Islamic History

ISLAMIC HISTORY
A NEW INTERPRETATION
2
A.D. 75O-IO55 (A.H. I32-448)
Downloaded from https://www.cambridge.org/core. Cambridge University Main, on 26 Apr 2018 at 06:08:12, subject to the
Cambridge Core terms of use, available at https://www.cambridge.org/core/terms.
https://doi.org/10.1017/CBO9780511622380
To my friend
Shaykh Ahmad ZakI Yamdni
a modern wa^tr whose statesmanship brings to life
many of the qualities of his great predecessors
https://doi.org/10.1017/CBO9780511622380
ISLAMIC HISTORY
A NEW INTERPRETATION
2
A.D. 75O-IO55 (A.H. I32-448)
M. A. SHABAN
Head of Dtpartmtnt of Arabic and Islamic Stadia
Vnivtrsily of Extttr
I CAMBRIDGE
UNIVERSITY PRESS
https://doi.org/10.1017/CBO9780511622380
CAMBRIDGE u n i v e r s i t y p r e s s
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
São Paulo, Delhi, Dubai, Tokyo, Mexico City
Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by

Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521294539
© Cambridge University Press 1976
This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 1976

First paperback edition 1978
Reprinted 1981, 1986, 1988, 1990, 1992, 1995
A catalogue recordfor this publication is available from the British Library
Library of Congress Cataloguing in Publication Data
Shaban, M. A.
Islamic history A.D. 750-1055 (A.H. 132-448)
Bibliography: p. 212
Includes index.
1. Islamic Empire — History— 750-1258.1. Title.
DS38.6.S48 909'.09’7671 75-39390
ISBN 978-0-521-21198-7 Hardback
ISBN 978-0-521-29453-9 Paperback
Cambridge University Press has no responsibility for the persistence or

accuracy of URLs for external or third-party internet websites referred to in
this publication, and does not guarantee that any content on such websites is,
or will remain, accurate or appropriate. Information regarding prices, travel
timetables, and other factual information given in this work is correct at
the time offirstprinting but Cambridge University Press does not guarantee
the accuracy of such information thereafter.
https://doi.org/10.1017/CBO9780511622380
CONTENTS
Acknowledgments pagevi
Preface vii
i The founding of the 'Abbasid regime i
2 Towards a civil war 2O
3 Dissolution under a new regime 4i
4 The origins of the iqta* 7i
5 Regional economic conflicts 89
6 The breakdown of the central government (I) "5
7 The breakdown of the central government (II) *37
8 The Buyid confederacy
9 The Fltimids 188
Works cited 212
Index 216
MAPS
1 Iraq, Jazira, Syria and tbugbSr 48
2 Trans-Saharan caravan routes 104
3 Iran and the East 176
Downloaded from https://www.cambridge.org/core. UCL, Institute of Education, on 26 Apr 2018 at 06:12:06, subject to the
https://doi.org/10.1017/CBO9780511622380
ACKNOWLEDGEMENTS
I am very grateful to my good friend and colleague Dr Richard
Hitchcock who generously gave time and effort to read every chapter
as I completed it. His sympathetic criticism and many useful sug-
gestions were of significant help to the final draft of this book.
Once more I am greatly indebted to Professor C. F. Beckingham
for reading the typescript with his usual thoroughness and for his
scholarly advice and invaluable suggestions. Professor T. M.
Johnstone took the great trouble of reading both the typescript and
the proofs with meticulous care, correcting many mistakes and
making many improvements for which I am most grateful. My
thanks are also due to Mr Hugh Kennedy for undertaking the
difficult task of compiling an accurate index for this difficult work.
My special thanks go to Lady Bruce Lockhart for typing the
manuscript with skill and efficiency. She soon discovered that my
spelling left a great deal to be desired and patiently corrected my
unpredictable mistakes. I am also grateful to the editorial staff
and the printers of the Cambridge University Press for the consider-
able patience they have shown and the great care they have taken in
the production of this book.
Thanks are due to the Government of the United Arab Emirates,
in particular to H.E. Ahmad al-Suwaidi, H.E. Muhammad Habrush,
H.E. 'Abdulmajid al-Qaysi and H.E. Dr 'Izz al-Din Ibrahim, who
with a generous grant made it possible for the Exeter University
Library to acquire most of the sources I needed for my research.
Finally, I must thank my wife and son for their understanding and
tolerance of my idiosyncrasies during a period when I was working
under great pressure to write this book.
[vi]
https://doi.org/10.1017/CBO9780511622380
PREFACE
This work is a continuation of my attempt to present a new interpre-
tation of Islamic history. It deals with the three centuries between
the rise of the 'Abbasids and the Saljuq invasion. This period
witnessed the establishment of a new regime, its failure to live up to
its revolutionary ideals and the gradual dissolution of a vast empire
into lesser political entities. The 'Abbasids failed to establish a
political structure supported by viable institutions to rule their
domains. They also failed to take measures to accomplish the
economic integration of their empire because they could not com-
prehend nor were they able to cope with the rapid expansion of
trade. This problem was the more baffling to them for it demanded
fundamental changes in the structure of the basically agricultural
economy they had inherited to allow for an economic structure in
which trade could be exploited to the advantage of all concerned.
New taxation systems were required to divide the burden of taxes
equitably between the prosperous inhabitants of the urban trade
centres and the long-suffering agricultural communities. The
entrenched vested interests in the cities took advantage of their
political power to oppose such measures and the rural communities
had no alternative but to revolt in the face of this continued
injustice.
The impact of the vast increase in the volume of trade went far
beyond the purely economic front, for it was not only internal trade
that was involved but also aa international trade of unprecedented
dimensions. The wealth of the empire attracted trade in all sorts of
goods, especially luxuries, from all parts of the world, and every
region of the empire became interested in being an intermediary for
importing such trade as well as for exporting the industrial output
of the empire. Each region devised its economic policies and its
attitude towards its neighbours within and outside the empire in
accordance with its own interests. This economic cacophony was
another major cause for the eventual political disintegration of the
' Abbasid regime because without a harmonizing influence from the
centre, the competition of the regions for a bigger share in trade soon
developed into uncontrollable inter-regional strife.
Lacking the political and economic organization to maintain its
integrity the regime resorted to military power to control its
domains. The recruitment of military forces to defend the empire
MO
https://doi.org/10.1017/CBO9780511622380.001
Vlll PREFACE
against external enemies was in itself a continual problem for
successive 'Abbasid rulers. Now the need for and the use of the
military to control the civilian population became the cause of
further major crises for the 'Abbasids. More men were needed to
subdue more revolts and the military leaders came to occupy the
centre of the political stage in the provinces as well as in the capital.
In some provinces military leaders were able to establish their
own rule in place of 'Abbasid misgovernment. But in the outer
provinces neither the 'Abbasids nor their military representatives
were capable of maintaining their rule in the face of the aggressive
opposition of the local populations under their own leaders. The
Fatimids were the exception because they established themselves in
Tunisia and then in Egypt, with the support of a Berber minority, to
replace the crumbling rule of dynasties of military leaders in both
provinces. However the revolutionary idealism of the Fatimids and
their detailed planning failed to take into consideration the vital
interests of the majority of their subjects whose support was
indispensable for the survival of their regime in Egypt.
There has been a conspicuous lack of a comprehensive study of the

history of the period under discussion. The existing studies deal
only with limited aspects of this eventful era. Some writers have
satisfied themselves with rehashing the stale conclusions reached in
the nineteenth century. More earnest scholars, who have applied
themselves to serious research on subjects such as the Ghaznavids,
the BQyids, the Fatimids or the 'Abbasid waîrs, have laboured in
vain. In most cases they have treated their subjects in isolation from
other issues of the time and therefore have failed to examine and
explain the interaction of these complex issues. Such studies have
reached conclusions much the same as those of the proverbial four
blind men who felt an elephant's leg, tail, ear and body respectively
and concluded that it was like a log, a rope, a fan and something
without beginning or end.
In this study an effort has been made to maintain a constant
surveillance of the main events in the Islamic domains and to
explain them in relation to each other, without being side-tracked by
insignificant occurrences, in order to present a clear analysis of the
history of the period as a whole. A deliberate attempt has also been
made to keep to a minimum the infinite number of names of per-
sonages involved, for to have put down the names and the titles of
all the BQyids, Hamdanids and 'Uqaylids would have served no
https://doi.org/10.1017/CBO9780511622380.001
PREFACE IX
purpose, and the close similarity of these names would only have
confused the reader. As in my previous works considerable care has
been taken to ascertain the exact meaning and to define the precise
use of important terms used in the sources. The meaning of such
terms varied from place to place and often enough changed over the
centuries; and it is of vital importance to examine systematically
these changes in order to obtain proper understanding of the source
material.
Finally, I have left out, as much as possible, the discussion of de-
velopments in North Africa and Spain because from the beginning
the 'Abbasids themselves did not concern themselves much with
these regions. Furthermore, the history of North Africa and Muslim
Spain warrant a separate study which I am now preparing in col-
laboration with Dr Richard Hitchcock. I hope it will not be long
before its appearance.
Exeter
December, i97j 0)
https://doi.org/10.1017/CBO9780511622380.001
https://doi.org/10.1017/CBO9780511622380.001
THE FOUNDING OF THE
'ABBASID REGIME
Abu al-'Abbas served as the first 'Abbasid Amir al-Mu'mimn for a
little over four years 749-54/132-6; a period that seems to have been
only a prelude to the coming to power of his older and stronger
brother Abu Ja'far. Indeed the figure of the latter loomed large
behind that of the man who was supposed to be actually in power.
Although other members of the 'Abbasid family were engaged in
various important functions, Abu Ja'far seems always to have been
at the centre of things at the crucial moment. This is not to say that
he was running the affairs of the empire in his brother's name, for
the latter did not have this power. From the moment of his elevation
to this high office the real authority remained in the hands of the
military leaders of the Revolution. In fact, his selection as the new
Amir al-Mu'minin and the imposition of this choice on the political
leadership was thefirstmanifestation of the usurpation of power by
the military. Abu Salama, the vtetfr al Muhammad and virtually the
head of the provisional government, had not been challenged when
he assumed the direction of affairs in Kufa after the arrival of the
revolutionary army there. Nevertheless, his vacillation and wavering
for many weeks on the important question of installing a new Amir
al-Mu'mimn was a curious political mistake. It is true that he was
endeavouring to find al-ridd mitt dl Muhammad, a member of the
family of the Prophet acceptable to all, who would command the
widest possible support throughout the empire. It is also true that
the implementing of this important ideal of the Revolution was of
vital importance if rule by consent was to be restored to the
troubled empire. However, the military leaders, assured by their
success, had no patience for the deliberations ofthe vastly experienced
Aba Salama. Their political experience was limited to their pre-
revolutionary struggle in Khurasan, but now that they were
suddenly thrown into the whirlwind of the politics of empire, they
decided that the best course to ensure the continuous success of
their revolution was to take matters into their own hands and hold
on to their newly acquired power.
W
https://doi.org/10.1017/CBO9780511622380.002
2 THE FOUNDING OF THE ABBASID REGIME
Of course they knew that the general Shi'ite opinion was in
favour of an Imam]Amir al-Mu'minin, an all powerful leader who
could bring justice to all. However, they also realized that by
installing such a leader they would have to give up their own power.
As they were not ready to accept this, they opted for an Amir al-
Mtfminin who would not be an Imam. Overruling Abu Salama they
made sure that the powers of the wasgr were not transferred to the
Amir al-Mu'minin. The latter would be only a figurehead with
minimum religious functions if any at all. The 'Abbasids, the
members of the House of the Prophet who had the weakest claims,
were ready to accept the office on any terms. The military leaders did
not find it difficult to secure the choice of Abu al-'Abbas, the
weakest member of the family. As he was called 'Abdullah, which
was the same as the name of his stronger brother Abu Ja'far, it was
made abundantly clear that their choice was Ibn al-Harithiyya - the
son of the Harithite woman; make no mistake 1 Other stronger
members of the family had no alternative but to accept and respect
this choice. Meanwhile, from the moment of his accession, when an
uncle had to finish his inaugural speech for him, until his death,
Abu al-'Abbas had the whole-hearted support of all the members of
his family who were also not oblivious of their self-interest.
Although the new office of wa^tr was a corner-stone in the
structure of government as planned by the political leadership of the
Revolution, the realities of the new situation made it redundant.
Neither the military leaders nor the impotent Amir al-Mu'minin had
any interest in continuing an institution that would have imposed
limitations on their power. Moreover, Abu Salama, by his behaviour
and his failure to achieve what he had set out to do, had made his
position intolerable. It was an easy matter for Abu al-'Abbas,
supported by his brother Abu Ja'far, to obtain the acquiescence of
the military leaders and get rid of the hapless waîr. However, while
the decision to nominate Abu al-'Abbas was taken on the spot by
Abu al-Jahm - the political commissar of the army and representa-
tive of Abu Muslim - in consultation with the other military leaders
present in Kufa, it is instructive to notice that the question of
eliminating the wasgr had to be referred to Abu Muslim, the Amir
al Muhammad, in Khurasan. The latter readily concurred and even
sent his own men to carry out Abu Salama's execution. At the same
time Abu Muslim decided, on his own, that Sulayman b. Kathir
al-Khuza'i, the elder statesman of the Revolution in Merv, should
also be eliminated. Among the original organizers of the Revolution
https://doi.org/10.1017/CBO9780511622380.002
THE FOUNDING OF THE ABBASID REGIME 3
it is clear that there was approval of Abu Salama's opposition to the
'Abbasids. But from Abu Muslim's point of view such latent
opposition to the newly established regime could only represent a
serious threat to the whole enterprise and therefore had to be dealt
with severely and decisively.
It should be noted that these developments took place in a relatively
short period and were initiated by men who were closely involved in
an all-consuming violent revolution. They still had to defeat their
main adversary in order to establish their authority over the greater
part of the empire. Abu Muslim, the organizer of the revolutionary
army, had to stay behind in Khurasan to safeguard the only secure
base of this army with their homes and families. It is probably
because of this factor that he was able to maintain his hold over the
Khurasaniyya in far away 'Iraq. It is not surprising that Abu
al-Jahm, Abu Muslim's representative and political commissar of
the army in Kufa, took over the responsibilities of the was(tr without
seeing the need of assuming the title. He thereby combined control
of the army with control of the administration. Under the circum-
stances there was manifest co-ordination between the headquarters
in Khurasan and the military leadership in 'Iraq. The latter also had
the complete co-operation of Abu al-'Abbas and all his brothers,
uncles and cousins.'
Such harmony continued for only a few months until the major
forces of the Marwanids were destroyed; cracks then began to ap-
pear in this united front. Some of the Khurasaniyya became anxious
to return to their homes.2 They might have considered themselves
on an extended expedition that had lasted over three years, but they
could not have contemplated permanent settlement in 'Iraq. Their
leaders did not seem to have had clear plans for them and even if
they did, events were moving too fast for any such plans to take
effect. The setting up of new camps at Hashimiyya and then Anbar
are not to be taken as attempts at establishing new capitals, but
rather as temporary measures to house the army as circumstances
required. Naturally Abu al-'Abbas would want some of the troops
to stay in 'Iraq, at least for the time being, in order to bolster his
regime, but he could not prevent those who wanted to return
from doing so. Indeed he could have no serious objection to reducing
1
For details of this phase see M. A. Shaban, Tb$' AbbSsid Revolution, Cambridge, 1970
pp. 138-68.
* al-Tabari, Muh. b. Jarir, TmkUal-Rjuul ma ol-AbdOk, ed. M. J. de Goeje et aL, Leiden
1879-1901, HI, p. 75.
https://doi.org/10.1017/CBO9780511622380.002
the size of an army that deprived him of any real power. In the event
it was decided to raise the stipends of the Khurasaniyya to eighty
dirhams a month and to allow some of them to return home. This
last measure resulted in two problems, however, one in 'Iraq and
the other in Khurasan.
In 'Iraq, and in Syria for that matter, there were many remnants of
the defeated Marwanid forces. While some of them had taken
advantage of the situation to settle down to a peaceful life, others
like those of Jazira, the hard core of Marwan's army, felt disgruntled
and waited for a chance to turn the tide. Yet others were ready to
throw in their lot with the new regime either out of conviction or
because of the lack of a better alternative.1 These last offered the
insecure 'Abbasids a golden opportunity of recruiting their own
forces to balance the power of the Khurasaniyya and to compensate
for the loss of those who had returned home. Members of the
'Abbasid family were appointed to lead newly formed contingents of
these men and charged with mopping-up operations in various parts
of the empire.2 At least for the duration of the reign of Abu al-
'Abbas this solution satisfied all parties concerned.
In Khurasan, the problem was far more complicated and created
serious trouble for Abu Muslim. It should not be assumed that, as
the Revolution had sprung from Merv, all the population of
Khurasan and the East had rushed in to support it. Indeed there is
no better indication that the situation there was precarious than the
fact that Abu Muslim himself had to stay behind to tend it rather
than lead the victorious army to reconquer the empire. Naturally
he kept enough recruits at hand to support, if not to protect, his
own position and very soon this proved to have been a wise decision.
It should be remembered that although radical change had taken
place in Merv, nothing much had happened to cause substantial
change in the rest of Khurasan and the East from the conditions
that had obtained after the conquest. The treaties concluded between
the Arab conquerors and the conquered populations of the various
localities were still inforce; and the taxes were allocated and collected
accordingly. The Principalities of Tukharistan were still intact and
still governed by their princes or military lords albeit under Arab
supervision. The city-states of Soghdiana kept their own socio-
economic structures under their own princes. In short, the system
1
Ibid., pp. $2-7.
1
Ibn A'tham al-Kufi, Kitib a/-F*ti$, Istanbul manuscript, Library of Ahmet m . no.
29)6, vol. 11, if. 227s, 228B, 23 JB.
https://doi.org/10.1017/CBO9780511622380.002
THE FOUNDING OF THE ABBASIO REGIME 5
of protectorates created by the Arabs in the East was still in force.1
Now, the sudden jolt of a revolution in their area caused them both
anxiety and consternation, and even aroused the hopes that some of
them had once entertained of regaining their long lost independence.
Fortunately for Abu Muslim the latter feeling was not widespread
in these parts and minor uprisings, such as the one in Bukhara, were
easily subdued.2 But as a leader, he was too vigorous and too
revolutionary to leave things as they were in his own backyard. He
embarked upon a grand plan designed to convert the population of
the entire East to Islam. Nezak the stubborn Hephthalite prince of
Badghis was but one example of such converts. He became a close
adviser to Abu Muslim and his followers were soon amongst the
latter's forces.3 The prince of Khuttal was not so co-operative; when
he felt the pressure he called on the Chinese for help and when it did
not materialize he went to China.4 However, Abu Muslim must
have shown a great deal of dexterity, and in his zeal perhaps tried
to be all things to all men.5 Nevertheless, during his short period of
power he was able to set in motion a policy that irrevocably brought
into the Islamic polity the mass of the indigenous population of the
East and, as we shall see, their role was of paramount significance.
The trouble came from the Khurasanian Arabs and particularly
from those who had returned from 'Iraq. After their great victories
in the west they seem to have had some delusions about their status
at home. At first they agreed to join those who had remained behind
with Abu Muslim to carry out his policy of the effective penetration
of the eastern principalities and Soghdiana. But when they, like the
others, were assigned to garrison duty in the various strongholds of
the area their hopes of returning to civilian life were dashed and they
immediately mutinied. Their concerted rebellion in Bukhara,
Tirmidh and Taliqan was so serious that Abu Muslim himself had
to lead the action against them. Although they seem to have had
allies in Abu Muslim's camp, he was able to ride the storm and
reaffirm his authority over his forces.6
When he saw that a relatively peaceful situation obtained in both
east and west, Abu Muslim decided that it was time to see for himself
what had happened in the heart of the empire. On the pretext of
going on the pilgrimage he journeyed to 'Iraq with a suitable
1
See M. A. Shaban, Islamic His/my A.D. (00-7jo (A.H. i}2): A mm lntirpritatiim,
2
Cambridge, 1971, p. 172. jabari, in, p. 74.
3
Ibid. pp. 100, 107. • Ibid., p. 74.
5
Ibid., pp. 74, 79, 119-20. ' Ibid., pp. 81-2.
https://doi.org/10.1017/CBO9780511622380.002
retinue. At Anbar he was cordially received by the Amir al-Mu'minin

and Abu Ja'far himself accompanied him to Makka. On their way
back after the pilgrimage they were informed of the death of Abu
al-'Abbas and his will that the succession should pass to Abu Ja'far.
It is significant that Abu Muslim, and all the members of the 'Abbasid
family, with one exception, accepted this nomination.1 This one
exception was 'Abdullah b. 'AlJ, an uncle of AbQ Ja'far. He was a
leading member of the family who had distinguished himself in the
fight against Marwan and in the subsequent campaigns in Jazira and
Syria. It was on his advice that the remnants of the Marwanid forces
had been recruited into the 'Abbasid armies. He commanded many
of them together with a small contingent of the Khurasaniyya. He
was, moreover, capable of using these forces to bring into line the
hard core of Marwan's army of Jazira which had lingering hopes of
resisting 'Abbasid control.2 Meanwhile, the Byzantines began to
take advantage of the situation on their Arab borders and it became
necessary for some action to be taken to discourage these attacks.
'Abdullah b. 'Ali and his forces were sent on a summer expedition
for this purpose.3 While on their way to the Byzantine borders the
news of the succession of Abu Ja'far reached them and they turned
back in revolt.
It is not enough to explain this revolt by attributing it to the
ambitions of its leader because he would certainly have needed to
convince the multitude of his followers that these ambitions were
worth risking their lives for. And why should they object so strongly
to the succession of Abu Ja'far? 'Abdullah's army was a good
example of reconciliation between victor and vanquished, and
hitherto had been serving the 'Abbasid cause even against former
comrades. Significantly their revolt was not against 'Abbasid rule;
it was against a particular 'Abbasid in whose succession to power
they must have seen a considerable threat to their own interests. Abu
Ja'far was undoubtedly a man of strong opinions backed with
powerful measures and he was bound to arouse strong objections.
Apparently he had his plans for reconstructing the 'Abbasid forces
within a new structure of the whole empire. He visualized a strong1
central government supported by a cohesive army that would give
it effective control over the provinces. While he had no objection
• Vnd., pp. 88-9.

2
Ibid., pp. 41, 52-7; Ibn A'tham, vol. 11, ff. 227B, 228B, 237A, B; al-Baladhuri, Fu/gj>
al-Buldin, ed. M. J. de Goeje, Leiden, 1866, p. 192.
T b l , HI, p. 91; Bal., Futuh, pp. 184, 189.
https://doi.org/10.1017/CBO9780511622380.002
THE FOUNDING OF THE ABBASID REGIME J
to recruiting men from Syria and Jazira, he did not want to see them
as a separate army operating from bases in their provinces. This
would not only lead to a revival of Marwanid forces but would also
enhance the autonomy of these provinces vis-a-vis the central
government. In this formative stage of 'Abbasid rule this could
become a precedent that would endanger future 'Abbasid policies.
Obviously 'Abdullah was of a different opinion, and, being closely
involved with the Syrians, his inclination would be to favour their
interests. Furthermore, he probably believed that Syria with its long
history of jealousy for its own autonomy, would resist Abu Ja'far's
plans and thus threaten the security of the 'Abbasids. Identifying the
interest of his family with his own and that of his army, he led them
in revolt.
Without any hesitation Abu Muslim, in full agreement with Abu
Ja'far, swiftly marched at the head of all the available forces against
the rebels. The Khurasaniyya in the rebel army did not have any
interest in supporting the rebellion; they quickly withdrew and
joined Abu Muslim's forces. Although some of the Syrians also
withdrew from 'Abdullah's army, it still constituted a formidable
force. When the two armies met the ensuing conflict took four
months before concluding in favour of Abu Muslim.1
At this point Abu Ja'far took a most unusual step, though one
very much in character and one which assumed great significance.
He sent his own representative to watch over the division of the
booty taken from the vanquished army. Considering that this booty
could not have been of any great value and that the share of the
central government was no more than one fifth, Abu Ja'far's action
could not conceivably have been taken simply for fiscal purposes.
He was asserting his authority as Amir al-Mu'minin over Abu
Muslim even at a moment when the latter had just won a great
victory on his behalf. Abu Muslim completely understood the
significance of this action. He expressed his objection in no uncertain
terms, and instead of going back to 'Iraq which would have been his
expected course of action, he headed towards Khurasan declaring
his dissent. * The dispute over the powers of the Amir al-Mu'minin
came to a head at a moment of Abu Ja'far's own choosing. His
future and that of the whole regime rested on the outcome and he
1
Ibn A'tham, vol n, ff. 223B-239B; al-Azdi, Yazld b. Muh., Tariki al-Mawfil, ed. 'A.
Habiba, Cairo, 1967, p. 178.
1
'{'abari, HI, pp. 98-150; Ibn A'fham, vol. n, ff. 238B-240A; Ya'qubi, Ahmad b.
Abi Ya'qub, Tirikb, Beirut, i960, vol. 11, p. 366.
https://doi.org/10.1017/CBO9780511622380.002
was determined to win. He took all precautions, and used every

means to persuade Abu Muslim to return to 'Iraq. When he returned
Abu Ja'far took the enormous but very much calculated risk of
having him executed without delay. Interestingly, the Khurasaniyya
did not raise any objections. The authority of the Amir al-Mu'minin
had acquired enough legitimacy to allow him to eliminate the Amir
dl Muhammad. As a consequence of this action Abu Ja'far came into
his own, and it was probably then that he decided to take the title
al-Mansur, the one destined to win. This was not an empty boast
nor was it a simple shot in the dark. Abu Ja'far certainly had a
genius for long term planning which characterized most of his
actions and this one was no exception. He decided on a title not only
for himself but also, in due course, the title of al-Mahdi for his
successor, with all its messianic connotations. It was no secret that
the 'Abbasids had set themselves apart from the Shi'ites, but Abu
Ja'far was aware that there was a great deal of sympathy for these
dangerous rivals in all parts of the empire. However, given time an
'Abblsid-Shi'ite reconciliation was not necessarily impossible and
it was certainly desirable. The Khurasaniyya had rejected the notion
of Imam/Amir al-Mu'minin, therefore he could not take the title
Imam. The next best thing that might appeal to the Shi'ites was a
title Mansur/Amir al-Mu'minin especially when it would be followed
by a title Mahdi/Amir al-Mu'minin in the not too distant future. For
almost a century, from the time of Mukhtar, the notion of a mahdi
had been in circulation; prophecies of messianic expectations had
been widely believed and there was no harm in exploiting them for
the purpose of reconciliation. • There was resistance to this idea from
some of the leaders of the Khurasaniyya and indeed from 'Isa b.
Musa, a most respectable member of the 'Abbasid family. But by the
end of Mansur's long rule 754-75/136-58 this opposition was easily
brushed aside at the crucial moment of the confirmation of
Mahdi.*
Mansur's long-term planning was at its most striking in the period
before the establishment of Baghdad. Practically every aspect of his
policies was involved in this project. Yet, without the benefit of
planning committees or consulting bodies, he single-handedly
planned its construction, taking into consideration strategic,
economic, administrative and demographic factors. In one year he
was living in his palace on the site and the whole Round City was
1
See Shaban, Islamic His/my, p. 95.
* Tabari m pp. J J I , 571, 544, 4JJ; Ya'qubl, Tarikb, vol. 11, pp. }79~8o.
https://doi.org/10.1017/CBO9780511622380.002
finished in no more than four years at the low cost of 4,000,882
dirhams.1 In his own lifetime it became a great metropolis.
The focal point of the new structure of government was Mansur
himself. Although he gathered all secular powers in his own hands
he had no delusions about the restricted functions of his office. He
was not much different from an Umayyad ruler, and being a realist
he laid no claim to religious authority to strengthen his position.
However, by bestowing titles with religious connotations upon
himself and his successor he was clearly indicating his thinking about
the future. Meanwhile, he was vigorously pursuing plans for a
highly centralized government projecting his authority from Bagh-
dad. The base of his power was the Khurasaniyya for whom
living space was provided around the city. Inside the walls were the
administrative departments, the heads of which were directly
responsible to him. These departments were to co-ordinate the work
of the various functionaries in the provinces.2 To make sure that
he was well informed about affairs in all corners of the empire and
to ensure that the representatives of the central government were
behaving properly, he appointed his own independent agents who
reported directly to him every day, even on such mundane matters
as food prices in their respective areas.3 The official title of such an
agent was sahib band, postmaster, but more important and to
emphasize his direct relationship to the ruler he was also given the
honorary status of mawla Amir al-Mu"minin. This is not to be
confused with mawld meaning a freed slave, a client or a non-Arab
member of the army related to a certain man or an Arab clan, as
was the case under the Umayyads. Under the 'Abbasids assimilation
had certainly done away with this type of relationship as far as the
army was concerned, and virtually deprived it of any meaningful or
significant sense in the rest of society. As we shall see, the term
mawla, like many other similar terms, went through many different
changes in various parts of the empire during the period of this
study.4 Knowing that mawld could mean so many things and
capitalizing on the special relationship that it used to denote,
Mansur simply gave it a different twist to establish a new relationship
between himself and members of his administration, and in this case
1
Tabari, in, p. 526.
* Jahshiyari, Muh. b. 'Abdus, Kitab al-Wit!(arS\ ed. M. al-Saqqa et al., Cairo, 1938, pp.
96-1 j j ; Azdi, Maw/il, p. 2 15; Ya'qubl, Kitab al-BuUdtt, ed. M. J. de Goejc, Leiden, 1892,
pp. 240, 245.
1
Tabari, HI, pp. 414, 4 j j ; Ya'quW. Tarikh, vol. 11, p. } 84.
4
See below, p. 31.
https://doi.org/10.1017/CBO9780511622380.002
IO THE FOUNDING OF THE ABBASID REGIME
it can only have meant confidant of the ruler. Ibn Khaldun carefully
explains this special relationship for us, calling it iffind' i.e. "choosing
another person for oneself for a special affair which this person is
required to accomplish in a sufficient manner".1
In Mansur's administration, there were more than fifty persons
given this title and all of them acquired very high posts - at least
seventeen were postmasters in ?reas they had originally come from
or about which they nad special knowledge.2 Some of them were
Arabs and as might be expected many were non-Arabs.3 Of the
latter some were given or assumed Arab names. Probably their
original names were ostentatiously un-Islamic or difficult to pro-
nounce. There are examples of many mawlds and their descendants
serving successive 'Abbasid rulers over a period exceeding a century
and in the same capacities/ At least one of them was a prince from
the eastern principalities, another was a brother-in-law of Man§ur,
and a third was his own step-son.5 This last example recalls the
English custom of kings raising their illegitimate sons to the higher
echelons of the nobility. Indeed under Mansur's successors there
were cases closer to the English custom, when brothers or relatives
of favourite concubines were appointed to high offices and accord-
ingly were given this honorary status.
Man?ur's internal espionage network was so successful that he
was reputed to have had a mirror in which he could distinguish
between friend and foe.6 Certainly, dismissed provincial governors
knew only too well that they could not get away with any riches
they might have extorted while in office. The special accounting
department that Mansur established for this purpose was fully
informed and was a good example of the strictness of his administra-
1
Ibn Khaldun, al-Muqaddima, Beirut, 1961, pp. 237-8, 326-7; also see E. W. Lane,
Arabic-English Lexicon, reprinted Beirut, 1968 s.v. fit', vol. iv, p. 173;.
1
Tabari,ni, pp. 139, 140, 145, I J I , 306, 319, 323, 367, 380, 392, 428, 454, 455, 456;
Jahsbiyari, pp. 101, 124, 125, 129, 134; Bal., Fu/ify, pp. 18;, 287, 293, 294, 310,401;
Azdi, Mawfil, pp. 198, 296; Ya'qubi,BuUdn, pp. 241, 242, 244, 245,247,249; Ya'qubi,
Tdrlkb, vol. 11, pp. 384, 392; Ibn al-Athir, 'las al-Din, al-KJmil fl al-Tarikb, ed.
C J. Tornberg, Leiden, 1866-71, vol. vi, pp. 22, 50, 213, 467; Tayfur, Ahmad b.
Abl Tihir, Kilab Baghdad, ed. H. Keller, Leipzig, 1908, p. 349; Yiqut, Mit'jam at-
Bu/a\3n, Beirut, 19)7, vol. m, pp. 25, 26.
> Tatar!, in, p. 323; Ya'qubi, BuUdn, p. 244.
• Tabari, in, pp. 367, 428, 454, 529, 576. 58*. 5«}> 979. 998, 1384, 1500, 1838; Tayfiir,
pp. 7, 16, 19, 120, 121, 142; Jahshiyari, pp. 124, 277; Bal., Fulib, pp. 293-4; Athir,
Kdmil, vol. vi, p. 200.
< Ibn al-jawzl, 'Abdulrahmin, al-Mmtazam, manuscript, biography of 'All. b. $ilih
229 A.H., cf. Miskawayh, To/drib al-Umam, ed. H. P. Amedroz, Oxford 1920-1, vol. 1,
p. 16; Tabari, in, p. 456; Ibn Hazm, Jambarat Aiuib at-Arab, ed. 'A. Hirin, Cairo,
1962, p. 21. * Tabari, HI, p. 166.
https://doi.org/10.1017/CBO9780511622380.002
THE FOUNDING OF THE ABBASID REGIME II
tion.' One mawld, whose descendants attained very high positions,

namely Khalid b. Barmak, seems to have been the fiscal expert under
Mansur. He accompanied the revolutionary army on its march and
was in charge of all its financial matters.2 Mansur made use of his
expertise, appointing him head of the central department of revenues
and then later governor of those provinces which had particularly
difficult taxation problems.3
It goes without saying that Mansur did not want to revive the
office of wa^tr; instead he had a kdtib who was more of an adminis-
trative assistant and strictly without any executive powers. The
choice of a native 'Iraqi, Abu Ayyub al-Muriyani, as a second
occupant of this office may indicate Mansur's awareness of his own
lack of knowledge about the intricate affairs of 'Iraq. However, it
seems to have been an experiment which was abandoned after the
death of Abu Ayyub. Mansur did not appoint anybody to replace
him and continued as always to exercise direct control over all
departments of his government.*
On the whole Mansur sought to incorporate the administrations
of the provinces into the central government over which he himself
had full control, hoping, perhaps, that this in turn would eventually
lead to the integration and unity of the whole empire.
The organization of the military forces was a more difficult
problem because of the unexpectedly increasing demands on many
internal and external fronts. Mansur would have been content to
rely on the Khurasaniyya alone to meet security requirements, for
the revolt of 'Abdullah b. 'All had shaken his confidence in troops
from Syria and Jazira even under the leadership of members of his
family. Although Abu Muslim had done his utmost to recruit new
forces from the indigenous population of the East, the process of
Islamization had not yet taken root enough to allow such an effort
to be wholly successful. Apparently the more powerful Hephthalite
and Soghdian princes stood aloof from the Revolution and saw no
reason to become involved in an adventure that might drag them as
far afield as North Africa. However Abu Muslim's political manoeuvr-
ing, persuasion and pressure had convinced some of the lesser princes
and military lords that they should join the cause and rush to
'Iraq as fast as they could mobilize their followers. In Baghdad,
living quarters were provided for small contingents from Balkh,
Khuttal, Isfijab, Ishtikhan, Bukhara, Faryab, Farghina, Khwarizm,
> Wd., p. 415. * J«hthiyiri, p. 87.
» Ibid., pp. 89, 99. IJ6. • Ibid., pp. 96-iJJ-
https://doi.org/10.1017/CBO9780511622380.002
1
Gurgan and Badghis. A combination of events forced Mansur to
fall back on other sources of recruitment to meet lesser tasks, leaving
the Khurasaniyya free to cope with serious emergencies.
Egypt was easily brought into the 'Abbasid domains and the
Arabs of this province co-operated with the new regime. They
helped suppress some minor Coptic uprisings and subsequently were
entrusted with the security of the province.2 Further west, North
Africa had been in a turmoil for decades; and so was Spain.3 Berber
revolts and differences between the Arabs in these regions were still
raging. Since it was obvious that it would require enormous military
forces to bring these vast territories under control, the decision was
soon taken to abandon Spain completely and leave the Arabs there
to their own devices. In North Africa the Berbers were not only
attempting to regain their independence but also were very bold on
the Egyptian borders with Libya.* This dangerous situation was
compounded by the possible threat of a Byzantine comeback. The
security of Egypt demanded that the 'Abbasids should hold as
much as possible of the North African coast. The presence of a
sizable Arab population in Tunisia decided the issue. It was hoped
that should they prove to be as co-operative as the Arabs of Egypt,
they could eventually uphold 'Abbasid authority in their region on
their own. But meanwhile the Khurasaniyya were chosen for the
difficult operation of pushing the Berbers back as far west as possible.
After considerable difficulties and successive campaigns over more
than ten years, 'Abbasid rule was established only as far west as
Qayrawan 772/15 5 and never beyond. Yet, the powerful thrust of
the Berbers of north-west Africa made it necessary that a large
contingent of Khurasaniyya should be stationed there. This was
done first on a rotational basis but eventually they were permanently
settled there.5
The revolt of 'Abdullah b. 'Ali encouraged the Byzantines to
resume their aggressive activities and in the following year 755/138,
they overran the stronghold of Malatya and razed its fortifications.
However it was possible to spare some of the Khurasaniyya to
restore the balance of power on this dangerous front. Malatya was
1
Ya'qubl, BUUOH, pp. 248-9.
1
Maqrlzl, Kbifal, Cairo, 1149 A.H., vol. ti, p. 261.
1
Shaban, Islamic History, pp. 149-51.
* Athlr, Kjmil, vol. v, pp. 257-40; Ibn 'Idharl, AI-Bay&t al-Mugbrib, ed. G. S. Colin and
R. Levi-Provensal, Leiden 1948-51, vol. 1, p. 65.
' Bal., Fu/ip, pp. 2 ) 2 - ) ; Tabarl, m, pp. 370-}; Ibn 'Idhari, pp. 73-9; Azdi, Mawsil,
p. 91.
https://doi.org/10.1017/CBO9780511622380.002
THE FOUNDING OF THE 'ABBASID REGIME IJ
recaptured and its fortifications were rebuilt, but no Khurasaniyya

could be spared to man this important stronghold. Men from Syria
and Jazira were recruited and encouraged to settle there and in
similar strongholds all along the borders. These men were granted
land allotments in these areas in addition to the regular army stipend
of eighty dirhams a month.'
In Khurasan and the East, 'Abbasid authority was severely shaken
after the execution of Abu Muslim. The loss of his effective leader-
ship deprived the 'Abbasids of the stability previously afforded them
in these vast regions. Even those who had been willing to support
the new regime had their apprehensions about its intentions. The
sudden turns of the regime appeared to have deviated from the put-
poses of the movement born in their midst. The disappearance of the
men who played such an important part in organizing the Revolution
could hardly have helped the situation. The successors of Abu
Muslim, as governors of the East, could not restore the sagging
confidence in the regime. The first of these governors carried out the
instructions of the central government to the letter. This approach
led to an outright mutiny of his forces and he was accidentally killed
in the scuffle.2 The following governor realized that because of the
demands of the central government he was put in an impossible
situation. Although he had been one of the most faithful sons of the
Revolution and indeed closely associated with Mansur against even
Abu Muslim, he nevertheless led an uprising against the central
government. He succeeded only in dividing the population of
Khurasan against itself, creating more reasons for instability.3
Another symptom of unrest took the shape of a popular revolt
amongst the indigenous population of western Khurasan between
Nishapur and Rayy, led by a certain Sunbad. Although it was a
minor affair in itself, it was potentially dangerous because it threatened
to cut off the vital northern route between Khurasan and the west.
In the event it proved to be even more serious for it resulted in the
opening of a new front, which the 'Abbasids could hardly contain,
in the mountainous region to the south of the Caspian Sea.4 Most of
this region was still independent under the rule of its own princes
who stubbornly resisted Arab advances. Alarmed by the Sunbad
uprising, the 'Abbasids rushed a newly recruited army from the
1
Tabari HI, pp. 111-5, I J J ; B»l., Futib, pp. 165-6, 187-190; Athlr, Klmit, vol. v,
pp. J7O-*.
1
Tabari, in, p. 154.
1
Ibid., pp. IOO-I, 134-5; Ibn A'tham, vol. 11, ff. IJJB, 237A.
b HI, pp. 119-20.
https://doi.org/10.1017/CBO9780511622380.002
Fars province which, interestingly enough, had both Arabs and
Persians among its members. Although the uprising was soon sub-
dued, this army, probably because of a dispute with the central
government over its share of the booty, started a revolt of its own.'
There was no alternative but to send out all the available Khura-
saniyya to deal with the situation, particularly as Sunbad was able
to flee and take refuge with the Prince of Tabaristan. The army of
Fars was soon dispersed and Sunbad was never heard from again,
but hosts of enemies were stirred up. The neighbouring Daylam
felt threatened and joined their brethren of Tabaristan to pose a
continual threat to the road to Khurasan. As there were not enough
Khurasaniyya to station there permanently Mansur once again had
to fall back on recruits from Syria and 'Iraq to form a big garrison at
Rayy under the command of his own son and heir apparent.2
In 762/145 the inevitable Shi'ite revolt broke out simultaneously
in Hijaz and the region of Basra. It had been no secret that the other
members of the House of the Prophet, namely the descendants of
'All, were increasingly disappointed by the usurpation of what they
considered their inherent rights, in addition to the apparent revision-
ism of the 'Abbasids. It was only a matter of time before they would
attempt to bring down 'Abbasid rule. But as usual with most
Shi'ite revolts, this one was poorly planned, and was more akin to
political agitation than a serious attempt to launch a new order. It
was as if its leaders were convinced that simply by declaring their
intention to take over the whole Muslim community would fall in
line behind them. They did not realize that although there was a
great deal of sympathy for them, the community at large was not
quite willing to go through another convulsion in so short a period.
Only four thousand Khurasaniyya under the militant anti-'Alawid
command of 'Isa b. Musa were enough to crush the revolt in both
Madina and Basra in a very short time. Two new names, Muhammad
the Pure Soul and his brother Ibrahim, were added to the register of
'Alawid Shi'ite martyrs.3 It is difficult to see why this revolt is given
such prominence in our sources; perhaps it was because it was the
last revolt of its kind. Henceforth Shi'ite revolts took a completely
different course, following new strategies and adapting themselves
to new causes. The simple cause of legitimacy proved to be insuffi-
cient to win for the descendants of the Prophet what they believed
to be their proper rights. It is significant that Ja'far al-Sadiq, a
descendant of 'Ali through his son Husayn, did not support the
1
Hid., p. 122. * Ibid., pp. 136-43. > Ibid., pp. 174-257-
https://doi.org/10.1017/CBO9780511622380.002
THE FOUNDING OF THE 'ABBASID REGIME 15
cause of his cousins although he was one of the most respected

leaders of the 'Alawids at the time. Later he was to become the foun-
tain of wisdom for the Isma'lli Shi'ites.1
The suppression of the Pure Soul's revolt was by no means the
end of Mansur's troubles. Watching the Arab difficulties on the
Byzantine front, the Khazars seized the opportunity to advance in
the Caucasus. Suddenly the whole northern frontier from the
Mediterranean to the Caspian was under attack and Mansur had to
mobilize all possible forces to meet these formidable enemies.
Surprisingly no call went to Khurasan for new recruits; it must
have been realized that there was no chance of help from this
troubled region. As the Arabs and Persians of Firs had just become
disgruntled, the necessary forces had to come from 'Iraq, Syria and
especially Jazira, which was particularly threatened by the Khazar
attack. Desperate as he was, Mansur required every man of Kufa
and Basra whose wealth amounted to ten thousand dirhams and
more each to volunteer to fight the enemy. The frontier was finally
stabilized and the threat averted, but garrisons had to be set up
across the Caucasus and recruits were paid regular stipends to
settle there.2 An attempt was made to establish a big base at Rafiqa
near Raqqa but it was soon abandoned and the bulk of the 'Iraqi
and Syrian recruits were stationed at Rusafa, across the river from
Baghdad, where living quarters were built for them. The only
man who could be trusted to be in charge of these new forces
was the heir apparent, and indeed they were called the army of
Mahdi.3
The eastern regions continued to show signs of unrest. The
Hephthalites of Harat, Pushang and Badghis who, under Nezak,
had joined Abu Muslim, decided that their co-operation was not
only useless but probably also harmful to their interests. Led by a
certain Ustad-Sis and supported by their brethren from Sistan
they revolted, trying to re-establish their own control over their
principalities. With some help from the central government but
mainly with his own forces, the governor of Khurasan was able to
defeat the rebels and punish their supporters in the south.4 An
important result of this victory was the widespread conversion to
Islam of the Hephthalites. One of their leaders was Mu?'ab b.
• Ibid., p. 254.
1
Ibid., pp. 318, 528, 353, 571; Ya'qubl, Tarikb, voL n , p. $71; Ibn A'thsun, vol. 11, ff.
241B-2B.
1 Tabari, HI, pp. 566-7, 372-3, 491. • Ibid., pp. 354-8.
https://doi.org/10.1017/CBO9780511622380.002
l6 THE FOUNDING OF THE 'ABBASID REGIME
Ruzayq of Pushang whose descendants, the Tahirids, were soon to

play an important part in Islamic history.1
The last few years of Mansur's rule were comparatively peaceful,
yet they were not particularly noted for important developments or
great innovations in the government of the empire. He seems to
have satisfied himself with setting up the machinery to rule his
domains as he saw fit. Hisham's abilities as a ruler are reputed to
have impressed Mansur and indeed his own rule can be best described
as a re-establishment of Marwanid rule with minor modifications.
His administrative and long term planning abilities were utilized to
restore the status quo in favour of the 'Abbasids without much
thought being given to the underlying problems of society. In short,
he was not a ruler who was profoundly concerned with reform and
he does not seem to have realized that his empire was going through
a period of fundamental change. As a leader of a supposedly
revolutionary regime he introduced very little change and indeed
was basically a conservative ruler.
Mansur inherited a fiscal policy based on an agricultural economy
but it must be pointed out that as yet there was no universal taxation
system for all the empire. Although in principle Muslims and non-
Muslims paid appropriate taxes on the land under their possession,
these taxes varied considerably from one province to the other, and
there were many unjust anomalies.2 Yet Mansur made no attempt to
remedy this situation and was content to leave it unchanged.3 This
is the more disappointing in that he showed some signs of awareness
of a major problem which was to bedevil the Islamic polity for
generations if not centuries; the problem of the taxation of urban
communities. Since in early Islam the great majority of tradesmen
and artisans were non-Muslims the graded poll-tax was in effect a
fair urban tax. The immense and rapid urbanization coupled with
the increasing conversion of the subject peoples created many
virtually tax-free zones. No system of taxation was introduced to
replace the defunct poll-tax in cities and towns and subsequently
all tradesmen and craftsmen were simply not paying any taxes.
Indeed the textile industry, which was primarily a cottage industry,
owed its fantastic development in every Islamic town to this tax
loophole. In Baghdad MansQr was asked to help a man who had a
1
See below, p. 4).
• Sbaban, Islamic"History,pp. 57-9, 48-9.
' Bal., F«tf$, p. 272; A. K. S. Lambton, Landlord and Ptasanl in Pirjia, Oxford, 1953,
pp. $*-), wrongly ascribes to Mansur tax changes that were introduced by Mahdl,
tee below, pp. 23-4.
https://doi.org/10.1017/CBO9780511622380.002
THE FOUNDING OF THE 'ABBASID REGIME 17
wife and three daughters to support; he refused exclaiming that

such a man should have four busy spinners in his household.1 This
could well have been a remark passed by a Scottish laird in Shetland
about the women knitters there. However two centuries had to pass
before such workers were taxed because of the obvious difficulties
involved.2 On the other hand, tradesmen who needed shops for
their businesses were easier targets for the tax collector. Emulating
the later Umayyads who had built shops and rented them to the
public for private investment, Mansur also had shops built in his
city.3 Tradesmen were charged a suitable rent, ghalla, according to
the size of the shops they needed. Those who wished to build shops
on state land were to pay a lesser rate and all payments were to go
to the state treasury.4 Unlike later 'Abbasids Mansur did not have a
private treasury and did not differentiate between his own revenue
and that of the state.5 These properties, mustagballdt, were to be an
ever expanding feature of every city. Of course this system of
mustagballdt was not enough to take care of the very complex
problem of urban taxation, but it was a step in the right direction.
But Mansur chose to overlook a more important tax loophole
resulting from the rise of the cities and the huge volume of trade
that passed through the hands of a relatively small number of rich
merchants. According to Islam these wealthy people had to pay
z%% of their income, after certain adjustments, as qakat. However
since the time of 'Uthman 644-56/23-35 the state had ceased to
enforce the collection of this modest tax and it was left to every
individual to dispose of it according to his conscience.6 In effect this
meant that while small businessmen had to pay some form of tax
most of the wealth of the big merchants was not taxed.
As for international trade, from early Islamic times the rule had
been that a tax should be paid at the port of entry, e.g. Basra, at the
rate of z\%, 5% or 10% according to the status of the merchant
concerned, whether he was a Muslim or not. There were many
exemptions and many ways of evading this tax altogether.7 Mansur
' Athir, Kamil, vol. vi, p. 19.
1
See below, p. i 8 j .
1
Bal., Fuiuh pp. 117-18; Jahshiyari, p. 47; Ya'qubl, BuM&i, p. J I I .
4
Tabari, HI, pp. 323-4, 579; Bal., Futiib, p. 295; Ya'qubl,huldan pp. 141-2, 246; Yiqut,
s
Bulddn, vol. iv, p. 484. Tabari, m, p. J90.
6
Qudama b. Ja'far, Kitab al-Kbaraj, part seven, published as part of A. Ben Shemesh,
Taxation in Islam, vol. 11, Leiden 196;, p. 103; Ibn Sallim, Kitab al-AmmH, ed. M. H.
al-Fiqi, Cairo, 13J3 A.H., pp. 457, 573.
•> Bal., Futib, p. 385; Ibn Sallim, Ammal, pp. 411, 4*7, J J I , 535, 5*8-9; MaqiW,
Kbitaf, pp. 121, 122, 123.
https://doi.org/10.1017/CBO9780511622380.002
l8 THE FOUNDING OF THE 'ABBASID REGIME
seems to have taken some measures to tighten the control for the
collection of this tax. We are told that he had walls built around
Basra and that the expenses for this project were exacted from its
residents to the tune of forty dirhams each.' As this measure did not
have any military justification at the time it is possible to conclude
that it was meant to control the volume of trade going into and out
of Basra.
The Umayyads had appropriated for themselves a great deal of
farm land all over the empire. Mansur simply confiscated it all and
redistributed it among members of his family. The only difference
between his action and that of the Umayyads was that he took into
consideration female members of the family who were also given
lands.* This seems to have established a tradition among the
'Abbasids that their women were as much entitled to share in their
wealth and exploit it as their men. While Mansur was very generous
with his own family he was quite stringent with his employees. The
highest salary he paid was 300 dirhams a month which was not much
by the standards of the time.3 This seems to have encouraged some
to indulge in corrupt practices such as taking commission from
favoured merchants or hoarding and speculating in food commodi-
ties; but these practices were not widespread, thanks to Mansur's
alert spying system.4
Under Abu al-'Abbas new gold and silver coins had been minted.
They had roughly 5% and then 10% less of the precious metals in
them than those of the last Umayyads. To meet his increasing
expenses Mansur issued yet cheaper money, stepping up this
percentage to about 15.s These steps must have had some inflationary
effect on the economy, but perhaps because of the gradual introduc-
tion of these coins, the effect went unnoticed and raised no problems.
There was, however, some slight objection when Mansur, in order
to withdraw the better coins from circulation, ordered his tax
collectors to accept only Umayyad coins for taxes paid in cash.6
The success of these limited economic measures is best shown in the
fact that at his death Mansur left a healthy reserve in the treasury.7
Thanks to the Revolution, assimilation was a fact of life that could
T»barl. MI, p. J74. » Bal.,"FutSb,p. 294.
Tabari, in, pp. 434-5; Jah»hiyiri, p. 1*6.
Jththiyiri, pp. 117, 118-19, 272.
Maqrlzl, al-Nuqid al-Istimiyya, ed. M. S. A. Baîr al-'Ulum, Najaf, 1967, p. 17.
Bal., FutiH), p. 469.
Azdl, Maw/il, p. i j o ; Mat'udi, al-Tanbib wa al-Isbraf, ed. M. J. de Goeje, Leiden, 189),
p. J42.
https://doi.org/10.1017/CBO9780511622380.002
not be undone and it was left to take its course according to the
circumstances of each province without much active interference on
the part of the central government. In Khurasan, where it struck
roots, it was bound to gather momentum and no measures were
taken to slow it down. On the other hand, in Egypt it was hardly
effective; the Egyptians remained Copts and the Arabs continued
their separate existence as they had done under the Umayyads.
Naturally there were exceptions but it was to be some time before the
two communities drew together. In the growing metropolis of
Baghdad assimilation was not an issue. It was simply taken for
granted as an accomplished fact, and the non-Arab converts were
engaged in all walks of life on the same basis as the Muslim Arabs.
It is worth pointing out that there is no justification for the belief
of some modern scholars that the non-Arabs received preferential
treatment at the hands of the new regime which, it is claimed, owed
its victory to them. This is not true. It was a natural result of
assimilation that more non-Arabs were brought into all government
services and it is unnecessary to observe that they were the ever
increasing majority over all the empire.
Finally, after a long, eventful reign and overcome by his worries
and ulcers Mansur died, well over sixty years old.
https://doi.org/10.1017/CBO9780511622380.002
TOWARDS A CIVIL WAR
If proof were needed that the foundations of the 'Abbasid regime, as
laid down by Mansur, were not as strong as they seemed, it was not
long in coming. Hardly a generation after his death the empire was
once more embroiled in a devastating civil war that explicitly pointed
up the weaknesses of the regime. Over a period of thirty-five years
three 'Abbasid rulers followed the political orthodoxy of their
founding-father without much deviation. Although some changes
occurred, these were, on the whole, ad hoc measures which were
attempted only as situations arose that threatened the central control.
The general policy line was to enhance the power of the ruler and
his central government, and that was thought to be enough to govern
his vast empire. Changes in taxation were introduced to enrich the
treasury without much regard to the interests of the tax-payers.
Modifications in the administration were attempted for the sole
purpose of tightening the grip of Baghdad over the provinces. The
rulers persisted in assuming titles with religious connotations in the
hope that this would give them some semblance of religious
authority. They continued to use the Khurasaniyya, their force de
frappe, to counter the protestations of any dissident subjects. How-
ever, they failed to see that such a militia did not constitute a
professional standing army or that it was no substitute for one. They
could not comprehend that the use of this part-time military force
to enforce the political will of the central government could only
lead to serious complications. The lessons to be learnt from earlier
precedents such as the dissolution of the Ba;ran and Kufan militia
or the revolt of the Syrian army against its Marwanid masters were
completely lost on these 'Abbasid rulers.' In a fast growing metro-
polis like Baghdad, allowing the Khurasaniyya to indulge in profit-
making enterprises was the one sure way of losing them to civilian
life altogether.
Assimilation was, in itself, an admirable achievement but it
needed to be followed up by measures to consolidate it in order to
integrate the peoples of the empire into one united community. In
1
Shaban, hlamit History, pp. m , 155.
[ao]
https://doi.org/10.1017/CBO9780511622380.003
TOWARDS A CIVIL WAR 21
a fast expanding economy and with a vastly increased volume of
internal and international trade, consideration should have been
given to the varying interests of all sections of the population and
to those of the different provinces, in order that the economy could
function under the most favourable circumstances. The relative
calm within most of the empire allowed for a good measure of
physical mobility for a considerable number of people as between
one part and another, but this in turn caused population pressures
to build up in areas of exceptional economic opportunities. The
central government was either blissfully unaware of these problems
or completely oblivious of its responsibilities.
Mahdi, the long-promised "divinely guided" ruler, finally came
to power for a reign of about ten years 775—85/158—69. Everything
had been prepared for his "coming", and provided that the quiet
state of affairs that had prevailed in the last years of his father con-
tinued, an era of justice for all was conceivable. Indeed he inaugur-
ated his rule with a grand gesture of reconciliation; he released from
jail all his father's political prisoners.1 Furthermore, as a "mahdi",
and probably seeing himself above politics, he followed what can
be described as a religious policy. He instituted a High Court,
maâlim, where he personally sat to hear the complaints of his
subjects against his subordinates, in order that people could see that
justice was being done. 2 He also sought to take advantage of his
position as a ruler whose ideals were closer to Shi'ite ones than
those of any of his predecessors, by pursuing a policy of reconcilia-
tion with, at least, the moderate Shi'itcs. One of those released from
Mansur's prisons was Ya'qub b. Dawud, a man from Merv whose
support of the Shi'ite cause needed no proof. His father and uncles
had worked for the last Umayyad governor of Khurasan and the
East, and in this capacity the father had been of good use, as a spy,
to Abu Muslim. Moreover, Ya'qub and his brother had taken part
in the revolt of the Pure Soul after which they had been imprisoned.3
Appointing Ya'qub to the unique position of "brother in God",
Mahdi made use of his circumstances and contacts to put into effect
the desired reconciliation.4 To facilitate his task, Ya'qub was given
the right to appoint his own agents, umand\ in all provinces, and
these were granted a free hand and authority even over the governors
in their respective areas.1 This scheme to create what amounts to a
1
T'bori, 111, p. 461. ' Ibid., p. 408; MaqrizI, Kiifaf, II, p. 107.
1
Tabari, n l j pp_ {06-7; Jahjhiyiri, p. 155.
4
jabari, m, p. 464; Jahshiyiri, p. IJJ. * T*b*ri. Hi, p. 486.
https://doi.org/10.1017/CBO9780511622380.003
22 TOWARDS A CIVIL WAR
one-party system through a complicated network of political
commissars did not bring about the desired results. First, the
differences between the parties concerned were too deep for such a
superficial attempt to succeed without real concessions on the part
of Mahdi. Second, these political commissars with their extraordinary
powers aroused the antagonism of the mawali post-masters and the
governors in the provinces. After five years of trying Ya'qub was
dismissed and returned to prison. •
Undaunted by this failure and probably as a reaction against it,
Mahdi embarked on a purgative campaign against the so-called
qanddiqa. Neither the origin of this term nor its derivation can be
determined, but this very vagueness and the absence of any common
cause among the people who were thus persecuted, lead us to believe
that it was a nomenclature intended to cover all those who disagreed
with Mahdi's religious policy. However, this campaign, couched as
it were in religious guise, was bound to generate opposition on the
same grounds even among members of his own family.2
In the administrative field Mahdi, not surprisingly, showed some
ability, perhaps because of the training he had received from his
father. Mansur had been presented with an illuminating report on
the art of governing by Ibn al-Muqaffa', an official of his administra-
tion who had also worked for the Umayyads.3 Trusting only in his
own judgment Mansur does not seem to have made any effort to
heed the advice contained in this report. In contrast, Mahdi seems
to have studied it very carefuUy and indeed to have implemented
some of its recommendations. Of course he was determined to retain
full control of his government, and any advice that recommended
measures to tighten this control would not be overlooked. In the
event, a comptroller was installed in every government department
and a chief-comptroller appointed to be in charge of all of them.
Mahdi's chief administrative assistant -wa\ir or kdtib, the difference
between the two was slight-had nothing to do with these new
officials who kept their independence even from his "brother in
God".*
There was another piece of advice in the report which was of
great importance, especially in the light of later developments. This
was the emphatic recommendation that the military forces should be
strictly confined to their duties and under no circumstances should
1
Ibid., pp. jo8, J17. * Ibid., pp. 499, 517-22, 534, 548-50. 588.
1
M. Kurd 'Ali, ed. Kasd'il al-hulagba', Cairo, 1946, pp. 117-34.
4
Tabaif, HI, p. 493; Jahshiyaci, pp. 146, 166.
https://doi.org/10.1017/CBO9780511622380.003
they be involved in civilian matters, particularly the collection and
administration of taxes. Mahdi understood the significance of this
important piece of advice and, unlike his father who sometimes left
both the department of revenues and that of the army under one
head, he, Mahdi, always assigned to these departments separate
heads and separate comptrollers in order to avoid any possible
interference or involvement of the military in matters of taxation.'
Furthermore, he was the first person who attempted to recruit men
specially to serve in the police force which hitherto had been drawn
from the army. To underline the importance of the duties of this
new force, its members were recruited from the prestigious popula-
tion of Madina.* Unfortunately this policy was not long continued
after Mahdi and, as we shall see, the result was disastrous.
The author of the report also recommended that tax anomalies
in the empire should be eradicated, and that inflation should be
taken into consideration in tax matters as well as in determining
the stipends of the army.3 Mahdi did not do anything to rectify the
situation with regard to the tax anomalies, neither did he adjust the
army stipends. But he did grasp the significance of inflation and
realized that the existing system of taxation was causing the treasury
a considerable loss of revenue. This system, as we know, was the
same as that imposed by the Arabs at the time of the conquest, by
which the conquerors took over in substance the systems they had
found in the provinces. If the Arabs acquired any lands at the time,
as in Syria and Jazira, they were required to pay the Muslim tithe,
i.e. one tenth of the produce. This arrangement kept the revenues
on a level with inflation and presented no problem. The question
was more complicated in the case of lands that had stayed in non-
Muslim hands, as in the Sawad. Although many of the people there
were eventually converted to Islam they continued to be required
to pay the same taxes on the lands in their possession. This was a
fixed rate in kind or cash assessed on the basis of area and kind of
product.4 This fixed rate, particularly in cash, was far behind the
current prices of the relevant products, and that was where the
treasury would stand to gain if a new system were to be introduced.
His kdtib suggested the sharing, muqdsama, system and it was soon
implemented. The tax was assessed on the actual produce and the
imposed rates varied from half to one third according to the method
1
Kaid'il, pp. 122-5; Tabari, Hi, p. 522.
» Jabari, in, pp. 485, 548, 535, 762-}.
1
Rasd'i/, pp. 124-6. * Shaban, Islamic History, pp. 48-9.
https://doi.org/10.1017/CBO9780511622380.003
and expense involved in irrigation. • Another situation where a new
system would benefit the treasury was where fruits and vegetables
were concerned. Islam had originally conceived of levying taxes
only on wheat, barley, dates and raisins, i.e. grapes; products that
were familiar in Arabia.2 Outside Arabia, in places such as Syria
where olives were a major product, the Arabs did not realize at
first that this was simply another taxable product. However, as
early as 'Abdulmalik, a tax had come to be levied on olives at the
rate of i dinar per ioo trees if these were within one day's distance
of a market town, or i dinar per 200 trees in remoter areas.3
In the Sawad there were no olives, but there were many kinds of
fruits and vegetables that presumably went untaxed for some time.
We do not know when exactly this situation was rectified, or indeed
whether it was at all rectified during the Umayyad period. But we
know that in the neighbouring Fars province Mahdi annulled a tax
levied on certain fruit trees.4 On the other hand we also know that
he extended the muqdsama system to cover all agricultural products in
the Sawad; the only concession to the producers was that their
expenses should be deducted before the share of the treasury was set
aside. Some adjustment was also made in consideration of distance
from the markets.5
In Egypt, where the land tax was basically a fixed rate to be paid
in cash, Mahdi instructed his governor to double the rate. This in
itself gives us some idea of the rate of inflation during this period,
and also about the increased level of taxation. A tax was also intro-
duced on the cattle sold in the market places, but it seems probable
that this was limited to cattle brought to big cities for slaughter.6
We do not have any information about changes in the land-tax
in the other provinces but these were mostly areas that had had
treaties of capitulation with the Arab conquerors, and accordingly
had been paying an agreed sum. There was therefore no opportunity
of introducing any increase in taxes there. It is reported that Mahdi
ordered the imposition of a new tax on the markets and shops in
Baghdad, but as we know that this practice was started under
Mansur, it would seem that Mahdi only enforced the collection of
Qudama, Shemesh, p. 119; Bal., ¥utub, p. ijz.
Ammal, pp. 474-5-
Abu Yusuf, Kjldb al-Khardj, Cairo, A.it. 1502, pp. 44-;.
Jahshiyarl, p. 151; Hilal al-Jibi, Kitab al-Wuzara', ed. H. F. Amedroz, Leyden, 1904,
p. 341.
Qudama, Shemesh, p. 118.
MaqrizI, Kiifaf, vol. t, p. 308; Shaban, Islamic History, pp. 38-9.
https://doi.org/10.1017/CBO9780511622380.003
TOWARDS A CIVIL WAR 2J
these "rents" and extended it to newly built shops outside the
Round Qty. 1
In accordance with the principle of engaging the army only in
military activities, and as the internal situation was under control,
Mahdi decided that his army would be best employed against the
Byzantines. Beginning in 778/161, a series of increasingly aggressive
summer expeditions alarmed the enemy and the situation soon
deteriorated into an all-out war.2 In 780/163 Mahdi himself went
out with his army, including the Khuraslniyya, and established a
new base at Raqqa from which his son Rashid led a successful
expedition into Byzantine territory.3 Returning to Baghdad Mahdi
left his son to conduct the campaigns against the enemy and in
addition put him in charge of all the western provinces of the
empire.4 This can only mean that the revenues of these provinces
were assigned to the war effort on this important front. In 781/165
Rashid put an army of 95,793 men - apart from volunteers - in the
field. This force reached the sea coast opposite Constantinople and
compelled the Byzantines to sue for peace. A three year truce was
arranged, and the Byzantines agreed to pay a tribute of about
70,000 dinars twice a year. Interestingly they also agreed to provide
Rashid's army with guides and promised to set up markets en route
where provisions could be bought.5 Indeed this was one of the most
peculiar features of Byzantine-Muslim wars in 'Abbasid times; both
armies in their advance into each other's territories actually bought
their provisions from the enemy. One can only conclude that booty
was not a primary objective in these wars, or that because of the
strongholds on both sides of the frontiers, acquiring booty was so
difficult that it was only attempted in exceptional circumstances.
In the East there were minor uprisings among the Soghdians in
Bukhara and Kish and also in the Hephthalite Principalities around
Pushang and Guzgin. With a little help from the central government
forces these areas were soon pacified.6 Mahdi then turned to the
stubborn Caspian provinces. In 783/167 a good part of the army
stationed at Raqqa was dispatched to this region under the command
of the heir apparent, Hadi. This expedition forced at least one of the
1
Maqrizi, Kiifaf, vol. t, p. 105; Ya'qubi, Tarikb, vol. 11, p. $99; Yiqut, Buldan, vol. iv,
p. 448.
* T»bari, HI. pp. 485-6, 491, 495.
' Ibid,, pp. 494-9; Azdl, Mav/il, p. 245-6.
4
Tabarl, m, p. 500.
• Ibid., pp. 503-5; Azdl, Maw fit, p. 247.
• Jabari, in, pp. 470, 484, 494, 517; Ya'qubi, Tarikb, vol. 11, p. $97.
https://doi.org/10.1017/CBO9780511622380.003
26 TOWARDS A C I V I L WAR
princes of Tabaristan to declare his submission before Hadi had to

turn back at the news of his father's death.'
As an important result of these campaigns, all the recruits that
had come from Syria, Jazira and 'Iraq during Mansur's reign were
moved from Baghdad; some to the new base at Raqqa, and others to
settle in the strongholds on the Byzantine borders. There must have
been a great desire on the part of these men to settle there, for they
volunteered to go even though they were given only stipends and
no land allotments as had previously been the case.2 This sudden and
curious enthusiasm on the part of these men to accept transfer to
these remote parts of the empire is contrary to their expected
behaviour. Admittedly an incentive had been offered at the begin-
ning, but the movement continued after that was reduced and even
without it at all; and as early as Mansur we find people from the
tribes of Azd, Tayy a "d Hamdan of Basra voluntarily moving to
Adherbayjan.3 This uncontrolled immigration in areas that hitherto
had hardly been noticed by the Arabs, eventually caused a serious
revolt by the indigenous population under the leadership of Babak.4
Further west, within a few years, a major reorganization was to take
place on the Arab-Byzantine borders.5
On the southern side of the empire another development was
quietly taking place. The governor of Basra, a cousin of Mahdi, was,
in addition, given charge of all the territories bordering the Persian
Gulf.6 This meant that all the possible inlets for the Indian Ocean
trade were put under one authority in order to tighten government
control over this trade. It is interesting to note that the reaction of
the Arabs inhabiting the desert on the west coast of the Gulf, was
to start attacking passing caravans.7 This was only a portent of more
serious trouble in this area.8
Mahdi was succeeded by his son Hadi whose rule lasted only for
a little over a year 785-6/169-70, and was practically a continuation
of his father's. A precedent was set at the time of this succession
which was to become a rule at the succession of every new ruler.
Hadi was away in Gurgan when his father died, and it seems that he
was not very popular with some of the Khurasaniyya in Baghdad,
or, at least, it was feared that some might object to his succession.
It was quickly arranged that all members of the army in Baghdad
i, H I , p p . 4 9 5 , 5 1 7 - 1 8 , 5 2 1 - 2 , 551. * Bal., Futup, p. 166.

