AHisWf

AHilMf
BYW. MONTGOMERY WATT AND
PIERRE CACHIA

S3 Routledge
jjj^^ Taylor & Francis Group

LONDON AND NEW YORK

First published 1965 by Transaction Publishers

Published 2017 by Routledge

2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
711 Third Avenue, New York, NY 10017, USA

Routledge is an imprint of the Taylor & Francis Group, an informa business

Copyright © 1965 by W. M. Watt and P. Cachia.

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 retrieval system, without permission in writing from
the publishers.

Notice:
Product or corporate names may be trademarks or registered trademarks, and
are used only for identification and explanation without intent to infringe.

Library of Congress Catalog Number: 2006048049

Library of Congress Cataloging-in-Publication Data

Watt, W. Montgomery (William Montgomery)
A history of Islamic Spain / W. Montgomery Watt and Pierre Cachia.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-202-30936-1 (alk. paper)
1. Spain—History—711-1516. 2. Muslims—Spain—History.
I. Cachia, Pierre. I I . Title..

DP102.W38 2007
946'.022—dc22 2006048049

ISBN 13: 978-0-202-30936-1 (pbk)

 FOREWORD

I n 1939 the prospect o f a w a r w h i c h w o u l d involve m a n y

Asian nations made m e n i n positions of responsibility i n
B r i t a i n suddenly aware of the meagre n u m b e r of our experts
i n A s i a n languages a n d cultures. T h e Scarbrough Commission
was set u p , a n d its report l e d to a great expansion o f O r i e n t a l
a n d A f r i c a n studies i n B r i t a i n after t h e war. I n the t h i r d
decade after 1939 events are m a k i n g clear t o ever-widening
circles of readers the need for something more t h a n a super-
ficial knowledge o f non-European cultures. I n particular the
blossoming i n t o independence of numerous A f r i c a n states,
m a n y o f w h i c h are largely M u s l i m or have a M u s l i m head o f
state, emphasises the g r o w i n g p o l i t i c a l importance of t h e
Islamic w o r l d , and, as a result, the desirability of extending
a n d deepening the understanding a n d appreciation o f this
great segment o f m a n k i n d . Since history counts for m u c h
a m o n g Muslims, a n d w h a t happened i n 632 or 656 m a y still
b e a l i v e issue, a journalistic familiarity w i t h present c o n d i -
tions is n o t enough; there must also be some awareness o f
h o w the past has m o u l d e d the present.
T h i s series o f "Islamic surveys" is designed t o give the
educated reader something more t h a n can be f o u n d i n the
usual p o p u l a r books. E a c h w o r k undertakes to survey a spe-
c i a l p a r t o f the field, a n d to show the present stage of scholar-
ship here. W h e r e there is a clear picture this w i l l be g i v e n ;
b u t w h e r e there are gaps, obscurities a n d differences o f
o p i n i o n , these w i l l also be indicated. F u l l a n d annotated
bibliographies w i l l afford guidance t o those w h o w a n t t o p u r -
sue t h e i r studies further. There w i l l also be some account of
the nature a n d extent o f the source material.
W h i l e the series is addressed i n the first place to the edu-
cated reader, w i t h l i t t l e or no previous knowledge of the
subject, its character is such t h a t i t should be of value also
to university students a n d others whose interest is o f a m o r e
professional k i n d .
vi FOREWORD

T h e transliteration of A r a b i c w o r d s is essentially t h a t o f
t h e second e d i t i o n o f The Encyclopaedia of Islam ( L o n d o n ,
i 9 6 0 , c o n t i n u i n g ) w i t h three modifications. T w o o f these are
n o r m a l w i t h most B r i t i s h Arabists, namely, q for k, a n d /
for dj. T h e t h i r d is something o f a novelty. I t is the replace-
m e n t o f t h e ligature used t o show w h e n t w o consonants are
to b e sounded together b y an apostrophe to show w h e n they
are t o b e sounded separately. T h i s means t h a t dh, gh kh, sh 9 9

ih ( a n d i n non-Arabic words ch a n d zh) are to be sounded

together; w h e r e there is an apostrophe, as i n ad'ham, t h e y
are t o be sounded separately. T h e apostrophe i n this usage
represents no sound, b u t , since i t o n l y occurs between t w o
consonants ( o f w h i c h the second is ft), i t cannot be confused
w i t h the apostrophe representing t h e glottal stop (hamza),
w h i c h never occurs between t w o consonants.

W . Montgomery Watt

GENERAL, EDITOR
 CONTENTS

Introduction:
The Interest of Islamic Spain xiii

Chapter i T h e M u s l i m Conquest
1. The conquest as a phase of Arab expansion 1
2. The weakness of Visigothic Spain 5
3. The course of the invasion, 711-716 8

2 T h e Province of the Damascus Caliphate

1. H i e organisation of the province 12
2. The end of the advance 15
3. Internal tensions of the province 20

3 T h e Independent U m a y y a d E m i r a t e
1. The establishment of the emirate 24
2. The crisis of the emirate 28

4 T h e Grandeur o f the U m a y y a d Caliphate

1. Umayyad Spain at its zenith 32
2. The economic basis 40
3. Social and religious movements 43
4. The ruling institution 48

5 C u l t u r a l Achievements under the

Umayyads
1. Intellectual life—the religious sciences 52
2. Intellectual life—poetry and belles-lettres 58
3. A r t 65
4. The source of Moorish culture 67

6 T h e Collapse o f A r a b Rule
1. The 'Amirid dictatorship and the
breakdown 70
2. The reasons for the breakdown 74
3. The "party kings" (1009-91) 78
viii CONTENTS

Chapter 7 T h e Berber Empires, T h e Almoravids

1. The foundations of the Almoravid state 82

2. The Almoravids i n Spain 84

8 T h e Berber Empires, T h e Almohads

1. Ibn-Tumart and the Almohad movement 89

2. Spain under the Almohads (to 1223) 91
3. The progress of the Reconquista from 1223
to 1248 95

9 C u l t u r a l Greatness i n Political Decline

1. Poetry 97
2. Prose literature and philology 105
3. The religious sciences and history 110
4. Philosophy and mysticism 118
5. The art of the eleventh and twelfth
centuries 125

10 T h e Last o f Islamic Spain

1. The Nasrids of Granada 127

2. The Muslims under Christian rule 130
3. Literature i n a period of retreat 134
4. The art of the thirteenth and fourteenth
centuries 140

11 T h e Significance of Islamic Spain

1. The Arab and Islamic colony 143

2. The stimulation of Christian Spain and
Europe 147

3. The intrinsic greatness of Islamic Spain 150

Notes 153

Bibliography 162

Index 168

[NOTE: Chapters 5, 2; 9, 1; 9, 2 and 10, 3 are by D r Cachial

 LIST OF ILLUSTRATIONS

Plates following page 64

1 D o o r a n d b l i n d arches i n the western facade o f the
mosque at Cordova
2 R e d a n d w h i t e striped arches o f the mosque at Cordova
3 Stone lattice a n d foiled arch i n the mosque at Cordova
4 Arches i n one o f the m a i n halls, possibly the throne
r o o m , at the Palace o f M a d i n a t az-Zahra*
5 D e t a i l o f i v o r y casket f r o m the t i m e o f the caliphate at
Cordova ( V . & A . )
6 O i l l a m p i n w r o u g h t bronze ( A l h a m b r a M u s e u m ,
Granada)
7 D e t a i l of delicate i v o r y w o r k o n a casket fashioned i n
Cordova ( V . & A . )
8 T o r r e d e l O r o , Seville
9 T h e Giralda, Seville

following page 104

10 A r c h e d entrance to the Sala de los Embajadores i n the
Alcazar, Seville
11 Some of the roofs o f the Royal Palace of the A l h a m b r a ,
Granada
12 Slender columns i n the Patio de los Leones at the A l -
hambra
13 D e t a i l o f stucco o n the walls of the Patio de los Leones
14 T i l e mosaic a n d stucco w a l l decoration i n the Salon de
los Embajadores, A l h a m b r a
15 Stucco w a l l decoration i n the Patio de los Leones, A l -
hambra
16 Pool o f t h e Partal a n d the T o r r e de las Damas, A l h a m b r a
17 T h e D o m e o f a chapel at the Mosque, Cordova
Maps
1 S h o w i n g v a r y i n g extent of Islamic influence A . D .
700-1400 17
2t M a i n trade routes and products i n Spain i n the
M i d d l e Ages 39
 ACKNOWLEDGMENTS

Plates i , 2, 3, 4, 6, 8, g, i o n , 12, 13, 14, 15, 16 and 17 are

?

reproduced from Moorish Spain b y E n r i q u e Sordo b y k i n d

permission of the publishers, E l e k Books, L t d . Photographer:
W i m Swaan. Plates 5 and 7 are reproduced b y k i n d permis-
sion of the Victoria and A l b e r t M u s e u m .
 INTRODUCTION

T H E INTEREST OF I S L A M I C SPAIN

F o r m a n y centuries M o o r i s h Spain has stirred the imagina-

t i o n o f Europe. Ballads gave glamour t o the courageous stand
of R o l a n d i n the pass of Roneesvalles, w h i l e legends sur-
r o u n d e d the figure of the C i d a n d m a d e h i m a great hero.
N o r was i t merely the struggle against the Moors w h i c h
caught the imagination. T h e better-informed inhabitants o f
the rude Christian kingdoms a n d dukedoms of western E u -
rope realised t h a t south of the Pyrenees was a l a n d o f higher
culture, where a m i d material l u x u r y m e n enjoyed the delights
of music a n d poetry; and they gradually took over w h a t they
could of this culture. W i t h the romantic movement something
of the o l d a d m i r a t i o n was revived, a n d i t is doubtless t h r o u g h
the influence o f W a s h i n g t o n I r v i n g t h a t " A l h a m b r a " has be-
come a familiar w o r d to m a n y w h o k n o w n o t h i n g o f the
fourteenth-century palace.
E v e n for the prosaic scientific historian, w h o speaks of
Islamic Spain—strictly, " M u s l i m " applies o n l y to persons—
the subject is n o t w i t h o u t its fascination. H e r e an oriental
culture has entered E ur ope a n d left b e h i n d magnificent
architectural remains. I t offers an i m p o r t a n t example o f the
close contact of diverse cultures, a n d one that has c o n t r i b u t e d
to m a k i n g the European and A m e r i c a n historian w h a t he
is. T h e chief monuments of this culture are relatively easy
to visit, a n d i n most seasons the visit is delightful. Moreover
the study of Islamic Spain gives answers t o questions regard-
i n g the general nature o f historical processes. Such questions
w i l l guide the treatment of the subject i n the present w o r k ,
a n d m a y be briefly indicated here u n d e r three heads.
F i r s t l y , Islamic Spain must be looked at i n itself. I t is com-
m o n l y regarded as h a v i n g great a n d magnificent achievements
to its credit. B u t i n w h a t d i d its greatness consist? T h e loveli-
ness o f the buildings i t has left? W o r k s of pure literature
t h a t are major contributions to t h e world's store? Philosophi-
cal, scientific, or religious w r i t i n g s w i t h a secure place among
xiv INTRODUCTION

the classics of the "one w o r l d " into w h i c h w e are moving?

O r is this image of Islamic Spain largely dependent on t h e
contrast between its l u x u r y a n d the bareness o f contemporary
life i n t h e rest o f western Europe, a n d on the fact that i t was
the channel t h r o u g h w h i c h elements of higher culture, b o t h
material a n d intellectual, entered Europe?
Secondly, Islamic Spain m u s t be looked o n as a part of the
Islamic w o r l d . I t shared i n the culture o f a vast area, a n d the
character of its links w i t h the heartlands must be looked at.
H o w d i d i t keep i n touch? Was i t m a i n l y a passive recipient
or d i d i t make any distinctive c o n t r i b u t i o n to Islamic culture
as a whole? M a y i t be regarded as an active cell i n the b o d y -
social of Islam? O n the other hand, h o w far h a d i t become
adapted t o the special circumstances of the I b e r i a n peninsula,
such as climate, geography a n d the m i n g l i n g o f religions?
D i d i t manage to integrate the various racial a n d social groups
i n t o a u n i t y , and t o impregnate the w h o l e society w i t h its
values? A subordinate question is t h a t of the relation of Spain
t o N o r t h Africa, especially the p a r t w h i c h is n o w M o r o c c o
a n d Algeria. H o w far were the t w o regions a single cultural
area d o m i n a t e d b y Spain?
F i n a l l y , Islamic Spain was i n close contact w i t h its E u r o -
pean neighbours. W h a t exactly d i d i t contribute to Europe?
I n h o w m a n y spheres can w e trace its influence, a n d see
w h e r e Europeans have learned f r o m Spanish Muslims? A g a i n ,
E u r o p e has clearly been influenced b y reacting against
Islamic Spain. T h e crusade is i n p a r t a retort to the jihad or
h o l y w a r o f the M o o r s ; and the Reconquista was a major
formative element i n the m a k i n g o f m o d e r n Spain. T h e an-
swer t o these last questions belongs t o the history of E u r o p e
a n d o f Christian Spain, b u t some i n d i c a t i o n w i l l be given of
the lines along w h i c h the answers must go.
A HISTORY OF I S L A M I C SPAIN
 1

T H E MUSLIM CONQUEST

1. The Conquest as a Phase of Arab Expansion

T o the inhabitants of Spain the A r a b conquest i n the years

f r o m 7 1 1 t o 716 came as a b o l t f r o m the blue. For the Arabs
themselves, however, the o v e r n i n n i n g o f Spain was merely
one phase i n a l o n g process of expansion. I t was an emi-
1

n e n t l y profitable a n d successful phase, a n d the success came

v e r y r a p i d l y ; b u t i n the process o f expansion w h i c h h a d
b e g u n at least as early as 630 there h a d been comparable
phases. D u r i n g the reign of the c a l i p h ' U m a r ( 6 3 4 - 4 4 ) the
embryonic A r a b state—at this t i m e a n alliance of many, b u t
n o t y e t a l l , o f the tribes o f the A r a b i a n peninsula—had de-
feated the Byzantine empire a n d wrested from i t the p r o v -
inces o f Syria a n d E g y p t , a n d h a d dealt such a crushing
b l o w to t h e Persian empire t h a t i t ceased to exist, leaving
w h a t w e n o w call I r a q a n d Persia t o be occupied b y the
Arabs as soon as t h e y c o u l d find m e n t o h o l d t h e m securely.
A n d this was o n l y a b e g i n n i n g . F o r about a century the Arabs
continued t o move onwards a n d outwards. One line of ex-
pansion was north-east along the golden road to Samarqand
a n d b e y o n d , a n d another south-east i n t o the Indus valley,
w h i l e i n the west they progressed t h r o u g h the coast-lands of
N o r t h Africa. T h e advance was n o t gradual b u t rather b y a
series of jumps. There were periods o f quiescence a n d con-
solidation, w h e n the Arabs paused i n t h e face of some serious
obstacle or i n order to deal w i t h i n t e r n a l tensions.
T o understand h o w this amazing expansion was possible
i t is necessary t o go back t o the career o f M u h a m m a d .
M u h a m m a d was b o t h p r o p h e t a n d statesman—a combination
w h i c h m o d e r n m a n w i t h his compartmentalised religion finds
difficult to understand. As a statesman he was interested i n
A r a b u n i t y ; b u t he m a y also have felt that political u n i t y was
i m p l i c i t i n the fact that his mission as a prophet was to the
2 THE MUSLIM CONQUEST

Arabs i n general a n d not simply to the m e n of Mecca. U n i t y ,

however, was v i r t u a l l y impossible w i t h o u t expansion, because
of the nature of nomadic life. T h e m a i n economic basis of
this life was the breeding and pasturing o f animals, w i t h i r -
regular movements from those areas where pasture was abun-
dant after r a i n to those where there were permanent wells.
W h e n t h e y were i n a position to do so the nomads exacted
payments for the safe conduct of m e n and merchandise. L i f e
was never easy i n the A r a b i a n desert, however, a n d a n o r m a l
feature o f i t was the r a i d or razzia, w h i c h was usually a i m e d
at d r i v i n g off an enemy's animals b u t occasionally i n v o l v e d
loss of h u m a n life. Deaths i n the razzias and other fighting
m u s t have done something to reduce pressure o n the l i m i t e d
f o o d supplies. A t some p o i n t i n M u h a m m a d ' s career i t m u s t
have become clear t o h i m that, i f there was to be a political
u n i t y o f the Arabs, fighting a n d razzias w o u l d have to be
suppressed; b u t this w o u l d increase the pressure o n the avail-
able food. H o w c o u l d this difficulty be surmounted?
I t is i n this context t h a t the Islamic conception o f the
jihad or h o l y w a r must be considered. I t was never a p u r e l y
religious phenomenon b u t always at least i n p a r t a political
instrument. I t was i n d e e d a transformation of the nomadic
custom o f the razzia, i m m e d i a t e l y relevant to the conditions
w h e n M u h a m m a d c o n t r o l l e d only M e d i n a a n d a f e w allied
tribes. T h e o r d i n a r y t r i b e m i g h t carry o u t a razzia against
any t r i b e or f a m i l y w i t h w h i c h i t was not for the m o m e n t
o n friendly terms. T h e l i t t l e state of M e d i n a functioned i n
m a n y ways as a tribe. A m o n g the nomadic tribes of the region
i t h a d its allies a n d friends and likewise its enemies. A t least
i n the latter p a r t of his career M u h a m m a d insisted that those
w h o w a n t e d to be f u l l allies must become Muslims and accept
h i m as prophet. I n this situation the conception of the h o l y
war meant t h a t the r a i d i n g propensities of M u h a m m a d ' s fol-
lowers were directed against non-Muslims; b u t as more of
the tribes near M e d i n a became M u s l i m s i t also meant t h a t
r a i d i n g parties h a d to go farther afield. There are indications
t h a t M u h a m m a d was aware t h a t the g r o w t h o f his alliance,
b y stopping r a i d i n g between the members of the alliance,
was increasing the pressure o n f o o d supplies, a n d that he
made preparations for more extended razzias i n t o Syria, the
 A PHASE OF ARAB EXPANSION 3
nearest comparatively r i c h country. Certainly his successors,
as soon as they h a d regained control over some disaffected
tribes, directed large r a i d i n g expeditions against b o t h Syria
and Iraq.
I t is a c o m m o n misapprehension t h a t the h o l y w a r meant
t h a t the M u s l i m s gave their opponents a choice "between
Islam a n d the sword". T h i s was sometimes the case, b u t o n l y
w h e n the opponents were polytheists and idol-worshippers.
F o r Jews, Christians and other "people o f the book", t h a t is,
monotheists w i t h w r i t t e n scriptures—and the phrase was v e r y
liberally interpreted—there was a t h i r d possibility; they m i g h t
become a "protected group", p a y i n g a tax or t r i b u t e t o the
M u s l i m s b u t enjoying i n t e r n a l autonomy. A member o f such
a g r o u p was k n o w n as a dhimmi. W i t h i n A r a b i a the nomadic
tribes w e r e nearly a l l idolaters, a n d w e r e therefore forced t o
become Muslims. Outside A r a b i a , however, the local i n h a b i t -
ants were expected to become "protected groups". There was
n o pressure to become Muslims, b u t rather t o remain as they
were. M o v a b l e booty captured o n expeditions c o u l d be
d i v i d e d among the participants i n the expedition, b u t w h e n
the A r a b M u s l i m s began to conquer l a n d , t h e y h a d no desire
to d i v i d e i t u p among themselves a n d t o settle d o w n t o an
a g r i c u l t u r a l life. I t was more advantageous to allow the ex-
isting cultivators to continue t o cultivate, w h i l e the t r i b u t e
a n d rents, d i v i d e d among the Muslims, enabled t h e m t o be
a f u l l - t i m e expeditionary force.
I t was i n this w a y that the Arabs were able to expand so
r a p i d l y a n d to keep o n expanding. T h e f u l l citizens or M u s -
lims received a stipend f r o m the treasury and were able to
devote themselves almost w h o l l y to soldiering. Since then-
stipend c o u l d be increased b y a share of any booty cap-
t u r e d , t h e y were eager t o go o n expeditions w h i c h promised
t o be lucrative a n d n o t too arduous or dangerous. W h e r e the
r a i d e d populations s u b m i t t e d a n d became "protected", h o w -
ever, i t was necessary to p l a n expeditions that w e n t further
afield, a n d also to leave garrisons i n the m a i n cities o f the
territories that h a d s u b m i t t e d t o the Muslims.
T h e A r a b expansion westwards h a d b e g u n almost as soon
as t h e y h a d obtained a lodgement i n Syria. F r o m Syria an
expedition h a d gone south-west i n t o E g y p t and between 640
4 THE MUSLIM CONQUEST

a n d 6 4 2 b r o u g h t the c o u n t r y under A r a b control. Almost

i m m e d i a t e l y afterwards there h a d been exploratory expedi-
tions along the coast i n t o Cyrenaica a n d T r i p o l i t a n i a . A n
a t t e m p t e d Byzantine come-back and their o w n preoccupa-
tions elsewhere slowed the progress o f the Arabs, b u t i n 670
t h e y were able to f o u n d the c i t y of Cairouan i n Tunisia. H e r e
again the advance was halted, chiefly because o f the resist-
ance o f the Berber tribes, t h o u g h the c i t y of Carthage also
remained i n Byzantine hands. B y p l a y i n g o n the rivalries
b e t w e e n the Berber tribes, especially t h a t between the no-
mads a n d the sedentary tribes, the Arabs were eventually
able t o establish themselves securely i n Tunisia, w h i l e most
o f t h e Berbers accepted Islam. I n 698 the Byzantines were
finally d r i v e n f r o m Carthage, a n d soon after 700 expeditions
of A r a b a n d M u s l i m ( p r o b a b l y n o m a d i c ) Berbers began to
penetrate t h r o u g h A l g e r i a into Morocco a n d the A t l a n t i c sea-
b o a r d . T h e resistance of the sedentary Berbers of these re-
gions was crushed, a n d they were forced to acknowledge
A r a b suzerainty. T h e closing stages i n this advance to t h e
A t l a n t i c were the w o r k o f M u s a ibn-Nusayr, w h o is said to
have been a p p o i n t e d independent governor of I f r i q i y y a (i.e.
T u n i s i a ) i n 708, responsible d i r e c t l y to the c a l i p h i n Damas-
cus; previously the head of the administration i n Cairouan
h a d been subject to the governor of E g y p t .
After these successes i n north-west A f r i c a i t is conceivable
t h a t the Arabs m i g h t have continued southwards. I n some
directions at least there was t e r r a i n of the k i n d to w h i c h t h e y
w e r e accustomed. U n d o u b t e d l y , however, the desire for
b o o t y was an i m p o r t a n t m o t i v e w i t h the rank and file of the
M u s l i m s ; a n d i t must soon have become clear that the re-
wards o f p u s h i n g further south-west or south w o u l d be
meagre. O n the other hand, there must have been rumours
a n d half-reliable reports o f the great w e a l t h and w o n d e r f u l
treasures o f Spain; a n d i t is therefore n o t surprising that the
M u s l i m s decided to risk the entirely novel a n d distinctly
hazardous operation o f crossing the strait i n order to discover
w h a t substance there was i n the reports. T h e expansion i n t o
Spain was thus entirely i n keeping w i t h the previous ex-
tension o f A r a b p o w e r i n N o r t h Africa, and m i g h t w e l l have
come about even h a d there been no factors i n the local situ-
 THE WEAKNESS O F VISIGOTHIC SPAIN 5
ation (such as the attitude and interests of Count Julian)
to encourage t h e m a n d give t h e m an opening.
W h i l e the chief control remained i n the hands of m e n o f
A r a b race (reckoned solely o n the male s i d e ) , after the sub-
mission o f t h e Berbers o f T u n i s i a a n d eastern Algeria about
700, m u c h o f the man-power i n the expeditions was Berber.
W i t h o u t this accretion o f man-power the conquest of Spain
w o u l d have been impossible. I t is thus more correct to speak
of M u s l i m expansion t h a n o f A r a b expansion. Yet the dis-
t i n c t i o n between Arabs a n d Berbers d i d n o t disappear w h e n
the latter became M u s l i m s , a n d was to prove a grave source
of i n t e r n a l tension i n Islamic Spain.

2. The Weakness of Visigothic Spain

T h e Spain w h i c h was conquered so easily b y the M u s l i m s

suffered f r o m grave internal weaknesses. N o t merely for an
understanding of the conquest, however, b u t also for a
proper appreciation o f the w h o l e c u l t u r a l development o f
Islamic Spain, i t is necessary to look at the condition of t h e
I b e r i a n peninsula i n the early years of the e i g h t h c e n t u r y . 2

T h e Visigoths first entered Spain i n 414 a n d occupied the

north-east o f the country—the R o m a n province o f Tarracon-
ensis. Thereafter they m a i n t a i n e d their h o l d under various
p o l i t i c a l arrangements, b u t there was no real u n i t y because
the Visigoths adhered to the heretical A r i a n f o r m of Chris-
t i a n i t y , w h i l e the majority o f the indigenous p o p u l a t i o n was
Catholic. A n i m p o r t a n t change took place, however, i n 589,
w h e n the k i n g and the leading Visigoths abandoned A r i a n i s m
for Catholicism. T h i s h e l p e d i n the establishment of a stable
a n d unified k i n g d o m embracing the w h o l e I b e r i a n peninsula
a n d the province o f Septimania i n the south o f France. B y
the b e g i n n i n g of the e i g h t h century the Visigothic aristocracy
a n d the Hispano-Roman nobles seem to have been fused t o -
gether i n a single p r i v i l e g e d group w h i c h m a y be referred
t o as the "upper classes" (clasas elevadas). There were p a r t y
divisions w i t h i n the upper classes, b u t they d o n o t seem to
have followed strictly racial lines. T o the upper classes be-
longed the ecclesiastical hierarchy. T h e archbishops a n d
6 THE MUSLIM CONQUEST

bishops p l a y e d a considerable p a r t i n the governing a n d ad-

m i n i s t e r i n g o f the k i n g d o m , b u t i t was not a theocratic state,
t h o u g h t h a t has sometimes been asserted. O n the contrary,
t h e bishops were d o m i n a t e d b y the k i n g a n d his advisers,
a n d h a d largely ceased to represent the interests of the or-
d i n a r y people.
T h e monarchy itself, however, was far f r o m strong. T h e
k i n g was supposed to be elected b y the upper classes f r o m
a m o n g t h e i r o w n number. T h e r e was thus n o fixed l a w of
succession, t h o u g h some kings t r i e d to secure the succession
of a son b y associating h i m w i t h t h e m i n p o w e r before their
deaths. T h i s was often resented b y the other members o f the
upper classes. I n d e e d there were constant intrigues over the
succession. T h e king's weakness was due also to the unsatis-
factory nature o f his army. I n theory, a l l free m e n capable o f
bearing arms h a d an o b l i g a t i o n t o serve, w h e n summoned
b y the k i n g . T h i s was n o t the t y p e o f o b l i g a t i o n a n d vassalage
f o u n d under the feudal system w h i c h arose i n western E u -
rope i n later centuries, b u t one w h e r e each m a n stood i n
direct relation t o t h e k i n g a n d o w e d h i m loyalty. T o w a r d s
the e n d o f t h e seventh century the kings were apparently
finding great difficulty i n collecting a n adequate army. These
unsatisfactory features of the monarchy appear to go back t o
t h e Germanic conception o f the " t r i b e " or p o l i t i c a l u n i t ,
w h i c h the Visigoths i n Spain were t r y i n g t o a p p l y to c i r c u m -
stances for w h i c h i t was not suited.
Besides the upper classes the p o p u l a t i o n comprised the
Hispano-Romans o f free status a n d also a large p r o p o r t i o n
of serfs as peasants, successors o f the R o m a n coloni. T h e
latter h a d a v e r y difficult life, b u t even the free m e n felt
themselves t o be u n d e r p r i v i l e g e d . There was therefore m u c h
discontent, a n d m a n y o f the o r d i n a r y people looked o n the
M u s l i m s as liberators a n d gave t h e m a l l the assistance t h e y
could. T h e cities i n particular h a d fared m u c h worse under
the Visigoths t h a n under the Romans, a n d h a d lost m a n y o f
their m u n i c i p a l privileges. Perhaps i t was difficult for the
Visigoths, because o f their p r i m i t i v e background, to appre-
ciate the benefits of commerce a n d o f c i t y life generally;
b u t the chief reason was p r o b a b l y the general economic re-
gression after the fall of the R o m a n empire.
 THE WEAKNESS O F VISIGOTHIC SPAIN J

This lack o f appreciation o f the need for commerce m a y

be one o f the factors u n d e r l y i n g the harsh treatment accorded
to the Jewish groups i n the k i n g d o m , for m a n y of the Jews
were merchants. A n o t h e r factor was the close association o f
the bishops a n d the k i n g ; m u c h of the w o r k o f governing the
k i n g d o m was transacted i n c h u r c h councils, a n d the eccle-
siastical authorities, influenced naturally b y theological con-
siderations, looked u p o n the Jews as enemies. Specially harsh
decrees o f t h e c o u n c i l of 693 made i t v i r t u a l l y impossible
for the Jews to continue as merchants. M a n y t h e n entered
i n t o conspiracy w i t h t h e i r Jewish associates i n N o r t h Africa,
a n d i n 694 a further decree enslaved a l l w h o d i d not accept
baptism. E v e n i f this was subsequently relaxed a n d n o t
strictly carried out, the strong feelings among the Jews m a y
have encouraged the M u s l i m s to invade, a n d the N o r t h A f r i -
can Jews w e r e doubtless ready to give w h a t information t h e y
possessed. Once the M u s l i m s h a d defeated the Visigothic
army, the Jews certainly gave t h e m a l l the h e l p they c o u l d .
T h e i m m e d i a t e p r e l u d e to the invasion was a customary
q u a r r e l about the succession, except t h a t this q u a r r e l h a d
p l u n g e d the peninsula i n t o w h a t was practically c i v i l w a r .
A father a n d son h a d reigned since 687. T h e son, W i t i z a ,
w a n t e d one o f his sons, A k h i l a , t o succeed, a n d to m a r k h i m
o u t a p p o i n t e d h i m dux o f the north-east province ( T a r r a -
conensis). W h e n W i t i z a d i e d i n 710, a strong group of nobles
seem to have elected Roderick k i n g , b u t A k h i l a appears t o
have retained his province and even to have m i n t e d coins
as i f he w e r e an independent sovereign. T h u s Roderick was
n o t i n secure c o n t r o l o f a l l the k i n g d o m w h e n he h a d to meet
the M u s l i m invaders, a n d i t is not surprising t h a t he was de-
feated, a n d t h a t after his defeat there was no i n d i v i d u a l or
group capable o f a c t i n g as central a u t h o r i t y o f the k i n g d o m .
T h e weakness o f the Visigothic k i n g d o m m a y thus be at-
t r i b u t e d to three m a i n factors: the divisions w i t h i n the upper
classes over the succession to the kingship; the discontent o f
the other sections o f society at the privileges o f the upper
classes, w i t h the consequent unreliability o f the a r m y ; a n d
the persecution of t h e Jews.
8 THE MUSLIM CONQUEST

3. The Course of the Invasion, 711-716

I t was i n A p r i l or M a y 7 1 1 that the first substantial b o d y o f

Muslims set foot i n southern Spain, a n d saw Andalusia i n
its most attractive m o o d . T h i s was n o t q u i t e the earliest
3

M u s l i m contact w i t h Spain, b u t w h a t w e n t before can be

discerned o n l y d i m l y t h r o u g h a haze of legend. Faute de
mieux w e must start f r o m this legend. T h e central figure is
a count Julian, about whose v e r y name there have been firece
arguments. H e was possibly a Byzantine exarch of Ceuta
(Septem) opposite Gibraltar. I f this is so, his comparative
isolation explains w h a t is tolerably certain—his close relations
w i t h one o f the parties contending for the throne o f the
Visigothic k i n g d o m o f Spain. T h e legend tells h o w Julian
became incensed because his beautiful daughter, w h o h a d
been sent for her education to Toledo, the Visigothic capital,
h a d been seduced b y Roderick, w h o , whether usurper or
not, was effectively b u t precariously k i n g o f Spain. I n his
rage Julian is said to have i n v o k e d M u s l i m help i n order to
gain his revenge. A p a r t from this story, a n u m b e r of small
indications suggest t h a t Julian a n d t h e Visigothic opponents
of Roderick deliberately t r i e d to interest the M u s l i m s i n Spain
and at first gave t h e m considerable h e l p .
A b o u t October 709 some o f Julian's m e n are said t o have
made a r a i d across the straits and demonstrated to the M u s -
lims that valuable booty was to be obtained. I n July 710 a
p a r t y o f four h u n d r e d M u s l i m s l a n d e d at the southernmost
t i p o f Spain (west o f G i b r a l t a r ) at the place n o w called
Tarifa, after their leader, Tarif. T h i s reconnaissance i n force
was successful, a n d the M u s l i m s w e r e sufficiently confident
to attempt a full-scale expedition i n the f o l l o w i n g year. Seven
thousand m e n were transported t o near Gibraltar i n vessels
p r o v i d e d b y Julian. T h e y were mostly Berbers, and their
leader was a Berber client of M u s i . ibn-Nusayr (the A r a b
governor of north-west Africa) whose name T a r i q i b n - Z i y a d
has been perpetuated i n Gibraltar, a c o r r u p t i o n o f "Jabal
T a r i q " , the m o u n t a i n o f T a r i q . T a r i q was a competent com-
 THE COURSE O F T H E INVASION, 7II-716 §

mander w h o h a d been placed i n c o m m a n d o f the forces gar-

risoning Tangier. O w i n g to the absence of K i n g Roderick i n
the n o r t h the M u s l i m s h a d t i m e to establish a base o n the
site o f the later t o w n of Algeciras. Roderick h u r r i e d south,
however, o n receiving news o f the expedition, a n d o n 19 July
attacked the M u s l i m s i n a valley or w a d i , n o w generally
identified w i t h t h a t o f the Rio Barbate. T h e M u s l i m s h a d
been reinforced b y a contingent of five thousand men, w h i l e
some o f Roderick's troops are said to have been disaffected
a n d t o have w i t h d r a w n f r o m the battle. T h e result was a de-
cisive v i c t o r y for the M u s l i m s . Roderick either was k i l l e d i n
battle or else he disappeared.
T h i s v i c t o r y demolished the central organisation of the
Visigothic k i n g d o m . T h e r e was further resistance, b u t o n l y
o n a local or regional basis. T a r i q q u i c k l y realised that Spain
was o p e n to h i m , a n d made for Cordova first. O n the w a y
he defeated a pocket o f V i s i g o t h i c fugitives at Ecija, a n d as
a result gained the support o f Jews a n d other discontented
groups over a w i d e area. H e thereupon decided to advance
w i t h t h e m a i n a r m y t o the V i s i g o t h i c capital, Toledo, w h i c h
he apparently occupied w i t h o u t serious opposition. H e was
p r o b a b l y also able t o make a reconnaissance to the n o r t h -
east i n the d i r e c t i o n of Saragossa before settling d o w n to w i n -
ter i n T o l e d o . M e a n w h i l e , i n October, Cordova surrendered
to a detachment o f seven h u n d r e d horsemen.
A c c o r d i n g to the sources, the p r o v i n c i a l governor of n o r t h -
west Africa, M u s a ibn-Nusayr, was filled w i t h jealousy o n
l e a r n i n g o f Tariq's successes, b u t this m a y be a distortion of
the facts. H i s conduct is compatible w i t h dispassionate p l a n -
n i n g to m a k e the most o f the o p p o r t u n i t y w h i c h h a d been
g i v e n to the M u s l i m s . W i t h eighteen thousand m e n , mostly
Arabs, he crossed the straits i n July 712 a n d advanced o n
Seville. Several smaller places were captured o n the w a y , a n d
resistance at Seville itself overcome. H e t h e n m o v e d n o r t h -
wards against a strong r e m n a n t of Visigoths w h o w i t h d r e w
i n t o M e r i d a , a n d h e l d o u t against the M u s l i m besiegers u n t i l
June 713. I t was apparently o n l y after this t h a t M u s a a n d
T a r i q met, t h e place b e i n g Talavera, a little w a y d o w n the
Tagus f r o m T o l e d o . A p a r t f r o m the q u e l l i n g o f a f e w revolts
lO THE MUSLIM CONQUEST

l i t t l e is recorded d u r i n g the year 713, b u t presumably t h e

M u s l i m armies were busy consolidating their h o l d o n the ter-
r i t o r y so far occupied.
D u r i n g the f o l l o w i n g year M u s a occupied Saragossa, a n d
m a y have sent o u t exploratory probes w h i c h reached as far
as Narbonne, for the Visigothic k i n g d o m h a d i n c l u d e d p a r t
of south-east France, i n c l u d i n g the Mediterranean littoral.
H e t h e n seems to have decided t h a t affairs i n the west were
more urgent a n d to have m o v e d westwards a n d penetrated
into the Asturias. T a r i q h a d already occupied L e o n a n d
Astorga, and F o r t u n of A r a g o n h a d s u b m i t t e d to h i m a n d
become a M u s l i m . I n the course of the year, however, M u s a
a n d T a r i q were summoned to the court of the c a l i p h at
Damascus. A fine story has been made o f M u s a s slow t r i u m -
p h a l progress t h r o u g h N o r t h Africa a n d E g y p t w i t h a great
t r a i n of captives a n d an unbelievably r i c h booty, and t h e n
of his harsh treatment b y the c a l i p h a n d his death i n prison
or at least i n poverty; b u t here again m u c h is legendary.
M u s a must have left Spain i n the a u t u m n of 714, for he
reached Damascus p r o b a b l y i n F e b r u a r y 715.
I n Spain the supreme c o m m a n d was left to M u s a s son
'Abd-al-'Aziz, w h o ably continued the w o r k of o c c u p y i n g
the country u n t i l he was assassinated i n M a r c h 716. T h e
M u s l i m h o l d o n the n o r t h and north-east was extended b y
the capture of Pamplona near the western e n d of the
Pyrenees a n d of Tarragona, Gerona and ( p r o b a b l y ) N a r -
bonne o n the Mediterranean seaboard. I n the south-east
M a l a g a and E l v i r a were taken, a n d a treaty made w i t h prince
T u d m l r ( T h e o d e m i r ) o f M u r c i a . A l l these events presumably
belong to 715, except that the treaty m a y have been i n 713.
W i t h the death o f 'Abd-al-'Aziz the phase of conquest and
occupation m a y be reckoned as h a v i n g come to an end. T h e
w h o l e o f the I b e r i a n peninsula h a d n o t been conquered nor
occupied. I n the north-west i n particular there was a large
area that was v i r t u a l l y untouched. Elsewhere, too, there were
probably localities where M u s l i m control was not effective.
Yet i n essentials the organisational u n i t y o f the country,
w h i c h h a d vanished i n the break-up of Visigothic power,
h a d been restored. There h a d been created a n e t w o r k o f
 THE COURSE O F T H E INVASION, 7II-716 11

administration, w i t h adequate m i l i t a r y b a c k i n g , covering

nearly the w h o l e peninsula, a n d the actual degree of con-
t r o l exercised b y the central M u s l i m a u t h o r i t y was p r o b a b l y
greater t h a n t h a t o f the later V i s i g o t h i c kings.
 2
T H E PROVINCE OF
T H E DAMASCUS CALIPHATE

1. The Organisation of the Province

T h e Arabs called their n e w d o m a i n i n the I b e r i a n peninsula

"al-Andalus". T h e w o r d is t h o u g h t to be a corruption of
"Vandalicia", a name derived from the V a n d a l invaders. I t
was used exclusively for that p a r t o f the peninsula under
M u s l i m rule, so that, as the Reconquista progressed, the
geographical area to w h i c h the t e r m was applied contracted.
I n m o d e r n usage Andalusia is applied to that region of south-
eastern Spain where the Moors h a d their last foothold i n the
thirteenth t o fifteenth centuries.
F o r the Arabs al-Andalus was o n l y a province, or p a r t of
a province, o f a vast empire w h i c h stretched from al-Andalus
and Morocco t o Central Asia and the Punjab. T h e ruler of
the empire was the caliph. This t e r m is an adaption of the
A r a b i c khaUfa, w h i c h means successor or deputy. T h e c a l i p h
was the m a n w h o h a d succeeded to the temporal b u t not
the spiritual powers of M u h a m m a d , T h e first four successors
of M u h a m m a d r u l e d from 632 to 6 6 1 a n d are k n o w n as "the
rightly-guided caliphs" (rdshidun). F r o m 661 to 750 the
caliphate was i n the hands of the f a m i l y of U m a y y a , a b r a n c h
of the tribe of Quraysh w h i c h i n h a b i t e d Mecca; some m e m -
bers of the f a m i l y h a d been among the leading merchants
of the t o w n d u r i n g Muhammad's lifetime. T h e U m a y y a d
caliphs h a d Damascus for their capital, t h o u g h the court was
often at one o f their palaces elsewhere i n Syria.
Despite the huge area controlled b y the U m a y y a d caliphs,
the organisation of government at the centre was still con-
ceived along the lines o f an A r a b i a n nomadic tribe. Above
all, i t was concerned w i t h persons rather t h a n places. T h e
caliph, too, was far from being an autocrat. O n the analogy
of a t r i b a l shaykh he was expected to consult the leading
 THE ORGANISATION OF T H E PROVINCE 13

m e n o f his entourage, and i n genera! to act towards t h e m as

first among equals. T h e weakness o f such a system, i n rela-
t i o n to the problems o f a large empire, was n o t far to seek,
a n d some of the later Umayyads tended towards the auto-
cratic Persian t r a d i t i o n o f statecraft—a t r a d i t i o n w h i c h was at
t h e basis of the succeeding ' A b b a s i d regime. A matter of
particular difficulty was t h a t of succession to the office of
c a l i p h . A c c o r d i n g t o A r a b ideas, primogeniture gave no spe-
cial privileges, a n d even the succession o f son to father was
n o t the o n l y possibility. T h e n e w chief or shaykh o f a t r i b e
was usually t h e best fitted a d u l t male f r o m a certain family,
agreed o n i n a meeting o f the leading members o f the t r i b e .
T h u s t h e r e t a i n i n g o f the caliphate i n the U m a y y a d family
was n o t achieved w i t h o u t manipulation, and was felt b y m a n y
Arabs to be a usurpation.
T h e caliphs, f o l l o w i n g the example o f M u h a m m a d , dele-
gated various duties t o individuals. T h e most i m p o r t a n t posi-
t i o n was t h a t of general o f an army. As w i d e territories came
to be conquered, the generals assumed the role o f provincial
governors. T h e change, i f such i t can be called, was a v e r y
simple one. W h e n a n a r m y r e t i r e d to w i n t e r i n a garrison-
c i t y i n recently acquired land—in Cairouan, for example—the
general continued i n his command, w h i c h became v i r t u a l l y
a c i v i l one, since the o n l y f u l l citizens o f the Islamic state
were the soldiers under h i m . F i n a n c i a l a n d j u d i c i a l matters
usually came under separate officials, w h o m i g h t be appointed
directly b y the c a l i p h ; b u t the general-governor was the m a n
i n charge.
T h e non-Muslims i n a province of the caliphate had, as
already noted, the status of "protected persons" or dhimmis
( t h o u g h i t is said t h a t i n al-Andalus dhimmis was a w o r d
t h a t came t o be a p p l i e d o n l y to Jews). T h e existing local
government was retained, a n d the head o f each c o m m u n i t y
became responsible to the M u s l i m a u t h o r i t y for the p a y m e n t
o f the t r i b u t e and other taxes and for the maintenance of
internal order. I n t h e M i d d l e East this responsible m a n was
usually the previous religious head of a religious group, such
as a p a t r i a r c h or bishop. T h i s seems also to have been the
n o r m a l case i n Spain, b u t i n 713 a treaty was made w i t h
T u d m l r ( T h e o d e m i r ) , the prince of M u r c i a , confirming h i m
14 DAMASCUS CALIPHATE

i n his rights as prince, and his subjects in—among other

things—the practice of their religion. W h e r e a c o m m u n i t y
refused the first summons t o surrender, and was t h e n de-
feated i n battle, the same status o f "protected persons" was
granted, b u t the conditions were liable to be more arduous
and the amount o f t r i b u t e and tax higher.
O r i g i n a l l y , a l l the M u s l i m Arabs were liable to m i l i t a r y
service a n d a l l received stipends f r o m the state. T h e y thus
constituted a superior m i l i t a r y caste. W h e n movable booty
was captured o n expeditions, i t was usually sold t o dealers
and the proceeds d i v i d e d among the participants i n the ex-
p e d i t i o n . Lands, however, were n o t sold, b u t left to the ex-
isting owners a n d tenants, a n d the rents p a i d into the central
exchequer. W h e r e the owners h a d fled, however—as was pre-
sumably the case w i t h some of the Visigothic noble families
—the governor o f the province h a d the r i g h t to donate these
lands t o M u s l i m s ; so i n course of t i m e m a n y of the Muslims
became landowners. I t is difficult t o follow i n detail the
transition f r o m a stipend-receiving class to a l a n d - o w n i n g
one i n any p a r t o f the caliphate, a n d i t is p a r t i c u l a r l y difficult
i n Spain. B y about 750 the system o f p a y i n g stipends h a d
apparently ceased to be important—presumably because i t
h a d come t o be a relatively small p a r t o f a m a n s income—
a n d i t p r o b a b l y fell into desuetude soon after that. Before
this happened, however, m a n y Arabs i n al-Andalus h a d be-
come landowners, usually residing i n towns close to their
estates.
U n t i l about 700 the non-Arabs i n the heartlands of the
caliphate were n o t encouraged to become Muslims. Some-
times, because o f the loss of revenue resulting f r o m conver-
sions (since M u s l i m s were not subject t o the p o l l - t a x ) , steps
were taken to discourage them f r o m leaving their religious
communities. T h e r e was always greater readiness to receive
those w h o were prepared to take p a r t i n m i l i t a r y expeditions,
a n d the presumption is that a l l the Berbers w h o entered
Spain h a d become Muslims. U p t o about 750, however, a
non-Arab, i n order t o become a M u s l i m , h a d to become a
client (mawla, p i . mawali) of an A r a b tribe. This was appar-
ently because the Islamic state was still regarded as a fed-
eration o f A r a b tribes. Since the status o f client was felt to be
 THE END OF T H E ADVANCE 15
an inferior one, and usually carried a smaller stipend t h a n
t h a t of the pure Arabs, there was discontent among the non-
A r a b M u s l i m s , w h o seem to have been increasing r a p i d l y
i n numbers d u r i n g the first half of the e i g h t h century. T h i s
discontent was an i m p o r t a n t factor c o n t r i b u t i n g to the fall
of the U m a y y a d caliphate of Damascus. T h e necessity for
non-Arab M u s l i m s to become clients of A r a b tribes seems
to have q u i e t l y disappeared soon after 750. I n al-Andalus
the Berbers, w h o mostly came from the mountainous parts
of N o r t h Africa, settled i n similar terrain and made a l i v i n g
b y pasturing animals.
W h i l e al-Andalus was part o f the Damascus caliphate, its
governors came under the Governor of I f r i q i y y a ( T u n i s i a ) ,
stationed i n Cairouan, a n d n o t d i r e c t l y under the caliph.
This was a reasonable arrangement i n v i e w o f the l e n g t h of
t i m e taken for communications a n d travel. I n the p e r i o d from
716 to 756 about t w e n t y m e n served as governor, some more
t h a n once. O n l y three h e l d the position for five years or more.
Some were o n l y temporary governors, replacing m e n w h o
h a d d i e d i n warfare against the Christians, or i n other ways.
Because they were so far f r o m Damascus a n d even f r o m
Cairouan, they were largely independent ( a n d this was doubt-
less one reason for the frequent changes). L i k e the caliphs,
however, they were not autocratic, b u t h a d t o show some
regard for the o p i n i o n of the leading Arabs i n al-Andalus.
T h e last of the governors, Yusuf ibn-'Abd-ar-Rahman a l - F i h r l ,
seems to have been appointed at an election i n the province
i n 747. T h i s was a t i m e w h e n the power of the c a l i p h i n
Damascus was already c r u m b l i n g away. I m m e d i a t e l y after
the conquest the p r o v i n c i a l capital h a d been Seville (instead
of the Visigoths' capital of T o l e d o ) , b u t about 717 i t was
transferred t o Cordova, as b e i n g more central.

