORE MARTYRS OF CORDOBA

good thin gs and enjoy the partnership of the Chaldeans'v."? Pelayo's reply
referred Oppa to the parabl e of the mu stard seed : " 'C hrist is our hop e that
through this little mountain, which now you see, the well-being of Spain
and the army of the Gothic people will be restor ed "'. H e stood firm and
achieved the first victo ry of the Reconquest.

The Passion of Argentea

The Passion if Argentea op en s another wind ow on Cordoba. Unfortunately,

as only on e copy survives and th ere is no evidence for Argentea's cult,
int erpretating the story of Argentea depends on the text alone. Argentea
was the dau ght er of a king nam ed Samuel and a queen nam ed Columba of
the city of Bibistr ense. She rejected the tr appings of royalty and dedicated
herself to chastity. When her mother died , she refused her fath er 's request
to carry out the duties of a royal consort a nd immured herself in a secure
cham ber below the palace. Even th is privation failed to satisfy her
aspirations and, hearing of a monk who was looking for martyrdom, she
wrote to him asking for his advice about following this way to perfection.
H e advised patience until the right oppo rtunity pr esent ed itself This
seemed to come with the overthrow of Bibistrens e, following which
Argentea and several of her fellow-citizen s made their way to Cordoba.
Argentea spe nt several years in Cordoba, appare ntly pr actising her
vocation of chas tity undisturbed . Then , th e Passion records, a Frank named
Vulfura was summoned in a dr eam to come to Cordoba to help her
achieve her goal of martyrdom. Vulfur a was immediately arrested and
thrown into pri son , where Argentea visited him assiduously. During one of
these visits, she was recognised as Samuel's daughter: 'she found herself
sur ro unded on all sides by pagans, and she heard, having been asked
ha rmful qu estions: "Are you not , 0 woman, the daughter of Samuel their
prince?" . . . Ther efore the blessed Argentea , wishing to be a participant in
the hop ed-for passion , rejoiced [and] intrepidly declared that she was not
only the daughter of the aforeme ntioned fath er, but trul y a guardian of the
ca tholic faith'. 128 Argentea and Vulfur a were brought before a judge,
qu estion ed , tortured and put to death . Argentea was buried in Cordob a
but the fat e of Vulfura's relics was unknown even to the hagiographer.
In the nin eteenth century, historians writing about the Passion if Argentea
dr ew its circumstantial det ails together to crea te a narrative which plac ed
Argentea not only in the geographica l heart ofUmayyad power, but at the
centre of the christian - m uslim conflict of the tenth century. Dozy, an
ind efatigable connector of bits and pieces of evidence, identified Argentea 's

101
 TWO MORE MARTYRS OF CORDOBA

native city, which the Passion names as Bibistrense, with a place which the
Arabic sources called Bobastro.!" This assumption made Argentea
the daughter of Ibn Hafsan, the head of a prominent family of converts
to Islam who led a long-running revolt against Cordoba. Bobastro was his
stronghold. Simonet stretched this association, saying that Argentea was
'carried to Cordoba with her brother Hafs and the other citizens of
Bobastro,' as though they were being taken into a Babylonian captivity. In
fact, the passion says that 'cumftatribus ceterisque concivibus Cordobensum urbem
petivit' which implies that she went willingly to Cordoba, accompanied by
some of her fellow citizens, among them perhaps monks rather than her
biological brothers. Add to this mixture an extra-textual reference to the
Cordoban martyrs of the 850s and the result is a neat representation of
religious conflict in al-Andalus, coming to a climax with Argentea's death
at the heart of islamic power.
The linking of Argentea and Ibn Hafsun may have deterred those
historians who have written so copiously on the ninth-century martyrs
from examining the Passion if Argentea, by placing it in the wrong
historiographical camp, with the Arabic sources . Indeed, the career of Ibn
Hafsun illustrates the problems of the Arabic sources only too well. He is
copiously documented, but in contradictory accounts. Although several
modern historians, most notably Acien , have tackled the problem of these
sources, it is still very difficult to arbitrate between them. 130 How relevant is
Ibn Hafsun to the martyrdom of Argentea?
According to Ibn 'Askar's History if Malaga, 131 Ibn Hafsun was the
descendant of a prominent Visigoth named Marcellus. His grandfather
had converted to Islam . Most of the Arabic historians related his
campaigns against Cordoba and his alliances with neighbouring rebels,
but with markedly different emphases. It appears that Ibn Hafsun
alternated periods of opposition to the Umayyads with service to the
regime in the army and as a governor. Ibn al-Qatiya says that, after many
years in rebellion, 'Ibn Hafsun died at the beginning of the reign of 'Abd
al-Rahman ibn Muhammad ['Abd al-Rahman III] after having made
friends with him and pledged his allegiance's!" His career points to a
weakness of central control of al-Andalus which is attested by frequent
rebellions in Toledo and elsewhere. 133 There are several similarities
between the career of Ibn Hafsun and his treatment by later historians and
a more famous medieval Hispanic adventurer, the Cid . 134 Only since the
nineteenth century has Ibn Hafsun been portrayed as a focus of nationalist
aspirations. Ibn Hafsun may have apostatised to Christianity as part of his
protest against the Umayyads. Yet modern historians' emphasis on this
apostasy shows them viewing their sources according to their own

102
 TWO MORE MARTYRS OF CORDOBA

perspectives. Cutler insisted that 'in the famous 'Umar ibn Hafsun the
dreams of the martyrs' movement were perfectly realised'. 135 Yet
Ibn al-Qntlya, who may be a contemporary witness, did not mention
Ibn Hafsun's apostasy and the earliest account of it appears the Anonymous
Chronicle if 'Abd al-Rahmdn al-Nastr, 136 an extract from the work of
Ibn Hayyan which survives in a copy of perhaps the fourteenth century.
Again, it was Ibn Idhari, who described the building of the Cordoban
mosque in such detail, who claimed to know the exact date oflbn Hafsun's
defection to Christianity!" Archaeologists claim to have found the church
where Ibn Hafsun was buried, at Las Mesas de Villaverde, Ardales, near
Ronda (Simonet supposed the ruins to be those of Ibn Hafsun's fortress)
but the identification of this site as Bobastro is still disputed;138 the church
has been dated only as 'early medieval', and could be Visigothic.P? It is
possible that the accusation that Ibn Hafsun was an apostate was made in
order to blacken his memory. The story is elaborated with the detail that
Ibn Hafsun's apostasy had been secret , and was revealed only when his
body was exhumed and found to be buried in the christian manner; 'Abd
al-Rahman III was said to have exhumed his corpse and crucified it on the
walls of Cordoba between those of two of his sons. It is likely that
Ibn Hafsun had christian support, but the question of his conversion is
much less certain.!"
To return to more immediate matters, none of the sources mention
Ibn Hafsun's daughter, and the dates for his career are difficult to square
with those given by the Passion if Argentea. Argentea was said to have left
Bibistrense after its overthrow. Ibn Hafsun died c.917, but this was not the
end of the rebellion .141 Bobastro faced a major Cordoban attack c.923/4
and was defeated c.928,142 when Hafs, one of several sons of Ibn Hafsun,
was in control, The Anonymous Chronicle says that 'Hafs ibn 'Umar was
brought before the emir with his family and hostages, who confirmed the
peace treaty and gave them good welcome'. It is unlikely, that Argentea
saw this as an opportune moment to be martyred. If, however, she had left
Bobastro earlier, during one of her father's periodic defeats, what the
Passion described as the 'turning cycle of her years' in Cordoba must have
seemed to stretch interminably if she had to wait until 931 - the date given
in the Passion - for martyrdom. It is impossible to mine the Passion if
Argentea for facts in any convincing way. The whole confection that
Argentea was the daughter of a proto-nationalist catholic rebel must be
rejected, leaving us with something more complicated.
Like the Passion if Pelagius, the Passion if Argentea is primarily a treatise on
the two principal christian virtues , described in this text as a twofold
handful of flowers - the white of chastity and the purple of martyrdom -

103
 TWO MORE MARTYRS OF CORDOBA

and is divided almost equally between the two aspects . Apart from the date
given for her martyrdom, the historical details are vague, and the story is
modelled on the anonymous hagiographical romances known as the gesta
martyrum rather than on contemporary debate with Islam. Argentea's
journey to Cordoba recalls the fourth-century martyrdom of Agape, Irene
and Chione at Salonika, who 'when the persecution was raging, ...
abandoned their native city, their family, property and possessions because
of their love of God and their expectation of heavenly things' . 143 When
Argentea made a public declaration of her faith ,
'the crowd excited into fury brought the follower of Christ before the
judge. Interrogated by the judge about the conditions of the faith, she
responded with constancy thus : "Why do you exasperate me with
your questions? Have I not testified that I was a follower of the
embraced christian faith? But because according to the apostolic
dogma, which believed in the heart leads to justice and confessed
through the mouth to deliverance, I confess before all: I believe in
one God in three persons, adored in indistinguishable substance, and
declare the personality to be unconfused'v.!"
Although holding ludicrously complicated beliefs about the nature of God
was one of the accusations made against christians by muslim polemicists,
this passage seems to have been lifted from a passion of a much earlier
period, where the Trinitarian statement is addressing anti-Arian concerns.
This impression is reinforced by the relatively good Latin of this text.
Equally anachronistic is the reference to an unnamed praeses, whose
judgement is enforced by tutors, surely out of place in tenth-century
Cordoba. Yet, if the hagiographer was merely reworking a conventional
passion for anti-muslim polemical purposes, the result is not a success. Some
of the new details work against the picture of Cordoba as a city blighted by
muslim persecution and the focus for opposition to Islam that he or she may
be trying to present. The text says of Argentea's arrival in Cordoba, 'united
with religious people in that city, .. . and firm in the usual fashion, she
emptied herself busily in continence, and thus for a long time she lived
through the turning circle of the years. With how many and what kinds of
virtues she flourished in her fashion and was illustrious in pious acts, if we
tried to set them all out with our pen we would seem to set out what was no
less decorative and boring'. The impression thus given that there was a
flourishing christian community in Cordoba is the only part of the whole
story which rings true. Churches built in the Visigothic period were still
active at the time of the Reconquest. Surviving inscriptions, and the copies
of others made by earlier antiquarians, commemorate nuns who died in the

104
 TWO MORE MARTYRS OF CORDOBA

tenth century!" The impression one gains from these fragments is that
Cordoba was a christian as well as a muslim city. A pious woman from
outside the capital would easily have found a convent to receive her. Being
martyred in the way outlined by the Passion might have been more difficult.
It is not clear when and for whom the text was written. Argentea's feast
day is not listed in the Calendar if Cordoba. In both the Arabic and Latin
versions of the Calendar if Cordoba, October 13 is given as 'the feast of three
martyrs put to death in the city of Cordoba. The sepulchre is in the District
of the Tower'. The Latin version adds : 'And their festival is in the [church
of] the Three Saints'. 146 This is assumed to be the church of Faustus,
Januarius and Martialis, where Argentea was supposed to have been
interred. The church, which was the most important in Cordoba after the
Arab conquest, and later changed its name to St. Peter, survived until after
the fall of the Umayyads. Yet the memory of Argentea does not seem to
have been preserved there.!" Two allusions in the text might indicate that
the author was a native of Cordoba - it says that Argentea 'made her way
to Cordoba as a stranger (advena) , of which her body would soon in the
future be an inhabitant' and after her death, her miracles were 'constantly
taking place among us up to now ', but neither of these allusions are
conclusive. The Passion if Argentea survives only in the copy written in a
northern monastery and, like the Passion if Pelagius, it should be interpreted
as relevant to a northern audience. Working from this assumption, it is
possible to speculate how the Passion if Argentea made sense as a product of
the christian kingdoms. Could it perhaps be a moral tale about the duties
of a princess, perhaps based on the existence of real martyrs in Cordoba,
but addressed to contemporary events in the north?
The names of the participants in the drama are unusual, to put it no
more strongly, and suggest that the protagonists of the Passion ifArgentea are
symbolic rather than real. Argentea's name recalls that of one of Eulogius '
martyrs, Aurea: silver and gold. Her father Samuel does not seem to be
have anything to do with the biblical Samuel. His name might, however,
bring to mind one of the contemporary rulers of Leon and Pamplona. The
rulers of Pamplona were called Sancho Garces or Garcia Sanchez
alternately. Sancho the Fat ruled Leon from 955-7 and from 960 -7.
Bibistrense, Samuel's royal city, may recall Barbastro, near Huesca, rather
than the elusive Bobastro of the Arabic sources. Argentea's mother does
not seem to have any connection with the female saint called Columba
who was commemorated in the Hispanic passionary, but in this context,
the name Columba may have reminded readers of a dove and of the burial
chambers called columbaria; it was an appropriate name for a protagonist
whose main function in the story was to die. Vulfura sounds like Wulfhere,

105
 TWO MORE MARTYRS OF CORDOBA

which is not a Frankish but an An glo-Saxon nam e, but suggests a foreign er

- although it also recalls th e scribe Vuilfurus.!" In this story his martyrdom
was not described , and his relics wer e apparen tly m islaid , so that he seems
no more than a literary device, like th e monk wh o told Argentea to wait
before she could fulfil her desire to be martyred . Perhap s Vulfura wa s
m eant to be downplayed in comparison with Argentea. Alternatively, he
became sup erfluous to the story. Argentea and her family and companions
are improbable, but their names may have help ed reader s of the Passion qf
Argentea to locate th e story in northern Hi spania around th e end of th e
tenth century.
It may be possible to pin down th e genes is of th e story mor e pr ecisely, as
an other manifestation of th e grow th of fem ale monasticism. Elvir a , wh o
sponsore d th e cult of Pelagius, seem s, like Argentea , to have been an
exe mplary princess. Perhaps othe rs were not. When in 957 , San cho th e Fat
lost his throne, th e Arabic source s say th at he was restored only with th e
support of a C ordoban army, after he, his gra ndmothe r Tota and Garcia
Sanchez I of N avarre had made an em bassy to Cordoba, and sur rende red
a number of frontier fortresses to th e muslims.!" Perhaps this deal involved
sending a royal princess to th e Umayyad harem. Several of th e Umayyads
had christian wives and con cubines, and th e family was fam ed for th e
blondness of its princes. A woman from N avarre called Subh in the Arabic
histories, became prominent at th e Cordoban cour t, after bearing two sons
to al-Hakam II. One of th e boys died young, but the othe r acce de d to th e
throne as Hisharn II at an early age and his mother was involved in his
regen cy, po ssibly becoming th e mistre ss of th e chambe rlain Ibn Abi Amir
who later usurped th e Umayyad throne and took th e title al-Mansur.P" She
is comm em ora ted in th e inscription on th e base of a fountain erec ted on
her orders in 977 . 1:> 1 Subh wa s a concubine who had been cap tured in a
campaign against Navarre and th ere is no suggestion that she was of royal
blood. Al-Mansur may hav e may have m arried on e of th e daughters of
Sancho the Fat , or possibly of one of th e kings of Navarre. The mother of
on e of his sons was 'the daughter of the ch ristian king Shanjuh', the Arabic
for Sancho. P? Sin ce th e Arabi c sources refer onl y to th e caliph's consor ts
who provided him with sons, th er e may have been other Navarrese or
Leonese princesses in his en tour age who achi eved less prominen ce in
Cordoba but wh ose defection to th e seduc tions of life with the infidel m ay
be the subject to whi ch th e Passion qf Argentea is a counterblast. The
hagiographer m ay be arguing that, rather than travelling to Cordoba to
consort with th e ene my, a christian princess, should she find herself in that
city, should prepare herself by a life of chastity and prison visiting to be
worthy of a martyr's crown .

106
 TWO MORE MARTYRS OF CORDOBA

Cordoba, seen from the perspective of the Passion if Pelagius and the
Passion if Argentea is not primarily a topos; it is not, as one might have
expected from modern reconstructions, the place of martyrdom, a far-away
city where christian unease about Islam received its ultimate expression.
Although the hagiographers do not evoke Umayyad splendour, not even by
mentioning the great mosque, their picture of a city where Christianity was
still openly practised, which must come from the tenth or eleventh
centuries, may be more realistic than those drawn by the Arabic historians.
The latter were remembering Cordoba long after its glories had passed,
when they were free to imagine how it might have been. For the
hagiographers, the significance of the tenth-century martyrs of Cordoba
was that they died as virgins, and this virginity was perhaps the highest
virtue at a time when virtue was threatened by the lure of life under
Islam .P" Pelagius the child saint served this metaphorical purpose just as
well as Argentea did and his name recalled the saviour of the christian
kingdoms. Both hagiographers showed that to go to Cordoba was to place
one's salvation in doubt. In Cordoba, christians might tempted by power,
riches and sexual pleasures to fall into sin and their princesses might be
diverted from the chastity and martyrdom of the ideal christian life. Both
texts are polemics, and as such they exaggerated the dangers which faced
christians in the land of Islam. The next chapter illustrates a more
measured christian response to life in Cordoba.

101
 CHAPTER SlX
--i_"-

Recemund and the

Calendar of Cordoba

'Conce rn ing astrology: Ibn ZaId the bishop , the Cordoban, wrot e on
this subject. He was th e favourite of al-M ustansir ibn a n-Nasir the
M arw anid [al-H akam II] for whom he composed the Book on the
Division if the Seasons and the Hygiene if the Body I in which he indicated
the sta tions of the moon and how these matters were relat ed, and
their approximate sign ificance'."

W hen Recemund, a christian official at the Cordoban court, was chosen

to lead a delegation from 'Abd al-Rahma n II I to Otto I, he earned
him self both a bishopric and a leading role in the history of the christians of
al-An da lus. Ther e ar e man y referen ces in the Arabic sources to christian
members of embassies from Cordoba to Europe and Byzantium," but the
Arabic histori an s supply little mor e tha n the names of these men ,
some times adding that they acte d as interpreters. Recemund's career is
mu ch bett er do cumented . Recemund's embassy to Otto was part of the
story of an ea rlier mission which Otto had sent to Cordoba, led by J ohn of
Gorze. Recemund met Liudprand, lat er bishop of Crem on a, who
dedicated a work of history, the Antapodosis, to him. In the nin eteenth
century, wh en the Calendar if Cordoba was rediscover ed , Recemund was
identified as its author and as the bishop Ibn Zatd , who composed a
calenda r for al-H akam II. Recemund's services to 'Abd al-Rahman III
make it perfectly plausible that he could have composed a calendar for his
son. Yet the components of this reading are two different versions of a
calenda r and a number of refer ences scatte red through the Ara bic sources
to bishop s who wrote calenda rs or went on embassies, non e of them nam ed
as Recemund. Recemund was identified as the author of the Calendar if
Cordoba as a result of two erro rs which characterise mod ern historiography
of al-Anda lus. The first was a failure to take an y acco unt of the
tran smission of the Calendar. The second was the ruthless appli cati on of
O ccam 's razor to dispar ate pieces of evidence, rend er ing them into a

108
 RECEMUND AND THE CALENDAR OF CORDOBA

composite Recemund. Depending on how many inconsistencies one is

prepared to skate over, some of the steps in the argument are plausible but
the conclusion is full of holes. The aim of this chapter is to reconsider the
evidence for Recemund, in order to assess the importance both of the man
and of the Calendar for the history of al-Andalus,

The Life ofJohn of GOTze

The main evidence for Recemund is the Life if John if Gorze/ a daughter
house of M etz in Lorraine. It survives in one tenth-century copy from
Metz, which was probably written by John of St-Arnulf in Metz, who
succeeded his namesake as abbot of Gorze. In his preface, the author said
that he began to write on the day after the saint's death . The first part,
which describes John's origins and education before he became abbot of
Gorze, was finished by 978 . It is followed by portraits of some of the monks
at Gorze and ofJohn himself. The final section is a long account of his
journey to Cordoba. The author did not give the information aboutJohn's
abbacy which he had promised in his introduction and the work seems to
have been unfinished rather than curtailed in transmission, since it breaks
off in the middle of a page. The Life is detailed and imm ediate and the
outline ofJohn of Gorze's embassy to Hispania is corroborated by other
sources .
According to the Life, John of Gorze's embassy to Cordoba was Otto's
response to a delegation from 'Abd al-Rahma n III congratulating Otto on
his victori es. Liudprand of Cremona reported a purpose which was much
less peaceful. A dispute had arisen between the two rulers over the attacks
on Otto's lands by pirates who had established their base at Fraxin etum
[La Garde Freinet in the Gu lf of St. Tropez]." They had probably been
preying on the surrounding area and on pilgrims crossing the Alps, for
about a century. The Latin sourc es called them Saracens, and later Arabic
histori ans agreed that they were muslims, but their origins are obscure. It is
not clear wh eth er th ey owed any allegiance to Cordoba, although Ibn
Hayyan reported that an unnamed christian prince, travelling to Cordoba
in 940 , asked 'Abd al-Rahman III to guarantee his safe passage through
Fraxinetum" and the Life says that th e embassy from Cordoba was intended
to forestall any reprisals by Otto against al-Andalus. The mission
found ered, said John of St-Amulf, because the sentiments expre ssed in
the caliph's letters to Otto were considered derogatory to the christian
faith . The ambassadors were kept waiting for three years at Otto's court,
then sent ba ck empty-handed . Wishing to reply to 'Abd al-Rahman's

109
 RECEMUND AND THE CALENDAR OF CORDOBA

insults, Otto asked Adalbert, bishop of Metz, to find someone to lead an

embassy to Cordoba. Adalbert turned to John of Gorze. His party was
escorted to Cordoba by merchants from Verdun who were active along
the routes connecting Hispania with northern Europe and Byzantium.7
The Lift does not give the date ofJohn's embassy, but it could be the one
mentioned by Ibn Idhari: 'In the year 348 [953-4] the ambassadors of
Otto, king of the Slavs arrived'." This evidence dates from three centuries
later, and there were almost certainly other embassies from Otto to the
Umayyads, but it is reasonable to suppose that John of Gorze's embassy
took place in the early 950s.
John of Gorze was not impressed by the moral standards of the
christians of Cordoba. He did not hand over Otto's letters immediately, in
spite of threats that if he did not divulge their contents to agents from the
palace, he would surely be killed for carrying letters blaspheming against
Islam . He had, however, already confided their contents to a presbyter, one
of the members of the unsuccessful Andalusi delegation to Otto. This man,
fearing for his life ifhe returned to the caliph empty-handed, passed on the
information to the Cordoban court. 'Abd al-Rahman was now in a
dilemma; honour forced him to kill Otto's ambassador to avenge the insult
to his faith, but this risked provoking war with the emperor. The only
solution seemed to be for all concerned to pretend that the letters had
never existed . John of Gorze could not be persuaded to agree to this
stratagem. First a jew named Hasdeu (perhaps Hasdeu ibn Shaprut, who
was prominent at the Umayyad court)" was sent to mediate with John of
Gorze, then the bishop of Cordoba, also called John. The bishop pleaded
with John of Gorze, saying that his intransigence was putting the whole
christian community of al-Andalus in danger:
"'Consider" he said "under what conditions we exist. Through our
sins we have been reduced to this, that we are subject to the sway of
the pagans. The Apostle forbids us to resist legitimate power. There
is only one consolation in this calamity, that they do not forbid us our
law . .. In the circumstances, therefore, it seems wise to us to comply
with all things which do not hinder our faith . .. "'10
John replied: " 'You, who appear to be a bishop, are the last person who
should speak thus .... It is a thousand times better for the christian to suffer
the cruel torment of hunger than to eat the food of the gentiles at the cost
of his soul"'. John declared that he would be torn limb from limb rather
than fail in his task of delivering Otto's letters . It was for God, not John, to
dictate what would happen to the caliph's christian subjects . At last 'Abd
al-Rahman, won over by John's courage, decided to send a new embassy to

110
 RECEMUND AND THE CALENDAR OF CORDOBA

Otto to seek a resolution of the crisis. The man commissioned to head the
delegation was Recemund.
Recemund comes out of this story little better than the other christian
collaborators condemned by John. The lift described him as a palace
official who diligently carried out his duties . His apprehension about being
the caliph's envoy to Otto was overcome only by the prospect of material
advancement. In the dramatic words of the Lift, Recemund asked the
caliph "'What reward are you going to give to the man who sells you his
soul?'" I I Recemund agreed to lead the embassy if the caliph recompensed
him with a see which had just fallen vacant: 'T his', the Lift continued, 'was
easily granted and he was suddenly advanced to the office of bishop from
the laity'. Recemund obtained the see of Elvira, near present-day Granada.
In the dedication of the Antapodosis, Liudprand referred to Recemund as
the 'bishop of Elvira in Hispania';' " and this appellation was repeated by
Sigibert of Gembloux, Trithemio and the false chronicler of Pavia,
although they may all have read the Antapodosis. Recemund could have
served as bishop from 958, succeeding Gapio, who died in that year.
Gapio's name is the last to appear in a list of the bishops of Seville, Toledo
and Elvira in the Codex Emilianense, I 3 begun in 962, and finished in 994. The
reason why the list stops with Gapio may be that, because it was a
necrology, Recemund was not included because he was still alive in 994.
But this was some forty years after his appointment and it seems that the
see of Elvira may have lapsed either during or after Recemund's
occupancy, because no more bishops were recorded. Perhaps he was not
accepted as bishop because of the uncanonical nature of his elevation.
Yet we should not accept every word ofJohn of St-Arnulf's description
of Recemund and the Cordoban christians. The hagiographer's working
of his material reflects the ideals of the Benedictine reform led by Gorze in
which John of Gorze played a prominent role ." The Lift is given over to
the great piety of its subject, and the structure of the portrait ofJohn in
chapters 72-94 echoes those of the Rule of Benedict and the Collection qf
capitularies of Benedict of Aniane." To some extent, the author pursued this
scheme in describingJohn's journey to Cordoba, which appears - perhaps
factitiously given that the Lift is incomplete - as a dramatic coda to John's
contemplative life as a soldier of Christ. Otto asked for monks to act as his
intermediaries in Cordoba, since, as the Lift put it, 'being already dead to
the world '!" they feared no earthly danger. The bishop of Metz's first
choice as leader of the delegation to Cordoba was Wido, but he committed
a most unBenedictine crime of disobedience: 'rebuked for having fallen
into some neglect of his monastic duties [he] burst out in insults against the
abbot and the whole community and enraged by his great wrath and folly,

111
 RECEMUND AND THE CALENDA R OF CORDOBA

he refused to subm it to any reprimand or mon astic discipline, so that at last

they rem oved his cowl and he was expelled from the monastery '. J ohn
volunteered to replace him , and although he was really too valuable to be
spared, his sacrifice was accepted in ord er that he might fulfil his desire for
martyrdom.John was not dism ayed by 'Abd al-Rahman's hostile reception.
Since the Andalusian delegation to Otto had been kept waitin g for thr ee
years, John in his turn was prepared to be detai ned for nin e. Now the
author's angle on his story becomes clear. It is yet another example of
th e saint's steadfastne ss in th e service of his faith, contrasted with the
transgressions of Wido. Once Recemund's embassy had been successfully
concluded, John was received at the Cordoban court. The impression
which the luxury and pageantry of this ceremony mu st have made is
obvi ou s from the biographer's long description of it but he em phasized
John's refusal to put on elaborate clothes for the occasion , yet another
instan ce of his sanc tity.
The complaints about the christians of al-Andalus which J ohn'S
biogr apher puts into his mouth may reflect John's views on th e
com promises involved in living under mu slim rule. Leyser suggested tha t
the pirates of Fraxin etum were merely the pr etext for a mission whose real
aim was 'religious reconnaissan ce' . I7 But this section of the lift read s as
though writt en to adher e to the conventions of the pa ssions of the fourth-
century martyrs as closely as the biographer could contrive, given that J ohn
returned from Cordoba unharmed and lived for another twenty years. To
this end, politi cal differences wer e played down and crisis was represent ed
as almo st entirely spiritual. H ence the Lift's stress on the two rulers'
inability to compose letters whose sentiments the other could tolerat e. The
christians who had compromised their faith to such an extent that they
wer e able to live under the domination of heretics had to be condemned.
The same bias may colour th e picture which the Lift paints of Recemund.
Ther e is another aspect of the Lift that contributes to our view of
Recem und. The hagiographer also praised John of Gorze for his learn ing,
devot ing a long section of his work to J ohn's education, his tra vels through
Italy and the scholars with whom he studied . It seems to have been J ohn's
eru dition rath er than his sanctity which prompted the lay and ecclesiastical
lead ers in M etz to appoint him to Gor ze. Even after he becam e abbot, and
in the face of his bishop's objection s, he continued to travel in pursuit of
wisdom . Under his abbacy, Gor ze attrac ted men from as far afield as
En gland, Ireland, Scotland and C alabria. " It is likely that Recemund
becam e part of this comm unity. The autho r of the Lift included Recemund
in his encomium of scholarship in spite of Recemund's mor al failings,
describin g him as 'most ca tholic' and as learned in Latin as in Arabic. 19

112
 RECEMUND AND THE CALENDAR OF CORDOBA

This furth er contradicts the view that the christians of al-Andalus had lost
their Latin a century earlier. Recemund's journey looks like a scholarly
enterprise as much as it does a diplomatic one . His party did not go directly
to Frankfurt but to Gorze, arriving in August and spending the autumn
and wint er th ere. M cCluskey proposed that Recemund's visit to Gorze was
one of the channels through which knowledge of Islamic astronomy was
transmitted to northern Europe, takingJohn of Gorze beyond the study of
computus which was part of the monastic curriculum.i" Recemund
stopped at Gorze on his return journ ey" There is the suggestion , although
it shou ld be no more than this, that he brought with him the Passion if
Pelagius.22 Unfortunately, there are no surviving examples of Recemund's
Latin, although he could have written the two anonymous letters, one of
them addressed to the counts Miro and Borr ell who ruled Barcelona in the
950s and 60s, of which only fragm ent s survive .P It would be a wrong to
attribute to Recemund every contemporary instance of contact between
the Latin cultures of al-Andalus and christian Europe. Yet is clear that
Recemund was not merely acting as a trans lator, nor as the venal
functionary of an alien ruler, but was able to hold his own among christian
scholars outsid e Hispania.

The dedication of Liudprand's Antap od osis

It is in this context th at the dedi cation to Recemund of Liudprand's

Antapodosis should be seen . Liudprand, a deacon of Pavia, spent his youth at
the courts of Hugh of Arles and Berengar of Ivrea , Hugh's successor as
ru ler of Italy. He desert ed Berengar for Otto, then the ruler of Saxony.
Liudprand served as Otto's ambassador to Constantinople in 960 and was
rewarded with the see of Cremona in th e following year. H e took part in
Otto's campaign against Berengar in Italy which culminated in Otto's
imp erial coronation in Rome in 962. Based on Ibn Idhari's date of 953/4
for John of Gorze's mission to Cordoba, Recemund's journey to Frankfurt
is usually dated 956, although it could have been two years later, when the
see of Elvira fell vacant. Liudprand was vague about when exactly he
himself became an Ottonian partisan but he could well have been at Otto's
court when Recemund arriv ed. The Antapodosis was written , according to
the preface to its third book , 'to reveal, declare and stigmatize the doings of
this Berengar, who now is not king but rather despot of Italy.. . . Let this
present page then be to them antapodosis, that is, repayment'." In the
opening paragraph Liudprand explained the dedic ation to Recemund and
his reasons for writing:

113
 RECEMUND AND THE CALENDAR OF CORDOBA

'To the reverend Lord Recemund, Bishop of Elvira , full of all

sanctity, Liudprand, the unworthy deacon of Pavia, sends greetings .
For two years, dearest father, through lack of skill I have postponed
compliance with your request, when you urged me, as one who did
not depend on doubtful hearsay but had the sure knowledge of an
eye-witness, to set down the doings of the emperors and kings of all
Europe'."
Liudprand then added three paragraphs on his unfitness to write the work,
before using as one of his examples of God's punishment of the wicked and
reward of the virtuous, an episode which seems to illustrate quite the
opposite, that is the atrocities committed by the Saracens of Fraxinetum.
Even though Recemund was representing the Umayyad caliph accused of
sheltering the pirates, who had attacked Liudprand's home town of Pavia ,"
this had not caused friction between Recemund and Liudprand. The
episode of the pirates was a shared concern and a subject to which
Liudprand returned several times in the Antapodosis. By beginning with this
flashforward to the matter of Recemund's embassy, Liudprand was closely
linking the text with his new friend . He had no more to say about Hispania
but he continued to address Recemund throughout the Antapodosis,27
although the work ends with Liudprand's first visit to Constantinople in
949, well before the meeting between the two men. The Antapodosis was
clearly written, or rewritten, with Recemund in mind.
Yet the general drift of the Antapodosis is difficult to follow. It appears
inaccurate, inconsistent, and scurrilous even by the standards of the time .
Liudprand's competence and even his mental stability have been
impugned. These problems are evident from the opening of the work.
Liudprand rather disingenuously elaborated his first reference to
Fraxinetum by saying :
'I imagine, my lord, that you are well-acquainted with Fraxinetum
and know it better than I do, since you have the information of those
who are tributary to your king Abderehamen. But for the benefit of
the general reader, I will say here that it has the sea on one side, and
on all others is protected by a close undergrowth of cactus'r'"
Liudprand discoursed on the defensive virtues of this cactus, which the
Saracens had encouraged to grow so large that to stumble against its
spines meant certain death. The passage illustrates one of the difficulties
in reading the Antapodosis, where Liudprand never let the truth get in
the way of a good tale. The archaeological remains, albeit from the
thirteenth century, suggest that the pirates' base at Fraxinetum owed its

114
 RECEMUND AND THE CALENDAR OF CORDOBA

impregnability not to cactus but to its elevated site and it is not near the
sea.
Next, Liudprand told several anecdotes about Byzantium and its rulers,
before coming to his ostensible purpose, a history of Germany and Italy
from 887 to 949, which does not in fact become an eyewitness account
until Book 4. Liudprand seems to have had two aims. The first was a
personal one ofjustifying his desertion of Berengar for Otto and the second
an attempt to show divine retribution, in which, for example, the Saracen
attacks on Italy can be read as punishment for the misdemeanours of Pope
John with Theodora.P It is not clear how well the work achieved either of
these goals. Neither does Liudprand seem to be thinking primarily about
Recemund. He spent too much time settling personal scores for the
Antapodosis to function well as the introduction to the recent history of
Europe he had promised Recemund and his moralizing explanations
became buried under the weight of anecdotes which filled his pages . Yet he
did not deal as harshly with Berengar and his spouse Willa as might have
been expected, and, whilst he portrayed Otto favourably, it is difficult to see
this text as being addressed to Otto's circle.30 Recent interpretations of the
work have made sense of it by focussing on Liudprand's praise of Otto,"
especially of his Italian policies.32
There is no evidence that the Antapodosis reached Hispania and one can
only speculate about how it might have been received. Recemund may
have known what sort of history to expect. He may have read an earlier
work by Liudprand, now lost, which is listed in a medieval catalogue as the
Deeds if the kings and principal parts if Europe 33 (unless this is the Antapodosis
under another name). Recemund may have anticipated using the
Antapodosis to show his fellow countrymen how divine judgement had
influenced the history of Europe. There may have been a demand in
al-Andalus for providential history in the tradition ofJohn of Biclar and
Isidore. As we shall see in the next chapter, Orosius' Seven Books if History
against the Pagans may have been translated into Arabic at the same period
to bring its message to a wider audience. Perhaps Liudprand's work had a
even deeper resonance for christians living under islamic rule. The style of
the Antapodosis, its mixture of heavy-handed satire and the heroic deeds
of kings, are not just the product of Liudprand's psychopathology but have
much in common with the Phrenesis of Rather of Verona." Liudprand
praised Rather's 'book describing in witty and elegant language the
sorrows of his banishment'P in Pavia . Liudprand claimed that part of
the Antapodosis was written in captivity on the island of Paxos, stretching the
facts a little, because he was merely absent from home on a mission for
Otto. Both Rather and Liudprand were referring back to Boethius , whose

115
 RECEMUND AND THE CALENDAR OF CORDOBA

prose style they adopted, and to the consolation of writing in adversity. As

Liudprand says in the opening paragraph of Book 6 of the Antapodosis, 'If he
concentrates on his writing and describes how some rise and some fall on
fortune's wheel, he will feel less acutely the troubles that now beset him ';"
Recemund would have understood these sentiments well. He was living in
spiritual exile under an islamic regime, however comfortable his life might
have been. He may have welcomed the Antapodosis as providing him with
some consolation in his own adversity.
The Life if John if Gorze and the Antapodosis show another side to
Recemund to the Cordoban functionary of modern histories of al-Andalus.
It is not possible to know how Recemund stands in comparison with
contemporary christians who have been forgotten. The survival of two
substantial sources referring to the one episode in which he played a
leading role may be no more than chance. Yet they show that for at least
one Andalusi christian, Latin culture was alive. In Gorze and in Frankfurt,
Recemund was accepted by christian scholars, and shared their intellectual
concerns.

The Calendar of Cordoba

The work now known as the Calendar if Cordoba consists of two versions,
one in Arabic, the other in Latin, of an astrological almanac of Arabic
origin supplemented by a christian liturgical calendar. The outlines of the
two works are very similar and there is no doubt that they are related,
although there are obvious differences in content, of which the most
striking is that the Latin version includes many more saints . Each begins
with a brief summary of its contents and an introduction, to which I will
return. This is followed by the calendar proper. Each month is listed first
by its Latin name, then by its Syriac and Coptic names, with the number
of days in that month, its zodiac sign and the position of the sun . Next
comes the name of the season and an indication of what weather might be
expected. There are a number of dietary and medical recommendations,
such as whether it is advisable to practise blood letting or to take medicines
on specified days. This advice is derived from Hippocrates and Galen.
Then the days are listed one by one, although neither version of the
Calendar has an entry for every day. Some entries note a christian festival,
others give snippets of astronomical or agricultural information. Much of
this was intended to be practical. The entry for 12 January recommends
that onions should be planted before the end of the month." Each month
concludes with a summary of the agricultural activities undertaken during

116
 RECEMUND AND THE CALENDAR OF CORDOBA

that month and a passage about nature which is often extensive and
lyrical.
The Latin text of the Calendar was published in 1838 from a manuscript
which may date from the thirteenth century. " In 1866 an Arabic text
written in Hebrew characters was rediscovered in the Bibliotheque
National in Paris and this was presumed to be the original, from which the
Latin translation was made by Gerard of Cremona and the school of
translators in Toledo. Seven years later, Dozy published the two texts side
by side and used each to elucidate the textual problems of the other. He
dated the Calendar to 961 because he identified these texts as the calendar
which Ibn Zaid dedicated to al-Hakam II, who acceded to the throne in
that year. In making the assumption that the Latin text was simply an
expanded version of the Arabic, Dozy took no account of the manuscript
transmission which invalidates his approach, as we shall see. Dozy's setting
of the two texts in parallel (which Pons rightly described as 'his curious little
work'?9 was the basis for Charles Pellat's modern edition. Pellat combined
the two versions to produce a single translation into French, favouring the
Arabic version when the Latin is not exactly equivalent. In doing so, he
gave the Arabic version priority, and obscured the differences between the
two texts. Pellat underlined his hypothesis that the Latin was a version of
the Arabic by 'correcting' the Latin from the Arabic and by reintroducing
into the Arabic version 'missing' passages translated from the Latin ,
although he did point out where he had done so. He also used a similar
work written by Ibn Qutayba of Baghdad (d.889)40 to elucidate difficulties
in the Arabic version . Most importantly for the historiography of the
Calendar, Pellat was perpetuating Dozy's mistake in combining two different
texts as the Calendar qf Cordoba. Fortunately, interpretation of the Calendar
does not entirely depend on this edition.

The Books oj Anwii'

The Calendar qf Cordoba belongs to a group of astronomical and

astrological works known as the Books qf Anuui'. It is not clear how the
Calendar fits into the development of this genre. The subject is complicated
by imprecise terminology. Islamic authors used varying names at different
times for their astronomical and astrological texts . An excursus into the
question of the development of the Books qfAnwa' may help to clarify some
aspects of the Calendar qf Cordoba which have been thought, quite
erroneously, to be original contributions by the Calendar's author, or
authors.