1
Ya'qubi, Tarikh, vol. ii, p. 571. * See below, pp. 34, j6.
» See below, pp. 29-50. * fabarl, HI, p. 501.
7
Athlr, Kami/, vol. vi, p. J I . * See below, p. 116.
https://doi.org/10.1017/CBO9780511622380.003
TOWARDS A CIVIL WAR If
should receive two years' stipend as an immediate grant and their
allegiance was thus secured.1 Another innovation at the time was
that Hadi separated what he considered as his private income from
that of the treasury. This was the income from the extensive con-
fiscated lands and properties of the Umayyads which he kept for
himself, in addition to the "rents" of shops and other similar
mustaghalldt that now included houses, public baths and grain mills.2
Two uprisings took place in the short reign of Hadi. A minor
Shi'ite revolt in Madina was very quickly suppressed and its leader
killed. The only notable result was that one of his supporters, his
cousin Idris,fledto North Africa where his son eventually established
the Idrisid dynasty.3 The other uprising, in Armenia, was more
serious. The continued influx of Arabs wishing to settle in the
thughur, border areas, met with resistance from the indigenous
population in alliance with the Arabs who were already settled there.
These disturbances encouraged the Khazars to resume their attacks
and the general situation there remained unsettled.4 Meanwhile, the
government was trying to persuade some of the newcomers to move
to other locations in order to relax the pressure on the more crowded
areas.5
It was now the turn of the celebrated Hariin al-Rashid who
succeeded his brother Hadi. In his reign 786-809/170-93 he followed
the same basic policies as his brother, father and grandfather, but
pressures were building up and strains began to show in the
whole structure. An explosion was imminent. A great deal of myth
surrounds this ruler, and modern scholars have preferred to leave
the mystery unsolved. One late noted scholar, in an attempt to explode
the myth, explained the whole thing in terms of the schizophrenia of
Rashid; fortunately this ingenious explanation was never published.
However, we must hazard a logical explanation, for without it later
developments can hardly be understood.
Like his father, Rashid came to power full of good intentions.
Gestures of reconciliation to the rival 'Alawids were freely offered
and gratefully received. As if to announce the good news he went
on a pilgrimage only a few months after his succession. He also
made a point of visiting Madina, and distributing a good deal of
money among the inhabitants of the two holy cities.6 Returning to
Baghdad he began to plan his government in the light of prevailing
1
Tat»«. "'> PP- 545-6- * H>id., P- 590.
» IbiJ., pp. 551-68. • Ya'qubi, Tdrikb, vol. H, pp. 416-8.
' Bal., Futib, p. 190. * Tahiti, HI, pp. 604-5.
https://doi.org/10.1017/CBO9780511622380.003
circumstances. The only troubled spots in the empire were the

regions of Armenia and Adherbayjan where internal conflict
encouraged enemy interference. On the Byzantine borders the truce
that had been concluded under Mahdi for three years had just come
to an end and the enemy were beginning to prepare for a possible
resumption of hostilities. Rashid, having been deeply involved in the
wars against the Byzantines during his father's time, and realizing
the dangerous situation on his right flank, decided upon his priori-
ties. He freed himself from all administrative responsibilities and
took personal charge of the army. For the administration of the
whole empire he fell back on his mentor and long-time associate
Yahya b. Khalid b. Barmak, a man whose loyalty, and that of his
family, to the 'Abbasids was absolutely beyond any shade of doubt.
As he and his father before him had had long experience in the
service of the 'Abbasids in various capacities, particularly on the
fiscal side, his ability was also proven. Rashid appointed him as his
waîr with full executive powers and, for seventeen years, this man
Yahya and all his sons, served Rashid faithfully in whatever assign-
ment he entrusted to them.1 Yet, as we shall see, when politics
required their elimination, they were brutally sacrificed.
When hostilities were resumed against the Byzantines, Rashid
satisfied himself for the four years 786-9/169-74 with defensive
action, in order to hold his territories.2 Meanwhile, a major re-
organization involving every aspect of the situation had been taking
place on these borders. It was realized that the advanced strongholds
in the tbughiir that were established, fortified and maintained at
great expense to the central treasury, were less than useless from a
military point of view. They had been fast losing any military
character that they might have had, or were intended to have had.
The enthusiasm of the men to settle there had an ulterior motive
which was far from that of fighting the enemy or defending their
own frontiers. They had become engaged in a most lucrative trade
with the Byzantines. Their proximity to the borders made these
tbugbiir ideal outlets and trade centres for both Islamic and Byzantine
products and any other merchandise that passed through their
respective territories. Indeed the exceptional status of these frontier
posts gave them almost complete immunity from taxes, a fact of
benefit to all concerned except the central treasury in Baghdad.
1
Ibs'J., pp. 545, 6oj, 606, 609, 651; Jahshiyarl, pp. 87-8, 177, 189-90, 204, 207, a i o - n ;
Ya'qCbl, Tdrlkh, p. 429.
* T»bari, m , pp. 521, j68, 605, 610.
https://doi.org/10.1017/CBO9780511622380.003
Furthermore, instead of obstructing enemy advances, they were
actually of considerable help to this very enemy. It will be recalled
that the early recruits to man these strongholds had been granted
land allotments as an incentive to settle there.' Although they were
supposed to pay the stipulated Muslim tithes, they more often than
not went tax-free. Naturally in their ample spare time they were able
to cultivate these lands. One can imagine that in these tax-free and
rather affluent localities prices would be low. Therefore any potential
outside market for the accumulating surplus would be highly desir-
able. The Byzantines provided an obvious solution. After crossing
Asia Minor and before entering Muslim lands, they could rely on
these strongholds to supply them with badly needed provisions. It
has been established that Muslim armies had to pay for their pro-
visions while passing through Byzantine territories even after the
conclusion of a rather successful campaign.2 Under the circum-
stances the Byzantines would be expected to pay the asking-price
for provisions, but that was not a difficult matter in view of the
services rendered. This situation was obviously dangerous and made
a complete mockery of the empire's system of defence against its
primary enemy. Rashid decided to dispense with all these advanced
posts on the whole northern frontier, and instead, established a new
line of defence to run along the southern slopes of the Taurus
mountains roughly on a line between Tarsus and Aleppo. Even the
term thughur was dropped from current usage and 'avdsim was
introduced instead. These were a chain of defensive positions
established along this line with the main base at the town of Manbij,
in charge of which he appointed a respectable 'Abbasid, 'Abdulmalik
b. Salih. Tarsus was refortified and a contingent of 3000 of the
Khurasaniyya was stationed there on a rotational basis. Another
2000 men disbanded from the thughur of Antioch and Massisa, who
had accepted to move to Tarsus, were given a rise of ten dinars in
their stipends in addition to land for their houses, but not agricultural
land. These arrangements were made for Tarsus because of its
exposed location.3 However no similar arrangements were made for
men stationed in other strongholds along the new defence line.
Presumably the new organization convinced most of the men in the
old thughur that their situation there was not tolerable and that their
1
See above, p. i j .
* See above, p. 25.
> Tabarl, m, p. 170; Azdl, Mavfil, p. z6a; Yaqflt, BuJJan, vol. iv, p. 165; BaJ., Futifi,
pp. 169-70; Mas'udI, Murij ol-Dbabab, ed. C Barbier de Meynard and P. de Courteille,
Paris, 1861-77, vol. VJ, p. 457.
https://doi.org/10.1017/CBO9780511622380.003
best course was to move voluntarily to the new 'awdsim. Of course,

as was to be expected, some refused to return and were given further
reason to complain when they were subsequently required to pay
their taxes in full. Their discontent added a new dimension to the
precarious situation in Armenia and Adherbayjan.'
In 790/174 'Abdulmalik b. Salih, the commander-in-chief of the
'cui'dsim^ had enough men to be able to resume and continue there-
after the customary summer expeditions from his base at Manbij.*
Needless to say, the trade that used to pass through the tbughur soon
found its way to the new 'awdfim, particularly to Aleppo, but the
government now had a tighter control over it.
To keep an eye on the Byzantine front and at the same time to be
close at hand if serious trouble broke out in Armenia and Adherbay-
jan, Rashid sought a central position in eastern Jazira. Although
thanks to the One Thousand And One Nights his name is eternally
linked with Baghdad, it is a curious fact that he scarcely ever lived
there. He had hardly come to power when he started looking for
another suitable place. After three moves in ten years he finally
settled in Raqqa which he made his headquarters.3 Of course the
Byzantine front had been on his mind, but he was also running
away from a problem which was becoming increasingly difficult -
the Khurasaniyya in Baghdad. Thefirstgeneration of these men had
justifiably taken great pride in their achievements for which they
were amply rewarded in every possible way. Their descendants had
inherited their positions and gradually increased their influence as
became clear at the succession of Had!. They were now reverentially
referred to as abnd' al-dawlajda'wa, the Sons of the Revolution, or
simply the abnd'. Of course they continued to draw their stipends but
they were also getting deeply involved in the commercial life of
Baghdad, utilizing their prestige to increase their wealth.4 As a
military forct they were hard to control and reluctant to forgo the
soft life of the city for the hardships of campaigning. It is not difficult
to imagine how frustrated a ruler would become in the face of such
an overbearing force to whom he and his ancestors owed their
very positions. Under the circumstances and until a substitute force
was constituted, Rashid preferred to stay away from his capital to
avoid any possible friction with the abnd'.
1
Ibn A'tham, vol. 11, f. 249B.
1
Tabari, HI, pp. 610, 612, 628, 629, 657, 645.
1
Ibid., pp. 606-7, 610, 64J-6; Azdi, Maw/il, p. 290.
4
T»ba«. m, pp. 826, 827, 829.
https://doi.org/10.1017/CBO9780511622380.003
'Abdulmalik b. Salih and his Syrian-'Iraqi forces were adequately

holding the Byzantine front. In spite of some differences between the
sitting population of Syria, particularly in Damascus, and some of
the men who had recently returned from the tbughur, which caused
some disturbance but which were soon ironed out, tht situation was
brought under control.1 However, the precarious conditions in
Armenia and Adherbayjan demanded that some action should be
taken; North Africa also was showing renewed signs of unrest. To
complicate matters, an 'Alawid who had taken refuge with the
Daylamites in the Caspian region began to agitate.2 Military forces
were needed to check all these threats but there were none available,
unless new sources of recruitment were to be found. The faithful
Barmakids were called upon to help, and Fadl was dispatched to
Khurasan for this important task.
Although our sources represent this operation as a military
expedition and Fadl was probably accompanied by some Khura-
saniyya, it was really more of a political mission whereby Fadl was
expected to utilize his contacts and those of his associates in order to
achieve his objective. They all knew the area very well, and they also
knew that the Hephthalites who, even before their conversion, had
co-operated with the Umayyads, were the best possible source of the
badly needed recruits. Now that Islam had struck roots amongst
them, it was probable that with some persuasion and incentive they
might provide an ideal solution. The incentive Fadl offered will be
explained presently, but the persuasion came from some of the more
co-operative princes of the Hephthalites themselves who were able
to convince their peers of the value of such co-operation. Fadl soon
succeeded in his mission and new sources of recruitment were
found not only among the Hephthalites, but also among the semi-
settled Soghdian population of Ushrusana. He called the new army
the 'Abbasiyya and grandly made them all mawdli of the 'Abbasid
ruler, thus giving a further new twist to the term mawld. We are told
that as many as 500,000 men were recruited, but actually only 20,000
of them went west with Fadl.3
Their arrival seems to have coincided with Rashld's move to
Raqqa in 796/180, for there was a noticeable upsurge of activity on
the borders. New strongholds were built and some men, apparently
new arrivals, were stationed there.4 Indeed in the following year
1
Ibid., pp. 624-5, 659; Athir, vol. vi, pp. 86-91, Ibn Hazm, Jamliara, p. 252.
1
Jabari, in, pp. 612-1 j .
5
Ibid., pp. 651, 6)4; Ya'qubi, Buldan, pp. 289-90. * Bal., Fuluh, p. 171.
https://doi.org/10.1017/CBO9780511622380.003
Rashid himself, as if to announce his newly acquired strength, led

the summer expedition deep into Byzantine territory.1 As new
troops continued to arrive, Rashid increased the pressure on the
Byzantines all through the following decade. This effort culminated
in a very big expedition in 805/190, again led by Rashid, but as it
had a limited success he was content to arrange for peace on favour-
able terms.2 As a result of this peace, and probably also because of
pressure from his Syrian-'Iraqi troops, Rashid allowed the re-
establishment of the discarded strongholds of the tbughur across the
Taurus mountains.3 Prisoners of war and gifts were exchanged
between the rulers. An unusual period of peace started on these
difficult borders and persisted even when a civil war broke out in the
Islamic empire.
While Rashid's attention had been so single-mindedly devoted to
the Byzantine problem, the affairs of the empire had been left in the
hands of the Barmakids. The father, Yahya, held the home front
while the sons, Fadl and Ja'far, were sent to deal with any troublesome
spots in the empire. Of course everything was done in the name of
the ruler, but the Barmakids themselves should be held responsible
for the success or failure of the policies they carried out. They seem
to have enjoyed very good public relations, because our sources
hardly voice any criticism of their actions even when these were
manifestly wrong. Naturally their major responsibility was the
fiscal administration of the empire which they and their appointees
in the provinces effectively controlled.
If the success of their fiscal policy is to be measured by the wealth
of the central treasury, then it was a huge success, but if it is measured
by its results and effects on the peoples of the empire, then it was a
dismal failure. It was certainly a strict conservative policy that never
took reform into consideration. Although some minor adjustments
were made as a result of complaints from some taxpayers their policy
as a whole was characterized by a harsh efficiency which had eyes
only for the interests of the treasury.
At an early stage of the Barmakid administration, it seems that
some zealous tax-collector had suggested that the converts of the
Sawad who, according to the new muqasama system, were paying as
much as half their produce in taxes, should also be subjected to the
Muslim tithes. As this amounted to double taxation and was indeed
a penalty for accepting Islam, it was quickly discarded.4 When the
1
Jibuti, «", P- 646. ' Ibid., pp. 708-11.
1 4
Ibid., p. 711. Ibid., p. 607.
https://doi.org/10.1017/CBO9780511622380.003
farmers of the Firs province complained that their tax rate was too
high compared to the more fertile Shiraz district of the same pro-
vince, their rate was reduced to two-thirds of that of Shiraz.1
If these measures seem lenient, the stringent methods of tax
collecting that accompanied them more than compensated for any
small loss the treasury might have suffered. Before the Barmakids'
time, it had been tacitly accepted that in a bad year the collection of
the full amount of taxes due should not be strictly enforced. There-
fore the taxpayer would remain "in debt" to the treasury for a small
portion of his tax, and in due course these arrears would be mag-
nanimously dropped at a suitable occasion. But that did not do for
Barmakid efficiency. Henceforth, not only had all taxes to be paid
in full regardless of any mitigating circumstances, but also all arrears
were vigorously claimed and special officials appointed for the
task.2 In some areas the taxpayers accepted the new rule with
resignation, but in others they protested and the outcome was
serious for the government.
In the region of Mawsil the situation became most serious, for
not only was the collection of arrears strictly enforced, but other
efficiency measures were also introduced. Taxes on livestock, which
had been in abeyance for all practical purposes, were demanded in
full, in addition to long-forgotten arrears.J Furthermore, where there
was some doubt as to whether a landowner should pay Muslim tithes
or muqdsama rate, he was asked to pay a lump sum equivalent to the
latter rate, which in effect meant that some of the Arabs there were
losing their acknowledged tax privilege.4 Consequently, it is not
surprising that Kharijite type revolts broke out all over the region.5
Attempts by the authorities to suppress these revolts by force only
worsened the situation. Some of the Arabs of Mawsil preferred to
leave the region altogether and moved to neighbouring Adherbayjan,
thus spreading dissent further afield.6
In Adherbayjan itself unrest was rampant as a result of the
pressure of earlier Arab immigration and the demobilization of the
thughur. Men who had been encouraged to settle in these frontier
areas and sometimes been given tax exemptions on their land allot-
1
Ijtakhri, KJldb al-Masalik ma al-Mamalik, ed. M. J. de Goeje, Leiden, 1870, pp. 157—8;
Maqdisi, Absan al-Taqdstm .. ., ed. M. J. de Gocje, Leiden 1877, p. 451.
» '(atari, l n > p . 649; Ya'qubt, Tarikb, p. 415; Azdi, Maw fit, p. 514; Athlr, vol. VI,
p. :e>5.
» Azdi, Mawsil, p. $14. * Ibid., pp. 27J-6.
» Tabarl, nr, pp. 631, 645, 649, 688, 711.
• A«di, Mawfil, pp. 279-81, 287; Athlr, vol. vi, p. IOJ.
https://doi.org/10.1017/CBO9780511622380.003
ments, suddenly lost this privilege and were asked to pay taxes.
Predictably they objected and resorted to armed revolt.1 The
ensuing instability in these exposed regions simply invited Khazar
attacks.2 The indigenous population, who had patiently tolerated
Arab encroachments on their territory and wealth, were certainly
alarmed by a further influx of Arabs. Nor were they reassured by the
presence of unruly elements in their midst who, instead of defending
them against Khazar attacks, actually brought closer this dreaded
danger. It is no wonder, therefore, that the so-called Khurramiyya
revolt, which broke out a few years later led by Babak, took root
at this time in the region.3 To cope with all this internal unrest and
the external threats the central government had to find troops to
contain the situation.
In Egypt efficiency measures were also responsible for unrest in a
province that hitherto had not suffered from any. Late in the Umayyad
period some Arabs from Mawsil had been settled in the area adjacent
to the desert on the east side of the Delta. Encouraged and sub-
sidized by the government, they had reclaimed lands from the
desert for agricultural purposes on which they had hardly paid any
taxes.4 Now, on instructions from the central government, these
lands were surveyed and assessed for tax purposes. Furthermore, the
measure used for surveying was a few inches shorter than the usual
length, and the assessments were made to favour the treasury. As
was to be expected, the result was armed revolt which could only be
suppressed by fresh troops sent from Baghdad.s Another measure
that antagonized Arabs and Copts alike was related to abbas, charit-
able and religious endowments: a purely Egyptian practice. The
Copts used to set up these endowments for the benefit of the Church
and the monasteries. As the management of such properties was left
in the hands of the benefactors or their agents there was much
leeway in the disposal of the income, in addition to the fact that they
were also exempted from taxes. The Muslims found this tax loophole
very convenient and had adopted the practice. Although the
Umayyads had tolerated this action, the 'Abbasids found it alien to
Islam and indeed Mahdi had tried to put an end to it. As these
endowments had to be authorized by a judge, Mahdi had appointed
a Hanafi judge to Egypt whose school of law did not permit the
Ibo A'tham, vol. u,ff.X48B-249B.
Tabari, in, p. 648; Azdi, Mavfil, p. 294.
Tabarl, m, p. 752; Azdi, Mavfil, p. J I J .
Shaban, Islamic History, p. 146.
MaqrizI, Kbifaf, voL 1, pp. 80, 508; Ukm, vol. 11, p. I 6 J ; Jtbttl, in, pp. 626-7, 629.
https://doi.org/10.1017/CBO9780511622380.003
principle of endowment. Since this move had met with objections,
the Barmakids decided to allow the practice to continue, provided
that such endowments were under the control of a judge who would
see to it that the income was actually spent on charity. For this
purpose a Maliki judge, whose school of law allowed the principle,
was appointed. •
Further west, in Barqa, Cyrenaica, where olives were an im-
portant product, a survey was made and a new assessment of
taxes on a lump sum basis was put into effect. Also the collection of
taxes on livestock was strictly enforced.1 In North Africa the
'Abbasids had been having difficulties holding on to Tunisia.
Differences between Berbers and Arabs and among the latter them-
selves were compounded by rivalries between the Arab-Berber
military forces recruited in the area and the Khurasaniyya brought
in on a rotational basis to control the region. Finally it was decided
that a large contingent of the new 'Abbasiyya troops would be
settled there permanently under the command of Ibrahim b. al-
Aghlab, who was also appointed governor of the region. Apparently
a deal was struck with him which gave him virtual autonomy in
exchange for an annual payment to the central treasury of 40,000
dinars. This net profit to the treasury was in contrast to the previous
arrangement whereby the treasury of Qayrawan had been receiving
an annual subsidy of 100,000 dinars from Egypt.1
The Barmakids' zeal in enriching the central treasury is, perhaps,
understandable, but it is difficult to understand their equal zeal to
enrich Rashid himself and the close members of his family. They
fervently used every means at their disposal, including corrupt
practices, with complete disregard for the interests of all concerned,
in order to achieve this unjustifiable objective. Shops, markets and
other mustaghalldt in many towns; pasture land under public owner-
ship; estates that belonged to rebels; lands that had been given in
allotments in frontier areas; deserted properties which the owners
had left because of the tax burden; properties whose owners died
without heirs; even confiscated Umayyad estates that had been
granted to members of the 'Abbasid family; all these properties
were appropriated and added to Rashid's private wealth.4 The name
1
Kindi, Govtrmrs and ]ud$u of Egypt, ed. R. Guest, Gibb Memorial Series, vol. xix,
London, 1912, pp. 371, 371, 583, 387, 395.
1
Ya'qubi, BuUdn, pp. 344-j.
5
Athir, vol. vi, pp. 105-8; Ya'qubi, Tdrtkb, vol. ir, p. 411.
i, 111, pp. 607, 749; Azdi, Maw/il, p. 287; Abu Yusuf, KbaraJ, pp 200-1; Bal.,
b, pp. 144, 151, 158, 179, 181.
https://doi.org/10.1017/CBO9780511622380.003
j6 TOWARDS A CIVIL WAR
of his son, who was a child then, was used in a tax-evasion device
to secure tithe status for lands on which magdsama'v/is due. The
owners of such lands formally sold their properties to this child but
continued to cultivate them, paying the required tithe to the treasury
in addition to a similar amount to the formal new owner. In these
cases the efficient Barmakid machinery would certainly help to make
the whole operation look legal. Needless to say, such lands eventually
ended up as the private property of Rashid's son, and the rightful
owners lost them for ever.' Estates that had been ceded to Marwan
II through the same device were now given to Rashid's daughters,
instead of being returned to their former owners. Zubayda, Rashid's
wife, had such vast interests in Egypt that an agent was appointed
to look after them.2 One has the feeling that the Barmakids did not
exclude themselves or their associates altogether from this wealth-
grabbing racket. Their life-style certainly betrays enormous wealth
and the everlasting allegiance they inspired among their entourage,
was due more to their celebrated "generosity" than to simple
loyalty.3
In contrast with their stringent fiscal policies over almost all the
empire, the Barmakids showed an unaccustomed leniency towards
the East; it was the only region where they made real tax concessions.
It will be remembered that Fadl went to Khurasan on a recruitment
mission and, significantly, he was successful. Of course this was
success at a price; he offered higher stipends to the new recruits and
their leaders, and when this was not enough all tax arrears were
forgiven, and the registers burnt.4 It was even suggested that, if
necessary, additional money would be provided from the central
treasury for this region, i.e. that the revenues of the East should be
spent there, and the central treasury should assume full responsibility
for all the expenses of the new army.5 It would seem, then, that the
Barmakids in order to ensure the regular supply of troops from the
East, made an agreement with the leaders of the area by which
these leaders secured financial and political concessions. As we know
that the Barmakids conceded autonomy to Ibrahim b. al-Aghlab in
Tunisia, there is no reason that they should not have conceded the
same to the indigenous leaders of the East, in return for their full
co-operation. As we shall sec, this very soon became formal policy
under Ma'mun.6
1
Bal., Futib, pp. )2j, 371. > Kindi, Governors, p. 392.
' T»bari, m , PP- 67}, 699-701. « Jahthiyiri, p. 191.
* Ibid., p. nt. * See below, p. 50.
https://doi.org/10.1017/CBO9780511622380.003
TOWARDS A CIVIL WAR )7
Meanwhile, this new eastern policy aroused tremendous opposi-

tion, particularly among the abnd' in Baghdad, an opposition that
finally brought about the downfall of the Barmakids. The abnd'
realized that the concessions accorded to the 'Abbasiya in the East
would eventually lead to concessions closer to home which would
probably be at their expense. They felt that the new forces would
ultimately ensure the erosion of their power in Baghdad and the
loss of their privileged position in the regime. Furthermore their
leaders, themselves from the East, had ideas as to how to deal with
their old neighbours without having to concede as much as the
Barmakids did. The successive indications of the failure of Barmakid
policies encouraged their opponents, and Rashid himself began to
lose confidence in them. Six years before their final downfall Yahya,
the senior member of the family, resigned. Although he was to be
in and out of office during this final period, his son Fadl was dis-
missed from office and obliged to travel to Raqqa to ask Rashid for
forgiveness; he was forgiven but was not returned to office.1 His
younger brother Ja'far was the only one who remained in office
until the very end. In other words, in the last six years of the
Barmakids' power, the number of offices they held was reduced to
a minimum, and other officials gradually took over their responsibili-
ties. One of these was Fadl b. al-Rabi', whose father was a mawld of
Mansur.2 It was around this figure that the opposition to the
Barmakids was built up, and in the light of his later behaviour, it is
clear that he was mainly opposed to their eastern policy. 'Ali b.
'Isa b. Mahan the leader of the abnd' in Baghdad was another figure
who was also opposed to this policy.3 Indeed, he was appointed as
governor of the East especially to put into effect a policy completely
opposite to that of the Barmakids. He tried to bring to heel the
princes and chiefs of the region, and to reimpose the full authority of
the central government on them. He also nullified any tax-conces-
sions and advantageous fiscal arrangements they might have gained
from the Barmakids. In contrast to the subsidy proposed by the
latter, we are told that he sent to the central treasury as much as 10
million dirhams a year.4 This new policy met with fierce resistance
and provoked numerous uprisings in the region.5 However, he
persevered with his harsh measures until a major revolt, led by one
Tabari, m, pp. 646-7, 651, 667; Jahshiyari, p. 227.
Tabari, lit, p. 638; Jahshyari, p. 269.
See below, p. 43.
Tabari, HI, pp. 649, 702, 716, 717, 727; Jahshiyari, p. 228.
b r i , HI, pp. 637, 645, 650-1; Ya'qubi, BUUUH, pp. 304-j.
https://doi.org/10.1017/CBO9780511622380.003
of his chief lieutenants, broke out in Samarqand. This leader was

Rafi' b. Layth who, interestingly enough, was a grandson of the last
Umayyad governor of Khurasan and the East. Although now a
staunch supporter of the 'Abbasids, he objected to the harsh treat-
ment that was dealt out to his fellow countrymen of the East with
whom he fully identified himself. He had the backing of other Arabs
of the region with similar convictions, and in addition enjoyed the
powerful support of the chiefs and princes of Soghdiana, Trans-
oxania and the Principalities of Tukharistan.' As this revolt spread
and threatened to end in the secession of the whole East, Rashid
hurriedly changed course. In an obvious move to placate the rebels
he sent back most of the recently recruited 'Abbasiyya whose main
task, significantly, was to arrest the unwanted governor 'All b.
'Isa b. Mahan.2 Of course they were also to try to persuade their
fellow countrymen to revert to peace, and it was hoped that they
would even fight against them if necessary. Any fighting that took
place however was perfunctory and the revolt continued unchecked.
Finally Rashid himself had to go east to try to contain this serious
revolt. As he did not have enough military forces, he planned to
seek the help of the princes of Tabaristan and Gurgan.3 However,
this hope did not materialize and he died as soon as he had reached
Khurasan. His death brought the crisis to a new climax and the
scene was fully set for the coming civil war.
In his last years, as he had been trying to strike a balance between
the various pressures weighing upon him and at the same time find
means of restoring stability to the empire, Rashid had unwittingly
helped the situation to deteriorate even faster. His overwhelming
preoccupation with the Byzantine wars was totally misguided
effort. There is very little evidence of any aggressive plans on the
part of the Byzantines; they were merely reacting to his own gran-
diose plans, and in fact were very willing to reach peace on any
reasonable terms. The problem was that his plans demanded more
military forces than he could muster. The abna" were not particularly
keen to join in what they may have considered an unnecessary war.
The Syrian-'Iraqi forces were in a demoralized state after their
withdrawal from profitable positions on the frontiers. To support
Rashid's efforts the Barmakids promoted their eastern policy in
order to recruit men from this vast region. However, the success of
1
Tabari, in, pp. 707-27; Ya'qubi, Tarikb, vol. 11, pp. 425, $35-6.
2
Tabari, HI, pp. 715-19; Ya'qubi, Tarikb, vol. 11, p. 425.
J
Tabari, in, pp. 705, 755.
https://doi.org/10.1017/CBO9780511622380.003
the scheme and the arrival of these men at the front antagonized
both the abnd' and the Syrian-'Iraqi forces. Even the commander-in-
chief 'Abdulmalik b. Salih showed enough opposition to force
Rashid to imprison him.'
The failure of the Barmakids' fiscal policies only added to the
need for new troops to cope with new situations. From the point
of view of everybody concerned the Barmakids were the culprits,
the villains of the piece. Although they were rapidly losing power,
that was not enough to placate their opponents who would not rest
as long as any Barmakid continued in office, and who focused their
rancour on the unfortunate Ja'far, the only one to remain in power
until the end. A scapegoat was required and in this particular case
Ja'far was the obvious target. When the moment of their downfall
came, it is significant that only Ja'far was executed. Furthermore,
and even more significantly, although he was executed near Kufa, his
body was sent to Baghdad where, on Rashid's specific instructions,
it was decapitated and cut in halves, one half hung on each of the
main bridges and the head put up in the city centre. The mutilated
corpse remained there for a year for everybody, particularly the
abnd\ to see as a proof of their ruler's repentance.2
Just before carrying out this hideous act, Rashid had been on the
pilgrimage to Makka where he took advantage of this most cele-
brated occasion to announce his plans for the succession. These
were not simply plans for nominating a successor. They included
a major re-structuring of the government of the empire, which
was to take effect immediately, during his lifetime. He seems
to have been thinking of a compromise that would satisfy his critics,
and at the same time keep the eastern policy alive albeit in a different
form. For all practical purposes the empire was to be divided into
two halves, the point of division being Rayy. While the western half
would eventually come under the control of Amin when the time
came to succeed his father, the eastern half would go to Ma'mun,
the next in line for the succession. A most important stipulation was
that the revenues and forces of each domain would normally be
used only for the benefit of its respective regions. Ma'mun would
have complete autonomy and the final say on fiscal matters in his
area. However, he was required to help his brother with military
forces if the need arose. A most binding agreement was con-
cluded, sworn and witnessed by hundreds of dignitaries, members
1
Ibid., pp. 688-94.
1
Ibid., pp. 675-80; Jahshiyari, p. 254; Azdi, Maw/il, p. 304.
https://doi.org/10.1017/CBO9780511622380.003
4<J TOWARDS A CIVIL WAR
of the 'Abbasid family, mawdli, government officials and army

generals. The only conspicuously missing witnesses were the
Barmakids, although they were on the same pilgrimage.'
Like all agreements this one could not foresee every circumstance,
and of course the parties concerned might put different interpreta-
tions on the same articles. Very soon it became clear that by dividing
the empire, Rashid had actually helped to set the opposing parties
against one another, and had provided them with sufficient resources
to become independent of each other. Furthermore, he was giving the
East a legitimate ruler around whom its forces could rally. Indeed
when Rashid died, RirV b. Layth saw no reason to carry on with his
revolt, and wholeheartedly joined Ma'mun's camp.2 This polariza-
tion in the East was matched by a similar process in the western
regions. 'Abdulmalik b. Salih was released from prison to help
muster his Syrian-'Iraqi troops, and 'AH b. 'Isa b. Mahan began to
mobilize his alma' for the expected fight. The inevitable confrontation
was to take place in a matter of a few months.
1
Tabari, in, pp. 655-65; Ya'qubi, Tarikh, vol. 11, pp. 416-20.
• jibiti, in, p. 777.
https://doi.org/10.1017/CBO9780511622380.003
DISSOLUTION UNDER A
NEW REGIME
In previous civil wars fighting had been limited to protagonists
whose interests had been directly involved, and on the whole other
sections of the population had chosen to remain out of the struggle,
even when their sympathies had been with one party or the other.
In this civil war, although the opponents were clearly identified,
almost every section of the population in every region became
involved and the prolonged struggle spread to all corners of the
empire. The abnd' were fighting a determined battle to conserve
their declining status in the 'Abbasid regime, while the Easterners
were struggling to consolidate the privileged new position created
by their active participation in support of the wider interests of the
empire. As Muslims and members of the army they took strong
exception to being pushed aside by the abnd', especially when these
were not fulfilling their military responsibilities. The latter were
determinedly opposed to any change which would give the Eastern-
ers a privileged place in the structure of the empire. Every group
in the empire with vested interests that were threatened by change
actively supported the abtid', while every group that aspired to
change sided with the Easterners. This situation created two
opposing blocs, one reactionary, symbolised by the abnd'; the
other progressive and represented by the Easterners. The problem
was that while the reactionaries knew exactly what they wanted, the
progressives had nothing in common beyond their desire to see the
abnd' brought to their knees. Once this objective was achieved their
apparently united front fell into disarray, and more and more new
disorders ensued. At the same time the abnd' became divided among
themselves as to how far they should accept their new situation.
Amin, Rashid's successor, was the leader of the reactionary forces,
while Ma'mun was the hero of the progressives. Perhaps not
surprisingly he had to contend with problems more serious and
complicated than those that faced his less fortunate brother.
When Amin came to power the treasury had more than ample
reserves and he had no problem paying his army the customary
[41 1
https://doi.org/10.1017/CBO9780511622380.004
42 DISSOLUTION UNDER A NEW REGIME
bonus.' At the same time the Byzantine borders were as tranquil as
they could be and there were no signs of disturbance there. Yet
Amin demanded from his brother financial or military help, and
according to some reports both. It can only be concluded that this
was a deliberate attempt on his part to invalidate the agreements
arranged by Rashid. It can also be gleaned that this was done under
pressure from the abna* and their allies in Baghdad, who had been
encouraged by their success against the Barmakids and the death of
Rashid.2 They now went all the way to restore the status quo and
erase any possible traces of the eastern policy.
Predictably Ma'mun objected, realizing that his brother's demands
were only a pretext for the more serious intentions of Baghdad.
Although Rafi' b. Layth had declared his allegiance to him, his
allies the princes of Soghdiana and the Principalities of Tukharistan
•were undecided as to their course of action.3 But Ma'mun was
fortunate enough to have in his service the two Sahl brothers, Fadl
and Hasan. Like many other senior officials in the empire, they were
proteges of the Barmakids who, in their long tenure in office, had
trained many such men. Naturally such men were not adherents of
all the Barmakid policies, but as these two brothers had been chosen
to serve Ma'mun it was to be expected that they were strong suppor-
ters of the eastern policy.* It has been wrongly believed that they
were originally from 'Iraq.5 This mistake stems from the misunder-
standing of a remark in a source that relates them to the district of
Sib where Fadl was granted an estate by Ma'mun. It also arises
because Hasan eventually made his home in 'Iraq where he was
practically exiled after his dismissal from office.6 However another
source clearly states that they came originally from the town of
Sarakhs in Khurasan.7 This is confirmed by the fact that their
cousins, of whom they had many in their service, were from
Khurasan.8 The brothers Sahl and their cousins formed the nucleus
of Ma'mGn's nascent administration in Merv. Fadl, the head of this
Tabari, HI, pp. 764-5.
Ibid., pp. 779, 784, 811-17; Jahshiyari, p. 289; Ya'qubi, Button, p. 305.
Tabari, m , p. 815.
Ibid., p. 7JO.
Dominique Sourdel, Lt Viîrat AbbaiitU, Damascus, 19)9, vol. i, p. 196.
Jahshiyari, pp. 229-)!, 506; Ya'qubi, BuM&t, p. )22.
Ibn Khalliltan, Wafajat al-A'yan, ed. Ihsan 'Abbas, Beirut, 1972, vol. iv, p. 41.
Jabari, in, pp. 97), 979, 985, 99), 996, IOOO, i o n , 1026, 1027; Jahshiyari, pp. 505,
J i 8 ; Ya'qubi, Buldan, pp. 307, $21, 406; Ya'qubi, Tdrikb, vol. n, pp. 448, 452;
Athlr, vol. VI, pp. 209, 215, 218, 220, 225, 246, 256; Ibn al-Abbar, Vtab al-Kultab, ed.
§. al-Ashtar, Damascus, 1961, p. 109.
https://doi.org/10.1017/CBO9780511622380.004
DISSOLUTION UNDER A NEW REGIME 43
administration, was significantly given the title dim al-riyasatayn

al-harb wa al-tadbir, i.e. indicating that he was in charge of the army
and of the fiscal administration.' In other words he was given full
powers by Ma'mun in order to muster badly needed support at this
critical point. On the fiscal side the taxes of all the region were
reduced by one quarter.2 With this kind of incentive and the wide
contacts of the Sahls and their cousins, the mobilization of the East
became feasible. They did not, however, rely on the 'Abbasiya
leaders amongst whom Harthama b. A'yan of Balkh was the most
prominent figure. Instead they cultivated their own new leaders
directly from the Hephthalite chiefs such as Tahir b. Husayn of
Pushang, whose own men were as many as five thousand. His
ancestors had supported and served the 'Abbasid cause since the
Revolution, and in the process had established useful connexions all
over the region. Although his own father advised him against the
venture he, probably after extracting more concessions from the
Sahls, decided to accept the assignment and to lead the Eastern
forces against the intransigence of Baghdad.3
Meanwhile 'Ali b. 'Isl b. Mahan mobilized an army of 20,000 men
from the abna" and marched to Rayy where the two armies met. The
result was a resounding defeat for the abna'; 'Ali himself was killed
and his army retreated in complete disarray to Baghdad.4 In pursuit
of his fleeing adversary Tahir advanced to the city where he was met
by unexpectedly fierce resistance. It took reinforcements from the
East, fourteen months of siege and practically street to street
fighting to capture a city that was largely destroyed. This obdurate
resistance was partly due to the last ditch fight of the abna\ but also
to the population of Baghdad itself which took a major part in the
fighting.5
Realizing Tahir's difficulties Ma'mun dispatched a strong con-
tingent of the 'Abbasiya under the command of their leader Harthama
b. A'yan. As he had taken part in previous operations in the west
under Rashid, it was hoped that Harthama's presence in 'Iraq might
attract some of the Syrian-'Iraqi forces to his side. The death of
their leader 'Abdulmalik b. Salih at that time must have helped a
little in this respect, in addition to the fact that some of them must
1
Jahshiyiri, pp. 30J-6; Tabari, in. p. 841.
» Tabari, in, p. 774; Azdi, Mawfil, p. 518.
' jabari, m , pp. 771, 92j, IO6J; Jahshiyiri, pp. 84, 290; Athir, vol. vi, pp. 29,104;
Ya'qubi, BUUOK, p. 306; Ya'qubi, Tirikb, vol. 11, p. 437; Ibn Jjtem, Jambara, p. 194;
Tayfur, pp. i i ) , 119, 129.
• j»bati, in, pp. 797-803, 820-30. » ttU., pp. 868-933.
https://doi.org/10.1017/CBO9780511622380.004
also have seen the writing on the wall. When Harthama arrived at
Baghdad, during the siege, a sizeable force of the Syrian-'Iraqi
troops also reinforced Tahir's side.1 Ma'mun's strategy was that
after the fall of Baghdad Tahir should proceed to capture the strong
base at Raqqa, while Harthama should advance to the south to
control the rest of 'Iraq. Meanwhile another army was sent under
the command of Hasan b. Sahl who was to deputize for his brother
Fadl in Baghdad and help to re-establish civil authority there.2 But
these arrangements did not work out quite as had been planned. In
the first place Tahir and Harthama seem to have been uneasy
partners even during the difficult period of the siege. Harthama
seemed to have wanted some measure of reconciliation with the
vanquished, while Tahir and Hasan were definitely against such a
policy. Indeed the unfortunate Amin sensed this attitude on the part
of Harthama, and tried to take advantage of it to save his own life
when everything else was lost. In the event he did not succeed and
was killed by one of Tahir's dose associates. When Tahir marched
to Raqqa, Harthama was recalled to Merv where he was imprisoned
and died.3 In the second place, when Hasan was proceeding to
establish his own authority in Baghdad, he realized that the first step
had to be the total demobilization of the abnd'.* This move created
a sharp division among them; some were willing to accept it while
others were determined to keep their positions. Nevertheless both
sides were adamant in their opposition to Hasan and what he stood
for, and once more fighting broke out between all parties.5 This
situation was not helped by the fact that quasi-Shi'ite uprisings
started in the vicinity of Kufa, Makka and Yaman. Although these
movements were a portent of more serious trouble later, at this
time they came to nothing, and after some time peace was restored
in these areas.6 However, in Baghdad chaos prevailed; the alma'
fought each other and at the same time they all fought the army of
Hasan which in itself was not united. The Syrian-'Iraqi troops that
remained with Hasan after Tahir had left to go to Raqqa, began to
waver as they saw no end to this interminable fighting.7
After the death of 'Ali b. 'Isa b. Mahan, the leadership of the
Ibid., pp. J I , 68, 846. » Ibid., p. 975-
lbid., pp. 903-50, 998; Ya'qubi, BulddH, p. 306; Ya'qubi, Tdrikb, vol. II, pp. 449-50;
Azdi, Mawfil, p. 541.
Tabarl, HI, pp. 998-9. > Ibii., pp. 999-1005.
lbii., pp. 976-9J, 1000-3, I O 1 7. 1019-xi, 1062-5, I09**> " 0 0 .
/ * / / , pp. 998-9, 1001-3; Ya'qubi, T&tkb, vol. 11, p. 450; Azdl, Mawfil, p. 341;
Tayfur, pp. 1, 14*.
https://doi.org/10.1017/CBO9780511622380.004
D I S S O L U T I O N U N D E R A NEW REGIME 45
abnd' had fallen to Muhammad b. Abi Khalid, who in turn was

killed in the new fighting. His brothers and sons continued in this
role but they changed sides so often that it is hard to keep track of
their activities.' Finally, as a device to get rid of Hasan, a reluctant
'Abbasid, Ibrahim b. Mahdi, was proclaimed the representative of
Ma'mun in Baghdad. Then, almost against his own will, it was
decided that he should be set up as a rival Amir al-Mu'minin there.2
Most important in this situation was the reappearance of the popula-
tion of Baghdad as a strong force in the fighting. Some were on the
side of the abnd' who objected to demobilization, others supported
those who accepted this measure. The latter were particularly
powerful on the east side of Baghdad where they seem to have
had extensive commercial interests. Here, the merchants played a
big part in arousing the population against the intrusion of Hasan,
and were even willing to pay for the recruitment of the mass of the
population as a kind of Home Guard to support the rival rule of
Ibrahim.3 Earlier, during the siege of Baghdad, Amin, after exhaust-
ing his financial resources and in an attempt to raise forces from the
same source, had had to find the means to pay them. Under the cir-
cumstances of the siege when food prices had rocketed and when
cash had not been available, paying these recruits in kind had been
the only possible way and indeed the best incentive.4 This same
method was used again to raise support for Ibrahim, and in fact
these "volunteers" proved to be very effective in the type of urban
guerilla warfare that took place in the city. These men were paid in
kind in certain measures and this practice gave rise to a new ver-
nacular term, 'ayydr, which is a derivative of 'tydr, i.e. measure. At
the time this was a rather laudatory term but eventually with the
deterioration of the uncontrolled activities of such men into brigand-
age, the term acquired a derogatory connotation and became
equivalent to " thug" or "rogue".* A similar process happened with
the term used to describe the men of the Syrian-'Iraqi forces,
%awdqil, which at first simply meant men who tied their turbans in a
special way leaving the edges around their heads. When these men
T i , H«, PP- S43, 883, 9)5, 978, 998, 999, 1000, 1003, >°°4. i°°7. 'S°*; Ya'qubi,
Tarikb, vol. 11, pp. 439, 447; Jahshiyarf, pp. 196, 312; Athir, vol. V', pp. 225, 227.
* Tabarl, in, pp. 1005-16, 1012-16, 1030-4, 1068.
1
Ibid., pp. 1008-11, 1016, 1023, 1032-5.
• Ibid., pp. 872, 881, 882, 890, 896, 899, 900; Mas'udi, Murij, voL VI, pp. 452, 461,
463, 465-
» Tabari, ««. PP- «««. *8j-6. 8 94. «9 6 . 899» 9°'-*. 9°7. 1008; Mas'udi, Mmrig, vol. vi,
p. 463; Athir, vol. vi, p. 197.
https://doi.org/10.1017/CBO9780511622380.004
did not come to the aid of the abnd\ the term %ao>dqil also acquired a
derogatory implication.1
The co-operation of the wealthy merchants with the abnd' to
raise such a force, was a clear indication of the determination of these
two groups to restore the status quo at any cost. Indeed the slogan of
"commanding good and prohibiting evil" - which has always been
used by reactionary groups in Islam to impose their own dogmatic
viewpoints - betrays their true intentions.2 The fact that this was the
time of the rise in Baghdad of Hanbalism, the most conservative
school in Sunni Islam, emphatically underlines the nature of the
prevailing atmosphere there. It is important to realize that the
various schools of law in Islam, and for this matter the different
sects, were not merely concerned with ritual and rites, but even more
with taxes which were usually referred to by the Islamic term yakat.
In this respect the Hanbalites represented the extreme right wing
that tried to perpetuate the inequitable taxation system which
allowed the majority of the urban communities, particularly the
wealthy merchants, to avoid paying any taxes. In clear contrast the
Shi'ites, the traditional opposition party, began to comprehend the
significance of the change in the economy and tried to adapt their
movement to fulfil the aspirations of the tax-oppressed communities.
Henceforth, Shi'ites became keen advocates of fair taxation and
indeed where they succeeded in establishing their own regimes,
there were always substantial and significant new fiscal arrange-
ments.3 In Baghdad, later developments show that while the small
shop-keepers, who had to pay the treasury " rents" for their shops,
turned to Shi'ism, the wealthy merchants continued to be avid
Hanbalites.4
The unabated fighting in Baghdad finally forced Hasan b. Sahl
to withdraw his army from the city, and Ma'mun had to take some
action to save the situation. He hurriedly announced the appoint-
ment of a respectable 'Alawid as his heir apparent, and for himself he
took the title of Imam.* This was a desperate tactic to appeal to the
Shi'ites in the empire, but it was too transparent to convince anybody
of the sincerity of Ma'mun's motives. In the event it only added to
the highly charged atmosphere of Baghdad and even strengthened
the determination of his opponents. Ma'mun, who must have been
1
T»bari, »i. PP- 84}, 844, 845, 847; Athir, vol. vt, pp. 178, 179; Ibn Man?ur, LJsdn
at-'Arab, Bulaq, 1891, s.v. %(jl, * Tabarl, m, p. 1009.
4
* See below, pp. 130, 148, 166. See below, p. 83.
» Tabari, m, pp. 779, 796, 1012-13, Ya'qubi, Tarikb, vol. 11, p. 450.
https://doi.org/10.1017/CBO9780511622380.004
one of the most supple politicians in Islamic history, made a complete
about-turn in his proclaimed policies. In the meantime he took the
most daring and risky step of his career; he decided to go to Baghdad
himself. He proceeded from Merv with such extreme caution that it
took him two years to reach Baghdad. Soon after he left Merv Fadl
b. Sahl was assassinated, almost certainly on Ma'mun's orders.1
Hasan b. Sahl, who was still in his camp to the south of Baghdad,
was declared insane.2 Most conveniently, the 'Alawid heir apparent
died in circumstances which can only be described as suspicious.3
Tahir and his forces were ordered to leave their unfinished task in
Raqqa and march on Baghdad; their arrival was to coincide with
Ma'mun's.4
Once he entered the city 819/204 everything there returned to
normal, as if the events of the previous decade had not occurred. All
symbols of his short-lived Shi'ite conversion were cast aside,5 with
one exception; Ma'mun held on to the title of Imam which continued
to be used by all the later 'Abbasid rulers. Another addition to the
lengthening list of his official titles was that oi khalifa which had long
been in popular use, but was now given a twist to signify that the
ruler was God's deputy on earth, instead of the simple earlier mean-
ing of successor.6 This was another attempt to consolidate the
authority of the Amir al-Mu'minin by giving it greater semblance of a
religious function.
It took some effort to bring the Syrian-'Iraqi forces of Raqqa
and the opportunists of the thughiir under control, but soon the
general situation in 'Iraq, Syria and Jazira looked promising,7
though troubles started in the other regions. The people of the East
who had stood by Ma'mun in his hour of need felt betrayed by his
sudden reconciliation with Baghdad, and began to show signs of
unrest.8 Tahir who had been given the responsibility of maintaining
order in Syria, Jazira, 'Iraq and Baghdad itself was the only obvious
person to bring stability to his own region.9 Accordingly a new
arrangement was introduced, one that in fact amounted to the
establishment of a new regime for the empire. The fact that Tatar's
Tabari, m, p. 1027.
Athir, vol. vi, p. 252; Tabari, in, pp. 1050, 1081-5. ' Tabari, in> p. IOJO.
Ibid., p. 1037. * Ibid., p. 1038.
Ibid., pp. 1097, 1112, 1117.
Ibid., pp. 1045-62, 1067-73; Tayfur, pp. 138-9; Ya'qObi, T**U, vol. II, pp. 4 j ; , 4 5 j ;
Hal., Fuluh, p. 185.
Ya'qubl, Tarikb, vol. n, p. 450; Tabari, in, p. 1043.
Tabari, in, p. 1039; Ya'qubl, Tarikb, vol. 11, pp. 4JJ-6; T«yfi^. P- *}•
https://doi.org/10.1017/CBO9780511622380.004
S Y R I A N - ' I R A Q
D E S E R T
x 'Iraq, Jazira, Syria and thugbur

Downloaded from https://www.cambridge.org/core. Cambridge University Main, on 26 Apr 2018 at 05:48:25, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511622380.004
JO DISSOLUTION UNDER A NEW REGIME
troops were the only force that could be relied on to control Baghdad,
and therefore had to stay there, provided the essence of the solution.
The governorship of the East was added to Tahir's responsibilities.
He himself would make his headquarters in Khurasan, while his
numerous sons, nephews and cousins would deputize for him in his
various other functions. His army would stay in Baghdad to police
the city and the Sawad, and to cope with any other disturbance
elsewhere.' In his home area Tahir should not have much difficulty
raising new forces for himself and for Ma'mun when necessary. The
price that he exacted for his services was considerable. In terms of
money, it was a subsidy of 10 million dirhams from the central
treasury.2 In political terms the price was no less than complete
autonomy for the whole region under Tahir and his descendants.
This bargain was struck through the good offices of Ahmad b. Abi
Khalid who had emerged as the strong arm of Ma'mun from the
moment he had decided to move to Baghdad.3 Although he con-
tinued in the service of Ma'mun until he died in 826/211, he actually
refused to accept a formal appointment as wa^tr or kdtib* Yet he
was deeply involved in the management of Ma'mun's government,
especially in the affairs of the East. He had considerable influence
with the Tahirids and influence almost equal to theirs in their own
region, where he was able to recruit the redoubtable Afshin of
UshrQsana and his men into the army.5 Ma'mun's choice of this
man, Ahmad, at that particular turn of his policy had not been with-
out reason. He was a member of the family of Abu Khalid, the
leaders of the abnd\ who had continued to resist Hasan b. Sahl's
policy in Baghdad.6 While other members of the same family had
taken and indeed changed sides in this struggle, Ahmad had chosen
to remain loyal to Ma'mun. However, being so closely connected
with the abnd', he was an ideal instrument for reaching an agreement
with them. Accordingly by choosing him as a close associate, Ma'mun
was practically announcing his willingness to reconcile this obdurate
group. Ahmad must have been a most valuable aid at the critical
moment of Ma'mun's arrival at Baghdad, and had probably arranged
for his own family's reconciliation with Ma'mun, in the same way
Tab"!, in, pp. 1059, 1045, 1044, 1102; Tayfur, p. 232.

Tabari, in, p. 104$; T»yfi»r, p. J2; Ya'qubl, BuU&t, p. $08.
Tayfur, PP- 5, J i - j ; Jahshiyiri, p. 318; Jabarl, HI, 1038.
Tayfur, p. 216.
jabari, m, pp. 1044, 1064-6; Bal., F«f«£, p. 430; Ya'qubl, Tartkb, vol. 11, p. 457.
Azdi, Mawfil, p. 341. 384; Ya'qubi, Tdrtkb, vol. 11, pp. 439, 447; Tayfur, p. 142;
Jahshiyiri, pp. joz, 312; Tabari, Hi, pp. 843, 883, 978, 1001, 1002, 1003, 1032.
https://doi.org/10.1017/CBO9780511622380.004
as he was to arrange for the setting up of Tahir in the East. Originally
Ahmad's family came from the district of Merv al-Rud, and it is
likely that, like the Tahirids, they were of Hephthalite origin. The
father Abu Khalid, whose personal name was Yazid, became a high
official in the administration of Hadi and Mahdi. It should be noted
that his service was, at least in part, associated with army activities
in the eastern regions.1 Again this would indicate where his useful-
ness lay and would suggest that he did not sever his connection
with the army. Indeed the emergence of his sons as leaders of the
abnd' is a proof that they had their interests at heart.
Perhaps it is worth pausing, at this point, to consider another,
different view of Ahmad b. Abi Khalid, in order that the reader may
get some idea of the problems involved in interpreting the source
material. In his most exhaustive study of one aspect of this period,
Sourdel has certainly collected all the available data. But it has to be
said that some of his footnotes are superfluous, and that his under-
standing of the source material is often wrong.2 It is not enough to
lump together all the information collected from the sources with-
out attempting a critical evaluation of it. In this case Sourdel, relying
on a badly edited text as a source, accepts a copyist's mistake as a fact
and thus asserts that Ahmad was of Syrian origin.3 This simple error
makes him miss the point altogether, misleading us in the process.
The author of this particular text was most hostile towards Ahmad,
and has nothing good to say about him. Furthermore, he accuses him
of every possible vice including gluttony, homosexuality, ugliness,
bad manners and even bad taste in food.4 Carried away by this rabid
hostility he also tries to cast doubt on his lineage. In this respect the
editor reads the sentence concerned as, "wa kana shdmiyjan, mawlan
li-Bani'Amir b. Lu'ajy".* The reading shdmiyyan, i.e. Syrian,is a mistake
of the editor or a copyist. Ibn Khallikan draws our attention to this
common mistake of copyists, and points out that this word should
read samiyjan, i.e. from the clan of Sama.6 The construction of the
sentence also confirms that the reading and the meaning should be,
" he was of the clan of Sama as a client of the Banu 'Amir b. Lu'ayy ".
The only Banu 'Amir b. Lu'ayy we know was a branch of Quraysh,
but our prejudiced author would not want to credit Ahmad with
such a distinguished relationship. Therefore he deliberately relates
1
Tabarl, m, pp. 21, 569, 455, 520; Athir, vol. vi, p. 51; Jahshiyari, pp. i8}-6; Ya'qubi,
Bsddan, p. 247.
* Sourdel, Viîrat, p. 218, e.g. footnote 3, especially the reference to Jihi?.
4
* Ibid., p. 219. Tayfur, pp. 220-8.
* Ibid., p. 216. * Ibn Khallikan, vol. m, p. 3)7.
https://doi.org/10.1017/CBO9780511622380.004
him to the clan of Sima whose claims that they belonged to Quraysh
through their ancestor Lu'ayy were disputable.' In other words this
author was indirectly accusing Ahmad of a false claim to a good
ancestral line.
Ahmad's arrangements for Tahir and the East proved to be
successful and the stability of the region was assured. Tahir stayed
in his position for two years until his death. Although there are
reports that he had dropped Ma'mun's name from the Friday sermon,
such reports should not be taken seriously for two reasons.2 Firstly
Tahir, who had been strongly opposed to Ma'mun's Shi'ite gestures,
may have had some objections or apprehensions about the continued
assumption of the title Imam, which would have had to be men-
tioned in the sermon. Secondly, Ma'mun himself did not pay any
attention to these reports and dutifully on the advice of Ahmad
confirmed the appointment of Tahir's son to succeed his father in
the same position and with the same arrangements.3
Ma'mun, then, turned to other pressing problems, the first of
which was to bring into line the Syrian-'IraqI troops who, in spite of
the capture of their base at Raqqa, were now taking refuge in the
strongholds of the thughur. It took the Tahirid forces of Baghdad
almost five years to restore the Government's authority in these
areas and destroy Kaysum, the stronghold of their most prominent
leader Nasr b. Shabath of whose clan 'Uqayl we will hear again.4
Once this was done the same forces were dispatched to Egypt, where
for the first time revolts had seriously shaken the government's
control. Since the last years of Rashid's reign the Arabs of the
Delta had stopped paying the newly imposed taxes. Amin had been
able to spare only a force of iooo men from the abnd' to help contain
the rebellion, but the civil war had put an end to all such efforts.5
To complicate the situation the army in Egypt had become divided
against itself and had carried on its own parallel civil war in the
province. After his victory Ma'mun had made all sorts of attempts
to pacify the country, including dividing it between the rival parties
there, but to no avail. Finally the Tahirid army was able to impose
law and order.* However, as nothing was done to reduce the
objectionable taxes, the revolt started again almost immediately
Ibn iiaxm, Jambara, pp. 12, 13, 166-72, 17$.
T»b«ri, i n , p. 1064; T«y<ur, pp. " 7 . 1 J°-i-
T>b*rf» ni, pp. 'o6j-6; T»yfii*, pp. j , 151,13*. i$4-
Tabui, m, pp. 1067-72; Athlr, YOI. VI, p. 198; Azdl, Mawfil, p. 366.
MaqrizI, KHfaf, rol. I, pp. 80, $10.
Ibid., pp. 172-3, 178-80, 310; Tabarl, m, pp. 1086-94.
https://doi.org/10.1017/CBO9780511622380.004
DISSOLUTION UNDER A NEW REGIME 5J
after the army's departure for Baghdad. This time it was even more
serious because the indigenous population gradually joined the
Arabs in their resistance and the whole country became enveloped
in the struggle.1 It took some time before the wrath of the central
government was turned against this unexpected resumption of the
revolt, because Ma'mun was busy with plans for reorganizing his
armies.
Perhaps the most important decision was to integrate all his
military forces in a homogeneous army which would be composed
of divisions to be deployed in accordance with the expected fields of
action. There would be three divisions and each would include
contingents from the abtia', the Syrian-'Iraqi troops and the new
troops from the East. The outcome of this plausible arrangement
would be a standing army balanced enough to facilitate its control,
and at the same time effective enough to diminish the likelihood of
any particular group being able to establish its special interests in
the areas of operation. The ultimate control of the army remained in
Ma'mun's hands, but of the three divisions one was under the
command of the Tahirid military governor of Baghdad; its function
was to maintain law and order in that city, the Sawad and the nearby
districts of southern 'Iraq and the Firs province. In addition it
would be held in reserve for any emergencies that might arise any-
where in the western regions of the empire. The other two divisions
were to be under the command of Ma'mun's son 'Abbas and his
brother, Mu'tasim, the heir apparent. The former was put in charge
of Jazira, the tkughwr and the 'awdsim, i.e. he was given the main
responsibility for the Byzantine borders and held in reserve for any
trouble in Adherbayjln. The latter was in charge of Egypt and
Syria which would also mean that he was available for help on the
Byzantine front.2 For his administration Ma'mun employed men
most of whom had been trained under the Barmakids. It is significant
that thejr appointments were merely as heads of departments
directly responsible to him. In other words he meant to keep, as
much as possible, the control of the army and the administration in
his own hands without the mediation of an overall factotum like
Fadl b. Sahl.
In the midst of all these arrangements Ma'mun made another
about-turn. We remember that he had taken the title of Imam, to
appeal to the Shi'itcs, and though it had not had much effect he had
continued to assume it. Now he proposed to take this function
1
See below, p. 60. " Tabarl, m, p. 1100.
https://doi.org/10.1017/CBO9780511622380.004
seriously and guide the whole community to what he considered
the right path; and if in the process he could effect a compromise
between the opposing blocs in the empire, so much the better. The
weakness of his previous attempt to reconcile the Shi'ites had been
its lack of any substantial new ideological proposition at a time when
every school of law, movement or sect was in the process of formulat-
ing its answers, not only to theological questions but also to the
pressing social and political problems of the time. The Hanbalites
had been steadily consolidating their position as the champions of
orthodox Islam, and their notion of the paramount nature of
revelation was being articulated to support their traditionalist inter-
pretation of Islam. In contrast, the Mu'tazilites, hitherto a relatively
obscure intellectual group, whose main concern had been with
speculative theology, were putting forward strong arguments
reconciling reason and revelation. This approach which clearly
would allow religious dogma to develop in accordance with the
needs of the times, had won them the sympathies of the Shi'ites. As
in the case of their predecessors, the Qadarites of the late Umayyad
period, the Mu'tazilites upheld tenets which were not void of
implications for the political issues of the time.1
Ma'mun seized on this system of ideas to effect the desired
compromise and reconcile Shi'ite and orthodox ideologies. At
first he encouraged and took part in elaborate discussions with the
intellectual elite about the fundamental principles of Islam and their
relation to all the contemporary issues, but always with a constant
focus on the political significance of these questions. Finally, his
decision was for the official adoption of the Mu'tazilite dogma.
However the only tenet which was particularly emphasized was
the Mu'tazilite insistence that the Qur'an, the Word of God,
was created and therefore could not be as eternal as God and was
certainly less divine.2 In other words the authority of revelation was
not as paramount as the conservative Hanbalites were claiming, and
in accordance with the Mu'tazilites, reason should be given its
proper place in order to allow religious thinking to develop without
undue hindrance. The logical political conclusion of this argument
was that change was possible without a divinely guided ruler. This
in itself shows that Ma'mun was aware of the considerable change
that had been taking place in the empire, and conscious of the strong
resistance to it. One can only conclude that he was not only a
1
Shaban, Islamic History, pp. 156-7,
* ftbul, in, pp. 1040, 1098, 1099, 1105; Athir, vol. vi, p. 288.
https://doi.org/10.1017/CBO9780511622380.004
DISSOLUTION UNDER A NEW REGIME JJ
shrewd politician but also a progressive ruler. For almost a decade
he continued to persuade his stubborn opponents of the feasibility
of this politically imposed theological compromise, but to no avail.
At last, and only months before his death in 833/218, while he was
actually on the battlefield against the Byzantines, he tried to enforce
religious conformity. He introduced what became known as the
"inquisition", mihna, which required that all dignitaries, especially
those of Baghdad, must declare in public their adherence to the
belief that the Qur'an was created. Significantly, Ahmad b. Hanbal
was one of the very few who stubbornly refused to make such a
declaration even in the face of torture. l Although this inquisition
continued for fifteen years, under Ma'mun's successors, it only
helped the Hanbalites to sustain and strengthen their position and
even finally enabled them to force the authorities to abandon this
policy once and for all.2 On the other hand the Shi'ites were content
to see their opponents turning against each other and were not
greatly interested in the whole controversy. However, after the
repeal of Mu'tazilism and the official re-instatement of orthodoxy,
the Hanbalites and the Shi'ites of Baghdad turned vehemently
against each other.
The progressive attitude of Ma'mun also shows to some extent in
his fiscal policies. As mentioned before he had reduced the tax
burden in the East by a quarter. Although it can be argued that this
had been done in special circumstances, other measures confirm this
was in accordance with his general policy. The taxes of Rayy were
reduced by 2 million dirhams to 10 millions, and in the Sawad the
muqasama system was adjusted in favour of the taxpayer; the
treasury's share was to be two fifths instead of half the produce.3
Nevertheless when the district of Qumm demanded the same treat-
ment as Rayy, this was not accepted, presumably because its taxes
were thought to be reasonably assessed.4 In Baghdad a new standard
measure for grain was introduced to protect the consumer from being
cheated by greedy merchants.5 The effects of inflation on people of
fixed incomes were recognized, and for the first time since the
Umayyads, the salaries of government employees were increased.6
These liberal measures did not prevent problems arising simul-
1
Tabari, »'. PP- m * - } : ; Tayfur, pp. 538-43.
1
See below, p. 75.
* Tabari, m, pp. 1030, 1039; Bal., F«/*$, p. 520. Azdl, Maw/il, p. 204.
4
Tabari, m, p. 1092; Azdi, Mawfil, p. 368.
5
Tabari, HI, pp. 1039, 1066; Athir, vol. vi, p. 254; Tayfur, p. 22.
6
Tabari, m, pp. 434-); Jahshiyarl, p. 126.
https://doi.org/10.1017/CBO9780511622380.004
tancously in three areas, the Byzantine front, Egypt and Adhcrbayjan.

