Main The Later Ghaznavids: Splendour and Decay

The Later Ghaznavids: Splendour and Decay

Categories: History
Year: 1992
Publisher: Munshiram Manoharlal
Language: english
Pages: 207
ISBN 13: 978-81-215-0577-2
File: PDF, 12.11 MB

You may be interested in

 

The Ghaznavids

Year: 1992
Language: english
File: PDF, 21.52 MB
 
You can write a book review and share your experiences. Other readers will always be interested in your opinion of the books you've read. Whether you've loved the book or not, if you give your honest and detailed thoughts then people will find new books that are right for them.
1

Strategic Thinking Skills

Year: 2012
Language: english
File: PDF, 2.87 MB
2

Acoma grammar and texts

Year: 1965
Language: english
File: PDF, 17.06 MB
~<'~;"'"SSPLE:NDOUR.AND DECAr

·~·~

-'Nartfiem ...... 1040-1186

This book is a sequel to Prof. Bosworth's
classic study of the origins and early history of
the Ghaznavid empire, The Ghaznavids, their
empire in Afghanistan and eastern Iran 9941040. It carries on the story of this originally
Turkish dynasty, based on Ghazna in eastern
Afghanistan, after its sultans had lost the·
western Iranian provinces of their empire to the
incoming Turkish steppe nomads of the Oghuz
- tribe, whose leaders then formed the Great
Seljuq state in the Middle East. The
Ghaznavids survived, however, as a stillpowerful empire, comprising eastern
Afghanistan, the Panjab, Baluchistan and Sind,
for almost a century and a half. Their court in
Ghazna and then, at a later date, in Lahore, was
a great centre for the beginnings and the
development of what was to be the IndoMuslim culture and literature, for it was from
the Panjab at this time that the gradual process
of the Islamisation of much of northwestern
India began.

Prof. C.E. Bosworth F.B.A. is Emeritus
Professor of Arabic Studies at the University of
Manchester and a former President of the
British Society for Middle Eastern Studies. His
many books cover the fields of the history of
the Iranian world and Central Asia and the
history, literature and culture of the Arab
world.

The Later Ghaznavids: Splendour and Decay

This book is a sequel to Prof. Bosworth's classic study of the origins and early
history of the Ghaznavid empire, The Ghaznavids, Their Empire in Afghanistan
and Eastern Iran 994-1040. It carries on the story of this originally Turkish
dynasty, based on Ghazna in eastern Afghanistan, after its sultans had lost
the western Iranian provinces of their empire to the incoming Turkish steppe
nomads of the Oghuz tribe, whose leaders then formed the Great Seljuq state
in the Middle East. The Ghaznavids survived, however, as a still-powerful
empire, comprising eastern Afghanistan, the Panjab, Baluchistan and Sind,
for almost a century and a half. Their court in Ghazna and then, at a later
date, in Lahore, was a great centre for the beginnings and the development
of what was to be the Indo-Muslim culture and literature, for it was from the
Panjab at this time that the gradual process of the Islamisation of much of
northwestern India began.
Prof. C.E. Bosworth F.B.A. is Emeritus Professor of Arabic Studies at the
University of Manchester and a former President of the British Society for
Middle Eastern Studies. His many books cover the fields of the history of the
Iranian world and Central Asia and the history, literature and culture of the
Arab world.

The Later Ghaznavids:
Splendour and Decay
The Dynasty in Afghanistan and
Northern hlClia 1040-1186

Clifford Edmund Bosworth

Munshiram Manoharlal
Publishers Pvt. Ltd.

ISBN 978-81-215-0577-2
Reprinted 1992,2015
First published 1977

© Munshiram Manoharlal Publishers Pvt. Ltd.
All rights reserved, including those of translation into other languages.
No part of this book may be reproduced, stored in a retrieval system,
or transmitted in any form, or by any means, electronic,
mechanical, photocopying, recording, or otherwise,
without the written permission of the publisher.
PRINTED IN INDIA

Published by Vikram Jainfor
Munshiram Manoharlal Publishers Pvt. Ltd.
PO Box 5715, 54 Rani Jhansi Road, New Delhi 110 055, INDIA

www.mrmlbooks.com

Contents
·--------- --·--------- - - - -

Map and Genealogical Table
bztroductio.! p.I
CHAPTER ONE p.6
Defeat in the \Vest and its Aftermath: The 'Time of Troubles'
1. The shrunken Gha:azavid emjJire
p.G
2.

3·
4·
5·
6.
7.
8.
g.
10.

The last months of Afas'zid's reign and
his retiral to India
The deposition of Afas'iid and
Mul;zammad's second sultanate
Maudzid's vengeance
A1audiid re-establishes the position
in the west
The campaigns in India
The intemalfimctioning of the emj1ire
Succession difficulties and the
accession of'Abd ar-Rashid
The usurpation of Toghri'l
Farrukh-Ziid'ssullarmte

p.~

p.1 i
p.:w
p.25
P<:iO
p. 33

P·3i
p:41
P·47

p.50
The Reign oflbrahim: Retrenchment and Continuity
CHAPTER TWO
1.

2.

3·
4·
5·
6.

7.

Ihriihim and tlze Seljuqs
The Glzaznavid army in tlze later period
Ibriihim' s campaigns in India and in Ghiir
The internal administration of the empire
Court life and culture
Relations with the Abbasid caliplzate
Ibriihim as ruler

p.50
p.58
p.61
p.6g
p. 74
p. 78
p. So

p.82
Mas' iid m and his Sons: Equilibrium and Incipient Decline

CHAPTER THREE
1.

Mas'iid'sreign

p.82

vi

CONTENTS
2. The strugglesfor power amongst Mas'ud's sons
3· Balmim Shiih' s reign: the Indian summer before the
Ghilrid i11vasion

p.8g
p.g8

CHAPTER FOUR p.III
The Struggle with the Ghiirids, and the last Ghaznavid Sultans
1. The Ghii.rid onslaught and Bahriim Shiili' s last years
p. 111
2. Khusrau Shiih and the retreat to India
p.120
3· Khusrau Malik's reign and the end of the dynasty
p.123
APPENDIX A p.132
lbn Baba al~Qashani on the History of the Ghaznavids
1. Introduction
2. Translation
3. Commentary

p.132
p.l34
p.144

APPENDIX B p.156
The Ghaznavid Rulers in Ghazna and India, 366-582/977-1186

Abbreviations p. 158
Notes and References p. 159
Bibliography of the principal works ctJnsulted p.187
Inde.~ p.192

Acknowledgements
It is obviously fitting that Edinburgh University Press,
who published my first book on the Ghaznavids, should
publish its successor volume, and I am accordingly grateful to the Press for undertaking the task in these difficult
times. I must also acknowledge valuable help in elucidating difficult Persian poetry from Mr Hasan Etessami,
Imperial Iranian Vice-Consul in Manchester.·
C.E.B •

•

Tla~ trlen~ •J.tht

9 HA '£NA 'VIb

6.M PI R £
er ~teA uoo
•r
'j• . ,. ..u.,

Ajrnere
RAJASTHAN

eMarwar

Ab1u

Anhilwllra
•

MALWA

fjjain

I, Sebi.iktigin, 366-,87/977-97

I

6,Maudiid
432-?440/1041-?1o4B

8. 'Ali
?440/ ?1048-g

10. Farrukh-Zii.d
443-51/1052-9

11.

Ibrii.him
451-92/1059-99

I

I

u. Mas'iid 111

7· Mas'iid n
?440/ ?1048-g

492-5o8j1ogg-1II5

I
13. Shir-Ziid
508-g/1115-16

14. Malik Arslan
sog-11j1116-I7

IS. Bahrii.m Shii.h
511-?552/1 I 17-?1157

I

16. Khusrau Shah
?552-5/ ?1157--6o
GENEALOGICAL TABLE OF THE GHAZNAVID SULTANS

I

I7. Khusrau Malik
555-82/ I 16o-86

Introduction

After my previops book, The Ghaznavids, their empire in Afghanistan and
eastern Iran 994-1040, was published by the Edinburgh University
Press in 1963, various people who had found the book useful asked
me when I was going to write its sequel, carrying the history of this
Turco-Iranian dynasty up to its demise in the last years of the 6th I
I 2th century. Over a decade elapsed before I was able to turn to this
project, but the present book now represents the fulfilment of this
expressed wish.
A cursory glance shows that the present work, covering some 150
years of history, is perceptibly shorter than the previous one, covering
less than half a century. The answer is, of course, that the sources for
the middle and later Ghaznavids are infinitely sparser than those for
the earlier period, an exposition of which filled. pp. 7-24 of my earlier
book. The triumvirate of authors, 'Utbi, Gardizi and Baihaqi,
provides for the earlier period a remarkably" rich conspectus not only
of military and high political affairs, the res gestae of the sultans and
their commanders, but also of the day-to-day running of tlte
machinery of state and more intimate, private lives of the monarchs;
Baihaqi's Mujalladiit, in particular, are only rivalled by the Tajiirib
al-umam of his older contemporary Miskawaih as a detailed chronicle
of the work of Muslim bureaucrats. Baihaqi carried his history-cumjournal of affairs up to the end of sultan Farrukh-Zad b. Mas'iid's
reign in 451 I 1059, and it is a matter of profound regret that the
richness of detail and the sapient observ11tion, which the ·latter
volumes of the Mujalladiit must have contained, are apparently lost
to us, for the extant part of his work breaks off with sultan Mas'iid's
ill-starred departure for India in Rabi' I 432INovember 1040. To
make matters worse, the final part of Gardizi's more laconic but still
valuable history, the Zain al-akhbiir, which went up to 'Abd arRashid's reign, se. till some time just after 44011049 1 has also

2

INTRODUCTION

disappeared; his narrative breaks off at the victory of Maud iid over
Mubammad and the murderers ofhis·father in Rajah or Shatban/
Marrl:t or April 1041.
Fortunately, something of the lost part of Baihaqi and perhaps of
other lost sources seems to have be.:n preserved in the final, historical
chapter of the Persian author lbn Baba al-Qashani's adab work in
Arabic, the Kitab Ra's mal an-nadim. The closing section of this
chapter deals with the Ghaznavids, carries the account of events up
to the author's own time and the accession of sultan Mastiid Ill b.
Ibrahim at the end of the 5th/ nth century, and is especially
detailed on the dark period of the 440s/ 105os, the 'time of troubles',
when the Ghaznavid state was racked by succession crises and by the
usurpation of the slave commander "Toghril.· The whole text of the
Ra's mal an-nadim has now been critically edited by my former student
Dr M. S. Badawi, but is so far unpublished; it has accordingly
seemed to me worthwhile to give a translation with commentary of
this section on the Ghaznavids. This forms Appendix A of the
present book, pp. 132-55 below.
From the accession of sultan lbrahim 9. Mastiid in 451/1059 onwards, the sources become yet scantier than for the preceding two
decades. Jiizjani's Tabaqat-i Na#ri continues to be of some value,
although his noqces of the successive reigns are fairly brie£ lbn alAthir's Kamil likewise provides useful information, especially for
Ghaznavid-Seljuq relations, of particular significanc;:e.in the reigns
of lbrahim and his grandson Bahram Shah; here too such Seljuq
sources as l;lusaini's Akhbar ad-daula as-iaijilqiyya give supplementary
material. Ibn al-Athir was further aware that the later Ghaznavii:l
sultans continued to fulfil the dynasty's historic mission by raiding
the shrines and palaces of infidel India; but he had great difficulty,
writing as he did in distant Iraq, in getting specific information and,
in particular, details of places and dates. Hence his notice oflbrahim's
Indian campaigns, discussed below in Ch.2, pp.61-3, is inserted
in the events of the year 472/ Io7g-8o, but it is quite uncertain
in which years of this sultan's forty-year reign the campaigns actually
fell, and the geographical location of the events is equally vague.
Later Persian and Indo-Muslim historians, like Mirkhwand in his
RaUt/at aNafa' and Firishta in his Gulshan-i lbrah.imi, largely
utilised such sources as jiizjani and lbn al-Athir for their sections.
on the later Ghaznavids empire, although, as noted below in
Ch. I, p. 33, Firishta has occasional items of information that do not

