Skip to main content

Full text of "Ahmed Riza Khan Barelwi_In the path of the prophet_Usha Sanyal"

See other formats


of the 



Ahmad Riza Khan 

In the Path of the Prophet 

Usha Sanjal 


prelims. 044 10/12/2004 4:09 PM Pag€Ai 


Ahmad Riza 
Khan Barelwi 




prelims. 044 10/12/2004 4:09 PM Pa 


Series editor: Patricia Crone, 
Institute for Advanced Study, Princeton 

Abd al-Malik, Chase F. Robinson 

Abd al-Rahman III, Maribel Fierro 

Abu Nuwas, Philip Kennedy 

Ahmad ibn Hanbal, Christopher Melchert 

Ahmad Riza Khan Barelwi, Usha Sanval 

Al-Ma'mun, Michael Cooperson 

r^\ Al-Mutanabbi, Margaret Larkin _T^_ 

Amir Khusraw, Sunil Sharma 

El Hajj BeshirAgha, Jane Hathawav 

Fazlallah Astarabadi and the Hurufis, Shazad Bashir 

Ibn Arabi, William C. Chittick 

Ibn FuJi, Ahmad Dallal 

Ikhwan al-Safa, Godefroid de Callatav 

Shaykh Mujid, Tumima. Bavhom-Daou 

For current information and details of other books in the 

series, please visit www. 

subjects/makers-of- muslim-world.htm 


prelims. 044 10/12/2004 4:09 PM Page-si i± 


Ahmad Riza 
Khan Barelwi 

In the Path of the Prophet 






prelims. 044 10/12/2004 4:09 PM Pag^-xiv 


Oneworld Publications 

(Sales and Editorial) 

1 85 Banbury Road 

Oxford OX2 7AR 


www. oneworld -publications, com 

© Usha Sanyal 2005 

All rights reserved 

Copyright under Berne Convention 

A CIP record for this title is available 

from the British Library 

ISBN 1-85168-359-3 

Typeset by Jayvee, India 

Cover and text designed by Design Deluxe 

Printed and bound in India bvThomson Press Ltd 



prelims. 044 10/12/2004 4:09 PM Pa 






prelims. 044 10/12/2004 4:09 PM Pag^-xvi 




prelims. 044 10/12/2004 4:09 PM Page-svii 


Preface x 
Acknowledgments xii 


The Mughal Empire 1 

The North Indian Successor States 4 

The History of Rohilkhand S 

British India under the East India Company 7 


-(J)- ShahWali Ullah 22 

Farangi Mahall : Training Employees for the 

Muslim States 26 
Nineteenth-century Reform Movements 28 


Rampur State S3 

Ahmad Riza's Education and Scholarly Training SS 

Scholarly Imprint of his Father 56 

Exemplary Stories 57 

Sufi Discipleship to Shah Al-e Rasul of Marehra 61 

The Importance of Dreams 61 

Savvids of the Qadiri Order of Sufis 62 

Going on Pilgrimage, 1 878 63 

Ahmad Riza as Mujaddid 64 

Fatwa Writing 66 


prelims. 044 10/12/2004 4:09 PM Pa 

g€~x viii 


Hidden Cues in a Fatwa, or what a Fatwa may not 

tell us 68 
Ahmad Riza's Fatawa 70 
Two Fatawa Written during Ahmad Riza's Second 

Pilgrimage to Mecca 73 
Political Issues in the Earlv Teens and Twenties 77 
Hijrat Movement 8 1 

Ahmad Riza's Popularity among Core Followers 83 
Passing on the Leadership 84 


Ahmad Riza as a Sufi 89 

The Perfect Pir 90 

Controversy about Sufi Intercession 91 

The Three Circles of Discipleship 92 

Shaikh' Abd al-Qadir Jilani and the Importance 

of the Qadiri Order 94 
Love of the Prophet 96 
Sufi Rituals 100 

Relations with other Muslims 1 02 
The Accusations of Unbelief 107 
Relations with Non-Muslims: Hindus and the British 109 


Seminaries (Madrasas) 1 1 1 
Printing Presses and Publications 113 
Voluntarv Associations and Oral Debates 1 1 S 
Generational Fissures in the Movement 118 
Assessment of the Importance of the Movement in 
Relation to other Movements 122 


prelims. 044 10/12/2004 4:09 PM Pa 



Ahmad Riza's 'Urs in India and Pakistan 1 29 
Ahl-e Sunnat/Barelwis in the Diaspora 131 

Glossary 133 

Major landmarks in South Asian history from 

the eighteenth to the twentieth centuries 1 35 
Bibliography 138 
Index 143 

"0" "& 


prelims. 044 10/12/2004 4:09 PM Pa 





prelims. 044 10/12/2004 4:09 PM Pa«Axi 



1 he subject of this book is an Indian Muslim scholar of the 
late nineteenth— ear lv twentieth centuries, Ahmad Riza 
Khan Barelwi (1856—1921). His writings and the interpret- 
ation of Islam they espouse laid the foundation for a movement 
known to its followers as the Ahl-e Sunnat wa Jama' at ("the 
devotees of the Prophet's practice and the broad community") 
and to all others as "Barelwi," an adjective derived from 
Bareillv, the town where Ahmad Riza was born and where he 
lived. It also forms the last part of his name. The movement was 
one of several reformist groups to have emerged in British India 
during the late nineteenth century. Like their rivals, the 
-(^y— Barelwis today have a large following in South Asia, as well as in — (^y- 

Britain and other parts of the world where South Asian 
Muslims have migrated. 

What distinguishes the Barelwis from the other reformist 
groups (Deobandis, the Ahl-e Hadith, and others) is their atti- 
tude to the relationship of the transcendant to this world. While 
the other groups reject sufism or Islamic mysticism either wholly 
or in part, and deny the importance of saintlv mediators, mir- 
acles, and other manifestations of the holy in the here and now, 
the Barelwis embrace everything associated with sufism as an 
intrinsic part of their identity. But they share with the other 
reformists a strong focus on the Prophet Muhammad as a model 
of correct behavior and an example of the virtues that every 

Muslim should strive to cultivate and that he or she should live by. 


Unlike some of the other Muslim reformist groups, Ahmad 
Riza defined religious communitv in cultural rather than 
political terms. When Indian Muslims began to engage in 


prelims. 044 10/12/2004 4:09 PM Page-sxii 


national politics in the early twentieth century, he advised his 
followers against it, arguing that the classical Islamic sources 
did not support political action against British rule in India, as 
the British had not interfered in the Muslims' internal affairs or 
religious institutions. This led to a split in the movement, with 
some Barelwi leaders following his advice and others rebelling. 
In some respects, Ahmad Riza and the Barelwi movement in 
general seem paradoxical. Thus, while Ahmad Riza's interpret- 
ation of Islam was deeply rooted in South Asian culture, he 
based his arguments on the classical Islamic sources and looked 
to the religious leaders of Mecca and Medina for validation 
and approval. And while he was a reformist in the sense of 
demanding that his followers be personally responsible for 
their own salvation, the kind of model Muslim person he 
visualized was one who embraced rather than shunned ritual 
intermediaries and a ritualistic style of "worshiping God. One 
might say that he "wanted his followers to use reformist reli- 
\7" gious methods so as to be better, and more individually driven, —kzj~ 



prelims. 044 10/12/2004 4:09 PM PaKxxlii 



i ill Roff, to whom I dedicate the book, guided the original 
'Ph.D. dissertation on "which this book is based, and has lent 
me his sage advice over many years even after he retired and 
returned to his home in Scotland. David Gilmartin has helped 
me think about the material in new ways which have found 
their way into this book. And Patricia Crone asked difficult 
questions and showed me that writing a little book can be 
harder than writing a big one. I would also like to thank the 
anonymous reviewer of the manuscript for making anumber of 
suggestions which I have incorporated here. 

I owe a great debt of gratitude to members of mv familv: mv 
-(zy— dear friend (and sister-in-law), Rupa Bose, who was mv first — (^3~ 

reader and urged me to try and make the material both inter- 
esting and relevant to as wide an audience as possible. Gautam 
Bose, mv husband, gave me time to write on weekends and 
holidays, while keeping our two bovs, Girish and Arun (who 
are eight and six, respectively) entertained and out of mischief. 
And to mv mother, Vina Sanyal, who has supported my aca- 
demic endeavors in myriad "ways over the years from far-awav 
New Delhi, a heartfelt thank you. 


prelims. 044 10/12/2004 4:09 PM Pagc-xxiv 


Centers of Ahl-e Sunnat (Barelwi) Influence in the 
late 19th and early 20th centuries in North India 

250 miles 


Muradabad. # Rampur 
DELHI ■ . 'Pilibhit 




# Sitapur 


1 I 





Map by MAPgrafix 


chl.044 10/12/2004 4:11 PM Page 1 





[n India, strong Muslim rule under the Mughal empire gave 
way in the course of the eighteenth century to weak central 
control and the establishment of a number of regional king- 
doms "which were independent of the Mughals in all but name. 
vv They in turn soon became indebted to the East India Companv, T7 

which had started out as a trading companv in 1 600 but bv the 
earlv nineteenth centurv had assumed a number of important 
political functions, the most important of which was the col- 
lection of land taxes. In 1858, after a failed Indian revolt against 
the East India Companv, the British Crown assumed formal 
control of India and the East India Companv "was dissolved. 


For three centuries (1526 to 1857), India was ruled by the 
Mughals, who were Sunni Muslims of Central Asian descent. 
The founder of the empire was Babur (r. 1 526—30), who swept 
into India from present-dav Afghanistan, but whose brief reign 
left him no time to consolidate his gains in north India. It was 
his grandson Akbar (r. 15 56—1605) who made a lasting 


chl.044 10/12/2004 4:11 PM Page 2 


impression on India and gave the empire a firm foundation. 
From their capital in the north (Delhi for the most part, though 
Akbar chose Agra and other cities as well), the Mughal emper- 
ors expanded the border in all directions. Starting from the 
northwest, including what are today Afghanistan and Pakistan, 
the empire expanded eastward to the Gangetic plain during 
Akbar 's reign, going as far as 'what is today Bangladesh. To the 
north, the Himalayan mountains constituted a natural border, 
preventing further conquest in that direction. Central and 
southern India "were ruled by independent kings, some Hindu, 
some Muslim, until well into Akbar 's reign. In fact, the south 
was not incorporated into the Mughal empire until about a cen- 
tury later, during Aurangzeb's long reign (r. 165 8— 1707), and 
even then the very southern tip of India remained independent. 
It was an agrarian empire, centered around the person and 
authority of the king. Land taxes constituted its main source of 
revenue. Since the majoritv of the Indian population was 
\7" Hindu, during his fifty-year rule Akbar set about winning —kzj~ 

hearts and minds by including Hindu princes in all branches of 
government and even bv marrying Hindu princesses. His 
eldest son, Salim (later Emperor Jahangir) had a Hindu mother, 
as did his grandson, Emperor Shah Jahan. At the same time, he 
showed his respect for popular Muslim religious figures. He 
paid homage to a particular lineage of Muslim mystics, or sufis, 
whose hospice was in the western Indian city of Ajmer. A story 
is told of how in 1 570 he walked from his capital Fatehpur Sikri 
(near Agra), in the north, to Ajmer in the west, a distance of 
about two hundred miles, in a gesture of thanksgiving after the 
birth of his son Salim. Ajmer was the burial place of a thir- 
teenth-centurv sufi whose intercession with God, the emperor 
believed, had been instrumental in his son's birth. In the first 
half of his reign, he also sponsored pilgrim ships from India's 
west coast to Mecca, sending generous gifts to that city. In 
sum, Akbar's religious eclecticism and inclusiveness helped 


chl.044 10/12/2004 4:11 PM Page 3 


Indianize the foreign Mughals and strengthened and stabilized 
the empire. (In the second half of his reign, Akbar encouraged a 
personality cult around himself, inventing a new "religion" 
with elements of different faiths, alienating a number of 
Muslims as a result.) 

Mughal decline began in the late 1600s during the reign of 
Akbar's great-grandson Aurangzeb (r. 165 8— 1707) and accel- 
erated throughout the eighteenth century. Although historians 
argue about what caused the decline, a number of factors "were 
at work: Aurangzeb 's reversal of Akbar's religious policvisheld 
by some to have been crucial, for he alienated a number of 
Hindu princely families bv excluding them from positions of 
power and imposing on them a tax which Akbar had abolished 
(the jizya). In fact, the second half of Aurangzeb 's reign was 
spent in incessant and, in the end, futile warfare against a minor 
Hindu chieftain, Shivaji (d. 1680) , who eventually carved out a 
small kingdom along India's west coast and expanded it bv war- 
\7" fare and diplomatic alliances with other Hindu rulers. In time, —kzj~ 

he and his successors (collectively known as the Marathas) 
were even able to challenge the Mughals in the north, the cen- 
ter of Mughal power. The financial drain of Aurangzeb 's mili- 
tary campaigns on the empire's resources contributed to the 

After Aurangzeb 's death in 1 707, the eighteenth century saw 
a succession of weak rulers. This encouraged foreigners to 
invade or try to take over. In 1 739 the military general-turned- 
emperor Nadir Shah invaded north India from Persia in the 
west. Taking Lahore (now in Pakistan) injanuarv of that year, he 
proceeded to march into Delhi a few months later. According 
to Juan Cole, "the savage looting of the capital later perpetrated 
by his troops constituted] one of the century's great disasters" 
(1989:41) .The next major attack was launched bv the Afghans, 
also from the northwest. In 1761, Ahmad Shah Abdali (later 
stvled "Durrani") fought the Marathas at Panipat, fiftv miles 


chl.044 10/12/2004 4:11 PM Page 4 


northwest of Delhi, where two important battles fought in 
earlier centuries had given the Mughals control over India. 
This, the Third Battle of Panipat, was "won by Ahmad Shah, and 
could have led to Afghan rule over India had Ahmad Shah's 
troops not been wearv of war and anxious to return home. The 
power vacuum in Delhi was soon to be filled bv vet another 
foreign power, the British East India Company. 


Apart from the foreign threats to the Mughal empire in the 
eighteenth century, it was also subject to internal fissures. In 
north India, one of the most significant new developments "was 
the rise of Shi 'ism as the state religion in two of the largest 
Muslim successor states, Bengal andAwadh (known as"Oudh" 
in British sources) . The kingdom of Awadh was founded by 
-(^y— Burhan ul-Mulk in 1722 and was centered in Lucknow in the — (^y- 

Gangetic plain. It grew in power under the first three gover- 
nors or nawabs (Burhan ul-Mulk, Safdar Jang, and Shuja ud- 
Dawla) over the next fiftv vears. After Nadir Shah's invasion in 
1739, the Mughal emperors were probably less powerful than 
the nawabs of Awadh. Although Shuja ud-Dawla, the third gov- 
ernor, stopped short of proclaiming Awadh's total independ- 
ence from Mughal rule, continuing to mint coins in the 
emperor's name and having the Friday sermon read in his 
name, for all practical purposes the state operated independ- 
ently. State affairs (diplomacv, economic policv, the appoint- 
ment of officials and successors to the governorship) were 
conducted without reference to the Mughal emperor. 

As both Cole and Francis Robinson (2001) explain, the cul- 
ture of the Bengal and Awadh courts was fed bv a constant 
influx of Shi'i Muslims from Iran and Iraq. Indeed, the govern- 
ors of Awadh were themselves of Iranian (Nishapuri) origin. 


chl.044 10/12/2004 4:11 PM Page 5 


This "was a time of political instability in Iran as well, and 
Iranians and Iraqis of all professions were eager to seek their 
fortunes in either Bengal or Awadh. Spear speculates that had 
the British East India Company not intervened in India in the 
mid- 1700s, the governors of both states would probablv have 
tried to consolidate their power at the expense of the Mughals 
and/or each other, but would then have had to deal with the 
Marathas (Spear, 1981: 76—77). But British intervention pre- 
vented the plaving out of this rather dismal scenario. 

Despite Awadh's fairly rapid political decline in the latter 
part of the eighteenth centurv, Shi'ism continued to influence 
the political and cultural landscape of the eastern Gangetic 
plain throughout the eighteenth and nineteenth centuries. 
Indeed, at the end of the nineteenth century Ahmad Riza Khan 
wrote frequently about the negative influence of Shi'ism in 
his home territory of Rohilkhand, west of Lucknow, urging his 
followers to refrain from participating in Shi'i rituals and 
\7" practices. ~x3~ 


Closer to home for Ahmad Riza Khan and his family is the his- 
tory of the Rohilla Afghans, after whom the region earned its 
name, Rohilkhand. The region around Bareillv, Ahmad Riza 
Khan's birthplace, 'was (and is) known as Rohilkhand, having 
been settled by the Rohilla Afghans in the seventeenth and eight- 
eenth centuries. In the mid-eighteenth century Rohilkhand 
came under the authority of Hafiz Rahmat Khan (d. 1774) who 
was a forceful and strong leader who might have succeeded in 
making Rohilkhand a lasting regional power had there been 
fewer players vying for control over north India. 

But this was not to be. Instead, the constant state of warfare 
finally forced the Rohillas to seek Awadh's help in order to beat 


chl.044 10/12/2004 4:11 PM Page 6 


back the Marathas. This in turn gave Awadh, under Shuja 
ud-Dawla, an excuse to take over in 1774 after Hafiz Rahmat 
Khan was killed in battle. Rohilkhand was thus absorbed into 
the state of Awadh. But bv this time Awadh had become 
financiallv dependent on the East India Companv (indeed, the 
latter had helped Awadh in its annexation of Rohilkhand). 
Consequently, in 1801 Awadh had to cede Rohilkhand to the 
East India Companv as part repayment of its debt. It was to 
remain under Companv rule until 1 858, when India became a 
part of the British Empire and Rohilkhand became part of the 
new state known as the Northwest Provinces. 

Economically, it is important to note, Rohilkhand had 
enjoved considerable prosperity in the early period of its his- 
tory under Afghan Rohilla rule. A rich alluvial plain in the 
foothills of the Himalayas, it was deemed one of the most fer- 
tile regions in the subcontinent in the earlv eighteenth centurv. 
But after it came under Awadh's rule, heavv revenue demands 
\7" — made by Awadh in order to pav off its own debts to the East —kzj~ 

India Company — impoverished its people. Subsequently, simi- 
lar demands by the East India Company led to indebtedness and 
rackrenting in the countryside. 

Meanwhile, another facet of the political situation in the late 
eighteenth and early nineteenth centuries is of interest to us as 
part of Ahmad Riza Khan's family background, namely, the cre- 
ation and growth of the independent state of Rampur, north- 
west of Bareillv. Rampur was the only Rohilla principality to 
survive the vicissitudes of the times and to continue to enjoy 
independence as a princely state under British rule. Rampur 
state was created by Faizullah Khan, who had fought by Hafiz 
Rahmat 's side for over twenty years "when the latter was killed 
in 1774, and had a reputation for bravery and leadership. Thus 
the mantle of leadership naturallv fell to him. 

However, as Rohilkhand had just been absorbed into Awadh, 
he had no territorial base of his own. Warren Hastings, then the 


chl.044 10/12/2004 4:11 PM Page 7 


Governor of Bengal, concluded a treaty with him, granting him 
the small estate of Rampur (about 900 square miles) situated 
north of the citv of Bareillv. Faizullah Khan thus became the first 
nawab of Rampur. Interestingly, although the Rampur nawabs' 
ancestors were Afghans — and Sunni Muslims — after the 1 840s 
most of the Rampur nawabs were followers of Shi 'ism. 

Bv around 1800, the Marathas were no longer a threat in 
north India, having retreated to western India and split up into 
four separate confederate states, each of whom owed alle- 
giance to the confederate chief or Peshwa in Pune. Bengal and 
Awadh had by now both come under the political and eco- 
nomic control of the East India Company: Bengal succumbed at 
the Battle of Plassev, thereby setting in motion East India 
Company rule over much of India for the next hundred years. 
Awadh, itself formally under Mughal rule, became increasingly 
indebted to the Company, and gradually, from 1775 to 
1801, ceded parts of its territory to the British after Shuja 
\7" ud-Dawla's death in 1774. Indirect rule over Awadh bv the —kzj~ 

British was to continue until its formal annexation in 1856, a 
year before the Revolt. 


If in the eighteenth century the British were one of several con- 
tenders for power in the wake of Aurangzeb's death and the 
weak rule of his successors, in the nineteenth they were 
unquestionably the most important power holders in India. This 
was not at all what they had intended, for the British had come 
to India as traders rather than as conquerers. The East India 
Companv, formed in 1600, was but one of several European 
trading companies to come to India in search of "exotic" items 
of trade — chiefly spices, but also silks, fine handspun cottons, 


chl.044 10/12/2004 4:11 PM Page 8 


saltpetre (which had military uses), and other items. The other 
companies were Portuguese, Dutch, and French. The weak- 
ening of the Mughal empire in the first half of the eighteenth 
centurv coincided with European rivalries at home (during the 
Napoleonic wars, the War of Austrian Succession in the 1740s, 
and the Seven Years' War in the 1750s and 1760s) between the 
British and French, leading them into proxy wars in Bengal and 
south India. In the 1760s Robert Clive of the East India 
Company secured the diwani or revenue-collection rights to 
large parts of Bengal and Bihar, after driving the French out of 
south India. Gaining the right to the diwani was a milestone, for 
it allowed the British to pay for their purchases with Bengal's 
tax revenue, thereby making the annual export of bullion from 
Britain to India unnecessary. 

Despite safeguards against abuse (for Clive and his men 
made small personal fortunes after their victories at Plassev 
and Buxar in the 1 75 0s and 1 760s) put in place bv the Board of 
\7" Directors in London, Company officials continued to enrich —kzj~ 

themselves personally until forbidden to engage in private 
trade in the 1790s. In 1813 missionaries and private traders 
were allowed into the country bv an Act of Parliament, and in 
1833 the Company lost its monopolv on trade in everything 
but opium and salt. Conquest of further territorv, some direct 
and some indirect, followed swiftlv during the first half of the 
nineteenth century. Under the pattern of indirect rule set in 
place by Clive, local rulers "were allowed to retain their thrones 
but forced to concede to certain vital annual demands for 
revenue, which ultimately drove them into crippling debt to 
the Company. This in turn led, in due course, to the East India 
Company's assumption of political power. 

A few dates will suffice to illustrate the pace of British 
annexation of territorv, for these events are well known: in 
1 801 Madras Presidency "was formed in the south, in 1 803 the 
British defeated the Mughal emperor in Delhi and made him a 


chl.044 10/12/2004 4:11 PM Page 9 


pensioner, in 1818 parts of the Maratha confederacy were 
taken over to form the bulk of Bombav Presidency, in 1 848 the 
Panjab was annexed, in the 1 840s seven princely states were 
taken over in as many years under the Doctrine of Lapse (which 
forbade a ruler from choosing an adopted son as successor in 
the absence of a natural-born son and held that such a kingdom 
came under Company rule by default), and in 18S6 the 
Nawab of Awadh was forced to give up his throne on grounds of 

In 1 857— 58, parts of the country rose in the anti-British 
rebellion known (in British accounts) as the Mutinv, though 
it was in fact much more broad-based than a mutinv, for it 
included peasants and landlords as "well as soldiers. When the 
Revolt was finally put down in 1858, the anomaly of East India 
Companv rule was replaced bv the more normal mode of 
government called Crown rule, and the East India Companv 
was dissolved. The last Mughal emperor, Bahadur Shah Zafar, 
\7" was banished to Rangoon, Burma, where he lived out the rest —kzj~ 

of his days, after the British had murdered his sons to make sure 
there would be no heirs. (In an interesting parallel to this sad 
episode, in 1 885 the ruler of Burma was banished to western 
India for the rest of his life after the British takeover of that 

Economic Consequences of British Rule 

Alongside the sweeping political changes indicated by these 
events were profound changes taking place in the areas under 
British control in the economic, legal, educational, and other 
spheres. The economic sphere was of course central to British 
concerns, and changes here began with the Permanent 
Settlement of Bengal in 1793. The British attempt to under- 
stand local land tenure systems "was motivated bv the desire to 
increase productivity and hence annual tax revenues — this 


chl.044 10/12/2004 4:11 PM Page 1C 


being the raison d'etre for the acquisition of territory by the 
Company. Seeing private property as the kev to creating a class 
of "improving" landlords on the British model and hence to 
ensuring future agricultural productivity, in 1793 Lord 
Cornwallis, as Governor-General of Bengal, conferred private 
ownership rights in perpetuitv on a number of Bengal 
zamindars or landlords, "who "were required in return to meet a 
"high and inflexible" annual revenue demand (Metcalf and 

However, the experiment failed. Under the indigenous svs- 
tem, the zamindar could "sell or transfer onlv his own revenue 
collecting rights, not the land itself, for that did not belong to 
him." If the peasants on his land felt overburdened, thev could 
move to another part of the countrv where conditions were 
better. However, under the new system, all the zamindars, now 
owners of the land and liable to high taxes in good and bad 
years, were under an onerous burden themselves. Manv were 
\7" unable to pav the taxes and had no choice but to put their —kzj~ 

estates up for sale. Far from being improving landlords, a num- 
ber of them sold their land to city-dwelling magnates who had 
the money to treat their estates as an investment (though with- 
out any incentive to "improve" them) at the tenants' expense. 
As for the tenants, they "were reduced to the status of "tenants 
with no rights" (Metcalf and Metcalf, 2002: 77) and few 
options. Squeezed bv the tax burden from above and unable to 
find better terms elsewhere (as all the landlords now enforced 
the same high revenue demand), in time they became a class of 
landless bonded labor. A third of the estates are believed to have 
changed hands in the first twenty years following the 
Settlement of 1 79 3 . 

In the years ahead, various alternatives were tried in other 
parts of India, ranging from assessments being revised every 
twenty or thirty vears to ownership being fixed on the tenants 
(ryots) rather than the landlords, in south India. Meanwhile, 


chl.044 10/12/2004 4:11 PM Page 11 


new crops 'were introduced in Bengal and elsewhere. A highly 
profitable trade in opium was started in Bengal in the 1 820s for 
export to China. This complicated three-way trade allowed the 
Company to pay for its exports from China with the proceeds 
of opium sales in China, once again making unnecessary 
the export of bullion from Britain. Opium "provided up to 
1 S per cent of the Indian Government's total revenue" in the 
1830s (Metcalf andMetcalf, 2002: 75). 

Another important economic change in India in the earlv 
1 800s was the substitution of factory-made British textiles for 
handmade Indian cloth, which put Indian weavers in Bengal 
out of work and increased pressure on the land. The destruc- 
tion of the textile industry followed — British-made textiles 
being cheaper to buy than the local product — initiating "the 
development of a classicallv 'colonial' economv, importing 
manufactures and exporting raw materials, [in a pattern] 
that was to last for a century, until the 1920s." The Metcalfs 
\7" conclude, "Overall, . . . the East India Companv during the early —kzj~ 

decades of the nineteenth centurv did little to set India 
on a path of economic growth" (Metcalf and Metcalf, 2002: 

Improvements in Infrastructure 

While the economy was dramatically affected in these and 
other wavs during the earlv nineteenth century, infrastructural 
developments had a lasting impact on Indians of all classes and 
communities. Railroads and the telegraph were introduced 
during Lord Dalhousie's governor-generalship in thel850s, 
and a postal system and print technology were introduced, 
making newspapers and periodicals available relativelv cheap lv. 
To give but one example of how significant a change the "penny 
post" represented, "[i]n the 1830s an exchange of letters 
between Britain and India could take two years; bv 1 870, with 


chl.044 10/12/2004 4:11 PM Page 11 


the opening of the Suez Canal [in the 1860s], a letter could 
reach Bombay in only one month" (Metcalf and Metcalf, 2002: 
97). Print technology, as we will see in subsequent chapters, 
greatly facilitated the growth of Islamic (as well as other 
religious) reform movements in the latter half of the nineteenth 

Change was occurring in almost every area of life at this 
time. From the point of view of the north Indian 'ulama, 
two areas were particularly significant, namely, education and 
the law. 

A British Model for India 

Having become the colonial masters of India, the British had to 
decide what direction they "wanted the country to take. What 
was the British purpose in being in India, what did it hope to 
achieve other than the economic and imperial goals of hegem- 
ony? Answering this question also involved assessing the Indian 
past. Did those who now governed India see anything of value 
in India's linguistic and literarv heritage, its educational trad- 
itions, its legal texts, and so on, or should Britain set in place a 
whollv Western system, a wholly new set of institutions that 
had no local roots "whatsoever? 