2. The End of the Advance

Since the Visigothic k i n g d o m extended into southern France,

i t was n a t u r a l for the Arabs to occupy this part, also, o f the
k i n g d o m w h i c h they h a d conquered. I t was indeed p a r t of
the v a c u u m they h a d created b y destroying the central ad-
i6 DAMASCUS CALIPHATE

m i n i s t r a t i o n o f the Visigoths. Unfortunately, information

about the M u s l i m occupation o f France a n d t h e i r expeditions
there is scanty, b u t i t is l i k e l y that, i f there h a d been serious
Visigothic resistance i n France, some m e n t i o n w o u l d have
been made o f i t . T h e raids into the region o f N a r b o n n e m a y
have b e g u n soon after the defeat o f t h e Visigoths i n Spain.
C e r t a i n l y b y 719 the A r a b governor o f the t i m e , Samh, was
able to occupy Narbonne a n d advance towards Toulouse.
Energetic action b y the d u k e o f A q u i t a i n e , E u d o , however,
l e d i n 7 2 1 to the repulse o f the M u s l i m s f r o m Toulouse and
t h e death o f Samh.
T h i s reverse d i d n o t prevent the M u s l i m s f r o m a t t e m p t i n g
to find other lines o f advance i n t o France. I n 725 Carcassonne
a n d Nimes were occupied, a n d f r o m there a force made its
w a y n o r t h w a r d s u p the v a l l e y o f the Rhone. I t is said t o have
reached A u t u n o n the Sadne, or even further. T h i s p r o b i n g
operation, however, does n o t seem t o have been f o l l o w e d
u p . Instead there was a n exploration o f t h e route to the west
of t h e Pyrenees b y 'Abd-ar-Rahman al-Ghafiqi. H e collected
his troops at Pamplona i n 7 3 2 a n d advanced i n t o France b y
the pass o f Roncesvalles. E u d o o f A q u i t a i n e was defeated
a n d Bordeaux occupied. T h e n the M u s l i m s pressed o n n o r t h -
wards, m a k i n g for Tours, w h e r e p l e n t i f u l b o o t y was to be
expected. E u d o , however, alerted Charles M a r t e l , t h e p r i n c e
of t h e Franks, whose p o w e r h a d been g r o w i n g a n d w h o at
once realised the seriousness o f the danger. H e marched south
t o counter t h e M u s l i m threat, a n d between Tours a n d Poitiers,
towards the e n d o f October 732, there took place a battle,
variously k n o w n as the battle o f Tours a n d t h e battle of
Poitiers. T h e M u s l i m s w e r e defeated a n d their leader k i l l e d .
Some o f the retreating force appears t o have m a d e for Nar-
bonne. There is n o record o f the M u s l i m s h a v i n g again at-
t e m p t e d to i n v a d e France b y this western route.
Before considering the significance o f the battle of Tours
i t w i l l be useful t o m e n t i o n some pieces of i n f o r m a t i o n about
events i n the years i m m e d i a t e l y f o l l o w i n g . I n 734 the M u s -
lims showed renewed interest i n the valley o f the Rhone, a n d
an e x p e d i t i o n f r o m N a r b o n n e occupied Aries a n d A v i g n o n .
A b o u t 738, however, t h e y were d r i v e n back b y Charles
M a r t e l . H e even advanced to Narbonne, a n d besieged i t for
i8 DAMASCUS CALIPHATE

a t i m e ; b u t he was unable to capture i t . N o t h i n g is heard o f

further a c t i v i t y i n this region u n t i l after the fall o f the Damas-
cus caliphate. T h e n , most p r o b a b l y i n 7 5 1 ( t h o u g h possibly
not u n t i l 7 5 9 ) t h e successor of Charles M a r t e l finally recap-
t u r e d this i m p o r t a n t base f r o m the M u s l i m s .
T h e battle o f Tours is often called one of the decisive bat-
tles o f w o r l d history. A l t h o u g h i n a sense i t was, i t w o u l d be
m o r e accurate t o describe i t as the p o i n t at w h i c h the t i d e
t u r n e d . There was no cataclysmic destruction o f the central
m i l i t a r y and p o l i t i c a l p o w e r o f M u s l i m Spain. T h i s continued
m u c h as i t h a d been, b u t the leaders n o w realised that the
route b y the west o f t h e Pyrenees was not a satisfactory line
of expansion. T h e defeats b y Charles M a r t e l i n 738 showed
t h a t expansion u p the Rhone valley h a d also ceased to be
either possible or desirable. A l l these expeditions i n t o France,
however, were t h o r o u g h l y i n keeping w i t h the p o l i c y w h i c h
h a d directed t h e M u s l i m advance t h r o u g h N o r t h Africa a n d
i n t o Spain. W h i l e the personal motives of some o f the ex-
peditionaries were religious, a n d w h i l e religious factors m a y
have entered i n t o the general strategy, the i m m e d i a t e a i m of
the expeditions was plunder. T h e M u s l i m s were chiefly inter-
ested i n regions where m u c h p l u n d e r was t o be h a d easily.
T h e y were p r e p a r e d t o fight, a n d to fight fiercely, b u t o n l y
to a l i m i t e d extent. I f advance i n a certain d i r e c t i o n meant
serious a n d long-continued fighting, the p l u n d e r ceased t o
be w o r t h the efforts r e q u i r e d to g a i n i t ; a n d p r o b i n g expedi-
tions w o u l d be sent o u t i n other directions. W h a t the v i c t o r y
of Charles M a r t e l at Tours showed was t h a t his strength
was n o w such t h a t this line of advance h a d ceased to be
lucrative; a n d his subsequent moves against Narbonne made
i t clear t h a t advances i n t o France no longer offered any hope
of profit.
T h e matter c o u l d be p u t i n another w a y , namely, t h a t the
M u s l i m s ' w i l l to advance was weaker t h a n the Franks' w i l l
to resist t h a t advance; b u t various i n t e r n a l factors c o n t r i b u t e d
to t h e w e a k e n i n g o f the w i l l to advance. Besides the increas-
ing "cost" o f t h e plunder, i t has to be remembered that the
Muslims, accustomed to a Mediterranean climate, m a y have
found that o f central France uncongenial. T h e y doubtless
also h a d some premonitions of the break-up of the Damascus
 THE END OF T H E ADVANCE 19
caliphate, a n d for this reason felt insecure. T h e i r available
man-power, too, A r a b and Berber, must already have been
stretched t o t h e utmost of its capacity. So for m a n y reasons
the M u s l i m s h a d l i t t l e desire to continue their attempts t o
advance i n t o France. T h e t i d e h a d b e g u n t o ebb.
I t was n o t o n l y i n France, however, t h a t the t i d e t u r n e d .
I n north-west Spain also a M u s l i m recession began. L i t t l e
is k n o w n about w h a t was h a p p e n i n g here i n the quarter-
century after 7 1 1 . Presumably there were M u s l i m garrisons
i n a l l the towns o f any size. Yet i n strongholds i n the m o u n -
tains there w e r e small groups w h o refused to submit. T h e r e
were possibly Visigothic nobles among t h e m , b u t the w i l l t o
resist apparently came chiefly f r o m the leaders of the local
populations, a n d i n the first place the GaMeians, since the 1

w h o l e outlook of these people was different f r o m t h a t o f t h e

Visigoths. A somewhat legendary story o f h o w at Covadonga
a M u s l i m force (accompanied b y the m e t r o p o l i t a n o f Seville)
was repulsed b y prince Pelayo m a y be dated i n 718 or be-
t w e e n 7 2 1 a n d 726. A p a r t f r o m this n o t h i n g is recorded u n t i l
the r e i g n o f Alfonso I , k i n g of the Asturias, 7 3 9 - 5 7 . H e re-
conquered m u c h of north-west Spain a n d Portugal. T h e M u s -
lims m a y have w i t h d r a w n f r o m nearly a quarter o f t h e
I b e r i a n peninsula, t h o u g h n o t a l l o f this was occupied b y
t h e followers of Alfonso. Some was left largely u n i n h a b i t e d
to constitute the "marches".
T h e reasons for this t u r n of the t i d e i n Spain itself w e r e
not u n l i k e those for that i n France. I n a d d i t i o n there w e r e
some special factors c o n t r i b u t i n g t o the result. T h e M u s l i m
settlers here w e r e chiefly Berbers, and, as w i l l presently be
seen, t h e y w e r e dissatisfied w i t h the w a y the Arabs treated
t h e m , a n d h a d risen i n revolt. T h e n , b e g i n n i n g i n 750, there
was a serious famine w h i c h , c o m i n g o n t o p o f e v e r y t h i n g
else, caused m a n y o f t h e m to leave t h e i r Spanish lands a n d
return to Africa.
T h e historian, especially the historian of E u r o p e w h o is
aware o f the importance o f t h e Reconquista i n the emer-
gence o f Spain, sees i n the successes o f Alfonso I the seeds
of the destruction o f M u s l i m p o w e r i n Spain; and i n a sense
this is so. F r o m the M u s l i m p o i n t o f v i e w , however, w h a t
happened i n this p e r i o d of the collapse o f the Damascus
20 DAMASCUS CALIPHATE

caliphate merely meant that al-Andalus was to have an u n -

t i d y frontier; b u t i t was no more u n t i d y than m a n y other
frontiers of the caliphate, a n d the existence o f the k i n g d o m
of the Asturias d i d not i n itself mean that this realm of a l -
Andalus was doomed t o extinction before i t h a d w e l l begun
to live. I t merely meant that the Muslims h a d constantly to
face a challenge f r o m the n o r t h . T h e real p r o b l e m is w h y i n
the long r u n Christian strength grew and M u s l i m strength
declined.

3. Internal Tensions of the Province

T h e r a p i d occupation o f almost the w h o l e I b e r i a n peninsula

together w i t h the attempts at further expansion into France
i n e v i t a b l y h a d repercussions among the agents o f these op-
erations, namely, the Arabs and their Berber allies. T h e con-
version of t h e local inhabitants to Islam h a d b e g u n before
750, b u t the numbers w e r e n o t sufficient to constitute t h e m
a separate factor i n the politics of the day.
M u c h of the tension w h i c h is f o u n d among the Arabs is
ascribed, i n the sources, to the r i v a l r y between tribes a n d
groups o f tribes. I n particular there was a b i t t e r f e u d be-
t w e e n t w o groups k n o w n as the Qaysites a n d the Kalbites,
a n d this f e u d sometimes extended to larger groups genealogi-
cally connected w i t h the original pair of tribes, u n t i l p r a c t i -
cally a l l the tribes of A r a b i a were i n v o l v e d . This r i v a l r y be-
t w e e n the tribes was over-emphasised b y D o z y i n his
presentation of the history o f Islamic Spain, and this has
been recognised b y his disciple Levi-Provengal. U n d o u b t -
edly, however, t r i b a l r i v a l r y d i d exist a n d was a factor to be
taken account o f i n politics. T h e difficulty is to k n o w h o w to
interpret i t . T w o aspects have to be distinguished: the sig-
nificance o f the facts at the centre of the caliphate, and their
significance i n al-Andalus.
I n the caliphate i n general the r i v a l r y was usually be-
t w e e n groups o f tribes, and the M u s l i m historians justify
a n d explain this b y alleging that the groups w h i c h associated
w i t h one another were genealogically l i n k e d . M o d e r n E u r o -
pean historians, however, t e n d to regard the genealogies as
 INTERNAL TENSIONS O F T H E PROVINCE 21

subsequent t o the groupings ( t h a t is, i n v e n t e d b y genealo-

gists o f the U m a y y a d p e r i o d ) ; a n d the w a y i n w h i c h the
groupings v a r y i n different regions is one confirmation o f
this. T h e actual groupings, too, appear to be d e r i v e d f r o m
conditions i n the garrison towns a n d conquered provinces
a n d n o t f r o m t h e pre-Islamic situation i n A r a b i a . Instead o f
b l a m i n g ancient feuds, m o d e r n historians see the basis o f t h e
tension i n Syria i n the fact t h a t m a n y Kalbites h a d been set-
t l e d there before the A r a b expansion, whereas those w h o
came w i t h the expansion were m o s t l y Qaysites. T h u s there
was a contemporary social a n d perhaps also economic differ-
ence u n d e r l y i n g the tension.
After 740, t r i b a l r i v a l r y became an i m p o r t a n t p o l i t i c a l
factor i n al-Andalus. I n p a r t w h a t happened i n the province
m a y simply have been a reflection o f w h a t was h a p p e n i n g
at the capital. T h e Qaysites a n d Kalbites functioned almost
like parties i n a m o d e r n state. W h i l e the c a l i p h was r e l y i n g
m a i n l y o n one party, most of the p r o v i n c i a l appointments
w e n t to its members. Social a n d economic differences be-
t w e e n the t w o groups no d o u b t influenced their support o f
different policies; b u t the meagre sources do not appear to
have been examined f r o m this p o i n t o f v i e w .
V e r y l i t t l e is k n o w n about w h a t was h a p p e n i n g to Spain
between 720 a n d 740. R a i d i n g expeditions were made i n t o
France; the pacification a n d settlement of the country con-
t i n u e d ; a n d local revolts were q u e l l e d . I n 740, however, a
revolt o f Berbers broke out i n N o r t h Africa, a n d Tangier
was captured b y the insurgents. Troops sent b y the gover-
nor f r o m his seat at Cairouan w e r e defeated, a n d i n 7 4 1 ,
despite reinforcements of Syrians f r o m Damascus, his a r m y
was again defeated. These successes l e d to a Berber r e v o l t
i n the north-west o f Spain. A l l t h e Berbers were t h o r o u g h l y
dissatisfied w i t h the treatment w h i c h t h e y received at t h e
hands o f the Arabs. T h e y were g i v e n inferior shares of w h a t -
ever was d i s t r i b u t e d , a n d the less desirable regions i n w h i c h
to settle, and, a l t h o u g h they w e r e Muslims, the Arabs d i d
n o t r e g a r d t h e m as equals. Since t h e y w e r e m o r e numerous,
a n d possibly also tougher fighters m a n for man, i t was n o t
surprising t h a t they were i n i t i a l l y successful.
T o w a r d s the end of 7 4 1 , however, an i m p o r t a n t n e w ele-
22 DAMASCUS CALIPHATE

m e n t came u p o n the scene i n al-Andalus. After the Berber

v i c t o r y i n N o r t h Africa seven thousand horsemen of the
Syrian reinforcements, under their leader Balj, took refuge
i n Ceuta a n d were besieged b y the Berbers. I n this somewhat
difficult situation they agreed w i t h the governor of al-Andalus
that, i f he supplied the transport, they w o u l d fight against
the rebels i n Spain and leave again w h e n the rebellion was
over. T h e y were accordingly ferried across, and defeated
three columns o f Berbers one after the other, and w o u l d
presumably have left h a d the governor not t r i e d to evade
carrying o u t his p a r t of the agreement i n f u l l This governor
was o f the Kalbite, or rather the w i d e r Yemenite, party,
w h i l e the Syrian Arabs under Balj were Qaysites. Instead of
leaving the country, therefore, they marched on Cordova
a n d expelled the governor, installing Balj i n his place. T h e
opposing Arabs reorganised themselves and gained some
Berber support, b u t were defeated b y Balj i n August 742,
t h o u g h he himself was k i l l e d .
T h e n e w governor sent from Cairouan t r i e d to quieten
the country b y settling the Syrians o n lands i n the valley o f
the G u a d a l q u i v i r and along the south coast. I n Syria they
h a d been jundis, t h a t is, m e n w h o received fiefs of l a n d i n
r e t u r n for b e i n g ready to serve i n the army w h e n required;
a n d their settlement i n al-Andalus was o n similar terms. This
d i d not prevent t h e m c o m b i n i n g w i t h certain other A r a b
tribes and f r o m 745 to 755 m a i n t a i n i n g i n power governors
favourable to their interests. B y 755 the opposing Yemenite
Arabs were showing signs of p r e p a r i n g to challenge the
d o m i n a n t coalition, w h i l e the n o r t h of die country was re-
covering from t h e famine w h i c h h a d afflicted i t since 750.
A t this juncture 'Abd-ar-Rahman ( b o r n 7 3 0 ) , a y o u n g m e m -
ber o f the U m a y y a d family, w h o h a d escaped from I r a q a n d
Syria w h e n a l l his relatives h a d been annihilated b y the
'Abbasids o n their c o m i n g to power i n 750, despatched an
emissary to al-Andalus. 'Abd-ar-Rahman h a d l i v e d for some
t i m e w i t h his Berber m o t h e r s t r i b e near the Mediterranean
coast o f Morocco, a n d his emissary was enthusiastically re-
ceived b y some o f the Syrian jundis, w h o were mostly clients
of the Umayyads. T h e leaders of the group i n power since
745 at first hesitated a n d finally rejected 'Abd-ar-Rahman's
 INTERNAL TENSIONS O F T H E PROVINCE 23

proposals. U p o n this the emissary successfully approached

the opposing Yemenite group. *Abd-ar-Rahman crossed t h e
straits, a n d w i t h an a r m y of Syrian jundis, Yemenites a n d
some Andalusian Berbers defeated the remnants o f the
Qaysite group i n M a y 756. T h e country as a w h o l e t h e n sub-
m i t t e d to h i m , and he was p r o c l a i m e d emir of al-Andalus i n
t h e mosque o f Cordova. T h e U m a y y a d emirate h a d been
established.
 3
THE INDEPENDENT UMAYYAD EMIRATE

1. The Establishment of the Emirate

'Abd-ar-Rahman i : 7 5 6 - 7 8 8
H i s h a m 1: 7 8 8 - 7 9 6
al-IJakam 1: 7 9 6 - 8 2 2
'Abd-ar-Rahman 11: 8 2 2 - 8 5 2

T h e proclamation o f 'Abd-ar-Rahman as emir created a novel

situation, t h o u g h the novelty was more theoretical t h a n prac-
tical. T h e title of "emir" or "commander" h a d hitherto been
used b y p r o v i n c i a l governors appointed b y the caliph; b u t ,
since the ' A b b a s i d caliphs were responsible for the massacre
of nearly a l l the U m a y y a d family, there c o u l d be no ques-
t i o n o f 'Abd-ar-Rahman recognising the c a l i p h i n any w a y .
O n the other h a n d , his position was never such that he c o u l d
c l a i m the office of c a l i p h for himself. So, for the first time i n
the Islamic w o r l d , there was a political e n t i t y w h i c h , w i t h -
o u t justifying itself b y heretical dogma, was organised i n
complete independence of the m a i n b o d y of Muslims. T h i s
was the theoretical novelty.
I n practice, however, the element of novelty was not great.
W h e n communications stretched over vast distances and were
slow, p r o v i n c i a l governors were largely left to their o w n de-
vices. T h i s h a d been p a r t i c u l a r l y the case i n the last decade
or so before the fall of the U m a y y a d caliphate i n 750. T h e
caliph h a d i n d e e d sent a large force from Syria to help
against the Berber revolt i n N o r t h Africa. A l t h o u g h from
this force the horsemen under Balj h a d crossed i n t o Spain,
this h a d been a more or less private arrangement between
the governor and the m e n concerned; a n d after this event
the M u s l i m s i n al-Andalus h a d been almost entirely o n their
o w n . T h e chief novelty i n 'Abd-ar-Rahman's position, then,
was that he h a d no superior to call o n h i m to d e m i t office,
 THE ESTABLISHMENT OF THE EMIRATE 2$

a n d t h a t he h a d a certain r i g h t to rule. Perhaps also the

knowledge t h a t al-Andalus was thus isolated emboldened
rebels i n their attempts t o seize power. T h e possibility that
the 'Abbasids m i g h t t r y to assert their a u t h o r i t y over this p a r t
o f the empire o f their predecessors h a d also t o be taken i n t o
account; b u t i t cost t h e m m u c h t i m e a n d effort t o secure even
a tenuous h o l d over N o r t h Africa, a n d t h e y were never a
serious threat to the n e w U m a y y a d regime i n Spain.
T h e m a i n p r o b l e m w i t h w h i c h 'Abd-ar-Rahman and his
immediate successors h a d t o cope, i n order to establish t h e
emirate securely, was the diversity of elements, m a i n l y racial,
i n t h e p o p u l a t i o n . F i r s t o f a l l came the Arabs w h o , t h o u g h
n o t numerous, h e l d a d o m i n a n t position; a n d they were fur-
ther d i v i d e d among themselves. T h e o l d opposition, however,
b e t w e e n Yemenites ( o r Kalbites) and Qaysites seems gradu-
ally t o have m e r g e d i n t o another, namely, that between
Arabs of the first wave, k n o w n as "old-established" (baladiy-
yun), a n d the later comers, k n o w n as "Syrians" (shdmiyyun).
Since t h e Syrians, as explained above, h a d been given fiefs,
this distinction was i n p a r t a social a n d economic one. T h e
Arabs, o f course, were a l l Muslims.
I n a d d i t i o n there were t w o other groups o f Muslims, t h e
Berbers a n d t h e local converts. T h e Berbers were m u c h the
more numerous, for they h a d supplied the b u l k of the i n -
v a d i n g a n d o c c u p y i n g man-power. T h e more significant o f
t h e Berber immigrants were those f r o m the sedentaries (as
distinct f r o m t h e n o m a d s ) , a n d i n Spain they t i n n e d again
t o a g r i c u l t u r a l activity. Despite their numbers they were, as
w e have noted, treated as inferiors b y the Arabs, and discon-
tent constantly smouldered among t h e m . T h e local converts,
i t w o u l d seem, came, after a t i m e , to be as numerous as
the Berbers, or even more so. T h e t e r m for "convert" was
musalim, b u t this seems t o have been restricted to those w h o
actually changed their religion. T h e n o r m a l name a m o n g
the Arabs for Spanish M u s l i m s was muwalladun, w h i c h m i g h t
be rendered " b o r n Muslims". T h e y are usually referred t o
b y Spanish writers as "renegades", a t e r m w h i c h doubtless
was n o t used u n t i l the Reconquista was w e l l under w a y . T h e
chief m o t i v e for the acceptance of Islam b y a large section o f
the Spanish p o p u l a t i o n was p r o b a b l y its association w i t h a
26 THE INDEPENDENT UMAYYAD EMIRATE

higher and v e r y attractive civilisation, coupled w i t h distrust

of the Christian bishops because of their close identification
w i t h the unpopular Visigothic rule.
T h e r e m a i n i n g large element i n the Islamic state—the
Christian p o p u l a t i o n w h o kept their religion—were called
Mozarabs (mustdribun), w h i c h m i g h t be rendered "arab-
izers", again p r o b a b l y a later t e r m used b y the Christians of
the Reconquista. T h o u g h Christian, these people also were
1

presumably attracted b y m a n y aspects of A r a b and Islamic

civilisation. T h e y were b y no means hostile to M u s l i m rule,
b u t learnt A r a b i c ( t h o u g h they also spoke a Romance dia-
l e c t ) and adopted m a n y A r a b customs. Besides the Chris-
2

tians there were m a n y Jews i n the chief towns, w h o , h a v i n g

suffered under the Visigoths, actively aided the M u s l i m con-
quest, and d o n o t appear later to have t h o u g h t of r e v o l t i n g .
W i t h a l l these diverse and often discordant elements, even
t o m a i n t a i n effective rule was difficult. There were numer-
ous revolts a n d risings of one k i n d or another. Sometimes
o n l y one o f the groups mentioned was i n v o l v e d ; sometimes
t w o or more w o u l d j o i n together. T h e o l d system b y w h i c h
every able-bodied M u s l i m was liable for service h a d broken
d o w n before 750, a n d w o u l d i n any case have been of little
use i n dealing w i t h the situation i n al-Andalus. One of *Abd-
ar-Rahman s methods of t a c k l i n g the p r o b l e m was to create
a professional army. This p r o b a b l y consisted largely of slaves,
w h o were easily obtainable from n o r t h of the Pyrenees. T h e
size of this mercenary army was increased b y his successors.
T h i s made the emir independent o f his subjects, b u t i t also
created serious problems for h i m .
I t has been suggested that the Umayyads achieved u n i t y
f r o m this heterogeneous b o d y b y i d e n t i f y i n g their cause w i t h
that of Islam, b u t there are complexities w h i c h this sugges-
t i o n does n o t explain. T h e matter w i l l be considered more
f u l l y i n a later chapter. Here let i t be n o t e d that this was at
best a long-term policy. For the m o m e n t the aim was to make
t h e emir the centre of u n i t y , b u t to b e g i n w i t h his authority
c o u l d o n l y be maintained b y sheer force. A notorious ex-
ample o f this is the so-called "day of the Foss" at Toledo,
p r o b a b l y i n 797 ( n o t i n 807) soon after al-Hakam began to
rule. A l l the notables of Toledo, m a i n l y Spanish M u s l i m s ,
 THE ESTABLISHMENT OF THE EMIRATE TTf

w h o h a d earlier shown signs of disaffection, were t r i c k e d b y

the governor into entering the castle o n the pretext of pay-
i n g t h e i r respects t o the heir to the throne; once inside t h e y
were beheaded one b y one a n d the bodies t h r o w n i n t o a
t r e n c h or foss.
T o w a r d s the end of the same reign, p r o b a b l y i n 818, there
was a n even more notorious blood-bath at Cordova. T h e
harshness o f t h e emir p r o v o k e d a rising among the inhabitants
of t h e "suburb'' south o f the G u a d a l q u i v i r . F o r a t i m e t h e
issue was i n doubt, b u t eventually the emir's troops gained
the u p p e r h a n d ; the "suburb" was p l u n d e r e d , three h u n d r e d
of the l e a d i n g survivors were executed, the rest o f the people
made to leave Cordova, a n d the w h o l e o f the "suburb'*
p l o u g h e d u p . T h e importance attached t o these events i n
t h e A r a b i c sources a n d i n some older E u r o p e a n accounts
should n o t make the m o d e r n reader suppose t h a t there was
n o t h i n g besides force t o support the central authority. I n t h e
r e v o l t o f the "suburb" one or t w o M u s l i m jurists were in­
v o l v e d , a n d this appearance o f a n e w class is at the same
t i m e a n i n d i c a t i o n t h a t t h e Umayyads were t r y i n g to develop
justice a n d fair treatment i n their realm.
W h i l e the U m a y y a d s were thus busy establishing their r u l e
t h r o u g h o u t t h e territories t h a t h a d come i n t o their hands,
there were no events o f p r i m a r y importance o n the n o r t h e r n
frontier, t h o u g h there was a certain amount o f activity. I n
the years f r o m 7 4 0 to 755 the l i t t l e k i n g d o m o f the Asturias
i n the north-west o f the I b e r i a n peninsula was able to ex-
p a n d somewhat a n d t o make itself relatively secure against
attacks. B e y o n d the Pyrenees, again, Charlemagne ( 7 7 1 - 8 1 4 )
was b u i l d i n g his p o w e r f u l empire. Occasionally he made i n -
cursions i n t o the peninsula, as i n 8 0 1 w h e n he entered Bar-
celona. H i s expedition o f 778 against Saragossa has been
made famous b y the Chanson de Roland. T h e central event
i n this p o e m , the defeat o f a rearguard at Roncesvalles, was
a v e r y m i n o r affair f r o m the m i l i t a r y standpoint; b u t t h e
more i m p o r t a n t aspect o f the campaign—the failure to take
Saragossa—may have influenced Charlemagne to leave Spain
alone for the most part.
Al-Andalus h a d no n o r t h e r n frontier i n t h e m o d e r n sense.
B e t w e e n the area securely h e l d b y the M u s l i m s and that se-
2,8 THE INDEPENDENT UMAYYAD EMIRATE

eurely h e l d b y the various Christian states was a tract of l a n d

more loosely h e l d a n d m e r g i n g i n t o a no-man s l a n d . These
were the Marches. T h e M u s l i m defence was based o n Sara-
gossa for the U p p e r M a r c h , on T o l e d o for the M i d d l e M a r c h ,
a n d o n M e r i d a for the L o w e r M a r c h . A t certain periods the
Muslims made a n expedition northwards every summer, b u t
these periods alternated w i t h periods o f apparent truce. One
of the most notable expeditions was t o Narbonne i n 793, a n d
there was another i n 8 4 1 to the same neighbourhood. Neither
the latter nor one i n 828 explicitly against Barcelona was
able to recapture t h a t t o w n f r o m t h e Franks.
B y the r e i g n o f 'Abd-ar-Rahman I I ( 8 2 2 - 5 2 ) the U m a y -
y a d emirate was w e l l established, a n d the c o u n t r y was
prospering. T h e r e were still revolts, b u t they were peripheral,
w h i l e at the centre a measure o f u n i t y h a d been achieved.
One index o f the general prosperity was the extensive b u i l d -
i n g programme carried out b y 'Abd-ar-Rahman I I . T h e series
of watch-posts set u p after 844 to g u a r d against Norse sea-
raiders shows the strength a n d practical efficiency o f the
regime. I n d e e d , *Abd-ar-Rahman I I felt sufficiently power-
f u l to enter i n t o t h e politics o f t h e various small a n d m i d d l e -
size states w h i c h occupied the region f r o m M o r o c c o to T u -
nisia, a n d t o support some of the smaller ones against their
larger neighbours. A f u l l discussion, however, of the basis of
U m a y y a d p o w e r a n d prosperity w i l l best come after w e have
seen the state at its z e n i t h i n the next century.

2. The Crisis of the Emirate

M u h a m m a d 1: 8 5 2 - 8 8 6
al-Mundhir: 886-888
'Abd-Allah: 888-912

W h e n 'Abd-ar-Rahman I I d i e d i n 852, the U m a y y a d state

was prospering, a n d appeared to be strong a n d securely es-
tablished. Yet the events of the next sixty years showed that
this appearance was deceptive, a n d that i n reality the struc-
ture was fragile a n d precarious. T h e troubles i n the earlier
p e r i o d h a d nearly a l l been due to t o w n s m e n w h o were dis-
 THE CRISIS OF THE EMIRATE 29
contented for one reason or another, a n d w h o h a d to v e n t
their discontent b y h i t t i n g o u t at a u t h o r i t y ; at the same t i m e
t h e y h a d l i t t l e idea o f a genuine alternative to the existing
r e g i m e a n d p o l i t i c a l system. Before the e n d o f the n i n t h cen-
t u r y , however, there h a d appeared a n u m b e r o f ambitious
individuals w h o f o u n d i n popular discontent an instrument
b y w h i c h t h e y c o u l d create an independent or semi-inde-
pendent l i t t l e state for themselves.
T h i s feature seems to have b e g u n i n the Marches. T h e
general conception of the Marches was good, a n d p r o v e d
a n effective means of defending the more densely p o p u l a t e d
parts o f the c o u n t r y i n the south a n d south-east. T h e system,
however, i m p l i e d g i v i n g a measure of p o w e r a n d independ-
ence to the governors o f the Marches a n d various subordinate
lords. As early as 842 one o f the latter, M u s a i b n - M u s a ibn-al-
Qasi, governor of T u d e l a , refused t o obey the emir, and suc-
cessfully resisted a series o f attacks b y t h e emir's troops. T h e
emir eventually accepted his profession o f loyalty, t h o u g h o n
M u s a s o w n terms. Before his death i n 862 M u s a was t h e
effective r u l e r o f most o f t h e U p p e r M a r c h , i n c l u d i n g Sara-
gossa, a n d even called himself "the t h i r d k i n g of Spain".
B e g i n n i n g i n 8 7 1 three o f his sons, w h o retained most of t h e
f a m i l y possessions, a t t e m p t e d to regain his power, b u t t h e
difficulties were too great, a n d i n 884 t h e sole survivor sold
Saragossa to the emir. T h e latter was n o t m u c h better off,
however, since he h a d to lean h e a v i l y i n this region o n a r i v a l
f a m i l y o f A r a b o r i g i n often k n o w n as the Tujibids, w h o also
demanded a measure o f independence.
T w o other n o t dissimilar series of events m a y be m e n -
t i o n e d briefly. I n the first case the muwallad (or Spanish
M u s l i m ) I b n - a l - J i l l i q l m a i n t a i n e d himself i n p a r t i a l i n d e -
pendence i n the region o f M 6 r i d a i n the L o w e r M a r c h f r o m
875 onwards, a n d his sons and lieutenants d i d n o t fully sub-
m i t t o the central government u n t i l 930. I n Seville, o n t h e
other h a n d , t w o A r a b families gained t h e u p p e r h a n d i n a
struggle w i t h the muwallads, a n d t h e n i n 899, after a quar-
r e l between t h e m , the head of one became semi-independent
ruler o f the region, recognised b y the emir a n d eventually
succeeded b y his sons.
T h e most threatening o f these attempts at independence,
30 T H E INDEPENDENT UMAYYAD EMIRATE

however, was t h a t o f I b n - I I a f s u n , another muwallad. I n 880,

w i t h a c o m p a n y o f malcontents, he raised a r e v o l t i n the
south, m a k i n g his centre the castle o f Bobastro. T h e country
was seething w i t h discontent, a n d he h a d l i t t l e difficulty i n
b u i l d i n g u p his o w n power and d e f y i n g the U m a y y a d armies.
W i t h his successes his ambitions g r e w , and he a l l o w e d no
principles to stand i n the w a y o f his efforts to increase his
power. A b o u t 890 he was negotiating w i t h the semi-
independent r u l e r o f Cairouan (recognised b y the 'Abbasids)
i n order to o b t a i n m i l i t a r y support a n d become emir of Spain.
A t this p e r i o d Ibn-IJafsun h a d m u c h support from the mu-
wallads w h o h a d gone over to h i m after b e i n g i n v o l v e d i n
fighting w i t h the Arabs of the region. I n 899, however, he
m a y have lost m u c h o f this support b y b e c o m i n g a Christian,
t h o u g h he doubtless also gained m u c h from the Mozarabs.
T h i s change o f religion d i d not prevent h i m i n 910 f r o m pro-
fessing his friendship for the F a t i m i d regime w h i c h h a d es-
tablished itself i n Tunisia i n the previous year. E v e n under
'Abd-ar-Rahman I I I the central government, t h o u g h i t weak-
ened Ibn-IJafsun's power, was unable to dislodge h i m f r o m
Bobastro; a n d after his death i n 9 1 7 his sons k e p t u p their
resistance for some ten years. T h e l e n g t h of this insurrection
is a clear index of the relative weakness of the central gov-
ernment.
One n o t e w o r t h y feature of these events and of the p e r i o d
i n general is the m i n g l i n g o f C h r i s t i a n i t y and Islam. T h e
f a m i l y of M u s a i b n - M u s a ibn-al-Qasi, i n the U p p e r M a r c h ,
h a d ties o f b l o o d a n d marriage w i t h the family w h i c h , at this
same period, was creating the k i n g d o m of Navarre r o u n d
Pamplona; indeed i t made contributions to the g r o w t h of that
k i n g d o m w h i c h were far f r o m negligible. T h i s p o i n t is p r o b -
a b l y to be connected w i t h the spread of Frankish feudal
practices, since feudalism placed the emphasis o n a m a n s
relation to his l o r d t o such an extent t h a t his religion almost
ceased to be relevant. M a n y incidents o f this p e r i o d i n w h i c h
3

m e n changed r e l i g i o n or swore allegiance to a l o r d o f the

other religion suggest that these n i n t h - c e n t u r y struggles were
n o t regarded p r i m a r i l y as struggles between the t w o religions.
I t w o u l d f o l l o w f r o m this that, t i l l this p e r i o d , i t cannot have
been the p o l i c y o f the Umayyads to make I s l a m the chief
 THE CRISIS OF THE EMIRATE 31
i n t e g r a t i n g force i n al-Andalus, or at least, i f t h e y h a d such
a policy, i t h a d not so far become effective. O n the other
h a n d , the Umayyads m a y have been b e g i n n i n g to be inter-
ested i n islamisation, for the emir ' A b d - A l l a h ( 8 8 8 - 9 1 2 ) is
said t o have been under the influence o f the jurists, and the
existence of jurists w o u l d be a m a r k o f islamisation.
I t is relevant at this p o i n t to m e n t i o n the theory of A m e r i c o
Castro i n The Structure of Spanish History ( 1 3 0 - 1 7 0 ) . H e
sees the cult of Saint James of Compostella, i n c l u d i n g t h e
pilgrimages, as incorporating an o l d Galician or I b e r i a n be-
lief i n the H e a v e n l y T w i n s (since James was regarded as
the t w i n - b r o t h e r o f Jesus), and also as g i v i n g the Galicians
a n d their neighbours f r o m the n i n t h century onwards the
firm conviction that they h a d d i v i n e h e l p i n their w a r w i t h
the M u s l i m s a n d that thereby t h e y w o u l d eventually be v i c -
torious. This c u l t is therefore the source o f the spiritual
strength u n d e r l y i n g the Reconquista. A p a r t f r o m the theory
i t is certain t h a t the c u l t existed i n the first half o f the n i n t h
century, a n d that under Alfonso I I I ( 8 6 6 - 9 1 0 ) , w h i l e t h e
Muslims w e r e busied w i t h their internal divisions, the j o i n t
Asturian-Leonese k i n g d o m expanded and grew stronger.
 4
T H E G R A N D E U R OF T H E U M A Y Y A D CALIPHATE

1. Umayyad Spain at its Zenith

'Abd-ar-Rahman n i : 9 1 2 - 9 6 1
al-Hakam 11: 9 6 1 - 9 7 6

T h e e m i r ' A b d - A l l a h was succeeded b y his grandson ' A b d -

ar-Rahman I I I , a y o u n g m a n o f twenty-one. W h e n the n e w
ruler came t o t h e throne the prospects for al-Andalus were
n o t b r i g h t . I n a d d i t i o n to w h a t was practically a c i v i l w a r
against I b n - H a f s u n , and t o the d i m i n i s h i n g control b y the
central government over the lords of the Marches, there
were t w o external dangers appearing o n the h o r i z o n : i n the
N o r t h , the k i n g d o m of L e o n and, i n w h a t is n o w Tunisia,
the n e w F a t i m i d power. Yet b y his gifts o f character a n d
statesmanship a n d b y the good fortune o f a l o n g r e i g n ' A b d -
ar-Rahman was able not merely t o overcome these weak-
nesses and threats, b u t also t o b r i n g al-Andalus to a h e i g h t
of greatness.
One o f his p r i m e concerns was the restoration o f i n t e r n a l
u n i t y . Vigorous a n d well-conducted campaigns d u r i n g the
first t w o years o f his r e i g n l e d t o the defeat of m a n y sup-
porters o f I b n - H a f s u n i n the outer section o f his sphere of
influence; to the reconciliation of waverers w i t h the govern-
m e n t i n Cordova; a n d to the encouragement o f those loyal
to i t . A large n u m b e r of castles a n d strongholds were placed
i n reliable hands. Advantage was taken of a dispute w i t h i n
the f a m i l y r u l i n g Seville i n o n l y n o m i n a l dependence, a n d
before the e n d o f 913 a governor obedient to 'Abd-ar-
R a h m a n h a d been installed. B y such tactics I b n - l j a f s u n s au-
t h o r i t y was greatly weakened, a n d after his death i n 917
his sons quarrelled and their p o w e r disintegrated. T h e sur-
render of Bobastro i n 928 m a r k e d the e n d of the threat to
u n i t y . I n the years i m m e d i a t e l y f o l l o w i n g , 'Abd-ar-Rahman
 UMAYYAD SPAIN AT ITS ZENITH 33
completed the w o r k of establishing effective control over the
Marches. I n the lower M a r c h this was m a r k e d b y the sur-
render to h i m , i n 930, o f Badajoz, b y a descendant of I b n - a l -
J i l l i q i . I n t h e m i d d l e M a r c h i t r e q u i r e d a two-year siege
before Toledo surrendered i n 932. I n the upper M a r c h , o n
the other h a n d , the T u j i b i d s h a d f r o m the first shown t h e m -
selves relatively faithful vassals of 'Abd-ar-Rahman, t h o u g h
i n 9 3 7 the l o r d o f Saragossa transferred his allegiance to the
k i n g o f L e o n , a n d i t was o n l y after a m i l i t a r y campaign i n
the r e g i o n a n d a siege o f Saragossa t h a t 'Abd-ar-Rahman
restored his control o f the upper M a r c h .
W h i l s t the first t w e n t y years of the r e i g n thus saw t h e
u n i t y o f al-Andalus re-established, they were also n o t e w o r t h y
for a large measure o f success against the Christian k i n g -
doms o f the n o r t h , L e o n a n d Navarre. I t m a y be that the
weakness o f these states was i n some w a y a repercussion o f
the collapse o f the Carolingian empire, or i t m a y be t h a t
the rulers o f the p e r i o d were less capable t h a n some of their
predecessors a n d successors. Certainly b y expeditions i n 9 2 0
a n d 924 'Abd-ar-Rahman was able to p u t a stop to the Chris-
t i a n raids i n t o M u s l i m territory. A h a l t was placed to the
extension of M u s l i m influence, however, d u r i n g the r e i g n o f
R a m i r o I I o f L e o n from 9 3 2 to 950. ( I t is convenient to speak
simply o f " L e o n " for w h a t is p r o p e r l y the k i n g d o m of t h e
Asturias a n d L e o n . )
T h e climax of the successes of Ramiro was i n 939. ' A b d -
ar-Rahman h a d marched against L e o n w i t h a larger a r m y
t h a n usual, allegedly about a h u n d r e d thousand men. H e was
met b y R a m i r o at Simancas, just south o f m o d e r n V a l l a d o l i d .
After several days o f p r e l i m i n a r y encounters the u n w i e l d y
M u s l i m forces w e r e p u t to flight, and m a n y lost their lives
o w i n g to the fact that R a m i r o h a d previously d u g a trench
(khandaq) at their rear. T h i s m i l i t a r y disaster was n o t i r -
reparable, b u t i t was a severe b l o w t o 'Abd-ar-Rahmans
p r i d e . Ramiro took advantage o f his success to resettle Chris-
tians i n the neighbourhood o f Salamanca. Presently, h o w -
ever, he became f u l l y occupied i n q u e l l i n g the attempted
assertion of its independence b y Castile; and 'Abd-ar-
R a h m a n soon restored his m i l i t a r y strength a n d his p o l i t i c a l
influence.
34 UMAYYAD CALIPHATE

A f t e r the death of Ramiro I I i n 950 internal disputes

greatly weakened the Christian states, a n d the years f r o m
9 5 1 t o 9 6 1 witnessed a great increase i n the p o w e r and influ-
ence of 'Abd-ar-Rahman. T h e suzerainty or hegemony of
*Abd-ar-Rahman a n d his successors was acknowledged b y the
k i n g of L e o n , the queen of N a v a r r e a n d the counts of Cas-
tile a n d Barcelona; a n d this acknowledgement was n o mere
f o r m a l affair, b u t was accompanied b y the payment o f an
annual i n d e m n i t y or t r i b u t e ; a n d failure to pay resulted i n a
p u n i t i v e r a i d . A t the same t i m e a n u m b e r of strongholds
were dismantled or handed over to the Muslims. Thus from
about 960 to the e n d of the century M u s l i m control of the
I b e r i a n peninsula was more complete t h a n at any other t i m e
before or after.
This makes i t pertinent to consider the v i e w expressed
b y A r n o l d Toynbee i n A Study of History ( v i i i . 3 5 1 ) that
the failure o f 'Abd-ar-Rahman a n d al-Mansur to complete
the conquest o f the peninsula at this p e r i o d w h e n they h a d
u n d o u b t e d m i l i t a r y supremacy m a r k e d the t u r n i n g of the
tide—the e n d o f M u s l i m expansion i n this direction and the
b e g i n n i n g o f Christian recovery. A discussion o f the p o i n t
w i l l t h r o w into relief some i m p o r t a n t features of U m a y y a d
Spain. I n one sense the conquest o f the peninsula was com-
plete, for a l l t h e north-west corner was i n c l u d e d i n t h e k i n g -
dom o f L e o n , a n d L e o n h a d acknowledged the suzerainty
of 'Abd-ar-Rahman. Yet the conquest was incomplete i n t w o
respects: no M u s l i m s w a n t e d to settle i n these n o r t h e r n
lands; a n d the local rulers remained vassals of a suzerain and
d i d not become heads of communities of dhimmis or "pro-
tected persons".
T h e reasons for the non-settlement o f the n o r t h e r n lands
are similar to those suggested above ( p p . 1 6 - 2 0 ) for the
lack of enthusiasm for the continued penetration of France
after the defeat o f 732, and for the failure to m a i n t a i n pres-
sure o n the north-west about the m i d d l e o f the e i g h t h cen-
t u r y . T h e Arabs almost certainly disliked the climate; and
most o f t h e m were town-dwellers, w h o f o u n d the towns of
the n o r t h small a n d lacking i n comfort. I t has been suggested
t h a t the Arabs were never h a p p y except where olive-trees
flourished. T h e Berbers w h o h a d originally settled i n parts
 UMAYYAD SPAIN AT ITS ZENITH 35

of the north-west h a d h a d unfortunate experiences before

t h e y w i t h d r e w , a n d this was doubtless remembered. T h e
hardness o f the life, coupled w i t h the hostility of the local
inhabitants, especially the mountaineers, made settlement i n
this region an unattractive proposition. W h e r e lands were
u n o c c u p i e d or sparsely populated, i t was rather the Chris-
tians w h o were prepared to undergo the rigours of resettle-
ment.
T h e treatment o f dependent p o l i t i c a l units i n a feudal-
like rather t h a n a t r a d i t i o n a l Islamic manner m i g h t be ex-
p l a i n e d b y stating t h a t the Muslims were only strong enough
t o impose a f o r m o f vassalage a n d n o t the complete Islamic
dhimmi-status; b u t i t is b y no means clear that this supposi-
t i o n is justified. I t is more l i k e l y that at m a n y points t r a d i -
tional A r a b a n d Islamic ideas were p r o v i n g less satisfactory
t h a n local Spanish ideas i n dealing w i t h the peculiar p r o b -
lems o f t h e n o r t h e r n frontier of al-Andalus. T h e conception
of the h o l y w a r or jihad, as already noted, was excellent for
1

u n i f y i n g the tribes o f A r a b i a a n d d i r e c t i n g their energies i n t o

the vast expansion o f the first Islamic century; b u t even i n
the east this conception was not suited to be the g u i d i n g
p r i n c i p l e o f a great empire i n its policies w i t h regard to its
neighbours. I n the west the problems h a d been intensified
b y the divisions a m o n g the M u s l i m s o f N o r t h Africa.
'Abd-ar-Rahman certainly made use o f the idea o f the h o l y
w a r i n s u m m o n i n g m e n to j o i n the army, b u t for most o f
his soldiers the p r i m a r y m o t i v e was p r o b a b l y materialistic
and not religious. T h e existence of strategic castles naturally
gave a special importance t o the relation o f the owner o f
such a castle t o his liege-lord; and Islamic political t r a d i t i o n ,
w h i c h was concerned rather w i t h the relation of political com-
munities to one another, h a d n o t h i n g t o say o n this p o i n t .
W h e n i t is further remembered t h a t m a n y of the leading
muwallads or Spanish M u s l i m s h a d close ties of kinship w i t h
some o f the Christian noble families, i t is n o t surprising t h a t
lords o f castles often stood i n relationships to one another
w h i c h w e r e conceived i n p u r e l y feudal terms w i t h o u t regard
for religion. I n short, there h a d been a failure to adapt the
distinctive Islamic political ideas t o a situation where, largely
for geographical reasons, castles a n d knights were the chief
36 TJMAYYAD CALIPHATE

feature. T h i s failure m i g h t conceivably be explained as due

to a lack o f religious fervour; b u t i t w o u l d be more realistic
t o a d m i t that, a l t h o u g h I s l a m is said t o be a p o l i t i c a l religion,
M u s l i m rulers after the early days have m a i n l y disregarded
religious precepts i n t h e i r conduct o f politics. There is a
certain autonomy about the w i e l d i n g o f p o l i t i c a l power, a n d
for most rulers raison d'etat outweighs a l l other considera-
tions. I t was o n l y to be expected that M u s l i m rulers i n t h e
W e s t e r n E u r o p e o f the t e n t h century should adopt the
principles a n d practices w h i c h h a d been f o u n d politically
effective there.
I n considering 'Abd-ar-Rahmans N o r t h A f r i c a n policy the
major fact t o be k e p t i n m i n d is the establishment of t h e
F a t i m i d dynasty first i n 909 i n Tunisia a n d t h e n i n 969 i n
E g y p t . F r o m one p o i n t o f v i e w this was the t r i u m p h o f the
sedentary Berbers over the nomads, whereas t h e earlier A r a b
conquest h a d been a v i c t o r y of the nomads w i t h the support
o f the Arabs over the sedentary tribes. Y e t this was n o t the
w h o l e o f the story. T h e F a t i m i d m i l i t a r y a n d p o l i t i c a l suc-
cesses were l i n k e d w i t h the p r o c l a m a t i o n o f a n o v e l set o f
religious ideas. Preached b y an enthusiastic agent (dai)
these ideas c o u l d w i n the active support o f m a n y simple
m e n w i t h the makings o f good soldiers. Theologically the
set o f ideas is described as the Isma'ilite f o r m o f S h f i s m . I t 2

declared t h a t the Islamic c o m m u n i t y h a d a n appointed leader

or imam—in the T u n i s i a n case one 'Ubayd-Allah—who was
a descendant o f M u h a m m a d and, as t r u e i m a m , was d i v i n e l y
inspired a n d supported. T h e p o l i t i c a l i m p l i c a t i o n was the
o v e r t h r o w o f existing rulers (since t h e y were n o t the r i g h t f u l
leaders o f the c o m m u n i t y o f M u s l i m s ) a n d their replacement
b y an autocratic administration under the true i m a m . I n the
particular case o f the F a t i m i d s , f r o m the t i m e t h e y began
t o r u l e i n Cairouan they took the claim t o universal sover-
eignty i n the Islamic w o r l d more seriously t h a n others w h o
h a d made similar claims before t h e m . Agents were sent i n t o
most parts o f the ' A b b a s i d empire a n d skilfully t u r n e d local
discontents into support for the F a t i m i d s .
M u c h o f the discontent w i t h w h i c h Spain was t h e n seeth-
i n g m i g h t thus have been t u r n e d to account b y the F a t i m i d s .
T h e o p p o r t u n i s m o f Ibn-Hafsun, too, h a d l e d h i m to profess
 UMAYYAD SPAIN A T ITS ZENITH 37

allegiance t o t h e Fatimids o n their attainment of power, a n d

thereby to direct their attention towards Spain. M o s t o f the
Berber settlers were from sedentary tribes a n d m i g h t be ex-
pected to welcome religious ideas similar to those favoured
b y the Berber followers o f the F a t i m i d s . A n i n c i d e n t i n the
m i d d l e M a r c h o f al-Andalus i n 9 0 1 h a d g i v e n an ominous
w a r n i n g of potentialities of this k i n d . Berber malcontents h a d
rallied r o u n d a m a n claiming t o be the M a h d i , the d i v i n e l y
g u i d e d i m a m , a n d he h a d l e d t h e m against Zamora, w h i c h
h a d recently been resettled b y Christians; there he was de-
feated b y the k i n g of L e o n a n d the m o v e m e n t d i e d away.
I n such matters race or t r a d i t i o n appears to have counted
for m u c h . Interest i n holy m e n was a n o r m a l feature o f the
religious life o f the Berbers of N o r t h Africa, b u t i t w o u l d ap-
pear t h a t Iberians were more interested i n k n o w i n g t h a t su-
p e r n a t u r a l powers n o t embodied i n particular persons were
w o r k i n g o n t h e i r behalf. A m e r i c o Castro contrasts the F r e n c h
and E n g l i s h belief t h a t the t o u c h o f a Christian k i n g c o u l d
cure scrofula w i t h the Spanish belief i n the "tangible, nearby
p o w e r " o f Santiago w h i c h gave m e n v i c t o r y i n battle—a p o w e r
t h a t was n o t incarnate i n persons, b u t m i g h t operate t h r o u g h
inanimate objects. I n so far as this was the case w i t h the
3

Spanish M u s l i m s there was no real threat to al-Andalus f r o m

F a t i m i d propaganda; b u t this w o u l d n o t be clear t o ' A b d -
ar-Rahman I I I .
T h e p a t t e r n of events i n N o r t h A f r i c a was n o t v e r y differ-
ent from t h a t o n the n o r t h e r n frontiers. T h e r e was first o f
a l l a p e r i o d of expansion w h e n several small principalities
accepted U m a y y a d suzerainty; b y 9 3 1 , after some m i l i t a r y
successes, most o f the r e g i o n f r o m Algiers to Sijilmasa rec-
ognised 'Abd-ar-Rahman as suzerain. Soon after this, h o w -
ever, his a t t e n t i o n was distracted f r o m N o r t h Africa b y the
aggressive p o l i c y o f Ramiro I I of L e o n ; a n d n o t l o n g after
the latter's death the F a t i m i d a l - M u i z z ( 9 5 3 - 7 5 ) began
t o w o r k vigorously for expansion. A f t e r an expedition i n 959
under his general J a w l i a r only T a n g i e r a n d Ceuta were left
to 'Abd-ar-Rahman. So things r e m a i n e d u n t i l towards the
e n d of the r e i g n o f a l - H a k a m I I ( 9 6 1 - 7 6 ) . A l - M u i z z decided
t o concentrate his energies o n eastward expansion. E g y p t was
conquered i n 969 a n d the seat o f government m o v e d there
38 TJMAYYAD CALIPHATE

i n 972. F r o m this t i m e o n F a t i m i d influence declined i n the

region from T u n i s i a to Morocco. Expeditions b y t h e . U m a y -
y a d general G h a l i b i n N o r t h Africa i n 973 a n d 974 recovered
some o f t h e lost g r o u n d , a n d the Umayyads retained an i m -
p o r t a n t stake i n N o r t h Africa u n t i l their central p o w e r began
t o decline.
T h e most i m p o r t a n t single event i n the i n t e r n a l history
o f al-Andalus under 'Abd-ar-Rahman I I I was l i n k e d w i t h the
threat f r o m t h e F a t i m i d s . T h i s event was the assumption
b y t h a t ruler i n 9 2 9 o f the titles of " c a l i p h " (khalifa) and
"commander o f the believers" (amir al-mumimn), together
w i t h t h e "throne-name" o f an-Nasir h - d i n - A l l a h ("defender
o f t h e r e l i g i o n of G o d " ) . I n m a k i n g this c l a i m w h a t was
asserted was n o t a universal r i g h t to rule a l l M u s l i m s b u t
the independence o f the ruler of al-Andalus o f a l l higher
M u s l i m p o l i t i c a l authority. T o support t h e c l a i m he c o u l d
p o i n t t o his descent f r o m the caliphs o f Damascus; even
before this the Spanish Umayyads h a d called themselves
"the sons o f the caliphs". T h e claim was thus n o t directed
against the 'Abbasids b u t was to counter the claim of t h e
F a t i m i d s , a n d t o give the p e t t y rulers of N o r t h Africa some
theological justification for recognising the suzerainty of the
Umayyads o f Cordova.
T h e increase o f d i g n i t y d e r i v i n g f r o m this t i t l e was also
appropriate i n v i e w o f the success of 'Abd-ar-Rahmans
policies. T h e h e i g h t e n i n g o f the separation between the ruler
a n d the subjects, however, was p r o b a b l y n o t due to the as-
s u m p t i o n o f the t i t l e o f caliph, b u t t o his general success a n d
prosperity. Sheer pressure of administration w o u l d contribute
to the change, just as i t h a d forced some o f the later U m a y -
y a d caliphs o f Damascus to t h i n k of m o d e l l i n g themselves
i n p a r t o n Persian imperialism. I t is thus n o t surprising t h a t
i n the last years of his r e i g n 'Abd-ar-Rahman I I I is said t o
have been more autocratic.
L i t t l e need be said about the r e i g n o f 'Abd-ar-Rahmans
son, a l - g a k a m I I ( 9 6 1 - 7 6 ) , w i t h the throne-name of A l -
Mustansir bi-'llah. T h e structure o f centralised p o w e r created
b y his father remained intact, a n d so b o t h i n t e r n a l l y a n d
externally the situation o f al-Andalus c o n t i n u e d to be m u c h
as i t h a d been. A n attempt b y the kingdoms of L e o n , Castile
M A P 2. Main trade routes and products in Spain in the Middle Ages
40 UMAYYAD CALIPHATE

a n d N a v a r r e to assert their independence was defeated b y

a n a r m y under G h a l i b i n 975—the same general w h o , as al-
ready noted, h a d b e g u n to restore U m a y y a d influence i n
N o r t h Africa. I n i n t e r n a l affairs the jurists appear t o have be-
come more i m p o r t a n t , b u t the w h o l e question of the jurists
is difficult a n d w i l l have t o be looked at more f u l l y later.
W h e n al-IJakam I I d i e d i n 976 the p o w er of the U m a y y a d
dynasty a n d t h e prosperity of their k i n g d o m were still at
their zenith, a n d there was l i t t l e to presage the sudden de-
cline after the year 1000.