INTRODUCTION

3

apparently appear in the earlier sources and whose origin is
unknown.
Where the historical chronicles fail, we can only fall back on
ancillary disciplines like archaeology and numismatics and on
literary sources such as the adab literature collections of anecdotes
and poetry. Apart from the valuable work of the Italian Archaeological Mission in Afghanistan in excavating and describing the
palace of Mas•ud 111 at Ghazna, the archaeological evidence on the
later Ghaznavids, whether in Afghanistan or in northern India, is
virtually non-existent. Nor do we have extant such a rich series of
coins for the middle and later Ghaznavid sultans as for the earlier
ones. From Maudiid's reign onwards, minting in the shrunken
Ghaznavid empire was concentrated on Ghazna for Afghanistan and
Lahore for India, whereas under the earlier rulers there had been a
rich variety of provincial mints operating in Afghanistan and
Khurasan. Moreover, certain of the comparatively ephemeral
sultans, such as Mas•ud 11 b. Maudiid, •Aii b. Mas•ud and Shir-Zad
b. Mas•ud 111, either did not reign long enough to mint their own
coins or else coins issued by them have not come down to us.
The literary sources are more illuminating. The anecdote collections, such as Fakhr-i Mudabbir Mubarak Shah's Adiib al-~arb wa-shshaja•a and •Aufi's Jawiimi• al-~ikiiyiit, present stereotypes of sultans
like Ibrahim and Bahram Shah, in their justice and beneficence. The
poetry, however, is of first-rate importance as corroborative material
for the more strictly historical sources. Much of this verse has clearly
been lost, for the names of many poets, together with exiguous specimens of their verses, are known only to us from the Persian and IndoMtislim tadlzkirat ash-shu•arii' literature and the literary anthologies.
But we have reasonably complete diwiins of such great poets as
•uthman Mukhtari, Abii 1-Faraj Riini, Sana'i and Sayyid I;Iasan;
these writers were attached to the court circles of the sultans, and
sometimes accompanied them on their Indian raids, so that their
verses provide details about certain episodes and campaigns other. wise little known or wholly unknown. The value of this poetry for the
historian has been demonstrated by the Indian scholar Gulam
Mustafa Khan in respect of one particular poet, Sayyid I;Iasan
Ghaznavi, in his monograph A history of Baltriim Shiih of Ghaznin, and
I have endeavoured to follow his path with reference to the other
poets and their verses. Much of this poetry is nevertheless difficult,
and more often allusive than specific in its historical references; I feel

4

INTRODUCTION

certain that a native Iranian scholar, thoroughly saturated in the
lore and literature of his own culture, will be able to extract further
items of information.
It remains to attempt a brief estimate of the historical significance
of the period and dynasty under review in this book. We are dealing
with the middle age and decay of the Ghaznavids. The great days of
the dynasty, when it rose to its peak of power under Mabmiid and
Mas•ud, had passed by the middle years of the sth/ IIth' century,
and the ascendant, dominating power in the Iranian east was now
that of another Turkish dynasty, the rulers of the Seljuq family and
their Tiirkmen followers. The Ghaznavids had to abandon Khurasan
and the western half of modern Afghanistan to the Seljuqs, and apart
from occasional outbreaks of irredentist aggression when the Seljuqs
appeared temporarily to be in difficulties, the sultans coming after
Maudiid generally acquiesced in what came to be the status quo, a
state of rough equilibrium between the two empires. India, however,
was left as the special war-ground of the Ghaznavids. The raids of
the middle and later sultans are singularly ill-documented from the
Islamic side, and the allusiveness and chronological vagueness of the
native Indian sources here provide no complementary dimension of
source material; but it is clear that pressure was substantially maintained on the Indian princes, who nevertheless resisted fiercely and
were never really overwhelmed by Ghaznavid arms. Hence, although the temple treasures oflndia continued to be brought back to
Ghazna for the beautification of palaces and gardens there, and
although the flow of bullion continued to keep the economy of the
Ghaznavid empire buoyant and its currency of high quality, there
were no major gains of territory beyond the eastern fringes of the
Panjab and that region of the Ganges-Jumna Doib which is contained today in the western half of Uttar Pradesh. The lasting
successes of Muslim arms in northern India were to be the work of
the Ghiirids and their slave commanders in the late 6th/ 12th and
the 7th/ I 3th centuries.
With the death of Mas•ud III in so8/ I I 15, a perceptible weakening in the fabric of the empire is discernible. Because of the succession
divisions after that sultan's departure, the eastern branch of the
Seljuqs, under the forceful and long-lived sultan Sanjar, was able to
extend its suzerainty over Ghazna, as the protector and helper of the
ultimately successful candidate for the throne, Bahrim Shah. Contemporary chroniclers felt this as a significant event in history, that

INTRODUCTION

5

the ancient and once-mighty Ghaznavid empire, which in its heyday
had absorbed or brought into vassal status so many of the local
dynasties in the eastern Islamic world, should now be subject in its
turn to a newer power. Ghazna was, however, very much on the far
periphery of the Seljuq empire, and Seljuq interference was minimal,
provided that the requisite amount of tribute was paid by Bahram
Shah. The real menace to the later Ghaznavid sultans, and the one
which finally engulted them, came from a local family within
Mghanistan, se. the Shansabani line of chieftains in Ghiir, one of the
most obscure and inaccessible regions of that country. Bahram Shah
endeavoured to exert the control that his forefathers had exercised
over the petty mountain lords in Ghiir at a time when Ghaznavid
resources were sh~inking and when the dynamism of the Shansabanis
was increasing. The resultant clash proved in the end disastrous for
the older dynasty. The last two Ghaznavids, Khusrau Shah and
Khusrau Malik, were forced to abandon Ghazna altogether, and
ruled only in the Panjab; then, once the Ghiirid leader Mu'izz
ad~Din Mul,lammad was ready for the next phase of expansionism
down to the plains of India, the Ghaznavid sultanate was finally
extinguished completely.
Culturally, the court life of the later Ghaznavids and the literary
work of their scholars and poets continued as an extension in the
east of the common Perso-Islamic culture; such poets as 'Uthman
Mukhtari and 'Abd al-Wasi' Jabali moved unhindered around the
courts of eastern Iran, Transoxania and Mghanistan, addressing
their eulogies to Seljuq, Ghaznavid, ~aflarid, Qarakhanid and other
patrons. Ghaznavid India constituted a further focus for Islamic
civilisation and literary activity, and 'Aufi already mentions such a
Persian poet of Lahore as Abii 'Abdallah Riizbih Nakati, panegyrist of sultan Mas'iid I. Aziz Ahmad, in his An intellectual history of
Islam in India, has surmised that the distinctive· Indian style of
Persian poetry, later called the Sabk-i Hindi and conventionally
traced back to the 1oth/ 16th century, began much earlier in
Ghaznavid India, so that a poet like Mas'iid-i Sa'd-i Salman exhibits
two styles, a straightforward 'Khurasanian' one, and a more intellectualised 'Indian' one. Be this as it may, the place of India as an
immensely fertile nurturing-ground for Persian literature clearly
begins in the middle Ghaznavid period.

ONE

Defeat in the West and its Aftermath:
The 'Time of Troubles'

I.

The shrunken Ghaznavid empire
The victory of the Seljuq family and their Tiirkmen followers in
Rama(,ian 431/May 1040 at Dandanqan, in the almost waterless
desert between Sarakhs and Merv, severed at a blow the Persian
provinces of the Ghaznavid empire from the capital Ghazna and the
heartlands, and speedily made the position in Khwarazm of sultan
Mas~iid's ally, the Oghuz ruler Shah Malik b. ~Ali of Jand, untenable. The damage to the fabric of the empire was indeed severe, but
not irreparable. It is true that Mas~iid's nerve now failed, and his
deposition and murder brought about a further temporary crisis for
the remaining parts of the empire; yet this was soon· surmounted,
thanks to the vigour and incisiveness of Mas~iid's son Maudiid.
Maudiid speedily took command of the situation; he established a
defensive bulwark against Chaghri Beg Da'iid and the eastern wing
of the newly-constituted Seljuq empire in Persia, and even made
plans to resume the offensive and recover the lost western territories.
Hence, whilst the Seljuqs inherited the Ghaznavid position in
Khw~razm, western Khurasan and Jibal, as far east as a line bisecting what is now modern Afghanistan and running through Tukharistan southwards to Sistan, the lands of northern and eastern
Afghanistan, plus the Indi~n conquests, remained intact for over a
century until the rise of the Ghiirids.
At the northern end of this Ghaznavid-Seljuq frontier zone, the
Transoxanian principalities ofChaghaniyan and Khuttal eventually
passed to the Seljuqs, but the main city of Tukharistan, se. Balkh,
remained in Ghaznavid hands under Maudiid, as did the important
Oxus crossing-point ofTirmidh (see below) ; it was only towards the
end of Maudiid's sultanate, or conceivably perhaps during the

THE SHRUNKEN GHAZNAVID EMPIRE

7

troubled decade of the I05os, that Tirmidh fell to the Seljuqs, and
the cession of Balkh was only formally recognised in the treaty which
sultan lbrahim b. Mas'iid made with the Seljuqs on his accession in
45I I I059 (see below, Ch.2, p.52 ). In central Afghanistan, Ghiir
seems to have remained as a buffer region between the two empires,
under its local chieftains. Mas'iid and other fugitive princes and
commanders from the field of Dandanqan had been kindly received
in Gharchistan and Ghiir during th~ latter half of Rama<!an 43 I I
first half of June 1040, and it had been at Ribat-i Karvan (modern
Rabat-Kirman in the region between the headwaters of the Heri Rud
and the Helmand River 1 ) that Mas'iid had halted to compose a
message, minimising the extent of his defeat, to the Qarakhanid
Arslan Khan Sulaiman b. Qadir Khan Yiisuf. 2 We then hear little of
Ghiir for the rest of the 5th I I I th century, apart from one episode of
intervention by sultan lbrahim (see below, Ch. 2, p. 69); it remained
in a state of loose vassalage to Ghazna, one which gave full play to
the internecine squabbles and rivalries of its petty chiefs.
The appearance of bands of Seljuq raiders in Sistan shortly before
Mas'iid's death eventually allowed representatives of the ancient local.
line of ~aflarids to throw off the control imposed by Mal;lmiid of
Ghazna in 392-31 I002-3 and to recover a reasonable degree of
self-government as amirs there under Seljuq suzerainty. For the next
century or so, the fortunes of the ~aflarids were largely bound up
with those of the Seljuq royal house, to whom they supplied contingents of the famed Sagzi infantry and at whose side they not infrequently fought in person. The town of Bust at the confluence of
the Helmand River and the Arghandab, though threatened briefly
at the t:nd of Mas'iid's reign, nevertheless remained firmly under
Ghaznavid control. The situation prevailing in what is now Baluchistan is obscure. The districts of Qu~ar, Makran, Walishtan or Sibi
and Kikanan were formally included in Mas'iid's territories. as
detailed in the investiture patent sent from Baghdad by the new
Abbasid caliph al-Qa'im in 42211o3I, and in this same year, Mas'iid
had provided military help for a local claimant to power in Makran,
who thereafter became a Ghaznavid vassal. 3 From subsequent odd
mentions in the sources of places under Ghaznavid control in the
more westerly parts of Baluchistan and in the coastal region of
Makran as far west as Tiz (modem Chabahar), the province seems
to have stayed within the Ghaznavid orbit, separated from the Seljuq
amirate of Kirman by a zone of extremely inhospitable and difficult

8

DEFEAT IN THE WEST AND ITS AFTERMATH

terrain, the haunt of savage and predatory peoples like the Kiifichis
and Baliich:'
We possess a valuable survey of the towns and districts of the
truncated Ghaznavid empire as it was in the second half of the 5th/
I Ith century under lbrahim b. Mas'iid. In an anecdote of Fakhr-i
Mudabbir Mubarak-Shah's Adab al-barb wa-sh-shaja'a, the lands
under the superintendence of the royal treasurer, the Sharif Abii
1-Faraj, foster-brother to the sultan, are enumerated. In the western
sector of the empire are mentioned the lands from the gates of
Ghazna to Tiginabad, Bust, Mastang, Quzdar, Kij or 'T"lz, 5 Makran,
the Garmsir, 6 Narmashir ( ? near Barn, in Kirman), the shores of the
Indian Ocean, Siwistan or Sibi, Daibul, Siiraj (?Broach), 7 Camba·r
and the whole of the adjacent Indian littoral. In the eastern (i.(:.
inland Indian continental) sector are mentioned Aror or Alor and
Bhakkar in Sind, Siwari ( ? Sibi again), Bhattiya or Bhatinda in the
Panjab, 8 Dava ( ?), Gujarbila ( ?), 9 Uchh, Multan, Karor and
Bannii, up to the gates ofGhazna again. 10
The above list shows that once the crisis of the middle years of the
century was surmounted, the Indian conquests ofSebiiktigin and his
successors were firmly held, and with the achievement of a virtual
stalemate with the Seljuqs in the west, the sultans could concentrate
on what might well be regarded as the historic mission of the
Ghaznavid state, se. the extending of Muslim arms into the northwestern parts of India and the laying of foundations for the later
spr\!ad of the Islamic faith there under the Ghiirids and their
epigoni. Hence with the great provincial centres of KhurasanNishapiir, Merv, Herat and Balkh-by then lost to the Seljuqs,
Ghazna itself and Lahore, the seat of administration for the Indian
provinces, became the two main centres of the later Ghaznavid
empire. 11 It may well be that the reduction of the sprawling early
Ghaznavid empire to a more manageable size was a source of
strength rather than of weakness; the sultanate survived in Ghazna
for some I 20 years after the cataclysm of Dandanqan, and in the
Panjab for 20 years further. The sultans were still able to tap the rich
resources of India, in the shape of temple treasures, tribute exacted
from Hindu rulers and slave manpower derived from the subcontinent. The importance of the spirit ofMuslimjihad in this period
should not be underestimated, even if secular motives for the spoliation of India loom more largely in our minds today than in those of
the tracHtional Islamic sources on the Ghaznavid campaigns in