This debate plaved out most famously in the fields of law and 
education. Among those who spoke for the liberal position (the 
term meant something different in the 1 800s than it does in US 
politics in the earlv twenty-first century) were Lord William 
Bentinck, Governor-General of India in the late 1 820s and early 
1830s, and John Stuart Mill "who worked for the East India 
Company from 1823 until 1858. Mill argued that different 
peoples were at different stages in the "ladder" of "progress" but 
could be advanced along the way by means of education and good 
government. Charles Trevelvan, who served in India in the 


chl.044 10/12/2004 4:11 PM Page i: 


1 830s, 'was also among this group. He believed that British learn- 
ing and institutions would put India on the path to "moral and 
political improvement" (Metcalf, 2001: 28—33). Thomas B. 
Macaulav is perhaps the best-known example of this position. In 
a well-known statement in 1835, he said he wanted to put in 
place a system of education which would "create not just a class 
of Indians educated in the English language, who might assist the 
British in ruling India, but one 'English in taste, in opinions, in 
morals and in intellect' " (Metcalf, 2001 : 34) . 

As Thomas Metcalf points out, this view was based on the 
belief that all "races" "were inherentlv educable and none had to 
remain perpetuallv on the lowest rung of the ladder of "civ- 
ilization ."But it was also based on a negative assessment of non- 
European cultures and their traditions of learning. Thus 
Macaulav is famous for his dismissal of the "entire literature of 
India and Arabia ... [as worth less than] a single shelf of a good 
European librarv" (Metcalf, 2001: 34). This negative assess- 
\7" ment of India stood in contrast with the views of an earlier gen- —kzj~ 

eration of officials and scholars such as Governor-General 
Warren Hastings and Sir William Jones, a judge in the East India 
Companv (d. 1794), who believed that India's rich textual 
tradition was worthy of studv bv Europeans (which in turn 
required the studv of languages, chieflv Sanskrit, but also 
Arabic for an understanding of Muslim texts) , and that the 
British could best rule by basing their laws on those of the 
countrv itself. 

"The outcome of British studv of the ancient texts, in Jones's 
view," Metcalf writes, "was to be a 'complete digest' of the 
Hindu and Muslim law, which could be enforced in the 
Companv's courts, and would preserve 'inviolate' the rights of 
the Indian people" (Metcalf, 2001 : 1 2). It was in this spirit — as 
well as a desire to be independent of Brahmin interpreters he 
considered unreliable — that Jones worked on his Digest 
(Metcalf, 2001 : 24). It was published in 1798 bv his successor 


chl.044 10/12/2004 4:11 PM Page 14 


H.T. Colebrooke.To be sure, Jones also graded European and 
"Oriental" learning and their laws on a hierarchical scale in 
which the European was superior to the Indian. Furthermore, 
in his view of things, India's glorious past or "golden age" had 
given wav to a state of decline in the present. Nevertheless, he 
differed fundamentally from Macaulav and others of like mind 
who devalued and belittled the Sanskritic and Arabo-Persian 
literary traditions altogether, and who sought to base British 
laws and education in India on Western traditions alone. 

In the end, Macaulav and Mill carried the dav, though with- 
out totallv rejecting Hastings' and Jones' vision. Since 1772 
civil law had been based on religious affiliation, for Hindus and 
Muslims were governed bv their own personal laws — Hindu 
law for the former, and "Anglo -Mohammadan" law for 
Muslims. In practice, the manner in which Islamic law was 
implemented was much altered under the East India Companv. 
As Zaman shows, certain medieval texts deemed authoritative 
\7" bv the British were "invested with almost exclusive authority as ~x3~ 

the basis of judicial practice in British courts, as far as Muslim 
personal law "was concerned" (Zaman, 2002: 22). Moreover, 
the manner of their application was more rigid than it had been 
in Mughal times, in keeping with the British desire to impose 
uniformity and predictabilitv in the law. (Zaman points out that 
the British were inconsistent in their application of the law too, 
but this was described as exercising "discretion" rather than 
being "arbitrary.") 

Macaulav was instrumental, in the 1 860s, in drafting a new 
penal code which replaced what the British saw as despotic 
"Oriental" rule with "predictable rules and regulations for the 
adjudication of disputes." Based on Jeremy Bentham's prin- 
ciples of utilitarianism, the new laws also sought to promote 
"unity, precision, and simplicity" (Metcalf, 2001: 37, 38). 
Islamic criminal law ceased to be applied in the courts after this 
time. Moreover, the muftis (and Brahmin pandits) who had 


chl.044 10/12/2004 4:11 PM Page 



been employed to help British judges on matters of personal 
religious law were no longer deemed necessary, and the pos- 
ition of "native law officer" was abolished in 1864. Qadis (judges 
who applied Islamic law) were frequently not appointed to 
British Indian courts either. Thus the application of Anglo - 
Muhammadan law in British Indian courts "was often in the 
hands of non-Muslim judges. This made even simple matters 
such as the dissolution of a marriage, for example, impossible, 
as such a decision "was invalid in Muslim eyes if made by a non- 
Muslim judge (Zaman, 2002: 25, 27). In the nineteenth cen- 
tury the Deobandi 'ulama tried unsuccessfully to create an 
alternative court of their own, but for a variety of reasons many 
Muslims continued to use the British Indian courts (Metcalf, 
1982: 147). As we shall see throughout this book, the primarv 
response of the 'ulama to the loss of access to the courts under 
British rule was to issue responsa (fatawa).The other alterna- 
tive was to take the issue under dispute to a Muslim princely 
\7" state "where British laws were not in place and where a qadi —kzj~ 

could be found. 

Education was also a significant issue for the 'ulama during 
British rule. During the late eighteenth century, Orientalist 
scholars such as William Jones had promoted schools for the 
education of maulwis and pandits who could assist Company 
officials in the interpretation of Hindu and Muslim law, respect- 
ively. Among the best-known schools of this period "were the 
Calcutta Madrasa (founded in 1781), the Sanskrit College in 
Benares (founded in 1 792) , and Delhi College, whichhad origin- 
ated as a madrasa during Aurangzeb's reign. Although the 
focus in all these schools was on "Oriental" learning, Delhi 
College also taught its students Western sciences and mathe- 
matics through works translated from English into Urdu. In 
addition, Lord Welleslev, Governor-General from 1798 to 
1805, established the College of Fort William in Calcutta in 
1 800 to teach young British recruits to the Company Indian 


chl.044 10/12/2004 4:11 PM Page 1( 


languages and Hindu and Muslim law, as well as Western sub- 
jects, before sending them out into the countryside as adminis- 
trators. Less well known is the College at Fort St. George, 
Madras, founded in 1812 by Francis Ellis, "which trained the 
British in Indian languages and Indians in Hindu and Muslim 
law simultaneously (Cohn, 1996: 47—53). 

When this Orientalist approach gave wav, from the 1820s, 
to the supporters of "reform" and "liberalism," the purpose of 
education became to instil British values. In 1835 Macaulav 
wrote in his policv statement or "Minute on Education" that the 
goal of British education in India should be to create a class of 
Indians who would be "English in taste, in opinions, in morals 
and in intellect." If this meant that down the road they would 
also "want self-rule — as he thought thev must — this was to be 
"welcomed, for the new political order would be one that 
represented "an imperishable empire of our arts and our 
morals, our literature and our laws" (Metcalf and Metcalf, 
\7" 2002: 81). In anv event, this was a distant prospect, not —kzj~ 

something that British policv makers needed to worrv about 
there and then. 

The immediate consequences of Macaulav's educational 
blueprint included, in 1835, the substitution of English for 
Persian as the language of government. Under the reform- 
minded Governor- General , Lord Bentinck (182 8—3 5 ) , several 
colleges were founded, though no effort was made to set up 
elementary schools. In Britain at this time, schools were run bv 
parochial (religious) bodies, not by government. Among the 
universities that date to this period are Patna College. 
Elphinstone College was founded in the 1820s in Bombav. 
Hindu College in Calcutta had been established in 1819, with 
private British and Indian financial support. Bv the 1830s 
English was being avidlv studied by "several thousand Indians" 
in Calcutta alone (Metcalf and Metcalf, 2002: 82). The first 
three Indian universities were inaugurated in 1 857. 


chl.044 10/12/2004 4:11 PM Page iy 


This embrace of Western learning was but an aspect of a 
wider reform movement under Raja Ram Mohan Rov, founder 
of the Brahmo Samaj in Calcutta, who sought to reform the 
Hinduism of his dav in the light of a perceived golden age 
accessed through the studv of ancient Sanskrit texts. That 
earlier form of Hinduism, for Rov, was characterized bv 
rationalism and simplicity rather than the idol worship of con- 
temporary times. David Kopf has characterized this era as the 
"Bengal Renaissance" on account of its spirit of enquirv and its 
openness to reinterpretation of received tradition (Kopf, 

"0" -& 


chl.044 10/12/2004 4:11 PM Page 18 




ch2.044 10/12/2004 5:11 PM Page 




Indians of all religions were keenly aware of Western criti- 
cisms of their religious customs and traditions. The Hindu 
reformer Raja Ram Mohan Rov had responded bv rejecting 
many aspects of contemporary ritual practice, arguing that the 
"pure" Hinduism of India's "golden age" 'was rational, simple, 
and devoid of practices which the British described as barbaric 
(such as idol worship, caste, widow immolation, child mar- 
riage, and other social practices deemed detrimental to 
■women). He also considered certain Sanskritic texts authorita- 
tive, and advocated their studv as a means of reforming reli- 
gious and social practices. 

In the Muslim case, religious leaders in the eighteenth and 
nineteenth centuries promoted internal reform as a response to 
Britain's rule of India. They reasoned that if Muslims had lost 
political power after so many centuries of rule, it "was because thev 
had been religiouslv negligent. Had thev been "good" Muslims, 
they "would have been strong and the British would never have 
been able to take over. Specificallv, Muslim reformers advocated 
greater individual adherence to religious precepts as set out in the 
shari'a, greater knowledge of the religious texts by the 'ulama and, 
to some degree, by ordinary believers, and a focus on the 
Prophet as a model of behavior in one's daily life. A related con- 
cern was with preaching (dawa), mainlv to other Muslims, to 
encourage greater religiositv. Their attitudes toward two other 



ch2.044 10/12/2004 5:11 PM Page 2( 


questions — sufism (Islamic mysticism) and British rule — varied 
widely. On both issues we find everything from complete 
acceptance to total rejection. 

The reformers' emphasis on authoritative texts, namely, the 
Qur'an which Muslims regard as the literal word of God, and 
secondarily the traditions of the Prophet (hadith), led to the 
first translations of the Qur'an. The Qur'an is learned and 
memorized in the original Arabic, but in India Muslims spoke 
Persian in the eighteenth and early nineteenth centuries, or 
Urdu starting in the mid-nineteenth centurv, or a regional 
Indian language such as Bengali or Tamil. Because Arabic was 
not spoken by Indian Muslims, the Qur'an "was poorly under- 
stood. Muslim reformers in the late eighteenth and nineteenth 
centuries translated the Qur'an into Persian, and much later 
into Urdu and other Indian languages. 

The Indian reform movements also highlighted the hadith lit- 
erature. The hadith are narratives (literally, stories or news 
\7" reports) about the Prophet (d. 632), relating to something he —kzj~ 

did or said, or which tell about his appearance, comportment, 
and so on. These narratives "were orally transmitted by his fol- 
lowers to successive generations of Muslims before being writ- 
ten down about a centurv after his death. A laborious process of 
evaluation over two centuries eventually resulted in six collec- 
tions of hadiths, named after the jurists who had collected them. 
The collection regarded as the most reliable is that of al-Bukhari 
(d. 870), with that of Muslim (d. 875) the next most reliable. 

The focus of the hadith literature is the Prophet, and all the 
Indian Muslim reform movements of the eighteenth and nine- 
teenth centuries were united in their strong emphasis on the 
personality and biography of the Prophet. They saw in him a 
model for how they could, and should, live their own lives. This 
made him an example one could hope to emulate. Together 
with the Qur'an becoming a subject of scholarly discussion and 
interpretation, the view of the Prophet as a model Muslim 


ch2.044 10/12/2004 5:11 PM Page 21 


meant that the 'ulama, and through their leadership other 
Muslims, were individually responsible for fulfilling their reli- 
gious obligations as Muslims and reiving much less than before 
on intermediaries and socially accepted, customarv wavs of 
behavior. This characteristic unites all the reform movements, 
despite their great diversity in other ways. 

While the political dominance of the British in India, and their 
debates about the intrinsic value or lack thereof of "Oriental" 
learning, were a powerful impetus for reform among Indian 
Muslims, there was also another source which came from the 
Islamic world itself, namelv, the Wahhabi movement in eight- 
eenth- and nineteenth-century Arabia. The influence of this 
movement on Indian reform movements "was felt through the 
annual pilgrimage to Mecca, through extensive periods of 
study bv a small number of Indian Muslims at Mecca and 
Medina, and bv the general improvement in communications 
which occurred in the eighteenth and nineteenth centuries. 
\7" The founder of the Wahhabi movement was Muhammad ibn —kzj~ 

'Abd al-Wahhab (1703—87). His message 'was an insistence on 
the unity of God (tawhid), which meant that all forms of super- 
stition (the veneration of saints' tombs, holy objects, and the 
like) "were contrary to the worship of the one God. He believed 
that the first generation of Muslims, namelv, the Prophet and 
his companions, were the models of true Islamic practice. He 
therefore rejected later developments in the history of Islam, 
particularly sufism and what he viewed as its excesses. Albert 
Hourani (1983: 37) describes his ideas as follows: 

The true Islam, stated Ibn 'Abd al-Wahhab, was that of the 
first generation, the pious forerunners, and in their name he 
protested against all those later innovations which had in fact 
brought other gods into Islam: against the later development 
of mystical thought, with its monist doctrines, its ascetic 
renunciation of the goods of the world, its organization into 
brotherhoods, its rituals other than those prescribed by the 


ch2.044 10/12/2004 5:11 PM Page 22 


Quran; against the excessive cult of Muhammad as perfect 
man and intercessor with God (although great reverence was 
paid to him as Prophet) ; against the worship of saints and 
reverence for their shrines; and against the return into Islam 
of the customs and practices of the [pre-Islamic age]. 

Although the precise influence of Ibn 'Abel al-Wahhab on 
Indian Muslim reformers of the eighteenth and nineteenth 
centuries is a matter of scholarly debate, there is no doubt that 
his ideas were well known and that thev plaved a major role in 
the thought of some religious thinkers in India. 


In the eighteenth century, the figure of ShahWali Ullah Dehlawi 
(1703—62) stands out as preeminent. The progressive collapse 
of central authority in Delhi caused him to plead with Muslim 
\7" leaders in Rohilkhand and in south India to do something to ~X7~ 

restore order. In his anxiety to see a strong Muslim ruler, Shah 
Wali Ullah even invited Ahmad Shah Abdali of Afghanistan to 
invade and take over. He must have been pleased with the out- 
come of the Battle ofPanipatin 1761 , which resulted in Ahmad 
Shah's victory over the Marathas and held out the hope of stable 
central government from Delhi. But he died the next vear, and 
as we know, that battle did nothing to settle the question of cen- 
tral rule as Abdali returned to Afghanistan, leaving a power vac- 
uum in his wake . 

However, ShahWali Ullah is remembered chieflv for his con- 
tribution to religious rather than political matters. His father, 
Shaikh 'Abd ur-Rahim (1644 — 1718), had established a sem- 
inary or madrasa, the Madrasa-i Rahimivva, in Delhi, and this 
was "where he spent his lifetime — as director of the school, as 
teacher, and as thinker and writer. His chief contribution to 
Islamic studies was to insist on the importance of the studv of 


ch2.044 10/12/2004 5:11 PM Page 21; 


hadith (pronounced hadis in Urdu), the traditions of the 
Prophet, and to argue that the 'ulamahad an obligation to study 
the original sources (the Qur'an and hadith) and draw on all 
four Sunni schools of law (madhhab, pi. tnadhahib) eclecticallv 
to make legal judgments. 

The four Sunni law schools ( Shi 'i Muslims have three of their 
own) came into being around the late tenth century. Named 
after their founders, thev are geographically based, such that 
different parts of the Muslim world have come over time to be 
associated with one or other of the four. In India, the predom- 
inant school is the Hanafi, named after Abu Hanifa of Iraq 
(d. 767) .The schools are distinguished bv minor differences of 
judgment between them. In this book, for example, we will see 
the case of a scholar combining the judgments of two different 
schools of law in a case relating to apostasv and marriage. 
However, most Indian Muslim scholars (including Ahmad 
Riza), frowned upon such practice. 
\7" The founding of the four law schools had the general effect —kzj~ 

of making it unnecessary for jurists to go directlv to the sources 
(the Qur'an and the prophetic traditions), allowing them 
to relv instead on the judgments of the founding jurists on 
major issues. Muslim scholars metaphorically refer to this 
development as the "closing of the gate of ijtihad," or inde- 
pendent reasoning. Thus, once the medieval jurists had 
judged something to be forbidden or permitted, based on the 
guidance of the Qur'an and prophetic traditions, all that later 
generations of scholars had to do was to follow in their 
footsteps. Thev no longer had to consult the original sources 

But while this was generally the case, in fact independent 
reasoning never ceased as new issues constantly arose, needing 
fresh interpretation and judgment by the 'ulama. Shah Wali 
Ullah contributed to Islamic reform in eighteenth-century 
India by reminding the Indian 'ulama of their obligation to 


ch2.044 10/12/2004 5:11 PM Page 24 


make legal judgments in light of the original sources, choosing 
between the judgments of the four schools when thev 
deemed this to be necessary, rather than following the one that 
was customary in their part of the world. "His espousal of 
jurisprudential eclecticism combined with consultation of 
Qur'an and hadis clearlv enhanced the responsibility of the 
'ulama for interpreting the Law to their followers" (Metcalf, 
1982: 38). It is important to note, however, that this obliga- 
tion was limited to the learned, the khawass, to the exclusion 
of the ordinary ( 'amm) believer. Most Muslims, including 
most 'ulama, were urged to follow Hanafi law exclusively; only 
a few were encouraged to engage in ijtihad along the lines 

In Islamic terms, the study of hadith is part of the branch of 
studv known as manqulat, or the traditional sciences (from the 
Arabic root nql, to transmit, hand down), in contrast with the 
ma 'qulat or the "rational" sciences (cf. Arabic 'aql, meaning rea- 
\7" son, rationality) which include subjects such as philosophy. —kzj~ 

ShahWali Ullah's espousal of the traditional sciences stands in 
contrast to other schools of religious scholars (including 
Ahmad Riza Khan) to be discussed shortly. In his view, the 
rational sciences were a source of confusion and should be 
avoided. The study of hadith, on the other hand, "would bring 
Muslims closer to the sources of their tradition and thereby 
strengthen and unite the communitv. Likewise, he encouraged 
the 'ulama to studv the Qur'an directly as "well, and to this 
end he translated it from Arabic into Persian. At the time, this 
"was an act of great courage which elicited much criticism from 
the 'ulama. 

Shah Wali Ullah is also known for his contributions to a 
major issue in sufism, namely, the theory of the unity of being 
(wahdat al-wujud) versus the unity of witness (wahdat 
al-shuhud) . This debate had been ongoing among sufis in India 
since the seventeenth century. The wujudi position is identified 


ch2.044 10/12/2004 5:11 PM Page 2 



with Ibn al-'Arabi (d.1240), the famous Andalusian sufi of the 
thirteenth centurv. Ibn al-'Arabi had argued that creation has 
no empirical existence in and of itself, that it is but an aspect of 
God Himself. It follows logically from this position 
that human beings themselves are but an emanation of God, not 
independent of Him. Critics of this theory, of "which there "were 
many, argued that this position denies tawhid, the Oneness of 
God, for it makes humans the partners of God. Shah Wali 
Ullah's view on this subject was to argue that the two positions 

were less at variance with one another than is commonlv 


believed. "The whole universe is pervaded by a common exist- 
ence, he argued, an existence both immanent and transcen- 
dent, but bevond that existence is the Original Existence of 
God" (Metcalf, 1982: 40). However, Shah Wali Ullah believed 
that the subject was too subtle to be discussed publiclv, and he 
urged caution in the matter. According to Metcalf, his espousal 
of the wujudi position led to its wide acceptance bv later gener- 
\7" ations of Indian sufis. —kzj~ 

Shah Wali Ullah also sought to reconcile Sunni and Shi'i 
Muslims, at a time of increased Shi'a influence in the regional 
courts at Awadh and Bengal. He venerated ' Ali, as did the Shi'i, 
but held that the first two caliphates (those of Abu Bakr and 
'Umar) were superior to the last two (those of 'Uthman and 
'Ali), because the Muslims had been politicallv united during 
their rule. Although this attempt at bridge -building was not 
verv successful, Shah Wali Ullah's achievements in other 
respects — his emphasis on hadith studies, his scholarlv 
output as an 'alim, and his high attainments as a sufi — were 
remarkable. Particularly important was his role in renewal of 
the law, as demonstrated bv his emphasis on ijtihad. In the fol- 
lowing centurv, his work was continued bv his four sons, espe- 
cially Shah 'Abd ul-Aziz, whom the Ahl-e Sunnat regarded as 
the Renewer of the thirteenth Islamic centurv. 


ch2.044 10/12/2004 5:11 PM Page 2{ 



While Shah Wali Ullah taught and wrote from his Madrasa-i 
Rahimivva in Delhi, another group of Sunni 'ulama, known as 
the Farangi Mahallis, were making their mark in Lucknow at 
the same time. Their residence in Lucknow began when, in 
1 695 , Emperor Aurangzeb granted the four sons of Mulla Qutb 
ud-Din (d. 1692) the house of a European merchant (hence the 
name "Farangi Mahall," or foreigner's house) in recompense for 
their father's murder and loss of the family's library to arson. In 
the eighteenth century the third son, Mulla Nizam ud-Din, 
devised a new madrasa curriculum which came to be known as 
the Dars-i Nizami. Madrasas all over India gradually adopted 
this syllabus. The madrasa at Farangi Mahall became a center 
for learning on a par with the Madrasa-i Rahimiyya. 

Unlike the latter, the Farangi Mahall madrasa focused on 
r~) ma 'qulat or rational studies. Francis Robinson, who has made — (-~\- 

an exhaustive study of the Farangi Mahall 'ulama, shows in 
detail the differences between the curricula followed by the 
two madrasas. Where the Madrasa-i Rahimivva emphasized 
hadith, the Farangi Mahall curriculum emphasized grammar, 
logic, and philosophy (Robinson, 2001: 46— S3). The Farangi 
Mahall 'ulama believed that knowledge of these sciences was 
"crucial to the studv of legal theory and jurisprudence (usul 
al-Jiqh) and of theology ('ilm al-kalam), and expertise in them 
helped make many ... other disciplines accessible" (Zaman, 
2002: 76). They also de-emphasized the study of sufism.The 
reason, Robinson explains, was that the 'ulama at Farangi 
Mahall were seeking to train future 

lawyers, judges and administrators . . . [whose] skills were in 
demand from the increasingly sophisticated and complex 
bureaucratic systems of seventeenth- and eighteenth-century 
India. . . . The emphasis of the [curriculum] on training 


ch2.044 10/12/2004 5:11 PM Page 21r 


capable administrators for Muslim states rather than 
specialists in "religion" per se may explain the dropping 
of mysticism from the course. Knowledge of Sufism was 
not what trainee administrators wanted. (Robinson, 

In practice, the curriculum was flexible within the overall 
framework initially set out bv Mulla Nizam ud-Din. Zaman 

Onlv in the latter half of the nineteenth century, and . . . 
possibly in response to a certain measure of influence 
exercised by Western styles and institutions of education in 
British India, did the Dars-i Nizami acquire a more or less 
standardized form that was widely adopted as a "curriculum" 
by madrasas of the Indian subcontinent. Madrasas have 
continued, however, to differ in their versions of this 
curriculum, which has scarcely been impervious to change 
~\_y~ even after its standardization in the late nineteenth century. K^J 

(Zaman, 2002:68) 

The point that the curriculum of the Dars-i Nizami was more 
rather than less flexible before British influence made itself felt is 
interesting and "worth noting. (It also accords with what histor- 
ians know of a host of other Indian institutions, such as caste 
itself, which became relatively "fixed" and inflexible in practice 
in the later nineteenth century.) 

However, if the purpose of the Dars-i Nizami was to train 
Muslim bureaucrats to work in the Indian Muslim states in the 
late eighteenth century, the political instability of the Muslim 
successor states made the princes rather undependable as 
patrons for prospective qadis (judges in Islamic law courts) or 
muftis ('ulama qualified to issue fatawa [sing, fatwa], juridical 
responses). The same may be said for those whose skills lay in 
writing poetry, in the musical arts, or even in the military, for 


ch2.044 10/12/2004 5:11 PM Page 2£ 


that matter. Metcalf writes as follows about the difficulties 
Farangi Mahallis encountered at this time: 

Wherever there was a prince, the Farangi Mahallis sought 
positions under him. Thus in the mid-eighteenth century . . . 
three members of the family joined princely armies. The 
travels, the varieties of employment, the violent deaths of at 
least one member in each of the first four generations of the 
family — all this suggests the difficulties facing the family in 
maintaining the pattern of dependence on princes. (Metcalf, 


The nineteenth -century reformists, of which there were many 
\7" groups, shared in the broad set of goals indicated earlier, "\7 

namely, better knowledge of the textual sources of Islam 
(mainly through the creation of new seminaries for the training 
of scholars), greater adherence to religious precepts by indi- 
vidual believers, and a close modeling of their lives on that of 
the Prophet. However, thev differed in significant ways. Based 
on their attitude toward British rule, we can distinguish three 
broad groups: the vast majority (Shah 'Abd ul-'Aziz, the 
Deobandis, the Ahl-e Hadith, the Nadwat al-'Ulama, the 
Ahmadis) were relatively uninterested in participating in the 
opportunities being opened up by British rule, although most 
of them accepted it without active protest. Of this group, the 
Ahl-e Hadith were the least accommodating toward the British 
while the Nadwa and the Ahmadis were the most so. The 
jihadists (Savvid Ahmad Barelwi and his followers), on the 
other hand, were actively opposed not onlv to British rule but 
to all forms of non-Muslim rule. Thev sought to restore Muslim 


ch2.044 10/12/2004 5:11 PM Page 2i 


rule through political means, fighing the Sikhs in the Panjab, 
and the British in northwest India generally. The Faraizi move- 
ment in Bengal falls in between the two, in that although the 
Faraizis did not declare a jihad against the British, they boy- 
cotted British-run institutions and refused to pav land taxes. 
Finally, the accommodationists (Sayyid Ahmad Khan) 
embraced British rule as a positive good from which Indian 
Muslims stood to benefit. 

Ahmad Riza belongs to the first group, though his story is 
not addressed until the following chapter. 


After ShahWali Ullah's death in 1762, his eldest son Shah'Abd 
ul-'Aziz (d. 1 824) took over the management of the Madrasa-i 
Rahimiyya. Shah 'Abd ul-'Aziz followed in the footsteps of his 
illustrious father bv studving and promoting hadith scholar- 
ly ship, but "widened the circle of those he addressed through the —kzj~ 
number of fatawahe wrote for individual Muslims who sought 
his advice. The subject matter of the fatawa ranged widely from 
details regarding the proper way to perform the ritual praver to 
relations with Shi'i Muslims, and to whether it was legitimate 
to seek emplovment under the British. The increased import- 
ance of fatwa writing was a direct result of the loss of political 
power bv Muslims, which led to a greater need for personal 
guidance bv the 'ulama, now that thev no longer had state- 
based shari'a courts. 

One fatwa bv Shah 'Abd ul- 'Aziz has been particularly com- 
mented upon bv later historians on account of its political 
implications. In 1 803, he "was asked whether it was permissible 
for a Muslim to give and take interest under British rule. The 
date is important, for Delhi had been occupied bv the East India 
Company that year. Would he take foreign occupation and the 
suspension of religious law in parts of the countrv to mean that 


ch2.044 10/12/2004 5:11 PM Page 3( 


the normal rules of conduct in other spheres of Muslim life no 
longer applied either? His answer to this question was equivo- 
cal. In a land ruled bv Muslims (termed dar ul-lslam in Islamic 
law), interest (sud, or, in Arabic, riba) is prohibited. However, 
in the troubled circumstances of the early nineteenth century, 
many Muslims had fallen on hard times and "were deeply in 
debt. If Shah 'Abd ul-'Aziz were to judge on the basis of the 
Islamic sources of law (Qur'an, hadith, and the principles of 
analogy [^jjus], and community consensus or ijma) that the 
legal status of British-controlled territorv had changed (or, in 
Islamic terms, that it was dar ul-harb rather than dar ul-lslam), 
the prohibition on taking and receiving interest could be tem- 
porarily suspended. 

Shah 'Abd ul- 'Aziz's response to the question was that in 
Delhi at that time, "the Imam al-Muslimin [the leader of the 
Muslims, perhaps a reference to the Mughal emperor] wields 
no authority, while the decrees of the Christian leaders are 
\7" obeyed without fear [of retribution] .... From here to Calcutta —kzj~ 

the Christians are in complete control" (Metcalf, 1 982 : 46 ; my 
interpolation in square brackets) .While he did not directly say 
that the legal status of Delhi had changed from the abode of 
Islam to that of "war, he implied that it had, so that he could be 
understood as tacitly permitting the questioner to engage in 
interest-bearing transactions without incurring sin (Mushir 
ul-Haqq, 1969). Or, to put it another wav," 'Abdu'l- 'Aziz thus 
appears to have wanted Muslims to behave politically as if the 
situation were daru'l-islam, for he gave no call to military 
action [against the British], vet he wanted them to recognize 
that the organization of the state was no longer in Muslim 
hands" (Metcalf, 1982:51). 