The Byzantines, who had not taken advantage of the civil war and
the subsequent disturbances, began to feel the impact of the recent
efforts of Baghdad to control the thughiir. The Byzantine emperor
started negotiations with Ma'mun for the purpose of allowing trade
to be resumed between their respective domains.1 Ma'mun's
response was not favourable and a tense situation began to develop
on this front. This tension was aggravated by conditions in Adher-
bayjan where unrest had been fast turning into open revolt. The
rebels were now making overtures to seek the help of the Byzantines,
and were certainly continuing to trade with them. This was the
formidable revolt of the Khurramiyya led by Babak.
Any attempt to explain this revolt in religious terms is pointless,
since the term Khurramiyya does not signify any specific relationship
to any particular religion; the Persian word khurram meaning only
"happy". It may have possibly been a component of a slogan of this
revolt which was seized upon and used as a derogatory term by their
opponents. It can also be noted that it was used to describe move-
ments between which there could not possibly have been any
relationship; very much as the term Kharijite had been used in
earlier times.2 The nature of this revolt can be explained by the
socio-economic conditions of the region. Reference has already
been made to the considerable influx of Arabs into this region and
the pressure this immigration had put on its resources.1 Although
there are extensive fertile plains in this region, there are also many
ranges of mountains that contain a wealth of mineral resources.
Significantly the Arabs who had settled in the plains of Adherbayjan
were on good terms with the rebels. On the other hand the Arabs
of the mountain ranges were most eager to fight the rebels, and
were indeed at the very root of the problem.4 These Arabs were in
conflict even with their own cousins in the neighbouring region of
Mawsil.5 The cause of this conflict was made crystal clear to Ma'mun
when he received a delegation from the region. They presented him
with no less than forty samples of minerals from the area and
plainly stated that these were the cause of all the fighting in Adher-
bayjan.6 Ma'mun did not need this vivid reminder of the long and
1
Bal., Futufr, p. 192; Tayfar, P- 284.
1
Mas'udi, Muruj, yol. vi, pp. 186-8; Miskawayh, vol. ti, pp. 278, 299; §ibl,
p. 543. » See above, p. 26.
* Azdi, Mavfil, pp. 356, 358, 364, 379-81, 383; Jihuil, in, pp. 1071-2, 1096; Ya'qubi,
J
TJrlkb, vol. 11, pp. 445, 473. Azdi, Mawsil, pp. 359, 366-77.
• an.. P- $54.
https://doi.org/10.1017/CBO9780511622380.004
complicated conflict. The central government, on its part, had long

been active in trying to lay its hands on a share of this wealth. Since
Mansur's time attempts had been made to tax the salt mines and the
well-known oil deposits of this region.l
It is a fact that crude oil was a well-known and widely used com-
modity in those days and even earlier. Surface seepages have for
long been a source of oil and bitumen and in certain areas of Persia
gas-flares were used by the Zoroastrians. In the eighth century such
deposits existed and the oil was referred to as naft. Geographers,
historians and jurists were absolutely clear about the distinction
between naft, crude oil, and qdr or qir, tar or pitch.2 Abu Hanifa,
the famous theologian who died in 767/150 defined naft as the fluid
of a mountain that trickled to the bottom of a well, and was used
as fuel for lighting a fire.3 Another jurist described it as a mineral
that appeared by itself, %ahir, in contrast to ore and gold that
needed to be mined.4 It seems, then, that the commonest way of
obtaining crude oil was to dig a well in areas where seepage had
appeared and simply skim off the substance which floated on top
of the water. Marco Polo was not exaggerating when he described
what he saw centuries later in Baku.* It is interesting to notice that
we have references in our sources to the existence of naft in many
areas where great oil deposits are known to exist today. Such areas
are Iran, 'Iraq, Arabia, Central Asia, Egypt and Adherbayjan.6 This
naft was not all of the same quality and its degree of purity differed
from one place to another; the determining criterion was the colour,
which varied from almost transparent to green and black. The best
quality crude oil known at the time was that of the troubled region
of Adherbayjan.7
While pitch, tar and similar bitumen products were used, among
other things, for caulking ships and curing camel skin disease, crude
oil was used for purposes which required a more combustible sub-
1
Bal., Fuf/ib, p. 210.
» Abu Yusuf, Kiaraj, p. 61; Yahyi b. Adam, Kbaraj, ed. A. M. Shikir, Cairo, A.H. I 547,
p. 32; Ibn tfawqal, $6rat al-Arf, ed. J. H. Kramers, Leiden, 1938-39, pp. 256, 299,
4*8, $ i j ; Ijtakhri, pp. 73, 92, 313; Maqdisi, pp. J J I , 352, 402.
1
LJian, s.v. nff.
* Mawardi, al-Abkam al-Sulldniyya, ed. R. Enger, Bonn, 18)3, pp. $41-2.
» Col. Henry Yule, Tbt Book of Sir Marco Polo tbt Vtiution, London, 187;, vol. v,
pp. 48-$ >-
* Maqdisi, p. 402; Iftakhti, pp. 75, 92, 155; tfawqai, pp. 190, 2J4, 256, 300; Mas'udi,
Murij, vol. 11, p. 25; Mas'udi, Taibfb wa al-Isbrif, ed. M. J. de Goeje, Leiden, 1894,
pp. 60, 61; Maqrizi, Kbifaf, vol. 1, p. 28; Yiqut, Buldm, vol. I, p. 328; Hamdulkh
Mustawfl, Nuzbal al-Qulub, ed. G. LeStrange, London, 1915, pp. 41, 207.
' Yiqut, Buldan, vol. (, p. 328; Mas'udi, Tanbib, p. 60; Istakhrl, p. 190.
https://doi.org/10.1017/CBO9780511622380.004
j8 DISSOLUTION UNDER A NEW REGIME
stance. The earliest example of such a use recorded in our sources

was in 737/119 but thereafter references begin to occur more fre-
quently in our sources, testifying to its use for many purposes and
notably in warfare.1 In 752/134, and significantly in 'Uman on the
east Arabian coast, it was used for shooting flaming arrows at the
enemy;2 in 777/160 by a naval expedition against the Indians;3 in
811/197 in the siege of Baghdad;4 and in 837/222 in the fight
against Babak.s With this increasing use new weapons were soon
developed similar to flame-throwers and what is popularly known
today as a Molotov cocktail.6 It was also widely used not much later
in naval warfare in the Persian Gulf and southern 'Iraq.7 Peaceful
purposes included its use for lighting the two great mosques in
Makka and Madina until it was replaced by candles in 860/246.8
Neither the Arabs nor the Byzantines were blind to the value of this
important commodity.
As suggested before, the question of taxation added a new
dimension to the issue of the minerals of Adherbayjan. It must be
realized that Islam had not originally contemplated any tax on
minerals. Although silver mines had existed close enough to Madina
they had not been of particular importance, and the question of
taxation had passed almost unnoticed. On the other hand, it had
been decided that treasure trove should be treated as booty and
therefore one fifth should go to the treasury. In legal terms the word
rikd^ had been used to denote treasure trove.' After the conquests
and their aftermath it was realized that there were many kinds of
mineral resources in the provinces that ought to come under some
form of taxation. Jurists of the various schools of law became engaged
1
Ash'ari al-Qummi, Kildb al-Maqaldt va al-Firaq, ed, M. J. Mashkur, Teheran, 1963,
1
Athlr, vol. v, p. 547; also Tabari, m, p. 78.
J
Tabari, in, p. 476.
4
Ibid., p. 869; Azdi, Mawfil, p. 350.
1
Jabarl, m, pp. 1202, 1211, 1215.
6
Athir, vol. vn, p. 362, vol. vni, pp. 129, 205; Miskawayh, vol. 1, pp. 282, j 11; ThSbit
al-§abi, Tdrlkb Akbbar aJ-Qardmifa, ed. Z. Bakkar, Damascus, 1970, pp. 19, 27, J9,73.
7
Tabari, m, pp. 1636, 1869, 1959, 2043, 2050, 2060, 2067; Athir, vol. VH, pp. 264,
266-7, 2 7 ° ; Miskawayh, vol. n, p. 46; MaqdisI, p. 12; Abu Bakr al-Suli, Akbbar al-
Rdctt.. ., ed. J. H. Dunne, London, 1935, p. 244.
• Tabari, HI, p. 1471. For other uses see: Tabari, 111, pp. 1236, 1415, 1511, 1578, IJ81
1693,1731, 2219; Athir, vol. vi, p. 188, vol. vn, pp. 44, 98, 124, 186, vol. v m , p. 222,
vol. ix, p. 423; Miskawayh, vol. 11, p. 80; Maqrizi, Kbifaf, vol. 11, p. 196; §SbI,
Wuâri', p. 19; Jahshiyari, p. 300; Abbar, p. 101; Hilal al-§abi, Rusum Dor al-Kbilafa,
ed. MIkhi'il 'Awad, Baghdid, 1964, p. to; Ibn al-Qalanisi, Dbayl Tarikb Dimasbq,
ed. H. F. Amedroz, London, 1908, pp. 20, 40, 47.
« Ibn Hanbal, Mutnad, ed. al-Sa'ati, Cairo, A.H. 1357, vol. ix, pp. 24-6.
https://doi.org/10.1017/CBO9780511622380.004
in lengthy discussions about the definition of the term rikd%, and

whether it should be extended to cover only some or all mineral
substances; and if so, how should these be taxed.1 Being more
practical, the treasury decided that all minerals must be taxed to the
tune of one fifth of the net produce. Of course, this process was put
into effect gradually when and wherever possible. As early as Mansur,
Adherbayjan, where taxes had first been introduced on salt and
crude oil, had become a major target for the execution of the plans
of the central treasury.2 With a variety of as many as forty mineral
substances and the increasing aggressiveness and efficiency of the
tax-collecting apparatus, the indigenous population of the region
had had good reason for their apprehensiveness towards and mis-
trust of the central government. When the Arab immigrants
forcibly appropriated some of their mines themselves,3 the people of
this unfortunate region had no alternative but to resort to armed
resistance to protect their rights. The Byzantines, who were eager
to have continued access to these minerals from a source so near
at hand, were understandably willing to support this rebellion.4
With this support and under the vigorous leadership of Babak the
rebels were able to defeat the successive expeditions Ma'mun sent
against them. Faced with this situation Ma'mun decided to go to the
Byzantine front and take command of it himself, in spite of the fact
that he had delegated this responsibility to his son and brother. In
fact the latter had his hands full in Egypt.
As mentioned before, the continued unrest among the Arabs of
Egypt and their division into factions supporting the opposite
sides in the civil war had created problems for Ma'mun which had
been temporarily brought under control by swift military action.5
However, Ma'mun's new governor had done nothing to alleviate
the effects of excessive taxation, and on the contrary had introduced
stringent measures for tax collection.6 It seems that Islam had begun
1
Ibn Sallam, AmaiH, pp. 537-42.
» Bal., Fulub, p. 210; Qudama, Shemesh, Kbarej, p. 104; Ibn Khurdadhbeh, KitSb al-
Masalik wa al-Mamalik, ed. M. J. de Goejc, Leiden, 1889, pp. 34, 38; Abu Yusuf,
KbarSj, p. 61.
5
Azdi, Mavfil, p. 3)8.
4
Fortunately I received the day I wrote this passage an offprint of an article by my
good friend M. Rekaya, "Mise au point sur Theophobc et l'alliancc de Babek avec
Theophile", Byxantion, T. xuv, Bruxelles, 1974, pp. 43-67, which contains all the
relevant references, although its author does not agree with the point of view expressed
here.
• See above, p. 53.
• MaqrizI, Kbitat, vol. I, pp. 8o, 311.
https://doi.org/10.1017/CBO9780511622380.004
to strike roots among the Copts but no consequential adjustments

had been made to allow for the new status of the converts. Further-
more, as a result of conversion, new ties and common interests had
been established between Arabs and Copts. Under the circumstances,
for the first time an Arab-Coptic revolt broke out and spread all
over the country.1 Mu'tasim tried to cope with the situation, but
when it became clear that he was unsuccessful, Ma'mun had to leave
the Byzantine front and come to Egypt with an army that included
the renowned Afshin and his men. With such a force the revolt was
soon suppressed and Ma'mun wisely proceeded to revise the land-
tax system that had been in operation for almost two centuries. He
realized that the doubling of the tax rate to two dinars under Mahdi
had not been unfair in the light of inflation. After all a judge's salary
in Egypt had risen from 10 dinars a month in 749/1 3 1 to 30 dinars in
772/15 5, to 163 dinars in 813/198 and to 200 dinars in 827/212.2 But
Ma'mun also realized that, because of the increased land holdings of
the Arabs and the gradual conversion of the Copts to Islam, the
Church, which had provided the basic machinery for the assessment
and collection of taxes, had lost its usefulness for this purpose.
Therefore, a new system, the qabala, was introduced, by which one
member of a local tax community contracted to undertake to pay to
the treasury a fixed sum, and was then given the responsibility of
collecting the tax himself. He would also undertake to make all
necessary repairs to the irrigation system, and the expenses would
be deducted from his payment to the treasury on an agreed basis. In
return he was promised that unusual circumstances would be taken
into consideration in case of arrears. To ensure the propriety of the
process, it was conducted in the main mosque in the capital, Fustat,
where the fixed sum for each locality was announced, but not
auctioned; and the person accepting would enter into a contract for
four years.3 Ma'mun also put Copts in charge of some districts
where they would have Muslims as their deputies.4 This step was
probably meant to fill the gap created by the absence of Church
officials, and at the same time to ensure the acceptance of these
appointees among all their subjects. Finally, while in Egypt he took
a seemingly unimportant action that had ramifications later. He
tried to normalize relations with the Nubians to the south who sent
1
Ibid., pp. 171-4, vol. n, p. 261; Ya'qubl, TSrtkb, vol. H, p. 466; Kindl, Wuldt,
PP- J94-4OO. 4*5-4.
1
Maqrizi, Kbifaf, vol. 1, p. 99; Kindi, Wulat, pp. 354, 369, 421, 435.
> Maqrizi, Kbifaf, vol. 1, p. 82.
4
Sa'Id b. al-Battiq, Najw al-jawabir, ed. L. Cheikho, Paris, 1909, pp. 58-9.
https://doi.org/10.1017/CBO9780511622380.004
DISSOLUTION UNDER A NEW REGIME 6l
a delegation to complain to him about Muslim encroachments on

their territories.1 From this point on Egypt's relations with the
south and indeed with the rest of Africa were to become an important
issue.
Having finished with Egypt Ma'mun wasted no time in returning
to the Byzantine front and starting action there in earnest. His
attacks were two pronged and his general aim was to establish new
'awdsim across the Taurus mountains, the centre of which would be
at the cross-roads of Tawana, Tyana, where he actually settled a
strong contingent.1 It is most important to note that the Byzantines
then asked again for peace, negotiating for the specific purpose of
establishing points of communication and allowing trade to flow
freely between the two empires.3 Ma'mun, however, refused and as
if to emphasize his determination to win on his own terms began to
conscript new troops from Egypt, Syria, 'Iraq and even Baghdad
itself.4 This last measure can only indicate that he was not getting
enough of the hoped-for recruits from the East, in spite of his
arduous efforts in this respect. No new development took place
before his death shortly afterwards in Tarsus in 835/218 and it was
left to his brother and successor Mu'tasim to continue with these
policies.
As soon as Mu'tasim was proclaimed the new ruler in Tarsus and
practically on the battlefield, an attempt was made to replace him by
'Abbas, Ma'mun's son. We do not know the principal instigators of
this attempt, but as 'Abbas himself supported his uncle it was
doomed to failure.5 Ma'mun's appointment of his brother as his
heir is a clear indication of his constant attempts to reach a com-
promise. During the civil war Mu'tasim had opposed his brother,
yet the latter, who had an adult son, had preferred to overlook the
lapse and appoint the person better suited to the circumstances. It
can be argued that it had been Rashid's will that his sons should
succeed ,each other, but it has to be admitted that the civil war had
changed all that. It is a fact that both Mu'tasim and 'Abbas had close
ties with the army but the former's association with Afshin indicates
that he had closer ties with the new arrivals from the East. It must
have been well known that he had certain plans for the army for he
proceeded to put them into effect as soon as he came to power. He
ordered the new stronghold at Tyana to be dismantled and the men
1
MaqrizI, Kbifaf, yol. I, p. 198. • T»bari, in, pp. 1102-n.
1 4
Ibid., p. m o . Ibid., p. IIIZ.
> Ibid., p. 1164; Ya'qubi, Tirikb, vol. 11, p. 471.
https://doi.org/10.1017/CBO9780511622380.004
settled there to be withdrawn before he and his army returned to

Baghdad.' When the Byzantines took this as a sign of weakness and
attacked the other strongholds across the Taurus range, he marched
back and inflicted a severe defeat upon them.2 He then sent his army
under the command of Afshin to fight Babak.
Before returning to Baghdad and again on the battlefield, Mu'tasim
discovered yet another conspiracy to replace him by 'Abbas. This
time it is clear that it was led by a combination of Tahirid men and
the abna\ It seems that the latter had not given up hope of regaining
their power, and had encouraged certain suspicions on the part of
the Tahirids of Baghdad about the new plans for the army. However,
the conspiracy was discovered in time and the leaders, including
'Abbas, were severely punished.3 Having firmly established his
authority, Mu'tasim pushed ahead with his plans for the army which
were embodied in building the new city of Samarra about sixty
miles to the north of Baghdad. It has to be remembered that his
father had found Baghdad unsuitable, and had therefore moved his
headquarters to Raqqa. During the civil war a good part of the city
had been destroyed and Mu'tasim decided that if he had to rebuild
anywhere, it was better to build anew altogether away from the
demands of population of the city. Although by moving the central
government departments to Samarra he meant it to be his capital, it
was significantly known by the public as the 'Askar, the army
camp.4 Indeed, this was the true purpose of building the new city.
There is every indication that Mu'tasim built Samarra as a clear
announcement of his willingness to accept new recruits from any-
where. This open invitation was accompanied by a massive campaign
to persuade more of the massive population of the East and the
various peoples of the Caucasus regions, mainly the Armenians and
the Khazars, to join the Muslim armies.5 Ma'mun's efforts in this
respect had been more successful than Rashid's; with a little more
perseverance especially when Islam had begun to strike roots
among these peoples, Mu'tasim was determined to give these sus-
tained efforts the best possible chance for ultimate success. The new
city, and indeed the new opportunities that these peoples saw in the
wealthiest and most powerful empire of the time, soon persuaded
!, HI, p. 1164.
1
Ibid., pp. 1234-62.
3
Ibid., pp. 1249-50, 1256-67.
• Ya'qubi, Buldan, p. 264.
» Hawqal, pp. 467-8; Mas'udI, Muruj, vol. vn, p. 118; Istakhrl, pp. 291-2; Athir, vol.
vi, p. 519; Bal., Fulih, p. 20}; Tabarl, m, p. 1194.
https://doi.org/10.1017/CBO9780511622380.004
many to enrol in its service. Reports that Mu'tasim had to move his
new troops from Baghdad because they were posing a threat to the
population there should not be taken seriously, because the same
reports tell us that it was the people of Baghdad that were harassing
these troops who, because of their small numbers, were not able to
defend themselves.1
These new troops were the so-called "Turks". It must be said
without hesitation that this is a most misleading misnomer which
has led some scholars to harp ad nauseam on an utterly unfounded
interpretation of the following era, during which they unreasonably
ascribe all events to Turkish domination. In fact the great majority
of these troops were not Turks. It has been frequently pointed out
that Arabic sources use the term Turk in a very loose manner. The
Hephthalites are referred to as Turks, so are the peoples of Gurgan,
Khwarizm and Sistan.2 Indeed, with the exception of the Soghdians,
Arabic sources refer to all peoples not subjects of the Sasanian empire
as Turks. In Samarra separate quarters were provided for new recruits
from every locality. The group from Farghana were called after
their district, and the name continued in usage because it was easy
enough to pronounce. But such groups as the Ishtakhanjiyya, the
Isbijabiyya and groups from similar small localities who were in
small numbers at first, were lumped together under the general term
Turks, because of the obvious difficulty the Arabs had in pronoun-
cing such foreign names.3 The Khazars who also came from small
localities which could not even be identified, as they were mostly
nomads, were perhaps the only group that deserved to be called
Turks on grounds of racial affinity. However, other groups from
Transcaucasia were classed together with the Khazars under this
general description.4
The mistake is compounded by another which is even more
astounding, and it is high time that it was utterly refuted; this is
the generally accepted belief that these troops were slaves. All
studies of this subject simply assume that the empire was run, ruled,
policed and defended by slaves; and that this practice spread to all
regions and continued for centuries. This is not only a gross mis-
understanding of human nature, but it also goes against the over-
l, •». PP- "79-8«; Aid!, Mawfil, p. 416.
* Shaban, Tit 'Abbdsid Rttolution, pp. 6-7; Mat'udi, Mmig, vol. a, p. 19;, vol. m,
p. 2 J4, vol. vi, p. 412; Khurdidhbeh, p. 178; T'bari, m, pp. 798, 891; Ibn Khallikan,
vol. vt, p. 40).
> Ya'qubi, BtUdn, p. 264; Mas'udI, Murig, vol. VII, p. 122.
• Ya'qubi, Bu/ttbi, pp. 25 8-62.
https://doi.org/10.1017/CBO9780511622380.004
whelming evidence to the contrary in our sources. This misconcep-

tion was quite obviously inspired by studies in Ottoman History
which were projected backwards to explain earlier developments. If
the Ottomans had had "slaves" forming the "ruling institution",
then it was assumed without any foundation that similar institutions
were almost inherent in Islamic society. Fortunately, modern
scholarship in the field of Ottoman history has come forward to
challenge these erroneous conclusions that have survived far too
long.1
In this case, it cannot be over-emphasized that Mu'tasim was
following the practice, started under his father, Rashid, and con-
tinued under Ma'mun on a wider scale, of persuading the chiefs and
princes of the East to enrol in the army of the central government.
Under Rashid all the 'Abbasiyya had been made mawdli of the ruler.2
Under Ma'mun only the leaders had been made his mawdli; thus
Tahir, originally a mawld, or client of the Arab clan of Khuza'a, had
been created mawld Amir al-Mu'minin, and had retained control over
his own men.3 Mu'tasim threw the door wide open for men to come
in groups under their own chiefs, or individually if they so desired.
In most cases they came in groups, and we find many instances
where their chiefs were native princes from the East. Afshin, who
was a prince of the region of Ushrusana in Soghdiana, was certainly
the prime example of such princes, but mention is also made of
others from the same area, such as Kaydar, Divdad, Bukhara-khuda,
Buzbareh, Siil-tegin, Marzban b. Tiirkesh, 'Ujayf b. 'Anbasa,
Khaqan Artug, Banijiir, Shir-i-Bamyan, and 'Amr of Farghana with
his retinue of princes.4 Following an ancient institution in the East
their followers formed corps that were called chdkars, a Persian word
meaning servants.5 These chdkars performed for their leaders the
same military service that the knights of the Middle Ages in Europe
performed for their kings. This whole institution was now trans-
1
M.I. Kunt, "Ethnic-regional solidarity in the 17th century Ottoman Establish-
ment", International Journal of Middle East Studies, vol. v, no. 3, June 1974, pp. 133-9;
see also the introductory note by the editor S. J. Shaw, pp. 251-2.
* See above, p. 31.
> Khatib, Tarikb Baghdad, Cairo, 1931, vol. ix, p. 55;; Ibn Khallikan, vol. 11, p. 309,
vol. in, p. 88.
• Tabarl, H l i pp. 1042, 1169, 1194, 1197, 1203, 1212, 1222, 1225, 1228, 1239, 1241,
1246, 1301, 1312, 1313, 1335, 1373, 1474, 1605, 1616, 1 6 5 7 , 1 0 2 2 - 3 ; Bal., F K / « $ , p.
431; Ya'qubF, Tartkh, vol. 11, pp. 474, 475, 477; Ya'qubl, Buldan, pp. 260, 290, 293-4;
tlawqal, p. 499; Khurdidhbeh, pp. 40, 180; Istakhrl, pp. 291, 323; Maqrizi, Kbifaf
vol. it, p. 251.
» Marvazi, On China, Tbt Turks and India, Arabic Text, ed. V. Minorsky, London, 1942,
p. 18.
https://doi.org/10.1017/CBO9780511622380.004
D I S S O L U T I O N U N D E R A NEW REGIME 6$
planted from the East into the heart of the empire. The chdkars
continued in the service of their own leaders, who in turn were
created mawall of the ruler.' Other men who came individually and
proved their abilities were made chdkars of the ruler himself. The
term used for this position was ghuldm (pi. gbilmdn) which is the exact
equivalent in Arabic of the Persian term chdkar. This latter term was
Arabicized as shdkiriyya. It is to be noted that the shdkiriyya in
Baghdad could only have been those of the Tahirid forces who
performed the task of policing the city itself and the Sawad.
Any ghuldm of the ruler who proved his military ability could be
promoted to the rank of qd'id, commander, and if he had special
administrative talents he could be made wasif and then khddim.*
These last terms should not be confused with the same words
meaning kinds of domestic servants. They must be understood to
denote certain ranks of employees of the ruler through whom he
was afforded a degree of control over the machinery of government.
In fact these were not much different from the mawdli of MansGr,
and the new titles were due to the fact that the term mawld had taken
another twist to cover leaders who rendered military support to the
ruler.
The highest rank a mawld or a khddim or indeed anyone subordinate
to the ruler could attain was 'abd (pi. 'abid), slave, of Amir al-
Mu'mimn. As early as Abu Muslim a new phrase had been coined to
describe loyal subjects, 'abid al-sam' wa al-td'a, i.e. (those like) slaves
who hearken and obey.3 As this phrase had continued in usage it is
possible to conclude that in the atmosphere of the time and con-
sidering the other titles discussed, the title of 'abd was meant to
denote a subject whose loyalty was beyond any doubt.4 There are
very many examples of men who could not have possibly have been
slaves taking pride in describing themselves as, or being addressed
with, this new title.5 In fact Sabi, a high official of later 'Abbasid
administration, who was in a position to know the rules of protocol,
tells us that khddim and 'abd were the highest ranks to which anybody
could aspire.6
l, HI, pp. 8, 928, 1373, 1427, 1605; Athir, vol. VII, p. 32; Ibn A'tham, vol. 11,
f. 271*.
1
Jahshiyari, pp. 258, 276; Ibn Man?ur, Usan, s.v. uld; Ibn Khaldun, MuqaMma,
p. 684.
1
Ibn A'tham, vol. 11, f. 255B.
* Abu Shuja', Dboyl. . ., ed. H. F. Amedroz, Cairo, 1916, p. 216.
» Tabari, HI, pp. 397, 572, 574, 700, 1096, I I J 8 , 122}, 1260, 1311, I J I 6 , 1327; Athir,
vol. vi, p. 375.
* §abi, Rusum, p. 107.
https://doi.org/10.1017/CBO9780511622380.004
Although we are in a relatively good position to know individual

names and the original places from which the princes of the East
had come, we unfortunately have no information about the re-
cruited Khazars. We are faced only with new names of army
commanders of whose origins we know nothing. Yet the existence
of Khazars in the army leads us to believe that men such as Bugha,
Yermesh, Wasif and Itakh were in fact their leaders.1 It is possible
that such men were chiefs of nomadic groups vaguely identified as
Khazars or Turks, without attachment to any specific place. The
fact that Bugha was Mu'tasim's brother-in-law indicates that he must
have been a man of considerable standing among his own people.2
Reports that Itakh and Ashnas had been bought as slaves and then
freed only to attain very high positions are almost certainly fictitious,
and were probably put forward by their opponents.3 It is highly
improbable that the prince of Bamyan would easily agree to be in
the service of Itakh if the latter had indeed once been a slave.4
Other reports that Mu'tasim commissioned the buying of Turkish
slaves from central Asia, and that as many as 4000 had been bought
to form the nucleus of his army even during Ma'mun's lifetime, are
surely greatly exaggerated since the Arabic sources are least reliable
when it comes to figures.5 Moreover slaves, particularly Turkish
slaves, men or women, were not cheap to say the least.6 It seems that
these isolated reports have confused the issue and in the process
confused scholars as well. Mu'tasim, in his efforts to take every
possible precaution against the absorption of the new recruits and
their involvement with local interests, had prohibited intermarriage
between the civil and the military communities. As a result of this
he had to buy female slaves and marry them to members of his
army; and to encourage the practice, he even allowed for such wives
in the stipends of their husbands.7 In short, the myth that Turkish
slaves were the mainstay of the army at that time has no foundation
whatever.
Mu'tasim was a man of considerable military ability, and he
certainly knew that slaves would not make good fighters. His
sound judgment in these matters is revealed by his choice of Afshln
and his men from the mountainous region of Ushrusana (Kfibbdmjya,
1
Tabari, in, p. 1383; Ya'qubi, Button, p. 262.
1
Athir, vol. vn, p. 62. ' Tabari, m, pp. 1017, 1383.
• Ibid., pp. 1335, 1373.
5
Ya'qubi, Buldan, p. 255; Mas'udi, Muruj, vol. vn, p. 118.
* tfawqal, pp. 435, 465.
7
Ya'qubi, BuUdn, pp. 258-9.
https://doi.org/10.1017/CBO9780511622380.004
mountaineers) to combat Babak and his followers who were defend-

ing a similar terrain.1 Although these rebels had been defying the
central government for two decades, Afshin was soon able to crush
their revolt. Returning triumphantly to Samarra, he seems to have
entertained some ambitions of getting the same price for his support
as the Tahirids, i.e. complete autonomy in his region in Soghdiana
and possibly some control over neighbouring princes.2 Mu'ta$im
could not afford to pay this price because the Samanids, proteges
of the Tahirids, were already in charge of the whole region and
were on very good terms with their patrons.3 This delicate balance
in the East which had been so recently established, would have been
completely destroyed if Afshin had been allowed to tamper with it.
Expecting great rewards for his services, he found himself instead in
prison where he died soon after. However, his men continued in the
service of Mu'tasim, and even his son was soon to be found among
the commanders of the army.4
In Tabaristan where Islam had spread since the time of Mahdi
there was an unexpected upheaval. Mazyar, the prince, who had
prided himself on being a mawla of Ma'mun and had been very co-
operative with his government, began to show signs of disagree-
ment, particularly with the Tahirids.* Being a member of one of the
ancient ruling families, he apparently objected to being subordinated
to the parvenu Tahirids.6 Furthermore he seems to have experienced
some difficulties in controlling his subjects. The merchants of the
prosperous ports on the Caspian, relying on their good connections
in the Tahirid domains, began to defy the authority of his tax-
collectors.7 In addition Ma'mun had established a garrison at Chalus,
a move that must have undermined the authority of Mazyar in his
territory.8 It also seems that members of this garrison had taken
advantage of their position to acquire lands that had been public
property.' When Mazyar's patience was exhausted, he simply
arrested, all members of the garrison who were mainly Arabs and
abnd'.10 He also ordered the walls of the defiant ports to be razed.11
1
Tabarl, in, pp. 1188, 1190, 1196, 1199, 1203, 1205, 1119.
1
Ibid., pp. 1305-18.
1
Athir, vol. vi, p. 362, vol. vn, p. 192; Hawqal, p. 467; MaqrizI, Kii/af, vol. I, p. 17).
4
Tabari, HI. P- 1664.
» Ya'qubi, Tdrlkb, vol. 11, p. 477; Mas'udl, Mmij, vol. vn, p. 137.
* Tabari, m, p. 1268.
7
Ibid., pp. 1271, 1272, 1278, 1298-9.
* Ibn Rustah, al-A'Iaq al-Nafija, ed. M. J. de Goeje, Leiden, 1892, p. 151.
» Tabari, m, pp. 1269, 1270, 1277.
10
Ibid., pp. 1273-4, 1278. " Ibid., p. 1275.
https://doi.org/10.1017/CBO9780511622380.004
Both Mu'tasim and the Tahirids moved fast, and Mazyar was quickly
arrested and replaced by a more pliable member of the same
family.1
In contrast to Ma'mGn, Mu'tasim was not a political animal. In
the internal policies of the empire he followed his brother's steps;
some tax anomalies were adjusted, and support for the Mu'tazilitcs
continued unabated.2 In the building of Samarra plans were made
for shops and such mustaghalldt that netted as much as 10 million
dirhams a year.1 In the administration there was no doubt who was
in charge of the army but in the fiscal field he introduced a new
element. Hitherto this responsibility had always been given to men
with long experience in government service, but Mu'tasim in his
straightforward military manner decided that it should be left to
financiers. It is probable that he also thought that such an important
section of the population should be represented in government. In
the light of the participation of the merchants of Baghdad in the
civil war and especially when the capital had been moved to Samarra,
such representation was a gesture towards these men whose economic
powers could not be ignored. One of the wealthiest merchants of
Baghdad, Muhammad b. 'Abdulmalik al-Zayyat, was chosen as a
kdtib\wa%tr, a position that he held for iz years throughout the
reigns of Mu'tasim and his son.4 It was during this period that many
of the rich 'Iraqi families of non-Arab origin began to move into
various departments, not only of the central government, but also
those of the provinces. Although many of them were big landowners,
their interests were certainly intertwined with those of the big
merchants in the capital and the provincial towns. Very soon and for
a long time they were to dominate and practically monopolize the
key positions in government, handing them down from father to
son for generations.1
However, the most influential figure in Mu'tasim's reign was the
chief judge of Samarra, Ahmad b. Abi Duwad. He was undeniably
an Arab, and of course a Mu'tazilite, but these facts do not explain
his power at that time. He seems to have been behind all Mu'tasim's
decisions, even in fiscal matters, to the extent that he was able to
persuade him to pay two million dirhams from his own treasury for
clearing and dredging a river in Farghana in Central Asia.6 Of
' Ibid., pp. 1*7}-$°}- * »«*, p. i*7*; Bal., Futit, pp. 145-4.
• Ya'qubl, BuUan, pp. 2J7-<S4. • Tatar!, m, pp. 1185-4.
» Ibid., p. 1)31.
• Ibid., pp. I2J5, 1526; Ya'qubi, TdrM, vol. 11, p. 478; Mat'udi, Murij, vol. VII,
p. 105.
https://doi.org/10.1017/CBO9780511622380.004
course one can argue that he was merely going along with the ruler's
plans of encouraging the people of the region to enrol in his service
by showing them how much they would gain from their co-opera-
tion. But then the question arises: what does a judge have to do
with such matters ? As it was an exceptional case, no firm conclusions
can be drawn from it other than that Mu'tasim had great faith in the
wisdom of this particular man.
One important feature of Mu'tasim's reign demonstrates the
increasing volume of trade through the Indian Ocean with East
Africa, South and South East Asia and the Far East. Indians had
started piratical activities throughout the Persian Gulf and many of
them had immigrated into the region of Basra and threatened its
communications with Wasit to the north. Again Mu'tasim moved
fast and put an end to this double threat to the prospering trade.'
Until the end of his reign he seems to have objected to a certain
measure which the merchant wa^tr, Ibn al-Zayyat, did in fact
introduce soon after his death when he must have had more freedom
to act. This measure was the abolition of all taxes on merchandise
coming in through the ports of the Persian Gulf.2 It is arguable that
this could have resulted in a reduction of prices, but as this trade was
mostly in luxury goods it would have benefited only the rich.
Moreover, Ibn al-Zayyat could not have been unaware of his fellow
merchants' gain from such a concession.
Mu'tasim was succeeded by his son, Wathiq, who ruled for five
uneventful years 842-7/227-32. Everything continued as it had been
under his father, and he does not seem to have been much interested
in ruling. During his short life he left the administration to Ibn al-
Zayyat. On the military side he divided the responsibility for the
defence of the empire between two of his most outstanding generals,
Ashnas in the west and Itakh in the east.3 This arrangement had
nothing to do with the domains of the Tahirids who were in charge
of the East in addition to policing Baghdad and the Sawad. When
'Abdullah b. Tahir died, his son was duly appointed to succeed him
in the same position with the same responsibilities.4
The only important development at this time was a delayed reac-
tion in Baghdad to the continued imposition of Mu'tazilism.s Some
of the leaders of the abnd', supported by the same conservative ele-
1
Tabari, m, pp. 1166-8; Mas'udi, Tanbib, p. 355; Bal., Futufi, p. 446.
1
jabarl, m, p. 136}; Ya'qubi, Tdrlki, vol. 11, p. 483.
J
Ya'qubi, Tartkb, vol. II, p. 479.
* Tabari, in, pp. 1338-9. » Ya'qubi, Tarikb, vol. it, p. 482.
https://doi.org/10.1017/CBO9780511622380.004
ments that had appeared during the civil war, organized a resistance
movement against the inquisition.1 Although this movement was
soon suppressed and its leaders executed or imprisoned, it seems to
have had repercussions far reaching enough to change the course of
events in the following years.
1
Jabtti, in, pp. i J45-9-
https://doi.org/10.1017/CBO9780511622380.004
4
THE ORIGINS OF THE 1QTA'
For the researcher in Islamic history the study of the next period of
almost a quarter of a century 847-70/232-56 ought to be an easy task.
There are detailed reports rich with information about the rapidly
developing changes, especially in the heart of the empire. Yet, these
very reports tend to confuse and blur rather than clarify the issues
involved. This period is characterized by long and complicated
fighting between the military forces of Baghdad and Samarra and
even among the forces in each of these centres. The names of the
principal leaders involved in this bitter struggle are mentioned in
the sources and so are the groups that supported them. As there
were no regimental names as such, army groups were given various
nomenclatures, but in some cases different groups were described
by the same name. What is important is whether the particular group
in question belonged to Baghdad or Samarra, e.g. the shdkiriyja.
Without identifying each group and its leader no real understanding
of the causes of the struggle can be obtained. As this period con-
tained the germs of military and administrative iqta', such an under-
standing is vital for the analysis of later developments. In their
valuable studies on this subject neither Cl. Cahen nor A. K. S.
Lambton has made the necessary effort to clarify these beginnings.'
Arabic sources like us to believe that Wathiq was so little interested
in ruling that he did not even want to take the trouble to nominate
a successor.2 A more plausible explanation is the fact that when he
died at the age of thirty-four his son was a minor, but perhaps he
also well knew that it was difficult to find a member of his family
who would be able or willing to continue as he himself had done
with the policies his father had initiated. After his death a meeting
of the men who held the reins of power was called to decide the
question of succession. The dominant member was the chief judge
Ahmad b. Abi Duwad; the others were the two most prominent
1
Cl. Cahen, "Involution de Viqfa' du ix° au x m e siecle. Contribution a une histoire
compare des societes mcd\£vi\ct ",Anna/es: Economic, sociith, civilisations, vm(i953),
pp. 25-52; Ann K. S. Lambton, "Reflections on the IQTA'" in George Makdisi, ed.,
Arabic and Islamic Studies in Honor of Hamilton A. R. Gibb, Leiden, 1965, pp. 358-76.
* Ya'qubl, TSrikb, vol. 11, p. 483.
https://doi.org/10.1017/CBO9780511622380.005
72 THE ORIGINS OF THE tqfd'
military commanders and three high civilian officials, including the
waqjr Ibn al-Zayyat. It is important to note that this was the first
time that such a situation had arisen in a century of 'Abbasid rule,
and that it set a precedent for later generations. In this meeting the
military commanders did not seem particularly interested in the
choice of any specific candidate, but surprisingly the civilians
favoured the selection of the minor son of Wathiq. Finally, in order
to observe legality, the will of the chief judge prevailed, and Wathiq's
brother who was twenty-six years old was selected and given the
title of Mutawakkil.1
Perhaps nobody was more surprised by the actions of the new
ruler 847-61/232-47 than the selectors themselves, for it soon
became clear that he had a mind of his own. In a matter of weeks he
proceeded to make sweeping changes in the personnel of his
administration. Ibn al-Zayyat who had been in office as wa%tr\katib
for twelve years, was dismissed, as were also his colleagues, some of
whom were very wealthy merchants indeed. Mutawakkil must have
realized that these men, who had continued their commercial
activities while in office, had used their positions to amass huge
fortunes for themselves and their fellow merchants, at the expense
of the general public and the central treasury.2 They were replaced
by a team headed by Abu al-Fadl al-Jarjara'i, a native 'Iraqi land-
owner of ancient lineage. The significance of this team is that it
combined members of the big land-owning families with other well-
trained men who had served in previous administrations.3 In other
words, after getting rid of the corrupt merchants, Mutawakkil
brought into government men with expertise in the affairs of the
local rural communities, and supported them with others of special
administrative talent.
On the military side he had even more radical plans which revolved
around two main ideas. The first was to change the structure of the
military forces in order to create a uniform standing army over
which he would have direct control without the mediation of chiefs
such as Afshin. The second idea was a logical extension of the first;
the autonomous powers of the Tahirids in the East and their func-
tions in Baghdad, the Sawad and Fars were to be abrogated. Of
course he knew only too well that he would meet strong opposition
from very powerful elements throughout the whole empire, but he
seems to have thought that by introducing them gradually he could
eventually succeed in effecting these substantial changes.
1 3
Tabarl, m, pp. 1)68-9. * VM., pp. 1376, 1379. Ibid., pp. 1373-9.
https://doi.org/10.1017/CBO9780511622380.005
THE ORIGINS OF THE tqfd* 73
Mutawakkil perceived that the official adoption of Mu'tazilism
and the subsequent fifteen years of inquisition had failed to produce
tangible results. On the contrary, it had aroused stubborn opposition
in the powerful conservative circles of Baghdad. In an attempt to
win support for his plans, especially among these circles, he decided
to drop this whole controversial issue and declare the official return
to orthodoxy.1 To emphasize his point he also made some anti-
Shi'ite gestures;1 and announced a discriminatory policy against
non-Muslims in the empire. While most of the announced measures
were of a symbolic nature, two were particularly offensive to non-
Muslims; they were to be excluded from any appointments in
government departments, and, more seriously, a tax was to be levied
on their houses.3 This last measure was widely resisted and when it
was vigorously imposed resulted in insurrections in Syria.* On the
other hand, the repeated announcements of these measures indicate
that the government had difficulties in enforcing this policy.5
Turning his attention to the army, Mutawakkil's first step was to
weaken its leadership. Fortunately Ashnas had died before Mutawak-
kil's accession to power and, with the acquiescence of the Tahirids,
the other powerful figure, Itakh, was arrested in Baghdad on his
way back from the pilgrimage.6 Then came a sensitive and poten-
tially dangerous problem, how to reduce the powers of the Tahirids
without appearing to renege on their agreement with Ma'mGn
which had been in operation for almost thirty years. Time had helped
to consolidate the Tahirid presence in Baghdad, the Sawad and
Firs to the extent that they actually governed these and even
adjacent regions also. The department of taxation in Baghdad was
referred to as the "greater" in contrast to that of Samarra.7 Under
the circumstances Tahirid authority in the heart of the empire
constituted a government within the government, and this was in
addition to their control of the East.
A combination of circumstances and rivalries between the
numerous members of the Tahirid family helped Mutawakkil to
implement his plans. Within a year no less than five Tahirids holding
key positions disappeared from the scene; two were murdered and
three died.8 Although the leading member of the family was alarmed
enough to come from Khurasan to Baghdad, Mutawakkil was
Ibid., pp. I4"-IJ- ' 1M-, PP- «JO7, IJ24, IJ»6, 1)79. 14OJ-4.
Ibid., pp. 1)89-94. * Ibid., pp. 1420-4.
Ibid., p. 1419.
Ya'qubt, Tarlkb, vol. 11, p. 486; Athir, TOI. VII, p. 50.
Yaq'ubl, TdrM, vol. II, p. 488. * T»barl, m, pp. 1405-6.
https://doi.org/10.1017/CBO9780511622380.005
74 THE ORIGINS OF THE iqtd*
astute enough to appoint other Tahirids to the vacant positions and,

without revealing his hand, to reduce the powers of the new
appointees.1 At the same time he proceeded with another subtle
move which was designed to weaken Tahirid authority in their
regions, and to give him more direct control over the whole empire.
He designated his three sons as his successors and decided on the
titles they should assume when they came to power. The eldest,
who was no more than thirteen years old, received the title Muntasir
and was the immediate heir apparent. In due course he would be
followed by his brothers Mu'tazz and Mu'ayyad respectively 2 This
was no innovation and could even be interpreted as a long term
measure to ensure stability. This designation was accompanied by a
declaration that had far-reaching and immediate consequences. The
actual government of the empire was divided between these three
children. Muntasir the heir apparent was put in charge of almost all
the western half of the empire, with the exception of Syria which was
assigned to Mu'ayyad. Mu'tazz was significantly given the respon-
sibility for the East and as a camouflage Armenia, Adherbayjan and
Fars were included in the bargain.3 These appointments were not
meant to be nominal because the declaration went into great detail
assigning all the affairs of these regions to their prospective new
rulers.4 Of course the idea of dividing the empire between sons was
not entirely new and a precedent had been set by Rashid. But in this
case the beneficiaries who were being charged with these respon-
sibilities were minors for whose actions Mutawakkil himself was
legally responsible. In other words he was using this as a means to
give himself the right to interfere in any way he wanted in all the
affairs of every province, particularly the East. The Tahirids, in their
already weakened positions, had no alternative but to accept the new
arrangements, especially as they were allowed to keep their own
military forces.
The next step was the dissolution of the army of Samarra, and
again this was done with great caution as part of the new arrange-
ments. It was almost logical that each son should have a strong
contingent to enable him to fulfil his responsibilities in his vast
domains. Therefore the army of Samarra was split into three divis-
ions, each of which was dispatched with its child-ruler to a location
in his regions. As all aspects of the administration of these regions,
1
Ibid., p. 1410; Athir, vol. VII, pp. 36, 43; Ya'qubi, Tdrikb, vol. 11, pp. 487-8.
* Jabari, HI, pp. 1394, 1396-7.
1
Ibid., pp. 1395-6. * Ibid., pp. 1398, 1494.
https://doi.org/10.1017/CBO9780511622380.005
THE ORIGINS OF THE iqtd' 75
including the fiscal affairs, were entrusted to these minors, it was
necessary to appoint capable administrators.1 Thus, in effect, the
commanding officer of each division and the chief administrator
were in charge of their respective regions albeit under the close
scrutiny of Mutawakkil. This precedent gave birth to what can be
best described as a military-bureaucratic alliance which was soon to
become a troublesome feature in the politics of the empire.2 Another
precedent set at this time eventually developed into an institution
that resulted in many serious problems. Although army stipends
were deducted from revenues before sending them to the central
treasury, the leaders of the divisions were given a land-grant, iqtd',
for their own personal expenses and as a means to entice them to
accept the move to their new regions. Even the Tahirids were given
such grants to compensate them for their losses, and probably also
to equate them with their new colleagues in the provinces.3 Of
course iqtd1 was no new feature, but hitherto it had been granted to
members of the ruling family and their favourite supporters for
services rendered. Now it was granted to military leaders in areas
where they exercised some administrative functions; this is the
genesis of military-administrative iqtd' to which we shall return
later.4
Having thus consolidated his own position, Mutawakkil continued
to carry out the rest of his plans. He dismissed his wtntfr Jarjara'i,
not because of any disagreements or faults but because he wanted to
replace him by a man more useful for his purposes. The new wasjr
was 'Ubaydullah b. Yahya b. Khaqan whose qualifications were long
experience and proven loyalty in the service of the 'Abbasids.
Although he was a Khurasanian by origin and therefore not from
the same category as his predecessor, his associates and the new
appointees to the various departments were men of the same back-
ground as Jarjara'i.5 In order to ensure the effectiveness of his own
control Mutawakkil revived the office of chief comptroller, under a
different'title, and gave it more authority. To this office he appointed
a well-trained administrator who was directly responsible to him.6
However, very soon this arrangement proved unworkable for it
gave the chief comptroller virtual authority over the trusted waqir
and his associates. The former was swiftly dismissed and the power
1 2
Ibid., pp. 1598-400. See below, pp. 94, 116.
1 4
Tabari, m, p. 1452. See below, pp. 95, 109.
s
jabari, m, pp. 1407, 1441; Ya'qubl, Tarikb, vol. 11, p. 485.
* Tabari, m, pp. 1440-1.
https://doi.org/10.1017/CBO9780511622380.005
76 THE ORIGINS OF THE tqtd"
of the latter vastly increased to include effective control over all
departments.»
Ibn Khaqan, with his new powers, was instrumental in helping
his master to proceed with a most important stage of his plans - the
recruitment and organization of a new army. All the members of
this army were directly attached to Mutawakkil as his shakiriyya, i.e.
their allegiance was only to him as their sovereign. For obvious
reasons such an army could not be recruited from Tahirid territory
in the East; furthermore this region was having its own new
troubles. Therefore, it had to be recruited from the western regions
such as Syria, Jazira, Jibal, 'Iraq, Adherbayjan, Armenia and the
Caucasus; even some abnd' from Baghdad enlisted alongside this
conglomeration of Arabs and non-Arabs from all these different
regions.2 While this operation was in progress Mutawakkil decided
on his next step; to move from Samarra altogether. His first choice
was Damascus, and he moved there in 858/244, but as this move
brought Samarra to the point of open revolt, he hurried back there
after only a few weeks in Damascus.3 However, a year later he
started building a new city just outside Samarra. It was called
Ja'fari, after his personal name, and its cost was two million dinars,
equivalent to fifty million dirhams at the exchange rate at the time.4
Although it is reported that most of the population of Samarra
moved to the new city where shops and similar mustaghallat were
constructed, it is also reported that because of an engineering mistake
the water supplies were not adequate, and it was subsequently
deserted, almost as soon as it had been built, after the death of
Mutawakkil.5
After the apparent success of all his plans and as he seemed to be
in full control, Mutawakkil wanted to take the final step to ensure
the complete success of his policies, but it was a step that brought
about his downfall. He thought that he could integrate the old
Samarra forces into his new shakiriyya army, but he realized that to
achieve this end he must first get rid of their leaders. He began by
attempting to take back the land-grants he had given them; a move
which only helped to reveal his hand.' As his intentions became very
• Ibid., pp. 1441-7; Ya'qubi, TSrikb, vol. n, p. 488.
* Tabari, m, pp. 1589, 1465; Athlr, vol. vn, pp. 52, 64; Mas'udI, Taubib, p. 361.
1
TaHari, m, pp. I4JJ-6; Ya'qubi, Tarikb, vol. H, p. 491.
4
Yaqut, Buldan, vol. 11, p. 14).
Tabari, m. P- Mj6; Ya'qubi, TariJkb, vol. 11, p. 492; Ya'qubi, Bulddn, p. 267; Bal.,
F $ , p. 298.
Ii pp. 1452-6.
https://doi.org/10.1017/CBO9780511622380.005
THE ORIGINS OF THE iqtd' ff
dear they decided that it was time to get rid of him before it was
too late, and through their accomplices they had him murdered in
his new palace in Ja'fari. Fath b. Khaqan, an uncle of the waqir who
happened to be with Mutawakkil at that time, was also murdered,
while his nephew the organizer of the sbdkiriyja remained unharmed
under their protection. Muntasir, the eldest son of Mutawakkil,
whose sympathies had been with the leaders of the Samarra army,
was promptly installed as his father's successor. He immediately
moved back to Samarra where he ordered a general withdrawal of
the Samarra forces from the provinces to the capital. He also received
the allegiance of the sbdkiriyya after some hesitation on their part.'
During the reign of Mutawakkil the Byzantine front which had
been quiet for more than a decade suddenly flared up. This was
mainly the result of the increased naval activities of the Byzantines
in the eastern Mediterranean which culminated in a major attack on
Dimyat on the Egyptian coast in 85 2/23%.z Consequently fortifica-
tions were built on the Syrian and Egyptian coasts; also the construc-
tion of a new navy was started in Egypt and naval operations against
the enemy began on a rather small scale.3 Meanwhile a counter-
attack was mounted on the Byzantine front. Summer expeditions
were resumed and continued every year under the vigorous leader-
ship of an Armenian Muslim, 'All b. Yahya.4 Part of the Samarra
army was stationed there at Sumaysat, and significantly the land-tax
there was confined to the Muslim tithes, probably because these men
or their leaders were granted agricultural lands there.5 On the other
hand, in other locations in the thughiir, a land-tax concession was
abolished. This was partial or complete tax exemption, ighdr, on
lands which had been previously granted to settlers in these
areas.6
In neighbouring Armenia and Adherbayjin there were fresh
troubles in the aftermath of Babak's revolt. One of the Arab leaders
who had supported Babak and had been imprisoned in Samarra
was able to escape, and he organized a new revolt against the central
government, probably because of the strict enforcement of taxes.
He did not seem to have much support and the region was soon
• Ibid., pp. 14JJ, 1457, 1459-79-
Ibid., pp. 1417-18; Ya'qubi, Tarikb, vol. n, p. 498.
Tabari, in, p. 1449; Maqrizi, Kbifal, vol. 1, pp. 211, 214, vol. 11, p. 191; Bal., Futit,
p. 118.
Tabari, lit, pp. 1414, 1419, 1420, 1426-8, 1434, 1436, 1437; Mas'udi, Tanbib, p. 191.
Tabari, in, pp. 1428, 1436, 1459; Bal., Fulib, p. 184.
Bal., Vutub, p. 171.
https://doi.org/10.1017/CBO9780511622380.005
78 THE ORIGINS OF THE iqtd'
1
pacified. Interestingly, his sons enlisted in the shdkiriyya probably
with some of their fellow-countrymen.2 However, in Armenia, when
the same tax-collecting methods were enforced, the result was a
combined Arab and Armenian revolt. The Muslim leaders of this
region, whether Arab or Armenian, had extensive connexions with
the chiefs of the adjacent Caucasian regions, and they probably relied
on their support of their cause. After considerable resistance from
the rebels and arduous efforts on the part of the government the
revolt was barely contained.3
In Syria the Christians of Him?, supported by the Muslims, rose
against the house-tax introduced by Mutawakkil. With little effort
their local insurrection was quickly suppressed.4 In Egypt the failure
of the 832/217 revolt convinced the Egyptians of the futility of
armed resistance, and at the same time allowed Muslims to control
more of the countryside.5 The introduction of the house-tax there-
fore had no repercussions there. This tranquil state of affairs gave
the government the chance to introduce a measure which was made
necessary as a result of the withdrawal of Church officials from the
administration. In 856/242 the governor of Egypt appointed a
headman, mukhtdr, to every village in the province; a move which
could only enhance the power of the administration.6
The good relations that Ma'mun had established with the Nubians
had been strengthened under Mu'tasim to the extent that the king
of Nubia had had a house in Cairo and another in 'Iraq, which
probably involved the presence of agents in both places.7 On the
other hand the Beja, the indigenous population of the Egyptian-
Sudanese Red Sea coast, who had had long established peaceful
relations with the Arabs, began to attack Upper Egypt. Since the
conquest of Egypt these people had adhered to a peace treaty with
the Arabs according to which they had agreed to pay an annual
tribute of 75 ounces of gold. They had also agreed to permit Arabs
to settle in their territory and exploit the silver, gold and precious
stone mines there. As is evident from the name of their king at this
time,' Ali Baba, some of them had accepted Islam, and there is every
indication that they had continued to abide by the terms of their
treaty. Nevertheless, it seems that as a result of the implementation
1
I'abari, in, pp. i)8o-z, 1587-8; Ya'qubl, Tarikb, vol. 11, p. 486; Khurdidhbch,
pp. 119-Z4.
1
Tabari, HI, p. 1589; Athlr, vol. vii, p. 52; see also above, p. 76.
1
Tabari, m, pp. 1407-8, 1414-16; Athir, vol. VII, pp. 58-9, 44-j.
* Tabari, in, pp. 1420-4. > Maqrizi, Kbifat, vol. n, p. 494.
* .Kind!, Wulil, p. 203. ' Maqrizi, Kbi/at, vol. I, p. 199.
https://doi.org/10.1017/CBO9780511622380.005
THE ORIGINS OF THE iqtd' 79
of the policy of imposing taxes on mineral produce they had been
obliged, together with the Arabs there, to pay the required fifth.
They must have considered this not only an "over-tax" but also a
breach of their treaty. Under Mutawakkil they ceased to pay the
tribute and stopped the Arabs from mining in their territories.
In 855/241 Mutawakkil had therefore sent an expedition which
successfully restored the position of the Arabs there.'
Another minor upset which seemed to have no great consequences
at the time but which was soon to have serious repercussions in the
heart of the empire, was the start of the Saffarid movement in the
rather neglected region of Sistln. As this was in Tahirid domains the
Tahirids dealt with it without great difficulty at this stage.2
On the economic side there were no great changes and the
finances of the empire were in good order as may be seen from the
fact that by the end of Mutawakkil's reign the central treasury
had a reserve of four million dinars and seven million dirhams.3
There is a report that Ahmad b. al-Mudabbir, who had been in
charge of the taxation department in Baghdad, was sent to "adjust"
the taxes of the Damascus and Jordan districts.4 We have no details
about this adjustment but later actions of this man in Egypt show
that he was a genius in devising new taxes.5 On the other hand there
is no indication of any change or unrest in these districts at this
time. It is possible that this adjustment was made necessary as a
result of a series of unusual natural disasters that happened around
that time. Violent earthquakes occurred almost everywhere in the
empire throughout the years 85 5-60/241-5. In Qumis 45,096 people
were killed in the earthquake of 856/242.' Some of the worst-hit
areas were in Jazira and Syria where whole towns were destroyed;
in 859/245 Antioch lost 1500 houses and on the Syrian coast the
sea level rose to a threatening height. Mutawakkil distributed three
million dirhams among the unfortunate people of Syria.7
We have some significant statistics that indicate the increasing
rate of conversion to Islam in 'Iraq. While the Arab Muslims of the
region of Basra paid six million dirhams in taxes a year, the non-
Muslims of Baghdad paid 130,000 dirhams and those of Wasit paid
only 30,000 dirhams in poll-taxes.8 With some simple calculations
1 2
Tabari, HI, pp. 1428-51. Athir, vol. VII, p. 4).
1 4
Mas'udi, Muruj, vol. VII, p. 227. Ya'qubl, Tarikb, vol. 11, p. 490.
»_ See below, p. 9}.
* Tabari, m, pp. 14)3-4.
T
Ibid., pp. 14)9-40; Ya'qubr, Tirlkh, vol. II, p. 49L
1
Khurdidhbeh, pp. 59, 125.
https://doi.org/10.1017/CBO9780511622380.005
80 THE ORIGINS OF THE tqfa'
it can be estimated that there were roughly 4000 non-Muslim families

in Baghdad and 1000 families in Wasit.
The murder of Mutawakkil at the hands of the military leaders
presaged a decade of unprecedented instability in the heart of the
empire. Muntasir was installed by the murderers of his father to do
their bidding and he was more than willing to concur. The army of
Samarra was reconstituted in the capital as it had been under
Mu'tasim, and all Mutawakkil's arrangements were declared invalid.
The meticulous system he had set up to rule over all his domains
without the cumbersome interference of the Tahirids was com-
pletely abolished, and they regained all their lost privileges. To
remove any remaining doubt Muntasir deposed his two brothers
from the line of succession in favour of his own child.1
Two men emerged as Muntasir's chief aids. The first, Utamish,
was one of the leaders of the Samarra army; and the other was
Ahmad b. al-Khasib, an administrator and a member of an 'Iraqi
land-owning family.2 The latter arranged for a solution to the
pressing problem of how to deal with the rival shdkiriyya army. The
decision was taken that it could be best deployed on the Byzantine
front, and Wasif, a prominent general from Samarra, was appointed
in charge of this operation. Significantly his chief lieutenant was an
uncle of Ibn Khaqan, the organizer of this army. Although our
sources represent Wasif's appointment to this task as an attempt by
Ahmad b. al-Khasib to get rid of a rival, it could not have been so.
There was neither the occasion nor the time for such rivalry to take
place. Their respective spheres of competence were completely
different, and time was too short to allow such rivalry to occur.
Furthermore, when Wasif was instructed to plan to stay four years
at his base in Malatya, he showed no sign of disagreement.3
It must have been a great shock to everyone concerned when
Muntasir died of a heart attack at the age of twenty-five after a reign
of only six months. Again a council was hastily convened to decide
on a successor. The desire to observe legality excluded Muntasir's
minor son and the unwillingness to have a strong ruler precluded
the choice of a capable candidate. With the implausible excuse that
the military leaders did not wish to have in power any son of
Mutawakkil, whom they had so recently murdered, a grandson of
Mu'tasim, who was barely eighteen years old, was selected and given
the title Musta'in 862-6/248-52.*
1
Tabarl, in, pp. 1485-95. * Ya'qubi, Tartkb, vol. II, p. 493; Tabarl, m, p. 1480.
1
Tabari, in, pp. 1480-1; Athlr, vol. vn, p. 72. * Tabarl, in, p. IJOI.
https://doi.org/10.1017/CBO9780511622380.005
THE ORIGINS OF THE iqtd' 8l
Utamish was elevated to the position of wa^lr, and Ahmad b.
al-Khasib was appointed kdtib.l In this arrangement there was still
a recognition of the principle of the separation of military and
financial affairs. Utamish, however, had different ideas about the
whole system of government, and decided that in order to put these
ideas into effect he should have supreme control of both. After being
in office for one month Ahmad b. al-Khasib was dismissed and
exiled to Crete; Utamish continued as waîr but appointed a kdtib
as his own administrative assistant.2 In this move lies the genesis of
the later office oiam'ir al-umard' and indeed, for all practical purposes,
Utamish held such a position though without this title, as it had yet
to be invented.3 He was determined to do everything in his power to
preserve the ascendency of the Samarra army, and consequently he
was not particularly keen on the continued existence of the shdkiriyya
army in any form. The problem acquired a certain urgency because
most of this army had returned to Samarra with Wasif after the
death of Muntasir. It was also compounded by the fact that other
military leaders, namely Wasif and Bugha, saw no reason why the
two armies should not be integrated into one powerful army. They
had worked with the shdkiriyya and found that with proper handling
and leadership they could only add to the total strength of the
military forces of the empire. For the moment a compromise was
reached by which some of the shdkiriyya were dispatched under the
leadership of Bugha to deal with some minor disturbance in the
Jibal region, while Wasif remained in Samarra with the rest.4
With remarkable professionalism Utamish proceeded to re-
organize the Samarra army in groups consisting of ten men, com-
panies of fifty, and battalions of a hundred, and all under appropriate
officers at various levels. He also declared that while all stipends were
assured and payment secured at regular times, all anomalies, inflated
family allowances and irregular increases, were annulled. Most
important, he proclaimed that no military personnel would be
permitted, under any circumstances, to take part in tax-collecting in
any form.5 Although this was a reasonable plan, it raised objections
from those affected by it. In less than a year Utamish was murdered
and all his decrees were revoked.6
At this stage Wasif and Bugha gained control of the situation. As
Ibid., pp. IJ02-3.
Tabari, in, p. 1508; Mas'udf, Murig, vol. vn, p. J14.
See below, p. 157.
Tabari, HI, pp. 1505-4, 1508, 1510-15; AthTr, vol. vu, pp. 79-81.
Tabari, in, p. 1799. * Ya'qubi, Tdrikb, vol. n, p. 496.
https://doi.org/10.1017/CBO9780511622380.005
82 THE ORIGINS OF THE tqtd'
they did not have the same notion of supreme control as Utamish,
a well-trained bureaucrat was appointed to take charge of the
administration.' Because of their own experience they were opposed
neither to stationing contingents of the army in the provinces, nor
to the still incipient military-administrative iqtd'. However they
became convinced that granting estates to military leaders would
only arouse bad feelings among their own men in addition to an-
tagoni2ing the general population, as indeed was the case in Kufa.
Baghar who had actually dealt Mutawakkil the last blow, had
been granted estates near Kufa and the result was a pseudo-Shi'ite
revolt there.2 When Bugha's allies, who were among the most
respectable men of the area and an early example of the military-
bureaucratic complex, tried to interfere with the affairs of the agent
of Baghar, the latter threatened mutiny.3 Bugha and Wasif had him
murdered and all hell broke loose.4 They lost any support they
might have had from the Samarra army and as they were supported
by the shakiriyya, an open conflict began between the two armies.5
Wasif used all possible means to assuage the feelings of the Samarra
forces, including releasing from prison the long forgotten son of
Afshin, but to no avail.6 Even the support of public opinion in
addition to that of the impotent Musta'in did not make much
difference to the declining position of Wasif and Bugha. Realizing
that their situation in Samarra was relatively weak they descended
on Baghdad where they hoped to have the support of the Tahirid
forces. Musta'in had no alternative but to join them.7
The conflict became one between Samarra and Baghdad. To their
credit the leaders of the former tried first to negotiate with their
adversaries, but the latter, confident in their strength, refused any
terms and began to mobilize all their forces.8 They called for
reinforcements from all the regions under their control and gathered
all the shdkiriyya in Baghdad. They also sent out instructions to the
various districts that the revenues should be carried to Baghdad
and not to Samarra.' Sima al-Sharabi emerged as the leader in
Samarra. The second part of the name of this man is an epithet that
expresses the real nature of the construction of the Samarra
army.
Sharabi is often taken to mean cup-bearer which is utterly wrong.
1
Tabari, m , pp. i ; i 2 , 1514, 1551. * Ibid., p p . 1 5 1 5 - 1 8 , I J J 5 .
1
Ibid., p p . 1 5 5 5 - 6 , 1 5 4 1 ; M a s ' u d i , Mitriij, v o l . VII, p . l i t .
* Tabari, m, pp. 1557-8. * Ibid., p. 1559.
* Ibid., pp. 1555. 155J. ' Ibid., pp. 1539, 1542.
* Ibid., pp. IJ4J-4- ' IM; PP- ijjo, 15J*.
https://doi.org/10.1017/CBO9780511622380.005
THE ORIGINS OF THE tqtd* 83
As it occurs frequently among the names of military leaders at this