THE LAST MONTHS OF MAS'iin's REIGN

9

India. All those sultans whose reigns were of any length or who were
not distracted by pressing internal problems seem to have led
campaigns into India, although these raids are generally more poorlydocumented than are those of Mal}.miid and Mas'iid in the earlier
part of the sth/ IIth century. The intensity of this spirit of jihiid is
seen in the florescence of the post-Firdausian Persian epic genre in
eastern Afghanistan, the region of Ghazna and Zabulistan; in this,
the poems of authors like Asad-i Tiisi and 'Uthman Mukhtari to a
considerable extent reflect contemporary struggles with the pagans
of India, equated with the enemies of the knights of ancient Iran
(for 'U thman Mukhtari's Shahriyar-niima, see below, Ch. 3, p. Bg).
The spoils of India enabled the later Ghaznavid sultans to maintain
the earlier traditions of their courts as centres of patronage for
scholars and literary men and also to build splendid public buildings
and palaces, such as the palace of Mas'iid Ill b. lbrahim at Ghazna,
recently excavated by Italian archaeologists, in which objects of clear
Indian provenance have beenl"ound (see below, Ch.3, pp.87-g).
2.

The last months of Mas'iid's reign
and his retiral to India
Mas'iid regained Ghazna, with the remnants of his forces, after a
journey through the mountains of Gharchistan and Ghiir to the
headwaters ofthe Heri Rud and thence to the capital, on 7 Shawwal
431/21 June 1040. He was doubtless already meditating the act of
vengeance and execution wrought only in the next month on the
luckless Turkish generals Begtoghdi, Siibashi and 'Ali Daya, whom
he considered to have failed him at Dandanqan, u and it was not
long after he had been back in Ghazna that the cloud of melancholy
and despair which had descended on him became worse and he made
his decision to retire to India, as is described below.
For the moment, however, there was an urgent necessity to establish a firm front, if that were possible, against a feared Seljuq advance
through Tukharistan to the Hindu Kush and the Kabul River
valley, whence an attack on the capital itself would present few
problems. In particular, reinforcements for the bastion ofBalkh had
to be organised in face of Seljuq raids through the surrounding
countryside and in face of a growing lack of confidence in the
Ghaznavid cause amongst the population of Khurasan; a large
proportion of these last, weary of the tramplings and extortions of the

10

DEFEAT IN THE WEST AND ITS AFTERMATH

Ghaznavid armies and of the incursions of the Tiirkmens into oasis
and agricultural land, were in the process of peacefully surrendering
their towns and rural districts to the Seljuqs. It was probably this
disaffected element in the urban population of Balkh, who would
have ultimately tried to come to an accommodation with the incorners, to whom Mas'iid's vizier Al).mad b. 'Abd a~-~amad referred
when he spoke of the large number of 'corrupt persons, evil-wishers
and malevolently-inclined people' within Balkh; at the same time,
the $abib-Barid or Chief of Intelligence of Balkh was writing to
Ghazna about the damaging activities of the 'ayyars swarming in
from the countryside. The strategic importance of Balkh for the
defence of the upper Oxus region and northern Afghanistan was
patent to the sultan and his advisers; as the $iibib-Barid's letter said,
'All Khurasan is bound up with this town, and if our opponents are
able to seize it, all the power and glory will immediately become
theirs'. 13 The governor of Khuttal on the Ghazna vids' behalf- he is
unnamed by Baihaqi, but he may have been the descendant of an
ancient ruling family there, like the Banijiirids or Abii Da'iidids 14 had evacuated the town and had presumably returned to his own
principality on the right bank of the Oxus, which was itself eventually
to come within the Seljuq sphere of influence.
The real organiser of resistance in Balkh against pressure from
Chaghri' Beg Da'iid's Tiirkmens proved to be the $iibib-Barid Abii
1-l:lasan Al).mad b. Mul).ammad 'Anbari, called Amirak Baihaqi.
According to the section on the 'Anbariyyiin family in Ibn Funduq's
local history of the town ofBaihaq, Amirak subsequently held out in
the fortress ofTirmidh for fifteen years, refusing to surrender it to the
Seljuqs; this period of time is clearly exaggerated, since this same
author states that Amirak eventually headed the Diwiin-i Risiilat or
Correspondence Department for Maudiid and 'Abd ar-Rashid, also
serving Farrukh-Ziid as a secretary and dying during that sultan's
reign in Shawwiil 448 f December 1056. That Amirak in fact surrendered Tirmidh in Maudiid's reign, perhaps in the mid-1040s, is
confirmed by the historian of the Seljuqs ~adr ad-Din al-l:lusaini.
This author relates that, after the failure of Maudiid's expedition to
Khurasan of 435/1043-4, Chaghri: Beg and his son Alp Arslan
appeared before the fortress of Tirmidh and demonstrated to the
kiitwiil or castellan·Amirak the hopelessness of his position and the
unlikelihood that the Ghaznavids would ever be able to afford him
any relief. Amirak therefore accepted the offer of an honourable

THE LAST MONTHS 01' MAs'{in's REIGN

11

surrender, made over his estates at Baihaq to Chaghri Beg's vizier
Abii 'Ali b. Shadhan, and departed for Ghazna. 15
Ibn al-Athir refers to the beleaguered position of Balkh after
Mas'iid's retreat from Dandanqan, but is confused over the details.
He states that Altuntaq I;Iajib (the Altuntash of Baihaqi) was the
Ghaznavid governor in Balkh at this time and was accordingly
besieged in the town when Chaghri Beg advanced on it at the same
time as his kinsman Blghu or Paighu 16 was attacking Herat. It was
the army under Maudiid and the vizier Al}.mad b. 'Abd a~-~amad
that the sultan then sent to relieve Altuntaq in Balkh in Rabi' I 432/
November-December 1040 (two months later than Maudiid's force
actually left Ghazna for Tukharistan, according to Baihaqi's detailed,
day-by-day chronology); but the vanguard of Maudiid's army, so
this account goes, was worsted by Chaghri Beg's forces, so that
Maudiid had to retreat and AltunHiq had willy-nilly to surrender
Balkh to the Seljuqs. 17
A much more detailed sequence of events is given by Baihaqi.
Either at the end of Shawwal or early in Dhii 1-Qa'da 43 I fend of
July or early August 1040, Mas'iid despatched a force of 1,000
cavalry to Balkh under the I;Iajib Altuntash, promising that a larger
army would follow on its heels and that he would then come personally to organise the defence of that region. The demoralisation and
indiscipline characteristic of the Ghaznavid troops at Dandanqan
was still in evidence; after leaving Baghlan, Altuntash's troops gave
themselves up to plundering the countryside there, with the result
that the wretched populace fled to the Seljuqs and warned Chaghri
Beg of the enemy's approach. The Seljuq amir was consequently able
to lure the Ghaznavid soldiers into an ambush, out of which Altuntash escaped to Balkh with only 200 men. 18
. Hence in Mul}.arram 432/September I040 Mas'iid had, as a
pressing obligation, to prepare a more powerful expeditionary force
to retrieve the position in northern Mghanistan. The commander of
this force was Mas'iid's son Maudiid, who had recently distinguished
himself on the battlefield at Dandanqan, where he had ridden round,
sword in hand, trying vainly to inspirit and rally the flagging
Ghaznavid soldiers. 19 To accompany him, and to provide weighty
military experience, Mas'iid detailed the Turkish generals Ertigin,
commander of the palace ghuHims or slave troops, and the l;liijib-i
Buzurg or Supreme Commander Badr (both of whom had j!lst
respectively acquired these elevated positions after the dismissal and

12

DEFEAT IN THE WEST AND ITS AFTERMATH

disgrace of Begtoghdi and Siibashi), and also appointed his own
vizier Al;tmad b. 'Abd a!ll-~amad as Maudiid's kadkhuda or adjutant.
But symptomatic of the declining faith of the sultan's servants in their
master, whose obstinacy and capriciousness seemed to have affected
the balance of his judgment, was the fact that Al;lmad insisted on
obtaining from Mas'iid a muwa~a'a or formal contract of service for
this expedition, wherein the vizier's position and rights vis-a-vis the
Diwan-i 'Ar~ or Department of the Army were carefully defined, and
·wherein his duties and responsibilities were unequivocally laid
· down. 20
A powerful and well-equipped force of elite soldiers was now
assembled in Ghazna. The sultan held a splendid farewell feast in the
Firiizi Garden, and the troops were reviewed on the greensward of
the 'Golden Field'. First came prince Maudiid's personal force of
200 palace ghuliims, armed with breastplates and spears, and with
numerous horses, which were to be led to the scene of battle and then
used for fighting in the actual encounter (the technical term for
such a horse beingjanibat), and swift riding camels; this force bore
the prince's ceremonial parasol ( chatr) and ample standards. Mter
these came a body of infantry, again with flowing standards, and a
group of I 70 ghulams, heavily-armed and with richly-caparisoned
horses. The J:Iajib Ertigin had ove( 8o of his own personal ghuliims,
after whom came another body of palace ghuliims and twenty senior
officers (sarhangs ), the greater part of them splendidly uniformed and
again with horses for conveying them to the battlefield, plus riding
camels. Finally, there came a further group of sarhangs. The total
force of cavalrymen, according to Gardizi, amounted to 4,000. The
original arrangement was for the sultan to follow closely behind with
the main Ghaznavid army, but by this time Mas'iid was already
meditating his move to India.21
Maudiid's army set off northwards for the Hindu Kush passes in
the middle ofMul;larram 432/later September 1040, 22 and then encamped at a place *Hibiin or *Hupyiin, which was evidently of some
note, although it has not so far been identified with certainty. The
name occurs several times in Baihaqi and Gardizi in connection
with Maudiid's expedition, and also in lbn Biiba's Kitab Ra's mal
an-nadim in connection with the events leading to 'Abd ar-Rashid's
killing (see below, p. I4I); but the consonant ductus l)f the name
varies, as does the dotting. 'Abd al-J:Iayy l;labibi, the most. recent
editor of Gardizi, has preferred *Hupyan, on the grounds that a

THE LAST MONTHS OF MAs•uo's REIGN

13

village of this name still exists in the Parwan area; and certainly
Baihaqi in one place links Parwan and *Hupyan together as the
initial goals of the expedition. 23 The general strategy envisaged was
that the Ghaznavid forces at Balkh, comprising the original garrison
plus the survivors of Altuntash's troops, would march out of Balkh
and unite with Maudiid's army, and the combined force would then
clear the Tiirkmens out of Tukharistan. These designs proved abortive, since the sultan altered his original plan to go to Balkh and
decided to leave for India, where he was deposed and murdered; the
news of these last events reached Maudiid when he was still encamped at *Hupyan. 24
At this time of general crisis for the Ghaznavid state, there was a
further need for Mas•ud to attend to certain spasms of unrest in the
heartlands of the empire and along the inner frontiers of the realm,
which had been for some time neglected through the preoccupation
with events in Khurasan. Hence within days of reaching Ghazna
from Dandanqan, Mas•ud in Shawwal431 /early July 1040 sent out
to Bust his former palace slave Niishtigin Naubati, with instructions
to hold that region firmly; the sultan realised that Tiirkmen raids
into Sistan or perhaps beyond were almost inevitable now that they
held western Khurasan. Shortly afterwards, it transpired that a
rebellious Sa~ib-Barid of the Ghaznavids, one Bii 1-Fa<;ll Kurniki, 25
who had been allegedly in treacherous correspondence with the
Seljuqs, had escaped to the region of Bust, and at the end of Dhu
1-Qa•da 431 /early August 1040 the head of the Diwan-i Risalat,
Abii Sahl Zauzani, was sent to Bust in pursuit ofBii 1-Fa<;ll. Baihaqi
observes at this juncture that, although Abii Sahl Zauzani had
aroused the sultan's wrath by allowing Bii 1-Fa<;ll Kurniki to escape,
if he had not been away at Bust he would have accompanied Mas•ud's
column to India and have been caught up in the mutiny of the army
at Marikala; moreover, Mu):lammad would certainly have had his
old enemy Abii Sahl put to death as his very first act of vengeance. 26
Soon after Abii Sahl was sent on this mission, in Mu):larram 432/
September 1040 an expedition was despatched against the rebellious
Afghans in what Gardizi calls 'the foothills of the mountains adjacent
to Ghazna', se. of the mountains of the modern Afghan Gardiz and
Pakhtiya provinces running eastwards to the Pakistan border;
whereas Gardizi describes these dissidents as 'Afghans', Baihaqi
interestingly calls them 'Khalaj'. 27 The leader of this expedition,
the kotwal or garrison commander of Ghazna, Abii •Aii returned