This fatwa is particularly interesting because of the "way it 
has been interpreted bv Muslims in the twentieth century. It 
has been read — bv Muslim nationalists as well as Muslim 
nationalist historians — as an endorsement of jihad (holy "war) 


ch2.044 10/12/2004 5:11 PM Page 31 


against the British. Their reasoning is that if it was no longer a 
sin to take on interest-bearing debt, it could onlv mean that the 
country was under non-Muslim rule, which in turn meant that 
holv war was justified against it. However, Metcalf suggests that 
Shah 'Abd ul-'Aziz may even have opposed the jihad that was 
launched shortly before his death. At anv rate, he is known to 
have encouraged his nephew and son-in-law 'Abd ul-Hayy to 
accept a job offered to him by the East India Companv — further 
evidence, it would seem, that he did not endorse jihad. 
However that mav be, a jihad movement "was launched in 1830 
by Savvid Ahmad of Rae Bareli (a town in Awadh) . To him we 
may now turn. 

Sayyid Ahmad Barelwi 

Savvid Ahmad Barelwi (not to be confused with Ahmad Riza 
Khan Barelwi, the subject of this book) was born in 1786 to a 
\7" family that claimed descent from the Prophet Muhammad. —kzj~ 

Among Muslims, such families (known by the title "Sayyids") 
enjoy high status by virtue of their ancestry. He traveled as a 
voung man from his hometown to Lucknow in search of 'work, 
and then to Delhi, where he studied under Shah 'Abd ul-Qadir 
(Shah 'Abd ul- 'Aziz's brother) of the Madrasa-i Rahimivva 
from 1805 to 1811 .Thereafter he left for central India, where 
he served as a cavalryman for one Amir Khan "who worked for 
the Marathas. In 1818, this Amir Khan was "forced to come to 
terms with the British who [awarded] him the principality of 
Tonk and stvled him a nawwab" (Metcalf, 1982: 54) . 

Savvid Ahmad then returned to Delhi the second time, now 
as a religious reformer determined to bring about greater 
observance of the shari'a. Some prominent younger members 
of the Shah Wali Ullah family accepted him as their spiritual 
leader (sufi shaikh). His ideas are set out in two influential 
books bv his close associate Muhammad Isma'il (d. 1831). 


ch2.044 10/12/2004 5:11 PM Page 32 


Entitled Taqwiyat al-lman (Strengthening the Faith) and Sirat 
al-Mustaqim (The Straight Path), the first was published in 
Persian, but soon translated into Urdu, while the second was 
actually written in Urdu. The central theme of the Taqwiyat 
al-lman is the claim that the Muslims of the time had deviated 
from the principle of tawhid, strict monotheism, bv a number 
of objectionable practices representing a form of: shirk (associa- 
tionism or polytheism, the opposite of tawhid). He divides 
them into three main groups, associating some with God's 
knowledge (ishrakji'l 'ilm), others with God's power (ishrakji'l 
tasarruf), and others with God's worship (ishrakji'l 'ibada), 
giving examples illustrating each type. Thus, belief in inter- 
cession is cited as an example of association of others with 
God's power. A host of popular practices, such as prostration 
before a tomb, going on pilgrimage to a holv person's tomb and 
making food offerings in honor of the deceased, and the like are 
cited as examples of the third kind of shirk. However, Sayvid 
\7" Ahmad did not condemn sufism per se, onlv its perceived —kzj~ 

excesses. In addition, he also promoted practices which he 
deemed Islamic, such as the remarriage of widows (the upper- 
caste Hindu prohibition on the remarriage of widows had no 
scriptural sanction in the Qur'an). He even helped to bring 
about the remarriage of women he knew. 

The second phase of Savvid Ahmad's career was overtlv 
political, for he decided in the earlv 1820s to wage a jihad 
against the new non-Muslim rulers of India (first the Sikhs in 
Panjab, then the British). He and his associates planned for it 
carefullv. First Savvid Ahmad went on the pilgrimage to Mecca, 
gathering followers along the way from his hometown in Rae 
Bareli to Calcutta, where a number of them boarded a ship for 
the long journey. After his arrival in Arabia, he had his follow- 
ers swear to follow him in the jihad to come. The model in these 
and other activities was the Prophet, who had led his followers 
to victory against the pagan Meccans from their base in Medina 


ch2.044 10/12/2004 5:11 PM Page 31; 


many centuries before. The oath of lovaltv had a double 
significance: at once a spiritual tie between master and disciple 
(and a promise to abide by certain principles of behavior 
which distinguished the devotee from the larger society 
around him), it was also a political act, presaging the coming 
jihad. His followers regarded him as the mujaddid (Renewer) 
of the new (thirteenth) Islamic century. As we shall see in 
subsequent chapters, the Ahl-e Sunnat movement disputed 
this claim. 

After his return to India in 1823, Savvid Ahmad toured 
the north for two years, organizing and making preparations. 
He proceeded in a westerly direction, intending to wage 
jihad from what is today Afghanistan. The shari'a stipulates 
(following the Prophet's example) that jihad be waged from a 
Muslim-ruled territorv adjacent to a non-Muslim one. 
Accordinglv, the target of the jihad movement "was the Panjab, 
then ruled by the Sikh leader Ranjit Singh rather than the 
\7" British. In 1831, after a series of military successes, Savvid —kzj~ 

Ahmad was killed along with six hundred others as a result of 
skirmishes with local Afghans who resented the reforms 
(and taxes) sought to be imposed on them. Leaderless, the 
movement lingered on for many years in northwestern India 
but finallv petered out in the 1 860s. 

Th e Fa ra 'i zi Movetn ent 

A very different Islamic reform movement, that of the Fara'izis 
in Bengal, unfolded during the 1 820s through to the 1 860s. The 
name derives from the word Jarz ( Arabic fard; pluraljara'iz or 
Jara'id), or duties of Islam. The leader of this movement 
was Haji Shari'at Ullah (d. 1840), who returned to Bengal in 
1821 after living in the Hijaz in western Arabia for many 
vears. Dismaved by what he saw as the laxitv of practice 


ch2.044 10/12/2004 5:11 PM Page 34 


among Bengali weavers and peasants, he preached renewed 
commitment to the duties of Islam (daily prayer, the Ramadan 
fast, and the pilgrimage, among other things). Shari'at Ullah 
also believed that sufism should be limited to the few, for its 
esoteric teachings were likelv to be misunderstood bv ordinary 
believers. His teachings have been compared to those of the 
Wahhabis, whose ideas were familiar to Shari'at Ullah from his 
long stay in Arabia. 

Rural Bengal at this time was in the midst of a severe eco- 
nomic depression brought about bv the Permanent Settlement 
of 1793, which changed landholding patterns and rendered 
many peasants landless. The introduction of British factory- 
made cloth at low prices was also driving Indian weavers out of 
business and forcing them on to the land. These circumstances 
help us understand the anti-British aspects of the movement, 
for Shari'at Ullah ruled that in the absence of functioning qazis 
and given the non-implementation of shari'a law, Bengal was 
\7" dar ul-harb (as some interpreted Delhi to have become after its ~x3~ 

occupation bv the British in 1 803) , and that the congregational 
noontime prayer on Fridays was therefore not permissible. For 
him the suspension of religious law in lands under British con- 
trol meant that the normal rules of conduct in other spheres of 
life no longer applied either. Under the leadership of his son, 
Dudhu Miyan, Fara'izis were urged to refuse to pay British land 
taxes. They also boycotted the British courts, settling their dif- 
ferences themselves. The movement "was highly successful in 
forging a sense of unitv and self-help among poor Bengali 
Muslims for a while. However, British reprisals, and the lack of 
strong leadership after Dudhu Mivan's death in the 1 860s led to 
the movement's decline (Metcalf, 1982: 68— 70; Ahmad Khan, 


ch2.044 10/12/2004 5:11 PM Page 3 



The Deobandi 'Ulama 

In 1867, a new seminary (madrasa) called the Dar al-'Ulum 
was founded in the small town of Deoband, about eightv miles 
north of Delhi. It was a new kind of madrasa: 

Its founders, emulating the British bureaucratic style for 
educational institutions, . . . acquired classrooms and a central 
library. It was run by a professional staff, and its students 
were admitted for a fixed course of study and required to take 
examinations for which prizes were awarded at a yearly 
convocation. Gradually an informal system of affiliated 
colleges emerged. . . .The school was, in fact, so unusual that 
the annual printed report, itself an innovation, made 
continuing efforts to explain the organization of the novel 
system. (Metcalf, 1982: 93-94) 

While this mav sound fairlv unremarkable to the modern 
reader, it has to be seen in the context of madrasa education at 
-(^y— the time. Traditional madrasas consisted of a building attached — t^y- 

to a mosque. The students did not have separate classrooms or 
libraries, and thev studied individual texts taught one-to-one, 
or in a small group, by a single teacher. The texts taught 
depended on the capacity of the student. When the student had 
mastered the texts, he received a certificate (sanad) from his 
teacher and could go on to study more advanced books if he so 
wished from the same or a different teacher. There were no 

The funding of the madrasa at Deoband was different as 
well. It was financed by private contributions from the resi- 
dents of Deoband and other well-wishers, not bv an endow- 
ment (waqf), as was customary. Nor "was it supported by the 
patronage of princely courts (as was the Madrasa-i 'Aliyya at 
Rampur, for instance) . 

Intellectually, the 'ulama at Deoband had much the same 
perspective as the Madrasa-i Rahimiyva in Delhi and Shah ' Abd 


ch2.044 10/12/2004 5:11 PM Page 3( 


ul-'Aziz(d. 1824). Two 'ulama who were central to the school's 
founding and early years were Maulanas Muhammad Qasim 
Nanautawi (1833-79) andRashid Ahmad Gangohi (1829-1905). 
Muhammad Qasim 's family had a long-standing relationship 
with the 'ulama of Delhi, as did Rashid Ahmad's. Both were of 
the reformist tradition; they were critical of the rituals cus- 
tomarily performed at saints' tombs, lavish weddings and 
feasts, and the payment of interest on loans, for instance. They 
were also ambivalent about rituals associated with the death 
anniversaries ('urs) of sufi saints, discouraging but not com- 
pletely condemning them. On the other hand, they "were punc- 
tilious about observing the ritual obligations of prayer, fasting, 
and performance of the pilgrimage. They also sought to 
encourage widow remarriage. "The follower was expected to 
abandon suspect customs, to fulfill all religious obligations, and 
to submit himself to guidance in all aspects of life" (Metcalf, 

\7" The fact that the Deobandis were reformist does not mean —kzj~ 

that they were opposed to sufism — on the contrary, both Qasim 
Nanautawi and Rashid Ahmad were disciples of the famous Haji 
Imdadullah — but it did mean that they disapproved of what 
they considered sufi excesses. The curriculum thev taught 
sought to be comprehensive: they "taught all the Islamic sci- 
ences and . . . represented] all the Sufi orders. Thev said that in 
this they followed Shah Walivu'llah. [However, unlike him, 
thev] emphasized reform of custom, not intellectual synthesis" 
(Metcalf, 1982:140). 

For the Deobandi 'ulama, as for those of the Ahl-e Sunnat 
movement, the writing of fatawa was an important means of 
disseminating the message. Although the subjects of these legal 
judgments varied widely, for the most part they steered clear of 
politics. Thev addressed questions related to sufism, the proper 
performance of ritual prayer, fasting, pilgrimage, and relations 
with other groups, both Muslim and non-Muslim. 


ch2.044 10/12/2004 5:11 PM Page 3y 


Zaman adds to the picture painted by Metcalf bv giving an 
interesting example of the approach to problems thrown up bv 
British rule in the late nineteenth and early twentieth cen- 
turies, such as the lack of qadis (judges of Islamic law) in British 
Indian courts. The judgments of the 'ulama were not enforce- 
able in court. For instance, without a qadi it now became 
impossible to have marriages annulled. As a result, women 
began to declare themselves apostates from Islam, since apos- 
tasv automatically terminated a marriage. In the 1930s, 
Maulana Ashraf 'Ali Thanawi (d. 1943), a famous Deobandi 
scholar ( 'alim) , tried to solve this problem bv arguing that apos- 
tasy had no effect on the marriage contract, 'while at the same 
time proposing both that the conditions under which mar- 
riages could be dissolved should be made less stringent and that 
in the absence of a qadi, 'ulama or other "righteous Muslims" 
acting together could dissolve a marriage in his stead. These 
ideas were accepted bv the political partv, Jamiyvat al-'Ulama-e 
\7" Hind, which had been founded after WorldWar I and which was ~x3~ 

dominated bv Deobandi 'ulama, and it became the basis for the 
Dissolution of Muslim Marriages Act of 1939 in British India. 
As Zaman points out, however, although this solved the prob- 
lem related to apostasv and made it easier to dissolve a bad mar- 
riage, it put no pressure on the British to appoint qadis in 
British Indian courts. 

The Ahl-e Hadith 

The movement known as the Ahl-e Hadith ("people of the 
[prophetic] hadith") derives from the fact that the 'ulama in this 
group advocated reliance on the Qur'an and hadith for 
guidance on matters of ritual and behavior. They denied the 
legitimacy of the four Sunni law schools (Hanafi, Shafi'i, 
Hanbali, and Maliki) that had emerged within some three hun- 
dred years of the death of the Prophet and which had long 


ch2.044 10/12/2004 5:11 PM Page 3£ 


reached so dominant a position that one could not be a Sunni 
without affiliation to one of them. Their rejection of the judg- 
ments of the law schools and insistence that each believer 
decide on an issue for him- or herself based on what the Qur ' an 
and hadith have to say about it presupposed a high level of liter- 
acy and familiarity with Arabic which the 'ulama were normally 
the only ones to possess; this made it highlv elitist. This was a 
reflection, perhaps, of their class status, for the leadership of 
the Ahl-e Hadith belonged to the well-born, people who had 
been employed by the Mughal court but had since fallen on 
hard times. 

Two additional features distinguished the Ahl-e Hadith from 
other Sunni Muslims. The first was a ritual matter: thev favored 
a certain manner of prayer that set them apart from everyone 
else. The second was more important, namelv that thev con- 
demned all forms of sufism, not just specific aspects of sufi 
practice after the fashion of the Deobandis.Thev opposed the 
\7" veneration of saints and pilgrimages to their tombs. In fact, —kzj~ 

they also opposed the practice of visiting the Prophet's tomb in 
Medina. Because of this and their condemnation of the four law 
schools, many Muslims compared them to the Wahhabis of 
Arabia. Like the Arabian Wahhabis, they read and admired the 
works of Ibn Taimivva (d. 1328), even translating his works 
into Urdu. 

The Ahl-e Hadith, however, claimed that they were intellec- 
tual descendants of the eighteenth-centurv scholar Shah Wali 
Ullah of Delhi. Shah Wali Ullah had, indeed, spoken of the 
importance of hadith scholarship, and of the precedence of 
hadith over the judgments of the law schools in cases of conflict 
between them. And unlike the Ahl-e Hadith, who "denied the 
legitimacv of ... the four major law schools" (Metcalf, 1982: 
270), at least for the educated elite, the Wahhabis followed the 
judgments of Hanbali scholars. Unlike Shah Wali Ullah, 
who had been eclectic in his use of the legal tradition, the 


ch2.044 10/12/2004 5:11 PM Page 3i 


Ahl-e Hadith preferred a narrow interpretation of the Qur'an 

Relations between the Ahl-e Hadith and the other Sunni 
Muslim reform movements were tense, leading on several 
occasions to lawsuits which the British were forced to 
arbitrate. Their relations with the British were also uneasy. 
The British suspected them of sedition until 1871 , when thev 
concluded the so-called Wahhabi trials conducted against 
the jihadists who had continued to fight the British in 
Afghanistan and along the northwestern border, following 
Savvid Ahmad Barelwi's lead. Thereafter relations between 
them improved. 

In terms of their theological positions on the Sunni law 
schools and sufism, the Ahl-e Hadith was perhaps the furthest 
from the Ahl-e Sunnat of all the movements considered 


The Nadwat al-'Ulama ("Council of 'Ulama," known as 
Nadwa, for short) was founded in the 1 890s in the hope of 
bringing Sunni and Shi'i 'ulama together on a single platform, 
despite their differences of opinion. It was hoped that, thus 
united, the Nadwa would be able to present to the British the 
views of its members on issues thev cared about. Annual meet- 
ings were planned at which all members would convene and 
decide on future action. As originallv conceived, its member- 
ship was to have consisted not onlv of Sunni and Shi'i 'ulama, 
but also of wealthv and powerful patrons such as Muslim 
"princes, government servants, traders, and lawvers" (Metcalf, 
1982: 345). It was also conceived as an all-India bodv, not a 
local one. It actively sought British recognition of its school, the 
Dar al-'Ulum, founded in 1898. After some hesitation, the 
British agreed to patronize secular learning at the school, 


ch2.044 10/12/2004 5:11 PM Page 4( 


contributed land for the fine building subsequently built in 
Lucknow, and in 1908 laid the foundation stone. 

The school curriculum was a source of considerable debate 
and discord from the verv outset. Some felt that English should 
be taught alongside Arabic and other subjects since it would 
allow the Nadwa to refute Western religion and culture all the 
more effectivelv. Although two of its earlv leaders, Savvid 
Muhammad 'Ali Mongiri and Maulana Shibli Numani, sup- 
ported English as a subject, the 'ulama opposed it, and the idea 
soon had to be given up. The opposition stemmed from fear 
that in the long run the introduction of English would lead to 
the secularization of the curriculum. 

Another goal of the new school was madrasa reform. In 
order to 

infus[e] the ranks of the 'ulama with fresh vigor, and . . . 
broaden the scope of their activities and their role in the 
s~~\ Muslim community ... it was deemed imperative to reform s--\ 

the prevalent styles of learning. .. .The Nadwa 's proposed 
curriculum sought to produce religious scholars capable of 
providing guidance and leadership to the community in a 
wide range of spheres: in law and theology, in adab (belles 
lettres), in philosophy, and in "matters of the world." (Zaman, 
2002: 69) 

The founders hoped that all Indian madrasas would follow its 
lead and adopt the curriculum that thev proposed to put 
together. Thev wanted to impart a "useful" education — bv 
which thev meant one that would create "a new generation of 
'ulama fit to lead the Muslim communitv."The studv of "exe- 
gesis [ofthe Qur'an],hadith,historv, and Arabic literature" was 
to be emphasized, while that of logic and philosophv — the hall- 
mark ofthe Dars-i Nizami svllabus thev were trving to reform 
— was downplaved (Zaman, 2002: 71—72). If this sounds coun- 
terintuitive, it has to be remembered that the Dars-i Nizami 


ch2.044 10/12/2004 5:11 PM Page 4: 


svllabus had been designed in the eighteenth century. The 
Nadwa considered it outdated and in need of revision. Exegesis 
of the Qur'an and hadith, on the other hand, required the stu- 
dent to study the sources at first hand, while the studv of Arabic 
literature and history were intended to broaden the student's 
knowledge of the Arab world more generally. 

In practice, it was hard to implement these changes, for the 
authority of the 'ulama ultimately rested on their mastery of 
the very texts that the Nadwa was trying to replace. (Indeed, 
Zaman points out that the authority of these texts had, if any- 
thing, increased during the colonial period.) The Nadwa's pro- 
posal to do away not onlv with many of these texts, but also 
with the discursive practices of the madrasa curriculum — in 
other words, with the whole system by which religious author- 
ity "was acquired and demonstrated — required the 'ulama to 
distance themselves from their tradition of learning, rather 
than embrace it. Another hurdle was the difficulty of getting 
\7" the 'ulama to put aside their differences. The challenge —kzj~ 

the Nadwa thus took on was enormous, and in the end the 
attempt failed. 

The Nadwa continues to flourish today, but its curriculum 
follows that of the Dars-i Nizami svllabus. 

Sayyid Ahmad Khan and MAO College, Aligarh 

Sayyid Ahmad Khan (d. 1898) "was not a religious scholar but 
an official in the judicial department of the British Indian 
government until his retirement in 1 877, and the college he 
founded in 1 875 had a very different purpose from those dis- 
cussed above. He is an important figure in the history of South 
Asian nationalism, particularly in Pakistan, where he is seen as 
the nineteenth-century "founder" of the idea of a separate 
homeland for South Asian Muslims. When the Indian National 
Congress was founded in 1885, Sayyid Ahmad spoke out 


ch2.044 10/12/2004 5:11 PM Page 41 


against the idea of an Indian nation that might be democratic 
when it became independent, as he believed this would be 
detrimental to Muslim interests, and founded an organization 
of his own, the Muhammadan Educational Congress (later 
renamed the Muhammadan Educational Conference). Shortly 
thereafter, the British honored him with a knighthood for his 
services to the empire in the aftermath of the Revolt of 1857, 
particularly the role he plaved in fostering mutual understand- 
ing between the British and the Indian Muslim community, and 
he became Sir Savvid. 

Savvid Ahmad Khan was a rationalist. His reformist ideas 
were in the tradition of Shah Wali Ullah, and were also similar 
to those of Muhammad Isma'il, the author of the Taqwiyat 
al-Iman, particularly in his disapproval of what he saw as accre- 
tions to Islamic belief and practice and different forms of 
associationism (shirk). He believed that Islam was a rational 
religion, one that was in full accord with human nature: 

I have determined the following principle for discerning the 
truth of the religions, and also for testing the truth of Islam, 
i.e. , is the religion in question in correspondence with human 
nature or not, with the human nature that has been created 
into man or exists in man. And I have become certain that 
Islam is in correspondence with that nature. (Quoted in Troll, 

And further: 

I hold for certain that God has created us and sent us his 
guidance. This guidance corresponds fully to our natural 
constitution, to our nature. ... It would be highly irrational to 
maintain that God's work [the natural world , including 
humankind] and God's word [the revelation of the Qur'an] 
are different and unrelated to one another. All beings, 
including man, are God's work and religion is His word; the 
two cannot be in conflict. ... So I formulated that "Islam is 
nature and nature is Islam." (Quoted in Troll, 1 978: 317) 


ch2.044 10/12/2004 5:11 PM Page 4: 


This formulation led some 'ulama, the Ahl-e Sunnat among 
them, to allege that Savvid Ahmad Khan worshiped nature 
rather than God, an allegation he vehementlv denied. 

In keeping with his modernist, rationalist thinking Savvid 
Ahmad Khan denied the possibilitv of miracles, interpreting 
the miracles surrounding the Prophet as later fabrications. He 
also interpreted belief in angels metaphoricallv rather than lit 
erallv, as a quality possessed by prophets. Thus, the angel 
Gabriel "stands for the . . . inherent possession of prophethood 
in the Prophet himself and thus stands for the cause of revela- 
tion" (Troll, 1978: 181). He was also critical of much of the 
hadith literature, dismissing it as being inauthentic. Like the 
Ahl-e Hadith, he denied the legitimacy of the four Sunni law 
schools, looking to the Qur'an and the example of the Prophet 
for guidance. On the power of personal praver (dua) to change 
one's ultimate fate, he believed that God "is pleased with such 
prayer and accepts it as He accepts anv other form of service. 
\7" ... Performance of this prayer brings about in man's heart —kzj~ 

patience and firmness" (Troll, 1978: 182). But he held that it 
did not change one's predetermined destiny. The concept of 
intercession and mediation between man and God were thus 
also denied. 

Savvid Ahmad Khan's reformist ideas were intimatelv con- 

j j j 

nected with the political context of late nineteenth-centurv 
British India. He came from a family which had been associated 
with Mughal rule, and he keenlv felt the loss of that rule. In his 
view, Muslims had lost out to the British because thev had failed 
to keep up with the scientific progress of the West and had 
allowed their practice of the faith to lapse as "well. Judging that 
British rule over India was there to stay for the foreseeable 
future, he set out on the one hand to cultivate good relations 
with the British and on the other to encourage Muslims to 
acquire the new linguistic and scientific skills necessary to suc- 
ceed in the new era. 


ch2.044 10/12/2004 5:11 PM Page 4' 


In the educational realm, Savvid Ahmad Khan's modernist, 
progressive vision expressed itself in the Muhammadan 
Anglo -Oriental (MAO) College, founded in Aligarh in 187S. 
The college was modeled on Oxford and Cambridge (he had 
spent two years, 1 869— 70, in Britain, studying everything from 
factories to schools). Not only would the curriculum offer an 
array ofWestern subjects (the natural sciences, mathematics, 
literature, and so on) , but it would also be residential. Over the 
years, as David Lelvveld (1978) eloquently demonstrates, the 
school fostered a strong sense of belonging — even brother- 
hood — among the students, many of whom had come from 
outside the immediate geographical area. Sayyid Ahmad Khan's 
goal of training a generation of Muslims who would become 
part of the new government structure was also partially real- 
ized, to the extent that three-quarters of school graduates got 
government positions. But there could be no sense of equality 
between the British and Aligarh's Muslims: "however skilled in 
\7" Western culture some Indians might become, the pall of arro- ~x3~ 

gant racism, inherent in the colonial situation, meant that full 
acceptance of Indians as equals never happened" (Metcalf, 

Savvid Ahmad Khan had to concede defeat on the religious 
front as well. So controversial a figure was he on account of his 
reformist ideas that the Muslims of Aligarh and elsewhere "were 
initially reluctant to support his new institution. The British 
stepped in not onlv with funds but in many cases "with profes- 
sors as well. Sayvid Ahmad did his best to reassure Muslim 
parents that their children would not be taught radical ideas bv 
hiring some of his fiercest critics as professors in the religious 
studies department. Consequently the program of religious 
education at MAO College, while reformist in the Deobandi 
sense, appears to have been uncontroversial. 

In sum, Aligarh's MAO College was aWestern-style institu- 
tion, unlike the Dar al-'Ulum at Deoband and that of the same 


ch2.044 10/12/2004 5:11 PM Page 4 



name started by the Nadwat al-'Ulama in Lucknow in the 
1 890s. It shared with them a sense that Islamic education 
needed reform in order to be meaningful in the late nineteenth 
centurv. Unlike the Deobandi madrasa, both the Nadwa and 
Savvid Ahmad Khan also aspired to some form of political asso- 
ciation with the British. In the early twentieth century, MAO 
College — which was recognized as a university and renamed 
Aligarh Muslim University in 1920 — fulfilled its promise by 
becoming the training ground for several prominent Indian 
Muslim nationalists. 

The Ahmadi Movement 

The Ahmadi movement, which was highly controversial, was 
founded by Mirza Ghulam Ahmad (d. 1908), in 1889. Ghulam 
Ahmad was born in the village of Qadiyan, Panjab, in the 
Qj 1 830s, to a familv that had prospered during Mughal times but — (^y- 

had lost much of its wealth during Sikh rule. He credited the 
British with an improvement in his family's fortunes, and in 
later years "was noticeably pro-British in his politics. His educa- 
tion was traditional (study of the Qur'an, Arabic, and other 
subjects) , but acquired at home, not at a madrasa. 

Unlike most of the other Muslim movements discussed in 
this book, the Ahmadis can date the beginning of their move- 
ment precisely, for in March 1 889 Ghulam Ahmad held a cere- 
mony of sufi initiation (bay 'a) at which he accepted his first 
disciples in the city of Ludhiana, Panjab. From 1891 onward, 
the group held annual meetings each December "to enable 
every Ahmadi to increase his religious knowledge by listening 
to speeches, ... to strengthen the fraternal bonds between the 
members, and to make plans for missionary activity in Europe 
and in America" (Friedmann, 1989: 5). The initial activities of 
the movement revolved around public oral debates with 


ch2.044 10/12/2004 5:11 PM Page 4( 


Hindus (the Arya Samaj) about miracles and eternal salvation, 
with Christians about the death of Jesus Christ and Christ's 
divinity, and with other Muslims (the Ahl-e Hadith), also about 
Jesus Christ. Ghulam Ahmad was also a prolific writer of books 
and articles in Urdu, Arabic, and Persian, and in 1902 began an 
English monthlv periodical, The Review of Religions, which has 
continued to be published ever since. The third significant 
thrust of the movement has been a missionarv one, with 
emphasis particularly on growth in Britain. 