117
 RECEMUND AND THE CALENDAR OF CORDOBA

Observation of the heavens may have begun with farmers in the Nile
valley, who noticed that the rising of the floodwater of the Nile coincided
with the first appearance of the star Sirius in the east just before dawn,
after a period during which the star rose only in daylight." This
observation was developed into a calendar based on the simultaneous
rising and setting of pairs of stars . The earliest written calendars date from
the seventh century BC, from Greece and Babylonia.t? During each period
of the calendar the weather was thought to be stable and predictable, so
that observations of the movement of the heavens could be used to direct
agricultural activities. Certain days were said to be lucky or unlucky. In
Greece, popular astronomy gave birth to literary, philosophical and
scientific traditions. The Phenomena of Aratus of Soli of c.275 BC, a poem
on the constellations and natural signs which could be used to predict the
weather, was very popular, and was translated into Latin at least three
times. Columella's treatise on farming of c.50 AD contains an astronomical
calendar and such works became widespread in the Middle Ages.
The people of pre-islamic Arabia divided their calendar into a number
of aniod' (singular naw'). The term naw' is imprecise but Ibn Qutayba of
Baghdad defined it as the simultaneous setting of one constellation in the
west and the rising of another star or constellation in the east." Some time
before the advent of Islam, the Arabs seem to have become familiar with
the stations of the moon, a system of stars or constellations where the moon
could be observed during each of the twenty-eight nights of its revolution.
This came to Arabia from Indian astronomy. The Arabs seem to have
transferred to these Indian lunar mansions the names of their amoa: Thus,
by the time that the first Books if Anuui' were written, probably in the
schools of Kufa and Basra in the eighth century," the term Books if Amoa'
was already something of a misnomer for astronomical works based on the
lunar mansions rather than on the paired constellations. To make matters
more complicated, alongside the Books if Anwa', islamic scholars developed
a second type of treatise, called the Books on Time. 45 These were
lexicographical works which assembled the names used for different
periods of the day, month and year and terms describing climate. The two
genres were often combined, and the resulting texts could be given either
title. Although most astronomers of the islamic period soon abandoned the
lunar mansions in favour of the more practical solar year, lunar
phenomena continued to be used in divination, and it is in this form
that they passed to western Europe in the later Middle Ages. Calendars
such as the Calendar if Cordoba were based on the solar year, but in the
entries for individual days and months, their authors included discussion of
the movements of the stars copied from the older Books if Anwa' and

118
 RECEMUND AND THE CALENDAR OF CORDOBA

lexicographical derivations from the Books on Time. Calendars too were

often misnamed either as Books qf Anioa' or Books on Time or given other
titles, making it very difficult to tell from a work's title alone what type of
text it is.
The information that the calendars contained came from many sources .
The islamic conquerors incorporated the almanacs which they came across
in the east, North Africa and Hispania. The best evidence for this is the
almanac of Sinan, which was used by the muslims of Iraq." The same
process is evident in the Calendar qf Cordoba, which lists some of the days
which the Egyptians regarded as inauspicious, and uses Coptic and Syriac
terms. Calendars similar to the Calendar of Cordoba were compiled
throughout the islamic lands. The Book qf Time of Ibn Musawayh of
Baghdad (d.857t 7 is the first surviving example, but the encyclopedia
compiled by the Ottoman scholar I:IajjI Khalifa in the seventeenth
century'" lists many such works, including some twenty written in the ninth
and tenth centuries. Although I:IajjI Khalifa did not mention either of the
versions of the Calendar qf Cordoba, nor any other Hispanc- arab work of this
type, it is likely that the Books qf Anuui' and related texts were introduced
into al-Andalus soon after their development in the east.t" The earliest
manuscript of such a work from al-Andalus now extant is the Book on the
Stars of 'Abd aI-Malik ibn Habib (d.853):>o It is based partly on a work of
the same name by the eastern scholar Malik ibn Arias (d.795), whose strict
interpretation of religious law was favoured in al-Andalus. Malikite
influence on the computational treatises written in al-Andalus is reflected
in their emphasis on the determination of the exact times of day for
prayers, calculations known as miqdi, or sacred astronomy. According to
later sources, Qasim ibn Asbagh (d.952 ~ possibly one of the translators of
the Arabic Orosius) introduced into Hispania the Book ofAtuod' of Ibn
Qutayba and Abu 'All al-QalI (d.967), who moved to Cordoba in 942,
brought with him a similar work by Ibn Durayd. The Andalusis were also
familiar with the writings of several other authors." It is not clear how
much of this information made its way into the Calendar qf Cordoba. The
complier of the Calendar does not seem to have used a Hispanic christian
model. The christians of al-Andalus preserved their astronomical knowledge
after the conquest and this may have been the only such lore available to
the new rulers until the arrival of astronomical tables from the east in the
reign of 'Abd al-Rahman II . But much of this material came from Isidore
and was of little practical value, although a treatise on astrology copied by
Hisham I's astrologer in the ninth century could be of Visigothic origin. 52
The origins of the Calendar qf Cordoba cannot be pinpointed on
astronomical grounds alone . Sams653 analysed the numerical data in the

119
 RECEMUND AND THE CALENDAR OF CORDOBA

two versions of the Calendar, starting with the hypothesis that the
measurements had been adapted by each compiler for the geographical
coordinates of the place where that version was compiled. He looked at the
values given by each version of the Calendar for the height of the sun at the
meridian, the length of its shadow at this point, the duration of night and
day and of dawn and dusk . Samso found his hypothesis to be untenable.
Some of the data given in both texts correspond to a latitude of 37;30 °,
which is nearer to Seville than Cordoba. The data reproduced in a
fifteenth-century astronomical text which Samso believed to be related to
the Arabic version of the Calendar if Cordoba, correspond to a latitude of 38°,
much closer to the actual position of Cordoba. Yet no definite conclusion
may be drawn from these discrepancies. A manuscript, dated before 1277,
one of more than one hundred surviving examples of the Toledan Tables, a
group of medieval astronomical texts, gives Cordoba the latitude of 37;30°,
but a near-contemporary copy of the same list of geographical coordinates
has 38;30°. Both lists were derived from the work of al-Khwarizrnl, who
introduced the concept of latitude and longitude early in the ninth century.
The Toledan tables suffer from many copying errors, and values quoted for
latitude and longitude continued to be wildly inaccurate during the Middle
Ages." Samso found that some of the information given by the Calendar if
Cordoba was grossly aberrant. He concluded that the two versions of the
Calendar used data from different sources with little concern for its practical
value to the reader." These sources of the Calendar give some clues to
dating the work. The Latin text cites the astronomer Albeteni/" probably
al-Battani (d.929) from the region around Harran in Northern Syria , since
some of the dates in the Calendar coincide with his observations for 882 .57
The Indian astronomical work called the Siddhanta, known to the Arabs as
the Sind Hind is mentioned twice. The astronomical tables compiled by
al-Khwarazmi (d.c.839) using this text were introduced into al-Andalus
shortly after 840 but were not adapted for the islamic calendar and the
meridian of Cordoba until al-Maslamah al-Majritr (of Madrid, d. 1009)
worked on them, sometime after 979, when he carried out a series of
observations/" The Arabic version of the Calendar if Cordoba seems to have
relied on the Sind Hind in al-Khwarizmi's version and could be dated
anywhere between 882 and 980 . Another Andalusian treatise which is
nearly contemporary with the Calendar if Cordoba, the Book ifAnw(i'and Time
of Ibn 'AsIm (d.1013), appears to be a resume of the works of several
eastern sources. 59 It is predominantly a lexicographical text but it includes
a calendar similar to the Calendar if Cordoba, based on the solar months
under their Syriac names, with the christian terminology added 'for those
who are not Arabs' . The two calendars have little in common apart from

120
 RECEMUND AND THE CALENDAR OF CORDOBA

the concluding paragraphs which both authors appe nded for each
month.
From a brief analysis of the material in the Calendar if Cordoba and
Ibn 'Aslm 's Book if Anwa and Time, we might conclude that th ey were
written to be read as a sort of encyclopaedia rather than to be used as
manuals. Their content was affected as mu ch by the sources which an
au thor had to hand as by its valu e to a local audience. Few later compilers
of calendars included that ofIbn 'Asim amongst their sources; his work was
ignored except that it was quoted in a somewhat random fashion by
al-Umawi of Cordoba (d .l 206).60 Nor was the Calendar if Cordoba mu ch
exploited. Al-Umawi and Ibn al-Banna' of Marrakesh (d.132 1) used data
which they may have taken from the Calendar but both also used other
sources. Although th e number of calendars composed in al-Andalus may
be small, each one seems to have been an indi vidual composition,
dependent on a number of different texts from Hispania and the cast. It is
not surprising that there are two different versions of the Calendar if Cordoba.
An swers to the qu estions wh ere, when and why the two versions of the
Calendar wer e made must be sought in a detailed comparison betw een the
two texts .

The Arabic version of the Calendar of Cordoba

The Arabic version of the Calendar has a long introduction which begins :
'This book was created to recall the periods and the seasons of the
year, th e number of its months and days, and the course of the sun
through the signs of the zodiac and the mansions, the limits of its
times of rising and the measure of its declination and elevation, the
varying size of its shadow at the meridian and the periodicity of time
and the succession of days in the waxing and waning of their length
and the cold and hot seasons and the temperate seasons which
separate them and the appointed date of each season and the
number of its days according to the do ctrine of the men of setting
straight and computation and the do ctrine of the first physicians who
determined th e seasons and their characteristics, for there were
disagreements between them on this subj ect which will be pointed
out in their place in this book, if God so wishes . . .'
After a discussion of the constitution of these divisions of the year according
to the practice of pr e-islamic and muslim astronomers, and the value of the
Books if Anwa' for agriculture and medicine ," the author continues:

121
 RECEMUND AND THE CALENDAR OF CORDOBA

'Then we will mention the months of the 'ojam [non-Arabs] and the
signs of the zodiac and mansions which correspond with them, the
number of days in each month, their place in the seasons, their
nature and what is beneficial to do during that month, and what care
every man should take for his comfort and bodily health. The
beginning of the christian year is January and they use the
chronology of the Bronze Age. 62 The beginning of the Syriac
calendar is October and the Syrians use the chronology of Alexander
the Great. But the christians have not made January the beginning
of their chronology because the first day of this month, as they say, is
the seventh after the birth of Christ, peace be upon him, and the day
of his circumcision. I have mentioned in this book all the christian
festivals so that this could be an increase in knowledge and help
towards their meaning. The christians have the festival of Easter,
which they call 'the Resurrection of Christ' and it is preceded by
their fast . .. [explanation of the dating of Lent and Easter]' .63
This version of the Calendar must have been compiled for readers who
were not christians. The entry for December 25 states: 'The christian Feast
of the Nativity of Jesus, and this is one of their greatest festivals' i'" The
entry for April 24 reiterates some of the information given in the
introduction on the calculation of Easter: 'This is the last of the appointed
dates for the christian Easter, the greatest of their festivals. And it is
absolutely not to be delayed past this point'.65 The way in which the
festivals are characterised differs from the Latin version. For May 3,
the Feast of the Cross, the Arabic has : 'and they claim that they found the
Cross of Christ abandoned in Jerusalem', whereas the Latin has: 'On this
day is the christian Feast of the Cross because on this day the Cross of
Christ was found buried in Jerusalem'.66 The author repeatedly used the
muslim salutations, such as 'peace be upon him' . This does not by itself
rule out a christian author, as the christians of al-Andalus used muslim
phrases when writing in Arabic but the frequency with which they are
included in the Arabic version of the Calendar is striking, and most do not
appear in the Latin version.
There must have been practical reasons why the Umayyad adminis-
tration wanted to know when their christian subjects celebrated their
festivals, but integration of the christian and muslim calendars went much
further than this. The muslims of al-Andalus and North Africa used both
the christian and the muslim calendars and celebrated some of the
christian festivals.v' In a passage quoted by al-Maqqari, reporting the
advice given to the people of Hispania by the ninth-century musician

122
 RECEMUND AND THE CALENDAR OF CORDOBA

and style guru Ziryab th at they should change their clothes according to
the season , th e christian festivals are cited as Ziryab's reference points.
A thirteenth-century theologian from Ce uta, al-'Azafi, composed a trea tise
condemning the dan gerous 'innovation' or heresy of celebra ting the
christian festivals. Muslims were joining in with so mu ch enthusiasm that
th ey risked falling into ap ostasy. They took th e oppo rtunity of
suppleme nting their own festivals with holidays borrowed from their
christian neighbours.
The Ar abi c version of th e Calendar could have been compiled by an
Andalu si christian or by a well-informed mu slim, who 'christiani zed ' an
islami c text for a muslim audien ce, perh ap s for th e caliph himself. Yet
th e pro cess of tr an sformation of an islamic calendar to a mu slim-
christian hybrid probabl y began in No rth Africa or the east. C hristian
saints were listed in calenda rs compiled outside H ispani a, such as the
Calendar of Ibn M asawayh .f" The Feast of th e Purification of M ary is
listed in the Ara bic version of th e Calendar but it appears in only one
Hispani c passion ary and is not in th e Latin version of the Calendar. This
impl ies th at th is festival was copied from th e exemplar used by the
compiler of the Arabic version and casts doubts on Dozy's assumption
tha t the Anda lusi comp iler was responsible for adding all the christian
festivals. It also makes it much less likely th at the Calendar was an unique
com mision for al-H akam II .
T he tran smission of the Arabi c version of the Calendar sheds some light
on these qu estion s." The manuscript, in H ebrew charac ters, appea rs to be
written in a single hand except for one brief addition. The codex in which
it was copied begins with a set of astro nomical tables calculated for the
Year 5 111 of the Hebrew World Era [1351] for the city of Huesca. This is
followed by a pharmacop oeia from M ontpelier in a H ebrew translation .
Then comes the text of the Calendar. The manuscript ends with an
incomplete copy of th e ph armacop oeia of M ose ibn Ardut of Huesca,
physician to th e In fante Alfon so of Arago n, who died during a campa ign
against Sardinia in 1323. This codex was copied by or for a fourteenth-
century jewish doctor from Huesca for his own use. Arabic learning
enjoye d high pr estige in jewish communities, in spite of their suspicion of
Islam?" but we canno t assume th at th e physician of Huesca tra nscribed the
Calendar witho ut altera tions. H e may have omitted some of the text as
irrelevant. This alone could explain some of the differences in con ten t
between this version and the Latin Calendar. As we have seen , the Arab ic
version does not list m an y saints' days and the absence of a nu mber of
saints celebra ted in Cordoba is particularly striking. It could even have
been th is compiler who interp olated th e par agraphs explaining the

123
 RECEMUND AND THE CALENDAR OF CORDOBA

inclusion of the christian festivals and outlining the calculation of Easter. It

is impossible to establish how extensively the physician of Huesca
abbreviated or interpolated but the idea that this manuscript is a faithful
copy of a tenth-century original is difficult to defend.
Yet the Arabic version does preserve the name of the tenth-century
author of the text which gave rise to the Huesca copy. It begins: 'Abu
I-Hasan 'Artb ibn Sa'Id the Secretary - may God pardon him and us -
said: .. .' Ibn Sa'td's authorship is confirmed by the colophon: ' End of the
Book of 'Artb On the Division if the Seasons and the Hygiene if the Bo4J'. This
title, yet another variant of those given to calendrical works, is the one
given to the work which Ibn Zaid the Cordoban bishop presented to
al-Hakam 11. 71 Pellat changed the name to Ibn Sa'd although the
manuscript has Ibn Sa'Id, because Dozy had assumed that when
the Arabic sources referred to a historian by the name of Ibn Sa'd , he was
the same as 'Artb ibn Sa'rd. Nowhere in the Arabic text does it say either
that 'Anb ibn Sa'Id was working for al-Hakam II, or that this version of
the Calendar was composed for the caliph. On the other hand, Ibn Sa'd
was cited several times by later historians writing in Arabic as a man of
letters who served as secretary to al-Hakam II . The two names are very
similar, and 'Anb ibn Sa'rd was also a historian. Dozy's edition of Ibn
Idhari includes fragments of a continuation of the history of al-Tabari
which recounts the history of Hispania and North Africa from 912-42.72
The title of this continuation is obscure, but Dozy believed the author to
be 'Arib ibn Sa'td, since, according to another Ibn Sa'td, the Maghrebi,"
'Arib ibn Sa'rd had written a history of this period." Al-Maqqari quoted a
passage from the same work" and described 'Arib as one of the greatest
of all historians." Because he was not identified as a christian, it is
probable that he was a muslim. Dozy believed that 'Arib himself rather
than one of his ancestors had converted to Islam, since none of the
references to him gave his genealogy, " He may be a representative of
the class of muslim bureaucrats of non-Arab origin who were so useful
to the Umayyads because their loyalty was not affected by inter-tribal
rivalries." 'Artb ibn Sa'Id was also a physician and the author of a treatise
on obstetrics, dedicated to al-Hakam, which survives" and a pharmaco-
poeia, which does not. Ibn al-'Awwam, in his Book on Agriculture, made
several references to the author of a calendar, under three different but
similar names. He could be 'Artb ibn Sa'td, although the quotations that
Ibn al-'Awwam extracted from this author do not match the Arabic
version of Calendar if Cordoba exactly'" There is enough evidence for 'Arib
ibn Sa'td to confirm that he was the compiler of a calendar of which the
surviving Arabic version is a copy.

124
 RECEMUND AND THE CALENDAR OF CORDOBA

The Latin version of the Calendar of Cordoba

The Latin version of the Calendar bears the title 'Liber anoe ', where 'Liber'
translates 'Kitab' and 'anoe' is a transliteration of 'Anwa' and opens:
'H ere begins the 'Liber anoc'. In this book is the remembrance of the
years and their days and the returning of the 'anoe' in their days and
the time for planting and the modes of agriculture and the
harmonizing of time and the storing of fruits."
As we have seen in the previous chapter, Hispanic christians commemorated
the annual round of festivals in liturgical calendars and passionaries. There
are no Visigothic texts which combine these festivals with astronomical and
agricultural data. It is not clear why Andalusi christians wanted their
calendars in this form . New agricultural practices introduced into Hispania
by the conquerors, particularly the use of irrigation, may have enhanced
the value of Arabic almanacs for the christian population. There seems
also to have been an attempt to update the saints included, particularly
those local to Cordoba. The Latin Calendar includes the ninth-century
Cordoban martyrs Adulphus and John, who died in 822, and Emila and
Perfectus, who died in 852. It is odd that only two of Eulogius ' martyrs
were included, and that Emila's companion Ierimias was not mentioned.
The calendar also lists Alvarus , and 'Esperende', presumably Alvarus'
teacher Speraindeo, (although neither of these men were martyred) from
the ninth century and Pelagius from the tenth. The entry for 6 January in
the Arabic version has: 'This is the christian Feast of the baptism ofJesus;
they say that on this night a star appeared above him '; the Latin entry for
the same day adds : 'This Feast is celebrated in the monastery of
Pinnamellaria'i'" This monastery, not far from Cordoba, was the home of
some of the ninth-century martyrs. The adaptation of an islamic almanac
as a christian calendar may be an expression of the dominance of Arabic
culture, rather than demonstrating a need for a new form in which the
saints could be commemorated. When the Calendar became a christian text
no attempt was made to hide its Arabic origins . The Latin version is full of
Arabic placenames and expressions." The entry for August 15, the Feast
of the Assumption of Mary, includes the response 'super quam sit salus', a
translation of the muslim salutation 'peace be upon hcr' .84 This is
repeated in a later entry, although no salutation appears at this point in
the Arabic version. The Latin 'villa ex villis' [one of the villas] represents a
typical Arabic locution. The naming of Jerusalem as 'domo almegdis'
translates the Arabic name ' bayt al-maqdis / al-muqaddas'; the translator
explained: 'id estJerusalem' . Alexander, 'habentis duo cornua' ['the possessor of

125
 RECEMUND AND THE CAl.ENDAR OF CORDOBA

two horns', because he ruled both Asia and Europe] is the Arabic name for
Alexander the Great. Sometimes the Latin version makes little sense.
When the translator wrote: 'In this season there are three scorpions in
which it is very cold', he was using 'scorpio' as a translation of the Arabic
"aqanb' [scorpion], which also means a period of nippy weatherf" Details
which gave special problems seem to have been omitted, such as the names
of stations of the moon, and the names given to the types of rain falling at
different seasons of the year. The writer did not seem to realise that 'tisirin
al-aunoai', which he translated as 'tisirin primus' was the Syrian name for
October, although this is noted at the beginning of the entry for that
month. The register of this translation is thus difficult to grasp. The
Arabicisms are sometimes used as though such hybrids were acceptable but
they also suggest that the translator had not mastered his material.
It is difficult to make sense of the Latin text without knowing when it
was translated from Arabic. A work entitled Liber Anoe is listed among the
translations of Gerard of Cremona," The surviving thirteenth-century
manuscript could be a copy of this version. Although it based on a text
which has many references to the churches and cults of Cordoba, the
manuscript cannot have been copied in Cordoba because the scribe made
many mistakes with the topography of the area, giving meaningless
transliterations of the Arabic names of the districts of Cordoba. Since it
appears that the writer did not understand either the Calendar or the Arabic
language very well, the balance of evidence favours a late translation or a
very bad copy by a scribe who had lost contact with the Arabic past of
Hispania.
The Latin version of the Calendar preserves, in a garbled form, the
attribution of authorship to 'Artb ibn Sa'Id. The opening does not translate
the phrase naming the author in the Arabic version exactly, but has 'Harib
filii Zeid episcopi composuit Mustansir imperatori [al-Hakam 11]'. It was the
manipulation of this phrase which allowed Saavedra to claim that the Latin
text of the Calendar was not a copy of that composed by 'Arib ibn Sa'Id, but
was compiled by a second man, Rabr' ibn Zald. Saavedra thought that the
use of the genitive indicated a copyist's lapse, and supplied some missing
words: 'Harib filii (Sad liber, cum additamentis Rabi filii) Zeid, episcopi. . . .'87
which the copyist is said to have omitted. Dozy rejected this hypothesis as
'completely without substance'i'" blaming simple confusion of two similar
names, although he conceded: 'I confess that this difference is what the
English call very puzzling'. There is much simpler explanation. 'ArIb was
correctly transliterated as Harib; Latin translators often used H for the
Arabic letter 'ain. The problem, if there is one, is in the confusion of Zard
with Sa'Id, To Arabists this seemed unlikely, since the orthography of the

126
 RECEMUND AND THE CALENDAR OF CORDOBA

two names is quite distinct. Yet, as we have seen, the Latin text was
distorted by ill-informed translation or bad copying. Some Arabic texts
were first translated into Hebrew, before a Latin version was made. If the
process of translation involved dictation to a writer of Latin who did not
know Arabic, the mistake is easy to explain, since Sa'td and ZaId might
sound the same to someone who did not know how they were spelled.
Equally, the mistake could have been made when one Latin manuscript
was copied from another by a scribe ignorant of Arabic. Elsewhere in this
manuscript there are many similar mistakes, so that al-Hajjaj becomes
Alahazez. Thus, it is almost certain that the Latin version of the Calendar is
a version of the work attributed to 'Anb ibn Sa'Id, although the author's
name was mangled in transmission, and there is no good reason to suppose
that a tenth-century christian was involved in its compilation.

SOUle Ulore calendars

More calendars of the same type as the Calendar if Cordoba have come to
light since Dozy's day. One is now a mere fragment of an Arabic papyrus
which could date to the tenth century'" A manuscript now in the
Escorial," a set of astronomical tables written in Arabic and attributed to
Muhammad ibn Abu Ashshokr the Maghrebi, contains a calendar whose
contents resemble the Latin version of the Calendar if Cordoba." Another
Escorial manuscript is a Book if Anwii' compiled by Abu 'Ali al-Hasan
al-Qurtubi (d. 1205/6) from the Calendar if Cordoba and the work of Ibn
'AsIm.92 An agricultural manual written by Ibn Wiifid in the eleventh
century contains a calendar which is purely agricultural, without any feasts,
either christian or muslim." The most enlightening comparison can be
made with another Latin calendar, now in Vich." It is dated 1235 and
mentions the Feast of Francis of Assisi who died in 1226 and was canonised
in 1228 and is probably Catalan. The calendar from Vich is also based on
an Arabic almanac but it must be independent of the Calendar if Cordoba, as
a different collection of saints has been added. The adaptation of the
model for this calendar seems still to have been in progress, because
although the astronomical and agricultural sections of the manuscript are
carefully written, the liturgical calendar has been both scratched out and
interpolated. The Vich version has even more saints than the Latin version
of the Calendar. January alone has twenty-seven festivals. The translator of
this version produced a text which is even more garbled than the Latin text
of the Calendar by trying to eradicate from the calendar traces of its origins
in Arabic. Although he was forced to used some of the Arabic names for

127
 RECEMUND AND THE CALENDAR OF CORDOBA

astronomical phenomena, he tried as far as possible to avoid Arabicisms

and references to muslims in general, so that, for instance, 'the Arabs say'
becomes 'certain people/the ancients said'. The name of the author does
not appear, but echoes of its dedication to al-Hakam II may be detected.
The codex containing the calendar is entitled Liber regius sive descriptio
temporum anni and begins: 'Here begins the Royal Book. The author of this
book says . . .' A fourteenth-century copy of the same work" has the
colophon: 'explicit Liber Egregius', surely a copyist's mistake for regius. It is
not possible to say whether the Vich calendar was a translation of a tenth-
century Arabic text, or whether it had evolved over several generations but
it belongs to the same family of texts as the Arabic and Latin versions of the
Calendar if Cordoba. Discussions of the Calendar should be widened beyond
the two texts edited by Dozy and Pellat ,

Recemund and Rabie ibn Zaid

How did Recemund's name come to be linked with the Calendar if Cordoba?
In 1871, Simonet published a Spanish translation of the liturgical entries
from the Calendar if Cordoba with the title: 'the Hispano-mozarabic
Calendar written in 961 by Rabi ben Zaid, bishop of Iliberis [Elvira]',96
claiming that: 'Recemund of Iliberis, mentioned by several contemporary
foreign writers, is none other than the bishop Rabi ben Zaid, celebrated by
the Arab authors for his astronomical knowledge and his journeys'.
Simonet identified Recemund with two bishops mentioned in the Arabic
histories of al-Andalus. The first was a certain Rabr' the bishop , said to
have brought from Constantinople an expensive enamelled basin for a
fountain, to ornament the palace which 'Abd al-Rahman built at Madinat
al-Zahra'. The second was the bishop Ibn Zaid who gave a calendar to
al-Hakam II . Although Dozy pointed out some of the problems of the
Arabic sources for Recemund, he accepted Simonet's hypothesis . Both
Dozy and subsequent historians skated over the vagaries of survival of
sources , and, quite unjustifiably, gave equal weight to such remnants as
supported their case. I would now like to unpack this historiographical
portmanteau. Who was, or were, Rabi' and Ibn Zaid, and how are they
connected with the Calendar if Cordoba?
Simonet postulated that sometime before the death of'Abd ar-Rahman
in 961 Recemund went on another mission for the caliph, to Jerusalem and
Constantinople. His journey could have coincided with Liudprand's
second embassy to the Byzantine capital in 960 , thus perhaps explaining
why Liudprand wrote so much about Constantinople in the Antapodosis.

128
 RECEMUND AND THE CALENDAR OF CORDOBA

Accepting for one moment the identification between Recemund and

Rabi", the Arabic sources preserve the story that he brought back a gift for
the emir:
'Ibn Hayyan said that among the marvels of al-Zahra' were two
fountains with their basins, so extraordinary in their form and
valuable in their workmanship, that in the opinion of that author
they were the principal ornament of the palace. The more precious
of the two was of gilded bronze and it was marvellously sculpted
with bas-relief representing human figures. . . . Concerning the
remarkable, expensive, engraved and gilded basin: Ahmad the Greek
brought it to him, together with Rabi' the bishop, coming from Iliya.
And as for the small green basin engraved with human images,
Ahmad brought it from Syria or, it is said, from Constantinople with
Rabr' the bishop and they said that it was unrivalled in singularity
and beauty. And it was carried from place to place until it reached
the sea. Al-Nasir ['Abd al-Rahman III] set it up in his sleeping
quarters in the eastern chamber called al-Munis and placed in it
twelve figures made of red gold , set with pearls and other precious
stones'."
This is part of a long quotation on the building of the palace of Madinat
al-Zahra' which al-Maqqari said he had taken from the work of
Ibn Hayyan. Citing the authorities for Ibn Hayyan's account, al-Maqqari
stated that: 'This historian obtained his information from the mouth of
Ibn Dahin the jurist, who obtained it from Muslamah ibn 'Abd Allah, the
teacher and architect'. The latter was said to have flourished during the
reign of 'Abd al-Rahman III. The three names make rather a short chain
of witnesses to cover nearly a century. The passage describes the materials
brought from all corners of the world, their great beauty, rarity and cost,
and the number of workmen involved. It is not clear from the syntax
whether Ahmad the Greek was responsible for acquiring both the basins,
nor whether Rabr' the bishop actually went to Constantinople rather than
meeting Ahmad on his return journey from the east, but the bishop could
have been acting for 'Abd al-Rahman, scouring the world for treasures to
adorn his new palace.
Madinat al-Zahra' was abandoned in 1008 and fell into ruins. As we
have seen, the sources for Madinat-al Zahra'?" differ over the site of the
palace, the date of its construction, its size, the number of men and beasts
involved in the work, and the materials used . The archaeological remains
of Madin at-aI Zahra' which have so far been excavated, impressive as they
are , do not substantiate all the claims made for it. Marble for the palace

129
 RECEMUND AND THE CALENDAR OF CORDOBA

was said to have come from Africa, Byzantium and the land of the Franks,
yet analysis of the marble used to pave the hall of 'Abd al-Rahrnan III
reveals that it was quarried in Estremoz, Portugal." Ibn Hayyan gave 329
[940/1] as the year in which construction began, but all the other accounts
have 325 [936]. This alone casts grave doubts over the rest ofIbn Hayyan's
evidence. The histories recorded several versions of story of the basin. The
earliest sources mentioned a fountain, thought to have been a present from
a foreign ruler. Later writers proposed either Jerusalem or Byzantium as
the origin of this fountain, which naturally gave rise to the idea that there
were two - one was said to be green, and the other made of gold. On the
other hand, many descriptions of Madinat-al-Zahra' failed to mention a
fountain and the chamber in which it was said to have been placed is not
otherwise known to have existed . Thus it is difficult to give much credence
to what this text said about Rabi' the bishop, other than to say that he was
worthy of being remembered as a christian who might have been involved
in the construction of Madinat al-Zahra'.
The evidence for Ibn Zafd the bishop is better than that for RabI'.
Further on in the passage about the calendar which Ibn Zaid dedicated to
al-Hakam II, al-Maqqari named Ibn Zaid as one of the famous authors of
al-Andalus. Each man was listed under the branch of scholarship in which
he was most prominent. Al-Maqqart's source was Ibn Sa'td the Maghrebi,
and al-Maqqari included Ibn Sa'rd's biography and quoted many pages of
his work .l'" Born in 1209 or 1214 in Alcala la Real, near Granada,
Ibn Sa'Id travelled in the east, before entering the service of the emir of
Tunis. According to some authors, he wrote more than four hundred
books , but he was principally remembered for the Book of the Sphere of
Literature encompassing the Language cf the Arabs. The first fifteen volumes of this
work were said to recount the history of the western islamic realms from
1152 to 1263, and the second half concerned the east . Only fragments
survive, although Ibn Sa'Id's history was much quoted. His resume of the
literary celebrities of Hispania is taken from an appendix which he added
to a Risala [Letter] written by Ibn Hazm (d. I 064).101 Ibn Hazm's message
was that the scholars of al-Andalus were equal with those of the east .
Ibn Sa'Id added weight to this argument with his own list. Although the list
is unusual in the attention which Ibn Sa'Id paid to non-muslims, this does
not guarantee the accuracy of his information about Ibn ZaId .
Al-Maqqari's account of Ibn Zaid continues: 'And it was in furthest
Ishbtla [Seville] that he worked on his writings, and the people of his
country accused him of atheism because of his devotion to this matter and
none of his writings were published'. Simonet carefully passed over the
accusation of heresy by omitting this part of the reference to Ibn Zaid, a

130
 RECEMUND AND THE CALENDAR OF CORDOBA

practice which, so far as I can tell, has been followed in all subsequent
discussions of this passage. Astronomy and astrology were not clearly
distinguished in the Middle Ages but both christians and muslims
recognised the difference between observation of the heavens made in
order to regulate the calendar and using the positions of the stars to predict
men's fate .l'" In the early centuries of Islam, astronomy and astrology were
considered part of the same body of knowledge and the movements of the
stars and planets gave insight both into the will of God and earthly events. 103
In al-Andalus, according to Sa'Id the Andalusi (1029-70), 'the practice of
astrology has met with some acceptance, both in the past and the present:
there were some well-known astrologers in every period, including our
own'.104 Yet books on astronomy were amongst those condemned as
heretical when the library of al-Hakam was broken up by al-Mansur at the
end of the tenth century. Christian authorities were even more suspicious of
astrology. The combination of patristic disapproval and the loss of Greek
astrological knowledge in the west meant that there was little christian
interest in astrology until the tenth century, further stimulated by translation
of texts from Arabic in the twelfth. The suspicion that Gerbert of Aurillac,
later pope Sylvester II (999-1003) had been to Cordoba to learn astrology,
which first appeared in the work of William of Malmesbury, was enough to
give him the reputation for necromancy. In fact, he probably travelled no
further than Ripoll in christian Catalonia. Here perhaps is the substance
behind the charge of heresy laid against Ibn Zatd. Ibn Zard was
remembered as involved in a science which was controversial even for
muslims and would not have been considered suitable for a bishop.
Although the sources for Ibn Zaid are late, it is plausible that such a
figure existed . The chain of argument which makes one person out of the
three bishops Rabt', Ibn Zaid and Recemund, however, has several links
which range from tenuous to preposterous. One reference to a bishop called
Rab!' ibn Zaid has been found in the Arabic sources. The thirteenth-
century author Ibn Abi Usaybi'a mentioned him in his biographical
dictionary of medical men, in the entry on Ibn al-Kattani who died in
Zaragoza in 1029.105 Ibn al-Kattani was the pupil of the bishop 'Abu
al-Harith, who in turn was a pupil of Rabr' ibn Zaid, 'the bishop, the
philosopher'. Ibn al-Kattani compiled a Book if Anwa' which survives in a
manuscript which may date to the eleventh century. 106 Its content is similar
to that of the Calendar if Cordoba. Yet Ibn al-Kauani did not cite either 'Artb
ibn Sa'Id or Rabi' ibn Zaid but gave other authorities instead, including Ibn
Qutayba of Baghdad and Abu Nuwas, the court poet of Hartin al-Rashid.
Two generations before Ibn al-Kattani, Rabr' ibn ZaId could have been
active at the court of al-Hakam II. He is certainly pr eferable to the second

131
 RECEMUND AND THE CALENDAR OF CORDOBA

candidate. Dozy suggested that Rabl' ibn Zaid was Aben Cayut, whom
Alfonso the Wise mentioned in his Books onAstronomy. 107 The prologue to the
Alfonsine Tables begins: 'Yhuda son of Marc son of Mosca and Rabicag
Aben Cayut said . . .' 108 but this Rabicag Aben Cayut was active in the
thirteenth century. In the prologue to the Book ifthe Clock called the Shadow if
the Stone, Alfonso stated: 'we ordered the aforementioned Rabicag to make
this book .. .'109 and Rabicag is mentioned several times in the same
context. He has been identified as a rabbi from Toledo, Isaac ben Sid.11O
The identification ofIbn Zaid, or perhaps Rabie ibn Zaid, as the author
of the Calendar if Cordoba is the most beguiling aspect of the nineteenth-
century interpretation of this work and the most difficult to disentangle.
The title of the work attributed to Ibn Zaid, the Book if the divisions if the
Seasons and the Hygiene if the Body is the same as the title given in the
colophon to the Arabic version of the Calendar. Both al-Maqqari and the
Latin version of the Calendar said that the work was given to al-Hakam II.
But in order to make Ibn Zaid the author of the calendar dedicated to
al-Hakam, it was necessary either to explain away the name 'Arib ibn Sa'Id
- which Saavedra tried, and failed, to do - or to accept that this solution
did not tie up all the loose ends. We are left with two men - 'Arib ibn Sa'Id,
the author of a calendar, and Ibn Zaid the bishop who presented a
calendar to the caliph. As Pellat acknowledged:
'The statement ofIbn Sa'Id [in al-Maqqart] and the colophon of the
MS are utterly contradictory. It has to be assumed therefore that an
understandable confusion has arisen between the names (which are,
it may be observed, anagrams one of the other) and that the blending
is so perfect that, towards the end of the introduction, a paragraph
relating to the christian festivals gives the impression that the work is
attributable to a single author. As for the information concerning the
agricultural activities, hygiene, daily life, etc., so precious in the view
of historians, it is not unreasonable to give the credit to 'Arib rather
than to Rabf", since the former was apparently more apt to respect
the tradition of Kutub al-Atuod', which themselves contain facts of this
type as well as material concerning astronomy and meteorology. In
view of the fact that a Kitiib Tqfti1 al-Aeman [Book if the Division if the
Seasons] etc., evidently as a result of confusion, is attributed to each of
the two authors, the problem remains unsolved '. III
Dozy had proposed that when the Calendar was translated into Latin the
name of Rabie ibn Zald was mistakenly written instead of 'Arib ibn Sa'Id
because the translators knew that both men had compiled calendars.
Recently, Van Koningsveld turned this argument on its head, arguing that

132
 RECEMUND AND THE CALENDAR OF CORDOBA

the Calendar if Cordoba must be the same as the work mentioned by

al-Maqqari; therefore Rabie ibn Zaid is its author, and the name 'Anb ibn
Sar'd is the result of the corruption of the text by a late copyist, influenced
by the greater fame of this author. I 12 There seems little reason to accept
either solution. Since there were at least two texts of the Calendar if Cordoba,
as well as other similar but independent calendars, it is not surprising that
the sources name more than one author of calendars. From the scant
biographical details which survive for 'Anb ibn Sa'Id and Ibn Zaid, it is
reasonable to suppose that both men composed calendars. Finally, given
the fame of al-Hakam's library,'!" it is likely that any book written, or
thought to have been written, in al-Andalus at this period would have been
validated by a dedication to the caliph.
There is one more link in the chain, the final piece of evidence which
Dozy and Simonet used to identify Rabi' ibn Zaid with Recemund. A
bishop called Rabi' is mentioned in a statement attributed to Ibn Khaldnn:
'Then came the embassy from the king of the Slavs, who at that time was
Otto, and 'Abd al-Rahman sent with the ambassador the bishop Rabie to
King Otto and he came back after two years' .114 This evidence, written at
least three centuries after the event and preserved by al-Maqqari,
contradicts the Lift ifJohn if Gorze, which says that John did not travel to
the Ottonian court with Recemund but remained in Cordoba. Ibn
Khaldun is not a very accurate witness to the history of al-Andalus, as we
shall see in the next chapter. This bishop sent to Otto could be Recemund,
but is difficult to share Dozy's certainty that 'this proves the identity of
Recemund with Rabt' because it fits him alone;'! " Other bishops played
a similar role as ambassadors to christian courts . Al-Maqqan mentioned a
bishop of Cordoba, Asbagh ibn Allah ibn Nabit, whom al-Hakam sent to
advise Ordoiio II of the Asturias on the implementation of a new peace
treaty between the two kingdoms.'!" The same man, now described as
'judge [qadi] of the christians' served as interpreter for a delegation sent to
Cordoba by Queen Elvira of Leon in 973. 117 The Rabie of Ibn Khaldun
could with equal plausibility be the bishop who joined 'Abd al-Rahman's
first delegation; the Lift ifJohn if Goree says that this bishop died at Otto's
court. 118 The statement by Ibn Khaldun, the closest we have to an
identification of Recemund with Rabie, must be put to one side, together
with the reference to Rabie the bishop and the enamelled basin . The
sources for Rabi' and Ibn Zaid merely serve to list a number of enterprises
for which christian bishops were remembered. There is no reason to
conclude that all these talents were embodied in Recemund.
'Artb ibn Sa'Id, a contemporary of al-Hakam II, compiled a calendar
which was later translated and interpolated by both christians and jews,

133
 RECEMUND AND THE CALENDAR OF CORDOBA

together with other works of the same genre. The existence of these texts
demonstrates the vitality of cross-cultural exchange in the tenth century
and later. Recemund could have been the author of such a calendar. He
was prominent in ecclesiastical and diplomatic circles both in al-Andalus
and in Europe, perhaps even in Byzantium. To pin all the evidence
reviewed in this chapter on Recemund is to underestimate the prominence
of educated christians in al-Andalus. Although no-one apart from John of
Gorze's biographer and Liudprand paid them much attention, the
achievements of Recemund and the bishop or bishops Rabt' and Ibn
Zaid suggest that some christians flourished under islamic rule. Bicultural
as well as bilingual, they adopted Arabic language and forms without
abandoning their christian faith or Latin literary heritage.

134
 CHAPTER SEVEN
-_4_ _-

The Arabic translation of Orosius

'In the year 337 [948/9], I believe, Armanius [Romanus], the ruler
of Constantinople, exchanged letters with him ['Abd al-Rahman III] .
He also sent him a large quantity of gifts, including the book of
Dioscorides on plants [the Materia Medica], with remarkable
illustrations in the style of ROm [Byzantium] . This book was written
in Greek, which is the same as Ionian. He sent with it the book of
Orosius, master storyteller, an admirable history of Rom, in which
are notices about the epochs, stories of the first kings and important
moral lessons. Armanlus wrote to an-Nasir that he would not be able
to profit from the remedies described in the book of Dioscorides
unless someone was able to put the Greek language into a better
form. For if there is someone in your land who is capable of this, you
will know the book's usefulness, 0 King. And as for the book of
Orosius, if there are among the Latins [Christians] of your country
those who know the Latin language, find them and let them translate
it into Arabic for YOU' .I

1 f Recemund ever received his copy of the Antapodosis, the Latin history
dedicated to him by Liudprand, it is not obvious where he would have
shelved it. There is no surviving history in Latin written in al-Andalus after
the two eighth-century chronicles, and no proof that these chronicles were
read in al-Andalus in the tenth century. But the Andalusis read the work of
one of Hispania's most famous historians, Orosius, although perhaps only
in Arabic. TIe Seven Books if Histories Against the Pagans, written in 417, is a
chronicle of the world from biblical to christian times in the tradition
of Eusebius. Orosius, a pupil of Augustine, tried to show the working out
of God's purpose in the history of the world, and the parallel fates of the
Roman and christian empires. The period beginning with the birth of
Christ was to be the last, triumphant phase of history? As he recounted one
disaster and persecution after another, Orosius found the task which he

135
 THE ARABIC TRANSLATION OF OROSIUS

had set himself increasingly difficult and the Histories end with the sack of
Rome by the Goths. In spite of this apparent failure to stick to his message,
the Histories became so popular during the Middle Ages that 'Orosius'
became a synonym for 'history'. 3 It is not obvious why the Histories were so
popular. Perhaps later readers accepted Orosius' version of events in spite
of its contradictions. His summary may have saved them the trouble of
ploughing through the works of drier historians. The text was widely
transmitted and more than two hundred copies survive." Orosius was
frequently quoted and translated. His work was turned into a very free
Anglo-Saxon version for King Alfred in the 890s. 5 For both the West
Saxons and the christians of al-Andalus, Orosius' Histories may have
represented a link between earliest antiquity and their own times." The
translation of Orosius' Histories from Latin into Arabic also bridged the
divide between the christians under islamic rule and the pre-islamic history
of Hispania. This chapter considers some of the problems of the Arabic
translation of Orosius in order to show the complexity of that bridge. For
the sake of clarity, I will refer to the Latin versions as the Histories and to the
Arabic translation by the Arabic rendering of Orosius - 'Urns/fls.
One manuscript of an Arabic translation of the Histories survives in the
library of Columbia University, New York." The manuscript, which will be
considered in more detail later, is incomplete. It seems originally to have
comprised a translation of the Histories with a continuation which brought
the history of Hispania up to the islamic conquest. Ibn AbI Usaybi'a
explained how Orosius' Histories came to Hispania and how the work was
translated. Ibn Khaldun quoted Orosius several times and made two brief
references to the translators." Yet the connection between the Columbia
manuscript and the references to the translation of the Histories in the
works of Ibn AbI Usaybi'a and Ibn Khaldun is not as obvious as it might
at first appear. Attempts to tie the pieces of evidence together have
occupied several scholars since the Italian Arabist Levi Della Vida wrote
on the' Urns/us in the 1950s.9 As this chapter will demonstrate, none of the
solutions they have proposed is entirely satisfactory. Yet it is the wider
question raised by the context for the translation, rather than the minutiae
of the problems so far worked over that make the 'Urns/us so interesting.