14 DEFEAT IN THE WEST AND ITS AFTERMATH
victorious from this punitive expedition in Rabi' I 432/ November
I040. 28
The most momentous decision made by Mas'ud in these last
months of his reign was that of the move to India in order, as he proclaimed, to spend the coming winter in the Indus valley fortresses of
Waihind (Sanskrit Udabhanda, modern Hund, on the Indus banks
15 miles north of Atak or Attock), M.r.manara (probably the
Ma'bar Mahanara 'Ford of [the villag~ of] Mah~ara' mentioned
by Biruni in his India as a ford across the Kabul River just above its
confluence with the Indus), Peshawar, and Girl (probably ShahbazGiri or Kapur-da-Giri 40 miles north-east of Peshawar, a place of
great antiquity on the ancient Kabul-India route) . 19 The vizier
Abmad b. 'Abd a11-$amad divined what was. really in the sultan's
mind at the same time as the latter was proclaiming his intention to
follow Maudud to Balkh, but the decision was announced to the
sultan's advisers at the end of Mul,arram-beginning of $afar 432/
October I040. The northern provinces of the empire were in effect
to be written off, and Balkh and Tukharistan ceded to the Qarakhanid Boritigin lbrahim b. Na11r, son of the Ilig Na~r (d. 403/
IOI2-I3) who had been Mabmud ofGbazna's rival over the partition of the Samanid lands, and the later Tamghach or Tabghach
Khan (d. 462/Io68). 30 Boritigin had been consolidating his position
in the mountains' north of the upper Oxus valleys and using such
fierce mountain peoples there as the Kumijis in order to harry the
valleys of Khuttal and Chaghaniyan~ it was Mas'ftd's hope that,
through the cession to him of territories on the south bank of the
Oxus, Boritigin would be set against the Seljuqs. Although an envoy
had been sent to the other leading Qarakhanid prince Arslan Khan
Sulaiman as soon as the sultan had reg;lined Gha.zna, Mas'ud knew
that the Seljuqs would lose no time in informing the Khan of the real
magnitude of their victory at Dandanqan and that he could consequently expect little direct military or diplomatic help from that
quarter. 31
A force of 2,ooo cavalry was sent around this time (the exact date
is not recorded) under the prince Majdiid b. Mas'iid to secure
Multan in the middle Indus valley, which we know to have been
chronically disaffected under Ghaznavid control, and which had a
substantial Isma'ili population that was shortly to break out in
rebellion during Maudiid's sultanate (see below, p.gi ).U·That the
tentative plan of a transfer to India had been maturing in the sultan's

THE LAST MONTHS OF MAS'iio's REIGN

I5

mind for some time seems proven by the fact that on I ~afar 432/ I I
October I040 the prince lzad-Yar b. Mas'iid arrived back from the
fortress called by Baihaqi Naghar and by Gardizi Barghund, which
was clearly not too far from.Ghazna. 33 lzad-Yar brought back with
him sultan Mas'iid's deposed brother Mul).ammad b. Mahmiid
(exactly when Mul).ammad had been transferred to Naghar fBarghund from his earlier imprisonment in the fortress of Mandish in
Ghiir is unknown) and Mul).ammad's four sons Al).mad, 'Abd arRal}.man (or 'Abd ar Ral).im), 'Umar and 'Uthman. All these
former captives. were now accorded a warm welcome at court.
Mul).ammad was awarded the mukhiitaba or form of address of 'The
Exalted Amir, Brother', 34 and his sons were given robes of honour
and presents of I ,ooo dinars each; the eldest, Al).mad, was married
to the princess l;lurra-yi Gauhar. In return, Mul).ammad's sons had
each to take oaths of allegiance (aiman al-bai'a), verbal and written,
to Mas'iid. It seems that in this time of vulnerability for the empire
during the aftermath of the Khurasan disasters, Mas'iid was hereby
endeavouring to conciliate the dispossessed branch of his family and
to restore dynastic solidarity, although it would in any case have been
dangerous to leave Mul).ammad and his sons in Afghanistan, as
possible rallying-points for disaffection, whilst he himself retired to
India. 35
The sultan's ministers, led by the vizier, protested that the situation
in Afghanistan was not so desperate as to warrant the abandonment
of the original Ghaznavid heartland. A strong military force could
secure Balkh and Tukharistan against the Seljuqs. It was unwise .to
assume that the Indians had any affection for the dynasty and would
provide a safe haven for them. Nor could the Ghaznavids' slave
soldiers, probably still demoralised and disgruntled after Dandanqan
and perhaps even fearing for the continuance of the dynasty's power,
be trusted with the safe conveyance of the Ghaznavid treasuries and
possessions to India. In the words of Al).mad b. 'Abd a~-~amad,
If my master decides on this transfer to India simply because the
enemy are fighting at the gates ofBalkh, this enemy nevertheless
has insufficient strength to reduce the town, since our defending
force there is so superior to them in martial ardour that they are
making sorties from the town and engaging the Tiirkmens. If
my master will only give the order for his servants to go forth
and clear our opponent~ from these regions, what need is there
to depart for India? It is better to spend this winter in Ghazni,

I6

DEFEAT IN THE WEST AND ITS AFTERMATH

since the position here is quite secure, praise be to God. On the
other hand, it is certain that if our master leaves for India and
transfers all his family and treasures thither, and the news becomes generally known to friend and foe alike, the glory of this
illustrious house will be wholly dissipated, to the extent that
every enemy will become greedy for a share in it. Nor should
any reliance be placed on the Indians in transporting so much
of your family and treasures to their land, for we ourselves have
not acted all that well towards the Indians. Furthermore, what
confidence can one have in the slave troops, to whom the
treasuries will have to be entrusted on the journey ~hrough the
open country?
Adding further weight to these arguments, the kotwiil Abii 'Ali
expressed the view of the military that, even if Ghazna itself were
threatened, it would be safer to guard the state treasuries and the
royal family in the fortresses of Afghanistan than to send them on the
uncertain journey to the plains oflndia. 36
.
But such sound advice was of no avail, and the sultan's self-will,
the istibdiid so often denounced among _themselves by his advisers,
would not allow him to be swayed by reason. His melancholia
included a fatalistic despair that the position in the west could ever
be retrieved. W~en the commander of the ghiizis in the Ghaznavid
army, the seasoned general 'AbdalHih Qaratigin, had offered to
raise in India a large army of cavalry and infantry and to bring it
back for offensive operations in Khurasan, Mas'iid had condemned
the plan as pointless, since it had been fore-ordained that 'We rose
to power at Merv [alluding to Mal;lmiid's victory over the Siimiinids
in Khurasan], and the power has gone from ·us at Merv'. He now
gave his ministers and officials in Ghazna formal· permission to enter
the service of the Seljuqs when they should arrive, noting that Abii
l-Qiisiin Kathir, for instance, had money enough for purchasing the
office of vizier, and Abii Sahl }:lamdiini enough for the office of
'Ari4 or Head of the Department of the Army under the putative
new regime. 37
By now, treasures and precious possessions from such outlying
fortresses of central and eastern Afghanistan as Didi-Rii ( ?),
Mandish in Ghiir, Niiy-Liimiin in Wajiristiin (the later place of imprisonment of the poet Mas'iid-i Sa'd-i Salmiin, see below, p.66),
Maranj (also known as a castle where Mas'iid-i Sa'd was once
incarcerated), and one other fortress whose name is not comprehen-

THE DEPOSITION OF

MAs•un

17

sible from Gardizi's text, 38 had been concentrated on Ghazna. Four
days after the failure of the sultan's advisers to dissuade him from his
plan, early in Rabr 1 432 /November 1040, all the stores of precious
metals, ornaments, fine clothes, etc. (detailed by l:fusaini as comprising 3,000 loads of ~IShiipiiri, Heriiti, Maghribi and Ma}.l.miidi
coinage, various kinds of bullion, jewels, precious vessels, etc.), together with members of the sultan's baram, were loaded on camels and
the whole assemblage departed for India. Also in the column were
Mu}.lammad's four sons and MuQ.ammad himself, just brought back
from Barghund, according to Gardizi; Baihaqi says that he was
intially kept in the citadel of Ghazna under the care of the Amir-i
lfaras or Commander of the Guard Sangiiy. 39
At this point of time, Baihaqi closes the ninth volume of his
Mujalladiit and interrupts his continuous narrative of happenings at
the Ghaznavid court to begin his tenth volume with accounts of
events in Khwiirazm under Ghaznavid rule and of events in Ray and
Jibiil during Mas•ud's reign; unfortunately, the part beyond the
history of Khwiirazm is no longer extant. We are now accordingly
dependent on Gardizi as the sole contemporary source, supplemented
by the quite detailed accounts ofMas•ud's deposition, MuQ.ammad's
brief second. sultanate and Maudiid's vengeance in lbn Biibii and
lbn al-Athir, and the more cursory mentions in l:fusaini andJiizjani.

3·
The deposition of Mas•UtJ and Muhammad's

second sultanate
Mas•ud's force presumably made its way across the mountains from
Ghazna, probably via Peshawar, to the Indus banks. The first
section of what must have been a lengthy column crossed the river,
with Mas•ud in the van, when the remaining part of the army, led by
the Turkish eunuch commander Aniishtigin Balkhi and a group of
the palace ghuliilns, mutinied and plundered the royal treasuries.
The rebels then set up MuQ.amrnad as sultan during the night of
I 3 Rabi' 11 432 / 2o-2 I December I 040, although only after MuQ.ammad had been threatened and possibly even forced physically to cooperate (according to Mirkhwiind, he was even threatened with
death) ; his sons, at least, had of course given their solemn oath of
allegiance to Mas•ud only a short time previously. 40 Using hindsight~ it was obviously an unwise decision of Mas•ud's to leave
Muhammad and his sons with the main body of the army whilst he

18

DEFEAT IN THE WEST AND ITS AFTERMATH

went on in front and crossed the river with a smaller force of soldiers.
He clearly misjudged the temper and morale of his army at this time,
already seen in his over-reaction to the defeat in Khurasan. The
soldiers had not only lost confidence in Mas'iid's powers as a success,ful war-leader in the field, but also in his overall judgment, and not
even the prospect of ghazw against the infidels of India and the
possibilities of plunder there could persuade them to retain their
allegiance. With the sultan's virtual abandonment of Ghaznawhatever excuses he might adduce of merely going to India in order
to collect troops for a revanche in Khurasan-it must have seemed
to the soldiers that the once-mighty Ghaznavid empire had broken
up. In effect, a sauve-qui-peut followed, for the rebellious troops cannot
have seen in Mubammad, whose tastes were predominantly literary
and studious 41 and who had already failed once in his bid for the
sultanate in 42 I I 1030, a military saviour who would restore the
empire to its ancient glory. Mubammad was therefore nothing but a
figurehead, raised to the throne in an attempt to give respectability
and, it was hoped, ultimate legitimacy through military success, to
the rebellion.
That Mubammad could never have been more than the puppet of
ambitious generals and other self-seekers would certainly have been
the case if the reports in certain sources of his blindness were true.
However, this question is obscure. There is no mention in the contemporary sources (Baihaqi, Gardizi), nor in the accounts of
I:Iusaini, lbn al-Athir or Shabankiira'i in his Majm4' al-ansiib that
Mas'iid had Mubammad blinded immediately after his deposition.
Indeed, there is evidence from Baihaqi that · Mubammad was
perfectly able to see when Mas'iid consigned him to captivity in the
fortress of Mandish in 42 I I I030 ; 42 he read the letter written by
Mas'iid to him in his own hand, notifying him of the fate of the
treacherous Turkish general 'Ali Qarib or Khwishawand, and he is
described as seeing from afar the· arrival of the swift camel bearing
this letter from the sultan in Herat. 43 Nor. does the contemporary
Gardizi state specifically that Mubammad was blind when he was
raised to the throne a second time; the mention of Mubammad as
having been blinded ( masmill, maklzul) comes only in such later
sources as lbn Baba, I:Iusaini, lbn Funduq, lbn al-Athir, Jiizjani,
Shabankara'i and Firishta. The information that Mubammad was
blind should accordingly be treated with~ certain amount of reserve,
though it is possibl~ that Mubammad had gone blind during his ten