2. The Economic Basis

A f t e r this brief description o f the more obvious events of

the reigns of 'Abd-ar-Rahman I I I a n d his son, i t is time to
look at some of the broader questions raised. A b o v e a l l , i t
has to be asked: w h a t made this p e r i o d a "great age"? W h e n
w e admire i t w h a t are w e admiring? Is i t the concentration
of p o l i t i c a l a u t h o r i t y a n d wealth? Is i t the fine buildings?
Or is i t something b e y o n d these—perhaps an upsurge of the
h u m a n spirit d e r i v i n g f r o m the prosperity a n d expressing i t -
self i n the art a n d architecture and literature? These are n o t
questions t h a t can be easily answered. I n d e e d i t is part of
the a i m o f this book t o raise such questions a n d leave t h e m
unanswered; a n d the questions just i n d i c a t e d w i l l be i n the
background for most o f the rest of our study of Islamic Spain.
A t the moment, however, w e t u r n to look at one particular
point—the material basis of the prosperity of the U m a y y a d
caliphate.
T h e agriculture o f al-Andalus was nearly a l l of the " d r y "
t y p e . T h i s was a l l that was possible i n the central plateau.
I n the south, however, i n w h a t is n o w Andalusia, irrigation
was possible. T h i s was n o t an A r a b i n v e n t i o n , b u t i t seems
to have been greatly developed b y the M u s l i m s , a n d they
m a y have i n t r o d u c e d i m p r o v e d techniques f r o m the east.
W i t h a higher level o f technical skill fresh crops became pos-
sible, a n d the M u s l i m s are credited w i t h b r i n g i n g to Spain
n o t o n l y oranges and several other species o f fruits and vege-
tables, b u t also rice, sugar-cane a n d cotton. A g r i c u l t u r a l l y
 THE ECONOMIC BASIS 41
al-Andalus was m o r e prosperous t h a n most other Islamic
lands. I t was also r i c h i n minerals, a n d t i e R o m a n processes
for o b t a i n i n g these were p r o b a b l y continued. Certainly M u s -
l i m Spain i n h e r i t e d from the Visigoths notable techniques
for artistic m e t a l - w o r k , a n d something o f this seems to have
been carried over t o the present t i m e .
I t w o u l d seem, however, t h a t the special c o n t r i b u t i o n o f
Islam m u s t be l o o k e d for rather i n t h e sphere o f urbanisa-
tion—in t h e development o f t o w n s a n d i n w h a t took place
i n t h e m . I s l a m has always been p r i m a r i l y a r e l i g i o n o f the
t o w n s m a n a n d n o t o f the peasant. I t g r e w u p i n Mecca,
w h i c h was a t h r i v i n g commercial a n d financial centre; t h o u g h
the caravans o f t h e M e c c a n merchants w e n t t h r o u g h the
deserts or steppes o f A r a b i a , the r e l i g i o n h a d l i t t l e t o d o w i t h
the desert, a n d the desert-dwellers have seldom been t h e
most zealous M u s l i m s . I t has, i f possible, even less t o do
w i t h the peasant. One i n d i c a t i o n o f this is the standard
Islamic calendar o f t w e l v e lunar months or 354 days—a cal-
endar w h i c h n o peasant r e l i g i o n w o u l d tolerate for a single
year.
T h e r e h a d been a decline i n m u n i c i p a l life i n the sixth
century u n d e r t h e Visigoths, a n d there h a d appeared a class
o f owners o f large estates, i n whose hands l a y most o f the
p o w e r a n d influence i n the country. T h e arrival o f the Arabs
w i t h t h e i r w i d e experience o f m u n i c i p a l administration i n
the east l e d t o a p a r t i a l reversal o f t h e process, i n t h a t there
was a gradual r e v i v a l o f t o w n life. T h o u g h t h e y h a d l i t t l e i n
the w a y o f democratic institutions, the Arabs seem t o have
encouraged a genuine feeling o f citizenship. O r d e r was
strictly m a i n t a i n e d . There w e r e officials w h o looked after the
markets a n d saw t h a t there w e r e no unfair practices. T h e r e
were corporations or gilds o f artisans, w i t h grades equivalent
t o master, j o u r n e y m a n a n d apprentice, a n d these w e r e care-
f u l l y regulated. T h e r e were inns t o give convenient accom-
m o d a t i o n t o t r a v e l l i n g merchants a n d their goods. T h u s there
were sound m a t e r i a l or economic reasons for the g r o w t h o f
towns; a n d the M u s l i m s w e r e b y n o means unaware o f the
opportunities offered b y towns for the c u l t i v a t i o n of litera-
ture, music a n d other artistic a n d intellectual activities.
T h e s t i m u l a t i o n of commerce resulting f r o m the general
4^ TJMAYYAD CALIPHATE

Islamic ethos m a y w e l l have been the chief source of the

prosperity o f al-Andalus. N o t merely d i d the merchants use
t h e l a n d routes o f t h e I b e r i a n peninsula a n d penetrate i n t o
France ( t h r o u g h w h i c h there was an extensive trade i n
slaves), b u t they almost certainly developed greatly the links
w h i c h t h e Visigoths already h a d w i t h N o r t h Africa. T h e pre-
cise importance o f the relationship t o N o r t h Africa, w h i c h
has been emphasised b y L6vi-Proven§al, is n o t altogether
clear, a n d is w o r t h y o f further study. I n the n i n t h a n d t e n t h
centuries the danger f r o m the N o r m a n sea-raiders led to t h e
creation o f a fleet, a n d as a repercussion o f this a merchant
m a r i n e g r e w u p , w h i c h gave direct communications w i t h T u -
nisia a n d E g y p t ; ships from these countries also shared i n
t h e traffic.
T h e u r b a n industries n a t u r a l l y catered chiefly for the needs
of t h e local p o p u l a t i o n , but, as the t o w n s m e n prospered a n d
commerce increased, there was also a market at home a n d
abroad for l u x u r y goods. T h e i n h e r i t e d skills a n d techniques
w e r e employed, a n d fresh ones elaborated. A l - A n d a l u s be-
came n o t e d for its magnificent textiles; a n d i t also p r o d u c e d
furs a n d ceramic objects.
I t is interesting t o note t h a t this general picture of M u s -
l i m economic influences i n Spain is confirmed b y a study o f
A r a b i c words i n m o d e r n Spanish. There is a large n u m b e r
of such words, b u t the i m p o r t a n t p o i n t is to notice the spheres
of life t o w h i c h t h e y belong. V e r y m a n y are connected w i t h
commerce, a n d w i t h various subsidiary aspects o f commer-
cial activity, such as travelling, w e i g h i n g a n d measuring, a n d
keeping order i n t h e markets and i n the t o w n generally; a
familiar example is aduana ( F r e n c h , douane) meaning a
customs-house, f r o m the A r a b i c diwan. A n o t h e r sphere w i t h
m a n y words is house-building; the words are mostly for parts
of t h e house or its furnishings w h i c h indicate a degree of
comfort b e y o n d the bare necessities. I r r i g a t i o n has a n u m b e r
of A r a b i c w o r d s ; a n d there is a plethora o f fruits, vegetables
a n d other species o f food. A r m y matters are i n c l u d e d , a n d
terms f r o m various industries a n d crafts. M o s t curious are a
f e w words like jarifo, "showy", w h i c h m i g h t be regarded as
proceeding f r o m a degree o f u r b a n sophistication i n the as-
sessment o f character. A n economic historian, not especially
4
 SOCIAL AND RELIGIOUS MOVEMENTS 43
of Spain, has summed the matter u p b y saying t h a t " i f the
n o r t h w a n t e d t h e best i n science, medicine, agriculture, i n -
dustry or civilised l i v i n g , i t m u s t go t o Spain t o l e a r n " .
5

F i n a l l y , t h e question must be raised—but n o t answered—

w h e t h e r there was an inevitable tension between this c o m -
merce-based turban civilisation o f southern Spain a n d the
m a i n l y a g r i c u l t u r a l a n d pastoral civilisation o f t h e n o r t h .
T h e r e was something about the economy o f the n o r t h a n d
its geographical features t h a t m a d e feudalism the best w a y
o f achieving a measure o f security, so t h a t i t was adopted
even b y t h e M u s l i m s . Islamic civilisation h a d its appropriate
m a t e r i a l basis i n t h e m i x e d economy o f Mediterranean " d r y "
f a r m i n g , i n d u s t r y a n d commerce. A r e w e justified i n seeing
the struggle between Christians a n d M u s l i m s i n Spain as
t a k i n g u p i n t o itself this tension between a p o l i t i c a l organisa-
t i o n suited t o a p r i m i t i v e economy a n d t h a t suited to a more
advanced u r b a n a n d mercantile economy? Was either c u l -
t u r a l system capable o f absorbing the other t y p e o f economy?

3. Social and Religious Movements

U n f o r t u n a t e l y there is n o t nearly as m u c h k n o w n as m i g h t
b e w i s h e d about social a n d religious movements i n a l -
Andalus. W h a t w i l l be said here, t h o u g h f o l l o w i n g the o p i n -
ions generally h e l d b y contemporary scholars, is based n o t
o n any t h o r o u g h study o f the sources b u t o n a comparatively
small n u m b e r o f facts w h i c h scholars h a p p e n to have no-
ticed. W i t h this w a r n i n g a conspectus m a y be given of t h e
different elements i n the society o f the t e n t h century.
T h e Berbers m a y be considered first, since t h e y are rela-
t i v e l y easy t o deal w i t h . T h e y seem t o have come mostly
f r o m t h e sedentary, n o t the nomadic Berbers. M o s t o f t h e m
p r o b a b l y w e n t to swell the r u r a l proletariat, t h o u g h a f e w
f o u n d their w a y to the cities a n d engaged there i n h u m b l e
crafts. One o r t w o , however, became n o t e d i n the religious
sciences. T h e y were a l l Muslims. Some o f their ancestors h a d
doubtless become M u s l i m s i n order t o share i n the A r a b con-
quests a n d the w e a l t h to be gained thereby. Those w h o
emigrated to al-Andalus after the first wave of conquest m a y
44 UMAYYAD CALIPHATE

have been attracted chiefly b y the higher standard of l i v i n g

there, a n d perhaps b y the slightly greater security. Once
settled i n al-Andalus there was need for solidarity w i t h the
other M u s l i m s , especially the Arabs, i n the face o f a certain
amount o f h o s t i l i t y f r o m the n o n - M u s l i m p o p u l a t i o n . I t is
perhaps largely for this reason t h a t t h e heretical forms o f
I s l a m p r o m i n e n t i n N o r t h Africa d i d n o t take root i n Spain.
K h a r i j i t e doctrine i n N o r t h Africa h a d become a n expression
o f t h e a n t i - A r a b feelings o f the nomadic Berbers, and was
therefore n o t attractive to the sedentary Berbers n o w i n al-
Andalus, w h o w e r e conscious o f their need for A r a b support.
S h f i t e doctrine, centred i n the conception o f the charismatic
leader or h o l y m a n , was perhaps more n a t u r a l l y attractive t o
t h e m ; b u t i n this case also the need for Arab-Berber solidarity
w o u l d h o l d t h e m back from f o l l o w i n g any leader w h o was
l i k e l y t o d i v i d e t h e m f r o m the Arabs. T h e m a n w h o claimed
to be the M a h d i i n 9 0 1 (cf. p . 3 7 ) l e d his followers against
non-Muslims a n d n o t against Arabs.
T h e Arabs, t h o u g h o n l y a small section o f the inhabitants
of al-Andalus, gave a certain colour to the w h o l e civilisation.
H e r e , however, w e come u p o n one o f the central problems
i n the history o f Islamic culture i n Spain. H o w strong was
the religious a n d c u l t u r a l influence o f the Arabs? I n w h a t
precise forms d i d i t manifest itself? H o w d i d i t come to be
so strong? I t is easy to understand i n a general w a y the basis
of A r a b influence. I n the heartlands o f I s l a m the p e r i o d o f
t h e U m a y y a d caliphate o f Damascus is regarded as one o f
A r a b d o m i n a t i o n . T h i s is i n contrast to the 'Abbasid p e r i o d
i n w h i c h Persian elements came to the fore. F r o m the con-
quest u n t i l 750 al-Andalus was a province o f the U m a y y a d
caliphate o f Damascus, a n d t h e n for another t w o a n d a half
centuries i t was r u l e d b y the U m a y y a d f a m i l y . There was
m u c h more t o i t , however, t h a n the mere fact of rule. T h e
Arabs w e r e characterised b y intense self-confidence or belief
i n themselves; a n d this, coupled w i t h t h e i r superior eco-
n o m i c position—they h a d lands i n the richest parts of the
country—must have l e d t h e other sections o f the p o p u l a t i o n
t o admire and emulate t h e m . A t first non-Arabs w h o became
Muslims were clients o f members of A r a b tribes, a n d often
adopted the patron's genealogy; i n course of t i m e the p o i n t
 SOCIAL AND RELIGIOUS MOVEMENTS 45

at w h i c h the genealogy became fictitious was forgotten, and

the clients began to pass themselves off as pure Arabs.
T h i s " a r a b i s i n g " o f Spain appears at m a n y points—the
name "Mozarabs" ( m e a n i n g "arabisers") for the Christian
l i v i n g under M u s l i m rule, the interest i n A r a b genealogy
a n d other questions o f A r a b origins, the dominance of the
distinctively A r a b i a n M a l i k i t e legal rite, a n d above a l l the
p o p u l a r i t y o f the language. I t is not clear t h a t the original
A r a b invaders a n d settlers b r o u g h t m u c h culture w i t h t h e m .
W h a t was p r o b a b l y more i m p o r t a n t was that they remained
i n contact w i t h the Arabic-speaking lands i n the M i d d l e East,
a n d w e r e therefore able to benefit f r o m the cultural advances
made there.
M u c h m o r e numerous t h a n the Arabs i n the strict sense
were the muwallads or M u s l i m s of I b e r i a n stock. I n course
of t i m e m a n y o f t h e m adopted p u r e l y A r a b genealogies;
the w r i t e r I b n - I J a z m ( p . 111) actually claimed descent from
a n arabised Persian. T h e confusion b r o u g h t about b y such
genealogies was n o t serious f r o m any practical p o i n t of v i e w .
Descent was reckoned only i n the male line, b u t the original
Arabs h a d m a r r i e d I b e r i a n w o m e n freely, so t h a t b y the t e n t h
c e n t u r y there was no clear racial distinction between Arabs
a n d muwallads. I t is therefore n o t surprising that the t w o
groups became more a n d more fused into one.
L i t t l e is k n o w n i n detail about the reasons for the accept-
ance o f the Islamic religion b y so m a n y inhabitants o f the
I b e r i a n peninsula. T h e existing state o f affairs m a y have h a d
something t o do w i t h i t . There was a h i g h degree of co-
6

operation—or should w e say "collusion"?—between the Visi-

gothic r u l i n g group a n d the ecclesiastical authorities, w h i c h
made life v e r y uncomfortable for those w h o for any reason,
material or theological, d i d not accept a l l the rulings o f the
ecclesiastics. A m o n g t h e latter m a y be reckoned m a n y slaves
a n d the r e m a i n i n g pagans; b u t i t is also possible that for
some the traces o f the A r i a n heresy so l o n g professed b y the
Goths (according t o w h i c h Jesus was little more t h a n a m a n )
eased the p a t h to conversion t o Islam o n the intellectual side.
F o r the Christian nobles and the numerous members of the
lower and m i d d l e classes of town-dwellers there was doubt-
46 UMAYYAD CALIPHATE

less a m i x t u r e of material and religious motives. A m o n g these

motives a p r o m i n e n t place w o u l d be h e l d b y desire for the
social advantages of b e i n g a M u s l i m a n d admiration for the
culture associated w i t h Islam.
Yet, w h e n allowance has been m a d e for a l l available facts,
there remains a puzzle about the acceptance o f Islamic c u l -
ture b y so m a n y of the people of Spain. O n the one h a n d , a
century before the conquest o f Spain b y the Arabs their
ancestors h a d been l i v i n g a very r o u g h life i n the steppes of
Arabia, and t h e actual invaders h a d h a d l i t t l e t i m e t o acquire
a h i g h level o f culture. O n the other h a n d , a t r a d i t i o n of
learning h a d been established at Seville b y Isidore ( d . 6 3 6 ) ,
w h i c h h a d made t h a t city one o f the leading intellectual
centres of Christian Europe. Yet this I s i d o r i a n t r a d i t i o n was
abandoned for t h a t of the Arabs a n d A r a b i c literature. H o w
is i t to be explained? D i d the association o f the ecclesiastical
scholars w i t h the rulers cut t h e m off f r o m the c o m m o n peo-
ple? Or was I s i d o r i a n culture always restricted to a small
handful of men? O r was there some other factor of w h i c h
w e are not f u l l y aware?
I t seems clear that the muwallads h a d n o t h i n g distinctively
I b e r i a n i n w h i c h they took p r i d e . T h e great H u n g a r i a n
Islamist, Ignaz Goldziher, after s t u d y i n g the Shu ubite move-
m e n t i n I r a q a n d Persia, examined materials about Spain t o
discover traces of i t there. I n the east this was a literary
movement w h i c h attacked the supposed superiority of the
Arabs and v a u n t e d instead the glories of the I r a n i a n peoples.
A l l that Goldziher was able to find was a couple of judges
of Huesca at the beginning of the t e n t h century w h o "pas-
sionately u p h e l d the cause of the muwallads", a n d a literary
epistle of the m i d d l e of the eleventh century w h i c h repeated
arguments of eastern Shu'ubite w r i t e r s . F r o m this the con-
7

clusion is that, t h o u g h the muwallads were sometimes an-

noyed at the A r a b assumption o f superiority, they h a d n o t h -
i n g positive o f their o w n to oppose to i t .
T h e next i m p o r t a n t group, the Mozarabs or Christians
l i v i n g under M u s l i m rule, shows this same fascination of
things A r a b . I n 854 a Christian w r i t e r described the situation
thus:
 SOCIAL AND RELIGIOUS MOVEMENTS 47
Our Christian young men, with their elegant airs
and fluent speech, are showy in their dress and car-
riage, and are famed for the learning of the gentiles;
intoxicated with Arab eloquence they greedily han-
dle, eagerly devour and zealously discuss the books
of the Chaldeans (i.e. Muhammadans), and make
them known by praising them with every flourish of
rhetoric, knowing nothing of the beauty of the
Church's literature, and looking down with contempt
on the streams of the Church that flow forth from
Paradise; alas I the Christians are so ignorant of their
own law, the Latins pay so little attention to their
own language, that in the whole Christian flock there
is hardly one man in a thousand who can write a
letter to inquire after a friend's health intelligibly,
while you may find a countless rabble of all kinds
of them who can learnedly roll out the grandiloquent
periods of the Chaldean tongue. They can even make
poems, every line ending with the same letter, which
display high flights of beauty and more skill in
handling metre than the gentiles themselves possess. 8

This passage makes i t clear to what an extent the Chris-

tians of al-Andalus, even while remaining Christians, were
carried away by admiration for Arab civilisation. I t should,
of course, be conceded that this description applies chiefly
to town-dwellers, and that a high percentage of Mozarabs
may not have been town-dwellers. The fact that the interest
is above all i n the language and in poetry argues that up to
the middle of the ninth century the distinctively Arab ele-
ments had been most prominent in al-Andalus, or at least had
appealed most to the Iberian population. Despite the accept-
ance of Arab civilisation, however, the Mozarabs were not
wholly satisfied. They supported Muwallad risings, like that
of Ibn-JJafsun; and from the latter part of the ninth century
many of them emigrated from al-Andalus to the Christian
kingdoms. Like the Muwallads they commonly spoke a ro-
mance dialect, though the more educated were able to write
i n Arabic as well as to speak it.
Other groups i n the country were the Jews and the slaves.
The Jews stood somewhat apart from the general life of the
country, though they came to have a share i n its intellectual
48 UMAYYAD CALIPHATE

life. Among the slaves and ex-slaves a distinction was made

between the negroes and the "Slavs" (Saqaliba). The latter
included not merely persons of Slav race but also Franks and
other slaves from the north. While the influx doubtless began
earlier, i t was especially under 'Abd-ar-Rahman I I I that large
numbers were imported for the army and for the service of
the palace. Some, but by no means all, were eunuchs. A
few rose to positions of power and influence. Most were even-
tually freed and usually settled i n the towns, so that by the
eleventh century they constituted an important element i n
the population, w i t h a place i n politics out of proportion to
their numbers. I t would appear that they normally became
Muslims. Christians also, however, seem to have had slaves,
and these would presumably be Christian.
These, then, were the main groups i n the population of
al-Andalus. The general social and religious trends can be
roughly discerned, and one thus gains some idea of the prob-
lems of creating and maintaining unity i n the Umayyad state.

4. The Ruling Institution

The Umayyad state of al-Andalus was first and foremost an

autocracy. Everything was, at least theoretically, i n the
hands of the emir or caliph, though according to his inclina-
tions he might delegate much of the detail of administration
or even, as i n the case of al-Hakam I I , much of the general
control of policy. I n the hands of the head of state was the
responsibility for internal and external affairs, and the su-
preme command of the army. His also was the power of life
and death, solely at his will. Various marks of dignity were
gradually adopted setting him off from his subjects and mak-
ing access to him difficult. This was especially the case after
the assumption of the caliphate i n 9 2 9 . Up to that date, for
example, the preacher at the Friday midday worship (or
prayers) had invoked God's blessing on the 'Abbasid caliph
in Baghdad as the rightful head of the community of Mus-
lims, although politically he was not recognised i n any way;
but from 929 i t was the name of 'Abd-ar-Rahman I I I an-
Nasir which was mentioned instead.
 THE RULING INSTITUTION 49
There was usually a chief minister with the title of hdjib
or "chamberlain". His office was much the same as that of
the vizier (wazir) i n the east; but i n al-Andalus the vizierate
was a minor dignity conferred on several subordinate minis-
ters. These men were i n charge of a large central secretariat,
housed i n the palace of the Umayyads at Cordova, and
mostly continuing there when the caliph and his court, about
the middle of the tenth century, moved to the new palace-
city of Madmat az-Zahra, some three miles out of Cordova.
There was a small replica of this central administration i n
each of the twenty-one or more provinces (kuwar, sing.
hard) into which al-Andalus (apart from the Marches) was
divided. Each province was under a governor or wall. The
non-Muslims had a certain measure of autonomy. They were
organised i n groups or communities i n the various provinces,
and at the head of each group, and responsible for its poll-
tax (jizya), was a comes or count (Ar. qumis). They also
had their own judges. 9

The system of three Marches for the defence of the north-

ern frontier has already been described (p. 2 8 ) . Little more
need be said about the armed forces. Mercenaries were first
used by al-IJakam I ( 7 9 6 - 8 2 2 ) , and the numbers gradually
increased. Many were Franks and Slavs, originally slaves;
and latterly there were also Berbers from North Africa and
negroes. Certain of the mercenaries constituted the personal
guard of the ruler. Apart from the mercenaries a considerable
number of citizens were liable to military service, the former
jundis (who had been given fiefs i n return for military serv-
ice) and townsmen. A third section was that of the volunteers
for the holy war, whom the ruler called for specially when
he was about to make an expedition against the Christian
kingdoms.
Though the jurists are not part of the administration i n
the strict sense, they belong i n a fashion to the ruling i n -
stitution. I n the east the jurists had acted as the leaders of
a sort of "constitutionalist" party within the caliphate, and
had insisted that i n many spheres government should be i n
accordance w i t h the Shana or revealed law. I n this way they
had created a counterweight to the autocratic tendencies of
rulers and given ordinary men a measure of security; there
5o UMAYYAD CALIPHATE

remained spheres, however, such as the relations of the ca-

liph to his courtiers, where the principles of the Sharfa were
not applied. I n certain respects, too, the jurists had come to
be dominated by the ruling institution, since i t was responsi-
ble for making the appointments to the best positions that
were open to jurists. This led to widespread worldliness
among the jurists and other scholars. 10

How far this situation, which was found at the centre of

the 'Abbasid caliphate, was paralleled i n al-Andalus is a ques-
tion which has not been adequately studied. The dominant
legal rite i n al-Andalus was the Malikite (as w i l l be seen i n
the next chapter), and the influence of the jurists seems to
have grown steadily. They are said to have had more influ-
ence under al-ljakam n (961-76) than under his father,
*Abd-ar-Rahman I I I . The general impression given by a rapid
survey of events is that the jurists of al-Andalus of the tenth
century had less influence than those of Iraq. I t likewise
seems doubtful whether (as is sometimes asserted) the
Umayyads deliberately used the Islamic religion to bring
unity and harmony to their heterogeneous realm. Various
points have been noted in previous chapters which suggest
that up to the later tenth century, at least, the Arabic ele-
ment i n the culture of the Muslims was more influential than
the Islamic.
Nevertheless there are some reasons for thinking that Is-
lamic influence was growing. Al-Andalus remained i n touch
w i t h the heartlands of the Islamic world, and the next chap-
ter w i l l show what this meant i n the intellectual and literary
spheres. The Muslims clearly thought of Baghdad as setting
the standard by which everything must be measured. A l -
though the claims of the 'Abbasid caliphs to political suze-
rainty were rejected, we find items of the court ceremonial
of Baghdad being adopted i n Cordova i n the later tenth cen-
tury, and perhaps even something of the form of administra-
tion practised by the 'Abbasids. I t would be wrong, however,
to conclude that this imitation of Baghdad was due to sheer
admiration; i t might have been much more a concern for
efficient administration. The administration of al-Andalus, i t
must be remembered, developed from the forms of the later
Umayyad caliphate of Damascus, where the weaknesses of
 THE RULING INSTITUTION 51
the old Arab system were being realised and interest was
being shown i n Persian methods. As time went on and the
problems facing the rulers of al-Andalus became more seri-
ous, they may well have been impressed by the practical
advantages of the Baghdad system; information about i t
could have been obtained directly from Baghdad or indi-
rectly from Cairouan, which, prior to its conquest by the
Fatimids i n 909, was the capital of a state partly dependent
on Baghdad.
There is practically nothing i n the organisation of the rul-
ing institution that is derived from the Visigothic tradition.
The former close association of the ecclesiastical hierarchy
with the rulers may have facilitated the growth of the in-
fluence of the jurists i n the Islamic state, but i t was not the
source of this tendency. Apart from this the chief point ap-
pears to be, as noted above, the adoption of quasi-feudalistic
ideas by the emirs and caliphs i n their dealings with the
Christian kingdoms in the north.
 5
CULTURAL ACHIEVEMENTS
U N D E R T H E UMAYYADS

l. Intellectual Life—The Religious Sciences

The focal point of Muslim intellectual activity has always

been law or jurisprudence. Yet a statement of this kind tends
to be misleading to the average reader, since the Islamic con-
ception of law differs in various ways from every other. The
word usually translated "Islamic law" is sharia, but the basic
meaning of this is "what is revealed". So the Sharfa i n the
modern sense is not to be compared with any code of posi-
tive law. There are a few definite enactments i n the Qur'an,
and these are taken up into the Sharfa, but this was far from
sufficing even the needs of the community of Medina during
the lifetime of Muhammad, still less the needs of a great
empire. So Muhammad's practice and the continuation of
this practice by his successors i n office was taken into con-
sideration also. The curious thing about the whole develop-
ment is that, besides those who were responsible for the
administration of justice, many men became interested i n
discussing questions of jurisprudence from a rather theoreti-
cal point of view. The root of their interest seems to have
been the desire to ensure that the Islamic community, as i t
was founded on a "revealed law", should remain entirely
faithful to that law. 1

This theoretical or religious concern of the jurists (if this

term may be applied prematurely to these early thinkers)
was pursued by them without any direct reference to the rul-
ing institution. Sometimes men from their ranks might be
judges i n the service of the institution; sometimes certain
jurists might feel very critical of the actions of the institution
and consider that i t had departed from the "revealed law".
This was particularly so under the Umayyad caliphate of
Damascus (up to 750); the 'Abbasids tended to defer to the
 THE RELIGIOUS SCIENCES 53
opinions of the jurists, at least ostensibly. This bifurcation be-
tween the jurists and the actual rulers meant that the con-
clusions of the jurists were not automatically put into prac-
tice, but only i n so far as the rulers decided that they were
to be the basis of action. The jurists indeed discussed much
more than legal matters i n the modern sense, for they i n -
cluded what we would call etiquette and liturgical forms.
The "revealed law" was a whole "revealed way of life".
At first the discussions were chiefly at the level of the ethi-
cal conceptions implicit i n the actual practice of the com-
munity, and i t was assumed that this practice was being con-
tinued unchanged. By about the beginning of the second
Islamic century (about 7 2 0 A . D . ) , however, i t was realised
that changes were beginning to creep i n , and that i n various
regions of the Islamic world there were different versions of
what Muhammad's practice had i n fact been. From this time
the activity of the jurists had two aspects. Firstly, they had
to decide whether, i n any given circumstances, the act was
i n accordance w i t h the "revealed law". Secondly, they had to
formulate the basic conceptions or "roots" of law i n such a
way as to justify i n self-consistent fashion all their particular
decisions. I t came to be universally accepted that the "re-
vealed law" was expressed not merely i n the Qur'an but i n
the regular practice—the "beaten path" or sunna—oi M u -
hammad. I t also came to be accepted that Muhammad's
practice was known only through duly authenticated Tradi-
tions ( i n the technical sense, marked here by a capital, of
anecdotes about Muhammad). Most jurists further held that
particular prescriptions could be derived from the Quran and
the Traditions by various rational procedures (such as analog-
ical argument), but there was considerable discussion about
which procedures were permissible. Yet a fourth "root" was
sometimes recognized, the consensus of the community
(ijma).
Between about 800 and 900 the main trends of thought
on legal matters hardened into schools or rather rites—the
latter word is preferable when referring to differences i n
practice rather than i n theory. Some of these rites, such as
the Zahirite which had a notable exponent i n Spain, died out
after a time. Among the Sunnites, or main body of Muslims,
54 UMAYYAD CULTURAL ACHIEVEMENTS

four rites came to be recognised as permissible variants—the

Hanafite, the Malikite, the ShafTite and the IJanbalite. So
far as al-Andalus is concerned the only one of these which
is important is the Malikite, which derives its name from
Malik ibn-Anas ( d . 7 9 5 ) , who belonged to the school of
Medina. I t is usually stated that at first the Muslims of Spain
followed the teaching of a Syrian jurist, al-Awza% and later,
about 800, went over officially to the school of Malik. This
is roughly true, but some refinements are necessary.
The administration of justice i n al-Andalus was formally
in the hands of the ruler, whether governor appointed from
Damascus, independent emir, or caliph; but he usually dele-
gated responsibility to special persons for this function. A t
first these were politicians rather than jurists. Even under
the emir 'Abd-ar-Rahman I there was no class of jurists,
though there were a few men who had studied jurisprudence
in the heartlands of the Islamic world. One such was Sa'sa'a
(d. 796 or 8 0 7 ) , who studied i n Syria under al-Awzai (d.
7 7 3 ) and others, presumably before the fall of the Umay-
yads there i n 750. Since al-Awzai was the leading jurist of
the province i n which the capital lay i t was natural that his
views should be followed in the distant province of al-
Andalus, and should continue to be followed when it became
an independent state under members of the Umayyad family.
After 750 al-Awzai, settled i n Beirut and outwardly,
though perhaps not inwardly, reconciled to the 'Abbasid
regime, was out of touch with the leading politicians and
much less influential than he had been. Students still seem
to have come from al-Andalus to hear his lectures, but they
also went to Medina and other places. I n Medina the views
2

on the principles of jurisprudence propounded by Malik and

several other teachers were similar to those of al-Awzai, but
slightly further developed; both were simple and primitive
3

compared w i t h the thinking of the Shafutes and Hanafites

in Iraq. I n this situation, i t probably did not much matter
whether the jurists of al-Andalus professed to follow al-
Awza'i or Malik, since in neither case would their views be
regarded as absolutely authoritative by the rulers.
A real change seems to have come about when two young
jurists from Cairouan, one of whom had studied i n Iraq, sys-
 THE RELIGIOUS SCIENCES 55
tematically arranged a large number of questions on particu-
lar points of law, and recorded the answers given to them i n
Cairo by one of Malik's pupils. The books of these two jurists,
containing the questions and answers, thus constituted a cod-
ification of the Sharfa on Malikite principles, and would be
very useful for practical purposes. The earlier of the two
books was apparently introduced into al-Andalus about 800
by Isa ibn-Dinar (d. 8 2 7 ) and Yahya ibn-Yahya al-Laythi
(d. 8 4 7 ) , the latter being a Berber. The emir al-Hakam I
(796-822) apparently gave some official recognition to this
codification of the Malikite rite, so that from this time on-
wards i t was the official rite i n Spain. I t was widely taught,
and the Malikite jurists came to be a cohesive group. There
4

was apparently little further discussion of the general prin-

ciples of jurisprudence at this period, but there was some in-
tellectual activity concerned with elaborating particular pre-
scriptions and applying them to Spanish conditions. Isa
ibn-Dinar was the author of a work i n twelve volumes. The
most notable production however, was by a slightly later jur-
ist, al-'Utbl ( d . 8 6 9 ) , and was apparently supplementary to
the early works of codification.
The establishment of a body of Malikite jurists may thus
be regarded as the chief intellectual activity i n the sphere of
religion i n al-Andalus under the Umayyads. The aims of these
jurists were practical rather than theoretical, and they
worked i n close association with the Malikite jurists of Cai-
rouan and other parts of North Africa. That Spain and North
Africa should have attached themselves to the Malikite rite
in preference to all others is not entirely fortuitous. I n Iraq
where the IJanafite and Shafilte rites were developed many
of the Muslims belonged to the pre-Islamic population of
the region and, before their conversion, had been under the
influence of Hellenistic culture. I n North Africa and the Ibe-
rian peninsula, on the other hand, the Arabs were the bearers
of the dominant intellectual culture; the Berbers had had
little of their own, and the Iberians, for reasons that are not
altogether clear, preferred that of the Arabs to the Latin
culture of the Isidorian revival. Because of the essentially
Arab outlook of these regions, without any admixture of the
speculative interest of Hellenism, i t was only natural that
56 UMAYYAD CULTURAL ACHIEVEMENTS

the simple and essentially Arab Malikite rite should have

most appeal, and that within Malikism the eminently practi-
cal form worked out i n Cairouan should be favoured. I t is
also possible that i n peripheral regions like these—one thinks
of the comparison w i t h the form of British culture in, say,
Canada or Australia—there was a tendency to cling to ortho-
doxy; but i t is difficult to say what could be meant by "or-
thodoxy" here, and, whatever meaning is given to it, it would
seem that the desire for something simple and practical was
more influential.
I t is commonly stated that the other legal rites had no
followers i n Spain, but this is not quite exact. I t may be that
they were not officially recognised nor provided for in the
courts, but on the other hand Ibn-Hazm (d. 1064) appears
to have received Shafiite teaching i n al-Andalus. The most
notable of those who are alleged to have held Shafiite views
was Baqi ibn-Makhlad (d. 8 8 9 ) . Like most of the leading
jurists and scholars of al-Andalus at this time he had studied
i n the heartlands of the Islamic world, but unlike most of
the others he had become interested i n the speculative as-
pects of jurisprudence and the derivation of legal prescrip-
tions from Traditions. On his return, whether because of his
general ideas or because he taught Traditions, he incurred
the hostility of the Malikites, but was able to remain active
in Cordova through the protection of the emir Muhammad
I ( 8 5 2 - 8 6 ) . Certainly the study of Traditions, however
frowned upon by the strict Malikites, began to take root i n
Spain.
Nearly a century later one of the leading figures in legal
circles i n al-Andalus had the reputation of being a Zahirite
—an adherent of the rite to which the later Ibn-Hazm be-
longed, but which eventually disappeared. This was al-
Mundhir ibn-Sald al-Balluti, who was the chief judge or
qadi of Cordova from 950 until his death i n 966 at the age
of eighty-two. I n the cases of al-Mundhir and Baqi the ideas
objected to by the Malikites must have been regarded as pri-
vate opinions on secondary matters, and cannot have been
propagated widely. The support that they received from the
rulers is a point to be noticed, for among other things i t sug-
 THE RELIGIOUS SCIENCES 57
gests that the rulers up to this period were not prepared to
allow the Malikites a monopoly.
There are slight traces i n al-Andalus of some of the other
intellectual currents found in the heartlands. I n the works of
biography, for example, i t is recorded of one or two men that
they held certain doctrines usually associated w i t h the
Mu'tazilites, a sect of semi-philosophical theologians promi-
nent in Iraq i n the first half of the ninth century. The bellet-
5

trist al-Jahiz ( d . 8 6 8 ) , who was greatly admired and read

in al-Andalus, was a Mutazilite, and admiration for him may
have contributed to the acceptance of Mu'tazilite ideas. As
a whole, however, Mu'tazilism did not take root i n al-Andalus.
More deeply influenced by Greek philosophy was Ibn-
Masarra (d. 9 3 1 ) , whose views, though not well known, ap-
pear to have contained Empedoclean elements. The opposi-
tion to him from the Malikites of Cordova was such that he
retired to a hermitage in the neighbouring Sierra, where he
was able to instruct and "form" a small number of disciples,
and perhaps to lay the foundations of Andalusian mysticism. 6

I n general, then, i t can be said that up to the end of the

tenth century the best developed field of study was the
Malikite doctrine of the "branches" or detailed legal pre-
scriptions. Of the other "religious sciences"—it is convenient
to use this translation for the Arabic word 'ulum, "knowl-
edges", which is also used for the natural sciences—a begin-
ning had been made w i t h the study of Traditions and w i t h
the exegesis (tafstr) of the Qur'an. Some individuals held
views on the "roots" of law and on matters of theology
(kaldm), but these can hardly be said to have existed as
disciplines. (Grammar and lexicography, which were needed
for Qur'anic exegesis, were studied, but they w i l l be dealt
w i t h i n the next section.) Of the "foreign sciences" (that is,
Greek thought), philosophy was not cultivated apart from
that of Ibn-Masarra, but from about 9 5 0 there was some ad-
vanced study of medicine, while al-IJakam I I ( 9 6 1 - 7 6 ) en-
couraged astronomy and mathematics.
Apart from the books on Malikite law, important works i n
the field of history and biography were also produced. Is-
lamic historiography had its roots partly i n the northern
Arabs' interest i n genealogy and the heroic achievements of
58 UMAYYAD CULTURAL ACHIEVEMENTS

their tribes, and partly i n Iranian (and to a lesser extent

Christian) historiographical tradition. I t may be said to have
come of age i n the heartlands by about 900. That the culture
of al-Andalus was still part of general Islamic culture is shown
by the fact that a native of Spain, 'Arib ( d . c. 9 8 0 ) , achieved
fame as the continuator for the Hijra years 2 9 1 to 320 ( A . D .
9 0 4 - 3 2 ) of the history of at-Tabari, the greatest of early
Arabic histories. Apart from this, most of the historical and
biographical writing i n Spain was at first on matters of local
interest. The first person who can claim to be called a his-
torian is Ahmad ar-Razi ( d . 9 5 3 ) , whose work is the basis
of the Spanish document known as the Cronica del Mora
Rasis. About the same time biographical works, now lost,
were compiled about the scholars of Cordova and other
towns. There has been preserved, however, a History of the
Judges of Cordova, written by a scholar from Cairouan who
settled there, by name al-Khushani ( d . 9 7 1 or 9 8 1 ) . These
7

scanty details are sufficient to show how, w i t h the access of

wealth and power under 'Abd-ar-Rahman I I I , the Muslims
of al-Andalus became aware of themselves as a distinctive
unit within the Islamic community, and thereby grew in self-
confidence.

2. Intellectual Life—Poetry and Belles-Lettres

When—with the advent of Islam and the conquest of an Em-

pire—the Arabs first strode on to a world stage, they already
had fine lines to declaim. Their poets were no fumbling be-
ginners, responding to the stimulus of a new-found greatness,
but had a treasury of resonant odes. Each of these followed
a fixed form; they usually started with an erotic prelude, then
moved through a succession of conventional themes—descrip-
tion of camels or horses, of hunting scenes and battles—to
culminate in the praise of some noble chieftain or valiant
tribe. Conventions and cliches abounded: at the outset, the
poet was assumed to be travelling through the desert w i t h
only one or two companions, and to have come across barely
discernible traces of an encampment "like tattoo-marks amid
the arteries of the wrist", i n which he recognised a site on
 POETRY AND BELLES-LETTRES 59
which once dwelt a woman he had loved. The odes also had
elaborate metres, and a single rhyme throughout. Everything
in them bespoke a well-established tradition, a tradition
which has seldom been challenged until modern times.
True, the life of the conquering Muslims was no longer
that of the desert nomads. Indeed, as the Arabs were de-
rided by their more sophisticated subjects as "camel-drivers
and lizard-eaters", they were not slow to adopt Persian man-
ners or to study Greek thought; but they were—as is natural
—slower to absorb aesthetic values and they showed very
little interest i n Greek literature, steeped as i t was i n pagan
legend. On the contrary, i n the battle of words that raged
between Arabs and non-Arabs i n the early Muslim empire,
the twin glories of which the Arabs could boast without fear
of being gainsaid by detractors who had accepted Islam were
the revelation specially vouchsafed to their race, and the lan-
guage i n which this revelation was couched and which was
essential to its understanding. W i t h this language, pre-Islamic
poetry was indissolubly linked. The conservatism inherent i n
the pre-Islamic tradition, born of the unchanging life of the
desert, was reinforced as i t became a bastion of racial pride.
Changes i n literary practice did occur, as was inevitable.
Subsidiary themes of the ode were developed into independ-
ent love-songs and wine-songs. The descriptive powers of the
desert poet who recorded with a strangely impersonal, photo-
graphic eye alike the beauty spot on a young woman's face
and the droppings of a gazelle upon the sands were enriched
by an emotional response to a more bountiful nature, and
extended to castles and ships and other man-made wonders.
Indeed the latter part of the eighth century—coinciding with
the establishment of the Umayyad emirate i n Spain—was i n
the East a period of bold innovation when Abu-Nuwas (d.
803) openly ridiculed and parodied the traditional ode and
Abu-l-'Atahiya (d. 8 2 3 ) brought the language of the market
place into court. Yet when all this effervescence had sub-
sided, what remained to slake the thirst of countless genera-
tions of Arabic-speaking intellectuals was the traditional bev-
erage w i t h but one strong new flavour added: rhetorical
artifice. The supremacy of the Ancients was admitted, and
the ideal formation of a poet was taken to be acquaintance
60 UMAYYAD CULTURAL ACHIEVEMENTS

w i t h and emulation of their odes. The structure and metres

of poems were traditional. Even the themes were mostly lim-
ited to conventional ones. What was expected of the living
poet was increasing refinement of expression within the con-
ventional framework. Some—like al-Mutanabbi (d. 9 6 5 ) i n
the East—were possessed of such genius that i t could burst
through these conventions without flouting them, while their
critics somewhat lamely tried to explain their greatness i n
terms of mastery of these very conventions. For most poets,
however, the task was one of refurbishing old motifs w i t h i n -
creasingly subtle variations, overlaying them w i t h recherche
similes and daring hyperboles, and the like. I n the words of
A. J. Arberry;

Precisely as Saracenic art and architecture, denied

the inspiration of the human body, tended inevitably
towards the elaboration of arabesque ornament, the
infinite subtle variation of geometrical design; so in
Arabic poetry the business of the creative craftsman
was to invent patterns of thought and sound within
the framework of his revered tradition. Poetry be-
came an arabesque of words and meanings. 8

Evidently this was a poetry for connoisseurs. The com-

mon people, even as they evolved regional dialects different
from the classical language, no doubt regaled themselves
w i t h folk-songs and folk-tales of which we have samples i n
the Arabian Nights; but these the intellectuals scarcely dig-
nified w i t h the name of literature. That which alone was
deemed worthy of studying and recording was a poetry
which had its outlet principally at the courts of princes,
princelings and other imitators. A poet's fortunes depended
on securing the favour of a wealthy and powerful patron
who would either reward him for a particular panegyric or
attach him to his service. I f he possessed other qualities as
well, his poetic attainments could open for him the door to
high office; but i n itself his position was a subservient one.
He addressed himself primarily not to the inarticulate masses
who needed to have their experience expressed, extended or
refined, but to a cultured elite well versed i n the rules of his
 POETRY AND BELLES-LETTRES 6l
craft and prepared to delight in, or pass judgement on, his
virtuosity.
Thus conditioned by the standards of an 61ite that valued
literary tradition and refinement of expression above all else,
Arabic poetry is remarkable for its continuity from genera-
tion to generation, and its homogeneity even under different
climes. I t is i n incidental details and subtle overtones that
reflections of local conditions and cultural changes are to be
sought.
Essentially, the Arabic poetry produced i n al-Andalus was
an offshoot of that produced i n the East; and i n the period
now under review, i t was little more than that. 'Abd-ar-
Rahman I had a country-seat identical w i t h his uncle's near
Damascus, like i t called ar-Rusafa, and to him are ascribed
a few lines of poetry which begin:

Emerging before us in the midst of ar-Rusafa is a

palm-tree
Far, in this Western land, from the home of palm-
trees;
I said: Like me is it in my expatriation and my
yearning
And in the length of my absence from my children
and my kin . . .