The Byzantine gift and the translators

When the Arabic historians looked back on the glories of Umayyad

Cordoba they made special mention of its libraries. Even a poor school
teacher travelled to the east in search of books'? and the biographical

136
 THE ARABIC TRANSLATION OF OROSI US

dictionary of Ibn Bashkuwal named three women who had famous

libraries. II Al-MaqqarI told the story of a bibliophile who failed to buy a
particularly fine volume in Cordoba's book market, outbid by a social
climber who did not know what the book was about but wanted it to fill
a gap in his new library which was exactly the same size.12 It was one of
the duties of islamic rulers to promote learning. According to tradition, the
chief patron of book collecting in Hispania was al-Hakam II. His father
'Abd al-Rahman III reigned for forty-nine years and, perhaps as an under-
employed crown prince, al-Hakam became a patron of the arts . A century
later Ibn Hayyan said:
'There was no caliph in all Islam to match al-Hakam in the
acquisition of books and poetry and his affection for them and the
importance he attached to them. He was a patron of sciences and
commended them to his people and they responded willingly. And
his gifts established links betwe en him and the scholars in the farthest
capitals'. 13
According to al-Hakam's librarian, the library's catalogue of titles covered
forty-four quires of twenty pages each." There were reputed to be 400,000
volumes . This large number was that judged appropriate for an important
library. The wazir Zuhayr (d. 1038), ruler of the small kingdom of Almeria,
was also said to have possessed 400,000 books. A catalogue from the
library of the Dar al-Hikma in Cairo compiled in 1045 lists a more modest
6,500 books ." Wasserstein deduced the titles of some fifty books which
al-Hakarn's library contained. The only surviving volum e which can be
traced back to Cordoba is a legal text discovered in Fes in 1934. Most of
the other titles in Wasserstein's list came from a description of the library
which Ibn al- Faradi (d. 1013) included in his History if the Scholars if
al-Andalus, although he was almost certainly writing after the contents of
the library had been dispersed." It is al-Hakam who may have sponsored
the translation of Orosius' Histories into Arabic.
Ibn AbI Usaybi'a's account of how a copy of the Histories came from
Byzantium to Cordoba comes from his Dictionary if Physicians, in an entry
on a native of al-Andalus active in the last quarter of the tenth century,
called Abu Daud Sulayman ibn Hassan, also known as IbnJuljul. IbnJuljul
was the author of a work based on the Materia Medica of Dioscurides which
was part of the Byzantine gift. Ibn AbI Usaybi'a quoted IbnJuljul as saying:
'The Book of Dioscurides was translated in the City of Peace
(Baghdad] during the Abbasid dynasty, in the days of Ya'far
al-Mutawakkil [caliph from 847-861]. Its translator was Istafan ibn

137
 THE ARABIC TRANSLATION OF OROSIUS

Basil, the interpreter from the Greek language into the Arabic
language. Hunain ibn Ishaq the translator scrutinized this translation
and corrected and authorized it, because Istafan had explained
in Arabic those Greek names [of plants] of which he did not know an
equivalent in the Arabic language, in their [original] Greek name.
[He did so] trusting that God would send someone after him who
would know those and explain them in the Arabic language . ..
Ibn Juljul said: And this book [the Arabic Dioscurides] entered
al-Andalus in the translation of Istafan, containing both the plant-
names known by him in Arabic and those unknown by him [in their
original Greek name]. And people, both in the Orient and in
al-Andalus profited from the disclosed part of it until the days of
al-Nasir 'Abd al-Rahman ibn Muhammad ['Abd al-Rahman III],
who at the time was the Lord of al-Andalus'.!'
IbnJuljul then explained how another copy of Dioscurides, in Greek, came
to al-Andalus, together with the Histories, and the significance of the new
text of the Materia Medica for Andalusi scholars. Unfortunately, IbnJuljul, or
Ibn AbI Usaybr'a, gave such a confused account of the Byzantine embassy
that it is difficult to date it, or to be sure that it was the Byzantine copy of
the Histories which gave rise to the Urasms. Apart from the inherent
improbablity of the biographer's being able to quote his subject at a
distance of three centuries, it is characteristic of the biographical
dictionaries that Ibn Juljul reported the Byzantine emperor's message to
'Abd al-Rahman III verbatim whilst giving a date for the mission which is
impossible - unless this is a copyist's mistake . Romanus, emperor (920-44),
could have sent the books to 'Abd al-Rahman, but the emperor who sent
ambassadors to Cordoba in the late 940s, possibly in 945/6 and again in
947 was Constantine VII Porphyrogenitus (944- 59).18
Most of the commentators on the UnlsfiIs paid little attention to the rest
C

of this passage . As Van Koningsveld pointed out, it should be read in the

context of the difficulties which physicians had experienced in using
Istafan's defective translation of Dioscurides. IbnJuljul went on to describe
the problem of translating from the new copy of the Materia Medica from
Greek into Arabic: 'when al-Nasir answered Romanus the King he asked
him to send to him a man who could speak both Greek and Latin to
instruct some of his servants, so that they would be able to act as
translators. Thereupon Romanus the King sent to al-Nasir a monk called
Nicholas, who arrived in Cordoba in the year 340 [951/ 2]' . Nicholas
worked with a group of scholars, including Ibn Juljul himself, to produce a
reliable translation. The Histories were not mentioned again, and it seems

138
 THE ARABIC TRANSLATION OF OROSIUS

that they were introduced to make the point that a Latin text was easier to
deal with because there were christians in al-Andalus who knew both
Arabic and Latin, even if they did not know Greek."
Although there is no evidence that he himself knew Latin, Ibn juljul
acted as an intermediary between Latin and Arabic scholarship.i" His
Dictionary survives in one incomplete manuscript dated 993 AH [1585V'
The passage quoted by Ibn AbI Usaybi'a does not appear in this
manuscript, and may have come from an autobiographical colofon to the
dictionary, now lost. Ibn juljul mentioned several christian physicians,
whom he said were the most eminent medical men in al-Andalus until the
middle of the ninth century, when medical texts arrived from the east, and
christian learning became obsolete." Ibn juljul listed jerome and Isidore
among his sources, and may perhaps have known something of christian
history. His insights into the linguistic problems of Andalusi scholarship,
and the switch from Latin to Arabic are fascinating, but his reference to
the arrival of Orosius's Histories in Cordoba should not be taken to mean
that this was the only copy of the Histories available in al-Andalus, and
therefore the text from which the 'Urastas was translated. At least one
copy of the Histories in Visigothic script survived in the peninsula.P The
catalogue of the library of San Zoylus in Cordoba listed one." Further, it
is unlikely that a Byzantine exemplar would have contained the
Visigothic material which was added to the Histories in the Columbia
manuscript.
Ibn Khaldun is the sole authority for the statement that the Histories
were translated for al-Hakam. Ibn Khaldun cited "Urusius the historian of
Rum, in his book which was translated for al-Hakam al-Mustansir of the
Banu Umayya by the judge of the christians and their translator, and
Qasim ibn Asbagh'v" Later in the same work, in an attempt clear up an
ambiguity, Ibn Khaldun said: 'the report of Orosius is preferable, because
its writers were two muslims who translated it for the caliphs in Cordoba,
and these two were well-known and they compiled the book'." Although
Ibn Khaldun referred to an embassy from Constantinople in 336 [947/8:
Ibn AbI Usabt'a said 337] and correctly identified the emperor involved, he
did not link the translation with the Byzantine gift. These two pieces of
information were put together only in the twentieth century. It is Ibn
Khaldun's brief and apparently contradictory references to the translators
of the' Urastas which have given modern students their most difficult puzzle.
A celebrated legal scholar called Qasim ibn Asbagh al-Bayyani was
mentioned several times by Ibn al-Faradt." Later authors also referred to
him, although the details of their accounts are different. 28 They did not
connect Qasim ibn Asbagh with the 'Urasius nor with translation in general.

139
 THE ARABIC TRANSLATION OF OROSIUS

According to Ibn al-Faradi, Qasim ibn Asbagh was born in 245 [859] and
was tutor to al-Hakam before his accession . He died in 341 [952-3]. By
the date of the embassy from Constantinople, Qasim ibn Asbagh was more
than ninety years old and reported by two authors to be suffering from
senile dementia." so it is most unlikely that he was involved in the
translation of the Byzantine copy of the Histories. Modern commentators
have dealt with this problem in an number of different ways, all of which
seem to involve arbitrary decisions to accept some parts of the evidence
while discounting others. Their approaches divide into two groups
depending on acceptance or rejection of the name Qasim ibn Asbagh . If
Qasim ibn Asbagh was involved, al-Hakam must have commissioned the
translation before 948 -9, so the account of the gift from the Byzantine
emperor is a red herring. The alternative is to look for another translator.
This has been done by scouring the biographical dictionaries and accounts
of later historians. Badawi put forward two candidates." The first was a
grandson of Qasim ibn Asbagh called Qasim ibn Muhammad (d.998), a
governor of Tudmir whom Ibn al-Faradi described as a literary man of
good character and mild temper." The second is another man mentioned
by Ibn Khaldnn, Asbagh ibn 'Abd-Allah ibn NabII al-jathltq." Badawi
suggested that Ibn Khaldun wrote his name as Qasim ibn Asbagh because
he was in a hurry and the name of his famous predecessor sprang to mind,
although Badawi concluded by rejecting all three names. Kuhayla put
forward two more names without being any more convincing, contending
that the famous historian mixed up the names because he was old and
forgetful. 33 None of the alternative candidates were noted translators.
It is even more difficult to establish the identity of the second translator.
Ibn Khaldun identified him only as 'the judge and translator [ta~iuman] of
the christians'. The word taijuman meant rather more than translator, for
such men also acted as interpreters of the differences betw een the laws and
customs of the two communities. The term is also used to describe the
ambassador Kartiyus the Greek whom Theophilus sent to 'Abd al-Rahman
III in 839 . Kartiyus was a christian'? but it is possible that christian
converts to Islam also carried out the various functions of a tasjumdn.
Simonet proposed two names for this second translator. 35 Hafs ibn Albar
al-Quti, who made an Arabic translation of the Psalms, is probably too
early, as we shall see, but the second, WalId ibn ]ayzuran, also known as
Ibn Mughith, was a contemporary of al-Hakam II. Hafs ibn Albar's
christian origin is indicated by his patronymic 'son of Alvarus' and his
nickname 'the Goth'. Ibn Mughith was also a christian although his name
is not obviously so. Ibn Hayyan listed four men who served al-Hakam as
interpreters of a deputation from Northern Hispania, and if he had not

140
 THE ARABIC TRANSLATION OF OROSlUS

labelled one of them as a judge of the christians, the others a bishop, an

archbishop and a count [qllmis], it would have been impossible to tell, from
their impeccably Arabic names, that they were christians." Thus the first
translator, if he was not Qasim ibn Asbagh, could also have been a
christian.
Yet, according to Ibn Khaldun both the translators were muslims,
which seems to rule out both Hafs ibn Albar and Ibn Mughith. Both
Kuhayla and Van Koningsveld argued that Ibn Khaldun was simply
wrong, since the office of judge of the christians' was reserved for
christians. Van Koningsveld proposed a neat solution to the problem which
hinges on the omission of the word 'and' from two manuscripts of the
History of Ibn Khaldun, one apparently corrected in his own hand." The
Arabic word 'and' is a single-letter prefix, easily omitted, but if the correct
text of the passage reads : Orosius 'was translated for al-Hakam
al-Mustansir the Umayyad, by the judge of the christians and their
interpreter in Cordoba, Qasim ibn Asbagh . ..' this amalgamates the two
translators into one, and the confusion is reduced. Van Koningsveld
thought that he was a different Qasim ibn Asbagh - not the famous
muslim scholar, but a christian judge and translator. By describing the
translators as muslims, Ibn Khaldun supposed that this is how the text of a
Latin author became available to an Arabic-speaking audience, although,
as we have seen, IbnJuljul expected christians to translate it. Ibn Khaldun,
or later copyists, may also have been misled into thinking that the post of
judge of the christians would have been held by a muslim but no t by the
famous scholar Qasim ibn Asbagh, thus turning the translator into two
people. It seems that little can be salvaged from Ibn Khaldun's notices.
One must conclude that the inconsistencies in the accounts of Ibn Juljul
and Ibn Khaldun are such that they cast doubts on the assumption that the
'Urastas which survives in the Columbia manuscript is a translation made
for al-Hakam from the Byzantine gift.
This conclusion suggests that the terms of the enquiry should be
widened. The 'Urusius is not the only translation of a christian history said
to have been commissioned for al-Hakam's library. In his book Meadoios if
Gold andPrecious Stones, the historian al-Mas'udt, who spent much of his life
in Baghdad and died at Fustat [Old Cairo] in 956 , gave a list of the
Frankish kings from Clovis to Louis IV Al-Mas'udi said he had read the list
at Fustat in an Arabic translation made for al-Hakam in 939 of a History if
the Franks by Bishop Godmar of Gerona.l" Godmar, or Gondemar, II was
bishop of Gerona from 943 to 951/2 and wrote a Chronicle ifthe kings if the
Franks which has not survived.l? Al-Mas'udt's use of it lends little credence
to his statement. His genealogy is a collection of garbled names. There are

141
 THE ARABIC TRANSLATION OF OROSIUS

only six kings between QJawda [?Clovis (cA81 ~ c. 5 1 1 ) ], 'the first christian
king' , and Qarla [Charlemagne (768~814)] , whose reign is truncated to
twenty-six years. Yet the passage may be based on a christian source, and it
is unlikely that al-Mas'udi read this text in Latin. Al-Mas'udt gave the
translation of Godmar's history a context by mentioning eastern
translations of non-muslim works into Arabic.t" The History if Franks, or
something similar, was known by muslim historians in the east, some of
whom, such as Ibn al-Athir, were well informed about Frankish history.
If the 'Urnsfils and the History if the Franks were indeed added to
al-Hakam's library, they may not remained there for very long. The reign
of al-Hakam's successor Hisham II was dominated by his vizir al-Mansur.
In a new climate of ultraorthodoxy, al-Mansur allowed religious scholars to
remove and burn much of the library's contents. The books which
offended them were works of what were described as 'the ancient sciences',
such as philosophy and astronomy, derived from the Greek heritage. It is
not known whether texts of christian origin also perished at the same time.
In the civil wars of the early eleventh century, Cordoba was threatened by
the Berbers and, in 1011, the minister Wadih is said to have sold the major
part of the library to raise money; what remained was seized by the
Berbers." The rulers of the kingdoms such as Zaragoza, Granada, Almeria
and Toledo which emerged from the collapse of the Umayyad caliphate
established important libraries but there is no further reference to
translations from Latin, nor to the possession of christian histories.

Orosius in Arabic scholarship

Yet an acquaintance with Orosius' Histories, or at least a part of it, survived

in the Arabic tradition. Citations from the work by several later writers
show how it fitted into Arabic scholarship and they may also help to date
its translation. With the exception of Ibn Khaldun all the Arabic scholars
who used the 'Urnsfils were highly selective. They copied only the first two
passages , on the geography of Hispania and its division into two parts:
'Spain, taken altogether, by its natural contour is a triangle, and is
almost made an island by the surrounding ocean and Tyrrhenian
Sea. Its first corner, looking towards the east, pressed in on the right
by the province of Aquitania, on the left by the Balearic Sea, is
inserted within the territory of the Narbonnese. The second corner
extends towards the northwest, where Brigantia, a city of Gallaecia,
is located and raises its towering lighthouse, one of the few

142
 THE ARABIC TRANSLATION OF OROSIUS

memorable structures, towards the watchtower of Britain. Its third

corner is where the Gades Islands, facing the southwest, look upon
the Atlas Mountains with the gulf of the ocean intervening.
The Pyrenean forest pastures form the boundary of Hither Spain,
beginning on the east and extending to the northern side to the
Cantabri and Astures and thence through the Vaccaei and the
Oretani, whom it has to the west; Carthage [Cartagena], situated on
the shore of our sea, fixes the boundary. Further Spain has the
Vaccaei, Ce1tiberi and Oretani on the east; on the north, the ocean;
on the west, the ocean; and on the south the strait of Gades; from
here our sea, which is called the Tyrrhenian Sea , flows in' .42
Molina analysed the variants of these passages appearing in the works of
Arab geographers and in later christian texts derived from them. " The first
paragraph appears in eleven versions , eight in Arabic, one in Romance and
one in Latin - the Chronicle qf Pseudo-Isidore, which will be discussed later."
These accounts are broadly similar but suggest that there were three
distinct lines of transmission. The version which is closest to the Columbia
manuscript is the one al-Maqqari said he was guoting from Ahmad al-Razi
(d.955).45 The guotation in the fourteenth-century Chronicle of the Moor
Rasis46 may also be a direct descendant from al-Razi, as may that of
Yaqut. " The other two groups of versions, although surviving in earlier
copies, diverge more than al-Razi's from the 'Urilsfils. Not all the authors
were working from a text identical to the Columbia manuscript. To take
just two examples: al-Bakri seems to share much of his information with
al-Razi but also included excerpts from the Histories which al-Razi did not
use, while the Chronicle qf Pseudo-Isidore has an altogether different version .
Analysis of the second passage, on the division of Hispania into two, is less
helpful for two reasons. The Columbia manuscript is badly damaged at
this point, and most of the geographers did not reproduce this passage in
full. Only the Chronicle of Pseudo-Isidore kept the basic structure of the
Histories, which the Columbia manuscript also followed, whilst al-Razi and
al-Nazzam, both guo ted by al-Maqqari, based their division of Hispania
not on the former Roman provinces, but on climatic and geographical
features. Molina appended to his study a family tree of the different
versions of Orosius' geography, concluding that a version of the 'Urilsfils
similar to that in the Columbia manuscript served all the geographers as
the basis for their descriptions of Hispania, but that the Chronicle qf the Moor
Rasis and the Chronicle qf Pseudo-Isidore have a common origin which is not
the same as the Columbia 'Urastas. As was his wont, Sanchez-Albornoz
identified this missing link, which he called the Book qf the Prophets, from a

143
 THE ARABIC TRANSLATION OF OROSIUS

reference in the biographical dictionary of Hajji Khalifa to 'the book of

Orosius, master of stories , which is a chronicle of the kings of Rum and
stories of the prophets sent to them. It was written in Latin'." Molina
described the geographers' use of 'Urusius as 'a reflexion of the cultural
situation of al-Andalus at the moment when the sorry remnants of the
classical world were appropriated by the flowering culture of islamic
Hispania'. The Arabic scholars may indeed have used only fragments of
the Histories, but they seem to have taken them from at least two versions
of the work available to them.
It is not surprising that most Arabic scholars extracted little from the
Histories apart from the introduction. Andalusi muslim scholars shared with
the majority of their counterparts in the east a lack of appreciation of
christian history. Although, by the tenth century, Cordoban learning
rivalled that of Baghdad, Andalusi scholars still travelled to Baghdad or
elsewhere in the east to sit at the feet of the masters there - or, at the very
least, it was important that their biographers should claim that they had
done so. The most advanced studies were of the Qu'ran, the sayings of the
Prophet, jurisprudence, and theology. History occupied a lowly position in
the islamic curriculum.t" The study of history was suitable for children,
especially young princes. It was not taught in the religious colleges, except
for the science of haduh, the transmission of the sayings of Muhammad,
which required knowledge of the biographies of the transmitters. Until the
tenth century, Andalusi authors were more likely to compile biographies of
Muhammad and the early history of Islam than to write about their own
part of the world .t" Lack of interest in the history of al-Andalus was
accompanied by almost total neglect of the pre-islamic history of Hispania;
Ibn Khaldun is an honorable exception. Sa'td the Andalusi (d. 1070)
described pre-islamic Hispania as a cultural desert: 'In anci ent times, prior
to the Arab occupation, al-Andalus was void of any scientific activity and
none of its inhabitants became known for any scientific contribution. A few
ancient inscriptions dealing with a variety of topics were found in this
country, but everyone is in agreement that they were left by the kings of
Rome, because al-Andalus formed part of their empire. It remained as
such, without any scientific activity, until the advent of the muslims '
conquest. Except for the study of islamic law and the Arabic language, the
lack of interest in science persisted until the Umayyads established their
authority' i'"
Sa'td's view was echoed by Ibn Hazm, who complained about the
dearth of histories of his native land." Ibn Hazrn recognised that history,
together with religious law and the study of language, was peculiar to each
nation and religion. This distinction affected the muslim attitude to

144
 THE ARABIC TRANSLATION OF OROSIUS

non-muslim history, which it was thought could only be imperfectly known,

so one should not waste too much time on it. Further, whilst muslim history
could be written with the intention of showing virtue rewarded and evil
punished, the history of the unbelievers was a mere diversion . Although
pre-islamic history, especially from Arabia, was part of muslim historio-
graphy from the earliest days, it was regarded as fabulous and retold as
historical romances embellished with poetry, often in very simple language,
or even in a local dialect. Such works, with the stockphrases and other
devices of the epic storyteller are still current in Egypt. 53 The pre-conquest
history of Hispania may have been regarded in the same light, although the
sparseness of the evidence makes interpretation difficult. 'Abd ai-Malik ibn
Habib's history of the world from creation to the Umayyads turned the
conquest of Hispania into a melodrama of treachery and revenge. 54 This
work has been a source of irritation to modern historians of al-Andalus
because later sources that they would like to regard as more trustworthy
repeated much of the same material. Ibn Habib's fables originated in
Egypt 55 but he elaborated his account, regarding this as a legitimate
approach to the history of Hispania before 711. Such works cannot be
taken as evidence of a serious interest in non-muslim history.56 The fact
that only Orosius' geographical data were copied underlines the nugatory
value of the rest of the Histories. If the Histories were translated for
al-Hakam, they would probably have been read as entertainment, not as
scholarship.
Only Ibn Khaldun quoted frequently from the writings of christian
historians. His approach to Orosius is worth considering at some length for
the light that it sheds on the transmission of the Histories. Although he was
born in Tunisia in 1337 and lived most of his life in North Africa,
Ibn Khaldun studied with Ibn al-Khatib and others in the kingdom of
Granada and returned there for three years as the favourite of Sultan
Muhammad V. His decision to write history seems to have resulted from
his need to understand his failure to negotiate political change in the
Maghreb, which he described in letters written to Ibn al-Khatib. Other
men had also failed; cities and empires lay in ruins . The history of the
Maghreb could be used to illustrate the principle that 'when there is a
general change in conditions, it is as though the whole world were altered
. .. Therefore there is a need at this time that some one should
systematically set down the situation of the world among all regions and
races'." Methodological explanations like these are Ibn Khaldun's passport
to acceptance by historians who reject other Arabic sources." Not all
contemporary scholars share the general admiration for Ibn Khaldun,
Al-Azmeh'" complained that attention is paid almost exclusively to Ibn

145
 THE ARABIC TRANSlATION OF OROSlUS

Khaldun's 'historical criticism' and 'social theory' to the neglect of other,

more 'oriental' aspects of his work. This bias is seen in the way his writings
have been rendered into western languages, where the language of
sociology makes his work more 'scientific'i'" In spite of Ibn Khaldun's
statement of intent, in the Muqaddimah, to divest history of its fabulous tales,
there was in effect no better way to establish the truth than to select
whatever appeared plausible, and to point to the number of reliable
authors who had used the same information. Ultimately, Ibn Khaldun had
the same view as his contemporaries when it came to the criteria for the
trustworthiness of a source.
The second volume ofIbn Khaldun's History, in which Orosius is quoted
repeatedly, is a survey of the biblical and post-biblical history of the jews
and christians. Ibn Khaldun introduced extracts about Hispania with the
words 'Orosius said', even when talking about the period after Orosius'
death. He entitled one section: 'The story of the Goths and what happened
to them from the time of the kingdom in al-Andalus until the time of the
islamic conquest, and its elements and destiny . . . this is the sequence of
events of those people the Goths; we quote it from the words of Orosius
and more properly from our opinion about that ... ' .61 He also credited 'the
scholar Isidore bishop of Seville and those who added to it after him '.
Ibn Khaldun's version of the history of Hispania is garbled and his
quotations of Orosius do not often match either the Histories or the cUrtIsfus.
Badawi listed all fifty-seven of Ibn Khaldun's citations of Orosius in the
appendix to his edition of the Columbia manuscript. He unravelled their
relationship with the Histories, the 'Urastas and other sources.f Six of
Ibn Khaldun's citations of Orosius can be found in the 'Urastas but not
in the Histories but almost half of Ibn Khaldun's citations are closer to
the original Latin than to the CUrasais. Many of these extracts, such as the
account of Nero's persecution of the christians and the murders of Peter
and Mark," are extensively elaborated. On sixteen of the occasions when
he introduced his material with the words 'Orosius said', Orosius said no
such thing, either in the Histories or in the cUnlsfus. Rather than speculating
about Ibn Khaldun's sources, it is instructive to look more closely at the
way he handled them.
Orosius is not the only non-muslim historian whose works Ibn Khaldun
used . He quoted two Coptic historians, ]irjus ibn al-'Amtd and Ynsuf ibn
Karyan, author of three books of the Maccabeans'" and the collection of
jewish sources known as the Chronicle rifJosippon.65 Ibn Khaldnn placed his
sources in a heirarchy in which the authority of a piece of information
depended on the religion of its author. He quoted the Chronicle rifJosippon
twenty-two times and reproduced almost verbatim nearly all the chapters

146
 THE ARABIC TRANSLATION OF OROSIUS

on the period up to the destruction of the Second Temple, because it was

the only material in his possession on this subject. Ibn Khaldun compared
Orosius with the Torah, praised for its meticulousness as a historical
sources in spite of its theological distortions." Yet Ibn Khaldun did not
quote Orosius, or any other non-muslim sources, when a muslim author
had written on the same subject. He took this policy to absurd extremes.
When recounting biblical history, which was important to muslims as the
pre-history of Islam, he did not go back to the Old Testament, but to
muslim historians. The scales had been tipped against the biblical material
by al-TabanY Even Ibn Khaldun, who was ideally placed to read the
scriptures and histories of the jews and christians of North Africa, does not
seem to have done so with the care demanded by his own theories of
historiography. Rather than copying from an Arabic translation of the
Histories, he used 'Orosius' as a form of shorthand for christian history.
Ibn Khaldun's quotations of Orosius do not establish in what form the
Histories were transmitted to him, although it suggests that the Columbia
manuscript represents only one of a number of Arabic versions of the
Histories.

The Colwnbia lDanuscript

The Columbia manuscript, the sole surviving Arabic version of the

Histories, consists of 129 loose pages, badly torn and chewed by insects,
especially at the edges . It is undated. Most of the table of contents survives,
listing each book and chapter, with a brief summary, starting with a
summary of chapter three.t" Levi Della Vida argued that two pages are
missing from the beginning, one side of the first being a translator's or
copyist's introduction, which would normally have included the date and
probably the place of completion and the name of the copyist and/or his
patron. The lemma for each book is repeated, sometimes in modified form,
at the beginning of that book. The manuscript gives out at Alaric's arrival
in Rome, just before the end of the last chapter of the Histories, and it may
be deduced that several pages are missing from the end. The lemma for
chapter 14 at the beginning of the manuscript indicates how the 'Unlsfils
went on:
'Chapter 14 in which are mentioned the emperors from Arcadius up
to the time of Heraclius Caesar and the empire of the Goths their
contemporaries up to the time of Rodrigo, at whose hand their
dominion was cut short, as well as the people who governed

141
 THE ARABIC TRANSLATION OF OROSIUS

al-Andalus before the Goths . .. [line missing] Caesar, Isidore the

learned bishop of Seville then after him was added ... [word
missing] . . . of the empire up to our time according to the extent of
their knowledge'J'"
The lemma at the beginning of book 7 confirms this:
'Book 7 in which there are accounts of the events of the empire of the
Romans, the Caesars, from the time of Augustus, during whose reign
Christ was born, up to the time when this book was written, and
what was added to it afterwards about the kingdom of the Goths in
al-Andalus up to the arrival of Tariq'.
Thus, unless these passages are misleading, the text from which the
translators were working was a copy of the Histories to which a continuation
of the history of Hispania had been added, using Isidore and extended to
711 from an unknown source. The division of the text into books follows
the Latin manuscript tradition, although chapter headings are not a
feature of the Histories and may have been added with the continuation.
A number of Arabic words of Hispanic provenance, many of them found in
the Latin Arabic Glossary now in Leiden," point to a peninsular origin for the
translation.
The date 712AH [1312] is given in the Columbia catalogue. The first
person to mention the manuscript in modern times was Silvestre de Sacy,
and he may be responsible for dating it. None of the later commentators
on the manuscript have confirmed this date." The manuscript is of paper,
which was used in Hispania from the ninth century. The oldest paper
manuscript is the one surviving book from al-Hakam's library, which is
dated Shaban 359 [9 June-7 July 970f2 but the Columbia manuscript is
not written in the same script as this. There are features which locate the
script in the Almohad or Almoravid periods, such as Kufic letters and full
vocalisation [the marking of the short vowels, which are commonly
omitted] which is uncommon in earlier manuscripts. Dating of Arabic
manuscripts is a controversial and under-researched area." It is difficult to
distinguish between Andalusi and Maghrebi scripts, and experts may
disagree by up to three hundred years in the dates they assign to them. It is
not always obvious whether a manuscript originated in al-Andalus, in
North Africa or even in christian Hispania. Sometimes christian manuscripts
in Arabic followed islamic codicological practices, for example in the
number of their gatherings, but at other times they adhered to Visigothic
norms. Most of the surviving manuscripts of this period were discovered in
North Africa but this does not mean that they were all muslim texts. There

148
 THE ARABIC TRANSLATION OF OROSIUS

was a large community of native christians in North Africa, and after 1126
there were mass deportations of christians from al-Andalus to Fes and
Meknes, following th e unsuccessful attack by Alfonso el Batallador on
Granada. " An Arabic translation of the Gospels now in Leon Cathedral
was copied in 1421 from an exemplar pr epared in 1175 by MiqII ibn 'Abd
al-Azlz th e bishop, 'in the Cit y of Fes, in the west part of the [North
African] shore in the eleventh year of the exodus of the christians of
al-Andalus towards it, may God restore them ';" Thus the Columbia
manuscript may have been copied in al-Andalus or in North Africa. The
palaeographic evidence is in favour of the latter and perhaps a fourt eeth-
century dating, although Van Koningsveld thought it a century older.76
This makes the date of the translation of which this manuscript is a copy
unc ertain.
The translators were obviously struggling with their material. Orosius'
convoluted style was difficult to understand, and they may have been
working with a corrupt version of the Histories. They took great liberti es
with the text, abbreviating and transposing it and sprinkling their effort
with disclaim ers, such as 'we have suppressed this, but through a love of
conciseness and not wanting to go on at length'. They left out almost the
whole of Orosius' prefaces to Books 5, 6 and 7. Some sections may have
been omitted because they were incomprehensible. Others were glossed
extensively. There are many spelling mistakes, particularly in the names of
places and people, although some of these may be the fault oflater copyists.
There ar e many instances of muslim influenc e on the text , which begins
with the bismillah, the opening words of the Qur an .77 The clear est example
of Arabicizing is the presentation of personal names in their Arabic form
X ibn Y This necessitated the invention of names for the forgott en fath ers
of the heroes of antiquity. The found er of Rome became Romulus ibn
Marcus, and Isidore's Homerus, clearly a Greek, was transliterated as
Mrrns, called 'the Italian poet' , and acquired a father called Marcionus.
Emperors were almost always made the son of their predecessor. The
emperor Constantine's genealogy was tra ced through seven of the previous
rulers" - an impressive accretion for a man whose origins are still obscure .
Some of the biblical characters were given the forms of their names as they
appear in th e Qur'an. Mount Ararat, where the Ark came to rest, was
identified with Mount Judi, according to islamic belief. In the Histories,
Orosius' narration of th e crucifixion made use of a passage from Virgil.
In the 'Urastas this was turned into a paraphrase in two verses written in
an Arabic poetic metre, to which three more verses were added from an
unknown source. The use of Arabic metre and islamic terminology was
pr esum ed by some commentators to indicate that it was translated for

149
 THE ARABIC TRANSLATION OF OROSIUS

muslims, but such features also characterise translations of christian texts

into Arabic. Tran slators used islamic religious vocabulary even in christian
theological texts, just as they adopted Arabic personal nam es.??
The Columbia text is not simply a translation of the Histories plus its
continuation. It is padded with material which does not come from
Orosius. The nature of these interpolations may help to pinpoint the origin
of the text from which the translation was made. Daiber'" concluded that,
apart from obvious biblical references, most of the interpolations could
have come from Isidore's Chronicle with a few echoes of the Etymologies. The
last Byzantine emperor named is Heraclius, who is also the last in Isidore's
account. Orosius had skipped over early times, merel y demonstrating how
the Fall of M an resulted in the Flood, befor e starting his history proper.
The l.Inis iiis fleshed this out with th e story of Creation, of Adam and of the
C

dispersal of the descendants of Noah, closely following the Old Testam ent
but using Creation Era dating, as Isidore had don e in the Chronicle. Islami c
universal histori es commonly began with the Creation; an Andalusi
example is the History of Ibn Habib. A later passage on the principate of
Augustus also came from Isidore, although the attribution was confused by
th e repeated interpolation 'Orosius said' . The majority of the other
additions were also deriv ed from the christian tradition , such as the
reference to St. Martin of Tours and toJerome, called 'the translator'. M ost
of these additions could have been made at any time between the early
seventh century and the date when the text was translated.
An interesting interpolation betw een the end of Book Six and the
beginning of Book Seven , is the bizarre story of Augustus' paving of
the river Tiber with bronze:
~d in the fourth year of his reign he impos ed upon the inhabitants
of the world a tribute in bronze, and collected the sam e amount
which everybody would have to pay in gold throughout the whole
world: th erefor e copper was sought in th e provinces at any price, so
that its pri ce went up above the pri ce of gold. Having collected a vast
amount of it, thick plates and plugs were stru ck out of it, and he
paved with them the river of Rom e and its bank s for a length of forty
miles and an impressive width. And the people went so far as to take
this as [the beginning of] a new era, which is the era of the christians
to the present timc' ."'
This story seems to be a conflation of two traditions. It may have
originated in the Arabic world, but it became mixed up with legend ary
material from other sources . Versions of the story app ear both in the works
of later Arabic historians includ ing Ibn Hayyan , al-Maqqari and al-Idrisl.

150
 THE ARABIC TRANSLATION OF OROSIUS

It was also reproduced by several jewish historians and in the Chronicle of

Pseudo-Isidore. Its inclusion in the 'Uriisfiis is an indication of how complex a
mixture of material this is, but it does not help to establish its provenance.
By far the longest interpolations in the 'Urusius are of the legends of
Constantine, especially those linking the emperor with pope Silvester. They
include the following story:
'At the beginning of his rule Qustantin adhered to the faith of the
pagans, and persecuted the christians and issued edicts against them.
And he came to the faith because of a learned christian called
Shalbashtur, [Silvester] patriarch of Rome. Qustantin had been
assiduous in his demands on the christians and afflicted them greatly.
And the sage claimed that [this was why] Qustantin became troubled
by leprosy, which was victorious over him. He grieved sorely because
of this and he gathered to him the people skilled in medicine and
with insight into illness and gentleness in medical treatment. And he
asked them to look into his illness and collected their opinions on it.
They told him that he should bathe in a cistern filled with the blood
of suckling infants one hour old. So he ordered that many children
be brought to him [so that they might] slaughter them in the cistern
on a day when he could come and bathe in that fresh blood. He went
out to the place which had been prepared. And when he emerged
from the palace, a hubbub of wailing was heard from the women
whose infants had been taken.
When Qustantm enquired about this he found that they were the
mothers of the infants whom he had gathered to be slaughtered.
Qustantin was merciful towards them and grieved with them for
their children and said: 'We do not order the murder of the children
of our defeated enemies, rather we order that they should be
protected. How then should we deem it permissible to kill the
children of our own citizens? I would rather tolerate the illness which
has recently come upon me than find it necessary to destroy this
group of human souls and their grieving mothers with them. And he
ordered the release of the infants and that no more should be
collected.
And when he came to his bed that night he saw in a dream an old
man who said to him : "As you have shown mercy on the children and
their mothers and borne your illness for [the sake on their liber ation,
so God shows mercy to you and grants you recovery from your
illness. Seek a man of the [christian] faith called Shalbashtur who will
banish the fear from you and instruct you and good health will come

151
 THE ARABIC TRANSLATION OF OROSI US

to you in body and spirit" . Qustantrn woke up am azed by wha t he

had seen and sent his servants for Shalbashtur. And they brou ght
him and he believed that [Con stantine] wanted to kill him . And he
met him and revealed the dr eam to him and disclosed his faith at
length in piety and deference; we shortened this passage and cut
what follows about the discussion of Shalbashtur with the jews and so
on and so forth , through a desire for conciseness'i'"
A Lift if St. Silvester circulated in the east in Greek and Syriac versions
before 500 , but th e elaboration of the legend of Constantine and Silvester
was an aspect of the propagation of the cult of Constantine in Rome, as the
popes assert ed the primacy of the western church over Byzantium,
culmina ting in the so-called Donation of C onstantine.P The Acts if Silvester,
dating from th e early-sixth century, wer e widely copi ed and more than
three hundred manuscripts survive. They wer e certainly read in Hi spani a,
although the earliest surviving manuscript is a fragm ent from Silos,
possibly from the eleventh century.'" The story of Constantine's leprosy
and his baptism by Silvester, which was alread y circulating in the sixth
century, was taken up by Byzantine chroniclers as an ortho dox alternative
to the ea rlier story that he had been bapti sed by the Arian bishop Eusebiu s
in Nicomedia." It is impossible to determine the dat e or proven an ce of the
version of Constantine's bapti sm in the 'Urustus although an elab orate
retelling of the legend is more likely to have been taken from a western
exem plar of the Histories than from a copy brought from Constantinople.
Another such legend which attracted the attention of the ' Urastas'
translators or the compilers of their source was the dream which inspired
the empe ror to build Constantinople, where the 'UnIsiils' version is similar
that recounted by Aldhelm ."
The Columbia manuscript's mixture of christian history and legends
with islamic elaborations is not unique. A second Latin history translated
into Arabic was discovered in the mosque of Sidi "Uqba in Q ayrawan ,
found ed in 829 .87 The Qayrawan manuscript, which is in an even worse
state than the Columbia manuscript, has been dated by Levi Della Vida to
th e lat e-thirteenth or early-fourt eenth centuries, and by Van Konin gsveld
to the twelfth cent ury'" It is in three parts, which may not always have
existed togeth er. One is a universal chro nicle which is bizarre, even making
allowances for th e numerou s lacun ae in the manuscript. The fables with
which the autho r elaborated his chro nicle includ e an account of the seven
disciples sent to Hi spani a by St. Peter to pr each the faith , who persuaded
the inh abi tants to shave off their beard s. Again , Con stantine is one of the
hero es of the chronicle but the version of his legend s pr esent ed here is not

152
 THE ARABIC TRANSLATION OF OROSIUS

the same as that in the 'Urusius. The Qayrawan manuscript has Constantine
undergoing a secret conversion before his accession to the empire, and
introduces the foundation of Constantinople at this point in the narrative.
After this, the chronicler leapt four centuries to the islamic conquest of
Hispania, with the briefest of introductions: 'then we come to the learned/
happy? man Tariq and how he unified al-Andalus'i''? The next section of
the manuscript is missing, and it resumes in the middle of the story of count
Julian's betrayal of Hispania to Tariq ibn Ziyad in revenge for Rodrigo's
violation of his daughter, which is follwed by another tale about a Visigothic
king called here Talul or Tulul. After another gap, the history concludes
with Tariq's troops eating their captives; this is clearly the end, since the
writer appended the formula 'the book is finished'. To judge by its language
and its spelling mistakes, this history is, like the 'Urasias of Hispanic origin.
The organisation of the chronicle confirms its Hispanic parentage, since it
divides history into the Six Ages of Isidore. Appended to the universal
chronicle are two works of christian-muslim polemic. One is the dialogue
said to have taken place between the Nestorian patriarch of Baghdad,
Timothy and the caliph al-Mahdi. This text was widely disseminated in the
east and North Africa. The second work is also in the form of a dialogue but
this time the protagonists are identified only as a patriarch and a muslim.
The association with these texts of undoubted christian origin points to a
christian provenance for the chronicle. The Qayrawan history is much less
learned even than the 'Urasius but it is very similar in conception. Histories
like this were written in Arabic by North-African christians at this period.?"
The earliest to survive is that of Sa'rd ibn Bitriq , patriarch of Alexandria
(933-40), a universal chronicle from the Old Testament through the islamic
conquests, ending in 938. Like the cUnlsiiis it is a mixture of scriptural and
historical sources sprinkled with legends. Another tenth-century history, by
Agapius (Mahbub) ibn Qustantin al-Manbiji, is based on the Byzantine
christian tradition but includes pre-islamic history and biographies of the
caliphs," a hybrid of christian and islamic history reminiscent of the
Chronicle if 741 and the Chronicle if 754.
Was the Columbia 'Urastus also a christian text? To judge by its glosses,
the Columbia manuscript seems to have been read in at least two different
circles. On folio 79r. there is a Latin gloss, suggesting that the 'Urastas was
being read in a part of Hispania reconquered from the muslims, by
christians who were moving from Arabic back to Latin ." On folios I lOr
and ll8v. an Arabic gloss obviously written by a muslim criticised the
'unbelief ' of the christians, especially the doctrine of the crucifixion . The
Qayrawan manuscript too is glossed in Latin, but it may later have been
read by muslims, hence its place of discovery. The complicated afterlife of

153
 THE ARABIC TRANSLATION OF OROSIUS

these manuscripts makes them difficult to pin down . They seem to be texts
which the christians produced for their own use, rather than commissioned
for islamic libraries, which later fell into muslim hands, probably in North
Africa, The origin of such histories might have been obscure even to their
late-medieval readers. Perhaps Ibn Khaldun, on finding the 'Urusius or
another translation of the Histories in the Maghreb, and knowing that it was
the work of a christian from Hisp ani a, supposed it to have been
commissioned for al-Hakam 's famous libra ry.