time, we are asked to believe that the natural order of affairs then
was to promote favourite cup-bearers to the rank of generals. Of
course this fits in very well with the mistaken theories of slave armies,
but it also shows almost deliberate negligence in handling the source
material. It has been noted that this name occurs in various forms,
Sarbani, Sharbani, Sharabi, Sharbamyani and others which cannot
be clearly identified, thanks to copyists and editors. It is not methodo-
logically sound in this area however to take one form of a name and
leave all others out of consideration.' In any case, in this instance all
forms lead to a precise identification and that is Sharbamyani, i.e.
a follower of the prince of Bamyan; other forms are the result of the
common Arabic practice of shortening long names when forming
relative adjectives, nisbas, from them as, e.g., 'Abdi and 'Abqasi from
'Abdulqays.
Reference has been made to the presence of the prince of Bamyan
earlier in Samarra, and Sima was obviously one of the leading figures
among his followers.2 In due course he attained a commanding
position in the hierarchy of the Samarra army. Having emerged as
the leader of these forces he started planning for the inevitable fight
against Baghdad. A son of Mutawakkil was brought out of prison
and proclaimed a rival ruler with the title of Mu'tazz. The new
ruler's brother Abu Ahmad, the future formidable Muwaffaq, was
put in charge of the expected military operations; he was given the
assistance of a wasjr though the effective control of the bureaucracy
in Samarra remained in the hands of Sima's secretary. Sima himself
was put in charge of the postal service and the seal which under the
circumstances can only mean that he had direct access to and control
over correspondence with the provincial governors concerning the
all important revenues. Finally, another general was appointed to
assist, and probably watch over, Abu Ahmad, and the whole of the
army advanced on Baghdad; only Sima and a few men stayed behind
to hold the fort.3
They converged upon Baghdad, a siege was begun and the city
was soon suffering from food shortages. When occasionally fighting
broke out neither side gained an advantage over the other and Ibn
Tahir, the governor of Baghdad, began to realize the weakness of his
position. Once more the 'ayydrs appeared on the scene as the local
1
Ibid., pp. 1350, 1687, 1688, and the variant readings in the footnotes; Sourdel, p. 294.
1
See above, p. 64.
m, pp. 1544, 1550, 1594.
https://doi.org/10.1017/CBO9780511622380.005
population was recruited to fight the enemy but to no avail.' Abu
Ahmad, on his part, saw the stalemate and started making overtures
to Ibn Tahir.* It took the latter some time to overcome the opposition
in his own camp, and only after threatening to throw in his lot with
Samarra, was he able to open negotiations.3 Finally terms were
agreed; the shdkiriyya would continue as a separate force in Baghdad,
and in return, the Tahirid governor would be given one third of the
revenues of Samarra for their expenses.4 Mu'tazz the rival ruler was
accepted by all parties, while Musta'in was deposed and exiled to
Wasit.5 The fate of Wasif and Bugha was not so easily decided. On
the one hand Ibn Tahir did not want them to remain in Baghdad;
on the other hand they understandably dreaded returning to Samarra.
Their prestige among their former colleagues came to the rescue,
and it was decided that they should be rehabilitated. Wasif was
appointed governor of Jibal and Bugha governor of Hijaz, but of
course they were kept in Samarra, and only their representatives
were sent to their respective regions.6 However, as they could not
desist from intrigues to regain their lost power they practically
brought about their own murder.7
In Baghdad, Ibn Tahir, who had apparently asked for troops from
the East when his position had been in jeopardy, was now faced with
a difficult situation. Although these troops arrived long after the
fight had finished, they still required to be paid. He knew that if he
were to pay them, this could only be done at the expense of either his
old troops or the shdkiriyya. When he turned to Samarra for more
financial help, he was plainly told that it was his own problem, to
which he had to find a solution from his own resources. He tried to
stretch these as much as possible to satisfy all these men, but this
only added to his problems. The shdkiriyya did not see why their
interests should be sacrificed for the interests of new arrivals. Open
fighting started but the Tahirid forces prevailed and the shdkiriyya
quickly dispersed.8
Mu'tazz seems to have thought that with some manipulation he
could be master of his army and wield real power. With astonishing
boldness he arrested his brother Abu Ahmad who had just won him
* Ibid., pp. 1552, 1564, 1586, 1588, for the term 'ayjtdr; for the details of the righting,
pp. 15 50-1628.
J
* Ibid., pp. 1628-50. Ibid., pp. 1650-7.
* Ibid., p. 1640. » Ibid., pp. 1640, 1645, 1670.
* Ibid., pp. 1658-60. ' Ibid., pp. 1687-96.
* Ibid., pp. 1660-71, 1725-55; Athir, vol. VII, pp. 115-14, 156; Ya'qubl, TaiiMb, vol.
n, p. 501.
https://doi.org/10.1017/CBO9780511622380.005
THE ORIGINS OF THE iqtd* 85
a victory, and had helped to maintain his position in Samarra. He
also deposed his other brother Mu'ayyad, his heir apparent according
to the will of their father Mutawakkil. A few months later both
were banished to Wasit, and eventually ended up taking refuge with
the Tahirids in Baghdad. Less fortunate was the deposed Musta'in
who was murdered in exile in Wasit on the orders of Mu'tazz.'
In order to control the army, the first step was to try to win the
loyalty of its most powerful generals. For this purpose Mu'tazz
introduced a new feature in government, which was not quite
military-administrative iqta* though not too far from it. Each general
was appointed as the governor of a province to which he sent, as his
personal representative, one of his trusted subordinate officers
accompanied by some of his troops. Neither general nor representa-
tive was granted any estates and indeed they were strictly kept out
of any intervention in the fiscal affairs of the province, for which an
independent official was appointed. The significance of this system
was that, while it maintained the principle of the separation of
military and fiscal affairs, it allowed the governor in the capital a
priority claim over the income from his province as a guarantee for
the pay of his men. The best known example of this system is
Baykbak whose deputy in Egypt was Ahmad b. Tulun, who was
obliged to leave the fiscal affairs in the firm hands of Ahmad b.
al-Mudabbir, one of the most astute experts in such matters.2
The liaison between the generals and the administration was
through the office of the waîr, Ahmad b. Isra'il, whose job was to
supervise the collection of taxes and the disbursement of revenues.
He also had to satisfy the generals which proved to be a difficult
task, for very soon he fell out with one of them, Salih b. Wasif, who
did not hesitate to arrest the hapless wayir. At this point an open
struggle started between Mu'tazz and the generals. He wanted to
keep the military out of fiscal affairs as much as possible, while they
demanded full control of both military and fiscal affairs. The
bureaucrats also became divided between those who supported
Mu'tazz, and others who were willing to serve under the generals.
In the event the military won and Salih b. Wasif took charge of all
affairs. In no time Mu'tazz was deposed and replaced by one of his
nephews who took the title of Muhtadi.3
1
Tabari, in, pp. 1668-70, 169J; Athir, vol. vn, pp. n 6 , 122.
1
Tabari, m, pp. 1685, 1697; Maqrizi, Kbilat, vol. 1, pp. 99, 103; Ya'qubi, Tarikb, vol.
il, p. 508.
' Tabari, in, pp. 1647, 1706-11; Athir, vol. vn, pp. 112, 131; Ya'qubi, Tarikb, vol. n,
p. 504; Mas'udi, Murij, vol. vn, pp. 379, 397-8; Mas'iidi, Tmbih, p. 365.
https://doi.org/10.1017/CBO9780511622380.005
There was some hesitation in Baghdad about accepting the new

arrangements in the capital and paying allegiance to the new ruler.
This was not important in itself, for the prompt arrival of the
customary bonus, paid at the accession of a new ruler, overcame any
objections.1 But the fact that there was some support in Baghdad
for Abu Ahmad to replace his deposed brother, was a portent of
developments that were to take place very soon.2 Meanwhile things
did not remain quiet in Samarra. Another general, Musi b. Bugha,
who was on a minor policing operation in the Jibal region, was
alarmed by the developments in the capital and decided to go back
there. On his arrival, Salih b. Wasif simply went into hiding and
Baykbak, yet another general, took over his responsibilities. The
situation was thrown wide open, and the struggle that had started
under Mu'tazz was resumed all over again. Furthermore, Musa b.
Bugha who was extremely unwilling to accept the succession of
Muhtadi, wanted to depose him but did not or could not offer an
alternative.3
At last, the men of the army themselves grew tired of the endless
differences of their generals, and many of them decided that in their
own interest they should support Muhtadi in order to enable him to
impose a solution on the continuing crisis. Their support was
conditional on his meeting certain demands. This emphatically
confirms that their main concern was that they should receive their
stipends regularly and on time. These demands, that clearly explain
the causes of all the problems, were that:
(1) all affairs of government should revert to the control of the
Amir al-Mum'inin;
(2) the re-organization that had been introduced by Utamish, to
put the army on a professional basis, should be reinstated;
(3) all allowances for women and children should be stopped;
(4) stipends should be paid regularly every two months;
(5) no general should be allowed to interfere in any form in the
fiscal affairs of the provinces;
(6) all land-grants should be annulled;
(7) the Amir al-Mum'inin himself should be responsible for pro-
motions in the army;
(8) he should also appoint one of his brothers as commander-in-
chief.*
1
Tabari, m, pp. 1714-15.
• See below, p. 92. ' Tabari, nr, pp. 1756-40, 1787-96.
4
Ibid., pp. 1796-9; Athir, vol. vn, pp. 151-3.
https://doi.org/10.1017/CBO9780511622380.005
THE ORIGINS OF THE tqtd' 87
As Muhtadi was not able to meet all these demands, fighting broke
out again and an attempt was made to depose him. He called on
those men who had sought to support him and, with surprising
ease, had some of the generals arrested and Baykbak killed. However,
in the ensuing fighting Muhtadi was wounded and died after a
reign of only eleven months 869-70/25 5-6.' Under the circum-
stances, and while the moderate elements had the upper hand in
Samarra, a son of Mutawakkil was brought out of prison and
proclaimed successor with the title Mu'tamid.2 The choice of
Mu'tamid was apparently a compromise that put an end to the
fighting and paved the way for reaching agreement on other
arrangements which will be discussed in the following chapter.
While all these disturbances were taking place in the heart of the
empire, and in spite of the gradual weakening of the Tahirids,
remarkably little happened in the provinces to decrease the authority
of the central government. In Hims insurrections occurred in
863/248 and 864/250 but were easily suppressed.3 In Mawsil, the
people of the region elected their own governor but as they con-
tinued to pay allegiance to the ruler in Samarra, there was no cause
for alarm.4 Summer expeditions were regularly carried out against
the Byzantines.5 In Fars, a take-over by the local commander of the
Tahirid forces did not upset the province. Realizing the growing
power of Saffar's movement in Sistan, the central government
confirmed him as governor of Fars, and added the threatened region
of Kirman to his domains.6
The only serious problem arose in Tabaristan, where Musta'in
had granted land which had been public property to the Tahirid in
charge of the region.7 Naturally this aroused the wrath of the local
population. A man of 'Alawid descent who had taken refuge among
them, led their rebellion and was able for a time, in 864/250, to
drive the Tahirid forces out of the area and even occupy Rayy
itself. Another 'Alawid captured Qazvin and Zanjan in 865/251 while
the central government forces were barely able to protect Hamadan.
A strong contingent of the Samarra army under the command of
1
Tabari, in, pp. 1799-1810, 1815-34; Athlr, vol. vn, pp. 154-61; Ya'qubi, Tdrikb,
vol. ii, pp. 505-6; Mas'udi, Tanbih, p. 366.
1
Tabari, m, p. I 8 J I .
3
Ibid., pp. 1508, 1533.
4
Azdi, Mawul, p. 88.
5
Tabari, in, pp. 1534, 1577, 1581, 1615, 1697.
6
Ibid., pp. 1698-9; Ya'qubi, Tarikb, vol. 11, p. 495; Istakhri, p. 144.
7
Tabari, in, pp. 1524-5; Athir, vol. VH, pp. 85-6.
https://doi.org/10.1017/CBO9780511622380.005
88 THE ORIGINS OF THE iqti'
Musa b. Bugha was dispatched to deal with these rebels. Both
'Alawids were forced to flee into the mountains, and government
control was temporarily re-established in this region.1 When the
rapid developments in Samarra alarmed Musa enough to force his
return to the capital in 869/25 5, fresh troubles started in the region
and indeed in the rest of the empire.
1
Tabari, m , pp. 1526-);, 1585, 1645, 1686, 1693, 1698, 17)6.
https://doi.org/10.1017/CBO9780511622380.005
5
REGIONAL ECONOMIC CONFLICTS
The lack of viable political institutions to rule the empire and the
failure to establish well organized military forces to defend it had
brought some traumatic experiences to Islamic society. Although
these problems had yet to be solved they were pushed into second
place by economic crises. It was not that the government was
suffering from monetary deficits or lack of reserves; nor was it that
the general population had to endure poverty because of a deficiency
of natural resources. These were ample in most regions, and the
occasional famines that occurred were not of long duration. Indeed
there were clear signs of prosperity and progress in all walks of life,
whether in industry and commerce or arts and sciences, that could
have justified calling this era the Golden Age of Islam.
The economic problems of this affluent multi-racial empire were
the result of the cumulative effects of bad management in all fields,
at all levels and of all regions. The symptoms were the absence of
equal opportunities not only as between the various sectors of
society, but also as between the diverse regions of the empire. The
central government acted very much like a colonial power whose
only interest was to exploit its domains without regard to the
interests of its subjects. It was a classic example of a loosely associated
economic bloc where the central areas reap all the benefits while the
outer areas make all the sacrifices. The government saw all its duties
and responsibilities in terms of enforcing tax-collection, the revenues
of which were to support a growing and corrupt bureaucracy and
an almost useless army. When it came to public services the govern-
ment did not seem to consider that these fell within the realm of its
responsibilities. Even necessary repairs to the delicate irrigation
systems which had been previously paid for by the government were
now charged to the users who had to pay for them over and above
the required taxes.1 We are told that hospitals were built in major
cities, but the cost of maintaining the only hospital in Baghdad was
the dismal sum of thirty eight and one third dinars a day, and this
was to cover all expenses including the wages of the resident
1
Tabari, in, p. 2155.
[89] 4-a
https://doi.org/10.1017/CBO9780511622380.006
C)0 RliGIONAI. ECONOMIC CONFLICTS
physicians in addition to the cost of medicaments. The insignificance

of this hospital becomes clear when we know that the daily wages of
the men in charge of the hunt amounted to seventy dinars, and the
daily cost of feeding and stabling the ruler's horses was four hundred
dinars.' It should also be realized that bimdristdn, hospital, must have
been a euphemism for such an institution, which was more of a
lunatic asylum than anything else. It can be easily discerned that the
patients were from the more affluent strata of society, so that in
effect the government was relieving them of the burden of taking
care of their less fortunate kin.
The corrupt practices started by the Barmakids continued to
permeate all ruling circles and were perfected to an art. Members of
the 'Abbasid family acquired more and more estates as their numbers
multiplied with amazing rapidity. Government officials, many of
whom came from families with big landholdings, took advantage of
their appointments to enhance their own wealth and that of their
masters. They continually devised illegitimate means of evading
their taxes at the expense of the small landowners, many of whom
were thus forced to give up their holdings and join the various
protest movements which were sprouting in the heart of the
empire. Meanwhile, the flourishing commercial life of the cities
attracted the accumulated wealth of both the big landowners and
government officials and the interests of both groups became
completely intertwined. The order of the day was hoarding and
speculation in commodities by men in government service, and even
female members of the ruler's household were involved in such
operations. These malpractices were so widespread that they were
officially accepted as the norm, and accordingly every high official
was obliged to pay a monetary fine or even had all his wealth
confiscated at his dismissal. This most unusual method of taxation
could not have been completely effective since it did not prevent the
same official from seeking office time and again, happily prepared to
accept the consequences.
The wealth of the empire and the private fortunes of the rich
generated an unprecedented volume of international trade, especially
in luxury goods, with all known parts of the world. China, South
East Asia, India, Africa, Western Europe as far north as Scandinavia,
Eastern Europe, Russia and Southern Europe from Byzantium to
Spain were all partners in this trade. Imports into the empire
included silks and brocades; furs and skins such as panther, sable,
1
§abi, iV/târd', pp. 18-20.
https://doi.org/10.1017/CBO9780511622380.006
REGIONAL ECONOMIC CONFLICTS 91
grey squirrel, ermine, mink, fox, beaver spotted hare and goat;
shagreen; fur-caps; felt; all sorts of spices and aromatics; drugs;
musk; camphor; aloes; woods, especially ebony and teak; amber;
rubies; gold; tin; wax; honey; hazel nuts; coconuts; fish-glue; fish-
teeth; ink; paper; swords; armour; arrows; locks; gold and silver
utensils; crockery; trinkets; slaves and all kinds of animals such as
tigers, panthers, elephants, peacocks, falcons, swift horses, sheep
and cattle. Much of this trade was for local consumption as well as
for re-export. Foi export also were the products of the most
developed industry of the empire, textiles of every kind be it linen,
cotton, silk or wool, piece-goods or finished items of all sorts.
Practically every region and every town had its own brand of cloth,
clothes, tapestries, cushions, upholstery or carpets.
The shores of the Indian Ocean, Persian Gulf, Arabian, Red,
Black, Caspian and Mediterranean Seas were studded with busy
ports, all involved in imports and exports to and from various parts
of the world. The flow of this trade was greatly helped by the
development of a highly sophisticated banking system that allowed
payment by cheques or letters of credit anywhere, even in enemy
territory. New overland routes leading into and out of the empire
were chartered by intermediaries who found fresh opportunities to
bring in the products of far-away lands. The regions through which
this trade passed naturally had an interest and competed for a bigger
share in such commerce, according to their respective locations at
the importing or exporting end. These competing, and sometimes
conflicting, interests demanded some sort of harmonization, if not
complete integration, into a wider economic entity. As this was not
done the competing regions began to look exclusively after their
own interests; alliances were made, and efforts were spent to secure
the flow of trade into or out of a given region. Merchant navies
were built and control over inland routes was sought; the more
inlets a region had the better for its prosperity, and if it could also
have some outlets so much the better.
The central government added to the complexity of the situation
by imposing inter-regional taxes in addition to the taxes imposed on
international trade at the port of entry. A new office was created,
sahib tariq, an official in charge of a route, who was not only charged
with ensuring public safety, but also with collecting the taxes on
internal trade.' Of course this new tax increased the wealth at the
centre of this economic complex but it further separated the interests
1
Jabari, in, pp. 1889-20)9, 2106, 2107, 2114.
https://doi.org/10.1017/CBO9780511622380.006
92 REGIONAL ECONOMIC CONFLICTS
of the regions from that of the central government. Moreover, the

central government did not make much effort to rectify the most
damaging fault in the tax-system; the inequity of the burden as
between rural and urban taxes continued and even worsened, because
of the enormous wealth of the ever-expanding urban communities.
Although sporadic attempts were made to remedy this situation, as
we shall see, vested interests were too strong to dislodge. The
tension between rural and urban communities mounted and were
certainly exploited by revolutionary movements in the empire.1
The compromise that brought Mu'tamid to office in 870/256
involved his brother Abu Ahmad. It will be recalled that the latter
had played an important role in the struggle between the forces of
Samarra and Baghdad. He had also played an essential part in the
final reconciliation of the two sides. Although Mu'tazz, the bene-
ficiary of this action, did not seem to have appreciated the help of
Abu Ahmad and had banished him to Wasit, the latter's going to
Baghdad is an indication of his ties there. As part of their plan to
reform the military forces, the men of Samarra had demanded that
Muhtadi should appoint one of his brothers in charge of the army.2
It fell to Mu'tamid to carry out this wish and Abu Ahmad's military
experience and association with both Baghdad and Samarra made
him an ideal choice for this task. For the first time, the responsibility
for the military forces was taken away from an 'Abbasid ruler and
given to another experienced member of the family, who was neither
a son nor in the direct line of succession to the current ruler. Further-
more he was given a title, in the same fashion as the 'Abbasid
rulers, Muwaffaq, i.e. successful with the help of God. This situation
has led some scholars to describe him as "regent", but this could not
have been the case because Mu'tamid was neither a minor nor
incapacitated; he was for twenty-two years the Imam, Amir al
Mu'mimin and Khalifa of God. Muwaffaq's appointment was made in
response to the demand of the army to keep the generals under strict
control, and to ensure the separation of military and fiscal affairs
even at the highest level. Another part of the compromise was to
appoint as waîr 'Ubaydullah b. Yahya b. Khaqan, the organizer of
the shakiriyya under Mutawakkil. As he had worked with the
military before, he was acquainted with the problems involved. His
full co-operation with Muwaffaq helped to bring some stability to an
otherwise chaotic situation.3
1
See below, p. 116. * Sec above, p. 86.
HI, pp. 1859, ijij.
https://doi.org/10.1017/CBO9780511622380.006
It was becoming clear that the Tahirid power base in the East was
being eroded, their position in Baghdad and the western regions
under their control weakened. Muwaffaq was put in charge of these
regions over the Tahirids with Baghdad as his base.' This necessary
step aroused the apprehensions of some of the Samarra forces who
saw it as a resurrection of the forces of the rival city. Under the
circumstances, Mu'tamid was forced to intervene to alleviate these
fears. He appointed his son Mufawwad as his heir apparent and gave
him a share of power in Baghdad with Muwaffaq, who himself was
appointed second in line of succession.2 However, the death of the
co-operative vaîr deprived the government of his experience in
dealing with the military, and made way for further complications.
To replace him Mu'tamid appointed Hasan b. Makhlad b. al-Jarrih
whose choice was apparently an attempt to appease a faction of the
Samarra forces. It did not take long for another faction, led by the
powerful general Musi b. Bugha, to arrive at the capital, at which
Hasan simply went into hiding. His replacement was an interesting
choice, Sulayman b. Wahb, a very experienced administrator who
came from a long line of officials that had been in government
service since the Umayyad period. Of more significance was the
fact that his son 'Ubaydullah had the unusual function of being the
kdtib of Musi b. Bugha, Mufawwad and Muwaffaq at the same time.3
This was not because of a sudden scarcity of administrators; it was
an attempt to co-ordinate if not to combine the administration of
Baghdad with that of Samarra. This ingenious solution did not seem
to lessen the fears of the opposing faction of the Samarra forces, and
they forced Mu'tamid to dismiss Sulayman and re-instate their
favourite Hasan. Muwaffaq's reaction indicates the seriousness of
this move, for he marched on Samarra and forced his brother, after
a week of negotiations, to reverse his decision. The result was that
the militant faction of the Samarra army, led by Musi b. Utamish,
left the capital and scrambled for positions on the thughtir of Jazira.4
This was practically the end of Samarra and its forces, 877/264. It
does not serve any purpose to go into the details of their last struggle
to survive. Suffice it to say that Mu'tamid, who had lost all but his
title, was sent down to Wasit five years later in 882/269.*
Of course, the fall of Samarra allowed Baghdad to regain its
position at the centre of the empire, but this did not mean a return
1 s
Ibid., p. 1841. Ibid., p. 1890.
J
Ibid., p. 1915. * Ibid., pp. 1916-7
* Ibid., p. 2068.
https://doi.org/10.1017/CBO9780511622380.006
to the situation as it had been fifty years earlier. The events, policies,
changes and developments intervening had had their impact on
everybody concerned. After some jostling for positions the outcome
was the emergence of two distinct parties, each backing completely
different policies. Supporters of these parties permeated all levels of
leadership in society from the ruling family down to government
officials, military leaders, merchants and landowners. Perhaps the
best representatives of these parties were the Jarrah and the Furat
families. Members of these two families succeeded each other in the
most important position in government for generations. In a real
sense they were the leading politicians of their times who, when in
power, had their proteges and disciples appointed to subordinate
positions in government in order to propagate their policies. On the
right stood the Jarrahs who had a conservative approach to all
matters, especially with regard to taxation. They wanted the land
to bear the main burden of taxes, and the towns to continue in their
privileged position. They were also for the exploitation of the
regions to the advantage of the central government. In order to
achieve this end and still have firm control over the regions they
were, in varying degrees, in favour of the wide application of
military-administrative iqtdi in all corners of the empire. It is no
wonder that the first prominent Jarrah, Hasan b. Makhlad, had been
the favourite of the militant Samarra forces. The military-bureau-
cratic complex, born in the provincial administrations as a result of
the manipulations of Mutawakkil, had gravitated towards the centre,
and found its promoters among this party. However, at this stage,
neither the military nor the bureaucrats dominated each other, but
it was almost inevitable that in due course the former should get the
upper hand over the latter.
To the left were the Furats who were definitely against military-
administrative iqta* in any form and were keenly aware of the
dangers of such a system. In order to keep the military out of fiscal
matters the Furats were willing to do their utmost to assure them of
regular and reasonable stipends. They were also determined to put
an end to the corrupt practices of their colleagues in government
service. These reformist ideas can equally be discerned in their
approach tofiscalaffairs where they tried to acknowledge regional
interests vis-a-vis those of the central government. In the reactionary
milieu of Baghdad it was almost a compliment that they should be
accused of being Shi'ites, and it is an indication that they were inclined
towards greater equality in taxation as between urban and rural areas.
https://doi.org/10.1017/CBO9780511622380.006
During the life time of Muwaffaq, who proved to be a most
pragmatic statesman, the two parties worked together and their
policies were applied simultaneously, albeit in different regions. On
the one hand there is no indication that new taxes were imposed in
Baghdad at this time. On the other hand a most extensive tax
programme was carried out in Egypt by Ahmad b. al-Mudabbir who
was in charge of its fiscal affairs. Furthermore, the central govern-
ment re-affirmed his authority in these matters and his complete
independence from Ahmad b. Tulun, the representative of the
incipient military-administrative iqta'; and against the latter's
opposition even salted fish was taxed.1 At this stage, Muwaffaq's
policy was that the military should be strictly kept out of the fiscal
affairs of the central government in Baghdad, and that all matters
whether military or fiscal should remain in the hands of a civilian
under his own firm control. For this task he appointed an efficient
administrator, Sa'id b. Makhlad, as his own assistant, kdtib.2 In the
regions, Muwaffaq tried to maintain the separation of military and
fiscal affairs as much as possible. However, he tolerated the applica-
tion of full military-administrative iqta* in some areas as it was the
only way to support government control there. This meant that some
trusted military commanders were appointed as governors of certain
regions with complete authority over all matters including fiscal
affairs, and in exchange, they undertook to pay fixed sums to the
central treasury.3 Meanwhile, he had to meet serious troubles
resulting from the breakdown of Tahirid power in the East and the
stand of some regions in defence of their own interests.
The first problem was in Sistan where the Saffarids had now
consolidated their strength. The origin of their movement has been
explained as the action of bands of brigands who, after establishing
themselves in their region, were finally able to take control of a
great part of the empire. We are told that they were Kharijites, yet
there is no indication whatever of Kharijite thought in their move-
ment. The valiant efforts of C. E. Bosworth do not prove any
tangible connections between the Saffarids and the early Kharijites
of the Umayyad period; and heresiographers' evidence should not
be taken as more than an attempt at making neat classifications of
sects and groups without regard to history.4 Indeed the descendants
1
Maqrfcl, Kbitat, vol. 1, pp. 103, 107, J I 6 ; Kindl, Wulit, p. 214; Ya'qubi, TarlAb, vol.
11, p. 508.
1
Tabari, in, p. 1930; Athlr, vol. vn, p. 265.
' Tabari, in, pp. 20)9, 2048, 2114, 2115; $ibi, Wu?ara\ p. n .
4
C. E. Bosworth, Sistm Under Tbt Arabs, Rome, 1968, pp. 37-42, 87-91, 109-23.
https://doi.org/10.1017/CBO9780511622380.006
of the early Kharijites were, at this very time, literally millionaires

and strong supporters of the ' Abbasids.' The sources that describe
the Saffarids as Kharijites use this term in its widest sense, simply
to mean rebels against the established government. This is exactly
what the Saffarids were.
It is important to realize that this movement took place in western
Sistan, one of the least inviting and most neglected regions of the
empire. Although it was first conquered in 652/32 Arab rule there
had never been very secure. This was due to the failure of the
conquerors to make any progress eastwards in the kingdom of
Zabulistan.1 For more than two centuries this kingdom had defended
its independence with great zeal against the repeated attempts of
the Arabs to gain a foothold in its territories. This resistance had
forced the Arabs to by-pass the region and concentrate their advance
eastwards through the coastal areas in the south into the province of
Sind where they had been able to establish a strong presence. Even-
tually, they were probably more than glad to leave Sistan to the
care of the Tahirids. The latter did not show any particular interest
in the region more than the usual obligation to defend it against
outside attacks. In fact we are told that they, the Tahirids, were
traditional rulers whose main concern was with the long established
families in their regions.3 In other words they contented themselves
with enforcing the treaties of capitulation concluded at the time of
the conquest with the dihqdns there. As these dihqdns were by defini-
tion the big landowners it is possible to conclude that the economy
was mostly based on agriculture. Although the east-west trade
route across Asia continued to bring some of this valuable trade to
Soghdiana whence it was carried west through Tabaristan to
Baghdad, a good deal of this trade had begun to find its way by sea
from China to the Persian Gulf and the ports of the Arabian Sea. At
the same time the trade route from Soghdiana to Balkh and then
across the Hindu-Kush into India had been obstructed by the
formidable power of Zunbil the king of Zabulistan. The conspicuous
absence of contacts between Sind and Khurasan suggests that an
economic boycott was mounted against the obstinate enemy. Such
a boycott would certainly have its repercussions in the neighbouring
region of Sistan and the Tahirids with their traditional concern
would not be disturbed enough to take measures to alleviate the
effects of the economic slump there. Under the circumstances a
p. 142. ' Shaban, 'Abbdsid Revolution, pp. 23, 28, 54.

1 Athir, vol. viit, p. 5.
https://doi.org/10.1017/CBO9780511622380.006
breakdown of public order was inevitable, and as a result of the
weakening power of the Tahirids the situation went out of control.
Bands of men spread all over the region either pillaging or demand-
ing protection money from their more fortunate countrymen. The
euphemism Kharijites is perfectly suitable to describe these men. On
the other hand the settled communities, in self-defence, began to
form their own militia which came to be known to us as 'ayydrs,
Home Guard, or mutatawwi'a, volunteers; and clearly there is no
contradiction between these two terms. As some of the bandits were
persuaded to join this militia the term Kharijites was used to
describe what was basically the same people. Accordingly, there is
no sectarian significance whatever in this episode, and this is
proved by the fact that after the quick eclipse of the Saffarids we do
not hear about any traces of Kharijism in Sistan.
The Saffarids, then, originated from these bands of men who went
through a process of change from outlaws to militia. The attain-
ments of one or other of the leaders of such groups must end by
giving him authority over his peers and in due course the talents
for leadership of the Saffarid Ya'qub were recognized in this way.'
He was able to organize his followers into a strictly disciplined
military force. This organization may explain the use of the term
'ayjdr in this situation. An able recruit who was willing to join
Ya'qub had to give up all his personal belongings which were
sold; the returns went to the treasury but were credited to the man.
He was then given all necessary equipment which, however, remained
the property of Ya'qub. As all his personal needs were taken care of,
such a soldier did not receive any stipends while in service. If he
were honourably discharged he was given his accumulated earnings,
otherwise he had to return all the equipment, receiving no money at
all except what had been credited to him at the time of recruitment.*
With such an organized army Ya'qub was able not only to establish
his uncontested control over Sistan but also to invade Rukhkhaj
and finally bring down the stubborn kingdom of Zabulistan,
870/256.3 Significantly the following step was to push down to the
Makran-Kirman coast and, as well put by one source, to control the
extremities of Sind and Hind.4
This coastal area, now known as Baluchistan, is even more arid
than Sistan, and any effort to expand in this direction would seem
1
Ijtalchri, pp. 146-7; Hawqal, pp. 419-20.
' Mas'udi, Murij, vol. VIII, pp. 47-9.
1
Athir, vol. VII, p. 226; '{atari, in, p. 1883. * Ijukhri, p. 247.
https://doi.org/10.1017/CBO9780511622380.006
certainly wasteful. Yet this coast and its extension to the Persian
Gulf had suddenly become the bone of contention between various
parties. We recall that the 'Abbasids had long been aware of the value
of the revenues from the taxes on the trade coming through the
southern shores of the empire. We also remember that Mu'tasim
had found it necessary to send land and sea expeditions to the
remotest parts of these coasts to remove any threat to the flow of
this prospering trade. The central government continued to con-
solidate its control over this trade which came mainly through the
ports of Suhar, Ubulla, Basra, Siraf and of course Daybul at the
mouth of the Indus River in Sind. The arid Kirman-Makran coast
between Siraf and Daybul was not considered worth the trouble of
a similar effort there, but this was where private enterprise came to
the rescue. Some of the seafaring Arabs from the Azd of 'Uman
moved into these areas, acquired good landing sites, built citadels
and established themselves at suitable points along this coast. Ships
unloaded their goods at these private harbours in return for a certain
fee which was probably lower than the usual tax-rate.1 To the
advantage of these entrepreneurs the central government did not
see any justification for interference especially as it did not have
firm authority in the area. Some of these Arabs, interestingly enough,
were called Saffarids. It is not suggested here that the Saffarids of
Sistan were of Arab origin, though the possibility cannot be totally
excluded. It may be too much of a coincidence that they should have
a two year war with their namesakes on the coast.2 It is not a coincid-
ence, however, that part of the population of wealthy Fars were also
after the control of arid Kirman. Sensing the danger of Saffarid
designs and realizing that no help was forthcoming from the central
government, these people decided it was time to defend their interests
and took matters into their own hands. Led by Muhammad b.
Wasil, they defied the government, deposed its representative,
started the attack on the Saffarids, joined forces with Baghdad
against them and continued the fight even after the government had
reconciled itself to accepting the terms of their common adversary.3
It is a remarkable fact that Muhammad b. Wasil was of a genuine
lineage of Umayyad Kharijites who had long since settled in Istakhr
in the heart of Fars, where they and their descendants had eventually
1
Ibid., pp. 116, 117, 140-1; Yaqut, Bulddn vol. n, pp. 543-4, vol. m , p. 298; vol. iv,
P-444-
1
I?takhri, p. 141; Yaqut, Bulddn, vol. 11, pp. 545-4.
J
Tabari, HI, pp. 1859, 1858, 1859, 1887-90, 1908, 1912.
https://doi.org/10.1017/CBO9780511622380.006
acquired tremendous wealth.' Of course, heresiographers and certain

modern scholars do not pay much attention to complicated details
which might interfere with their simplistic outlook.
It cannot be over-emphasized that at the time of the rise of the
Saffarids there were many other regional uprisings. Although there
may be no apparent connection, the Saffarid movement cannot be
understood in isolation from the others. Another most important
movement at the time was the Zanj revolt. It is a sad comment on
research in Islamic history that, in spite of the proximity of the
territories where these two movements took place, no attempt has
been made to examine their relationship, and instead efforts have
been wasted on examining imaginary relationships with movements
whose force was long spent. It is a curious fact that the two move-
ments never made any attempts to ally themselves against their
common enemy, the central government, and instead actually
fought each other.2 This seemingly illogical behaviour can only mean
that each movement considered the elimination of the other to be of
advantage to it. As we will soon see, their origins were different
but their objectives were not entirely dissimilar - control of a
longer stretch of the shores of the Arabian Sea and the Persian
Gulf.
The rise of the Saffarids brought about the final collapse of the
Tahirids in the East. With the exception of Soghdiana, in almost all
other districts local strong men tried to emulate the Saffarids and
gain control of their own regions.3 It was practically impossible for
the central government to cope with this situation by its own
military forces alone. The presence of the Samanids, the faithful
proteges of the Tahirids, in Soghdiana and the stability of their rule
in this region offered a viable solution to the government's predica-
ment. They were acknowledged to be directly in charge of their
region and its vicinity down to the River Oxus. Ya'qub the Saffarid
was given charge of the rest of the Tahirid domains to the south of
the river, in addition to a subsidy from the central government to
enable him to carry out the fight against the rebels.4 To establish his
authority in these troubled regions, Ya'qub knew that he first had
to deal with the Shi'ite Hasan b. Zayd who was actively trying to
re-assert his position in Tabaristan with the help of the local popula-
tion. As this proved increasingly difficult and as he was faced with a
hostile population in every direction, Ya'qub realized that he was
1
I$takhri, p. 142. * Tabari, m, pp. 1914-15.
' Ibid., pp. 1874-5. i*8j, 1931, 2017, 2039. « Ibid., p. 1841.
https://doi.org/10.1017/CBO9780511622380.006
IOO REGIONAL ECONOMIC CONFLICTS
at the losing end of the bargain with the central government.1

Furthermore, he was actually shielding the Samanids, the real
winners in the whole situation. He decided to turn back and pursue
his plans in the south. The central government tried to negotiate
with him and entice him to co-operate by offering him some nominal
appointment in Baghdad. He over-estimated his bargaining position
and asked for too high a price which the government could not
have given even if it had agreed to it. Ya'qub demanded authority
over the Samanids in order to obtain their help against the rebels.
When negotiations failed he boldly advanced on Baghdad, but again
this was a gross over-estimation of his own power and Muwaffaq
had no difficulty in inflicting a severe defeat on him not too far from
Baghdad in 875/262.2 Turning back, Ya'qub tried to demonstrate
his rule over Firs. However the resistance of the local population
led by Muhammad b. Wasil, a defeat by the Zanj, and trouble in his
own backyard put an end to his ambitions.3 Yet, after his death in
879/263 his brother 'Amr was able to negotiate with the central
government and secure the terms Ya'qub had sought. This was
because the Samanids were not doing anything to check the efforts
of former Tahirid proteges and remnants of their forces to regain
the domination of Khurasan.4 The Shi'ites of Tabaristan took
advantage of this situation to extend their influence in this region.
To the disappointment of the central government 'Amr failed to
make any progress against these rebels especially since they had
joined forces in opposition to him. On the other hand this very
failure, in addition to the mounting pressure from Tabaristan,
persuaded the Samanids to act. With a little effort they restored
order in these regions and drove 'Amr out of Khurasan.5 Thus they
replaced their patrons, the Tahirids, as masters of the whole East, while
the Saffarids continued for centuries in their little enclave in Sistan.
In the meantime the other movement in southern 'Iraq had been
gathering momentum to the extent of becoming a real threat to
Baghdad. This was the Zanj revolt, one of the most misunderstood
episodes in Islamic history. Since Noldeke first wrote about it almost
a century ago and introduced the notion that it was a slave revolt,
this idea has prevailed and has never been reconsidered.6 Although
he was writing only " sketches ", these outdated misconceptions have
1
Ibid., pp. 1840, 1880, 1885-5, 1940-1, 1993. * Ibid., pp. 1891-4, 2113.
> Ibid., pp. 1895-6, 1908, 1914, 1993, xi 12. * Ibid., pp. 1931-2.
» Ibid., pp. 1941, 1993, 2039, 2123; Athir, vol. vn, pp. 256-7.
» Th. Nfildeke, "A Servile Wat in the East" in Sktttbts from Easttn History, tr. by
J. S. Black, Edinburgh, 1892, pp. 146-7;.
https://doi.org/10.1017/CBO9780511622380.006
REGIONAL ECONOMIC CONFLICTS IOI
been slavishly regurgitated by modern scholars. Perhaps the tempta-
tion of the romantic idea of a slave revolt in a slave-ridden society is
too much to resist, or that wading through the considerable amount
of valuable material in the sources is thought to be an unnecessarily
cumbersome task. As it happens this is one of the best reported
events of the period and by no lesser authority than Tabari himself,
who witnessed some of its developments and quotes directly the
accounts of some of the participants, notably Muhammad b. Hasan
b. Sahl, who was a leading figure in the revolt. •
It was not a slave revolt. It was a %anj, i.e. a Negro, revolt. To
equate Negro with slave is a reflection of nineteenth-century racial
theories; it could only apply to the American South before the
Civil War. The facts are that in Islamic society there were white as
well as black slaves and that slave labour was not a factor in the
economy as it had been in Rome. In the Islamic world slaves were
mostly employed in domestic housework and of course as concubines.
All the talk about slaves rising against the wretched conditions of
work in the salt marshes of Basra is a figment of the imagination
and has no support in the sources. On the contrary, some of the
people who were working in the salt marshes were among the first
to fight against the revolt.2 Of course there were a few runaway
slaves who joined the rebels, but this still does not make it a slave
revolt. The vast majority of the rebels were Arabs of the Persian
Gulf supported by free East Africans who had made their homes in
the region. Along with such Negroes there were Arabs from the
clans of Bahila, Hamdan, Iyad, 'Ijl, Qays, 'Abdulqays, and the numer-
ous clans of Tamim were strongly represented.3 Furthermore, this
Arab-Negro alliance was well represented in the leadership of the
revolt.4 The founder of the movement was undeniably an Arab
whether he was of 'Alawid descent as he claimed, or from 'Abdul-
qays as his opponents asserted.s More important is the fact that
among the leaders we find Negroes as well as Arabs and the astound-
ing fact that the latter were from the 'Abbasid "establishment"
itself. Mention has already been made of Muhammad b. Hasan b.
Sahl, who, though not exactly an Arab, was none other than the son
of one of Ma'mun's closest associates.6 But the various Muhallabite
Tabari, m , pp. 1742-2098. • Ibid., p. 1752.

lbid.,pp. 1745,1744. 1746, 1 7 5 ' . 1756. "759. >7<>°. ' 7 7 ° . >777. 1850, 19°*. ' 9 ° S . »9*5.
1998, 2016; Athir, vol. VII, pp. 202, 204, 248, 255, 283; Mas'udi, Tanblb, pp. 3 9 2 - ) .
Tabari, m , pp. 1751, 1763, 1766, 1769, 1778, 1779, 1095.
Ibid., pp. 1742-3; Mas'udi, Miirij, vol. v m , p. 31; Ibn Hazm, jambara, p. 57
Tabari, i n , pp. 2135-6.
https://doi.org/10.1017/CBO9780511622380.006
leaders represent a prestigious Arab line who had served the

Umayyads and the 'Abbasids with great distinction in many capaci-
ties.1 Even Jews were among the supporters of the revolt.2 If more
proof is needed that it was not a slave revolt, it is to be found in the
fact that it had a highly organized army and navy which vigorously
resisted the whole weight of the central government for almost
fifteen years. Moreover, it must have had huge resources that
allowed it to build no less than six impregnable towns in which there
were arsenals for the manufacture of weapons and battleships.3
These towns also had in their mammoth markets prodigious wealth
which was more than the salt marshes could conceivably produce.
Even all the booty from Basra and the whole region could not
account for such enormous wealth.4 Significantly the revolt had the
backing of a certain group of merchants who persevered with their
support until the very end. Tabari makes it very clear that the
strength of the rebels was dependent on the support of these
merchants.5 Under the circumstances it is hardly surprising that the
rebels did not have a concrete ideology, albeit they had a purpose
which certainly was not the freeing of the slaves, for they continued
to have their own.
The bone of contention was the African trade, and this issue had
its effects not only on the Persian Gulf but also in all the regions of
North Africa, Egypt, the Red Sea coasts, Syria and even the tbughur
on the Byzantine borders. The demand for African products was not
a new feature of this era. Carthage had made its fortunes by trading
in these products, the Abyssinians had sent them to Byzantium and
the Makkans had greatly benefited from this trade. The demand for
these products had never diminished, but the routes of the trade
had changed because of wars, political situations and the interests
of middlemen. Before the Arab conquest of North Africa the
Byzantines had control of all the coast, and the trade had been easily
carried north across the Sahara by many routes. New excavations
at Garama in Libya testify to the existence, before Islam, of a great
trade centre there, almost halfway between the terminals of the
shortest north to south route from Chad to Tripoli.6 The conquest
and the subsequent century of revolts, unrest and fighting had put
an end to the flow of this trade to the coast.
Ibid., p. 1745; Athir, vol. vii, p. 285. ' Tabari, m, p. 1760.
Ibid., pp. 1878, 1899, 1932, 195$, 1971, 1974, 1042-$, 2062.
Ibid., pp. 1959, 1964, 1966, 1971, 1975, 2032, 2053, 2055-6, 2068, 2078.
Ibid., pp. 1783, 1835; Athir, vol. vii, p. 262.
M. S. Ayyub, Jerma, Tripoli, 1969, pp. 91-127.
https://doi.org/10.1017/CBO9780511622380.006
REGIONAL ECONOMIC CONFLICTS IO3
When the 'Abbasids came to power and the Aghlabids established

their rule over Tunisia - then called Ifriqiyya - 800/184, this region
like Libya and Egypt continued its basically agricultural economy,
and no attempt was made to revive the long-forgotten trade. But
the Berbers, the indigenous population of North Africa, remember-
ing its lucrative profits, were now free to resume this activity. This
manifested itself in the rise of new political entities in north-west
Africa; the Rustamids in what is now western Algeria in 777/160;
and the Idrisids in what is now Morocco in 789/172. These small
states were little more than trading companies formed at the ter-
minals of well-known trade routes across the Sahara. The Rustamids
were Kharijites and the Idrisids Shi'ites, the main difference in
economic policy between them being their respective systems of
taxation. The Kharijites were for a measure of laissez-faire, i.e. lower
taxes on trade. They obviously needed the intermediaries to bring
the trade to them and in such a case the imposition of taxes would be
counter-productive.1 The Shi'ites favoured a higher rate of taxation
either because some local minerals were involved or because they
had better access to the trade that came to them along easier routes.2
Another difference between the Shi'ites and the Kharijites was the
fact that among the former there was a noticeable Arab element
while the latter were almost purely Berbers. It is worth noting that
while Egypt and Tunisia were under the firm control of the 'Abba-
sids and the Aghlabids, Tripolitania, which was at the terminal of
the shortest but most difficult route across the Sahara, remained a
strong bastion of the Kharijite Berbers.3
It should be pointed out that the most guarded trade secrets of
the time were the places where the product originated and the routes
across difficult terrains. Intermediaries knew only too well that these
secrets constituted all their know-how, and accordingly many
practices and customs were developed to preserve them. In some
areas foreigners were not allowed to cross certain territories although
they were welcome to trade at market places established for this
purpose close to the borders.4 In other places goods were brought
across a river and left there over-night to enable the foreign trader
1
Hawqal, pp. 67-9, 155-5; Maqdisi, p. 29; Ya'qubi, Bulddn, pp. 252, 345; Bakri,
Al-Mu&krib . . ., cd. Dc Slanc, Paris, 1911, pp. 68, 1)9.
1
Ya'qubi, Bulddn, pp. 356-7; Bakri, pp. 76-7, 161; 'ldhari, vol. 1, pp. 232-3; see also
my forthcoming study on Spain and North Africa.
> Kindi, UPulot, p. 222; Maqrizi, Kbitat, vol. 1, p. 320; Ya'qubi, Buidan, p. 352; Hawqal,
pp. 67-9.
4
Maqrizi, Kbitat, vol. 1, pp. 190, 199; Hawqal, p. 51.
https://doi.org/10.1017/CBO9780511622380.006
104 REGIONAL ECONOMIC CONFLICTS REGIONAL ECONOMIC CONFLICTS 105
(
s
Tangier i
Mariyya
^_^——
Mai
•^ r
r
Wuhrin
^F -
T
u \
til
•J^VValila Naki / TakrTt

MEDITERRANEAN SEA
/
c
\ 1 /
V /
^•Marrakish \ / y ^-—s^
t ' / f \ An
),'(
\ / / = ^ i SYRIAN
I / / x
V / Bajra-V
/ y< N
/
S
r
//
^ \ Sijilmas* If hbub Oisit^ _jf^ e - .^ TV M Aylah v ^ ~ \
/ / >
r \ /
/ 1
1
;
/
/ / /
^ / / 1 / \ F rTra i^V T /
*
y 1
* • *
i f 1 Garam* • " • ^
i \ 1 Dikhla OasisJtw _ , ±
/ 1
/ / 1 TM
/ 1
J
i
\
\ ^ — " • " " "
•Al w i n \ v i /
1
1
I /
( / 1 kks
/
/ f
/ \f
1
/ "'} " J" / If \ n
H
w A R N
\i
i A; D E S m
R T / ^» S\ \ RED)A >
v\ x AL-RUB'al-KHALl / /
i
/ y \ \SEA \ / /
r 7 imbuktu ^ U ^ Najrin
l \v »
" \
>
\ 'n Z A \ G 'H A W A /'^
V. \
1
V N —L --" \ vv^
^ i — »—'
\ / \
J J V-O
\
\
G * s
1
\
\
f
1
/ 1 INDIAN OCEAN
a Trans-Saharan caravan routes
Downloaded from https://www.cambridge.org/core. Cambridge University Main, on 26 Apr 2018 at 05:52:47, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511622380.006
io6 REGIONAL ECONOMIC CONFLICTS
to leave what he considered a fair amount of his own goods in

exchange for what he desired. In other words secrecy was observed
to the extent that bargains were struck and goods changed hands
between partners who did not actually meet face to face.'
Arab travellers and geographers, who have left us detailed
descriptions of all the trade routes within the domains of Islam,
give us as much information as possible about the feasibility, use
and extent of outside routes from which they were often themselves
excluded.2 The picture that emerges is of a wide network of routes
which spread over the northern half of Africa not only from north
to south but also from east to west. Admittedly most of these routes
passed through difficult terrain under extremely hazardous condi-
tions and were usable only for certain times of the year. All the same
the incentive of profits from the trade made it worthwhile to take
these risks. With the help of expert guides who knew how to sniff
out sources of water and avoid the perils of these barren areas,
travellers were able to go through the various stages of their journey
from one destination to another. The north to south routes were
well known but the east to west were the most difficult to chart
because they passed through many lands under the control of many
peoples, none of whom wanted to divulge their secrets. This
difficulty is compounded by the fact that our sources use various
names to indicate the same groups of places and, to add to the
problem, these names can be confused with other things since
ancient differs from modern usage, e.g. the Nile and the Sudan.
Therefore, to avoid this confusion and for the sake of clarity,
modern names will be used here to describe the several stages of
these routes. It was possible to start from any point on the Atlantic
coast of West Africa, be it in Senegal or Ghana, and advance to
Kano in Northern Nigeria, and from there to Chad. Here the
traveller had a multiplicity of choices; he could go north across the
Sahara to Fazzan and then to Tripoli; or he could advance north-east
to the Oasis in the Egyptian desert, and thence either to the Nile
valley or the Egyptian-Libyan coast. An easier choice from Chad
was to proceed slightly south-east in order to skirt the desert and
then almost straight east to Darfur in the Sudan. From here it was
1
Marvazi, p . 49.
1
Yaqut, Bu/Jan, vol. m , pp. 142, 159-60, 261, vol. i v , pp. 260, 4 ) 2 , 486, vol. v , pp. 242,
592; Khaldun, pp. 94, 9 8 - 9 ; Ya'qubi, BuUdn, pp. 334, 3 3 ; , 345; Maghribi, Jugbrdfijd,
pp. 9 1 , 9 6 ; Mas'udi, Tanbib, p. 331; I$takhri, p. 4 ; ; Hawqal, pp. 15, 16, 60, 6 1 , 9 2 ,
101, 155, 154. M 5 i Bakri, pp. 9, 14, I J , 180, 182; Maqrizi, Kbitat, v o l . 1, pp. 4, 182,
193-7, vol. " . PP- *57. 163.
https://doi.org/10.1017/CBO9780511622380.006
an easy journey into Nubia and then north to Egypt or east to the
ports of the Red Sea. From 'Aydhab, the most important of these
ports at the time, it was an easy crossing to the ports of Hijaz, which
were only a short distance from Makka and Madina, and then
ultimately to Baghdad. Of course, it was also easy to sail from
'Aydhab to any other port on the East African, Arabian and Persian
Gulf coasts. All the peoples along these routes were interested in the
profits from the trade passing through their territories, and although
relations were usually good, conflicts frequently arose. Piratical
activities in the Persian Gulf, increasingly vicious attacks on trade
caravans at all points en route from Makka to Baghdad by impover-
ished nomads, the sudden rise of antagonisms between Makka,
Madina and Jidda, and above all the Zanj revolt, were all con-
temporaneous manifestations of such conflicts.'
In the case of the Persian Gulf the seafaring Arabs of the region
had an astonishingly wide range of international relations. They
had an established colony as far east as Canton; and an enterprising
merchant from Siraf managed to overcome all the barriers and pay a
visit to west-central Africa.2 Arab relations with East Africa since
antiquity need no proof. Over the ages many Arabs have settled
and become part of the population there, and although we do not
know when this process started, there is no reason to believe that it
was a one-way practice. On the contrary, the secrecy on the African
side, the open-door policy of the empire, the expansion of trade and
the clamour for African goods, certainly stimulated the setting-up
and growth of East African colonies in all the trade centres of the
Gulf. Many of these immigrants became Muslims and established
close relationships with the indigenous population of the region.
Others were not assimilated to such an extent, but their interest in
trade would have been stronger owing to their bonds with their
native lands.3 Such a situation gave these people and their partners
in the Gulf a virtual monopoly over the importation of African goods
into the area. Any outside competition in this field through newly
activated trade routes such as that via 'Aydhab would certainly
represent a very serious threat to the prosperity of these monopolists.
To add to their distress a change in the government's taxation policy
was put into effect at this time. We remember that early in Wathiq's
reign 847-61/227-31, taxes on trade coming through the Gulf ports
T . m, pp. I9)i> 1941,1008, 2025, 2026, 20)9.
> MarvazI, p. 42.
l8
, 1744, «75«. 175*. «757. '75*. 176J-4. >766. 1769, «77»-*. 37-
https://doi.org/10.1017/CBO9780511622380.006
IO8 REGIONAL ECONOMIC CONFLICTS
had been abolished. The new taxation policy of MuwafFaq did not
allow such leniency and, if the new taxes imposed in Egypt were a
guide, it can be easily perceived that stringent new taxes were re-
imposed on all these goods, and considering that they were imported
from non-Muslim territory, the rate could have been as much as
20%.' The appointment of Ibrahim b. al-Mudabbir, the brother of
the organizer of the new tax system in Egypt, to supervise tax
collection in Basra is a vivid attestation to the tax-collecting methods
there.2 Any merchant who is suddenly confronted with such a tax
in addition to a drastic decline in his trade, cannot be expected to sit
back and hope for the best. The Arabs of the region whose interests
were certainly involved and who, under similar circumstances
earlier, had begun to attack trade caravans would surely rise again
in defence of these interests. Without this combination of wealth
and manpower, the Zanj revolt would not have been possible.
With remarkable efficiency and expedition the rebels swiftly
established their control over most of the Persian Gulf coast, and
extended it inland to secure their food supplies. Special vehemence
was reserved for the port of Basra which they practically destroyed.
Their choice of sites fo rtheir own new towns and their meticulous
knowledge of the intricate waterways of the region in addition to
their great skill in naval warfare were all utilized to strangle the
Basran economy and drive all the in-coming trade through their own
channels.3 Wasit, the major bottle-neck on the way north to Baghdad,
was completely cut off from any road or waterway leading south to
the Gulf coast. Furthermore, the rebels occupied Kufa in order to
secure the alternative inland route to the north.4 They expelled
government forces from all these areas and easily withstood the
onslaught of the successive expeditions that MuwafFaq sent against
them. Realizing the grave dangers of this situation, he decided to
mobilize all his financial and military forces against this audacious
enemy. For the army he introduced yet another re-organization which
was in effect an amalgamation of previous measures. It is characteris-
tic of his pragmatism that his plans took into consideration all
aspects of the existing conditions. Perhaps the most important step
was that he was able to integrate most of the military forces in such
a way that racial differences were set aside. This involved re-grouping
1
See above, p. 95; Sabi, Wuâra', p. 248.
» Tabari, in, p. :8}8; Abbar, p. 162.
J
jabari, in, pp. 1767-8. '77*. "776. 1783, 1835, 1836, 1848-9, 1855-6, 1866, 1991,
2032, 20J3, 2096. • Ibid., pp. 1883, 2013.
https://doi.org/10.1017/CBO9780511622380.006
the men, placing non-Arabs under Arab commanders and vice

versa.1 This simple device put an end to the internal squabbles
within the army. All that remained to ensure its stability was to
secure the financial resources in order to guarantee to the men
regularity of their stipends. For this purpose Muwaffaq allowed for
the first time the full application and formal institution of military-
administrative iqtd1. As explained before this meant that military
commanders were appointed governors of certain regions and given
complete authority over the all important fiscal affairs there and, in
return, they assured the central treasury of the payment of an agreed
sum from the revenues of their respective regions. This was applied
in areas where government rule needed to be more firmly established,
or in areas such as Egypt where the revenues were of vital import-
ance to the central treasury. Thus Ahmad b. Tulun was given a free
hand in Egypt, and relieved of the restrictive presence of the fiscal
supervisor Ahmad b. al-Mudabbir who was transferred to Syria.2
The Saffarid attack and the Zanj revolt helped Muwaffaq to carry out
his plans against considerable opposition spear-headed by his own
son, the future Mu'tadid. The latter's subsequent imprisonment on
his father's orders is a clear sign of the gravity of the situation.3
The problems of Ibn Tulun in his domains were no less com-
plicated than those of Muwaffaq and also had a great deal to do with
the African trade. The revenues of Egypt were ample enough to
satisfy Ibn Tulun and the central treasury to the extent that he
abolished all the tax innovations that had been so recently introduced
by Ibn al-Mudabbir. The central government was so happy with
his management of Egypt that it added southern Syria up to
Damascus to his domains. His problems stemmed from the suddn,
i.e. the Negroes, the thughur, his son 'Abbas and Tripolitania. At a
first glance these various problems seem entirely unrelated, yet they
were all directly connected with the sources, routes and markets of
the African trade.
The sudden and conspicuous appearance of the sudan amongst the
armies of Ibn Tulun in Egypt calls for an explanation. Some sources
like us to believe that he bought as many as 40,000 Negro slaves
and made soldiers out of them to build up an empire of his own.4
Buying such a number of slaves, let alone training them to be an
effective fighting force in a completely unfamiliar territory, would
1
Ibid., pp. 1920, 1921, 1975, 1977, 1981, 1988, 2016, 2020, 2059, 208J, 2097.
• MaqrizI, vol. I, p. 319; Ya'qubi, Tarikb, vol. 11, pp. 508-9; Kindi, Wkldl, pp. 214-17.
4
» See below, p. 113. MaqrizI, Kbifaf, vol. 1, p. 31J.
https://doi.org/10.1017/CBO9780511622380.006
IIO REGIONAL ECONOMIC CONFLICTS
certainly have required more time than the few years that preceded
their appearance in Egypt and subsequently in Syria and on the
Byzantine borders in the early years of Ibn Tulun's rule 868-84/
254-70. Other sources more accurately inform us that lie "enlisted"
these siiddn in his army.1 Curiously enough Muhammad 'AH, the
ruler of Egypt in the early nineteenth century, did the same thing
and we know that when he unsuccessfully deployed such men in
these very areas against the Ottomans, the failure of the experiment
was attributed to the fact that these men had no motive to fight in a
foreign land. Under Ibn Tulun, however, these suddn fought well
and for a good reason not too difficult to understand. The key is in
the use of the term suddn to describe people who were as negroid as
the Zanj. In the terminology of the time these designations were not
used at random; they were meant to define certain groups of man-
kind. The Zanj were the population of East Africa and extending
from there into Central Africa. To the north there were the Habasha,
i.e. the Abyssinians, and still further north was the land of the Beja.
West of these, to the south of Egypt, the country of the Nubians
extended south to Central Africa and west into West-Central
Africa. From what is now the western Sudan to the shores of the
Atlantic were the lands of the siiddn, many of whom were also known
by their tribal names or according to the names of their localities.
We are here concerned with the Zaghawa, the name of a tribe and
its territory which bordered the south of the Sahara and extended
west from what is now the western Sudan across Chad, Niger and
Northern Nigeria to Upper Volta.2 Through these regions passed
an important trade route that started from Ghana and continued all
the way to the Egyptian Oases and then either to the Nile valley or
to Tripolitania. It is significant that this route had been in use until
Ibn Tulun cut it at the Oases.3 He did so, not because he did not
want the trade, but because he wanted more control over it. The
vast Egyptian desert and the troublesome Kharijites of Tripolitania
demanded constant surveillance, and this would be much easier if
the trade could be diverted to a more controllable route. The good
relations with the king of Nubia, who had had his Nubia House in
Fustat since the days of Mu'tasim, provided the solution. Although
he did not allow Egyptians to enter his territory and all trading with
them was done at the borders, it was a different matter to solicit his
1
Kind!, Wuldt, p. 214.
' Khaldun, pp. 94, 98; Yaqut, Bulddn, vol. Hi, p. 142.
' Hawqal, p. 15}.
https://doi.org/10.1017/CBO9780511622380.006
REGIONAL ECONOMIC CONFLICTS III
co-operation in allowing the Zaghawa to pass through his lands. After

all they had been by-passing Nubia altogether and going from Chad
to the Egyptian Oases, and it could only be to his advantage to
bring more trade to his kingdom. For Ibn Tulun, Aswan was an
ideal control point from which merchandise could be transported
north up the Nile or east to 'Aydhab, a secure outlet to the Red
Sea and further. For the Zaghawa the Nubian route was a much
safer one that would save them from the hazards of the desert. Once
this was established, their increasing presence in Egypt was almost a
logical consequence and a clear indication of their interest in widen-
ing the scope of their trade. Ibn Tulun would have no objection to
such an expansion which could only enhance the wealth of his
domains. This common interest created the opportunity for military
as well as economic co-operation which explains the enlistment of
the sudan in the army of Egypt.
Between them these two parties commanded major sources for
African goods, the greater part of an important trade-route and
outlets on the Red and Mediterranean Seas. The addition of southern
Syria to Ibn Tulun's governorship gave them more outlets, as it was
no secret that the ports on the Syrian coast were trading with
Byzantium through Cyprus. To take full advantage of these sea
outlets, no time was wasted in building a strong merchant navy for
Egypt.1 However, it was soon realized that the best exporting
outlets were through the markets of the thughiir. While Baghdad
was busy fighting the SafFarids and the Zanj there was a perfect
chance to try and capture these areas. For this task Ibn Tulun dis-
patched an expedition of the siiddn led by their powerful chief Lu'lu'.
The intrepid resistance of the population of Tarsus and Antioch
to these intruders is the most telling proof that they knew what their
losses were likely to be.* On the other hand the Byzantine emperor
showed his goodwill towards Ibn Tulun by sending him gifts and
releasing from captivity a nephew of Afshin.3 This last personal
gesture can be taken to mean that the emperor knew that there was a
special relationship between the Afshin and Tulun families. This
affinity in addition to the fact that Ibn Tulun had in his service both
the son of tSxcjabghu, the ruler of Tukharistan, and Toghj, a prince
from Farghana, ought to put to rest the fictitious story that Tulun
was of slave origin.4
• Kindr, Wttlaf, p. 254.

* Tabari, m, pp. 1929, 19)9-40, 2025, 2028-9; Mas'udi, Muri$, vol. vm, pp. 68, 71.
> T»bari, ni, p. 19)1. * Ibid., p. 20)7; Maqrizi, Kbifat, vol. I, pp. 51), 319.
https://doi.org/10.1017/CBO9780511622380.006
The resistance of the thughur convinced Lu'lu' of the futility of

continuing his attacks on this front. At the same time he also
realized that if the Zanj were to win their fight against Baghdad,
they would be in a much better competitive position to dominate
the greater share of the African trade. Without any hesitation and
against the wishes of Ibn Tulun, Lu'lu' and his followers turned
around and joined Muwaffaq against the Zanj.' Ibn Tulun's efforts in
person to capture the thughur came to naught, especially as he had
opposition to his plans from within his own ranks. His brother had
refused to take part in the whole venture.2 His son 'Abbas had a
positive plan of his own which he proceeded to put into practice
while his father was on the Byzantine frontier. He marched west into
Tripolitania and proclaimed Tulunid rule in this troublesome region,
and from there he vied with the Aghlabids to command as much as
possible of the North African coast.3 His logic must have been
inspired by the fact that the latter had at last got involved in the
lucrative African trade, and started exporting direct to Europe
from the bases in Sicily which they had conquered a few years
earlier.4 In his judgment it was probably better to spend the effort
overcoming a competitor rather than to secure outlets for a market
which was accessible already. His father had to cut short his cam-
paign and return to Egypt to deal with him.
The reconciliation of father and son and the death of the former
soon after, in 884/270, did not put an end to Tulunid ambitions to
capture the thughur. Ibn Tulun was succeeded by another son,
Khumarawayh, who immediately resumed the campaign, but this
time the dauntless population there had the active support of
Baghdad. Muwaffaq, who had suppressed the Zanj revolt only
months earlier in 883/270, was able to spare some of his troops and
dispatch them to the rescue. After three years of fighting neither side
had won a decisive victory, and the stalemate led to a compromise
agreement. Muwaffaq realized that the regions of Damascus and
Palestine preferred to remain under Tulunid rule and that their
economic interests lay with Egypt. On the other hand it became
clear to Khumarawayh that capturing the thughur was out of the
question, and that the interests of the people there were in the flow
of trade coming from the Gulf and Baghdad. Therefore he gave up
T b u l , i. P- 2°*5; Kind!, Wulat, p. 124.