Indifferent as the lines are in quality, they show clearly

enough where the spiritual home of Spain's earliest poets lay.
For 'Abd-ar-Rahman and other newcomers to al-Andalus
already had an intellectual formation of which they had
cause to be proud, and the Visigoths whom they displaced
had no culture so obvious i n value as to call for immediate
integration into the conquerors'. Indeed a ninth-century
Christian writer had cause to complain that fellow-Christians
were so taken w i t h the Arabic language and its literature as
to neglect, and even express contempt for, Latin texts. For 9

those born and bred to Arabic there was not even a need to
choose, and the massive fact about Umayyad Spain is that
for all the political rivalry that existed between i t and the
East, i t was to the East that i t looked for cultural guidance.
Not only had Andalusian literature grown out of a cutting
62 UMAYYAD CULTURAL ACHIEVEMENTS

from the East: i t was constantly reinforced and modified by

grafts from the East. From Baghdad came the singer Ziryab
(d. 8 5 2 ) , disciple and then rival of IsTiaq al-Mawsili, w i t h
his trained children and slave-girls, to found the Andalusian
school of music and song and incidentally to teach the man-
ners of Baghdad society. From Baghdad also came the noted
scholar Abu-'Ali al-Qali ( d . 8 6 7 ) , to be received with great
honour and to dictate his voluminous Amali. These were dis-
connected discourses of primarily lexicographical or gram-
matical interest, but studded, as is the Arab tradition, w i t h
poetic quotations often conveying, i f only by implication, lit-
erary preferences; they were to have incalculable influence
on succeeding generations of Andalusian intellectuals. W i t h
surprisingly little lag i n time, the compositions of the greatest
poets of the East were being studied and imitated i n al-
Andalus. Indeed, although the older poetry was studied, i t
was the new taste for rhetorical embellishments that was
most strongly reflected i n the production of Andalusian poets.
Not surprisingly, al-Andalus did not immediately produce
poets of great distinction. There are not a few poets of the
late eighth century and of the ninth whose names are re-
corded and samples of whose compositions have been pre-
served; but they are no more than competent. Indeed many
of them were princes of the Umayyad house who owed to
their rank and station the attention they have received from
literary historians and anthologists. I t was not until late in the
Umayyad period—when Cordova had become a centre of
learning and the court offered worthy patronage to men of
talent and of learning—that al-Andalus produced two men of
letters of lasting reputation.
Ibn-'Abd-Rabbih ( 8 6 0 - 9 4 0 ) , a poet whose amatory verses
are not without charm, is famous above all for his literary
thesaurus Al-Iqd al-Fand ("The Peerless Necklace"), which
proved immensely popular for many centuries both i n the
East and i n the West. He took for his model the work com-
piled i n the East by Ibn-Qutayba (d. 8 8 9 ) and entitled The
Fountains of Story, and drew his material mostly from the
East; i n fact, he quoted no Andalusian compositions other
than his own. Ibn-'Abd-Rabbih was also the author of a 450-
line urjuza—a, type of metrical composition less demanding
 POETRY AND BELLES-LETTRES 63
than the classical ode, the hemistiches usually being treated
as independent lines and rhymed i n couplets—on the warlike
exploits of 'Abd-ar-Rahman I I I ; i t deserves mention because
narrative poems are rare i n Arabic i n any age.
Undoubtedly the most accomplished Andalusian poet of
his time was Ibn-Hanf (d. 9 7 3 ) , but accusations of heresy
forced him to leave Spain when he was about twenty-seven
years of age and to seek his fortune with the Fatimids. He
was styled "the Mutanabbi of the West", but this should be
taken as an indication of his standing among Andalusians and
of his grandiloquence and sententiousness rather than of his
genius. Illustrative of his style is this description of the war-
ships of the Fatimid ruler al-Muizz:

. . . They are proud mountains, except that they move;

Some of them are lofty peaks and some forbidding [heights].
They are birds—but birds of prey
That have no quarry other than [human] souls.
They are [flints] struck to start an all-consuming fire
Which when battle is joined proves unextinguishable.
When they breathe hard in anger they fling out a blaze
Like that with which the fire of Hell is fit.
So their hot mouths are thunderbolts
And the breath they exhale is iron.
They hold firebrands over the deeps
Like blood being caught in black sheets.
They embrace the waves of the sea as though it were
Oil in which they dip trimmed wicks.
The water there, though dark in the main,
Is like a skin touched [and stained] by the saffron of a
perfume.
They know no reins other than the winds,
And no rough terrain other than the foam . . .

This profusion of unassorted images, like debris tumbling

down a mountainside, is characteristic of the style that by
then prevailed throughout the Arabic-speaking lands. Two
features of i t w i l l seem strange to Western taste. The first is
its fragmentation: the poet does not attempt to build up one
consistent image, or induce one consistent mood; this is a phe-
nomenon observable i n Arabic poetry from its earliest stages,
64 UMAYYAD CULTURAL ACHIEVEMENTS

each idea or conceit being rounded off i n one self-sufficient

line. The other is the use of imagery i n ways which some-
times seem strangely lifeless and aesthetically indifferent, as
when awesome fire-carrying ships are compared to night-
fights.
Louis Massignon has made the arresting observation that
the artistic spirit of Islam tends to "de-realise", to petrify
objects so that the metaphor follows a descending gradation:
man is compared to animal, animal to flower, flower to pre-
cious stone. As regards the past, the poet does not try to re-
live emotion. He takes the memory as memory, he works
w i t h dreams, shades, ghosts. For the idea i n Muslim art is
not to idolise images, but to go beyond, to the One who
makes them move as i n a magic lantern or a shadow play,
towards the only One who endures. Attractive as this analy-
10

sis is, i t is by no means always true that i n the Arabic meta-

phor the progression is from animate to inanimate. Rather, i t
seems that the poet is indifferent to the way i n which the
progression is made. He is content w i t h a juxtaposition of
shapes or colours provided only that they coincide i n some
one respect—the reflection of yellow fire i n water and the
stain of saffron on human skin—without concern for the very
different experiences with which either image may be as-
sociated.
Both these features point to an atomism, a concern with
independent detail, deeply ingrained i n the cultural tradition
of the Arabs i f not i n their make-up, and contrasting sharply
with the concern w i t h unity displayed by European men of
letters since the days of ancient Greece.
I n this important respect, there was as yet no noticeable
difference between the poets of al-Andalus and those of the
East. Soon, al-Andalus would emerge as the fulcrum of Mus-
lim intellectual life west of Egypt. Relations w i t h Christian
courts i n the North of Spain and w i t h Byzantium, toleration
of Jewish scholars who were to act as translators and inter-
mediaries, access to Greek and even to some Latin sources—
these were to enable al-Andalus to form its own cultural
blend. I t would even seem that intermarriage with Iberians
and daily contacts with a population that remained largely
Christian and Romance-speaking were colouring the mentality
Plate 1. Door and blind arches in the western fagade of the
mosque at Cordova, showing tile and bas-relief decoration charac­
teristic of the art of the Caliphate.
Red and white striped arches of the mosque at Cordova.
Plate 3. Stone lattice and foiled arch in the mosque at Cordova
with characteristic arabesques of stylized floral motifs.
Plate 4. Arches in one of the main halls, possibly the throne room,
at the Palace of Madinat as-Zahra\
Plate 6 Oil lamp in wrought bronze (Alhambra Museum, Gra-
nada).
Plate 7. Ivory casket from the time of the Caliphate at Cordova.
Mozarab craftsmanship.
Plate 8. Torre del Oro, built by Almohads at beginning of thir­
teenth century, at the entrance of Seville on the River Guadalquivir.
Plate 9. The Giralda, Seville, once the minaret of the Almohad
Mosque, now Cathedral Tower.
 ART 65
of the Arabs, who were numerically far from dominant i n
the population. But this was not as yet apparent i n the lit-
erary production. A few books, now lost, appear to have been
written about Andalusian men of letters, and these may be
an indication of nascent "national" pride. But the one sig-
nificant new departure is the appearance of a new form of
poetic composition, the muwashshah (to be discussed i n
Chapter 9 ) , said to have been invented by either Muqaddam
ibn-Mu afa or Muhammad ibn-Mahmud both of whom lived
in Cabra, near Cordova, at the beginning of the tenth century.
As for fine prose or belles-lettres, the one form that was
practised was the epistle, and the scope for this mostly i n
official correspondence. Secretaries appear to have been nu-
merous i n al-Andalus, and their prestige considerable, for
we know of three books written about them; but none of
these books is extant. I t is likely that here again the Anda-
lusians followed the lead of the East: a preference for pithi-
ness i n official correspondence, and a growing taste for rhyme
and other verbal embeUishment.

3. Art

Though the study of art lies somewhat apart from the usual
fields of the historian, i t is important, i n dealing w i t h Islamic
Spain, to take into consideration the conclusions of the ex-
perts i n these matters, since the development of Islamic or
Moorish art is complementary to the literary and intellectual
history of al-Andalus. 11

The period up to 9 7 6 was the creative or formative period

of Moorish art during which there appeared the distinctive
features and characteristic spirit which are its glory. The out-
standing production was the Great Mosque of Cordova, be-
gun by 'Abd-ar-Rahman I and successively enlarged by *Abd-
ar-Rahman I I , al-^akam I I and (shortly after 9 7 6 ) by
al-Mansur. A section i n the middle was pulled down after the
Reconquista to make way for a Christian cathedral, but a
vast area remains much as i t was at the end of the tenth
century.
The oldest part of the mosque marks the unexpected
66 UMAYYAD CULTURAL ACHIEVEMENTS

emergence of a new architectural tradition. No exact proto-

type has been discovered, though there are reminiscences of
Umayyad and Syrian buildings. The horse-shoe arch which
is used throughout is now known to have been taken over
from Visigothic architecture; but the doubling of the arches
to give extra height—possibly suggested by Roman aqueducts
—is something new. The peak of artistic achievement is usu-
ally held to have been i n the part added by al-IJakam I I
with its more ornate arches and elaborate decoration, espe-
cially round the mihrab or niche indicating the direction of
Mecca.
The other notable artistic achievement was the palace-
city of Madinat az-Zahra. Though i t was ruined and ren-
dered uninhabitable i n 1013, i t has recently been i n part
excavated (and a small section restored), so that i t is now
possible to have some idea of what i t was like during the
brief period when i t was i n use. I t was not strictly utilitarian,
though i t was well used, but was rather an expression of the
self-confidence of the first caliph. No effort was spared to
make i t a thing of matchless beauty. The lay-out of the walls
is i n the Roman and Byzantine tradition, and so is the deco-
ration. Byzantine sculptors may even have been brought to
Cordova for the purpose. Al-IJakam I I is likewise said to have
brought Byzantine workers i n mosaics.
Besides these great architectural works and a number of
fortified castles belonging to the same period, there were also
small articles i n various media from ivory and marble to i n -
laid gold work and crystal (the latter having been discovered
at Cordova about 8 5 0 ) . A good tradition i n metal work was
inherited from Visigothic Spain, but the other crafts were
mainly developed tinder the Muslims, sometimes on the basis
of oriental techniques.
Moorish art seems a fusion of oriental and occidental, al-
though i t is difficult to pin-point the separate elements. As
was only natural most of the materials and techniques used
were those of Visigothic Spain. The decorative motifs show
considerable Hellenistic influence, some of which certainly
came through Visigothic Spain, though some possibly came
from the hellenised art of Syria. For long, indeed, the art of
Islamic Spain showed traces of the Umayyad art of Syria,
 THE SOURCE O F MOORISH CULTURE 6 7

w i t h all the influences which i t had absorbed. As late as the

time of 'Abd-ar-Rahman I I I the new minaret of the Mosque
of Cordova was square i n cross-section like the minarets of
Syria which had been modelled on church towers. I t is i n the
additions to the Mosque by his successor al-IJakam I I that
Baghdadian influences become clearly marked; and this was
the period, of course, when definite efforts were being made
to assimilate the intellectual culture of the heartlands.
How is this fusion of cultures to be described? A t one
point Henri Terrasse speaks of an expression " i n an Islamic
dress of the soul of Spain"; but a few lines lower he says that
"Umayyad Spain attained, for all its temporary political co-
hesion, an ever more affirmed unity of thought and l i f e " , 12

and this would imply that the civilisation was primarily Arab
and Muslim. Is i t then i n essence something Spanish or Arab
or Muslim? The energy to expand into Spain and there create
the Umayyad state came from the drive of Arab peoples
guided by the conceptions of Islamic religion. Yet into this
community were incorporated not only Berbers but also
many of the Iberian population; and the incorporation was
a genuine integration into a "unity of thought and life". Each
of the two formulations seems to express an aspect of the
truth.

4. The Source of Moorish Culture

The reflections on Moorish art of the previous pages require

to be complemented by a consideration of similar questions
i n the literary and intellectual sphere. Here the distinction
between Arab and Islamic elements comes to the fore.
As has been noticed at various points above there is much
to suggest that at first the Arab element was dominant. This
was only natural i n a state which had originally been a prov-
ince of the caliphate of Damascus, characterised above all by
Arabism. The chief aesthetic form among the Arabs was lan-
guage and poetry, and somehow or other the fascination for
the Arabs of their own language communicated itself to many
of the other inhabitants of the peninsula, including the Moz-
arabs. As a result of this i t was natural that there should be
68 UMAYYAD CULTURAL ACHIEVEMENTS

a strong interest i n poetry and philology. The acceptance of

the Malikite legal rite, too, was i n keeping with the practical
and non-speculative Arab outlook.
Yet there was also a w i l l to belong to the great commu-
nity of Muslims, and to keep i n touch with the distinctively
Islamic developments of thought i n the heartlands. Through-
out the Umayyad emirate and caliphate nearly all the lead-
ing jurists made the "journey" to the East to sit at the feet
of the great teachers there. I t has been noted how, as early
as 822, the aesthete Ziryab was given a warm welcome at
the court of Cordova, and became an arbiter of fashion i n
many fields until his death i n 857.
The most important steps, however, towards maintaining
continuity w i t h the literary and intellectual outlook of Bagh-
dad were taken i n the tenth century, first of all by 'Abd-ar-
Rahman I I I but chiefly by al-IJakam I I . Poetry and belles-
lettres had attracted attention before this, and the anthology
of Abu-Tammam ( d . 8 4 6 ) and a work on rhetoric by al-
Jahiz (d. 8 6 9 ) had been brought to Spain i n the reign of
Muhammad I ( 8 5 2 - 8 6 ) by returning scholars, while the
13

works of Ibn-Qutayba (d. 8 8 9 ) were known by 9 1 0 . I t was

14

*Abd-ar-Rahman I I I , however, who welcomed the great phi-

lologist al-Qall i n 9 4 1 . I t must have been about this time,
too, and so probably w i t h some encouragement from the
court, that 'Arib was making his summary and continuation
of the history of at-Tabari. Advantage was also taken of the
Fatimid occupation of Cairouan to encourage Sunnite schol-
ars, uneasy i n the new conditions, to settle i n al-Andalus.
Al-IJakam I I had himself a deep interest i n scholarship
and bibliography, and set about making the caliphal library
one of the largest and best i n the Islamic world. I t is said
eventually to have had over four hundred thousand volumes.
He employed the historian 'Arib as a secretary, though
whether for the library is not known. I n 963, just after the
beginning of the reign, there settled i n Cordova a scholar
from the east who was an expert in the text of the Q u r a n 15

—a subject which had recently been of political importance

i n Iraq. This man had some knowledge of Shafi'ite jurispru-
dence, but was presumably unable to make use of it because
of Malikite opposition. The arrival i n al-Andalus of another
 THE SOURCE O F MOORISH CULTURE 69

eastern expert i n the same subject just before the death of

the first shows how i t was possible to establish in al-Andalus
a living tradition of scholarship i n the various branches of the
"religious sciences".
The above remarks would seem to contradict the view
that the distinctively Islamic aspects of the culture of the
Muslims were prominent early i n the emirate. On the con-
trary i t becomes more and more clear that to begin with the
main cultural influence was what has already been called
Arabism. Only about the middle of the tenth century through
the exuberance following on the successes of the first caliph
did the men of al-Andalus contrive to make the Islamic sci-
ences take firm root among them. The suggestion of Levi-
Provengal that 'Abd-ar-Rahman I I I cultivated Byzantine
16

artistic forms i n order to reduce his dependence on Baghdad

is surely mistaken, for this is the very period when al-Andalus
was making the greatest efforts to establish i n itself the Sun-
nite intellectual traditions of the east.
 6
T H E COLLAPSE OF ARAB R U L E

1. The 'Amirid Dictatorship and the Breakdown

Caliphs: Hisham I I : 9 7 6 - 1 0 1 3
other six Umayyads: 1009-1031
three ljammudids: 1016-1027
Chamberlains: al-Mansur (ibn-Abi-'Amir): 9 7 8 - 1 0 0 2
al-Muzaffar ('Abd-al-Malik): 1002-1008
al-Mamun ('Abd-ar-Rahman): 1008-1009

When al-IJakam I I died i n 976 he was succeeded by his son

Hisham I I , then eleven years old. There were some influen-
tial men who wanted a younger brother of al-IJakam, since
they realised that a regency would be disadvantageous to
themselves; but Ja'far al-Mus'hafi, the minister to whom the
caliph had entrusted the management of affairs during his
illness, acted energetically, and secured both the boy's suc-
cession and the continuation of his own power.
I n these events al-Mus'hafi was supported by a man of
thirty-eight, often known as Ibn-Abl-'Amir, who came from
an old Arab family w i t h lands near Algeciras. Ibn-Abi-'Amir
had come to Cordova to study jurisprudence and literature,
and had been appointed steward to the princess Subh to
look after the properties and revenues of her son Hisham,
whom al-]Jakam had regarded as his successor. From this
fairly humble position Ibn-Abl-'Amir was able to scheme and
intrigue his way up the ladder of the civil service until i n
9 7 6 his influence was an important factor i n securing the
succession of Hisham. His ambitions were not yet satisfied,
however. W i t h uncanny insight into the movement of events
and a deep understanding of men's reactions to them, he
schemed and calculated—at times i n an utterly cold-blooded
fashion—until i n 978, w i t h support from the general Ghalib,
whose daughter he married, he ousted al-Mus'hafi and him-
self became hajib or chamberlain.
 THE 'AMIRID DICTATORSHIP 71
I n the three following years he was busy strengthening
his own position further. One side of his policy was to gain
the support of the jurists. A plot against the caliph gave him
the opportunity to order the execution of a "Mu'tazilite" op-
ponent of the jurists who was implicated i n i t . To gain further
support he himself copied out the Qur'an with his own hand,
and had many heretical works removed from the library of
al-llakam I I and burned. Another side of his policy was to
reduce the young caliph to impotence. He was encouraged
i n sensual indulgences and was kept away from contacts and
activities which might have prepared him for assuming per-
sonal control of state affairs. The princess Subh, seeing what
was happening to her son, was filled w i t h bitter hatred for
her former proteg6, said to have been also her lover; but
her efforts to alter the situation were easily countered by the
astute political climber. The final blow was i n 9 8 1 when he
moved the administration from the caliphal palaces of the
Alcazar i n Cordova and Madinat az-Zahra' to a new palace
built by himself and called Al-Madina az-Zahira. The caliph
was virtually cut off from outside contacts; and i t was given
out that he had decided to devote himself to piety and to
hand over the entire supervision of the affairs of the realm
to Ibn-Abl-'Amir.
I n this same year of 9 8 1 a quarrel developed between I b n -
Abi-'Amir and his father-in-law Ghalib. The latter had some
help from the Christian princes of the north, but Ibn-Abi-
*Amir w i t h keen foresight summoned from Africa another
general w i t h his Berber troops, and placed his reliance on
these and on Christian mercenaries. The jundis or men of the
levies were grouped i n artificial regiments instead of i n tribes
as previously, and this meant that they were much weaker
than they had been. Despite his Christian help Ghalib was
defeated and killed. Ibn-Abi-'Amir on his return i n triumph
to Cordova took the title of al-Mansur bi-Tlah ("the one ren-
dered victorious by God"), usually shortened to al-Mansur or
i n the Romance dialects Almanzor. He also was given or
took the privilege of having his name mentioned after that
of the caliph at the Friday midday worship—a sign that he
was of nearly equal rank. As time went on, he received other
marks of sovereignty, but he was wise enough not to make
7% THE COLLAPSE O F ARAB RULE

any claim of any kind to the caliphal dignity itself. The pe-
riod from 9 8 1 to the death of al-Mansur's son al-Muzaffar
in 1008 is thus justifiably referred to as the 'Amirid dictator-
ship, but apart from the fact that al-Mansur was both strong
and efficient his rule was no more autocratic than that of
most other Muslim regimes of the time.
Al-Mansur's "reign" is known to have been one of great
military activity, but the records are scanty. He is said to
have led fifty-seven victorious expeditions. The result of all
this activity was an extension of the region definitely held
and occupied by Muslims, and the maintenance of a degree
of suzerainty over the Christian kingdoms. Christian rulers
attempting to break agreements with al-Mansur met w i t h
severe reprisals. Most of the expeditions were against Leon
and Castile or against semi-independent feudal lords within
this general area. I n 985, however, the Muslims marched
against Barcelona, while 9 9 7 saw the great expedition which
pillaged and destroyed the church and shrine of Saint James
at Compostella i n the north-west corner of the peninsula.
Only the actual tomb of the saint was left intact. Was i t
superstition that led al-Mansur to make this exception? I t
certainly enabled the Christians to claim a little later that the
saint had been too strong for the Muslims. I n the year 1000,
1

however, when several Christian rulers were driven by the

disaster to join together to resist the Muslims, the result was
one more severe defeat for them. I n the closing years of al-
Mansur Muslim military power was thus supreme practically
to the Pyrenees. Indeed, such was the vitality and vigour of
al-Andalus that during the same years i t was extending its
influence i n North Africa, so that i n 998 al-Mansurs son, the
future al-Muzaffar, was able to establish himself i n Fez as a
kind of viceroy.
When al-Mansur died i n 1002, apparently worn out by the
anxieties of his strenuous career, his son 'Abd-ai-Malik had
little difficulty i n succeeding to his fathers position by ob-
taining from the caliph Hisham I I a grant of similar powers.
During his "reign" of six years the position of the Muslims
w i t h regard to the Christian kingdoms i n the north was more
or less maintained, though with increasing difficulty. Constant
military activity was required. After a successful expedition
 THE 'AMIRID DICTATORSHIP 73
in 1007 the caliph decreed for 'Abd-al-Malik the honorific
title of al-Muzaffar, "the victor", but less than a year later
the ruler of al-Andalus was dead i n somewhat mysterious
circumstances. He had proved, even i f not the equal of his
father, a competent administrator and a first-class general.
The years from 1008 to 1031 are i n some ways one of the
most tragic quarter-centuries i n all history. From the pinnacle
of its wealth, power and cultural achievements al-Andalus
fell into the abyss of a bloody civil war. No central authority
remained which was able to maintain order throughout the
country. Everywhere was confusion. One leader after another
appeared, each with his supporting group, and tried to estab-
lish effective government at the centre; but one after another
had to admit failure. Some lasted only a month or two, none
more than two or three years. The nominal (and often also
actual) leaders of the various attempts to re-establish cen-
tral control claimed the dignity of caliph. Besides Hisham
I I , who was forced to abdicate i n 1009 but restored i n 1010,
six members of the Umayyad family held the caliphate i n
this period, as well as three members of a half-Berber family
known as the Qammudids. The farce was ended in 1031
when a council of ministers meeting i n Cordova decreed the
abolition of the caliphate and the setting up of a council of
state. This council, of course, ruled only the region of Cor-
dova.
The sorry sequence of events was started off by the
younger brother who succeeded al-Muzaffar. He quickly an-
tagonised the people of Cordova, not least by persuading
the caliph to declare him heir to the caliphate; and while
he was absent i n the north they deposed the caliph and set
up another Umayyad i n his place. Even now the young
'Amirid did not understand how to retain the loyalty of his
army, and he soon perished. Before long, however, the new
caliph had lost most of his original support, and i t was the
turn of a group of Berber officers to gain power for an Umay-
yad nominee to the caliphate. Next there came to the fore a
group of Saqaliba or Slavs, mostly civil servants or mercenary
soldiers, likewise w i t h a candidate. So i t went on. I t is impos-
sible here to give all the details, but i t may be noted that
74 THE COLLAPSE O F ARAB RULE

the three chief eaKph-making groups were the populace of

Cordova, the Berbers and the Saqaliba.
By 1031 the thirty towns of any size had a more or less
independent ruler. This was the state of affairs which caused
the period from 1031 (or from 1009) to be known as the
era of the "party kings" or reyes de taifas (Ar. muluk Ot­
tawa if) .

2. The Reasons for the Breakdown

Although the breakdown of the caliphate has a central place

i n the history of Islamic Spain, the reasons for that break-
down have not been thoroughly investigated. What is to be
said here is therefore once again of a provisional character.
The immediately obvious fact underlying the breakdown
was what is called "particularism", and this was both local
and racial. Difficulties of communication, due to the various
mountain ranges probably encouraged the tendency for each
region to become an independent political unit. Effective
power was i n the hands of the local ruler, and only a vast
expenditure of energy by the central government could keep
the local rulers i n check. Also, from about the middle of the
tenth century the mixture of races was becoming more of a
problem. I t is possible that up till then the foreign elements
which entered i n the eighth century had been largely as-
similated. Even where there was little physical intermingling,
there seems to have been a degree of cultural homogeneity.
I n the tenth century however, i t became usual to import
many slaves from the north and east of Europe, known as
Saqaliba or "Slavs", to serve as soldiers and to fill posts i n
the civil service. Their chief came to have considerable i n -
fluence. I n addition, al-Mansur in his rise to power had
brought over from Africa fresh contingents of Berbers, whose
attitudes were different from those of the long-established
Berbers. A l l these facts indicated that there had been an
accentuation of racial divisions.
Though this state of affairs is tolerably clear, i t is not clear
why i t should have become so much more difficult i n the
early eleventh century to maintain unity. Even i f some of
 THE REASONS FOR T H E BREAKDOWN 75

those who attempted to re-establish the central government

were incompetent, surely they were not all incompetent. Or,
if they were, had some change come over the character of
the people? We know that wealth had greatly increased un-
der 'Abd-ar-Rahman I I I , and i t may be that the bulk of the
population had become so materialistic i n outlook that few
were capable of the sacrifices needed for unity. This mate-
rialistic outlook of the leaders or of their supporters, or both,
was very probably one of the factors which led to the break-
down.
Another line of thought is suggested by keeping i n mind
the similarities of the situation i n Baghdad. There the power
had for some time been slipping from the caliph, and i n 945
it finally passed into the hands of a family of military leaders;
but neither they nor their successors managed to keep con-
trol of the full extent of the caliphate. Though the break-
down was never so complete as i n al-Andalus and was fol-
lowed by a partial revival, there is something comparable
both to the 'Amirid "dictatorship" and to the loss of unity.
Could i t be, then, that there was either some fundamental
defect i n Islamic civilisation, or i n the whole medieval struc-
ture of society? Two points seem to be specially relevant:
the failure to adapt Islamic ideas to contemporary prob-
lems, and the absence of a strongly-based middle class inter-
ested i n maintaining an effective central government.
W i t h respect to the first of these points i t may be noted
that Islam, though i t has the name of being a political re-
ligion, has not been conspicuously successful i n its political
ideas. Things went well enough during Muhammad's life-
2

time because he was able to adapt existing ideas and insti-

tutions to the needs of his growing community. He and his
immediate followers, however, were virtually restricted to
the political conceptions associated w i t h the Arabian tribe.
I n one or two matters these proved capable of useful de-
velopment, namely, i n regarding the community of Muslims
as comparable to a tribe and communities of non-Muslims
as subordinate tribes. Yet such ideas alone were not sufficient
in a great empire, and inevitably Persian ideas of statecraft
were borrowed, tentatively under the Umayyads of Damas-
cus, and without reserve under the 'Abbasids.
7 6 THE COLLAPSE OF ARAB RULE

Some of these Persian ideas also permeated to Spain.

What was noteworthy i n Spain, however, as has been al-
ready mentioned, was the acceptance of some of the feudal
conceptions of Western Europe. D i d the Muslim rulers who
were prepared to have Christian princes under their suze-
rainty (while leaving them local autonomy) think they were
acting according to Muhammad's model for the treatment of
subordinate communities and fail to notice the differences?
Or were they aware of the differences but happily decided
to accept local practices? I f the latter alternative is near the
truth, is i t possible that the Islamic religion was less effective
than the Christian i n giving support to the relationship of a
man to his liege-lord (even i f the Christian was far from
perfect)? Because the political ideas according to which Mus-
lims acted were not closely linked w i t h the basic ideas of
Islam as a religion, political activity had little religious sanc-
tion and men tended, therefore, to follow self-interest or
raison d'etat. I n other words, the chief concern of any regime
came to be its own maintenance and not the welfare of those
ruled.
The considerations about ideas of a feudal type are most
relevant to a discussion of why the Muslims failed to expand
in the Iberian peninsula or even to hold what they possessed.
Yet they also affect other aspects of the problem, such as
military policy. The reason for having fresh Berber and Slav
immigrants was to make it possible to keep the Christian
princes i n check and to expand i n North Africa. But were
these genuinely Islamic policies conducive to maintaining a
body politic i n which men could freely worship God and pre-
pare themselves for facing the Last Judgement? The rulers
of al-Andalus certainly spoke about the holy war; but was
this any more than a way of raising the spirits of their troops?
The relationship of politics to religion is never an easy matter.
Politics has a proper autonomy, and political activities must
be guided by political considerations. A t times, however, i n
the Islamic world politics was carried on wholly within a
framework of religious ideas; and this was so at some of the
more successful periods. Yet elsewhere, as i n al-Andalus,
polities burst out of the religious framework, and one won-
ders how much this had to do with the political failure.
 THE REASONS FOR T H E BREAKDOWN 77

The second point for discussion was the absence of a mid-

dle class interested i n maintaining a strong central govern-
ment, and some of the matters just mentioned are again
relevant. The question of class structure i n the medieval
orient is not easy. Roughly speaking, i t seems to have been
the case that there were two classes, an upper class and a
lower class. The latter consisted of the urban and rural
proletariats; the former, of the rulers, the civil servants and
other administrators, the land-owners (often also administra-
tors) and perhaps the great merchants. The intellectuals, of
whom the Sunnite jurists are the chief representatives, stood
apart, but had come to be largely dependent on the rulers
and subservient to them. I n so far as the intellectuals per-
formed their function of safeguarding the intellectual basis
of Islam they had some influence over the urban proletariat
Otherwise only the upper class was politically active and i n -
fluential.
I t would appear, however, that the effect on the upper
class of the increasing wealth of the country (whether al-
Andalus or Iraq) was to accentuate the division into groups
or cliques, each trying to better itself materially at the ex-
pense of the other groups. Seldom i n Islamic history had
the upper class found its primary motivation i n the ideas of
religion; and non-religious motivations were certainly strong
in al-Andalus at the end of the tenth century. While those
in power were ready to use religious ideas like the holy war
to spur the proletariat to greater efforts, other members of
the upper class probably recognised this exploitation for what
i t was. The expansionist military policy of the 'Amirids was
doubtless regarded by rival cliques as aimed at increasing
their own power and glory, rather than simply at keeping
the Christian princes i n check. I n such circumstances there
would be little support for the policy among the upper class,
and something of this attitude might filter down through the
society. The increase of luxury would i n any case make
many men unwilling to expose themselves to the discomforts
and dangers of military campaigns.
A t the root of some of the difficulties inherent i n this con-
dition of affairs was the failure to develop a religiously-based
conception of the function of the upper class i n the commu-
78 THE COLLAPSE O F ARAB RULE

nity of Muslims. There was some idea of the special place

of the imam or leader, but there was really nothing between
him and the ordinary Muslim. The consequence i n practice
was that relations between the caliph and the upper class
were not guided by any religious ideas but by sheer self-
interest. There was nothing to foster i n any members of the
upper class a loyalty to the central government as a struc-
tural principle of the community. I f they opposed those i n
power, i t could only be from self-interest i n the hope of i n -
creasing their own share of power. The wealthier members
of the proletariat were likewise without any grounds for ex-
erting themselves to preserve the structure of society. The
masses could indeed be stirred to vigorous action by religious
ideas such as opposition to heretical innovation, but such
ideas were largely irrelevant to contemporary circumstances
and their application to such circumstances merely op-
portunist.
Also, i t was perhaps the case i n al-Andalus that, hidden
under its Sunnite exterior, there was an almost Shiite re-
spect for the person of the rightful imam. This would be i n
keeping w i t h the outlook of the peoples of North Africa.
That this factor was important is suggested by the reports of
the dismay of many when the younger son of al-Mansur
(who of course had none of the blood of the Prophet's fam-
ily) got himself declared heir to the caliphate.
These then are some of the factors i n the situation of al-
Andalus just before the breakdown of the caliphate and of
the central government. Before a final evaluation of them
can be given much further research is needed.

3. The 'Tarty Kings" (1009-91)

Though a semblance of the caliphate continued until 1031,

the break-up of the unity of al-Andalus began i n 1009. As
soon as the central government lost control local governors
or other leaders were virtually compelled to take authority
into their own hands. On the frontier areas or Marches the
disintegration was not so great, since much power was al-
ready concentrated i n the hands of the commanders there.
 THE "PARTY KINGS" (1009-91) 79

Thus political units of moderate size continued to exist w i t h

their capitals at Badajoz, Toledo and Saragossa i n the Lower,
Middle and Upper Marches respectively. I n the rest of the
country the situation was rather different, and i n the earlier
part of the eleventh century nearly thirty separate political
units can be found i n or near the south and east coasts.
Some did not maintain for long any degree of independence.
There were constant intrigues both between the small states
and within each, and also constant fighting. A ruler often
could not trust his chief minister, and yet had to rely o$ h i m ;
and the members of a ruler's family were frequently plotting
to bring about his downfall and replace h i m . The political
3

history of the period is thus a confused and tangled mass of

petty events.
The "parties" (taifas, tawaif) from whom the "party kings"
take the name are the three ethnic groups of Berbers,
Saqaliba ("Slavs") and "Andalusians"; the latter included
all Muslims of Arab and Iberian stock (and perhaps some
descendants of early Berber settlers), who were now almost
fused into one so that the Arabs were not reckoned as a sepa-
rate "party". I n any region one "party" tended to be dominant
and to rule primarily i n its own interests without much
thought for the welfare of the rest of the population. Thus
there was a lack of unity even i n the small states into which
Spain was now divided.
The Berbers controlled the south coast from the Guadal-
quivir to Granada w i t h its seaboard. One notable dynasty
was the IJammudid which before 1031 produced three claim-
ants to the caliphate and which ruled Malaga and Algeciras
until after the middle of the century. Still stronger was the
Zirid dynasty i n Granada, which shortly after mid-century
added Malaga to its domains. About the same time Algeciras
and the small towns between i t and the Guadalquivir be-
came subject to Seville. The Saqaliba mostly moved east-
wards at the breakdown of the central government, and some
of their representatives gained power i n coastal towns like
Almeria, Valencia and Tortosa; but they did not form dynas-
ties like the Berbers.
Among the "Andalusians" the strongest dynasty was that
of the 'Abbadids i n Seville. Its founder was the qadi or judge
8o THE COLLAPSE OF ARAB RULE

Muhammad ibn-'Abbad, who held the supreme power from

1013 to 1042. He was succeeded by his son and grandson,
usually known by their honorific titles of al-Mutadid ( 1 0 4 2 -
6 8 ) and al-Mutamid ( 1 0 6 8 - 9 1 ) . Al-Mutadid greatly ex-
tended the small kingdom of Seville to the west and south-
west, and engaged i n fighting against Cordova and Granada
in the east. Cordova was eventually included i n his kingdom
by al-Mutamid. Despite the political upheavals art and let-
ters flourished under the "party kings", since each little ruler
imitated the splendour of the former caliphal court as far as
his resources permitted. The court of Seville, however, under
al-Mu'tadid and al-Mutamid, was undoubtedly the most bril-
liant i n Spain.
4

Cordova, after the fall of the 'Amirids i n 1009, was first

of all the chief scene of the struggle for the caliphate. I n the
course of this i t was pitilessly sacked i n 1013. Jahwar, the
man responsible for the declaration abolishing the caliphate
in 1031, virtually held supreme power thereafter, though he
tried to insist that rule was i n the hands of a council. He was
followed by his son and grandson; and historians sometimes
speak of a dynasty of Jahwarids. Cordova, as just mentioned,
5

was finally added to the domains of Seville, after a short in-

terlude when i t was under Toledo.
The disintegration of al-Andalus was, of course, the op-
portunity for the Christian princes i n the north, and, though
occasionally still quarrelling among themselves, they did not
fail to make use of the opportunity. Instead of themselves
paying tribute to the caliph, they were able to demand tribute
from the "party kings". First it was the turn of the rulers on
the Marches—at Badajoz, Toledo and Saragossa—to reach this
degree of subordination. The most vigorous of the Christian
rulers, Alfonso V I of Leon and Castile ( 1 0 6 5 - 1 1 0 9 ) , was
able to exact tribute even from the relatively strong kingdom
of Seville. The "party kingdom" of Toledo was the weakest
of the three on the Marches and succumbed to Alfonso i n
1085. This was an important stage i n the Reconquista, since
Toledo was never afterwards in Muslim hands. Yet one won-
ders how far a man like Alfonso was consciously fighting as a
Christian against Muslims, and how far simply strengthening
his own kingdom. I t has been suggested that the Christian
 THE "PARTY KINGS" (1009-91) 81

Spaniards and the Arabo-Iberian Muslims (called "Andalu-

sians" above) felt themselves to be essentially a single people.
One point corroborating this suggestion is the Muslim ac-
ceptance of "feudal ideas" (discussed i n the previous section).
Another illustration would be the career of the Cid; this name
is an Arabic title (sayyid or sid meaning "lord") which is
used par excellence of Rodrigo Diaz de Vivar, a Castilian
noble, who about 1081 after a quarrel w i t h Alfonso V I
offered his services as a military leader to the Muslim king
of Saragossa and ended as independent ruler of the Muslim
town of Valencia. Despite his close association w i t h Muslims
he was adopted by Christian Spain as a paragon of manly
prowess. 6

The fall of Toledo and the generally threatening situation

caused al-Mu tamid of Seville to seek the help of the ruler
of the powerful Almoravid state i n North Africa, Yusuf ibn-
Tashufin (or Tashfin). Yusuf brought an army across the
straits to Spain and defeated Alfonso V I at Zallaqa near
Badajoz i n 1086; then he and his men returned to Africa.
Despite the Muslim victory, however, the threat continued;
Yusuf was again summoned, and arrived i n 1088. The cam-
paign d i d not go so smoothly as he had hoped, and under
the influence of the Malikite jurists of al-Andalus he decided
not simply to pursue the limited objectives of those who i n -
vited him, but to make an all-out effort on his own account
to retrieve the fortunes of Islam. A t the end of 1090 he
moved forward to dispossess the Muslim rulers, and in the
course of 1091 Cordova and Seville fell into his hands. This
may be taken as the beginning of the Almoravid period.
 7
T H E BERBER EMPIRES, T H E A L M O R A V I D S

1. The Foundations of the Almoravid State

The North African state to which the eyes of Spanish Mus-

lims turned after the fall of Toledo i n 1085 had grown to a
vast size i n less than half a century. I t included not merely
the whole of Morocco and Mauritania, but also the basin of
the Senegal river i n the south and the western part of Algeria
i n the north. Only meagre accounts have been preserved of
the outward history of the religious movement of the A l -
moravids, and these give little understanding of the funda-
mental causes of its rapid success. This is one of many sub-
jects which deserve fuller investigation.1

The movement began i n a group of camel-breeding, no-

madic Berber tribes known collectively as Sanhaja. Their
home had been the steppes of the Sahara, but some of them
had moved south to the basins of the Senegal and the upper
Niger. They are the ancestors of the modern Tuareg, and
i t may be that "Senegal" is derived from their name through
a dialectal variant "Sanaga". The story of the movement
begins with the performance of the pilgrimage to Mecca by
some notables of one of the Sanhaja tribes, led by their chief
Yahya ibn-Ibrahim. On their way back they spent some time
i n Cairouan, then the intellectual centre of North Africa
(apart from E g y p t ) . Here they were much impressed by the
teaching of a Malikite jurist, Abu-'Imran al-Fasi, who prob-
2

ably died a few months after their visit. He realised how

much they and their fellow-tribesmen needed instruction, and
w i t h his support they prevailed on the pupil of a pupil of
his to accompany them as teacher. The man was Ibn-Yasin
(more fully, 'Abd-Allah ibn-Yasin al-JazuK), and the date
April 1039.
The subdivision of Sanhaja to which Yahya ibn-Ibrahim
belonged did not like the teaching of Ibn-Yasin, and at
 FOUNDATIONS O F T H E ALMORAVID STATE 83

length the latter, w i t h some disciples from another sub-

division, retired to an island i n the Niger, and is reported
to have given himself up to religious studies and pious ex-
ercises. European conceptions, however, must not be allowed
to mislead us, for despite the undoubted mysticism and as-
ceticism these men presently emerged as fierce and efficient
soldiers. The Arabic word for such a "house of retreat" as
they had is ribat, and from this is derived the usual name
of the adherents of Ibn-Yasm's movement, al-Murabitun,
which has come through Spanish and French into English
i n the form "Almoravids". These facts help to make i t clear
that what attracted the Sanhaja i n the first place was not
simply Malikite jurisprudence, but the mystical teaching as-
sociated w i t h i t ; and i t is noteworthy that the man who first
stirred the Sanhaja, Abu-'Imran al-Fasi, was regarded as a
saint by later sufis or mystics.
I t was about the year 1055 that the Almoravid army en-
tered on a phase of expansion by conquering the small state
w i t h its capital at the oasis of Sijilmasa. The military leader
here was Yahya ibn-'Umar, but Ibn-Yasin was still the ac-
knowledged spiritual head. When Yahya was killed a year or
so later, Ibn-Yasin saw to i t that he was succeeded by his
brother Abu-Bakr ibn-'Umar, and (since Ibn-Yasin i n turn
was killed about 1058) Abu-Bakr remained supreme head
(amir) of the movement t i l l his death i n 1087. Successes now
came rapidly, and i n 1061 Abu-Bakr gave his cousin, Yusuf
ibn-Tashufin, a semi-independent command i n the northern
regions, while he himself devoted his attention to the south.
From the new capital of Marrakesh (Marrakush), which he
founded i n 1062, Yusuf ibn-Tashufin extended Almoravid
rule over the fertile areas of Morocco and the western half
of modern Algeria.
This expansion of the Almoravids and their growth i n
power is partly explained by the fact that the regions they
conquered were at that period divided up into many small,
weak states; but i t was presumably the combination of re-
ligious and political aims which gave the Almoravids their
power by making possible a measure of unity between the
various subdivisions of Sanhaja. The rapid growth of empires
from insignificant beginnings has been a not infrequent fea-
84 THE ALMORAVIDS

ture of nomadic life, and a comparison at once springs to

mind with the religious and political movement i n Arabia
under Muhammad. Apart from the final outcome, however,
there were various differences. One was that the Almoravids
found a ready-made system of law, and worked as much as
possible through the existing Malikite jurists. Another was
that they acknowledged themselves part of a greater unit
by professing loyalty to the 'Abbasid caliphs i n Baghdad.