THE DEPOSITION OF MASfUD

19

years' period of captivity, and one could thereby harmonise the
saurces. 44
,Jbn Baba says that Mas'ud crossed the Indus at Waihind just before the troops mutinied. 45 Mas'ud and the troops still loyal to him
took refuge in the ribal or fortress of Marikala, modern Marigala,
situated in a pass of the low hills between Attock and Rawalpindi, a
few miles to the east of J:iasan Abdal; according to Raverty, these
hills were notoriously full of robbers and brigands, whence apparently a folk-etymology marri-kala 'fo.rtress to protect travellers'. 46
Fighting took place between the besieged and the attacking rebels,
and the superiority was clearly. with the latter. lbn al-Athir's
account has the anecdotal touch here that Mas'ud surrendered himself voluntarily on his mother's advice; Gardizi says that the besiegers, with their troops and elephants, broke into Marikala,
fetched out Mas'ud and bound him. Mul}.ammad enjoined good
treatment for his brother, and in the middle of Rabi' n 432 f late
December 1040 conveyed him and his wife Sara Khatun, daughter
of the Qarakhanid Qadir Khan Yusuf, to the fortress of his own
choice, that of Girl, frequently mentioned by Baihaqi as one of the
principal Ghaznavid strongholds of northwestern India. 47 There for
the moment Mas'ud remained, lamenting, according to Ibn alAthir, the contrast between his former state and his present one. 48
Actual power during Mul}.ammad's short second sultanate was
largely in the hands of his sons, and above all, in those of Al}.mad,
whose behaviour is described in the later sources (though not in
Gardizi or Ibn Baba) as unbalanced ( J:iusaini and lbn al-Ath}r,
'reckless and unbalanced' and even ma'tilh 'mad'); of course, these
later sources may well have been influenced by the picture subsequently formed of Mas'ud as the martyr-sultan. Al}.mad was
backed by AnU.Shtigin Balkhi and other leaders whose families had
suffered discrimination or disfavour during Masfud's reign and were
now able to taste the sweetness of revenge. These included Sulaiman
b. YU.Sufb. Sebiiktigin, whose father had been removed from power
by Masfud in 422/1031 and imprisoned till he died, and the son of
the general fAJi b. 11 Arslan, called Qarib or Khwishawand, whose
father had been initially a prominent supporter of Mul}.ammad's
during the latter's first sultanate, had then betrayed his master, but
had been very soon jailed and killed for his treachery by Mas'ud. 49
That Masfud had already drawn upon himself lasting hatreds by
his vendetta against the Ma/.llnildiyan or Pidariyan, the leading figures

20

DEFEAT IN THE w·EST AND ITS AFTERMATH

of his father's reign whom Mas'ud wished to remove from positions
of influence in the state, had already been shown by the desertion to
the Seljuqs during the Dandanqan campaign of former ghulams of
Yusuf, of'Ali Qarib and of the two other Turkish generals ruined by
Mas'ud, se. Eryaruq and Asightigin Ghazi. 50
The deposed Mas'ud remained at Giri for about a month, and
then was killed at Al}.mad b. Mul}.ammad's instigation, either unbeknown to Mul}.ammad himself after a forged execution order had
been sent to the custodian of the fortress (Gardizi) or after Al}.mad
had persuaded his father to agree to the deed (l:lusaini and Ibn alAthir, both of whose accounts have many anecdotal touches).
According to these latter two sources, Mas'ud was either killed and
then his body thrown into a well which was then sealed up, or else
thrown into the well alive and buried there; according to lbn Baba,
his head was simply chopped off. The date of Mas'iid's death is
given by Gardizi as 11 J umada I 432/ I 7 January I04I (erroneously
as $afar 433/0ctober I04I in Ibn Baba, see below, Appendix A,
p. I40 ). Gardizi's date was apparently also that given by Baihaqi
in the lost part of his Mzgalladiit, on the evidence of the marginal
gloss in one of the manuscripts oflbn Baba's Kitiib Ra's miil an-nadim,
who must have had access to Baihaqi's work (see further, below,
p. 22). J uzjani states that Mas':iid was forty-five when he died. 51
Mul}.ammad subsequently wrote to Maudud in Tukhiristan that his
father had been killed as an act of private vengeance by the sons of
Mas'ud's former commandeJ:-in-chief in India, Al}.mad Inaltigin,
who had unsuccessfully r~e.Ued there against the sultan in 421/
1033·51

4·
Maudild's v_engeance
Whoever may have been the prime mover in Mas'ud's sla~g, the
fact remained that he was dead and that his son Maudiid had at his
disposal a sizeable military force in northern Afghanistan; he was
accordingly bound to constitute himself his father's avenger and the
punisher of those who had broken the oath of fealty to Mas'ud so
recently taken.
From sporadic mentions in the pages of Baihaqi, it is possible .to
piece together something of Abii 1-Fatl}. Maudud's career before he
gained the throne in 432/ I 041. He was indeed closely associated
with his father in various military enterprises and was entrusted with

MAt:DUD S VESGEANCE

21

several responsible tasks. Thus he accompanied Mas'iid on the
expedition which left Ghazna at the end of Shawwal 425/ midSeptember 1034 and went via Bust, Herat and northern Khurasan
to Gurgan and Tabaristan, and during the course of the fighting in
Gurgan, he commanded a detachment of 4,ooo cavalrymen. 53 At
the Mihrgan festivities at the end of 427/autumn 1036, it was
decided to send Maudiid and the general 'Ali Daya with a strong
army to Balkh, after disturbing reports of Tiirkmen activities in the
Ray area had come in; the two commanders did not return to
Ghazna till the middle of J umadii 11 429/ end of March 1o38Y His
name was put forward by the sultan in Mul;larram 430 f October
1038 as the possible commander of a powerful force destined for
Khuttal, where the Qarakhanid Boritigin (see above, p. 14) was
laying waste the upper Oxus valleys to such an extent that Mas'iid
described his depredations as worse than those of the Tiirkmens; but
the prince's name was withdrawn, on the advice of the vizier Aiimad
b. 'Abd a~-Samad, who was in any case opposed to the diversion of
such great resources to Khuttal when the Tiirkmen menace in
Khurasan was so pressing. 5 5 As noted above, p. 11, 1viaudiid fought
valiantly but vainly at Dandanqan.
As the eldest son, Maudiid was always one ofMas'iid's favourites;
hence he is named with his brothers Majdiid, 'Abd ar-Razzaq and
Sa'id as enjoying a specially-favoured closeness with Mas'iid at the·
Shabahar army review festivities of 428/1037· When the sultan's
official heir and favourite son, the amir Sa 'id (unless this is not a
personal name at all, but a designation, Amir-i Sa'id 'the Fortunate
Amir', on the lines of former Samanid practice? 56 }, died in Rabi' 1
430/December 1038, Maudiid was made wali 'ahd or heir in his
place, regaining the position which he had held at the beginning of
Mas'iid's reign but had apparently lost at some point subsequent to
then. 5 7
"Vhat we lack is any clear indication ofMaudiid's date ofbirth and
therefore age, though Baihaqi states that he was the eldest son (see
above, n. 56). There is in Baihaqi considerable information about the
protracted negotiations over Maudiid's projected marriage to a
Qarakhanid princess, part of 1\.f.as'iid's grand strategy in securing
Qadi'r Khan YiisufofKashghar and Khotan as an ally and dividing
him from his brother and rival in Transoxania (and also enemy of
the Ghaznavids}, 'Ali b. l;lasan or Hariin Bughra Khan, known as
'Alitigin. 58 As early as 422/1031 negotiations were begun for the

22

DEFEAT IN THE WEST AND ITS AFTERMATH

marriage of Mas'iid himself with the Khan's daughter Shah Khatiin
and of Maudiid with the daughter of the Khan's eldest son and heir
Bughratigin Sulaiman (the later Arslan Khan, frequently mentioned
in the later pages of Baihaqi, see above, p. 7 ). Matters dragged on
for a long time, and Maudiid's intended bride eventually died en
route for Ghazna in 42511034. 59 Children could of course be
betrothed before puberty and married at an early age, but Maudiid's
military charges during the course of his father's reign indicate that
he had reached adulthood by the early years of Mas'iid's sultanate.
Accordingly, we should probably accept the statement· of Jiizjani
that Maudiid was thirty-nine when he died, that is he was born in
401 l1oio-11 or 402 I 1011-12, rather than that of Ibn al-Athir that
he died at the age of twenty-nine only. 60
The news of all these events reached Maudiid in Tukharistan. All
thoughts of reinforcing the Ghaznavid forces at.Balkh and of undertaking operations against the Seljuqs had for the moment to be
abandoned. He first of all returned to the base of *Hupyan in the
Hindu Kush, and then, on Al).mad b. 'Abd a~-$amad's advice,
crossed the mountains swiftly to secure Ghazna; it could be expected
that Mul).ammad and the rebel army would march on the capital
once the finer spring weather came round. Maudiid was accorded
an enthusiastic reception in Ghazna, hence he spent the latter part
of the winter (mid-432learly 1041) there, holding the requisite
ceremonies of mourning for his father (whose designation henceforth
was invariably to be that of the Amir-i Shahid 'Martyr-King') and
assembling his forces. 61
The opposing armies did not therefore clash in battle until the
spring, the date of the actual battle being given as 13 Raj ab 432 I I 9
March 1041 by Ibn Baba, and 3 Sha'banl8 April (a discrepancy of
20 days) by I;lusaini and Ibn al-Athir. The two surviving manuscripts of Gardizi's ,:Zain al-akhbiir break off immediately at the end of
the historian's description of the battle and of Maudiid's vengeance
on his father's murderer's, and doubtless his noting of the date of the
battle came just at the beginning of the lost part. In so far as Ibn
Baba seems to have had access to the lost sections of Baihaqi's
Mujalladiit, more credence should perhaps be attached to the earlier
date. 62 Mubammad's forces had established themselves in vicinity of
Peshawar for the winter, but Mul).ammad was helpless in the face of
their indiscipline and of their excesses. Ibn al-Athir records in this
connection that

MAUDUD'S VENGEANCE

23

Mul}.ammad's army raised all sorts of demands against him, and
he lost all kingly authority. They made tyrannical confiscations
of the people's property and plundered it, so that the land became ruined and its inhabitants fled, this above all at the town
of Peshawar, whose populace was massacred and their goods
despoiled. A slave was sold there for a mere dirham, whereas
this same sum bought a man of wine.
He then goes on to record that Mul}.ammad's forces left Peshawar on
28 Rajah /3 April, according with his later date for the battle with
Maudiid's army. u
The most detailed account of the battle is given by Gardizi, although it is only the much later source of Jiizjani who mentions
where it actually took place, se. in the district of Nangrahar (the
modern, post-1964 re-organisation Mghan province of that name,
lying along the middle reaches of the Kabul River, with Jalalabad
as its centre). 414 It seems that there was a third figure who might
potentially have become involved in the clash, in addition to the two
protagonists Maudiid and Mul}.ammad. Present near the battlefield-either having come with one of the opposing armies or else
having arrived independently with a force of his own -was Maud iid's
uncle •Abd ar-Rashid, who had in fact also been present with
Maudiid and Mas•ud on the field at Dandanqan. 415 As the sole
surviving son-so far as we know-of the great sultan Mal}.miid,
•Abd ar-Rashid had obviously a powerful claim to the headship of
the Ghaznavid dynasty, and was indeed ultimately to become sultan
shortly after Maudiid's own death. Maudiid had therefore at least to
secure •Abd ar-Rashid's neutrality, if he could not gain his active
support. Maudiid accomplished this, firstly by promising •Abd arRashid a dominant share in the exercise of power and bestowal of
honours, if he himself successfully obtained the throne, and secondly,
by reminding his uncle of the solemn oath which the latter had
given to his brother Mas•ud that he would not harm the interests of
Mas•ud's sons. •Abd ar-Rashid's neutrality was thus gained, and the
danger that he might conceivably come to an agreement with
Mul}.ammad and join up with his forces, or even perhaps that he
might watch the two opposing sides destroy each other and then step
in, as a tertius gaurlens, to seize the fruits of victory for himself, was
averted. 4141
Feeling now safe from such possibilities, Maudiid led personally an
assault on the enemy line which proved decisive; Muhammad'11