Christian Arabic JDanuscripts

The 'Urusius and the Qayrawan history can be placed in a wider context,
among a group of christian texts written in Arabic. Hispania was unusual
amongst the islamic conquests in failing, in the long run, to lose its native
language to Arabic. As th e Reconquest proceeded, the primacy of Latin
was restored and Arabic manuscripts, now unreadable, were destroyed. In
such circumstances, it was easy to forget how Arabic scholarship had
dominated al-Andalus, even in chr istian circles. Some fragm ent s of
christian Arabic manuscripts have been preserved in book-bindings.
Others were copied by muslim and jewish scribes for jewish scholars."
or in muslim circles in North Africa, for polemical purposes. The colophon
of a copy of the Gospels dated 1493 said that its purpose was 'to take notice
of the traditions of the jews and the christians and of their despicable
beliefs .. . so that . .. the excessive errors they committed will become clear
to those who look into this book . .. They will then believe th at the religion
of Islam is the most superior of all religions'.94 Most of the manuscripts date
from the twelfth century or later, although some ar e copies of translations
which were probably made before 1000.
The existence of other christian Arabic texts can only be inferred. Som e
of the Arabic glosses that have been discovered in Visigothic manuscripts
quoted passages which the readers appear to have copied from translations
of the works th ey were glossing. A manuscript of Isidore's Erymologies,95 with
about 1,500 Arabic glosses, provides examples of this practice. A map of
the world in this manuscript has an Arabic legend which could come from
chapter 2 of the 'Unlsftls. To a Latin manuscript of the Visigothic law code?"
was added lengthy interlinear interpolation in Arabic, perhaps from the
translation which Ibn Hazm read." Van Koningsveld's recent studies of
this material have shown how much of their literary heritage the christians
of al-Andalus made available in Arabic and the following summary is
based on his work.

154
 THE ARABIC TRANSLATION OF OROSI US

Nearl y all the transl ations were of sacred texts. During this period the
Bible was extensively and repeat edly translated into Ara bic throughout
the islami c lands, to make the scriptures available to those christians who,
like th e christians of al-And alus, spoke their own language at home, but
used Arabic rath er than Latin or Greek as their written language. As texts
were copied to be read aloud, it mad e sense to tran slate them into
whichever language their audience better understood . It is also possible
that during periods of tension between christians and their rulers it might
have been dan gerous to read Lat in aloud even if one could do SO.98 By the
eleventh century, when Ibn H azm compiled his History if Religion, he was
able to consult texts of all sections of the New Testam ent in Arabic; he
describ ed the script and number of leaves of each exemplar." Ibn H azm
could have read a copy of the tra nslation of the Gospels mad e by Ishaq ibn
Balask of Cordo ba in 946 in which each Gospel begins with the bismillah. roo
Ishaq used a Hispani c version of the pre-jerom e Latin bible; a codex now
in Munich' ?' contains Ishaq 's text correc ted from an an onymous Arabi c
translation of the Vulgat e. Another Arab ic translation of the Go spels,
mentioned by Rodrigo Jimenez de Rad a, bishop of Toledo (d. 1247), is
attributed to bishop John of Seville. 102 This is usually taken to be the J ohn
of Seville who att ended the Council of Cord oba in 839 and corresponded
with Alvarus, altho ugh Van Koningsveld thinks that Jimenez de Rada was
referring to the twelfth-century l ohannes H ispalensis, who was famo us for
his tran slations from Latin into Arabic. In one bilingual manu scrip t, of
whieh only a fra gme nt of Paul's Epistle to the Ga latians surv ives.l'" the
Arabic version seems to have been written first, with the Latin abbreviated
and cra mmed in where it would fit. No complete Old Testam ent text in
Ara bic of H ispani c origin survives (Ibn H azm had a ninth-century
Egyptian version) but m argin al glosses in Latin manus cript s make it clear
that such translations existed.' ?' There are, however, three manuscripts of
Arabic versions of the Psalm s. The earliest of these is a metrical translation
into the metr e of th e rojar, one of th e commo nest used in Ara bic poetry.
The same metre was used in the poetry interpolated into the ' Unlsfils. This
translation of the Psalm s was completed by H afs ibn Albar, probably in
889 , although the manuscript may have read 989. It survives in a copy
made by David Co lville (16 17- 27) of a manu script from EI Escorial, now
lost. 105 H afs ibn Albar said he was working from the Bible of J erome to
replace an earlier, un satisfactory translation of the psalter. He took care
to consult authorities on Arabic,' ?" in order to avoid the criticism of 'the
ignorant, obstin ate and nar row-minded' . 10 7 U rvoy interpreted this to mean
those who were opposed to Arabicizing chr istian texts. It is clear that H afs
saw no incompatibility between good Arabic and strong chr istian faith ; the

155
 THE ARABIC TRANSLATlON OF OROSIUS

Arabicizers were 'scholars, trustworthy men of our religion' .108 Hafs wrote
a work of polemic which was quoted by al-Qurtubt.l'" who prais ed him as
'one of the most intelligent and excellent of men . . . because he wrote
under the protection of the muslims and learned from their sciences things
which surpassed those of the christians'. His career testifies to the demand
for translations of christian biblical and theologi cal texts into Arabi c and
shows that such translations were sometimes read by muslim scholars .
Another christian Arabic translation which al-Qurtubi read was the
Collection qf councils which survives in one manuscript from EI Escorial ."? In
the Arabic translation, the canons were presented under a series of subject
headings, like a collection of th e sayings of Muhammad, rather than
chronologically as they ar e in the Latin manuscripts of the same council s.
There are several other instances of Arabicizing of the material. As Kassis
pointed out: 'the bismillah is rather curious ... when placed on the lips of
Reccared in his speech opening the third council of Toledo'.'!' The
manuscript was copied out by Vinc enti us for a bishop named 'Abd
ai-M alik and completed in Era 1087 [1049], although th e translation itself
may be earlier. In 1090, a book of canons written in Arabic was donated to
Coimbra on the death of bishop Paternus; two fragments which may be
part of this manuscript are now in Lisbon .!' " Vincentius' copy of the
canons is particularly interesting because it has a number of Latin glosses
and a long passage in Latin which Vincentius himself may have copied.
Perhaps Vinc en tius, in the middle of the eleventh century, represents the
point of tru e bilingualism for the christians of al-Andalus. As the tide
turned and the christian armies gradually displaced the Arab ruler s, Latin
texts were copied in greater numbers and christian Arabic texts were
gradually discarded, or in some cases retranslated into Latin .
A bri ef reference to the Chronicle qf Pseudo-Isidore serves as a postscript to
this discussion of translation in early islamic Hispania and closes the circle
containing the Histories and the 'Urastas. In this chronicle the people of
Hispania could onc e again read their history in Latin, but in a text which
faintly echoes the way that history had been pres ented to them during the
islamic period. As with so many of these works, th e date and provenanc e of
the Pseudo-Isidore ar e controversial and interpreta tion has been made more
difficult by several generations of errors in transmission, including those of
the nin eteenth-century editors. I 13 The surviving manuscript probably dates
from the first half of the twelfth century, certainly after 1115. Like the
earlier histories , this is a type of universal chronicle, although very shor t,
based on Orosius and Isidore with the addition of material on the fall of
th e Visigoths and the Arab conquest. Gautier-Dalche identified the Pseudo-
Isidore as the translation of history originally written in Arabic by its

156
 THE ARABIC TRANSLATION OF OROSIUS

misspellings of the names of people and places . Some of these errors were
compounded when the text was rendered back into Latin. The Chronicle
may have been copied again in the area around Narbonne, producing
further misidentifications of places . The Pseudo-Isidore reproduced the
passages from Orosius on the geography of the peninsula which the Arabic
geographers had used . In his description of Hispania, the compiler put
south at the top and north at the bottom, following islamic practice. One of
the christians involved in its transmission produced a revised version of an
episode of Visigothic history taken from Isidore. The Visigoth Gesalic was
no longer described as seeking help from the Vandals in North Africa
against the Burgundians; the Pseudo-Isidore has him going to Corinth and
returning via Italy, perhaps because the Maghreb could no longer be
perceived as a potential ally. The Pseudo-Isidore reads like a christian work
whose transmission went through an intermediary stage in an Arabic
translation, made in similar circumstances to those which generated a
version of the 'Urastas.
The outline I have presented removes from the Columbia manuscript
some of the historiographical baggage it has accumulated. It is not unique,
but just one manifestation of the process of translation into Arabic of
christian texts, both sacred and secular. The evidence does not rule out a
commission from al-Hakam II, but the text was probably translated for a
christian audience, although muslims also read it. Whilst the christians of
al-Andalus never lost their Latin entirely, the Umayyad period saw an
increasing demand for christian texts in Arabic and amongst these was at
least one translation of Orosius' Histories. This is how 'Orosius' became
synonymous with christian history for Ibn Khaldun. When christian
Arabic texts could no longer be read in the peninsula, most of them were
destroyed, leaving the Columbia manuscript as their most important
representative .

157
 CHAPTER E1GHT
_.
Sara the Goth and her
descendants

:.\lamundo [one of the sons of Witiza, the penultimate Visigothi c

king of Hi spania, 698 -710] died , leaving a daughter called Sar a the
Goth and two young sons, one of whom was the metropolitan of
Seville, and the oth er Oppas, who died in Galicia. Artubas [another
son of Witiza] enlarged his possessions, seizing tho se of his neph ews,
at the time of the beginning of the calipha te of Hisham ibn 'Abd
al-Malik [724-43]. Sh e [Sara the Goth] ordered the construc tion of
a boat in Seville, which was the city where her father Alamundo had
fixed his residence, since the thousand villas which had fallen to him
were in the west of H ispani a ...
Then Sara the Goth sailed with her brothers towards Syria,
disembark ed in Ascalon and continued on her journey until she
stopped in front of the gate of Hi sham ibn 'Abd al-Malik. She
informed him of her arrival and of al-Walid's pledge to her fath er,
presenting her complaints against th e injusti ces committed by her
uncl e Artubas, The ca liph received her and she saw 'Abd
al-Rahman ibn Mu'awiya , a young man who was standing in the
pr esen ce of the caliph. 'Abd al-Rahrnan would often recall this in
Hi spania, when Sara went to C ordob a and was allowed to visit the
monarch's family. Hi sham, to show his favour to Sara, wrote to
Hantala ibn Safwan al-Qalbr, governor of Africa , ord erin g him
to carry out th e provisions of al-Walid ibn 'Abd ai-M alik and to
tran smit the order to the governo r of Hi spani a Husam ibn Dirar,
usually known as Abu Khatab al-Qalbi, who would carry out the
orde r.
The caliph Hisham marri ed her [Sara] to 'I sa ibn Mu zahim, who
wen t to Hi span ia with her and regained the possession of her villas.
This ' Isa was the forefather of the Ban u al-Q uuya. From this
marri age she had two sons, Ibrahim and Ishak '. I

158
 SARA THE GOTH AND HER DESCENDANTS

T his extract is taken from the work of a historian, Ibn al-Qutrya (d. 977),
who was not a christian but the descendant of a christian convert to
Islam. The title History if the Conquest if al-Andalus which appea rs in the
manuscript of his work is misleading. It begins with a fairly detailed
account of the first years of islamic ru le in Hispania, followed by an
account of the problems of maintaining order in the peninsula until the
arrival of 'Abd al-Rahman I in 756. The rest of the work is a series
of eulogies of the Umayyads, mad e up of anecdotes about the reign of
H isham (788-96), the revolts against al-Hakam I (796-822) and bri ef
accounts of the reigns of 'Abd al-Rahman II (822- 52), Muhammad
(852- 86), al-Mundir (886- 88) and 'Abd Allah (888- 912). It was probably
compiled after the death of'Abd al-Rahman III in 961, since, although his
reign is not covered, he is referred to with the formula 'May God be
pleased with him ', which impli es that he had died. In man y respects it is
the sort of propaganda that th e Umayyad caliphs wou ld ha ve been happ y
to read. Yet, unlike oth er historian s of al-Andalus, Ibn al-Qunya also
rem embered the role which the christians had played in the conquest and
in the development of Andalusi society. Ibn al-Q ntrya's account of the
islam ic settleme nt is based on a number of stories about the descend ants of
Witiza , the penultimate Visigothi c king. They includ e the story of his
granddaughter Sara, from whom Ibn al-Qutlya may have taken his title, or
nickname 'son of the Gothic wom an '. The hou se of Witiza played a
leading role in several accounts of the islami c conquest, both in Arabi c and
in Latin . Ibn al-Qjitlya said that the sons of Witiza were the first christians
to make their peace with the invad ers. Histor ians writin g in Latin took a
more jaundiced view. According to the Chronicle if 754, Witiza's br other
betrayed Toledo to the invaders," and, from the Chronicle if Alfonso III
onw ards, Witiza and his sons shared with Rodrigo the respo nsibility for the
'ruin of Spain' . T he Chronicle ifAlfonso III not ed that 'the Saracens entered
Hispania on account of the treachery of the sons of Witiza' .3 The stories
became ever more elaborate (and discouraging to historians) in the Arabi c
histori es. It is generally assumed th at such legends have no explana tory
value. Yet the fact that th e stories are contradictory and some times frankly
incr edibl e does not invalidate them as historical sour ces. It is possible to
examine their contemporary meaning, regardl ess of their truth content.
This chapter discusses some of the problems of the History if the Conquest if
al-Andalus, before focussing on Ibn al-Qunya's stories abo ut the house of
Witiza as a source of information about the way in which christians may
have interpreted their transition from Visigothi c to islamic nobility.

159
 SARA THE GOTH AND HER DESCENDANTS

Ibn al-Qiipya

The family ofIbn al-Qunya held high rank in al-Andalus," suggesting that
their relationship with their patrons, members of the Qurayshi tribe, was of
long standing. Ibn al-Qutlya's father was a judge in Seville and Ecija . Most
of our information on Ibn al-QiltIya himself comes from his pupil
al-Faradi," who was born in Cordoba in 962, served as judge in Valencia
and was killed during the Berbers' capture of Cordoba in 1013.6 Al-Faradi
was most famous for his biographical dictionary, which survives in a single
late-medieval manuscript discovered in Tunis in 1887. It includes a
comparatively long entry on Ibn al-Qntiya:
'Muhammad (Ibn 'Umar) Ibn 'Abd al-Aziz ibn Ibrahfrn ibn 'Isa ibn
Mazahim, a client of 'Umar ibn 'Abd al-Azfz, known as Ibn
al-Qutlya of Cordoba, whose family came originally from Seville,
known by the surname Abu Bakr, studied in Seville with
Muhammad ibn 'Abd Allah ibn al-Qarn and Hasan ibn 'Abd Allah
al-Zabart and Sa'Id ibn jabr and 'AlI ibn AbI Shiba and in Cordoba
with Tahir ibn 'Abd al-Aziz and Ibn AbI al-Walid al-Arj and
Muhammad ibn 'Abd al-Wahab ibn Mughlth and Muhammad ibn
'Umar ibn Lubaba and 'Umar ibn Hafs ibn AbI Tamim and Aslam
ibn 'Abd al-Aziz and Ahmad ibn Jild and Muhammad ibn Masur
and Muhammad ibn 'Abd ai-Malik ibn Ayman and 'Abd Allah ibn
Yanis and Ahmad ibn Bashir ibn al-Aghbas and QasIm ibn Asbagh
and others.
He was a learned grammarian and more advanced in this subject
than the people of his time. He was unsurpassed and no-one
matched him. He wrote excellent books on this art, among them the
Book on the Corljugation if Verbs and the Book on the Shortened and the
Extended Alif[another grammar] and others. He was the guardian of
the stories [akhbar] of al-Andalus, dictating stories of the lives of the
emirs and the affairs of the scholars and poets . These he dictated
from memory. His grammar books were more often studied. He did
not adhere to the rules of hadith' or fiqh [islamic jurisprudence] in his
narratives and he did not return to the original sources ; what he
recounted conveys meaning but not literal truth. What he said often
lacked verification.
He lived to a great age and the people listened to him; generation
after generation all the scholars and elders who had charge of the law
quoted him. He came to the council and he acted according to the
guiding principles of the sons of the kings and others. Feats of

160
 SARA THE GOTH AND HER DESCENDANTS

speculation about the Arabic language are attributed to him upon

hearing? the Kiimil of Muhammad ibn Yazid al-Mubarrad, which Ibn
al-Qjniya would transmit as he had heard it from Sa'rd ibn Qahir,
Learned assemblies gave testimony of this.
He died, may Allah have mercy upon him, before we had finished
with him on a Monday, seven days before the end of [the month of]
Rabi' al-Awwal in the year 367 [977]. He was buried on Wednesday
at the hour of afternoon prayers in the burial ground of the Quraysh
and Abu Ba'far ibn 'Awn Allah was entrusted with leading the
funeral prayers for him .
This combination of a surviving text and a biography of its author is an
exciting one for the modern historian, and those who struggle with obscure
Latin authors such as 'Fredegar' must envy the Arabists their biographical
dictionaries. In the case of Ibn al-Qjitiya, the witness was close to his
subject. Although al-Faradi was only 15 years old when his master died,
there is no reason to suppose that he could not have written an accurate
account of Ibn al-Qunya's life. Yet there are problems with this passage.
Some of the details may be corroborated but there is one striking omission.
Al-Faradi did not list a book of history [tar'i1ch] among Ibn al-Qjitfya's
achievements, and he appears to have disparaged his teacher as a historian.
Ibn al-Qutiya was merely a teller of stories [akhbiir] which he passed on
with scant regard for their accuracy. As we shall see, the distinction
between tar'i1ch and akhbiir has important implications for the status of the
History as evidence for al-Andalus, In an attempt to resolve the disjunction
between al-Faradi's comments and the existence of the History, considerable
emphasis has been placed on the exact wording of al-Faradi's grumble that
Ibn al-Qjltiya did not repect the rules of haditli when writing history.
Al-Faradi's statement, rather than raising insuperable obstacles to the
acceptance of the History as genuine, may account for some of the
difficulties posed by the surviving text. Before examining this passage in
detail, it is useful to consider the problems of biographical dictionaries
in general, and of this passage in particular, which suggest that it may be a
mistake to take the biographer too literally.
Biographical dictionaries were among the first products of Arabic
literacy. In the cast, the earliest surviving texts date from the middle of the
ninth century. Al-Faradi's compilation is the first known to have been
written in al-Andalus. At first, the dictionaries concentrated on religious
scholars, but by the tenth century their subjects included poets and other
men of letters , judges and physicians. The entries in these dictionaries
followed a formula, as can be seen from a comparison between al-Faradi's

161
 SARA THE GOTH AND HER DESCENDANTS

biography of Ibn al-Qjttiya and the account of the judge and scholar
al-Ghafiqr by Ibn al-Abbar (d.1260), analysed by Humphreys," Like
al-Faradi, Ibn al-Abbar listed several generations of the genealogy of his
subject and the place of origin of his family. He named some of the men
with whom he studied, the fields of scholarship in which he excelled,
outlined the virtues of his character and gave the date of his death and a
brief account of his burial. Such biographies shed more light on the
educated classes as a whole than on any individual, but although they are
stereotyped and selective biographical dictionaries are thought to be less
susceptible to later interpolation and falsification than other sources .
Yet biographical dictionaries could be misused . In a passage describing
the swearing of an oath of loyalty to Hisham in 976, after the death of his
father al-Hakam II, the fourteenth-century author Ibn al-Khatib gave a
long list of the people present. Number eighteen was 'Abii Bakr ibn
al-Qjitrya, imam and writer'." This list is at first sight a valuable source for
the prosopography of al-Andalus, making up to some extent for the
absence of charters. Unfortunately, when the people mentioned are
checked against their biographies, it turns out that some had died , others
were not yet born and many were outside al-Andalus at the time of
Hisham's accession. Ibn al-Khatib said that he had taken the first
paragraph of the account from Ibn Hayyan. This extract cannot be found
in other citations of Ibn Hayyan. It is, however, very close to a passage in
Ibn Idhari which may have been copied in the late thirteenth century.
Thus, although the earliest authority for the episode is a text copied a
century later than the event reported, the transmission of the text seems
reliable. It is perfectly plausible that a ceremony of declaration of
allegiance to Hisham took place along the lines which Ibn al-Khaub
described. Yet almost all but the first ten and the last two of the people who
were supposed to have been there could have been assembled by
Ibn al-Khatib from a grand trawl of a biographical dictionary covering
three centuries. Whilst one would like to use this passage as showing
Ibn al-Qjitiya as a prominent Umayyad loyalist, closer examination of the
evidence undermines this interpretation. Similar care must be taken in
reading all the sources for this period.
Ibn al-Quttya was listed in several other dictionaries.!" The similarities
between their entries on Ibn al-Qptiya and al-Faradi's makes it likely that
they are all derived from al-Faradi. Biographical compilers were even more
likely than other writers to copy from earlier works, and it is unusual to find
two accounts of the same subject which are independent of each other. I I
Yet the authors did not copy their predecessors word for word.
Ibn Khallikan, in a biographical dictionary completed in Cairo in 1274,

162
 SARA THE GOTH AND HER DESCENDANTS

included an entry on Ibn al-Qjitlya which is much longer than al-Faradr's."

It begins with a summary of the information attributed to al-Faradi but
goes on to include a different version of the story of Sara the Goth from the
one that appears in the History if the Conquest if al-Andalusl? This story will
be considered later. Ibn Khallikan attributed the name Ibn al-Qpttya not
to Sara the Goth, but to Ql1t ibn Ham ibn NI1!), one of the grandsons of
Noah. Ibn Khallikan also told the following anecdote:
'[a writer and poet] relates that he was on his way one day to an
estate he owned at the feet of the mountain of Cordoba, one of the
most agreeable and lovely spots on earth, when he happened to meet
Abt; Bakr ibn al-Qntiya returning from the place, where he also
owned an estate . He says, 'On meeting me he halted beside me and
expressed pleasure at our encounter. I playfully improvised the
following verse: "Where are you coming from , unique one , sun for
whom the world's a sphere?" He smiled and promptly answered:
':-\ place whose solitude astonishes the hermit, and where rascals can
be themselves in privacy" . I could not refrain from kissing his hand,
as he was my old master whom I honoured and revered'.
It is possible that this is a genuine memory of Ibn al-Qnttya which the
copyists of the other dictionaries, including that of al-Faradi, chose not to
include in their entries on this scholar - a representative anecdote which
sums up the personality of the man in a few lines. It is much more likely,
however, to have been interpolated from a model, perhaps by Ibn
Khallikan himself.
On the question of Ibn al-Qjinya's status as a historian, not all the
biographers shared al-Faradi's harsh judgement. Ibn Khallikan said of
Ibn al-Qutiya that 'he was the guardian of hadua andfiqh ...'. "Iyad was
able to describe Ibn al-Qjrtiya both as 'the guardian of the stories [akhbar]
of al-Andalus and the lives of the emirs and the affairs of the scholars ' and
as the author of 'an excellent book on her history'. 14 Comparison between
other entries in 'Iyad's dictionary and the surviving copy of al-Faradi's
indicates that 'Iyad was using a manuscript of al-Faradi's dictionary, but
was either adapting it, or using a copy which is not the same as the
surviving text. 15 Three compilers of biographical dictionaries - Yaqut,
writing half a century before Ibn Khallikan, and the later authors
al-Dahabi (d. I348) and al-Safadi (d.1363) all said that Ibn al-Quuya wrote
a book of history [tanKh] . It is possible that these authors have all
misunderstood al-Faradr's comment. Yet Ibn al-Qjinya was remembered as
a historian in other, independent, sources . Ibn al-Abbar quoted Ibn
Hayyan, who in turn had cited Ibn al-Qjuiya, as the source of a story

163
 SARA THE GOTH AND HER DESCENDANTS

which does not, however, appear in the History if the Conquest if al-Andalus. 16
Ibn al-Qjltlya was also cited several times by al-Khushani (d.971) in his
Book if the Judges if Cordoba." although here again the extracts do not
correspond with the History. In al-Maqqan's collection of sources for
al-Andalus, Ibn al-Qjitlya is praised at one point for his industry and
veracity, but criticised at another because he took notes from a visitor to
Cordoba who later turned out to be a fraud." Only one passage in the
History of Ibn al-Qnttya is reproduced in the work of a later historian. It is
an account of the return of Miisa ibn Nusayr to the east which appears
in the last three folios of a manuscript from El Escorial, which has been
attributed to Ibn Fayyad." Since the author of this text did not name
Ibn al-Qjitiya as his authority for this episode, and his narrative is more
detailed than that of Ibn al-Qptiya, it is equally plausible that the two men
were using a common source. One should have reservations about relying
on the surviving manuscript of al-Faradi as establishing either that
Ibn al-Qjitlya did not write history, or that he did not transmit it correctly.
The manuscript of the dictionary may itself be a miscopied or altered
version. The statement that Ibn al-Qjitfya was a poor historian may have
been interpolated by a later copier of the dictionary who had his own
opinions. Similar questions about textual transmission arise when
considering the History if the Conquest if al-Andalus.

The History of the Conquest of al-Andalus:

history or fable?

The History survives in only one manuscript.i" now in Paris, which dates
from the fourteenth century. In the nineteenth century, Cherbonneau
made a partial translation from this codex and another manuscript, from
Istanbul, which has since disappeared." The Paris codex also includes the
sole surviving copy of the Akhbar Majmll'a 22, a collection of historical
traditions which may have been written down at the same time as the
History.23 The manuscript names the author twice. It begins: ~bii Bakr
Muhammad ibn 'Umar ibn 'Abd al-Aziz related to us .. .'. The work ends :
'Here ends the history of Ibn al-Qptlya, Praise be to God'. Thus the
attribution of the text to Ibn al-QiiJIya seems secure . The problems arise
from the structure of the work and the way in which it was transmitted.
The History if the Conquest if al-Andalus is taken less seriously as a source for
the tenth century than other Arabic sources, even though it may be one of
the earliest. This is less because of the anecdotal nature of the work, which
in this respect is merely more extreme than others of this genre, than

164
 SARA THE GOTH AND HER DESCENDANTS

because Ibn al-Q iitlya, contrary to th e usual practice, did not quot e long
passages from his predecessors. At the beginning of th e History Ibn
al-Qjitlya cited four scholars as his autho rities. T wo of these were two of
th e teachers listed by al-Faradi, a third is known from other biographical
dictionaries but th ey wer e not describ ed as historians. Ibn al-Q utiya did
not quote from any of th ese men exac tly but used th e vague formulae, 'it
was said', 'some of th e scholars said' etc. For this rea son , mod ern scholars
have echoed th e criticisms levelled at Ibn al-Qiitlya by al-Far adi. Chalmeta
rejected the History's account of the conquest. H e classified th e History as
akJzbar,24 distinguishing it from tar'ikh, which he defin ed as the careful
transmission of hadflh - exac t copying from on e's predecessors in a cha in
stre tching back to th e events being describ ed . Akhbar, in contrast, is the
m er e selecting of stories from oral tradition . Chalm eta described
Ibn al-Qiitlya as 'an extreme case .. . of ora l transmission .. . a walking
fossil, a living anachronism'. The term 'ora l tradition' has been used to
acc ount for a multitude of th e problems po sed by narrative sources written
down many years after the events the y portray. It is not clear exactly what
is meant here but Chalmeta implies that, as a man who eschewed written
history, Ibn al-Qutlya felt at liberty to rework old stories as it suited him.
Less pejoratively, it could also me an that Ibn al-Q iitlya's stories continued
to be passed on by word of mouth for several gene ra tions befor e they were
written down. This reduces th eir potential as evide nce for 'wha t actually
happened'.
Ther e m ay be som e sub stan ce in th e acc usa tion th a t the History qf the
Conquest of al-Andalus owes its imp erfections to a reliance on or al tradition .
One of th e episodes whi ch Ibn al-Qutiya recounted ab out Artubas, one of
the sons of Witiza, begins , 'and one of th e stories [min al-akhbar] about
Artubas . . .' 25 , said to be cha rac teristic of akhbdr. So is the History's paucity
of dates and th e inclusion of po etry. Anthropologists have shown, however,
that oral transmission is not synonymous with inac curacy, mu ch less with
fantasy. Although Ibn al-Qjtttya may have be en th e first to write down
th ese stories, th ey may nevertheless hav e been known in standa rd versions.
Furthermore, it is likely th at he was not plu cking th em at random, but in
orde r to fill in some of th e gaps in th e texts available to him . This is
impossible to prove, because th e texts he nam es have not surv ived. On the
marriage of Sara the Goth he said: 'this acco unt, or most of it, is found
in th e Bookon the Conquestqf al-Andalus of 'Abd ai-M alik ibn H abib and in th e
po etry of th e vizier Tamtrn ibn Alqama. ' 26 The story of Sara th e Goth, for
which Ibn al-Qjiuya cited Ibn Habib as his source, is not mentioned by
Ibn Habib in the on e surviving copy of his History, but som e of th e poetry of
Tarnim ibn Alq ama (803-86) was quoted by Ibn al-Abbar." It covers th e

165
 SARA THE GOTH AND HER DESCENDANTS

period from the arrival of Tariq ibn Ziyad in 7 II to the reign of 'Abd
al-Rahman III, and could have served as the basis for this section of the
History.
Ibn al-Qjttiya included several stories which are almost certainly
without historical foundation; these include the betrayal of Hispania to the
invaders by Julian as revenge for Rodrigo's violation of his daughter, and
episodes from the conquests of Tariq, the arrival of Miisa and the rivalry
between the two men. There is a fund of stories about the conquest.
Rubiera attempted to make historical sense of one of these which Ibn
al-Qutiya did not relate, the discovery of the Table of Solomon in Toledo,2a
but most are clearly legendary. Some of these fables were borrowed in their
entirety from outside Hispania. At least one of the episodes in
Ibn al-Qjttiya's History has a prototype in eastern lcgend.P They may
have been based on topoi which were adapted to the history of al-Andalus
to glorify the Umayyads. Many of these stories came to Hispania from
Egypt in the eighth century and later.'" although the stories of the house of
Witiza's descendants cannnot be found in any of the surviving Egyptian
manuscripts. Ibn al-Qunya may have heard some of the stories of the
conquest from his own family, but he did not mention any such family
traditions. Further it is precisely at the point where he introduced the
exploits of the sons of Witiza and the descendants of Sara that he said he
was using written authorities." Nor did he quote the charter which
al-Wand was said to have given to the sons of Witiza , confirming them in
possession of their lands . One might have expected that this document, or
a memory of its terms from which its wording could be reconstructed,
might have been preserved by his family. The fact that he did not have
access to such material suggests that there had been a period of forgetting
between the conquest and the tenth century which Ibn al-Qjitlya had to fill
from other sources. Although the question whether he used written sources
remains unresolved, to dismiss Ibn al-Qjitlya's History as purely anecdotal is
too harsh, and places too much emphasis on modern scholars' ability to
distinguish between akhbdt and tar'ikh.
Further inaccuracies arose from the process of transmission from Ibn
al-Qjitlya to the manuscript. The formula 'he related to us' [akhbar-na] with
which the History opens, often implies oral transmission, although this is not
invariably the case. This fits with al-Faradr 's statement that Ibn al-Qutfya
dictated his stories from memory. The biographical dictionaries have
preserved the names of several of the students who may have taken down
Ibn al-Qatiya's words. The History could have been transmitted by Ibn
al-Qnuya's son, called Abu Hafs 'Umar, who referred to traditions which
he heard from his father, or by his nephew 'Abd al-Malik ibn Sulayman

166
 SARA THE GOTH AND HER DESCENDANTS

(d.429/1037) 'a narrator of histories who would refer for authority to his
uncle Abu Bakr' .:12 Yet the surviving text was written down by someone who
was not a relative of Ibn al-Qutlya because it says of ' Isa ibn Muzahim, the
first husband of Sara the Goth that 'he was an ancestor oflbn al-Qppya',"
where the writer might have been expected to say 'my ancestor' if he
belonged to the same family," The apparently inconsequential nature of
some of the anecdotes, at least for modern readers, and the random way in
which they have been collected have been attributed by modern historians
to a none-too-bright student. Since the surviving manuscript of the History is
so late, it is impossible to say what relationship it bears to what might have
been dictated and perhaps written by Ibn al-Qutrya.
In the context of Arabic literacy, dictation does not in itself invalidate
the History as a source, since, according to later authors, this method of
transmission was more valid than copying. Students sat at the feet of a
master, who dictated from memory. Even though the resulting work could
be read by all, a student gained status by becoming part of the chain of
witnesses to a master's own words . Ibn Bashkuwal mentioned verses by
another Cordoban scholar, Ibn Sa'Id al-Tamimi, in which he expressed his
satisfaction at seeing himself surrounded by a thousand students in the
great mosque, each one taking down what he had dictated." Having
written the text down, the student was to obtain a certificate that his copy
was accurate, and that he was licensed to teach from it. This seems to have
been the ideal, and the practice may have fallen short. Both teacher and
pupil could be at fault. Ibn al-Qutrya was not the only historian criticised
for sloppiness. The work of another historian of al-Andalus, Ibn Habib
(d.852), poses similar problems of interpretation to the History ofthe Conquest
cf al-Andalus, and may usefully be compared with it. Ibn Habib's History
survives in one copy which, like the History of Ibn al-Qjttlya, is thought to
be a pupil's notes rather than the original. In this case, the supposition
stands on a firmer basis, since the surviving manuscript carries the name of
Yusuf ibn Yahya al-Maghami, who died in 901. To an even greater extent
than the History cf the Conquest cf al-Andalus, the History of Ibn Habib is a
mish-mash of anecdotes with little apparent basis in reality. The pupil may
not be entirely to blame. Al-Faradi accused Ibn Habib himself of a number
of breaches of the rules of transmission, including an inability to distinguish
true haditlt from false, and of soliciting authorisation to teach the works of
his masters by obtaining copies of them, without going through the process
of hearing, reading and checking them with their authors." If al-Faradi, or
later interpolators of his dictionary, could be so critical of the reliability
even of the authors themselves, it seems that the process of transmission of
information about the past was already considered unsatisfactory. This

167
 SARA THE GOTH AND HER DESCENDANTS

judgement is increasingly being adopted by modern historians. It is

possible that Ibn al-Qjltiya was indeed telling his students stories which
have been condemned as fabulous by his contemporaries and that he did
not always tell them the same way. If, in addition, his students were not
taking down their master's every word, the interpretation of a single
surviving manuscript becomes extremely difficult. It is very probable that
several versions of the stories attributed to Ibn al-Qutfya were in
circulation, interpolated by his pupils and later generations.
Thus there are enormous problems in accepting the surviving copy of
the History as a tenth-century work, even at one remove from its author.
Any interpretation of the History must be provisional, and must always bear
these reservations in mind. I am working on the assumption that the History
has a tenth-century perspective on the conquest and its aftermath. Like a
christian chronicle, this was an edited version of history. The compiler of
the History who, for the sake of argument, I will call Ibn al-Qjitlya, was
being selective. By his inclusion of so many anecdotes about the christians
who participated in the establishment of al-Andalus, he was giving form
to the history of those christians whose descendants were later to convert to
Islam and became part of the establishment. The context for such an
attempt was an awakening of interest in the writing of history and
genealogy in al-Andalus. Before studying some of the episodes recounted
by Ibn al-Qjitiya, it is necessary to consider this context in more detail.

History and genealogy

Although Ibn Habib wrote his History in the ninth century, the real
historiographical horizon for al-Andalus, the period when histories and
genealogy were being collected and formalised for written rather than oral
transmission, was the middle of the tenth century. The first Andalusi to be
dignified with the title 'the historian' was Ahmad al-Razi (d.955) whose
son 'Isa said of him that he 'collected data from old people and
transmitters of reports, which he collated and organized into a history. He
was thus the first to codify the rules of historical composition in Hispania.
His work brought him closer to the sovereign and earned himself and his
son a greater measure of royal favour. Together they endowed the
Andalusis with a science they had not hitherto practised with success' Y
Like his contemporaries, ar-Razi concentrated on the exploits of the
Umayyads. Most of his work does not survive, but it was quoted
extensively, and his description of th e famous towns of al-Andalus was
translated into Romance." Interest in history was stimulated by a request

168
 SARA THE GOTH AND HER DESCENDANTS

from al-Hakam II for scholars to work on the history of al-Andalus,

probably made during the reign of his father. Al-Khushani said of
al-Hakam that he 'conceived the excellent plan of initiating the study of
history [and] the knowledge of genealogies, and he wished for the merits of
the ancestors to be published'. It was necessary to write down 'the
dispersed traditions destined to be lost'. 39 Ibn al-Abbar reported
al-Hakam's demand 'that all those Arabs whose lineage, tribe or family
had been lost or had been relegated to oblivion before his reign should
reconstitute the memory of it and, once the nobility of their ancestors had
been established by persons competent in this subject, they should inscribe
it in special registers so that everyone was aware of his own lineage'.'? The
context for the Umayyads' promotion of history can easily be found in the
establishment of the caliphate in 929 . In a letter to his provincial governors
stating his intention of taking the caliphal title , 'Abd al-Rahman
complained that two centuries of silence had cast doubt on the legitimacy
of the Umayyads. The letter was reproduced by Ibn Hayyan and the
anonymous author of a chronicle on the reign of'Abd al-Rahman, Neither
text survives in a contemporaneous manuscript, but the episode testifies to
the importance of history and genealogy to the Umayyad caliphate."
Books of genealogies became very popular in the Middle East in the
ninth century. The earliest to survive is the work of Hisham ibn
Muhammad al-Kalbi, who died c.8l9. 42 Although the islamic genealogies
have been used in prosopography, their construction has only recently been
studied using the methodology which had already been applied to similar
works written down in northern Europe at about the same time. General
conclusions have been drawn from comparisons between written
genealogies of the early Middle Ages and the oral traditions of pre-literate
peoples gathered by anthropologists.'? These elucidate the effect of
ideology on the process of transmission. Genealogies may be accurately
transmitted from one generation to succeeding generations without being
true. They are reconstructions of the past, which may be used in a number
of ways to describe the present; they may state the supposed ethnic origins
of the group and evoke political stability by giving a sense of historical
inevitability to current alliances. Important events in the history of the
group, such as migrations, were commemorated in the name of a famous
ancestor, who might be mythical." Thus it is necessary to know the context
of a genealogy, when and where it was written down, in order to make
sense of it. Kennedy examined the work of Hisham al-Kalbi from this
perspective, assuming that it marks the transition of islamic genealogies
from oral to written record and that the Bedouin of the twentieth century
remember their genealogy in the same way as the tribes of the pre-literate

169
 SARA THE GOTH AND HER DESCENDANTS

centur ies of Islam had done." Yet he also pointed out th at islami c
gen ealogi es are very differ ent from those of the Anglo-Saxons, for exa mple.
The scale of gen ealogical writing, in particular, is unlike anything preserved
in northern Europe, (except in Irel and) where early medieval genealogists
wer e mainly concerned with documenting royal descent. H isham al-Kalbi
recorded some 35,000 names in his book. Nor can th e writt en genealogies
easily be compared with th e memories of their ancestors which the
twentieth-century Bedouin preserve. Yet som e of the same considerations
can be seen to apply. An important function of the islamic genealogists was
to extend the chain of ancestors back to th e pre-islamic period. In his Lift rif
Muhammad." Ibn Ishaq gave his subject the most illustrious of an cestors,
tracing the prophet's descent from Adam through forty-eight gener ations,
just as Anglo-Saxons were doing in the genealogies of their kings. Although
such evide nce was used to arbitrate in legal disputes, it is clear that
genealogy was not merely a political instrument but formed a branch of
knowledge for its own sake. Kennedy liken ed genealogies to hadflh - cha ins
of knowl edge whi ch did not allow for any gaps. The conce pt of genea logy
as a science m ay be particularly relevant to gene alogical writing and
history in al-Andalus,
In al-Andalu s, the earli est known writers of genealogies were Ibn H abib,
who was said to hav e written two books, a Genealogy andhistory of the OJirashis
and Genealogies, lawsandstudies ofthe Arab s , ~ 7 and QasIm ibn Asbagh (d.95 I),
who wrote a Book ofGenealogies. None of th ese works ha s survived, but th ey
are mentioned in the biographies of th eir authors and severa l other
scholars of this period wer e remember ed as genealogists. Ibn 'Abd Rabbihi
(d .940) dedicated a chapter of a work called The Unique Necklace to
genealogies." Despite al-Hakarn's initi ative , all these gen ealogies so far
mentioned seem ed to have be written about th e Arabs of th e Middle Ea st,
not about al-Andalus, and relied heavily on Ibn al-Kalbi and others . The
earli est work to include Andalusi genealogy is the ]amhara, written after the
fall of th e Umayyads by Ibn Hazrn, who claimed to be relyin g on two
books of genealogies, now lost, by Ahmad al-Razr, Like his predecessors,
Ibn Hazm began with pre-islamic Ar abia and th e ge nealogy of
th e descendants of 'All, the Abbasids and the Umayyad s. H e included
th e names of the men of the se famili es who had passed to al-Andalus. Yet is
surprising how few gen ealogies of th e em inent m en of al-Andalu s could be
reconstructed in full; among th em wer e th e Banu H ajja] and Banu
Khaldun of Seville and th e Tujibis of Zaragoza.
From th e ]amhara and other sources , it is clear that th e pit falls of
gen ealo gy as a hi storical so urce a re only too well illustrat ed in
Hi spania.P Natives of the peninsula manufactured islami c genealogies

170
 SARA THE GOTH AND HER DESCENDANTS

for themselves, and these genealogies were adapted to ch anging

circumstances.50 Bizarrely, the inhabitants of two castles in the Duero
valley who were said to have made a treaty with the conqueror Musa ibn
Nusayr, claimed Arab origins in spite of being christians.51 Ibn Hazm
himself may have been guilty of reinventing his origins. He said that his
ancestors had been muslims since the seventh century, when they became
clients of the Umayyads, and that one of them passed to al-Andalus in the
eighth century and was given land in the Algarve . Yet the historian Ibn
Hayyan, a contemporary oflbn Hazm, maintained that Ibn Hazm's family
were recent converts to Islam, and that Ibn Hazm's father was the first to
achieve any status in al-Andalus." Even for the muslims whose origins in
Arabia did not have to be fudged, there was a problem in claiming any role
for their family in the conquest of Spain, since it is likely that the majority
of the men who entered Spain in 711 were Berbers, and that few Arabs
arrived in Spain until well after the Umayyad takeover of 756. 53 Many of
the genealogies were more illustrious in their reconstructions than in
actuality and the subject of ethnicity in al-Andalus became hopelessly
confused. The fourteenth-century author Abu-l-Walid ibn Nasr of
Granada made an attempt to clarify the matter: 'In the beginning, there
were the Banu Hashim [the family of Muhammad ], who came from
Arabia, Iraq, Syria, Egypt and north Africa; secondly the noble Arabs and
their clients; thirdly the Berbers, who arrived from North Africa in large
numbers; and in fourth place the natives of the country, many of whom
converted to Islam , but others kept their faith and lived as conquered
subjects . To this race the Romans, Galicians, Castilians, the people of
Aragon, the Goths and Franks, etc, belonged. The jews also belonged to
this group; some of them had resided in the Peninsula since before the
muslim invasion, but others arrived later'.54 This jumble of anachronisms -
there was no Castile or Aragon at the time of the conquest, and terms such
as 'Roman', 'Frank' and 'Galician' had different meanings in different
periods and between one author and another - accurately reflects the
confusion of the sources . More than two centuries had elapsed between the
conquest of al-Andalus and the writing of the History of Ibn al-Qjtttya, Its
author is unlikely to have been as certain of his genealogy as he claimed; if
he was insisting on his christian origins, it must have been for a purpose.
The role played in the 'ethnogenesis' of the Islamic world by adherents
of other faiths who converted to Islam and became clients of islamic rulers
is difficult to state clearly. From the earliest years oflslam, there was tension
between the integration of outsiders into the new religion, and the
demands of maintaining an illustrious lineage. By the eighth century,
converts to Islam who became clients of the Umayyads or of the islamic

171
 SARA THE GOTH AND HER DESCENDANTS

nobility were participating in all levels of government, both in the east and
in al-Andalus, and tribal structures were beginning to break down in
favour of equality between those of Arab and non-Arab origin.P This was
reinforced by the adoption of Arabic as the official language of
bureaucracy and scholarship, so that to some extent the use of Arabic,
in conjunction with conversion to Islam, became synonymous with being
an Arab - although this identification broke down in those conquered
lands, such as Hispania, where the indigenous populations adopted Arabic
without necessarily accepting Islam. The rhetoric of the genealogies,
however, insisted on purity of descent as the criterion for judging a man's
right to call himself an Arab. The idea that the Arabs were a single nation
developed in the east in the first half of the eighth century. Arabs were to be
distinguished on the basis of biological continuity with those to whom
Muhammad's revelation had been granted. Other nations might share this
honour by virtue of having joined the Arabs, but they remained
newcomers. Such ideas led to learned argument about the exact status
of clients. Since the main focus of Ibn al-Qjitiya's History is the
Umayyads, the fact that he wove into this story the history of the family
of one of their clients is a significant comment on the status of such
people in al-Andalus.