> Kindl, Widit,?. 21 j .
1
Ibid., pp. 220-4; Tabari, HI, pp. 1932, 2011; MaqrizI, Kiitaf, vol. 1, )2o; Athir, vol.
VII, pp. 195-225. • Athir, vol. VII, p. 196.
https://doi.org/10.1017/CBO9780511622380.006
REGIONAL ECONOMIC CONFLICTS II3
all his ambitions there, and in return MuwaHaq confirmed him and
his descendants as rulers over Damascus and Palestine as well as
Egypt for thirty years. Eventually the Tulunids agreed to pay the
central treasury either 200,000 or 300,000 dinars a year.' Compared
to the 750,000 that Ahmad b. Tiilun had paid for Egypt alone, it
meant a great loss for Baghdad.2 However the real casualties of this
agreement were Lu'lu' and his Zaghawa sudan; he was imprisoned
and his followers absorbed into Muwaffaq's army.3 The remaining
sudan in Egypt reacted to these developments by rebelling in 887/273
but they were easily subdued.4
Having freed himself from all these major problems, Muwaffaq
began to reassert the authority of the central government wherever
possible. His change of policy was signified by the dismissal of
Sa'id b. Makhlad in 885-6/272. In his thirteen years in office, this
man, who had been first appointed only as a kdtib, had been allowed
a good deal of authority to enable him to handle the affairs of the
central government on behalf of Muwaffaq in his protracted absences
on various campaigns. Indeed he had had an advantage over the
military leaders as is evident from the fact that he had led military
expeditions, and had been granted the title dbii al-riyasatayn, i.e. one
having authority in two capacities.5 In his stead Muwaffaq appointed
Isma'il b. Bulbul, a man long associated with Sa'id's administration
as wa^tr to the impotent Mu'tamid. All the same, on his appoint-
ment, Isma'il assigned the Furat brothers to various departments
concerned with fiscal affairs.6 This new team was well suited to
apply Muwaffaq's new policy, the core of which was the repudiation
of military-administrative iqtd'. He was so determined to revoke this
system which had been practically forced upon him that he used his
army against such governors as refused to give up their positions.7
When his son Ahmad opposed these measures, he did not hesitate to
throw him into prison.
In the last months of his life Muwaffaq was a very sick man and
his long-forgotten brother, the nominal ruler Mu'tamid, was in no
position to exercise any authority. This confused situation allowed
1
Kindi, WuJdl, pp. 233-40; Maqrizi, Kiilat, vol. I, p. 321; Tabari, in, pp. 2106, 2108,
2116, 2132.
Kindi, VmldJ, p. 214.
Tabari, in, p. 2112; §abl, Wizard', pp. n - 1 2 .
Athir, TOI. vn, p. 298.
Tabari, m , pp. 1950, 2048-9, 2083, 2109; Mai'udI, hAurig, vol. vin, pp. 61-2.
Tabari, in, pp. 1931, 2104, 2110, 2123; §ibl, R/UMM, p. J I ; Abbir, p. 180.
Tabari, HI, pp. m j - 1 8 , 2131.
https://doi.org/10.1017/CBO9780511622380.006
the supporters of Ahmad to bring him out of prison and install him
as a de facto ruler. Within a few months after the death of both
Muwaffaq and Mu'tamid in 892/278 he was formally invested as a
successor under the title Mu'tadid. Isma'il was arrested, the Furats
went into hiding and the Jarrahs eventually came to power.'
1
Ibid., pp. 2 1 1 8 - 2 5 , 2 1 5 1 2 1 ) 3 ; Abbar, p . 175.
https://doi.org/10.1017/CBO9780511622380.006
THE BREAKDOWN OF THE
CENTRAL GOVERNMENT (I)
So far the central government had managed to muddle its way
through successive crises and challenges from the provinces and
had continued to rule without any solid foundations. Its wealth had
helped it to recruit military forces adequate to impose temporary
solutions for the many problems in its vast domains. When military
measures had failed half-hearted political concessions had been made
in the hope that they could be revoked in due course. Any long term
policies that had been introduced had had the single aim of strengthen-
ing central control without taking regional interests into considera-
tion. The maintenance of this political order depended on the
continued flow of revenues from the provinces into the coffers of
the central treasury. The increasing tendency in these provinces to
take a stand in defence of their economic interests had begun to
have its impact on the revenues of the central government and this
in turn had affected its ability to mobilize its forces to maintain its
rule.
It is most significant that when Mu'tadid came to power in 892/279,
there were no reserves in the central treasury for the first time in its
history.1 This critical situation demanded immediate rectification.
In his ten years' reign, his efforts were doomed to failure because he
did not or could not depart from the basic principles of the political
orthodoxy of his ancestors. It is no wonder that the seeds of self-
destruction inherent in these policies continued to germinate, and
caused his successors even more insoluble problems that eventually
brought about the collapse of the central government. Military-
administrative iqtd' became the only vehicle of maintaining control
over the provinces and led to the appearance of mutaghallibiin or
ashdb al-atrdf, i.e. military leaders who defied the central government
and for all practical purposes became independent in some of the
outer regions. To contain this movement the government needed
more money to pay more men to fight for the preservation of its
integrity in the remaining territories under its direct control. As the
1
§abi, Wtatari', pp. 9-10.
https://doi.org/10.1017/CBO9780511622380.007
Il6 THE BREAKDOWN OF CENTRAL GOVERNMENT (i)
money had to come from these latter regions a new method of tax
collecting, tax-farming, was introduced to ensure the flow of the
badly needed revenues. Members of the military-bureaucratic
complex were quick to endeavour to utilize this new system to their
advantage, and allowed each other to exploit the situation to enrich
themselves. Their growing land-holdings in addition to the increas-
ing burden of taxation forced many small farmers out of their lands.
The outcome of this dislocation of the rural population was a grow-
ing horde of nomads who vented their grievances by attacking
caravans and attempting to dominate the network of the inland
trade routes. This serious threat to the all-important trade created
new conflicts and unusual alliances between unexpected partners.
Prompted by vested interests among its own circles, the central
government had to act to find forces to protect trade. Military-
administrative iqta' was extended to include even the income from
inter-regional taxes imposed on the trade passing along the important
trade routes. It was also used to pay for the services of military
groups from the outer regions who flocked into the heart of the
empire, attracted by these free-for-all conditions. One of these
groups, the Daylam, from the indigenous population of the Caspian
regions were to play a major role in the downfall of the central
government. Meanwhile, revolutionary movements of a Shi'ite
character found fertile ground among the rural communities and
hastened the break up of the empire.
The symptoms of all these ailments had been present during the
rule of the strong man Muwaffaq, whose pragmatism had helped to
save the empire as well as to check these weaknesses. Having over-
come his formidable adversaries, he had begun to think in terms of a
long term policy but gout and elephantiasis had only added to his
difficulties in this respect. We do not know what this policy would
have been, yet his moves to put an end to military-administrative
iqtd' indicate his awareness of the dangers of a system that he had
tolerated for the sake of expediency. He must have been aware also
that there would be strong resistance from powerful conservative
circles to any attempt to change the prevailing conditions. It is
ironical that he himself had, to some extent, helped to consolidate
the powers of these circles in order to win their support in his
earlier difficult years. The resulting balance of power in the empire
was to the advantage of these circles and they did not like to see it
disturbed.
Mu'tadid, who had opposed his father's departure from his first
https://doi.org/10.1017/CBO9780511622380.007
THE BREAKDOWN OF CENTRAL GOVERNMENT (i) 117
policies, was also a strong man, but pragmatism was certainly not
one of his characteristics. Indeed he was essentially a conservative
who was willing to accept only minor tactical changes in order to
preserve the established order. In his reign 892-902/279-89 he
completely failed to take note of the very obvious signs of unrest in
his domains, and made no attempt to meet the crying need for change
everywhere. Instead, he satisfied himself by using every possible
device to tighten his personal control over every component of his
government. To such a conservative ruler the policy of the separation
of military and fiscal affairs was not only a political dogma but also
a means to acquire the management of both. Nevertheless, in spite
of all the measures he took to secure this objective, his short-sighted
policies procured just the opposite result.
Mu'tadid inherited a relatively strong and unified army which
had been rid of the scourge of racial differences, and he had a grand
plan for the proper use of it. This plan involved dividing the army
into three main forces, each of which had a special function. The
first division constituted the elite of his forces and was to act as a
standing army, stationed in Baghdad to take part in ceremonial
occasions and to be dispatched to meet any emergency. These men
fell into two categories, infantry and cavalry. The infantry were
mainly from the siiddn, and eventually some Daylamites joined them.
As they were stationed in barracks built for the purpose of housing
them and because of their particular duties, they were called masaf-
fijjya. The cavalry who, by the nature of things, were more involved
in ceremonies and more effective in fighting, were not only better
equipped but also a more distinguished calibre of men. Many of
them were the sons of generals, chiefs and comparable leaders;1
some of them were Arabs while others were not, and amongst the
latter there was a noticeable Khazar element.2 They were all housed
in the palace and performed the duties of the personal guards of the
ruler. Because of these various factors they were described by different
appellations, fursdti, cavalry; khdssa, elite; hujariyya, living in rooms;
gbilman, servants and mamdlik, slaves of the ruler. This last term gave
rise to some reports that they were bought, but as we know that
they were of such distinguished origins such reports can be easily
dismissed. Any one of these men who showed special abilities was
promoted to the rank oifatd Amir al-Mu'minia, i.e. a brave youth in
service of the ruler.3 They were all in the charge of a khddim, a high
' $ibl, RuiSm, pp. 7X-2.

* Athlr, vol. VII, p. $76; Tabari, HI, p. 2x63. > Tabari, m, p. 2265.
https://doi.org/10.1017/CBO9780511622380.007
118 THE BREAKDOWN OF CENTRAL GOVERNMENT (i)
ranking official of the palace. Their stipends were paid from the
central treasury, and to begin with this cavalry force was no more
than a hundred men, although their numbers increased very rapidly.
The second division of Mu'tadid's army was also composed of
cavalry and infantry, but of men of less ability. They were known as
'askar al-kbidma, servicemen and also as sburta, police force. Their
duties were to maintain public order in Baghdad and its immediate
vicinity. They were also charged with the responsibility of securing
the safety of the main routes leading into the city. In charge of them
was also a kbddim, and this means that they were under the direct
control of the ruler. Their stipends, which were less than those of the
elite forces, were also paid by the central treasury.
The third division comprised the least able men, a fact which is
confirmed by the appellation 'askar al-diiti, inferior forces. Their
function is indicated by their other name ma'iina, auxiliary body.
Under the leadership of their generals, they were sent out to the
provinces to enforce government authority, police their areas, and
support the fiscal administrators when needed, to enable them to
collect overdue taxes from recalcitrant subjects. This does not mean
that they were allowed to interfere in fiscal affairs, because although
they were to receive their stipends from the revenues of the provinces
in which they were stationed, a special civilian official munfiq,
bursar, was appointed to supervise the distribution of stipends in
order to ensure the strict separation of military and fiscal affairs.1
In charge of the central government Mu'tadid appointed a watfr,
'Ubaydullah b. Sulayman b. Wahb, an old hand associated with
Muwaffaq's early administration. He also happened to be a grand-
father of Mu'tadid's wife and could therefore be expected to be a
more loyal agent.2 Significantly this va%ir was given authority over
the generals of the third army division in the provinces.3 In other
words, a trustworthy civilian representing the ruler and in charge of
fiscal affairs, was assigned the responsibility and given the power to
keep the military out of fiscal affairs in the provinces where they
were stationed. To fulfil these strenuous duties the old wa^tr had the
assistance of his son Qasim, the ruler's father-in-law, who was
appointed as liaison officer between his father and his son-in-law.4
These meticulous arrangements were thought to be enough to bring
1
§abi, WiC(_ara\ pp. n - 1 7 . where the details o f this organization ate elaborated in the
current official language; for the term munfiq see, $abi, WuârS', p. 158; Miskawayh,
vol. I, p. 153. * Batriq, p . 74.
> Tabari, H I , pp. 2152, 2155; Mas'udI, Murig, vol. v i n , p . 114.
• 5abi, Wtiâri', p. *o.
https://doi.org/10.1017/CBO9780511622380.007
THE BREAKDOWN OF CENTRAL GOVERNMENT (i) II9
stability to the empire, but as they were not accompanied by any

measures to tackle the various regional problems, the whole plan
fell apart at the first shock.
The first problem that was particularly pressing was an empty
central treasury. By coincidence the death of the Samanid governor
of the East early in 892/279 offered a solution, as it seems to have
encouraged many groups to encroach on Samanid domains. Once
more, remnants of Tahirid forces in Khurasan and the Shi'ites of
Tabaristan tried to take advantage of what seemed to them to be a
lapse of Samanid power, in order to re-assert themselves in their
regions.» Further south, the Saffarid 'Amr b. Layth also thought that
he had a chance of dislodging the Samanids altogether. The central
government, which had earlier confirmed the rule of the Samanids
over the whole east in lieu of the Tahirids and presumably on the
same conditions, namely granting them a subsidy from the central
treasury, was now more than ready to strike a bargain with the
Saffarids. These were ready not only to forsake the subsidy but also
to pay Mu'tadid four million dirhams a year in exchange for assign-
ing Samanid domains to them.2 Although the Saffarids had some
initial successes against the combined forces of Khurasan and
Tabaristan they were eventually routed by the Samanids. 'Amr was
captured and pointedly sent to Baghdad. Mu'tadid was put in the
humiliating position of having to confirm the Samanid Isma'il b.
Ahmad over the East, and resume paying him the subsidy of between
three and ten million dirhams to enable him to continue his efforts
against the Saffarids.3 Although this venture did not bring much of
the hoped-for revenue to Baghdad, it brought there some of the
remainder of the Tahirid forces of Khurasan; a factor which added
to thefinancialdifficulties of the central government.4
The astute wa^tr had to find the badly needed money especially as
the central government had to face new challenges which will be
explained presently. Although, by all accounts, he was a capable
man he seems to have been of limited imagination. Or at least, his
abilities were more in the administrative than in the fiscal field.
First, he saw that because of the complexity of the lunar calendars
taxes were being demanded in April for crops that were harvested
in June. As this anomaly was bound to interfere with the revenues
it was easily adjusted to the satisfaction of everyone concerned.5 He
i, lit, pp. 2135, 2151. * Ibid., pp. 2155, 2151, 2185, 2188.
4
> Ibid., pp. 2194, 2205, 2204. Ibid., p. 2141.
1
Ibid., p. 214); Mas'udi, Tanbib, p. 216.
https://doi.org/10.1017/CBO9780511622380.007
I2O THE BREAKDOWN OF CENTRAL GOVERNMENT (i)
took another step which involved a further complex matter; the

inheritance laws of Islam. Before the stringent administration of
Muwaffaq these matters had been left to the discretion of the judges,
but the government had then created a special department for
inheritance affairs. Its purpose had been to acquire for the treasury
any legacies which, according to the strict application of the law,
would have had no claimants. As there must have been many ways
to evade such meddling by government, the expenses of running
such departments would have been more than the revenues. Sensibly,
they were abolished and inheritance problems returned to the
discretion of the judges.' Another interesting measure introduced
for the sake of economy at that time was afive-dayweek for govern-
ment employees. In addition to Fridays they were also allowed
Tuesdays, unfortunately without pay.2 All these measures might
have saved the government some money, but obviously more action
was needed to satisfy the needs of the central treasury.
With surprising agility the old wa^tr brought out of prison the
Furat brothers, who had gone into hiding at the accession of
Mu'tadid and had subsequently been arrested, and appointed them
his advisers on fiscal matters.3 They went to work and almost
instantly revenues began to roll in. As we know they were not in
favour of military-administrative iqtd\ but they knew that the
qabdla system had been working efficiently in Egypt since it had
been introduced there by Ma'mun. It has to be remembered that this
was not a tax-farming system and that safeguards were injected into
it to avoid the harmful results of tax-farming.4 Under the compelling
circumstances in Baghdad and because of the availability of men of
immense fortunes whose interests were tied to those of the govern-
ment, it was thought appropriate to introduce tax-farming in the
areas under the direct jurisdiction of the central authority. The
Furat brothers approached a certain Ahmad b. Muhammad of the
tribe of Tayy and offered him the concession for the districts around
the capital, including the trade routes leading into the city, in
exchange for the guaranteed payment of 7000 dinars a day to the
central treasury; the total being 2,520,000 dinars per annum.* They
also took advantage of the succession of a new TulQnid governor in
Egypt and southern Syria in 896/283 to renegotiate the terms of the
1
Tabari, , „ p. 21 ? i; 'Arib b. Sa'd, Silat Tarikb al-Tabari, ed. M. J. de Goeje, Leiden,
1897, p. 18. » Sibi, Wmpra', p. n.
1
Ibid., p. 10. • See above, p. 120.
* Sibi, Ww(arS\ p. n .
https://doi.org/10.1017/CBO9780511622380.007
previous agreement with Muwaffaq. The lump sum that the Tulunids
had to pay for their military-administrative iqtd' was practically
doubled to 450,000 dinars per annum, and they had to give up their
right of access to Qinnasrin, a trade link with Jazira.' These appar-
ently innocuous arrangements introduced in good faith by the Furats
had far-reaching and certainly unintended results in all corners of the
empire. They provided the shock that tumbled down the house of
cards that Mu'tadid had so meticulously built. Almost every one of
the military leaders stationed in the provinces demanded or seized
control of fiscal affairs in his region. Long-established local digni-
taries, nomadic chiefs, wealthy merchants and big landowners led a
flood of uprisings hoping to get the same privileges or to protect
their local interests. Mu'tadid first tried to regain control by using
his elite force but this proved impossible, and in one case after
another he had to yield to the full application of military-administra-
tive iqtd'.1 The tax-farmers made use of their profits to have their
own private armies which were probably a necessity for them.3 Thus
the distinction between tax-farming and military-administrative
iqtd' became theoretical. The members of the military-bureaucratic
complex were not slow to introduce new corrupt practices to
enhance their wealth and power. First, government officials went in
as partners of tax-farmers and in no time the wa\ir himself was a
tax-farmer.4 Needless to say the Furat brothers were replaced by the
Jarrah brothers who were better suited to the new conditions.5 It is
also significant that, at this point, the commander of the second
division, the police force of Baghdad, was allowed to intrude into
the fiscal affairs of the central government.6
As would be expected, the tbugbur were a major target for leaders
who had any significant military strength and wanted to establish
their dominions in these outlying regions. Another Afshin, a
descendant of the princes of Ushrusana, by name Muhammad b.
Divdad b. Divdast but better known as Ibn Abi al-Saj, who along
with his followers had been in the service of the 'Abbasids since the
early days of Samarra, forced Mu'tadid to concede to him Armenia
and Adherbayjan as a military-administrative iqtd'.'' Local Arab
Tabari, m , pp. 2185-7; Athir, vol. VII, p. J40.
Tabari, ux> pp. 2136-46.
Muhammad b. 'Abdilmalik al-Hamdani, Takmilat Tarikb el-Jabarl, ed. A. Y. Kan'in,
Beirut, 1961, p. 20.
MaqrizI, Khitat, vol. 1, p. 82; Misluwayh, vol. 1, p. 16.
6
Tabari, in, p. 2190. Ibid., p. 2192.
Ibid., pp. iz22, 218).
https://doi.org/10.1017/CBO9780511622380.007
122 THE BREAKDOWN OF CENTRAL GOVERNMENT (l)
chiefs like the Hamdanids, about whom we shall hear later, captured
strongholds at profitable points in the thughiir of Jazira.1 Other
Arabs of the region satisfied themselves by simply collecting their
own taxes on the passing trade.2 In Tarsus, one of the most important
trade centres of the thugbur, where merchants from every town in the
empire had their own resident agents, a new situation developed
which involved the whole northern frontier from the Mediterranean
to Central Asia.3 Also involved were all the peoples interested in
trade between the Islamic empire, Byzantium, Russia and Europe as
far afield as Sweden. The wealthy merchants of Baghdad who had
their fingers in every pie were also interested, and so was Mu'tadid
whose mother, grandmother and great grandmother were of
Byzantine origin. It is not surprising that he spoke fluent Greek.*
The problem in Tarsus was that some elements in the community
had started and were actively pursuing piratical activities against the
Byzantines in order to discourage them from trading with the
Syrian ports through Cyprus.s These activities aroused the ire of the
Syrians, the Egyptians and the Byzantines. Further east the Arab
chiefs who had established themselves at Sumaysat, Mardin and
Amid were contending with each other for a bigger share in the
trade and were also in direct competition with Tarsus. As they had
better access to the trade coming up the Tigris and the Euphrates
they were inclined to try to maintain good relations with Baghdad.
Their neighbours to the east in Armenia and Adherbayjan were more
interested in the trade coming from Europe through the regions of
the Caucasus. Although some of this trade had always been shipped
across the Caspian to the flourishing ports of Gurgan and Tabaristan
it had never represented a serious threat to the busier inland route
across the Caucasus. The threat came from a completely unexpected
quarter, the domains of the Samanids. Previously the Tahirids had
never shown any interest in trade and had witnessed the decline of
the trans-Asian traffic almost with equanimity. The Samanids with
their bases in the land of the Soghdians, the enterprising people who
had dominated commercial intercourse with China, had a different
outlook towards trade. As it was obvious that the sea-route around
Asia had attracted most of the Chinese trade, these entrepreneurs
directed their attention to the other trade that had been flowing
1
Ibid., pp. 2141-5, 2185. * Ibid., p. 2 i j 8 ; Athir, vol. VII, p. 252.
J
Istakhri, p. 04.
Istakhri, p. 64.
§abl, Kusum, p. 89; Mas'udi, Muruj, vol. vni, pp. 108, I I J .
7'abari, HI, p. 2185; Mas'udi, Muruj, vol. vm, p. 77.
https://doi.org/10.1017/CBO9780511622380.007
through the Caucasus regions from areas as far away as the Baltic
shores. The thousands of Islamic coins, dating from the first to the
fourth centuries of the Islamic era, that have been found in Scand-
inavia and along the course of the Volga, testify to the continued
interest of various peoples in this extensive trade. As in the case of
the east-west African trade, the European trade was carried out
through various intermediaries amongst whom the Slavs, the
Bulghars and Khazars weie most prominent. Like their contempora-
ries in Africa, these peoples were just as jealous of their trade secrets
and just as determined to prohibit foreigners from crossing their
lands. Each intermediary group benefited from transporting the
merchandise across its own territory, in addition to the taxes
imposed at the point of entry. With their special expertise and being
in close proximity to the terminus of this trade at the mouth of the
Volga, the Samanids could not have had much difficulty diverting
this trade around the east side of the Caspian to the entrepot of
Khwarizm.1 As good merchants they would have had an innate
abhorrence of taxes and tariffs especially if they wanted to capture
the greater share of this trade. Indeed we hear only about a tax
imposed at the crossings of the Oxus, which at this time was not
considered a border. Furthermore as it was at the negligible rate of
one dirham per donkey, mule or horse load and two dirhams per
camel load, and considering that these loads were mostly luxury
goods, it could hardly have been more than the ferry fare. This is
confirmed by the fact that the same report speaks about a payment or
half a dirham to one dirham at the various overnight stopping-
places en route.1
Significantly these crossing points were used to discourage some
of the slave trade. A special licence from the authorities in addition
to a levy of 70 to 100 dirhams was required for the transport of
every young male Turkish slave. For female Turkish slaves no
licence was needed but a levy of 20 to 100 dirhams was assessed
according to age. As this could not have been introduced on
humanitarian grounds, the only conceivable reason is that the Sama-
nids were particularly anxious not to offend the Turkish nomads
surrounding their territories, who were involved in their trade
schemes, and from amongst whom these slaves were mostly captured.
Indeed the fact that only Turkish slaves were singled out for this
1
Iitakhri, pp. 218, 221, 22;, 226, 299; Khurdadhbeh, p. 154; Rusteh, p. 141; Ibn
Fadlin, RitdJa, ed. S. Dahhan, Damascus, 1959, pp. 74, 98, 105, 14); also see below,
p. 149. * Maqdisi, p. 540.
https://doi.org/10.1017/CBO9780511622380.007
T H E
124 BREAKDOWN OF CENTRAL GOVERNMENT (i)
treatment supports this conclusion. Another interesting use of these
crossing points was to stop the smuggling of silver ore from the
rich mines of Central Asia into the heart of the empire. This makes
economic sense because the value of silver was rising rapidly as
against gold. Therefore, all baggage was inspected and silver was
returned to Bukhara where it could better serve the purposes of the
expanding trade.'
Faced with this complex situation and in order to be able to cope
with it, Mu'tadid had to decide on priorities. While he was interested
in the flow of trade, he was not concerned with nor was he in favour
of any particular route. His main responsibility was to contain the
situation and stop it from further deterioration. There was no reason
to fight the Samanids; he had conceded Armenia and Adherbayjan
to Ibn Abi al-Saj and had no desire whatever to fight the Byzantines,
and he had just concluded a reasonable agreement with the Tulunids.
Therefore his immediate problems were limited to securing the
vital link with the eastern regions from Baghdad to Rayy, to restor-
ing some sort of order in the troublesome tbughur of Jazira, and to
terminating the unnecessary provocations of the Byzantines by
Tarsus. With remarkable energy he proceeded towards these
objectives. His own military efforts did not bring about the desired
results on the eastern front, and he had to fall back on the political
acumen of his waîr who was dispatched to Rayy. There, through
further application of military-administrative iqtd\ the waîr was
able to bring some stability back to the surrounding regions.2
Mu'tadid himself turned to Jazira where he had a measure of military
success mainly because his opponents were divided amongst them-
selves. The strongholds of Sumaysat, Amid and Mardin easily fell
into his hands, and the alliance between the Arabs and the Kurds of
the region could not sustain the pressure from his elite forces. The
Arab Shi'ite Hamdanids who had co-operated with the "Kharijite"
Kurds turned against their allies and joined forces with Mu'tadid
to subdue the latter. For this flexibility they were rewarded by a
military-administrative iqtd' in the region of Mawsil, where they
had had their strongholds. The Hamdanid forces thus became part
of the central government's third division, and in this role they were
soon to play an important part in support of Baghdad's and, of
1
All this information is included in Maqdisi's report, quoted in the previous note,
which was completely misunderstood by W. Barthold, "Turktstan Down to tbt Mongol
Imarion, London, 1928, pp. 239-40.
* Tabarl, nit pp. 31140-1, 2147, 2152, 21J5-6, 2161, 2178.
https://doi.org/10.1017/CBO9780511622380.007
course, their own interests. Other Arabs who had been in control
of strongholds of the thugbiir followed the Hamdanid example and
surrendered to Mu'tadid. To ensure their continued allegiance he
stationed a strong contingent, led by his own son the future Muktaf i,
at Raqqa, and returned to Baghdad.'
As for the problems posed by Tarsus, Mu'tadid thought that with
some persuasion the pirates there would desist from their con-
troversial activities. For this task and to sort out the problems of the
aftermath of his military operations, he sent, as his own representa-
tive, his best informed high official, the post-master general of all
the north-west regions.2 Nothing happened in Tarsus, but a serious
disturbance started in Adherbayjan. Wasif, probably an Armenian
Muslim, who had a considerable following in Ibn Abi al-Saj's army,
disagreed with his chief about the new arrangements for the region.
Being a native he was in a better position to imagine the extent of the
loss they would suffer as a result of the drying up of their trade.
Therefore he led his followers in a march on Malatya, one of the
forward strongholds of the thugbiir, and had no difficulty in establish-
ing himself there. The Byzantines had no objection to anybody
literally setting up shop closer to their territory, and thus took no
action against Wasif. But Mu'tadid was quick to realize that Malatya's
prosperity would be at the expense of the other markets further
south and east in Jazira. The upshot would be the dislocation of the
arrangements he had so recently contrived there. Without wasting a
moment he mobilized all his forces, including the Arabs of Jazira,
and led them in person against Wasif. In desperation, Wasif tried
to take refuge with the Byzantines but was swiftly intercepted and
captured, and his army quietly absorbed into the government's
forces.3 Mu'tadid then made a grand tour to inspect all the thughur.
In Tarsus he took a most extraordinary step, yet a most significant
one; he ordered the destruction of its entire Muslim fleet that had
operated against the Byzantines. Fifty expensive warships went up
in flames, and many parties were satisfied at least for a time.4
In the last three years of his life Mu'tadid was confronted with a
new problem, but one that had been simmering for generations. He
must have been puzzled by the sudden explosion in southern 'Iraq,
an area which had been so recently pacified. Perhaps in his own way
1
Ibid., pp. 2108, 1141-5, 2149-51, 2185-7.
* Ibid., p. 2184.
1
Ibid., pp. 2195-8; Mas'udf, Muruj, vol. vni, pp. 196-9.
* Tabari, in, pp. 2199-200.
https://doi.org/10.1017/CBO9780511622380.007
he, like Marie Antoinette, wondered why people were shouting for
bread. He had just stumbled on a new political order for the empire
and was probably hoping for an era of tranquillity. Instead, there
was an explosion that was to reverberate in almost all corners of the
empire. It was bound to happen and there had been signs of its
coming, but reactionary circles were too blinded by their interests
to take note of these forewarnings. Confident in their power and
certain of government support they were too arrogant to consider
the slightest change which was not to their advantage. Thus, they
practically asked for the coming revolution.
Although this revolution took place in separate regions and
almost certainly under different leaders, there were three distinguish-
able common features in its eruptions; regional interests, rural
disaffection, and Shi'ism. The establishment of military-administra-
tive iqtd' had not been motivated by the consideration of regional
interests. It was promulgated for the benefit of the central govern-
ment and the satisfaction of the military-bureaucratic complex.
Moreover, the presence of separate and strong military forces in the
regions, such as those of the Tulunids in Egypt, would surely work
to the disadvantage of the weaker regions. To protect their vital
interests the latter had no alternative but armed revolt against the
established order.
As always, the rural communities had patiently endured the
injustices imposed upon them in the hope that some day, somehow,
salvation would come. They had had to find new ways and means of
evading the constantly increasing burden of taxation, while the
affluent members of the urban population had been unscrupulously
busy multiplying their untaxed wealth. Any attempt to redress this
glaring inequity had been easily brushed aside by the powerful
vested interests. The introduction of tax-farming must have caused
the long suffering tax-payers to despair of any prospects of relief.
The rapid spread of big land-holdings had been mostly at the expense
of the peasantry and the small landowners, many of whom had been
driven out of their properties. The upsurge in the number of nomads
in areas bordering the fertile lands of Syria and 'Iraq was a direct
result of this dislocation of population. Many of them had supported
the Zanj revolt against their oppressors and had been eagerly waiting
for another opportunity to vent their grievances.
From its very inception Shi'ism had been the traditional opposition
party, and as such had been the vehicle for various protest move-
ments. It has already been explained and it cannot be over-stressed
https://doi.org/10.1017/CBO9780511622380.007
THE BREAKDOWN OF CENTRAL GOVERNMENT (i) I27
that during the Umayyad period the various Shi'ite sects had had
nothing in common, other than the necessity of an Imam, to be
chosen from the members of the House of the Prophet to occupy the
office of Amir al-Mu'mimn.1 The absence of a concrete Shi'ite
ideology, and the diversity of their opinions on the many aspects of
the one thing upon which they had agreed, had helped to emphasize
the sectarian nature of Shi'ism. On the other hand, the collective
will of the Muslim community to preserve its unity had been so
strong that it had forced the 'Abbasids to set themselves apart from
Shi'ite sectarianism.2 Furthermore, the fact that Shi'ite sects had
identified themselves with regional interests, whether in Kufa or
Merv, had deprived Shi'ism of the universality that Sunni orthodoxy
had so strongly claimed. The descendants of 'All through his
martyred son Husayn, the grandson of the Prophet, had had the
strongest claims for the latter's legacy, and until the failure of the
revolt of Zayd in Kfifa in 740/122,3 they had been willing to fight
for their rights. The appearance of other members of the House of
the Prophet leading separate Shi'ite movements had not affected the
legitimate rights of the Husaynids. The victory of the 'Abbasids,
themselves heirs to the claims of a third son of 'All who was not
even a grandson of the Prophet, had allowed the descendants of Hasan
to overcome their ancestor's lapse in abdicating in favour of the Umay-
yads, and put forward their own claims. This had manifested itself
in the revolt of the Pure Soul in the reign of Mansur.* Henceforth the
Husaynids, satisfying themselves with their legitimacy, remained the
quietist party, while the Hasanids continued to lead the activist party.
The last spasm of the Husaynids had been expressed in their co-
operation with Ma'mun in his shilly-shallying attempts to effect a
general compromise. However, the imposition of orthodoxy by
Mutawakkil had, at last, convinced this anachronistic main line of
Shi'ite leaders of the futility of their claims. The "disappearance"
of their twelfth imam in 878/265 had, in effect, marked the giving up
of these claims, and accordingly their party had become known as
the Twelvers. In contrast, the Hasanids had already established a
dynasty in Morocco, the Idrisids, beginning 789/172, which can
barely be described as Shi'ite but had succeeded, to some extent, in
acquiring bases in Yaman and Tabaristan. In these latter regions
these Hasanids had called themselves 'Alawids in order to emphasize
1
Shaban, Islamic History, pp. 179-81. * See above, p. 8.
> Shaban, Tbe 'Abbasid Revolution, pp. 132, 148.
4
See above, p. 14.
https://doi.org/10.1017/CBO9780511622380.007
their claims as sole heirs to their ancestor's legacy, or Zaydites, an

epithet that evoked memories of the struggle of their cousins the
Husaynid Zaydites of Kufa, and stressed their determination to
continue to fight for their claims. Significantly the 'Abbasids applied
the term Talibids to all descendants of 'All indicating that, in their
view, there was no meaningful difference between any of them, thus
in effect equating the grandsons of the Prophet through his daughter
Fatima with their half-brother who could not claim such a dis-
tinguished lineage.1
Amidst all this an insignificant Shi'ite sect, which had deviated
from the main line of the Twelvers as a result of a disagreement about
the succession of the seventh imam, was thrust on to the scene. They
were called Seveners and also Isma'ilis after the name of their chosen
imam, and it is with these that we are now concerned. Their move-
ment, which eventually developed into a radical social, economic,
philosophical and religious revolution, started as a secret society
soon after the occultation of the twelfth imam of the main line.
Subsequently it manifested itself in five separate regions; Kufa,
Syria, Bahrayn, Yaman and North Africa in that order. At the same
time there was supposed to be a headquarters in Salamiyya, a small
village bordering the desert not far from the site of ancient Palmyra.2
While in Kufa, Syria and Bahrayn they were called Qaramita, in
Salamiyya they called themselves Isma'ilis, and some of them in the
Syrian desert called themselves Fatimids. In North Africa they
encouraged the latter appellation.
A great deal of research has been done on the origins, and nature
of these groups, the connexions between them or independence
from each other, yet we are still very much in the dark about all
these things. Indeed little progress has been made since they were
first discussed in 1886.3 Perhaps it serves a purpose to remember
that these movements were active over eleven centuries ago under
circumstances where their survival, let alone their success, ultimately
depended on the maintenance of absolute secrecy at the incubatory
stage. It is more important, and perhaps more instructive, in trying
to solve these vexing questions, to examine the particular milieu of
each of these movements and to establish the nature of the support
each had, a most important ingredient in any revolution. It is just
as important to scrutinize all the economic conditions of Isma'ili
1
SabI, Wt*ara\ p. 10.
1
Yaqut, BuUati, vol. m , pp. 55, 240-1.
1
M. J. tie Goejc, Mfmoirt sur lej Carmatbes du Bahrain it hi FatimiJej, Leiden, 1886.
https://doi.org/10.1017/CBO9780511622380.007
strongholds in relation to the wider economic situation of the

Islamic and the international scenes. In this respect a little more than
lip service has been paid to the Fatimids, while the other Isma'Ilis
have not been considered worthy even of such meagre treatment.
The callousness of the central government and its obdurate
refusal to take cognizance of the two burning issues of the time,
regional interests and rural disaffection, gave the Shi'ites the
opportunity to revive their movement in a revolutionary form as
champions of both causes. Their long association with regional
uprisings and their complete identification with the oppressed made
them ideally suited for this role. Nonetheless, the sectarian nature of
Shi'ism, in addition to the inherent connection with regional
interests, were bound to allow for various approaches to solve the
problems involved. The highly sophisticated Fatimid ideology was
a far cry from the crude notions of the Qaramita of Kufa. Other
contemporary non-Isma'ili Shi'ites like the Hamdanids, the Zaydis
of Tabaristan and the Buyids, had their own ideas and solutions for
the same problems. It is most significant that all these Shi'ite regimes
which, for some decades, dominated the greater part of the Islamic
lands, found no room for co-operation among themselves, and
indeed were busy fighting each other in defence of their respective
local interests.'
There is no reason to doubt that there were Isma'ili headquarters
in Salamiyya from which agents were sent to exploit the situation in
troubled areas; this fits well with Shi'ite activities before this time.
But it is not possible to detect a master-plan according to which these
agents would operate, nor is it easy to piece together the confused
reports about the identity and relations of the members of this sub-
versive network. They all moved wide and fast, had disciples and
relatives who perpetuated their activities, and appointed their own
agents even in government circles,2 with little apparent direct con-
nexion with Salamiyya. Hamdan Qarmat, the most famous of these
sub-agents, is reported only to have kept up correspondence with his
supposed masters which could not have amounted to much in the
way of contact.3 It has been wrongly assumed that he gave his name
not only to his own movement but to others as well. Endless
speculations have accordingly been made about the Arabic, Aramaic
or Nabati origins of the various forms of the word Qarmat. The
1
See below, pp. 167-8.
1
Tabari, m, p. 1179.
> Maqrfcl, llti'iz al-Humt/i', ed. J. Shayyil, Cairo, 1948. p. 22$.
https://doi.org/10.1017/CBO9780511622380.007
term first appeared to describe a group who had supported the Zanj
revolt, the reference being to the Qarmatiyyun and to Nubians who
could hardly speak Arabic. • The geographer Maqdisi also associates
these two peoples and considers that they were siiddn.1 Reliable
lexicographers tell us that the Qarmatiyyun (sing. Qarmati) were a
specific race.3 It does not need much imagination or etymological
explanation to realize that the people in question were from Garama,
the ancient Libyan trade centre.4 It becomes clear, then, that these
Qaramita were remnants of the Zanj revolts who, under the circum-
stances, were ready to take part in any revolt, especially in this area
and of course Bahrayn. It should be pointed out, however, that this
participation was only a minor element, but being a common one, it
was enough to give these movements this peculiar name.
The movement in Kufa, where the first Qaramita appeared and
disappeared, spread among the peasants of the Sawad but did not
have the support of the Arab nomads in the desert.5 It was specifically
a reaction to the newly imposed tax of one dinar a year per man, which
had been decreed by Ahmad al-Ta'i who had the tax-farming
concession and held a military-administrative iqtd1 of most of the
region.6 The preachings of their leader Hamdan seem close enough
to those of the Isma'ilis. His grandiose plans to establish a fortified
base were probably inspired by the Zanj tactics and were not enough
to provoke the central government to take action.7 But one measure
is most important as it is an indication of Isma'ili thinking and a
portent of things to come. For the first time in Islamic history an
income tax of 20% was introduced and this was effected by this
relatively insignificant man.8 The movement was tolerated for
roughly ten years 890-901/277-88 but then Ta'i and his men were
able with little effort to bring it to an end. Hamdan disappeared with
his movement.9
The second movement of the Qaramita broke out in Bahrayn in
899/286 under the leadership of Abu Sa'id al-Jannabi. This man
who was from Firs, certainly had no direct or indirect relation with
Salamiyya, and the attempts of some sources to establish a connexion
between him and Hamdan are not convincing.10 Intellectually he
may not have been as talented as Hamdan, but tactically he was
Tabari, HI, pp. 1756-7. * Maqdisi, p. 241.
LJsan, s.v. qrml. * See above, p. 102.
Tabari, in, pp. 2117, 2124, 2198, 2202; Maqrizi, ////'aj, pp. 208-9.
7
Tabari, HI, pp. 2126, 2127, 2198. Maqrizi, ///»'aj, p. 215.
Ibid., p. 210. • Jabari, m, pp. 2202, 2206.
Ibid., p. 2188; Maqrizi, ItJi'dz, p. 214; Hawqal, p. 29}.
https://doi.org/10.1017/CBO9780511622380.007
certainly more gifted. His support came mainly from amongst the
nomads of East Arabia whom he organized into an army powerful
enough to maintain a base in Bahrayn and pose a threat to Basra.1
The central government was not particularly alarmed but was
prompted to take action by pleas from its representatives there. In
891/278 Mu'tadid sent an expedition of 2000 men who were utterly
routed by Abu Sa'Id and his Qaramita.2 Surprisingly the latter did
not follow up their victory by advancing towards Basra, and instead
they satisfied themselves with consolidating their strength in
Bahrayn. In 900/287 the old wasjr died and a year later Mu'tadid
himself also died. Both were succeeded by their sons, Qasim b.
'Ubaydillah and Muktafi respectively, Qasim being the grandfather
of Muktafi. Needless to say the policies of the central government
continued unchanged, but it had to cope with the third movement of
the Qaramita.
This time there is no doubt that the movement had no direct
relation with Salamiyya and indeed was probably in competition
with it. It was organized by a disciple of Hamdan, Zakruyeh, who
had gone underground after the collapse of the Kufa peasant move-
ment.3 He found his support among the nomads of the Syrian-
'Iraqi desert and particularly from the clans of Kalb, once a pillar of
Umayyad rule. They were now reduced to transporting goods along
the caravan route from Kufa to Damascus, skirting the desert via
Palmyra.4 It is significant that other nomads of the clans of Asad,
Tayy and Tamim from the vicinity of Kufa, who had no interest in
this particular route, refused to join the movement.5 It is also
important to realize that this route was part of Ta'I's responsibility,
and it is more than likely that the men engaged in work along it
were also subjected to the tax of one dinar per year.
Taking advantage of the death of Mu'tadid, these men, led by
Yahya son of Zakruyeh, rose in arms and swiftly established their
domination of this important trade route. In the process they
defeated the government forces on the 'Iraqi side of the desert and
the Tulunid army of Damascus on the Syrian side.6 The latter must
have shown some apathy in fighting the Qaramita because the
Damascene merchants, whose interests were directly involved,
pleaded with their colleagues of Baghdad to persuade the govern-
i, HI, pp. 2188, 2192; Maqrizi, Itti'az, pp. 215-16.
Tabari, m, pp. 2188-9, 2192-3; Maqrizi, Itti'az, P- 2 l 8 -
4
Tabari, m, p. 2217; Maqrizi, Itii'az, p. 224. Tabarl, HI, pp. 2217-18.
Ibid., p. 2217.
Ibid., pp. 2219, 2221; Maqrizi, Uti'3%, pp. 225-6.
https://doi.org/10.1017/CBO9780511622380.007
ment to intensify its efforts against this relentless enemy.' While the
authorities were trying to comply with this request, the Qaramita
were joined by a respectable group called Banu al-Asbagh.2 A few
years earlier a member of this family had been influential in maintain-
ing good relations between Mu'tadid and the thughiirJ* It can there-
fore be inferred that members of this family had some interest in the
flow of trade through at least the 'Iraqi side of the route controlled
by the Qaramita. The important thing here is that the adherence of
this group coincided with a methodical change in the movement.
First, the leader, Yahya, was proclaimed as the Shaykh, Elder, a fact
that indicates the existence of some sort of collective leadership.
At the same time he claimed a Husaynid descent and the Asbaghids
began to call themselves Fatimids, which was almost certainly
meant to include all the leadership of the movement.4 We do not
have any evidence to substantiate the claims of the Asbaghids that
they were descendants of Fatima, the daughter of the Prophet, or
for this matter of any other significant person of that name. But the
fact that this claim was not disputed, even by a Jarrah member of
the administration, is in itself a proof of its validity.5 The meaning
of this is that these men had the combined prestige of being descend-
ants of Husayn and Hasan, the only sons of Fatima. The significance
of such a claim from a Shi'ite point of view is that it associates the
legitimacy of the Husaynids with the activism of the Hasanids to
the exclusion of all other 'Alawid claimants, and to the utter abase-
ment of the 'Abbasids.
When the Shaykh Yahya was killed, soon afterwards, in the
fighting against the government forces, he was succeeded by his
brother Husayn, the Man with the Mole, who took not only the
title of Amir al-Mu'mimn, but also nine other messianic titles includ-
ing that of Mabdi. Furthermore, he gave honorific and significant
titles to at least two of his aides.6 After reaching an agreement with
Damascus and imposing the payment of tribute on it, he directed
his attacks towards the north occupying Salamiyya and destroying
it in the process.7 Alarmed by this threat, the government mobilized
all its forces, and Muktafi himself went out to Raqqa to muster
support for these efforts against the Qaramita. The Arabs of the
thughur, especially the Hamdanids, who were cut off from both
1
Tabari, in, p. 2222. * Ibid., p. 2219.
3 4
Ibid., pp. 2i)7, 2148. Ibid., pp. 2118-19.
» Ibid., p. 2257.
4
Ibid., pp. 2219, 222;, 2232-4; Maqrizi, Ilti'a%, pp. 231-2.
7
Tabari, m, pp. 1225-6; Maqrizi, Uli'ij^ p. 227.
https://doi.org/10.1017/CBO9780511622380.007
Damascus and Kufa, gave their whole-hearted aid and effective

military help.' The wasjr who remained behind in Baghdad appointed
the hdtib Muhammad b. Sulayman, who was the head of the depart-
ment of army affairs, to put together and lead an expedition which
constituted men recalled from the regions for this purpose.2 A
contingent was sent down to Bahrayn as a precaution against any
move by the Qaramita there.1 The Tulunids seem to have been in
two minds as to what attitude they should take towards a movement
which did not seriously threaten their interests, and was in fact
directed against their competitors in Syria. After some internal
struggle, there emerged a party in favour of co-operation with
Baghdad, and accordingly a Tulunid army led by the prominent
general, Badr, advanced against the Qaramita from the south.4 This
pincer-movement was more than enough to defeat them. Their
leader was captured, his forces dispersed and the government's
authority restored over the Syrian-'Iraqi desert at least for the time
being.5
The next move, emanating from Baghdad almost immediately,
was rather surprising. The army of Muhammad b. Sulayman was
ordered to attack Egypt and in a matter of months was able to
terminate Tulunid rule there in 905/292.6 Two remarkable factors
contributed to this easy victory; the defection of Badr and his men
to the central government's side, and the deployment of the rebuilt
naval forces of Tarsus in co-ordination with the army's attack on
Fustat, the Tulunid capital.7 The presence of the Hamdanids in the
army atucking Egypt indicates that they had as much interest in
bringing down the Tulunids as the people of Tarsus, who did not
hesitate to sail all the way across the Mediterranean for the same
purpose.8 The leader-organizer of the expedition, Muhammad b.
Sulayman, had, himself, an interesting history. Twenty years earlier,
he had been in the service of Lu'lu' who had defected with his
sudan from the army of Ahmad b. Tulun to fight the Zanj along with
Muwaffaq.9 His intimate knowledge of Egypt and the intricate
problems involved must have qualified him for this particular task.
After his victory, he was particularly interested in destroying any
ni, pp. 2231-2. • Ibid., pp. 2236, 1240-2.

1
Ibid., p. 2232; §abi,Qaramita, p. 23.
4
Tabari, in, pp. 2219, 2251, 22)3; Maqrizi, Kbital, vol. i, p. 322; Kindi, Wulit, pp.
242-3.
* Tabari, in, pp. 2232-45. * Ibid., p. 2248.
7
Ibid., pp. 22 j 1-2; Maqrizi, Kbifat, vol. 1, p. 322. * Kindi, Wtdat, pp. 246-7.
• Maqrizi, Kbifaf, vol. 1, p. 327.
https://doi.org/10.1017/CBO9780511622380.007
remaining power of the siiddn in Egypt.' He also made a special

effort to dislodge the rest of the Tulunid forces in the country,
including Badr whose defection had facilitated the government's
victory.2 The latter was known as Hamami and although one source
explains this epithet as pigeon-keeper, it is mort credible to relate
this powerful leader to Dhat al-Hamam, a big market town on the
coast to the west of Alexandria.3 Earlier there had been some
troubles to the west of the Delta in which the Berbers of the nearby
desert had been involved in support of an army faction, and appar-
ently the latter had also been Berbers.4 All this leads us to believe
that Badr was a Berber who represented interests different from
those of the siiddn. In other words the controversy involving Ahmad
b. Tulun and his son 'Abbas had not completely subsided.5 Six
months after the fall of the Tulunids, the party supporting the cause
of the siiddn revolted. The government forces withdrew to Alexand-
ria and significantly Badr was called upon to subdue the rebels.6 In
terms of general policy Baghdad, aware of the crumbling power of
the Aghlabids and the incipient Fatimid movement in North Africa,
decided to fortify the position of its forces in Egypt. Before his
return to Baghdad, Muhammad b. Sulayman laid the foundations
for direct control over all Egyptian affairs by the appointment of
two independent representatives of the central government to take
charge of military andfiscalmatters.7 This was, of course, a departure
from military-administrative iqtd\ but it was an exception forced by
the circumstances.
Meanwhile, the Qaramita of Syria thought that they had an
opportunity to recover their position. They regrouped their forces
and were even joined by some of the government troops of Damas-
cus.8 In a series of lightning attacks they terrorized all the urban
communities bordering the Syrian-'Iraqi desert, and for two years
905-6/293-4 they attacked and plundered every possible passing
caravan, including those returning from the pilgrimage. The great
wealth of goods transported by the latter was far more than that
needed for the simple functions of this religious occasion. Once
' Ibid., p. $12. * Ibid., p. 527; Kind!, Wulat, p. 248.
' Al-Sam'ani, Amab, ed. D. S. Margoliouth, Leiden, 1912, f. 174; Athir, vol. vm, p.
17; Ya'qubi, Bu/ddn, p. $42; Bakir, p. j .
4
Tabarl, in, p. 2155; Kindl, Wulat, p. 242; Maqrizi, Kiitaf, vol. 1, p. 522.
• Sec above, p. 112.
4
Tabari, in, pp. 225j, 2267; Kindt, Wulat, pp. 259-6}; Mas'udi, Muruj, vol. vm.
pp. 2}6-7.
» Kind!, Wuldt, p. 258; Tabari, HI, p. 2253. • Tabari, in, p. 2257.
https://doi.org/10.1017/CBO9780511622380.007
THE BREAKDOWN OF CENTRAL GOVERNMENT (i) 13J
more the Hamdanids distinguished themselves in supporting the

government forces to bring a final defeat to these impudent maraud-
ers. This time old Zakruyeh himself was wounded and captured.
His death put an end to this particular movement. • The Qaramita of
Bahrayn, who had not made a move to support their fellow revolu-
tionaries nor to take advantage of the government's plight, were
left alone, and it was to be some years before they made their
presence felt.
Throughout his reign 902-8/289-95 Muktafi was as interested as
the Byzantine emperor Leo VI the Wise 886-912, in finding a
modus vivendi between their respective domains. In 902/289 he sent
presents to the latter, and in 903/290 and 907/294 Leo responded by
sending a peace mission headed by his own brother-in-law.2 These
cordial relations were not disturbed by the hostile activities of the
hawks on both sides of the borders.3 Indeed measures were taken to
bring such elements under control. On his part Muktafi appointed
a Hamdanid over all the Mawsil regions, and the latter proceeded to
subdue the Kurdish and Arab nomads of the area.4 On the other
hand, Leo cut off the support of the Byzantine military base at
Konya to the extent that its garrison was forced not only to cease
operations, but also to seek refuge in Tarsus, of all places.5
In the eastern regions Muktafi was not so fortunate, because the
difficulty of the problems there was compounded by the absence of
any agreement on a unified approach to tackle them, even within the
parties concerned themselves. To begin with, his wa^tr, who was
also his own grandfather, happened to have had the tax-farming
concession of Fars and Kirman in exchange for the sum of four
million dirhams.6 This kind of interest prompted him to persuade
Muktafi of the advantages of reaching an accommodation with the
neighbouring Saffarids.7 The powerful military commander in
charge of the security of Fars opposed this approach, and seems to
have had the support of influential figures in Baghdad including
Muktafi's uncle. In the ensuing struggle, the wasjr won and managed
to have his two strong opponents murdered.8 However, attempts
to reach an agreement with the Saffarids failed because of opposition
within their own ranks, and soon enough defecting Saffarid forces
arrived at Baghdad.*
1
Ibid., pp. 2255-78. * Ibid., pp. 2223, 2236, 2277.
1
Ibid., pp. 2249, « } ' • 2268, 2269, 2275; 'Arib. p. 13.
* Athlr, vol. VII, pp. 293-4. * Tabari, in, pp. 2251, 2275-6.
* Miskawayh, vol. 1, p. 16. ' Tahiti, in, p. 2222.
* Ibid., pp. 2209-15. • Ibid., pp. 2233, 2255.
https://doi.org/10.1017/CBO9780511622380.007
I36 THE BREAKDOWN OF CENTRAL GOVERNMENT (i)
When the aged waqir Qasim died in 903/291, he was replaced, on
his own recommendation, by 'Abbas b. Hasan who had a clear
interest in the affairs of the Samanids. The new wa^tr was a loyal
disciple of his patron and therefore his policies were in essence a
continuation of the latter's.1 The central government had never
been completely satisfied with the arrangements with the Samanids.
Encouraging the Saffarids had been one way of changing these
arrangements, but as this had failed the authorities in Baghdad had
been watching for other means to bring about the desired effect.
Soon, the opportunity presented itself in the form of a series of
insurrections in Rayy, Tabaristan and Gurgan.2 The diversion of the
northern trade to Khwarizm had adversely affected the prosperity
of the regions under Samanid domination. With some subtle
manipulation, the central government could at least try to regain
these regions. Such hopes were curtly dashed as the Samanids were
quick to use their military forces to subdue the insurgents.3 As a
result Pars, the former Samanid governor of Gurgan and 4000 of
his men, had to flee to Baghdad. Their arrival there had been
anxiously expected by the disconcerted wavyr when Muktafi died
and more urgent problems now had to take precedence*
1
§abi, Wuâri', pp. 229, 360-1. • Tabari, m, pp. 2208-9.
1 4
Ibid., pp. 2216 2220-1. Miskawayh, vol. I, p. 4.
https://doi.org/10.1017/CBO9780511622380.007
THE BREAKDOWN OF THE
CENTRAL GOVERNMENT (II)
The anxiety of the wa%ir, 'Abbas, before the death of Muktafi had
not been without good reason; he had known, for some time, that
there would be a crisis of succession. Muktafi, who died at the age
of thirty-two, had been in failing health for several months, if not
longer. As his children were too young, he had expressed the wish
that he should be succeeded by his brother, Ja'far, who had just
reached puberty.' Having been in office for four years, the experienced
wayir was convinced that the absence of strong leadership would
simply take the lid off the whole situation. There were three separate
military forces; one in the provinces and two in Baghdad. The
former, which constituted the third division, was dispersed in various
regions and was not a cause for compelling worry. But the other two,
the elite force of the first division and the police force of the second
division, were concentrated in the capital. From the time of their
formation they had been under the direct command of strong rulers
who had used them effectively as instruments of the central govern-
ment. A boy of thirteen could not be expected to take personal
charge of these vital forces; that would leave their commanders in
virtual mastery of the government. The fact that one of these
commanders was the maternal uncle of the boy ruler could only add
to the apprehensions of the wayjr. The latter had hoped that the
arrival of Pars and his army of 4000 men would enable him to
enforce the choice of a more mature ruler. When Muktafi died before
the arrival of this army, the waîr had no alternative but to accept the
proclamation of Ja'far to succeed his brother under the title of
Muqtadir, 908-32/295-320.*
In less than four months all the watfr's fears had come true. Some
army commanders arranged a coup d'etat, killed the unfortunate
waîr, deposed Muqtadir and replaced him by another 'Abbasid
with the title of Murtada. In 24 hours a counter-coup by the elite
first division reversed the situation and reinstated Muqtadir, the
1
'Arib, pp. z i - 2 ; Miskawayb, vol. I, pp. 2-3.
1
Miskawayh, vol. 1, p. 4.
f>37l
https://doi.org/10.1017/CBO9780511622380.008
I38 THE BREAKDOWN OF CENTRAL GOVERNMENT (il)
child ruler.1 From this turmoil emerged four strong men who
retained their positions and held the reins of government between
them for almost twenty-five years - Nasr al-Qushuri as commander
of the first division; Gharib al-Khal, a leading general of this army,
who was closely allied to Nasr; Mu'nis al-Fahl, as commander of the
second division police force; and Mu'nis al-Khasiyy as commander
of the third division. Nasr, whose official titles were hdjib, chamber-
lain, and mawId Amir al-Mu'minin, was of Khazar origin from a
village called Qushura in the vicinity of Balanjar.2 His recent
arrival from his native land was betrayed by the fact that he did not
speak Arabic well.3 Gharib was Muqtadir's maternal uncle, hence
al-Khal, the Uncle.4 His association with Nasr and the fact that
neither he nor his sister are described by any source as Armenians or
Greek, makes it more than probable that he was also a Khazar.
The other two personages together are interesting. As they were
namesakes various epithets were used to distinguish them from one
another. One was khayin, bursar, versus khddim, high official; both
terms allude to their respective previous activities. Because of his
military achievements Mu'nis, the commander of the third division,
was also called mtqaffar, victorious, and in his old age he became
known as kabir, elder. However, they were popularly known as
al-Fahl, the virile, and al-Khasiyy, the emasculated. Although this
had nothing to do with their sexual prowess, scholars, failing to see
the vivid contrast of these names, have taken khasiyy to mean literally
"eunuch". Consequently another aspect of the slave society theories
has been blown up out of all proportions and has created a good deal
of confusion. The use of the official title khddim to describe Mun'is
in addition to the term khasiyy has led to the unfounded conclusion
that both words mean "eunuch". This confusion is compounded by
the fact that in North Africa and Muslim Spain some officials had
the status of kliddim along with the military rank of ghuldm or fata.
As they also had the physical attribute oi khasiyy, all these terms have
been taken to mean "eunuch" also. Therefore we find "eunuchs"
as generals, high officials and even rulers. The scholars who have
introduced this idea and come to this astounding conclusion have
been influenced by the institutions of the Roman, Byzantine and
Ottoman courts where harems and eunuchs participated in palace
1
Ibid., pp. j - 8 ; Txbari, m, pp. 2280-2; §abl, Wizard', pp. 87-8; 'Arib, pp. 16-9.
1
§abi, Wuâri', p. 154; Maqdisi, pp. 51, 355 and variant readings in the footnotes
thereof; Yaqut, Bu/dait, vol. 1, p. 489; Takmila, p. 29.
> §abi, Wuzpra', p. 92.
* Suli, Akhbar al-Kadl, cd. J. H. Dunne, London, 193;, p. 5.
https://doi.org/10.1017/CBO9780511622380.008
THE BREAKDOWN OF CENTRAL GOVERNMENT (il) I39
conspiracies and attained a measure of political power in these
societies. Be that as it may, a eunuch is the embodiment of man's
cruelty and inhumanity towards his fellow man. Such a frustrated
and abused male may crave for power, conspire or be entrusted with
confidential tasks, but he certainly cannot lead armies or rule govern-
ments. Anyone who has any doubts about that would only have to
see a eunuch to realize this, and until recently it was possible to see
in Egypt such pitiable specimens who had outlived their Ottoman
masters.
What is being disputed here is the wide use of eunuchs in Islamic
society and what is denied is their attainment of any power. The
palace of Muqtadir in Baghdad was run by women and there is no
indication whatever that Mu'nis had anything to do with it.1 He
was called khasiyy not because he was castrated but because he could
not grow a beard.2 To the Arabs male virility was, and to some
extent still is, represented by wearing a beard. Men were punished
and humiliated by publicly shaving off their beards. Arab geo-
graphers and travellers never tire of telling us of the various habits
of wearing or shaving hair in other parts of the world.3 In short,
beardless men were considered less than normal and this abnormality
had to be attributed to the physical defect characteristic of eunuchs.4
Therefore, any man who did not grow a beard was commonly
described as khasiyy. Ibn Khayyat, an author who died 854/240, tells
us about an Arab, Sa'd of the tribe of Azd, who had been in charge
of Kufa in 745/127 and who was called khasiyy. Then this careful
author explains that he had been called thus because he had not had
a beard.5 The most famous and certainly not castrated khasiyy was
Sa'd b. Qays b. 'Abada. His father had been a powerful leader in
Madina during the Prophet's lifetime and he himself stood by 'All's
side until the very end. However, this unusually big and brave man,
like Mu'nis, had no beard and accordingly he was known as khasiyy
al-ansdr, the eunuch of the Madlnan helpers of the Prophet I!6
We have no indications at all about the origins of the bearded
Mu'nis, but the ready links the beardless Mu'nis had with the
Byzantines suggest that he may have originated in their territories.
Takmila, p. 31; Athir, vol. vm, p. 74.
Athir, vol. vm, p. 123; Takmila, p. 51; Miskawayh, vol. I, p. 160.
Rusteh, p. 129; Bakri, pp. 175, 179; Hawqal, pp. 397, 482; I?takhri, p. 226; Mas'udI,
Tanbib, p. 168.
Maqdisi, p. 242.
Ibn Khayyit, Tdrikb, ed. A. D. Umary, Baghdad, 1967, p. 430.
Iffahani, Maqdlil al-Jalibiyyin, ed. S. S. §aqr, Cairo, 1949, pp. 71-2.
https://doi.org/10.1017/CBO9780511622380.008
I4O THE BREAKDOWN OF CENTRAL GOVERNMENT (il)
Perhaps it is of interest to note that there were no conflicts between
these two leaders. However, their conflicts and alliances with the
other two leaders Nasr and Gharib, dominated the scene and in fact
created the conditions for the Buyid take-over. These conflicts were
complicated by the fact that during Muqtadir's minority his mother,
probably prompted by the desire to protect him, took an active part
in the affairs of state. In addition to her alert brother Gharib, her
sister and uncle too did not remain idle either. In the turbid atmo-
sphere of the long reign of Muqtadir there were no less than fifteen
changes of wayirs and five coups d'itat. In order to be able to present
the history of this period; make sense and keep things in perspective,
the following analysis will be made in terms of two different policies
as represented by the protagonists the Jarrahs and the Furats. This
is not to over-simplify the issues but rather to clarify them, and not
at the expense of leaving out any waîr. For if he who was holding
office was not a Jarrah or a Furat by name, he was so by association.
In the very few appointments when this was not exactly the case, the
policies followed then can be identified to the right or left of the two
main parties.
Reference has already been made to the opposing views of these
two groups and it is clear that their differences had remained basically
the same.1 Some changes had become unavoidable because the
domination of the government by the military had introduced a new
factor in the situation. The Jarrahs, the conservatives who had always
been for the wide application of military-administrative iqtd', had
become completely subservient to the military. In order to serve
their masters better they became converts to the cause of "good
management", almost with the same degree of conviction as the
Conservatives in Britain today. The strict measures of the Jarrahs
generally worked to the advantage of the rich and powerful who
could easily evade them. The Furats were not exactly socialists
although they were accused of worse things by their opponents. It
is farcical that Nasr, who could not even pronounce the word
properly, should accuse them of being Qaramita.2 Surely they did
not and could not introduce any revolutionary changes, yet by trying
to lessen the woes of tax-farming and military-administrative iqta1,
they were in effect supporting the cause of the tax-payer. They
openly declared that tax-farming was evil and that in any case it
should not be granted to the military or government officials.1
Furthermore, they strongly asserted their conviction that the
1
See above, p. 94. » §ibl, XTiqari', p. 92. » Ibid., pp. 71, 258.
https://doi.org/10.1017/CBO9780511622380.008
THE BREAKDOWN OF CENTRAL GOVERNMENT (il) 141
military should be subject to the civil authorities.1 On the other