2,. The Almoravids in Spain

Yusuf ibn-Tashufin: ( 1 0 9 0 ) - 1 1 0 6
' A l l ibn-Yusuf: 1106-1143
Tashufin ibn-'Ali: 1143-1145

Nothing but the desperate situation i n which they found

themselves after the fall of Toledo i n 1085 could have i n -
duced al-Mu'tamid of Seville and other rulers i n al-Andalus
to invite Yusuf ibn-Tashufin to Spain. Before finally conclud-
ing an agreement w i t h him they made stipulations about
his return to Africa after the hoped-for defeat of the Chris-
tians; and he, while accepting these i n principle, made some
counter-stipulations. Algeciras was eventually given to him
as a base, and i n late summer 1086 he and his men set out
to meet the army of Alfonso V I . The armies met at Zallaqa
near Badajoz, and the result was a complete victory for the
Muslims; the Christians were either killed or put to flight
i n disorder. I n due course Yusuf and his men returned to
Africa i n accordance w i t h his undertakings.
The victory of Zallaqa, however, though a setback for
Alfonso V I , did nothing to alter the basic situation i n Spain,
namely, that the Muslims, because they were so divided
(and perhaps for other reasons), were much weaker than
the Christians and unable to parry attacks. I n particular the
eastern coastal regions from Valencia to Lorca were virtually
still controlled by Alfonso, and, w i t h a Castilian garrison i n
the strong fortress of Aledo between Lorca and Murcia, his
hold on the region was increasing. Further appeals were there-
fore made to Yusuf ibn-Tashufin both by the Malikite jurists
 THE ALMORAVIDS IN SPAIN 85
and by al-Mutamid and other princes. Yusuf and his cap-
tains for their part had tasted something of the luxury of al-
Andalus, and were probably not loth to return. I n addition
they believed that they were promoting the cause of Islam
by fighting against its enemies. I n the spring of 1090, there-
fore, Almoravid forces landed for the second time at Alge-
ciras, and Yusuf led them and the contingents from al-
Andalus against the fortress of Aledo. The siege dragged on
for several months. When Alfonso V I approached with a re-
lieving army, Yusuf retired to Lorca, but Alfonso found the
fortress to all intents indefensible and razed i t to the ground.
Yusuf had thus gained one important objective.
During the siege Yusuf had also been taking stock of the
general political situation i n Spain. He realised that i n most
of the petty states the control of affairs was i n the hands of
members of the Arabo-Andalusian aristocracy, and that these
persons, though Muslims, were not deeply attached to the
Islamic religion but were chiefly interested i n poetry, belles-
lettres and the arts generally. On the other hand, he was
aware that he had a great volume of support from the ordi-
nary people and from the Malikite jurists. I f i t may be pre-
sumed that he originally intended to retire again to Africa
after setting the Muslims of Spain on their feet, i t had become
clear to him by the closing months of 1090 that this was not
feasible. The controlling aristocracies i n the petty kingdoms
and principalities were too suspicious of one another to be
able to resist Alfonso. The interests of the Muslims as a whole
called for Yusuf to unify al-Andalus under himself, and he
was presumably moved i n the same direction by his own
ambitions coupled w i t h the expansionist ethos of the Almo-
ravid polity, of which he was now supreme head.
Yusuf lost no time i n acting on the decision he had taken.
Before the end of 1090 he had occupied Granada without
fighting. I n March 1 0 9 1 Cordova fell to him, and soon after-
wards he began a siege of Seville which led to the surrender
of that city and of al-Mu tamid himself i n September. Various
smaller towns had also passed under his control. The south
of Spain had thus been incorporated into the Almoravid em-
pire, and, as opportunity offered, its grip was extended north-
wards. The most important stages were the capture of
86 THE ALMORAVIDS

Badajoz i n 1094, of Valencia i n 1102 and of Saragossa i n

1110. I n Valencia the Cid had died i n 1099, but his widow
had been able to maintain independence for some time
longer. The fall of Valencia, as also that of the other petty
states, was an indication of the great military superiority of
the Almoravids over Alfonso. Despite this the Almoravids
were not strong enough to occupy any territory that had
been effectively occupied by the Christians, who had been
pursuing a policy of resettling empty lands with Mozarabic
Christians from al-Andalus. I n particular the Almoravids were
unable to recapture Toledo.
The power of this Berber dynasty did not long remain at
its zenith. Generals, other officers and men were filled w i t h
admiration for the culture and material luxury of al-Andalus,
which far surpassed that of the cities of North Africa and
still more that of the steppe environment from which they
originally came. This admiration led, i f not to moral corrup-
tion, at least to a weakening of moral fibre. A l l began to
consider only their own interests, and senior officers lost con-
trol of those under them. There was a loss of cohesion i n the
ruling institution as a whole. Financial difficulties were su-
perimposed on the arrogant behaviour of the Berber soldiery
to produce disaffection among sections of the ordinary peo-
ple; and this disaffection was sufficient to produce a change
of fortune for the regime.
The decline began i n 1118 w i t h the loss of Saragossa to
Alfonso I of Aragon (el Batallador); the disloyalty of much
of the populace contributed to this setback. The same Chris-
tian king was able to make excursions far into the south of
the country i n 1125 and 1126, and to remove numbers of
Mozarabs for resettlement i n newly acquired Christian re-
gions i n the north. Alfonso V I I of Castile was able to make a
similar expedition into the south i n 1133. Eventually the
growing disaffection and discontent of the common people
led to the rebellions of 1144 and 1145 which ended Almora-
vid rule i n Spain.
There is not yet any general agreement among scholars
on how to assess Almoravid Spain. The view of Dozy, which
has mostly held the field so far, was that Yusuf ibn-Tashufin
and his generals were semi-barbarians, and the Malikite
 THE ALMORAVIDS IN SPAIN 87
jurists mere narrow-minded bigots, with the result that the
sparkle and splendour of the culture of al-Andalus was
changed to gloom, while the poets and other writers were
denied free expression. There are a number of facts which
fit in with this view, but on the whole i t appears to be too
one-sided, though the point is another of those which re-
8

quire further study. For one thing there is an element of

"class war" involved. The dominant class of the previous
period, which has here been called the Arabo-Andalusian
aristocracy, had lost its power to the Almoravid dynasty col-
laborating w i t h the Malikite jurists, and having the favour of
the common people (who at first were probably more con-
tented than during the period of party kings). Our informa-
tion about the Almoravid rule, however, comes mostly from
members of the former dominant class—the very class for
whom life had become difficult. Yet even i f secular poets
could find few patrons, the decorative arts are now known
to have been flourishing, and the same seems to be true of
popular forms of poetry and song.
I t would further appear that i t was during the Almoravid
period that the Muslims of Spain first became fully con-
scious of the distinctive character of their religion and reli-
gious community. Up to this time Spanish Islam had been
often, perhaps mostly, a formal and official religion, accepted
as a matter of course, but without any burning enthusiasm.
Now for many men i t became a matter of intense inner con-
viction. I t was doubtless because of this emphasis on Islam
as a religion that the jurists made life difficult for Jews and
Christians. I n part the new awareness of Islam may have
been a response to the growing self-awareness of the Chris-
tians. Such opposition, too, as there was to poetry and belles-
lettres may have been because they were secular and Spanish
and not sufficiently Islamic. A large area of common culture
was shared by the Christian aristocracies and the Arabo-
Andalusian aristocracy. One piece of evidence for this is the
readiness of Muslims to remain (under legal safeguards) i n
the towns where they had lived after these came into the
control of the Christians.
After the effective end of Almoravid rule i n Spain i n 1145,
there was a period of much confusion before al-Andalus
88 THE ALMORAVIDS

came to be firmly i n the grasp of the Ahnohads about 1170.

These years are sometimes known as the second period of
"party kings", but the name is not altogether a happy one.
There was indeed a measure of breaking up into small states
under "petty kings", but these did not represent "parties"
(tawaif) as had been the case w i t h the rulers of the small
states which appeared when the Umayyad caliphate broke
up. After 1145, too, some of the rulers of small states ac-
knowledged the suzerainty of the Almohads, others that of
various Christian kings. For the purposes of a survey such as
the present i t is sufficient to regard the Almohad period as
beginning i n 1145 w i t h their first intervention i n the affairs
of al-Andalus.
 8
T H E BERBER EMPIRES, T H E A L M O H A D S

1. Ibn-Tumart and the Almohad Movement

There are various similarities between the Almohad and

Almoravid empires. Both came into existence i n north-west
Africa, and then later included al-Andalus i n their territories.
Both were ruled by a Berber dynasty, and found their origi-
nal supporters among Berber tribesmen. Both were i n origin
religious movements, or, perhaps rather, had a religious basis.
I t was only natural, of course, that the Berbers, who sup-
ported the Almohads, should be the centuries-old enemies
of those who supported the Almoravids. The latter were no-
mads of the group of tribes known as Sanhaja, whereas the
former were mountaineers from the Atlas belonging to the
Masmuda. From the scholar's point of view, moreover, there
is far more source material about the beginnings of the Almo-
hads than about that of the Almoravids. 1

The founder of the Almohad movement is commonly

known as Ibn-Tumart (Tumart being a Berber diminutive
of his father's name *Umar). His family belonged to a branch
of the tribe of Hintata, and he was born i n a village i n the
Atlas i n the years round about 1082. As a student he visited
Cordova, then went east to Alexandria, Mecca and Baghdad.
I t is doubtful whether he heard lectures from the great i n -
tellectual leader of the time, al-Ghazali; but he probably
learnt something of the philosophical theology known as
Ash'arism i n the Nizamiyya college at Baghdad and i n Alex-
andria. I n the course of his studies and travels he became
filled with reforming zeal. As a basis for his reforms, however,
he had elaborated Islamic dogma i n a new form. I n this
much emphasis was placed on taw'htd, "unity" or rather "as-
sertion of unity", and as a result his followers came to be
known as al-Muwahhidun, "the assertors of unity", or (in
European languages) Almohads or Almohades.
 THE ALMOHADS

Ibn-Tum art's enthusiasm led him to commence preaching

to the crew and passengers of the ship from Alexandria, and
then to the inhabitants of the towns through which he passed.
The results were not altogether promising. Sometimes vio-
lent opposition was roused, and Ibn-Tumart had to leave hur-
riedly. I t was after his expulsion from Bougie (Bijaya) i n
1117 or 1118 that Ibn-Tumart met the man who was to carry
the movement to political success, 'Abd-al-Mumin. The latter
was himself a student and had had thoughts of going east,
but from this time onward he remained attached to I b n -
Tumart. After spending some time i n Marrakesh, where there
was the usual mixture of support and opposition, they with-
drew to the comparative solitude of the remote town of T i n -
melal (Tinmal, etc.). This became a centre of propaganda
for the new doctrine, which rapidly gained supporters and
was given a hierarchical organisation. About 1121 Ibn-
2

Tumart put forward the claim to be himself the Mahdi, the

divinely guided and inspired leader. Soon he had sufficient
men under him to challenge Almoravid power at the local
level. A defeat i n 1123 d i d not halt his progress. He himself
was military as well as spiritual leader, and found his death
i n battle i n 1130.
Ibn-Tumart had designated 'Abd-al-Mu min as his succes-
sor, but i t was not until 1133 that the latter was proclaimed.
A t first he had to confine himself to guerrilla tactics, but
eventually he gained sufficient support i n the mountainous
regions to venture into the plains to meet the main Almo-
ravid armies. He had somewhat the better of an encounter
near Tlemcen i n 1145, and soon after the battle the Almo-
ravid ruler died as the result of an accident. The ensuing
weakness of the Almoravid administration led i n 1147 to the
loss of their capital Marrakesh and the establishment of the
Almohads there. This was to all intents and purposes the end
of the Almoravid state.
Though 'Abd-al-Mumin had interfered i n al-Andalus as
early as 1145, he d i d not direct his main military effort after
1147 t o
taking over the Almoravid domains there, but was
content w i t h diplomatic activities. He had realised that there
were opportunities for expanding eastwards i n Africa far be-
yond the limits reached by the Almoravids. Here, however,
 SPAIN UNDER T H E ALMOHADS 91
there was a Christian threat to be met, namely, from Roger
I I of Sicily. Nevertheless, the eastern half of modern Algeria
fell to him as the result of a carefully prepared campaign
i n 1151, while a further campaign i n 1159-60 gave h i m T u -
nisia including the towns of Tunis, Cairouan and al-Mahdiyya
(the former Fatimid capital), and the coast of North Africa
as far east as Tripoli.

2. Spain under the Almohads (to 1223)

'Abd-al-Mumin: 1130-1163
Abu-Yaqub Yusuf 1: 1163-1184
Abu-Yusuf Ya'qub al-Mansur: 1184-1199
Muhammad an-Nasir: 1199-1213
Abu-Ya'qub Yusuf 11: 1213-1223

The position of al-Andalus after the loss of control by the

Almoravids i n 1145, and still more after the Almohad cap-
ture of Marrakesh i n 1147, is obscure. Effective power was
i n the hands of a number of petty local rulers. Some of these
may for a time have continued to be, in some respects, de-
pendent on the Christian kings, but more and more they
tended to acknowledge the suzerainty of the Almohad caliph.
The greatest measure of independence was shown by Ibn-
Mardanish who ruled Seville and had some control over
much of the west of al-Andalus.
The founder of the Almohad empire, 'Abd-al-Mumin,
after incorporating Tunisia and Tripolitania, began to think
once more of the Iberian peninsula and from 1162 was mak-
ing preparations for a large-scale campaign there. Before the
plans could come to fruition his death intervened, and the
son who (after a short dispute) succeeded him, Abu-Ya'qub
Yusuf, did not carry out these plans. I t was not indeed until
1171 that the new caliph attempted to take a firmer grip of
al-Andalus. There was opposition from Ibn-Mardanish, but
his death i n 1172 left his successors w i t h little choice except
to surrender Seville to the Almohads. The caliph proceeded
northwards after this, and for a time besieged Toledo; but
eventually he realised the great difficulty of the operation
9 2 THE ALMOHADS

and his folly i n prosecuting this enterprise, and therefore

abandoned i t . After this, however, the Almohads seem to
have had effective control of most of al-Andalus. I n a later
campaign Abu-Ya'qub Yusuf was able to carry the "holy war"
into the enemy's territory. Unfortunately he was wounded i n
besieging the fortress of Santarem (near Lisbon), and shortly
afterwards died from his wound ( 1 1 8 4 ) .
The first task of his son and successor, Abu-Yusuf Ya'qub,
was to recover the town of Bougie and the neighbouring part
of the Algerian littoral from a descendant of the Almoravids
who had set himself up there as an independent ruler. This
series of events shows that, though the Almohads could bring
together, when i t was required, an extremely powerful army,
the forces by which they maintained order i n normal times
were insufficient to restrain adventurers. There was usually
thus some part of his empire which needed the special at-
tention of the Almohad caliph. I t was not until 1189 that
Abu-Yusuf Yaqub was able to give attention of this kind to
al-Andalus. Various successes enabled him to get the agree-
ment of the kings of Castile and Leon to a five years' truce
i n 1190. After further operations against fortresses in Portu-
gal i n the same year and the following one, he was able to
leave al-Andalus i n a comparatively settled condition while
he attended to urgent matters i n Africa. The end of the truce
led to a further period of activity i n the Iberian peninsula.
Early i n the campaign i n July 1195 the Almohads won a
great victory over Alfonso V I I I of Castile at Alarcos (about
halfway between Cordova and Toledo). The victory was
partly exploited then and i n the following year, but the
Almohads seem to have lacked the resources to bring about
any fundamental change i n the balance of power between
Christian Spain and al-Andalus.
The Christians, on the other hand, were goaded into i n -
creased activity by this reverse, which came precisely at the
time when they thought the reconquest was moving smoothly
forward. Bishops and archbishops played an active part i n
ironing out the differences between political leaders, patch-
ing up quarrels, and removing mutual suspicions. A crusade
was preached not only i n Spain but also beyond the Pyrenees,
and this brought many reinforcements. These preparations
 SPAIN UNDER T H E ALMOHADS 93
were made easier by the truce which was signed after the
battle of Alarcos, and by the lack of vigour of the new Almo-
had caliph, Muhammad, who succeeded to his father i n 1199.
For a time the situation was stable, with the Almohad do-
minion i n Spain apparently at its zenith; but when the Chris-
tians eventually moved to the offensive it was soon obvious
that a decisive shift had taken place i n the balance of power.
I t was in July 1212 that a combined force from Leon, Castile,
Navarre and Aragon moved south from Toledo and met the
Almohads at Las Navas de Tolosa. The Almohads suffered
such a defeat that their power i n Spain was virtually broken
for good and all, though this did not become manifest t i l l
after 1223 owing to the Christians being once again dis-
tracted by internal problems. The caliph Muhammad died,
perhaps accidentally, i n 1213, and his son of fifteen who suc-
ceeded him, Abu-Ya qub Yusuf I I , was incapable of restoring
the fortunes of the tottering regime.
I t would be interesting to know the reasons for the Almo-
had debacle, but the matter has received little consideration,
and anything that may be said here must be of a tentative
nature. There are a few points, however, that are relatively
certain. Ibn-Tumart had grown up i n the Almoravid empire,
and from one point of view his doctrines are seen to be d i -
rected against the Almoravids. The Almoravid regime was
closely linked with the Malikite jurists for whom jurispru-
dence meant knowledge of the details of law as applied i n
practice, that is, knowledge of what recognised authorities
had decided. Ibn-Tumart, however, held that the jurist must
be able to derive his decisions about the application of the
law from first principles, that is, from actual texts i n the
Qur'an and the Traditions, or from the consensus of the Mus-
lims. Another important matter was Ibn-Tumart's claim to be
the Mahdl (the divinely-sent leader and restorer of order).
Though this was a Shi'ite rather than a Sunnite belief, Ibn-
Tumart was doubtless thinking not of any tie-up with the
Shfites i n the east, but of providing a focus for the deep-
seated Berber reverence for saints and holy men (of which
the popular respect for marabouts is evidence).
A regime with a dogmatic basis of this kind was bound
to oppose the class of Malikite jurists who had had so much
94 THE ALMOHADS

influence under the Almoravids. Yet they do not seem to

have provided themselves w i t h any alternative body of j u -
rists, though some of those who supported them were Zahir-
ites (cf. p. 112i. below). Elsewhere Shfite leaders like the
Fatimids claimed to be able to give authoritative decisions
on points of law, though they seldom i n practice produced
any extensive legal novelties. Ibn-Tumart does not even seem
to have claimed so much. On certain points he insisted on a
strict carrying out of the prescriptions of the Qur'an and the
Traditions, but i n general he seems to have been content w i t h
the law as currently administered. Thus there was no basic
conflict between the new doctrines and current legal prac-
tice such as to exclude the existing jurists automatically. The
Almohads d i d indeed claim the title of caliph for themselves,
and abandoned even the formal acknowledgement of the
'Abbasids i n Baghdad; but the Malikite jurists were presum-
ably not greatly concerned about the 'Abbasids. More and
more, then, the corps of Malikite jurists continued to exercise
their functions under the Almohads i n al-Andalus.
The further fact that the Almohads had at times to make
concessions i n order to retain the goodwill of the jurists is
perhaps a pointer to the most serious weakness of the Almo-
hads—the lack of popular support. The Almoravids had re-
moved the old Arabo-Andalusian aristocracy from power and
found their support i n the Malikite jurists and the common
people. The Almohads may to some extent have had active
support from the dispossessed aristocracy; but they had little
from the populace once the momentary feeling of relief from
Almoravid abuses had passed. The figure of the Mahdi which
had appealed to the Berbers had probably little attraction
for the inhabitants of al-Andalus, just as i t may have had
little attraction for the Arabs i n the eastern regions of the
Almohad empire i n North Africa. I f this is so, i t means that
the Almohad empire was first and foremost a military state
requiring only the minimum degree of consent from the per-
sons ruled. There was no enthusiasm among the people, and
no sense of being associated with the ruling 61ite i n a mo-
mentous cause. I n the difficult days after 1223 at least one
member of the ruling family abandoned his allegiance to the
doctrines of Ibn-Tumart; so the state of affairs among the
 RECONQUISTA (1223-1248) 95
rank and file may be imagined. This more than anything else
seems to have brought about the decline of the Almohad
regime.
A contributory factor was doubtless the Berber origin of
the ruling elite. This meant that the figure of al-Mahdl had
probably little attraction for the non-Berber Muslims of al-
Andalus; the basic hope of the native Iberian stock seems to
have been for direct supernatural intervention. Again, there
3

was apparently an embryonic form of national or racial

awareness, perhaps chiefly for linguistic reasons. The Almo-
hads were at first successful in gaining the allegiance of the
numerous Arabs i n eastern Algeria and Tunisia; but i t would
be only natural if, after a time, the differing fundamental
orientations of the linguistic groups caused political reper-
cussions.

3. The Progress of the Reconquista from 1223 to 1248

The Almohad caliph who died i n 1223 left no son, and dy-
nastic quarrels ensued which virtually ended resistance to the
Reconquista i n al-Andalus with one exception. I n North Af-
rica, despite the quarrels within the ruling family, something
of the Almohad empire continued to exist, but i t was gradu-
ally whittled away as subordinates made themselves inde-
pendent and new states were established, until it became
finally extinct i n 1269. I n al-Andalus some of the Almohad
rulers retained a measure of authority in limited areas for a
few years after 1223, but the central administration had
disappeared.
I n this period of confusion a descendant of former princes
of Saragossa seemed for a time to be gaining a notable
measure of success i n the east and south of al-Andalus. After
the union of Leon and Castile i n 1230, however, the Chris-
tians once more took the offensive, and he was defeated more
than once in battle and finally assassinated. After this no sin-
gle Muslim leader had much support, though some preserved
a precarious measure of semi-independence for two or three
decades. The leader of the Christian offensive was Ferdinand
I I I , king of Castile from 1217 and of Leon also from 1230
96 THE ALMOHADS

(until his death i n 1 2 5 2 ) . I n a series of campaigns from 1231

onwards he finally conquered for the Christians the very
heart of al-Andalus. The outstanding events were the occu-
pation of Cordova i n 1236 and of Seville i n 1248. After that
"mopping-up operations", especially i n the east, may be said
to have taken twenty years. By the end of this period Muslim
rule i n Spain had disappeared with one exception.
The one exception—the one gleam of light for the Muslims
—was the appearance of the Nasrid kingdom of Granada.
About 1231 a man of Arab descent (from Medina), Muham-
mad ibn-Yusuf ibn-Nasr, founded a little state for himself
round Jaen, and then in 1235 seized Granada and made i t
his capital. He kept his head above water by careful diplo-
macy—getting Christian help against local Muslim rivals, and
then North African Muslim help against the Christians. The
geographical features of this small kingdom, together with
other factors, enabled i t to maintain itself for two and a half
centuries.
Apart from this small, though important exception, how-
ever, the fall of the Almohads meant the end of Islamic Spain.
 9
C U L T U R A L GREATNESS
I N POLITICAL DECLINE

1. Poetry

The poetry which under the Spanish Umayyads had been

establishing itself i n new soil, growing and building up en-
ergy, burst into flower i n the late tenth and the eleventh
centuries. This was a period of political division and insta-
bility, but poetry does not immediately rise and fall w i t h
every fluctuation of political fortune, and the fact that under
the "party kings" there were several courts, each rivalling the
others i n the patronage of arts, gave scope for many poets
to prove their talents and reap their rewards.
There was no lack of talent. The most celebrated of all
Andalusian poets was Ibn-Zaydun ( 1 0 0 3 - 7 0 ) who expressed
w i t h much tenderness and delicacy his unhappy love for the
princess Wallada, herself a poetess. The severe theologian
Ibn-Hazm ( 9 9 4 - 1 0 6 4 ) also composed a treatise on love, The
Ring of the Dove, illustrating each of the aspects of love and
the experiences of lovers with verses by himself and by oth-
ers. Particularly distinguished was the court of the 'Abbadids
i n Seville, where the tone was set by princes who were
themselves gifted poets, by al-Mutadid ( 1 0 1 2 - 6 9 ) , a poet
of vigorous spirit capable at his best of coining some striking
similes, and even more by his son al-Mu'tamid ( 1 0 4 0 - 9 5 ) .
Al-Mutamid gave fine expression to the fashionable themes
of his time before he was reduced to abject captivity by the
Almoravids, and poured out his heart i n pathetic laments. I n
his service also was a close friend whom he advanced to high
office but who eventually fell out of favour and was put to
death, Ibn-'Ammar ( 1 0 3 1 - 8 3 ) , whose verses have a certain
grandeur and resonance; and to his court came Ibn-Hamdis
( 1 0 5 5 - 1 1 3 2 ) when driven out of Sicily by the Normans.
This golden age of Andalusian poetry is generally held to
98 CULTURAL GREATNESS

have come to an end with the eleventh century, and this is

true i n the sense that no significant new upsurge took place
thereafter; but the models then created continued to be imi-
tated with great skill and refinement for several centuries.
Fine poets continued to thrive under the Berber dynasties,
not least i n the provinces, the most distinguished being I b n
Khafaja of Alcira ( 1 0 5 0 - 1 1 3 9 ) , particularly renowned for his
descriptions of gardens.
I t is tempting to connect this flowering of poetry w i t h life
in the secular, tolerant principalities of the Reyes de Taifas,
"turbaned Italian republics" as they have been dubbed, or
rather with the leisure of their courts. I t is also easy to ascribe
its subsequent loss of vitality to the reactionary and repres-
sive character of the Berber dynasties, whose rulers are often
depicted as semi-barbarians incapable of appreciating the
subtleties of the Andalusian mind. There is indeed a measure
of truth i n these assertions, but against their unqualified ac-
ceptance must be set the facts that life under the "party
kings" was not without its seamy side of base intrigue and
gnawing insecurity, that the successors of Yusuf ibn-Tashufin
were quick to adopt Andalusian attitudes and offer their pa-
tronage to poets, and that poetry did not in fact significantly
alter its character as one dynasty succeeded another.
No more i n this age than in any other can poetry be neatly
explained i n terms of political and social conditions. W i t h
perhaps nothing more precise to stimulate i t than the bounty
of nature and the Umayyad Caliphate's patronage and prom-
ise of grandeur, Andalusian poetry developed along lines that
quickly became stylized, and its life was thus protracted even
when very different times had come not only upon Spain
but upon the whole of Islam, times when three Caliphates
had gone into decline, when enemies were threatening from
without and orthodoxy was reacting w i t h i n . 1

I t is in fact difficult to imagine any society where the dis-

tinctive poetry of al-Andalus could for long be anything but
a conventional or escapist expression. Its main themes were
those of fastidious, luxurious pleasure-seeking. Other themes
honoured i n Arabic literature—the gnomic, the ascetic, the
mystic—did have their spokesmen, but they were i n poor
voice. Instead, we read of drinking parties held on the river
 POETRY 99
at night, or i n a grove or flowery meadow either in the cool
of the evening or at dawn "when night washes off its kohl i n
the morning dew"; of fawn-like ephebes or girls slim-waisted
and round-hipped inducing inebriation by the wine they
pour, by the glances they give, by the kisses they yield;
while to the accompaniment of a lute a slave-girl sings or
another dances, finally slipping out of her dress to appear
"like a bud unfolding from a cluster of blossoms".
2

There was refinement as well as sensuality i n these pleas-

ures. The Andalusian poets showed minute and lively care
i n every object of beauty, and the ideal expressed by I b n -
Khafaja was not of a man so sturdy or stoic or staid as never
to lose a night's sleep because of love, but of one reacting
immediately to every stimulus, angered or depressed by ad-
versity, but also trembling w i t h joy at the sight of beauty,
"like the branch of a ben-tree wet with dew, which the
blowing of a gentle breeze bends this way and that". This
3

refined sensuality is manifested i n the special interest that

Andalusian poets took i n nature and i n love.
Nature provided the idyllic background for bacchic scenes
and lovers' meetings, but i t was then only vaguely sketched
in. I t was usually single items that came i n for minute ob-
servation—dusk, the wavelets on the surface of the river, and
especially single flowers, the description of which was so
popular that i t filled some anthologies and was recognised as
a genre on its own. Flowers especially were often somewhat
artificially depicted as arrangements of gems—the nenuphar
as white pearls w i t h a black bead in the centre, the w i l d
jasmine on its green stem as a yellow hyacinth on an emerald
rod; but there were also comparisons of great variety and
ingenuity, often involving personification, the flowing river
being seen as the tremulous flanks of a thinly clad dancer,
the night as a negro king with the moon for a diadem and
Gemini for ear-rings. Nor did Andalusian poets fail to project
emotion into nature, sensing i n the cooing of doves and the
sighing of the breeze expressions of sympathy with parted
lovers.
I n love-poetry too the Andalusians ranged very widely,
from consuming passion to dalliance, from expressions of ab-
ject submission to one's mistress to mock adjudication on
100 CULTURAL GREATNESS

compensation due for the wounds of love, from the uncom-

plicated sensuality of an Ibn-Khafaja—

My hands travelled about his body

Now to the waist and now to the breast,
Descending to the Tihama of his flanks
And rising to the Najd of his breasts.
4

—to Ibn-Hazm's protestation that he preferred to meet his

beloved i n a dream lest i n reality the touch of his hand
should make her fade away. 5

Yet the range does not extend significantly beyond the

sensual, no matter how refined. I n The Ring of the Dove
Ibn-EIazm did reflect the (ultimately Neoplatonic) view that
love is the reunion of the two halves of a spirit created as
one sphere, but the recognition was always through physical
attraction. "Platonic" love i n this context is related to a pe-
culiar psychology of ambiguous chastity, whereby the erotic
orientation was a morbid perpetuation of desire since, i n
6

physical love, fulfilment and satiety must coincide. Woman

is said to have had much freedom and to have been much
exalted i n Andalusian society, but i n poetry i t was always
her physical charms that were sung, the only character trait
mentioned—though chastity i n the man was sometimes
praised—being her caprice and cruelty i n denying her lover.
One passage is quoted by Henri Per&s i n which ar-Radl,
7

the son of al-Mu tamid, told his beloved: "Thou art beautiful
both physically and morally" but i t is also a passage i n which
occurred a striking—and rare—indication of Christian influ-
ence, for the beloved was also spoken of as "an angel".
The question is thus raised of a possible kinship between
certain attitudes observable i n Andalusian and i n European
literatures. Peres more than any other has meticulously ex-
amined the poetic production of the eleventh century. He
has made much of instances of personification of nature as
against the Arab tendency to de-animate discussed i n earlier
chapters. He has seen as distinctive a number of features of
which he judges the fundamental one to be: a love of nature
which, despite some artificiality of expression, is rooted i n
reality; and an underlying melancholy which gives the Anda-
 POETRY 101
lusian poets a preference for half-tints. They prefer spring
to summer, evening and night to noon. Even when they are
among boon companions, one feels they prefer silence and
solitude. Perhaps because of the instability of life, they feel
that no pleasures are wholly pure, and "a disquiet—rare
among Orientals—troubles every act that is a tribute to life".
Their humble attitude to woman and their conception of love
are almost Christian. A l l i n all, "for the virtues of strength
and of action—which constitute the fitting ideal of Arab so-
ciety i n Islam—they substitute qualities of mildness, of hu-
mility, of tenderness, of reflectiveness and of musing. They
tend to become more fully human by developing all faculties,
by allowing, perhaps, precedence to the heart over the mind
and w i l l . "
Indeed Peres sees i n the Andalusians such a preponder-
ance of Ibero-Roman blood as to make them, despite the
presence of undeniably Oriental elements, a continuation of
the aboriginal people; and to the Judaeo-Christian element
he ascribes a major role i n the development of their dis-
tinctive features.8

Impressive and instructive as Peres' study is, i t is difficult

to agree with his conclusions, or at least with the weight he
gives to the distinctiveness of Andalusian poetry. Substan-
tially, this poetry is one w i t h that of the Islamic East. The
Andalusians never lost their interest i n the literary produc-
tion of the East. Their habit of characterising their own poets
as "the Mutanabbi" or "the Ibn-ar-Rumi of the West" indi-
cates deference to the standards of the East; Ibn-Hazm even
complained that he would have been better appreciated i f
he had not been born i n Spain. I n fact, Andalusian themes
all grew naturally out of classical poetry, and very few of the
features singled out by Peres are without echo elsewhere i n
Islam. I n expression also the Andalusians yielded to none i n
their search for recherche images. The taste for rhetorical
word-play was so deeply ingrained that when al-Mutamid,
languishing i n fetters, heard that two of his sons had been
killed, i t was i n bitter puns on their names that he gave voice
to his undoubtedly genuine grief. A n d i n the search for ever
more extravagant ways of expression, neither artificiality nor
bad taste were always avoided, so that a poet as refined and
102 CULTURAL GREATNESS

discriminating as Abu Amir ibn-Shuhayd ( 9 9 2 - 1 0 3 5 ) could

yet write:

Such was my kissing, such my sucking of his mouth

that he was almost made toothless.

Yet the distinctive lyricism is there, and has been sensed

by many since Baron MacGuckin de Slane remarked on i t ; 9

but i t lies i n the preponderance of certain themes, i n emo-

tional overtones, perhaps also i n a somewhat greater cohesion
of mood and homogeneity of imagery, as i n Ibn-HiafajVs
magnificent description of a lofty and forbidding mountain 10

or i n these lines on a party of horsemen, i n which all the

metaphors are derived from water:

On coursiers like torrent courses did they ride, and tied

To their dun lances the likes of clear blue waters;
Into their scabbards they deposited streams; and chose
For helms, bubbles out of the foam;
They were clad in coats of mail that were pools
Quivering—except over their shoulders. 11

What is significant is that where Andalusian literary practice

diverges from that of the East, i t moves i n a direction ac-
ceptable to European taste.
Miscegenation does not by itself explain this phenomenon,
and as has already been seen the Muslim conquerors d i d not
find in Spain an advanced culture worthy of imitation. But
the indigenous population had at least a folk-literature, and
i n the society that evolved—a society which could maintain
few barriers between muwallads and Mozarabs who were
often close kinsmen, a society that had so adapted itself to
the co-existence of different cultures that, despite different
calendars, some popular Muslim festivals were made to co-
incide with Christian holidays—this folk-literature must have
mingled and fused with that of the newcomers. There is i n
fact a thirteenth-century text which asserts that the early
Andalusians sang "either i n the manner of the Nazarenes, or
in that of the Arab camel-drivers". I t is not difficult to
12

imagine that once hybridisation had taken place at the popu-

 POETRY 103
lar level some of the attitudes that ran through the original
folk-literature worked themselves by osmosis even into the
convention-ridden compositions of the £lite. 13

Some such process is particularly likely i n the development

of the strophic forms of poetry, the muwashshah and the
zajal, which are the undisputedly original contribution of al-
Andalus to Arabic poetry. The muwashshah i n its commonest
form is a poem of five or more stanzas, w i t h an initial couplet
which may have been used as a refrain but which i n any
case provided a recurring element, i n that each stanza there-
after would consist of three lines w i t h a rhyme of their own
followed by two lines reproducing the rhyme of the initial
couplet. The rhyme scheme would thus be:

AA bbbAA(AA) cccAA(AA) etc.

but many elaborations and variations of this scheme were

also worked out. The final couplet, called the kharja, was to
be the "punch line" of the entire poem, and early sources on
the theory of muwashshah specify that i t ought to be com-
posed first, but i t could also be borrowed. Again according
to these early sources, the kharja ought to be snappy and
spicy, and ought to be composed not i n classical Arabic but
i n the colloquial idiom or i n Romance. I n recent years, some
such kharjas have i n fact been discovered; they are i n a mix-
ture of colloquial and Romance, and almost always i n the
form of words put into the mouth of a woman.
The zajal is entirely i n the colloquial, and usually has a
simpler rhyme scheme, w i t h a single line at the end of each
stanza reproducing the rhyme of the initial couplet—i.e.:

AA bbbA(AA) etc."

Some Andalusian muwashshahs were composed i n metres

identical, or nearly so, w i t h those of classical Arabic poetry.
Some on the other hand struck Ibn-Sana-al-Mulk (d. 1212)
—the propagator of the genre i n the East—as having, when
read, no metre that the ear could detect although presum-
15

ably when sung they were made to fit into a rhythmic pat-
tern by arbitrary contraction or lengthening of the syllables.
104 CULTURAL GREATNESS

Between these two extremes are many that can be scanned

i n accordance w i t h a syllabic system possibly derived from
the quantitative classical metres, but noticeably different
from them; they raise intriguing problems on which the last
word has yet to be said. 16

The muwashshah and the zajal are but two of a number of

non-classical verse-forms which came to be used i n Arabic-
speaking lands. One of these at least—the
17
mawdliyd—is
known to have been practised i n Iraq as early as the eighth
century A . D . There are also strophic compositions ascribed to
the Baghdad poets Abu-Nuwas ( d . c. 8 0 3 ) and Ibn-al-
Mu'tazz (d. 9 0 8 ) which i t is not impossible to regard as
forerunners of the muwashshah. 18
There is no positive evi-
dence, however, that these solitary examples served as mod-
els for any later poets. Certainly i n the East the development
of strophic poetry remained very largely confined to folk-
literature until the Andalusians brought the muwashshah and
the zajal to the point where they excited the admiration and
then the emulation of cultured Muslims elsewhere, so that
the pre-eminence of al-Andalus i n this field is uncontested.
The earliest known Andalusian muwashshahs are from the
eleventh century, but their "invention" is ascribed to a poet
of Cabra who died early i n the tenth. "Invention" here need
mean no more than that this poet was first to give the mu-
washshah a form acceptable to men of letters, for the char-
acter of the kharfa bespeaks a popular origin—possibly songs
sung by native women entertainers before Arabic-speaking
audiences. Garcia G6mez has shown how an Arab poet whose
ear had caught a snappy line from a popular song might well
have written the first muwashshah as a setting for i t , but
1 9

i t seems likely that the indebtedness of this poet must have

been more extensive, so different is the entire structure of
the muwashshah from that of classical Arabic poems.
The genre acquired a great vogue, and was practised by
many of the best poets, some—like Abu-Bakr Muhammad
ibn-Zuhr (d. 1110/1)—building their literary reputations on
their mastery of i t . I t lent itself well to the prevailing taste
for artifice. A t the same time, w i t h the stanza rather than
the line as its unit, its patterns both adaptable and challeng-
ing, the form produced, when handled by masters, some of
Plate 10. Arched entrance to the Sala de los Embajadores in the
Alcazar, Seville, Mudejar style.
Plate 11. Some of the roofs of the Royal Palace of the Alhambra,
Granada.
Plate 12. Slender columns in the Patio de los Leones at the Alham-
bra, Granada.
Plate 13. Detail of stucco on the walls of the Patio de los Leones
at the Alhambra, Granada.
Plate 14. Tile mosaic and stucco wall decoration in the Salon de
los Embajadores, Alhambra.
Plate 15. Stucco wall decoration in the Patio de los Leones, Al­
hambra, Granada.
Plate 16. Pool of the Portal and the Torre de las Damas in the
Alhambra.
Plate 17. The intersecting binding arches of a Dome at the
Mosque, Cordova; a style which influenced later Italian Renaissance
architecture.
 PROSE LITERATURE AND PHILOLOGY 10$

the most charming of Andalusian love-poems and some of

the most delicate descriptive pieces.
Zajal did not make its appearance until quite late, and its
principal exponent Ibn-Quzman or Abenguzman ( d . 1160 or
1 1 6 9 ) claimed most of the credit for having perfected i t .
2 0

That is not to say that i t had not had a lengthy history before
then: as a truly popular form, i t would have been beneath
the attention of the men of letters who might have recorded
it. Ibn-Quzman was the first whose verve and coarse good
humour earned h i m acceptance among men of letters, and
hence into posterity.
So i t was that i n Spain, alone among Muslim lands, the
vigorous spirit of the common people breached the wall of
convention erected by the classicists.

2. Prose Literature and Philology

Linguistic and literary studies have always been particularly

closely related i n Arabic. The earliest grammarians and lexi-
cographers sought to establish what early, pure Arab usage
was, and for this purpose their raw material was the poetry
of the Ancients, so that their works were studded w i t h cita-
tions used not only i n illustration but as authority. Indeed i t
was they who stimulated the recording of such poetry, and i t
was an extension and development of their interests that led
to the compilation of anthologies, the collection of literary
anecdotes, the writing of commentaries on literary texts, the
growth of literary criticism itself.
The Muslims i n Spain apparently had a distinct educa-
tional system—a more rational one than that of the East, as
Peres has been quick to point out —whereby the language
21

was mastered first and the religious "sciences" studied after-

wards. But the "sciences" themselves were very much i n the
tradition of the East. They were first brought over, as has
been mentioned, by such learned men as 'al-Qali, and the
link was maintained by a constant traffic of scholars and of
scholarly works, a two-way traffic that emphasises the essen-
tial unity of Islamic culture.
Thus Arab grammarians i n al-Andalus as elsewhere made
io6 CULTURAL GREATNESS

their contributions to a common fund of knowledge by writ-

ing commentaries on standard works composed i n the East.
The lexicographer Ibn-Sida (ioo6?-66) compiled among
other works two great dictionaries of no small renown alike
i n the East and i n the West.
Books of adab--collections of miscellaneous items often
quite encyclopaedic i n their range, for the genre embraced
everything of interest to the cultured man—were produced
i n Spain. One entitled The Light of Kings (Siraj al-Muluk)
by A b u Bakr at-Turtushi, also known as Ibn-Abi-Randaqa
( 1 0 5 9 - 1 1 3 0 ) , consists almost entirely of anecdotes on royal
behaviour, and could perhaps be assigned to the special class
of "mirrors for princes". Another by Yusuf ibn-ash-Shaykh
of Malaga ( 1 1 3 2 - 1 2 0 7 ) dealt w i t h a wide variety of sub-
jects i n alphabetical order, and came to be used as a manual
of general culture.
There also began to appear anthologies where Andalusian
poets had pride of place, and i n particular anthologies on a
single theme often reflecting the Andalusians' delight in na-
ture; the earliest now extant is Abu-l-Walid al-IJimyarfs
(1026-c. 1 0 8 4 ) , which is concerned entirely w i t h spring and
w i t h spring flowers. Ibn-llazm's The Ring of the Dove, al-
ready mentioned as an anthology of poetry and i n fact in-
spired by an Eastern anthology, Abu-Dawud al-Isfahanfs
Kitab az-Zahra, is at the same time a treatise on love of some
considerable originality and of unsurpassed reputation.
I t was w i t h the declared aim of doing full justice to Anda-
lusian genius and of breaking away from the subservience to
Eastern models that Ibn-Bassam (d. 1147) composed his lit-
erary history, well-named The Treasury (Adh-Dhakhira),
for i t is to this day one of our richest treasure houses of lit-
erary information. A younger contemporary, al-Fat'h ibn-
Khaqan, drew heavily upon The Treasury i n his two literary
histories, which have nevertheless earned themselves a dis-
tinguished reputation for the excellence of their prose style.
Finally, not out of rivalry w i t h the East but i n response to a
Berber challenge issued at the court of the Almohad ruler
al-Mansur, ash-Shaqundi ( d . 1231) composed an Epistle
(Risala) i n defence of Andalusian culture remarkable for its
measured argumentation, its acumen, and its elevated style.
 PROSE LITERATURE AND PHILOLOGY 107

Although the stylistic excellence of both al-FatTi ibn-

Khaqan and ash-Shaqundi has been noted, fine prose as such
—prose aimed primarily at creating aesthetic effect rather
than conveying information—was a literary activity that grew
out of official correspondence. The secretaries of princes were
no mere amanuenses, but senior "civil servants", men of re-
sponsibility and prestige whose skill w i t h words was deemed
not unworthy of the attention of literary critics.
Eventually prose writing was extended beyond official
correspondence to epistles on various themes, descriptive
pieces, imaginary debates between the sword and the pen or
between various kinds of flowers.
Once again, the trail had been blazed i n the East, where
a number of writers—of whom the most gifted was al-Jahiz
(d. 869)—had developed a lively and direct style of prose;
these had their admirers and imitators i n al-Andalus. The
most renowned was the poet Ibn-Zaydun: his two epistles-
one, written when he was i n disgrace, pleading for the rulers
•clemency; the other, addressed to his rival Ibn-'Abdus, i n
satirical vein—are fine pieces of craftsmanship, vigorous, and
full of learned allusions.
Before long, the taste for rhetorical artifice had invaded
prose as well as poetry. I n East and West alike, a cleverly
ornate rhymed prose, aimed at commanding the reader's ad-
miration rather than engaging his sympathetic emotions, be-
came the rule i n all writing laying claim to literary distinction.
This kind of prose found a particularly suitable outlet i n
the maqdma. This was a short story told with consummate
i f self-conscious artistry, almost always of a fraud perpetrated
by a witty and likeable rogue to secure a free meal or some
other modest prize. I t was reputedly invented in the East by
Badf-az-Zaman al-Hamadhani ( 9 6 9 - 1 0 0 8 ) , but once again
the statement need mean no more than that he was the first
to give i t a polished literary expression. The form and the
name of the maqdma 22
are consistent with its having grown
out of a kind of anecdote which i n the ninth century told of
some beduin, chaste of speech and austere of manner, ha-
ranguing an assembly of notables on the pristine virtues
threatened by a fife of luxury, but which a century later had
io8 CULTURAL GREATNESS

reduced the hero to the role of an uninvited guest insinuating

himself among his betters to cadge a meal.
Badi-az-Zaman displayed some ingenuity and finesse i n
the frauds he related, and some lively observation of human
behaviour; but his own interest was i n the opportunities the
maqdma offered for fine diction and clever expression, and
i t is to these that the genre owed its success. Al-Hariri of
Basra ( 1 0 5 4 - 1 1 2 2 ) went further i n that his hero hardly ever
used any means other than his eloquence to bemuse and then
to dupe his audience; the narrative thus became no more
than a slender frame for verbal tours de force of breath-
taking brilliance.
The writings of Badf-az-Zaman—epistles as well as maqd-
mas—were quickly known, much admired, and soon imitated
i n Spain. No less favoured were those of al-IJariri, which
some Andalusians are reported to have heard from the lips
of the author himself; i t was i n fact an Andalusian, Abu-
1-'Abbas Ahmad ash-Sharishi—i.e. of Jerez—(d. 1222) whose
commentary on the maqdmas of al-Hariri came to be ac-
cepted as the standard one throughout Islam. Maqdmas are
known to have been composed i n al-Andalus from the late
eleventh century onward, probably the outstanding contribu-
tor to the genre being Abu-'t-Tahir Muhammad at-Tamimi
as-Saraqusti al-Ashtarkum—i.e. of Estercuel—(d. 1 1 4 3 ) , who
in the composition of his fifty maqdmas imposed upon him-
self onerous arbitrary standards of verbal ornamentation, ap-
parently i n emulation of what the Eastern poet A b u - V A l a
al-Ma'arrl ( 9 7 3 - 1 0 5 7 ) had done i n his Luzumiyydt.
23

Concerned as i t is less w i t h a story to be told than with a

writer's virtuosity, the maqdma is nevertheless—if we except
the short anecdote—the only narrative form developed i n
Arabic literature, prose or verse. A n early translation of the
fables of Bidpai by the Persian Ibn-al-Muqaffa' (d. 7 5 7 ) had
gained acceptance as a literary classic. Persian tales in a
translation of similarly high quality appear to have formed
the core of the Arabian Nights, but the anonymity and in-
different language of the later accretions bespeak a popular
origin. The masters of Arabic prose i n fact left story-telling
almost entirely to the common people, and i t is consistent
w i t h the atomism already noted in Arab literary practice that
 PROSE LITERATURE AND PHILOLOGY 10Q

though they polished their means of expression to a diamond

brilliance, they evolved no epic and—until modern times—no
drama and no novel, no genre requiring sustained invention
i n accordance with a unified conception. The one exception
of note i n the East has been a prose work by the poet al-
Ma'arri, the Risdlat al-Ghufrdn (The Epistle of Forgiveness)
describing a visit to heaven and encounters there w i t h poets
of previous ages; its superficial similarities with the Divine
Comedy have attracted incidental attention i n the debate on
whether Dante owed any part of his inspiration to Islamic
sources. 24

I n Muslim Spain, the picture was not substantially differ-

ent. Tales and legends—some of which have been preserved
—were part of the heritage of the common people, and these,
unhampered by any conscious attachment to a hidebound
tradition, freely picked threads from whatever lore offered
delight and instruction, threads of Arab or Greek or local
origin, and as freely wove them into the patterns suggested
by their cross-fertilised imagination. But no literary narrative
developed, although one Ibn-al-Kinanl, an eleventh-century
physician, was the author of a book entitled Muhammad wa-
Su'dd, now lost, but which would appear to have been some
kind of a romance.
There are, however, two Andalusian works of the greatest
interest i n this connection, even though the narrative element
i n both was of secondary importance i n the author's mind.
The first is the Risdlat at-Tawdbiwa-z-Zawdbi by Abu-
'Amir ibn-Shuhayd ( 9 9 2 - 1 0 3 5 ) . I n this the author describes
a journey i n supernatural realms where he meets the tawdbi,
the genii (jinn) who—according to pre-Islamic belief—in-
spired and were individually attached to poets. He has inter-
views with the genii of three pre-Islamic poets and a number
of "moderns". I n most instances there is no more than a brief
dialogue i n which the particular tdbi hears, usually with ap-
proval, an ode after the manner of his own poet but com-
posed by Ibn-Shuhayd. But there are also passages which
trace the use made by successive poets of particular conceits
and figures of speech; and the book ends with a discussion
of the merits of poetic compositions by a mule and by an ass,
no CULTURAL GREATNESS

and w i t h a discourse by a goose—possibly satires on unidenti-

fied literary personalities.
This is the first book of the kind known in Arabic, for i t
ante-dates by a few years al-Ma'arri's Risdlat al-Ghufrdn, al-
though there is nothing to show that al-Ma'arri knew of i t
Peres has suggested that i t was itself possibly inspired by
29

some of the Dialogues of Plato or of Lucian, but of this again

there is no positive evidence. I t is at all events primarily lit-
erary criticism presented i n fanciful form. Nevertheless the
story is told with some verve, at times w i t h a tongue-in-the-
cheek humour rare indeed i n Arabic literature; and the fact
that i t is the genii of poets and not the poets themselves who
are encountered gives opportunities for characterisation by
physical description which are only occasionally, but then
quite deftly, taken.
The second work i n narrative form is the story of Hayy
ibn-Yaqzdn by Ibn-Tufayl ( d . 1 1 8 5 ) , which is considered as
philosophy i n a later section (p. 1 1 9 ) . From the standpoint
of literary history i t may be noted that Ibn-Tufayl took over
the exact title and the broad theme—the attainment of the
highest Truth by the exercise of human faculties—of an ear-
lier work by the Eastern philosopher Avicenna or Ibn-Sina
( 9 8 0 - 1 0 3 7 ) . But whereas Avicenna's work was a philosophi-
cal discourse and the hero's name—Alive, son of Awake—no
more than an obvious symbol, Ibn-Tufayl combined i t w i t h
a popular tale concerning a boy brought up by a gazelle,
and gave flesh and blood to his speculations by depicting the
mental development of an island recluse unencumbered w i t h
traditional beliefs and untrammelled w i t h social bonds—a
forerunner of Rousseau's Entile. I n i t the unlikely marriage
of philosophy w i t h popular story-telling has given Arabic lit-
erature its most cogent and arresting narrative until modern
times.