24

DEFEAT IN THE WEST AND ITS AFTERMATH

army crumbled, and the general Ertigin and the palace ghulams
battered them in an attack from the rear. Mu]:J.ammad, his sons, and
the rebel generals Aniishtigin Balkhi and 'Ali Qarib's son all fell into
Maudiid's hands. The latter now took exemplary vengeance on those
whom he deemed responsible for his father's death. lbn Baba names
only Aniishtigin Balkhi, the son of Amir Yiisufb. Sebiiktigin, a son
of Muhammad's (presumably A]:J.mad) and three other commanders,
as suffering death. Other sources, however, speak of a more general
slaughter of the military leaders and of the whole of Mu]:J.ammad's
family (including, by implication, Mu]:J.ammad himself), with only
'Abd ar-Ra]:J.man (or 'Abd ar-Rai;!Im) being spared because he had
shown compassion towards the imprisoned Mas'iid in Giri and had
condemned his brother Ahmad's leading role in the killing. 67 To
mark the site of the battle and to commemorate his victory, Maudiid
now built there a settlement ( qarya, qa,raba, what Raverty in the
notes to his Tabaqiit-i Nii#ri translation calls a 'Bazar and emporium') and a ribiit, and named the place appropriately as Fat]:J.abad before returning in triumph to Ghazna. Fakhr-i Mudabbir
states that Fatl;labad, which was in the same area as a ribiit built by
Sebiiktigin to celebrate his victory in Lamghan over the HindiiShahi Raja Jaipal, subsequently prospered and became noted as a
resort for ghiizis (perhaps for warfare against the pagans of the
adjacent Kafiristan? ). The place is mentioned by the early nineteenth-century traveller Charles Masson as being situated four miles
south ofBalabagh and twelve miles fromJalalabad, and 'Futtehabad'
was also occupied by the British forces advancing towardsJalalabad
under Sir Robert Sale during the First Afghan War. 68
lbn al-Athir gives the date of Maudiid's state entry into Ghazna
as 23 Sha'ban 432/28 April 1041, and lbn Baba has Sha'ban [4]33
(read 432) f April-May 1041 as Maudiid's official accession date. 69
According to Gardizi's information, Maudiid bor~e the honorific
titles of Shihab ad-Din wa-d-Daula and Qutb al-Milla, and these
are confirmed by his coins. Some of Maudiid's bring the further
titles ofJamal ad-Daula and Fakhr a1-Umma, with the variant Qutb
ad-Din for that given in the literary source ofGardizi. Since Maudiid
on his coins acknowledges the supreme overlordship of the Abbasid
caliph al-Qa'im, it may be assumed that these titles were obtained
from Baghdad; some of them may possibly have been acquired by
.him during his father's lifetime. 7°
Internal threats to his position from within the Ghaznavid family

MAUDUD RE-ESTABLISHES IN THE WEST

25

ilad ostensibly been scotched by the defeat of Mul}.ammad, but
M:audud was taking no chances; in defiance of his solemn promise to
'Abd ar-Rashid of a share in the royal power, as mentioned above,
he ilnJilediately had his uncle arrested and imprisoned in the fortress
orMandish, where he remained all through Maudiid's sultanate. 71

5·
MaudUd re-establishes the
position in the west
The taSks facing the new sultan were nevertheless formidable. There
was still the Tiirkmen threat in northern Khurasan which Mas•ud
had faced with such a lack of vigour, together with a new fear that
the Seljuqs might take over Sistan and outflank the Ghaznavid
dominions from the south. There had been intermittent unrest in
Ghaznavid India during Mas•ud's reign, seen in the serious rebellion
of A!].mad Inaltigin; Mul}.ammad had drawn support from there
during his brief second sultanate; and we shall see that Maudiid had
very soon to face further rebellions in India.
We have seen that Maudiid's intended campaign into Tukhiristan
was rendered abortive by his need to return and secure the succession; as a result, the Ghaznavid general Altuntash was unable to
defend Balkh any longer, and it now suffered a severe plundering by
the Tiirkmens. Herat also fell to the Oghuz, but with Maudiid's
firm e<~tablishment on the throne of his fathers after the Nangrahar
victory, confidence in the Ghaznavids revived to some extent in
eastern Khurasan. The people of Herat rose against their occupiers
and restored the town once again to Ghaznavid allegianc~, although
by 434/ 1042-3 we hear of it again as besieged by Chaghri Beg
Da'lid; in the end it fell to the Seljuqs, and Herat and the surrounding region of Badghis passed definitively into the Seljuq orbit. There
is further evidence of a resid:mm of pro-Ghaznavid feeling in the
report of ~adr ad-Din l;lusaini that Chaghri Beg had to send an
expedition against the local leaders of Farazbaj or Qarabaj ( ?),
where people were still paying taxes to the Ghaznavids; one of these
local lords had to be attacked and besieged before he agreed to
recognise the Seljuqs. 73 Maudiid may have recovered Balkh for a
while, and Tirmidh on the Oxus certainly held out under Amirak
Baihaqi till after 435/1043-4 (see above, p. 10 ).
Maudiid's energetic policies and his determination not to accept
that the former Ghaznavid territories in the west were irretrievably

26

DEFEAT IN THE WEST AND ITS AFTERMATH

lost gained for him an access of prestige among contemporaries at
this time; it must have been difficult for these last to accept that so
mighty an edifice as the empire of Mal)mii.d and Mas•ud could be
permanently damaged by a horde of nomadic barbarians from the
Central Asian steppes. It is said that, early in his reign, Maudii.d
received an embassy and offers of allegiance from the 'King of the
Turks' in Transoxania, obviously a Qarakhanid and probably
Boritigin, who on numismatic evidence was already ruling in Bukhara
in 433/I04I-2 as Co-Qaghan of his brother Arslan Khan •Ain adDaula Mul)ammad b. Nasr of Ozgend, with the corresponding title
of Bughra Khan, and in Samarqand by 438/ I046-7; concerted
Ghaznavid-Qarakhanid military action against the common foe of
the Seljuqs did not, however, materialise till the very end ofMaudii.d's
reign, see below. 13
In 435/ I 043-4, hearing that Chaghri Beg Da'ud had fallen ill,
Maudii.d sent an army into Tukharistan. This attack was parried by
the Seljuq amir's son Alp Arslan, who was at that time based on
Balkh; in the ensuing battle, the Ghaznavid forces were defeated
with considerable losses, and the remnants returned to Ghazna. It
seems to have been this reverse which finally convinced the castellan
of Tirmidh, Amirak Baihaqi, ·that it was hopeless to hold out any
longer against the encircling Seljuqs. Chaghri Beg could legitimately
assume from this failure ofMaudii.d's that the Ghaznavid sultan now
lacked the resources ever to mount a serious and sustained war for the
recovery of Khurasan, and l;lusaini re<;ords that at this point,
Chaghri Beg formally made over the governorship of all northeastern Khurasan as far as the Oxus headwaters, comprising Balkh,
Tirmidh, Tukharistan, Qubadhiyan, Wakhsh and Walwalij, to his
son Alp Arslan. 74
Even so, Maudii.d still dreamed of regaining the lost territories,
and towards the end of his reign he tried to organise a military coalition against the Seljuqs, expending large smns of money in subsidies
and promising rule over the different regions of Khurasan, under a
general Ghaznavid suzerainty, to various anti-Seljuq powers of
eastern Islam to whom he now made approaches. These last included
the Dailami prince from the Kakuyid dynasty of Jibal, Abii. Kalijar
Garshasp b. •Ala' ad-Daula Mul)ammad, who had in 437/ 1045-6
finally lost his appanage of Hamadan to the Seljuq leader Ibrahim
Inal, and who spent the last years of his life in exile with his brother
Faramurz in I~fahan or with the Buyids of ~ars. 75 Maudii.d corn-

MAUDUD RE-ESTABLISHES IN THE WEST

27

municated with him in I~fahan, and persuaded Abii Ka.Iijar
Garshasp to raise an army and march eastwards; but the army
perished in the Great Desert, and the Kakiiyid returned, ill, to
western Persia. Maudiid further made approaches to the. Qarakhanids, to 'the Khaqan, King of the Turks', probably Tamghach
Khan lbrahim b. N~r, the former Boritigin, again. The Khan sent a
contingent from Bukhara to the vicinity of Tirmidh and plundered
and devastated the district (this fact confirms the indications of
I;Iusaini and lbn Funduq that Tirmidh had by now lapsed from
Ghaznavid control), and another force under his general Qashqa
was sent against Khwarazm, which had been abandoned by Mas'iid's
old ally Shah Malik and taken over by the Seljuqs. However, both
these attacks were repulsed, and Chaghri Beg and Tamghach Khan
lbrahim eventually met on the banks of the Oxus and made peace.
By this time, Maudiid himself may well have been dead. He had set
out from Ghazna with an army, but was immediately taken ill,
returned to the capital and died, so that all his grand strategy came
to naught. 7 8
Nor could Maudiid in the end retain Sistan within the Ghaznavid
sphere of. influence, as it had been in the days of Mabmiid and
Mas'iid, and by the end of his reign, the ruling family of Safla.rid
amirs had constituted Sistan as a largely autonomous unit, although
subject now to ultimate Seljuq suzerainty. This was nevertheless a
reasonably favourable outcome of affairs for the Ghaznavids, since
the Safla.rid amirate did at least form something of a buffer-state,
reducing the danger of Tiirkmen incursions through southern
Afghanistan against Ghaznavid Bust and Zamin-Dawar and possibly
even against Ghazna itself, a fear shown by Mas'iid's despatching his
general Niishtigin Naubati and then his official Abii Sahl Zauzani
immediately on his return from Dandanqan to secure the region of
Bust (see above, p. I3)·
Until Mas'iid's last years, the Ghaznavids had sent out officials to
Sistan in order to collect the tribute and taxes due to the sultan as
suzerain of Sistan. These officials operated latterly side-by-side with
the l>afla.rid amir Abii 1-Fac;ll Na~r b. Abmad, who had been
appointed regent in Sistan by Sult!m Mal;lmiid just before his death
in 42 I/ I030. 77 Mas'iid for a time worked through officials of his own
appointed to collect the taxation from Sistan, but at the opening of
429/0ctober I037 he dismissed the.two officials responsible and replaced them by amir Abii 1-FaQ.l once more, who now became

28

DEFEAT IN THE Wt;S'i' ANJJ ITII AJ:o"l'ERMATH

directly responsible to the sultan for the 'amal or financial yield of the
province. 78 The social and political situation in Sistan had been since
the time of the Arab governors of the caliphs, predecessors of the
~affiirids, a complex and divided one, in which the bands of 'ayyars
and sarhangs, who in general expressed local patriotic Sagzi feeling
and opposition to outside domination, frequently played a dominant
rOle."
Early in 432/ autumn 1040 a Sagzi rebel called Al;lmad-i Tahir
raised a revolt at Karkiiya against the authority of amir Abii 1-Fa(ll,
and summoned in the Tiirkmens as allies. Abii 1-Fac;ll sought military
assistance from sultan Mas'iid in Ghazna, but this request came at a
highly unfavourable time, when the Ghaznavid was preparing for
the move to India. He did, however, finally in Rabi' I 432/November-December 1040 send a force of 5,ooo cavalrymen under Ba
Na~r (presumably the l;lajib Bii n-Na~r mentioned in various places
by Baihaqi, for example, as being amongst the combatants at
Dandanqan and amongst the sultan's group that fled through Ghiir
back to Ghazna) 80 to relieve Abii 1-Fa(ll, by now beleaguered in the
capital Zarang by the rebels and by the Seljuq leader Ertash, who is
described as the brother of Ibrahim Inal and a cousin ofToghril Beg.
Abii 1-Fa(ll saw no way out but capitulation to the Seljuq, and he
came to an agreement with Ertash that the khutba in Zarang should
be made for the more senior Seljuq chief Bi'ghu or Paighu. 81 Ba
Na~r's force could only withdraw to Bust. Bighu appeared personally
in Sistan in Rabi' n 432/December 1040, and the united Tiirkmen
bands advanced on Bust and laid waste the countryside there; however, differences arose between Ertash and Bighu, and these compelled aretreat from Bust. In the end, the Tiirkmens evacuated the
province ofSistan and returned to Khurasan. a:z
Thus the position at Maudiid's accession was that Sistan, under
amir Abii 1-Fac;ll, was temporarily clear ofoutside forces, but Maudiid
resolved immediately to send an army to restore Ghaznavid influence
and to establish there a barrier against further Tiirkmen raids. A
force under the .Leadership of the commander Qaimas was sent later
in 432/spring-summer 1041, but it was, however, defeated, and
Abii 1-Fac;ll later intercepted letters from Maudiid to various notables
in Sistan; hence in Jumada n 433/February 1042 Abii 1-Fa(ll
arrested and imprisoned a number of Ghaznavid sympathisers,
including men of religion, faqihs and an imam, and military commanders. It may well have been that the religious classes in Sistan