hand they accepted corruption in government circles as a fact of life.
Therefore they allowed it as a lesser evil, but at the same time tried
to utilize it to the advantage of the treasury by demanding for the
latter a share in the amassed gains.2 While there is no record that the
Furats did anything to remedy the iniquities of urban and rural
taxation, the Jarrahs added to the gravity of the situation by abolish-
ing taxes on the import-export trade.3 The merchants of Baghdad
were more than ready to lend money to the government while the
Jarrahs were in power, but of course at the handsome interest of at
least 12 % a year. The strict prohibition of usury in Islam was over-
looked by the severest of the Jarrahs, the so-called "good vizier"
'Ali b. 'hi*
The Furats first came to power in 908/296 in the wake of the
counter-coup that reinstated Muqtadir. As the Jarrahs had sup-
ported the first coup against the latter, it was almost natural to
appoint a Furat as waytr* The finances of the central government
were in a comfortable state; the public treasury had a reserve of
600,000 dinars while the private treasury of the ruler had 15 million
dinars.6 With some necessary manipulation the waqjr and his
brothers used the latter to meet the demands on the former.7 Although
Furat satisfied the needs of the ruling circles in Baghdad, he fell out
with some of the military commanders as he showed some un-
expected determination in trying to bring them in line. One of them
was summarily executed as a result of his attempts to interfere in the
watfr's domain.8 But the beardless Mu'nis was too strong a figure
for such treatment and to his credit he was not averse to letting the
waqtr carry out his responsibilities. Mu'nis was only interested in
securing the wages of his men but the watfr had different ideas
about how this should be done, and indeed questioned the wisdom
of having the third division as a separate force from the first. This
had been an acceptable arrangement when the ruler himself had been
able to command the first division. Under the new circumstances
Furat left the second division, now numbering 9000 men, to police
Baghdad and its vicinity, and took some steps to combine the other
two divisions.9 His plan was to reduce the number of the first
division in Baghdad to the minimum needed for ceremonial occasions
1
Ibid., p. 72. * Ibid., pp. 210-7.
1
Ibid., p. 286; Miskawayh, vol. i, pp. 228-9. * $abi, Wuâri', p. 81.
> Ibid., p. 88; Tabari, in, p. 2282; Miskawayh, vol. 1, p. 5.
• 'Arib, pp. 22-3. ' Miskawayh, vol. 1, p. 108.
* Miskawayh, vol. 1, p. 12. * Ibid., p. 20.
https://doi.org/10.1017/CBO9780511622380.008
142 THE BREAKDOWN OF CENTRAL GOVERNMENT (il)
and scatter the rest in the provinces, whereby the central treasury
would be relieved of their expenses. Indeed when Pars and his 4000
men who had been anxiously awaited in Baghdad, finally arrived
there, they were quickly directed to Jazira where they could balance
the Hamdanids' growing strength and extract their stipends from
them.' The appointment of the Khazar Tigin al-Khassa, i.e. of the
elite force, as governor of Egypt in 910/297 was not a coincidence.2
It means that some of the forces of the first division were now given
the function of the third division in the provinces. General Mu'nis
would not be expected to object to a move that would ultimately
enhance his own forces, as long as acceptable arrangements were
made to secure the high stipends of the elite force. But he and Furat
did not see eye to eye on this crucial point, and the latter was forced
out of office in 912/299 after holding power for three years and
eight months.3
At this point a solution was urgently needed for this delicate
problem, and for this task Muhammad b. 'Ubaydillah b. Yahya b.
Khaqan was brought out again and appointed wa^tr. Although
aged and almost senile he was the descendant of a long line of
experienced administrators and his own father had been involved in
a similar situation fifty years earlier in Samarra and Baghdad.4 To
counteract his weaknesses, his son was appointed as his assistant
and, most unusually, a trusted khddim with strong army connections
was made an overseer, not only over all departments but also over
the wayir himself.5 Within a year the old idiot, who had tried to
please all parties concerned, had achieved just the opposite and
exasperated everybody.' His dismissal in 913/301 coincided with
the death of the bearded Mu'nis, the chief of police and also a strong
force among the proverbial powers behind the throne.7 This
temporary lapse in the strength of the latter caused them to lose the
initiative. Although the Mother, the Uncle and Nasr were in favour
of the return of the Furats, they could not press their preference
against the wish of the other Mu'nis, who had already decided on his
choice. This was to be a Jarrah team headed by the "good vizier"
himself, 'Ali, b. 'Isa, who had been taking refuge in Makka for the
previous five years.8
Ibid., p. 16.
Ibn Taghribirdi, al-Nujum al-Zabira, Cairo, 1965-72, vol. in, p. 171.
Miskawayh, vol. 1, pp. 20, 25; Athir, vol. VIII, p. J I .
See above, p. 7;. » Miskawayh, vol. 1, pp. 2 ) - j .
Ibid., pp. 24-6; 'Arib, p. 39; Sibi, Wuâra', pp. 262-3. ' 'Arib, p. 45.
Sab!, Wuâri', pp. 30, 263; Miskawayh, vol. 1, pp. 25, 27; 'Arib, p. 42.
https://doi.org/10.1017/CBO9780511622380.008
Soon enough, the "good management" of the Jarrahs came into

play to the satisfaction of Mu'nis and his swelling forces. It has been
pointed out that this military commander was one of the few of his
kind who was not particularly keen on interfering with the fiscal
affairs of the provinces wherein he operated; all he wanted was to
secure the stipends of the men under his command. This reasonable
attitude made the wa^r's work less complicated and was the basis of
the good relationship that lasted between the two men for two
decades. By the device of appointing the infant son of the teenage
ruler in charge of Egypt and Syria, with Mu'nis as his deputy, the
revenues from these provinces were allocated to meet the latter's
needs.1 Nasr was allocated the revenues from some districts in
'Iraq for the additional expenses of his police force which was
mainly paid by the central treasury. With remarkable efficiency the
waspr proceeded to clear up the mess left by his predecessor. His
principal aim was to balance the budget and to this end he was
determined to take all the necessary steps. Cuts in all salaries and
expenses paid by the central treasury were his major weapon in
economizing on expenditure.2 As for the receipts, he introduced a
mixed bag of minor tax adjustments and major tax concessions. The
adjustments were concerned with re-assessing the land-tax in some
areas of Fars and 'Iraq, sometimes in favour of the treasury and
sometimes in favour of the tax-payer.3 The concessions amounted
to abolishing taxes on imports and exports, particularly at points
where such trade was bound to pass through the hands of 'Iraqi
merchants. Hence trade taxes imposed at Makka, all the ports of the
Persian Gulf and the thughiir of Jazlra were annulled.4 Our sources
tell us, along with the copious praise they heap upon the pious
wa^tr, that all these just and desirable tax reforms had cost the central
treasury the very moderate sum of 500,000 dinars.5 However, as
these sources are always reticent about taxes on trade which they
considered illegal, it can be concluded that this figure is meant to
cover only the minor land-tax reforms. Considering that these very
sources estimate the population of Baghdad, about that time, at
96 million, the validity of their calculations regarding the other
figure comes very much into doubt.6
1
Miskawayh, vol. i, pp. 32, 153.
1
Ibid., pp. 29, 34; §abl, Wuzara", p. 282; Athlr, vol. v m , p. 51; Takmila, p. l j .
> §abi, Witâra', pp. 286, 340, 3 4 3 - 6 ; Hawqal, p. 303; Istakhri, p. 158; Takmila, pp.
1 2 - 1 3 ; Miskawayh, vol. I, pp. 27, 2 9 - 3 0 ; §abf, Rusum, pp. 6 7 - 8 .
4
Miskawayh, vol. I, pp. 2 8 - 9 ; Takmila, p. 13; Athlr, vol. VTII, p. j r .
6
» Takmila, p. 13; Miskawayh, vol. 1, p. 29. §abi, Rjuimr, p. 20.
https://doi.org/10.1017/CBO9780511622380.008
T H E
144 BREAKDOWN OF CENTRAL GOVERNMENT (il)
In spite of the good reputation he enjoyed in conservative circles,
our waqjr does not seem to have gone far enough to satisfy the
rigid Hanbalites of the capital.1 He got himself involved in the
controversial issue olwaqf, endowment, establishment. We remember
that the central government had objected to such institutions
established in Egypt about a century earlier under the name ahbds.2
Although over the years they had begun to gain some general
acceptance, the issue had not been completely resolved. Trying to
establish a precedent, the waîr persuaded Muqtadir to endow some
of his properties in Baghdad and outside, for religious purposes.3
The amount of money involved was insignificant but the principle
was of great importance, for once it was accepted for religious
purposes then it could be used for personal and family endowments.
This innovation did not appeal much to the Hanbalites, nor was it
easily accepted by the less conservative elements in Baghdad.
Indeed, when the Mother, a little later, took advantage of it to
protect some of her extensive properties against the unknown, or for
that matter known and foreseeable, ha2ards of the time, there were
riots in the city and demands that the deeds be invalidated.4
Another controversy, and one that the weqir did not ask for, was
over the question of the extravagant mystic Hallaj. His preachings,
about which a great deal has been said, do not concern us here; what
concerns us is the fact that he was involved in the highly charged
political atmosphere of Baghdad. The conservatives saw in him a
venomous enemy who was out to destroy the establishment. Others
were intrigued by his theopathic phrases and saw him as a harmless,
possibly deranged man who could not have brought down the
established order even if he wanted. Such "saints" were common
enough and could gather a considerable following but it would all
soon dissipate itself. The wa%tr was torn between the two attitudes
but to be on the safe side he decided to imprison this agitator. To
the Hanbalites, who wanted his head, this punishment fell too far
short of the offence; to others it was too much.5 Meanwhile the
controversy continued to simmer and was to be thrown wide open
after a lapse of nine years.6
As if this was not enough to diminish the credentials of the
wasgr with his conservative supporters, he, of his own volition,
decided to add to their suspicions. Without any apparent reason, he
1
Sabi, XPu&ra', p. $55. * See above, p. $4.
» Sibi. Wtc&ri', p. 286. « Miskawayh, vol. I, p. 245.
1
jabari, HI, p. 2x89. * See below, p. 152.
https://doi.org/10.1017/CBO9780511622380.008
opened negotiations with the Qaramita of Bahrayn who had been
quietly busy with their own affairs for over seven years, and indeed
were to continue to do so for another decade. When they asked for
spades, the request was generously met and almost incidentally they
were allowed to trade with Siraf across the Gulf.' The only possible
explanation, and it is not quite convincing, is that he had the fore-
sight to take precautions against the forthcoming Fatimid attacks
on Egypt and thwart any concerted action by the Qaramita. To be
sure, within a year the Fatimids had occupied Cyrenaica and in
915/302 attacked Alexandria. Mu'nis mobilized an army of more than
40,000 men and marched to Egypt where he was able to expel the
attackers.2 Wliile the Qaramita made no hostile moves, the Ham-
danids tried to take advantage of the situation to extract from the
central government more territory and privileges in Jazira. Having
finished with the Fatimids, Mu'nis marched straight against the
ambitious Hamdanids who quickly dispersed.3
All these military operations required money, and in order to
balance his budget the wa^tr had to tighten his control over expendi-
ture. In his zeal he went over the tolerance limits of those affected
by his austerity measures and began to feel their resistance. He tried
to resign and was persuaded to carry on, but the pressures against
him that had been building up, soon reached breaking point. After
almost four years in office 301-4/913-17 the Jarrah 'Ali b. 'Isa was
dismissed, and his team was replaced by a Furat team.4
The new team was headed by the same Furat wav$r who had held
office in 908-12/296-9 but the circumstances this time were different.
While Mu'nis and his men were satisfied with their pay arrangements,
Muqtadir and his Mother were clamouring for the Palace expenses
to be paid in full by the public treasury. Pledging to them a daily
payment of 1500 dinars in addition to all other expenses, Furat
resorted to a most extraordinary method of raising the necessary
money. He knew that corruption was rampant among all govern-
ment officials, and that it was no use trying to stop it or simply
hiding one's head in the sand and pretending the trouble was not
there. He also knew that there had been a long tradition, reputedly
introduced by 'Umar in Madina, that every government appointee
should submit an account of his wealth upon dismissal and would be
1
§abi, VMâra', pp. 292-5; 'Arib, p. 59.
1
Tabari, m, PP- 229-50; 'Arib, pp. 51—j.
1
Miskawayh, vol. 1, pp. 56-9.
* Ibid., pp. 41-2.
https://doi.org/10.1017/CBO9780511622380.008
I46 THE BREAKDOWN OF CENTRAL GOVERNMENT (il)
heavily fined for any irregularities. However, as this had been left
to the discretion of the ruler, it had not been always enforced.
Accepting corruption as the order of the day, Furat decided to
standardize the practice of taxing it, and for this purpose he set up a
Department of Bribes.1 Ingenious as it was, this method by its very
nature could not assure a regular flow of revenue. After being in
office for eighteen months and being unable to meet his commit-
ments, Furat and his team were dismissed in 918/306.
To satisfy the needs of Mu'nis and the demands of the rulers a
new team, that combined the efficiency of the Jarrahs with the
wealth of the tax-farmers, was brought into office. It was headed by
Hamid b. 'Abbas the biggest tax-farmer of his time, whose power
and wealth were as excessive in Baghdad as in his home region of
Wash. Although a very genial and successful businessman he had
had no past experience in government. His first assistant, who was
appointed in charge of all departments, was the Jarrah 'All b. 'Isa,
the "good vizier".2 The essence was that while the waîr, Hamid,
should concern himself with finding better means to collect more
revenues, the assistant watfr, 'All, should devote all his efforts to
devising measures to cut down the expenditure. As was to be
expected tax-farming was applied wherever possible, and the
distinction between the tax-official and the tax-farmer disappeared
for they became one and the same. The wa^tr contracted for himself
the tax-farming of all the lands under the immediate jurisdiction of
Baghdad for the total sum of 54 million dirhams. Then, at a hand-
some profit, he sub-contracted these to his own representatives in the
areas concerned.3 The tax-farming of Egypt and Syria was granted
to the tax-official there, who had been a long-time associate of the
assistant watpr, for three million dinars.4
The flow of revenues into the treasury did not offset the draw-
backs of the system and it soon began to crumble. The wa^tr, who
had his own private army, ran the government as a private business.s
The newly established Department of Tax-farming was under the
efficient direction of Kalwadhanl, a man whose expertise must have
been widely recognized. His presence in the administration for the
following decade represented a rare element of stability and con-
tinuity which may explain the durability of the central government.
However, the water's right-hand man was Ibn al-Hawwari, whose
1
Ibid., p. 42; Athlr, vol. VJII, p. 7j. * Miskawayh, vol. I, pp. j8, 70-1.
» Ibid., pp. 59-60, 154. • Ibid., p. 107.
> Takmila, pp. lo, 53.
https://doi.org/10.1017/CBO9780511622380.008
THE BREAKDOWN OF CENTRAL GOVERNMENT (ll) I47
appointment truly reflects the corruption of the time. This man, as
his name indicates, was a flour merchant whose interests were fully
involved with the interests of the waqir, who was also a grain mer-
chant. Having both official power and private means, this absolutely
profit-orientated pair applied all their talents to making gains
beyond their wildest dreams. The new corrupt practices they intro-
duced in government were bad enough in themselves, but were less
serious in their results than the monopolies set up to speculate in
commodity prices. Inflated prices of flour and grain in particular
brought about riots in Baghdad, not only by the civilian population
but also by the military. Although these riots were contained, steps
had to be taken to rectify the situation. After some struggle on the
part of the waîr to hold on to his powers, his tax-farming contract
was cancelled and his Jarrah assistant was given the responsibility of
collecting the revenues. Tax-farming continued in operation with the
strict proviso that government officials should be excluded from
accepting such contracts. It is significant that while the assistant
wat(tr was given a free hand to dominate all departments, the wa^tr
himself was not dismissed and continued in office for at least three
more years.' In other words the rulers realized that they could not,
at that stage, alienate the powerful circles the rva^tr represented.
This political manoeuvring to maintain a united front was inspired
by developments in the eastern and western regions of the empire.
To take the west first, another Fatimid expedition arrived in Egypt
in 919/307 and successfully occupied bases in Alexandria, Giza and
Fayyum. It took Mu'nis almost two years and the help of the Tarsus
navy to dislodge the enemy, a success for which he was rewarded by
being given control over the affairs of Egypt and Syria.2 In the
eastern regions rapid developments had been taking place every-
where from Central Asia to Adherbayjan. Disconnected as they may
seem, these developments once more revolved around the northern
trade. We have already explained the changes that had been taking
place with regard to the routes of this trade. The Samanids, the main
beneficiaries of these changes, were not at one in how to deal with
the new situation. The prosperity of Khwarizm was at the expense of
Tabaristan and both were parts of their domain. Any unrest in
Tabaristan would echo in Khurasan, particularly in Nishapur,
where the merchants of Tabaristan had had good relations since the
1
Miskawayh, vol I, pp. 56-100; Athlr, vol. vm, pp. 82-6; Takmila, p. j i ; 'Arlb,
pp. 70-9.
» Miskawayh, vol. I, pp. 75-6; 'Arlb, pp. 79-86.
https://doi.org/10.1017/CBO9780511622380.008
I48 THE BREAKDOWN OI CENTRAL GOVERNMENT (il)
days of Tahirid power.' This difference of opinion manifested itself
in the defection of general Pars and his troops to Baghdad in 907/295.
Three years later the Sarnanids sent to Muqtadir an unusual present
of furs, especially sable, which was in effect like sending samples to
a prospective customer.2 At the same time they began attempts to
dominate Rayy, a vital link on the route from Tabaristan to the west.
Although these attempts were officially disavowed by the Samanids
they soon produced adverse results.3
In 914/301 a Shi'ite revolt broke out in Tabaristan and at the
same time the Samanid ruler was assassinated. The Shi'ite revolt was
that of the famous Hasanid Utrush who had been living among the
Daylamites for thirteen years. The regime he set up had two interest-
ing features; the imposition of an income-tax of 10%, and the
undermining of the power of the big landowners.4 He established
his headquarters at Chalus, a strong point that dominated a northern
route from Tabaristan via Qazvin.s This move brought about a
chain reaction in Adherbayjan. There the main interest in this trade
was in dealing with the Khazars through the trans-Caucasian land-
route. As this was in direct competition with the sea-route from
Khazar country to Tabaristan, the interests of the latter were
incompatible with those of Adherbayjan.6
The assassination of the Samanid ruler allowed the opposing
factions the opportunity to push foward their claims. This was
vigorously expressed in two rebellions; one in trade-conscious
Samarqand and the other in Tabaristan-orientated Nishapur.
Fortunately the Samanid regime had the inner strength to overcome
such difficulties. The son of the assassinated ruler, a boy of eight
years, was proclaimed successor to his father, but the real power
fell to Jayhani, the head of the administration and one of the most
sagacious men of his time.7 After attending to his internal problems
and pacifying the Samanid domains, the external issues became his
concern. There was every reason to encourage the prospering trade
and since Baghdad was a main entrepot for this trade, it was only
sensible to reach an understanding with the interested parties there.
The co-operation of the two capitals Bukhara and Baghdad would
1 2
Sec above, p. 67. 'Arib, p. 35.
1
Athir, vol. vm, p. 76; Miskawayh, vol. 1, pp. 39, 45.
4
Athir, vol. vm, pp. 61-2; Miskawayh, vol. 1, p. 63.
* 1'abari, m, p. 2292; Isjakhri, p. 206.
' Miskawayh, vol. 1, p. 4 ) ; Khurdadhbch, p. 154, Istakhri, p. 218; Hawqal, pp. 388,
7
394, }9 8 -
T»bari, u t i p. 2289; Athir, vol. viit, pp. 58-60, 86-8.
https://doi.org/10.1017/CBO9780511622380.008
surely be enough to overwhelm the minor powers in the regions in
between, be it in Tabaristan or Adherbayjan. The authorities in
Baghdad were of varying degrees of enthusiasm about the extent
of such co-operation. The merchant wa%tr Hamid might be expected
to encourage and indeed be involved in such commercial transac-
tions. Nasr, a real power in the capital, although a Khazar, under-
stood all the implications of co-operating with the Samanids and
threw his support behind it.1 On the other hand Mu'nis, who had
always shown a particular interest in the thughur and Adherbayjan,
was watchful for the interests of these regions, and did not want to
see them sacrificed for the interests of the East.2 He was also aware
that the Byzantines, who sent a trade mission to Baghdad in 918/305,
with the ostensible reason of exchanging prisoners of war, were also
interested in this trade, part of which passed through their hands.3
The culmination of all these schemes was a full-fledged trade
mission that started from Baghdad on 21 June 921/11 Safar 309 and
arrived at the capital of the country of the Bulghar, up the middle
Volga at its confluence with the River Kama, on 12 May 922/12
Muharram 310. Every aspect of this most unusual journey, its timing,
starting point, route, destination and the identity of every person
involved, helps to focus the attention on the concern of all parties
with the northern trade. It also brings together all the threads
needed to understand the relations between these complex events
that involved peoples, not only from within, but also from outside
the empire.
This mission was a response to a combined approach by Jayhani
and the chief of the Bulghar. A Muslim Khazar representing the
latter accompanied by a man from Khwarizm as a messenger from
Jayhani, had journeyed to Baghdad not long before 921/309. 4 A
fellow countryman from Khwarizm, Tigin, who as a blacksmith in
his native land had been acquainted with the country of the Bulghar,
had helped to introduce these messengers to his superior in the
Palace.5 The latter, in turn, had persuaded higher personages to
listen to their persistent pleas.6 At last, it had been decided to
respond to this important message by sending a delegation repre-
senting Muqtadir to negotiate with the chief of the Bulghar. The
members of this delegation were carefully chosen because of their
special knowledge of and interest in these faraway lands. Leading
• 'Arlb, p. 51; Mislcawayh, vol. t, p. 52. * Miskawayh, vol. 1, pp. 82-3.

1 4
Ibid., pp. 55-5; $4bl, Ruswn, pp. 11-14. Fabian, pp. 69, 77.
< Ibid., pp. 80-1, 136. ' Ibid., p. 81.
https://doi.org/10.1017/CBO9780511622380.008
them was a high-ranking officer in the service of Muqtadir who was

descended from the Russians of the Volga.1 Second to him was
general Pars, who had earlier defected from the Samanid ranks and
who also happened to be a Slav.2 Third was, of course, the inter-
mediary, Tigin of Khwarizm. Fourth was Ibn Fadlan, to whose
diligence we owe the fascinating description of their voyage. He
pointedly related himself to Muhammad b. Sulayman, who had
played a major role in bringing down the Tulunids, and who until
917/304 had been in charge of the fiscal affairs of Rayy. He was to
act as the secretary for this important mission, and his interest is
indicated by his association with Rayy.3 Of course the emissary of
the Bulghar accompanied the delegation back but the representative
of Jayhani was detained in Baghdad.4
Starting from the capital they advanced east on the main route to
Rayy, where they anxiously waited for eleven days in expectation of
some protection on their way further east.s This seems not to have
been available, for they then joined the trade caravan and had to
conceal their identities while passing through Tabaristan territory
controlled by the Shi'ites of the region.6 Arriving at the Samanid
capital Bukhara, they, in company with the defector Pars, were
cordially received by Jayhani and were introduced to the boy ruler.
Moreover, they were given all the help they needed for the following
stages of their journey to Khwarizm and then around the Caspian
to the Volga.7 Apart from the usual hazards of travelling, their main
difficulty was to secure permission to pass through the territories of
the nomadic Turkish tribes of the Ghuzz and Pecheneg. Their
escorts, credentials and familiarity with the area helped them to
convince the reluctant nomads to permit such an unprecedented
breach of their code of secrecy.8 Finally reaching the Bulghar
capital they saw for themselves the cause of the strength of these
people and the justification of their own mission. The Bulghar had
just begun to settle astride the crossroads of east-west and north-
south trade; the former by land from Khwarizm to Kiev, and the
latter from the Caspian up the Volga to the Baltic shores. They did
not need to go anywhere for the trade because it all came to them,
and they had no difficulty collecting their 10% tax.9 However, as
they were not politically united they were to some extent subject
1
Ibid., p. 69. * Sec above, p. 136; Fadlan, p. 69.
> Fabian, pp. 65, 69; Miskawayh, vol. 1, p. 51; see also above.
!
• Fadlan, p. 77. Ibid., p. 74.
* Ibid., pp. 74-5. ' Ibid., pp. 76-7.
• Ibid., pp. 98, IOJ-4, 107. « Ibid., p. 145.
https://doi.org/10.1017/CBO9780511622380.008
T H E B R E A K D O W N OF C E N T R A L G O V E R N M E N T (il) 151
to the more powerful Khazars, their neighbours to the south-west

who also controlled the estuary of the Volga. To keep the peace the
Bulghar were obliged to pay to the Khazars a tax of one sable-fur
per household.1 Pressured from both the Khazars and the Turks,
the Bulghar were looking for allies. The Samanids were just as
eager to find allies who would help them bring in line the trouble-
some Turkish nomads and shake the Khazar domination off the
Caspian. The presence of Muslims among the Bulghar and the
willingness of more of them to be converted, helped to cement their
alliance with the Samanids.2 The subsequent prosperity of both was
at the expense of the Khazar, whose power collapsed in 965/355,
and of the Shi'ites of Tabaristan, who in a decade were reduced to
insignificance.
The rulers of both Bukhara and Baghdad must have been greatly
surprised by the results of the economic demise of Tabaristan.
Their hopes that between them they could overcome any resistance
to their concerted policies from the region to the south and the south-
west of the Caspian, were speedily dashed by the Daylamites. These
people who had previously confined their energies to their own
mountainous territory must have been hard hit by this economic
decline. Hitherto, they were hard to find amidst the multitude of
races in the various armies of the empire. Although Islam had long
struck roots among them, there had not been much change in their
socio-political organization. Their adherence to Shi'ism, in its
Zaydi form, was due more to missionary activities than to any
particular need for a vehicle of protest. The new economic situation
now gave them reason for protest and forced them out of their
isolation. First, every chief of a clan or a tribe, every leader of a
village, a town or district, and every prince from the many ruling
families in the region, supported by his men, rose up in arms to
defend their interests and establish their control at every possible
point on the secondary route from Rayy to Baghdad via Qazvin, or
on the main route via Hamadan. The people of Adherbayjan, who
were just as badly affected by the decline of their own trade, jumped
into the ring to defend their territory against the encroachments of
the Daylamites. The rulers of Bukhara and Baghdad were extremely
anxious to safeguard the link between their respective territories,
and were willing to go to any length for this purpose. Their attempts
merely complicated the situation and worked to the advantage of the
Daylamites. The Samanids tried to exploit the internal differences of
' Ibid., pp. I I J , 145. » Ibid., pp. 67-8, 117-18, izo-2, 13J.
SIH
https://doi.org/10.1017/CBO9780511622380.008
IJ2 THE BREAKDOWN OF CENTRAL GOVERNMENT (il)
the latter to use some of them against their fellow-countrymen. The
result was a series of bewildering alliances in which the partners
changed sides with surprising agility to the confusion of all con-
cerned. The central government tried to intimidate some and placate
others. When military expeditions failed, military-administrative
iqtd1 was extended to include any leader, Daylamite or otherwise,
who could establish his control over any area. Then outright land-
grants, i.e. simple iqta1, were given to persuade some leaders even
just to accept military-administrative iqtd1. But the dike was breached
and all efforts to contain the Daylamites failed. Three of their
leaders were able to establish short-lived dynasties that were of no
significance except for the fact that they were the precursors of the
Buyids who, in due course, were to occupy Baghdad itself. *
Unable to cope with all these disturbances, the authority of the
central government began to weaken and the reactionary Hanbalites
of the capital were quick to take advantage of the situation to assert
their own strength. To placate them the ineffectual wayir Hamid
offered a scape-goat. The harmless mystic Hallaj, who had been
languishing in prison for nine years, was brought out, put to an
elaborate trial and finally executed in 922/309.* Perhaps there is
nothing better to illustrate the atmosphere of Baghdad than the fact
that when the great scholar Tabari died there in 923/310, public
demonstrations were allowed to be held against him and he was
subsequently denied a decent burial simply because he had refused to
recognize Ibn Hanbal as a jurist.3 Nevertheless, these convulsions
did not do the government any good because they were only
attempts to sidetrack the main issue - lack of money. Revenues had
decreased because of the turmoil caused by the Daylamites in the
central regions, and expenditure had increased because of the
military measures taken against them. When the assistant wa%ir,
'Ali b. 'Isa, tried to tighten his hand on expenditure, the military
objected.4 Accordingly the whole Jarrah team headed by Hamid was
dismissed in 923/311 and replaced by the ever ready Furat team.
Significantly our sources refer to this year as the Year of Ruina-
tion.5 The events of the year themselves justify such an epithet, but
there is also justification in the events of the preceding and following
years; it was the beginning of the end. Four days after the change of
1
MUkawayh, vol. i, pp. 117, 161-2, 115-4; Athlr, TOI. VHI, pp. 1x8-9, 1 J*-4J I
199; TaJkmila, pp. 38, 51-2; 'Arib, pp. 137-8. 154, 161.
1
» Athlr, vol. vin, p. 98; Miskawayh, vol. 1, p. 84.
4
Miskawayh, vol. 1, pp. 8j-8. » 'Arib, p. n o .
https://doi.org/10.1017/CBO9780511622380.008
THE BREAKDOWN OF CENTRAL GOVERNMENT (il) I}}
waîrs, and as if it was timed to coincide with it, the Qaramita of
Bahrayn started a series of devastating and lightning attacks on
towns and caravans. The first attack was on Basra, where a force of
only 1700 men was able to occupy and loot the town for seventeen
days. • Ten months later they attacked a pilgrimage caravan on its
way back to Baghdad, plundered it and captured many distinguished
pilgrims.2 These two attacks were to typify all the others that
occurred spasmodically over a period of almost half a century. For
seventeen years these Qaramita had chosen to remain quiet and had
contented themselves with receiving some spades and minor trade
concessions from Baghdad. In 913/301 their first leader Hasan
Jannabi had been assassinated and succeeded by his son Sa'id. As
the latter had not been able to assert his leadership, he had been
replaced by his younger brother Sulayman who started to give the
movement a new life. Although in 923/311 he was only seventeen
years old, he must have had considerable talent for leadership. The
constant appearance of the figure 17 in his first moves suggests that
he was trying to stress that the past quiet 17 years had been wasted,
and that he was determined to celebrate his own seventeenth
birthday with a bang. It is futile to try to establish whether or not his
movement was related to that of the Fatimids in North Africa. Any
reports or arguments to this effect are far-fetched and are always
contradictory. Such reports might even have been encouraged and
circulated by the parties concerned themselves. After all it does
not hurt to let one's enemy think that one has support from other
quarters. However, the fact remains that the activities of the Qara-
mita of Bahrayn and the Fatimids were not co-ordinated, and that the
first time they came into touch with each other they fought against
each other.3
The designs of Sulayman and his Qaramita were aimed at captur-
ing a share in the profits of the trade of the Persian Gulf, the Syrian-
'Iraqi desert and the Arabian desert. They had been granted trading
rights in Siraf, but they had soon come to realize that this was
literally a drop in the ocean.4 Therefore, under the vigorous leader-
ship of Sulayman, they proceeded to occupy the ports of 'Uman and
thus dominated all the Arabian coast of the Gulf. They next tried
to establish points of control on the Persian side of the Gulf. Their
attacks on Basra were enough to frighten the trade away from this
1
Miskawayh, vol. I, pp. 104-5; "Arib, pp. 110-11.
1
' See below, p. 204. • See above, p. 14}.
https://doi.org/10.1017/CBO9780511622380.008
flourishing port and divert it to their own ports where they imposed
and collected their own taxes.' For the land routes they first turned
to an easy target, the pilgrimage caravans, which were also important
trade caravans. The sanctity of the occasion did not deter the
Qaramita, though they were not against the pilgrimage in particular;
they were against its use by the merchants to enhance their profits, in
much the same way as many of us now are against the commercial
exploitation of Christmas. Having thus announced their intention,
the Qaramita turned to the north-south trade route skirting the
Syrian-'Iraqi desert, and tried to attack and dominate the market-
towns along this route. Their intention, which was eventually
achieved, was to collect what they considered taxes and what their
opponents preferred to call protection money.1
The rulers in Baghdad were naturally alarmed by this new threat
to trade, the more so because of the ease with which the Qaramita
won their military successes. It was not that the latter had formidable
forces - the biggest number mentioned is 2700 men - and it was
this very fact that daunted them.3 It clearly indicated that the Qara-
mita were able to win their victories only because they had the
sympathy of the local population of the areas in which they operated,
especially in 'Iraq. Something was manifestly wrong with a govern-
ment that lost the support of its subjects to that extent. If it were to
continue to govern some radical changes had to take place, and fast.
It is no wonder that our sources associate the Year of Ruination
923/311 with a military near take-over of control of government.
That this did not happen at this point was due only to the fact that
general Mu'nis, the only military leader who could accomplish it,
at the time, did not want it.4 Over the years he had acquired more
strength, as men realized that the one way of ensuring their payments
was to put themselves under his command.5 He had arrived at a
working arrangement with the administrators to guarantee the
regularity of these payments, and in exchange had supported them
as long as they had kept their end of the bargain. He was not against
military-administrative iqti? per se, but he took care that it should not
be applied in areas where the revenues had been allocated for the
payment of his men, i.e. Syria and Egypt. Military leaders in other
provinces believed that this type of iqta1 was the only way to salva-
1
Hawqal, p. 2 5 ; Mas'udi, TanbJb, p. 393; Miskawayh, vol. 1, p p . 139, 284.
2
Miskawayh, vol. 1, pp. 145-6, 173-82, 2 0 1 ; 'Arib, pp. 118, 134; Takmila, p. 5 3 ;
§abi,Qaramita, p. 56; Mas'udi, Tanbib, p. 390.
1
îbi, Qaramita, p. 49. « §5bi, Wizard', pp. 45-6.
> Miskawayh, vol. t, p. 115.
https://doi.org/10.1017/CBO9780511622380.008
THE BREAKDOWN O F C E N T R A L G O V E R N M E N T (il) IJJ
tion, especially under the circumstances of the Daylamite torrent.

The most important military commander in Baghdad, Nasr, was
convinced that a military take-over was inevitable and acted as if it
had already happened.'
The civilian officials, whether in the capital or the provinces, were
divided in their opinions about their own relationship to the
military. While most of the Jarrahs and their associates were willing
to continue to co-operate with the military, some of them had no
compunction about being completely subservient to them. The
Furats remained adamant in their convinction that military leaders
should be subjugated to the civilian authorities. For this the Furat
B>a%tr and his son had to pay with their lives, on the insistence of the
military leaders, after their dismissal from office in 924/312.2 The
remaining eight years of Muqtadir's reign were dominated by the
struggle between various factions of the military and civilians to
establish a new order. The situation was not helped by the fact that
Muqtadir, who had come of age, saw no reason why he should not
exercise his prerogative and actually rule. In this period of extreme
instability there were nine waîrs, one of them staying in office no
more than two months. In 929/317 Mu'nis, worried about the
diminishing revenues and the effect on the payments of his men,
organized a coup against Muqtadir. The meddlesome ruler was
deposed and a more mature 'Abbasid was installed in his place under
the title Qahir. However, the elite forces of Baghdad, whose func-
tions were now mostly ceremonial, supported by some of the police
force, were soon able to reverse the situation and reinstate Muqtadir.3
Confident in his renewed strength, the latter audaciously proceeded
to establish civilian supremacy over the military. For this purpose,
Husayn b. Qasim, a descendant of a long line of nadirs, was put in
charge of the whole administration in 931/319.4 To emphasize his
responsibility and authority a glorified title was bestowed upon him,
and it was decreed that his name should appear on the coinage along
with that of the ruler.5 To begin with, he took advantage of a
developing tendency in some provincial areas to set his policy in
motion. Some of the civilian administrators in the provinces had
begun to recruit their own private amies from among the available
Daylamites, and had used such forces to set up a semi-autonomous
administration in their respective areas. Such officials offered an
1 2
Vakmila, p. 57. ibid., p. 137.
3 4
Ibid., pp. 187-200. Sec above, p. 118.
5
'Arib, p. 165; Miskawayh, vol. 1, p. 223.
https://doi.org/10.1017/CBO9780511622380.008
Ij6 THE BREAKDOWN OF CENTRAL GOVERNMENT (il)
ideal opportunity to demonstrate the principle of civilian supremacy.

They were therefore recognized and confirmed as appointees of the
central government in charge of ail affairs in their regions.1 This was
easy enough, but when the wayir turned his attention to the military
it proved a different matter. His attempts to divide and rule were
too transparent to achieve the desired results and succeeded only in
pushing Mu'nis against the wall. When the latter began to show
definite signs of disagreement, the glorified waîr was hastily dis-
missed after only seven months in office. But it was too late; Mu'nis
marched on Baghdad and Muqtadir was killed in the ensuing fight in
932/320.*
Once more Qahir, who had been Mu'nis's choice three years
earlier, was proclaimed the new ruler. Mature he might have been;
stupid he certainly was, for as soon as he came to power he got
himself immersed in conspiracies to support one military faction
against another.3 In 933/321 he succeeded in having Mu'nis elimi-
nated, but the murder of the old and respected general only helped
to remove from the scene one more element of stability. As all this
strife was taking place in Baghdad, the population there, with its
Hanbalite and Shi'ite tendencies, was drawn into the conflict.4 The
conspiracies multiplied and finally the conspirators turned against
Qahir and had him deposed after a turbulent reign of only eighteen
months, 932-4/320-2.
At this point the military commanders, realizing that none of them
had enough strength to dominate the others, decided to give the
politicians, so to speak, the opportunity to resolve their own
dilemma. Towards this end they agreed on the succession of a son
of Muqtadir, who was proclaimed as Radi.s Though an inexperienced
young man, at most in his mid-twenties, he made a valiant effort to
meet the crisis. He appointed a waîr, who included in his team a
coalition of Jarrahs and Furats, and gave him a free hand to do as
he saw fit.6 It was almost inevitable that, under such circumstances,
the waylr should find himself in conflict with one military leader or
another. After a year of hard work trying to reconcile the interests of
the government with those of the military leaders in Baghdad and
the provinces, he had to take some stern measures against one
particularly officious commander. As this action did not bring about
1
Miskawayh, vol. i, pp. 147, 158-9, 166; Athir, vol. v m , p . 135; 'Arib, p. 158.
2
3
Ibid., pp. 2597-64. * Ibid., p. 522; Athir, vol. v m , p. 204.
5 6
Miskawayh, vol. I, p. 290. Ibid., pp. 2 9 2 - ; ; Athir, vol. v m , p . 2 1 1 .
https://doi.org/10.1017/CBO9780511622380.008
any immediate reaction from the latter's supporters, the waîr

thought his position was secure enough for him to take matters into
his own hands and put the military in their place. He then dismissed
his coalition team and replaced them witii his own son as a second
vaîr.1 Obviously he overestimated his power because in a few
months the military had forced him out of office, and replaced him
with their favourite Jarrahs.2 In two months they too had to resign
and were succeeded foi another few months by another two
watfrs. The civilians had declared their bankruptcy, and Radi had no
alternative but to call in the military to take over all affairs of
government.3
This year 936/324 marked the beginning of another decade of, if
it were possible, even more turmoil. A new office was created, the
holder of wliich was called Amir al-Umara\ supreme commander,
and was invested with complete powers over all military forces as
well as over all government departments. He was not only made the
superior of the wayir but also put in charge of all fiscal affairs in the
capital and the provinces. This was all very well and might have
worked if the central government had had anything left to give. The
treasury in Baghdad could hardly claim to be a central treasury and
in any event it was empty. The "provinces" had been reduced to the
immediate vicinity of Baghdad. The Samanids were still in control
of most of their domains in the East but were busy trying to stem the
Daylamite tide. The Daylamites were roaming over the central
Iranian Plateau fighting each other, and everybody else, to make
room for themselves. Those who were squeezed out by their fellow
Daylamites drifted to southern 'Iraq where they enlisted in the
service of the Baridis, the former tax-officials who now controlled
the region.4 The Qaramita had mellowed a little but were still
dominating east Arabia and making their presence felt in the rest of
the Peninsula. Jazira and the thughur were in the hands of the
Hamdanids, while Egypt and Syria were virtually autonomous under
Ikhshid.s
The installation of a strong man and the sacrifice of the long
cherished principle of the separation of military and fiscal affairs, did
not save the empire because there was no empire to be saved. These
desperate measures had produced nothing but more chaos. In a
period of ten years 936-45/324-34 five military commanders fought
1
Miskawayh, vol. 1, pp. $05-6, J I 8 - I J .
1 J
Ibid., p. 3)6. Ibid., pp. 351-2.
• Ibid., pp. 254-8, $01. » Ibid., pp. 566-7.
https://doi.org/10.1017/CBO9780511622380.008
over and succeeded each other in the supreme office. At one point
the men in the ranks demanded that an 'Abbasid should be appointed
to this office, but this was only an echo of things past.1 Neither the
Qaramita nor the Daylamites would take much cognizance of such a
figurehead; the former were brought into the fighi as allies of one of
the contenders and the latter were recruited by all of them.2 The
main forces of the Daylamites under the effective leadership of the
Buyids were closing in on 'Iraq from the east and the south-east.3
Their objective was to capture Baghdad and, in the circumstances,
this was a more realistic aim than saving it. They needed only to bide
their time until the forces of the capital had exhausted themselves in
futile fighting. When Radi died in 940/329, he was succeeded by his
brother Muttaqi who got himself too involved in the struggle; he
was easily deposed and was blinded for his efforts in 944/333.* He in
turn was succeeded by another brother Mustakfi who witnessed the
end of this episode. It came soon after his succession when the
office of the supreme commander fell to a civilian official, which in
effect meant that the military had given up trying to save the
central government of Baghdad.
• §uli, Radi, p. 204.

1
Miskawayh, vol. it, p. 24; Athir, vol. v m , p. 278.
1
Miskawayh, vol. 11, pp. 57, 55.
• Ibid., pp. 67-72; §uli, Radi, pp. 248-9.
https://doi.org/10.1017/CBO9780511622380.008
8
THE BUYID CONFEDERACY
The collapse of the centrd government and the emergence of the
Daylamites mark the beginning of the division of the Islamic lands
into two distinct spheres, the line of demarcation running along the
western slopes of the Zagros mountains down the east coast of the
Persian Gulf. To the east of this line the Iranian world, though by
no means united, consistently regained its identity. To the west, the
Arab world, also disunited, continued to survive almost on the same
lines as we know it today. The Arab conquerors had succeeded in
pushing the Buddhist wall as far back as the Chinese borders and
thereafter the regions of Outer Iran had become an integral part of
the Islamic domain. Yet these regions of what the Arabs had called
the East, because of the circumstances of the conquest, had con-
tinued to have a rather special status within the empire.1 On the
other hand, western Iran, the seat of the Sasanians, had been more
affected by Arab penetration and subsequent control. From the
point of view of the Arabs the areas extending eastwards into the
Iranian central plateau had been considered their property. In the
first two centuries of Islam, Tabasayn, to the north-east of the Dasht-
i-Kavir desert, had been the border point and the "door" to the
East.2 Under the Tahirids, this point had been pushed westward to
Rayy, and now it was the turn of the Daylamites to push it further
west into 'Iraq. These people, mainly the highlanders of the Elburz
mountains, who had remained virtually independent under both
Sasanians and Arabs, swept down through the central plateau and
almost unintentionally revived the claims of the Iranians to their own
territory. It would be incorrect to give the impression that this was a
nationalist movement, because it was not. Indeed the first reaction of
the Daylamite leaders, after settling down in their new domains, was
to immerse themselves in Arab culture and adopt it as their own. All
the same, they gave other Iranians, especially those of Fars, the
opportunity to reassert themselves over the more Arabicized
population of 'Iraq. It is interesting to note that later non-Iranian
1
Shaban, Tbt 'AbbSsid Rttolution, pp. 1-15.
1
Bal., Fmeb, p. 40J.
[159]
Downloaded from https://www.cambridge.org/core. University of Warwick, on 02 Aug 2017 at 22:26:31, subject to the
https://doi.org/10.1017/CBO9780511622380.009
l6o THE BUYID CONFEDERACY
peoples, the Saljuqs and the Mongols, moved further in this direction
and thus helped to buttress the claims of the Iranians.
In their own territory the Daylamites were ruled by various
princes, some of whom were of ancient lineage; or by local tribal
chiefs and there were as many chiefs as there were hills in the Elburz
range.1 These simple but important facts were reflected in the three-
pronged outburst of these people into neighbouring regions. One
group led by the prestigious line of the Musafirids, while maintaining
their home base, branched out and established control over an
enclave on the Caspian coast in Adherbayjan. Their influence with
their peers among the autonomous or independent rulers of the
Caucasus region, helped them more than their military forces to
preserve their interests for almost seventy years.2 Other Daylamite
leaders of less distinction but with more forces led their followers
east where for a time they served in the Samanid armies. Eventually
they were able to occupy the prosperous regions of Tabaristan and
Gurgan where they set up the Ziyarid dynasty and upheld it for
over a century until these regions were taken over by the Saljuqs in
1030/421. Significantly the Ziyarids made a point of staying on good
terms with the Samanids and their successors the Ghaznavids, yet
they had a protracted conflict with their fellow Daylamites, the most
important of whom were the Buyids.
It is ironic that the Buyids first made their name in the service of
the Ziyarids. Three brothers whose father's name was Buyeh com-
manded between them a few hundred men. Whether by design or by
accident their energies were diverted southwards where they
achieved some measure of success. This initial success and the
opportunities it offered persuaded other leaders to join them with
their men.3 Under this kind of collective leadership the Buyid forces
grew steadily and soon they found themselves in control of most of
southern and western Iran. Their pressure against southern 'Iraq
helped to speed the downfall of the central government and when
this happened they entered Baghdad unopposed in 945/334. They
did not think this feat had required special effort or meant a change
of plan. The peculiar arrangement they had had, almost from the
beginning of their successful adventure, continued in operation, as
if acquiring this big metropolis was of no more importance than
conquering a provincial town. Indeed this arrangement lasted,
almost unchanged, throughout their hundred and ten years in power.
1
Hawqal, pp. $76-7- ' U>tf., pp. 348, 554.
1
Miskawayh, vol. i, p. }5j, vol. II, pp. 5, 122; Athlr, vol. v m , p. 201.
https://doi.org/10.1017/CBO9780511622380.009
THE BUYID CONFEDERACY l6l
It was a Buyid Confederacy under which the regions remained

separate, each with its own forces and resources. Co-operation for a
specific purpose was desirable, and it was given, but unification was
out of the question. When it happened by force of circumstances in
978/367, it was only for five years and no effort was made to per-
petuate it when it lapsed in 983/372. The corollary here is that the
Buyid brothers must have been awire of the spirit of regionalism
prevalent in these territories and it was decided therefore to allow
every region to pursue its own interests with the least possible
interference from others. They must have also been aware of the
nature of their own position among their followers. Although other
chiefs had placed themselves under Buyid leadership, their individua-
lity and their recent jealously guarded independence in their own
mountains would still have to be taken into consideration. Accepting
the principle of their own equality under this system of collective
leadership, the brothers were in effect setting an example to be
followed in the regions, where each one of them would be no more
than an equal among equals. Thus other chieftains would also have
some measure of power in their new domains. Accordingly, one
brother was set up in the northern sector of the plateau from Rayy
to the Jibal province. Another, the eldest, was in charge of the Fars
province. The youngest was given the southern coastal provinces of
Kirman and Khuzistan, whence he was able to march into southern
'Iraq and finally Baghdad. The fact that he was left in charge of all
these regions has its economic and political significance. From an
economic point of view these regions controlled almost all the inlets
of the Indian Ocean trade in addition to Baghdad, the main entrepot.
Politically, the principle of equality as between the brothers was
emphasized by leaving the youngest in charge of 'Iraq. On the other
hand, this says a great deal about the contempt the Buyids felt for
the seat of the defunct central government.
Having decided the general lines of their rule the Buyids had to
turn to the infinitely more complicated question of how to apply
them in detail to the existing circumstances in their territories. It
has to be recognized that these rather unsophisticated people had
thrust themselves upon a developed society. Their task was more
difficult than that of the Arab conquerors. The latter had inherited
functioning governments and their concern had only been to keep
them operating. In contrast, the Buyids had inherited a brokendown
system of government for which they now had to provide a sub-
stitute. Rescue came from the provincial oligarchy of the long
https://doi.org/10.1017/CBO9780511622380.009
l6l THE BUYID CONFEDERACY
subdued regions who saw in the advent of the Buyids a chance to

press for their interests. It was from Fars, the centre of the ancient
Sasanian civilization, and particularly from Shiraz that most of the
members of the Buyid bureaucracy were recruited. Generations of
families serving in the same functions helped the Buyids to administer
their three domains.' In this situation the office of waîr acquired
more power than it had ever had before. The 'Abbasid rulers could
have governed without nadirs and had done so; the Buyids could
not, for their experience was very limited.
It was fortunate for the Buyids that they had arrived at Baghdad
via Fars, because as soon as they entered the former they badly
needed experienced advice. Their first problem was what to do with
the 'Abbasid Amir al-Mu'minin who was also the recognized Imam
and Khalifa of most of the community. Their first instinct was
probably to get rid of such an unnecessary figurehead, but they were
reminded that this would create more problems than it would solve.
They could not be expected to stop to think about the finer points
of the juridical differentiation between his various functions. They
themselves were Shi'ites only because they had recently accepted
Islam in this form. As they had been converted at the hands of
Zaydis they were supposed to belong to this sect. But at this stage
they could not have had more than the most rudimentary knowledge
of the development of Shi'ism. It is very possible that they did not
know of the existence of other Shi'ite sects, let alone of the contend-
ing schools of orthodoxy, until they had left their native land. It is
no wonder, then, that throughout their rule they tolerated all sects
and schools and tried only to keep the balance between them. They
did not consider the Qarimita extreme heretics, nor did they worry
unduly about Isma'ilt missionaries in their domains. They fought
against and allied themselves with the Shi'ite Hamdanids but
certainly not on religious grounds. They did not have as much
ditnculty with the Hanbalites of Baghdad as these themselves had
with their orthodox Shafi'ite fellows. However, despite their
differences, all orthodox circles were agreed on the necessity of the
continuation of the office of Amir al-Mu'minin. Over the years
the 'Abbasids had succeeded in hammering into the conscience of
the community the fact that the administration of justice and the
conduct of the religious life of the community must be the respons-
ibility of the holder of this office. As this did not interfere with the
Buyids' political objectives they were only too happy to oblige and
1
MUkawayh, vol. n, pp. 120-1, 124, 147-8, 260, 169, 301; Iâlchri, pp. 147, 149.
https://doi.org/10.1017/CBO9780511622380.009
THE BUYID CONFEDERACY 163
to allow the office to continue, but only for these specific purposes.'
All secular powers remained in their hands, a fact which was signalled
by their bestowing on themselves an endless list of honorific titles.2
The question of the military forces of the fallen central govern-
ment was another problem that confronted the Buyids in the pro-
vinces, and which became acute in Baghdad. They were keenly
aware of their own shortcomings. They knew better than anyone
that there were not many Daylamites left behind in their native
mountains. If they were to control their vast domains they would
have to supplement their forces from another source. They also
knew that as mountaineers their fighting capability was only as
infantry, and therefore in the southern and western plains they
needed the protection of a cavalry force.3 The forces of the fallen
adversary offered a solution to both problems. In the provinces, as
there were not many of them, they were quietly absorbed in the
Buyid forces and treated on an equal footing with regard to pay. But
this in itself represented another problem for there was not enough
cash on hand to pay all concerned. The only assets available to the
conquerors were the extensive land-holdings of the 'Abbasid family
and their associates. These were taken over by the Bviyid administra-
tion and a new type of iqtd' was introduced for the benefit of the
men and their commanders for whose pay there was not enough
ready cash. Such men were assigned the income from a specific
land-holding in lieu of their fixed stipends or as a supplementation
to them. This was meant to be a temporary arrangement which
could be changed at the wish of these men or the administration, and
the ownership of the land remained in the hands of the latter.
In Baghdad the situation was different because of many factors.
First, the remnants of the armies of the capital, although divided
among themselves, were still numerous enough to make them a force
with which to contend. Second, of course there was no farm-land
in the city, but there was much of it nearby and still much more to
be taken over in other parts of 'Iraq. Third, there was more cash
available than in the provinces. Fourth, while the new Buyid ruler
would need his Daylamites, for obvious reasons, to remain in
Baghdad, the old forces could be utilized to extend his rule in the
rest of 'Iraq. As these forces outnumbered the Daylamites, there was
1
Miskawayh, vol. 11, p. 86; Mas'udi, Tcaiblb, pp. 399-400.
* See the very interesting list included by Cl. Cahcn, Encyclopaedia of Islam (New
Edition), Leiden, 1954, s.v. "Buwayhids".
• Athir, vol. VIII, p. 47}; Miskawayh, vol. u, pp. 77, 529.
https://doi.org/10.1017/CBO9780511622380.009
164 THE BUYID CONFEDERACY
no question of absorbing them into the latter and thus the former
remained a separate force under their own commanders. To pay
them the Buyid iqtd' system was extended to 'Iraq, to include not
only the farm-lands there, but also inter-regional trade taxes.1
Naturally this arrangement would help to establish the administra-
tion's authority over more territory and more revenues, and at the
same time secure the pay of these forces. As for the Daylamites, it
was determined, almost as a favour, to pay them'directly from the
treasury in Baghdad. Although the Amir al-Mu'minin himself was
now paid an allowance from the same treasury after his properties
had been taken over by the Buyid administration, the Daylamites
were in a better position to demand what they wished.2 They con-
sidered it an affront to give the defeated forces an assured pay while
they had to rely on a possibly empty treasury. The Buyids had to
give in and grant their leaders the same privilege in areas more
accessible to Baghdad. These ad hoc arrangements, which were
supposed to be temporary, continued throughout the Buyid domains
until their downfall, despite attempts by the rulers and even demands
from the beneficiaries themselves to revoke them. They created a
two-tier system for the payment of the military forces and neither
side was completely happy with its deal. Competition for more and
better farm-land was intensified by demands for a change from one
system to the other. In most cases the collateral security rights were
passed from father to son and consolidated into outright ownership.
Some of the holders of iqtd' neglected the irrigation systems and
when these were destroyed, they simply demanded to exchange the
ruined lands for others still in production. Other holders used their
iqtd' as means to acquire more land, using their influence to intimi-
date both the tax-collectors and neighbouring landowners.3 The
land-tax collection continued under the tax-farming system and
gradually the tax-farmers gave up the impossible task of collecting
their dues from such influential land-holders.4 The harmful effects of
tax-farming were greatly increased by the instability of the iqtd'
system, to the utter detriment of the treasury. The attempts of some
Buyid administrations to win the support of a particular army group
did not help the cause of the treasury either. One of the most curious
examples occurred in Baghdad in 958/347, where the ruler had
never been happy with his Daylamites since they had exacted iqtd'
from him in 945/334. In order to win over the old troops completely
1
$abi, Rdjd'i/, pp. )4}, 558. * Miskawayh, vol. II, p. 87.
1
Ibid., pp. 96-100. • Hawqal, pp. 2)9-40.
https://doi.org/10.1017/CBO9780511622380.009
and use them to restrain the Daylamites, the Buyid authorities

devised an ingenious way of giving them a bonus. Those men whose
pay was assigned to farm-land in southern 'Iraq were to travel there,
one group after another in rotation, to collect their pay. This was not
exactly necessary but to arrange for them to receive free hospitality
en route in addition to a travel allowance of ten dirhams per day for
every man with the rank olghuldm and twenty for the higher rank of
naqlb, defies reason. As our source puts it, they found it both con-
venient and profitable to prolong their travels; the allowances gave
them a comfortable income and accordingly they invested their
salaries in all sorts of commercial enterprises.1 Later when the
authorities wanted to stop this mockery they were met by very
stubborn resistance.2
The teething difficulties, even though some of them turned out
to be chronic diseases, did not hinder the Buyids from looking
outside their domains to examine their situation vis-a-vis their new
neighbours. The interesting fact here is that their attention focussed
simultaneously on all four corners. Of course their regional division
helped them to cope with more than one front at a time, yet there
was, as was to be expected, a good deal of co-ordination of effort
especially at this early stage. Sitting astride the Iranian plateau and
extending into 'Iraq, the Confederacy was in an advantageous
position to dominate most of the east-west and north-south trade
routes. Furthermore, its southern shores could be encouraged to
attract a good deal of the Indian Ocean trade. Obviously the Buyids
had good advice and also knew their own limitations, for they
proceeded systematically to make the best of their situation with a
minimum of military adventures, using peaceful means whenever
possible. Therefore, they ruled out of court any conflict with either
Byzantium or Egypt and throughout their rule refrained from inter-
fering with these two powers even when the opportunity to do so
offered itself.3
The activities of the Buyids in the Gulf are the best illustration of
their methods and intentions. Soon after occupying Baghdad they
set out to capture Basra, which had been in the hands of the tax-
farmers, the Baridis, since the last years of the central government.
This was an easy task; the important port was captured and the
Baridis fled to the Qaramita in Bahrayn. The Buyids then reimposed
1
1
§abi, Tarikb, part 8, ed. H. F. Amedroz, and published as vol. iv in Dbayl Tajirib al-
3
Umam, Cairo, 1916, pp. j 74, 587. Qalinisi, p. 11.
https://doi.org/10.1017/CBO9780511622380.009
166 T H E BUYID CONFEDERACY
the trade-taxes that had been abrogated by the Jarrahs over thirty
years earlier. At the same time they reduced the land-tax that had
been increased by the greedy tax-famers.' Having established them-
selves in Basra the Buyids proceeded to woo the Qaramita. The
latter had reached the peak of their agitation under Sulayman when
in 930/317 they had committed the unforgivable sacrilege of looting
the Ka'ba and carrying away the Black Stone. This act is not quite
as bad as it sounds for they could also have destroyed the Ka'ba
itself had they wanted to do so. They had nothing against the
pilgrimage itself; their action was the clearest possible assertion of
their objection to the use of this religious occasion for commercial
purposes, from the benefits of which they had been excluded.
Attacking the pilgrimage caravans had not always been possible and
had not achieved the desired results. Until such time as arrangements
were made with them the whole enterprise had better be stopped
altogether. The best way to accomplish this was to take away the
Black Stone which was, and still is, indispensable to the rituals of
the pilgrimage.
Although the Qaramita seem to have been agreed on this act, they
had had their differences with regard to their attempts to establish
control over the Persian side of the Gulf. These differences had
manifested themselves in a doctrinal and political struggle among
them until the death of Sulayman in 944/332. Therefore when the
BQyids captured Basra four years later, the Qaramita were not averse
to being wooed. The arrangements reached by the two sides made
it possible for each to live at peace with the other. It is surprising to
find that the Buyids allowed the Qaramita to have their own customs-
house alongside theirs in Basra. It is almost certain that similar
arrangements of mutual convenience to both sides were made for the
all-important pilgrimage land-routes leading in and out of Baghdad,
Ktifa and Basra. We are fortunate to have examples of the duties
levied at some of these points; four dirhams a sheep in Basra; sixty
dirhams a pilgrim for every camel-load, and fifty for a mule-load in
Kufa and Baghdad; in Basra one hundred dirhams for any load, and
in all three places one hundred dirhams on any load of linen cloth. 2
The Buyids, who were willing to go to some lengths to win over the
Qaramita, were also ready to discourage trading with the port of
Slraf and divert all trade to Basra where both could share its benefits.3
Soon all the East African Zanj, well-established in 'Uman, felt the
1
Miskawayh, vol. 11, pp. 25, 88, 112, u j , 127-9.
1
MaqdisI, pp. 135-4. ' Ibid., p. 426.
https://doi.org/10.1017/CBO9780511622380.009
THE BUTID CONFEDERACY 167
pressure of this competition and persuaded a faction of the Qaramita

to co-operate with them in an attack on Basra. The Buyids over-
looked the participation of the Qaramita and directed their rancour
against the Zanj. In 965/354 an army from Basra and a navy from
Siraf successfully occupied 'Uman for a time. When this was not
enough another expedition was dispatched there in 974/363 and put
an end to the whole affair, while the Qaramita did not have to lift a
finger.' Moreover, the Buyids took it upon themselves to police the
waters of the Gulf and curtail the activities of the pirates who were
operating from Qatar, next door to Bahrayn.2 On the other hand the
Qaramita actively supported the Buyids in their fight against the
Samanids to maintain their hold over Rayy.3 It is worth noting that
this is the only area where the Qaramita twice made the effort to help
their partners in this uneven alliance. One can only conclude that
they, the Qaramita, were interested in retaining their access to Rayy,
an important link on the east-west trade-route. When it came to the
repeated fighting between the Buyids and the Hamdanids to the
north, the Qaramita remained absolutely neutral. Indeed they kept
on very good terms with their former enemies the Hamdanids, who
in appreciation of this stand were quick to send them a present of
iron ore when it was requested.4
It was under these cordial circumstances that the Qaramita
decided, of their own accord, in 951/339 to return the Black Stone
to Makka after keeping it for 22 years.* Their newly acquired gains
and respectability in addition to their good relations with the
Buyids, who had their politics to worry about in Baghdad, con-
vinced them that keeping the Stone did not serve any purpose. Any
reports that they returned it on the orders of the Fatimids in North
Africa have been invented to give credit where it was not deserved;
and any arguments put forward by modern scholars to support these
reports are without foundation.6 Indeed the Ikhshidids of Egypt and
southern Syria were in a better position than the Fatimids to exert
influence with the Qaramita to whom they paid the handsome
amount of 300,000 dinars a year.7 This offering was not to encourage
the religious principles of the Qaramita, neither was it a result of any
common cause between the two sides. It was to secure the safety of
1
Miskawayh, vol. n , pp. 46, 145-4, 196, 213-18.
Ibid., pp. 300-1; Hawqal, p. 48; Ijtakhri, p. }}.
Ibid., p. 205. * Ibid., pp. 126-7.
Bernard Lewis, The Origins oflsmailism, Cambridge, 1940, pp. 81-2.
Athir, vol. vni, p. 4; 2.
https://doi.org/10.1017/CBO9780511622380.009
l68 THE BUYID CONFEDERACY
the caravans travelling on the land-routes from Egypt and Syria to

Hijaz. As we shall see, when the Fatimids, upon their appearance in
Syria 971/360, cut off this subsidy, they invoked the wrath of the
Qaramita who did not hesitate to attack them.1 Significantly, the
Buyids remained in the background, almost unconcerned, satisfying
themselves and their allies the Qaramita by supporting them with
money and arms.2 In the event the Qaramita were defeated, but in
the process they went through a turmoil of organizational change.
They installed their own model of collective leadership in the form
of a council of six men presiding over a community with some
remarkable progressive features. Communal property, no taxation
and social security were the principles of this new regime.3 Sur-
prisingly, even the orthodox circles of Baghdad did not object to
the constitution of such a dangerous precedent in an area so near.
If anything the relationship between the two sides was strengthened.
The Qaramita were given land-grants in southern 'Iraq, allowed a
considerable presence in Kufa and influential representation in
Baghdad itself.4 However, they were soon drawn into the internal
affairs of the Confederacy and supported one side against another in
a conflict over succession. The Buyids did not tolerate this unwar-
ranted interference and inflicted a military defeat upon the intruders
in 986/375. s For all practical reasons this was the end of the Qaramita
who withdrew to Bahrayn where they led a quiet life.
In southern 'Iraq the Buyids had a problem reminiscent of that
which the central government had had with the Zanj a century
earlier, and in the same area. This was in the Batiha, literally the
marshland that covered the lower course of the Tigris and Euphrates
between Wasit and Basra.6 Swampy as it was, this area had con-
siderable fertile lands of which the main produce was rice.7 It also
controlled the river traffic, between Basra and Baghdad, which had
to pass through an intricate system of canals. During the years of the
decline of the central government and almost immediately before
the arrival of the Buyids on the scene, the natives of the Batiha had
decided to take their destiny in their own hands.8 The independent
little republic they established there was able to defy the repeated
attempts of the Buyids to subjugate it, and for over a century it
1
See below, p. 204. * Qalanisi, p. 1.
1
Athir, vol. vni, p. 506; Iâwqal, pp. 23-7; Nifii-i-Khusraw, Safar-Name, ed. C.
Schefer, Paris, 1881, p. 82.
* Aba ShujS', pp. 102, 109; Sibl, ROJJ'II, pp. }6o-8j.
' AbQ Shuji', pp. 109-10; Athir, TOL DC, pp. 27-9. * Mas'udi, Tanbtb, p. 40.
7
Maqditi, p. 119. * Athir, vol. VIII, p. 362.
https://doi.org/10.1017/CBO9780511622380.009
survived and flourished under its own native rulers. It was able not
only to impose its taxes on all the trade going up and down the river,
but also to exact tolls from all travellers, including military person-
nel. • The ideological orientation of this republic seems to have suited
all parties; while it maintained good relations with all Shi'ites, the
conservative elements of Baghdad found refuge in it when they
needed to escape Buyid intimidation.2
In the north-western corner of their domains the Buyids had a
difficult problem, the troubled and troublesome Hamdanids. These
Kharijitcs turned Shi'ites were as divided in their loyalties as they
were amongst themselves. Their interests were more involved with
the Byzantines than with their fellow Muslims. These native Arab
chieftains had less support from their fellow Arabs than from the
conglomeration of races in the region, including the Greeks. There
were two Hamdanid power centres whose interests were incom-
patible, the line of division running north-south from Armenian
Qaliqala (Erzurum) to Mawsil. To the west of this line the Hamda-
nids of Halab (Aleppo) were in competition with Tarsus, and for this
matter, southern Syria, in the Byzantine trade. They were willing
to fight their fellow Muslims as much as the imperial forces of
Constantinople to achieve their ends. Yet they had more allies
within the eastern Byzantine territories than they had in Syria or
'Iraq. As they had access to the desert land-route along the Euphrates
they had very little to do with Baghdad. The rulers of the latter were
quite happy with this relationship because they had no intention of
getting involved in controversies with the Byzantines.
To the east of that line, the Hamdanids of Mawsil were busy
promoting their interests in Armenia and Adherbayjan. They had
no interest in fighting the Byzantines and were satisfied to trade with
them through the Armenian intermediaries.1 At the same time they
were all intent on expanding their network into Adherbayjan, reach-
ing out north into the Caucasus regions, and east into the Jibil
province.4 In their relationship with Baghdad they were torn between
two factors. On the one hand, they appreciated the volume of their
trade with the metropolis down the Tigris and were reassured by the
strong support they had there.' On the other hand, they did not
want to be subservient to the Buyids to the extent of paying them a
Athir, vol. VIII, p. 47); §ibi, Tarikb, pan 8, p. 41 j .
Hawqal, p. 22j; Maqdisi, p. 120.
Athir, vol. v m , p. JIO.
Miskawayh, vol. n, p. 108.
https://doi.org/10.1017/CBO9780511622380.009
I70 THE BUYID CONFEDERACY
tribute, or indeed accepting the taxes on trade the latter levied at