Christians in the History of the Conquest of al-Andalu«

The History ifthe Conquest ifal-Andalus is an exemplary text and many of the
stories are qualified by explanatory notes, such as 'the reason for this was'.
The example which its author was putting forward was that of the
Umayyads, whose worthy deeds are the focus of the text. Ibn al-Qjitfya
managed to maintain his pro-Umayyad line even when recounting
episodes in which his sympathies might have lain with rebels against the
emirs, such as the revolt of the Arrabal in Cordoba against al-Hakam I,
which included many of his fellow scholars.56 He exonerated al-Hakam I for
the severity with which he suppressed the revolt because of the piety of his
later life. Ibn al-Qjitlya had little time for the lists of campaigns and
description of magnificent building programmes which are the stock-in-
trade of other histories of the Umayyads. He demonstrated the nobility of
the emirs and caliphs by the wisdom which characterised their
appointments of judges and military commanders and by a number of
anecdotes illustrating their justice and clemency and generosity, showing
the importance of faithful courtiers and scholars for good government.
Hisham I (788 -96) was described thus: he 'undertook the shepherding of

172
 SARA THE GOTH AND HER DESCENDANTS

his subjects with the utmost that a governor could practise of benevolence,
equity, modesty, visitation of the sick and attendance of funerals; also he
caused the abolition of tithes and the collection of alms and chose
simplicity in his dress and mount' r'"
Some of these tales appear extremely slight, and to read the History as a
continuous narrative invites irritation with the trivial nature of the author's
concerns. The following episode is typical:
'One day an accident befell Hisham (God have mercy upon him) as
he was returning from the funeral of Ta'laba ibn 'Ubayd to the dead
man's home. A dog rushed him from a house which bordered the
well-known cemetery of Quraish. It seized the lined cloak of Nerv
cloth that he was wearing and ripped it. He said, "The governor of
Cordoba is ordered to fine the owner of this house one dirham
because he has let loose a dog in a place where muslims may be
harmed by it" . But after he left the house of Ta'laba ibn 'Ubayd he
commanded the cancellation of the fine saying, "We have surely
distressed the owner of the house more than we are bothered by the
tearing of the garment"'.58
The reader may be forgiven for feeling disappointed that this is how Ibn
al-Qjitiya chose to represent the emirs . The History, however, invites
comparison with the hagiography of christian saints. Just as the
hagiographer illustrated the christian life with the actions and miracles
of his subject, so the author of the History is putting forward the Umayyads
for emulation, showing how their worthy deeds kept the dynasty in power,
and their realm at peace. Power depended on just government, which
meant, amongst other things, good relations with all one's subjects, even in
such a small matter as a torn cloak. This extended to the emirs' christian
subjects and converts to Islam. In particular, Ibn al-Quttya was making
specific claims for his own ancestors, which have not been appreciated in
previous accounts of the History. Chalmeta saw the anecdotes about Sara
the Goth and the sons ofWitiza as 'the strongest proof of the ignorance of
Ibn al-Qptiya, and the most powerful demonstration of the extremes to
which the fatuous desire to aggrandise his anodyne ancestors could drag
him',59 erasing from his mind everything he had ever known about the
'real' history of al-Andalus. The truth content of these stories is indeed
negligible . Yet they illustrate different aspects of the process of coming to
terms with the conquest. At the risk of becoming as anecdotal as the
original, I will illustrate this point with a number of extracts from the History,
where each reference to a christian makes a point about the integration of
the christians in Andalusi society.

173
 SARA THE GOTH AND HER DESCENDANTS

The first episode relates to th e defeat of Rodrigo:

'The last king of the Goth s of al-And alus, Ghitisha [Witiza] left on
his death three sons: the eldest was called Alamundo, after him came
Romulo, then Ar tubas, Because at the time of the death of their
fath er they were min ors, their moth er remain ed in Toledo rulin g the
kingdom. Rodrigo, who was a genera l appointed by the former king,
left the cour t, followed by the military men who were under his
command, and established himself in Cordo ba . . ..
Wh en Tariq ibn Ziyad entere d Hi spani a, in the time of the caliph
al-Walid ibn 'Abd aI-Malik, Rodrigo wrote to the sons of the king,
who were alrea dy handsome youths and able to man age a horse,
inviting them to help him and unite with him against the commo n
enemy. They gathe red the fronti er troops and march ed out, camping
in Secunda; they did not dare to en ter Co rdoba becau se they did not
have complete confidence in Rod rigo: the latter had to leave the city
to join up with th em .
They set out immediately to meet Tariq, and when the two ar mies
were on th e point of joining bat tle, Alamundo and his br oth er agree d
amo ngst them selves to betray Rodrigo. T he very same night , they
sent m essengers to Tariq to inform him that Rod rigo was no more
than one of their fathe r's vilest vassals, and to ask him for his
prot ection in orde r th at th ey might cross to his camp on the following
morning. They asked him to confirm and assure their possession of
the inheritan ce that their fathe r had held in Hi spani a. T hey were
three th ousand villas, which would later be know n as "the royal
villas". At dawn they crossed over to Ta riq's camp with the troops
which they had brought with them . This was the cause of the
co nq uest' i'"

R ath er th an seeing the defection of the sons of Witiza as part of the 'ru in of
Sp ain ' set in motion by Rodri go, Ibn al-Qjniya, as might be expec ted,
portrayed their action as legitim ate. The prin ces' new status was confirmed
in writing, first by Tariq, then by M usa ibn Nusayr in No rth Africa and
finally by the caliph him self, in terms which left neith er their landholdings
nor their roya l status in doubt :
'When the three brothers came before Tariq, they said to him, "Are
you your own chief or is there a chief over you?" Tariq answered
them , "T here is a chief over my head an d over his ano the r". H e
permitted them to join M usa ibn Nusayr in Ifriqiya so that their
relation with him might be confir med. They requested of Tariq a

174
 SARA THE GOTH AND HER DESCENDANTS

letter to Musa regarding their affairs with Tariq and the pledge that
he had given them. Tariq complied. They set out toward Musa
carrying Tariq's letter explaining their agreement of submission and
the conditions granted them. They met Musa in his descent toward
al-Andalus in the vicinity of the land of the Berbers. Musa ibn
Nusayr sent them to al-Walid ibn 'Abd al-Malik, They went to the
caliph who ratified for them the pledge of Tariq ibn Ziyad. Along
with that he concluded a written contract with each one of them'."
Under the terms of the settlement, each of Witiza's sons received a
thousand villages. The claim that one's ancestor had sworn allegiance to
the invaders was the nearest equivalent available to an indigenous family of
having an ancestor who had come to the peninsula with the conquerors.
Yet Ibn al-Qjitlya was doing far more in this passage than claiming that his
family had always been loyal to the regime . He is showing that this
situation was acceptable to the heirs of the Visigoths because they did not
lose all their land to the invaders.
This episode echoes the treaty ofTudmir which was drawn up between
'Abd al-Aziz, the son of Musa ibn Nusayr and Theodimir, the lord of seven
towns in south-east Hispania. The earliest surviving copy of this treaty is an
interpolation into the tenth-century manuscript of the Chronicle if 754.62
The text was also preserved in three Arabic works, including a fourteenth-
century biographical dictionary. Theodimir was a powerful warrior who
finally decided to make peace with the invaders. Like the sons ofWitiza, he
went east to have his settlement with 'Abd al-Aziz ratified by the caliph:
'The Amir Almuminin [prince of the believers] found him to be more
prudent than the rest, and treated him with honour. The pact which
he had made with Abd al-Aziz a short time before was firmly
renewed by him. So he returned to Spain rejoicing and remained
secure from then on, so that in no way were these powerful bonds
dissolved by succeeding Arab rulers '.
In the Chronicle if 754, the treaty is dated 5 April 713. There is, however, no
firm evidence that it is genuine. The terms of the treaty, protecting the
persons of those who submit peacefully, and allowing them freedom
to practise their faith in return for the payment of taxes , are similar to those
quoted in histories of the conquest of oth er islamic lands , such as the
Conquests of al-Baladhuri (d.892).63 These treaties were already the subject
of controversy in the eleventh century and continue to provid e subject for
debate in the twentieth. It is not clear whether they reflect uniformity in
the terms of treaties being imposed throughout the islamic lands from early

175
 SARA THE GOTH AND HER DESCENDANTS

in th e conquest, or a later feeling that such terms ought to have been

imp osed . T he Ara bic historian s may have been m akin g retrospective
claim s for the civilised manner in which th e conquests had been carried
out by invokin g th e conditions for surre nde r said to have been laid down
by ' Umar II . Ibn al-Qjitiya's story of the preservation by the sons ofWitiza
of th e Visigothic royal esta tes is very sim ilar to that of Theodimir and he
m ay have known th e story or have seen a version of the treaty. Ther e may
be a histori cal basis for Theodimir but th e preservation of thi s and other
similar passages shows bo th christian and muslim writ er s explaining the
founding of a sta te share d between peopl es of differ ent or igins and faiths.
The charter given to th e sons of Witiza fulfils th e same fun ction in the
History if the Conquest if al-Andalus.
Ibn al-Q upya mention ed another Visigothic famil y which wea thered
th e conq uest by allian ce with the muslim governors. These wer e the
desc endants of th e count Cass ius, known as the Banu Q asI, who ruled in
the Ebro valley from th e ninth century.'" According to Ibn al-Q ntlya, QasI
[Cassius] joined forces with M usa ibn Nusay r and Tariq ibn Ziyad,
per haps in 713 or 714, went to Damascus, and convert ed to Islam.f
Ibn H azm listed his sons as For tu n, Ab u Tawr, Abu Salam a, Yimus and
Yai)ya.66 Caiiada J uste tr aced their descen t th rou gh Fortun and M usa ibn
Fortu n to M usa ibn Musa, the rul er of Zaragoza from 852 . Their origins,
as recounted by Ibn al-Q ullya , may be a product of 'the spurious
antiq uarianism of th e lat er Umayyad period ' , as Collins put it,67 ra ther
th an reliable genealogy bu t, like the deeds of the sons of Witiza, they
satisfied a need for stories which bridged th e conquest.
The seco nd episode to be studied in detail is a story abo ut Artubas, the
third son of Witiz a, and relates to th e period of th e esta blishme nt of 'Abd
al-Rahman I in Hi spani a, wh ich m ark ed a new stage in th e conquest and a
second turning point in the history of Ibn al-Qullya's famil y. The evidence
for Artubas outside the History is very slight, altho ugh he could be the Ardo
(713- 20) who succee ded Achila III (710- 13) as th e last Visigothic ruler of
Septimania." Acco rding to Ibn al-Qjniya , Artubas lived in Cordoba, and
held lan ds in th e cent re of th e pen insula which 'Abd al-Rahrna n coveted :
'Amo ng th e stories of Artubas , [it is said th at] 'Abd al-Rahman ibn
M u'awiya or de red th at his land ho lding be confiscated and the cause
of th is was that this m on arch looked over Artubas' m an or one da y
while he was out on an exped ition, in which this man was
accompanying him , and round abo ut the same he saw not a few of
the gifts which the vassals would usually have offered to him during
the halts whi ch he m ade in the villages of his dom ains. This made

176
 SARA THE GOTH AND HER DESCENDANTS

him jealous. They were , therefore, confiscated and Artubas had to go

and live with his nephews, ending up in penury. He then made his
way to Cordoba, visited the chancellor Ibn Buht and said to him :
"Do me the favour of asking the emir, God protect him, permission
to see him, because I have come to say goodbye". The chancellor
went to 'Abd al-Rahman and he ordered Artubas to come before
him. On his entry he saw that he was dressed in rags. He asked :
"Hello, Artubas! What brings you here?" To which he answered:
"You do, who has come between me and my possessions, defaulting
on the treaties which your ancestors made with me , without any fault
on my part which might authorise it" . 'Abd al-Rahman continued:
"What's all this about wanting to say goodbye to me? You're surely
not going to Rome?" Artubas answered: "Quite the opposite, man!
I heard that you were going to Syria!" 'Abd al-Rahman replied: "And
who will allow me to return there, seeing that I had to leave to escape
being murdered?" Then Artubas asked him: "Have you the intention
of consolidating your hold on this land so that your son can inherit,
or do you want to deprive him of what has been given to you?" 'Abd
al-Rahman replied: "0, no, by God! not only do I want to strengthen
my dominion, but also that my son should inherit". Then Artubas
told him: "Then see how this matter can be arranged". Then he
denounced roundly, without beating about the bush, all the matters
which offended the country and 'Abd al-Rahman was so grateful that
he ordered that twenty of Artubas' villas should be returned to him,
he lavished on him splendid vestments and gifts and gave him the
office of count, the first to occupy this dignity in al-Andalus'i'"
The story goes on to underline the equation between landholding and
justice. A certain poor man, a client of a family of Arabs from Syria, asked
Artubas if he could work one of his villages in return for a share of the
produce because, after the overthrow of the Umayyads in the east , he
found himself unexpectedly obliged to settle in al-Andalus. Artubas gave
him the land outright, together with its livestock and slaves and a castle.
Ten Syrian nobles who were present at this transaction were appalled that
Artubas was squandering his property on a poor man rather than giving it
to them. After delivering a short homily on the importance of charity both
in Islam and in Christianity, Artubas gave each man a hundred villages;
Ibn al-Qjitiya listed some of these gifts. It is not clear what relationship
these events had to actual landholdings in either the eighth or the tenth
centuries. The Chronicle if 754 reported that th e governor al-Hurr
(716-18) 'restored to the christians the small estates that had originally

177
 SARA THE GOTH AND HER DESCENDANTS

been confiscated for th e sake of peace so as to bring in revenue to the

public treasu ry' ." There is no other evide nce that Ibn al-Qunya or other
clients of th e U mayyads claiming descent from th e Visigothic royal hou se
still held an y of th e royal estates. Indeed , Ibn al-Qjitfya noted that the
suppo rte rs of th e Umayyads were settled throughout Sp ain 'a t the expense
of th e non-muslim protected peoples, and th e first immigrants and th e
Berbers [also] retained all of th eir spo ils'." Perhaps th e descendants of
th e Visigothic nobility were susta ined in th eir present impoverishment by
th e fiction th at , like Artubas, they had proudly handed over th eir
patrim on y. H ere again, Ibn al-Qjr ttya was talking up th e status of his family
by evoking tr eati es whi ch his ance stors had made with th e conquero rs.
The further progress of good famili es of christian origin may have been
in Ibn al-Q u pva's mind in recounting th e followin g an ecdote." When 'Abd
al-Rahrna n II died , his son Muhammad kept on his father's min isters and
servants, including his secre tary 'Abd Allah ibn U mayya. As th e secretary
was ill, his duties were performed tempor arily by a christian, Go mez ibn
Antoni an. When 'Abd Allah died , Muhammad declared tha t 'if only
Gom ez were a muslim he would not wish to change him for anyo ne .
H earing this [Gom ez] testified in public tha t he was a mu slim, and he was
appo inted secre tary. Gomez discharged his duties with eloquence and a
lively intelligen ce'. Nevertheless, he mad e enem ies, one of whom wro te to
Muhammad:
'It would be one of th e strangest and most amazing things, sho uld the
Abbasid caliphs of the east get to hear of it, that th e Umayyads of
th e west need , in order to fill their highest secreta rial post and for
excellence in th e writing of royal letters, to nam e an indi vidu al like
G om ez th e christian, son of Anton ian , son ofJuliana, wh o was also a
christian. Oh, m y lord! Why do you not cho ose better peopl e, who
would honour the post not just in th e person of th e man wh o
occ upies it, but also through his illustrious descent . ..'
The writer nam ed himself as one of th ese worthies togeth er with severa l
othe rs. One of them was eventua lly appo inted, but not until he had
completed several tests, including the writing of a spec ime n letter, which he
passed round several friends before submitting th e best version. Ibn
al-Qjitiya did not say th at Gomez was dismissed , nor th at Muhammad had
been wro ng in his judgem ent to appo int him. Ibn al-Q uttya treated
G om ez' expedient conversion to Islam as a m atter of cou rse. Perhap s many
christians took thi s route to preferment at court. Ibn al-Q u!Iya seems
mer ely to be ridi culing th e oppos ition of muslims of lon ger standing to th e
int egration of th e native population int o th e Umayyad regim e.

178
 SARA THE GOTH AND HER DESCENDANTS

Yet, as we have seen in the case of Recemund, one did not need to
convert to Islam in order to prosper. Although, as a muslim, Ibn al-Qjitfya
might not have wanted to spell this out, he listed Hafs ibn Albar as one of
his christian forebears worthy of being cited in the History if the Conquest:
'Romulo held a thousand villas in the east of Hispania, having chosen
Toledo as his place of residence. Amongst his descendants are Hafs ibn
Albar, judge of the christians'.73 As we have seen, 74 Hafs is remembered
today for his verse translation into Arabic of the Psalms . Hafs, who was
known by the nickname 'the Goth', could be the son of Alvarus of
Cordoba." His translation of the Psalms was commissioned by a bishop
called Valens, possibly the Valentinius who corresponded with Alvarus."
Alvarus himself boasted of his Visigothic ancestry, although he is not
known to claim descent from Witiza and it seems that he was making a
general claim about the inhabitants of Hispania rather than about his
own lineage." Hafs was both a pious christian and an 'Arabizer'. Ibn
al-Qptiya may not have endorsed this position but, by writing Hafs into
the History, he provided another example of christian integration in
islamic society.

Sara the Goth

Sara the Goth eponymises a key moment in the integration of the house of
Witiza into Andalusi society. As we have seen, Ibn al-Qptrya told how,
when Alamund died, leaving a daughter Sara and two young sons, her
uncle Artubas seized their inheritance." Sara sailed to Syria to complain to
the caliph Hisham ibn 'Abd al-Malik, reminding him of al-Walid 's pledge
to her father. Sara's journey to the east, however, may also be a literary
device to get Sara to meet the caliph, because this meeting had two further
consequences. The first was Sara's meeting with 'Abd al-Rahman, the
future ruler of Hispania. The second was that Hisham arranged Sara's first
marriage, to 'Isa ibn Mozahim, from which Ibn al-Qptrya was descended.
Conveniently for Sara, or for the story that Ibn al-Quttya was telling:
'In the same year that 'Abd al-Rahman ibn Mu'awiya ['Abd
al-Rahman I] came to Hispania, she was widowed and Hayya ibn
Mulamis al-MadhijI and 'Umayr ibn Sa'Id competed for her hand,
but Talaba ibn Ubayd al-Khudami interceded for 'Umayr ibn Sa'Id
with 'Abd al-Rahman ibn Mu'awiya and he gave her to him in
matrimony. From this union were born Habib ibn Umayr, forefather
of the Banu Sad, the Banu Hajjaj, the Banu Muslima and the Ban a

119
 SARA THE GOTH AND HER DESCENDANTS

al-jirz, illustrious lineages, whose nobility the other sons of Umayr,

born in Seville of other women, could not achieve . ..'
Few women are mentioned in the Arabic histories, and the prominent role
which Ibn al-QutJya gives to Sara must have particular significance .
Women appear in the foundation myths of many cultures, usually relating
to the identification between a group and a territory and the allocation of
land. Wanda, the daughter of Krak, the legendary founder of Cracow,
overcame a tyrant making an unsuccessful bid for her hand and
landholdings." Such women often have magical powers, which they
deploy to perpetuate the ruling dynasty. One other Gothic woman appears
in the History - Umm 'Asim, the wife of'Abd al-Aziz, son of the conqueror
Musa.80 This marriage had catastrophic consequences for 'Abd al-Azlz, Ibn
al-Qntiya echoed the story which is found both in the Arabic tradition and
in the Chronicle if 754, that 'Abd al-Aziz married the widow of Rodrigo
(called Egilona in the Chronicle), who incited him to usurp the throne of
Hispania."' All three accounts say that 'Abd al-Aziz was assassinated. Sara
the Goth had no such aspirations, and no magical powers. Her significance
lies in her two sets of descendants. Note the biblical resonance of the names
Sara, 'Isa [Jesus], Ibrahim and Ishak [Isaac] . Though they are not used in
strict correspondence with the Book of Genesis, they evoke the patriarch
Abraham, whom both the christians and muslims revered as the
forerunner of all the nations. Ibn al-Qntiya may not be talking about
actual marriages but the descendants of Sara the Goth symbolise the
development of post-conquest society, tying the Visigothic past to the
structure of al-Andalus in the tenth century.
There are no Andalusi parallels to story of Sara the Goth, but it is not
unique among the Arabic sources . The family of Abu 'AlI al-Hasan
al-Masarjisi of Nishapur (d.853/4), recorded in the biographical diction-
aries from that city, offers some points of comparison.f Abu 'All al-Hasan's
grandfather Masarjis was a leader of the christian community in Nishapur,
possibly a bishop, and his father was also a christian, called 'Isa. Abu 'All
al-Hasan was converted to Islam by a prominent muslim of non-Arab
origin. His son became one of the most eminent patricians ofNishapur. The
family's smooth transition from christian to muslim without loss of status is
plausible. Abu 'All al-Hasan's biographer described his conversion to Islam
as the result of divine intervention, but more worldly factors were also
involved . Perhaps Abu 'All al-Hasan's advancement was being hindered by
his faith, like the secretary Gomez ibn Antonian. But the name of Masarjis
was transmitted in the genealogies of the descendants of Abu 'All al-Hasan's
daughters, preserving the memory of their noble non-muslim ancestor.

180
 SARA THE GOTH AND HER DESCENDANTS

The nickn am e Ibn al-Qnttya applied only to descendan ts of Sara and

her first husband 'Isa ibn Mu zahim . H e may not have been of Arab origin,
since he is mention ed with out a nisba giving the names of his gra ndfat he r,
and the nam e ' Isa could be christian. Acco rding to Ibn al-Qjltiya, Sara's
brothers both rem ain ed christian; one was the metropolitan of Seville, the
othe r a bishop called Oppa who defected to the christian north.
Ibn al-Q uttya, th e Chronicle if 754 and the Chronicle if Alfonso III tell rela ted
bu t inco mpatible stories abo ut Oppa; in the christian sources Oppa was
described variously as eithe r the brother or the son ofWitiza, and as bishop
of Seville or Toledo. P If Sara' s first hu sband had been a mu slim, by islamic
law all the descendants of this marriage would have been mu slim, whet her
or not Sa ra herself conve rted. The fact that they continued to be known as
the Banu al-Q utfya could impl y that this famil y rem ained chr istian for
some considerable tim e after the conq uest, altho ugh it may mer ely be a
reminder that th e family's ancestry was more noble in the fem ale than in
the male line. Perhaps one of th eir descendants converted in circumstances
similar to those of Gomez ibn An tonian, By the time tha t Ibn al-Q unya
was writ ing, however, they were good mu slims. This may be the reaso n
that Ibn al-Qjitiya rem ember ed his family as also being related, altho ugh
less directly, to a second lineage of the descen da nts of Sara the Go th, the
offspr ing of a second marr iage who were muslims from the outset becau se
Sara's second hu sband was a mu slim . Ibn al-Q utfya seems to be implying
th at altho ugh his family had not always bee n mu slim, they had long had
good mu slim connec tions.
The acco unt of Sara's defence of her inh erit an ce in Ibn Khallikan 's
biography of Ibn al-Q ii!Iya,8+ is similar to that found in the History.
Ibn Khallikan, however, men tion ed only the first of Sar a's marriage. Hi s
narrative continues:
'Her life lasted into the reign of emir 'Abd al-R ahman ibn Mu' awiya
ibn Hi sham ibn 'Abd al-Malik, the immigrant into al-Andalus from
the family of th e Umayyad s. When she would enter into his pr esence
he would satisfy her needs. H er na me prevailed in her descend ant s
and th ey are known by it to this day. That is related in the book
Celebrations if RemarkableMen selected and composed of information on
the later j urispru dents and savants of Co rdoba imparted by the j urist
Abu 'Umar Ahmad ibn Muhammad ibn 'Afif al-Ta'rikhi [the
histori an , d. after 1039]. T ha t material was explained an d written
down by the jurist Abu Bakr al-Hasan ibn M uhammad ibn Muffarigh
ibn 'Abd Allah ibn M uffarigh al-Ma'afiri the Co rdovan known as
al-Q ubbashi, who carried al-Ta'rikhl's teachings in his mem ory'.

181
 SARA THE GOTH AND HER DESCENDANTS

Not only does Ibn Khallikan give a different account of Sara, but he also
cites a different source. The passage is a reminder of the unreliability of
biographical dictionaries and of their authors' predilection for finding
explanations for matters whose real origins had been forgotten .
Unlike Ibn Khallikan, several sources mentioned the Banu Hajjaj, one
of the lines of descent from Sara's second marriagc.P Only Ibn al-Qjttiya
referred to their Visigothic origin. Other writers either knew nothing about
it or they paid the subject no attention because they did not feel that their
muslim credentials were in question. Yet the inclusion of christians in the
History is not an example of sha'ubija - a tendency to denigrate Arabic
culture and history and glorify the past of the subject peoples which was
widespread throughout the islami c lands." Ibn al-Qjttfya was not
challenging the status quo. Ribera's interpretation of Ibn al-Qntrya -
'in the core of his being moves and sparkles the Spanish spirit and the
exaggerated sense of honour of his Gothic nobility"? - is perfectly in tune
with Spanish concerns with the loss of empire after 1898, but far from the
spirit of the History. In Ibn al-Qutlya's version of his past, the two lines of
descendants of Sara the Goth stand for the ethnic complexity of tenth-
century Cordoba. Ibn al-Qutlya had not finished with the question of
ethnicity. Fierro has pointed out the significance of another passage from
the History. Al-Sumayl ibn Hatim expressed surprise that Qu'ran III, 134
reads that Muhammad's revelation was handed down to 'the people'
rather than 'the Arabs'. " 'By God!'" he exclaimed, '''I can see that we.will
be associated with slaves and the lowest of the low''' .88 Fierro interpreted
this anecdote relating to a protagonist with a negative image in the History
as 'a criticism of the association between religion and ethnicity . . .'89 Yet the
distinction had not quite lost its significance. For Ibn al-Qntiya, both his
christian past and his status as a client of the Umayyads were live issues.
The two marriages of Sara represent the ambiguous position which
christian converts, the clients of muslim rulers, occupied.
The History if the Conquest ifal-Andalus may be no more than a collection
of stories from different sources, both oral and written, collected at an
unknown date, and attributed to Ibn al-Qptrya because the compiler of the
stories had read in the biographical dictionaries that Ibn al-Qjitiya was
interested in such matters. As in the best family sagas, advantageous
marriages, treaties and involvement in important historical events ensured
the family of Ibn al-Qut1ya their place in history. But the History if the
Conquest if al-Andalus is more interesting than this. It is possible to connect
the stories about the descendants of Witiza with anecdotes like the one
which Ibn Khallikan told about Ibn al-Qunya's talent for poetry. These
stories are literature, not historical truth, but they can be useful for modern

182
 SARA THE GOTH AND HER DESCENDANTS

historians because they show how medieval authors represented the past of
an individual or of a group in a way which their intended audi ence found
credible. The stories of Ibn al-Qjttiya, and above all the story of Sara the
Goth, showed that there was a close and long-standing kinship between
those who had com e to Hispania with the conquerors, and the Visigothic
nobl es who had made peace with them. Reading these stories, christian
converts to Islam could understand how they had reached their present
position, and what role their anc estors had played in the islamic conquest.

183
 ---
CHAPTER N1NE

Afterword

1 t is appropriate that Ibn al-Qutlya, with his story of a christian family's

integration into islamic society, should have the last word in this survey
of the christians of al-Andalus - not because the Banu Qiit1ya converted to
Islam, but because they preserved the memory of their Visigothic past.
Such continuity with Visigothic Hispania links the literary sources we have
been considering. It is difficult to say how significant this continuity is. After
a life time of work on the christians of al-Andalus, Simonet saw them as a
beleagured fifth column, lying low and waiting for deliverance. For
Simonet, the future of christian Hispania lay with the Asturias and the
monarchy founded by Pelagius, where 'the faith, patriotism, arts, sciences
and letters took refuge and, in sum, whatever elements and hopes of
fortunate restoration remained to the unhappy Spanish nation'. I Yet, this
pessimistic conclusion may be too harsh. Early medievalists are always
faced by patchy evidence and they must decide how much weight to give to
the surviving material. We have so few copies of christian texts from
al-Andalus that it is legitimate to wonder whether they were read only by a
tiny section of the christian community. The isolation in which some of
these works languished is quite clear. Although the eighth-century
chronicles were copied in northern Hispania, Asturian chroniclers writing
at the end of the ninth century had not read them. Yet christian learning in
al-Andalus survived the onslaught of Islam .
Despite the continuing vitality of christian culture, the use of Arabic and
changing social customs and architecture - in Cordoba, although perhaps
not in Toledo - were transforming al-Andalus into an islamic state . All the
Andalusi christian texts show some awareness of the history and language
of the conquerors. The eighth-century chronicles, compiled at a time when
there was almost certainly little writing in Arabic in Hispania, retold news
from the east to a much greater extent than earlier Hispanic chroniclers
had done, and the author of the Chronicle if 741 used this material to
underline the parallels between islamic defeats in the east and in Hispania.

184
 AFTERWORD

When Arabic became the dominant literary language in the peninsula,

christian texts were translated and these translations were islamicized in
form and vocabulary. Greenblatt cited this type of 'reproduction of
mimetic capital' as evidence of assimilation." Yet this acculturation was
very superficial, because the texts translated were nearly all sacred or
ecclesiastical. Even the Calendar if Cordoba, adapted from an Arabic model
compiled by a muslim, could have functioned as a christian liturgical
calendar. At the period when these texts were being translated, Recemund
was received as a christian scholar in Gorze. Rather than making the facile
connection between the use of Arabic and conversion to Islam , one should
perhaps bear in mind that Arabic-speaking christian communities have
survived in the Middle East to the twentieth century and conversely, that
countries such as Iran and Pakistan have adopted Islam without its
language. Alvarus excepted, the christians of al-Andalus may have
experienced few qualms during the transition to Arabic as their principal
literary medium. Change of language and form did not affect the context
in which sacred texts and christian histories were used and understood.
The language of sacred texts and providential history was not one in
which the question of adapting to a new religion could easily be addressed.
Thus it is not surprising that these texts do not say anything about the
process of conversion. It is also possible that religious affiliation was not the
dominant issue for the early islamic period in Hispania that it has since
become. The propagandists for the Asturian kings and for the emirs and
caliphs of Cordoba did their work well. Their versions of what happened in
the centuries immediately after 711 still dictate that the history of this
period is skewed towards the confrontation between Cordoba and the
Asturian kingdom, between Christendom and Islam . It is assumed that
the christians under islamic rule defined themselves only in relationship to
the conquerors and their religion. Historians of early medieval Hispania
continue to use a framework of oppositional pairs such as conversion or
resistance and continuity or change. Debates about the nature of conquest
are particularly susceptible to these generalisations, seeking the 'fatal
impact' of the colonisers . Historians of the islamic period in Hispania must
be particularly careful to avoid such prejudice, whether in favour of or
against islamicization, and to say no more than the sources permit about
the process of conquest in any given time and place." The evidence we
have been considering, by its very nature, comes down on the side of
continuity, the survival of christian learning in an islamic context.
The chronicles of the eighth century show providential history being
written in a way which would have been familiar to Isidore a century
earlier. Yet it was the last history of this kind to be written by the christians

185
 AFfERWORD

under islamic rule. Clearly, it would have been difficult for Andalusi
christians to continue writing abo ut the 'ru in of Hi spania' in the terms used
by the compiler of the Chronicle if 754, altho ugh in the north , the compiler
of the Prophetic Chronicle of 883 did produce such a text. 4 It is possible tha t
the continued popularity of O rosius indi cates that ther e was an audience
for pr ovidenti al history, but no mean s of bringing th e story up to date. Such
history was not an objective acco unt of the past but an incitem ent to
overcome the sins of th e peopl e in orde r to bring about th eir temporal as
well as their spiritual salvation, and the expulsion of the Arabs was not on
th e horiz on in the early islami c period.
The ada pta tion of th e Rom an mart yr ac ts by Eulogius and by the
authors of the Passion if Pelagius and the Passion if Argentea show that it was
still possible to write hagiography in the old style, ada pted to a
contem po rary situa tion. Yet this way of representing the clash of cultures
may have meant more to outsiders looking in, like the nor thern christians
who read the Passions of Pelagius and Arge ntea, tha n to christians in the
U mayyad capital. Hi stori ans have spent too long looking at the christians
of al-Anda lus th rou gh the eyes of Eulogius and Alvarus. The neglect of
Eul ogius' martyrs suggests th at the conflict between christian and mu slim
documented in the martyr stories ma y be oflittle relevan ce to the chr istians
of al-Anda lus as a who le. No doubt And alusi christians were clear abo ut
their religiou s and cultural identity bu t they may not have let it assume
overriding imp ortan ce in dictat ing their life choices . Indeed , even the
christian sources, as unrepresent ative as they may be, show christians
takin g a number of different options. In the ninth century, one could be a
martyr, or a churc hman loyal to the islami c regime. In the tenth, one might
chose to becom e a quiet member of a mon astic community in the islami c
capital, as Argentea was said to have been. Or on e might be like
Recemund, a fun ctionary in islamic govern me nt and a memb er of calipha l
delegation s to northern Europe and Byzantium. Such a position did not
prevent Recemund from also being a Latin scholar, and his service to the
calip h was rewarded with a bishopric. O ne might decide to emigra te to
the nor th , perhap s to take up opportun ities offered by the expa nsion of
mon asticism. O r, like the secretary Go me z ibn Antonia n, one might decide
th at , after all, conversi on to Islam was the most expedient path . T hat all
th ese opt ions were possible sho uld be seen as a mark of the interest and
complexity of the period. T here may have bee n no concerted christian
resistance movement in Cordoba, but nei the r was th e Visigo thic
inhe rita nce forgo tte n . Hen ce, when the mu slim rulers were evicted from
the peninsula, it would be easy to por tray the Reconquest as a chris tian
reawakening.

186
 ---
Notes

Notes to chapter 1

I Historia Silense: 8, trans Russell 1938: 226.

2 Chroniques Asturihmes: 21.
3 Conde: 1820.
4 Do zy: 1861.
5 Levi-Provencal: 1950-3.
6 Kennedy: 1996.
7 Lapiedra: 1997.
8 Fernandez and Fierro .
9 Barkai: 1984.
10 Bulliet 1990: 130- I.
II Epalza 1992: 160-1.
12 Reilly 1992: 17-21.
13 Bulliet 1979: 116; ibid 1979b: 30-51.
14 Young 1990: 168-87 ; see p. 161.
15 Marin 1992: 379- 93.
16 The Formation qf al-Andalus.
17 Guichard 1976.
18 Simonet 1903.
19 Manzanares 1972: 133.
20 Cagigas 1947.
21 Castro 1948; Sanchez-Alborn oz 1956; Gotor 1995: 31- 52.
22 Burns 1989: 307-30.
23 ibid: 325.
24 Lineh an 1993: 4.
25 The Formation ofal-Andalus 2: xxvii-i.
26 Southern 1962.
27 Manzano 1994: 83-99; Christys 1999; Christys 2001b; for the changing
meaning of frontiers in this period see Goetz 200 I.
28 Martinez-Gros 1997: 116-1 7.
29 Triki 2000.
30 Menocal 1992: 484-5.

187
 NOTES TO CHAPTER 2

31 Collins 1989: especially at p. 4.

32 Al-Maqqari,
33 AI-Tabari: 6-7 ; Humphreys 1991: 73-4.
34 See for example Fierro 1995: 1-37 .
35 Ferre 1967: 320-6;Jimenez de Rada ed. Fernandez 1987: Prologue 6.
36 See for example Molina 1998; Manzano 1992; Chalmeta 1994.
37 Pohl 1997: 6.
38 Sanchez-Albornoz 1945: 105.
39 See for example Burman 1994 and ibid 1996: 110.
40 CSM I: 15-54; LOpez Pereira 1980; Wolf 1990: 28-45 and 111-158.
41 MGH AA II : 323-30.
42 Collins 1989: 57.
43 CSM I: 68-77.
44 CSM I: 314; Wasserstein 1991: 3.
45 t» Pasion de San Pelayo.
46 Pasionario Hispanico 2: 382-7.
47 Fletcher 1992: 62.
48 MGH SS 4: 335-77.
49 Liudprand of Cremona.
50 Van Koningsveld 1977: 45-52.
51 Ribera 1928 I: 228.
52 Levi 1962.
53 Van Koningsveld 1977: chapter 3; ibid 1992; ibid 1994.
54 Ibn al-Qjitlya Historia.
55 See for example Bitterli 1989.
56 Gruzinski 1988.
57 Wachtel 1971.
58 Guha 1992: 210-15.
59 Crone 1980: 30.
60 Fontaine and Pellistrandi 1992.
61 Manzano 1998.

Notes to chapter 2

I Al-Maqqarf I: 297-362.
2 Ibn Hawqal trans Romani 1971: 63-4.
3 Barcelo 1991.
4 Christys 200 Ia.
5 Times Atlas of World History 1993: 108-9.
6 Ibn 'Idharl ed Levi-Provencal and Colin 1948-51 2: 247.
7 Pavon 1988: 169-98 and 403-26.
8 Al-Azmeh 1992: 260-1 .
9 Hillenbrand 1992: 123.
10 Ibn Idhari ed Levi-Provencal and Colin 1948-51 2: 244seq; the account in
al-Maqqari, who said he was quoting Ibn Idhari, is even longer: Al-Maqqari I:
368seq.
II Ruggles 2000 135-6.

188
 NOTES TO CHAPTER 2

12 Labarta and Barcelo 1987 and see pp . 129-30.

13 Pavon 1988: 189.
14 Scales 1997.
15 Al-Maqqari Analeetes: I: 302-3.
16 Ibn Hayyan trans Garcia Gomez 1967: 211; Garcia Gomez 1965: 352 .
17 Ibn Hawqal trans Romani 1971: 64.
18 Ibn l:Iawqal trans Romani 1971: 69 .
19 Lapidus 1967: 79.
20 AjbarMaehmua: 9 ~ 10; Fath al-Andalus: 23.
21 Molina 1989.
22 CSM I: 32.
23 Chroniques Astunennes: 38; Wolf 1990: 164.
24 Chroniques Astunennes: 18.
25 Chroniques Asturiennes: 24.
26 Linehan 1993: 99 .
27 Porres 1985.
28 AjbarMaehmua : 40; trans: 50; CSM I: 46; Porres 1985: 17.
29 Porres 1985: 20.
30 AjbarMaehmua : 103-5; trans: 97-8.
31 Porres 1985: 23.
32 Granja 1967: 19~20 and 22-4.
33 Ibn al-Qiitlya trans Nichol: 36-9.
34 Manzano 1992.
35 Porres 1985: 31.
36 e.g Ibn al-Qjttlya Historia: 95-7 ; Manzano 1991: 290 and n.65.
37 Chroniques Astunennes: 56 ; Wolf 1990: 175.
38 Chroniques Astunennes: 28.
39 Porres 1985: 35-40.
40 Ibn Hawqal trans Romani 1971: 60.
41 Ewert, Gladiss, Golzio and Wisshak 1997: 87-93.
42 Collins 1998: 277 .
43 Epalza 1992: 151.
44 Linehan 1993: 217 .
45 Reilly 1992: 37.
46 DHEE 3: 1310-20; Vogel 1986: 277 ~80 .
47 Linehan 1993: 214 .
48 CSM I : 32.
49 CSM I : 59-66; AASSJanuary 3: 149-51 ; Fear 1997: xxiii-xxxiv,
50 CSM I: xvii-i n.1 O.
51 Fear 1997: xxxiii.
52 CSM I : 38.
53 CSM I : 45.
54 PL 96 : 43-4.
55 CSM I : 62-3.
56 CSM I : 64 .
57 Collins 1989: 75.
58 Vives 1963: 507 -9.
59 Wolf 1990: 153; this passage may be an interpolation and is not in CSM.
60 PL 86 : 1083; Perez 1926: 131-2.