The disagreements between the Ahmadivva and other Sunni 
Muslims in South Asia are mainlv over Ghulam Ahmad's claims 
to religious authority. He believed he was the "mujaddid, 
renewer (of religion) at the beginning of the fourteenth cen- 
tury of Islam; muhaddath, a person frequently spoken to by 
Allah or one of His angels; and mahdi, 'the rightlv guided one, 
the messiah,' expected by the Islamic tradition to appear at the 
end of days" (Friedmann, 1989: 49). Of the three claims made 
\7" here, the second, that of being spoken to by Allah, was particu- ~x3~ 

larly controversial, as the rank of muhaddath is considered to be 
only slightly below that of prophethood and implies direct 
communication with God. No Sunni reformer had ever 
claimed it before. By contrast, the claim to the status of Mahdi 
is relatively common in Sunni history, and several claimants 
appeared in the eighteenth and nineteenth centuries, associ- 
ated with anticolonial jihad movements against British or 
French rule. It was not, however, as a militant Mahdi that 
Ghulam Ahmad cast himself. On the contrary, he denied not 
only the obligatory nature, but also the very legitimacy of 
jihad in the sense of armed confrontation, an extraordinarily 
bold heretical move only partly explained in terms of his 
positive attitude to British rule. In his view, jihad was to be 
interpreted as the peaceful attempt to spread the faith through 

Ghulam Ahmad was fierce in his denunciation of the Indian 


ch2.044 10/12/2004 5:11 PM Page Ay 


'ulama, who in his view had allowed Islam to fall into a sorry 
state : 

Like leaders of other revivalist and messianic movements in 
Islam, Ghulam Ahmad was convinced that Islamic religion, 
Islamic society, and the position of Islam vis-a-vis other 
faiths sank in his times to unprecedented depths. Corruption, 
blameworthy innovations (bida'), tomb worship (qabr parasti) , 
worship of Sufi shaykhs (pir parasti) , and even polytheism 
became rampant. The Islamic way of life was replaced 
with drinking, gambling, prostitution, and internal 
strife. The Qur'an was abandoned, and (non-Islamic) 
philosophy became the people's qibla [guide]. (Friedmann, 
1989: 105) 

More specifically, Ghulam Ahmad accused the 'ulama of failing 
to stem the tide of Christian influence in India. Ghulam Ahmad 
propounded a number of anti-Christian arguments. In agree- 
Qj ment with the Qur'an (4: 1 57), he maintained that Christ had — (^y- 

not died on the cross, but whereas most Muslims believe that 
he is alive and will return together with the Mahdi, Ghulam 
Ahmad claimed that he had died at the age of a hundred and 
twenty and was buried in Srinagar, the capital of Kashmir. For 
Ghulam Ahmad, belief in the death of Jesus was important in 
light of the Christian missionaries' denunciation of the Prophet 
Muhammad as a dead prophet, in contrast to Jesus Christ who, 
they said, was alive in heaven and would one day return (as the 
Sunnis agreed). In Ghulam Ahmad's depiction of the second 
coming, he, Ghulam Ahmad, would be the messiah, not Jesus 
Christ. "By claiming that Jesus died a natural death, Ghulam 
Ahmad tried to deprive Christianity of the all-important cruci- 
fixion of its founder. In doing this he was following classical 
Muslim tradition. Bv claiming affinitv with Jesus, he went one 
step further: he tried to deprive Christianity of Jesus himself" 
(Friedmann, 1989: 118). 


ch2.044 10/12/2004 5:11 PM Page 4i 


Two further theological ideas need to be understood in this 
brief summary, namelv, Ghulam Ahmad's ideas about prophecy 
and his claim to be a "shadowy" (zilli) prophet himself. As 
Friedmann makes clear, these ideas — and indeed other aspects 
of Ghulam Ahmad's thought — are based on sufi concepts trace- 
able to Ibn 'Arabi (d. 1240). Ibn 'Arabi believed that the total 
cessation of prophecy after the death of the Prophet 
Muhammad would have left the Muslim community utterly 
bereft. This was impossible in his view, so he postulated that 
prophecy had continued in a new form. There were two differ- 
ent tvpes of prophecv, he said, the legislative, which is superior 
and which had ceased on the death of the Prophet, and the non- 
legislative, which is given to sufis of extraordinary caliber and 
insight and which he claimed for himself. Friedmann sums up 
the difference, in Ghulam Ahmad's view, between the Prophet 
and himself as follows : 

^~A while it is true that no law-giving prophet can appear after f"\ 

Muhammad, prophetic perfections are continuously 
bestowed upon his most accomplished followers, such as 
Ghulam Ahmad, to whom Allah speaks and reveals his secrets. 
However, since Ghulam Ahmad attained this position only by 
his faithful following of Muhammad, his prophethood does 
not infringe upon Muhammad's status as the seal of the 
prophets. (Friedmann, 199S: 56) 

Furthermore, after Muhammad's mission had been com- 
pleted, Muslims were the only ones favored with direct com- 
munication from God bv having people among them who were 
muhaddath .This proved their superiority over Christianity. 

A few years after Ghulam Ahmad's death in 1 908, the move- 
ment split into two factions, subsequently known as the 
Qadivanis and the Lahoris (after the places where they have 
their headquarters; Qadivan is now in India, Lahore in 
Pakistan). The Qadivanis, led bv Ghulam Ahmad's son, were 


ch2.044 10/12/2004 5:11 PM Page 4< 


more numerous and supported Ghulam Ahmad's prophetic 
claim, while the Lahoris watered it down, rejecting his claim to 
prophethood and only accepting him as a Renewer (mujaddid) 
rather than a prophet. (In the 1 970s and 1 980s, the Ahmadis of 
both factions 'were declared non-Muslims in Pakistan bv a con- 
stitutional amendment and other legislative means.) 

"0" "& 


ch2.044 10/12/2004 5:11 PM Page 5C 




ch3.044 10/12/2004 5:17 PM Page 5] 



hmad Riza Khan 'was born in Bareillv, in the western 

.United Provinces, in 18S6, just a year before the great 
Indian Revolt. A story is told about his grandfather, Maulana 
Riza 'Ali Khan (1809—65/66), relating to the British resump- 
tion of control over Bareillv after the Revolt had been put down 
in that town: 

After the tumult of 1 8S7, the British tightened the reins of 

power and committed atrocities toward the people, and 

everybody went about feeling scared. Important people left 

their houses and went back to their villages. But Maulana Riza 


'Ali Khan continued to live in his house as before, and would 
go to the mosque five times a day to say his prayers in 
congregation. One day some Englishmen passed by the 
mosque, and decided to see if there was anyone inside so they 
could catch hold of them and beat them up. They went inside 
and looked around but didn't see anyone. Yet the Maulana was 
there at the time. Allah had made them blind, so that they 
would be unable to see him. . . . [When] he came out of the 
mosque, they were still watching out for people, but no one 
saw him. (Bihari, 1938:5) 

Bihari goes on to quote the Qur ' anic verse, "And We shall raise 
a barrier in front of them and a barrier behind them, and cover 



ch3.044 10/12/2004 5:17 PM Page 52 


them over so that they will not be able to see" (36: 9, Ahmed 
'Ali translation). 

The storv is interesting at many levels. It casts Maulana Riza 
'Ali as a fierce opponent of the British who put his trust in God 
instead of fleeing and who was so holv and so good that God pro- 
tected him, blinding the enemy to his presence. This miracle, for 
so it "was described (karamat) , was a sign of his eminence as a sufi 
(mvstic).The title of Maulana before his name shows that he was 
also a religious scholar (faqih). Or, to put it another way, he 
didn't just practice his faith bv meticulously adhering to the Law 
(shari'a), he also lived it and breathed it in his inner being. 

Ahmad Riza Khan's familv had not always been associated 
with religious learning. His ancestors were Pathans who had 
probably migrated from Qandahar (in present-day Afghanistan) 
in the seventeenth century, joining Mughal service as soldiers 
and administrators. One familv member eventually settled 
down in Bareillv, where he was awarded a land grant bv the 
\7" Mughal ruler. There followed a brief interlude in Awadh, when —kzj~ 

Ahmad Riza Khan's great-grandfather served the nawab in 
Lucknow, probably in the late 1700s, when Mughal power was 
in decline and Awadh in the ascendant. The nawab is said to have 
given Hafiz Kazim 'Ali Khan, Ahmad Riza's great-grandfather, 
two revenue-free properties. These properties were in the fam- 
ily's possession until 1 954 (Hasnain Riza Khan, 1986:40^+1). 

We know that Hafiz Kazim 'Ali later returned to Bareillv, for 
that is where his son Riza 'Ali (Ahmad Riza's grandfather) grew 
up. It was Riza 'Ali who made the break from soldiering and 
state administration to become a scholar and sufi. In the earlv 
nineteenth centurv, at a time when Muslim states all over India 
were bowing to British power, the opportunities for a soldier 
who sought a Muslim patron were diminishing rapidly. Riza 
'Ali was educated atTonk, the onlv Muslim state in central India 
(where, as mentioned in the previous chapter, Sayyid Ahmad 
had been a soldier in the ruler's army in the 1820s). After 


ch3.044 10/12/2004 5:17 PM Page 51; 


completing his study of the Dars-i Nizami svllabus there by 
the age of twenty-three, he returned to Bareillv and made his 
reputation as a scholar. 

Ahmad Riza's father, Naqi 'Ali Khan (1 83 1—80), carried on 
the scholarly tradition begun bv his father, while also looking 
after the family properties. By this time the family owned sev- 
eral villages in the adjoining districts of Bareillv and Badavun. 
The Revolt of 18S7 did not affect the familv significantlv, 
though some property in Rampur was lost in its aftermath 
because of failure to find the title deeds and prove ownership to 
the British. Relations with the British appear to have been 
indirect but cordial. Ahmad Riza's nephew Hasnain Riza owned 
a printing press which later published many of Ahmad Riza's 
writings. Hasnain Riza reportedly collected certain fees from 
the police tribunal for the British, acted as arbitrator between 
Muslims in the town, and mediated between them and the 
British on occasion. He did not, however, -work for the British 
\7" in an official capacity. —kzj~ 

The family also had close ties with officials in Rampur state, 
which, as noted in chapter 1 , retained its independence under 
a Muslim nawab throughout the period of British rule. Thus, 
for instance, Ahmad Riza's father-indaw "was an employee at 
the Rampur Post Office, and attended the nawab 's court 
(Hasnain Riza Khan, 1986: 152). Rampur 's nawabs had been 
Shi'is since the 1840s — all but one, that is: Kalb 'Ali Khan 
(r. 1 865— 87) "who was a Sunni. 


As noted earlier (pp. 6—7), Rampur state was founded by 
Faizullah Khan in the 1 770s bv treaty with Warren Hastings, then 
the Governor of Bengal. It was all that was left to the Rohillas 
after the absorption of Rohilkhand by the up-and-coming state 


ch3.044 10/12/2004 5:17 PM Page 54 


of Awadh to the east. Having acquired a little state of his own, 
Faizullah Khan put down his arms and devoted the remaining 
vears of his life to developing Rampur as a center of Muslim 
cultural life and sought to attract writers, poets, and other men 
of literary or scholarlv talent to his court. There is some evi- 
dence that he founded the Raza Library, which is in operation 
to this dav, home to a large collection of valuable manuscripts 
in Arabic, Persian, and Urdu. 

Awadh became increasinglv indebted to the East India 
Companv over the course of the early nineteenth centurv, and 
was finallv forced to cede power to the Companv altogether in 
1 856. The Rampur court then rose as an alternative source of 
patronage to which people would travel in search of emplov- 
ment. "Mulla Hasan [of Farangi Mahall] went from Lucknow 
to Shahjahanpur, and thence to Rampur via Delhi; Mawlana 
'Abd 'Ali Bahr al-'Ulum (1731-1810) from Lucknow to 
Shahjahanpur, to Rampur, to Buhar in Bengal and finallv to 
\7" Madras" (Robinson, 2001 : 23). The 'ulama of Farangi Mahall, it —kzj~ 

should be noted, were Sunni bv persuasion. The Rampur court, 
which became Shi'i in the 1 840s, was hospitable to both Sunnis 
and Shi 'is. 

The court welcomed a number of poets, most famouslv, in 
the nineteenth century, Mirza Ghalib (d. 1869), who taught 
poetry to Rampur 's nawab,Yusuf 'Ali Khan (r. 1855— 65).Yusuf 
'Ali was himself a poet. From 1859, he began to send Ghalib a 
regular monthly grant for correcting his poetry and writing 
occasional panegvrics on important state occasions. Contrary 
to custom (andYusuf 'Ali's preference), Ghalib was permitted 
to live in Delhi, making onlv occasional visits to the Rampur 
court. Ghalib, like manv of his contemporaries, wrote not only 
in Persian — the language of choice for the educated elites of all 
communities, Muslim as well as Hindu, throughout the 
eighteenth and the first half of the nineteenth centuries — but 
also in Urdu, which rapidlv began to replace Persian in the 


ch3.044 10/12/2004 5:17 PM Page 5 



second half of the nineteenth century. Thus, Ahmad Riza Khan's 
writings, which I will examine in later chapters, were almost 
entirely in Urdu. 

The madrasa at Rampur known as ' Alivva also attracted well- 
known 'ulama from other parts of north India. Among them 
were Maulanas Fazl-e Haqq Khairabadi (d. 1861) and 'Abd 
ul-Haqq Khairabadi (d. 1 899) , both specialists in the rational sci- 
ences (ma 'qulat). It was founded in the eighteenth century with 
endowment (waqf) funds from two villages, and enjoved state 
patronage under the nawabs. However, it never achieved the sta- 
tus of other madrasas in the country, such as Farangi Mahall in 
Lucknow or the Madrasa-i Rahimivva in Delhi, where ShahWali 
Ullah taught in the eighteenth century. Rampur 's Raza Librarv, 
on the other hand, was an institution of great renown. For seven 
years, from 1 896 to 1 903, it was managed bv the famous Indian 
nationalist leader, Hakim Ajmal Khan (1863—1927), who 
expanded the library's holdings on medicine (tibb), enabling it 
\7" to become one of the best in the country. A new library building —kzj~ 

was also constructed at the end of the nineteenth century. 


Ahmad Riza's most important teacher was his father. He stud- 
ied the Dars-i Nizami svllabus under his direction, and imbibed 
from him the rationalist tradition. The pattern of a student 
studying specific books under a single teacher, whether in an 
institution such as a madrasa (seminary) or at the teacher's 
home, was traditional throughout the Muslim world. At the 
end of the period of study, the teacher would give the pupil a 
certificate (sanad) stating that the student had studied certain 
books under his direction (including glosses and commentaries 
thereon) and giving him permission (ijaza) to teach these in 


ch3.044 10/12/2004 5:17 PM Page 5( 


turn. Thereafter, if he so wished, the student could continue his 
studies under another teacher, with whom he would remain 
until he had obtained another certificate testifying to compe- 
tence in another set of books. Chains of transmission of author- 
ity — recorded in writing at the end of a period of studv — were 
thus established between individual teachers and their stu- 
dents, for each teacher received the authority to teach from the 
one who had taught him. Over time, these chains of authority 
linked a vast network of 'ulama in different parts of the countrv 
(for an example of such a chain of ma'qulat scholars, see 
Robinson 2001: 52-53). 

Not surprisinglv, in view of the strong ties between teachers 
and their students, the intellectual positions taken by the for- 
mer often stamped themselves indelibly on the minds of the 
latter. So it was with Ahmad Riza Khan. His father's stand on a 
number of theological issues in the mid-nineteenth centurv 
later also became his own. 


One of the well-known debates of the early nineteenth centurv 
dealt "with God's omnipotence. Some 'ulama argued that God 
had the power, should He so wish, to create another prophet like 
Muhammad. Thus, Muhammad Isma'il, author of the Taqwiyat 
al-Iman (Strengthening the Faith), had written in the 1 820s: 

in a twinkling, solely by pronouncing the word "Be!" [God 
could], if he like[d], create crores [tens of millions] of 
apostles, saints, genii, and angels, of similar ranks with 
Gabriel and Muhammad, or produce a total subversion of 
the whole universe, and supply its place with new creations. 
(MirShahamat'Ali, tr. (modified), 1 852: 339) 

This statement — known as imkan-e nazir, the possibility of an 
equal (of the Prophet) — was made in the context of tawhid, as 


ch3.044 10/12/2004 5:17 PM Page 5y 


an illustration of God's power. It was strongly opposed by 
Maulana Fazl-e Haqq Khairabadi, whose presence at the 
Madrasa ' Alivva at Rampur and association with the rationalist 
position in 'ulama circles were mentioned earlier. Maulana 
Fazl-e Haqq — taking a position known as imtina'-e nazir, or 
impossibility of an equal — argued that even God could not pro- 
duce another prophet like the Prophet Muhammad. 

A generation later, in the 1 85 0s and 1 860s the two views were 
expressed again, both verballv and in print, with Naqi ' Ali Khan, 
Ahmad Riza's father, echoing Maulana Fazl-e Haqq Khairabadi 's 
position. In the 1 890s, Ahmad Riza Khan himself wrote a respon- 
sum (fatwa) in which the focus of discussion was no longer on 
God's transcendental power but rather on the uniqueness of the 
Prophet. Arguing that it 'was impossible for anyone ever to equal 
the Prophet (not only in this world but in any of the six levels of 
the earth believed to exist apart from this one), he declared that 
to maintain otherwise amounted to denial of the finality of his 
\7" prophethood and thus to kufr, unbelief. Although the terms of ~x3~ 

debate had shifted from a discussion of God's powers to 
Muhammad's prophethood, Ahmad Riza's stance on this issue, as 
on others as well, was clearly influenced bv his father. 


Ahmad Riza's biographer, Zafar ud-Din Bihari, records a 
number of stories about Ahmad Riza's spiritual and intellectual 
accomplishments as a child. Each of them illustrates a 
distinctive aspect of the way his followers came to see him in 
later life. Thus, when learning the Arabic alphabet from his 
grandfather, Ahmad Riza is said to have instinctively under- 
stood the deeper significance of the letter "la"— a composite let- 
ter with which the attestation of faith (the kalima or shahada, lit. 
"witness") begins. He grasped not onlv its outward meaning, 


ch3.044 10/12/2004 5:17 PM Page 5£ 


that related to the Oneness of God, but also its inner, gnostic 
meaning, communicated to him bv his grandfather. This story is 
significant in light of the fact that Ahmad Riza went on to 
become both an 'alim or scholar of Islamic law, and a sufi or 
mystic seeker of God. 

Other stories claim that at four, Ahmad Riza had memorized 
the entire Qur'an by heart, and at six he addressed a gathering 
of worshipers at the mosque from the pulpit on the occasion of 
the Prophet's birthday (an annual celebration at which he 
addressed large crowds from the mosque in later years) .When 
studying the Dars-i Nizami from his father he showed that he 
had outstripped him in knowledge bv answering a criticism 
noted by him on the margins. His father was very happy to see 
this and embraced him. And when he was fourteen — much 
vounger than most scholars in a comparable situation — and had 
finished his studies in both the rational (ma 'qulat) and copied 
(manqulat) sciences, his father entrusted him with a great 
\7" responsibility, that of writing fatawa(Bihari, 1938: 1 1 , 31—33). —kzj~ 

This was to be the hallmark of his later career as a scholar. The 
number of fatawa he wrote from then until his death in 1921 
was said to be in the thousands. 

Ahmad Riza's superiority of intellect to other 'ulama far 
older than him is also illustrated in several stories. Shortly after 
his marriage, when he was about twenty, he gave an opinion 
that contradicted that of a famous scholar at the Rampur court, 
Maulana Irshad Hussain Rampuri.The nawab noticed this and 
upon enquiry discovered that Ahmad Riza was the son-indaw 
of one ofhis courtiers. So he asked to meethim (Bihari, 1938: 
1 35). Accordingly, Ahmad Riza Khan came to court. Impressed 
by both his youth and his erudition, the nawab suggested that 
Ahmad Riza would profit by studying under the famous 
Maulana 'Abd ul-Haqq Khairabadi, who had a reputation as a 
scholar of logic and "who attended the Rampur court. Ahmad 
Riza replied that if his father gave his permission, he would be 


ch3.044 10/12/2004 5:17 PM Page 5? 


happv to stay in Rampur for a few days and study with 'Abd 
ul-Haqq. Just then 'Abd ul-Haqq himself came into the room. 
The story continues: 

Maulana 'Abd ul-Haqq believed that there were only two and 
a half 'ulamain the world: one, Maulana Bahr ul-'Ulum ['Abd 
al-'Ali of Farangi Mahall, d. 1810—1 1], the second, his father 
[Fazl-e Haqq Khairabadi, d. 1 861], and the last half, himself. 
How could he tolerate this young boy being called an 'alim? 
He asked Ahmad Riza: Which is the most advanced book you 
have read in logic? 

Ahmad Riza answered: Qazi mubarak. 
He then asked: Have you read Sharah tahzib? 
Ahmad Riza Khan, hearing the derision in his voice, asked: 
Oh, do you teach Sharah tahzib after Qazi mubarak over here? 
['Abd ul-Haqq decided to try a different approach. He asked:] 
What are you working on right now? 
Ahmad Riza: Teaching, writing of fatawa, and writing. 
f~^\ 'Abd ul-Haqq: In what field do you write? f"\ 

Ahmad Riza: Legal questions (masa'il), religious sciences 
{diniyat), and rebuttal of Wahhabis (radd-e wahhabiyya) . 
'Abd ul-Haqq: Rebuttal of Wahhabis? [A discussion about the 
best authority in this field of disputation followed, at the end 
of which 'Abd ul-Haqq fell silent.] (Bihari, 1938: 33-34) 

The tone of the exchange leaves the reader in no doubt as to the 
winner. Ahmad Riza Khan had defeated 'Abd ul-Haqq 
Khairabadi, who belonged to an eminent family of 'ulama in 
the ma'qulat tradition, with links to Farangi Mahall. Robinson 
goes so far as to say that the Farangi Mahalli family's "impact 
in northern India ... was intensified by the development of 
a powerful offshoot, another great school specializing in 
ma'qulat scholarship, that of Khavrabad in western Awadh, 
whose notable scholars [included] Fazl-e Haqq Khairabadi" 
(Robinson, 2001: 67). Given that Ahmad Riza's family also 
adhered to the tradition of ma 'qulat studies rather than the 


ch3.044 10/12/2004 5:17 PM Page 6C 


hadith scholarship emphasized bv the Shah Wali Ullah familv in 
Delhi, there was no philosophical difference between the two 
men. Moreover, Ahmad Riza's vouth and his own family's rela- 
tive obscurity in the world of 'ulama scholarship ('which only 
went back two generations) compared to ' Abd ul-Haqq's at this 
time, would lead one to expect him to be deferential to the 
older man. Instead, the conversation as reported by Zafar 
ud-Din Bihari indicates that Ahmad Riza had alreadv mastered 
the works of logic (standard texts of the Dars-i Nizami 
svllabus) that the nawab of Rampur had suggested he study 
under 'Abd ul-Haqq.The only person who ever corrected any 
of Ahmad Riza Khan's writings, Bihari reports, was his father, 

Apparently Ahmad Riza Khan took a personal dislike to 'Abd 
ul-Haqq Khairabadi, for we are told that on another occasion 
when Ahmad Riza was traveling to Khairabad with a revered 
friend of the family, who was planning to visit 'Abd ul-Haqq 
\7" Khairabadi, Ahmad Riza refused to accompanv him, saying that —kzj~ 

'Abd ul-Haqq was in the habit of saving things "detrimental to 
the glorv (shan) of the . . . 'ulama", and that he would therefore 
prefer to visit someone else (Bihari, 1938: 176). 

The fact that Ahmad Riza's visit to the nawab 's court was 
occasioned bv his writing an opinion that contradicted 
Maulana Irshad Hussain Rampuri's is also part of this pattern. 
If the exchange with Maulana 'Abd ul-Haqq tells the reader 
about the depth of his learning and the range of his scholarship 
(I will examine "what he meant by "rebutting Wahhabis" in a 
subsequent chapter), his contradiction of Maulana Irshad 
Hussain is intended to show that he had an independent 
mind, was a skillful logician, and had outstripped his elders 
early on in his career. The spirit of competition demonstrated 
here was also to characterize the claims and counterclaims 
made by rival Muslim movements in the later nineteenth 


ch3.044 10/12/2004 5:17 PM Page 61 



If the responsibility for writing fatawa at age fourteen at the end 
of his Dars-i Nizami studies marked a watershed in Ahmad 
Riza's life, so too did his discipleship to Sayyid Shah Al-e Rasul 
in 1877, when he was twenty-one. Shah Al-e Rasul was in his 
eighties at the time and died two years later, so the tie between 
them was not close — for Ahmad Riza had not spent time with 
him prior to his discipleship, not even the customary forty-day 
period (chilla) of waiting and training. Shortly before his death, 
however, Shah Al-e Rasul entrusted Ahmad Riza's spiritual 
development to his grandson, Shah Abu'l Husain Ahmad, 
known as Nuri Mivan (1839—1906), who was Ahmad Riza's 
senior by about fifteen years, and the relationship between the 
two men did become close. 


Ahmad Riza's biographv indicates the importance of the tie 
between Shah Al-e Rasul and Ahmad Riza bv reference to 
dreams. Thus it is recorded that before his journey to Marehra 
with his father, Ahmad Riza experienced a period of painful 
spiritual longing. His grandfather appeared to him in a 
dream and assured him that he would soon be relieved of his 
pain. The prophecv was fulfilled when Maulana 'Abd ul-Qadir 
Badavuni came to their house and suggested that both father 
and son affiliate themselves to Shah Al-e Rasul. Shah Al-e 
Rasul was also awaiting his arrival, for he already knew (we 
are told) that this new disciple would be the gift he could 
present to God after his death, when God would ask him "what 
he had brought Him from this world (Hasnain Riza Khan, 
1986: 55— 56). Because he was alreadv so well advanced 


ch3.044 10/12/2004 5:17 PM Page 62 


spiritually, the forty-day waiting period had not been 


The decision as to whom Ahmad Riza and his father should bind 
themselves (for they did so together) in this all-important 
relationship was probably dictated in part by Shah Al-e Rasul's 
genealogical history. The Barkativya family of Marehra to 
which Shah Al-e Rasul belonged were Sayvids, or descendants 
of the Prophet through his daughter Fatima and son-in-law 
' Ali. His very name "Al-e Rasul," meaning "[the] family of the 
Prophet," indicates as much. Other males in the family had 
similar names. Shah Al-e Rasul's younger brother, for example, 
"was calledAwlad-e Rasul, or"children of the Prophet." Women 
in the family "were often named Fatima or a compound thereof, 
such as Khairiyat Fatima, "Fatima's well-being." Although such 
names were not limited to Savvid families, in this case they 
were indicative of such status. 

The Barkativva Savvids had migrated to India, via Iraq and 
Ghazni (in present-day Afghanistan), in the thirteenth century. 
They had settled down in Marehra, a small country town 
(qasba) about a hundred and twenty miles southeast of Delhi, in 
the seventeenth century, after an earlier period of residence in 
Bilgram, western Awadh.The Mughals had awarded religious 
families such as the Barkativva Savvids revenue-free (mu'aji or 
madad-e ma 'ash) lands to support them. The family name prob- 
ably referred to their illustrious seventeenth-century ancestor, 
Savvid Barkat Ullah (1660—1729), who founded the hospice 
(khanqah) around which later generations of the family lived 
and grew up. In time, their settlement came to be known as 
"Basti Pirzadagan" (Qadiri, c. 1927) . 


ch3.044 10/12/2004 5:17 PM Page 63 


The sufi affiliation of the Barkativva Savvids was with the 
Qadiri order, one of the three major sufi orders in India since 
the eighteenth century (the others are the Chishti and the 
Naqshbandi). The Qadiri order traces its origins to 'Abd 
al-Qadir Jilani Baghdadi (d. 1166), and has been popular in 
South Asia since the fifteenth centurv. I take up the significance 
of this sufi affiliation to Ahmad Riza in the next chapter. 


Shortly after Ahmad Riza became Shah Al-e Rasul's disciple in 
the ritual known as bai 'a, he and his father undertook another 
important journev, namelv, the pilgrimage to Mecca. By per- 
forming this ritual, Ahmad Riza was fulfilling one of the 
so-called "pillars" of Islam, a necessary step before he could 
assume his role as the leader and Renewer of his community. In 
-(^y— this sense, he was undertaking a rite of passage, a transforma- — (^y- 

tive event which allowed him to return to Bareillv with greater 



Mecca and Medina, the two holiest cities for Muslims, were 
under Ottoman control at this time. Mecca is the center of the 
Muslim pilgrimage because it houses the sanctuary which 
Abraham is believed to have built with his son Ishmael in 
antiquitv and also because it is the city in which Muhammad was 
born. Bv the nineteenth century it "was first and foremost as the 
Prophet's birthplace that it "was revered. Medina, the city where 
Muhammad lived in the second phase of his career and where he 
is buried, is not a part of the pilgrimage. But because he is buried 
there, many Muslims making the pilgrimage visit it too. Ahmad 
Riza and his father, not surprisinglv, went to both places. 

While Ahmad Riza was in Mecca he received recognition 
from 'ulama in high positions of authoritv. Savvid Ahmad 
Dahlan, the mufti of the Shafi'i law school, gave him a certificate 


ch3.044 10/12/2004 5:17 PM Page 64 


(sanad ) in several fields of knowledge — hadith (the traditions 
of the Prophet), exegesis of the Qur'an (tafsir) , jurisprudence 
(Jiqh), and principles of jurisprudence (usul-ejiqh) .The other 
scholar to do so was the mufti of the Hanafi school of law. 
Although Ahmad Riza had not studied under these scholars 

formally thev authorized him to teach in the fields thev had 

j j j 

specified and to cite their names when doing so. 