Takrit to the north of Baghdad.1 The tense relationship between
these two parties contributed to the complexity of the situation in the
neighbouring regions to the east.
During the last years of Muqtadir's reign and the following years
of deterioration in Baghdad, the Byzantines stepped up the pressure
against the thughur and in the process captured Malatya.2 It was
mainly due to the Bulghar and Magyar attacks in the Balkans that
this pressure was not sustained. It is interesting to learn that, in
Constantinople, around that time, the emperor used to invite
Muslim prisoners-of-war to a Christmas dinner where pork was
specifically excluded from the menu, and actually gave them presents
of cash and clothes.3 With such a show of goodwill one begins to
understand the frequent incidence of the changing of sides and
religions as between Muslims and Christians on both sides of the
Arab-Byzantine borders.4 The toleration of one side was matched
by the magnanimity of the other. It was not quite an entente
between Christendom and Islam, but more in the nature of a
decision to co-exist taken by two adversaries. The almost continual
fighting on the borders was due more to the meddling interference
of importunate groups on both sides, especially the Arabs of the
thughur, than to the deliberate policies of either power. They both
realized the benefits that could accrue from trade and accordingly it
was encouraged to flow. The problem was that neither side was in
complete control of its own borders. Under the circumstances,
lasting agreements and arrangements of mutual convenience could
not be reached despite numerous embassies between Baghdad and
Constantinople.
When the Buyids came to power the Hamdanids had been well
established in Mawsil and the surrounding regions of Jazira for
almost forty years. During this period the latter had been deeply
involved in the politics of Baghdad. In 942/330 a Hamdanid had
attained the office of Amir al-Umard' but, after realizing the hope-
lessness of the situation of the central government, had decided to
withdraw to his safer seat in Mawsil.5 As heirs of Baghdad's domains
the Buyids, after some difficulty, were able to assert their authority
over the reluctant Hamdanids. The agreement they reached with
1
Ibid., pp. 115, 204-6, *8j. » Ri<#, p. 251; Hawqal, p. 181.
1
Rusteh, pp. 122, I2j, 125.
4
Hawqal, p. 211; Miskawayh, vol. II, p. J54; Athir, vol. vm, pp. 415, 447, 491.
5
Miskawayb, vol. 11, pp. 28, 44.
https://doi.org/10.1017/CBO9780511622380.009
THE BUYID CONFEDERACY I7I
them provided for a tribute to be paid by the latter, but as it was not
their intention to keep to its terms it was more often broken than
not. • While this uneasy relationship continued with the Buyids, the
Hamdanids earnestly applied themselves to the business of con-
solidating their position in their territories. Although they had
control over all the tbughurt with the exception of Tarsus, the effect
of the fall of Malatya to the Byzantines had been to drive the trade
further south and east. This presented no problems to the Hamdanids
and in fact helped them by diverting the trade to the more control-
lable towns of Aleppo, Nasibin and Mawsil.2 However the Byzantines
had different ideas, and for the first time they put into effect a plan
that would enable them to take full charge of the trade with all their
eastern neighbours. Henceforth all this trade was confined to two
ports of entry. One was Trebizond on the Black Sea coast where it
was decreed that all the Muslim merchants should gather to enter
Byzantine territory. The other was Antaliya on the Mediterranean
coast to the west of Tarsus, and here it was that the sea-trade would
be carried out.3 This simple plan had far-reaching repercussions for
all concerned. It enabled the Byzantine authorities to enforce the
collection of the taxes due on all the incoming and outgoing trade,
and while the Egyptian and Syrian ports could continue to trade
with Antaliya, the role of the thughur as trade centres was clearly
ended. Furthermore, the prosperity of the Hamdanids of Aleppo and
Nasibin was put in great peril. Those in Mawsil were not particularly
affected except in so far as their relationship with the Armenians
to the north was concerned. Under these new arrangements the
latter, with the full support of the Byzantines, acquired a sudden
importance. Instead of being subservient to all their powerful
neighbours, they became major partners in a highly desirable trade.
Indeed their interests were now more involved with those of the
Byzantines than they had been for centuries.4
As was to be expected, this change in the pattern of trade created
a political upheaval of great magnitude. The first to react were those
who were most threatened, the Hamdanids of Aleppo. Starting in
948/336, for twenty years they tried desperately to re-establish the
position of the thughitr by incessant military campaigning, but to no
avail.1 It is significant that while the Hamdanids of Maw?il sent
Ibid., pp. 108, 115, 204-6.
Athir, vol. vin, p. 164; §ulf, RJJJ, p. 251; Hawqal, p. 214.
Hawqal, pp. 176, 178, 197-8, 344; lstalchri, p. 188.
Athir, vol. m , pp. 173, 377, 407, vol. ix, p. 67; Qudama, Kbaraj, p. 2)4.
Athir, vol. VIH, pp. 361, 365, 37J, 381, 387, 388, 393, 396, 399-402.
https://doi.org/10.1017/CBO9780511622380.009
17* THE BUTID CONFEDERACY
presents of horses, wine and golden crosses to the emperor in

Constantinople, those of Aleppo received presents of mules, clothes
and golden trinkets from the Byzantines of the eastern territories.'
This contradictory behaviour on the part of the latter can only mean
that they approved of the Hamdanid attempts to re-establish the
thugkur. Such consorting with an actively hostile enemy did not
escape the eyes of the authorities of Constantinople, and could have
been considered a factor contributing to the prolongation of this
unnecessary war. In 969/358 a major Byzantine counter-offensive
succeeded in capturing Antioch and Aleppo. As a result Antioch was
permanently occupied while in Aleppo the Byzantines set up their
own protectorate.2 From this point on the Hamdanids of Aleppo
ceased to be of any significance. They deteriorated into factions
fighting each other for survival, only serving as pawns in the ensuing
struggle between the Byzantines and the Fatimids over Syria.3 It
will be noticed that neither the Buyids nor indeed the Hamdanids
of Mawsil interfered to save their brethren of Aleppo from the fate
that befell them.4 Now that all the thugkur were in the hands of the
Byzantines, curious developments began to take place on both sides
of the borders. First, the clan of Habib, cousins of the Hamdanids
and numbering as many as 12,000 men, defected to the enemy, were
converted to Christianity and helped the Byzantines to capture more
Arab strongholds.5 Then,brewing discontent inthe eastern Byzantine
territories broke out in a rebellion led by two distinguished members
of powerful families in the area, Bardas Phocas and Bardas Sclcrus.
It could have been a coincidence that the latter held Hisn Ziyad, one
of the former Arab strongholds in the tbughur, but the fact that he
allied himself and inter-married with the Hamdanids of Mawsil
throws a different light on the matter. However, as his in-laws
failed to provide him with the support Bardas expected, Basil II,
eventually and soon after his succession to the throne in 976, was
able to subdue the rebellion with the help of the Russians. Bardas
Sclerus fled to the Hamdanids only to end up in Baghdad taking
refuge with the Buyids and seeking their help against his own
compatriots in 980/370. The Buyids, who had never wanted to get
involved with the Byzantines, had other more pressing problems
and kept Bardas under guard for five years until it suited their
1
§abi, Rasd'i/, pp. 152-3; Miskawayh, vol. 11, p. 208.
1
Miskawayh, vol. II, pp. 177, 180-1, 190-5, 210; Athir, vol. vm, pp. 404, 414-1;,
425, 434. 444-5. 4471 Antiki, p. 358; JHawqal, pp. 177-8.
' See below, p. 204.
• Athir, vol. vin, p. 4JJ. » Hawqal, pp. 211-12, 221.
https://doi.org/10.1017/CBO9780511622380.009
THE BUYID CONFEDERACY I73
purposes to send him back after some unsuccessful negotiations

with the Byzantines.'
Within the Confederacy, another group, the Buyids of Jibal, had
its problems too. Their geographical location left them in control of
the east-west trade-routes across the Iranian plateau and in a position
where conflict was possible with those who had a special interest in
this trade. As there was also trade coming down from the north
through the Caucasus, other interests were involved, such as those
of the Musafirids. These were the branch of Daylamites who had
managed to keep a foothold in their native lands while establishing
another in Adherbayjan on the Caspian coast. Their conflict with the
Buyids was limited to fighting for ascendancy in their home territory
and it was of no significance. On the other hand the Musafirid
enclave in Adherbayjan was of considerable importance at least in
its long-term effects. They managed to reach reasonable agreements
with the local princes and leaders of the region according to which
each group had an equitable share in the benefits of the trade; judging
by the revenues from one point, where taxes were collected on the
outgoing trade from Adherbayjan to Rayy, which amounted to
100,000 dinars a year, these benefits were considerable.2 The
Musafirids also maintained good relations with the all important
Armenians, but they did not reckon with the Kurds and that was a
serious mistake. Perhaps it was also a justifiable one; for centuries
nobody had bothered to take Kurdish interests into consideration.
These people had been left alone to carry on their semi-nomadic
way of life in their mountains without much outside interference.
Sometimes they had taken part in minor uprisings in the Jibal,
but hitherto one can hardly see any traces of them outside their own
territories. With the breakdown of the central government they had
begun to show signs of defiance such as pillaging passing caravans. 3
The rise of the Daylamites and the fact that they were now surround-
ing the Kurds on all sides must have not only given the latter reason
for alarm but also caused them to think about their own situation.
It was not difficult for them to realize that they were not very
different from the Daylamites, and if these people could assert
themselves so strongly, there was all the more reason why the Kurds
should control at least their own territories. With remarkable
dexterity they set out gradually and under various leaders to establish
1
Athlr, vol. viii, pp. 491, 506-ip, 517-18, vol. ix, pp. 12, JO-I; Abu Shuji', 58,
m - 1 2 ; Ba;riq, p. 82; An tiki, pp. 130, 166-7.
» tfawqal, p. 35}. ' Athir, vol. vm, pp. 44J-6. J'«; §ull. R*(0, pp. 192-3.
https://doi.org/10.1017/CBO9780511622380.009
174 T H E BOYID C O N F E D E R A C Y
their domain. They made a point of not antagonizing the Buyids

too much while at the same time strengthening their ties with the
Armenians and, accordingly, the Byzantines. > The Buyids on their
part tolerated the Kurdish advances, especially as these were increas-
ingly directed towards the north. Even when the Kurds set up their
own trade centres and improved the roads leading to them, the
Buyids did not take these actions as a direct threat to the prosperity
of their own trade activities.2 On the other hand the Musafirids
were greatly alarmed by the increasing Kurdish rivalry, particularly
when it began to show its effects in Adherbayjan.3 This inevitable
conflict ended with the disappearance of the Musafirids from their
enclave on the Caspian coast in 984/374, and the rise of more
Kurdish powers in Adherbayjan. Almost immediately the latter
clashed with the Hamdanids of Mawsil who were at that time under
pressure from the Buyids and the 'Uqaylids. The result was the
final collapse of the Hamdanids and the division of their Mawsil
domains between the Kurdish Marwanids who dominated the
eastern half of Jazira, and the Arab 'Uqaylids who occupied Mawsil
and the western half of Jazira.4 It is worth noting that these Arabs
had been among the first to settle in the thughiir more than two
centuries before.5 The Byzantine drive had forced them out of their
favourite haunts into the troubled domains of the Hamdanids.
Although the 'Uqaylids were to survive in Mawsil until after the
Saljuq invasion, their presence and behaviour were responsible for
many disorders in the intervening years.6
In the north-eastern corner of the Confederacy the most important
issue was, in one word, Rayy. The Samanids desperately needed to
control it, the Ziyarids of Tabaristan and Gurgan badly wanted it,
and the Buyids captured it.7 From this central position directly in
the centre of the bottleneck between the Elburz chain and the
desert, the Buyids were in an ideal situation to control the flow of
almost all the east-west trade. The Samanids, who had just been able
to attract the northern trade to their territories, came to be at the
mercy of the Buyids and their whole trade policy was a shambles. It
» Athir, vol. VIII, pp. 130,189, 37J-7, 519, vol. ix, p. 67; Miskawayh, vol. 11, pp. 150-1.
2
Abu Shuja', p. 290; §abi, Tarikb, part 8, pp. 433-4.
1
Athir, vol. vm, pp. 289-91; Hawqal, p. 334.
4
Hawqal, pp. 22J-6; Abu Shuja', pp. 83-6, 144, 175-9; Athir, vol. ix, pp. 2$-6,
49-50, idem, vol. X, p. 11.
» See above, p. 52. • See below, p. 204.
7
Miskawayh, vol. 11, pp. 100, 115, 119-20, 123, 129; Athir, vol. VIII, pp. 333-4, 351,
J6J-8, 578-9, 460-1.
https://doi.org/10.1017/CBO9780511622380.009
THE BUTID CONFEDERACY I75
was no use having all the goods of Europe if they could not have an
outlet for them. The Ziyarids, Daylamites as they were, found them-
selves in the odd position of being allied to the Samanids against
their fellow-countrymen. However, as the magnitude of the difficul-
ties of the Ziyarids was not as great as that of the Samanids, they
managed to survive and live through the Ghaznavid domination and
the Saljuq invasion. It was the Samanids, whose political structure
could not take the strain of the economic crisis, that began to
crumble. Admittedly, they had a very sophisticated government
machinery but such things were not unusual at the time - even the
unsophisticated Daylamites had been able to set up at least five
governments. One can appreciate Barthold's pioneering preoccupa-
tion with the details of Samanid administration, but it has to be
realized that he was looking at "Turkestan" from an insular point
of view. These two factors put severe limitations on the validity of
his interpretation of the broader issues. In the seventy-five years
since his study was first published, none of his conclusions has ever
been questioned. His disciples, perpetuating his narrow outlook,
have simply taken them up and reiterated them in variant forms.
Although Barthold recognized some elements of the Samanid
power-base, such as the Khwarizm-shahs whose origin "goes right
back into mythological times", and the Muhtajids of Chaghaniyan
who continued to rule their domain regardless of the rise or fall of
the Samanids, he could not relate these people to the political
structure of the Samanid regime.'
The Samanids, the heirs of the Tahirids, were basically presiding
over a federation of the ancient Principalities of Tukharistan, the
city-states of Soghdiana and the little Khurasan of the Sasanians, a
fact that Ibn Hawqal understood and clearly stated.2 The difference
between the Tahirids and their successors was a question of the
emphasis on the interests of the various areas, and the ascendancy of
one group over another. Nothing had happened since the Arab
conquest to necessitate a change in the geo-political structure of
these regions. Indeed the local leaders had continued to exercise
their authority, albeit under Arab suzerainty, long after the 'Abbasid
revolution. Afshin is but one example of such rulers, and the rise of
the Tahirids is a confirmation that families of old lineage had gained
even greater power under the 'Abbasids; just as the Muhtajids of
Chaghaniyan had done, the Farighunids of Guzgan held their own
1
Barthold, Turkestan, pp. 233-4.
» Shaban, Tbt 'Abbdsi4Knolution, pp. J - I J ; Hawqal, p. 430.
https://doi.org/10.1017/CBO9780511622380.009
176 THE BUYID CONFEDERACY THE BUYID CONFEDERACY
mul 1N
TABARISTAN ""1(1U
# Simnan
QUMIS QijiHISTAN
p shl
DASHT-i-KAVIRDES.ERT "
5 Iran and the East

Downloaded from https://www.cambridge.org/core. University of Warwick, on 02 Aug 2017 at 22:26:31, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9780511622380.009
I78 THE BUTID CONFEDERACY
under the Samanids, and for long after that.' The Simjurids had their
enclave between Harat and Nishapur.2 The family of 'Abdulrazzaq
from the local nobility of Tus were just as prominent in Khurasan
under the Samanids as the Kanarang had been under the Sasanians.3
It is not surprising that it was one of the 'Abdulrazzaqs who encour-
aged the poet Daqiqi, the forerunner of Firdawsi, to write the
Shdhndmeh, the Book of Kings and epic of the Iranian past. It is
significant that Daqiqi describes his patron as of dihqdn origin. If
such people of the original local nobility had survived in Khurasan,
there is no reason to assume that other local leaders had ceased to
exist in the other lands of the East. Indeed there is evidence in
abundance of their continued existence under the Samanids and
after. The rise in the fortunes of one family or the fall of the other is
in the same vein as the rise and fall of the Tahirids, Samanids or
Ghaznavids. The local rulers supported and served under the leaders
of the federation as long as it served the interests of their respective
areas. Of course there were conflicts and rivalries, but it was the
responsibility of the leaders of the federation to pursue a policy
which would reconcile these conflicts. Such a policy would have to
take into consideration the agricultural economy of Bukhara, the
commercial interests of Samarqand, the mineral resources of Khuttal,
the industrial output of Merv and Nishapur, the livestock production
of Guzgan and even the meagre produce of the semi-nomads of the
Hindu-Kush mountains. It would also have to recognize the long
established trade contacts between the various components of the
federation and the neighbouring lands in all directions, and har-
monize these with the interests of the federation.4 The Tahirids had
tried to steer a conservative course in favour of the dominant
agricultural interests. As we have seen, this had aroused the ire of
the people of Sistan and had contributed to the downfall of the
Tahirids.5 The Samanids were no less conservative than their
predecessors, but they were in a different situation. The industrial
expansion, in addition to the fact that their power-base was in
Soghdiana, obliged them to give trade an equal place with agriculture
in their economic policy. This equilibrium would have been main-
tained had it not been for the opening of the new trade-route into
1
Athir, vol. ix, pp. 69, 103; Maqdisi, p. 5)7; Ijukhri, p. 172.
• Athir, vol. vm, p. 292, vol. ix, pp. 18, 69.
» Miskawnyh, vol.ir.p. 117; Athir, vol. vni,pp. $53-4; Shaban, The'Abbasid Ktvolutton,
p. 5.
* Hawqal, pp. 432, 449, 430, 477; lstakhri, pp. 166, 244, 25;, 26;, 280, 281; Maqdisi,
p. 469; Athir, vol. ix, p. 305. * See above, p. 96.
https://doi.org/10.1017/CBO9780511622380.009
THE BUYID CONFEDERACY I79
Europe. The vast profits that this generated created a new economic
situation. Faced with this change the Samanid Nasr b. Ahmad was
forced, before his death in 943/331, to deviate from the orthodoxy of
his ancestors to the extent of being accused of Isma'ilism.1 We do
not know exactly what he did to deserve such an extravagant
indictment, but perhaps a clue can be found in the developments in
nearby Rayy and Tabaristan. There, about the same time, the flexible
Shi'ite Daylamites, who had just captured the region, had introduced
a novel tax of one dinar per head. Our source explains that this tax
was imposed specifically to balance the burden of taxes between the
merchants and the landowners who, hitherto, had had to carry the
load alone.2 It is possible that Nasr had in mind such a reform.
However, the economic decline that soon set in in the Samanid
domains as a result of the Buyid blockade at Rayy, could not be
remedied by a simple tax reform. It required an urgent solution; a
political one on which all members of the federation would agree.
The disagreement that followed only hastened the disintegration of
the federation. Every constituent in every corner became increasingly
concerned with its immediate interests. Every leader had his own
approach to a solution that favoured his area. The 'Abdulrazzaqs of
Khurasan did not see why they should jeopardise the interests of
Nishapur and fight the Buyids simply to protect the commercial
interests of Samarqand. Such an attitude in an area so close to Rayy
would certainly encourage the ambitions of the Buyids. Other
leaders did not take kindly to this negation of responsibility and
proposed to dislodge the delinquent 'Abdulrazzaqs.3 Others still
supported the view that it was easier to find another opening for the
trade via the southern route through Sistan and Kirman, than to fight
the intrepid Buyids.* All these differences represented themselves
in a fruitless internal struggle within the federation. The frequent
change of the commander-in-chief of the Samanid army only
indicates that a particular group had temporarily won the upper hand
in this endless struggle.5 The resulting economic plight forced the
government to exact more taxes from its subjects.6 As it worsened
as many as 20,000 men were persuaded to leave their homes and seek
1
Ibn al-Nadim, Al-Fibrisl, Cairo, A.H. 1)48, p. 266.
* Mas'udi, Muruj, vol. ix, p. 14.
1
Miskawayh, vol. ti, pp. 117-19; Athir, vol. vm, pp. 553, 578-81.
4
Athir, vol. vm, p. 482.
' Miskawayh, vol. 11, pp. IOI-J, 154-6, 177, 191-2; Abu Shuja', p. 27; Athir, vol.
vm, pp. 292, 544-*. JJ9. 37°. J9*. 4°4. *ol. ix, pp. 7, 18-20, 69, 102.
* Maqdisi, p. 540.
https://doi.org/10.1017/CBO9780511622380.009
I8O THE BUYID CONFEDERACY
their fortunes elsewhere in the more prosperous west. • The Buyids
were ready to offer generous peace terms and to conclude a truce for
ten years, but they would not give up Rayy.2 By 961/350 any hope
of saving the federation had vanished and what followed was a
scramble for the remnants. Any leader who could muster some
tangible military strength tried to lay his hands on as much territory
as possible.1 Out of this situation the militaristic regime of the
Ghaznavids was born in the remote region of Ghaznin, 7280 feet up
the slopes of the HindQ-Kush mountains.
The rise of the Ghaznavids has been explained as the fulfilment of
a sudden urge of a slave to carve himself an empire.4 This ludicrous
interpretation of history completely ignores the circumstance in
which these developments took place. Furthermore, Sebiiktigin, the
man responsible for this episode, had never been a slave. Reports to
the contrary are almost certainly fictitious, and indeed come only in
a source of more than three centuries later.5 The reliability of this
particular source has never been established and has actually been
doubted by one scholar who put it under close scrutiny.6 According
to this source Sebiiktigin was kidnapped as a boy alone in his home
in Central Asia, whence he was brought and sold in Samanid
territory. Yet, it is rather striking to learn from a more reliable
source that much later in his life he had with him a brother and a
sister who apparently had not gone through the process of slavery.7
Of course it can be argued that as he acquired power he brought
them to share his glory, but under the circumstances this was hardly
likely. In fact we are told by the same contemporary source that
Ghaznin was the nest of the clan and relatives of Sebiiktigin.8
Furthermore when he died his body was transported for burial to
Ghaznin in the same tradition that required the body of a Muhtajid
to be transported for burial to his home in Chaghaniyan.' It is now
clear that Sebiiktigin was a native leader of the region of Ghaznin
who came to fame during the period of the disintegration of the
Samanid federation. Like many other local leaders, he was looking
1
Miskawayh, vol. n, pp. 222-9.
1
Ibid., p. 100.
3
Athir, vol. vm, p. J98.
4
Barthold, p. 81; C. E. Bosworth, The Gbaxjiavids, Edinburgh, 1963, p. 40.
* Shabankart, Ma/ma' al-Ansdb fi al-Tawarikb, Istanbul manuscript, Yeni Cami 909,
ff. 164A-6A.
* Bosworth, pp. i i , 39-40; M. Nazim, "The Pand-Namah of Subiiktigin ", Journal of
tbt Royal Asia/it Society, London, 1933, pp. 369-622.
» Al-'Utbl, al-Tmikb a/-YamM, Cairo, A.H. 1286, vol. 1, pp. 236, 255.
* Ibid., p. 274. • Ibid., p. 256; Athir, vol. vin, p. 384.
https://doi.org/10.1017/CBO9780511622380.009
THE BUYID CONFEDERACY l8l
for a solution for the economic dilemma within the framework of the
interests of his own region. Being on the periphery of the Samanid
domains and on the trade-route into India, Ghaznln must have
suffered considerably from the general economic decline. On the
other hand its location on the very borders of the pagan sub-
continent offered an easy way out of the problem. Taking matters in
his own hands, Sebiiktigin started and continued over the twenty
years 977-97/366-87, a long campaign of plundering and looting the
plains of India. His success attracted more men to join his campaigns
and more men brought back more wealth. Eventually his son,
Mahmud, 998-1050/388-421, was to invade India and establish an
empire that extended from the Oxus to the Ganges. But it was an
empire that fizzled out almost as soon as it was established. It had a
highly organized army that congregated from all directions, includ-
ing Arabs who came out of the blue.' Any claims that these were
slave armies are based on the wrong interpretation of the terms
'abld (sing, 'abd) and mamdllk (sing, mamliik). Although these words
originally mean slaves, over the years they have acquired different
connotations. We have already seen 'abd used to denote a loyal
subject of the highest rank in the service of the ruler.2 Indeed, the
great Mahmud himself and the powerful Buyid leaders, whose many
honorific titles are too long to repeat here, were proud to adopt the
title of 'abd Amir al-Mu'minin, in spite of the fact that the impotent
Amir al-Mu'minin was more of their 'abd than anything else.3 In an
age when empty and lengthy titles were handed out to everybody in
sight, the words were debased and came to mean very little. The
Buyids and the Ghaznavids considered themselves 'abidpar excellence,
and it was only appropriate to apply this term to the loyal members
of the army in relation to their leaders.* Therefore the term 'abid
was used to signify the members of the standing army in Spain,
North Africa, Egypt, Syria, 'Iraq, Jazira, Iran and Central Asia. It
has to be remembered that all these 'abid were well paid and their
demands for regular pay always represented major problems for
every regime.
As well as an organized army, the Ghaznavids also had an efficient
administration fashioned after the Samanid pattern which has been
admirably described by Bosworth. However, this was not enough
1
Ibid., p. 387. * See above, p. 65.
» §abi, Tarikh, part 8, p. 54$; Idem, Kasa'il, p. 119; idtm, Rumm, p. 109; Athlr, vol.
VIII, p. 480.
• Abu Shuja', p. 50; Ra(U» p. 15).
https://doi.org/10.1017/CBO9780511622380.009
182 THE BUYID CONFEDERACY
to maintain an empire that did not have the allegiance of its subjects.
On the Indian side the Ghaznavids represented plunder. Whatever
has been said about their service to the cause of Islam by spreading
it into India ignores the fact that they introduced to the Indians the
worst facet of Islam. It was much in the same vein as the unattractive
image the Ottomans presented to the Europeans. It is instructive to
compare the relatively strong position that Islam has in Malaya and
Indonesia where it spread by peaceful means and its position in India
where the Ghaznavids left a long trail of blood.
On the Central Asian side, the Samanids were fast crumbling, and
it did not take the Ghaznavids under Mahmud much effort to advance
to the Oxus but no further. For here they were checked by another
power marching from the north from the steppes of Central Asia.
These were the semi-nomadic Turks who had been living outside
the Samanid borders and amongst whom Islam had quietly spread.
Taking advantage of the collapse of their neighbour's power they
easily occupied Soghdiana and established a dynasty there known to
us as the Qarakhanids. Their only contribution to Islamic history is
that in their internal quarrels they enlisted the help of yet another
branch of Muslim Turkish nomads, the Ghuzz, and these in turn
paved the way for the Saljuq invasion. Replacing the Samanids in
the lands to the south of the Oxus, the militaristic regime of the
Ghaznavids failed to win the loyalty of the population who had no
interest whatever in invading India. On their part, the Ghaznavids
behaved as an occupying power in an area which had had a long
tradition of autonomy.1 Furthermore, in 1025/416, Mahmud allowed
the Saljuq nomads to cross the Oxus and roam around the plains of
Khurasan. Letting such a force loose in a densely populated urban
region was bound to create disorders. When Mahmud tried to quell
these disturbances he succeeded only in driving them west into the
central plateau. In 1035/428 after Mahmud's death, the main bulk o
the Saljuq forces crossed the river and wrought havoc in Khurasan.
The helpless population of Merv and Nishapur had no option but to
make the best of a bad situation. Recognizing the suzerainty of the
invaders the Khurasanian oligarchy placed themselves in their service
and used them to drive out the hated Ghaznavids in 1040/431. In
Cl. Cahen's words, and there is no better way of putting it, the
Ghaznavids "fled to India. Khurasan was lost, and the Iranian
plateau was wide open. The evolution of the Iranian and Turkish
worlds had led the former to admit the Turks into its own bosom.
1
Athlr, vol. ix, p. 247; Hawqal, p. 450.
https://doi.org/10.1017/CBO9780511622380.009
Like that of the German empire, the conquest by the Turks, from
then on, was accomplished from inside".1
The struggle to save the Samiinid federation created some ripples
in Sistan and Kirman. The Buyids, who also had a foothold in
Kirman, were not unduly perturbed. Taking the necessary precau-
tions they decided to wait for their adversaries to exhaust themselves.2
But it was a different matter when Scbiiktigin attained power in
Ghaznin, especially as he immediately occupied Bust in 977/367.*
The rise of an aggressive military power in an area where complete
disintegration had been expected alarmed the Buyids who reacted
almost instantly. The Confederacy had had its share of internal
problems, whether as a result of fights over succession, or from the
demands of refractory armies. However, these problems did not
necessitate a change in the basic structure of die body politic. But as
soon as the spectre of the Ghaznavid menace appeared on the
horizon, there was an immediate effort to muster all the military and
financial resources of the Confederacy. The Buyids were not to know
that Sebiiktigin was going to concentrate his attacks in a completely
new direction, and his advance on Bust must have convinced them
of an imminent attack on their own territory. The most threatened
Buyid ruler, Fana-Khusraw, who also happened to have held power
in Fars for twenty-nine years, took it upon himself to impose his
authority over all Buyid domains. It was a temporary policy that
lasted only for five and a half years 978-83/367-72, and was never
attempted again in the remaining seventy years of the life of the
Confederacy.
Fana-Khusraw's objectives were to stamp out any nagging
problems which any one region had been unable to handle by itself,
and rectify the anomalies which were bound to exist under such a
system, especially with regard to the ruinous arrangements for
military pay. In short he wanted stability and he set out to establish
it. Baghdad was a hot-bed of sedition, and obviously the Buyid in
charge had not been able to control the riotous Hanbalites, Shafi'ites,
and Shi'ites, let alone his own army which had got itself involved in
their almost daily squabbles.4 Therefore it was occupied, and law
and order were swiftly imposed. The Hamdanids of Mawsil were
more of a threat than die Byzantines. Indeed, when Fana-Khusraw
1
Cl. Cahen, "The Turkish Invasion: The Selchiikids", A History of tie Cnuadtt, ed.
K. M. Sctton, London, 1958, pp. 135-76, the quotation is from p. 142.
1
Miskawayh, vol. 11, p. 249; Athir, vol. vm, pp. 416-17, 4)1, 481-3; MaqdisI, p. 472.
3
Athir, vol. vm, p. 504.
• Ibid., pp. 4JJ-6, 462-4, 466-9; Miskawayh, vol. 11, pp. 323-52.
https://doi.org/10.1017/CBO9780511622380.009
184 THE BUTID CONFEDERACY
attacked them they sought the help of the Byzantines and the
Fatimids, but none came to their rescue. They were driven out of
Mawsil and lost all their citadels in the region which had contained
a fabulous amount of wealth.1 The Kurds had just lost one of their
strongest leaders, so they were in no position to resist the Buyid
strong man.2 Although both Kurds and Hamdanids were to make
a come-back they presented no problem for a time. The Arabs close
to the desert borders of 'Iraq were brought into line and so were
some isolated Kurdish groups in Jibal.' The independent republic of
Batiha offered strong resistance when it was attacked, and Fana-
Khusraw pragmatically arranged peace terms with its ruler.4 In the
Gulf strict surveillance was organized to drive away the pirates,
secure control over 'Uman, and ensure the friendship of the Qara-
mita in Bahrayn. Kirman was fortified and Fana-Khusraw sent his
own son to reside there.5
The most interesting moves made in this hectic period were
in the fiscal field. Fana-Khusraw tried, not without success, to cut
down to a minimum the iqtd' system for paying the military. He had
to be careful not to interfere too much with the interests of some
powerful leaders of the Daylamites.6 For those men who were paid
directly from the treasury, measures were taken in order that they
should receive their stipends on time. On the other hand, it was
decided that no pay increases should be allowed except as a reward
for military victory.7 While tax-farming was allowed to continue in
operation it was brought under the close supervision of the regional
governments.8 Under the long rule of Fana-Khusraw in Fars, it
seems that he allowed his officials to experiment with some radical
tax innovations. There is no doubt whatever that the purpose of these
radical measures was to make the urban population pay its proper
share of taxes. These officials must have resorted to their remote
Iranian past to devise means for new taxes, and it is more than likely
that some of these measures were echoes of ancient Sasanian
practices. It was certainly un-Islamic to impose taxes on brothels, and
it was against all the rules of Islamic law to exact water rates.9 The
hold of the tax-collector over the towns was considerably tightened,
to the extent that anyone leaving town was required to have a
Miskawayh, vol. n , pp. 382-9;.
Athir, vol. VIII, pp. 519-21.
Ibid., p. 522; idtm, vol. ix, p. 5; Miskawayh, vol. 11, pp. 598-9, 414-15.
Miskawayh, vol. 11, p. 412. » Abu Shuja', p. 28.
7
Ibid., p. 47. Ibid., p. 45.
Ibid., p. 171. » Maqdisi, p. 441; Hawqal, p. 302; Istakhri, p. 157-
https://doi.org/10.1017/CBO9780511622380.009
THE B 0 T I D CONFEDERACY 185
passport.' On every load of goods entering town a tax of thirty

dirhams was collected.2 Inside the town the authorities set up market
places, shops and arcades and rented them for double the usual
rate.3 Brokers' houses were also built where important commercial
transactions were to take place, and we are told that the govern-
ment's revenue from one of these houses was 10,000 dirhams a day.*
The government also monopolized a wide range of industrial
activities, such as the making of ice and the production of raw silk,
and in every town it had its own clothes factory. All grain mills
and rose-water factories were taken over by the authorities. New
taxes were imposed on the sale of all kinds of cattle in the markets.1
For the import and export trade old rates were increased, and
hitherto untaxed items brought under taxation.6 The one con-
spicuously absent measure, which was fashionable with other Shi'ites,
was the imposition of an income tax. Perhaps all those urban tax
increases were supposed to compensate for this default; or perhaps
it was because the Buyids did not have a recognized Shi'ite Imam in
whose name the fifth or the tenth was collected as income tax. Or
was it the other way around? That is to say, was it that they did not
want to recognize such an Imam because they did not want to impose
the tax ? It is probably for this reason that they took the easy way out
and recognized the 'Abbasids and unmistakably declared their
intentions of safeguarding the vested interests of the wealthy.
As for the rural communities, measures were taken to enforce die
strict collection of taxes. The nomads were required to pay taxes on
their herds, and for this purpose an elaborate system of regrouping
them under appointed chiefs was instituted and registered. In return
these unruly people, whose numbers were over 2,000,000, were
drawn into the system and were made to serve the purposes of the
government. They were given the responsibility of guiding and
protecting the trade caravans, and naturally these services were not
free of charge.7 For the land-tax, an infinitely more complicated
three-tier system was introduced. The complication arose because
the system was designed to exact maximum amounts while taking
existing conditions into consideration. Accordingly the assessments
were made by way of a fixed proportion of the produce, by a fixed
sum, or by measurement of the land. Nevertheless, the amount of
1
Maqdisi, p. 429. * Ibid., p. 400.
» Ibid., pp. 413, 429; Isfakhri, p. i } 8 . « Maqdisi, p. 4)4.
9
Ibid., pp. 413, 431; Istakhri, p. 158; Aba Sbuji', p. 71.
* Athir, vol. IX, p. 16.
7
Haw<j»l, pp. 26), 269; Ispkhii, pp. 99, 113, I I J .
https://doi.org/10.1017/CBO9780511622380.009
l86 THE BUYID CONFEDERACY
taxes due differed from one district to another even when assessed
on the same basis.1 On the other hand, the government did not
shrink from the responsibility of maintaining and indeed improving
all irrigation systems. Astonishing engineering skills were utilized
and great sums of money were spent to build dams and dig new
canals to help agriculture.2 With regard to minerals, the tax of one
fifth continued in operation except for pissasphalt which became a
state monopoly.3
When Fana-Khusraw extended his authority to include Baghdad
and 'Iraq, most of these fiscal measures were put into effect as soon
as possible. While expenditure was not spared to improve irrigation,
a general 10% land-tax increase was decreed. Monopolies were set
up and a grand plan to rebuild Baghdad was initiated. Any house
owner who wanted to renovate his property was offered a free loan
from the treasury.4 One can imagine that the hospitals built at that
time were of a better standard than the lunatic asylums of previous
days.
It is almost sad to have to stop writing about this unexpectedly
imaginative and promising period, but like all good things it had to
come to an end. It is even sadder that the end came so abruptly and
that these reforms, belated as they were, were not given an oppor-
tunity to take root. The death of Fana-Khusraw in 983/372 was
followed by a reversion to conditions as they had been before his
energetic last five and a half years. It was not because of the absence
of a strong man that this happened; it was because of the absence of
the threat that had given the impulse to those urgent measures. The
Ghaznavid danger proved to be a false alarm and it was to be half a
century before it materialized in the form of an attack on Rayy which
the Ghaznavids occupied in 1029/420, only to give it up soon after
to the Saljuqs. During this half century the Buyid Confederacy
continued willy-nilly its struggle for survival. The increasing
demands of the military and the accumulative bad effects of the iqtd'
system had had less harmful consequences on its prosperity than the
disastrous drying up of trade. It was in this period that the Pecheneg
Turks started to attack the Volga basin from their territory to the
north of the Caspian. These attacks coincided with the emergence of
the Russians whose good relations with the Byzantines helped to
1
Hawqal, pp. 302-}; Ijtakhri, pp. 157-8; Maqdisi, pp. 448, 451-2.
1
Abu Shuja', p. 69; Maqdisi, pp. 404, 419, 444.
3
Hawqal, p. 302; l;talchri, pp. 154-5.
4
Miskawayh, vol. 11, pp. 405-6; Abu Shuja*, pp. 45, 71; Athir, vol. ix, p. 16.
https://doi.org/10.1017/CBO9780511622380.009
divert the whole northern trade to Kiev and Constantinople. The
Italian merchant states at the same time began to attract a good deal
of this business and to export goods across the Mediterranean.
Central Asia and the Caucasus region lost completely in this race, and
subsequently Iran, east and west, the main duct for this trade, began
to suffer severe economic decline. In fact, the hitherto all-important
Rayy suddenly became of no importance and its loss to the Buyids
in the last decades of their existence did not make much difference to
their fortunes. As if the loss of the northern trade was not enough,
the Buyids also lost all their southern trade. Thanks to the Fatimids
who managed to attract all the Indian Ocean trade, the Bflyid
Confederacy was practically bankrupt when it fell to the Saljuqs in
105 5/447. Significantly the new conquerors did not have much of a
trade policy.'
1
Athir, vol. x, pp. J I , IOJ, m .
https://doi.org/10.1017/CBO9780511622380.009
THE FATIMIDS
It is surprising to realise that more research has been done on the
Fitimids than on any other aspect of Islamic history. Conveniently
M. Canard has summed up the results of all this research, including
his own, in an article of thirteen large and crowded pages in the new
edition of the Encyclopaedia of Islam to which he and others have also
contributed numerous other relevant articles. Therefore, the basic
facts of Fatimid history have been well established. Bernard Lewis,
who started working on the subject almost forty years ago, has
recently given us his interpretation of Fatimid history.« All these
researches have in common one principal weakness; they do not
take into consideration earlier developments nor all of the important
circumstances in the areas the Fatimids controlled. They are also
unbalanced in that more has been said about the birth certificate of
the first Fatimid ruler than about economic conditions in his domains.
His genealogy, interesting as it may be, is unimportant because
enough people believed in his cause to make possible his rise to
power. They also believed his claim to be a descendant of the
Prophet through his daughter Fatima, a claim which was an easy
target for the attacks of his opponents. It has to be realized that, at
the time, there were hundreds if not thousands of such descendants
and it was well nigh impossible to prove or refute their claims. The
important thing is that this particular leader put forward an Isma'ili
Shi'ite ideology of radical revolutionary dimensions.
We cannot comprehend the history of the Fatimids without
understanding the full implications of their wide-ranging economic
policies about which little has been said. Although Lewis had long
recognized that the trade activities of the Fatimids reached as far as
India, they were judged to be blind to the African trade in their own
backyard.1 Their taxation system, a cornerstone in their success and
failure, has not been discussed and has gone as unnoticed, as if they
1
Bemud Lewii, "An Interpretation of Fatimid History", Colloqm Inltrnatimal star
UHiittirt it Cain 27 Mars-; Avril if if, Cairo, n.d. pp. 287-95.
1
Htm, "The Fipmidj and the Route to India", Rent it la Faculti dt Scimeis Economiqms
itl'VmttrsiUi'lstatbul, 11 annee. No. 1-4, Istanbul 195}, pp. 1-5.
[188]
Downloaded from https://www.cambridge.org/core. University of Groningen, on 13 Apr 2018 at 13:23:15, subject to the
https://doi.org/10.1017/CBO9780511622380.010
THE FAJIMIDS 189
had none. From scholars like Canard and Lewis, this woeful neglect
is surprising. Furthermore, it is indefensible for Lewis to compare
the rise of the 'Abbasids in Khurasan with that of the Fitimids in
Tunisia.1 Any accidental similarities are superficial and should be
treated with the greatest caution because there were major differences
between the two movements in the nature of their support, their
institutions and organizations, and their purposes and ideol-
ogies.
The Fitimids first established themselves in Tunisia and sixty
years later they took a most extraordinary step, which has been
looked on as if it were an everyday occurrence. They moved
literally lock, stock and barrel to Egypt, never to glance back again.
While in Egypt their relationship with their former subjects was of
little importance to the Fitimids, and indeed within a few decades
these subjects became independent and even hostile to them. This
transplantation of the Fatimids has been explained as part of their
grand plan to overthrow and supersede the 'Abbasids. If that was
the case then the Fatimids went about it in the strangest possible
way, spending more effort on controlling northern Syria than in
achieving their primary objective. They might have dreamt about a
universal empire that would envelop all Islamic domains, but their
efforts in this respect never went beyond sending missionaries to
BaghdSd as well as to India and Central Asia. Any such ambitions
they may have entertained remained in the realm of ultimate objec-
tives rather than practical policies, and had no effect whatever on the
situation at the time. Indeed, it is inaccurate to speak about a vast
Fifimid empire of which the "centre was Egypt; its provinces at its
peak included North Africa, Sicily, Palestine, Syria, the Red Sea
coast of Africa, the Yemen, and, of special importance, the Hijiz,
possession of which conferred great prestige on a Muslim ruler and
enabled him to use the potent weapon of the pilgrimage to his
advantage."2 At no point did such an empire exist. For soon after
the Fatimids left Tunisia, their western borders retracted to a point
hardly west of Alexandria and their interest in Sicily had evaporated
before they could even establish a foothold in Syria. Palestine was
more often than not out of their control and they never posstssed
Hijiz. To say that Yaman was part of the Fitimid empire requires a
long stretch of the imagination; thus describing a relationship by
correspondence as an actual marriage. It is more realistic to say that
after leaving Tunisia, the Fitimids dominated Egypt but despite
> Utm, Inlerpntatim, p. 188. > Ibid., p. 189.
https://doi.org/10.1017/CBO9780511622380.010
190 THE FATIMIDS
desperate attempts failed to establish more than a foothold on the

Palestinian coast. Considering the chaos that prevailed in Syria and
Palestine at the time this was a conspicuous failure.
In 893/280 an Isma'ili missionary, Abu 'Abdillah, arrived in
Little Kabylia in present-day Algeria, in the company of pilgrims
from the Berber tribe of Ivutama whom he had met in Makka. It is
significant that such an astute man should have chosen to go straight
to North Africa to propagate his cause and by-pass Egypt altogether.
It indicates that the Isma'ilis, who saw possibilities for success in the
central province of 'Iraq, did not believe that such possibilities
existed in Egypt. They knew that outside Fustat and Alexandria the
Egyptian population was overwhelmingly Coptic, and they reckoned
that they would have better chances of converting Muslims to their
cause than converting of Copts to Islam. North Africa was perfectly
suited to the purposes of the Isma'ilis, because most of the population
there had long been Islamized. Furthermore, heterodoxy was not
new to the area; if anything it was the norm. The Shi'ite Idrisids
and the Kharijite Rustamids had successfully maintained their
regimes for over a century and were still very stable. The only
instability emanated from Tunisia where the Aghlabids had had one
trouble after another. From the very beginning their regime had
been based on military control. However, as their forces were con-
stituted of Khurasaniyya, Berbers and Arabs, there had been a great
deal of disharmony and quarrelling amongst them. The problems of
the Aghlabids had been aggravated by their complete dependence on
an agricultural economy while their neighbours, east and west, had
been busy reviving and capturing the ancient trade across the
Sahara. This situation had forced upon the Aghlabids an unhealthy
isolation for which they had had to find a solution. Hence the long
and expensive military campaigns to occupy Sicily 827-902/212-91.
To meet these expenses it had been decreed that the land-taxes
should be paid in cash and not in kind, and significantly this was
revoked only when the Fatimid threat became imminent.1 Mean-
while, the Aghlabids had begun, at last, to try to attract some of the
trade by encouraging the use of an east-west route through their
territories.2 Nevertheless, this attempt had not been particularly
successful, as sailing along the coast was a less expensive and easier
means of communication. Towards the end of the rule of the
s, the Byzantines, who had lost Sicily to them, made some
1
'Idhari, vol. i, pp. 95, 1 JI ; Athir, vol. vi, p. 231.
2
Maqrizi, Kbitat, vol. I, p. 174.
https://doi.org/10.1017/CBO9780511622380.010
THE FAflMIDS I9I
overtures to establish trade relations with them through the island,

but by then it was too late.'
The task of the missionary Abu 'Abdillah was not a difficult one.
It was easy to incite and organize the Kutama Berbers of Algeria
against the Aghlabids with whom they had no particular affinity.
Sensing that the enterprise was about to bear fruit the Isma'ili
Imam, 'Ubaydullah, in whose name it was launched, began cautiously
to journey from his base in Syria to North Africa in 902/289. His
surreptitious route led him to Sijilmasa where he was ignominiously
imprisoned. After bringing down the Aghlabids in 909/296 Abu
'Abdillah dutifully marched on Sijilmasa, freed his leader and
brought him to the conquered capital of the Aghlabids, Raqqada, in
910/297. There, the Imam was installed as the legitimate Shi'itc
Amir al-Mu'minin and proclaimed himself the Mahdi, the long
awaited leader who would initiate an era of divinely guided rule.
Almost immediately, differences of opinion on the course of action
to be pursued began to appear among the victors. Abu 'Abdillah
disagreed with his own brother, a senior advisor to the new regime.
The latter wanted to take extreme measures against the intellectuals
who had supported the fallen adversaries.2 Then Abu 'Abdillah
himself fell out with the ruler over the question of taxation. The
former advocated a return to the most elementary form of taxation
in Islam, that is tithes in kind on some agricultural produce and
cattle.3 Realizing that this would be impractical and might in fact
raise problems with many Berbers who were not accustomed to
paying taxes on their live-stock, the Mahdi introduced a different
system of taxation. A fixed sum in cash was levied on acreage, and
no measures were taken to tax cattle. Although this was in effect a
return to the system which the Aghlabids had abandoned before
their downfall, some adjustments were made to alleviate its worst
effects.4 Perhaps these differences could have been reconciled if
another major discordant note had not been struck. It is possible
to discern that Abu 'Abdillah, who was obviously less ambitious
than his master, was convinced that the best course of action after
victory was to turn quietly to the west and consolidate Fatimid
power as much as possible at the expense of the Rustamids and the
Idrisids.5 Furthermore, having spent almost twenty years among the
Berbers, he had become familiar with their problems and interests to
' 'Idhirl, vol. 1, p. 144.

1
Ibid., pp. 150-1. > Ibid., pp. 141-2.
Ibid., p. 17}; Maqrizi, Kbifal, vol. I, p. }J2. * MaqrizI, Kiifaf, vol. I, p. 351.
7 SIH
https://doi.org/10.1017/CBO9780511622380.010
192 THE FATIMIDS
the point of identification with them. On the other hand the Mabdi,
a relative newcomer to the area, was free from such attachment.
Moreover, as the leader of a sectarian movement which was putting
forward claims to universality, he could not support a policy that
would effectively make his regime no more than a successor to the
pronouncedly Berber sectarianism of the Rustamids and the Idrisids.
To him turning towards the east was the logical solution to the
dilemma. Therefore, Abu 'Abdillah and his brother had to be
sacrificed for the loftier objectives of the movement, and when some
Berbers rebelled they were swiftly subdued. It is significant that the
Mabdi, and his successors, practically ignored the existence of the
Umayyads in Spain. The latter had extensive trade interests over all
of North Africa as is evident from many emporiums on the North
African coast, and the presence of commercial colonies as far east as
Alexandria and the Island of Crete.' To interfere with these vested
interests was not only inviting the wrath of a powerful enemy, but
also stirring up a hornets' nest. Therefore the Fatimids, the claimants
to all Shi'ite heritage, chose of all things to leave alone an Umayyad
Amir al-Mu'minin in nearby Spain and tolerate the challenge to their
universal claims.
At this stage it is difficult to see how far east the Fatimids con-
templated extending their domains. Their well-known long term
and detailed planning did not come into play at this time. Knowing
that their military forces were still narrowly based and limited only
to the Kutama, themselves a minority among the Berbers, the
Fatimids sensibly confined their objectives to the possible. As heirs
to the Aghlabids they captured, though with some difficulty, Sicily
and Tripolitania. Once these were secured an aggressive policy
developed simultaneously on two fronts, one at sea and the other on
land. For the first an impressive ship-yard with at least two dry-docks
was constructed in the newly built capital of Mahdiyya, and in no
time it had 900 warships. This powerful fleet was soon engaged in
wide-ranging and far-reaching attacks against Italian ports as far
afield as Genoa, and against the islands of the western Mediterra-
nean.2 Of course such attacks could be expected to yield a certain
amount of booty, but this could not have been expected to com-
pensate for the immense expenses involved in maintaining such a
large fleet. It is noteworthy that the Fatimid fleet did not clash much
1
'Idhari, vol. I, pp. n 6 , 136, 169, 175, vol. II, pp. 242-3; Athir, vol. vi.pp. 281-2,
vol. VIII, p. 459; Hawqal, pp. 74, 104.
> 'Idhari, vol. 1, pp. 174, 187-94, 209; Athir, vol. vm, pp. 116-17, 212, 232.
https://doi.org/10.1017/CBO9780511622380.010
THE FATIMIDS 193
with the equally powerful fleet of Muslim Spain. It is also remarkable
that while the Italians did not take any counter measures against
these attacks, the Arabs of Sicily strongly objected, to the extent of
attacking the naval base in Tunisia.1 This strange quadrangular
entanglement suggests that Sicily with its mixed population and its
position as a trade intermediary between the Italian and the Spanish
ports, was the loser in this situation. In other words, in announcing
their presence the Fatimids were trying to reach out directly to the
Italian market at the expense of Sicily and without causing unneces-
sary conflict with Spain.
On the land front the Fatimids' objective was just as prudently
planned. There was no question of conquering Egypt or confronting
the 'Abbasids, the Fatimids knew very well that they did not have
the military power to embark on such adventures which were
obviously beyond their resources. What they aimed for was the
domination of the coast as well as the desert between their holdings
in Tripolitania and the Nile Valley. The expeditions of 913-15/301-2,
919-21/307-9, 935/323 were not meant to occupy Egypt, they were
intended to drive Egyptian influence out of Cyrenaica, and as far as
possible to limit it to the Valley itself. In these expeditions Fatimid
forces came as close as possible to the city of Alexandria and then
proceeded through the desert to a point opposite Fustat. From there
they continued to advance through the desert to the FayyQm Oasis
and then to the rest of the Egyptian Oases before returning home. 2
The clear conclusion is that the Fatimids were striving to gain control
of the African trade flowing through the north-south routes across
the Libyan-Egyptian desert to the flourishing ports on the coast.3
Achieving this end in addition to having access to the European
market would have put the Fatimids in a commanding position as a
major trading power in the Mediterranean basin. One can see the
necessity of a powerful navy for such an enterprise. There is no
doubt that trade was the most important motive in almost every
action taken by the Fatimids throughout their history.
Contrary to their expectations the frantic activities of the Fatimids
produced just the opposite results. The long-established equilibrium
in the rest of North Africa was radically upset. The fall of the
Rustamids in 909/296 put an end to the prosperity of their com-
mercial centre in Tahart and all its trans-Saharan trade was driven
1
'Idhiri, vol. I, pp. 168-74.
» Athir, vol. vm, pp. jo, z$o; MaqrizI, Kii/af, vol. I, p. 174,vol. 11, p. 27; 'Idhiri, Tol. 1,
J
pp. 188-9. tfawqal, PP- 6 7. 69, 103-4.
8-3
https://doi.org/10.1017/CBO9780511622380.010
194 THE FAJIMIDS
much further west. The result was a series of Berber revolts that
were encouraged by the Umayyads of Spain. The more the Fatimids
tried to contain these revolts the further they themselves got involved
and to no avail. At one point a revolt became so serious that it
threatened the very existence of the Fatimids in Tunisia.' On their
eastern flank the first expedition of 913-15/301-2 gave the Fatimids
a foothold in Cyrenaica, but it also created such a panic there that
much of the population moved to Alexandria and its vicinity.2
Those who remained in Cyrenaica took part in the revolts that
occurred in Tripolitania against the Fatimids. Under these circum-
stances the trade which had found its way back to the Oases gradually
reverted to Nubia. It is interesting to note that the king of Nubia,
who had been worried about the drying-up of trade through his
territory, found it necessary to attack the Oases in 950/339.3 This
action could not have helped the cause of the Fatimids and indeed
induced them to change their plans. On the coast neither Tripolitania
nor Cyrenaica was content with the outcome of the Fatimid adven-
tures. However, the unrest in Cyrenaica and its close contacts with
Alexandria drew the attention of the Fatimids to the possibilities in
this direction. On the other hand Fatimid attempts in the western
Mediterranean only helped to bring the Byzantines and the Umayyads
of Spain closer together. In 945/334 and 949/338 the emissaries of
Constantinople arrived at Cordova seeking its friendship.4 The
rising star of the Umayyads persuaded the emperor Otto the Great
to send his own embassy to Cordova in 953/342.5 l n spite of their
many allies in North Africa the Umayyads thought it prudent to
occupy Ceuta and establish a stronghold on the African coast.6
Perhaps it was this move, more than anything else, that convinced
the Fatimids that their plans had rebounded upon them. The Mahdi
and his two successors had persevered with their efforts to push for
the success of these plans for over forty years. The new ruler Mu'izz
who came to power in 953/341, had different ideas which he gradually
put into effect. He began to widen the military base of the regime. It
is no coincidence that it is in his reign we find a noticeable element of
Sicilians and probably also Italians in his army. When this army
arrived in Egypt it had two separate contingents of "Rum". 7 He
also brought Berbers from tribes other than the Kutama into his
1
Athir, vol. vm, pp. 515-22; 'Idhari, vol. 1, p. 191.
2
Maqrlzl, Khilat, vol. 1, p. 174; Kindl, Wuldt, p. 274.
3
Antaki, p. 112; Maqrizi, Khitat, vol. 1, p. 236.
4 5
'Idhari, vol. n, pp. 215, 215. Ibid., p. 218.
6 7
'Idharl, vol. 1, pp. 203-4. Maqrizi, Kbitat, vol. 11, p. 8.
https://doi.org/10.1017/CBO9780511622380.010
THE FATIMIDS I9J
forces, such as the Zuwayla of Tripolitania.' But his major achieve-
ment in this respect was to win over the powerful tribe of Sanhaja by
giving them a share in the power structure of the regime. In 958/347
an apparently unnecessary expedition was organized under the
command of the redoubtable Sicilian general Jawhar and with the
full participation of the Sanhaja, to pacify the regions of central
North Africa.2 Militarily this expedition did not accomplish much;
it even failed to recapture Ceuta from the Umayyads. Yet politically
it was an unqualified success for it involved the Sanhaja in the
defence of these regions. Mu'izz now became free to pursue his
plans elsewhere, plans that took ten years of preparations.
The fall of the Tulunids in 905/292 was followed by thirty years
of direct control of Egypt from Baghdad. The remnants of the
Tulunid forces were transferred to Syria and instead a strong con-
tingent from Baghdad under a succession of military governors was
stationed in Egypt to hold the province. The real power, however,
remained in the hands of the fiscal administrators both in Egypt and
in Syria. As has already been explained the revenues of these two
provinces were allocated for the needs of general Mu'nis's armies.
In case of emergency these armies were quick to come to the rescue,
as happened when the Fatimids attacked Egypt.1 The downfall of
Mu'nis, the breakdown of the central government and the rise of the
the Hamdanids made necessary a new arrangement for southern
Syria and Egypt. It was felt that the continued threat of Fatimid
attacks demanded a strong military presence in Egypt. Ironically
the only available troops were the remnants of the Tulunid forces
which were consolidated in Syria under the leadership of Muhammad
b. Tughj, better known as Ikhshid, himself a former associate of the
Tulunids. As it happened he was also a descendant of a long line of
loyal 'Abbasid generals that had been recruited from Central Asia
over a hundred years earlier in the heyday of Samarra. These qualifi-
cations and his knowledge of Egypt and its problems offered the
desired solution. In 935/32} Ikhshid was appointed governor of
Egypt in addition to his governorship of Damascus. As he was
granted these territories as a military-administrative iqtd*, this
appointment was in effect a return to the Tulunid system by which
Ikhshid had full control over the military and fiscal affairs of his
domains. Arriving in Egypt with his army and navy, he was soon
able to establish his rule and integrate whatever forces had been
1
Ibid., p. 4- * Maqrfci, Khifaf, vol. I, p. JJ2.
> See above, p. 14;; Kindl, Vmlat, pp. 278-86.
https://doi.org/10.1017/CBO9780511622380.010
I96 THE FAJIMIDS
1
there into his army. During his eleven years in power 935-46/
323-34, he managed to hold his own against the encroachments of
the Hamdanids and maintain the integrity of southern Syria, includ-
ing Damascus, in a period when the central government was in
utter disarray. He also kept the Qaramita at bay by paying them
300,000 dinars a year.2
Although Ikhshid was supposed to have started a dynasty and
indeed was succeeded by three of his descendants, yet the real power
throughout the years of his dynasty was firmly in the hands of
Kafur. He is supposed to have been a black slave, a eunuch, with a
pierced lower lip and flat-footed to boot. He is reported to have
been brought to Egypt and sold as a slave when he was ten years old.
It is also reported that at that point he expressed the hope that one
day he would rule Egypt. According to these reports Ikhshid had
bought him at the bargain price of eighteen dinars which could have
happened only after the arrival of the former in Egypt in 935/323.
Yet in ten years' time we find Kafur as a general of Ikhshid's army
in Syria and subsequently as the virtual ruler until his death in
966/355.* It is important here to point out that Kafur was unfor-
tunate enough to encounter one of the greatest Arab poets of all
times. Apart from his poetical gift, this man was extremely vitriolic
as well as a great liar, as is evident from the name by which he was
known, Mutanabbi, he who claims false prophethood. When he
eulogized Kafur he made him the greatest of the great, and when
Kafur did not pay him his price the poet satirized him as the scum
of the earth. Although one finds echoes of the praise in the sources,
the colourful terms of the satires reverberated far more and were
soon taken as facts. Naturally such facts would come to be embel-
lished with the appropriate stories, and it is left to the researcher to
apply his common sense.
Kafur recruited suddn in his army, a fact that would make it more
than probable that he was of the same origin.4 These new recruits
were needed primarily to hold Syria and strengthen Egyptian
defences in the west. However, Kafur had to cope with a new front
which hitherto had been very peaceful. In 955/344 the Nubians
attacked Aswan and accordingly an army had to be sent to protect
the southern borders.5 The sudden anger of the Nubians which had
1
Maqrizi, Kbilal, vol. i, pp. 328-9; Kindl, Wulit, p. 287.
1
Athir, vol. VIII, p. 452.
1
Ibid., p. 343; Maqrizi, Kbilat, voL 1, p. 329, vol. 11, p. 26.
4
Maqrizi, Khilal, vol. I, p. 94.
» Ibid., p. 198; Antaki, p. 112.
https://doi.org/10.1017/CBO9780511622380.010
THE FATIMIDS I97
provoked this attack must be related to their attack on the Oases

only six years earlier. After making the considerable effort of cutting
the important links of the trade-routes at the Oases, they attacked
Aswan in order to put pressure on the authorities in Egypt to take
Nubian interests into consideration. The dilemma of these authori-
ties was that if they were to co-operate with the Nubians to restore
the trade to Nubia, they would add to the pressures in Cyrenaica
which were having serious effects on the stability of Alexandria.
Besides, there was always the Hamdanid weight on Syria to worry
about. These increasing dangers demanded more forces, and if the
suddn were to shoulder this responsibility they would have had to
have a major say in the affairs of the country. Hence the accession to
power of Kafur who must have been the most prominent leader of
the suddn in the army.' His task was not only to defend Egypt and
Syria against outside attacks, but also to keep the balance between
the various groups within these territories, the Copts, the Arabs of
Alexandria and the Berbers of Cyrenaica who moved in with them,
the suddn, the rest of the army, the population of Damascus and the
Syrian ports, and the increasingly troublesome Arabs of Palestine
who were just beginning to constitute a force of their own. There is
all the evidence that both the Hamdanids and the Fatimids were
using all possible means to stir up discord within Kaffir's domains.
Yet with remarkable political acumen he contrived to preserve the
integrity of his sway until his death in 966/355.* Within three years
of this the Fatimids had occupied Egypt.
Arriving at Alexandria in 969/358 the Fatimid army met with no
resistance and probably had not expected any. Its commander general
Jawhar declared an amnesty and assured the people of the safety of
their lives and property. Of greater importance and more significance
is the fact that he also assured them that they would be allowed
freedom of religion.3 This was not the rash promise of a victorious
general, it was in accordance with a general line of policy that had
already been declared by Mu'izz. The latter must have realized that
after many decades of proselytizing, Isma'ilism had not really taken
root in North Africa, though that was a favourable milieu. Egypt
with a great majority of Christians divided into at least three sects,
and a minority of Muslims following three schools of orthodoxy,
would not be fertile ground for conversion. In fact Mu'izz had
decided to curtail the preaching of the esoteric teachings of Isma'ilism
1
Maqtlzl, KJrifat, vol. i, p. 329. ' Ibid., p. 451; Urn, vol. n, p. 430.
1
IJtm, Illi'az, vol. 1, pp. I J I - 2 .
https://doi.org/10.1017/CBO9780511622380.010
198 THE FATIMIDS
and limit that only to the initiated.1 The writings of his chief
theoretician Qadi Nu'man, which are extant, do not differ much
from orthodox writings except on the question of the position of the
imam, and some other very subtle points which will be presently
explained. This rather neutral religious policy was maintained in
effect, with very few exceptions, throughout the rule of the Fatimids.
This pragmatic attitude from a missionary and proselytizing sect
allowed them to employ in their service not only Muslims of all
shades of opinion, but also many Christians and Jews. On the other
hand, as they did not penetrate to the roots of society, no trace of
Isma'ilism whatsoever remained in Egypt after two centuries of
Fatimid rule.
Although building new capitals was a universal practice of this
age, the fact that almost as soon as Jawhar arrived at Fustat he
proceeded to build Cairo in 970/359, confirms the determination of
the Fatimids to have the least possible interference with the existing
state of affairs. Everything had been planned in advance to the
extent of laying the foundations of the Azhar, the intellectual centre
of Isma'ilism in Egypt. Although space was provided within the
walls of the new city for reasonable future expansion it was designed
to be strictly the administrative and military capital of the regime.
Fustat, which came to be known as Old Cairo, flourished and
expanded as the commercial capital of Egypt.2 When Mu'izz arrived
from Tunisia in 973/362 everything was ready for him to put into
effect his well conceived and detailed plans.
These plans were the culmination of Shi'ite experience during the
previous three centuries. The continuity and unity of this experience
were emphasized by the shining prominence given to the sixth
Imam Ja'far al-Sadiq in all Isma'ili traditions. The Fatimids were
determined to benefit from the history of the past Shi'ite movements
and to avoid their mistakes. They put forward a clear ideology and
an elaborate theology which continued to develop by virtue of the
presence of a divinely guided leader. Being also the ruler, this leader
would ensure justice for all. Although they knew only too well that
they were a minority sect, they behaved as if their universal accept-
ance was an accomplished fact. They operated from one region and
certainly pushed hard for its interests, yet their vast international
relations enhanced their universal image. They were keenly aware of
rural disaffection and the privileged situation of the urban com-
munities, and they tried to plan for a more just society. There is no
1
Athir, vol. vm, p. 389. > Maqdisi, p. 199.
https://doi.org/10.1017/CBO9780511622380.010
THE FAJIMIDS 199
doubt that Egypt was the field chosen for this unique experiment in
Islam. Indeed the rule of the Fatimids in Tunisia was on a completely
different basis, and when they moved to Egypt they simply did not
care what happened thereafter in the west. Tunisia was left to the
care of the Sanhaja Berbers, their recently cultivated allies. The only
stipulation was that Mu'izz reserved for himself the right to appoint
judges and tax-administrators, but there is no trace that this condi-
tion was observed by the Zirids, the leaders of the Sanhaja.l Sicily
was entrusted to the Kalbids who had kept it under control, and
Tripolitania was given to members of the Kutama Berbers, the
original supporters of the Fatimids, while Cyrenaica remained in
the hands of the latter.2 Theoretically all these territories remained
under the suzerainty of the Fatimids, but in practice it was certainly
a tenuous relationship. In 975/365 the Zirids sent the Fatimids
badly needed financial help; in 983/373 it was only a present that
arrived at Cairo; in 987/377 the Zirids annexed Tripolitania, and in
1012/403 the Fatimids conceded Cyrenaica to them.3 In 1015/406
the Zirids expunged any traces of Isma'ilism that there might have
been in their domains, and in 1041/433 they dropped all pretence
and became actively hostile to the Fatimids.4 In the meanwhile the
Zirids had been busy promoting their own interests and competing
directly with the Fatimids for a greater share in the African trade.
It is interesting to note that elephants and the giraffes which used to
adorn Fatimid ceremonial processions in Cairo began to appear in
the Zirid court in 1032/423 as presents from the king of the suddn>
In Egypt the Fatimids started their rule by setting up the most
centralized and hierarchical administration ever known in Islam.
The architect of this elaborate plan was Ya'qub b. Killis, a converted
Jew who had originated in Baghdad, moved to Egypt, served
Kafur in Palestine, fled to Tunisia and had come back with Mu'izz.
Naturally the focal point of the whole system was the Imam who was
unequivocally acknowledged as God's representative on earth.
From him all authority emanated and to him all duties were due. The
management of the affairs of the country was divided between three
discernible branches, administrative, judiciary and missionary; the
last two terms do not denote the exact function of the respective
branches. The judiciary was the charge of a Chief Justice whose job
1 2
MaqrizI, Khitat, vol. i, p. 3}}. Athir, vol. vm, p. 456.
1
'Idhari, vol. 1, p. 250; MaqrizI, Kbilaf, vol. 11, p. 158; Athir, vol. ix, pp. 25, 37-8.
• 'Idhari, vol. 1, p. 275; Athir, vol. ix, pp. 180, 366-7.
1
'Idhari, vol. 1, pp. 246, 247, 249, 275; MaqrizI, Kiifaf, vol. 1, pp. 451, 461.
https://doi.org/10.1017/CBO9780511622380.010
2OO THE FAJIMIDS
was not only the administration of justice but also included such
things as supervising the mint.1 On a par with him was the Chief
Missionary who had his representatives in every district of the
Fatimid domains, as well as in all parts of the Islamic world. His
functions were at two levels. Internally, he was in charge of educa-
tion, and surprisingly his agents were given the responsibility of
collecting certain taxes which were more like membership dues of the
Isma'Ili sect.2 Externally, he headed what was very similar to the
Congregation of the Propaganda of the Catholic Church in charge of
foreign missions. As such he was very much involved in the external
relations of the Fatimids which most certainly included trade. It was
possible for one person to hold at one and the same time the offices
of Chief Justice and Chief Missionary. The administrative branch
comprised various departments, each under its own head, and all
under the direction of one person. This particular official was called
wdsita, i.e. intermediary between the Imam and the rest of the
administration. On two occasions when this high official was given
the responsibility of overseeing the military forces, he was given the
rank oiwaîr.3 Normally the military were under the direct command
of the ruler and sometimes there was a commander-in-chief, but the
constitution of the army did not often allow this to happen. It was
also possible for the wasgr or the wdsita to be Chief Justice and Chief
Missionary at the same time but this was only in times of crisis.4
Having established their administration they turned to the
economy with much the same zeal for planning in an effort to
exploit the resources of Egypt to the best advantage. On his arrival
in Egypt general Jawhar, needing money for his immediate expenses,
had simply doubled the land tax from t,\ dinars per acre to seven.5
But this had been a temporary measure and it was now time to think
about a more permanent system. Here again the position of the
Imam dictated the solution. As God's representative on earth he
owned all the lands in his domains. He could give land-grants to
whomsoever he wished and his fortunate favourites would own such
lands. As for the rest of the people who had been in possession of
land, they lost their titles to their properties but were allowed to
continue to cultivate them. No one was forced off his land and it
was passed from father to son, but as was well put by some astute
observer at the time, such a person was no more than qinn, a heredi-
1
Maqrizi, Khitat, vol. i, p. 404. * Ibid., p. 39:.
5
Mawardi, Ahkam, pp. 41-3; Maqrfei, Kbitat, vol. H, p. 5.
s
• Maqrizi, Khitat, vol. i, p. 464. Hawqal, p. 163.
https://doi.org/10.1017/CBO9780511622380.010
THE FAJIMIDS 2OI
tary slave or serf. The land-tax these serfs paid was aptly called
kird\ rent.l For the purpose of this tax a survey was made and the
rates were reassessed.2 Judging from the figures we have of the total
revenues collected these rates do not seem to have been excessive.3
Land-tax collecting continued on the basis oiqabdla, i.e. contracting
to pay a fixed sum for a given district. But there were crucial
differences from the original system introduced by Ma'mun a
century and a half earlier. First, these contracts were auctioned and
given to the highest bidder to the detriment of the tax-payer.*
Second, the contractor was not required to be a member of the tax-
community concerned. Indeed government officials and, more
seriously, army commanders were allowed to undertake such con-
tracts.5 In most cases these contracts took the character of tax-
farming. In due course a powerful contractor was able to pass his
contract to his descendants and the district concerned became
virtually an iqta', land-grant. The third difference produced cata-
strophic results. Under the original system the contractor had had to
undertake the responsibility for maintaining and repairing any
damage to the intricate network of canals and dikes without which
the irrigation system would collapse absolutely, and allowances had
been made for the expenses involved.6 Whether out of ignorance or
on principle the Fatimids overlooked this vital factor in Egyptian
agriculture. No provisions were made at all for the maintenance of
the irrigation system, and for the first time in millennia of Egyptian
history this task was left to the individual peasants. The only pre-
caution the Fatimids took was against floods. The major dikes were
divided into stretches, for the maintenance of which a levy of ten
dinars per stretch was imposed, and the government itself undertook
to do the necessary work.7 Needless to say, no work was done and
the treasury found other urgent needs for spending the money.8 In
a few decades the inevitable and rapid deterioration of agriculture
came about and the result was not only famines year after year, but
also a drastic fall in taxes.9 Maqrizi never ceased to lament this
situation, repeatedly reminding us that from time immemorial one
fourth if not one third of the revenues of Egypt had always been
1
Maqrizi, Kbitaf, vol. i, p. 8;; Maqdisi, pp. 64-5 and the variant reading in the foot-
notes, also p. zi2.
1
Hawqal, p. 143; Maqrizi, Kbitaf, vol. 1, p. 405.
1 4
Maqrizi, Kbitaf, vol. 1, p. 82. Idem, vol. H, pp. 5-6.
' Idem, vol. 1, p. 85. ' S e e above, p. 60.
7
Maqrizi, Kbitaf, vol. I, p. n o . * Ibid., p. 101.
• Ibid., pp. *49-J°-
https://doi.org/10.1017/CBO9780511622380.010
2O2 THE FATIMIDS
allocated to the care and upkeep of the irrigation system.' He also