189
 NOTES TO CHAPTER 3

61 Caba llero 1994; Ga ren 1995; Dodds 1992: 61-3.

62 CSM I: 1- 6.
63 Co llins 1989: 68-9.
64 CSM 1: 71.
65 CSM I: 55- 71; Collins 1989: 70.
66 Cava dini 1993.
67 CSM I: 68- 78; Cavadini 1993: 10-1 8; Collins 1989: 221- 3.
68 CSM I: 139.
69 Cavadini 1993: 20.
70 CSM I: 78- 80 .
71 Cava dini 1993: 5 I.
72 CSM I: 114- 16.
73 CSM I: 181.
74 PL 96: 887-94.
75 Cavadini 1993: 45 and 167 n5.
76 PL 96: 893- 1030.
77 PL 96: 9 18.
78 MGH Epistolae 3: 636-50; Cancilia 2: 122- 9.
79 McWilliam 1990.
80 Bullough 1983: 8.
81 MGH Epistolae 4: 242-5.
82 CSM I: 82-93.
83 Pelikan 1978: 62.
84 MGH Epistolae 4: 284-5 .
85 Wolf 1996: 85.
86 Boniface trans Talbot 1954: 124.
87 See p. 59.
88 CSM I : xvii-i n 10.
89 Ibn Hayyan ed MakkI 1973 2: 368.

Notes to chapter 3

1 CSM I: 7-1 4 and 42.

2 Co llins 1989: 89- 9 I.
3 CSM I: 43; Wolf 1990: 144.
4 Muhlberger 1990: 2- 4.
5 Wood 1984.
6 Wood 1992.
7 Hillgarth 1970.
8 J ohn of Biclar: Prologue.
9 Diaz 1962.
10 Andres 1976.
II BM Egerton 1934; Biblioteca de la Academia de la H istoria 81.
12 Diaz 1962: 71.
13 Biblioteca de la Universidad Compl utense 116-Z-46 (Villa-Amil catalogue
no I34).
14 Cardelle 1999; LOpez 1980: 8.

190
 NOTES TO CHAPTER 3

15 Library of the Arsenal 982 .

16 Muhlberger 1990: 200 ~ 3 ; Burgess 1993.
17 Mommsen edited Hydatius' chronicle as well as the eighth-century chronicles.
18 John of Biclar: 100.
19 Lop ez Pereira 1980: 83-92.
20 Collins 1989: 59; D iaz 1970: 313-43; Vallve 1978.
21 Collins 1989: 57-8; Lop ez Pereira 1980: 13-16, and 64 .
22 CSM 186 and 88; Wolf 1990: 149 and 152.
23 Or therabouts; the writ er knew that the exact date was the subject of debate:
CSM 1: 77.
24 CSM 1: 29.
25 Lopez Pereira 1980: 104.
26 CSM 1 20.
27 CSM 1: 21.
28 CSM 1: 17; Wolf 1990: ll 3.
29 CSM 1: 3 1; Wolf 1990: 131.
30 CSM 1: 33; Wolf 1990: 132- 3.
31 Wolf 1990: 75.
32 Wolf 1990: 76.
33 Wolf 1990: 78.
34 Wolf 1990: 77 and 79.
35 Wolf 1990: 77.
36 Wolf 1990: 79.
37 MGH AA ll: 289-90; Wolf 1990: 104-5.
38 Witt eric assumed th e kingship to himself which he had usurped from Liuva
and held it for seven years; because he lived by the sword he died by the sword .
The death of th e innocent Liuva , son of Reccared was not unavenged; he was
killed by his men between the courses of a meal .
39 Wolf 1990: 105-6.
40 CSM 1: 7.
41 ibid .
42 MGH AA 11: 291; Wolf 1990: 106.
43 MGH AA 11: 291-2; Wolf 1990: 106-7 ; cf CSM 1: 19.
44 Collins 1994.
45 Wolf 1990: 13.
46 MGH AA 11: 292; Wolf 1990: 107.
47 MGH AA 11: 294; Wolf 1990: 110.
48 CSM I: 20; Wolf 1990: 115.
49 Haldan 1990.
50 Sec for example Palmer 1993; Brock 1976; ibid 19922: 51-75.
51 Brock 1976; some of these are translated in Palm er 1993.
52 Palmer 1993: 1-4.
53 Palm er 1993: 29 -35.
54 Palmer 1993: 36-42.
55 Conrad 1990: 32.
56 Conrad 1990 .
57 Proudfoot 1974.
58 The Chronicle ofTheophanes.
59 The Chronicle ofJohn, bishop qf Nikiu.

191
 NOTES TO CHAPTER 3

60 Constantelos 1972.
61 Wallace-Hadrill 1982: 59; Lamoreaux 1996.
62 Brock 1976: 15.
63 McCormick 1995: 319.
64 Nikephorus Patriarch 0/Constantinople Short History.
65 Dubler 1916.
66 MGHAAII :367.
67 Palm er 1993: 29-35.
68 CSM I: 10; see also CSM I: 21; Wolf 1990: 121-23.
69 Collins 1989: 56.
70 LOpez Pereira 1980: 107.
71 ibid: Ill.
72 ibid: 96-9.
73 CSM I : II; CSM I : 24; Wolf 1990: 122.
74 Palmer 1993: 54.
75 CSM I: 7.
76 'Era 649. The 56th emperor of the Romans, H erac1ius was crowned, who
rebelled against Phocas in Africa for [the love of] Flavia, noblest of virgins to
whom he was betrothed in Africa, and who was ordered by Phocas to be
deported from Libya to the territory of Constantinople; for this reason the
princeps attacked with all the armed men from the west, launched a naval attack
on the republic with more than a thousand ships, and made a pact with
Nicetas, the magister militum of the Romans, who assembled a land army, that
whoever arrived in Constantinople first would rule the empire. Herac1ius,
sailing from Africa, arrived first in the royal city. Which he attacked, soon
overcoming any resistance. Thus the Byzantines presented Phocas to Herac1ius
with his throat cut' .
'n CSM I : 8-9; CSM I: 18; Wolf 1990: 113-4.
78 CSM I: 9.
79 CSM I: 9.
80 CSM I: 13.
81 CSM I: 39; Wolf 1990: 140.
82 CSM I: 38-9 and 46-7; Wolf 1990: 139-41 and 146-8.
83 Collins 1989: 55.
84 Kaegi 1992: 110-12; Haldon 1994.
85 See p. 28.
86 'In the western regions they ach ieved many military victories through their
leader, as-Samh. Having held power for a little less than three years he
undertook, on his own initiative, a census of Hispania Ulterior and Hispania
Citerior. He divided by lot among his allies the booty, arms and whatever else
in the way of plunder the Arab people in Spain had as yet kept undivided, and
added part of all the moveable and immoveable goods to the fisc. Afterwards
he made Narbonne his own and harassed the people of the Franks with
frequent battle. He placed garrisons of Saracens in the city of Narbonne to
oversee its defence more effectively. Assembling his forces, as-Samh came to
Toulouse and surrounded it with a siege, trying to overcome it with slings and
other types of machines. Informed of this turn of events, the Franks gathered
together under Eudes, their commander. There, at Toulouse , while the battle
lines of both armies were engaging with one another in serious fighting, the

192
 NOTES TO CHAPTER 4

Franks killed as-Samh, the leader of the Saracen forces, along with that
portion of the army gathered with him, and pursued the remaining part as it
slipped away in flight. Abd ar -Rahman accepted the command of the
Saracens for onc month, until Anbasah, by order of the prince , came to serve
as governor'; Wolf 1990: 137-8.
87 Collins 1989: 54.
88 CSM I: 41-3; Wolf 1990: 143.
89 CSM I: 14; CSM I: 35; Wolf 1990: 134.
90 Kennedy 1986: 106-8.
91 Alexander 1978; ibid 1985.
92 Palmer 1993: 222-42; Reinink 1992.
93 the Annenian History : 132.
94 'The Apology of Timothy the Patriarch'.
95 Brock 1992: 57.
96 Alexander 1985: 152- 3.
97 Cook 1992.
98 BL Ar. 9449 , f.130b.
99 the Chronicle ofTheophanes: 570.
100 Palmer 1993: xxvii.
10 1 Burgess 1993: 9.
102 Landes 1988.
103 MGH AA I I: 481.
104 CCSL 115: 3: 10.
105 Chroniques Asturiennes: 1-9.
106 Alexander 1985: 85.
107 CSM I: 52-3; Wolf 1990: 156-8.

Notes to chapter 4

I CSM 2: 392.
2 CSM 2: 397-8.
3 CSM 2: 402.
4 CSM 2: 408-15.
5 Jimenez 1970: 494 .
6 Ibn Rushd al:Jadd XV I: 378-9, 396,438-41; cited by Fernandez and Fierro
2000 .
7 Netton 1996: 5.
8 Wolf 1988: 38-47 .
9 Dubois 1965; Christys 1998.
10 AASS March 2: 87-95; Chroniques Astunennes 30.
II Wolf 1988: 23.
12 Coope 1995: xii.
13 CSM I: 330-43.
14 CSM 2: 497 -503.
15 PL 115: 945.
16 Morris 1972.
17 CSM I: 330.

19 3
 NOTES TO CHAPTER 4

18 CSM I: 33 1.
19 CSM I : 332.
20 CSM I : 333.
21 Frend 1984: 742.
22 CSM I : 334.
23 CSM I: 335.
24 CSM I : 335.
25 CSM 2: 497.
26 Menend ez 1992: 56.
27 Andres 1974.
28 Lambert 1953; Mado z 1945; see map 2.
29 Orl andi s 1976: 362.
30 Gomez Moreno 1911: 3 1.
31 CSM I : 335.
32 Real Academia de la H istoria de Mad rid 80.
33 Fontaine 1983: n69.
34 Epalza 1994.
35 CSM I : 340.
36 Fontaine 1983: 29 n33.
37 CSM I: 332.
38 CSM I : 332; Wolf 1988: 71.
39 CSM I : 333.
40 CSM I : 334.
41 CSM I : 336.
42 CSM 2: 433: Wolf 1988: 57.
43 CSM 2: 339.
44 CSM 2: 483- 6; Wolf 1990: 96-9.
45 CSM 2: 406-8.
46 M illet-Gerard 1984: 125- 37.
47 CSM 2: 483.
48 ibid .
49 Diaz 1970: 149- 68.
50 Escorial d.1.2.
5I Escorial d.L I.
52 Biblioteca de la Real Acad emia de Madrid 78.
53 Wolf 1990: 99- 100.
54 Diaz 1970: 161.
55 ibid: 155.
56 CSM I : 197- 201.
57 Diaz 1970: 151.
58 CSM I : 333.
59 Madoz 1945: 420.
60 CSM I : 200.
61 CSM 2: 487.
62 Kedar 1984; Ducellier 1971; Newma n 1993; Samir and Nielsen 1994.
63 Sahas 1972.
64 CSM 2: 375.
65 CSM 2: 375-6.
66 CSM 2: 487.

194
 NOTES TO CHAPTER 4

67 Wolf 1996: 100.

68 CSM I: 310-11 ; Waltz 1970.
69 See p. 153.
70 Dick 1961.
7I Griffiths 1985.
72 Peeters 1939.
73 Lapiedra 1994.
74 Wolf 1988: 9-10.
75 CSM 2: 441.
76 See for example Wolf 1988: 31.
77 CSM 2: 721.
78 CSM 2: 486; Wolf 1988: 87.
79 CSM 2: 406-8.
80 CSM 2: 397.
81 Gil 1970; AASS October 9: 626-47.
82 Pasionano Hispdnico I: 242.
83 AASS October 9: 645-6.
84 Gil 1970: 123- 35.
85 Viguera 1981: 67.
86 Lopez 1964.
87 Gil 1970: 113.
88 AASS October 9: 645.
89 MGR SS 15: 239-64; Geary 1978: 148.
90 Gil 1970: 126-34.
91 Gil 1970: 123.
92 Collins 1986: 163-5.
93 Bishko 1948.
94 Moral 1975.
95 LOpez 1964: 164.
96 Lewis and Short 1966.
97 LOpez 1964: 397.
98 Gil 1970: 110.
99 Gil 1970: 127.
100 Ibn Hayyan trans Levi-Provencal and Garcia Gallo: 1954.
10I Archivo General deNavarra, Brevario de Leire fol 95 v.
102 Manzano 1991: 93.
103 Viguera 1981: 67; Granja 1966: 67-8.
104 CSM 2: 500.
105 CSM 2: 411.
106 CSM I: 332-3: Wolf 1988: 66.
107 CSM 2: 473: Wolf 1988: 71.
108 Wolf 1988: 57.
109 CSM 2: 406.
110 Coope 1995: 17.
III CSM2 : 412 .
112 Le Calendrier: 143.
113 CSM 2: 364; Wolf 1988: 55.
114 CSM I: 211-14.

195
 NOTES TO CHAPTER 5

Notes to chapter 5

I MGH SS Rer. Ger. 34: 53; Wilson 1998: 29.

2 PL 115: 948 .
3 Levi-Provencal 1950: 230 -2.
4 Ibn al-Faradl Tarikh : n081.
5 AI-Khushani: 231.
6 Cagigas 1947: 325.
7 Simonet 1903: 391.
8 Ibn Idhiiri ed Levi-Provencal and Colin 1948-51 2: 201-3.
9 Simonet 1903: 626 .
10 Vivancos 1988.
II DHEE 3: 1541-2.
12 Diaz 1969.
13 BM Add . 25.6000; Pasionario Hispanico I: 27ff.
14 Diaz 1969: 103.
15 fo1258v.
16 Gaiffier 1937: 271-2 .
17 Pasionano Hispanico I : 28.
18 MBN 1005.
19 Gaiffier 1938: 369.
20 Guerreiro 1992: 138.
21 fo1225 .
22 BN n.a .l. 2180; MBN 494 foI23-59; Pasionano Hispanico I : 33; Diaz 1983:
424-425.
23 MBN 822, foI29-59.
24 Diaz 1969: 97.
25 BN n.a.l. 2179.
26 Silos Biblioteca Monasteria I; Diaz 1969: 99.
27 Silos Biblioteca Monasteria 3 (dated 1052), BN n.a.l. 2.171 (dated 1067) and
2.169 (dated 1072).
28 except that the Silos collection does not include the Passion qf Castissima; Diaz
1969b : 99.
29 BN n.a.l. 239 fol 68v-83.
30 Diaz 1969b : 102.
31 Tuy Archivo Catedralicio fol 182seq; La Pasion de San Pelayo 22-24; Oporto,
Biblioteca Pub lica Municipal, codex XX fol75seq.
32 Guerreiro 1992.
33 BN n .a.l. 3879; Dubois and Renaud 1976; Quentin 1908: 130.
34 Pasionario Hispdnico : 265.
35 ibid: 289.
36 ibid: 120 and 159.
37 Chroniques Asturiennes: 23.
38 Vives and Claveras : 1946.
39 Pasionario Hispdnico : 287-9.
40 Akeley 1967: 103.
41 Perez de Urbel 1926.
42 ibid: 231.

196
 NOTES TO CHAPTER 5

43 La Pasion de San Pelayo: 46 and 48.

44 ibid: 64-5; trans Bowman 2000: 232-3.
45 See p. 97.
46 La Pasion de San Pelayo : 17.
47 Espana Sagrada 23: III .
48 La Pasion de San Pelayo: 34-7; Bowman 2000: 231.
49 Ibn Hayyan ed Chalmeta et al 1979: 103-10; Kennedy 1996: 89.
50 Chalmeta 1972.
51 See p. 97.
52 Historia Silense: 163.
53 Ibn Hayyan ed Chalmeta et al 1979: 121-7.
54 Diaz 1969: 108-9.
55 Wright 1994.
56 Gomez Moreno 1911: 347-8.
57 Williams 1993.
58 Floriano 1951: II no.192, pp . 362-72.
59 Aguilar 1994.
60 Oliver 1994.
61 Mediano 1994.
62 They were consigned to different volumes of Fontaine 1995: 86-9 and 1973:
329-31.
63 Dodds 1992: 50-2.
64 Dodds 1993.
65 Williams 1994 I: 132.
66 Werckmeister 1997.
67 Werckmeister 1993.
68 MGR Capito I : 261; Woolf 1965.
69 MGH Capito I: 169; Collins 1981: 178-81.
70 MGH Poet. lat. I : 484; Fontaine 1992.
71 Perez 1934: 254-5.
72 Perez 1934: 278; Isla 1992: 54-61.
73 Chroniques Astunennes: 55; Wolf 1990: 175.
74 Gomez Moreno 1911: 107.
75 DHEE 4: 1640.
76 ibid: 1648.
77 ibid: 1651-2.
78 Gomez Moreno 1911: 141.
79 Pastor 1995: 25.
80 Chroniques Asturiennes: 23.
81 Manzano 1991: 174.
82 Garcia de Cortazar 1969: 10 I.
83 LOpez and Rodriguez 1984.
84 u Calendrier: 103.
85 Escorial a.I.l3; Diaz 1983: 89-114.
86 Escorial a.II .9; Salisbury 1991.
87 Elliot 1993: 364-72.
88 Isidore Etymologiarum II, XI I I.
89 Jordan 1997: 10-28.
90 Boswell 1980: 194-97 .

197
 NOTES TO CHAPTER 5

91 King 1972: 127, 157 and 235.

92 Jordan 1997: 16.
93 MGH AA I I: 1329, trans Ian Wood.
94 PL 125: 692.
95 Jerome Commentaria in Hiezechielem 5.16.48-51.
96 La Pasion de San Pelayo: 52-3; Bowman 2000: 232.
97 Wilson 1998: 2; ibid 1984: 30-65; Haigh t 1965.
98 MGH SS Rer. Ger. 34: 52-62.
99 Wilson 1998: 30.
100 Bartlett 1986: 22; Geary 1978: 63-4.
101 Historia Silense : 171.
102 ibid: 168.
103 Fletcher 2000: 151.
104 Wreglesworth 1995: 106-21.
105 La Pasion de San Pelayo: 16 and n12.
106 Lasko 1972: 158.
107 Hayward 1993.
108 Historia Silense: 171; Collins 1983: 242.
109 ibid: 34.
110 Diaz 1983: 5 15.
III La Pasion de San Pelayo : 16 n13.
112 BM Add. 30845.
113 BN n.a.l. 239.
114 Diaz 1969: 108.
115 Isla 1992: 100-3.
116 See p. 84.
117 PL85: 1041-50; Sand oval 1610.
118 PL 85: 1046.
119 hunc nostrum patronum martyrum tuum, corpore passum Cordubae, Tuda
suscepit.
120 La Pasion de San Pelayo : 16 n 12.
121 La Pasion de San Pelayo: 17 n 13; DHEE 4: 598-602.
122 La Pasion de San Pelayo : 13 n 4.
123 Isla 1992: 80.
124 Isla 1992: 90-1 .
125 La Pasion de San Pelayo : 128.
126 Chroniques Astunennes: 38-42; Wolf 1990: 164-8.
127 Wolf 1990: 166.
128 Pasionario Hisprinico : 386.
129 Simonet 1903: 596 n2.
130 Acien 1994.
131 Vallve 1966.
132 Ibn al-Qjinya trans Nichols: 255.
133 Marin 1995.
134 Fletcher 1989.
135 Cutler 1965: 333.
136 Una cronica anonima.
137 Ibn Hayyan trans Guraieb 1958: 172.
138 Vallve 1965; Acien 1992: 22.

198
 NOTES TO CHAPTER 6

139 Manzanares 1972: 143.

140 Fierro 1995.
141 Una cronica anonima : 119; Ibn Idharf ed Levi-Provencal and Colin 1948~51 2:
I 71 has 305 [917-18].
142 Una cronica anonima : 140-1 , 146 and 148.
143 Musurillo 1972: 281.
144 Pasionario Hispdnico : 386.
145 Perez 1934: 266 , mentioned the nuns Ikilio, Justa and Rufina from the
monastery of Santa Eulalia, south of Cordoba, but did not give his sources.
146 Le Calendrier: 151.
147 DHEE 3: 859 .
148 See p. 84.
149 Ibn Idharf ed Levi-Provencal and Colin 1948-51 2: 251; al-Maqqart I: 235.
150 Ibn Idhari ed Levi-Provencal and Colin 1948- 51 2: 268.
151 Levi-Provencal 1931: 37-8.
152 Ibn Idhari ed Levi-Provencal 1930: 38.
153 Cameron 1989: 191.

Notes to chapter 6

I T qftfl al-'Azman toa-masalili al-'Abdan.

2 Al-Maqqari 2: 125-6.
3 EI-I:Iajji 1970.
4 Vita Johannis Goreensis.
5 Liudprand of Cremona: 8-9; Senac 1982: 20 on the site of Fraxinetum.
6 Ibn Hayyan ed. Makki 1973: 154.
7 Sabbas 1934: 183.
8 Ibn Idharr ed Levi-Provencal and Colin 1948-51 2: 218.
9 Scheindlin 1992: 188-200.
10 Vita Johannis Goreensis: 122.
II ibid: 128.
12 Liudprand of Cremona: I .
13 CSM I: xvii-iii nlO .
14 Parisse 1993.
15 Barone 1993: 141-58.
16 Vita Johannis Gorzmsis: 116.
17 Leyser ed Reuter 1994: 125-42.
18 Vita Johannis Gorzensis 17,33,41 ,43 and 52; MGH SS4: 246, 486 and 659;
Thompson 1929.
19 Vita Johannis Gorzensis: 128.
20 McCluskey 1998: 169-70.
21 ibid: 130.
22 Cerulli 1970.
23 Collins 1990: 113.
24 Liudprand of Cremona: 71.
25 ibid: 7.
26 ibid: 54 and 102.

199
 NOTES TO CHAPTER 6

27 ibid: 54, 71 and 10l.

28 ibid: 8.
29 ibid: 54-5; Sutherland 1988: 35-9.
30 ibid: xiii.
31 Leyser 1985.
32 Buc 1995.
33 Sutherland 1988: 5 1.
34 Staubach: 1989~90 , 1991.
35 Liudprand of Cremona: 95.
36 Liudprand of Cremona: 151.
37 Le Calendner: 29 .
38 ibid: vii.
39 Pons 1893: 89.
40 Ibn Qutayba IGtab al-Anwii'; Ie Calendrier: xi.
41 North 1994: 12.
42 Evans 1998: 4-19.
43 Ibn Qutayba: 6; Ibn 'Asim: 19-20.
44 Young 1990: 275; Hill 1993: 32~3 .
45 Kutub ai-Azmina.
46 EI6: 523 -4.
47 Troupeau 1968.
48 I:Iajji Khalifa Kasbfai-zunun 5: 53.
49 Forcada 1970.
50 Kunitzsch 1994.
51 Ibn 'Aslm : 39.
52 Samso 1983: 149-6l.
53 Samso 1983: 125-38.
54 Toomer 1968.
55 Vernet 1965: 545.
56 u Calendner: 31 .
57 Vernet 1965: 545.
58 Samso 1978: 178; EI6 : 599 -602.
59 Ibn 'Aslm : 26 and 44.
60 Ibn 'Asim: 46.
61 Le Calendner: 1-19.
62 See p. 150.
63 Ie Calendrier: 23 .
64 ibid: 73.
65 ibid: 183.
66 ibid: 8l.
67 AI-Maqqari 2: 368 and 463.
68 u Calendner: 4 1.
69 Flick 1965.
70 Nirenberg: 167-8.
71 See p. 108.
72 'Ani> ibn Sa'td trans Castilla.
73 See p. 130.
74 Pons Boigues 1893: 88-9.
75 Al-Maqqari 2: 93.

200
 NOTES TO CHAPTER 7

76 Al-Maqqari I: 643 and 661 .

77 Dozy trans Custodio 1957: 185.
78 For th e Umayyad secretaries in the east see Khalidi 1994: 89-96.
79 'Arib ibn Sa'td trans Castro 1983.
80 Dozy trans Custodio 1957: 186-90.
81 Le Calendrier: 3.
82 ibid: 27.
83 Espana Sagrada 53: 13.
84 Simonet 1871: 127-32.
85 Le Calendrier: 14 n.3.
86 McCluskey 1998: 170.
87 Saavedra 1892: 15.
88 Dozy trans Custodio 1957: 196.
89 Levi 1981.
90 Escorial cod.ar. 297.
91 Espana Sagrada 53: 134- 5.
92 Forcada 1992: 17-18.
93 Millas 1943.
94 Martinez and Samso 1981.
95 MBN 6.036.
96 Simonet 1871.
97 Al-Maqqari I: 373-4.
98 Labarta and Barcelo 1987.
99 Dodds 1992: 29 and 39 n6.
100 Pons 1893: 306-10.
10I Pellat 1954.
102 Tester 1987: 109-23 and 151.
103 Stanton 1990: 108.
104 Sa'Id th e Andalusi: 77.
105 Ibn Abi Usaybi'a: 492 .
106 Forcada 1992: 16- I 7.
107 Dozy trans Custodio 1957: 188.
108 Alfonso X 4: III.
109 Alfonso X 4: 1.
110 Roth 1990.
III £/6: 351-2.
112 Van Koningsv eld 1991: 701.
113 See p. 137.
114 Al-Maqqari I: 235.
115 Dozy trans Custodio 1957: 196.
116 Al-Maqqari Analectes I: 160.
117 EI-I:J:ajji 1970: 92.
118 Vita ]ohannis Gorzensis : 11 5.

Notes to chapter 7

I Ibn AbI Usaybi'a: 47.

201
 NOTES TO CHAPTER 7

2 Alonso-Nunez; Lacroix 1965.

3 Fabrini 1979: 10.
4 Ross 1955: 35 n2; Bately 1961.
5 Orosius ed Bately 1980; ibid ed Lund and Fell 1984.
6 Smyth 1995: 538 -9.
7 Columbia University library MS X-893 712 H .
8 Ibn Khaldiin ed Dagher 1958-682: 169 and 402 .
9 Levi Della Vida 1954.
10 Ibn al-Abbar Al-Takmila no3 12; Ribera 19282: 181-228.
II Ibn Bashkuwal Al-Sila nos 1412, 1415 and 1417.
12 Al-Maqqari I: 302.
13 Ibn Hayyan ed Antuna 1937: 197.
14 Al-Maqqari 1: 250.
15 Wasserstein 1990-1 .
16 Ibn al-Faradi: no6 .
17 Ibn AbI Usaybi'a: 2: 47; Van Koningsveld 1977: 58.
18 Toynbee 1973: 491-2.
19 Van Koningsveld 1977: 59.
20 Vernet 1968; ibid 1978: 72.
21 Van Koningsveld 1977: 59.
22 Vernet 1968: 453 -4.
23 Van Koningsveld 1977: 57.
24 CSM II: 707.
25 Ibn Khaldun ed Dagher 1958-68 2: 169; Orosius ed Badawi 1982: II .
26 ibid: 2: 402; Orosius ed Badawi 1982: II .
27 Ibn al-Faradi: noslO68 , 1070, 1071 and 1079; Pons 1893: 105-8.
28 for example I:Iajji Khallfa: 5: 172-3.
29 Ibn al-Faradi: no1068.
30 Orosius ed Badawi 1982: 13-15.
31 Ibn al-Faradi: no1079 .
32 Ibn Khaldun ed Dagher 1958-682: 315.
33 Kuhayla 1985-6.
34 Levi-Provencal 1937.
35 Simon et 1903: 111-112, 171 and 622.
36 Ibn Hayyan ed El-Hajji 1965: 64.
37 Van Koningsveld 1977: 57.
38 Al-Mas'ndi 3: 69-72.
39 DHEE 2: 1019.
40 Al-Mas'udi 8: 291.
41 Sa'Id al-Andalusi: 61; Pedersen 1984: 120.
42 Orosius, trans Deferrari 1964 I: 15.
43 Molina 1984; Vallve 1967.
44 MGH AA II : 377- 88.
45 Molina 1984: 74.
46 Cronica del Moro Rasis: 13-15.
47 ibid: xxiii.
48 Hajji Khallfa : 5: no10625.
49 Rosenthal 1952.
50 See p. 168.

202
 NOTES TO CHAPTER 7

51 Sa'Id al-Andalusi: 58.

52 Pellat 1954.
53 Cachia 1990: 18.
54 Ibn J:labib.
55 Makki 1957.
56 Lewis 1962.
57 Ibn Khaldiin, trans Rosenthal 1958: I: 65.
58 Humphreys 1991: 135.
59 Al-Azmeh 1981: 50.
60 AI-Azmeh 1981: 159; Khalidi 1994: 22.
61 Ibn Khaldun ed Dagh er 1958-68: 2: 493; Orosi us ed Badawi 1982: 496.
62 Orosi us ed Badawi 1982: 469~97 .
63 Ibn Khaldun ed Dagher 1958-68 2: 411.
64 Fischel 1967.
65 Fischel 1954.
66 Fischel 1967: 139.
67 Rosenthal 1962.
68 Unfortunately, Badawi omitted this section from his edition .
69 Levi 1954: 268.
70 Van Koningsveld 1977; Orosi us ed Badawi 1982: 498.
71 Daiber 1986: 202.
72 Garcia Gomez 1947: 293.
73 Van Koningsveld 1994.
74 Serrano 1991.
75 Van Koningsveld 1994: 428, two fragments now in Fes may be part of the
earlier manuscript.
76 ibid: 1994: 445.
77 Kuhayla 1985-6: 122.
78 Orosius ed Badawi 1982: 457.
79 Griffiths 1990: 18.
80 Daiber 1986.
81 Levi Della Vida 1971: 117.
82 Orosius ed Badawi 1982: 457-8.
83 Smith 1971: 217-236; Linder 1975; The Book qf the Pontiffi: 14-26.
84 BN lat. 2178.
85 Scott 1994.
86 Aldhelm 83-4; William of Malmesbury I: 372-3.
87 Levi 1962.
88 Van Koningsveld 1994: 445.
89 Levi 1962: 190.
90 Troupeau 1971.
91 Rosenthal 1968: 138.
92 Van Koningsveld 1994: 445 n83.
93 ibid: 1992: 89-91 and 100-3.
94 ibid: 1994: 431.
95 ibid: 1977: 45-6.
96 ibid: 17 .
97 Ibn Hazm Kittib al-jisal 2: 3.
98 Wright 1994: 3.

203
 NOTES TO CHAPTER 8

99 Ibn Hazrn Kitiib al-jisal 2: 3-4.

100 Van Koningsveld 1977: 55; ibid 1992: 702; ibid 1994: 425.
10 I ibid : 1994: 426.
102 Jimenez de Rad a Historia 4: 77.
103 Tisserant and D e Bru yne 1910.
104 Van Koningsveld 1977: 54.
105 Le Psaultier Mozarabe; Urvoy 1994; Van Koningsveld 1977: 54 .
106 l..e Psaultier Moearabe; verse 98.
107 ibid : verse 108.
108 ibid: verse 100.
109 Van Koningsveld 199 1: 699 .
110 Escorial MS Ar. 1623; Van Koningsveld 1991: 704.
III Kassis 1994: 4 18.
112 Van Koningsveld 1994: 442 .
113 Gautier-Dalch e 1984.

Notes to chapter 8

I Ibn al-Qunya Historia: 4-6.

2 CSM I : 47.
3 Chroniques Asturiennes: 37-8.
4 Avila 1985: 156.
5 Ibn al-Faradi: no 1316.
6 Ibn al-Bash kuwal: no 567.
7 I quote th e last two sentences from Ibn al-Qutlya trans Nicho ls: xix, who
commented: 'T he sense of this phrase is obscure.'
8 Ibn al-Abbar: no 792; Humphreys 199 1: 190.
9 Ibn al-Khatib: 48-58; Avila 1980.
10 Fierro 1989: n5 .
II Humphreys 1991: 189.
12 Ibn Khallikan trans MacGuckin de Slane 3: 79-83 .
13 See p. 181.
14 "Iyad 6: 297.
15 Fierro 1989: 488 .
16 Ibn al-Abbar I : 45.
17 Al-Khushanl: 34, 65, 80-1 , 87-8, 134 and 136.
18 Al-MaqqarIAnalectes I: 43 1-2 and II 6.
19 Escorial 1.654; Fierro 1989: 490 .
20 Paris BN 1.867 foI2-50 .
21 Cherbonneau 1853.
22 Ajbar Machmua .
23 Molina 1989.
24 Chalmeta 1973: 40.
25 Ibn al-Qutrya Historia: 36.
26 Ibn al-QiitIya Historia : 6-7 .
27 Ibn al-Abbar I : 144.
28 Rubiera 1980.

20 4
 NOTES TO CHAPTER 8

29 M an zan o 1992.
30 D ozy 188 1 I 32- 4; Makkf 1957.
31 Ibn al-Qjl trya Historia: 5.
32 Ibn Bashkuwal: no s 849 an d 765 .
33 Ibn al-Qjr tfya Historia : 6.
34 Fierro 1989: 489.
35 Ibn Bashkuwal: no 769.
36 Al-Far adi no 816 ; Ibn H abib: 42-3.
37 Levi-Proven cal 1950-3 3: 504.
38 Cronica del M oro Rasis.
39 Al-Khu shani : 5.
40 Ibn al-Abbar: 202 - 3.
41 Una cronica anonima: 78-9; M artinez G ros 1992: 29.
42 Ibn al-KalbL
43 Dumville 19 79; Pohl 1994; Balandier 19 72: 69- 70 and 8 I.
44 Dumville 1985.
45 Kenn ed y 199 7.
46 Ibn Ishaq: 3.
47 Ibn H abib : 6 1 and 63.
48 Ibn 'Abd R ab bfhi 3: 44seq .
49 C hrisrys 2002 .
50 Pefiarroj a 1993: 6'1; M anzano 1991: 228 .
51 M an zan o 1991: 19 7.
52 C hej nc 1982: 20 .
53 C ha lmc ta 1994: 37.
54 Vallve 1992: 54 .
55 Duri 1987: 35; Lapi edra 199 7: 285.
56 Ibn al-Q uttya Historia : 52-6.
57 Ibn al-Qjr nya Historia: tran s Nichol: 94-5.
58 Ibn al-Q iitfya Historia tr an s Nic ho l: 9 1- 2.
59 C ha lmeta 1994: 141 .
60 Ibn al-Qjitfya Historia : 2-3,.
61 Ibn al-Qji tlya Historia: 3-4, trans Nichol 6 -7 .
62 CSM I: 37; Wolf 1990: 149- 50 ; Lopez 1980: 40-3 .
63 Al-Baladhurl I: 187,223, 246-7 , 249, 271- 5, and 338- 40 .
64 Cana da 1980 ; Granja 1966: 470 -1 ; Viguera: 1981: 68; sec p. 19.
65 Ibn al-Q uttya Historia : 47.
66 Ibn H azrn J amhara: 467.
67 Collins 1989: 204 .
68 M GH AA II : 46 1.
69 Ibn al-Q u ttya Historia : 37-8.
70 CSM I : 36; Wolf 1990 : 136.
71 Ibn al-Qjir tya Historia: trans Nicho l 45.
72 Ibn al-Qjittya Historia: 82- 5.
73 Ibn al-Q uttya Historia: 5.
74 See p. 155.
75 Dunlop 1954.
76 Dunlop 1955: 2 11- 13.
77 CSM I: 269 -70.

205
 NOTES TO CHAPTER 9

78 Ibn al-Qjnrya Histona 4-6; see p. 158.

79 Slupccki 1994: 186-95 .
80 Ibn al-Quttya Historia : II .
81 Ajbar Machmua: 31-2; CSM I : 78.
82 Bulliet 1972: 106-1 0.
83 Collins 1989: 33.
84 Ibn Kallikan ed 'Abbas 1970-1 3: no44 7; Ibn al-Qutiya Historia trans . Nichols
xxii-iii.
85 e.g. Ibn Hayyan ed Antuna 1937: 67 and Ibn Hazm ]amhara : 398-9.
86 Duri 1987: 102-4.
87 Ibn al-Qntrya Historia : xxii.
88 Ibn al-Qjttrya Historia : 40-1 .
89 Fierro 1989: 511.

Notes to chapter 9

I Simonet 1903: 212.

2 Greenblatt 1991: 6.
3 Thomas 1994: ix, 3 and 15.
4 Chroniques Asturiennes: 1-9.

206
 -.-.-

Bibliography

Acien Alman sa, M . (1994-) Entre el fiudalismo y el Islam: 'Umar ibn Hofsan en los
historiadores, en las fu entesy en la historia J aen: Universidad de Jaen .
Aguilar, V. (1994-) 'O nornastica de origen ar ab e en el reino de Leon (siglo X )'
Al-Qgntara 15 pp . 35 1- 63 .
AjbarMachmua (Coleccion de tradiciones) ed . and trans. Lafuent e y Alcantar a, E. (1867)
M adrid reprinted facsimile (1984-) M adrid: Edicione s Atlas.
Akeley, T C. (196 7) Christian Initiation in Spain c.300- 1100 London: Darton,
Longman and Todd.
Aldhclm TIe Prose Works tran s. Lapidge, M . and Herren , M . (1979) Ipswich: Boydell
Brewer.
Alexander, P. (19 78) 'The M edieval Legend of the Last Rom an Emperor and its
M essian ic Origin' Journal of the Warburg and Courtauld Institutes 16 London
pp .I -1 5.
- - (1985) TIe By zantine Apocalyptic Tradition Berkeley London: U niversity of
C alifornia Press.
Alfon so X ed . Ri co y Sinobas M . (1866 - 73) Libros del saber de Astronomia del Rf[Y
D. Alfonso X de Castilla 5 vols Madrid: Aguado .
D 'Alverny, M- Th. ed . Burnett, C . (1994-) La connaissance de l'lslam dans l'Occident
Medieval Aldershot : Variorum.
Andres, G. de (1974-) ' Los codices visigoti cos de la catedra l de Oviedo' Cuademos
Bibliogr4ficos 31 Madrid: CSIC pp . 5-29.
- - (19 76) ' Los codices visigotico s deJorge de Beteta en la Biblioteca del Escorial '
Celtiberia 5 1 Soria : Seme stral pp . 10 1-7 .
'T he Apology of Timothy th e Patri arch befor e the C aliph al-M ahdi ' ed. and tran s.
Min gan a, A. (1928) Bulletinofthefohn Ryl£mds Library 12 M anchester pp. 137-226.
'Arib ibn Sa'Id trans . Castilla Brazales,]. (undated) La Cronica de ;,lno sobre al-Andalus
Granada: Impredisur.
- - trans. Arjona C astro , A. (1983) El libro de la generaaon delfeto Cordob a.
TI e Armenian History attributed to Sebeos trans. with no tes by Thompson, R .W and
comme nta ry by Howard:Johnson,]' and Greenwood, T. (1999) Liverp ool:
University of Liverp ool.
Ashtor, E. trans. Klein , A. and Klein,]' (1973) TIleJ ews ofMoslem Spain Philad elphia:
J ewish Publication Society of Amer ica.
Avila, M .L. (1980) 'La pro clam acion (bay'a) de H isham II. Afio 976 d.c.' Al-Qgntara
I pp . 79-114- .

207
 BIBLIOGRAPHY

- - (1985) La sociedad hispanomusulmana alfinal del califato (aproximaci6n a un estudio

demogr/fico) M adrid: CSIC.
Al-Azrneh, A. (198 1) Ibn Khaldun in modem scholarship: a study in orientalism Lond on :
Third World Centre for Research and Publishin g.
- - (1992) 'Mortal Enemies, Invisible Neighbours: Northe rn ers in Andalu si Eyes,'
in The Legacy qf Muslim Spain pp. 259-72.
Al-Baladhuri tran s. Hitti, P. and Murgott en , F. (19 16- 24) the Origin; of the Islamic
State 2 vols New York: Columbia U niversity Press.
Balandier, G. tran s. Sherid an Smith, A.M . (1972) Political Anthropology Harmonds-
worth: Penguin.
Barcelo, M .'EI ca lifa patent e: el ceremonial om eya de Cordoba 0 la escenificaci6n
del poder' in Pastor R ., Kieni ewicz I, Garda de Ent erri a E. et al eds (199 1)
Estructuras y formas del poder en la historia, Salam anc a: Ediciones Universida d
Salamanca pp . 51-71 .
Barkai, R . trans. Bar-Ko chba, M. and Komay, A. (1984) Cristianosy musulmanes en la
Espana medieval (El enemigo en el espqo) Madrid: Ediciones Rialp.
Barone, G. J ean de Gorze, moin e benedictin ' in Parisse, M. and Oe xle, O. eds
(1993) L'abbaye de Goree au X ' siecle Nancy: Presses Un iversitaires de Nan cy
pp . 141-58.
Bartl ett, R . (1986) Trial by Fire and Water Oxford: Oxford University Press.
Bately, J.M. (1980) ' King Alfred and the Latin MSS of Orosius' History' Classica et
Medievalia 22 Copenhagen : Glydendal pp . 69-105.
Bishko, CJ. (1948) 'Salvus of Albelda and Frontier Monasticism in Tenth-Century
Navarre' Speculum 23 Cambridge, Massachussets: Medi eval Acad em y of
America pp . 559- 90.
Bitterli, U. (1989) Cultures in Corfiict: Encounters between European and non-European
Cultures, 1492-1800 Stanford, California: Stanford Univ ersity Press.
Boniface trans. Talbot, C . (1 954) the Anglo-Saxon Missionaries in Germany London :
Sheed and Ward.
The Book qf the Pontiffs (Liber Pontficalis): the Ancient Biographies qf thefirst ninety Roman
bishops to AD 715 trans. Davis, R . (1989) Liverpool: Liverpool Univ ersity Press.
Bosch-Vila, J. (1974) 'La ciencia de los linajes y los genealogistas en la Espana
musu lmana' in Miscelaneade estudiosdedicados al pnfesor A. Marin Ocete I pp. 63-77 .
- - and Ho enerbach, W. eds (1980) Andalucia isldmica: Textos y Estudios I Granada:
Universidad de Granada.
Boswell, J. (1980) Chnstianuy, Social Tolerance and Homosexuality. Grry People in Western
Europefrom the beginning ofthe Christian Era to theFourteenth century Chicago London:
University of Chi cago Press.
Brock, S. (19 76) 'Syriac Sources for Seventh-Century History ' Byzantine andModem
Greek Studies I I Oxfo rd : Blackwell Publishers pp. 17- 36.
- - (1992) Studies in Syriac Christianity 2 vols Aldershot: Variorum.
Brooks, E.W (1906) 'T he sources ofTheophanes' Byrantinische Zeitschrift 15 Munich :
Beck pp . 578 -87 .
Buc, P. (1995) ' Italian Hussies and German Matrons. Liutprand of Cr emona on
Dynastic Legitim acy' Friimittelalterliche Studien 29 Berlin: Gruyter pp . 207-25.
Bulliet, R. W (1 9 72) the Patricians qf Nishapur Cambridge, Massachussets: Harvard
University Press.
- - (1979) Conversion to Islam in the Medieval Period Cambridge, Massachu ssets
London: Harvard University Press.