Equally important, though in a different way, was his 
encounter with Husain bin Saleh, the ShafTi imam. The latter 
noticed him one dav during the evening praver and took him 
aside. We are told that he held"his forehead for a long time, sav- 
ing at length that he saw Allah's light in it. He then gave him a 
new name, Zia ud-Din Ahmad, and a certificate in the six col- 
lections of hadith, as well as one in the Qadiri order, signing it 
with his own hand" (Rahman 'Ali, 1961: 99). This encounter 
emphasized the spiritual (sufi) rather than the scholarly sources 
of Ahmad Riza's authority. So too did another — Medinan — 
\7" experience, a dream in which Ahmad Riza was assured that -\^r 

he was absolved of all his sins. As most Muslims believe that 
this assurance is granted to very few, this vision can be read as 
a claim to leadership of the Ahl-e Sunnat movement in 
coming vears. 


Ahmad Riza's proclamation as the mujaddid of the fourteenth 
Islamic centurv occurred in unusual circumstances and in an 
unusual manner. Throughout the 1 890s the Ahl-e Sunnat had 
been busy organizing meetings opposing the Nadwat 
al-'Ulama. Ahmad Riza had plaved an active part in this oppos- 
ition movement, writing some two hundred fatawa on this 
issue alone. Starting in 1897, the Ahl-e Sunnat also published a 
monthlv journal (Tuhfa-e Hanajlyya, the Hanafi Gift) from 


ch3.044 10/12/2004 5:17 PM Page 6 



Patna, Bihar, which brought together anti-Nadwa articles, 
poems, and news reports about the annual meetings. It was in 
print until about 1910. 

Ahmad Riza's stature was heightened when one of his fatawa 
was published in 1900 with the approval and certification of 
sixteen 'ulama from Mecca and seven from Medina. In October 
of that vear the annual meeting of the Ahl-e Sunnat 'ulama took 
place in Patna, during which time a new madrasa, the Madrasa 
Hanafiyya, was formally opened. The Nadwa was holding its 
own annual meeting in a different part of town. In fact the 
Ahl-e Sunnat appears to have deliberatelv chosen to hold its 
meeting in the same place and at the same time as the Nadwa, 
in order the better to undercut its message. 

It was during the weekdong meetings that occurred at Patna 
that one of the 'ulama present referred to Ahmad Riza in his ser- 
mon as the "mujaddid of the present centurv." According to 
Zafar ud-Din Bihari, all those present seconded the idea, and 
\7" later thousands of others, including several 'ulama from the —kzj~ 

Haramain (Mecca and Medina) did so as well. As he writes, 
there was thus consensus among the 'ulama of the Ahl-e Sunnat 
on the question. Zafar ud-Din adds that Ahmad Riza fulfilled 
the requirements of a mujaddid, namelv, that he (it could not be 
a woman) be a Sunni Muslim of sound belief, endowed with 
knowledge of all the Islamic "sciences and skills," the "most 
famous among the celebrated of his age," defending the faith 
without fear of "innovators" "who would criticize him, and also, 
according to Zafar ud-Din, a profound sufi. He also had to sat- 
isfy the technical requirement that he be well known when one 
centurv ended and the other began (or, as Bihari puts it, at the 
end of the century in which he was born and the beginning of 
the centurv in which he "was to die) .The thirteenth Islamic cen- 
tury had ended on 1 1 November 1882, and Ahmad Riza had 
indeed begun to establish a reputation among the 'ulama of 
north India bv then. The fact that 'ulama in Mecca and Medina 


ch3.044 10/12/2004 5:17 PM Page Si 


were ready to append their names to his commentary on the 
Prophet's knowledge of the unseen (see below) was taken by 
his followers as confirmation that he was indeed the mujaddid of 
the fourteenth Islamic century. 


Ahmad Riza's scholarly reputation rested primarily on his writ- 
ing of fatawa, a responsibility entrusted to him by his father when 
he "was fourteen and carried out until his death in 1 92 1 . A fatwa 
is written in answer to a question asked by a Muslim man or 
woman to a mufti, a scholar of Islamic law, about a legal or moral 
problem, such as an inheritance dispute, a debate about vari- 
ations in the prayer ritual, or questions of faith and belief. The 
legal questions are not usually of the type posed to lawyers in the 
West, for the law in which the mufti is an expert is religious law. 
-yy— The nearest equivalent in the West to a fatwa is rather the answers —yy- 

to the questions posed to "the Ethicist" in the New York Times 
Sunday Magazine. In a Muslim city, there are hundreds of 
"ethicists," all willing to answer questions. They are the religious 
scholars, known as muftis when they act as authors of fatawa. 

To qualifv as a mufti, a scholar needs to have expert know- 
ledge of sources of the law — the Qur'an, the sunna (the 
example of the Prophet), the consensus of the community 
(ijma), and analogical reasoning (qiyas) — as well as familiarity 
with the legal tradition of the school (madhhab) to which he 
himself and the questioner belong. If no direct answer could be 
found in the sources, a person endowed with such knowledge 
was qualified to applv his judgment (ijtihad) to the question at 
hand. The latitude permitted to a mufti — or that he permitted 
himself — in interpreting the sources has varied considerably 
throughout Muslim history. For many centuries ijtihad had 
been downplayed, and following one's school of law (taqlid) 


ch3.044 10/12/2004 5:17 PM Page 67^ 


had been the norm. This was also the case in colonial India. 
But regardless of the mufti's theoretical stand on ijtihad, the 
activity had never ceased in practice, since new problems and 
questions constantly needed answers. The mufti's answer, 
while considered authoritative (on account of his knowledge), 
did not have the force of law. 

The question-and-answer format of a fatwa is also worth 
noting. Some hold it to go back to the Prophet himself, on those 
occasions when he acted in his own capacity when asked a ques- 
tion: "It is reported, for example, that [a believer] asked the 
Prophet, 'O Messenger of God, is the pilgrimage to be per- 
formed every year or onlv once?' He replied, 'Onlv once, and 
whoever does it more than once, that is an [especiallv meritori- 
ous] act' " (Masud, Messick, and Powers, 1996: 6). Such reports 
are recorded in the hadith literature, which complements the 
Qur'an as a secondarv source. In later generations, the activitv 
of the mufti was seen as a continuation of the Prophet's 
\7" example. Thus the fourteenth-centurv scholar al-Shatibi wrote —kzj~ 

that "the mufti stands before the Muslim community in the 
same place as the Prophet stood" (Masud, Messick, and 
Powers, 1996:8). 

Because the work of writing fatawa "was "religious" in nature 
— in other words, it was a means of guidance and benefit to 
other Muslims — muftis were forbidden to take bribes or gifts 
of any kind from the person who had asked the question. Even 
private muftis were expected to render their judgments for 
free (muftis who worked for the state received salaries, like 
the qadis in Islamic courts) . Whether all did so is unlikelv. In 
some cases, the problem of compensation was solved bv the 
creation of pious endowments (awkaf) specificallv for muftis 
and teachers. 

In colonial India, as noted in previous chapters, the loss of 
state power and the lack of qadis in British Indian courts 
increased the need for muftis, as thev were the sole authoritv 


ch3.044 10/12/2004 5:17 PM Page 6£ 


left to guide the community. The latter half of the nineteenth 
century also saw a rapid increase in communications networks 
and new and inexpensive print technologies that allowed 
'ulama such as Ahmad Riza to reach a wider group of people 
and forge a network of relationships beyond the immediate 
local area. This created competition for followers, especially as 
different reform movements made their appearance, so that 
the activity of writing and publishing fatawa became highlv 
competitive. They "were a way of reaching the hearts and minds 
of Sunni Muslims throughout the subcontinent, since thev 
dealt with practical issues rather than academic problems of an 
erudite nature. 


Fatawa vary from the verv short and simple to the long 
and complex, depending on their intended audience — those 
written for ordinary believers tend to be simple, straight- 
forward, and without citation of sources, while those written 
by scholars for scholars were naturally likelv to be complex. 
However, even "when simple in form, a fatwa often contains 
hidden cues about the scholar's point of view. An example 
from a Deobandi fatwa about the pilgrimage, written by 
Maulana Rashid Ahmad Gangohi (d. 190S) in the 1890s, is 
instructive : 

Query: What of a person who goes to Noble Mecca on hajj and 
does not go to Medina the Radiant, thinking, "To go to Noble 
Medina is not a required duty, but rather a worthy act. 
Moreover, why should I needlessly . . . risk . . . property and life 
[in view of the marauding tribes along the way] . . . and [spend] a 
great deal of money?" ... Is such a person sinful or not? 


ch3.044 10/12/2004 5:17 PM Page 6? 


Answer: Not to go to Medina because of such apprehension is a 
mark of lack of love for the Pride of the World [the Prophet 
Muhammad], on whom be peace. No one abandons a worldly 
task out of such apprehension, so why abandon this pilgrimage? 
. . . Certainly, to go is not obligatory. [But] some people, at any 
rate, think this pilgrimage is a greater source of reward and 
blessing than lifting the hands in prayer and saying "amin" out 
loud. Do not give up going out of fear of controversy or concern 
for your reputation. . . . Even if not a sinner, this person lacks 
faith in his basic nature. (Metcalf, 1996: 184) 

At first sight the fatwa seems only to be answering a simple ques- 
tion, namelv, what the mufti thinks of a nonobligatory ritual act, 
that of paving homage to the Prophet bv visiting his grave at 
Medina while performing the pilgrimage to Mecca (the latter 
being incumbent upon all adult Muslims, men and women, to 
perform once in their lifetimes). But on second reading vou 
notice that it engages in polemics against the Ahl-e Hadith. 
Xy Since Medina is not far from Mecca (about 270 milesnorth), ~xt~ 

manv Muslims make the journey there either before or after 
the pilgrimage itself. But the reply contains several clues that 
tell us that the question and answer were directed against the 
Ahl-e Hadith. The practice of "lifting the hands in praver and 
saving 'amin' out loud" was specific to the Ahl-e Hadith and dis- 
tinguished them from other Sunni Muslims in South Asia. It was 
also the Ahl-e Hadith who "opposed pilgrimage (ziyarat) to the 
Prophet's tomb in Medina, as thev opposed pilgrimage to all 
tombs," sharing the orientation of the Wahhabis who "had gone 
so far as to destroy the tomb of the Prophet" in the early nine- 
teenth century (Metcalf, 1996: 186-187). 

Now let us look at a very different case, also from Deoband. 
Masud's (1996) studv of two Deobandi fatawa shows how the 
'ulama sometimes initiated a process of change in the shari'a 
(bv applving their independent reasoning), but used the cit- 
ation of respected medieval sources to present their judgment 


ch3.044 10/12/2004 5:17 PM Page 7( 


as an exercise in submission to authority (taqlid), that is, the 
authority of their particular school of law, which in British 
India "was (and is) overwhelmingly Hanafi. Bv comparing two 
fatawa bv Maulana Ashraf 'AliThanawi (d. 1943) on "whether 
the apostasv of a Muslim woman annulled her marriage, Masud 
shows that Thanawi changed his position between 1913, the 
date of his first fatwa, and 1931, when he revised his opinion. In 
1913, he had ruled that apostasy did result in annulment, 
whereas in 1931, applying Maliki law (thus having recourse 
to legal opinion in another school, or taljiq), he argued that 
"apostasv did not annul the marriage contract and could not be 
used as a legal device [to terminate the marriage]" (Masud, 
1996: 193-203; cf. p. 37, on the legal issue). 

The fatwa is a clear case of the application o£ ijtihad, but is not 
presented as such. Had the argument been seen as an instance of 
ijtihad being exercised bv a single mufti rather than one which 
had the weight of traditional jurisprudential authority behind it , 
\7" it might not have been accepted. As this instance shows, ijtihad— —kzj~ 

far from being something the mufti could be proud of engaging 
in —had to be wrapped up in the guise of taqlid. 

This case — dealing with apostasv and the difficult v Muslim 
women experienced in initiating a divorce — is clearlv more 
complex than the first. The 1931 fatwa (the revised one) was 
published as a book of over two hundred pages. Its publication 
led to a political effort for marriage reform bv the national 
partv representing Deobandi and other 'ulama, the Jamiyvat 
al-'Ulama-e Hind, and in 1939 resulted in the enactment of 
national legislation in British India to facilitate the dissolution 
of Muslim marriage on specific legal grounds. 


Like the other Muslim movements of the late 1800s, the Ahl-e 
Sunnat movement established a Dar al-Ifta, a "house for issuing 


ch3.044 10/12/2004 5:17 PM Page 71 


fatawa." Unlike the other movements, however, that of the 
Ahl-e Sunnat "was attached to Ahmad Riza's house rather than to 
the school established in 1904. It was from here that, assisted bv 
his closest and ablest students, he responded to the questions 
that came in dailv from all over the countrv. 

Zafar ud-Din Bihari, Ahmad Riza's disciple and biographer, 
relates that every evening Ahmad Riza would set aside some 
time to meet people at his home. The day's mail would some- 
times be opened and read out loud. Depending on the nature of 
the question, Ahmad Riza would either answer it himself or 
pass it on to one of his students to do so. Thus, if it dealt with 
sufism (tasawwuf), was particularly complex, or had not come 
up before, he would answer it himself. Subjects deemed less 
difficult were handled bv a small group of students. He worked 
in the privacv of his personal library or in his familv living quar- 
ters (zenana khana) , and took pride in answering everv question 
as quicklv as possible. Regarding it as a religious (shar 'i) dutv, he 
\7" was offended when someone offered him payment for his ~x3~ 

fatwa. So devoted was he to the task of responding (istifta), 
wrote Zafar ud-Din, that he did so even when he was sick. We 
are told that on one remarkable occasion he was seen dictating 
twentv-nine fatawa to four scribes while sick in bed: while one 
scribe wrote down the answer to one question, he dictated the 
answer to the second one to another, and so on, until all 
twenty -nine questions had been answered (Bihari, 1938: 

It was by writing down and copving fatawa dictated bv 
Ahmad Riza that his students learned his stvle of fatwa- writing. 
Once thev had mastered the skill, Ahmad Riza was able grad- 
ually to entrust some of the work to them. He considered his 
student Amjad 'Ali A'zami to be the most skilled, and asked 
other students to learn from him. 

Many — though bv no means all — of Ahmad Riza's fatawa 
were published in a twelve-volume collection known as the 


ch3.044 10/12/2004 5:17 PM Page 72 


Fatawa-e Rizwiyya, some at the Hasani Press owned by his 
brother. Only two appear to have been published during his 
lifetime. Publication of the others did not begin until the 
1950s, and was still ongoing in the 1980s. The process was 
begun bv Maulana Mustafa Riza Khan (d. 1981), Ahmad Riza's 
younger son. Perhaps the lack of funds held back further publi- 
cation. Unfortunatelv, when publication was finally resumed it 
was found that many of the handwritten fatawa were damaged, 
and laborious effort was required to assemble the later vol- 
umes. Nonetheless, most of them were published. A different 
problem arose when a printer kept delaying publication on one 
pretext or another until the editors caught on to the fact that he 
had Deobandi views! 

Some fatawa discuss a range of issues related to the question 
but nevertheless distinct from it, especially when they are long 
and complex. Ahmad Riza tended to expand, rather than 
restrict the range. So too did the Deobandis in the same period. 
\7" As Metcalf savs, "Any categorization of the topics covered in —kzj~ 

[Rashid Ahmad Gangohi's] pronouncements is necessarily 
crude, for a single fatwa could often illustrate at once a variety 
of issues concerning belief, practice, jurisprudential prin- 
ciples, and attitudes toward other religious groups" (Metcalf, 
1982: 148). In a fatwa responding to the question as to whether 
a Muslim who had become an Ahmadi "was an apostate, Ahmad 
Riza raised issues relating not only to apostasy and marriage but 
also to the nature of prophecy. 

Ahmad Riza's opinions were alwavs forcefully expressed. 
He was decisive in his judgments, giving clear guidance to 
his followers on right and wrong and backing up his 
opinions by citation of an arrav of scholarly writings that added 
to his religious authority. At a time when so many different 
points of view were being expressed, one imagines that the 
ordinary believer would have found this note of certainty 


ch3.044 10/12/2004 5:17 PM Page 7C- 



In 1 90S— 6, Ahmad Riza went to Mecca and Medina for the sec- 
ond time. In 1906, Mecca was a place "where diverse opinions 
flourished. The Wahhabi movement (consisting of an alliance 
between the followers of Muhammad ibn 'Abd al-Wahhab and 
the Saudi family) "was based in Najd in central and eastern 
Arabia. In the Hijaz (as coastal northwestern Arabia is known), 
however, power was in the hands of Sharif 'Ali (r. 1905—8) of 
Mecca. Although technicallv the Sharif (also known as Amir) 
was an appointee of the Ottomans — the Hijaz being a province 
of the Ottoman empire — in fact the amir exercised 
autonomous control. Sharif 'Ali died in 1908, whereupon 
Sharif Husavn — memorablv portraved in the film Lawrence of 
Arabia for his part in leading the Arab Revolt against the 
Ottomans — came to power. 
r~) Ahmad Riza's views found a receptive audience among some — £^- 

Meccan 'ulama who disliked the Wahhabi perspective. By this 
time he "was a well-known Indian scholar, one who had been in 
correspondence with the 'ulama of the Hijaz during the 1890s 
when he had sought confirmation for his fatawa in opposition to 
the Nadwa.Two Meccan 'ulama now asked for his opinion on the 
status of paper money. In response he wrote a fatwa entitled 
Kafl al-Faqih al-Fahimfi Ahkam Qirtas al-Darahim (Guarantee of 
the Discerning Jurist on Duties relating to Paper Monev) . One 
scholar reportedlv stated, "Although he was a Hindi [an Indian] , 
his light was shining in Mecca" (Malfuzat, vol. 2, p. 17). There 
were other marks of respect: confirmation of his opinion on a 
ritual related to the pilgrimage (despite a contrary opinion bv 
some Meccan scholars) and visits to his home. Bearing in mind 
that onlv a segment of the 'ulama was involved, we might even say 
that relations between center and periphery, Mecca and India, 
had been reversed during Ahmad Riza's three-month stay. 


ch3.044 10/12/2004 5:17 PM Page 74 


Ahmad Riza wrote three fatawa while in Mecca. While the 
first is the one on paper money, the second, Al-Dawlat 
al-Makkiyya bi'l Maddat al-Ghaybiyya (The Meccan Reign onThat 
Which Is Hidden), deals with the Prophet, particularly his 
"knowledge of the unseen" ('ilm-e ghaib), which had been an 
object of debate between Ahmad Riza and the Deobandi 'ulama 
for some time. (The third fatwa, Husam al-Haramayn 'ala Manhar 
al-Kufr wa'l Mayn [The Sword of the Haramavn at the Throat of 
Unbelief and Falsehood] , is discussed in the next chapter.) 

Ahmad Riza made two related arguments in Al-Dawlat 
al-Makkiyya. The first 'was that God's knowledge is distinct from 
that of the Prophet. As he wrote: 

One is the masdar or source, from where knowledge 
emanates, and the other is dependent upon it. In the first 
case, knowledge is zati, that is it is complete and independent 
in itself. ... In the second case, it is ata'i, that is "gifted" by an 
outside source. Zati knowledge is exclusively Allah's. ...The 
second kind is peculiar to Allah's creatures. It is not for Allah. 
(Al-Dawla al-Makkiyya, IS, 17, 19) 

Having made this fundamental distinction between God's 
knowledge and the Prophet's, Ahmad Riza then proceeded at 
great length (the fatwa is approximately two hundred pages 
long) to lav out the scope of the Prophet's knowledge of the 
unseen. He began bv saving that some knowledge of the unseen 
is possessed even by ordinary human beings : Muslims believe in 
the resurrection of the dead, heaven and hell, and other unseen 
things, as commanded by God. The knowledge possessed bv 
prophets was of course much greater than that of ordinary 
people, and although it was but a drop in the ocean compared 
to what God knows, it was itself "like an ocean beyond count- 
ing, for the prophets know, and can see, evervthing from the 
First Dav until the Last Dav, all that has been and all that will be" 
(Al-Dawlat al-Makkiyya, 57, 59). 


ch3.044 10/12/2004 5:17 PM Page 7 



As for the Prophet, his knowledge kept growing as the 
Qur'an was revealed to him over a twenty-two-year period 
(610—32 CE).Thus, Qur'anic passages that refer to his lack of 
knowledge about something refer to a time when knowledge of 
the particular matter still had not been revealed, and were 
abrogated by later verses on that subject. By the end of his life, 
however, God had told him about 

the tumult of the resurrection (hashr o nashr), the accounting, 
and the reward and punishment. So much so that he will see 
everyone arriving at their proper places [at the end of times], 
whether heaven or hell, or whatever else God may tell him. 
Undoubtedly, the Prophet knows this much, thanks to God, 
and God alone knows how much else besides. When He has 
given his beloved [Muhammad] so much, then it is apparent 
that knowledge of everything in the past and the future, 
which is recorded in the Tablet (lawh-e mahfuz) is but a part of 
his knowledge as a whole. (Al-Dawlat al-Makkiyja, 77) 

The Prophet also knew what was going on inside people's 
minds: "He knows the movement and glance of the evelid, the 
fears and intentions of the heart, and whatever else exists" 
(Al-Dawlat al-Makkijja, 90). 

And, most controversially (for the Deobandis, among 
others, denied this), the Prophet had knowledge of the five 
things referred to in Qur'an 3 1 : 34: 

Only God has the knowledge of the Hour. 

He sends rain from the heavens, 

and knows what is in the mothers' wombs. 

No one knows what he will do on the morrow; 

no one knows in what land he will die. 

Surely God knows and is cognisant. 

(31: 34, Ahmed 'Ali trans.) 

Ahmad Riza argued that apart from the resurrection, the other 
four things — knowledge of when it would rain, of the sex of a 


ch3.044 10/12/2004 5:17 PM Page 7f 


yet unborn child, of what one would earn on the morrow, and 
of the land where one would die — were not all that significant 
in themselves. In fact, they were rather minor in scale of import- 
ance compared to knowledge of the attributes of God, heaven 
and hell, and the like. (In fact, Ahmad Riza argued, knowledge 
of these five things had been given not only to the Prophet, but 
also to Shaikh 'Abd al-Qadir Jilani, the qutb or "pivot" at the 
head of the invisible hierarchy of saints on "whom the govern- 
ment of the world depends.) The reason God had singled out 
these five things for mention in the Qur'an was that the sooth- 
sayers (kahins) of early seventh- centurv Arabia — the age of the 
Prophet, when the Qur'an was revealed — believed thev could 
predict such things. God "wanted them to know that these 
things were "hidden" (al-ghayb) and that none could know them 
but He and those He favored. The Prophet had been favored 
with this knowledge (including the hour of the resurrection) 
but had been commanded not to reveal it. 
\7" Ahmad Riza cited two Qur'anic verses in defense of his ~x3~ 

views. They "were 3: 179, "nor will God reveal the secrets of the 
Unknown. He chooses (for this) from His apostles whom He 
will", and 72: 26— 27, "He is the knower of the Unknown, and 
He does not divulge His secret to any one other than an apostle 
He has chosen" (Ahmed ' Ali trans.) . 

In keeping with the sufi dimensions of Ahl-e Sunnat belief 
and practice (discussed in the next chapter), Ahmad Riza also 
held a number of related beliefs about the Prophet, some of 
which are found in Shi'ism: that he "was God's beloved for 
whom God had created the world, that Muhammad had been 
created from Allah's light and therefore did not have a shadow, 
and, most importantlv, that he mediated between God and the 
Muslim believer in the here and now — one did not have to wait 
for the last day and the resurrection for such mediation to 
occur. Ahmad Riza's views about the Prophet's knowledge of 
the unseen were in keeping with his overall perception of the 


ch3.044 10/12/2004 5:17 PM Page 7 7^ 


Prophet as one who was uniquely endowed bv God. Also note- 
worthy in this regard is the hierarchy of levels of knowledge 
laid out in the above fatwa: after God, Muhammad's knowledge 
was greatest, then followed the knowledge of various 
prophets, that of the 'ulama and sufi shaikhs and pirs (Shaikh 
' Abd al-Qadir Jilani foremost among these), and finallv, that of 
ordinary believers. 

In 1911, Ahmad Riza's translation of the Qur'an, entitled 
Kanz al-Imanji Tarjuma al-Qur'an (Treasure of Faith relating to a 
Translation of the Quran), was published in Muradabad, a 
north Indian citv where some of his followers were based. 
Although an English translation "was subsequently published bv 
the Islamic World Mission in Britain, it has yet to receive schol- 
arly attention. 


In the years leading up to World War I the Indian nationalist 
movement united behind the British Crown by sending troops 
all over the world to fight on behalf of the British, but with high 
hopes that after the 'war was over the process of self-rule would 
be speeded up. Into this mix were added fears on the part of the 
Muslim leadership that they might not fare too well in demo- 
cratic elections in a Hindu-dominated India, and that steps 
needed to be taken to safeguard Indian Muslim interests. This 
led a small group of Muslim leaders to form the All-India 
Muslim League in 1906. 

The 'ulama had to decide whether or not they should take a 
political stand as "well, and if so, whether they should throw 
their support behind the Indian National Congress, which was 
the dominant nationalist party, or the Muslim League, or 


ch3.044 10/12/2004 5:17 PM Page 7£ 


whether thev should form a party — or parties — of their own. 
And if they did form their own party, should they join with 
Congress in anti-British agitation, or act independently? As 
may be imagined, there were many different opinions among 
them, expressed once again in fatawa, commentaries, and 
other scholarly writings, not to mention oral debates and 
speeches made during Friday prayers. Ahmad Riza's opinion 
that there was no religious justification for Indian Muslims tak- 
ing an anti-British stand was challenged by 'ulama from other 
movements, "who accused him of being pro-British. 

During the prewar years a number of Indian Muslims had 
begun to organize around an international issue, that of helping 
the Ottoman caliph, whose empire was in danger of complete 
dismemberment by the Allies after the war. This pan-Islamic 
movement was supported bv Indian Muslim leaders such as 
Abu'l Kalam Azad (1888—1958), who owned and contributed 
regularly to the influential Urdu journal Al-Hilal, and Maulana 
'Abd ul-Bari Farangi Mahalli (1878-1926), who was involved 
in efforts to raise money for Turkish relief from India. In 1913, 
'Abd ul-Bari began an association called Societv of the Servants 
of the Ka'ba (Anjuman-e Khuddam-e Ka'ba). Ahmad Riza's 
support was sought, but he refused — not because he was 
unsvmpathetic to the plight of the Turks or because he did not 
want to protect the Ka'ba, but because he objected to the 
composition of the Anjuman. Because it strove to be an 
inclusive body, welcoming all Muslims, whether Shi'a, Ahl-e 
Hadith, modernist, or other, Ahmad Riza refused to be associ- 
ated with it. He did so on grounds similar to those he had 
expressed against the Nadwa in the 1890s, namely, that he 
could not support a body 'which included people he deemed 
"bad" Muslims (bad-mazhab) or those who had "lost their way" 
(gumrah), his terms for the groups mentioned above. 

Although he 'was all in favor of helping the Turks financially, 
Ahmad Riza believed that given the straitened circumstances of 


ch3.044 10/12/2004 5:17 PM Page 7? 


Indian Muslims there was not much that thev could do, and he 
was critical of what he saw as the wasteful expenditure of 
resources bv politically active 'ulama. In a 1913 fatwa he 
expressed his svmpathv for the plight of the Turkish people, 
quoting Qur'an 13: 11: "Verily God does not change the state 
of a people till thev change it themselves" (Ahmed ' Ali trans.). 
After suggesting that both the Turks and the Indian Muslims 
would ultimately have to depend on their own resources rather 
than external help, he went on to suggest that if every Muslim 
donated a month's salarv, living for twelve months on eleven 
months' earnings, they would be able to render the Turkish 
Muslims substantial help. 

In addition, he proposed a fourfold course of action aimed at 
making the Indian Muslim community economically and polit- 
ically self-sufficient: first, by boycotting the British Indian 
courts (as he was to do in 1917) they would save money on 
stamp duties and legal fees. Secondly, they should buv whatever 
\7" goods they needed from fellow Muslims, thereby keeping —kzj~ 

money "within the Muslim communitv (and not allowing them- 
selves to go into debt to Hindu moneylenders). Thirdly, 
wealthy Muslims in large cities such as Bombay should open 
interest-free banks for use bv Muslims. And finally, all Indian 
Muslims should strengthen themselves bv acquiring the know- 
ledge of their faith (Ahmad Riza Khan, 1913). 