3. The Religious Sciences and History

For the various intellectual disciplines, as for poetry, the po-

litical disorder which followed the breakdown of the Umay-
yad caliphate was a period when what had been quietly
 THE RELIGIOUS SCIENCES A N D HISTORY 111

growing burst into flower. That this should happen in these

disciplines was no doubt due i n large part to what had been
done i n the third quarter of the tenth century—in the later
years of 'Abd-ar-Rahman I I I and under al-Hakam II—in
building up libraries and encouraging scholars from the heart-
lands to settle i n Spain. The temper of the age must also
have helped, for the self-awareness and self-confidence cre-
ated under 'Abd-ar-Rahman I I I seem to have persisted even
long after the Umayyads had disappeared. I t is therefore not
surprising that i n the eleventh and twelfth centuries Spain
produced men of wide-embracing scholarship, of whom the
greatest achieved fame i n the heartlands, while at the disap-
pearance of most of Islamic Spain i n the thirteenth century
scholars from al-Andalus were able to find employment i n
North Africa, Egypt and Syria.
Far and away the most outstanding scholar of al-Andalus
in the eleventh century was Ibn-Hazm of Cordova ( 9 9 4 -
1 0 6 4 ) , whose name is sometimes europeanised by Aben-
hazam. There is a curious point about his genealogy. He
26

claimed that an ancestor of Persian descent had come to

Spain as a client of a member of the Umayyad family; and
certain later biographers accept this claim. Yet some of his
contemporaries twitted him on being really of an indigenous
Spanish family from west of Seville. He certainly felt him-
27

self identified with the Arabs i n general and the Umayyad

family i n particular, and was a bitter opponent of the Chris-
tians. One of his minor works was a treatise on Arab gene-
alogies.
His father held various prominent posts under al-Mansur
the 'Amirid and his son al-Muzaffar (d. 1 0 0 8 ) , and doubt-
less continued to be immersed i n the troubled affairs of al-
Andalus between that date and his own death i n 1012. Ibn-
Jlazm's education must have been almost complete by this
time, since the disintegration of the ruling institution began
i n 1009. There are scattered references to the family moving
from one mansion or estate to another, until in 1013 they
settled at Jativa near Valencia. By 1016 Ibn-IJazm seems
to have been involved i n politics as a supporter of the Umay-
yads, but his career as an administrator was very unsettled
and included fighting and imprisonments. I n December
112 CULTURAL GREATNESS

1023 he became chief minister to one of the short-lived

Umayyad caliphs, but seven weeks later the caliph had been
assassinated and he himself once more thrown into prison.
For a short time within the period 1 0 2 7 - 3 1 he is reported to
have held another political appointment, but before long he
withdrew completely from politics and devoted himself en-
tirely to scholarship.
His first prose work, The Ring of the Dove, 28
is a distinct
surprise coming from one who subsequently attained high
rank as a jurist and theologian, for i t is about love and lovers
w i t h copious vivid anecdotes. While i t expresses something
of his own youthful experiences and interests, i t is perhaps
best understood as arising from his interest i n the Arabic
language and as exercising his skill i n the use of i t ; the prose
is carefully composed, and elegant verses of his own are i n -
serted. This concern for language is a connecting link be-
tween The Ring of the Dove and the later theological system.
Ibn-IJazm's first instruction i n the religious sciences had
been derived from the Malikism dominant i n Spain, and was
mainly limited to considering the details of the legal system
i n its practical application; but i t was impossible to keep an
acute student from asking questions about the justification
of particular prescriptions, for i n Spain there were now both
Traditionists of great learning and ShafTite jurists, and the
latter must have lectured about the "roots" or basic prin-
ciples of the legal structure. Ibn-IIazm went deeply into the
study of Traditions, and actually wrote a large book on law
from the ShafTite standpoint. By 1027, however, he was be-
ginning to be dissatisfied w i t h the ShafTite system, and was
corning under the influence of one of his earlier professors
of literature who now introduced him to the Zahirite school
of jurisprudence, henceforward his spiritual home. Zahirism
was the creation of Dawud al-Isbahani (d. 8 8 4 ) , and made
its central principle the retention of the "plain meaning"
(zahir) of the words of the Quran and the Traditions. This
central principle was relevant primarily to the attempt to
harmonise apparently contradictory statements i n the primary
texts without falling into metaphorical interpretation (which
was condemned i n conservative circles).
The attraction of Zahirism for Ibn-IJazm must have been
 THE RELIGIOUS SCIENCES A N D HISTORY 113

great. Hitherto i t had been merely a legal school or rite, but

Ibn-Hazm now attempted on the same central principle to
establish also a system of dogmatic theology. I n this attempt
there would seem to be evolving a combination of the funda-
mental Arab conception of language (and of the relation of
language to knowledge) with a speculative tendency, which
was probably of Iberian origin (since i t may also be exempli-
fied in the philosophical achievements of the following cen-
t u r y ) . Language for the Muslim Arabs was no mere human
convention but something created by God to be appropriate
to the things to which i t refers. The Qur'an as the speech
of God must be a perfect vehicle to convey to men what he
wanted to convey. So the chief work of the scholar is to un-
derstand what God meant by the Qur'an and, secondly, what
is meant by the various sayings of Muhammad. This under-
standing, however, presupposes wide familiarity w i t h the
Qur'an and Traditions (including knowing the Qur'an by
heart), and the most scrupulous care by succeeding gen-
erations of scholars i n transmitting accurately the precise
verbal form of these scriptural texts. Such a conception was
very general among Muslim scholars and by no means con-
fined to the Zahirites, but it was the Zahirite Ibn-Hazm who
took i t most seriously and tried to work out its implications.
Ibn-Hazm seems to have had considerable influence on the
later intellectual outlook of al-Andalus, but i n theology he
had no professed followers. Some of his pupils adopted
Zahirism in law, at least for a time, and the Zahirite school
continued for a century or two, though always with only a
tiny body of adherents. Ibn-IJazm's tongue was proverbially
sharp, and he made many enemies. For a time he found ref-
uge i n Majorca, but disputes there led to his expulsion. The
closing years of his life were spent on the family estate near
Niebla (west of Seville), and he died there i n 1064.
Apart from his works on legal theory most attention has
been given by recent scholars to his book on sects. I t has been
hailed as the first work ever on comparative religion, but
this is not quite accurate. For one thing there are several
earlier heresiologies i n Arabic, and for another the aim was
polemical and not descriptive. The intermingling of religions
i n Spain presumably drove a scholar like Ibn-IIazm to think
ii4 CULTURAL GREATNESS

out the position of his own religion over against its rivals.
Altogether some three dozen works by Ibn-IJazm are still
extant, though some indeed are little more than pamphlets,
and they cover a wide range. I n addition to those mentioned
there is a work on ethics (which has been translated into
Spanish), a critique (based, according to the biographers,
on inadequate understanding) of Aristotelian logic, and an
epistle defending al-Andalus against the charge of neglect-
ing the biography of its intellectuals. 29

Because Ibn-IJazm is outstanding he was not altogether

typical. A first-rate scholar who became known i n the heart-
lands but was more i n accordance w i t h the Andalusian ethos
is Ibn-'Abd-al-Barr ( 9 7 8 - 1 0 7 1 ) . Though he studied only
30

i n Cordova, he corresponded w i t h scholars i n the East (the

heartlands) and became the greatest Traditionist of his time
i n Spain and North Africa. His studies and writings included
various cognate disciplines such as Arab genealogy and the
life and campaigns of Muhammad. A collection of biographies
of the Companions of Muhammad—it was always a Com-
panion who was first i n the chain of transmission of a Tradi-
tion—attracted much attention and prepared the way for more
complete collections.
His most original work, however, was one of which an
abbreviated title would be, The Exposition and Excellence
of Knowledge. I n method this is a work typical of the Tradi-
tionist movement, for it consists almost exclusively of
anecdotes about the sayings and doings of Muhammad and
various distinguished early Muslims, together with reports of
the views held by scholars; he never gives his own views
directly, though they may sometimes be inferred from such
statements as that all jurists and scholars, w i t h two excep-
tions, agree that analogical reasoning (qiyds) is permissible
in legal decisions but not i n theology. The choice of topics,
however, is of great interest, and includes matters connected
w i t h the Arab conception of knowledge (as just described
above), such as: the obligation to seek knowledge; the su-
periority of scholarship to piety and to martyrdom; whether
it is desirable to write down knowledge (a procedure abhor-
rent to the oral culture of the pre-Islamic Arabs); travelling
i n quest of knowledge; the respect due from pupil to teacher;
 THE RELIGIOUS SCIENCES A N D HISTORY 11$

the attitude of scholars to rulers; the sources of legal and

religious knowledge; and the types of argument possible i n
these fields. I n this he shows a tendency to move away from
the rigid Malikism of the tenth century, and indeed explicitly
condemns taqlid or the blind following of authority. I t is not
surprising to learn that he first belonged to the Zahirite
school, and that, though he later adhered to the dominant
Malikite school—rising to be qadi of Lisbon—he was thought
to have had ShafTite leanings. I n the middle period of his
long life he lived in various towns on or near the eastern
seaboard, and his death also occurred there at Jativa.
Further light is thrown on the conditions in which scholars
lived at this period by the life of al-Humaydi. He was born
i n Majorca before 1029, his father having gone there from a
suburb of Cordova, presumably because of the unsettled state
of affairs. I n Majorca al-Humaydi came under the influence
of Ibn-Hazm, and gained an excellent knowledge of Tradi-
tions and kindred subjects. He also studied under Ibn-'Abd-
al-Barr, possibly i n Cordova. I n 1056 he set out on a study-
journey to the East, which included making the pilgrimage
to Mecca and spending some time i n study there. He is also
said to have learnt Traditions in Tunisia, Egypt, Damascus
and Baghdad. Whether he returned to Spain for a time is
not clear. He accepted the Zahirite outlook of his teacher
Ibn-Hazm, and suffered on account of the general opposition
to this rite. I t was because of this opposition that he eventu-
ally settled in Baghdad. Besides some competent but not
noteworthy works on Traditions he was also prevailed on by
friends i n Baghdad to write an account of the learned men
of Spain, and this he did from memory. He died in Baghdad
i n 1095, one of the first emigrants from Spain i n an easterly
direction.
The continuing high standards of scholarship i n al-Andalus
despite the troubles of the times are vouched for not merely
by the names and biographies of several scholars during the
next century and a half and by the lists of surviving manu-
scripts but above all by the fame attained i n the heartlands by
Qadi Tyad ( 1 0 8 3 - 1 1 4 9 ) . He is probably the chief figure of
the Almoravid period. Born i n Ceuta in the African part of
the Almoravid empire, he studied in Cordova, but eventually
 CULTURAL GREATNESS

returned to Ceuta as qadi. He was promoted to a similar

post i n Granada, but after a short time there moved on ( i n
1137) t o
Cordova. The troubled situation as the end of the
Almoravid dominion approached was doubtless the reason for
his withdrawal to Marrakesh, where he died i n 1149.
He had the reputation of being the greatest Traditionist
of the Islamic west i n his day, and has left behind several
books of average quality on jurisprudence and the study of
Traditions, as well as a collection of biographies of Malikite
jurists. One work, however, known as the Shifa ("cure"),
raised him far above the average; there still exist more than
twenty commentaries on it, written from the fourteenth to
nineteenth centuries. Though the full title speaks of the
"rights" of the Prophet, i t is essentially a book i n praise of
him i n which his figure is raised to supernatural stature.
The existence of miracles performed by him is emphasised
(in opposition to the theologians who insisted that his only
miracle was the Qur'an itself); his moral perfection is de-
scribed; his preservation from error and blemish is asserted.
The book thus marks a large step forward i n the development
of the theological doctrine of the person of Muhammad,
and i t was presumably for this reason that i t attracted so
many commentators. I t is tempting to connect the general
attitude of the book w i t h the saint-worship prevalent i n
North Africa. Though lyad's grandfather is said to have
emigrated from al-Andalus first to Fez and then to Ceuta, the
family may well have been originally Berber and almost cer-
tainly had North African blood on the female side. Thus
heredity probably combined w i t h environment to foster the
outlook characteristic of North Africa.31

The MaliMsm of the Almoravid period was thus changing

from that of the late tenth century, and Malikite scholars
were including i n their scope much more than practical legal
details. As the threat from the Almohads began to appear i n
North Africa (from about 1125 onwards), the Malikites sup-
porting the Almoravids felt impelled to attack not merely
the teaching of Ibn-Tumart and his Almohad followers but
also the Ash'arite theology now dominant i n Baghdad which
was thought (not altogether correctly) to have been the
major influence i n the formation of his views. There were
 THE RELIGIOUS SCIENCES A N D HISTORY 117

particularly vicious attacks on the distinguished Ash'arite

theologian and mystic, al-Ghazali; books were written criticis-
ing him, his views were officially condemned as heretical,
and his great work, The Revival of the Religious Sciences,
was publicly burned (apparently on the order of the Qadi
lyad).
W i t h the establishment of Almohad power in Spain from
1145 onwards, the Malikite jurists ceased to have the official
support they had enjoyed under the Almoravids, but some
at least retained their appointments. As time went on the
Almohad rulers found they could not do without the Malikites
and the section of the population whose mouthpiece they
were. Though the Almohads started off with an official the-
ology, this was not necessarily connected with any one system
of jurisprudence. Abu-Yusuf Yaqub, who ruled from 1184
to 1199, is said to have favoured the Zahirites. The most illus-
trious jurist of the time supporting the regime was Ibn-Rushd
( 1 1 2 6 - 9 8 ) , who, because he became one of the world's
great philosophers, is seldom thought of as a jurist. Yet he
came of a family of jurists, himself rose to being qadi of
Seville and then of Cordova, and i n 1188 seems to have
completed an important work on jurisprudence, apparently
written for the most part twenty years earlier. The book deals
w i t h the "differences" between the various legal rites (or
schools), and pays special attention to the types of argument
used by each to justify its particular decisions. This sub-
32

ject of the "roots" of law was one originally avoided by the

Malikites of Spain, but cultivated by the smaller groups of
Zahirites and Shafi'ites, and it seems likely that it was these
groups who were the chief supporters of the Almohads in
jurisprudence.
Finally, historical writing in Spain may be mentioned
briefly. I t is not inappropriate in this chapter, since many
33

of the historical and biographical writers were also jurists.

The most important of the earlier historians, in the view of
Levi-Proven§al, is Ibn-IIayyan (d. 1 0 7 6 ) , whose work,
though surviving only i n parts, contains much valuable and
reliable information. A contemporary, Said, qadi of Toledo
(d. 1 0 7 0 ) , left a compendium of universal history (trans-
lated into French in 1935 as Livre des categories des nations)
n8 CULTURAL GREATNESS

which divides the peoples into those which cultivated science

and those which did not, and indicates the extent of Muslim
knowledge of non-Muslim cultures at this period. Many col-
lections of biographical notices were also produced i n al-
Andalus, mostly of local scholars. Practically all this historical
writing is of more concern to modern scholars for its content
than as a literary production.
There is a sense i n which the culture of Islamic Spain was
continued i n North Africa, and i t is therefore fitting to con-
clude by mentioning the name of Ibn-Khaldun ( 1 3 3 2 - 1 4 0 6 )
who, though born i n Tunis, belonged to an Arab family
which had lived i n Spain since the eighth century and had
latterly played an important part i n the life of Seville until
shortly before the capture of that city by the Christians. He
is the author of an extensive historical work whose later
volumes deal i n detail w i t h the history of the various dynas-
ties of North Africa. He has attracted much attention, how-
ever, by the Introduction or Muqaddima to this history
which is generally regarded as a pioneer work i n sociology.
Though he lived mostly i n various parts of North Africa, i n -
cluding Egypt, and only spent a little over two years i n Spain
at Granada ( 1 3 6 2 - 5 ) , his most recent translator has judged
that "his basic loyalty to Spain and its civilisation' was a
deep influence underlying his w o r k . He thus illustrates
34

one of the channels through which al-Andalus made con-

tributions to the Islamic world as a whole.

4. Philosophy and Mysticism

I t has been noted above (p. 5 7 ) that philosophy was culti-

vated i n al-Andalus i n the early tenth century by Ibn-Masarra
and his school. Though there are traces of a continuing i n -
terest in philosophical ideas, no scholar of note can be de-
scribed as a philosopher until we come to Avempace, also
known as Ibn-Bajja ( d . 1 1 3 8 ) . He was born at Saragossa,
but later lived for some years at Seville and Granada. To-
wards the end of his life he moved to the Almoravid capital
at Fez, where he was apparently poisoned by a distinguished
physician. His chief work was called The Rule of the Solitary,
 PHILOSOPHY AND MYSTICISM 119
and i n general may be said to express a moral protest against
the materialism and worldliness of the ruling classes of the
day. Because society is so corrupt, he maintained, the man
who has seen its true condition must keep himself detached
from i t , at least i n thought. There is genuine ethical passion
behind this work of Avempace, but one is bound to remem-
ber that the conservative Malikite jurists were i n control of
intellectual life under the Almoravids, and that little else
than retirement and solitude was open to Avempace. For the
pure philosopher as distinct from the historian the great i n -
terest of his work is i n his analysis of the "spiritual forms" or
ideas present i n human thought. 35

The seed sown by Avempace bore wonderful fruit after

the fall of the Almoravids and the establishment of Almohad
rule i n al-Andalus. Reasons can be discerned for the pos-
sibility of great philosophical achievements under the Almo-
hads. The founder of the Almohad movement, Ibn-Tumart
(c. 1 0 8 0 - 1 1 3 0 ) , had been a theologian and not a philosopher,
but he had favoured the more philosophical forms of theol-
ogy. I t is almost certain that he had not been a pupil of al-
Ghazali himself ( d . 1 1 1 1 ) , the great theologian who had
36

mastered the Neoplatonic philosophy of the day and, while

criticising i t vigorously, had shown that much of it was com-
patible w i t h sound theological doctrine; but he had been i n
touch with this line of thought. Moreover the intellectual
defence of Malikism under the Almoravids (as has been seen)
had included attacks on al-Ghazali as well as on Almohad
theology. Thus once the intellectual opposition came into
power there was likely to be a climate of opinion favourable
to philosophy; and this indeed happened.
The first important philosopher under the Almohads was
Ibn-Tufayl (c. 1 1 0 5 - 8 5 ) , also known i n medieval times as
Abubacer from his "father-name" (kunya) of Abu-Bakr. He
was born at Guadix near Granada, and after being secretary
to the local governor rose to the position of vizier and court-
physician to the Almohad ruler Abu-Ya'qub Yusuf ( 1 1 6 3 -
8 4 ) . His philosophy was given to the world i n a partly alle-
gorical form (perhaps to parry opposition) i n the romance of
Ifayy ibn-Yaqzdn ("Alive, son of Awake"), which has been
translated into various European languages. 37
120 CULTURAL GREATNESS

The hero of the romance, Hayy, grows up from babyhood

on a deserted island without human contact, suckled and
nurtured by a gazelle. By his own reflections on what he sees
around him he gradually works out a complete philosophical
system, including a doctrine of God, and attains some meas-
ure of mystical ecstasy. Another young man Asal now comes
to the island seeking withdrawal from the world i n order to
engage i n mystical contemplation ( i n the tradition of Avem-
pace). When the two meet and compare notes, they find that
the philosophical religion of Hayy is identical with the posi-
tion of Asal, reached by philosophical criticism of traditional
religion. Hayy is now filled w i t h enthusiasm for converting
the common people on the inhabited island to his philosophi-
cal religion, but when the two set out and attempt this they
find that the people do not want what they have to give and
w i l l not accept i t .
This story, which is told with great charm, clearly refers
i n the first place to the contemporary problems of the rela-
tion between philosophy and religion. IJayy may be said to
stand for pure philosophy, and Asal for philosophical theol-
ogy, perhaps such as that of Ibn-Tumart. The ruler of the
inhabited island is Salaman, who is said to favour the literal
meaning of texts (zahir) and to be averse to metaphorical
interpretation (tawil), and may therefore stand for the
Zahirites and other jurists supporting the Almohads (but per-
haps not for the Malikites). Ibn-Tufayl's solution of his prob-
lems is thus a negative one. The philosophical religion is true
but i t cannot be used directly to guide the affairs of state or
the fives of the ordinary people. A few privileged individuals
may through philosophy fulfil the highest potentialities of
human life, but they only do so through withdrawal from
the active life of the world. One is reminded of the philoso-
pher-kings of Plato's Republic, who found their true life i n
the contemplation of the Good i n itself. What Ibn-Tufayl
does not explain is how the contemplation or mystical ecstasy
of the philosophers is able to contribute to the well-being
of the worldly state consisting of ordinary people.
The other great philosopher of the Almohad period—in
some ways the greatest philosopher of all who wrote i n Arabic
 PHILOSOPHY AND MYSTICISM 121
—is Averroes or Ibn-Rushd ( 1 1 2 6 - 9 8 ) , who has already
been mentioned as a jurist. He was a friend of Ibn-Tufayl,
whom for a short time (about 1183) he followed as physician
at the Almohad court. He had originally been introduced by
Ibn-Tufayl, probably about 1153, to the Almohad prince and
future ruler Abu-Ya'qub Yusuf. Though the young Aver-
roes was already well-versed i n the Greek sciences, he was
afraid and denied his knowledge of such matters when the
prince asked him whether the philosophers thought the
heavens eternal or created. I t was only when the prince
turned to Ibn-Tufayl and spoke freely of Plato, Aristotle and
others, that he ventured to join i n the conversation. Despite
this inauspicious beginning he formed a close friendship w i t h
the prince. The latter's son and successor, Abu-Yusuf Ya'qub
al-Mansur, was also friendly w i t h the philosopher, but about
1195, to gain the support of the Malikite jurists i n the cam-
paign against the Castilians, had to remove Averroes from
his post as qadi of Cordova and order his books to be burnt;
soon afterwards, however, he made up for this by installing
him i n the court at Marrakesh.
The most important philosophical work of Averroes is con-
tained i n the commentaries he wrote on many of the writ-
ings of Aristotle. He had penetrated deeply into the thought
of Aristotle, and for this reason was able to interpret his writ-
ings i n a genuinely Aristotelian fashion. Previously the Mus-
lim thinkers' understanding of Aristotle had been largely
dependent on the Neoplatonic tradition which distorted his
teaching i n various ways, and minimised the difference be-
tween him and Plato. Much confusion had been caused by
the circulation of an Arabic version of a Neoplatonic book
under the title of The Theology of Aristotle. One of the
great merits of Averroes was thus to recover the true Aristotle
and to transmit his thought to Europe. This came about when
Christian and Jewish scholars i n Spain translated the com-
mentaries of Averroes into Latin or Hebrew. This introduction
of Aristotle to Europe was one of the chief factors contribut-
ing to the great achievement of Thomism, though Averroes
is not to be blamed because his view of the relation of reason
and revelation was distorted by the so-called Latin Averroists
into the theory of the "double truth".
122 CULTURAL GREATNESS

Although under the Almoravids the Malikites had com-

bined an attack on al-Ghazall with that on the Almohads,
Averroes felt constrained to defend philosophy against the
criticisms made by al-Ghazali" i n his book The Inconsistency
of the Philosophers (written about 1 0 9 5 ) . I n his defence
and reply, The Inconsistency of the Inconsistency, Averroes
examined the earlier work paragraph by paragraph, refuted
in detail its strictures on the philosophers, and incidentally
expounded his own belief i n the ability of reason to compre-
hend the ultimate secrets of the universe. This book was
38

of the highest quality and had some influence on European

thought (a Latin translation having been made by 1 3 2 8 ) ;
but i t was too late and too much on the periphery to bring
about any revival of philosophy i n the heartlands of Islam.
I n so far as philosophy was now cultivated there, i t was by
theologians, especially of the Ash'arite school; and their
philosophy was subordinate to theological dogma. Though
the work of Averroes was known in the east, its outlook was
so foreign to these men that i t had nothing to say to them.
Apart from his purer AristoteHanism Averroes differs from
Ibn-Tufayl i n his more positive attitude i n respect of the
relation of philosophy to religion. Averroes was deeply con-
vinced that both philosophy and religion are true; this was
the basis of his own life in which he combined philosophical
writing with the work of a qadi. I n a short essay, now trans-
lated into English under the title The Harmony of Religion
and Philosophy, he insists that, since philosophy is true and
the revealed scriptures are true, there cannot be any dishar-
mony between them. He then shows i n some detail how ap-
parent contradictions are to be reconciled. Though philoso-
phers may occasionally be mistaken i n points of detail,
philosophy is i n general sound; and therefore the reconcilia-
tion has to be effected by finding interpretations of scriptural
statements which are in harmony w i t h accepted philosophi-
cal doctrine.
From this standpoint Averroes justifies the participation
of the philosopher i n the active life of the day. The religious
ideas of the ordinary people are valid when properly under-
stood and interpreted; and therefore the philosopher does
not avoid contact w i t h popular religion, but "should choose
 PHILOSOPHY AND MYSTICISM

the best religion of his period", accept its formulations and

explain them. I n so doing he is contributing to the life of
the state, i n which religion has an important function. So fully
aware is Averroes of the place of religion i n society, that
he considers a revealed religion—when philosophically un-
derstood, of course—superior to the religion of pure reason.
Perhaps the difference between Ibn-Tufayl and Averroes on
this point is due to the moulding of the thought of the first
by Almoravid and Malikite hostility to philosophy, i n con-
trast to the toleration and friendship generally experienced
by philosophers under the Almohads.
W i t h Averroes, philosophy reached its highwater mark in
Islamic Spain. The period of toleration was over by the time
of his death, for the serious political situation had brought
the Malikites back into power. There is sometimes talk of
the "philosophy" of Muhyi-d-Dm ibn-ai-'Arabi, but i t is rather
a theosophy. A somewhat younger man—probably born a few
years before 1200—but w i t h a greater claim to be a philoso-
pher is Ibn-SabIn ( d . 1 2 7 0 ) . He was born and brought up
i n Spain and belonged to the Spanish tradition, but soon
preferred to go to North Africa, and ended his life i n Mecca
by opening his veins, i t is said. A book of some interest has
been ascribed to h i m (though doubtfully), known as Answers
to Sicilian Questions; these were questions asked of the Mus-
lim scholars i n Ceuta by the emperor Frederick I I (by the
intermediary of the Almohad ruler of the t i m e ) . The philoso-
phy was less Aristotelian and more Neoplatonic than that
of Averroes. This is not surprising since Ibn-Sabln was also
a mystic. 39

I t is difficult to write explicitly about sufism or mysticism

i n Spain, partly because i t is intertwined with the philoso-
phy studied i n Spain, and partly because i t is dependent on
the development of mysticism i n North Africa and elsewhere.
The connection of mysticism w i t h philosophy goes back to
Ibn-Masarra (p. 57 above). The names of some of his dis-
ciples are known who were active i n the later tenth and early
eleventh centuries; but the distinctive Masarrite movement
came to an end when one man put himself forward as a
wonder-working leader and became immersed i n polities.
The Masarrite mystics had had a centre i n Pechina as well
124 CULTURAL GREATNESS

as that i n the capital, Cordova; and i n the neighbouring

Aimeria another mystical movement is found i n the second
half of the eleventh century. This movement was given
definite form by Ibn-al-lrrif (d. 1 1 4 1 ) , and had secondary
centres at Seville, Granada and i n Algarve (Portugal). The
Almoravid rulers apparently became suspicious of the possible
political implications of the movement as their hold on al-
Andalus weakened, and summoned both Ibn-al-lrrif and his
supporter i n Seville, Ibn-Barrajan, to North Africa, where
both died. The leader i n Algarve managed to maintain him-
self i n political independence from 1141 to 1 1 5 1 .
40

According to the most distinguished student of the mysti-

cism of Islamic Spain, Miguel Asin Palacios, the influence
of Masarrite ideas is also to be seen i n the greatest of the
mystics of al-Andalus, Muhyi-d-Dln ibn-al-'Arabl ( 1 1 6 5 -
1 2 4 0 ) . He was born i n Murcia, and studied both i n Spain
41

and North Africa, at this time united under the Almohads.

I t was probably i n Seville that he came under the influence
of Ibn-al-lrrif and Ibn-Barrajan. I n 1 2 0 1 , however, he went
east on pilgrimage, and spent the rest of his fife i n such
places as Mecca, Baghdad and Damascus, thereby coming
into touch w i t h the various strands of mysticism i n the heart-
lands, including the IJanbalite. His literary production was
vast, and took up into itself much of the material found i n
earlier mystical writers. Asin Palacios also asserts Christian
influences, though other scholars tend to minimise these. Ibn-
al-'Arabi makes much of the conception of the Logos, which
he identifies w i t h Muhammad or the reality of Muhammad;
but this is not parallel to the Christian doctrine of the Logos,
for Ibn-al-'Arabfs system i n general is pantheistic and mo-
nistic. To go into such matters i n detail, however, belongs
rather to the general study of Islamic mysticism.
Finally i t is to be noticed that men born i n al-Andalus
made important contributions to the growth of the dervish
or mystical order of the Shadhiliyya. Abu-Madyan of Tlem-
cen (d. 1 1 9 3 ) , sometimes regarded as co-founder of the order
along with ash-Shadhili (d. 1 2 5 8 ) , came originally from
Spain, as did also the latter's chief disciple and successor,
Abu-'l-'Abbas al-Mursi, the man from Murcia (d. 1 2 8 7 ) .
The most important of these Andalusians who spent their
 ART O F 11TH A N D 12TH CENTURIES 12$

maturity i n North Africa was Ibn-'Abbad of Ronda ( 1 3 3 3 -

9 0 ) , who spent much of his life i n Rabat and was latterly
i n Fez. He is chiefly remembered for a commentary on
42

one of the basic works of the order and for a collection of

"letters of spiritual direction".

5. The Art of the Eleventh and Twelfth Centuries

The art of Islamic Spain under the "party kings", the Almo-
ravids and the Almohads is worthy of the attention of the
historian i n that i t both throws light on some of the general
questions he is trying to answer and also introduces further
complexities. One of the complexities is that the course of
development i n the visual arts differs considerably from that
i n literature. The reason may be that the two forms proceed
mainly from two distinct social milieus. Works of art, espe-
cially of architecture, were commissioned by the ruling £lite
i n any given period, and were executed, for the most part,
by members of the body of craftsmen who had inherited
the necessary skills.
The most notable remaining architectural work of the pe-
riod of "party kings" is the Aljaferia of Saragossa, a palace
built by the local ruler Abu-Ja'far al-Muqtadir ( 1 0 4 9 - 8 1 ) .
As compared w i t h works of the tenth century i t shows an
increasing interest i n decoration. Arches are elaborately
lobed, and the interlacing geometrical designs become more
subtle. There is a certain love of contrast apparent i n the
alternation of plain areas and panels filled w i t h elaborate
decoration. A l l this is i n a natural line of development from
the art of the Umayyad period.
I n Spain itself there are no outstanding works of the
Almoravid period, but the general character of their art can
be discerned from various buildings i n North Africa now
that—comparatively recently—the plaster with which they
were covered i n the Almohad period has been removed. The
Almoravids made use of craftsmen from al-Andalus, so that
the style of the architecture of Islamic Spain was transferred
to the southern shore of the Mediterranean. The tendency
to cover an area entirely with ornamentation is specially
126 CULTURAL GREATNESS

noticeable i n many examples of Almoravid work. On the

other hand, the religious beliefs of the Almoravids, and their
supporters the Malikite jurists, do not seem i n any way to
have affected their artistic productions.
W i t h the Almohads, on the contrary, the element of pu-
ritanism i n their religious outlook led to a reaction against
the luxuriant decoration of the immediately preceding pe-
riod. I t was for this reason that some of the Almoravid work
was plastered over. The greater simplicity of Almohad art
is immediately apparent especially i n the North African ex-
amples, where i t also is felt to have an air of grandeur. This
is not so evident i n Almohad works i n Spain itself. One of
the chief Almohad monuments, the Giralda of Seville (now
the bell-tower of the cathedral and illustrated i n Plate 9 ) is
closer to earlier Spanish styles than to the contemporary
style of North Africa.
Among the points of wider import to be noticed there is
the ready acceptance by the new Berber rulers of the ar-
chitectural tradition of al-Andalus. Because their previous
material culture was of the simplest, they had really no
alternative to taking over the tradition and no way even of
contributing to i t . The theological ideas underlying the
Almoravid and Almohad movements, though emphasising the
distinctively Islamic moment (and the superiority of Islam
to Christianity), had little direct influence on artistic produc-
tions. Because the tradition of al-Andalus was definitely
Islamic and anti-Christian, the Muslim zealots had no qualms
about taking i t over. They may indeed have felt that there
was something specifically Islamic about its geometrical and
stylised floral ornamentation.
 10
T H E LAST OF I S L A M I C SPAIN

1. The Nasrids of Granada

The founder of the Nasrid dynasty, as already noted (p. 9 6 ) ,

was Muhammad ibn-Yusuf ibn-Nasr, also known as Ibn-al-
Ahmar. Though i t was originally at Jaen that he set himself
up as a ruler (about 1 2 3 1 ) , the progress of the Reconquista
under Ferdinand I H of Castile, and i n particular the loss of
Jaen itself i n 1245, forced him to retire southwards and to
make Granada, which he had occupied i n 1235, the seat of
his government. When i t became clear to him that he could
not indefinitely keep the forces of Castile at bay with his
slender military resources, he decided to become a vassal of
Ferdinand, as several other local Muslim rulers were doing.
I n this capacity he gave support to his liege-lord i n the cam-
paigns which led to the capture of Seville and the lower
valley of the Guadalquivir, and i n other subsequent cam-
paigns against Muslims. The state which was thus created
extended from Tarifa (just beyond Gibraltar) i n the west
to some twenty or thirty miles beyond Almeria in the east—
in all about 240 miles. I n the north the frontier was probably
nearer to Jaen than to Granada, at a distance of sixty or
seventy miles from the sea as the crow flies.
When Muhammad I of Granada became the vassal of
Castile, he was not the only Muslim ruler i n this position.
The others, however, gradually disappeared and were re-
placed by Christian governors, the last to go being the emir
of Murcia i n 1264. I t is therefore worth asking why the king-
dom of Granada should have managed to keep its independ-
ence for two centuries and a half. Various reasons may be
suggested, though no single one of them seems to be decisive.
Muhammad I appears to have been a good vassal to Ferdi-
nand and his son, and thereby to have merited generous
128 THE LAST OF ISLAMIC SPAIN

treatment; and by the time of his death i n 1273 the accept-

ance of an independent Granada may have been a fixed
point i n the policy of Castile. Castile, too, with many Muslim
subjects, may have felt that i t was useful to have a Muslim
state near to which the more discontented could flee for
refuge. Perhaps, however, the main emphasis should be laid
on two geographical factors—the mountainous nature of the
country, and the nearness to Africa. Most of the territory of
Granada consisted of relatively high mountains, and this
natural defence was reinforced by strong fortresses and for-
tress towns (like Honda) at the points where attack was
easiest. So long as things went well, the rulers of Castile
probably felt that no attempt to advance their frontiers was
worth the military cost. Moreover the nearness of Africa made
i t possible for the Nasrids to appeal for help from time to
time to the new Muslim rulers of Morocco, the Marinid
dynasty. This meant that they were never completely at the
mercy of Castile, even though they were careful to give the
Marinids no opportunity of adding Granada to their African
domains.
The state of Granada was very consciously Islamic. A
welcome was given to refugees from the rest of Spain. Arabic
was the only language used. Though there were Jews in the
state, there were no Mozarabic Christians; but i t is not clear
whether this was because of some definite enactment, or be-
cause the attitude of ordinary Muslims made life too un-
pleasant for them. This emphasis on Islam and on the defence
of Islam is understandable after the concern for the holy war
shown by the Almoravids and the Almohads and after
the growing self-consciousness of the Christian Reconquista
during the period of success from 1212 to 1248.
On many aspects of the history of the Nasrid state little
information is available. Most is known about its relations
to the Christian states, since there is material about this in
their chronicles. The period of greatest brilliance was from
1344 to 1396, during which time the finest parts of the A l -
hambra were built. On the whole the state was very pros-
perous, through its intensive agriculture, its urban handicrafts
and its trade. There were many internal difficulties, however.
 THE NA§RIDS OF GRANADA 129
Quarrels about the succession among members of the ruling
family, each with supporting interests, were frequent, espe-
cially from the last decade of the fourteenth century. The
consciously Islamic attitude of the state favoured an increase
i n the power of the jurists; and they, together with the African
mercenaries and certain urban elements, tended to favour
war. Opposed to them were the ruling elite, the merchants
and the peasants, whose interests were much better served
by peace.
The end of the Nasrid realm came about as much through
its own internal weaknesses as through the growing strength
of the Christians. This strength was very much increased by
the union of Aragon and Castile through the marriage of
Ferdinand and Isabella, Isabella ascending the throne of
Castile i n 1474 d Ferdinand ( I I ) that of Aragon i n 1479.
a n

Even before this, however, the capture of Gibraltar i n 1462

showed that the Christian powers were again on the move.
Yet the final debacle might have been postponed had the
Muslim leaders not lost their nerve and given way to i m -
patience. I n 1481, before the end of a period of truce, some
of them seized the castle of Zahara from the Christians, and
this provocative pinprick doubtless led Ferdinand and Isa-
bella to the resolve to make an end of Granada. A n out-and-
out military attack was avoided. Instead Ferdinand availed
himself of the Muslims' divisions, and kept one party of
them at peace by supporting them while his armies were
launched against the other party i n isolation. I n this way he
managed to capture Ronda ( 1 4 8 5 ) and Malaga (1487) i n
the west, and then Almeria ( 1 4 8 9 ) i n the east. The final
campaign against Granada was launched i n 1491, and before
the end of the year the defenders had recognised the hope-
lessness of their position and agreed to surrender. Honourable
terms were granted, and the surrender became effective i n
the first days of 1492. Romance has played round the scene
when from a vantage-point on the hillside the last of the
Nasrids—the last Muslim ruler i n Spain—Abu-'Abd-Allah or
Boabdil, bade farewell to al-Andalus.
130 THE LAST OF ISLAMIC SPAIN

2. The Muslims under Christian Rule

From one point of view the fall of the Nasrids was the end
of Islamic Spain; and yet from another point of view it was
far from the end. To the historian of Islamic culture what
was happening i n some non-Muslim parts of Spain in the
thirteenth and fourteenth centuries is at least as important
as what was happening i n Nasrids Granada. To complete
the story of Islamic Spain, then, it is necessary to look at the
life and achievements of the Muslims who remained i n
Christian states.
Scholars' familiarity with the reputation of Christian Spain
for intolerance, together w i t h the prominence given to the
idea of Reconquest, has sometimes suggested that, once a
province came under Christian rule, there ceased to be any
Muslims i n it. Although Castile in particular favoured the
policy of settling Christian colonists i n unoccupied lands, i t
was far from the case that there were no Muslims under
Christian rule. When Toledo was captured in 1085, many
craftsmen remained; with them were also some scholars, and
these played an important part i n the transmission of Islamic
science and philosophy to Europe. After 1248 there were
many Muslims i n the Christian kingdoms. I n the new A n -
dalusian province of Castile they constituted a majority of
the population, while i n Aragon proper and the province of
Valencia the Christians were a comparatively small minority.
This state of affairs was inevitable; the rulers had to retain
the Muslims because they were an essential part of the econ-
omy of the country, while for the Muslims there was no other
region where they could exercise their skills adequately.
Those Muslims who continued in their former homes after
the change of rule are known as Mudejars (Spanish, mude-
jares), the word being derived from the Arabic mudafjan,
"permitted to remain", w i t h a suggestion of "tamed, domesti-
cated". They had a position similar to that of the protected
minorities i n the Islamic states. They followed their own re-
ligion, laws and customs, and were free to carry on their
crafts and to trade. Each local community had a Muslim
 THE MUSLIMS UNDER CHRISTIAN RULE I3I

head appointed by the king. I n return for their privileges

they paid a poll-tax or tribute. They constituted distinct
communities, sometimes forced to mark themselves off by
dress, and inhabited special quarters i n the chief towns.
Many were hard-working peasants i n the country districts.
Certain crafts were almost entirely i n the hands of Mudejars.
The presence of the Mudejars i n Christian Spain produced
little i n the way of the "historical events" on which nine-
teenth-century historians tended to concentrate. Yet their
uneventful life is an indication of an important historical
phenomenon—the existence of an economic structure and a
material culture common to Christian and Muslim. I n the
Mudejar age—the thirteenth and fourteenth centuries—this
economic structure and material culture reached out to em-
brace even those parts of northern Spain which had had
practically no direct contact with Muslims. The most impor-
tant pieces of evidence for this are artistic productions of
the period.
A t the same time i t must be realised that there was selectiv-
ity i n the assimilation of this culture by the Christians. What
was clearly at variance with basic Christian conceptions was
never taken over; and this applied to the higher levels of
the intellectual life, particularly i n the case of the ordinary
Spanish Christian. Christian scholars from outside Spain, i n -
deed, came to make contact with the living tradition of Greek
philosophy i n its Islamic dress, and much ink had to be spilt
in separating what could from what could not be Christian-
ized; but that does not belong specifically to the history of
Spain.
By the beginning of the fifteenth century a change of at-
titude can be detected among the Spanish Christians. I n part
it may be due to economic grievances, for many of the M u -
dejars were wealthy. Certainly anti-Muslim prejudice began
to appear among the common people. W i t h the union of
Spain under Ferdinand and Isabella such prejudice came to
have some influence on policy. The old policy of religious
tolerance still governed the terms of surrender granted to
the inhabitants of Granada i n 1492, but i n the same year an
edict was promulgated forcing the Jews throughout Spain to
be baptised or to leave the country. Earlier, i n 1478, the
13^ THE LAST O F ISLAMIC SPAIN

Inquisition had been "nationalised" in that the inquisitors

for Spain were to be appointed by the king and queen and
not by the Pope. This appears to have had the result that
what was promoted, or even enforced, was not Christian
orthodoxy simpliciter, but Christian orthodoxy as conceived
by leading Spaniards.
Another facet of the new policy was seen i n 1499 when
the powerful Cardinal Xim^nez de Cisneros visited Granada
and had discussions with the jurists there. This was followed
by the burning of specifically Islamic books and by forced
conversions. The result was an insurrection which began i n
the following year and lasted into the one after that. By way
of punishment the Muslims of Granada i n 1502 were given
the choice of baptism or exile. Many chose baptism, though
without any change of basic conviction. I n 1525 and 1526
similar measures were taken against Muslims i n other prov-
inces. Thereafter there were officially no Muslims in Spain,
but for nearly a century the rulers of Spain had to cope w i t h
the problem of the Moriscos.
Various reasons can be given for this growth of intolerance
in Spain after the long tradition of tolerance. There had been
a time when i t had been hoped that members of the three
faiths could be welded into a unity; but i t was becoming
clear that this was not possible. Political unity under Ferdi-
nand and Isabella, with its new potentialities for disunity,
made a unity of spirit all the more necessary. The intransi-
gence and defensive attitude of Granada i n its closing decades
doubtless contributed to the resolve of Ferdinand and Isabella
to work for a genuine unity of outlook. By 1525 Spain, now
growing into an imperial power, was aware of the Islamic
threat to Europe by the Turkish advance towards Vienna
(which was actually besieged i n 1 5 2 9 ) . Commitments there
and across the Atlantic were making the utmost demands on
Spanish man-power; so that common-sense suggested the re-
moval of potentially hostile elements from the base while so
many of the reliable men of fighting age were out of the
country.
There is said to have been some understanding that the
Muslims were to be allowed forty years to carry out the
 THE MUSLIMS UNDER CHRISTIAN RULE 133

measures of 1525 and 1526. Whether this is so or not, there

was a revival of anti-Muslim legislation i n 1566. The previ-
ous measures had not been effective. For one thing, Islamic
religious practice i n the heartlands and elsewhere had often
permitted and justified taqiyya or the concealing of one's
true religious beliefs where to reveal them would endanger
one's life; and the Moriscos had apparently got official legal
opinions from Muslim jurists outside Spain to the effect that
in the circumstances of sixteenth-century Spain such conceal-
ment of their attachment to Islam was permissible. Manu-
scripts are preserved known as aljamiados, which are written
i n the Spanish language but i n Arabic characters; these con-
tain expositions of Islamic faith and practice by and for
Moriscos. A n additional reason for intolerance was the fac-
tor, now making itself felt, of the relatively higher birth-rate
among the Muslims, which was raising the proportion of
Muslims in the population. On the other hand, the Moriscos'
essential contributions to the economic and material life of
Spain led many nobles i n Aragon and Valencia to support
them.
From 1566, then, the pressure on the Moriscos increased.
Some of them revolted i n 1569 and received help from the
Ottoman governor of Algiers. Yet, despite the pressure, their
communities remained largely intact i n the towns owing to
the self-contained character of their lives. They were thus
proving an unassimilable element i n the population. Finally
between 1609 and 1614 came edicts of expulsion, as a result
of which about half a million are said to have gone to North
Africa. What happened i n some places there throws interest-
ing light on conditions i n Spain. Though now among fellow-
Muslims, some Moriscos showed themselves just as unassimi-
lable as they had been i n Christian Spain. Their culture was
that common material culture of Spain, Islamic and Chris-
tian. I n the Islamic atmosphere of North Africa, they became
more than ever aware of their Spanishness and of their su-
periority to African Muslims, Berbers or others. Indeed some-
thing of Islamic Spain is preserved i n certain cities of North
Africa until the present day, so that Islamic Spain to this
limited extent is still alive.
134 THE LAST OF ISLAMIC SPAIN

3. Literature in a Period of Retreat

This period i n the literary history of Andalusia has been

characterised, not unjustly, as a mere epilogue. There was
no great inventive impulse, no startling poetic innovation.
Rather was it—as indeed i t was i n the Islamic East also—
an age of compilation and wide erudition, when the scholar-
ship accumulated by succeeding generations was recorded
or given final expression, often by expatriate Andalusians.
One such expatriate was the grammarian Ibn-Malik ( 1 2 0 8 -
7 4 ) who expounded the whole of Arabic grammar in two
urjuzas of which the shorter, running to somewhat less than
a thousand lines of doggerel verse, is used to this day as a
1

teaching text. Another who taught i n the East was Abu-

Hayyan ( 1 2 5 7 - 1 3 4 4 ) , celebrated primarily as a grammarian
but also well versed i n a number of Islamic sciences; it is a
measure of his erudition that he was the author of an impor-
tant Turkish grammar, and is said also to have written an
Ethiopic one. Yet another was Ibn-Sald al-Maghribi ( d .
1274), a m a
of many parts who compiled a much-prized
n

anthology to which he contributed not a few verses of his

own.
Those who remained within the shrinking Muslim domains
i n Spain were also living on accumulated capital, but the
capital was large enough to enable them to live with dignity
and elegance. Under the rule of the Nasrids of Granada, and
especially at court, i n that Alhambra whose exquisitely
carved palaces and pavilions rest lightly on slender pillars
as deceptively frail-looking as a ballerina on her points, poets
and prose-writers more conscious of their heritage than of
approaching doom could yet display some of the sparkle of
an earlier age.
Towering above all other literary figures of the time was
that of Lisan-ad-Dln ibn-al-Khatib ( 1 3 1 3 - 7 4 ) . Several times
vizier, chronicler, author of a useful biographical dictionary,
he was also a master of ornamental prose who wrote numer-
ous maqdmas, and a highly accomplished poet who composed
both traditional odes and fine muwashshahs. He also com-
 LITERATURE IN A PERIOD OF RETREAT 135

posed an urjuza entitled Raqm al-Iiulal fi Nazm ad-Duwal

recounting the history of Islam in the West, which one mod-
ern Arab considers worthy of the title of an Arab Shah-
Nameh, though it never struck root i n the imagination of
2

the people as did the Persian epic. On the whole, he was

notable not for the blazing of new trails but for his delicate
workmanship, easy lilt, charming sentiment and felicitous
imagery, as he sang of lovers trysting in idyllic surroundings,
watched only by envious narcissi or by lilies that prick up
their ears at their passing, sharing a destiny "faultless, except
that i t spent itself i n the twinkling of an eye".
I t was his protege and successor Ibn-Zamrak ( 1 3 3 3 - 9 3 ) ,
the last of the great Andalusian poets, whose verses were
used i n decorating the walls of the Alhambra.
The Moriscos who found themselves under Christian rule
had of course vastly different preoccupations. Not for them
a literature of refinement dependent on the patronage of the
great. Instead, they circulated among themselves works de-
signed to counter the pressures to which their faith was sub-
jected: expositions of canonical law; polemical tracts; some
poetry i n praise of the Prophet and on other Islamic themes;
and not a few tales and legends derived from universal folk-
lore, or inspired by ascetics, or growing out of Quranic refer-
ences to Moses and Joseph and Solomon and Alexander, or
magnifying the deeds of Muhammad and other Muslim
heroes. This aljamiado literature, worded i n Spanish although
written i n Arabic characters, can command no more than a
passing reference in this survey; i t is worth noting, however,
that many popular tales were thus recorded which i n a
classical-minded Arabic-speaking society might have escaped
the attention of the literate.
The profusion of popular narratives thus revealed is not
without relevance to a consideration of the influences ex-
erted by Andalusian literature outside Islamic Spain, influ-
ences by no means restricted-to the period surveyed i n this
section, but which the Reconquista hastened and swelled.
On the one hand, the Islamic East had always had the
opportunity of savouring the finest literary productions of
al-Andalus, and to i t the vogue of the mutvashshah had
spread. Now, as the Muslim hold on the Peninsula was re-
136 THE LAST OF ISLAMIC SPAIN

duced and loosened, more and more of the leading families

of al-Andalus and of her scholars sought the protection of
fellow Muslims and the patronage of Muslim princes else-
where, especially i n North Africa. Tunis and then Fez be-
came repositories of Andalusian culture, so that i t is to a
North African who studied i n Fez, al-Maqqari (d. 1631)
that we owe a very great deal of the information we now
have on al-Andalus and its literature.
On the other hand, Christians were bringing under their
sway a mixed population with a culture some aspects of
which they came to appreciate and assimilate into their own.
Already when the fortunes of the Muslims were i n the
ascendant, their learning had attracted scholars of all faiths.
Spanish Jews i n particular were directly indebted to Arab
thought, and many of them—including the great Maimonides
(1135-1204)—sat at the feet of Arabic-speaking teachers
and wrote their books i n Arabic. Even i n belles-lettres,
Hebrew poets—such as Ibn-Gabirol (1021-52)—were influ-
enced by Arab prosody, and Hebrew muwashshahs were
written which not only adopted the conventions of the Arabic
form but were " i n the overwhelming majority of cases . . .
imitations of particular Arabic poems". Not a few prose-
3

writers of the twelfth and thirteenth centuries also wrote

maqamas i n Hebrew; one of them, Judah ben Solomon al-
Harizi ( 1 1 6 5 - 1 2 2 5 ) actually translated those of al-Hariri i n
1205. Kalila voa Dimna was translated into Hebrew both by
a certain Rabbi Joel and by Rabbi Eleazar ben Jacob ( 1 2 8 3 ) ,
who was also a maqama-wxitex.
The Christians also, even when they were not won over
to Islam, were attracted by Arab learning; annotations i n
Arabic on Latin texts show that their clerics were at home
i n the language. But the link between Arabic and the litera-
tures that subsequently developed i n Christian lands is not
so precise or direct as i t has been for Hebrew.
The clearest such link is the transmission of narratives.
Early i n the twelfth century Pedro Alfonso, a convert from
Judaism baptized i n 1106, put together thirty-three tales,
Arab by origin or by transmission, which he translated into
Latin and published under the title of Disciplirw clericalis,
indicatng that they were intended for the edification of the
 LITERATURE IN A PERIOD OF RETREAT I37

literate. The book had a very wide diffusion i n Europe, was

translated into many languages, and echoes of i t are found
in such famous works as Don Quixote and the Decameron.
Rabbi Joel's Hebrew version of Kallla wa Dimna was also
translated into Latin under the title of Directorium Vitae
Humanae by John of Capua, himself a converted Jew.
Spanish narrative literature i n particular owes a great deal
of its initial impetus to early translations of three Oriental
works that reached i t at least i n part through Arabic. The
first was again Kallla wa Dimna, translated from the Arabic
by order of Alfonso X towards the middle of the thirteenth
century. The other two are the Syntipas, also known as the
Sindibad-Name and the Sendebar, another collection of tales
of Indian origin, translated i n 1253 under the title of Libro
de los engannos et los asayamientos de las mujeres, and the
story of Barlaam and Josaphat, based ultimately on the life
of Buddha.
These stories echo and re-echo through the later literature
of Spain and of other European countries, as do some of the
tales from the Arabian Nights. Here and there other specific
instances of indebtedness can be found; thus Turmeda's Dis-
puta del Asno ( 1 4 1 7 ) , i n which the Ass is made to rebut
arguments for the alleged superiority of man, closely follows
one of the Epistles of the Pure Brethren, a tenth-century
collection of treatises on science and philosophy. I n his
Criticdn, Gracian also appears to have drawn i n part on the
same popular tale that Ibn-Tufayl used i n his Hayy ibn-
Yaqzdn* A n account of Muhammad's ascent into heaven,
the Miraj, translated into Castilian by order of Alfonso X
though now known only i n French and Latin versions, may
even have played a part i n shaping Dante's imagination. 5