MAUDUD RE-ESTABLISHES IN THE WEST

29

were especially favourable to the Ghaznavid connection because of
the sultans' reputation as upholders of the Sunna and because of fears
of Tiirkmen anarchy. In Rajah of the same year/March I042 a
Ghaznavid army, numbering 2,ooo cavalry and Io,ooo infantry, reappeared in Sistlin and joined forces with various dissident elements
there, including the partisans of the earlier rebel Al}.mad-i Tlihir and
the 'ayytir group of the Shangaliyan. 83 Fierce fighting ensued, with
Abii 1-Fa<;l.l besieged in the citadel ofZarang for four months, till the
latter wrote to Ertash for help. A relieving force eventually appeared,
defeated the Ghaznavid coalition forces, killing many of their leaders
and pursuing the fugitives through the desert of the Dasht-i Margo
back to Bust, where the soldiers of Ertash and Abii 1-Fa<;l.l plundered
the region before returning to Sistlin in Rabi' I 434/0ctoberNovember I042. It may well be this expedition that Ibn al-Athir
also mentions and places in ~afar 435/September-October I043
(read rather ~afar 434, which would fit better with the Ta'rikh-i
Sisttin's very detailed and exact chronology of events?), stating that
Maudiid's forces repulsed from Bust an Oghuz attack at that time. 84
The ensuing events of the year 434/ I042-3 and thereafter in
Sistlin are inter alia notable for the appearance, as a leading figure on
the stage of history, of the Ghaznavid Turkish slave commander
Toghrll, who was to play such a maleficent role in 'Abd ar-Rashid's
sultanate. At this juncture, Toghrllleft Bust with 2,ooo troops and
marched towards Sistan, having the good fortune to capture en route
a member of the ~aflarid family, the amir Abii n-Na~r or Bli Na~r
b. Man~iir b. Al}.mad, at the valley of H.n.danqan. He then entered
Sistlin in Jumada n 434/January-February I043, occupying Karkiiya and causing there indiscriminate slaughter amongst the Muslim
and Zoroastrian population. Abii 1-Fa<;l.l, however, sent a body of
troops to defend the citadel of Karkiiya, and Toghri:l decided to
return to Ghazna, taking with him the amir Abii n-Na~r. This last
was later to be exchanged by Abii 1-Fa<;l.l for a son of the great
Ghaznavid vizier Al}.mad b. l;lasan Maimandi 85 and other captured
Ghaznavid commanders, and he continued thereafter to play a part
in the tortuous, often internecine strife of Sistlin. The Ta'rikh-i Sisttin
records under 437 f I045-6 a further clash ofErtash ~nd Ghaznavid
troops, in which the Seljuq leader was defeated and fled for safety to
the citadel ofZarang; he was killed at Tabas three years later, having
been no longer mentioned in connection with events in Sistlin. 86
Maudiid's reign thus closed with a reasonably favourable and

_30

DEFEAT IN THE WEST AND ITS AFTERMATH

stable position on his southern flank. Abii 1-Far;ll-who continued to
reign as Amir of Sistan till his death in J umada 11 465/ March
I073 87 -was now balancing himself between the two rival great
powers, although he gradually became more and more drawn into
the Seljuq orbit as the Seljuqs consolidated their power in Khurasan
and Kirman; by Rajah 439/December I047-january I048 he felt
secure enough to release most of the Ghaznavid sympathisers whom
he had arrested six years before. 88 Bust still remained as the bulwark
of Ghaznavid power in southern Mghanistan, and was never in fact
to be relinquished by the Ghaznavids till the rise of the Ghiirids in
the following century (see below, Ch.4, p. I22 ).

6.
The campaigns in India
Mter the suppression of Al}.mad Inaltigin's revolt in India, Mas'iid
had in Dhii 1-Qa'da 427/ August-September I036 appointed his son
Majdiid to be commander-in-chief there, fitting him out with a robe
of honour appropriate for the viceregal office, and attaching to him
three military commanders and a splendidly-equipped army, a
secretary from the Diwan-i Risalat for chancery business and a
mustarifi or accountant as his financial clerk (the latter is named by
Baihaqi as Sa'd-i Salman, obviously the father of the poet Mas'iid-i
Sa'd-i Salman, and this may well have been the origin of the family's
settling in the Panjab ). Majdiid thus began an association with
India, doubtless building up there a power-base for himself with an
entourage of officials and troops personally loyal to him. He is, however, mentioned as being back in Ghazna on occasion, for example
for the 'Id al-Fip- celebrations in 428/ I037, and Gardizi (though
not Baihaqi) says that he was in the autumn of early 432/ I 040 sent
by sultan Mas'iid to secure Multan (see above, p. I4). 89
Majdiid was thus in India when his father was deposed and
murdered. What his attitude was to Mul}.ammad's elevation to the
throne is unknown, but he refused to recognise the succession of his
brother Maudiid in Ghazna. He raised the standard of revolt in both
Multan and Lahore, but was mysteriously found dead three days
after the 'Id al-Ac;lba (presumably of 432, when this festival fell on
I I August I04I ). Majdiid's revolt thereupon collapsed, and Maudiid
was able to make firm his authority throughout Ghaznavid India.
The army which he had already despatched from Ghazna to quell
Majdiid's revolt must be that mentioned in an anecdote of Fakhr-i

THE CAMPAIGNS IN INDIA

31

Mudabbir's Adiib al-barb wa-sh-shajii'a as being under the command
of the SaHir Al}.mad b. Mul}.ammad as f:liijib-i Buzurg, with the Faqih
Saliti nominated as governor of Lahore (it is very likely that these
two names conceal, under other forms, personages mentioned in the
pages of Baihaqi, but exact identification is impossible). The tale
recounts how this force reached Lahore, where Maudiid's authority
was recognised by the army there. 90
Having joined up with the Lahore garrison, the Faqih Saliti left a
deputy in Lahore, and the combined Ghaznavid army undertook a
campaign against Multan. In the course of the 4th/1oth century, the
extremist Shi'i Isma'ili da'wa or propaganda movement had enjoyed
a signal success amongst the MusliiOS of the old Arab colonies in Sind
and Multan; these regions had recognised the supremacy of the
Fapmid caliph in North Africa and Cairo, al-Mu'izz (341-65/
953-75), and the famous idol-temple of Multan, dedicated to the
Sun-God Aditya, had been destroyed by one of the Isma'ili dii'is or
missionaries. 91 Mal}.miid of Ghazna, the zealot for orthodoxy and
upholder of the Sunna, had in 396/ 1006 attacked the local Isma'ili
ruler of Multan, Abii 1-Fatl}. Da~iid b. Na~r, stormed the town and
conducted a savage massacre oflsma'ili sympathisers. 92 Thereafter,
we hear nothing particulaqlbout events in Multan, but Isma'ilism
there was clearly not dead, and rebellion broke out there, probably
in 432/1040-1 when the news ofMas'iid's deposition and capture at
Marikala became generally known in India, and Ghaznavid authority was at a low ebb. Once Lahore was secured for Maudiid, the
Faqih Saliti marched against the rebels, who were headed by Abii
1-Fatl}. Da'iid's son, whom the Isma'ilis ( Qariimifa in contemporary
Ghaznavid phraseology) 93 addressed as their Shaikh. The Isma'ili
forces were unable to withstand the powerful Ghaznavid professional
army, and withdrew to Man~iira ·in southern Sind. Multan itself was
compelled to surrender, and the khutba there was now made for the
Abbasids and Maudiid (implying that it had been made by the
rebels for the Fatimids once again). The Faqih Saliti then appointed
Mul}.ammad-i I:Jalimi 94 as governor of Multan before returning to
Lahore, harrying the Jhats and other infidels of the middle Indus
region en route. 9 '
The news of Mas'iid's end likewise emboldened various Indian
princes into launching an attack on the Muslims, and Fakhr-i
Mudabbir goes on to say that the army returning from Multan to
Lahore was attacked by a coalition of Indian rulers, 'Rays, Ranas

32

DEFEAT IN THE WEST AND ITS AFTERMATH

and Thakkurs of the hill tracts', under Sandanpiil, described as the·
grandson of the K.iibul-Shiih (read Hindii-Shiih, se. the HindiiShiihis ofWaihind, old opponents of the Ghaznavids?).u A battle
took place at Q.d.r.j.w.r ( ?) in which the Faqih Saliti's army
routed the infidels and in which Sandanpiil was killed. 117
Despite his unavoidable concerns with the Seljuqs in northern
Khurasan and with Sistiin, Maudiid found some time to fulfil the
traditional role of the Ghaznavid sultans as hammers of the pagan
Hindus and as bringers into circulation within the eastern Islamic
economy of the temple treasures of India. After the loss of a rich
province like Khurasan, warfare in India was now especially vital for
financing the administration of the Ghaznavid empire and for providing the standing army with plunder and with an outlet fj>r its
energies. Maudiid had shown himself personally as a brave fighter
during his father's lifetime, and is said to have been particularly skilful with the bow. The invention of a particular kind of arrow-head,
the paikan-i MaudUdi one, is ascribed to him, this being allegedly
made from gold so that anyone shot by an arrow thus tipped could
either have his shroud bought out of the value of the arrow-head or
have treatment for the wound out of its value. Perhaps we have here
a reminiscence of the costly equipment and bejewelled weapons used
on ceremonial occasions, such as official receptions, by the Ghaznavid
palace ghuliims, although the legend of Maudiid's golden arrowheads, iflegend it was, did engender the verses
"ie!o The sultan of the age, Shiih Maudiid, the one who makes
arrow-heads ofgold for his enemy,
So that whoever is killed by it can thereby obtain his shroud,
or if wounded. can get treatment through it. 98
'ihe principal passage in the Islamic historians relating to
Maudiid's activities in India is a regrettably vague passage of.Ibn
al-Athir's (substantially repeated, but with some embroidery, in
Mirkhwiind ), with no topographical.information and with the names
of two Indian princes in corrupt form. This passage states that, in
435/1043-4, three Indian princes attacked Lahore and besieged it
for a considerable period of time. The governor there (whether this
was still the Faqih Saliti of Fakhr-i Mudabbir's anecdote is unknown) sent to Maudiid for help and received reinforcements, by
means of which the Ghaznavid army then took the offensive. One of
the Indian princes had by then returned to Ghaznavid allegiance.
One of the remaining two, D.w.biil (Devapiila) H.r.biita ( ?) was

THE INTERNAL FUNCTIONING OF THE EMPIRE

33

besieged in one of his fortresses till he surrendered; the other, Tab.t
( ? ) Ray, was slain in battle with s,ooo of his men, and the fates of
these persuaded other Indian rulers of doubtful allegiance to reaffirm their loyalty to the Ghaznavids. It seems that we have reference here to the confederacy of Indian potentates who at this time
reconquered from the Muslims Hansi and Thanesar to the north-west
of Delhi, Nagarkot and other places, and who besieged Lahore for
seven months. One of the leading members of this coalition was the
great Paramara Raja ofMalwa, Bhoja; the Devapala mentioned by
lbn al-Athir is probably the Kachchhapaghata Raja of that name,
son of the ruler of Gwalior Kirttiraja who may have been the prince
who submitted to Mabmiid of Ghazna in 4I3/ I022.'' Explicit
confirmation of this is, however, lacking. 100
To supplement this, we have only some additional information
from the Deccani historian Mubammad Qasim Hindii-Shah, called
Firishta, who wrote his Gulshan-i lbrahimi at the beginning of the I I th f
I 7th century and is accordingly a very late source. Exactly whence
Firishta derived his information is unknown, and the authenticity of
his information cannot be checked through earlier sources; it does
not, for instance, figure in Mir'khwand, According to Firishta, when
Mas'iid was murdered, his son Majdiid, at the instigation of his
adviser Ayaz Kh~!.!, marched from Multan and occupied fqr himself
territory in the valley of the Indus and its tributaries as far east as
Hansi and Thanesar before his sudc;Jen and mysterious death. He
then further relates how, in 435/I043-4, the Raja of Delhi and
other rulers recaptured Hansi, Thanesar and their dependencies
from Maudiid's governor in India and besieged Nagarkot for four
months. The Rajput princes of the Panjab were stirred up, and three
of them attacked Lahore, but were beaten off. These latter details fit
grosso modo with those of lbn al-Athir, but Firishta is nevertheless
even vaguer than his predecessor in that no names are given at all
for the Indian princes involved in these events. 101