pointed out that after the coming of the Fatimids the northern half
of the Nile Delta was lost to agriculture and became waste
land.2
As for trade the Fatimids had two objectives. One was to use it as
a means to tax urban communities and thus introduce a more
equitable tax system in their domains. As a consequence of their
usual methodical approach not a single product, trade or profession
escaped taxation. Taxes were imposed on all items entering any town
and collected at gates, road-blocks, quays, market places, ports and
custom-houses. Pottery makers, home spinners, trinket producers,
leather tanners, vinegar distillers, bakers, oil pressers, beer makers
and prostitutes all had to pay their taxes. Fish were taxed at the sea-
side and when they were preserved. Cattle were taxed on grazing
grounds, in transport and at the slaughter house.3 A value-added
tax was imposed on the all important textile industry at every stage
of production from raw material to final product.4 And of course
custom duties were collected on imports and exports.s Also,
broking houses were established wherever goods in large quantities
changed hands and government agents collected stipulated commis-
sions in addition to excise tax.6 All these taxes were paid by Muslims
and non-Muslims alike. However as the latter had to pay a poll-tax,
the former were required to pay their %akdt. We have no information
about the rate of this Muslim tax and it seems that it was difficult to
enforce.7 All the same, members of the Ismaili sect had to pay what
was called najwd, collection, and in fact it was not much different
from a Sunday church collection. Living in a Christian environment
the Fatimids seem to have learnt a great deal from their subjects. All
Isma'ili believers were required to attend exclusive meetings held
twice a week on Mondays and Thursdays, and under the guidance of
a representative of the chief missionary they read the lesson. At
every meeting this representative collected the najwd which varied
from three and a third dirhams to 3 3 and a third dinars according to
the wealth of the believer.8 One can imagine that such faithful
members would receive special treatment from the state in much the
same way as nowadays card-carrying members would expect special
treatment in a one-party regime.
1 2
Ibid., pp. 61, 76, 100. Ibid., p. 171.
3 4
Ibid., pp. 104-5. MaqdisI, p. 215.
s 6
Maqrlzl, Khitat, vol. i, pp. 109, i n . Ibid., p. 89.
7 8
Ibid., p. 368. Ibid., p. 391.
https://doi.org/10.1017/CBO9780511622380.010
THE FATIMIDS 203
The Fatimids' other objective with regard to trade was to capture
a major share of international trade. Geographically Egypt was in
an ideal location for this purpose, and it had access to African
produce. The Fatimids went all the way and, unabashed, used all
possible means to attract the Indian Ocean trade from the Persian
Gulf to the Red Sea. They had long had missionaries in Yaman who
had not yet achieved much result. Now, with increased efforts the
Isma'ili missionaries in India were more successful, not particularly
in converting the Indians to Isma'ilism, but more as a kind of
consular service. Gradually the Yamanis began to realize the advan-
tages of giving a sympathetic ear to the Isma'ilis, but it took some
time before an Isma'ili regime was established in Yaman in 1048/439
under the Sulayhids. As suggested before this did not mean that
Yaman became part of a Fatimid empire. Indeed by this time the
Fatimids had already begun to decline and the Sulayhids were of no
help to them. On the other hand Hijaz, which was dependent on
Egypt for its grain, was more amenable to co-operating with the
Fatimids, although Isma'ilism never gained a foothold there. The
fact remains that it was of no vital importance for the Fatimids to
dominate the east coast of the Red Sea, because the ports of the west
coast were more than enough to handle all the available trade from
the Indian Ocean. The co-operation of the Nubians was more
important not only because it gave them access to the African trade,
but also for the safety of the route from 'Aydhab to Aswan.'
Securing the sources of trade was comparatively easy, but securing
markets was a different matter. In their long term planning the
Fatimids had justifiably counted on controlling Syria, but in the
event these calculations proved costly. Their arrival in Egypt
coincided with the devastating attack of the Byzantines in 969/3 5 8
and their domination of northern Syria. The outcome of this dubious
victory was that while they were able to advance their borders well
to the south of the Taurus Mountains, they had to sustain inordinate
efforts to exercise any control over their conquests. The chaotic
conditions that had existed in the relatively isolated thughiir in the
north were simply transferred to Syria, where they were impossible to
contain. In this densely populated province every town, city and
district rose up in arms to defend its own interests. The Arabs of the
Syrian-'Iraqi desert and those driven back from the thughiir roamed
around establishing themselves wherever possible. They fought
amongst themselves, sometimes in support of the Byzantines and at
1
Hawqal, p. 150.
https://doi.org/10.1017/CBO9780511622380.010
204 THE FATIMIDS
other times against them. The appearance of the Fatimids on the
scene added to the confusion. The people of Damascus hired 400
defectors from the Buyids to fight for them against the Fatimids.1
They were even willing to accept the suzerainty of the Byzantines
over Damascus if they could only establish themselves there. Taking
advantage of this situation the Qaramita of Bahrayn attacked the
Fatimids in Syria and advanced into Egypt but were repelled in
971/361. Two years later they attacked Cairo itself but were again
defeated. It does not serve any purpose to recount the details of this
endless fighting or even to mention the names of the various Arab
petty dynasties that fell almost as soon as they came to power. The
important thing here is to realize that while the Buyids stayed out of
this situation, neither the Byzantines nor the Fatimids were able to
establish any dominant presence in Syria.
The difficulties that the Fatimids encountered in Syria obliged
them to try to strengthen their army. As there were no more
Berbers available they tried to entice defectors from the armies of the
Buyids and the Hamdanids, and soon Daylamites, Arabs and many
others were serving in the Fatimid army.2 When these were not
enough the siiddn provided another source of new recruits. Members
of this heterogeneous army spent more effort fighting amongst
themselves than against the enemy. In the army there were no less
than thirty groups; each group was known by a certain name, and
of course they were all considered 'abid, slaves of the Imam* How-
ever, there was another group who really were of slave origin who
were known as tardbi, i.e. those who are fostered. As Maqrizi
explained it, they were young captives put into the charge of palace
officials who would teach them writing and archery, and eventually
some of them became commanders of the special guards of the
ruler.4 This is the first example of actual slaves in the army, and
interestingly enough they were not even called 'abid.
The unsettled situation in Syria and the confrontation there
between the Byzantines and the Fatimids did not stop the two powers
from trading with each other. The Syrian ports continued to trade
with Constantinople through Cyprus and Byzantine merchant ships
brought goods to and carried them from Old Cairo.s Indeed the
emperor Basil II 976/1025 sent 'Aziz 975-996/365-386 the magnifi-
cent gift of 28 enamelled trays inlaid with gold, each one valued at
1 1
Maqrizi, Khitat, vol. II, p. 8. Ibid., p. 12.
3
Ibid., p . 21. * Ibid., p. 194.
1
Maqdisi, p. 194; Hawqal, p. 176; Athir, vol. via, p. 388.
https://doi.org/10.1017/CBO9780511622380.010
THE FAJIMIDS 2OJ
3000 dinars.' On the other hand the Italian city states were quick to
send their agents to Alexandria and Cairo to establish their commer-
cial interests. This alternative venue for trade encouraged 'Aziz to
press for a military advantage in Syria even against the advice of his
experienced wa^tr Ya'qub b. Killis.2 For this purpose 'Aziz started
the practice of recruiting men from the eastern regions, although his
Berber troops were very much opposed to such a move which
represented a threat to their own position within the regime.3 This
antipathy between the westerners, the Berbers, and the easterners
eventually developed into open hostility which was in turn
aggravated by the appearance of more suddn in the Fatimid armies
after the death of 'Aziz.
To encourage trade the ingenuity of the theoretician Qadi Nu'man
was utilized to find a solution to legitimise the practice of usury. We
know that although Islam strictly forbids usury, it had long been
practised in Baghdad, though it had continued to be a controversial
issue.4 In North Africa some obscure Shi'ite group had also found it
necessary to practice usury to stimulate their trade, and had argued
that it was not different from any other business transaction.5 The
ingenious Fatimid theoretician was more subtle, offering his own
definition of usury. He argued that it was certainly illegal to make a
loan in gold dinars and pay it back with interest in gold dinars. But
if a loan of iooo silver dirhams and one gold dinar was repaid with
2000 silver dirhams then that was not usury, and the same applied to
any transaction that involved any goods, as long as the repayment
was made in different kind from the loan.6 In his own way 'Aziz
was no less subtle than his advisers, and this shows very clearly in
his manipulation of the Christians of Egypt. The majority of the
peasants of Egypt were Monophysite as against the Greek Ortho-
doxy of the Byzantine Church.7 In Cairo and particularly Alexandria,
however, there were Melkites who recognized the supremacy of the
Patriarch of Constantinople. 'Aziz, who had a Christian wife, had
her brother Orestes appointed as the Melkite Patriarch of Jerusalem
and another brother Arsenius as the Metropolitan of Cairo in 986/
375.* This intriguingly close association of a Muslim ruler with his
Christian subjects is a good illustration of the nature of Fatimid rule
1
Maqrizi, Khilal, vol. I, p. 415. * Athlr, vol. ix, p. j<f.
1
Maqrizi, Kbilal, vol. 1, p. 451. vol. II, p. iz.
5
• Athir, vol. x, p. 155; Mawardi, Abkam, p. 423. Bairn, p. 161.
* Al-Qadi al-Nu'man, Kilab al-lqliidr, cd. M. W. Mirza, Damascus, 1957, p. 84; idem,
Da'a'im al-hlim, ed. A. Fayzi, Cairo, i960, vol. u, pp. 37, 40.
7
Maijrizi, Khilal, vol. it, p. JOI. • AntakI, pp. 164-5.
https://doi.org/10.1017/CBO9780511622380.010
2O6 THE FATIMIDS
in Egypt. It also indicates that 'Aziz was encouraging the Melkites

in order to build new bridges between his domains and Byzantium.
Such links would help the cause of trade. • The existence in Egypt
of a colony of Takrur, siiddn from Mali where they had their own
quay in Cairo, signifies that Aziz was not oblivious to the interests
of the suppliers of the merchandise.2
When 'Aziz died he was succeeded by his son Hakim, 996-1021/
386-411, one of the most controversial figures in Islamic history.
He has been accused of schizophrenia, melancholia, mental and
emotional instability and cruelty, among other things. On the other
hand he has been credited with idealism, generosity, benevolence
and even genius. Yet the best description of him is by a good
historian who lived about two centuries later. He said that Hakim
was trying to emulate Ma'mun.3 We know the difficult situation
Ma'mun had faced and how supple he had had to be to survive.4
Hakim was no different, he had problems and tried to cope with them
to the best of his ability. The difference was, perhaps, that he came
to power at the age of eleven. The Berbers of Kutama thought that
they had an opportunity to purge the army of the easterners. The
Christian wasita of 'Aziz was murdered and replaced by Hasan b.
'Ammar, a prestigious Berber leader. Nevertheless they seem to have
overestimated their strength because within a year they were dis-
lodged by the easterners who then installed Barjawan as the new
wasita. Here again we have another white khasiyy who was signifi-
cantly described as "physically perfect", i.e. definitely not castrated.s
For four years Barjawan, with the help of a Christian administrator,
was the real ruler of Egypt. He banished the Kutama to Syria and
continued to press for military advantage there. His last act however,
was to respond to an approach by Basil II, made through Orestes
the Patriarch of Jerusalem, to conclude a peace treaty for ten years
beginning 1001/391.6 Having reached the age of sixteen Hakim
decided that it was high time for him to exercise his rights and
accordingly he arranged for the assassination of Barjawan.
The first important action of Hakim in 1000/390 was to promote
his uncle Arsenius the Metropolitan of Cairo to the office of Patriarch
of Alexandria.7 This was an auspicious sign that he intended to follow
his father's policy, but soon things began to go wrong. In 1005/395
1
Taghribardi, vol. iv, pp. 151-2. * Maqrizi, Kbilat, vol. 11, p. 326.
1
Idem, Itli'dz al-Himaja'', ed. M. Hilmi Ahmad, Cairo, 1971, vol. n, p. 117.
4 5
Sec above, p. 55. Maqrizi, Khifal, vol. 11, p. 3.
* Antaki, p. 184. ' Ibid., p. 185.
https://doi.org/10.1017/CBO9780511622380.010
THE FATIMIDS Z07
a revolt broke out in Cyrenaica which must have had an economic
recession as a result of the loss of its trade through the Egyptian
Oases. The Arabs in the vicinity of Alexandria also supported this
revolt which was led by a descendant of the Umayyads. It took over a
year to suppress this revolt. The unfortunate leader fled and ended
up in Nubia where he was arrested. Dutifully the Nubian king
delivered his worthwhile captive to Cairo. However, as Cyrenaica
continued to be unsettled it was written off and conceded to the
Zirids of Tunisia.'
The agricultural policy of the Fatimids that had left the mainten-
ance of the irrigation system to the public had begun to show its
disastrous effects. After thirty years of neglect the canals gradually
clogged up, and the inevitable result was not only a drop in the
harvest but also the shrinking of the area of arable land. This was
the worst crisis of the Fatimids and it was the major concern of
Hakim. The measures he introduced and for which he has been
accused of madness will all make sense when seen in this light. His
target was to increase necessary food supplies, especially for the
Cairo metropolis, and to stimulate the declining agricultural pro-
duction as much as possible. His measures included the prohibition
of beer and wine, the destruction of vines around Cairo, the prohibi-
tion of eating lupine, water-cress, muluki)iyya and fish without scales
and the killing of all dogs except hounds.2 In a country where grain
was in short supply and where bread was the basic diet, it was
certainly absurd to use the grain for making beer which only the
rich could afford. By the same token it was advisable to use the
available land particularly around Cairo to grow grain rather than
vines. Lupine is until this day an appetiser associated with drink in
Egypt. Every Egyptian and every visitor to Egypt knows how
popular mulukiriyya is in the country: at one point the Egyptian
Cultural Office in Washington D.C. turned its garden into a field for
it. It is a mucilaginous vegetable, Corchorus oiiforius,with the common
name Jew's Mallow, and is essentially a weed. In the country it
grows everywhere, but near the cities, because of the demand for it,
it is cultivated like any other vegetable. Water-cress is almost in the
same category as mulukinyya, and Hakim's reason for prohibiting the
eating of both was the conservation of land for more essential foods.
As for the fish without scales, the reference must have been to the
barbel, the only kind of this species which is abundant in every
1
Ibid., pp. 188-91; 'Idhiri, vol. 1, pp. 257-8; Maqrizi, Kbilal, vol. 11, p. 286.
1
Antaki, p. 188; Maqrizi, Kbilal, vol. II, pp. 286, 287 288, 342.
https://doi.org/10.1017/CBO9780511622380.010
2O8 THE FAflMIDS
stretch of water in the country. It is a mud-fish which burrows into

the mud and survives on the sediments at the bottom. Therefore it
actually performs the environmental function of clearing up the
waterways. This service is still recognized in Egypt in the fact that
there is a self-imposed prohibition against fishing it out of the wells
of water-wheels which are impossible to dredge. Hakim extended
this prohibition to all canals to protect this useful species and allow
it to do what the people did not do. It is interesting to know that
this fish is known today as qarmiit, a clear reference to the Qaramita
As masters of the art of propaganda, the Fatimids introduced this
association to make this fish unpopular, and the term stuck in much
the same way as the name 'Umar - the Devil incarnate from the
Fatimid point of view - is still used as a term of abuse, particularly in
the old parts of Cairo. The killing of all dogs, with the exception of
hounds, is not as abhorrent as it may seem. In the first place there are
millions of stray dogs in Egypt and many of them are infected with
rabies. In the second place, stray or not, these animals are just as
starved as the rest of the human population. They are not given
tinned dog food and there are no scraps to be had; they eat bread.
In the time of Hakim the dog situation must have been worse than
it is now, and bread was scarce. There could not have been many
hounds in Egypt at the time for there has never been much to hunt.
However these hounds were spared for the meagre amount of game
they might have helped to run down. Another measure Hakim
introduced was to prohibit bakers from using their feet in kneading
the dough, and this proves his sanity beyond any doubt.1 He was
not a genius but he certainly was doing his best in a difficult situation.
Turning to the political arena Hakim took a number of steps that
were designed to curtail the influence of the Melkite Church in his
domains. He confiscated the properties of his Melkite mother and in
the process annulled many of the religious endowments of the
Church.2 A year later in 1010/400, he arranged for the assassination
of his uncle Arsenius, the Melkite Patriarch of Alexandria.3 Almost
at the same time he ordered the destruction of the church of the
Holy Sepulchre in Jerusalem where his other uncle Orestes had been
the Melkite Patriarch a few years earlier. It will be remembered that
it was this uncle who had arranged for the treaty with Byzantium.
Of course the Byzantines were disturbed by this needless act of
destruction and the more so because it amounted to a declaration
1 1
Maqrizi, Khitat, vol. n, p. 341. Antiki, pp. 195, 196, 197.
1
Ibid., p. 197.
https://doi.org/10.1017/CBO9780511622380.010
THE FATIMIDS 209
that their treaty with the Fatimids was at an end and was not going
to be renegotiated. None of these moves was meant to eradicate the
Melkite Church, their purpose was to appeal to the Coptic majority
in Egypt and rally its support. These Copts were the peasantry of
the country whose enthusiasm was a decisive factor in fighting the
agricultural decline.1 It was also for this purpose that certain taxes
were abolished, although the frequency of Hakim's orders for the
abolition and re-imposition of these taxes was rather a result of the
economic malaise. We do not know exactly what these taxes were,
but we know that land-tax was not involved in these orders. Yet,
the fact remains that some of the Christians were satisfied with the
abolition of some of the taxes.2
It was almost inevitable that Hakim's turn against the Melkites
should adversely affect Fatimid trade with the Byzantines. However,
the first reaction came from the Arabs of the Tayy in Palestine. They
mounted a considerable revolt and in 1012/403 brought an obscure
'Alawid, who was ruling Makka, and proclaimed him their Amir
al-Mu'minin.* Although this attempt came to nothing the Fatimids
lost control of most of Palestine, and were only able to hold on to
part of the Syrian coast. In 1015/406 Basil II severed all commercial
relations with the Fatimids and confined all trading to Aleppo which
was under Byzantine domination.4 This marked the beginning of
the decline of Fatimid trade with Europe and Africa. The Zirids
in Tunisia who controlled practically all the Mediterranean African
coast were in a better position to trade with the Italian city states.
Furthermore they succeeded in attracting all the African trade to
their territories, and maintained good relations with the sudan. It was
as an advertisement of their commercial success that the Zirids
sent Hakim in 1014/405 a present of sables, of all things.5
Hakim's preoccupation with his Christian subjects did not pre-
clude his concern for the Muslims. Although the latter were in the
minority they provided the power base of the regime. Realizing that
the Isma'ilis were themselves a minority among the rest of the
Muslims, and trying to increase the popularity of his regime, he
started in 1008/399 a n e w policy of toleration towards non-Isma'ilis
by relaxing some of the restriction on certain ritual practices.6 This
deviation from strict Isma'ilism angered the Old Guard of the move-
1
Ibid., p. 195. » Ibid., pp. 205-6.
1
Athlr, vol. ix, p. 86; Maqrizi, Kbilat, vol. i, pp. 185, 211.
• Antiki, p. 214. > 'Whirl, vol. I, pp. 260-1.
* Maqrizi, Kiifaf, vol. 11, pp. 540, 54).
https://doi.org/10.1017/CBO9780511622380.010
2IO THE FAJIMIDS
ment almost to the point of open revolt. Undaunted, Hakim per-

severed with his policy and, to intimidate the opposition, he went to
the extent of executing, in 1011/401, two of the most illustrious
leaders of the time - the son of general Jawhar and the son of the
founding theoretician Qadi Nu'man who had both inherited their
fathers' positions.1 In 1012/403 Hakim made a most astounding
declaration, which was in effect a repudiation of the primary axiom
of Isma'ilism. Henceforth, he wanted to be considered only as the
Amir al-Mu'minin and no more, i.e. he ceased to be the Imam. To
emphasize his new position he declared that he should be addressed
only by this title in the simplest possible form, and that none of his
subjects should consider himself as 'abd, slave of the ruler.2 In the
following year he departed further from Isma'ili dogma which strictly
stipulated primogeniture, by appointing a cousin as his heir apparent.3
These radical shifts confused some of the Isma'ilis to the extent that
some believed that Hakim had transcended the human confine and
had become the incarnation of the Divinity. Out of this confusion
rose the movement, in 1017/408, of the Druze who still exist in
various parts of Syria. However, the Fatimid establishment was
further upset when Hakim abolished the twice-weekly lesson and
collection.4 Accordingly he was assassinated, almost certainly with
the connivance of his sister, in 1021/411.
At this point there was a return to orthodox Isma'ilism, yet the
role of the Imam became less prominent, especially as many of the
later Fatimid rulers came to power as children. Hakim was succeeded
by his son Zahir who was sixteen years old. For the first five years
of his reign his aunt was the effective ruler and after her death he
continued to rule on the same lines until he died in 1036/427. He in
turn was succeeded by his seven year old son Mustansir 1036-94/
427-87. The affairs of state were run by a succession of wdsitas whose
main efforts were to try to maintain the status quo. But this was an
impossible task. The annals of this period are simply repeated
accounts of continuous chaos in Syria, endless fighting between the
various groups within the Fatimid army, and constant deterioration
in the economic conditions of Egypt. Although a new treaty was
concluded with the Byzantines in 1038/429 and renewed in 1048/439,
it had no effect either on the stability of Syria or as a stimulant to
Fatimid trade.5 The agricultural output declined so rapidly that in
1
Ibid., p. 15. » ibid., pp. 285,288.
1
Ibid., pp. 288, 289. ' Idem, vol. I, p. 391.
s
AthTr, vol. ix, pp. 526, 580.
https://doi.org/10.1017/CBO9780511622380.010
THE FAJIMIDS 211
1054/446 the food crisis forced the Fatimids to beg the Byzantines
for grain.1 This "Greatest Crisis" continued unabated until 1074/
466, when the Fatimids had no option but to allow a military take-
over to save themselves and the country from total disintegration.2
Ironically it was an Armenian mercenary, who happened to have
been in control of Acre during the prevailing anarchy in Syria,
whom the Fatimids invited to perform this rescue operation.
It was under these circumstances that the so-called "high water
mark of Fatimid expansion came in the years 1057-9/448-51, when
a Turkish general in 'Iraq called Arslan al-Basasiri went over to the
Fatimid side and proclaimed the Fatimid Caliph first in Mawsil and
then, for a year, in Baghdad itself. Despite the efforts of the Chief
Missionary, however, the Fatimid government was unable to
provide effective support, and the strongly Sunni Saljuqs drove
al-Basasiri out of Baghdad."3 The general in question had been a
member of those crumbling Buyid forces who had fled in all direc-
tions when the Saljuqs had first entered Baghdad in 1055/448.
Taking advantage of a temporary absence of the Saljuq forces from
the city he entered it with no more than 400 bedraggled followers.4
As soon as the Saljuqs returned in 1060/451, he fled for his life but
was ignominiously captured and executed. If that was the "high
water mark of Fatimid expansion" one wonders what would be
considered the low water mark. The fact of the matter is that the
high water mark of the Fatimids, as well as that of the Buyids, had
long since passed, and Basasiri had had no choice of side open to
him. The 'Abbasids had no power, the Buyids had disappeared and
the Fatimids were in total decline. After 1055/448 there was only
one side, the Saljuqs.
1
Maqrizi, Kbitat, vol. I, p. 535. * Ibid., p. $57.
1
Lewis, Ir.ttrprtlalton, p. 292. • Athir, vol. ix, p. 441.
https://doi.org/10.1017/CBO9780511622380.010
WORKS CITED
SOURCES
ibn al-Abbar, Vtdb al-Kuttdb, ed. S. al-Ashtar, Damascus, 1961.
'Arib b. Sa'd, §ilat Tdrikh al-Tubari, ed. M. J. dc Goejc, Leiden,
1897.
ibn A'tham al-Kufi, Kitdb al-Futuh, Istanbul manuscript, Library
Ahmet III, no. 2956.
ibn al-Athir, 'Izz al-Din, al Kami/ft al-Tdrikh, ed. C. J. Tornberg,
Leiden, 1866-71.
al-Azdi, Yazid b. Muhammad, Tdrikh ai-Mawsil, ed. 'A. Habiba,
Cairo, 1967.
al-Bakri, Abu 'Ubayd, al-Mughrib . . ., ed. De Slane, Paris, 1911.
al-Baladhuri, Ahmad b. Yahya, Futiib al-Bulddn, ed. M. J. de Goeje,
Leiden, 1866.
ibn al-Batriq, Sa'ld, Na%m al-Jawdbir, ed. L. Cheikho, Paris, 1909.
ibn Fadlan, Risd/a, ed. S. Dahhan, Damascus, 1959.
al-Hamdani, Muhammad b. 'Abdilmalik, Takmilat Tdrikh al-Tabari,
ed. A. Y. Kan'an, Beirut, 1961.
Hamdullah Mustawfi, Nu%hat al-Qulub, ed. G. le Strange, London,
1915.
ibn Hanbal, Musnad, ed. al-Sa'ati, Cairo, A.H. 1357.
ibn Hawqal, Siirat al-Ard, ed. J. H. Kramers, Leiden, 1938-9.
ibn Hazm, 'AH b. Muhammad, Jamharat Ansdb al-'Arab, cd. A.
Harun, Cairo, 1962.
ibn 'Idhari, Ahmad b. Muhammad, al-Bajdn al-Mughrib, ed. G. S.
Colin and E. LeVi-Provencal, Leiden, 1948-51.
al-Isfahani, Abu al-Faraj, Maqdtil al-Tdlibiyyin, ed. S. S. Saqr,
Cairo, 1949.
al-Istakhri, Abu Ishaq Ibrahim, Kitdb al-Masdlik wa al-Mamdlik, ed.
M. J. de Goeje, Leiden, 1870.
al-Jahshiyari, Muhammad b. 'Abdus, Kitdb aI-Wu%ard\ ed. M. al-
Saqqa et al., Cairo, 1938.
ibn Khaldun, 'Abdulrahman, al-Muqaddima, Beirut, 1961.
ibn Khallikan, Wafqydt al-A'jdn, ed. Ihsan 'Abbas, Beirut, 1972.
ibn Khurdadhbeh, Kitdb al-Masdlik wa al-Mamdlik, ed. M. J. de
Goeje, Leiden, 1889.
al-Kindi, Muhammad b. Yusuf, Kitdb al-Wuldt, ed. R. Guest, Gibb
Memorial Series, vol. xix, London, 1912.
https://doi.org/10.1017/CBO9780511622380.011
WORKS CITED 2X3
al-Maghribi, 'All, b. Musa, Kitdb al-Jughrdfya, ed. Isma'il al-'Arabi,
Beirut, 1970.
ibn Manzur, Lasdn al-Arab, Bulaq, 1891.
al-Maqdisi, Shams al-Din, Ahsan al-Taqdslm . . ., ed. M. J. de Goeje,
Leiden, 1877.
Maqrizi, Ahmad b. 'Ali, al-Khitat, Cairo, A.H. 1249.
Itti'dg al-Httnafd'', vol. 1, ed. J. Shayyal, Cairo, 1948; vol. 11,
ed. M. Hilmi Ahmad, Cairo, 1971.
al-Nuqiid al-Isldmiyya, ed. M. S. A. Bahr al-Ulum, Najaf, 1967.
al-Marvazi, Tahir, On China, The Turks and India, Arabic text, ed.
V. Minorsky, London, 1942.
al-Mas'udi, 'Ali b. al-Husayn, al-Tanblh wa al-Ishrdf, ed. M. J. de
Goeje, Leiden, 1893.
Muruj al-Dhahab, ed. C. Barbier de Meynard and P. de Courteille,
Paris, 1861-77.
al-Mawardi, al-Ahkdm al-Sultdniyya, ed. R. Enger, Bonn, 1853.
Miskawayh, Ahmad b. Muhammad, Tajarib al-Umam, ed. H. F.
Amedroz, Oxford, 1920-1.
Nasir-i-Khusraw, Safar-Ndme, ed. C. Schdfer, Paris, 1881.
ibn al-Qalanisi, Dhayl Tdrlkh Dimashq, ed. H. F. Amedroz, London,
1908.
Qudama b. Ja'far, Kitdb al-Khardj, part seven, published as part of
A. Ben Shemesh, Taxation in Islam, vol. 11, Leiden, 1965.
al-Qummi, Ash'ari, Kitdb al-Maqdldt wa al-Firaq, ed. M. J. Mashkur,
Teheran, 1963.
Rasd'il al Bulaghd', ed. M. Kurd 'Ali, Cairo, 1946.
ibn Rusteh, al-A'ldq al-Nafisa, ed. M. J. de Goeje, London, 1892.
al-Sabi, Hilal, Kitdb al-Wwyird', ed. H. F. Amedroz, Leiden, 1904.
Rusiim Ddr al-Khildfa, ed. Mikha'il 'Awad, Baghdad, 1964.
Tdrikh, part 8, ed. H. F. Amedroz, and published as vol. iv in
Dhayl Tajarib al-Umam, Cairo, 1916.
al-Sabi, Thabit, Tdrikh Akhbdr al-Qardmita, ed. Z. Bakkar, Damascus,
1970.
ibn Sallam, Abu 'Ubayd al-Qasim, Kitdb al-Amwdl, ed. M. H. al-
Fiqi, Cairo, A.H. 1353.
al-Sam'ani, 'Abdulkarim, al-Ansdb, ed. D. S. Margoliouth, Leiden,
1912.
al-Shabankari, Majma" al-Ansdb Ft al-Tawdrikh, Istanbul manuscript,
Yeni Cami, 909.
abu Shuja', Dhayl. . ., ed. H. F. Amedroz, Cairo, 1916.
al-Suli, Abu Bakr, Akhbdr al-Rddi..., ed. J. H. Dunne, London, 195 3
https://doi.org/10.1017/CBO9780511622380.011
214 WORKS CITED
al-Tabari, Muhammad b. Jarir, Tarikh al-Rusul wa al-Muluk, ed.
M. J. de Goeje et al., Leiden, 1879-1901.
ibn Taghribardi, al-Nujum al-Zdhira, Cairo, 1963-72.
Tayfur, Ahmad b. Abi Tahir, Kitdb Baghdad, ed. H. Keller, Leipzig,
1908.
al-'Utbi, al-Tdrikh al-Yamini, Cairo, A.H. 1286.
Yahya b. Adam, Kitdb al-Khardj, ed. A. M. Shakir, Cairo, A.H. 1347.
al-Ya'qubi, Ahmad b. Abi Yaq'ub, Tarikh, Beirut, 1970.
Kitdb al-Bulddn, ed. M. J. de Goeje, Leiden, 1892.
Yaqut, Shihab al-Din, MM1 jam al-Bulddn, Beirut, 1957.
abu Yusuf, Ya'qub, Kitdb al-Khardj, Cairo, A.H. 1302.
BOOKS AND ARTICLES

Ayyub, M. S., Jirma, Tripoli, 1969.
Barthold, W., Turkestan Down to the Mongol Invasion, London, 1928.
Bosworth, C. E., Slstdn Under the Arabs, Rome, 1968.
The Ghavyiavids, Edinburgh, 1963.
Cahen, Cl., " L'e volution de Yiqta* du ixe au xme siecle. Contribution
a une histoire comparee des societes medidvales ", Annales:
Bconomie, societes, civilisations, vin (1953)-
"The Turkish Invasion: the Selchukids", A History of the Crusades,
ed. K. M. Setton, London, 1958.
de Goeje, M. J., Memoire sur les Carmathes du Bahrain et les Fatimides,
Leiden, 1886.
Lambton, A. K. S., Landlord and Peasant in Persia, Oxford, 1953.
"Reflections on the IQTA"' in George Maqdisi, ed., Arabic and
Islamic Studies in Honor of Hamilton A. R. Gibb, Leiden, 1965.
Lane, E. W., An Arabic-English Lexicon, reprinted, Beirut, 1968.
Lewis, Bernard, The Origins of Ismd'ilism, Cambridge, 1940.
"The Fatimids and the Route to India", Revue de la Faculte de
Sciences Economiaues de I'Universite d'Istanbul, 11 annde, no. 1-4,
Istanbul, 1953.
"An Interpretation of Fatimid History", Colloque International sur
I'Histoire du Caire 27 Mars-} Avril 1969, Cairo, n.d.
Nazim, M., "The Pand-Nameh of Subiiktigin", Journal of the Royal
Asiatic Society, London, 1933.
Noldeke, Th., Sketches from Eastern History, tr. by J. S. Black,
Edinburgh, 1892.
Rekaya, M., "Mise au point sur Theophobe et l'alliance de Babek
avec Theophile", Byântion, t. XLIV, Bruxelles, 1974.
https://doi.org/10.1017/CBO9780511622380.011
WORKS CITED 215
Shaban, M. A., The ' Abbasid Revolution, Cambridge, 1970.
Islamic History A.D. 600-yjo (A.H. IJI), A New Interpretation,
Cambridge, 1971.
Sourdel, D., Le Vizirat Abbaside, Damas, 1959-60.
Yule, Colonel Henry, The Book of Ser Marco Polo, The Venetian,
London, 1875.
https://doi.org/10.1017/CBO9780511622380.011
INDEX
'Abbas b.Hasan, 136, 137 Armenians, 173, 174, 211; in 'Abbasid
'Abbas, son of Ma'mun, 53,61,62 service, 62, 76, 77
'Abbas, son of Ibn Tulun, 109, i n Arsenius, Metropolitan of Cairo, 205, 206,
abm' al-dawlajdawa, 30, 37-46, 50, 52, 53, 208
62 Arslan Basasiri, 211
Abu 'Abdillah, 190-2 Asad, clan of, 131
'Abdullah b. 'Ali, 6, 7, n , 12, Ashnas, 66, 69, 73
'Abdullah b. Tahir, 69 Aswan, HI, 196-7, 203
'Abdulmalik b. Salih, 29, 30, 31, 39, 40, 'avifim, 29-30, 53
43 'Aydhab, 107, i n , 203
'Abdulaqays, clan of, 101 Abu Ayyub al-Muriyani, 11
'Abdulrazzaqs, 178, 179 Azd, tribe of, 26, 98
Acre, 2it Azhar, 198
Adherbayjan, 26, 28, 30, 31, 33, 76, 77, 'Aziz, 204-6
122, 125, 148, 149, 151, 169, 174;
governors of, 74,121,124,173; mineral BSbak, 34, j6, 58, 59, 62, 67, 77
resources of, 56-9 Badghis, 5, 12, 15
Afshin, 50, 60, 61, 62, 64, 66-7, 72, 82, Badr, 133-4
«75 Baghdad, 47, 50, 73, 93, 100, 107, 108,
Aghlabids, 103,112,134,190-1 119, 120, 135, 136, 139, 144, 146, 149,
Ahmad b. Abi Duwad, 68, 71 151, 211; foundation of, 8-9, 11; army
Ahmad b. Hanbal, 55 in, 11, 26, 30, 34, 37,42, 45, 52, 71, 82,
Ahmad b. Isra'il, 85 84. 9*. " 7 . 137. '4». «4«. "95! battles
Ahmad b. Abi Khalid, 50-2 in, 44, 46, 82-3; under the Buyids, 152,
Ahmad b. al-Khasib, 80, 81 158, 160-70
Ahmad b. al-Mudabbir, 79, 8j, 95, Bahila, clan of, 101
109 Bahrayn, Qarami?a in, 128,130,131,133-
Ahmad b. Muhammad of Tayy. 120, 130, 5,153,165,168,204
Baku, 57
Aljmad b. Tulun, 85, 95, 109-12 Balkh, 11, 49
Alexandria, 134. 147. ' 9 0 . *9*. 193. '94. Baluchistan, 97
196, 205, 206 Bamyan, 83
Aleppo, 29, 30, 169, 171, 172, 209 Banijur, 64
Algeria, 103, 190, 191 Banu al-Asbagh, 132
'Ali Baba, 78 Bardas Sclerus, 172
'Ali b. "Isi b. Mahan, 37, 38, 40, 43, 44 Bardas Phocas, 172
'Ali b. 'Isa, the "good" vizier, 142-5, Barjawin, 206
146, 152 Baridis, 157, 165
'Ali b. Yahya, the Armenian, 77 Barqa, 35
Amid, 122, 124 Barmakids, 31, 32, 33, 35-9, 40, 42, 90
Amin, 39, 41, 44, 45 Barthold V., 175
Amir al-Mu'mirin, 1, 2, 6, 7, 8, 45, 47, 86, Basra, 14, 15, 20, 26, 98, 101, 102, 131,
92,162,164,191, 209, 210 153,165-6; trade and taxation of, 17-18,
Amir al-Umari', 157, 170 69,79,108,166,169
Anbir, 3, 5 Basil II, Byzantine Emperor, 172, 204,
Antioch, 29, 79, i n , 172 206,209
Arabia, 24 Batflja 168, 184
Armenia, governors of, 121, 124; rebel- BSykbik, 85-7
lions in 27, *8, jo, 31, 77, 78; trade of, Berbers, 12, 35,134,190-2,194,195, 20$,
122 206
https://doi.org/10.1017/CBO9780511622380
INDEX
Beja, 78, n o Egypt: under the 'Abbisids, 12,19, 24,34,

Black Stone of the Ka'bn, 166-7 35. 36, $2, J3,56,57, 59-61, 77, 78. '43.
Bosworth, C. E., 9; 157; under the Tulunids, 109-13, 126,
Buddhists, 159 133-4; under the Fatimids, 145, 147,
Bughi, 66, 81-2, 84 189, 193-211; taxation of, 24, 34, 35,
Bulghar, 123, 149-51, 170 59-60, 95, 108, 120, 146, 168, 200-2;
Bukhiri, 5,11,25,124,148, 150, 151,178 trade of, 202-3
Bukhara-Khuda, 64 Eunuchs, 138-9
Bust, 183
Buyeh, 160 Fadl b. al-Rabf, 37
Buyids, 129, 140, 152, 158, 159-87, 204, Fa(jl b.Sahl, 42,47, 53
211 Fadl b.Yahya, the Barmakid, 31, 32, 36, 37
Buzbareh, 64 ibnFadlan, 150
Byzantines, 5, 58, i n , 135, 199, I6J, 174, Fana-Khusraw, 183-6
190; wars with, 6, 12, 25, 32, 38, 55, Farghana, n , 63, 68, i n
61-2, 77, 87, 124; trade with, 102, 122, Farighunids of Guzgin, 175
12;; relations with Hamdinids, 169- Firs, 130; under the 'Abbisids, 14, 24,53,
73,183-4; relations with Fitimids, 203- 74,98,100,135; under the Tahiridi, 72,
4, 208-9, * i o - n 77, 87; under the Buyids, 161,162,183;
taxation of, 24, 33, 135
Cahen, CL, 71, 182 Faryab, 11
Canard, M., 188 Fatimids, origins of, 128,129,132,188; in
Cairo, 78, 198, 205-6 North Africa, 153, 167, 189-94; in
Caspian Sea, 13, I J , 31, 67, 122, 150, Egypt, '45.147.187.197-*";»"Syr"
151 172, 204, 210-11; economic policies
Caucasus, 15, 62, 76, 122, 113, 169, 173, of, 188-9, »9". »93» 2 °o-J. *o$, 207-8
186 Fath b. Khiqin, 77
Ceuta, 19] Fayyum, 147, 193
Chad, 102, 106 Fezzan, 106
Chilus, 67, 148 Firdawsi, 178
tbakars, 64, 71, 76-8, 80-1, 82, 85,92 Furat, family of, 94, 113, n o , 121,140-7,
China, 5, 90, 122 I5*~7.193
Christians, status of, 73, 208-9; *** *"" Fusta?, 60, n o , 133, 190, 198
Copts
Coinage, 18, 123 Garama, 102, 130
Copts, 12, 19, 34, 60, 209 Genoa, 192
Crete, 81, 192 gbalU, at mHslagbttllit
Cyprus, i n , 122, 204 Ghana, 106
Ghaznavids, 160, 175,178, 180-6
Damascus, 31, 79, 112, 113, 131, 132-3, Ghaznin, 180, 181, 183
19J, 196, 197, 204 Gharlb al-KMl, 138, 140, 142
Daqiqi, 178 **«/&», 65
Dirfur, 106 Ghuzz, 150
Dasht-i-Kavir, 159 7
Daybul, 98 Gurgin, 12, 26, 28,63,122,136,160,174,
Daylam, 14, 116 178
Daylamites, 31,117,148,151-2,15J, 157, Guzgan 25
IJ8
Dhit al-Qamim, 134 Habasha, t i o
dibqans, 96, 178 Hablb, clan of, 172
Dimyit, 77 Hid!, 'Abbisid ruler, 25, 26-7, 30
Dlvdid, 64 Hakb, jw Aleppo
Druzes, 210 Hallij, 144, 152
Hikim, Fitimid ruler, 206
earthquakes, 79 Hamadin, 87
https://doi.org/10.1017/CBO9780511622380
218 INDEX
Hamdin, tribe of, 26, 101 Isma'ilis, 15, 128-30, 190, 203, 210
Hamdanids, 124-5, 129, 133, 155, 142, Isma'ilism, 179, 188, 197, 198, 199, 203
157, 162, 167, 169-73, 174, 183, 195, Itakh, 66, 69, 73
196 Italian ports, 192, 193
Hamdan Qarmat, 129, 130, 131 Iyad, clan of, 101
Hanafi, law school, 34
Hanbali, law school, 46, 53, 55, 144, 152, Abu Ja'far al-Mansur, sit Mansur
156, 162, 183 Ja'far al-§adiq, 14, 198
Himid, b. 'Abbas, 146, 149, 151 Ja'far b. Yahya, the Barmakid, 32, 37, 39
Abu Hanifa, 57 Abu al-Jahm, 2, 3
Hasan al-Jannabi, 130, 153 Jarjari'i Abu al-Fadl, 72, 75
Hasan b. Makhlad b. Jarrah, 93, 94 Jarrah, family of, 93,121, 132,140-7,152-
Hasan b. Sahl, 42, 44, 45, 46, 47, 50 7, 166
Hasan b. Zayd, 99 Jawhar, 195, 197, 200, 210
Hephthalites, 5, 25, 31, J I , 63 Jayhani, 148, 149, 150
Harthama, b. A'yan, 43, 44 Jerusalem, 205, 208
Jews, IOI, 198, 199
Ibn al-Hawwari, 146 Jidda, 107
Hijiz, 14, 84, 107, '68, 203 Jordan, 79
Hims, 78, 87 Jibal, 76, 81, 84, 86, 161, 173, 184
Hindu-Rush, 178, 180 Jazira, 47, 93, 122, 123; troops from, 4, 6,
Hisn ZiySd, 172 7, 11, 13, 15, 26, 76, 181; taxation, 23;
Husayn, the Man with the Mole, 132 trade of, 121, 125; Hamdanids, in, 142,
157.«7°
Ibrahim, brother of Muhammad the Pure
Soul, 14 Kafur, 196-7, 199
Ibrahim b. Al-Aghlab, 35, 36 Kalb, clans of, 131
Ibrahim b.al-MahdJ, 45 Kalwadhini, 146
Ibrahim b. al-Mudabbir, 108 Kano, 106
Idrisids, 103, 190-2 katib, 11, 22, 23, 50, 68, 81, 93,113,133
Idris b. 'Abdillah, 27 Kaydar, 64
Ifriqiyya, see Tunisia Ibn Khaldun, 10
'Ijl, clan of 101 Khalid b. Barmak, n
Ikhshid, Muhammad b. Tughi, 155, 157, Ibn Khalliqan, 51
167,195-6 Khaqan-Artug, 64
%**. 77 Kharijites, 33,95,96,97,98,103, no, 169,
iqti\ 71, 75, 82, 84; use by 'Abbisids, 94, 190
95, 113, 115, 116, 120, 121, 124, 130, Vuf'XI, 138-9. 2 °6
140, 152, 154. 195; u se ty Buyids, 163- Khazars, I J , 27, 34, 123, 148, I J I ; in
4, 184 'Abbasid service, 62, 63, 66, 117, 138,
Imam, title used by 'Abbisids, 2,8, 46, 47, 142
53, 92, 162; by Shi'ites 127-8, 185; by Khurasan, under the 'Abbasids, 1, 4, 13,
Fatimids, 191,199-200, 204, 210 14, 15, 19,11, 36, 38,42,189; under the
India, 90 Tahirids, 50, 73, 96, 100; under the
'Iriq, 5, 7, 8, 57, 58; under 'Abbisids, 3,4, Samanids, 175,178-82
11, 43-4, 47, 53, 78, 100, 126, 143, 157, Khurasaniyya, 3, 4, 6, 7, 8, 9, 11, 12, 13,
181; soldiers of, 14, 15, 26, 76; Qara- 14, 20, 25, 29, 30, 190
mitain, 131-4, 154, 168; Buyids in, 157, Khumarawayh, 112
159, 161, 162-5, 186, 2 i i Khurramiyya, 34, 56
'Isa b. Musi, 8, 14 Khuttal, 5, 11, 178
Isma'il b. Ahmad, the Simanid, 119 Khuza'a, clan of, 64
Ismi'il b. Bulbul, 113 Khwarizm, 11, 63,123,136,147,149,150
Isfijab, 11, 63 Khwarizm-Shahs, 175
Ishtikhanj, 63 Khuzistin, 161
Ishtikhan, 11, 98 Kiev, 150
https://doi.org/10.1017/CBO9780511622380
INDEX
Kirman, 98, 135, 161, 179, 183 Muhammad b. Wasil, 98, 100
Kish, 25 Muhtadt, 'Abbasid ruler, 85-7
Kufa, 1, 15, 39, 44, 82, 108, 128, 130, 168 Muhtajids of Chaghaniyan, 175, 180
Kurds, 124, 135, 173-4, 184 Mu'izz, Fatimid, ruler, 194-j, 197-8
Kutama, tribe of, 190-2, 206 Mukhtar, 8
Muktafi, 125, 131, 132, 135-7
Lambton, A. K. S., 71 Mu'nis al-Fahl, 138,142
Leo, VI, Emperor of Byzantium, 133 Mu'nis al-Khasiyy, 138,139,141,142,143,
Lewis, Bernard, 188 145. 147.149. '54-6>>95
Libya, 12, 102-3, '3° Muntasir, 'Abbasid ruler, 74,77,80,81
Lu'lu', i u - 1 2 , 113, 133 Ibn al-Muqaffa', 22
Muqtadir, 'Abbasid ruler, 137-56
Madina, 14, 23, 27, 58, 107 Murtada, 'Abbasid ruler, 137
mabdi, 21, 132, 191, 192, 193 Musa b. Bugha, 86, 88, 93
Mahdi, 'Abbasid ruler, 8, 15, 21-6, 27, Musa b. Utamish, 93
28, 34, 67 'Musafirids, 160, 173-4
Mahdiyya, 192 Mus'ab b. Ruzayq, 15
Mahmud of Ghazna, 181-2 Abu Muslim, 2-8, 11, 13, 65
Makka, 6, 39, 44, 58, 107, 142, 143, 167 mustagballat 17, 27, 35, 68
Malatya, 12, 80, 125, 170 Musta'in, 'Abbasid ruler, 80,82,84
Milikite, law school, 35 Mustakfi, 'Abbasid ruler, 158
mamlik, 181 Mustansir, Fatimid ruler, 210
Ma'mun, 'Abbasid ruler, 36, 39, 41, 43, Mu'tadid, 'Abbasid ruler, 113-26, 131,
46, 50, 52-61, 64, 66, 67, 68, 78, 120,
127,206 Mu'tamid, 'Abbasid ruler, 87,92,113
Manbij, 29 Mutanabbi, 196
Maqrizi, 201 Mu'tasim, 'Abbasid ruler, 53, 60, 61-9,
Mansur.1'Abbasid ruler, 8, 9, 10, 12-19, 80, 85, 87, 98, n o
20,21,22,24,26,37,59 Mutawwakil, 'Abbasid ruler, 72-80, 82,
Massisa, 29 83. 85, 91. i*7
Marco Polo, 57 Mu'tazilites, 54-5, 68, 69
Mardln, 122, 124 Mu'tazz 'Abbasid ruler, 74, 83, 84, 86,
Marwan, Umayyad ruler, 4, 6 92
Marzban, b. Turkesh, 64 Muttaqi, 'Abbasid ruler, 158
Marwanids, 174 MuwafTaq, Abu Ahmad, 83-4, 86, 92-5,
Marocco, 103, 127 100, 108, 112, 113-14, 116, 117, 133
mawld, mawdli, 9, 10, n , 22, 31, 64,67,138 maxalim, 21
Mawsil, 33, 34, }6, 87, 135, 169-71, 174,
183-4 naff, 57
Mazyir, 67 Nasibin, 171
Melkite Christians, 205-6 Nasr b. Ahmad, the Samanid, 179
Merv, 4, 11, 42, 47, 127, 178, 181 Nasr b. Shabath al-'Uqayli, 52
Merval-Rud, 51 Nasr al-Qushuri, 138, 140, 142, 143, 149,
mi/ma, 55 >55
Mongols, 160 Nishapur, 13, 147, 148, 178, 179, 182
Mu'ayyad, 'Abbasid ruler, 74, 83 North Africa, n , 12, 27, 31, 35; trade of,
Muhallabites, 101-2 102-3; Fatimids in, 128, 134, 181, 189,
Muhammad, b. Hasan b. Sahl, 101 190, 192
Muhammad b. Abi Khalid, 45 Nubia, 60-1, 78, n o , 194, 196-7, 203, 207
Muhammad b. 'Abdulmalik al-Zayyat, 68,
O i l see naff
Muhammad the Pure Soul, 14, 15, 21, 57 Orestes, Patriarch of Jerusalem, 205,106,
Muhammad b. Sulayman, 133-4,130 208
Muhammad b. 'Ubaydillah b. Yahyi b. Otto the Great, 194
Khaqin, 142 Oxus River, 123, 182
https://doi.org/10.1017/CBO9780511622380
22O INDEX
Palestine, 112, 113, 189, 197, 208 Sanhaja Berbers, 195, 199
Palmyra, 131 Sarakhs, 42
Pars, 136, 137, 142, 148, i j o Sawid, 23, 24, 53, 69, 72, 73, 130
Pechenegs, 150 Scandinavia, 90, 123
Pushang, 15, 25, 43 Scbiiktigin, 180-1, 18 j
Senegal, 106
qabila, 60, 120, 202 Severners see Isma'ilis
Qadarites, 54 Shafi'ite, school of law, 183
Qadi Nu'man, 198, 205, 210 Shakiriyya, see Chakars
qa'id, 65 ShFites, 2,46,94,116,126-9,13 2,15 6,162,
Qahir, Fatimid ruler, IJJ—6 169, 183, 185, 198; relations with
'Abbasids, 8, 21, 46, 53-5, revolts, 14-
Qaliqala, 169
15, 27, 44; in North Africa, 103; in
Qaramita, 128-35, 152-4, 157, 158, 162,
Tabaristan, 99, 100, 119, 148,150, 151
165-8, 196, 204
Shabmmeb, 178
Qarakhanids, 182
Qasim b. 'Ubaydillah b. Sulayman, b. Shiraz, 33, 162
Shir-i-Bamyan, 64
Wahb, 131, 135, 136
Qatar, 167 Sib, 42
Qays, clan of, 101 Sicily, 189, 190, 192-3
Qazvin, 87, 148, 151 Sijilmassa, 191
Qinnasrln, 121 Simjurids, 178
Qumis, 79 Sind, 96, 97, 98
Qumm, 55 Siraf, 98, 107, 153, 166, 167
Qur'an, 54, 55 Slstan, 15, 63, 87, 95, 96, 100, 179, 183
slavery, I O I , 123
Spain, 12, 181, 192, 194
Ridi, 'Abbasid ruler, 156-8 Soghdiana, 4, 5, 38, 42, 64, 96. 98, 175.
Rafi' b. Layth, 38, 40, 42 182
Rifiqa, 15 Sourdel, 51
Raqqa, 15,15, 31,37,44,47.5*.62,125,132 Sudan, 106
Raqqada, 191 sudan, 1 0 9 - n , 113, 130,133-4,196-7. >99
Rashid, Harun, 'Abbasid ruler, 25, 27-40, 204, 205
41, 42, 52, 61, 62, 64, 74 Suhar, 98
Rayy, 13, 14, 39, 43, 55, 124, U 6 . ' 5 ° . Sulayhids, 203
159. 167, 174, 179. l 8 6 « l 8 7 Sulayman al-Jannabi, 153, 166
Rukhkhaj 97 Sulayman b. Kathir, 2
Rusifa, 15 Sulayman b. Wahb, 93
Russia, 90, 122, i ] o, 186 Siil-tegin, 64
Rustamids, 103, 190-2, 193 Sumaysat, 77, 122, 124
Sweden, 122
Sibi, 65 Syria, 4, 47, 53, 74, 79, 102, 120,126,143,
Sa'id b. Makhlad, 95, 113 195; disturbances in, 6, 7, 31, 52, 73;
Saffarids, 79, 87, 95-100, 109, 135 troops from 11,13,14,15,20,26,61,76,
Sahara, 102, 106 181; taxation in, 23, 24, 73; trade of,
Salih b. Wa?if, 85, 86 169, 171; Qaramita in, 128, 134; Fafi-
Sahl, family of, 42, 43 mids in 168, 172, 189, 191, 203-4, 205,
Sa'id b. al-Jannabi, 153 210
Abu Salaraa, 1, 2, 3
Salamiyya, 128, 130, 132 Tabaristan, 67,119,136,160,179; princes
Saljuqs, 160, 174, 175, 186, 211 of, 14, 26, 38, 67; ShiMtes in, 87, 99,
Siminids, 67, 100, 119, 122-4, I35~7i 100, 127, 148-9, 150, 151; trade of, 96,
147-8, 151, 157, 160, 167, 174-82 122,147-8
Samarqand, 38, 148, 178 Tabari, 152
Simarra, 62-3, 67, 68, 71, 73, 74, 76, 77, jabasayn, 159
80,82,93 Tahart, 193
https://doi.org/10.1017/CBO9780511622380
INDEX 221
Tahit b. Husayn, 4 3 , 4 4 , 4 7 , 5 0 , 5 1 , 5 2 , 6 4 , Tyana, 161

109
Tahirid family, 16, 72-5, 75, 80, 87, 95, Ubulla, 98
159; in Baghdad, 62, 69, 82-4; in •Ubaydullah al-Mahdi, 191-4
Khurasan, 67, 68, 69, 76, 79, 96, 99, 'Ubaydullah b.Sulayman b. Wahb, 118
100, 119, 175 'Ubaydullah b. Yahya b. KhSqan, 75-6,
Takrit, 170 77.80,92
Takrur, 206 'Ujayf b. 'Anbasa, 64
Taliqin, 5 'Uman, 58, 98, 153, 166, 167
Tamim, clan of, 101, 131 Umayyads, 9, 22, 24, 34, 98; in Spain 195
Tarsus, 29, 61, m , 122, 124-5, '33. '35. 'Uqaylis, 52, 174
147,169 Ushrusana, 31, 50, 64, 66, 121
Tawana, 61 Ustad-Sis, 15
Taxation: under the 'Abbasids, 16-18, 23- Utimish, 80-1, 82, 86
6, 3*-3. 55.58-9. 73. 9«-2> 94. 119-21, Utrush, the Hasanid, 148
140-1, 143, 146-7; in Iraq, 23-4, 143;
in Syria, 23-4; in Egypt, 24, 34-5, 60; Volga River, 123, 149-51
in thugbur in North Africa, 35; in Wasif, 66, 80, 81-2, 84
Khurasan, 36; Shi'ite attitudes to, 46, Wasif the Armenian, 125
130, i68;'under the Buyids, 179,184-6; waslf, 65
under the Fatimids, 191, 200-3 Wisit, 69, 79, 84, 85, 92, 108, 146, 168
Tayy, tribe of, 131,209 was/fa, too, 206, 210
Thousand and Ont Nights, 30 Withiq, 'Abbasid ruler, 69-71, 72, 107
tbugbir: under the 'Abbasids, 27-32, 47, wasjr, 1, 2, 3, 22, 85, 118, 124, 135, 136,
52, J3,56, 93, 121, 124, 149; under the 137. »4°-i. 143-7. >J*t »5J-7.162, *oo
Tulunids, 77, 112; under the Ham-
danids, 132, 157, 171; Byzantines in, Yahya b. Khalid b. Barmak, 28, 32, 37
172, 203; trade of, 109, i n , 122,143 Yahya, shaykh of Banu al-Asbagh, 132
Tigin al-Khissa, 142 YahyS b. Zakruyeh, 131
Tigin of Khwarizm, 149-50 Yaman, 44, 127, 128, 189, 203
Tirmidh, 5 Ya'qub b. Dawiid, 21-2
Toghj, m Ya'qub b, Killis, 199, 205
Trade, 21, 56-8, 90-1, 96, 148-151, 173, Ya'qub, the Saffiurid, 97-100
178; of Indian Ocean and the Gulf, 17, Yermesh, 66
26, 69, 96, 98, 100, 107-8, 122, 153,
187; of Mediterranean, i n , 171; with Zabulistan, 96, 97
Byzantines, 28-9, i n , 122, 170-1, 204- Zaghawa, tribe of, n o , n i, 113
6, 209; of Africa, 101-7, 110-11, 190, Zahir, Fatimid ruler, 210
"94. '97; with Khazars and Russia, Zakruyeh, 131, 13$
122-3, ' 4 8 . I S°- I > 187; Fatimid policy, Zanadiqa, 22
202-3, 2 °4~6. 1 O 9 Zanj. 9*. 99> 100-2, 107, 108, n o , 112,
Transoxania, 38 126, 1 jo, 133,166-7
Tripoli, 102,106 Zanjan, 87
Tripolitania, 103, 109 Zawaqil.45,46
Tukharistan, 4, 38,42, 117, 175 Zaydls of Tabaristan, 129, 162
Tulunids, 124, 126,131, 133-4,195 Zirids, 199, 207, 209
Tunisia, 12, 35, 36, 103, 189, 190, 198, Zoroastrians, 57
199,207,209 Zuwayla Berbers, 195
Turks, 63-66, 123, 150-1, 182-3, '86 Zubayda, 36
Tus. 178 Zunbil, King of Zibulistan, 96
https://doi.org/10.1017/CBO9780511622380

Islamic History - A New Interpretation Vol.2

Uploaded by

Copyright:

Available Formats

Islamic History - A New Interpretation Vol.2

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Islamic History - A New Interpretation Vol.2

Uploaded by

Copyright:

Available Formats

Cambridge University Press

Published in the United States of America by

© Cambridge University Press 1976

This publication is in copyright. Subject to statutory exception

First published 1976

Reprinted 1981, 1986, 1988, 1990, 1992, 1995

A catalogue recordfor this publication is available from the British Library

Library of Congress Cataloguing in Publication Data

Cambridge University Press has no responsibility for the persistence or

There has been a conspicuous lack of a comprehensive study of the

retinue. At Anbar he was cordially received by the Amir al-Mu'minin

• Vnd., pp. 88-9.

was determined to win. He took all precautions, and used every

tion.' One mawld, whose descendants attained very high positions,

recaptured and its fortifications were rebuilt, but no Khurasaniyya

cause of his cousins although he was one of the most respected

Ruzayq of Pushang whose descendants, the Tahirids, were soon to

wife and three daughters to support; he refused exclaiming that

princes of Tabaristan to declare his submission before Hadi had to

i, H I , p p . 4 9 5 , 5 1 7 - 1 8 , 5 2 1 - 2 , 551. * Bal., Futup, p. 166.

circumstances. The only troubled spots in the empire were the

best course was to move voluntarily to the new 'awdsim. Of course,

'Abdulmalik b. Salih and his Syrian-'Iraqi forces were adequately

Rashid himself, as if to announce his newly acquired strength, led

Meanwhile, this new eastern policy aroused tremendous opposi-

of his chief lieutenants, broke out in Samarqand. This leader was

of the 'Abbasid family, mawdli, government officials and army

administration, was significantly given the title dim al-riyasatayn

abnd' had fallen to Muhammad b. Abi Khalid, who in turn was

x 'Iraq, Jazira, Syria and thugbur

Tab"!, in, pp. 1059, 1045, 1044, 1102; Tayfur, p. 232.

tancously in three areas, the Byzantine front, Egypt and Adhcrbayjan.

complicated conflict. The central government, on its part, had long

stance. The earliest example of such a use recorded in our sources

in lengthy discussions about the definition of the term rikd%, and

to strike roots among the Copts but no consequential adjustments

a delegation to complain to him about Muslim encroachments on

settled there to be withdrawn before he and his army returned to

whelming evidence to the contrary in our sources. This misconcep-

Although we are in a relatively good position to know individual

mountaineers) to combat Babak and his followers who were defend-

astute enough to appoint other Tahirids to the vacant positions and,

it can be estimated that there were roughly 4000 non-Muslim families

As it occurs frequently among the names of military leaders at this

There was some hesitation in Baghdad about accepting the new

physicians in addition to the cost of medicaments. The insignificance

of the regions from that of the central government. Moreover, the

of the early Kharijites were, at this very time, literally millionaires

p. 142. ' Shaban, 'Abbdsid Revolution, pp. 23, 28, 54.

acquired tremendous wealth.' Of course, heresiographers and certain

at the losing end of the bargain with the central government.1

Tabari, m , pp. 1742-2098. • Ibid., p. 1752.

leaders represent a prestigious Arab line who had served the

When the 'Abbasids came to power and the Aghlabids established

•J^VValila Naki / TakrTt

a Trans-Saharan caravan routes

to leave what he considered a fair amount of his own goods in

the men, placing non-Arabs under Arab commanders and vice

co-operation in allowing the Zaghawa to pass through his lands. After