The inspiration thus appears to be very largely from popu-

lar story-telling, but i t has been suggested that the picaresque
novel of the sixteenth and seventeenth centuries, which re-
counts sympathetically the rogueries and adventures of a
low-born hero, may owe its existence to the maqdma.
Ribera has also conjectured the existence of an early popu-
lar epic, and sought to trace its possible influence on the
Chanson de Roland and the Poema del Cid.
Much debated among Arabists and Romance scholars has 6
X38 T H E LAST OF ISLAMIC SPAIN

been the possible influence of Arabic poetry, more specifically

of Andalusian strophic forms on the art of the juglares and
troubadours, and hence on the whole lyrical tradition of
Romance literatures.
Opportunities for such influence to be exerted were many
where Christians and Muslims lived side by side and ex-
changes between courts were common; nor were contacts
made i n war necessarily barren i n this respect. These oppor-
tunities are not merely matters of conjecture. Captive Morisco
women singers are known to have performed i n Christian
courts; an illustration i n Alfonso X's Cantigas de Santa Maria
(1265) shows a Moor and a Christian singing together and
accompanying themselves on identical lutes; and the Coun-
cil of Vailadolid i n 1322 was condemning no imaginary evil
when i t forbade the employment of Muslim singers and per-
formers i n churches. One troubadour, Garcia Fernandez, is
known to have married a Morisco singer and to have moved
between Christian and Muslim possessions i n Spain. 7

Significant similarities can i n fact be detected between

Andalusian and Romance lyric poetry. The most firmly
demonstrable are those based on the structure of Andalusian
strophic poetry. Indeed Guillaume I X appears to have i n -
corporated some Arabic lines i n his Chanson g. I t is also
8

argued, however, that there are similarities of substance es-

pecially i n attitudes towards women and love, the troubadours
perpetuating the amatory commonplaces of Andalusian poetry
—the cruelty and tyranny of the beloved, the servility and
sufferings of the lover, the intervention i n a love intrigue of
stock characters such as the confidant, the calumniator, the
guardian—and all i n all reflections of a more refined concep-
tion of love than was known i n Europe until Ibn-IJazm.
That an Andalusian poetic tradition was one of several
forces at work i n the formative period of Romance litera-
ture i t would be difficult to gainsay, especially when one
remembers the undoubted contributions made by the culture
of Islamic Spain i n all other fields, including the closely
allied one of music. The degree and extent of the influence
exerted, however, can be determined only by detailed ex-
aminations of texts and close discussion of standards of com-
parisons on which there is as yet no agreement. The argu-
9
 LITERATURE IN A PERIOD OF RETREAT 139

merit cannot therefore be usefully summarised. Suffice i t

here to reproduce the view of Menendez Pidal, a discriminat-
ing and well-informed Romance scholar: that Arab influence
was one of several that can be detected i n the earliest com-
positions of Provengal troubadours, notably Guillaume I X ,
Marcabru and Cercamon, but that i t waned rapidly, surviv-
ing mostly i n the popular anonymous songs of France and
Italy; that i t d i d not significantly penetrate Galician-
Portuguese poetry except i n Alfonso X's Cantigas de Santa
Maria; that i t was predominant i n Castilian poetry until the
sixteenth century when i t was supplanted i n all but the most
vulgar forms, although traces of i t subsist i n the classical
Spanish theatre which is rich i n traditional songs.10

I t yet remains to be stressed that what the Andalusians

handed down to their Romance successors was not their
classical tradition—not their monorhyme odes w i t h their suc-
cession of conventional themes, not their trope-laden prose
—but imaginative tales, strophic forms of poetry and perhaps
the particularly refined lyricism of their love-songs. They
were features remarkably close to those which have been
held to distinguish their literature from that of their co-
religionists of the East. They were also, i n no small part,
features that belonged to popular literature or derived from i t .
This is consistent w i t h a total picture of Islamic Spain i n
which the literature of the elite, strengthened by patronage,
sought to perpetuate traditional standards no doubt valid
but derived, and reaped the reward of steadfastness i n glit-
tering masterpieces. But co-existing w i t h i t , penetrating i t
only partially but more profoundly than i n other lands where
Arabic held sway, was another literature which was the
natural expression of a population ethnically and culturally
mixed, i n which the Arab element gave and took, merged
and moulded, was stimulated and i n turn elaborated and
developed. The fusion of these ethnic and cultural elements
produced the distinctive features of Andalusia's literature; i t
also favoured their survival.
Andalusian literature was like the progeny of an expatriate
aristocrat who had had two wives: one a free woman of his
own race and background, the other a local bondwoman.
The sons of the first wife shone best at court; those of the
I4O T H E LAST OF ISLAMIC SPAIN

bondwoman were more adaptable, even when political change

swept away their fathers privilege.

4. The Art of the Thirteenth and Fourteenth Centuries

Despite the worsening political situation i n the twelfth and

thirteenth centuries the artistic tradition of Islamic Spain
was still alive and indeed producing some of its greatest
works. I t may be regarded as having split up into two
branches: Mudejar art, and the art of Granada (or Grena-
dine a r t ) .
The Mudejars, as already explained, were the Muslims
who elected to remain i n territories ruled by Christians.
Their numbers were considerable i n certain areas, and i t
was through this new form of association of Muslim and
Christian that much of the culture of al-Andalus passed to
Christian Spain, or rather was assimilated by Christian Spain
i n its expansion. This coming together of two societies i n
what is almost a single cultural organism is what is some-
times called "the Mudejar fact". Mudejar art, however, is
not simply the art produced by the Mudejars, but rather
the art arising out of this new cultural unity. I t was an art
which continued the tradition of Islamic art from the inde-
pendent Muslim kingdoms, yet was genuinely at home i n this
new composite Christian-dominated society. So much was
this the case that the craftsmen who carried out the work
were sometimes Christian and sometimes Muslims from in-
dependent Granada, and yet the work had the Mudejar
stamp; while Mudejar craftsmen who went elsewhere pro-
duced work i n other styles.
W i t h i n the Mudejar style the experts distinguish "court
Mudejar" from popular Mudejar. The former is exemplified
above all i n the Alcdzar of Seville, and tends to follow Umay-
yad patterns. The latter has been described by Henri Ter-
rasse as the national art of Spain at this period. I t varied
somewhat from region to region, since i t incorporated aspects
of the regional tradition. There are fine examples of Mudejar
work i n the churches of Toledo from the end of the twelfth
century onwards. A l l this is extremely interesting i n itself.
 A B T O F I3TH AND I4TH C E N T U H I E S 141

From the point of view of the present survey, however,

the great interest of Mudejar art is the evidence i t provides
for the symbiosis of the societies, resulting i n the taking up
into the life of the Christian kingdoms of much of the ma-
terial and intellectual culture of al-Andalus.
The art of Granada is known to us especially from its great-
est achievement, the Alhambra, though this was far from
being the only product of the artistic activity of the Nasrid
kingdom. We can see i n the Alhambra how the general out-
look of the kingdom affected its art. I t was a society on the
defensive, conscious of being a bastion of Islam against a
hostile world, and intent on preserving what i t had inherited.
I n art this meant that i t remained firmly within the earlier
artistic tradition of al-Andalus, and introduced no novelty;
that i t rose to new heights of achievement was through the
perfection of its craftsmanship.
The Alhambra is a complex of buildings on a spur of the
Sierra Nevada above the town of Granada. Like other Moor-
ish palaces i t was also what would now be called "govern-
ment offices". I n addition i t was a strong fortress. From the
artistic point of view, however, interest centres on the palace
proper, consisting of a comparatively small number of rooms
and halls round two large and other smaller patios or court-
yards. These were constructed mainly during the second
two-thirds of the fourteenth century. The passion to cover
whole surfaces with intricate finely-wrought decoration was
given rein, with exquisite results. Full use was also made of
the water, which was available in abundant supply from
higher up the mountains, to place fountains, formal pools
and vegetation i n the courtyards—a great source of pleasure
i n the warm climate. The chief impression, however, is that
of perfect ornamentation and perfect grace resting lightly
on, even descending on, the slender columns.
This was an art that, i n its exuberance, built for present
enjoyment, and had no thought of creating enduring monu-
ments. I t used fragile materials superimposed on a basic
structure that was of the slightest. I t is mainly through a
series of happy chances and the continuing care of successive
generations that the Alhambra has been preserved for us. By
the fifteenth century, however, the artistic impulse of the
 THE LAST OF ISLAMIC SPAIN

craftsmen of Granada was weakening doubtless as a result

of the inward-turning attitude of the kingdom and its politi-
cal decay. Before Nasrid Granada was finally extinguished,
its distinctive art was virtually defunct.
So i t was that, partly through the actions of men (Chris-
tians) and partly through the natural processes of growth
and decay, the art of Islamic Spain came to its end. The
Spaniard of modern times has tended to regard the Moorish
past of the country as something foreign to him, yet for
those who have eyes to see the artist and craftsman of today
still draw some of their inspiration from Islamic sources.
 11
T H E SIGNIFICANCE O F I S L A M I C SPAIN

i. The Arab and Islamic Colony

One of the lines of thought about Islamic Spain suggested

at the beginning of this book is that which considers it as
part of the great community of Muslims. I t was a limb or
organ of the vast cultural and religious body whose terri-
tories stretched from the Atlantic coasts of Spain and North
Africa at first to Samarqand and the Punjab and latterly to
the East Indian archipelago. What happened to the life of
the body as it poured into this isolated limb? Were the forms
of Islamic life suitable for the circumstances of the Iberian
peninsula, or had there to be further measures of adapta-
tion? Had the culture of al-Andalus anything to contribute to
the culture of the heartlands? These are some of the ques-
tions now to be looked at again after this survey of the his-
tory of Islamic Spain. I t should be clear by this time that,
where there are so many gaps i n our knowledge, any an-
swers given can only be provisional.
To the student of culture one of the most interesting fea-
tures of the Islamic society of the heartlands is the way in
which the Muslim Arabs, through their language and reli-
gion, provided a mould into which could be poured most of
the cultural heritage of the Fertile Crescent and neighbour-
ing lands. The Arabs came from the desert and from towns
largely dependent on the desert, and had thus a low standard
of material culture, though they had attained—it may be
claimed—a high degree of human excellence and of skill i n
personal relationships. The peoples they conquered i n Iraq,
Syria and Egypt had for centuries enjoyed a high level of
material and intellectual culture, the latter including Greek
philosophy and the Christian theology based on it. Yet i t
was the culture of the Arabs which became the matrix of
144 T H E SIGNIFICANCE OF ISLAMIC SPAIN

the new Islamic civilisation, and all that was best i n the older
and higher culture was assimilated into the new culture.
I n the case of Spain i t has to be remembered that its main
contacts w i t h the heartlands were under the Umayyad
caliphate (up to its demise i n 7 5 0 ) . After that time Islamic
Spain was cut off i n certain respects from the main centres
of Islamic life, since these were under the rule of the \Ab-
basids, who had destroyed the Umayyads, whereas Spain
continued under a branch of the Umayyad family for over
250 years longer. Now the Umayyad caliphate i n the heart-
lands was characterised by the dominance of Arab elements
rather than of the distinctively Islamic. The Umayyads were
practising Muslims, but they did not show the same defer-
ence as the 'Abbasids did to the self-appointed exponents of
Islamic religion and religious law. I n administration they at-
tempted, though w i t h increasing difficulty, to adapt Arab
political ideas (derived from tribal institutions) to the run-
ning of an empire, whereas the 'Abbasids openly leant on
the Persian imperial traditions. The assimilation of Hellenistic
thought by Muslims had begun under the Umayyads, but
this had been almost exclusively i n Iraq, and Syria had been
unaffected.
Thus the culture of the first Muslims i n al-Andalus was
much more Arab than Islamic and the dominance of the
Arab element continued to be characteristic. Evidence for
this is the interest i n Arabic poetry, i n grammar, i n the writ-
ing of commentaries on such typically Arab works as the
Maqamat of a l - ^ a i i r i , i n details of Arab genealogy. The
adoption of the Malikite legal rite points i n the same direc-
tion, for this was the most truly Arabian rite; the other main
rites originated i n Iraq where the intellectual milieu had
been permeated by Hellenistic ideas. Similarly the philosophi-
cal theology of the East had no real foothold i n Spain. This
predominance of the Arab and anti-intellectual element
makes i t all the more remarkable that there should have been
such a flowering of philosophy under the Almohads, and the
reasons suggested above do not solve the mystery.
The evidence summarised i n the body of the work tends
to the conclusion that the Arab element continued to be
dominant until the eleventh century, and that the Islamic
element only exerted its full influence under the Almoravids
 THE ARAB AND ISLAMIC COLONY 145

and the Almohads. This was not because al-Andalus was

entirely shut off from the heartlands. On the contrary, travel
was usually easy, and at some periods it was normal for
the scholars of al-Andalus to have studied in the great cen-
tres of Islamic learning such as Medina and Baghdad. The
introduction of canons of taste from Baghdad by Ziryab i n
the ninth century does not seem to have greatly affected
the intellectual and religious life. More important was al-
Hakam II's formation of a great library and the encourage-
ment given about the same period to scholars from the heart-
lands to settle in al-Andalus. This created the foundation on
which a more comprehensive structure of distinctively Islamic
learning could be raised. The growth of specifically Islamic
thought and feeling which this made possible was fostered
by both Almoravids and Almohads because of their religious
outlook.
I n no part of the Islamic world after the first half-century
or so did specifically Islamic ideas have much influence on
the sphere of administration in general. The conception of
the "holy war" could from time to time rouse the enthusiasm
of the masses and swell the ranks of an army; and for this
reason i t was found useful by politicians. For the most part,
however, the actual rulers of Islamic states found i t neces-
sary to follow secular traditions of governmental practice. The
'Abbasids had modelled their practice largely on pre-Islamic
Persia, and something of this penetrated into the court-life
and administration of al-Andalus. The next striking diver-
gence of al-Andalus, however, from the theoretical Islamic
norms is i n the arrangements that came to be made with
even greater frequency, whereby non-Muslim local rulers be-
came tributary to Muslims and Muslim local rulers tributary
to Christians. This appears to be an acceptance by the Mus-
lims of local practices, perhaps specially appropriate to
geographical conditions.
I n this brief sketch of cultural developments i n al-Andalus,
these appear to have been determined (except in the political
sphere) almost entirely by the culture of the heartlands,
though by different strands at different times. Yet this is a
case where appearances are deceptive, and where i n fact
I46 T H E SIGNIFICANCE OF ISLAMIC SPAIN

Iberian influences are greater than at first sight they seem

to be. One point that is clear is that nothing of the Christian
intellectual culture of pre-Islamic Spain made any real con-
tribution, despite the high level attained by Isidore of Seville
and his school. On the contrary great numbers of the local
inhabitants became Muslims, and i n course of time were as-
similated to the Arab section of the population. Even more
surprising—and a point worthy of deep reflection—is the at-
traction of Arab culture for the indigenous inhabitants who
remained Christian, and who because of this attraction were
known as Mozarabs. One may surmise that perhaps, somehow
or other, the Carthaginian heritage influenced them i n this
way. Certainly much of the culture of al-Andalus (apart
from specifically religious dogmas) was shared by all the in-
habitants of al-Andalus of whatever racial origin and religious
confession. I t was from popular aspects of this amalgam (it
seems) that the new poetic forms developed which were one
of the chief original contributions of al-Andalus to the heart-
lands (though the scholars working i n the traditional dis-
ciplines also made contributions of value).
Indeed, the more these matters are reflected on, the more
i t is clear that there was a genuine symbiosis of indigenous
and adventitious elements of population w i t h their respec-
tive cultures. Many of the outward forms which are most
easily apprehended by the modern scholar were contributed
by the Arabs, and also something of the spiritual drive. Yet
the creative energies which produced the great aesthetic
achievements i n architecture, literature and other arts, seem
to have come at least as much from the Iberian element or
some section of i t . Here we are up against one of the great
mysteries. Just as the Persian genius found i n Islam something
which fertilised i t and enabled i t to flower luxuriantly, so
also i t seems to have been the case w i t h the Iberians.
I t should further be kept i n mind that there may have
been Berber contributions to the cultural amalgam, but these
are hard to detect. The most obvious would be an interest
i n the saint or charismatic leader, which has long been a
feature of North African religion. This, however, as explained
above, could not have much influence i n Spain because of
 CHRISTIAN SPAIN AND EUROPE 147
the dependence of the Berbers on the Muslim rulers. I t con-
tributed to the set of ideas underlying the Almoravid and
Almohad movements, but the extension of the culture of al-
Andalus into North Africa under the dominion of these move-
ments suggests that the Berbers had little of their own to
contribute. I n any case, a discussion of the Berbers would
belong rather to the study of North Africa.

2. The Stimulation of Christian Spain and Europe

A second group of questions is concerned with the relation

of Islamic Spain to Christian Spain and more generally to
Christian Europe. W i t h regard to Christian Spain in particu-
lar there is little doubt about the fact. Somehow or other i t
was the necessity of struggling for very existence against
the Muslims that made Christian Spain great. I n the Recon-
quista Spain found its soul. The difficulty is to explain just
how this came about. A view commonly held has been that
there was an essential continuity between the Catholic Spain
of Visigothic times and that of Ferdinand and Isabella. The
difficulty about this view is that the kingdom of the Asturias,
which was the centre from which the Reconquista began,
was not part of Visigothic Spain i n any important sense, but
rather a rebel on its borders. Much nearer to the truth seems
to be the view of Americo Castro in The Structure of Spanish
History—SL view which is summed up i n the words: "Christian
Spain 'became'—emerged into being—as she incorporated and
grafted into her living process what she was compelled to by
her interaction w i t h the Moslem world" (p. 9 6 ) .
The fire of the Reconquista was certainly kindled i n the
Asturias first and then i n the other little northern kingdoms.
To begin with, however, i t was not so much a hope of re-
conquest as merely a fierce desire for independence; nor did
it arise consciously from zeal for the propagation of Chris-
tianity and its defence against Muslims. Care must be taken
not to read back into the past what belongs to later centuries.
Time and time again the actors on both sides are found pay-
ing scant attention to the differences of religion. I t seems to
148 T H E SIGNIFICANCE OF ISLAMIC SPAIN

have been about the middle of the eleventh century before

the movement of expansion of the northern kingdoms became
consciously identified w i t h the extension of Christendom; and
it was a little later that the Muslims became consciously
defenders of the territories where Islam was dominant.
I n their bitter struggle, however, first for independence
and then for the extension of their power, the men of the
north turned more and more to the Christian faith. I n particu-
lar they found i n the cult of Santiago (Saint James of Com-
postella) a source of supernatural power to sustain them i n
trials and give them hope of ultimate victory. A t the same
time they could not afford to neglect the material founda-
tions of military victory; and as far as they could they took
over the weapons and techniques which seemed to give the
Muslims superiority. Along w i t h these they adopted many
other aspects of the superior civilisation w i t h which they
were at grips. The process of assimilation was hastened by
the practice, begun at least as early as the tenth century, of
settling Mozarabs from the Muslim territories i n waste lands
on the borders of the Christian domains. Gradually these
people and their culture became an integral part of the
Christian kingdoms. A t a later stage a second factor still
further promoted assimilation—the existence of large bodies
of Muslims, the Mudejars, within the Christian states. As the
Christians from the north became familiar w i t h Toledo (after
1 0 8 5 ) , with Cordova (after 1 2 3 6 ) , with Seville (after 1248)
and with many lesser Islamic cities, they accepted the way of
life which had developed i n al-Andalus—all except its reli-
gious aspect. Thus the w i l l to reconquest—the w i l l to be
Christian Spain—found i n the culture of al-Andalus the ele-
ments which, compounded, gave i t its outward embodiment;
but these elements were far from foreign, since they were
taken from the symbiosis of Arabo-Islamic and Iberian so-
cieties.
This complexity of cultural relationships also underlies the
phenomenon of the Troubadours and some of the concep-
tions associated with chivalry. I t is impossible to say that any
particular feature is due either to the Orient or to Europe,
for there had been such a fusion of the two strands that i t
 CHRISTIAN SPAIN AND EUROPE 149
is no longer possible to make an absolute distinction between
them. Yet somehow or other i n this new unity there was
creative fire, which kindled further fires.
Likewise i t was through the symbiosis or cultural fusion
that i t was possible for Greek philosophy—both translations
of Greek books and original works i n Arabic of Muslim
thinkers—to reach Christian Europe. There was no "iron cur-
tain" between Christian Toledo and Islamic Cordova i n the
later twelfth century when Averroes was at the height of his
power; and the thought of the great Aristotelian penetrated
more easily into Christian Europe than into the Islamic heart-
lands, and constituted a large part of the stimulus which pro-
voked the greatest intellectual achievement of medieval
Christendom, the philosophy of Saint Thomas Aquinas.
While a long fist can be made of particular things which
Christian Europe owes to al-Andalus—a list which goes from
pieces of scientific knowledge and philosophical conceptions
through techniques of applied science to aspects of form i n
literature and the visual arts—it is important not to lose sight
of the general situation. Islamic culture was the main higher
culture w i t h which Western Christendom was i n direct con-
tact for much of the period under review; and behind this
culture was the most powerful political organisation of which
Western Christendom had experience. I t was really only i n
the period of the crusades that there was any close contact
with Eastern (Byzantine) Christendom—and the very con-
ception of "crusade" probably owes much to the jihad or
"holy war" of the Muslims. Because of this special relation-
ship of Western Christendom to the Islamic world—a rela-
tionship of which the focal point was Spain—it was only
natural that the Christians should feel both strong attraction
and strong repulsion. Islam was at one and the same time the
great enemy and the great source of higher material and in-
tellectual culture. Would it be fanciful to see i n this a parallel
to the relationship to contemporary Europe of the new na-
tions of Asia and Africa? I f there is a parallel, then the Euro-
pean who is able to enter imaginatively into his own history,
can experience something of what i t feels like to be a member
of one of these new nations.
150 THE SIGNIFICANCE OF ISLAMIC SPAIN

3. The Intrinsic Greatness of Islamic Spain

The third and most difficult set of questions is that con-

cerned w i t h Islamic Spain i n itself. I n particular there is the
question whether i t has an intrinsic greatness or whether its
reputation is a reflection, through the centuries, of the out-
ward magnificence which impressed the somewhat backward
medieval Christians.
There can be no doubt that the Europeans were impressed
i n this way. Some of the contributions of al-Andalus to the
life of Europe have just been described. As the point has
been made—more graphically—by Americo Castro (p. 8 7 ) ,
"those victorious armies (sc. i n 1248) could not repress their
astonishment upon beholding the grandeur of Seville; the
Christians had never possessed anything similar i n art, eco-
nomic splendour, civil organisation, technology, and scientific
and literary productivity". When allowance is made for all
this, however, and when we admit that we are still i n part
influenced by a cultural memory of former admiration for
material luxury and intellectual sublimity, are there any rea-
sons left for considering the Islamic period of Spanish history
as being among the great ages of mankind?
One test would be to ask how many of the writers and
thinkers of al-Andalus are worthy of a place among the clas-
sics of the "one world" into which we are moving. (This is,
of course, a rough and ready test, which begs many ques-
tions.) There are undoubtedly a few names which at once
suggest themselves as possible. Most obvious, perhaps, is
Averroes, partly because of his influence on Christian philoso-
phy, though he is worthy of a place in his own right. Although
i t is somewhat slight, the charm of IJayy ibn-Yaqzdn by
Ibn-Tufayl gives i t a claim to be among the immortals. Ibn-
Khaldun, i f he can be reckoned as belonging to the tradition
of al-Andalus, has also a place; but Ibn-ljazm is rather on
the border-line, since his works are closely linked w i t h the
whole dogmatic intellectual milieu and have less of a univer-
sal quality. Among the poets i t is doubtful i f any can be
said to be of universal appeal; but perhaps some of the
 INTRINSIC GREATNESS 151
mystics, like Muhyi-d-Dm ibn-al-'Arabi, w i l l be included i n
the world's pantheon of mystic "saints". Thus al-Andalus has
a few men of the very first rank; and behind these one can
discern many others i n the second rank—of whom a few are
known fairly well—whose achievements i n the strange busi-
ness of living were of no little merit. The life of al-Andalus
is indeed a noble facet of the total experience of mankind.
Apart from literature there is also the loveliness of the
Moorish buildings that have been preserved. There is some-
thing of transcendent value in a beautiful object; and i t may
be argued that a civilisation which can produce such ob-
jects must have a quality of greatness. I n general this argu-
ment may be allowed to be sound. Yet i t is instructive to
contrast our attitude to the Parthenon w i t h our attitude to
the Alhambra. Many people who admire both would be in-
clined to say that they see i n the Parthenon a thing of beauty
which is at the same time an expression of the Greek spirit,
whereas the Alhambra is for them just a thing of intrinsic
beauty without any reference to the culture which produced
it.
This contrast is worth looking at more closely. I t is in-
deed natural that we should have a much greater apprecia-
tion of Greek culture than of Moorish culture. The former
—or at least a selection from it—is part of our own heritage,
part of the tradition into which we enter; but the latter, for
all that i t has contributed to the culture of Europe, was i n
its essence something alien, the great enemy, to be feared
even i n the moment of admiration. Our inherited "image" of
Islam was framed i n the twelfth and thirteenth centuries un-
der the domination of this fear of the Saracen; and even now
few western Europeans can regard Islam w i t h impartiality.
Yet should our appreciation of a beautiful object be affected
by our lack of appreciation of the culture from which i t
springs? May i t not, on the contrary, be the case that appreci-
ation of a beautiful object is able to provide a key to the
appreciation of the alien culture? May i t not even be that the
beautiful object is the measure and the validation of the cul-
ture? Because of lovely buildings like the Great Mosque of
Cordova and the Alhambra of Granada must not the culture
of Islamic Spain be a great culture?
152 T H E SIGNIFICANCE OF ISLAMIC SPAIN

This point w i l l bear expansion. There is an obvious dif-

ference between the Parthenon and the Alhambra. When we
admire the Parthenon, we mainly do so from outside, whereas
it is only from inside that the Alhambra can be admired. This
has nothing to do with the contrast of religious and secular
purposes, for i t is also true that the glories of the Great
Mosque of Cordova are mainly within. I t has further been
suggested that the slender pillars of the Alhambra with the
elaborate and massive overstructure express the descent from
a supernal realm of that which is of eternal value and signifi-
cance, whereas other buildings express rather mans attempt
to rise up to heaven. Now suggestions of this type may be
multiplied and elaborated, and some w i l l no doubt w i n wider
approval than others. Even the best, however, is bound to
be to a great extent inadequate, for the appreciation of beauty
can never be reduced to conceptual terms. Nevertheless, if
there is something i n the above suggestion which even dis-
tantly approximates to the essence of the appreciated beauty,
then man in the Western-European tradition who finds the
beauty of the Alhambra touching a responsive chord i n him-
self is acknowledging the high intrinsic worth of this expres-
sion of the soul of Islamic Spain, and providing himself with
a key to a deeper understanding of this whole culture.
 NOTES

1
THE MUSLIM CONQUEST

Abbreviations
E I , E I , E I ( S ) ; Encyclopaedia of Islam, four volumes and sup-
1 2

plement.
GAL, GALS: Geschichte der Arabischen Literatur.

1. For a general account of Arab expansion, cf. Bernard Lewis,

The Arabs in History, London, 1950, chs. 3, 4.
2. The best and most complete account is the composite volume
with the separate title Espana visigoda, which constitutes
volume I I I of the Historia de Espana directed by Ram6n
Menendez Pidal (Madrid, 1940). This contains an important
section on Visigothic art and is lavishly illustrated. I n English
the doctoral dissertation of Aloysius K. Ziegler, entitled Church
and State in Visigothic Spain (Washington, 1930), is to be
highly recommended. The first chapter, pp. 1-25, of Islam
d'Espagne by Henri Terrasse (cf. p. 188) is also worthy of
mention.
3. Detailed justification of the statements here, will be found in
Levi-Provencal, Histoire de TEspagne musulmane, i . 8-34 (cf.
p. 162 below).

2
T H E PROVINCE OF T H E DAMASCUS
CALIPHATE

1. Cf. Americo Castro, The Structure of Spanish History, 77-9,

and contrast Levi-Provencal, Histoire, i . 66, 68, which appears
to show that Pelayo and Alfonso I were of Visigothic descent.
154 NOTES

3
THE INDEPENDENT UMAYYAD EMIRATE

1. The form (in Latin) muztarabes is quoted from a document

of Alfonso V I dated 1101 by Isidro de las Cagigas, Los
Mozdrahes (Madrid, 1947), i . 72 n. 31.
2. The continuance of Latin or Romance speech is emphasised by
Armand Abel in Unity and Variety in Muslim Civilization, ed.
G. E. von Grunebaum, Chicago, 1955, 207.
3. The word "feudalism" is used here in a loose sense by a his-
torian of Islam to indicate relationships which to his inexpert
eye are more like European feudalism than anything in the
Islamic world. What is here said is therefore compatible with
the common view of historians of Europe that in Spain there
was no feudalism in the strict sense.

T H E GRANDEUR OF T H E U M A Y Y A D
CALIPHATE

1. Cf. Montgomery Watt, Islam and the Integration of Society,

London, 1961, 158 f.
2. Cf. Montgomery Watt, Islamic Philosophy and Theology,
Edinburgh, 1962, 100-3.
3. The Structure of Spanish History, 167-9; cf. 149.
4. For this evidence of language see Castro, Structure of Spanish
History, 96-100; cf. T. W. Arnold and A. Guillaume (eds.),
The Legacy of Islam, Oxford, 1931, 19-24.
5. H . Heaton, Economic History of Europe, 2nd ed., New York,
1948, 78.
6. Cf. T. W. Arnold, The Preaching of Islam, 3rd ed., London,
1935, 131-4.
7. Zeitschrift der deutschen morgenlandischen Gesellschaft, liii
(1899), 601-20; the two judges are on p. 605, the phrase
used being shadid al-asabiyya li-l-muwalladin. I n n. 2 the
figure should be "1147".
 NOTES 155
8. Alvar, Indiculus luminosus, § 35; quoted from Arnold, Preach-
ing of Islam, 137 f.
9. Dr E. Bosworth calls attention to a similar adaptation in India
in the eleventh and twelfth centuries, where the population
was too extensive to convert (private communication).
10. Cf. Montgomery Watt, Muslim Intellectual, Edinburgh, 1963,
88-108.

CULTURAL ACHIEVEMENTS
U N D E R T H E UMAYYADS

1. Cf. Montgomery Watt, Integration, 199-209. For a general ac-

count of Islamic law, cf. N. J. Coulson, Islamic Law, Edin-
burgh, 1964.
2. E.g. al-Ghazi ibn-Qays (Ibn-al-Faradi, no. 1013).
3. Cf. J. Schacht, The Origins of Muhammadan Jurisprudence,
Oxford, 1950, 288 f., 311-14.
4. M . Talbi, "Kairouan et le malikisme espagnol", in Etudes
Levi-Provengal, Paris, 1962, i . 317-37; cf. R. Brunschvig in
Al-Andalus, xv (1950), 401. The books were the Asadiyya of
Asad ibn-al-Furat and the Mudawwana of Sahnun (d. 854),
the latter being by far the more important.
5. Cf. Islamic Philosophy and Theology, ch. 7.
6. M . Asin Palacios, Obras escogidas, i (Madrid, 1946), 1-216,
"Ibn Masarra y su escuela"; Appendix I I (179-84) deals with
the first Mu'tazilites in Spain.
7. Spanish tr. by J. Ribera, Historia de los Jueces de Cordoba pot
Aljoxani, Madrid, 1914.
8. Moorish Poetry, introd., xiii.
9. See ch. 4, n. 8 (and p. 47).
10. "Les M6thodes de realisation artistique des peuples de
l'lslam", Syria (1921), 19.
11. What is said in this section (and the others dealing with art)
follows closely the treatment of Henri Terrasse in Islam
d'Espagne (see Bibliography).
12. Op. cit. 101.
13. 'Uthman ibn-al-Muthanna (Ibn-al-Faradi, no. 889) and Faraj
ibn-Sallam (ibid. no. 1036).
156 NOTES

14. Ibid. no. 199; R. Guest, The Governors and Judges of Egypt,
548. An example of how far Spanish Muslims had assimilated
the ideals of pre-Islamic Arabs will be found in A. Abel's
article in Unity and Variety, ed. von Grunebaum, 214.
15. Abu-'l-Hasan al-Antaki (Ibn-al-Faradi, no. 132; as-Subki, i i .
313; Shadhardt adh-Dhahab, iii. go); he is said to have lived
from what his slave-girl earned by spinning, but no date is as-
signed to this. For the other scholar see Ibn-al-Faradi, no. 933.
16. Histoire, ii. 147 f.

6
T H E COLLAPSE OF ARAB R U L E

1. Cf. A. Castro, The Structure of Spanish History, 158: "The

damage wrought by this Islamic thunderbolt increased the faith
in the holy relic, so holy that not even Almanzor himself had
succeeded in destroying i t " .
2. Cf. Islam and the Integration of Society, ch. 5; and p. 43
above.
3. For a general picture of conditions a little later, cf. Levi-
Provencal, "Memoires de 'Abdullah . . . de Grenade", Al-
Andalus, iii, iv (1935, 1936).
4. Cf. E I , art. Abbadids.
2

5. Cf. E I , art. Djahwarids.

2

6. There has been much discussion of the historical basis of the

legends clustering round the Cid, but this belongs primarily
to the history of Christian Spain. The best work on the subject
is R. Menendez Pidal, E l Espana de Mio Cid, 1929, etc. Eng-
lish tr. The Cid and his Spain, London, 1934, etc.

T H E BERBER EMPIRES—
THE ALMORAVIDS

l . The fullest account is that of Alfred Bel in La Religion musul-

mane en Berberie, I , Paris, 1938, 211-31 (who had previously
 NOTES 157
been responsible for the article "Almoravides" in E I ; the cor-
1

responding article in E I will appear as "Murabitun").

2

2. GALS, i . 660. "Fasl" means "coming from Fez".

3. Cf. A. Gonzalez Palencia, Historia de la Espana musulmana*
Madrid, 1945, 96-8.

8
T H E BERBER EMPIRES—
THE ALMOHADS

1. General references: I . Goldziher, "Materialien zur Kentniss

der Almohaden-Bewegung", Zeitschrift der deutschen morgen-
landischen Gesellschaft, xli (1887), 30-140; also his "Intro-
duction" to J. D. Lueiani, he Livre de Moliammed ibn
Toumert, Algiers, 1903; these are still unsurpassed. Articles:
"Ibn Tumart" in £I(S); " ' A b d al-Mu'min" in E I ; "Muwali-
2

hidun" (not yet available) in E I ; R. Brunschvig, "Sur la doc-

2

trine du Mahdi Ibn Tumart", Ignace Goldziher Memorial

Volume (ed. S. Lowinger), i i . 1-13. A. Huici Miranda, His-
toria politica del imperio almohade, Tetuan, 1956-7.
2. Cf. J. F. P. Hopkins, "The Almohade Hierarchy", Bulletin of
the School of Oriental and African Studies, xvi (1954), 93-112.
3. A. Castro, Structure of Spanish History, 130 ff., esp. 167-9.

C U L T U R A L GREATNESS I N
POLITICAL DECLINE

1. G. Margais, La Berberie musulmane et FOrient au Moyen

Age, 12.
2. See Nykl, Hispano-Arabic Poetry, 135-6.
3. Ibid., 229-30.
4. See a somewhat different translation in Nykl, p. 228. Najd is
the central upland of Arabia, Tihama the coastal strip. It was
a widespread convention in Arabic poetry to speak of the loved
woman in the masculine gender.
158 NOTES

5. Peres, PoSsie andalouse, 424.

6. Garcia G6mez, Poemas ardbigoandaluces, 43-4.
7. Op. cit. pp. 465-6.
8. Op. cit., particularly pp. 473-5.
g. I n intro. of Ibn Khallikan's Biographical Dictionary I (Paris,
1842), p. XXXV.
10. See Peres, op. cit. pp. 159-60.
11. The lines are by Ibn-Quzman (d. 1160), the vizier of al-
Mutawakkil of Badajoz. Nykl gives a different translation in
Hispano-Arabic Poetry, p. 302.
12. Garcia G6mez, Poesia ardbigoandaluza, pp. 30-31.
13. This is, in substance, the hypothesis first put forward by Ribera
and taken up with modifications by Garcia G6mez in "Sobre
un posible tercer tipo de poesia ardbigoandaluza", Estudios
dedicados a Menindez Pidal, I I (Madrid, 1951), pp. 397-
408, to account for the appearance of strophic forms. I t does
not seem far-fetched to extend it to the modest changes ob-
servable in conventional poetry.
14. See S. M . Stern, Les Chansons mozarabes, xiii-xvii.
15. Ddr at-Tirdz (Damascus, 1949), 37.
16. See Garcia G6mez, "Sobre un posible tercer tipo de poesia
arabigoandaluza", 400-1.
17. These non-classical forms were first brought to the attention
of Orientalists by von Hammer in "Notice: Sur dix formes de
versification arabe dont une couple a peine etait connue
jusqu'a present des orientalistes europeens", Journal Asiatique,
3 serie, Tome vni (aout, 1839), 162-71. More recently, a
e

great deal of information scattered through the sources has

been usefully brought together by 'All al-Khaqani in Funun al-
Adab ash-Sha hi (Baghdad, 1962).
18. See Nykl, Hispano-Arabic Poetry, p. 269.
19. Poesia ardbigoandaluza, 43-9.
20. Almost certainly not the same man as the one mentioned in
n. 11, but there is some confusion in the sources.
21. PoSsie andalouse, 24-5.
22. The word is usually, justifiably, translated "assembly", but it
can also be used of a dramatic situation, a prowess, a harangue,
etc. See R. Blachere, "fitude semantique sur le nom maqama",
Mashriq, 47 (1953), 646-52.
23. The entire output of Andalusian maqdmas has yet to be ade-
quately studied. The material available has been conveniently
indicated by Dr. Ahmad Mukhtar al-'Abbadi in "Maqamat
al-'Id . . .", Revista del Instituto Egipcio de Estudios Isldmicos,
2 (1954), 159-73.
 NOTES 159
24. The broad question of Muslim influences on Dante was first
raised by Miguel Asin Palacios in La escatologia musulmana
en la Divina Comedia (Madrid, 1919) and gave rise to a lively
polemic, reviewed by Asin Palacios in the second edition of
his book (Madrid, 1943). A direct link between al-Ma'arri
and Dante is, however, entirely conjectural, and receives little
credence either from Orientalists or from modern Arab scholars
—see 'Abd-ar-Razzaq Humayda, Fl %Adab al-Muqaran (Cairo,
1948), 91-101. A likelier channel through which Muslim
eschatological ideas may have reached Dante is mentioned
p. 137 below.
25. Op. cit. 37-8.
26. For Ibn-Hazm see article in E I . Also: I . Goldziher, Die
Zdhiriten, Leipzig, 1884, esp. 116-72, which discusses his juris-
prudence and theology. M . Asin Palacios, Abenhdzam de
Cordoba, y su historia critica de las ideas religiosas, five vols.,
Madrid, 1927-32, has (in the first vol.) a useful discussion of
his life and of his thought in general; the rest is a translation
or summary, at times misleading, of his book Al-Fisal on the
sects. The important study by Roger Arnaldez, Grammaire et
theologie chez Ibn Hazm de Cordoue, Paris, 1956, links up
his theological position with particular views about the nature
of language.
27. L^vi-Provencal, Histoire, iii. 182.
28. (Tawq al-hamdma), translated by A. J. Arberry, London,
1953. This is to be preferred to the earlier English translation
by A. Nykl. There are also translations in several other Euro-
pean languages.
29. For a list of his works cf. GALS, i . 692-7.
30. GAL, i . 453 f.; GALS, i. 628 f.
31. GAL, i . 455 f.; GALS, i . 630-2; Tor Andrae, Die Person Mti-
hammeds in Lehre und Glauben seiner Gemeinde, Stockholm,
1918, 60, 112, 118f., 131, 147-50, 204-12, etc. (the Shifa
has been an important source for Andrae of earlier views);
Andrae is mistaken in calling l y a d an Ash'arite—it was as
Malikite that al-Raqillani was his "Gesinnungsgenosse".
32. There is a full account in R. Brunschvig, "Averroes juriste",
Etudes L4vi-Provengal, i (Paris, 1962), 35-68.
33. A conspectus is given by Charles Pellat, "The Origin and
Development of Historiography in Muslim Spain", in B. Lewis
and P. M . Holt (eds.), Historians of the Middle East, Lon-
don, 1962, 118-25.
34. Ibn Khaldun, The Muqaddimah, an Introduction to History, tr.
by Franz Rosenthal, 3 vols. London, 1958, p. xxxvi.
i6o NOTES

35. The Rule of the Solitary, ed. and tr. by M . Asin Palacios,
Madrid, 1946. English tr. of first section by D. M . Dunlop in
Journal of the Royal Asiatic Society, 1945, 61-81.
36. For al-Ghazall see Islamic Philosophy and Theology, ch. 13;
and Watt, Muslim Intellectual, Edinburgh, 1963.
37. Translated by S. Ockley as The Improvement of Human Rea-
son, London, 1708; revised by A. S. Fulton, London, 1929.
Also by P. Bronnle, as The Awakening of the Soul, London,
1904. See also GAL, i . 602 f.; GALS, i . 831 f.
38. There is an excellent English translation by Simon van den
Bergh, with the title: Averroes Tahafut al-Tahafut (The In-
9

coherence of the Incoherence), London, 1954 (Gibb Memorial

Series). Another useful translation is Ibn Rushd (Averroes)
on the Harmony of Religion and Philosophy, by George F.
Hourani, London, 1962. For further references see Watt,
Islamic Philosophy and Theology, 144 f.
39. Cf. GAL, i . 611; GALS, i . 844.
40. M . Asin Palacios, " E l mistico Abu-l-'Abbas ibn al-'Arif de
Almeria . . .", Obras escogidas, i (Madrid, 1946), 217-42.
The form I r r i f is preferable to 'Arif according to GAL, i . 559,
where there is also a notice of Ibn-Barrajan.
41. Obras escogidas, i . 145-51. In general cf. A. J. Arberry, Sufism,
London, 1950, 97-101; A. E. Affifi, The Mystical Philosophy
of Muhyid Din Ibnul 'Arabi, Cambridge, 1939; Tarjumdn al-
Aswdq, tr. by R. A. Nicholson, London, 1911; H . Corbin,
Vlmagination crSatrice dans le goufisme d'lhn 'Arabi, Paris,
1958; A. Jeffery, "Ibn al-'Arabfs Shajarat al-Kawn" (transla-
tion), Studia Islamica, x. 43-77, xi. 113-160.
42. GALS, i i . 358: M . Asin Palacios, "Un precursor hispano-
musulman de San Juan de la Cruz", Obras escogidas, i . 243-
326 (from Al-Andalus, i [1933]. 7-79); Lettres de direction
spirituelle, tr. P. Nwyia, Beirut, 1958.

THE LAST OF I S L A M I C SPAIN

1. Illustrative passages are to be found in Khuldsat Ta'rikh Tunus

by Hasan as-Sumadihi (Tunis, 1925), pp. 141 ff. (reference
communicated by G. M. Wickens).
2. Muhammad Mzali, "Al-Butula kama yusawwiruha 1-adab al-
 NOTES

'arabi ffl-Andalus wa shamal Ifriqya", al-Fikr, v. 4, n. 5 (Feb.

1959), 22-33.
3. S. M . Stern, " 'Asiqayn Ftanaqa—An Arabic Muwassah and its
Hebrew Imitations", al-Andalus, 28, 1 (1963), 155-70.
4. See Angel Gonzalez Palencia, Historia de la literatura ardbigo-
espanola, 334-48.
5. See Enrico Cerulli, Libro della Scala e la questione delle fonti
arabo-spagnole della Divina Commedia (Vatican, 1949), and
Francesco Gabrieli, "Daw* jadid 'ala Danti wa l-lslam", Revue
de VAcadSmie Arabe de Damns, 33, Pt. 1 (Jan. 1958), 36-55.
6. Among those who put forward the view that Arabic has been
influential are P. Juan Andre, Ribera, Burdach, Singer, Mulertt,
Nykl and Menendez Pidal. Among those who favour rival
theories are Gaston Paris, Jeanroy, Pillet, Schrotter and Vossler.
7. Ahmad Lutfi 'Abd al-Badf, "At-Trubadur Gharsiyah Firnan-
dith", Revista del Institute Egipcio de Estudios Isldmicos, i i ,
1-2 (1954), 85-92.
8. See L6vi-Provencal, "Arabica Occidentalia, I I " , Arabica, i
(1954), 201-11.
9. If, for example, the essentials of the zajal are reduced to stanzas
with the rhyme scheme bbbA, cccA, etc., parallels will be
found as far afield as in Scottish and Irish ballads; see Gon-
zalez Palencia, op. cit. 357 ff. Nykl and Menendez Pidal, how-
ever, base their conclusions on far more complex analogies.
10. Poesia arabe y poesia europea, 68-78.
 BIBLIOGRAPHY

A. D E T A I L E D HISTORY

For the period up to 1031 there may be said to be a standard his-

toriographies tradition, that established by Reinhardt Dozy and
continued by fivariste Levi-Provencal. Both did a vast amount of
work in making available the basic source-material, including that
for the later period. Dozy's Histoire des musulmans d'Espagne,
711-1110 was first published at Leiden in 1861, and translated
into English as Spanish Islam (London, 1913). Levi-Provencal
brought out a revised edition in 1932, but later decided that an en-
tirely new work was needed. Unfortunately only three volumes of
his Histoire de YEspagne musulmane had appeared before his
death. These take the story to 1031, and completely supersede the
earlier works mentioned. Their full bibliographical references are
indispensable for the serious student, and make it unnecessary to
include references in the earlier part of the present survey. Since
the standard view of the earlier history of Islamic Spain is thus
mainly the work of two closely associated men, it may be that when
some scholar with a different perspective familiarises himself with
all the material the general line of interpretation will be modified.
For the period after 1031 there is no such historiographical tra-
dition and no single work covering the period in any detail. This
is a serious gap in modern historical studies, and is reflected in the
sections on this period in large composite works and in the smaller
popular books on Islamic Spain. For the rest of the eleventh cen-
tury there is the third volume of Levi-ProvengaFs revision of Dozy,
but he would probably have altered the treatment considerably in
the projected fourth volume of his own history. The Almohad
period is dealt with in some detail by Ambrosio Huici Miranda in
his Historia politica del imperio almohade, Tetuan, 1956-7. There
are numerous works in Spanish on particular matters, usually
from a specifically Christian angle; for example, the four volumes
of A. Gonzalez Palencia on Los Moz&rabes de Toledo en los sighs
XII y XIII, Madrid, 1926-30.
Of shorter popular works one of the best is that by the last au-
thor, entitled Historia de la Espana musulmana, fourth revised edi-
 BIBLIOGRAPHY 163
tion, Barcelona, 194S; but this is now beginning to be out of date.
A Political History of Muslim Spain (Dacca, 1961), by a Pakistani
scholar S. M . Imamuddin, is competent but apparently based on
the older work of L6vi-Provengal, and rather fades away when
that ceases. Much better than either of these, though not strictly a
history, is Islam d'Espagne by Henri Terrasse (to be mentioned in
section B ) . Philip K. Hitti in his History of the Arabs (London,
1937, etc., various editions) devotes most of Part I V (over a hun-
dred pages) to Spain, and is specially strong on the intellectual
contributions. I n the large composite Historia de Espana under
the direction of R. Menendez Pidal, vols. I V and V are simply
translations of L6vi-Proven§al and the later volumes are not yet
published.
There are two further ways in which the gaps still left may some-
times be filled. There are many articles concerning Islamic Spain,
sometimes containing material not otherwise available in European
languages, in The Encyclopaedia of Islam: four volumes and sup-
plement, Leiden, 1913-42; second edition, vol. 1, Leiden and Lon-
don, i960, continuing. Articles of religious interest from the first
edition, often with some revision, were reprinted in a separate
volume as Handwdrterbuch des Islam (1941) and A Shorter En-
cyclopaedia of Islam (1953). There is naturally a tendency for the
older articles to be out of date. I n the second place there is a vast
number of articles in periodicals of all kinds; and a classified list of
these is to be found in J. D. Pearsons Index Islamicus, 1906-55
(Cambridge, 1958) and the first supplement, 1956-60. The sec-
tions dealing with Islamic Spain are XXXV (history) and XXXVII,
h (literature).