This is the onlv fatwa known to me in which Ahmad Riza 
addressed himself to practical issues rather than religious 
ones. It is interesting that he concentrated entirely on eco- 
nomic self-sufficiency, and said nothing about political action. 
To the end of his life he remained convinced that the Indian 
Muslim communitv needed internal reform rather than 
political independence. His reference to Hindus in this fatwa 
is also revealing. In his view, political alliances forged "with 
Hindus for the sake of overthrowing the British were mis- 


ch3.044 10/12/2004 5:17 PM Page 8( 


Taking Ahmad Riza's cue, leaders of the Ahl-e Sunnat move- 
ment formed their own associations and organizations address- 
ing such issues as helping theTurks, instead of joining nationally 
prominent ones such as the Anjuman. In fact, several other 
Muslim groups formed associations of this kind in the teens and 
twenties of the twentieth century. But fissures began to appear 
in the Ahl-e Sunnat movement as a younger generation of 
Ahl-e Sunnat 'ulama challenged his apolitical stance. I follow 
this development in chapter S bv studying a debate on a matter 
of religious ritual, the call to praver, which culminated in a 
court case in 1917. 

Not surprisingly, the politicization of the Muslims was 
speeded up bv the war. The Khilafat movement, launched in 
1919 to preserve the caliphate after the Ottoman defeat in 
WorldWar I, was the first national movement in which Hindus 
and Muslims struggled side bv side against the British in sup- 
port of a specifically Muslim issue. By this time Mohandas 
\7" K. Gandhi (known as Bapu ["father"] to his followers) had —kzj~ 

returned to India after many years in South Africa and had 
assumed leadership of the Indian National Congress. 
Determined to "work toward Hindu— Muslim unitv, he saw in 
the khilafat issue an opportunity to bring the two sides 
together. In 1920, the Muslim leadership reciprocated bv 
urging Indian Muslims to join with the Indian National 
Congress in its nationwide Noncooperation movement 
(1920—2) to oust the British from India. The Noncooperation 
movement involved everything from giving up British honors 
(titles bestowed on eminent Indians, for example) to 
boycotting British courts and schools and the nonpayment of 

On the Muslim side of the Khilafat movement were 
leaders such as Maulana 'Abd ul-Bari, the 'Ali brothers 
(Shaukat 'Ali and Muhammad 'Ali), Maulana Azad, Mufti 
Kifavatullah, 'Abd ul-Majid Badavuni (a sufi disciple of 


ch3.044 10/12/2004 5:17 PM Page 81 


Maulana 'Abd ul-Muqtadir Badayuni), and a number of 
Deobandi 'ulama, including Shabbir Ahmad 'Usmani and 
Husain Ahmad Madani. In 1919 they had created the first 
national political organization of 'ulama, namelv, the Jam'iyyat 
al-'Ulama-e Hind (Society of the 'Ulama of India). Its goals 
were at once pan-Islamic (protection of Arabia, particularly 
the holy cities of Mecca and Medina) and national (the 
promotion of Muslim Indian interests and pursuit of freedom 
from British rule). Deeming the British rulers the greater 
enemy, it was willing to cooperate with Hindus on the national 

Ahmad Riza, characteristically, opposed the Khilafat move- 
ment. Part of his objection related to his insistence that the sul- 
tan of Turkey could not claim the title of caliph as he was not 
of Quravsh descent (there were other shar 'i conditions as well, 
though this was the most important). The other had to do "with 
his view that Muslims could not seek the cooperation of kajirs 
\7" (unbelievers) in the pursuit of a religious (shar 'i) goal — a clear —kzj~ 

indication that he was looking at the Khilafat movement in reli- 
gious rather than political terms. 


In the late summer of 1920, Maulana 'Abd ul-Bari launched a 
new movement, known as the Hijrat (Emigration) movement. 
He issued a fatwa declaring that Muslims should abandon 
British-ruled India and migrate to a neighboring Muslim terri- 
tory. Hoping that thev could acquire land in Afghanistan, some 
twentv thousand people — most of them Pathans from what is 
today the Northwest Frontier Province in Pakistan, but also 
peasants from the United Provinces and Sind — sold their pos- 
sessions and marched toward Kabul. However, Amir 
Amanullah Khan (r. 1919—30) had just come to power in 


ch3.044 10/12/2004 5:17 PM Page 


Afghanistan in the previous vear, having launched a jihad 
against the British (an event known to historv as the Third 
Afghan War) with the help of Afghan religious leaders, to get 
rid of British control of the country. Although defeated, he had 
concluded a settlement with the war-wearv British that 
accorded Afghanistan full independence, including control 
over foreign affairs. Fearing the economic consequences of the 
influx of so many people, Amanullah closed Afghanistan's fron- 
tiers to the emigrants, forcing most of them —now destitute — 
to go back to their homes. 

This was the context for Ahmad Riza's fatwa, published in 
the Rampur newspaper Dabdaba-e Sikandari in October 1920, 
declaring that India was dar ul-hlam, or a land of peace, not 
dar ul-harb, a land of "war. In fact the fatwa had originally been 
written in the 1 880s, but it was as relevant as ever. He wrote: 

In Hindustan . . . Muslims are free to openly observe the two 
^~A 'ids, the azan, ... congregational prayer ... which are the signs ("\ 

of the shari'a, without opposition. Also the religious duties, 

marriage ceremony, fosterage There are many such 

matters among Muslims ... on which . . . the British 
government also finds it necessary to seek fatawa from the 
'ulama and act accordingly, whether the rulers be Zoroastrian 
or Christian. ... In short, there is no doubt that Hindustan is 
dar al-Islam. (Ahmad Riza Khan, 1 888-9) 

Despite the anti-British sentiment among Indian Muslims at 
this time, he continued to insist that the fundamental shar'i sta- 
tus of the countrv had not changed. There was thus no justifica- 
tion for either jihad or hijrat. 

A flood of accusations of his pro-British svmpathies fol- 
lowed, including an allegation that he had met with the 
Lieutenant Governor of the United Provinces, Sir James 
Meston, while in Nairn Tal, the hill retreat where he went in the 
last few years of his life to observe the Ramadan fast. He also 


ch3.044 10/12/2004 5:17 PM Page 8: 


had to answer charges of lack of concern for the Turks and the 
holy cities of Mecca and Medina. 


While Ahmad Riza's views on national issues mav not have 
enjoved widespread support outside the Ahl-e Sunnat 
movement, he continued to be revered and loved bv his core 
group of followers to the end of his life. An event in 1919 illus- 
trated this clearlv. That year, he undertook a long journey by 
train from Bareillv to Jabalpur, in central India, to perform the 
dastar-bandi (tying of the turban) ceremony ("which marks the 
end of a student's career, akin to a student's graduation cere- 
monies), for one particular student, Burhan ul-Haqq Jabalpuri 
\7" (d. 1 984) . Bv this time his health was poor, and the journev of "\7 

about six hundred miles took two davs. He was greeted like 
royalty not only at the Jabalpur station, but at smaller stations 
along the way. People thronged to kiss and touch his feet, and 
lined the streets on the "way to the station. 

Once arrived there, Ahmad Riza was surrounded bv well- 
wishers and distributed lavish presents to all and sundry, not 
just to his hosts. Zafar ud-Din Bihari writes about everybody's 
amazement at the money, gold ornaments, and clothes which 
he had brought as gifts. In return, they gave nazar, a token gift 
given to a sufi pir, and feasts throughout his one-month stay. 
Bihari also reports that at a series of public meetings people 
came forward to seek his pardon for sins of omission and com- 
mission — some of them minor, such as shaving the beard or 
dveing the hair black, both of which he disapproved of. 
Spiritual matters of deeper import were discussed in private 
sessions (Bihari, 1938: 56—57). 


ch3.044 10/12/2004 5:17 PM Page 84 



In the years before his death in 1 92 1 , Ahmad Riza made a series 
of decisions about the leadership of the movement in the 
future. Already in 1 9 1 S , as reported bv the Dabdaba-e Sikandari, 
he had chosen his older son, Hamid Riza Khan (1 875— 1943), as 
his sufi successor (sajjada nishin). After 1921, Hamid Riza 
became the head of what came to be known as the Khanqah-e 
' Alivva Rizwivva, the new sufi order named after Ahmad Riza. 
Ahmad Riza's vounger son, Mustafa Riza Khan (1892—1981), 
had been active in the Dar al-Ifta during the teens of the twen- 
tieth century. In the twenties, he was involved in organizational 
activities centered on defense of the Arab holv cities and rebut- 
tal of the Arva Samaj. In addition, he was a scholar in his own 
right and did a great deal to collect and publish his father's 
works. In the 1930s, he started a second school in Bareillv, 
which is still functioning today. 
(") In 1921, Ahmad Riza passed on to both his sons (and a — C\- 

nephew) the responsibilitv for writing fatawa. Responding to a 
question whether India would ever gain its freedom from the 
British, and if so how qadis and muftis would be appointed, he 
told his audience that one day: 

The country will definitely become free of English 
domination. The government of this country will be 
established on a popular basis. But there will be great 
difficulty in appointing a qadi and a mufti on the basis of 
Islamic shari'a law. ... I am today laying the foundation for this 
[process] so that ... no difficulty will be experienced after 
independence. (Rizwi, 198S: 20—21) 

He then proceeded to appoint one of his close followers, 
Amjad 'Ali 'Azami, as the qadi, and two others — Mustafa Riza 
Khan and Burhan ul-Haqq Jabalpuri — as muftis to assist him. 
This qadi would be the qadi for all India, he said. The fact that he 


ch3.044 10/12/2004 5:17 PM Page 8 



believed he was choosing an all-India qadi speaks to the way he 
viewed the Ahl-e Sunnat movement, as part of the worldwide, 
universal umma or communitv of Sunni Muslims. To his mind its 
reach and status were pan-Islamic, not merely local. That these 
arrangements were not in fact realized reflects the realitv on 
the ground, in that the future of the Indian Muslim communitv 
"was largely determined bv people and events far removed from 
Bareillv. The Ahl-e Sunnat movement, though bv no means 
absent during the momentous events of the 1930s and 1940s in 
British India, was but a small part of a larger whole. 

"0" "& 


ch3.044 10/12/2004 5:17 PM Page 




ch4.044 10/12/2004 5:19 PM Page 8y 





i y the 1880s, Ahmad Riza had begun to establish an identity 
'of his own as a mufti who wrote erudite works, including 
dailv responsa (fatawa) in response to questions from Bareillv 
Muslims and others in distant places, and as a sufi surrounded 
bv a close group of disciples. His perspective was markedly 
hierarchical. In the spiritual sphere, what mattered most was 
"closeness" to God, just as in the scholarly one it had been the 
amount of knowledge the person had. Bv both measures, the 
Prophet came first, followed bv the founder of the Qadiri 
order, and finally the sufi master to whom the individual 
believer was linked through discipleship. 

In his personal life, Ahmad Riza took pains to follow the 
sunna (the "wav") of the Prophet down to the smallest detail. It 
was because thev gave primacv to the Prophet in their lives that 
Ahmad Riza and his group of followers referred to themselves 
asthe"Ahl-e Sunnatwa Jama' at," or "devotees of the Prophet's 
practice and the broad communitv." Ahmad Riza's biographer, 
Zafar ud-Din Bihari (who was also part of his inner circle of 
disciples) , gives us the following picture of Ahmad Riza: 

He wouldn't put any book on top of a book of hadith 
[traditions of the Prophet]. . . .When reading or writing, he 
would draw his legs together, keeping his knees up. . . . He 



ch4.044 10/12/2004 5:19 PM Page 8£ 


never stretched his legs out in the direction of the qibla [the 
direction of prayer in Mecca]. He offered all his daily prayers 
in the mosque [not just the Friday noontime prayer, as 
required by law] (Bihari, 1938: 28). 

Elsewhere, Zafar ud-Din relates that when Ahmad Riza 
entered the mosque, he did so with his right foot first, while 
upon leaving it, he did so with his left foot. Even within the 
mosque, he made sure he stepped up to the mihrab (the praver 
niche) with his right foot first (Bihari, 1938: 177). 

As should be clear from this, the dailv lives of the 'ulama 
were governed by strict etiquette (adab) . Like Ahmad Riza, the 
famous 'ulama of Farangi Mahall were always mindful of the 
example of the Prophet. Maulana'Abd ur-Razzaq (1821—89), 
for example, "is portrayed as following the Prophet in 
almost every possible respect. When he drank water, he did so 
in three gulps. When he ate, he did so sparingly. . . . And before 
-(^y— he began he always said 'Bi'sm allah' " (Robinson, 2001: 83). — (^y- 

Veneration of the Prophet also caused many 'ulama to be 
verv respectful of sayyids, descendants of the Prophet: thus, 
Maulana 'Inayat Ullah, one of the Farangi Mahall scholars, 
"revered the Prophet's family, excusing a savyid hundreds of 
rupees rent he owed for the sake of his ancestor. For the same 
reason ... he even went so far as to always use the respectful 'ap' 
rather than the usual 'turn' when he spoke to the sayyids 
amongst his pupils" (Robinson, 2001: 84). Similarlv, when 
Ahmad Riza discovered that a young man hired as household 
help was a savvid, he forbade everyone in the house to ask him 
to do anything, asking that thev take care of his needs instead. 
Uncomfortable with all the attention, after a while the man left 
of his own accord. 

Ahmad Riza was not just a strict Sunni in the sense of imita- 
tor of the Prophet's conduct, however, but also a sufi of the 
Qadiri order. 


ch4.044 10/12/2004 5:19 PM Page 



Ahmad Riza became Shah Al-e Rasul's disciple (murid, lit. 
seeker) in 1 877. He seems to have thought of the relationship 
between master and disciple as unbreakable bv the disciple even 
after the master's death, even though it had not necessarily been 
close in his lifetime. That at least is how he treated the relation- 
ship with his own master, Shah Al-e Rasul, who had died a mere 
two years after it had been formed. As mentioned alreadv, Shah 
Al-e Rasul's grandson, Nuri Mivan, took over as Ahmad Riza's 
spiritual director (though technically they were sufi "brothers" 
or pirbhai, being disciples of the samepir), and Ahmad Riza con- 
tinued to pav his respects to his deceased master by commem- 
orating his death every year at his home in Bareillv. 

The reason the relationship with the pir was so important, 
according to Ahmad Riza, was that the pir had a unique insight 
into his disciple's mental frame of mind, and was alwavs on 
(") hand to guide him: — C\- 

Sayyid Ahmad Sijilmasi was going somewhere. Suddenly his 
eyes lifted from the ground, and he saw a beautiful woman. 
The glance had been inadvertent [and so no blame attached to 
him]. But then he looked up again. This time he saw his pir 
and teacher (murshid) , Sayyid . . . Abd al -Aziz Dabagh. 
(Malfuzat, vol. 2, p. 45) 

On the second occasion the pir had intervened to prevent Savvid 
Ahmad Sijilmasi from looking — intentionally, this time — at a 
"woman outside the circle of relatives with whom social intim- 
acy was permitted, and possibly being led astray. Scrupulous 
Muslims hold the very act of looking at an unrelated woman as 
sinful because it enables impure thoughts to arise. The Muslim 
standard is therefore more stringent than the Christian one. For 
Christians, a sin is committed when the viewer is lustful, but not 
before: "He who looks at a woman to lust after her has alreadv 


ch4.044 10/12/2004 5:19 PM Page 9C 


committed adulterv with her in his heart." (Matthew, 5:28). 
Therefore without a pir's guidance the believer was likely to fall 
into error. Or, as Ahmad Riza put it elsewhere, "To try [to go 
through life without a pir] is to embark on a dark road and be mis- 
led along the wav bv Satan" (Ahmad Riza Khan, 1901: 9—1 1 ) . 

However, such acts of dav-to-dav guidance were but a small 
part of the pir's role in the disciple's life. The most important 
reason whv a person should bind himself to a pir, Ahmad Riza 
explained, was that pirs are intermediaries between the believer 
and God in a chain of mediation that reaches from each pir to the 
one preceding him, all the way to the Prophet and thence to 
God. Hadith (prophetic traditions) proved, he said, that 

there was a chain of intercession to God beginning with the 
Prophet interceding with God Himself. At the next level, the 
sufi masters (masha'ikh) would intercede with the Prophet on 
behalf of their followers in all situations and circumstances, 
including the grave (qabr). It would be foolish in the extreme, 
\y therefore, not to bind oneself to a pir and thus ensure help in vJ 

times of need (Ahmad Riza Khan, 1901: 12). 


The pir, in turn, should conform to four exacting standards: 
he should be a Sunni Muslim of sound faith (sahih 'aqida), 
should be a scholar ('alim) qualified to interpret the shari'a, his 
chain of transmission (silsila) should reach back from him in an 
unbroken line to the Prophet, and finallv, he should lead an 
exemplarv personal life and not be guilty of transgressing the 
shari'a (Maljuzat, vol. 2, p. 41). 

If both master and disciple conformed to these high stand- 
ards, the disciple would eventually attain a state of complete 
absorption in his pir, a condition known as fanaji'l shaikh. Nuri 
Mivan was cast as a perfect illustration of the model of/ana: 


ch4.044 10/12/2004 5:19 PM Page 91 


[Nuri Miyan] loved and respected his [pir, Shah Ale Rasul]; 
indeed, he loved everyone who was associated with him, and 
all the members of his family He followed his commands, he 
presented himself before him at his court (darbar) , he sought 
his company, he was completely absorbed in him. His face had 
the same radiance [as Shah Ale Rasul], his personality had the 
same stamp (hat), he walked with the same gait, when he 
talked it was in the same tone. His clothes had the same 
appearance, he dealt with others in the same way. In his 
devotions and strivings, he followed the same path (maslak). 
The times set apart for rest in the afternoon and sleep at night 
were times when he went to him particularly, receiving from 
him guidance in every matter and warning of every danger. 
(Ghulam Shabbar Qadiri, 1968: 91) 




Belief in the intercession of saintly persons with Allah on behalf 

of the ordinary believer is controversial in Sunni Islam. Indeed, 
Muslim reformers have often spoken out against it on the 
grounds that it is a form of: shirk or associationism and an accre- 
tion to "pure" Islam . Year s before , Muhammad Isma 'il had writ- 
ten against this very belief (and the practices that arise from it) 
in his book Taqwiyat al-Iman, classifying it as the second of 
three types of: shirk (see p. 32). Ahmad Riza, for his part, wrote 
extensively in favor of such belief, declaring that Muhammad 
Isma'il's position was contrary to the Qur'an, which gives the 
prophets the power to intercede with God's "permission" (izn), 
and that it detracted from the Prophet's power, which included 
the ability to perform miracles. 

For Ahmad Riza and the Ahl-e Sunnat movement, which saw 
sufism as a necessary complement to the law, the intercession 
of sufi masters and, ultimately, of the Prophet himself was 


ch4.044 10/12/2004 5:19 PM Page 92 


crucial to the relationship between master and disciple, for the 
living hope that the dead pir (here the ordinary dead are less 
central than the holv, exalted dead) will intercede for them 
both in the here and now and when they face Judgment Dav. 
But the living can do something for the dead too : the pravers of 
the living can increase the dead person's chances of a favorable 
judgment on Judgment Dav through the concept of the trans- 
fer of merit (isal-e sawab). Haji Imdad Ullah "Muhajir" Makki 
(1817—99), one of the most famous sufis of the nineteenth cen- 
tury — who belonged to the Chishti order and was respected bv 
'ulama from a number of rival movements, including the Ahl-e 
Sunnat — wrote in his book Faisla-e Haft Mas'ala (Solution to 
Seven Problems) that the pravers of the living could help the 
dead person answer the questions of the two angels Munkar 
and Nakir correctly when thev visited the dead in the grave and 
therebv ensure his or her ultimate entry into heaven. 

The spiritual power or grace (baraka, barkat) of the pir is 
\7" believed to be especially strong at his tomb, and indeed to grow —kzj~ 

over time. As Ewing writes: 

[When a saint dies] his spirit is so powerful and so dominant 
over the body that the body itself does not die or decay but is 
merely hidden from the living. The baraka of the saint is not 
dissipated at the saint's death. It is both transmitted to his 
successors and remains at his tomb, which becomes a place of 
pilgrimage for later followers. The pir does not actually die in 
the ordinary sense of the term. He is "hidden," and overtime 
he continues to develop spiritually, so that his baraka increases, 
as does the importance of his shrine. (Ewing, 1 980: 29) 


As Ewing points out, a pir's followers fall into three distinct 
groups which can be visualized as a series of concentric circles. 


ch4.044 10/12/2004 5:19 PM Page 93 


In the first, outermost circle are the large number of people 
who come to the pir with evervdav problems to be solved, such 
as curing an illness, ensuring the birth of a son, or answering a 
request for an amulet to be worn for good luck. Ahmad Riza 
would pass on all such people to his students, unless their prob- 
lem had to do with sufism (Bihari, 1938: 68). 

Within this outer circle was a smaller "inner circle" of fol- 
lowers in whose training he took great interest. All were 
known as khalifas (deputies) .They were divided into "ordinary" 
( 'amm), the second group, and special (khass), the third group, 
also the smallest. Some of Ahmad Riza's ordinarv khalifas went 
on, in the 1920s, to become prominent leaders of the Ahl-e 
Sunnat movement during the Khilafat and Indian nationalist 
movements. He looked upon them as lieutenants or right-hand 
men who could be counted upon to debate with an opponent, 
run a newspaper or school (madrasa), and generally promote 
the goals of the movement in their hometowns, but did not 
\7" regard them as spiritual disciples. This relationship, Ahmad —kzj~ 

Riza said, ceased upon the death of the teacher. His relationship 
with the khalifa-e khass, on the other hand, was of primarily 
religious significance and was continuous, not ceasing with the 
death of the teacher. Those in this small group experienced 
Jana of the pir and saw themselves as tied to their master even 
after he had died, as described above. Out of this select group 
the pir would choose one as his successor (sajjada nishin). 
Ahmad Riza chose his eldest son, Hamid Riza Khan — authoriz- 
ing him, in November 1915, to continue the chain of sufi dis- 
cipleship (silsila) named the silsila Rizwivva (from the "Riza" in 
his name). The sajjada nishin also bore worldlv responsibilities 
for the maintenance of properties and management of funds 
(Ahmad Riza Khan, 1901 : 14). This ensured the continuity of 
the sufi master's spiritual and worldlv network over time. 


ch4.044 10/12/2004 5:19 PM Page 94 



Ahmad Riza was affiliated to the Qadiri order (tariqa), one of the 
three major sufi orders in nineteenth-century India (along with 
the Chishti and Naqshbandi) . The Qadiri order was founded in 
the twelfth century by Shaikh ' Abd al-Qadir, a native of the town 
of Jilan in Iran, who later became a scholar and preacher in 
Baghdad. His tomb in Baghdad is visited bv pilgrims from all 
over the Muslim world, particularly from South Asia. To his fol- 
lowers, he is a saint, an intercessor with God, and the occupant 
of a place of honor in the hierarchy of saints "between this world 
and the next, between the Creator and the created" (Padwick, 
1996: 240). One of his most popular epithets is "Ghaus-e 
A'zam," the "Greatest Helper." Qadiris regard him as the Qutb, 
axis or pole of the invisible hierarchv of saints who rule the spir- 
itual universe. This spiritual "government" is as follows: 

Every ghaus has two ministers. The ghaus is known as Abd 
Allah. The minister on the right is called 'Abd al-Rab, and the 
one on the left is called 'Abd al-Malik. In this [spiritual] 
world, the minister on the left is superior to the one on the 
right, unlike in the worldly sultanate. The reason is that this is 
the sultanate of the heart and the heart is on the left side. 
Every ghaus [has a special relationship with] the Prophet. 
(Malfuzat, vol. l,p. 102) 

The first ghaus, Ahmad Riza said, was the Prophet. He was fol- 
lowed by the first four caliphs (Abu Bakr, 'Umar, 'Uthman, and 
'Ali), each of whom was first a minister of the left before he 
became ghaus upon the death of the previous incumbent. They 
were followed bv Hasan and Husain ('Ali's sons, the second and 
third imams, respectively, in Shi'ism) . The line continued down 
to 'Abd al-Qadir Jilani. He was last "great" ghaus (ghausiyat-e 
kubra) . All who followed after him were deputies (na'ib) . In this 


ch4.044 10/12/2004 5:19 PM Page 9 



chain of spiritual authority, the sources of spiritual knowledge 
are united with those of shari'a knowledge — for the source of 
the latter is none other than the Prophet, followed bv the first 
four caliphs of Sunni Islam. This is a fitting image for one who, 
like Ahmad Riza Khan, saw himself as embodying the path of 
both shari'a and sufism (tariqa). 

Shaikh 'Abd al-Qadir Jilani was also a relative of the 
Prophet, being descended on his mother's side from Husain 
('Ali's vounger son bv the Prophet's daughter Fatima) and on 
his father's from Hasan ('Ali's older son bv Fatima). This is the 
source of the epithet "Hasan al-Husain."This double genea- 
logical connection mattered greatlv to Shaikh 'Abd al-Qadir 's 
followers, for thev believed him to have inherited the spiritual 
achievements of all his ancestors. 

Ahmad Riza's views on Shaikh 'Abd al-Qadir are expressed 
in several poems, some of which relate to his exalted status: 

Except for divinity and prophethood 
you encompass all perfections, O Ghaus. 

Who is to know what your head looks like 
as the eye level of other saints corresponds to the sole of 
vour foot? 


You are mufti or the shar', qazi of the community 
and expert in the secrets of knowledge, 'Abd al-Qadir. 

Or again: 

Prophetic shower, 'Alawi season, pure garden 
Beautiful flower, your fragrance is lovely. 

Prophetic shade, 'Alawi constellation, pure station 

Beautiful moon, vour radiance is lovelv. 

' j j 

Prophetic sun, 'Alawi mountain, pure quarry 
Beautiful ruby, your brilliance is lovely. 
(Ahmad Riza Khan, 1976: 234) 


ch4.044 10/12/2004 5:19 PM Page 9( 


All the adjectives refer to specific persons, namely, Muhammad 
("prophetic"), his cousin and son-in-law 'Ali ('"Alawi"), his 
daughter Fatima ("pure"), and his grandsons Hasan ("beauti- 
ful," the literal meaning of hasan) and Husain ("lovelv," husain 
being the diminutive of hasan). These five figures, popularlv 
svmbolized by the human hand in Shi'ism (the panj), are parti cu- 
larlv holy to Shi'i Muslims. 

This emphasis on Shi'i figures of authority in the poetry of a 
religious leader who prided himself on his Sunni identity mav 
seem odd to readers familiar with the Sunni— Shi'i divide in 
Muslim historv. Ahmad Riza's own writings are on manv occa- 
sions fiercelv anti- Shi'i in tone. Nevertheless, the sufi chains of 
authority in all the South Asian orders — the Chishti and 
Naqshbandi as well, though the Qadiri more emphatically so — 
bring the two sides together bv their emphasis on genealogv. 


Muhammad became an object of devotion early in Islamic his- 
tory, perhaps as early as the eighth century, within a hundred 
vears of the birth of Islam. It displaved itself, among other things, 
in the birth of the concept of the Prophet's light (nur-e muham- 
madi), the idea that Muhammad was created out of God's light 
and that his creation preceded that of Adam and the world in gen- 
eral. In the tenth century, the famous Baghdadi mystic al-Hallaj 
(d. 922) wrote that the Prophet was the "cause and goal of cre- 
ation." He supported his assertion bv quoting the hadith qudsi (a 
hadith in which the Prophet reports a statement by God but 
which does not form part of the Qur'an), that "If you had not 
been, I would not have created the heavens." The idea of the 
prophetic light (on which, see Schimmel, 197S: 215—16; 
Schimmel, 1987) has been developed in both Sunni and Shi'i 
mvsticism, though with an important difference. In Sunni Islam, 


ch4.044 10/12/2004 5:19 PM Page 9y 


the prophetic light belonged to the Prophet alone, whereas in 
Shi'ism it was inherited and carried forward by each of the twelve 
Imams. Among Sunni mvstics, it eventually came to be connected 
with the concept of "annihilation in the Prophet" (Janaji '1 rasul ) . 
"The mystic no longer goes straight on his Path toward God: first 
he has to experience annihilation in the spiritual guide, who func- 
tions as the representative of the Prophet, then the ... 'annihila- 
tion in the Prophet,' before he can hope to reach, if he ever does, 
Jana Jl Allah [annihilation in Allah]" (Schimmel, 197S: 216). 
Somewhat later, in the thirteenth century, the Spanish mvstic 
Ibn al-'Arabi (d. 1240) developed the concept of Muhammad as 
the Perfect Man (insan kamil) , "through whom His consciousness 
is manifested to Himself. . . . [T]he created spirit of Muhammad is 
... the medium through which ...the uncreated divine spirit 
[expresses itself and] through which God becomes conscious of 
Himself in creation" (Schimmel, 197S: 224). 