208
 BIBLIOGRAPHY

- - (1979b) 'Conversion to Islam and the emergence of a Mu slim society in Iran'

in Levtzion, N. ed . Conversion to Islam New York London: Ho lmes and Meier
pp .30-51.
- - (1990) 'Conversion stories in early Islam ' in Conversion and Continuity
pp. 123-34.
Bullough , D. (1983) 'Alcuin and the kingdom of heaven: liturgy, theology, and the
Carolingian age ' in Blumenthal, U.R . ed. Carolingian Essays Washington: D.C.:
Catholic University of America Press pp. 1- 70, reprin ted in Bullough (199 1)
Carolingian Renewal: sources and heritage Manchester New York: Manchester
University Press pp . 16 1-240.
Burgess, R .W (1993) The Chronicle if Hydatius and the Consularia Constantinopolitana
Oxford: Clarendon Press.
Burm an , T.E. (1994) Religious Polemic and the Intellectual History if the Morarabs,
c.l 050- 1200 Leiden New York London: Brill.
- - (1996) 'The "Tathlith al-wahdaniyah" and the twelfth-century And alusian
approach to Islam' in Medieval Christian Perceptions pp . 109-30.
Burns, R.I. (1989) 'T he significance of the fronti er in the Middle Ages' in Bartlett ,
R . and Mack ay, A. eds Medieval Frontier Societies Oxford: Clarendon Press
pp .307-30.
Caballero Zoreda, L. (1994) 'Un canal de transm isi6n de 10 clasico en la alta eda d
media espanola. Arquitectura y escultura de influjo om eya en la peninsu la
iberica entre mediados del siglo VII e inicios del siglo X (I)' Al-Qgntara 15
pp .32 1-48.
Cachia, P. (1990) An overview if modern Arabic literature Edinburgh: Edinburgh
University Press.
Cagigas , I. de las (1947-8) M inorias etnica-religiosas de la Edad Media espaiiola I : Los
Moouabes Madrid: CSIC .
Le Calendrier de Cordoue ed. Dozy, R . (186 1) 2nd edn and trans. Pellat, Ch . (196 1)
Leiden: Brill.
Cameron, A. (1989) 'Virginity as Metaphor: wom en and the rhetor ic of early
christianity' in ibid . ed . History as Text London : Du ckworth pp. 181-205.
Canada juste, A. (1980) 'Los Banu Q asi (714- 924)' Principe de Viana 41 Pamplona
pp.9-96.
Cardelle de Hartmann, C . (1999) 'T he textual transmission of the Mozarabic
chronicle of 754' Early Medieval Europe 8.1 Oxford: Blackwell Publishers
pp . 13- 29.
Castro, A. (1983) Espafia en su historia: cristianos, moros y judios 2nd edn Barcelona:
Editorial Critica.
Cavadini, J-M . (1993) The last Christology if the West. Adoptionism in Spain and Gaul
785-820 Philadelphia: Univ ersity of Pennsylvani a Press.
Cerulli, E. (1970) 'Le Calife 'Abd ar-Rahman III de Cordoue et le martyr Pelage
dans un po erne de Hrotsvitha' Studia Islamica 32 Paris: Semanal pp . 69 -76.
Chalmeta Gendr6n, P. (1972) 'Historiographia medi eval hispana: arabica' Al-Andalus
37 pp . 353-404.
- - (1973) 'Una historia discontinua e intemporal (jabar)' Hispanic 33 Madrid:
Instituto J er6nimo Zurita pp . 23-75.
- - (1994) Invasion e islamizacion: La sumisum de Hispania y laformacion de al-Andalus
Mad rid : Mapfre.
Chejne, A.G. (1982) Ibn Haem Chicago: Kazi.

209
 BlBUOGRAPHY

Ch erbonneau,]. A (1853) ' Extrait du livre d'Ibn EI-Khouthiya intitule "Fo utouh
Elandalous lil-moslimin'" Joumal Asiatique 5e serie I Paris pp. 458-74.
Chri stys, A (1998) 'St-Ge rmain des Pres, St. Vincent and the Martyrs of Cordoba'
Earry Medieval Europe 7.2 Oxford: Blackwell Publishers pp. 199-216.
- - (1999) 'Christian-Muslim frontiers in early medieval Spain' Bulletin qf
Intemational M edieval Research 5 Leeds : Int ernational Medieval Institute
pp .I - 19.
- - (200 Ia) 'Cordoba in the Vita Argentea' in De Jong M. and Theuws F. eds
Topographies ofPower in the Early Middle Ages Leiden : Brill pp. 119-36.
- - (200 Ib) 'Crossing the frontier of ninth- century Hispania' forthco ming in
Abulafia D. and Berend N. eds. MedievalFrontiers: Concepts andPractices Aldershot :
Ashgate .
- - (2002) 'The History oflbn Habib and ethnogenesis in al-Anda lus' forthcoming.
The Chronicle ofJohn, bishop qf Nikiu trans . Charles,]. (19 16) Oxford: Williams and
Norgate.
The Chronicle qf Theophanes trans. Mango , C. and Scott, B. (1997) Oxford: Oxford
University Press.
Chroniques Astunennes (fin IX siecle) ed. and tran s. Bonnaz, Y (1987) Paris: Edition s du
Centre National de la Recherche.
Co lbert , E.P. (1962) the Martyrs qf Cordoba (850-859): A Study of the Sources
Washington: Catholic University of America Press.
Co llins, R . (198 1) 'Charles the Bald and Wifred the Hairy' in Gibson , M. and
Nelson,]. eds Charles the Bald: Court and Kingdom Oxford: British Archaeologi cal
Reports Intern ational Series 101 pp. 169-87 .
- - (1983) Early Medieval Spain: Unity in Diversity 400- 1000 London Basingstoke:
Macmillan Press.
- - (1986) The Basques Oxford: Blackwell Publishers.
- - (1989) the Arab Conquest qf Spain, 710-797 Oxford: Blackwell Pub lishers.
- - (1990) 'Literacy and the laity in early medieval Spain ' in McKitterick, R. ed.
The Uses qf literacy in Early Medieval Europe Cambridge: Cambridge University
Press pp. 109-1 32.
- - (1994) 'Isidore, Maximus and the Historia Gothorum' in Scharer, A and
Scheibelreiter, G. eds Historiographic im friihen Mittelalter Munich: Oldenbourg
pp.345-58.
- - (1998) Spain: an O:ifOrd Archaeological Guide Oxford: Oxford University Press.
Conde,].A (1820) Historia de la dominacion de los arabes en Espana Madrid: Garcia .
Conrad, L.I. (1990) 'T heophanes and the Arabic historical tradition: some
indica tions of intercultural transmission' Byeantinische Forschungen 15 Amsterdam:
AM. H akkert pp. 1- 44.
Cons tantclos, DJ. (1972) 'T he Mo slem conquests of the Near East as revealed in
the Greek sources of the seventh and eighth centuries ' Byrantion 42 Paris:
Champion pp. 325-57 .
Cook , M. (1992) 'The Heraclian dynasty in Muslim eschatology ' Al-Qgntara 13
pp.3-24.
Coope,].A (1995) the Martyrs qf Cordoba: Community and Famil» Corfiia in an Age qf
Mass Conversion Lincoln London: University of Nebraska Press.
Crone, P. (1980) Slaves onHorses Cambridge New York: Cambridge University Press.
Cronica Anonima de 'Abdal-Rahman III al-Nasir ed. and trans . Garcia Gom ez, E. and
Levi-Provencal, E. (1959) Madrid: CSIC.

210
 BIBLIOGRAPHY

Cumica del M ora Rasis ed Catalan, D. and Andres , M .S. de (19 74) Madrid:
C redos.
C utler, A. (1965) 'The ninth -centu ry Spa nish martyrs' movem ent and the or igins of
Western C hristian mission s to th e Mu slims' TI e Muslim World 55 H ar tford,
Co nnecticut: H ar tford Seminary Foun dat ion pp . 32 1~39 .
Daib er, H. (1986) 'Orosius' Histonae Adoersos Paganos in arabischer Uberlieferu ng' in
Van H en ten , JW , D e l onge, Hj. , Van Rood en P.T. and Wesselius JW eds
Traditionand Re-interpretation inJ ewish and Early Christian literature: essays in honour of
J iirgen G.H. Lebram Leid en: Brill pp. 2 10-49.
Dani el, N. (1962) 'Spani sh Christian sour ces of informat ion about Islam (Ninth-
Thirteenth C enturies) Al-Qgntara 15 pp. 365- 84.
David , P. (1947) Etudes historiques sur la Galice et la Portugal du VI' au XII' siecle Lisbon:
Livraria Portu galia Editora.
Diaz y D iaz, M .C . (1959) Index Scriptorum Latinorum Medii Aevii Hispanorum 2 vols
M ad rid: CS IC .
- - (1962) 'La tran srnision textua l del Biclaren se' Analecta Sacra Tarraconnenis 35
Tarragon a pp . 57- 78.
- - (1969) 'La Pasion de S. Pelayo y su d ifusion' Anuario 6 pp. 97- 116.
- - (1969b) ' La circulation des man uscrits dans la Penin sule Iberique du vue au
X I'· siecle' Cahiers de Civilisation midieoale 22 Poitiers: Univ ersite de Poitiers
pp. 2 19-41 and 383-92 .
- - (19 70) 'Los textos an timah ometan os mas antiguos en codices espaiioles'
Archives d'histoire doctrinale et luteraire du Moyen Age 45 Paris: Vrin pp. 149- 68.
- - (19 76) De Isidoro al Siglo X I: ocho estudiossabre la vida literana peninsular Bar celon a:
Ediciones EI Albir.
- - (19 79) Libros y librerias en la Riqja altomedieual Logroiio: Institut o de Estud ios
Rioj an os.
- - (1983) Codices visigoticos de la monarquia leonesa Leon: Centro de Estudios y
ln vestigacion San Isidro.
Dick, 1. (196 1) 'La passion arabe de S. Antoine Ruwah , nco-m ar tyr de Dam as (d. 25
dec 799)' u Museon 74 Leuven pp. 109- 33.
Dodds, JD. (1990) Architecture and Ideology in Early Medieval Spain University Park
London : Penn sylvania Stat e University Press.
- - (1992) ed. Al-Andalus: TI e Art qf Islamic Spain New York: M etropolitan Mu seum
of Art.
- - (1993) 'Islam, Christiani ty and the problem of religious art' in TI e art qf
medieval Spain A.D.500-1 200 (1993) New York : M etr opolitan Mu seum of Art.
pp . 27- 40.
Dozy, R . (188 1) Recherches sur l'histoire et la litterature des arabes d'Espagnependant le Moyen
Age 2 vols Leiden: Orient al Press.
- - ed. Levi-Pr ovencal, E. (1932) Histoire des musulmanes d'Espagnejusqu'a la conquite
de l'Andalousie par les Almoraoids (711- 1110) 2 vols Leide n: Brill.
- - tran s. C ustodio Vega, A. (1957) 'Articulo de Dozy sobre Recemun do de
Cordo ba ' in ibid . ed De la santa iglesia apostolica de Ilibens (Granada), Espana Sagrada
56 M adrid: Imp rent a y Edit or ial E. Maestre pp. 181- 96.
Du bler, C .E. (1946) 'So bre la cronica arab igo-bizantina de 74 1' Al-Andalus I I
pp. 298 - 322 .
D ub ois, J and Ren aud, C. eds (1976) Edition practiques des mariyrologes de Bede, de
l'anony me lyonnaiset de Horus Paris: Centre Nati onal de la Recherche Scient ifique.

211
 BlBLlOGRAPHY

Ducellier, A (1971) Le Miroir del'Islam: Muslims et Chretiens d'Orient au Moyen Age(VII'

- XII' siecles) Paris: Julliard.
Duran Gudiol, A (1975) De La Marca Superior de al-Andalus al reino deAragon Hu esca:
Caja de Ahorros de Zaragoza, Arag6n y La Rioja.
Dumville, D. (1979) 'Kingship, Gen ealogies and Regnal Lists' in Sawyer, P. and
Wood , LN. eds Early Medieval Kingship Leeds: University of Leeds.
-~ (1985) 'The West Saxon Genealogical Regnal List and the Chronology of
Early Wessex' Peritia 4 pp. 21-66 reprinted in Britons andAnglo-Saxons in the Early
MiddleAges Aldershot: Variorum.
Dunlop, D.M. (1954) 'Hafs ibn Albar - the last of the Goths' Journal if the Royal
Asiatic Society if Great Britain and Ireland 3-4 London 137-51.
- - (1955) 'Sobre Hafs ibn Albar al-Quti al-Qurtubi' Al-Andalus 20 pp . 211-13.
Duri, AA ed. and trans. Conrad L.1. (1983) The Rise if Historical Writing Among the
Arabs Princeton: Princeton University Press.
- - trans. Conrad L.1. (1987) TIe Historical Formation if the Arab Nation London:
Croom Helm.
Elliot,].K. (1993) TIe Apocryphal New Testament Oxford: Clarendon Press.
Epalza, M . de (1992) 'Mozarabs: an Emblematic Christian Minority in Islamic
Spain ' in TIe Legacy if Muslim Spain pp. 149-175.
- ~ (1994) 'Falta de obispos y conversi6n al Islam de los cristianos del-And alus'
Al-Qgntara 15 pp . 385-400.
Evans,]. (1998) TIe Theory andPractice if Ancient Astronomy Oxford: Oxford Univ ersity
Press.
Ewert, C ., Gladiss, A.v., Golzio , K.-H. and Wisshak, ].-P. (1997) Hispania Antiqua.
Denkmiiler des Islam: von den Anfiingen bis zum 12. Jahrhundert Mainz: Phillip von
Zabern.
Fabrini, E (1979) Paolo Orosio, Uno Storico Rome : Edizion e di storici e letteratura.
Fath al-Andalus ed . Molina, L. (1994) Madrid: CSIC .
Fear, A:I: (1997) Lives if the Visigothic Fathers Liverpool: Liverpool University
Press .
Fernandez Felix, A and Fierro, M.I. 'C ristianos y conversos al islam en al-Andalus
bajo los Omeyas. Una aproximaci6n al pro ceso de islamiz aci6n a traves de una
fuente legal andalusi del s.III/lX' forthcoming in Anejos de Archivo Espaiio! de
Arqueologia.
Ferre, E. (1967) 'Une source nouv elle pour I'histoire de l'Esp agn e musulmana'
Arabica 14 Leiden: Brill pp . 320 - 26.
Fierro, M .1. (1989) 'La obra hist6rica de Ibn al-Qutiyya' Al-Qgntara 12 pp . 485-512 .
-~ (1995) 'La falsificaci6n de la historia: Al-Yasa' b. Hazrn y su kitab al-Mughrib'
Al-Qgntara 15 pp . 1-37.
Fischel , WJ. (1954) 'Ibn Khaldun andJosippon' in Homenaje a Millas- Vallicrosa 2 vols
Barcelona: CSIC I pp . 587 -98.
- - (1967) Ibn Khaldun in Egypt Berkley Los Angeles: University of California Press.
Fletcher, R. (1989) TIe OJtest for El Cid Oxford: O xford Universit y Press.
- ~ (1992) Moorish Spain London: Weidenfeld and Nicholson .
-~ (2000) 'A Twelfth-Century View of the Spanish Past' in Maddicott, ].R. and
Palliser, D.M. eds TIe lvledievalState: essoys presented toJames Campbell London Rio
Grande: Hambledon Press pp. 147-62 .
Floriano, A C . (1949 - 51) Diplomatica Espanola delPeriodo Astur2 vols Oviedo: La Cru z.
Fontaine,]. (1973) L'Art preroman hispanique La Pierre-qui-vire: Zodiac.

212
 BlBUOGRAPHY

- - (1983) 'Mozarabie Hispanique et monde Carolingien: les echanges culturels

entre la France et l'Espagne du VIlle au X" siecle' Anuario 133 pp. 17-46.
- - and Pellistrandi, C. eds (1992) L'Europe Haitiere de l'Espagne Wisigothique
Madrid: Casa de Velazquez.
- - 1995 L :Art Mozarabe 2nd edn La Pierre-qui-vire: Zodiac.
Forcada Nogues, M . (19 70) 'Los libros de Anwa' en al-Andalus' in Vernet Gines ,
J and Samso,J eds Ellegado cientifico andalusi Madrid trans . Kennedy M . in The
Formation ofal-Andalus 2 pp . 305-28.
Franke, F.R. (1953) 'Die freiwilligen Martyrer von Cordova und das Verhaltnis des
Mozarabes zum Islam ' Spanische Forschungen des Gorresgesellschafl 13 pp. 1-170.
Frend, W (1984) The Rise ofChristianuy London: Darton, Longman and Todd.
Fuck , J (1965 ) 'Zum Kalendar von Cordoba von Jahre 961 ' Orientalische
Literatureeitung 7/8 Berlin : Akademie-Verlag pp. 333-7.
Gaiffier, B. de (1937) 'Les notices hispaniques dans Ie martyrologe d'Usuard'
Analecta Bollandiana 55 pp . 268-83.
- - (1938) 'L'inventio et Translatio de S. Zoilo de Cordoue' Analecta Bollandiana 56
pp.361 -9.
- - (1942) 'La Passion de SS Cyriaque et Paule ' Analecta Rollandiana 60 pp. 1-15.
- - (1954) 'La lecture des Actes des martyrs dans la priere liturgique en Occident.
A propos du passionaire hispanique' Analecta Bollandiana 72 pp. 134-66.
Garcia de Cortazar, JA. (1969) El dominio del monasterio de San M illan de la Cogolla
(Siglos X a XIII) Salamanca: Universidad de Salamanca.
Garcia Gomez, E. (1947) 'Algunas precisiones sobre la ruina de la Cordoba Omeya'
Al-Andalus 12 pp . 267-93.
- - (1965) 'Notas sobre la topografia cordobesa en los «Anales de Al-Hakam
II» por Isa al-Razi' Al-Andalus 30 pp . 319-79.
Garcia Moreno, L.A. (1989) Historia de Espana Visigoda Madrid: Ediciones Catedra,
Garen, S. (1995) 'Santa Maria de Melque and church construction under Muslim
rule' Journal qf the Society ofArchitectural Historians 5 I Charlottesville, Virginia
pp. 288-305.
Gaurier-Dalche, P. (1984) 'Notes sur la Cronica Pseudo-Isidoriana' Anuario 14
pp. 13-32.
Geary, P. (19 78) Furta Sacra: Thefls qf Relics in the Cental Middle Ages Princeton:
Princeton University Press.
Gil, J (19 70) ' En Torno a las Santas Nunilon y Alodia ' Revista de la Universidad de
Mndrid 19 pp . 74-135 .
Glick, '[ E (1979) Islamic and Christian Spain in the Barty Middle Ages. Comparative
Perspectives on Social and Cultural Formation Princeton: Princeton University Press.
- - (1995) From Muslimfortress to Christian castle: social and cultural change in medieval
Spain Manchester: Manchester University Press.
- - and Sunyer, 0. (1969) 'Acculturation as an explanatory concept in Spanish
history' Comparative Studies in Society andHistory II The Hague: Mouton 136-54.
Goetz, H .-W. (200 1) 'C oncepts of realm and frontiers from Late Antiquity to the
early Middle Ages' in Pohl W, Wood I and Reimitz H. eds The Transformation qf
Frontiers Leiden, Boston, Cologne: Brill. pp. 73-82.
Gomez Moreno, M (19 11) Iglesias mozdrabes Madrid. reprinted facsimile Granada:
Imprenta Sta . Rita. Monachil.
Gotor, JL. (1995) 'La polemica con Arnerico Castro' Atti dei Convegni Lincei 113
Rome: Accademia Nazionale dei Lincei pp. 31-52.

213
 BIBUOGRAPHY

Granja, E de la (1966) 'La Marca Superior en la Obra de al-Tldrf Estudios de Edad

Media de la Corona de Aragon 8 Salamanca: Universidad de Salamanca
pp . 447-545 .
- - (1969 - 70) 'Fiestas cristianas en al-Andalus' Al-Andalus 34 pp. I-53 and 35
pp . 119-42 .
Greenblatt, S. (199 1) Maroellous Possessions: the wonder qf the New World Oxford:
Clarendon Pr ess.
G riffiths, S.H . (1985) 'The Arabic account of'Abd al-Masih al-Nayrani al-Ghassani
Ie M useon 98 Leuven pp . 331-74.
- - (1990) 'The first Summa Theologica in Arabic: Christian Kalam in ninth-century
Palestine' in Conversion and Continuity pp . 15-32 .
Gruzinski, S. trans. Corrigan, E. (1988) the Conquest ofMexico: the Incorporation qf
Indian Societies into the Western World, I 6th-18th centuries London: Polity Press.
Guha, R . (1992) ed Subaltern Studies IV: Writings onSouth AsianHistory and Society Delhi :
Oxford University Press.
Gu erreiro, R . (1992) 'Le rayonnement de l'hagiographie hispanique en Gaule
pendant Ie haut Moyen Age: circulation et diffusion des Passions hispaniques' in
Fontain e, E and Pellistrandi, C . eds L'Europe Hentiere de l'Espagne Wisigothique
Madrid: Casa de Velazquez pp . 137-57.
Gui chard, P. (19 76) Al-Andalus: Estructura antropologica de una sociedad islamica en
occidente Barcelona: Barra!.
H aight, A.L. (1965) Hrotswitha ofGandersheim: herlift, times andworks anda comprehensive
bibliography New York: Hrotswitha C lub.
El-Hajj i, AA (19 70) Andalusian Diplomatic Relations with Western Europe during the
Umayyad Period (AH 138-366/ AD 755-976) An Historical Stu4J Beirut: Dar
al-Irshad.
H ajji Khalffa ed and trans. Fhigel, G. (1835-58) Kashfal-eunun 'an Asamial-kutub wa-
l-Juniin 7 vols Leipzig London: Bentley.
Haldon, J.E (1990) Byzantium in the Seventh Century Cambridge: Cambridge
University Press.
H arv ey, L.p. (1992) 'British Arabist s and al-Anda lus' Al-Qgntara 13 pp. 423- 36.
Hayward, P.A (1993) 'The Idea of Innocent Martyrdom in Late Tenth- and
Eleventh-century Eng lish H agiology' Studies in Church History 30 London: Nelson
pp.81 -92.
Hill, D.R. (1993) Islamic Science and Engineering Edinburgh: Edinburgh Univ ersity
Press.
Hillenbrand, R . (1992) "The ornament of th e world " : medieval Cordoba as a
cultural centre' in the Legacy ofMuslim Spain pp. 112- 135.
Hillgarth,J.N. (1970) 'Historiography in visigothic Spa in' La stonogrofia altomedievale,
Settimane pp. 261-31 I.
- - (1985) 'Spanish historiography and Ib erian reality' History and Theory 24.1 The
Hague: Mouton pp . 23-42.
Historia Silense ed Perez de Urbel, J. and Gonzalez Ruiz-Zorilla, A (1959) Madrid:
CSIC .
H itchcock, R . (198 1) 'Q uienes fueron los verdaderos mo zar abes ' Nueva reoista de
filologia hispanica 3 1 M exico City: Sem estral pp . 574- 85.
Humphreys, S. (199 1) Islamic History: A Frameworkfor Inquiry London: LB. Tauris.
Ibn al-Abbar ed Codera, E (188 7-90) Al-Takmila Madrid: Biblioteca Arabico-
Hispana.

214
 BIBLIOGRAPHY

~- ed Mu'nis, H. (1963) Al-Hulla al-siyard' 2 vols Cairo.

Ibn 'Abd al-Rabbihi (1940-9) Kitab al-'Iqd al-Farid8 vols Cairo: Matb a' al-Lajna t
al- Ta'IIf wa-I-Tarjama wa-l-Nashr,
Ibn AbI Usaybi 'a (1965) 'Uyun al-anbd' jt tabaqdt al-atibbii' Beirut: Dar Maktabat
al-Haya.
Ibn 'AsIm ed and trans. Forcada Nogu es, M . (1993) Tratado sobre los Anwa' y los
tiempos - capitulo sobre los meses Barcelona: CSIC .
Ibn Bashkuwal ed Codera, F. (1882) Al-Sila jt ta'rikh a'immat al-Andalus Madrid:
Biblioteca Arabico- Hispana.
Ibn al-Faradi ed Codera, F. and Ribera y Tarrago, j. (1891-2) Ta'rikh. 'ulama'
al-Andalus Madrid: Bibliotheca Arabico-Hispana.
Ibn Hablb Kitab al-Ta'nkli ed with commentary Aguade,j. (199 1) Madrid: CSIC .
Ibn Hawqal trans. Romani Suay, MJ. (1971) Corfiguracion del mundo (fragmentos
alusivos al Magreb y Espana) Valencia: Antonio Ubieto Arteta.
Ibn Hayyan ed Antuna M.M. (1937) Al-Muqtabis, tome troisieme, chronique du regne du
a
caliph umayyad 'AbdAllah Cordoue Paris: Larose .
~- trans. Gurai'eb ,j. (1958) CHE 28-9 pp. 166-73.
~- trans. Levi-Provencal, E. and Garcia Gallo , A. (1954) 'Textos inedito s del
Muqtabisde Ibn Hayyan sobre los origenes del reino de Pamplona' Al-Andalus 19
Madrid pp . 295-315.
~ - ed Al-Hajji A.A. (1965) Al-Muqtabisfi akhbdr balad al-Andalus (al-Hakam II)
(Al-Muqtabis VII) Beirut: Dar al-Tahqafah,
~- trans. Garcia Gomez, E. (1967) Anales Palatinos del Califa de Cordoba al-HakamII
Madrid: Sociedad de Estudios y Publicaciones.
~ - ed Makki, M .A (1973) Al-Muqtabas min anbd' ahl al-Andalus Beirut: Dar
al-Kutub al-'Arabi.
~ - ed Chalmeta Gendron, P. Corriente, F. and Subh, M . (1979) Al-Muqtabas V
Madrid: CSIC.
~- trans. Corriente, E and Viguera Molins , M .:J. (198 1)Al-Muqtabas VZaragoza:
Anubar.
Ibn Hazm Kitab al-jisal (1996) Beirut: Dar al-Kutub al-'IlmIya.
~ - ed Levi-Provencal, E. (1983) Yamharat ansab al-'arab Beirut : Dar al-Kutub
al- "Ilmiya.
Ibn Idhari ed Dozy, R . (1848- 5 1) Histoire de l'Aftique et de I'Espagne intitulie al-Bayano-
l-mogrib par Ibn AdharideMaroc etfragments dela Chronique deArib de Cordoue Leyden .
~- revised ed Levi-Provencal, E. and Colin , G. (1948- 5 1) Al-Baydn al-Mughnb
2 vols Leyden: Brill.
~ - ed Levi-Provencal, E. (1930) Al-Baydn al-Mughrib, tome troisieme, Histoire de
I'Espagne Musulmane au XI' siecle Paris: Geuthner.
Ibn Ishaq ed Wtistenfcld F. (1858-60) Das Leben Muhammeds reprinted (196 1) 3 vols
Frankfurt am Main: Minerva.
Ibn Juljul ed Said, F. (1950) Tabaqdt al-atibbd' ioa-l-hukamd' Cairo.
~- trans. Vernet Gines, j. (1968) 'Los medicos andaluces en el Libro de las
generaciones de medicos de Ibn Yulyul' Anuario 5 pp . 445-62.
Ibn al-Kalbl ed. Caskel, W. (1966) Gamhara al-Nasab: dasgenealogische Werkdes Hisham
ibn Muhammad al-Kalbi 2 vols Leiden: Brill.
Ibn Khaldun ed Dagher, j. (1958- 68) Kitab al-'Ibar 7 vols Beirut.
~- trans. Rosenthal, F. (1958) The Muqaddimah, an Introduction to History 3 vols New
York: Bollingen Foundation.

215
 BIBLIOGRAPHY

Ibn Khallikan ed. 'Abbas , I. (19 70- 7 1) Wqfqyat al-a'yiin Beirut.

- - trans. MacGuckin de Slan e ( 1 8 4 2 ~ 7 1 ) Ibn Khallikan 's Biographical Dictionary
Paris reprinted (19 70) 4 vols Beirut: Librairie du Liban .
Ibn al-Khatib ed Levi-Provencal, E. (1956) A'mal al a'lam fi man buyi'a qabla 'l-ihtiliim
min muluk al-Isliim Beirut: Dar al-Makshuf,
Ibn Qutayba ed Hamidullah, M . and Pellat, Ch. (1956) Kitab al-anuui' Hyd erabad:
Dairatu'l-ma'arif il-usmaniya.
Ibn al-Qjitfya ed Gayangos, P., Saavedra, E. and Codera , E (1868) Historia de La
conquista de Espana deAbenalcotia de Cordoba Madrid reprinted with trans . Ribera y
Tarrago,]. (1926) Madrid: Real Academia de la Hi storia.
- - trans. Nichols, ].M. Ibn al-Qutiyya, The History of the Conquest of
Al-Andalus. Unpublished PhD Dissertation University of North Carolina 1975.
Ibn Rushd al-jadd ed. I:IajjI M . et al (1988- 9 1) Kitab al-bayan ioa-l-tahsil toa-l-sharh
wa-l-taU!Jih uia-l-ta-Iil jt masd'il al-mustahraja 19 vols, 2nd edn. Beirut .
Isidore ed Lindsay, W.M . (19 11) Isidorii Hispalensis Episcopi Etymologiarum siveoriginum
libri XX 2 vols Oxford: Clarendon Press.
Isla Frez, A. (1992) La sociedad gallega en la Alta Edad Media Madrid: CSIC.
J erome ed Glorie, F. (1964) Commentaria in Hiezechielem CCSL 75~ 7A.
Jimenez de Rada, Rodrigo ed Fernandez Valverde,]' (1987) Historia de rebus
Hispaniae Turnholt: Brepols.
- ed Fernandez Valverde ,]. and Estevez Sola,]. (1999) Opera Omnia III. Historiae
minores. Dialogus libri vitae Turnholt: Brepols .
Jimenez Pedrajas, R. (1960) 'Las relacion es entre los cristianos y los mu sulmanes en
Cordoba' Boletin de la Real Academia de Cordoba de Ciencias, Bellas Letras y Nobles
Aries 80 Cordoba pp . 107~236.
- - (1967) Las datas del martirio y traslado de las santas Nunilo y Alodia Cordoba:
Tipografia Catolica,
- - (19 70 'San Eulogio de Cordoba, autor de la pasion fran cesa de los martires
moz arabes cordobeses Jorge, Aurelio y Natalia' Anthologica Annua 15 Rome:
Insituto Espafiol de Historia Ecclesiastica pp. 465 -583.
Jordan, MJ. (199 7) The Invention if Sodomy in Christian Theology Chicago London :
University of Chicago Press.
John of Biclar ed. C ampos,]. (1960) Juan de Biclaro Obispo de Gerona, su Vida y su obra
M adrid: CSIC .
Kaegi, W.E. (1992) Byzantium and the EarlY Islamic Conquests Cambridge: Cambridge
University Press.
Kassis, H . (1994) 'Arabi c-speaking christians in Al-Andalus in an age of turmoil
(fifth / elevent h century until a.h. 478/a.d. 1085)' Al-Qgntara 15 pp . 401-22.
Kedar, B.Z. (1984) Crusade and M ission: European Attitudes towards the Muslims
Princeton: Princeton University Pr ess.
Kennedy, H . (1986) the Prophet and the Age if the Caliphates London New York:
Longman.
- - (1996) Muslim Spain and Portugal London: Longman .
- - (199 7) 'Fro m O ral Tradition to Written Record in Arabi c Genealogy' Arabica
44 Leiden : Brill pp . 531 -44.
Kh alidi , T (1994) Arabic historical thought in the classical period Cambridge: Cambridge
University Pre ss.
Al-Khushani ed. and trans. Ribera y Tarrago ,]. (19 14) Historia delosjueces de Cordoba
porAijoxani Madrid: Imprenta y Editorial E. Maestre.

216
 BIBLIOGRAPHY

King, PD. (1972) Law and Society in the Visigothic Kingdom Cambridge: Ca mbridge
University Press.
Kuh ayla, "U, (1985- 6) ' Kitab al-tawa rikh li-Bawlus Urusiyus wa tarjamaruhu'
Reoista del Instituto Egipcio de Estudios Islamicos 23 Madr id pp. 119-37.
- - (1993) T ar'ikl: al-Na sdra fi- l-Andalus Cai ro: Dar al-Kutub al-Islamiya ,
Kunitzsch , P (1994) 'Abd al-Ma lik b. H abib 's Rook on the Stars' Zeitschriflfiir die
Geschichte der Arabisdi-Islamischen Wissenschaften 9 Frankfurt pp. 161-94 and II
pp. 179-88 repr inted in The Formation qf al-Andalus 2 pp. 277- 304.
Labarta, A. and Barcelo, C . (1987) 'Las fuentes arabes sobre al-Zahra': estado de la
cuestion' Cuademos de Madinat al-Zahra I Cordoba pp. 93- I06.
Lacroi x, B. (1965) Orose et ses idees Montreal: Institut d'etud es medi evales,
Lamb ert, E. (1953) 'Le voyage de Sain t Euloge dan s les Pyrenees en 848' Estudios
dedicados a Menendez Pidal 4 Madrid: CSIC pp. 557- 67.
Lamor eaux j.C . (1996) ' Early eastern christian responses to Islam ' in Medieval
Christian Perceptions pp. 3-32.
Landes, R. (1988) ' Lest the Millennium be fulfilled: Apocalyptic Expectations and
the Patt ern of Western Chronography 100-800C E' in Verb eke, W., Verhelst, D.
and Welkenhuysen , A. eds The VIe and Abuse of Eschatology in the Middle Ages
Leuven: Leuven University Press pp. 137-211.
Lapiedra Guti err ez, E. (1994) 'Los mart ires de Cordoba y la politica anticristiana
contemporanea en Oriente' Al-Qgntara 15 pp. 453-62.
- - (1997) Como los musulmanes llamaban a los crtstianos hispanicos Alicant e: Baiiul
Imp resores.
Lap idus, I.M. (1967) Muslim Cities in the LaterM iddle Ages Cambridge, Ma ssachussets:
H arvard University Press.
Lasko, P (1972) Ars Sacra 800- 1200 Harmondsworth: Penguin .
Levi Della Vida , G. (1954) 'La traduz ione arabe delle storie di Orosio' Al-Andalus 19
pp. 257-93.
- - (1962) 'Un texte Mozarabe d'Histoire Universelle' in Etudes D'Onentalisme
dediees a la memoire de Lem-Prooencal 2 vols Paris: Maissoneuve et Larose I
pp. 175- 83.
- - (1971) Note di Storia Letteraria Arabo-Ispanica Rome: Instituto per I'Oriente.
- - (198 1) Arabic Papyri in the University Museum in Philadelphia (Pennslyoania) Rom e
repr inted (1989) Atti della Academia Narionale dei Lincei series 3 vol 25 fasc.1 Rome .
Levi-Provencal, E. (1931) Inscriptions Arabes d'Espagne Leiden: Brill.
- - (1937) 'Un echange d'amb assades entre Cordoue et Byzance au IX" sieclc'
Byrantion 12 Paris: Champion pp. 1-24.
- - (1950- 3) Histoire de l'Espagne musulmane 3 vols Paris: Maisonneuve et Laro se
and Leiden: EJ. Brill.
Lewis B. 'The use by Muslim historians of non-Muslim source s' in Historians
pp.35-45.
Lewis, C.'!. and Short, C . (1969) A Latin Dictionary Oxford : Clarendon Press.
Leyser K. (1985) ' Liutprand of Cr emon a, Preacher and Homilist ' in Walsh, K. and
Wood , D. eds The Bible in the Medieval World Oxford : Blackwell Publishers
pp. 43-60.
- - ed. Reuter, T. (1994-) K Leyser: Communications and Power in Medieval Europe: The
Carolingian and Ottonian Centuries London : Hambledon Press.
Linder, A. (1975) 'T he myth of Constant ine the Gr eat in the west: sources and
hagiographi c commemorat ion' Studi Medievali 16 Turin : Loescher pp. 43-95.

217
 BIBLIOGRAPHY

Linehan, P. (1993) His/myand theHistorians qf Medieval Spain Oxford: Clarendon Press.

Liudprand of Cremona trans. Wright, EA. ed. Norwich, JJ. (1993) The Embassy to
Constantinople and Other Writings London: j.M. Dent.
LOpez, C.M. (1964) ' Problemas Historicos en torno a las Santas Nunila y Alodia'
Principe de Viana 24 Pamplona pp. 155-63.
- - (1965) 'En torno a la patria de las Santas Nunilo y Alodia ' Principe de Vzana 26
Pamplona pp . 385-404.
- - (1970) 'M as sobre la problematica en torno a las San tas Nunilo y Alodia'
Principe de Viana 28 Pamplona pp . 101-32.
LOpez Pereira, J.E. (1980) Estudio critico sobre la cronica mazarabe de 754 Zaragoza:
Anubar.
Lopez Quiroga, J. and Rodriguez Lovelle, M . (1984 ) 'U na aproximacion
arqueologica al problema historiografico de la «Despoblacion y Repobl acion
en el valle del Duero> s.VIII- X I' Anuario 21 pp . 3-10.
Madoz,j. (1945) 'EI viaje de San Eulogio a Navarra y la cronologia en el epistolario
de Alvaro de Cordoba' Principe de Viana 20 Pamplona pp. 415 -23.
Makki, A. (1957) 'Egipto y los origines de la historiografia arabe-espafiola' Revista
del Instituto Egipcio de Estudios Isldmicos 5 Madrid pp . 157-248.
Manzanares del Cine, M. (1972) Arabistas espaiioles del siglo XIX Madrid: CSIC.
Manzano Moreno, E. (199 1) La Frontera de Al-Andalus en Epoca de los Omeyas Madrid:
CSIC.
- - (1992) 'Oriental Topoi in Andalusian Historical Sources' Arabica 39 Leiden:
Brill pp . 42-58.
- - (1994) 'Christian-Muslim Frontier in Al-Andalus: Idea and Reality ' in Agius,
D. and Hitchcock, R. eds The Arab Influence in Medieval Europe Reading: Ithaca
Press pp. 83-99.
- - (1998) 'Arabes , bereberes y indigenas: al-Andalus en su primer periodo de
forrnacion' in Barcelo, M . and Toubert, P. eds L'Incastellamento Rome: Ecole
Francaise de Rome and Escuela Espanola de Historia y Arqueologia en Roma.
Marin, M . (1992) 'Arabistas en Espana: un asunto de fam ilia' Al-Qgntara 13 Madrid
pp . 379 -93.
Marin Guzman, R . (1995) 'The Causes of the Revolt of 'Umar ibn Hafsun in
Al-Andalus 880 -928: A Study in Medieval Islamic Social History' Arabica 42
Leiden: Brill pp. 180- 221.
Martin, G. (1984) 'La chute du royaume visigothique d'Espagne dans l'historio-
graphic chretienne des VII " et IX " siecles' Cahiers de linguistique hispanique
medieuale IX Paris: Klincksieck pp. 207-33.
Martinez Gazquez, j. (1981) 'Algunos observaciones al texto del Calendario de
Cordoba' Al-Qgntara II pp. 319 -42.
- - and Sarnso, j. (198 1) Una nueva traduccion latina del Calendario de Cordoba (siglo
XIII) in Vernet Gines j. cd . Textos y estudios sobre astronomia espanola en el siglo XIII
Barcelona pp. 9-78.
Martinez Gros, G. (1992) L'ideologie omeyyade: La construction de la ligitimite du califat de
Cordue X ' -Xl' siecles Madrid: Casa de Velazquez .
- - (1997) Identite Andalouse Aries: Sindbad/Actes Sud.
Le Martyrologe d'Usuard cd . Dubois, j. (1965) Brussels: Societe des Bollandistes .
Al-Mas'udi cd . and trans. Barbier de Meynard, C . and Pavet de Courteille, A.
(1864) Mung al-Dhahab (Les Prairies d'Or) 9 vols Paris: Imprimerie Imperiale et
Imprimerie Nationale.