B. G E N E R A L INTERPRETATIONS

There have been vehement debates within Spain and beyond its
frontiers about the general interpretation of Spanish history and
the significance of the Moorish period for the whole. The romantic
aspect of Moorish Spain seems to have caught the imagination of
Europe as a result of the publication of Tales of the Alhambra by
Washington Irving in 1832. In this spirit Stanley Lane-Poole,
who admired the Arabs but disliked contemporary Spaniards,
thought Spain's greatness was due to the Moors and that her de-
cadence began when she expelled them {The Moors in Spain,
London, 1888). Catholic Spaniards, on the other hand, have some-
times tended to regard the period of Islamic domination as a mere
164 BIBLIOGRAPHY

interruption in the continuing life of a single entity, Catholic Spain.

With many refinements this is the idea behind the writings of
C. Sanchez Albornoz. A more exciting and apparently more bal-
anced treatment of the complex questions at issue—and one more
congenial to the Islamist—is that of Americo Castro in The Struc-
ture of Spanish History (tr. by E. L . King, Princeton, 1954), a
revision of his Espana en su historia: cristianos, mows y judios
(Buenos Aires, 1948). His general thesis is that there was no con-
tinuity between Visigothic Spain and later Christian Spain, but that
the latter was something new which was born and grew up in the
mixed culture (largely Arab) which developed under the Muslims.
Of books specially devoted to Islamic Spain one is outstanding:
Islam d'Espagne, une rencontre de VOrient et de YOccident by
Henri Terrasse, Paris, 1958. The author's primary interest is in
the history of art and archaeology, and he makes full use of ma-
terials from this sphere in putting forward a view of Islamic Spain
which is roughly that of Americo Castro. Valuable, too, are the
three lectures, first delivered in Egypt in 1938, in which Levi-
Provencal shares his general reflections on the subject to which he
had given his life: La Civilisation arahe en Espagne, vue generate
(new edition, Paris, 1948).

C. L I T E R A T U R E

References in Section " A " to the Encyclopaedia of Islam and to

J. D. Pearson's Index Islamicus are equally relevant here. There
are also important contributions to the understanding of Spanish-
Arabic literature scattered in the historical works of Dozy and
Levi-Provencal, and in Julian Ribera y Tarrago's Disertaciones y
opusculos, 2 vols. (Madrid, 1928).
The fundamental work for all Islamic source material in Arabic
is Carl Brockelmann's Geschichte der Arabischen Literatur, consist-
ing of two original volumes (second edition, Leiden, 1943, 1949)
and three supplementary volumes (1937-42). This work lists all
the manuscripts of Islamic Arabic works known to Western scholars
up to the time of publication, as well as the printed editions. I t
also gives the dates of each author, where known, and a few
biographical details, with references to notices in Arabic biographi-
cal dictionaries and to modern books and articles. The primary
division is into chronological periods. I n each period there are
 BIBLIOGRAPHY 165
subdivisions according to subject, but each author appears only in
one place.
Not to be overlooked are the sections on Spain in general his-
tories of Arabic literature, notably: Reynold A. Nicholson, Liter-
ary History of the Arabs (second edition, Cambridge, 1930), pp.
405-41; Sir Hamilton .Gibb, Arabic Literature, an Introduction
(second edition, Oxford, 1963), pp. 108-17, 136-41; Francesco
Gabrieli, Storia della letteratura araba (Milan, 1951), pp. 163-76,
236-8, 247-54, 262-69; J.-M. Abd-el-Jalil, Histoire de la UttSrature
arabe (second edition, Paris, i960), pp. 191-203.
The one comprehensive book on Spanish-Arabic literature is
Historia de la literatura ardbigo-espanola, by Angel Gonzalez
Palencia (second edition, Madrid, 1945); it sums up in a con-
venient form the results obtained by scholars, mostly Spanish and
French, until twenty years ago.
On poetry, Adolf Friedrich von Schaek's Poesie und Kunst der
Araber in Spanien und Sicilien (Berlin, 1865; second edition, Stutt-
gart, 1877) is largely out of date, but remains useful for its ex-
tensive verse translations.
Most active of contemporary students of Andalusian poetry is
Emilio Garcia Gomez, editor and translator of a number of impor-
tant texts, author of perceptive studies of which the following are
available in book form: Poesia ardbigoandaluza (Madrid, 1952),
Cinco poetas musulmanes (Madrid, 1944). He also surveys the
entire development of Andalusian poetry in his introduction to
Poemas ardbigoandaluces (Madrid, 1930, 1943).
In La Poesie andalouse en arabe classique au XI siecle (Paris,
6

1937, 1953), Henri Peres is inclined to overstress non-Arab and

non-Islamic elements, and his painstaking tabulation of poetic
motifs may need to be corrected at some points for the effects of
convention on one hand, of the peculiarities of some poets on the
other; he does nevertheless sift and present in a masterly form an
enormous amount of material pertaining to a key period.
The bulk of A. R. Nykl's Hispano-Arabic Poetry and its Rela-
tions with the old Provencal Troubadours (Baltimore, 1946) is a
survey of Andalusian poetry period by period, with biographical
notices of a great many poets and translations of selected passages.
Surprisingly—for he gives ample evidence of extensive erudition—
NykFs translations from the Arabic are often very wide of the mark.
His selections, too, appear to be influenced by strong personal likes
and dislikes, and by his overriding interest in a link with Trouba-
dour poetry—see also his Troubadour Studies (Cambridge, Mass.,
1944). Yet there is a wealth of information in this work, not least
in the bibliographical notes.
i66 BIBLIOGRAPHY

A pioneer work on strophic poetry is Martin Hartmann's Das

Arabische Strophengedicht. I : Das Muwassah (Weimar, 1897).
S. M . Stern's Les Chansons mozarabes (Palermo, 1953)—an at-
tempt to decipher all the known kharjas composed in a mixture of
Romance and Arabic—is one of several important contributions
this author has made to the study of the muwashshah, and he has
announced his intention of publishing a comprehensive work on
the subject.
Modern Arab scholars are showing a growing interest in al-
Andalus. Among their books specifically dealing with literature may
be noted: az-Zajal ft 1-Andalus, a series of lectures by 'Abd-al-'Azfz
al-Ahwani (Cairo, 1957); Udaba* al-Arah ft I-Audalus wa Asr ah €

Inbtdth by Butrus al-Bustani (Beirut, 1947); Al-Adab ahAndalust

min al-Fail} ild Suqut al-Khilafa by Ahmad Haykal (Cairo, 1958);
Tarikh al-Adab al-Andahm-Asr Siyadat Qurtuba by Ihsan 'Abbas
(Beirut, i960); Ft %Adab al-Andalusl (Damascus, 1957) and At-
Tahta ft 'sh-Shir al-Andalusl (Damascus, 1959) by Jawdat ar-
Rikabi. Vol. 3 of Ahmad Amin's Zuhr al-Isldm (Cairo, 1955) is
devoted to al-Andalus.
Of Andalusian literary works in translation, one of the most use-
ful is an anthology entitled The Pennants (Rdydt al-Mubarrizln)
compiled by Ibn-Sa'id in 1243. Although unequal in quality, and
although only about half of the poems in it are by Andalusians, it
is representative of the taste of the time. I t forms the basis of
Garcia G6mez's Poemas ardbigoandaluces (Madrid, 1930, 1943),
and has been very ably translated into English verse by A. J. Ar-
berry under the title Moorish Poetry (Cambridge, 1953). Several
translations also exist of Ibn-Hazm's treatise on love, The Ring of
the Dove, including two in English: one by Nykl in 1931, and a
superior one by Arberry (London, 1953). Some fifty poems and
one epistle by Ibn-Zaydun are translated by Auguste Cour in Un
Podte arabe d'Andalousie (Constantine, 1920), and the poems of
Ibn az-Zaqqaq have been edited and translated by Garcia G6mez
(Madrid, 1956).
Abu Bakr at-Turtushfs Siraj al-Muluk has been translated by
Maximiliano Alarc6n under the title of Ldmpara de los principes
(two volumes, Madrid, 1930). The Risala of ash-Shaqundi by
Garcia G6mez under the title of Elogio del Islam espanol
(Granada, 1934), and by A. Luya in Hesperis, xxii (1936).
Ibn-TufayFs Hayy ibn-Yaqzdn was first translated into English
by S. Ockley as The Improvement of Human Reason (London,
1708), and this was revised by A. S. Fulton (London, 1929). A
popular, abridged version entitled The Awakening of the Soul, by
Paul Bronnle, was first published in 1904. I t has also been trans-
 BIBLIOGRAPHY 167
lated into Spanish, as E l filosofo autodidacto, by Gonzalez Palencia
(Madrid, 1934) and into French by Leon Gauthier (Beirut, 1936).
A popular traditional narrative has been edited and translated
by Garcia Gomez under the title Un texto drabe occidental de la
Leyenda de Alejandro (Madrid, 1929).
 INDEX
The Arabic article al-, with its variants such as an-, ash-,
etc., is neglected in the alphabetical arrangement
'Abbadid dynasty, 79-80, 97
'Abbasid caliphate
Almohad attitude towards, 94
Almoravid loyalty to, 84
armihilation of Umayyads,
22, 24
autocracy of, 13
imitation of administrative
practices in Spain, 50
jurists under, 52-53
Persian influences on, 44, 51,
75, 144, 145
weakness of, 75
*Abd-al-'Azfz, 10
'Abd-Allah
dates of reign, 28
reliance on jurists, 31
'Abd-al-Malik, see al-Muzaffar
'Abd-al-Mu'min
dates of reign, 91
designated successor to Ibn-
Tumart, 90
plans for occupation of Spain,
9i
successes in North Africa, 90-
91
'Abd-ar-Rahman, see al-Ma*-
mun
'Abd-ax-Rahman I
administration of justice un-
der, 54
as a poet, 61
buildings erected by, 65
dates of reign, 24
early career of, 22-23
novelty of situation, 24-25
proclaimed emir of al-
Andalus, 23
racial problems under, 25-26
raising of professional army,
26
'Abd-ar-Rahman I I
building programme, 28, 65
dates of reign, 24
part in politics, 28
prosperity of his reign, 28
Abd-ar-Rahman I I I
assumption of titles, 38, 48
autocracy of, 38, 49
Christian kingdoms under,
33-34
control of the Marches, 32,
33
cultural achievements under,
58
dates of reign, 32
economic basis of prosperity
of reign, 40-43
encouragement of cultural
activities, 68, 69, 111
influence of jurists under, 50
North African policy, 36-38
ode on, 63
prosperity of reign, 38, 40, 75
restoration of internal unity,
3^-33
slaves recruited for army of,
48
strength of central govern-
ment under, 30
use of the idea of jihad, 35
Abu-'Abd-Allah, 129
Abu-'Amir ibn-Shuhayd, 101-2,
109-10
Abu Bakr at-Turtushi, 106
Abu-Bakr ibn-'Umar, 83
Abu-Bakr Muhammad ibn-
Zuhr, 104
 INDEX
Abu-Dawud al-I§fahani, 106
Abu-Hayyan, 134
Abu-lmran al-Fasi, 82, 83
Abu-'l-'Abbas Ahmad ash-
Sharishi, 128
Abu-VAbbas al-Mursi, 124
Abu-'l-'Atahiya, 59
Abu-l-Walid al-Himyarl, 106
Abu-Madyan, 124
Abu-Nuwas, 59, 104
Abu-Tammam, 68
Abu-'t-Tahir Muhammad at-
Tamml as-Saraqusti al-
Ashtarkuni, 108
Abu-Yaqub Yusuf I
dates of reign, 91
death of, 92
friendship with Averroes, 121
Abu-Ya qub Yusuf I I
dates of reign, 91
succession of, 93
Abu-Yiisuf Ya'qub al-Mansur
campaigns in Spain, 92
dates of reign, 91
friendship with Averroes, 121
leanings towards Zahirism,
117
adah, 106
administration
as a province of the Damas-
cus Caliphate, 12-23
consultation with leading
Arabs, 15
establishment of, 10-11
governors of the province, 15
influence of Persian methods,
50-51, 145
of justice, 54
organisation of the province,
12-15
under the Umayyad Cali-
phate, 48-51
aduana, 42
agriculture
advances in, 42
in al-Andalus, 40-41
irrigation, 40
Ahmad ar-Razi, 58
Akhila, dux of Tarraconensis, 7
169
Aledo, fortress of, 84
Alfonso I , 19-20, 86
Alfonso I I I , 31
Alfonso V I
campaigns against the Mus-
lims, 81, 84-85
capture of Toledo, 80-81
Alfonso V I I , 86
Alfonso V I I I , 92
Alfonso X, 137, 138
Alhambra
architecture of, 141-42, 151-
5^
decoration on walls of, 135
'All ibn-Yusuf, 84
aljamiados, 133, 135
Almohads
Almoravids and, 89
art of, 125, 126
campaigns in Spain, 91-92
Christian successes against,
9^-94
continued existence in North
Africa, 95
defeat of the Almoravids, 90
era of, 87-88
Ibn-Tumart and, 89-90
Islam and, 144-45
Malikism and, 94, 122
origin of name, 89
philosophy under, 118-23
religious sciences under, 116—
Spain under, 91-95
theology of, 119
Almoravids
Almohads and, 89
art of, 125, 126
assessment of, 86-87
Christian attacks on, 86
combination of religious and
political aims, 83-84
decline of, 86
defeat by Almohads, 90
dissatisfaction of ordinary
people with, 86—87, 94
effective end of rule in Spain,
87-88
expansion of state, 83-84
170 INDEX
Almoravids (cont'd)
foundations of state, 82-84
in Spain, 84-88
Islam and, 87, 144-45
Malikite jurists and, 81, 82,
83, 84, 85, 86-87, 93, 94,
115, 11&-17, 119
origin of name, 83
philosophy under, 118-19
religious sciences under, 115—
17
state in North Africa, 81
suspicion of mysticism, 123-
24
amir al-mumiriin, 38
'Amirid dictatorship
and the breakdown, 70-74
reasons for the breakdown,
74-78
al-Andalus
internal tensions of, 20-23
organisation of, 12-15
origin of name, 12
Andalusia, 12
Andalusians
defined, 79
spheres of influence in the era
of the "party kings", 79-
80
anthologies, 106
Arab
and Islamic colony, 143-47
conquest of Spain: as a phase
of expansion, 1-5; charac-
ter of expansion, 1-2; end
of the advance, 15-20; ex-
peditionary forces, 2-4;
need for Arab unity and,
1-2; North African ad-
vance, 4-5; reasons for ex-
tension to Spain, 4-5
domination of influence in
Umayyad dynasty, 44,
143-45
landholders in al-Andalus,
15, 44
rule, the collapse of: absence
of middle class, 77-78;
materialism and, 74-75;
particularism and, 74; po-
litical events of, 72-74;
politics and religion, 75-
77; reasons for, 74-78; the
"party kings", 78-81
unity: and jihad, 2-3; and
nomadism, 2
Arabian Nights, 60, 108, 137
Arabic
poetry and, 59, 67, 112
popularity of, in Spain, 45,
62, 67-68, 151
theology and, 113
Arabising, 45
Arabs
and Berbers, 5, 20, 36-37,
43-44, 79
and Spanish Muslims, 29, 45
and Syrians, 25
as town-dwellers, 34
cultural contacts with Middle
East, 45
influence on literature, 67-68
Aragon
Fortun's submission to Islam,
10
union of Castile and, 129
Arberry, A. J., 60
architecture, see art
Arianism, 5, 45
'Arlb, 58, 68
Aristotle, 121
army
Arab expeditionary force, 2-4
Arabic words in modern
Spanish form, 42
control with head of state, 48
expansionist military policy
and, 76, 77
liability for military service,
14, 49
mercenaries in, 49, 73
organisation of, 10-11
position of generals in, 13
professional, 26
slaves recruited for, 48, 49,
73, 74, 76
stipends for service, 14
 INDEX
Army (cont'd)
Visigothic, 6, 8
volunteers for, 48
art
distinctive features of, 65-66
formative period of, 65
greatness of Moorish, xiii, 65¬
67, 151-52
in era of "party kings", 80,
125-26
influences in, 66—67
interest of Arabo-Andalusians
in, 85, 87
of Granada, 141-42
of the Almohads, 126
of the Almoravids, 125-26
of the Mudejar age, 131,
140-41
of the 11th and 12th centu-
ries, 125-26
of the 14th and 15th centu-
ries, 140-42
under the Umayyad Cali-
phate, 65-67
Visigothic, 66
asceticism, 83
Ash'axism, 89, 116-17, 122
Asin Palacios, Miguel, 124
Astorga, capture of, 10
astronomy, 57
Asturias
consolidation of Asturian-
Leonese kingdom, 31
expansion of the kingdom, 27
independence of, 20
Muslim penetration into, 10
reconquest of, 19
the Reconquista and, 147
atomism, 64, 108
Avempace, 118-19, 120
Averroes (Ibn-Rushd)
and Ibn-Tufayl, 122-23
as a jurist, 117
career of, 120-21
commentaries on Aristotle,
121, 150
defence of philosophy, 121¬
22
greatness of, 150
171
Avicenna, 110
al-Awza'I, 54
Radajoz, 79, 86
Badf-az-Zaman al-Hamadhani,
107, 108
Baghdad, as capital of the 'Ab-
basid Caliphate, 50-51
haladiyyun, 25
Balj, 22, 24
ballads, xiii
BaqI ibn-Makhlad, 56
Barcelona, 34, 72
belles-lettres
interest of Arabo-Andalusian
aristocracy in, 85, 87
of the Umayyad period, 65,
68
Berber(s)
and the Fatimid dynasty, 36
-Arab solidarity in al-
Andalus, 44
contributions to culture, 147
conversion to Islam, 4, 14
dissatisfaction with Arab rule,
19, 21, 25
distinction between Arabs
and, 5
Empires, see Almohads, A l -
moravids
nomadic, 36-37* 43, 82, 89
part in conquest of Spain, 5,
8-10, 36-37
recruitment of, in army, 49,
71, 74, 76
return to North Africa, 19
revolt in North Africa, 21-22,
24
revolt in north-west Spain,
21, 34-35
revolt of 901, 37
rulers and the "party kings",
73-74, 79
sedentary, 4, 25, 36-37, 43
settlement in al-Andalus, 15,
19, 25, 44
tensions between Arabs and,
20, 25-26
 INDEX

biography and Islamic political tradi-

in 10th century, 57, 58 tion, 35
in 11th and 12th centuries, custodianship of captured,
114, 116 32, 34
Bobastro, 30, 32 Castro, Americo
booty on Christian Spain, 147
as a factor in Arab expan- on Spanish belief in "tangi-
sion, 3, 4, 8, 18 ble, nearby power", 37
disposal of movable, 14 on the greatness of Islamic
Bordeaux, Muslim capture of, Spain, 150
16 theory on the cult of Saint
building programmes, 28, 40 James of Compostella, 31
Byzantine Empire Catholicism
conflict with Arabs in North conversion of Visigoths to, 5
Africa, 4 ecclesiastical hierarchy un-
influence of in Islamic art, 66 der, 5-6, 7
loss of Syria and Egypt, 1 ceramics, 42
Cercamon, 139
Cairouan chamberlain
Fatimid rule in, 36, 51 'Amirid, 70-74
foundation of, 4 duties of, 49
Ibn-Hafsun's negotiations Chanson de Roland, 27, 137
with, 30 Charlemagne, 27
seat of the governor of Ifri- Christian(s)
qiyya, 4, 51 al-Mansur's campaigns
caliphate against, 71-72
meaning of, 12 campaigns against Granada,
problem of succession to, 13 129
Spain as a province of the culture and the Mudejar,
Damascus, 12-23 131
see also 'Abbasid, Umayyad emigration of, 47
Carcassonne, Muslim capture exaction of tribute from Mus-
of, 16 lim states, 80
Castile fascination with Islamic cul-
al-Mansur's campaigns ture, 46-48, 61, 67-68
against, 72 feudalism and the, 76
attempt to assert independ- growth of intolerance, 131-
ence, 38-40
relations with Granada, 127- 33
28 in the era of the "party
revolt against Ramiro I I , 33 kings", 80-81
submission to 'Abd-ar- literature, 136
Rahman I I I , 34 mingling of Muslims and,
truce with Abu-Yusuf Ya'qub, 30-31
92 -Muslim struggle in terms of
union of Aragon and, 129 differing economies, 43
union of Leon and, 95-96 offensive in the 13th century,
casdes 95-96
of the Reconquista, 26
 INDEX
Christians ) (cont'd)
relations with muwallads, 35,
46
resettlement of, 33, 35, 86
rule, Muslims under, 130-33
selectivity in assimilation of
culture, 131
Spain and Europe, the stimu-
lation of, 147-49
submission to *Abd-ar-Rah-
man I I I , 33
successes against the Almo-
hads, 9^-93
tolerance towards Muslims,
130-31
treatment of Jews, 131
under the Almoravids, 87
Cid, El, xiii, 81, 86
cities of Visigothic Spain, 6
civil wars of 1008-31, 73-74
comes, 49
commerce
and the Christian-Muslim
struggle, 43
as the economy of southern
Spain, 43
attitude of Visigoths towards,
6-7
stimulation of, 42
words associated with, 42
conquest of Spain, question of
completeness of, 34-36
conversion to Islam
conditions of, 3, 14
converts as clients, 44
loss of revenue through, 14
motives for, 25-26
Muhammad's views on, 2
of local Spanish population,
25, 26
policy in 700-50, 14-15
protected groups, 3
reasons for among muwal-
lads, 45-46
social advantages of, 46
Umayyad policy on, 30-31
Cordova, city of
as capital of al-Andalus, 15,
32
173
during the era of "party
kings", 80, 81
Great Mosque of, 65-66, 67,
151
massacre at, 27
part played by populace in
civil wars, 73
surrender of, 9, 85, 96
corporations of artisans, 41
crops, introduction of new, 40
crusade and jihad, xiv, 149
crystal, 66
culture
Almoravids and, 86-87
Andalusian art and poetry,
64-65, 67
Christian fascination with
Islamic, 46-48, 61, 68
cultural achievements under
the Umayyads, 52-69
cultural greatness in political
decline, 97-126
dominance of Arabic over
Islamic, 50, 67-68, 69
Eastern influences in, 68-69
encouragement of scholar-
ship, 111
in the era of the "party
kings", 80
Isidorian, 46, 55, 146
Mudejar, 131
preference for Arabic to
Latin, 55, 61
problem of the acceptance of
Islamic, 46
sources of Moorish, 67-69
symbiosis of, 149
unity of Andalusian, 144-46
unity of Islamic, 105-6
Damascus caliphate, Spain as
the province of the, 12-23
Dante, 137
Dawud al-Isbahani, 112
day of the Foss, 26—27
de Cisneros, Cardinal Ximenez,
132
dervish order, 124
174 INDEX
dhimmt
conditions offered to, 3
defined, 3
exclusion of inhabitants of
the north-west, 34
garrisoning of, 3
Jews as, 13
status and feudalism, 35
Dozy, Reinhardt, 20, 86
economic
basis of the prosperity of the
Umayyad Caliphate, 40—43
differences and tribal rivalry,
21
distinction between Arabs
and Syrians, 25
position of Muslims in Chris-
tian Spain, 130-31
superiority of the Arabs, 44
Elvira, capture of, 10
Emirate, see Umayyad emirate
Eudo, Duke of Aquitaine, 16
Europe
appreciation of Islamic cul-
ture, xiii
contacts with Islamic Spain,
xiii
relations with Spain, xiv
stimulation of Christian
Spain and, 147-49
famine, 19, 22
al-Fat'h ibn-Khaqan, 106
Fatimid dynasty
'Abd-ar-Rahman's assump-
tion of titles and, 38
discontent raised by agents
of, 36-37
establishment in Tunisia and
Egypt, 36, 51
expansion eastwards, 37-38
religious ideas behind, 36
threat to al-Andalus, 32
Ferdinand I I of Aragon and
Isabella of Castile, 129,
132
Ferdinand I I I of Castile and
Leon, 95-96, 127
Fernandez, Garcia, 138
feudalism
and Islamic political tradi-
tions, 35, 50-51, 76
as a feature of Northern
Spain, 43
dhimmt status and, 34
effect on religion, 30-31, 35
Muslim acceptance of, 81
al-Fihri, Yusuf ibn-'Abd-ar-
Rahman, 15
fleet, creation of, 42
Fortun of Aragon, 10
Fountains of Story, The, 62
France
defeat of Muslims at Tours,
16, 18
invasion of the Rhone valley,
15, 16, 18
invasion of southern, 10
invasion of western, 15-16,
18-19
Islamic trade with, 42
Muslim raids into, 16, 21
significance of the battle of
Tours, 16, 18, 34
Visigothic province in, 5, 15
Franks, as mercenaries in Span-
ish army, 48, 49
furs, 42
Galicians
and the cult of Saint James
of Compostella, 31
revolt of, 19
genealogy
adoption of Arab, 45
interest in Arab, 45, 57-58,
144
works on, 111, 114
Gerona, capture of, 10
al-Ghafiqi, 'Abd-ar-Rahman, 16
Ghalib, General
death of, 71
expeditions in North Africa,
38
relations with al-Mansur, 70,
71
 INDEX
Ghalib, General (cont'd)
suppression of revolts in al-
Andalus, 40
al-Ghazall, 89, 117, 119, 122
Gibraltar
capture of, 129
origin of name, 8
gilds, 41
Goldhizer, Ignaz, 46
Gomez, Garcia, 104
governors of al-Andalus, 15
grammar, 57, 105, 106, 134
Granada, Nasrid kingdom of
art and architecture of,
128, 141-42
end of, 129
establishment of, 96, 127-28
Islamic character of, 128
reasons for survival of, 127-
28
see also Alhambra
Guadalquivir
massacre at, 27
settlement of Syrians in val-
ley of, 22
Guillaume IX, 138
haib, see chamberlain
al-Hakam I
dates of reign, 24
day of the Foss, 26-27
introduction of mercenaries,
49
recognition of Malikite rite,
55
al-Hakam I I
assumption of titles, 38
building programme, 65, 66,
67
dates of reign, 32
delegation of powers, 48
encouragement of cultural
activities, 57, 68
establishment of caliphal l i -
brary, 68, 71, 145
influence of jurists under, 50
interest in scholarship, 68,
111, 145
175
North African policy of, 37-
38
political events of reign, 38,
40
Hammudid caliphs, 70, 73, 79
Hanafite rite, 54, 55
Hanbalite rite, 54, 124
al-Hariri, 108
Hellenism, 55, 144
Hisham I , 24
Hisham I I
abdication and restoration, 73
dates of reign, 70
succession of, 70
Hispano-Roman nobility, 5, 6
historians
interest of, in Spain, xiii-xiv
of 10th-century Spain, 57-58
of later centuries, 117-18
house-building, words asso-
ciated with, 42
Huesca, 46
al-Humaydl, 115
Ibn-'Abbad, 125
Ibn-'Abd-al-Barr, 114-15
Ibn-'Abd-Rabbih, 62
Ibn-Abi-'Amir, see al-Mansur
Ibn-Abi-Randaqa, 106
Ibn-al-lrrif, 124
Ibn-al-Jilliqi, 29, 33
Ibn-al-Kinanl, 109
Ibn-al-Muqaffa', 108
Ibn-al-Mu'tazz, 104
Ibn-'Ammar, 97
Ibn-Bajja, see Avempace
Ibn-Barrajan, 124
Ibn-Bassam, 106
Ibn-Gabirol, 136
Ibn-Hafsun
Christian support of, 47
relations with Fatimids, 36—
37
revolt of, 30, 32
Ibn-Hamdis, 97
Ibn-Hanf, 63-64
Ibn-Hayyan, 117
176 INDEX
Ibn-Hazm
adherence to Shafi'ite and
Zahirite rites, 56, 111-14
book on sects, 113
claim of arabised Persian de-
scent, 45, 111
concern for language, 112,
113
death of, 113
early career of, 111-12
interest in genealogy, 111
poetry of, 112
range of scholarship, 113-14,
150
studies in theology, 112-13,
US
The Ring of the Dove, 97,
100, 101, 106, 112
Ibn-Khafaja, 98, 99, 100, 102
Ibn-Khaldun, 118, 150
Ibn-Malik, 134
Ibn-Mardanish, 91
Ibn-Masarra
influence of Greek philosophy
on, 57, 118
mysticism of, 123-24
Ibn-Qutayba, 62, 68
Ibn-Quzman, 105
Ibn-Rushd, see Averroes
Ibn-Sab'in, 123
Ibn-Sa'Id al-Maghribl, 134
Ibn-Sana al-Mulk, 103
Ibn-Slda, 106
Ibn-Tufayl
Averroes and, 122-23
greatness of, 150
narrative writing of, 110
philosophy of, 119-20
Ibn-Tumart, 89-90, 93-94,
116, 119, 120
Ibn-Yasln, 82
Ibn-Zamrak, 135
Ibn-Zaydun, 97, 107
Ifriqiyya, 15
ijma', 53
industries, 42-43
inns, 41
Inquisition, 131-32
intellectual life under the
Umayyads, 52-65
irrigation, 40, 42
Isa ibn-Dlnar, 55
Isidore, 46, 55, 146
Islam
and politics, 75-76
and the tribe, 75
dominance of Arabic element
over, 50, 143-45
feudalism and, 76
mingling of Christianity and,
30-31
new awareness of, under the
Almoravids, 87, 126
relations with Islamic Spain,
xiv, 50, 56, 67, 68-69,
143-47
Islamic Spain
course of the invasion, 8-11
in North Africa, 118, 133,
147
interest of, xiii-xiv
internal tensions of, 20-23
intrinsic greatness of, xiii,
150-52
military and political power,
10-11, 16, 18
relations with Europe, xiv,
147-49
relations with the Islamic
world, xiv, 50, 56, 67, 68-
69, 143-47
significance of, 143-52
the last of, 127-42
islamisation
beginnings of, 31
failure to adapt ideas to con-
temporary problems, 75-77
increasing influence of, 50
jurists and, 31, 49-50
political traditions and feu-
dalism, 35-36, 50-51
Isma'ilite form of Shi'ism, 36
ivory, 66
Ja'far al-Mus'hafl, 70
al-Jahiz, 57, 68, 107
Jahwar, 80
 INDEX
Jahwarid dynasty, 80
jarifo, 42
Jaw'har, 37
Jews
as dhimmis, 13
assistance in Muslim inva-
sion, 7, 9, 26
Christian treatment of, 131
in Islamic Spain, 26
in Nasrid Granada, 128
in 10th century, 47-48
literature of, 136
under the Almoravids, 87
Visigothic persecution of, 7,
26
jihad
'Abd-ar-Rahman Ill's use of
idea of, 35
and the crusade, xiv, 149
and the razzia, 2-3
as a political instrument, 2-3,
76, 77, 145
unsuitability of concept of,
35
volunteers for, 49
jizya, see poll-tax
John of Capua, 137
Judah ben Solomon al-Harizi,
136
Julian, Count
legend of, 8
part in the Arab conquest of
Spain, 5, 8
jundis, 22, 23, 49, 71
jurisprudence, 52, 53
jurists
al-Mansur's policy towards,
71
Almohads and Malikite, 94,
117
and islamisation, 31, 49-50
and the Cordova massacre,
27
bifurcation between ruling
class and, 53
contact with East, 68
dual aspect of activities, 53
function in the caliphate, 49
177
increasing influence of, 40,
49-50
in Nasrid Granada, 129
in society, 77
interpretation of Sharf a, 49-
50, 52-53
Malikite and Almoravid state,
81, 82, 83, 84, 85, 86, 87,
93, 94, 116-17, 119, 120
position in 10th-century
Spain, 50
rise of a class of, 54, 55-56
theoretical concern of, 52-53
kaldm, 57
Kalbites
effect of rivalries in al-Anda-
lus, 20, 21, 22, 25
effect of rivalries in caliphate,
20, 21
khalifa, see caliphate
Kharijite doctrine, 44
al-Khushani, 58
km a, 49
kuwar, 49
land
disposal of conquered, 3, 13-
14
gifted to Muslims, 14, 44
-tenure of the Syrians, 22, 25
Las Navas de Tolosa, 93
law
administration of, 49, 54
as the focal point of intellec-
tual activity, 52-53
codification of, 54-55
dominance of the Malikite
rite, 45, 50, 144
Islamic conception of, 52
legal rites, 53-54, 55-57
roots of, 53, 112-13, 117
legends, xiii
Leon
'Abd-ar-Rahman Ill's rela-
tions with, 32, 33-34
al-Mansur's expeditions
against, 72
178 INDEX

Leon (cont'd) form favoured in Spain, 55,

attempt to assert independ- 68
ence, 39-40 jurists and the Almohads, 94,
capture of, 10 122
consolidation of kingdoms of reasons for adoption of, 55-
Asturias and, 31 56
defeat of Berbers by, 37 recognition of, 54, 55, 56
truce with Abu-Yusuf- Malikism, 112, 115-16, 119
Ya'qub, 92 al-Mamun, 70, 73
union with Castile, 95 al-Mansur
L6vi-Provengal, Evariste, 20, building programme of, 65
42, 69, 117 career of, 70
lexicography, 57, 105, 106 dates of chamberlainship of,
linguistic 70
groups, 95 military activities, 72, 74
studies, 105-6 policy of, 70-72
Lisan ad-Din ibn-al-Khatib, question of completeness of
134-35 conquest, 34
literature maqdma, 108, 134, 136, 137
Andalusian and Romance, al-Maqqari, 136
137-40 marble, 66
greatness of Andalusian, xiii, Marcabru, 139
150-51 Marches
influence of Spanish, 135 'Abd-ar-Rahman Ill's policy,
narrative, 137 3^-33
of Christians, 136 administration of, 19, 20
of era of the "party kings", as semi-independent states,
80 29-30
of Spanish Jews, 136 defence of, 28, 29, 32-33, 37>
of Umayyad period, 41, 58- 49
65 in the era of the "party
philology and prose, 105-10 kings", 78-79
al-Maarrl, Abii-'l-'Ala, 108-9, revolt of Berbers in, 37
110 tribute paid by rulers of, 80
MacGuckin de Slane, Baron, Marrakesh, 83, 91
102 Martel, Charles, 16, 18
al-Madina az-Zahira, 71 Masarrite movement, 123-24
Madinat az-Zahra, 49, 66, 71 Masmuda, 89
Maimonides, 136 Massignon, Louis, 64
materialism, 75, 77
Malaga, 10
mathematics, 57
Malik ibn-Anas, 54
mawdli, 14
Malikite rite
mawaliya, 104
Almoravid state and, 81, 82,
mawla, 14
83, 84, 85, 86, 87, 93, 94,
medicine, 43, 57
116-17, 119, 120 Merida, 28, 29
codification of, 54-55 metal-work, 41, 66
development of study of, 57 middle class, 75, 77-78
dominance of, 45, 5°> 44
1

minerals, 41
 INDEX
monarchy in Visigothic Spain,
5-6, 7
Moriscos, 132-33, 135
mosaics, 66
Mozarabs, 26, 45, 67, 102, 128,
146, 148
Mudejars, 130-31, 140-41* 148
Muhammad, 1, 2
Muhammad I , 28, 56
Muhammad an-Nasir, 91, 93
Muhammad ibn-*Abbad, 80
Muhammad ibn-Ma^anud, 68
Muhammad ibn-Yusuf ibn-
Nasr, 96, 127
Muhyi-d-Din ibn-al-'Arabi, 123,
124, 151
al-Mu'izz, 37
muluk at-tawaif, see "party
kings"
al-Mundhir, 28
al-Mundhir ibn-Sa'ld al-Balluti,
56
Muqaddam ibn-Mu'afa, 65
Murcia, 10, 127
Musa ibn-Musa, ibn-al-Qasf,
29, 30
Musa ibn-Nusayr, 4, 9-10
musdlim, 25
music, xiii, 41, 62
Muslims under Christian rule,
130-33
musta'ribun, 26
al-Mutadid, 80, 97
al-Mutamid, 80, 81, 84, 85, 97,
101
al-Mutanabbi, 60
Mu tazilism, 57, 71
al-MuwaWdun, 89
muwalladiln
acceptance of Islam, 45-46
adoption of Islamic culture,
46, 102
belief in "tangible nearby
power", 37
defined, 25
fusion of Arabs and, 45
language spoken by, 47
relations with Christian no-
bles, 47
42 1
179
semi-independent status of,
29-30
muwashshah, 65, 103-4, 134>
135-36
al-Muzaffar, 70, 72, 73
mysticism
foundations of Andalusian,
57
of Almoravid movement, 83
of 11th and 12th centuries,
123-25
Narbonne, 10, 16, 18, 28
an-Nasir li-din-Allah, 38
Nasrid, kingdom of Granada
art of, 128, 141-42
culture of, 134
end of, 129
foundation of, 96, 127-28
Islamic character of, 128
reasons for survival of, 127-
28
Navarre, 30, 33, 34, 40
negroes, 48, 49
Nimes, 16
nomadism
and Arab unity, 2, 36
and the Umayyad caliphate,
12-13
literature and, 59
rise ©f empires and, 83-84
Norse invaders, 28, 42
North Africa
Abd-ar-Rahman Ill's policy
towards, 36-38
Arab expansion across, 3-5
art of, 125-26
Berber revolt in, 21-22, 36
culture of Islamic Spain con-
tinued in, 118, 133, 147
extension of 'Amirid power
in, 72, 76
heretical forms of Islam in,
44
relations with Islamic Spain,
xiv
Shilte practices in, 78
trade with, 42
i8o INDEX
Pamplona, 10
particularism, 74
"party kings"
action of Christian princes,
80-81
art under, 125
ethnic groups representing,
79
events of era of, 73-74, 78-
81
poetry under, 97, 98
second period of, 88
spheres of influence of, 79-
80
Pedro Alfonso, 136
Pelayo, Prince, 19
Peres, Henri, 100-1, 105, 110
Persia, 44, 51, 75-76
philology, 68, 105-6
philosophy
and religion, 122-23
and theology, 118-21
Hellenistic, 57, 149
in the 11th and 12th cen-
turies, 118-23
value of Andalusian, xiii-xiv,
149
Pidal, Menendez, 139
poetry
and the common people, 105
Arab elements in, 58-59, 60,
67
as court poetry, 60-61, 68,
97-98
belles-lettres and, 58-65
continuity of Arab, 61
de-animation in, 100
distinctiveness of Andalusian,
64, 100-1
early Andalusian, 61-64
Eastern influences on, 61-63,
68
golden age of Andalusian,
97-98
imagery of, 63-64, 102
influence of Arabic, 138, 146
influence on Romance litera-
ture, 138-39
interest of Arabo-Andalusian
aristocracy in, 85, 87
in the 11th and 12th cen-
turies, 97-105
Islamic and Europe, xiii,
100-1, 102
Islamic elements in Andalu-
sian, 101-2
love, 99-100
new forms of, 65
nomadism and, 59
pre-Islamic traditions in, 58-
59
refinement of, 99
rhetorical artifice in, 59, 62,
101
strophic forms of, 103-5,
138, 139
stylisation of, 98
themes of, 58-60, 98-100
under the Berber dynasties,
98
under the "party kings", 97-
98
under the Umayyad cali-
phate, 58-65
poll-tax, 49, 131
Portugal, 19
prosperity
materialism resulting from,
74-75
of reign of 'Ahd-ar-Rahman
I I I , 38, 40-43, 75
"protected groups", see dhim-
mls
provinces, 49
Qadi lyad, 115-17
al-Qali, Abu-All, 62, 68, 105
Qaysites
and tensions in al-Andalus,
20-23, 25
and tensions in the Damascus
caliphate, 20, 21
qiydSy 114
Quran, 93~94, 112-13
Qu'ranic exegesis, 57, 68
Quraysh, 12
 INDEX
Rabbi Eleazar ben Jacob, 136
Rabbi Joel, 136, 137
racial particularism, 74
raiding, 2-3
Ramiro I I of Leon, 33~34> 37
rmhidun, 12
razzia, 2
Reconquista
and the age of tolerance,
130-31
and the kingdom of Granada,
127
beginnings of, 19-20
changes in the area of al-
Andalus, 12
cult of Saint James of Com-
postella and, 31, 72, 148
effect of the fall of Toledo, 80
motives behind, 19-20, 147-
48
progress of, 95-96
successes of, 92-93
terms used by Christians of,
25-26
religion
Almoravid policy on, 85
and philosophy, 122-23
and the unity of Spain, 50-
5i» 143
combination of politics and,
83-84
movements of the 10th cen-
tury, 43-48
reasons for acceptance of Is-
lamic, 45-46
religious sciences
Berber contributions to, 43
in 11th and 12th centuries,
110-17
living tradition of scholarship
in, 68-69
other than law, 57-58
study of, 105
under the Umayyad Cali-
phate, 52-58
value of writings on, xiii-xiv
ribdt, 83
Roderick, King, 7, 8, 9
Rodrigo Diaz de Vivar, see Cid
l8l
Roger I I , of Sicily, 91
Roland, ballads of, xiii
romance dialect, 47
ruling institution
and legal rites, 56-57
bifurcation between jurists
and, 52-53
description of, 48-51
the upper classes and, 77-78
Sa'id, qadi of Toledo, 117
Saint James of Compostella
cult of, 31, 72, 148
destruction of shrine of, 72
saint-worship, 116, 146
Salamanca, 33
Samh, 16
Sanhaja, 82, 83, 89
Saqaliba, 73, 74, 76, 79
Saragossa
Aljaf6ria of, 125
and the defence of the
Marches, 28, 29, 33, 79
capture of, 10
Charlemagne's attack on, 27
recapture of, 85-86
Sa'sa'a, 54
science, xiii-xiv, 43
Senegal, 82
Septimania, 5
Seville
Alcazar of, 140
as capital of al-Andalus, 15
capture of, 9, 85, 91, 96, 127
Giralda of, 126
kingdom of, 79, 80, 81
semi-independent status of,
29, 32
ash-Shadhili, 124
Shadhiliyya, 124
ShafTite rite, 54, 55, 56, 68,
112, 115, 117
slwmiyyun, 25
ash-Shaqundl, 106
Shfism, 36, 44, 78, 93
Shu ubite movement, 46
slaves, 47-48, 49
Slavs, 48, 49, 73, 74
i8n INDEX
social structure
advantages of conversion to
Islam and, 45-46
Arab and Syrian, 25
of Visigothic Spain, 6-7
religious movements and, 43-
48
the breakdown of Islamic
Spain and, 77-78
tribal rivalry and, 75
sociology, 118
Spain
conquest of, 1-11
course of the invasion, 8-11
making of modern, xiii-xiv
weakness of Visigothic, 5-7
Subh, princess, 70
sunna, 52, 53
Sunnites, 53-54
Syria, 3, 21, 66-67
Syrians, 22-24, 25
tafsir, 57
taqiyya, 133
taqlid, 115
Tariff, 8
Tariq ibn-Ziyad, 8-10
Tarraconensis, 5, 7
Tarragona, 10
Tashufm ibn-'All, 84
taw'hid, 89
tdtvtl, 120
taxes, 3, 14
Terrasse, Henri, 67, 140
textiles, 42
Theodemir, 10, 13
theology
Ash'arite, 119-20
Ibn-Hazm's works on, 112-13
Qadi Tvad's work on, 116
S hilosophy and, 118-20
sdo
and the defence of the
Marches, 28, 33, 79
art of churches in, 140
day of the Foss, 26-27
fall of, 80, 81, 84, 86
surrender of, 9, 33
Visigothic capital at, 9, 15
Thomism, 121, 149
Toulouse, 16
Tours, significance of the bat-
tle of, 16, 18
towns
and the arts, 41
Arabs as dwellers in, 34
Christians as dwellers in, 47
citizenship, 41
commerce and, 42
Islam as a religion of, 41
municipal government of An-
dalusian, 41
urbanisation of Muslim, 41
Toynbee, Arnold, 34
Traditions, The
and the Sharfa, 53
study of, in Islamic Spain, 56,
57, 93-94, 112-16
tribal rivalry in the Umayyad
caliphate, 20-21
tribute
paid by Christian states, 34
paid by dhimmis, 3, 13
paid to Christian states, 80,
131
Tuareg, 82
Tudela, 29
Tudmir of Murcia, 10, 13
Tujfbids, 29, 33
Tunisia, conquest of, 4
'Ubayd-Allah, 36
'ulum, 57
'Umar, caliph, 1
Umayyad
caliphate at Damascus: and
the Quraysh, 12; discon-
tent of non-Arab Muslims
and the fall of, 15; effects
of tribal rivalries on, 20,
21; jurists under, 52-53;
Muslim expansion and the
break-up of, 18-19, 20; no-
madic ideal, 12; Persian in-
fluence in, 75; supersession
by 'Abbasids, 22-23, 24
caliphate of al-Andalus: abo-
lition of, 73; autocratic ba-
 INDEX
Umayyad (cont'd)
sis of, 48; cultural achieve-
ments under, 52-69; domi-
nance of Arab elements i n ,
50-51; establishment of,
38, 48; government of, 48-
51; grandeur of, 32-51;
members ruling i n 1009-
31, 73; organisation of, 12-
15; powers of the caliph,
48
emirate of al-Andalus:
achievement of unity i n ,
26; crisis of, 28-31; estab-
lishment of, 23, 24-28;
islamisation i n , 31, 144;
novelty of, 24-25; racial
problems i n , 25-26; revolts
and risings i n , 26-27, 28-
29; the independent, 24-31
Spain at its zenith, 32-40
upper classes
i n Islamic Spain, 77-78, 85
i n Visigothic Spain, 5, 6, 7,
45-46
urbanisation, 41
a l - U t b i , 55
Valencia, 86
Visigothic
and Islamic culture, 61
art, 66
Spain: conquest, 5; decline
of towns i n , 41; ecclesias-
tical hierarchy i n , 5-6, 7,
26, 45-46, 51; end of, 9.
10, 51; persecution of Jews
183
in, 7, 26; privileges of up-
per classes, 5, 6, 7, 45-46;
problem of succession to
the kingship, 6, 7; religion
i n , 5-6; resistance of no-
bles, 19; trade w i t h North
Africa, 42; unreliability of
army, 6, 7
vizier, 49
wdli, 49
Washington Irving, xiii
wazir, 49
Witiza, 7
words, from Arabic i n modern
Spanish, 42-43
Yahya ibn-Ibrahfm, 82
Yahya ibn-'Umar, 83
Yahya ibn-Ya&ya al-Laythi, 55
Yemenite party, 22, 23, 25
Yusuf ibn-ash-Sbaykh, 106
Yusuf ibn-Tashufin
campaign against Alfonso V I ,
81, 84
command i n North Africa, 83
conquest of Spain, 84-87
dates of reign, 84
decision to unify Spain, 85
poetry under, 98
zdhir, 112, 120
Zahirfem, 112-13, 115, 117, 120
Zahirite rite, 53, 56, 94, 112-13
za\al 103, 105
7
Zirid dynasty, 79
Ziryab, 62, 68, 145