Of the relationship between God and the Prophet, Ahmad 
\7" Riza said: —kzj~ 

Only the Prophet can reach God without intermediaries. This 
is why, on the Day of Resurrection, all the prophets, saints 
(auliya), and 'ulama will gather in the Prophet's presence and 
beg him to intercede for them with God. . . . The Prophet 
cannot have an intermediary because he is perfect (kamil). 
Perfection depends on existence (wujud) and the existence of 
the world depends on the existence of the Prophet [which in 
turn is dependent on the existence of God]. In short, faith in 
the preeminence of the Prophet leads one to believe that only 
God has existence, everything else is his shadow. (Ahmad Riza 
Khan, Malfuzat, vol. 2, p. 58) 

To those who argued that belief in the perfection of the Prophet 
was contrary to belief in the Oneness of God (tawhid) , Ahmad 
Riza replied that "everything comes from God," that only God 
is intrinsic (zat) while everything else is extrinsic or depend- 
ent. This said, however, God chose Muhammad as "His means 


ch4.044 10/12/2004 5:19 PM Page 9£ 


of bringing the extrinsic ( ghair) world to Him. . . . Muhammad 
distributes what He gives. What is in the one is in the other." 
And on Muhammad as God's light, he said: 

God made Muhammad from His light before He made 
anything else. Everything begins with the Prophet, even 
existence (wujud). He was the first prophet, as God made him 
before He made anything else, and he was the last as well, 
being the final prophet. Being the first light, the sun and all 
light originates from the Prophet. All the atoms, stones, 
trees, and birds recognized Muhammad as prophet, as did 
Gabriel, and the other prophets. (Bihari, 1938: 96—98) 

Being made of light, the Prophet Muhammad had no shadow. 
Ahmad Riza wrote in a fatwa, "Undoubtedlv the Prophet did 
not have a shadow. This is clear from hadith, from the words of 
the 'ulama, of the [founders of the four Sunni law schools], and 
the learned" (Ahmad Riza Khan, 1405/ 1985:51-52). He cited 
vy numerous hadith to prove the luminous qualitv of the Prophet's T7 

face and bodv, to show that flies did not settle on his body, that 
after he had ridden on the back of an animal, the animal did not 
age any further, and so on. Such miracles associated with the 
Prophet also have a long historv in popular literature through- 
out the Muslim world. 

Ahmad Riza wrote a number of eloquent verses about the 
Prophet. One, entitled Karoron Durud (Millions of Blessings), is 
well known in Pakistan today, and is recited on the Prophet's 

I am tired, you are my sanctuary 
I am bound, you are my refuge 
My future is in your hands. 
Upon you be millions of blessings. 

My sins are limitless, 

but you are forgiving and merciful 


ch4.044 10/12/2004 5:19 PM Page 9? 


Forgive me my faults and offenses, 
Upon you be millions of blessings. 

I will call vou "Lord," for vou are the beloved of the Lord 
There is no "yours" and "mine" between the beloved and the 

And like poets all over the Muslim world, Ahmad Riza also cele- 
brated the Prophet's Night Journev to Jerusalem in his poetrv: 

You went as a bridegroom of light 
on your head a chaplet of light 
wedding clothes of light on your body. 
(Ahmad Riza Khan, 1976: 9, 13) 

The Prophet was a personal presence in Ahmad Riza's life. 
When he went on his second pilgrimage in 1 90S— 6, he spent a 
month in Medina, where the Prophet is buried. Ahmad Riza 
was in Medina during the Prophet's birthday celebrations. 
According to his own statement, he spent almost the entire 
period at the Prophet's tomb ; he even met the 'ulama of Medina 
there. He considered this the holiest place on earth, even sur- 
passing the Ka'ba, as he wrote in the following verse: 

O Pilgrims! Come to the tomb of the king of kings 
You have seen the Ka'ba, now see the Ka'ba of the Ka'ba 
(Ahmad Riza Khan, 1976: 96; Maljuzat, vol. 2, p. 47-48) 

Ahmad Riza believed that the Prophet could help whoever he 
wished, in whatever "way he saw fit, from his tomb. (He also had 
the capacity to travel in spirit to other places.) While most Sunni 
'ulama believe that the Prophet will intercede with God on 
Judgment Dav for ordinarv Muslims, Ahmad Riza believed that 
the Prophet's intercession is ongoing from the grave. (The 
Prophet lives a life of sense and feeling while in his grave and 
spends his time in devotional praver.) He mediates with God 
everv dav; his abilitv to do so is not limited to Judgment Day. 
Ahmad Riza had undertaken this second hajj particularly in the 


ch4.044 10/12/2004 5:19 PM Page 1C 


hope of being blessed with a vision of the Prophet. And according 
to Bihari, this did indeed occur after he had presented the 
Prophet with a poem (ghazal) he had composed to him. In 
Bihari's words, "His fortune (qismat) awoke [on the second night 
of waiting]. His watchful, vigilant eves were blessed with the 
presence of the Prophet" (Bihari, 1938: 43—44). He also reported 
having seen the Prophet in a dream (Maljuzat, vol. 1 , p. 82—83) . 

Ahmad Riza also expressed his love of the Prophet in small, 
everyday acts. For instance, in all correspondence, fatawa, and 
other writings he signed himself as 'Abd al-Mustafa, meaning 
"Servant of the Chosen One," the latter being an epithet of the 
Prophet.Andonone occasion he told a follower that if his heart 
were to be broken into two pieces, one would be found to say, 
"There is no God but Allah," and the other would say, "And 
Muhammad is His Prophet" (Maljuzat, vol. 3, p. 67). Together, 
the two phrases constitute the profession of faith for a Muslim. 


In addition to daily acts of devotion to the sufi pir, Shaikh 'Abd 
al-Qadir Jilani, and the Prophet, special rituals marked their 
birth or deathdavs. It was a time when the community came 
together, affirming not only their shared beliefs but also their 
group identity. Some of the rituals were particular to them, not 
being favored bv the other groups. 

The ritual celebration of a pir's deathdav ( 'urs) "was frowned 
upon bv 'ulama such as the Ahl-e Hadith whom Ahmad Riza 
called "Wahhabi." Others, such as the Deobandis, held that it was 
in order as long as the celebrations did not involve anv forbidden 
activities such as singing, dancing, and the use of intoxicants. 
Ahmad Riza would mark the occasion by recitation of the entire 
Qur'an (khatma), poetry in praise of the Prophet (na't), and ser- 
mons bv the 'ulama. He himself would deliver a sermon at the 


ch4.044 10/12/2004 5:19 PM Page 1C 


mosque, speaking not only about Shah Al-e Rasul but also about 
Shaikh 'Abd al-Qadir, the founder of the Qadiri order to which 
he belonged, and the Prophet. The event would be reported in 
Rampur's Urdu newspaper, the Dabdaba-e Sikandari. 

It lasted anvwhere between four and six days. In 1 9 1 2 , a year 
in which the Dabdaba-e Sikandari reported on an 'urs celebrating 
Nuri Mivan on his death anniversary, it lasted five days and was 
attended by four to five thousand people, some from distant 
parts of the countrv (this was a much smaller turnout than the 
usual twentv thousand, on account of confusion as to the dates 
of the event). Apart from the Qur'an readings and recitation of 
poetrv in praise of the Prophet, Nuri Mivan's 'urs featured the 
viewing of prized relics (tabarrukat) such as ahair of the Prophet 
or 'Ali's robe, which had come into the famih/'s possession. 
These objects were also viewed fortv days after the pir's death, 
when his successor (sajjada nishin) was formallv installed in a 
ceremonv known as the dastar-bandi ("tving of the turban"). The 
\7" symbolism of this and other rituals, it is fascinating to note, —kzj~ 

bears close similarities with ceremonies associated with royalty. 

Ahmad Riza's veneration for Shaikh 'Abd al-Qadir was ritu- 
ally expressed through the eating of consecrated food and the 
drinking of consecrated water on the eleventh of everv month 
(gyarahwin) in memory of his birthdate.This was done to the 
accompaniment of certain pravers (durud ghausia) and the 
recitation of the Qur'an while facing Baghdad (Bihari, 1938: 
202—203). As with the celebration of the 'urs in memorv of 
one's pir, the observance of gyarahwin was frowned upon bv 
some 'ulama, including those of Deoband. 

The Prophet's birth anniversary was the occasion for a big 
joyous celebration every year (majlis-e milad or milad al-nabi) . It 
was one of the few annual occasions when Ahmad Riza gave a 
sermon at the mosque in Bareillv, addressing a large gathering 
that overflowed the mosque's seating capacity (Bihari, 1938: 
96—98). Like the other ritual occasions mentioned above — the 


ch4.044 10/12/2004 5:19 PM Page 1C 


'urs for pirs or sufi masters and the gyarahwin for Shaikh 'Abel 
al- Qadir Jilani — some ' ulama obj ected to the milad celebrations 
on the grounds that it could lead to worship of the Prophet, and 
hence shirk or association of partners with Allah. As Metcalf 
reports, the 'ulama of Deoband tried to "avoid fixed holidays 
like the maulud [milad] of the Prophet, the 'urs of the saints," 
and other feasts (Metcalf, 1982: 151). The Ahl-e Hadith were 
even more disapproving than the Deobandi 'ulama. Not onlv 
did thev prohibit the 'urs and gyarahwin, but they even "prohib- 
ited all pilgrimage, even that to the grave of the Prophet at 
Medina. ... In their emphasis on sweeping reform, thev under- 
stood sufism itself, not just its excesses, to be a danger to true 
religion" (Metcalf, 1982: 273-274). 

However, not all were as willing to condemn such ritual cele- 
brations. Haji Imdad Ullah, mentioned earlier, had addressed 
the issue in his book Faisla-e Haft Mas'ala. In his view, the per- 
missibility of the event depended on the intention of the par- 
\7" ticipants. If a person equated the ritual with ibadat or worship, —kzj~ 

on a par with obligations such as ritual prayer (namaz) or the 
fast during Ramadan, then it was reprehensible. However, if it 
was seen as a means of honoring and respecting the Prophet, it 
was acceptable (Faisla-e Haft Mas'ala, 50—76). Another contro- 
versial issue had to do with the ceremony known as aiyam or 
"standing up" during the milad. This was a point at which the 
Prophet's birth was recalled during the sermon. Ahmad Riza 
justified the act of standing up as a mark of respect for the 
Prophet, and also quoted a scholar from Arabia who said that 
the Prophet's spirit was present in the room at that time. 


Ahmad Riza's relations with the other Indian reform move- 
ments are best understood with reference to his 1906 fatwa, 


ch4.044 10/12/2004 5:19 PM Page 10c- 


Husam al-Haramain, written while in Mecca. In his Maljuzat, 
Ahmad Riza explains that when he arrived in Mecca he found 
that the judgment of unbelief was about to be passed on an 
Indian scholar for having supported the argument that the 
Prophet had knowledge of the unseen. He suggests that had it 
not been for the presence of a Deobandi scholar, Maulana 
Khalil Ahmad Ambethwi (a disciple of Maulana Rashid Ahmad 
Gangohi) in Mecca, this judgment would not have been arrived 
at. He therefore hastened to write a fatwa of his own to avert 
the expected pronouncement of kufr (or takfir). As Ahmad Riza 
said: "the Wahhabis had arrived before [me] , among them Khalil 
Ahmad Ambethwi. ...Thev had obtained access to the minis- 
ters of the kingdom, right up to the Sharif. And thev had raised 
the issue of the [Prophet's] knowledge of the unseen" (Maljuzat, 
vol. 2, p. 8). 

Ahmad Riza was anxious to present his arguments to the 
highest authorities in the Sunni Muslim world while he was 
\7" there, for confirmation of these arguments bv the Meccan —kzj~ 

'ulama would bolster his standing at home while undermining 
that of his opponents. The fatwa begins by describing the sorrv 
state of Sunni Islam in India at the time : 

The school (madhab) of the Ahl-e Sunnat is a stranger in India. 
The darkness of dissension (Jltna) and trial is fearful; wicked- 
ness is in the ascendancy; mischief has triumphed. ... It is 
incumbent on [you] to help the religion and humiliate the 
miscreants, if not by the sword, then at least by the pen. 
(Husam: 9-10) 

Later in the fatwa he makes the same point by citing a hadith in 
which Abu Bakr, the first Sunni caliph, is said to have heard the 
Prophet say that a time will come when things are so bad that a 
person who was a Muslim in the morning will be a kajlr in the 
evening, and vice versa. This is how bad things are in India, he 
tells the Meccan 'ulama. 


ch4.044 10/12/2004 5:19 PM Page 1C 


What do vou think of mv judgment, he asks them in urgent 

Tell me clearly whether you think these leaders . . . are as I 
have portrayed them in my commentary, and if so, whether 
the judgment [of unbelief ] that I have passed on them is 
appropriate, or whether, on the contrary, it is not permissible 
to call them kafirs — even though they deny the fundamentals 
of the faith (zaruriyat-e din), . . . areWahhabis, and . . . insult 
Allah and the Prophet. (Husam: 10) 

We must pause to consider two terms used in this passage. 
First, what is meant bv "fundamentals of the faith," and second, 
what exactly did Ahmad Riza mean when he called the people 
he accused of unbelief "Wahhabis"? 

The first term is easily explained, as it was the subject of pre- 
vious fatawa bv Ahmad Riza which had dealt primarily with the 
Nadwa. As he explained there, the fundamentals (or essentials) 
Qj included beliefs based on clear verses (nusus) of the Qur' an (as — (^y- 

against verses open to a variety of interpretations), on accepted 
and "widely known hadith, and on the consensus (ijma) of the 
Muslim community. Such beliefs include: the unitv of Allah, 
the prophethood of Muhammad, heaven and hell, the delights 
and punishments of the grave, the questioning of the dead, the 
reckoning on the dav of judgment, belief in the prophets, in the 
corporeal existence of the angels, including the Angel Gabriel 
through whom Muhammad received the revelations contained 
in the written Qur' an, in the jinn and Satan, and the occurrence 
of miracles. All these beliefs were "articles of faith" or aqida, and 
had to be accepted. As Friedmann comments with regard to 
Mirza Ghulam Ahmad, "Faith is . . . indivisible : even the rejec- 
tion of one essential article places the person bevond the pale" 
(Friedmann, 1989: 160). It is ironic that when this was 
applied bv other Indian 'ulama to Mirza Ghulam Ahmad, he 
"was judged an unbeliever. He is the first person so judged in 


ch4.044 10/12/2004 5:19 PM Page 10€- 


Husam al-Haramain. Unlike the other people named in the 
fatwa, however, he was not described as a"Wahhabi." 

The term "Wahhabi" has been encountered in previous chap- 
ters with reference to Indian 'ulama such as Muhammad Isma'il 
and Savvid Ahmad Barelwi (leaders of the Tariqa-e 
Muhammadivva in the 1 820s), for instance, or the Ahl-e Hadith 
and Savvid Ahmad Khan (the founder of MAO College in 
Aligarh). Ahmad Riza was not specificallv suggesting that the 
'ulama he called Wahhabi had anv direct link with the nineteenth- 
centurv Wahhabi movement in Arabia, though he did think that it 
had influenced these Indian 'ulama. He used it as a general term 
of abuse for anyone he deemed to be disrespectful of the Prophet. 

In the rest of the fatwa, Ahmad Riza proceeded to name four 
groups of Indian 'ulama and explain whv he considered the 
leader of each group to be an unbeliever. The first, as noted 
above, was Mirza Ghulam Ahmad, whose followers Ahmad Riza 
calls the Ghulamivva (rather than Ahmadivva, as thev called 
\7" themselves) , in a plav on words — the literal meaning of the word —kzj~ 

ghulam is "slave "though here it is probablv better understood as 
"knave," as Ahmad Riza accused him of making a number of mis- 
leading claims about himself (claims we examined in chapter 2) . 
Reversing Ghulam Ahmad's claim that he was "like the Messiah" 
(Jesus Christ), Ahmad Riza denigrated him as the Antichrist 
(dajjal), inspired bv Satan. However, it was Ghulam Ahmad's 
statement that he was a"shadowy" prophet that incensed Ahmad 
Riza the most. His unbelief was said to be greater than that of 
any of the other scholars named in the fatwa. 

Ahmad Riza's second group consisted of "Wahhabis" who 
believed that this world was only one out of seven, and that 
there were prophets like Muhammad in the other six worlds as 
well, making seven in all. He referred to this group by the 
home-made term Wahhabiyya Amthaliyya , "likeness Wahhabis." 
According to him, most of them held that the likenesses of 
Muhammad were the last prophets in their respective worlds, 


ch4.044 10/12/2004 5:19 PM Page 1C 


as Muhammad 'was in this one, but there 'were also some "who 
denied it: in the other six worlds the "seal of the prophets" 
(Arabic, khatim al-anbiya') would be someone else. Ahmad Riza 
called these people, whom he found particularly offensive, 
"seal Wahhabis." Ahmad Riza appears to have been referring to 
debates about God's unlimited power which had been ongoing 
since the early nineteenth century. In Taqwiyat al-Iman 
(Strengthening the Faith) , Muhammad Isma'il had written: 

In a twinkling, solely by pronouncing the word "Be!" [God] 
can, if he like[s], create tens of millions of apostles, saints, 
genii, and angels, of similar ranks with Gabriel and 
Muhammad, or can produce a total subversion of the whole 
universe, and supply its place with new creations. 
(Mir Shahamat 'Ali trans. , 339) 

Since that time, the Indian reformist 'ulama had been debating 
among themselves whether this meant that there could hypo- 
\j- thetically be other final prophets in the six other worlds they —kzj~ 

believed to exist apart from the one we know. 

All three of the 'ulama Ahmad Riza described as leaders of 
the "likeness" or "seal" Wahhabis were from Deoband. One 
'alim was quoted as saving that the discerning among the 
'ulama know that prophetic superiority is unrelated to being 
either first or last in time. Ahmad Riza declared that they were 
unbelievers because they had implicitly denied the finality of 
the Prophet Muhammad, "which of course "was a "fundamental" 
belief on which all Muslims agreed. 

The third group (whom Ahmad Riza called the "Wahhabivva 
Kadhdhabivva," "the lie Wahhabis,") also from Deoband, "were 
said to believe that God can lie should He wish to. The leader of 
these 'ulama was said to be Maulana Rashid Ahmad Gangohi, 
the Deobandi'alim whose fatwa on pilgrimage we examined in 
chapter 3. Bv saving that God can lie, Ahmad Riza said that 
Rashid Ahmad was casting doubt on the verv profession of 


ch4.044 10/12/2004 5:19 PM Page 1C 


faith, the shahada or kalima. The first part of the profession savs, 
"There is no God but God," and belief in it is, once again, neces- 
sary if one is to be considered a Muslim. Once again, Ahmad 
Riza's discussion ignored the hypothetical nature of Gangohi's 
statement, which was also about God's absolute power. 

He called the last group the "Wahhabiyva Shavtaniyva", "the 
Satanic Wahhabis." Allegedly led bv Rashid Ahmad Gangohi, 
like the third group, thev were said to believe that Satan's 
knowledge exceeded that of the Prophet and that the Prophet's 
knowledge of the unseen was onlv partial. Rashid Ahmad was 
said to have cited a controversial hadith to the effect that the 
Prophet Muhammad did not even know what lay on the other 
side of a wall, claiming that highlv respected authorities also 
accepted it, which Ahmad Riza doubted. In support of his own 
argument Ahmad Riza cited a Qur'an verse: 

He is the knower of the Unknown, 

and He does not divulge His secret to any one 

Other than an apostle He has chosen. 

(72: 26-27, Ahmed 'Ali trans.) 

The suggestion that the Prophet Muhammad's superiority to 
preceding prophets since the beginning of time was even hvpo- 
theticallv denied, or that the finality of his prophethood was 
being denied, or that his knowledge of the unseen was not 
acknowledged led Ahmad Riza to declare that the 'ulama con- 
cerned were kajlrs and apostates (murtadd) from Islam. 


This accusation was not lightly made. In earlier fatawa on 
Muhammad Isma'il and his statements in Taqwiyat al-Imam, for 
example, Ahmad Riza had cited seventv different grounds for 
declaring Muhammad Isma'il to be an unbeliever, but had not 


ch4.044 10/12/2004 5:19 PM Page 1C 


in fact done so. He had believed it prudent to "restrain the 
tongue" (kaff-e lisan) and had given Muhammad Isma'il the 
benefit of the doubt, as he believed one should. In 1896, he had 
written a fatwa in which he characterized a number of contem- 
porary Muslim movements —from Savvid Ahmad Khan's mod- 
ernist Aligarh movement, to the Ahl-e Hadith, Deoband, and 
the Nadwa, not to mention the Shi 'a — as having "wrong" or 
"bad" beliefs (bad-mazhab) and being "lost" (gumrah). These 
people were misleading ordinarv Muslims, he said. In 1900, he 
had sent this fatwa (most of which was against the Nadwa) 
to certain Meccan 'ulama, asking them to confirm his opinions 
(sixteen Meccan 'ulama had signed their assent to this fatwa). 
But "with the exception of the Aligarh modernists (whom he 
described as "kajirs and murtadds," he had stopped far short of 
calling the other groups unbelievers, even though thev had, in 
his view, denied the "essentials" of the faith (zaruriyat-e din) . 
Much had changed bv 1906,apparentlv. In 1900 a number of 
\7" his followers had declared him to be the Renewer (mujaddi d) of ~x3~ 

the fourteenth Islamic century. Not surprisingly, the claim was 
not accepted bv rival movements who elevated their own 
'ulama to the title. Perhaps this helps explain whv it was that 
when Ahmad Riza went on pilgrimage in 1905—6, he was pre- 
pared to write a fatwa against a small group of Deobandi 
'ulama, as well as Mirza Ghulam Ahmad, naming them all as 

For the Ahl-e Sunnat, this effort was crowned with success 
when twenty 'ulama from Mecca and thirteen from Medina 
certified Husam al-Haramain, giving it their support. They 
belonged to three different law schools, namely, the Hanafi, 
Shafi'i, and Maliki. One of them (whose title was Shaikh 
al-' Ulama) appears to have been a scholar of great standing in 
Mecca. Khalil Ahmad Ambethwi, the Deobandi scholar who 
had preceded Ahmad Riza to Mecca and had been trying to get 
a fatwa declaring an Indian scholar to be an unbeliever because 


ch4.044 10/12/2004 5:19 PM Page 1C 


of his belief in the Prophet's knowledge of the unseen, had to 
leave Mecca two weeks after his arrival because, Metcalf savs, 
some people "objected to his visit." Back in India, the 
Deobandis got busy writing fatawa of their own responding to 
Ahmad Riza "point bv point," leading to what Metcalf calls a 
Jatwa war" (Metcalf, 1982:310). 


With regard to Muslims' relations with Hindus, Ahmad Riza's 
assessment was that the interests of Hindus and Muslims were 
intrinsicallv opposed. He argued that the Muslim leaders of the 
Khilafat (and Noncooperation) movements had lost their sense 
of balance, as thev wanted to cut off relations with one set of 
+ unbelievers, the British, while seeking close relations with 

another, the Hindus. In religious terms, this was tantamount to 
"pronouncing that which was indifferent (mubah; neither good 
nor bad) to be forbidden (haram) , and that which was forbidden 
to be an absolute duty (Jarz qati ' ) ." The Christians were at least 
people of the book, whereas the Hindus were pagans. 

In a 1920 fatwa about the Noncooperation movement (one 
of his last), he argued that even in political terms it made no 
sense for the Muslims to throw in their lot with the Hindus, for 
whereas the British had refrained from interfering in Muslims' 
internal (and religious) affairs, the Hindus had done the very 
opposite. Here he cited the incidence of recent Hindu— Muslim 
riots in the United Provinces, and Hindu refusal to allow the 
sacrifice of cows during the 'Id festivities. Criticizing the 
Muslim leadership bitterlv, he wrote: 

What religion is this that goes from its [previously] 
incomplete subservience to the Christians to completely 


ch4.044 10/12/2004 5:19 PM Page 1] 


shunning them, and immerses itself wholly in following the 
polytheists (mushrikin)? They [the Muslims] are running 
from the rain only to enter a drainpipe. (Ahmad Riza Khan, 

In the same fatwa, Ahmad Riza went on to argue that social rela- 
tions (mu 'amalat) "with the British were permissible according 
to the shari'a as long as unbelief or disobedience to the shari'a 
were not promoted thereby. But the leaders of the Khilafat and 
Noncooperation movements had prohibited such relations, 
while simultaneously advocating intimacy with Hindus. If all 
relations with the British were to be cut off, he argued, then 
whv did the Muslim leaders continue to use the railways, the 
telegraph, and the postal system, all of which benefited the 
British Indian government's revenues? 

Ahmad Riza was not alone among the 'ulama to make such 
arguments on the basis of his interpretation of the sources. 
According to I. H. Qureshi (1974: 270—271), some 'ulama of 
Deoband were also opposed to the Noncooperation move- 
ment, but such was the atmosphere in the countrv that their 
voices were not heeded. 


ch5.044 10/12/2004 5:22 PM Page 1 






TheAhl-e Sunnat (or"Barelwi") movement began to take shape 
in the 1 880s, and came into its own in the 1 890s in the context 
of its anti-Nadwa campaign. Thereafter it grew steadily in 
different parts of the country, as Ahmad Riza's followers them- 
selves began schools, published journals, held oral disputa- 
tions, and organized around specific issues in different parts of 
the country. Being built around scholarly interpretation of the 
Qur ' an and prophetic sunna together with sufi practice and rit- 
ual, its participants also encouraged the kinds of annual calen- 
drical observations described in chapter 4. 

In this chapter I look at the organizational features of the 
movement, and then turn to a divisive debate that split its mem- 
bers along generational and political lines during Wo rid War I. 


Unlike the Deobandis and the Nadwat al-'Ulama, the Ahl-e 
Sunnat movement did not have a central Dar al-'Ulum at 


ch5.044 10/12/2004 5:22 PM Page 11 



Bareillv or in any of the other small towns in the United 
Provinces where they were influential, although thev did have 
small-scale, relatively modest madrasas in Bareillv and else- 
where. An important madrasa associated with the movement 
later on was the Dar al-'Ulum Hizb al-Ahnaf in Lahore, 
founded in 1924 by Sayyid Didar 'Ali Alwari (1856-1935), 
who belonged to the Chishti (Nizami) order of sufis. He 
counted Ahmad Riza as one of his teachers, having received a 
certificate from him in jurisprudence (fiqh) and hadith, among 
other things. Like all the other Ahl-e Sunnat madrasas, the Hizb 
al-Ahnaf taught the Dars-i Nizami svllabus. Amplv supported 
by financial contributions by Panjab-based pirs, it trained large 
numbers ("hundreds of thousands," according to Sayyid Didar 
'Ali's grandson) of 'ulama and teachers throughout Panjab. 
Organized along the same lines as Deoband's Dar al-'Ulum, it 
also had specialized departments of preaching (tabligh) and 
debate. Preachers were needed both to counter the influence 
\7" of rival Muslim movements (described in the Ahl-e Sunnat lit- —kzj~ 

erature as "Wahhabis") and the Arya Samaj. The Arya Samaj, a 
Hindu reformist organization founded bv Swami Davanand in 
the Panjab in the 1860s, had become a matter of concern for 
Muslim reformers in the earlv twentieth century, on account of 
its shuddhi or reconversion movement, that is, the effort to con- 
vert Hindus who had converted to Islam back to Hinduism. 
Debaters had a similar competitive function, namely, to 
increase Ahl-e Sunnat influence and curtail that of its rivals. 

A number of smaller Ahl-e Sunnat madrasas dotted the north 
Indian plains: the Madrasa Manzar al-Islam in Bareillv founded 
bv Ahmad Riza in 1904, and managed bv his brother and bv 
Zafar ud-Din Bihari, was perpetually short of funds, particu- 
larly during the war years (191 4—1 8) . In Badavun, the Madrasa 
Shams al-'Ulum was founded in 1 899 , and fared well because of 
a grant from the Nizam of Hyderabad (which lasted until 1948, 
when Hyderabad was incorporated into independent India). It 


ch5.044 10/12/2004 5:22 PM Page lLc- 


also received British support in the form of land and buildings. 
This madrasa had a separate publications wing, and its graduates 
went on to pass BA examinations in Urdu and Persian at 
Panjab and Allahabad universities, earning the title of Maulawi 
'Alim (Urdu) and Munshi Fazil (Persian). In Muradabad, 
Ahmad Riza's close follower Na'im ud-Din Muradabadi 
founded a school, the Madrasa Na'imivva, in the 1920s. Bv the 
1930s it had become large enough to earn the title of Jam'ivva 
(center of learning). It had a Dar al-Ifta (center for the writing 
of fatawa) and handsome buildings. In the late 1 980s, I found it 
in the heart of Muradabad, surrounded bv narrow lanes and 
bustling commerce. The classrooms surrounded an open court- 
vard, to one side of which was Na'im ud-Din's simple tomb. 

These and countless other madrasas like them (many are 
flourishing in the Pakistani cities of Karachi, Lahore, and else- 
where todav) are "modern" in the sense that thev no longer fol- 
low the one-to-one stvle of instruction practiced until the 
\7" mid-nineteenth centurv, and also in that thev have annual —kzj~ 

examinations, classrooms, libraries, and all the other organiza- 
tional features of regular public schools. But the svllabus is still 
based on the Dars-i Nizami curriculum. 