7·
The internal functioning of the empire
Lacking as we do for Maudiid's reign such a detailed account of the
day-to-day working of the diwins as we derive from the pages of
Baihaqi for the bureaucracy during Mas'iid's sultanate, we must
perforce assume that the essential continuity in personnel over the
two reigns implied little significant change in governmental ethos or

34

DEFEAT IN THE WEST AND ITS AFTERMATH

in the practical running of the empire after Mas'iid's death. Certainly, the wise and experienced Abii Na~r Al)mad b. Mu}.lammad b.
'Abd a~-~amad Shirazi, 101 who had been Maudiid's adviser in
Tukharistan at the time of his father's murder and whose counsels
had probably contributed much to Maudiid's eventually securing
the throne for himself, continued in office as vizier for the early part
of the new reign. Al_lmad had been kadkhudii (chief executive or
adjutant, in effect vizier) to Mal_lmiid's governor in Khwarazm,
Altuntash, and then after the death of Al_lmad b. l;lasan Maimandi
in 424/1032, he became vizier to Mas'iid, exercising a moderating
influence on the sultan's erratic ways without, however, being able
to restrain him in the end from the ill-starred decision to retire to
India, described above. 103 It was from this time onwards, se. his
appointment as vizier, that he became the mamdill) of such of the
great contemporary poets as Maniichihri, who praises him as
"il; The sun of viziers, Al_lmad-i 'Abd a~-~amad, the one who is
not merely the sun of viziers but the sun of both the heavy
creations [se. of men andjinr].
He is the oustanding leader of all the outstanding leaders of
the world, just as the iron point is the foremost part of a khalli
spear.
He is superior to all mankind through his possession of two
small things, se. through his [stout] heart and his [eloquent]
tongue. 104
After these eight years as Mas'iid's vizier, A}.lmad served Maudiid
for two years, but then fell into disfavour through the jeajousy of the
military commanders; Baihaqi states that he died only a short time
after dismissal, but the much later biographical sources state that his
enem~s had him ppisoned. 1 05
Maudiid now appointed to the vizierate Tahir, ·who had been
accountant (mustarifi) in the Diwiin-i Istifii', the accounting section of
the Diwiin-i Wazir. Baihaqi mentio~s him as being still in charge of
this department in 424/1033, and when Mas'iid departed for India
just before his deposition and death, he gave Tahir formal permission
to take office under the Seljuqs, whom he fully expected would occupy
Ghazna after his retiral to India (see above, p~ I 6). However, it soon
transpired that narrow financial expertise was not enough for the
onerous job of vizier, for Tahir was a complete failure as chief
minister and, after only two months, Maudiid dismissed him. 106
The sultan was more successful with his third vizier, 'Abd ar-

THE INTERNAL FUNCTIONING OF THE EMPIRE

35

Razzaq b. Al}.mad b. I:Iasan Maimandi, for 'Abd ar-Razzaq served
Maudiid for the remainder of his reign; he then played a decisive
part in setting up 'Abd ar-Rashid as sultan and served him as chief
executive during his brief tenure of power (see pp. 39-40 ). There
was, of course, always a preference for someone like 'Abd ar-Razzaq,
whose family background was one of service to the dynasty, for it was
widely held that the arcana of such professions as secretary or
financial official were handed down within families; whence both the
Maimandi and the Shirazi families ~ere active in the Ghaznavid
administration for at least three generations. It seems that 'Abd arRazzaq had fallen from grace tog~ther with his father in the later
part of Mal;lmiid;s sultanate, for when the ·new ruler Mas'iid ordered
the release of Al}.mad b. J:lasan from jail, 'Abd ar-Razzaq was likewise set free at the beginning of 422/1031 from incarceration in the
fortress of Nandana or Nardin on the Jhelum river in the Panjab.
Thereafter he is mentioned sporadically by Baihaqi, and must have
served in the central administration; he was also present at Dandanqan. Although he did not apparently serve Farrukh-Zad as
vizier, he was still active then, for Baihaqi speaks of him as still in
Qfficial employment when he himself was writing in 450/ 1058, and he
served as an informant for Baihaqi over various items of information
handed down from his father Al}.mad b. I:Iasan's time. 107
Having the spoils of India at their disposal; the Ghaznavids were
great builders of palaces and kiosks and enthusiastic layers-out of
gardens and polo-grounds. 108 Unfortunately, we know nothing of
Maudiid's efforts in this direction, although it is likely that hj::
endeavoured to follow in his predecessors' footsteps here. Similarly, it
seems that the co11rt continued to be organised on the familiar formal
and hierarchical pattern of Perso-Islamic monarchs, with a household of court officials and eunuchs and a group of boon-companions
around the sultans for entertaining him during leisure periods and
drinking-sessions. We know that the prince Kai Ka'iis b. Iskandar b.
Qabiis, from the Ziyarid family of Gurgan and Tabaristan, spent
seven or eight years at his kinsman Maudiid's court as a commensal
( nadim-i khiiu); in his 'Mirror for Princes', the Qfibiis-nama, he mentions the wine-drinking sessions of Maudiid and the vizier 'Abd
<tr-Razzaq. 109 We may further assume that Maudiid kept up the traditions of his house in encouraging poets and literary men, although
once again, specific information is lacking; the great poetic figures of
MaQ.miid's and :Mas'iid's reigns, such as 'U~uri, Farrukhi and

36

DEFEAT IN THE WEST AND ITS AFTERMATH

Maniichihri, were either dead or silent by Maudiid's sultanate. The
great scientist and polymath Abii Rai):tan Biriini was nevertheless
alive until after 442/1050- 1, dying an octogenarian, probably at
Ghazna; his treatise on mineralogy, the Kitiib al-Jamiihir fi ma'rifat
al-jawiihir, was certainly written during Maudiid's reign, and perhaps
also his last major work, on pharmacology, the Kitiib a1-$aidala fi
1-libb,llO
. As noted above, p. 2 7, Maudiid died when about to lead a revanche
against the Seljuqs in Khurasan. According to lbn Baba, he was
struck down by an internal disorder ( qulanj) shortly after leaving
Ghazna; he had only time to despatch the vizier 'Abd ar-Razzaq to
Sistan in order to avert a threat there before he died. Firishta, alone
of the sources, has the information that Maudiid had set out via
Sakawand and the Loghar valley making for a fortress called
Sankot ( ? ), where he intended to collect some treasure stored up
there, when he was taken ill. 1 11
·The data on the date of Maudiid's death and the duration of his
sultanate are incomplete and somewhat contradictory, and are
further confused by uncertainty in the sources over the reigns of his
two ephemeral successors Mas'iid 11 b. Maudiid and 'Ali b. Mas'iid.
Of the standard reference books, S. Lane Poole's The Mohammadan
dynasties (London 1893), 289, adopted 440/1048-9 as the date for
Maudiid's death, the two short reigns after him and 'Abd arRashid's accession. E. de· Zambaur, in his Manuel de glnealogie et de
chronologie pour l'histoire de l'lslam (Hanover 1927 ), 282, placed
Maudiid's death in Rajah 440/December 1048-January 1049, the
two short reigns in this same year 440 and 'Abd ar-Rashid's accession in 441/june 1049-May 1050. The present author, in his The
Islamic dynasties, a chronological and genealogical handbook (Edinburgh
1967 ), 181, basing himself on lbn,at-Athir's information, chose 441
as the year of Maudiid's death, the two short reigns and 'Abd arRashid's accession, but as will emerge from what is said below, this
dating is controversial.
lbn al-Athir states that Maudiid died on 20 Rajah 441/18
December 1049 at the age of twenty-nine, agreeing to within a day
with lbn Baba's date of 21 Rajah 441/rg December 1049, but lbn
al-Athir's information here that Maudiid reigned for nine [lunar]
years and ten months is too long, and we should probably read eight
years and ten months. Building upon the date of 23 Sha'ban 432/28
April 1041 for Maudiid's victorious entry into Ghazna and his

37
foanal accession to power (see above, p.24), we arrive at a duration
for his reign of approximately eight lunar years, eleven months I eight
solar years, seven months, three weeks. J:Iusaini gives no exact dates,
but states that Maudiid ruled for seven [lunar] years, ten months
and two days; this would place his death on 25 Jumada n 440l5
December 1048. 112 We thus have two possible dates for Maudiid's
death, with a disparity of just over one year. The dates given by lbn
Baba and lbn al-Athir have about them the ring of definiteness; the
dating based on a calculation involving J:Iusaini's figure for Maudiid's
period of rule involves the computation of a date of death not
directly confirmed by any written source. Nevertheless, in favour of
the second date is a piece of numismatic evidence to which D. Sourdel
has drawn attention. He points out that two dinars of •Abd arRashid are extant, one in the British Museum (it was presumably
because of this coin that Lane Poole, the cataloguer of the British
Museum's Islamic coin collection, gave in his Muhammadan dynasties
the year 440 as that of •Abd ar-Rashid's accession) and the other in
the Kabul Museum, which both clearly have the date 440, and these
seem therefore to confirm the dating based on J:Iusaini. 113 If one
adopts this last system, the brief reigns of Mas•ud u and •Aii will
have to be placed in the third quarter of 440lwinter of 1048-g in
order to allow •Abd ar-Rashid to begin his reign in say the last
quarter of 440 I spring I 049·
SUCCESSION DIFFICULTIES

8.
Succession difficulties and the accession
oj•Abd ar-Rashid

We do not possess much firm information about Maudiid's two
immediate successors. Many of the later sources do not even mention
their existence. 114 According to jiizjani, he left behind three sons.
lbn Baba says that when on the point of death, Maudiid appointed
his five-year-old son (named elsewhere as Mas•ud) as successor, but
Mas•ud reigned only for five days before the great men of state raised
to power Maudiid's brother Abii 1-l;lasan •Aii b. Mas•ud. The latter
is an obscure figure, unmentioned by Gardizi and Baihaqi, for
instance-at least in the extant part of their works.!....as having
played any earlier part in affairs. According to lbn Baba again, he
reigned for only forty-five days ·before the army leaders deposed him
and consigned him to imprisonment in a fortress, but during this
time he assumed the honorific title of Baha' ad-Daula, if the literary

38

DEFEAT IN THE WEST AND ITS AFTERMATH

sources are correct; no coins of his are known. 115 Jiizjiini has the information that Mas'iid (erroneously called here Mu.l)ammad) b.
Maudiid and 'Ali b. Mas'iid ruled in concert (bi-sh-shirka), which
may perhaps relate to some regency provisions for the child Mas'iid;
but once their ineptitude was revealed, they were removed from
power. 116
It may have happened that 'Ali b. Mas'iid showed signs of being
a potentially strong ruler, and was therefore deposed in favour of one
whom the military considered more pliable; but it is more probable
that the indolence of harem life and subsequent incarceration had
rendered 'Ali little fitted for the exercise of power. Indeed, the
traumatic experiences of the succession disputes in 42I I I030 and
432 I I040, added to the general climate of fear and suspicion prevailing in a despotic state like the Ghaznavid one, 117 seem to have
brought about the adoption of a policy rather like that of the later
Ottoman qtifes or 'cage', the precautionary jailing of all male relatives who might have designs on the throne. As Ibn Biibii notes, the
three successive sultans 'Abd ar-Rashid, Farrukh-Ziid and Ibriihim
all had to be fetched out of the fortress where they had been imprisoned and then raised to power. This state of affairs forms a clear
contrast to the position in the early, formative stages of the Ghaznavid empire, when brothers or uncles of the sovereign, such as Ma.l)miid's brothers Abii 1-Mu~affar Na~r and 'A9ud ad-Daula Yiisuf,
were regularly employed as military commanders or in provincial
governorships.
It is difficult to know what to make of the information ofFirishta,
who despite his lateness as a source has interesting material on these
two reigns not found in the much earlier authorities. Thus he speaks
of what seems very probable, that the real power during Abii
Ja'far Mas'iid b. Maudiid's brief reign was exercised by the great
Turkish military commanders, amongst whom he names two rival
parties he