218
 BIBLIOGRAPHY

Al-Maqqari ed. Dozy, R ., Dugat, G., Krehl, L. and Wright, W. (1855- 6 1) Analectes
sur l'histoire et la liuerature des Arabes d'Espagne par al-Makkarf 2 vols Leiden ,
reprinted 1967 London: Oriental Press.
Mazzoli-Guintard, C. (1997) 'Remarques sur Ie fonctionn ement d'un e capitale a
double polarite: Madinat al-Zahni'-Cordue' Al-Qgntara 17 pp. 43-64.
McCluskey S.C . (1998) Astronomies and Cultures in Early Medieval Europe Cambridge:
Cambridge University Press.
McCormick, M. (1995) 'Byzantium and the West 700-900' in McKittterick, R. ed.
New Cambridge Medieval History II c.700-c.900 Cambridge: Cambridge University
Press pp. 349-82.
McGinn, B. (1979) Visions iftheEnd: Apocalyptic Traditions in theM iddle Ages New York:
Columbia Univer sity Press.
McWilliam ,j.E. (1990) 'The context of Spanish Adoptionism: a review' in Conversion
and Continuity. pp. 75-88.
Medi ano, ER. (1994) 'Acerca de la pobla cion arabizada del Reino de Leon (siglos x
y xi)' Al-Qgntara 15 pp. 465-72.
Menend ez Pidal, G. (1992) Los Caminos en la Histona de Espana Madrid: Cultura
Hispanica.
Menocal, M.R. (1992) 'AI-Andalus and 1492: the ways of remembering' in The
Legacy if Muslim Spain pp. 483-504.
Millas Villicrosaj.M. (1943) 'Ellibro de Agricultura de Ibn Wiifid y su influencia en la
agricultura del Renacimiento' Al-Andalus 8 pp. 281- 332.
Millet-Gerard, D. (1984) Chretiens mozarabes et culture islamique dans l'Espagne des
VIII'- IX ' siecles Paris: Etudes Augustienn es.
Mishin , DJ. (1999) 'Saqlabl servants in islamic Spain and North Africa in the early
Middle Ages' PhD Dissertation: Central European University, Budap est.
Molina, L. (1984) 'O rosio y los geografos hispanomusulmanes' Al-Q.antara 5
pp.62-92.
~~ (1989) 'Los Ajbar Majmua y la historiografia arabe sobre el periodo omeya en
AI-Andalus' Al-Qgntara 10.2 pp. 513-42.
~~ (1998) 'U n relato de la conquista de al-Andalus ' Al-Qgntara 19 pp. 39-65.
Moral, T. (1975) 'EI mon asterio riojano de las santas martires Nunilo y Alodia'
Principe de Viana 36 Pamplona pp. 435-45 .
Morris, C . (1972) The Discovery ifthe Individual/ 050-/200, reprinted (1987) Toronto,
Buffalo London: University of Toronto Press.
Muhlberger, S. (1990) The Fiflh-century Chroniclers Leeds: Francis Cairns.
Musurillo, H. (1972) The Acts if the Christian Martyrs Oxford: Clarendon Press.
Netton, I.R . (1996) Text and Trauma: An East- West Primer Richmond, Surrey: Curzon
Press.
Newman, N.A. ed. (1993) The early Christian-Muslim Dialogue: a collection if documents
from the first three Islamic centuries (632- 900 AD) Hatfield, Pennsylvania:
Interdisciplinary Biblical Research Institute .
Nikephorus Patriarch if Constantinople Short History ed. and trans . Mango, C. (1990)
Washington D.C .: Dumbarton Oaks.
Nir enb erg, D. (1996) Communities if Violence; Persecution if Minorities in the Middle Ages
Princeton: Princeton University Press.
North, j. (1994) The Fontana History if Astronomy and Cosmology London: Fontana.
Noth, A. (1994) The Early Arabic Historical Tradition: A source-critical study Princeton :
Darwin.

219
 BIBLIOGRAPHY

Olague, I. (1971) La Reoolucion Islamica en Occidente Barcelona: Guadarrama.

Oliver Perez , D. (1994) 'Sarraceno: su etimologi a e historia' Al-Opntara 15 pp. 99-130.
Orlandis,]. (1976) La iglesia en la Espana visigotica y medieval Pamplona: Universidad
de Navarra.
Orosius trans. Deferrari, RJ. (1964) Orosius, TIle Seven Books of History Against the
Pagans Washington D.C .: Catholic University of America Press.
- - ed . Bately, ].M. (1980) the Old English Orosius Oxford: Early English Text
Society.
- - ed. Badawi, 'A. (1982) 'Unisius, Td'rtkl: al-'dlam Beirut: Al-mu'assasa al-'arabiya
li-dirasat al-nashr.
- - ed . Lund, N. trans. Fell, C . (1984) Two Vqyagers at the Court ofKing Alfred: The
Ventures qf Othere and Wulfstan together with the Description ofNorthern Europe from the
Old English Orosius York: Sessions.
Palmer, A. (1993) The Seventh Century in the West-~rian Chronicles Liverpool: Liverpool
University Press.
Parisse, M. 'Introduction' in ibid . and Oexle, O. eds (1993) L'abbaye de Goree au X '
siecle Nancy: Presses Universitaires de Nancy pp. 9-24.
Pasionario Hispanico ed . Fabrega Grau, A. (1953- 5) 2 vols Madrid Barcelona:
Instituto P. Enrique Florez.
La Pasion de San Pelayo ed . and trans. Rodriguez Fernandez, C. (199 1) Santiago de
Compostela: Universidad de Santiago de Composteia.
- - trans. Bowman, ].A. 'Raguei, the Martyrdom qf St. Pelagius' in Head, 1: ed.
(2000) Medieval Hagiography, an Anthology New York London: Garland Publishing
pp. 227-36.
Pastor de Togneri, R . (1995) 'EI feudalismo en la obra de Sanchez-Albornoz y las
interpretaciones actuales' Atti dei Convegni Lincei 113 Rome pp . 19- 30.
Pavon Maldonado, B. (1988) 'Entre la historia y la arqueologia. EI enigma de la
Cordoba califal desaparecida' Al-Qgntara 9 pp . 169-98 and 403-26.
Pedersen,]. trans. French, G. (1984) the Arabic BookPrinceton: Princeton University
Press.
Peeters, P. (1939) 'La Passion de Saint Pierre de Capitolias' Analecta Bollandiana 57
pp.67-74.
Pelikan, j. (1978) the Growth qf Medieval 7heology (600-1300) Chicago London:
University of Chicago Press.
Pellat, C. (1954) 'Ibn Hazm, Bibliographie et Apologiste de I'Espagne Musulmane'
Al-Andalus 19 pp. 53-102.
- - (1962) 'The Origin and Development of Historiography in medieval
Hispania' in Historians pp . 118-25.
Peiiarroja Torrejon, L. (1993) Cristianos bajo el islam: los mozdrabes hasta la reconquista de
Valencia Madrid: Gredos.
Perez de Urbei, j. (1926) 'O rigen de los himnos mozarabes' Bulletin Hispanique 28
Bordeaux: Feret pp . 5-21, 113-39,209-45 and 305-20.
- - (1934) Los monjes espaiioles en la edad media Madrid: Imprenta y Editorial E.
Maestre.
Pohl , W (1994) 'Tradition, Ethnogenese und literarische Gestaltung: eine
Zwischenbilanz' in Brunner, K. and Merta, B. eds Ethnogenese und Uberluferung
Vienna Munich: Oldebourg pp . 9-26.
- - ed . (1997) Kingdoms qf the Empire: the Integration ofBarbarians in Late Antiquity
Leid en: Brill.

220
 BIBLIOGRAPHY

Pons Boigues, E (1893) Ensayo Bio-Bibliogr4fico sobre los Historiadoresy Geogr4fos Arabigos
Espaiioles reprinted facsimile (1993) Madrid: Ollero y Ramos.
Porres Martin-Clcto, j. (1985) Historia de Tulqytula (711- 1085) Toledo : Instituto
Provincial de Investigaciones y Estudios Toledanos.
Proudfoot, A. (1974) 'The Sourc es of Theophanes for the Heraclian dynasty'
Byeantion 44 Paris: Champion pp. 367-439.
Le Psautier Mozarabe de Haft le Goth ed. and trans . Urvoy, M-Th . (1994) Toulouse :
Publications de l'Universite de Toulouse .
Quentin, H. (1908) Les martyrologes historiques Paris: LecofTre.
Reilly, B. (1992) the Contest qf Christian and Muslim Spain, 1031-1157 Oxford:
Blackwell Publishers .
Reinink, G. (1988) 'Pseudo-Methodius: a concept of history in response to the rise
of Islam' in Cameron, A. and Conrad, L.1. eds the Byzantine and Early Islamic
Near East I, Problems in theliterary Source Material Princeton: Darwin pp. 149-88.
- - (1988b) 'Pseudo-M ethodius und die legende vom Romische Endkaiser' in
Verbeke, W., Verhelst, D. and Welkenhuysen, A. eds 771e Use and Abuse qf
Eschatology in theMiddle Ages Leuven : Leuven University Press pp. 82-111.
Ribera y Tarrago, j. (1928) Disertaciones y opusculos 2 vols Madrid: Imprenta y
Editorial E. Maestre.
Riche P. (1992) 'Les refugies wisigoths dans Ie monde carolingien ' in Fontaine, j.
and Pellistrandi, C. eds L'Europe Heritiere de l'Espagne Wisigothique Madrid: Casa
de Velazquez pp. 177-84.
Rosenthal, E (1962) 'T he influence of Biblical Tradition on Muslim Historio-
graphy' in Historians pp. 35-45 .
- - (1968) A History qf Muslim Historiography Leiden: Brill.
Ross, DJ.A. (1955) 'Illustrated Manuscripts of Orosius' Scriptorium 9 Anvers:
Staandard Boekhandel pp. 35-56.
Roth , N. (1990) jewish collaborators in Alfonso's scientific work' in Burns, R.1. ed.
Alfonso X the Learned of Castille and his thirteenth-century Renaissance Philadelphia:
University of Pennsylvania Press pp. 59-71 .
Rubiera Mata, MaJ. (1980) 'La Mesa de Salomon' Awraq 3 Madrid: Instituto
Hispano-Arabe de Cultura pp. 26-31 .
Ruggles D. E (2000) Gardens, Landscape and Vision in the Palaces qf Islamic Spain
University Park, Pennsylvania: Pennsylvania University Press.
Russell, j.C. (1938) 'Chroniclers of Medieval Spain ' Hispanic Review 3 Philadelphia
pp.218-35.
Saavedra, E. (1892) Estudio sobre la invasion de losArabes Madrid: EI Progreso.
Sabbas, E. (1934) 'Quelques types de marchands des IX'" ct X'" siecles' Revue beige de
philologie et d'histoire 13 Brussels: Falk pp. 176-87 .
Sahas , DJ. (1972) John qf Damascus on Islam: the 'Heresy qf the Ishmaelites' Leiden : Brill.
Sa'id the Andalusi cd. and trans . Salem, S.1. and Kumar, A. (1996) Science in the
Medieval World: 'Book cftheCategories ofNations' Austin, Texas: University of Texas
Press.
Salisbury, j.E. (1991) Church Fathers, Independent Virgins London: Verso.
Samir, K. S. and Neilsen,j.S. eds (1994) Christian Arabic Apologetics during the Abbasid
Period (750-1258) Leiden : Brill.
Samso,j. (1978) 'La tradicion classica en los calendarios agricolas hispanoarabes y
norteafricanos' in Second International Congress ofStudies on Cultures ofthe Western
Mediterranean Barcelona: Universidad Autonoma de Barcelona pp. 177-86.

221
 BIBLIOGRAPHY

- - (1983) 'Sobre los materi ales astronomicos en eI « Calendario de Cordoba»

y en su version latina del siglo VIII' in Vern et Cin es J. ed. Nueoos estudios sobre
astronomia espanola en el siglo de Alfonso X Barcelon a: CSIC pp. 125-38.
- - (1983b) 'La primitiv a version arabe del Libro de los Cru ces' in Verne t ibid.
pp. 149-61 .
San chez-Albornoz y Menduifia C . (1945) 'Donde y cuand o murio Don Rodrigo,
ultimo Rey de los Codos?' CHE 3 pp. 5-105.
- - (1956) Espaiia: un enigma hist6rica 2 vols Buenos Aires: Sudam erica.
- - (1967) Investigaciones sobre histonogrofia hispana medieval (siglos VIlJ al XI/) Buenos
Aires: Univer sidad de Buenos Aires.
- - (1977) En tomo a los origenes delfiudalismo 2 vols Buenos Aires: Universidad de
Buenos Aires.
Sanchez Alonso, B. (194 1- 50) Historia de la Historiograjia espaiiola 3 vols Madrid:
CSIC.
Sandoval, P. de (16 10) Antiguedad dela cibdady iglesiade T uyy delos obispos que sesave que
avido en ella Braga reprinted facsimile (1974) Barcelona : EI Albir.
Scales, P.C. (1994) The Fall ofthe Caliphate ofCordoba Leiden: Brill.
- - (1997) 'Cordoba under the Umayyads: a 'Syrian' garden city?' in Roe, C. de
and Verhaege, E eds (1997) Urbanism in Medieval Europe Zellik: LA.P.
Rapporten.
Scheindlin, R. P. (1992) 'TheJ ews in Muslim Spain ' in The Legacy ofMuslimSpain pp.
188-200.
Senac, Ph. (1982) Provence et piraterie sarrasine Paris: Maissoneuve et Larose.
Serrano, D. (199 1) 'Dos fetuas sobre la expulsion de mozarabes al Magreb en 1126'
Anaquel de Estudios Arabes II Madrid: Universidad Complutense de Madrid
pp . 162-82.
Simon et, EX. (1871) 'Santoral hispano-mozarabe escrito en 961 por Rabi ben
Zaid, obispo de Iliberis' reprinted in Custodio Vega, A ed. (1957) De la santa
iglesia apostolica de lliberis (Granada), Espaiia Sagrada 56 Madrid: Imprenta y
Editorial E. Maestre pp. 123-59.
- - (1903) Historia de los Moouabes de Espana 2 vols Madrid reprinted (1967)
Amsterdam: Ori ental Press.
Slupecki, L.P. trans. Szymanska, 1. (1994) Slavonic Pagan Sanctuaries Warsaw: Polish
Academy of Sciences.
Smith ,J.H . (1971) Constantine the Great New York; Scribn er.
Smyth, A.P. (1995) King Alfred the Great Oxford: Oxford University Press.
Southern, R . (1962) Western Views rif Islam in the M iddle Ages Cambridge,
Massachuss ets: H arvard University Press.
Stanton, C .M. (1990) Higher Learning in Islam: the Classical Period, A.D. 700-1300
Savage, Maryland: Rowm an and Littlefield.
Staubach, N. (1989- 90, 199 1) 'H istoria oder Satira? Zur literarischen Stellung
Liudprands von Cremona' M ittellateinischesJahrbuch 24-5 Cologne : Selbstverlag
pp.46 1-87 .
Stern, S.M . (196 1) 'A Letter of the Byzantine Emp eror to the Court of the Spanish
Caliph al-Hakam (96 1- 76)' Al-Andalus 26 pp. 37-42.
Sutherland,J.N. (1975) 'The idea of revenge in Lombard society in the eighth and
tenth centuries: the cases of Paul the Deacon and Liudprand of Cremon a'
Speculum 50 Cambridge, Massachussets: Medie val Acad emy of America
pp. 391-410 .

222
 BIBLIOGRAPHY

- - (1988) Liudprand cf Cremona, Bishop, Diplomat, Historian. Studies qf the Man and his
Age Spoleto : Centro italiano di studi sull'alto medievo.
AI-TabarI ed. De Goeje, MJ. (1897- 190 1) Ta'rikh al-rusal ioa-l-mulak Leiden .
Taha, A.D. (1989) the Muslim Conquest and Settlement qfNorth Africa and Spain London
New York: Routledge.
Teres, E. (1957) 'Linajes arabes en Al-Andalus segun la «Yamhara» de Ibn Ha zm'
Al-Andalus 22 pp. 54-103.
Tester,]' (1987) A History qf Western Astrology Woodbridge: Boydell.
Thomas, N. (1994) Colonialism 's Culture: Anthropology, Travel and Government Oxford:
Polity Press.
Thompson,].W (1929) 'The introduction of Arabic Science into Lorraine in the
Tenth Century' Isis 12 Brussels: Weisenbach pp. 184-93.
The Times Atlas of World History ed. Parker, G. (1993) 4th edn London: Harper
Collins.
Tisserant, E. and De Bruyne , D. (19 10) 'Une feuille arabo-latine de l'Epitre aux
Galates' Revue Biblique NS 7 Paris: LecofTre pp. 321-43.
Toomer, GJ. (1968) 'A Survey of the Toledan Tables' Osiris 15 Bruges: Saint
Catherine Press pp. 5-174.
Toynbee, A. (1973) Constantine Porphyrogenitus andhis world London New York: Oxford
University Press.
Triki, H . (2000) 'Al-Andalus, ce jardin des poetes . . . VII"-XI e siecles' in Les
Andalousies de Damas Ii Cordoue Paris: Hazan/lnstitut du Monde Arabe.
pp.208-25.
Troupeau, G. (1968) 'Le Livre du temps de Ibn Masawayh' Arabica 15 Leiden : Brill
pp. 113-42.
- - (1971) 'La litteraturc arabe chrcticnne du X" au XII" siccle' Cahiers de
civilisation medieuale 14 Poitiers: Universite de Poitiers pp. 1-120.
Urvoy, D. (1997) 'Les consequences christologiques de la confrontation islamo-
chreticnne en Espagne au VIle siecle' Das Frankfurter Koneil von 794 Mainz trans.
Garden KJ. in the Formation ofal-Andalus 2 pp. 37-50.
Urvoy, M.-Th. (1994) 'Influence islamique sur Ie vocabulaire d'un psautier arabe
d'Al-Andalus' Al-Qgntara 15 pp. 509-18.
Vallve Bermejo,]. (1966) ' Una fuente importante de la historia de Al-Andalus, La
«Historia» de Ibn 'Askar, Al-Andalus 31 pp. 237-65.
- - (1967) 'Fuentes latinas de los gcografos arabcs ' Al-Andalus 32 pp. 241-60.
- - (1967) 'Sobre algunos problemas de la invasion musulmana' Anuario 4
pp.361 -7.
- - (1978) 'Espana en el siglo VIII: ejercito y sociedad ' Al-Andalus 43 pp. 51-112.
- - (I989) Nuevas ideas sobre la conquista drabe de Espana: Toponomia y onomastica
Madrid: Real Academia de la Historia.
Van Koningsveld, P. Sj. (1977) the LatinArabic Glossary qf the Leiden University Library
Leiden : Brill.
- - (1991) 'La literatura cristiano-arabe de Ie Espana medieval y el significado de
la transmision textual en arabe de la Collectio Conciliorum' in Concilo III de
Toledo, XIV Centenano, 589-1989 Toledo pp. 695-710.
- - (1992) 'Andalusian-Arabic Manuscripts from Christian Spain : a Comparative
Intercultural Approach' Israel Oriental Studies 12 pp. 75-110.
- - (1994) 'Christian-Arabic manuscripts from the Iberian Peninsula and North
Africa: A historical interpretation' Al-Qgntara 15 pp. 423-49.

223
 BlBUOGRAPHY

Vernet Gines,]. (1965) 'La ciencia en el Islam y en occidente' L'Occidente e l'Islam

nell'Alto Medioevo, Settimane pp. 535-72.
- - (1978) La Cultura Hispanoarabe en Orient y Occident Barcelona: Ariel.
- - (1986) La ciencia en Al-Andalus Seville: Editoriales Andalu zas Unidas .
Viguera Molins , MJ. (198 1) Aragon Musulmana Zaragoza: Mira.
Vita ]ohannis Gorrensis cd. Pertz MGH SS IV pp. 335-77.
Vivancos Gomez, M.C. (1988) Documentacuin del monasterio de Santo Domingo de Silos
(954-1254) Burgos: Ediciones].M. Garrido Garrido.
Vives Gatell,]. and Claveras,]. eds (1946) Oracional Vtsigotico Barcelona.
- - Marin Martinez, T. and Martinez Diaz , G. (1963) Concilios visigoticos e hispano-
romanos Madrid Barcelona: CSIC.
Vogel, C . revised and trans . Storey, W.G. and Krogh Rasmussen, N. (1986) The
Medieval Liturgy: An Introduction to the Sources Washington, D.C .: Pastoral Press.
Wachtel, N. (1971) La vision des vaincus: les Indiens du Pirou devant la Conquete espagnole
1530-1570 Paris: Gallimard.
Wallace-Hadrill, D. (1982) Christian Antioch. A Study if Early Christian Thought in the
Eastern Church Cambridge: Cambridge University Press.
Waltz,]. (1970) 'T he significance of the voluntary martyrs of ninth-century
Cordoba' Muslim World 60 Hartford, Connecticut: Hartford Seminary
Foundation pp. 143-59 and 226-36.
Wasserstcin, DJ. (1985) The Rise and Fall if the Parry Kings Princeton: Princeton
University Press.
- - (1990- 1) 'The library of Al-Hakam II al-Mustansir and the culture of
Islamic Spain' Manuscripts if the Middle East 5 Leiden: Ter Lugt Press
pp .99 -105.
- - (1991) 'T he Language Situation in al-Andalus' in Jones, A. and Hitchcock, R.
eds Studies on the Muioashshali and the Khatja Oxford: Ithaca Press pp. 1-15
reprinted in The Formation if al-Andalus 2 pp. 3-18.
- - (1993) The Caliphate in the West Oxford: Clarendon Press.
Werckmeister, O.K. (1993) 'Art of the frontier: mozarabic monasticism' in The art if
medieval Spain A.D.500-1200 (1993) New York: Metropolitan Museum of Art,
pp. 121-66.
- - (1997) 'The Islamic rider in the Beatus of Girona' Gesta 36.2 pp. 101-6.
William of Malmesbury cd. and trans . Mynors, R.A.B, Thomson, R.M. and
Winterbottom, M . (1998) 2 vols Oxford: Clarendon.
Williams,]. (1993) 'Biblia Hispalense' in The art if medievalSpain A.D.500 -1200 New
York: Metropolitan Museum of Art pp. 162-3.
- - (1994) The Illustrated Beatus: A Corpus if Illustrations if the Commentary on the
Apocalypse 3 vols London: Harvey Miller Publishers.
Wilson, K. (1984) ' Hrotswith of Gandersheim' in ibid. cd. Medieval Women Writers
Manchester: Manchester University Press pp. 30-63.
- - (1998) Hrotswit if Gandersheim: aJlorilegium if her works Cambridge: Cambridge
University Press.
Wolf, K.B. (1988) Christian Martyrs in Muslim Spain Cambridge: Cambridge
University Press.
- - (1990) Conquerors and Chroniclers if EarlY Medieval Spain Liverpool : Liverpool
University Press.
- - (1990b) 'The earliest Latin lives of Muhammad' in Conversion and Continuity
pp.89-102.

224
 BIBLIOGRAPHY

- - (1996) 'Christian views of Islam in early medieval Spain ' in Medieval Christian
Perceptions pp . 85-108.
Wood, LN. (1984) 'The End of Roman Britain: Continent al Evidence and Parallels'
in Lapidge , M . and Dumville , D. eds Gildas: New Approaches Woodbridge:
Boydell Press pp . 1-25.
- - (1992) 'Continuity or calamity?: the constraints of literary model s' in
Drinkwater, J. and Elton , H . eds Fifth-century Caul: a crisis ofidentity Cambridge:
Cambridge Univ ersity Press.
Woolf, P. (1965) 'L'Aquitaine et ses margcs' in Braunfel s W. ed. Karl der Crosse,
Lebenserk undNachleben Dusseldorf: Schwarm I pp . 269-309.
Wreg lesworth , J. The Chronicle of Alfonso III and its sign ificance for the
historiography of the Asturian kingdom 718-910 AD Unpublished PhD :
University of Leeds 1995.
Wright, R . (1994) 'La muerte delladino escrito en A1-Andalus' Euphrosyne. Revista de
Filologia Clasica Nova Serie 22 (Home naj e a Manuel C . Dia z y Diaz) Lisbon :
University of Lisbon tran s. Wright in the Formation ofal-Andalus pp. 219-36.
Young, MJ.L. , Lat ham , J.D. and Sergeant, R .B. eds (I990) TIle Cambridge History of
Arabic Literature I Religion, learning and science in the 'Abbasid period Cambridge New
York: Cambridge University Press.

225
 -~-
Index
'Abd al-Aziz, son of Musa ibn Nusayr,
175, 180
'Abd ai-Malik ibn J:Iabib see Ibn J:Iabib
'Abd al-Rahman [ibn Mu'awiya] I,
Umayyad emir (756- 88)
deal ings with house of Witiza 158,
176, 179
'Abd al-Rahrnan III, Umayyad emir
and caliph (9 12-6 1)
campaign against north (920) 90
embassy to Otto I 108-9
proclaims caliphate 14
Abu Bakr Muhammad see Ibn al-
Qil!iya
Adoptionism 24- 7
Ahmad al-Raz! (d.955), historian 168
Chronicle ofthe Moor Rasis 143, 168
Akhbiir Majmu'n 18, 164
akhbdr versus tar'ikh 165-6
Alcuin and Adoptionism 26
Alvarus 9
Life qf Eulogius 55-61
excom munication of 79
polemic against Islam 65
Ambrosio de Morales 54, 57, 63, 81
Anthony of Damascus, eastern martyr
66
Arabic historiography, problems of
5-6
Arabic manuscripts, dating of 148-9
Arabic translation of Orosius (chapter 7)
135-57
Arabic translations
of canons of church councils 156
of Gospels 154
of Psalms 155
226
Argentea, martyr 10
Passion qf Argentea 82-3,85, 101-6
'Arib ibn Sa'Id, author of Calendar of
Cordoba 124, 126, 132-3
Artubas, son of Witiza , 176-7
astrology, suspicion of 130- I
Asturias , kingdom of the
art 91-2
churches 91-2
monasteries, foundation and
refoundation of 93
propaganda 1- 2, 185
Aurea, martyr of Cordoba, 78, 105
Banil Hajjaj , decendants of Sara the
Goth 170, 179, 182
Banil Qasi 19-20, 75, 176
Beatus of Liebana 9, 25
Commentary on the Apocalypse 25,50,92
Against Elipandus 25
Berbers in conquest and settlement of
al-Andalus 13, 178
bilingualism 156
biographical dictionaries as historical
evidenc e 3, 161-2, 182-3
blasphemy against Islam, meaning of
53,66
Bobastro , stronghold of Ibn Hafsun
102-3
Boniface, letter to Aethelred of Mercia
26-7
Book qf Daniel 47, 65, 91
Books ofAnwii' 117-121
Books on Time 118-9
Bronze Age, chronology of 122, 150
 INDEX
Calendar of Cordoba 16, 78,94, 105, 108,
116-28
Arabic version I 21-4
astronomical data in 119-20
date of II 7, 120
Latin version 125-7
Calendar rif Vich I 27-8
Cardefia see San Pedro de Cardcfia
Chalmeta Gendron, Pedro on Ibn al-
Quttya 165, 173
christian-muslim frontier 4, 19
Chronicle rif 7418,29-33,35-8,40-7
manuscripts of 30, 32
Chronicle of 754 8, 28-35, 40-7
manuscripts of 31-2
relationship to Chronicle rif 741 41-2
Chronicle rifAlbelda I
Chronicle ofAlfonso III 31
Chronicle rifPseudo-Isidore 143, 156-7
Chronicle ofSampiro 97
Chronicle riftheMoor Rasis see Ahmad
al-Razt
chronicles as evidence 29, 32
Cixila, bishop of Toledo (c.744- 54) 21,
23, 90
clients, status of 171-2
Codex Emilianensis 21, 27, III
Collins, Roger
on value of Arabic histories 6, 176
on eighth-century chronicles 8-9,
41,44-5
on History rif the Goths 37
Constantine, Byzantine emperor, and
pope Sylvester, legends of 151-3
Constantine VII Porphyrogenitus,
Byzantine emperor (944- 59),
embassies to 'Abd al-Rahman III
138
Constantinople, siege of (717) 43, 46,
49 -50
Continuatio Byzantia-Arabica see Chronicle cf
741
conversion to Islam , meaning of 2-3
Bulliet, Richard, on 2-3
of Abu 'All al-Hasan al-Masarjisi of
Nishapur (d .853/ 4) 180
convivencia 10
Cordoba 14-18
Arabic writers on 14-17
227
churchesof67,104- 5
council of (839) 27
mosque of 14- 15
size of 15
Cordoba and Toledo (chapter 2) 14-2 7
Cosmas and Damian, monastery of
(Abellar, near Leon) 90, 93
Covadonga, battle of see Pelayo
Dhabha, blasph em es against Islam
80-1
Dioscorides, Materia Medica 137-8
Dozy, Rheinhardt 2, 8
and Passion ofArgentea 101- 2
edition of Calendar rifCordoba II 7
on author of Calendar rif Cordoba 126,
132
on Recemund and Rabi" ibn Zald
128, 133
Duero, depopulation and repopulation
of 93-4
Elipandus, bishop of Toledo (754- after
800) 9, 24-7
Elvira, daughter of Ramiro II of Leon ,
abbess of San Salvador de Leon
97, 106
emigration, christian 2
to the Asturias 91, 93-4
to Francia 92
Ermogius, bishop in Passion rif Pelagius
88 ,90,99
Ermogius, 'confessor' of Celanova 99
ethnicity and religion, problems of
171-2,182
Eugenia, supposed tenth-century
martyr of Cordoba 81
Eugenius II, archbishop of Toledo
(646- 57) 22, 24
Eulogius 9, 52-62, 68-70, 73, 75-9
Apologetic Book cfthe Martyrs 54, 64
elected bishop of Toledo 27, 59
journey to Navarre 55-9, 75
letter to Wiliesindu s 55, 57, 75
Lift rifEulogius 55- 62
martyrdom of 60 -1
Martyr Document 56, 76-7
Memorial ofthe Saints 53, 68, 73, 76-9
Evantiu s, deacon of Toledo 22-3
 INDEX
Felix of U rgel 26
Flora and M aria, martyrs of Cordoba
52-3, 56, 60, 76-7
Fortun abbot of San Salvador de Leir e
57,70- 1, 74
Fraxinetum, pirates of 109, 114
genealogy
Anglo-Saxon 170
as evidenc e 169- 70
in al-Andalus 168-71
George ofSt-Sabas, martyr of Cordoba
53
tr anslation to Paris (with
Aurelius and Nathalia/Sabigotho)
80 see also Usua rd of
St-Germain
glosses
Arabi c, in Visigothi c manuscripts II ,
154
Lat in, in Arab ic manuscrip ts 156
in ' UnlsflIs 153
Godmar /Gondemar II , bishop of
Gerona (943-95 1/ 2), History qf the
Franks 141
Gomez ibn Anton ian , secretary to em ir
Muhammad, 178
Gorze, monastery of 109, 112-13
Hadrian I, pope, letter s to Spanish
bishop s 25- 6
H afs ibn Albar al-Q utl, 179
proposed as tran slator of 'Urasus
140-1
translator of Psalm s 155-6
al-Hakam I, U mayyad emir (796- 822),
Ibn al-Qjltlya on 172
al-Hakam II, U mayya d em ir (96 1- 76)
108, 137, 169
library of 131, 133, 137, 142
H asdeu ibn Shaprut I 10
H ashim ibn 'Abd al-Aziz , ca mpaigns
again st Toledo (882) 20
H eraclius, Byzantine emperor
in Chronicle qf74/ 42-4, 51
in Chronicle qf 754 34, 42-4
prophecies concerning 49
H isham I, Umayyad emir (788- 96),
Ibn al-Q u nya on 172-3
228
'H ispania' , as alternative to 'Spa in' or
'Iberia' 8
Hispanic Passionary 82- 3, 85-7
Histories see Orosius Seven Books of
Histories Against the Pagans
history, providenti al 30, 35, 38, 50-I ,
135, 185-6
Hrotswit of Gandersheim 10, 96- 7
al-Hur r, governor of al-Andalu s
(716- 18) 28,1 77-8
Hydatiu s, Chronicle 32, 50
Ibn Abi Usaybi'a , author of
biographical dicti onary 131,
136- 7
Ibn Abi Amir see al-Mansur
Ibn al-Abbar (d. 1260), autho r of
biogra phical dictionary 162, 169
Ibn 'Asim (d. 1013), Book qf Anwa' and
Time 120-1
Ibn al-Fara di (d. 1013), biographical
dictionary 137, 160-1
Ibn l:Iabib (d.853)
as a historian 167
Book on the Stars I I 9
works on genealogy 170
History qf al-Andalus 145, 165
Ib n H afsun 102-3
Ibn H awq al 14, 16-17
Ibn H ayyan on M adinat al-Za hra'
129- 30
Ibn H azrn (d. 1064)
false genealogy of I 71
History ofReligion 154- 5
Jamhara 170
on history 144- 5
Risala 130
Ibn Idh ari on mosque of Cordoba 15
Ibn juljul [Abu D aud Sulayman ibn
H assan] on D ioscurides and
Histories 137-8, 141 Dictionary
139
Ibn Kh aldun 136, 145- 7
cites Coptic and j ewish historian s
146- 7
cites Orosius 146-7
on history of Hispan ia 146
on translators of ' Unlsfiis 139, 141
view of history 145-6
 INDEX
Ibn Khallikan, author of biographical
dictionary
on Ibn al-Quttya 162-3
on Sara the Goth 181-2
Ibn Qutayba of Baghdad (d.889) Book if'
Anwa' II 7-8, 131
Ibn al-Qutlya 12, 159-6, 164-8
as historian 160, 163-4
History of the Conquest if' al-Andalus
159, 164-8
Ibn SaId see 'Anb ibn SaId
Ibn SaId the Maghrebi (1209 or
1214- ), historian 130
Ibn Wafid , author of an agricultural
calendar 127
Ibn Zard the bishop 108, 130-2
Ildefonsus, archbishop of Toledo
(657- 67) 21-4, 26
Lift of Ildefonsus 21- 3
Inigo Arista of Navarre 71-2, 74-5
Isaac , martyr of Cordoba 52
Isidore, archbishop of Seville
Chronicle 34
History ofthe Goths, Sueves and Vandals
30-1,36-8
in Chronicle if' 754 34
Jimenez de Rada, bishop of Toledo
(d. 1247) Arab History 6
John bar Penkaye, World Chronicle, on
islamic conquest 47
John of Biclar, Chronicle 30-1, 35- 6
John of Damascus (c.652- c.750) On
Heresies 64
John of Gorze II, 108-113
Lift if'John if' Gore» 109-112
John of Nikiu , Chronicle 39
John of Seville, letter to AIvarus 63 -4,
155
al-Khushani (d.9 71) Book of the fudges if'
Cordoba 81, 164
Last Emperor, legends of 47- 5 I
Latin, loss of, in al-Andalus 10, 90
Latin-Arabic Glossary II , 148
Leire see San Salvador de Leire
Leo III, Byzantine emperor (7 17- 4 1)
43, 49
229
Leon , charters from 91
libraries of al-Andalus 136-7
of Dar al-Hikma, Cairo 137
Liudprand of Cremona II, I 13
Antapodosis, 108, 113-6
embassi es to Constantinople I 13,
128
Madinat al-Zahra' 15-17, 128-30
al-Mansur, wazlr of al-Andalus
(961-1002) 97-8, 106, 131, 142
al-Maqqarr, historian 6, 15, 130, 143,
164
nostalgia for Cordoba 15
Martyrology ofLyon, Hispanic saints in 86
martyrs of Eulogius (chapter 4) 52-79
Maslama ibn 'Abd ai-Malik , Umayyad
commander 43, 46, 49-50
al-Mas'udi (d.956), historian 141-2
mauidli see clients
Merida 17
Migetius 9, 24-5
Morales see Ambrosio de Morales
Mozarabic Chronicle see Chronicle if' 754
'Mozarab' , meaning of 8
Muhammad
in eighth-century chronicles 43
Lift ofMuhammad from Leire 62-4,
67-8
Muhammad, Umayyad emir (852- 86)
67, 178
Musa ibn Nusayr 18, 28, 164, 166,
174-6
Musa ibn Musa of Banu QaSI 19, 75,
176
al-Mutawakkil, Abbasid caliph (847-
61), decrees against christians
66-7
names, problems of2, 91-2
news from the east in the eighth-
century chronicles (chapter 3)
28 -51
Nikephorus, Short History 40
Nu'aym ibn Hammiid (d.843) Book if'
civil wars, on Last Emperor 48-9
Nunilo and AIodia, martyrs of Huesca
62,68-78
date of martyrdom 73-6
 INDEX
date of translation 74-6
dedications to 72
Passion qf Nunilo and Alodia written by
Eulogius 68 -9
Passion qf Nunilo and Alodia from
Cardefia 68-9, 84
Translation ofNunilo andAlodia 70
Oppa, bishop of Toledo, brother, son or
grandson of Witiza 100- I , 181
Orosius' Seven Books ofHistory against the
Pagans II, 135-40, 145-7, 157
quoted by Arabic geographers
142-4
Otto I (936- 73) 108-10
Pelagius, bishop oflria Flavia (-985)
100
Pelagius , tenth-century martyr of
Cordoba 10
cult of 94, 98 -100
dedications to 98-9
homosexuality and 95 -6
mass in honour of 99
Passion qf Pelagius 82-5, 88-91 ,
94-6,98-100
translation to Leon 89
Pelayo, bishop of Oviedo (110 1-30) 97
Pelayo, hero of Covadonga I, 18,
100-1
Perfectus , martyr of Cordoba 52, 54, 64
Peter, deacon of Toledo, author of
treatise on Easter and letter to
Felix of Cordoba 23
Peter of Capitolias, eastern martyr 66
Phocas, usurping Byzantine emperor
(602-10) 41 -2, 49
polemic against Islam 5, 47, 64-5, 69,
95, 153
Prophetic Chronicle 50, 63, 186
Pseudo-Methodius, prophecies of
47 -8
Qasim ibn Asbagh al-Bayyant (d.95 1)
6, 119, 139-41, 160
Brilliant OJtalities qf the Umayyads 6
Book qf Genealogies 170
Qayrawan manuscript, christian
history in Arabic 152-3
230
Rabt' the bishop 128-30
Rabt' ibn ZaId, supposed author of
Latin version of Calendar cfCordoba
126, 131-3
RabI ibn Qays ibn Yazld al-Ghassini,
eastern martyr 66
Raguel, author of Passion qf Pelagius
89-91
Ramiro III of Leon (96 7- 84) 97-8
Recemund and the Calendar qf Cordoba
(chapter 6) 108-34
Recemund 10, 108-9, 111-16, 128-9,
131, 133
identified as RabI' ibn ZaId 128,
131, 133
supposed visit to Constantinople 128
Rodrigo, Visigothic king (7 10- 11) 7,
34,174
Sa'Id the Andalusi (1029-70) on
science in al-Andalus 131, 144
Sa'Id ibn Bitrtq, patriarch of
Alexandria (933- 40), author of
christian history in Arabic 153
al-Samh, governor of al-Andalus
(718- 21) 28, 45-6
San Juan de la Pefia 58, 72
San Miguel de la Escalada 91, 93
San Pedro de Cardefia 82
Passionaries ofCardeiia 83-5, 87
San Pedro de Valeranica 98
San Salvador de Leire 57-8, 70-2
Sanchez-Albomoz, Claudio
on christian-muslim frontier 4, 93
and Arabic sources 5
Sancho Garces I of Navarre (905- 25),
usurper 72
Sancho the Fat of Leon (955- 7 and
960 -7) 97- 8, 106
Santa Maria de Melque 23
Santo Domingo de Silos 82, 87
Passionary qf Silos 84-5
Sara the Goth and her descendants
(chapter 8) 158-83
Sara the Goth 158, 179-83
Secunda 17, 174
shu'ubrya 182
Silos see Santo Domingo de Silos
Simonet, Francisco 4, 8, 102, 128, 184
 INDEX
on Recemund and Rabt' ibn Zatd
128, 130, 133
on translator of Orosius 140
Sindered, bishop of Toledo ( c. 7 0 4 ~ 2 1 )
21
Sisibert, bishop of Toledo (690~3 ) 22~3
Sophronius, bishop ofJerusalem, on
islamic conquest 40
'Spanish Islam ', concept of 3
Speraindeo, abbot of Cordoba 56, 64,
78
Tabanos, monastery of 52
al-Tabart, historian 6
Tamtm ibn Alqama, poet ( 8 0 3 ~ 8 6)
tar'fkh see akhbiir
Tariq ibn Ziyad 18, 28, 166, 174~6
Tarragona, Oracional of Tarragona 87
Theodimir see Tudmir
Theodulf, bishop of Orleans, emigrant
to Francia 92
Theophanes, Chronicle 39, 49
Timothy I, patriarch of Baghdad
(775- 85), author of anti-muslim
polemic 47, 65, 153
Toledo 18-27
bishops and archbishops of 21, 27
buildings of 20
hymnary 23, 83 , 87
reconquest of 20-1
Tours/Poitiers. battle of 28 ~9
translation of Gre ek to Arabic ,
problems of 138-9
treaties between conquerors and
christians 175-6
Tudmir, treaty of 175
Tultusceptru de libro domni Metobii 63
Tuy 93, 99
231
Two mor e martyrs of Cordoba (chapter
5) 80-107
'Unlsfiis. (Arabic translation of Orosius '
Seven Books ofHistory Against the
Pagans) 137 ~40 , 146-7 , 153-4,
157
Columbia manuscript 147-9
date of manuscript 148
Usuard, monk of St-Germain 53-5
Valdeavellano, monastery of 83
Passionary cf Valdeavellano 83, 85
Valeranica see San Pedro de
Van Koningsv eld
on authorship of Calendar qf Cordoba
132
on christian Arabic manuscripts
154-6
Viliulfo, bishop of Tuy (c.1000) 99
virginity 88, 95-6, 103, 107
as rust or mildew 65
Vuilfurus , scribe of Silos 84 , 99
Vulfura , companion of Argent ea 101,
105-6
al-WalId I, Umayyad caliph (705- 15)
66, 158, 166, 174-5
Wiliesindu s, bishop of Pamplona 71, 74
letter to see Eulogius
Wistremirus, bishop of Toledo (-c.858)
27, 75
Witiza , Visigothic king (698- 710)
158-9
sons of, in Chronicle of 754 and
Chronicle qf Alfonso III 159
sons of, in Ibn al-Qjittya 174-5
wom en, in found ation myths 180