In the early nineteenth centurv, most printing presses in India 
were owned bv Christian missionaries who used them to pub- 
lish copies of the Bible in Indian languages and a few classical 
Indian texts. Bv the 1 880s, however, the situation had changed 
dramatically, as Indian-owned printing presses grew in number 
and Indian-language publishing blossomed. In Bareillv, two 
printing presses published most of Ahmad Riza Khan's fatawa 
and other writings between them. They were the Hasani Press, 
owned bv Ahmad Riza's brother Hasan Riza Khan (and later by 


ch5.044 10/12/2004 5:22 PM Page IV 


his nephew Hasnain Riza Khan) , and the Matba' Ahl-e Sunnat wa 
Jama'at, run bv Ahmad Riza's close follower Amjad'AliA'zami. 
The earlier books published date from the late 1870s. Some 
were only fifteen pages long, others had hundreds of pages, and 
most fell somewhere in between. They had print runs of 
between five hundred and a thousand copies, and popular titles 
were sometimes reprinted as many as three times. For instance, 
an anti-Deobandi fatwa on the need to respect graves (entitled 
Ihlak al- Wahhabiyyin 'ala Tauhin Qubur al-Muslimin, or Ruin to the 
Wahhabis for their Disrespect toward Muslim Graves) was first 
published in 1904, and reprinted for the fourth time in 1928 
with a print run of a thousand. Given the fact that books were 
often read aloud (since there were many more people who 
could not read than those who could), the reach of a single copy 
"was much greater than is apparent from the numbers. 

Much care was expended in finding an appropriate title, the 
beginning and end of which not onlv rhvmed but also poked fun 
\7" at the opponent. For example, in 1896 Hasan Riza Khan, the —kzj~ 

owner of the Hasani Press, wrote an anti-Nadwa work entitled 
Nadwe ka Tija — Rudad-e Som ka Natija (The Nadwa's Tija —The 
Result of Its Third Report). The word tija means the third dav 
after a person's death. Thus, Hasan Riza implied that the 
Nadwa's third report showed that the Nadwa was dead as an 
institution. (Ahmad Riza alone wrote more than two hundred 
fatawa against the Nadwa.) Furthermore, the numerical value 
of the individual letters (in accordance with the abjad svstem, 
which assigns each letter of the Arabic alphabet a number) 
vielded the date of the work when added together. 

Journals were another category of publication. In Patna, Qazi 
'Abd ul-Wahid Azimabadi, a close follower, started publishing a 
monthly journal, the Tuhfa-e Hanajiyya (Hanafi Gift) in 1897—8, 
with the primary purpose of rebutting the Nadwa. It carried art- 
icles about tenets of the faith, jurisprudence, hadith, stories about 
the prophets and the first caliphs, and reports about rival 'ulama 


ch5.044 10/12/2004 5:22 PM Page 1L£- 


organizations, particularly the Nadwa. It had a small circulation 
(of two hundred to two hundred and fifty), its subscribers being 
the educated elite (ru 'asa, sing, ra'is) of towns in Bihar, the United 
Provinces, Bombav, Ahmadabad, and Hyderabad. 

Newspapers constituted vet another kind of Ahl-e Sunnat 
publication. Here I am thinking particularly of the Rampur- 
based Dabdaba-e Sikandari (in translation, Alexander 's [Awesome] 
Majesty), which began weekly publication in the 1860s. I 
examined issues dating from 1908 to 1917. The paper had a 
pro-British perspective, which paralleled that of the Nawab of 
Rampur (though it was privately owned by a scholar of the 
Chishti (Sabiri) line of sufis; the nawab probably patronized the 
paper, but he did not own it). In its international political 
coverage, the Dabdaba reported on the war and other major 
events in Europe, as well as national events in India (such as the 
constitutional devolution of power to Indians in the early 
twentieth centurv), particularly those of interest to Muslims. 
\7" In addition, it devoted space to purely "religious" events, such —kzj~ 

as 'urs announcements, detailed reports on a divisive debate 
within the Ahl-e Sunnat movement about the call to prayer (see 
pp. 118—22), and, during Ramadan, the exact time of sunrise 
and sunset as determined by the 'ulama. Starting in 1910,italso 
devoted two full pages (out of sixteen) to fatawa by the Ahl-e 
Sunnat 'ulama in answer to questions. Bv February 1 9 1 2 , it had 
published two hundred such fatawa. 


Finallv, let us look at two other kinds of activities less depend- 
ent on the written word, namely, voluntary associations and 
oral debates, common to all the reform movements of the late 
nineteenth centurv. The organizational structure of voluntarv 


ch5.044 10/12/2004 5:22 PM Page \\A-~ 


associations, like that of the madrasas, showed clear British 
influence, in that each had office bearers, annual reports, fund- 
raising committees, and so on. To quote Kenneth Jones on the 
Arya Samaj, "Battles were fought, victories won, and defeats 
suffered according to the proper forms of parliamentary pro- 
cedure" between the rival "sabhas, samajes, clubs, anjumans, 
and societies" which proliferated in this period (Jones, 1976: 

Thus in 1921, the Ahl-e Sunnat 'ulama formed an organiza- 
tion called "Ansar al-Islam" (Helpers of Islam, the word 
"Ansar" being a reference to the seventh-century helpers of the 
Prophet at Medina) in order to raise money for the Ottomans 
after their defeat at the hands of the Allies in World War I. It was 
but one of several such Indian organizations, and was in compe- 
tition with the Farangi Mahall-led (and much better known) 
Anjuman-e Khuddam-e Ka'ba (Societv of the Servants of the 
Ka'ba). Another organization, the Jama'at-e Riza-e Mustafa 
\7" (Society Pleasing to the Prophet Muhammad), was formed -\^r 

around 1924 in order to counter the conversion efforts of 
the Arva Samaj . 

Oral disputations (munazara) "were perhaps the oldest form 
of contestation between rival groups, both Hindu and Muslim. 
In the earlv nineteenth centurv, the contestation had been 
between Christian missionaries on the one hand and Hindu or 
Muslim learned men on the other. In the latter half of the cen- 
tury, by contrast, the contestants were often adherents of the 
same religion, challenging each other's version of reform. The 
disputations "were highlv public events observed bv large num- 
bers of onlookers, and thus they had the air of a fair (me/a). They 
lasted several hours, sometimes several days. Although neither 
side was ever won over, both usually left feeling thev had won. 
Here is the description of a disputation between an Ahl-e 
Sunnat contestant and a Deobandi one, from the Ahl-e Sunnat 


ch5.044 10/12/2004 5:22 PM Page 11 



In 1 9 1 9-1 920, [Ahmad Riza] sent Hashmat 'All to debate 
with [a Deobandi 'alim] at Haldwani Mandi, all bv himself. 
He was only nineteen years old. He harassed his opponent 
and silenced his argument [in defense of a Deobandi book] . 
And on the question of the Prophet's knowledge of the unseen, 
[the opponent] was left astounded. This was his first debate. . . . 
After successfully defeating his opponent, he returned to 
[Ahmad Riza, who] was very pleased with his report, embraced 
him, and prayed for him. He gave him the name "the father of 
success," as well as a turban and tunic, and five rupees. He also 
said that henceforth [Hashmat ' Ali] would get five rupees every 
month. . . . And, by the grace of Allah, [Ahmad Riza's] favor was 
always with him, and he won a debate on every occasion. 
(Mahbub 'Ali Khan, 1960: 7-8) 

The following reports on a disputation between the Ahl-e 
Sunnat and Swami Shraddhanand, leader of the Arva Samaj, 
that never took place : 

\J" When Shraddhanand began [the conversion movement], ^ J 

Hazrat [Na'im ud-Din Muradabadi] invited him to a debate. 
He accepted the invitation. Hazrat went to Delhi [to debate 
with him]. He ran from there and came to Bareilly. Hazrat 
went to Bareilly and challenged him to debate. He ran from 
there and went to Lucknow. When Hazrat went to Lucknow, 
he went to Patna. Hazrat followed him to Patna, but he went 
to Calcutta. Hazrat went there too, and caught [up with] him. 
He then clearly refused to debate. (Na'imi, 19S9: 9) 

The point was made: Swami Shraddhanand knew he would lose 
if he debated with Na'im ud-Din Muradabadi but could not 
refuse the challenge thrown at him. The disputations had an 
element of "social inversion," a term used by scholars who have 
studied public theater to describe occasions when the normal 
social etiquette observed between equals is dispensed with, 
giving each partv license to insult the other. Occasionally, the 
disputations led to violence. 


ch5.044 10/12/2004 5:22 PM Page 11,4 




In January 1914, alongside news of impending war in Europe 
and national events in British India, the Rampur newspaper 
Dabdaba-e Sikandari began to report a verv local story. The 
issue, which had evidently been agitating the 'ulama of Bareillv 
and a number of local country towns — all of whom identified 
with the Ahl-e Sunnat movement — for some time before it 
began to be reported in the paper, had to do with the Fridav 
noontime praver, the most important of all the weeklv prayers. 

The Fridav noontime prayer is distinguished bv the fact that 
the call to it is sounded twice rather than once. The question 
"was: should the muezzin, the one who issues the call, be stand- 
ing inside the mosque or outside it when he makes the second 
call? According to Ahmad Riza, he should be standing outside 
the mosque, for this had been the practice since the time of the 
-yy— Prophet and the first two caliphs. He cited a hadith from Abu —yy- 

Daud in support of his view. Opposing him were 'ulama from 
towns near Bareillv, such as Rampur, Pilibhit, and Badayun. 
They argued that the second azan had been sounded from 
within the mosque since the beginnings of Islam and that there 
was no reason to change the practice now. 

The space devoted to this dispute in the Dabdaba-e Sikandari 
indicates that it had been brewing for some time. In January 
1914, Ahmad Riza addressed a number of related questions: 
what had been the precedent and model (sunnat) set by the 
Prophet and his closest companions? What should be done 
when the prevailing practice contravened this ideal? Should 
people change their practice to conform to the ideal? 

Ahmad Riza's response — that the current practice of sound- 
ing the call from within the mosque was mere custom (rawaj) 
and had no basis in either the Qur'an or the hadith, and that it 
was the 'ulama's moral dutv to "revive a dead sunnat" — was 


ch5.044 10/12/2004 5:22 PM Page 11* 



followed bv a request that supporters of his view let the Dar 
al-Ifta at Bareillv know. They were also asked to collect the 
signatures of those who had decided to follow his lead, and to 
send them on to the Dar al-Ifta. 

Two weeks after Ahmad Riza's Januarv fatwa, his opponents 
countered that the practice of sounding a second azan had not 
existed during the Prophet's time.Thev said it had been started 
by the third caliph, Caliph 'Uthman (r. 644—56) , and that it had 
been done from within the mosque for thirteen hundred years: 
far from reviving a dead sunnat, Ahmad Riza and his followers 
were "kill[ing] a living sunnat. And far from getting a reward, 
[thev] would be punished." Ahmad Riza Khan was basing his 
view on his own independent reasoning (ijtihad), thev claimed; 
it was the consensus of the communitv that ought to prevail, 
not the opinion of a single scholar. Given that Ahmad Riza — like 
the majority of Indian 'ulama — laid no claim to exercising 
ijtihad, and that he believed firmlv in staving within the confines 
\7" of the Hanafi law school, eschewing even the mixing of law —kzj~ 

schools after the fashion of Ashraf 'AliThanawi (see p. 70), this 
was a particularly offensive charge. 

In subsequent months the debate grew more heated as accus- 
ations proliferated on both sides. Ahmad Riza argued that the 
'ulama opposing him were misleading the people, "turning 
their backs on religion [din] ," "slandering the shari' a," "follow- 
ing a bid V rather than a sunna, and "committing a grave sin ."In 
a later issue of the Dabdaba, he accused a particular scholar of 
being influenced bv certain Deobandi 'ulama. A follower of 
Ahmad Riza's offered a Deobandi scholar a fiftv-rupee prize if 
he was able to satisfactorily answer a list of questions related to 
the second azan. The debate was thus widening beyond the 
original group of contestants. 

In February 1915, Ahmad Riza successfully secured the 
signatures of a small number of 'ulama from the Haramain 
assenting to his point of view. One of the Medinan scholars 


ch5.044 10/12/2004 5:22 PM Page 12 


wrote : "There is no advantage to giving the azan in the mosque . 
Those people who are outside [are alerted by the azan that 
thev] should strive after the remembrance of Allah." It was also 
reported that pamphlets were now being written bv both sides. 
The opponents now included 'ulama who considered them- 
selves Ahl-e Sunnat as well as Deobandis, as some Ahl-e Sunnat 
'ulama began to defect to the other side. The use of pamphlets 
was also significant, as a pamphlet might reach more people 
than either a fatwa or a newspaper. Being intended for a wider 
audience than a small erudite circle of 'ulama, it might also be 
written in a looser, more informal stvle. (On the other hand, it 
must be admitted that fatawa were often published in the form 
of little booklets or pamphlets for general circulation as well.) 

The next stage in the debate was quite dramatic: sometime 
in 1916 a court case was instituted against Ahmad Riza in 
Badavun on a charge of libel. The details are unclear, but the 
plaintiff charged that one of Ahmad Riza's pamphlets (entitled 
\7" Sad al-Firar, A Hundred Flights [i.e., Defeats]) was libelous of ~x3~ 

Maulana 'Abd ul-Muqtadir Badavuni, "who had recently died. 
This was a surprising development, because the latter came 
from a familv with close ties with Ahmad Riza's own. 
Moreover, Maulana 'Abd ul-Muqtadir had plaved a prominent 
part at the 1900 meeting in Patna at which Ahmad Riza had 
been proclaimed mujaddid, having initiated that proclamation. 

In 1917 Ahmad Riza "was summoned to court, but failed to 
appear. This was a clear indication that he did not acknowledge 
the authority of the court. His reasons included the public set- 
ting of a British Indian court, in "which British Indian law rather 
than shari'a law was applied, and one in which the judge him- 
self was usuallv a non-Muslim (in this case it was a Hindu) . 
Some months later the judge dismissed the case, saving the 
plaintiff had no grounds for his case. Ahmad Riza's supporters 
interpreted this judgment as a victorv, and the event was cele- 
brated with the group recitation of verses in praise of the 


ch5.044 10/12/2004 5:22 PM Page 1 



Prophet (na 't) and victory processions in Bareillv. There ended 
the "azan debate," as no appeal was filed. (Nevertheless, the 
actual practice of calling the second azan from within rather 
than outside the mosque did not change either.) 

This debate shows us how the community, in the sense of the 
people who actually interacted and took account of each other, 
had become vastlv bigger than it used to be and now included 
people who were linked bv newspapers and other modern 
communications. It probablv started as an oral discussion in 
Bareillv, then moved on to debate in the Rampur newspaper, 
then widened further still when Ahmad Riza received approval 
for his point of view from Mecca and Medina, and finallv moved 
to a British Indian court where he was charged with libel. The 
audience increased substantially after the Rampur paper began 
reporting on it in 1914. The paper was probablv read bv the 
literate Muslim classes throughout the modern state of Uttar 
Pradesh — by 'ulama, landed gentrv, and urban professionals. 
\7" These people, who had probablv heard of Ahmad Riza even if —kzj~ 

thev did not know of him personally, became part of a 
"consuming public" — following Benedict Anderson's insights in 
Imagined Communities — through their act of reading the paper. 

It was characteristic of this public that it was anonymous, 
unlike the initial group of people close to Ahmad Riza, whose 
lovaltv he could count on. Public opinion had to be won over. 
Presentation of validating opinions from Mecca and Medina 
was important in the Indian context precisely because it could 
be expected to carry weight outside the circle of people bound 
to Ahmad Riza bv personal ties. In the final stage of debate, that 
centered on the courtroom, the issue became even more pub- 
lic and, for the first time, political as well, in the sense that the 
authority of the colonial state was being pitted against that of a 
traditional scholar. 

The fissures in the Ahl-e Sunnat movement, evident at the 
conclusion of the azan debate, were occurring at a time of 


ch5.044 10/12/2004 5:22 PM Page 12 


enormous political change in British India. The politicization of 
the Muslim community had begun even before World War I — 
witness not onlv the organizational efforts of 'Abd ul-Bari 
(see p. 78), but also the Kanpur mosque dispute and riot in 
1 9 1 3 , in which the Sunni Muslims of Kanpur, a major city in the 
United Provinces, angrilv protested against the demolition bv 
the British Indian government of part of a mosque in order to 
make "way for a road. The issues that arose after the war — 
whether or not to join the Indian National Congress, or form a 
joint partv of 'ulama from different movements, or abstain 
from politics altogether — were to grow in urgencv after 
Ahmad Riza's death in 1921 .The new leaders of the movement 
adopted different solutions and led their followers in different 
directions. While no single person "was able to unite the Ahl-e 
Sunnat movement as Ahmad Riza had done, this "was perhaps a 
sign of the success of the movement rather than the reverse, 
illustrating its geographic spread and growth far beyond 
\7" Bareillv, its original birthplace. —kzj~ 


There are no statistics to tell us which of the rival reform move- 
ments of the late nineteenth and earlv twentieth centuries had 
the most followers, particularly in the early twentieth centurv. 
Most scholars believe that the Deobandis were influential in the 
urban areas, while the "Barelwis," as the Ahl-e Sunnat are 
widely known, "were popular in the countryside. If this were 
true, it would make the Ahl-e Sunnat vastly more influential 
than the Deobandis, and probablv the erudite Ahl-e Hadith as 
well, not to mention the followers of Sayyid Ahmad Khan, 
as the South Asian population was and continues to be 


ch5.044 10/12/2004 5:22 PM Page 12^~ 


overwhelmingly rural. However, this judgment arises from the 
general identification of the "Barelwis" with sufism, and with 
unreformed Islamic practice among the population at large. 
But since we have no way of knowing whether Muslims who 
prayed at the sufi shrines that are ubiquitous throughout South 
Asia thought of themselves as "Barelwi," we cannot make this 

How we name things affects how we think about them. 
Those who think of "Barelwis" think of a general sufi-oriented 
group of people with a vast popular following. However, if we 
keep the self-image of "Ahl-e Sunnat" before us, we see the 
movement as more focused and less diffuse. In my view, we 
have to start by looking at those who identified with the move- 
ment by attending its schools, subscribing to and buving its 
journals, attending its meetings, and participating in other 
ways in the issues that engaged its leadership. In addition, con- 
sidering that the people who did these things were part of the 
\7" literate elite, by definition a small minority, we can assume that —kzj~ 

a larger number of people around them were influenced by 
being read aloud to, and by constituting a silent audience that 
attended and participated in events. Even when we add these 
people in, the membership of the Ahl-e Sunnat movement 
could not have exceeded thousands, perhaps tens of thousands, 
particularly in the late nineteenth century. 

Some examples will help put this in perspective. Thus, 
as noted earlier in the discussion of Ahl-e Sunnat publications, 
in the 1 890s a strong anti-Nadwa campaign was waged by a 
follower of Ahmad Riza's in Patna, Bihar, through the journal 
Tuhfa-e Hanajiyya (Hanafi Gift). Its circulation at its height 
was about two hundred and fiftv. Most of its subscribers in its 
early davs were from Bihar (72 out of 119), followed by the 
United Provinces (23). Their professions included legal 
representatives, revenue collectors, students, mosque leaders, 
and school administrators, among others. Another example is 


ch5.044 10/12/2004 5:22 PM Page 12 


provided by the printing history of an anti-Deobandi fatwa by 
Ahmad Riza dealing with the Deobandis' alleged disrespect 
for graves and gravesites (Ihlak al- Wahhabiyyin 'ala Tauhin Qubur 
al-Muslimin, Ruin to the Wahhabis for Their Disrespect 
toward Muslim Graves). It went through four printings 
between 1904 and 1928; a thousand copies were printed in the 
fourth printing (we have no numbers for the earlier editions) . 

Given that Ahmad Riza Khan was less interested in schools 
than were the Deobandis, the school network of the latter was 
wider and more influential than that of the Ahl-e Sunnat.To cite 
some rough numbers: the Madrasa Manzar al-Islam in Bareillv 
graduated between four and ten students per year between 
1 908 and 1917. Resources "were poor, with few teachers, class- 
rooms, and inadequate library and boarding facilities. Schools 
run bv Ahmad Riza's followers in other north Indian towns also 
tended to be relativelv modest, though as indicated above they 
increased steadily through the years. Bv comparison, bv 1900 
\7" the Dar al-'Ulum at Deoband had about a dozen teachers and —kzj~ 

between two and three hundred students in a given vear, new 
buildings, including classrooms and boarding facilities, and it 
graduated about fifteen thousand students in its first hundred 
vears (1867—1967). It also had a wide network of affiliated 
schools, starting in 1866 with a school at Saharanpur, just north 
of Deoband in the western part of the Northwestern 
Provinces. Although its student numbers were small (about a 
hundred), the network 'was constantly growing. 

Numbers for other aspects of the two movements are hard 
to specifv, though thev can be assumed to have been similar. In 
general, the two were often paired as oppositional groups: 
thus, "Deobandi— Barelwi" was a common term for sectarian- 
ism within the Indian Muslim fold. 

The Ahl-e Sunnat side gained additional strength from 
another quarter, namelv, reformist sufi groups which sup- 
ported them on specific issues. Reformist sufis (of the 


ch5.044 10/12/2004 5:22 PM Page 12^~ 


Naqshbandi and Chishti orders) were distinguished from the 
vast populace bv their insistence on adherence to the shari'a 
and a general concern for reform. In the Panjab, a state with 
wealthv sufi hospices, such sufis had great influence. To cite an 
example, Pir Mehr 'Ali Shah of Golra (18S6— 1937), a small 
town in the Panjab, who was directlv associated with the Ahl-e 
Sunnat movement, went to the Northwestern Provinces to 
studv Qur'anic exegesis (tafsir) and hadith under reformist 
'ulama, then returned to Golra to transform it into a reformist 
Chishti center. Anti-British in his politics, he instructed his fol- 
lowers to be personally observant and promoted knowledge of 
religious law among his followers. He often issued fatawa on 
points of religious law. His self-identification with the Ahl-e 
Sunnat added to the influence of the movement in Panjab state. 
To sum up, the reformist groups had different regional 
emphases but more or less equal overall importance in the 
countrv as a whole, parti cularlv in the Northwestern Provinces 
\7" (renamed the United Provinces of Agra and Oudh in 1900). —kzj~ 

However, because the Deobandis emphasized schools more 
than the Ahl-e Sunnat, in the long term thev had greater influ- 
ence in the urban areas than the Ahl-e Sunnat. 


ch5.044 10/12/2004 5:22 PM Page 12 




ch6.044 10/12/2004 5:26 PM Page 12 


Ahmad Riza Khan, leader of the Ahl-e Sunnat or "Barelwi" 
movement, was quintessentiallv South Asian. The movement 
he led made universalist claims, as its very name makes 
clear. Translated, the term Ahl-e Sunnat wa Jama'at means 
"the devotees of the Prophet's practice and the broad com- 
munity." It resonates with Sunni Muslims the world over, and 
has been used in the past bv Sunni Muslim movements in dif- 
-(^y— ferent historical contexts and geographical settings as a means (^J 

of identifying their own communitv with that of the first 
Muslims established by the Prophet in seventh-century Arabia. 
It was Ahmad Riza's firm belief that he was following in the path 
of the Prophet, and in everything he did and said he considered 
the Prophet his model. To those who agreed, this made him, 
Ahmad Riza, a model for emulation in his turn. 

Ahmad Riza's interpretation of the sunna of the Prophet "was 
informed bv ideas of hierarchy and religiositv derived from sufi 
notions of "love" for the Prophet, and expressed itself in ritual 
worship centered on sufi shrines and calendrical anniversaries 
of sufi pirs, Shaikh 'Abd ul-Qadir Jilani, and, of course, the 
Prophet's birthdav. It "was thus informed bv personal devotion 
to a wide arrav of pious and holv ancestors. This was its hall- 
mark and its source of strength. A warm, loving (and simultan- 
eously demanding) relationship between each believer and 
his or her pir lav at its heart. Such a relationship is particularly 



ch6.044 10/12/2004 5:26 PM Page 12 


resonant in the South Asian context, for it mirrors similar ties 
among other religious communities in the subcontinent, par- 
ticularly Hindu followers of the bhakti tradition. Bhakti or devo- 
tional "worship of God emphasized the individual believer's 
relationship with a personal god (forms of Vishnu or Shiva). 
"The devotee's . . . adoration "was often focused on the person of 
a human guru or spiritual preceptor who "was revered as a liv- 
ing manifestation of the god" (Bayly, 1989: 41). In fact, south 
Indian sufi texts since the fifteenth century have frequently 
interwoven Hindu and Muslim sufi motifs, enabling the 
Muslim saint to "leap the boundaries between 'Hindu' and 
'non-Hindu', 'Islamic' and 'un-Islamic' "(Bavlv, 1989: 120). 
Critics of the Ahl-e Sunnat also claim that ritual practices dur- 
ing the Prophet's birth celebrations (milad) resemble Hindu 
worship practices. Indeed, despite some major differences 
between the two traditions, such as the lack of images and of 
priests in the Islamic context, there are many similarities: for 
\7" instance, food and water offered to and consecrated by the —kzj~ 

saint, then consumed by the worshiper, the sprinkling of rose 
petals in the sanctum, the recitation of religious texts and the 
telling of exemplarv stories about the Prophet and the saints 
are similar to Hindu worship practices. 

Nevertheless, I take seriously the Ahl-e Sunnat claim to be a 
reformist movement. While critics might argue that the Ahl-e 
Sunnat were too accommodating of local practice, too local, 
and too parochial to be considered "reformist" — unlike the 
Deobandis or the Ahl-e Hadith or the Nadwa, for example — I 
would argue that the Ahl-e Sunnat movement "was reformist in 
the self-consciousness of its practice, and in its insistence on 
following the sunna of the Prophet at all times. In paving atten- 
tion to every detail of their comportment on a dailv basis, 
members of the Ahl-e Sunnat were no different from followers 
of rival movements at the time. What set them apart from the 
other movements "was their interpretation of "what, in practice, 


ch6.044 10/12/2004 5:26 PM Page 12 


was entailed bv following the Prophet's example. While they 
interpreted this in more customdaden terms than their rivals, 
in their view thev never transgressed the boundaries of the 
shari'a at any time. 

While the Ahl-e Sunnat movement was certainlv more 
inclined toward the emotional or magical than the Deobandi, 
both shared a common wo rldview. Ahmad Riza was punctilious 
about observing the sunna, as he interpreted it, in every detail 
of his life, and taught his followers to do likewise. Frowning on 
what he considered be-shar' (without shari'a) behavior, he 
dressed, "walked, and conducted himself with others in ways 
that conformed with what he took to be the shari'a. Public 
events such as the milad and 'urs were also conducted within the 
bounds of shari'a — without use of drugs and intoxicants and 
qawwali singing (though the latter was allowed in small groups 
by some 'ulama), and emphasizing Qur'an readings and the 
recitation of poetrv in honor of the Prophet. Like the other 
\7" reform movements, he and the Ahl-e Sunnat 'ulama in general —kzj~ 

also encouraged their followers to fulfill the five "pillars" of 
Islam and to refrain from antisocial behavior of anv kind. 


Since his death in 1921 , Ahmad Riza's 'urs has been celebrated 
bv his followers every vear in Bareillv. In 1987,1 was in Bareillv 
during the 'urs. Here is a transcription from my notes: 

I attended one session of the three-day annual 'urs 
celebrations for Ahmad Riza and his son Mustafa Riza. 
Women are discouraged, though not prohibited, from going. I 
was amazed at the size of the crowd. A newspaper report the 
next day said that lakhs, or hundreds of thousands, of people 
had attended. The program consisted of three days of 


ch6.044 10/12/2004 5:26 PM Page l: 


speeches, interspersed with Qur'an readings and recitation of 
na't poetry in praise of the Prophet. 

The venue is a large open ground adjacent to the local 
college, which suspends classes for the duration of the 'urs 
and makes the classrooms available to people to sleep in at 
night. So for three days the place is like a large camp, with 
provision for food and shelter for some thousands of people. I 
didn't get to see what went on in the khanqah itself, because 
of the crowd. It is down in the heart of the city, accessed 
through narrow lanes and alleys, and there was no way one 
could force one's way through — my host and guide was most 
reluctant to attempt it. 

In Pakistan, I found that Ahmad Riza's death anniversary was 
also commemorated with conferences at five -star hotels at 
which speeches were made and na 't poetry recited. There are a 
number of Pakistani organizations which sponsor events 
honoring Ahmad Riza's life and work throughout the vear as 
\7" well as publishing his books. One of the most prominent of ~X7~ 

these, called Idara-e Minhaj al-Quran, was headed bv a law 
professor,Tahir ul-Qadiri, in the late 1 980s.Tahir ul-Qadiri 'was 
a well-known public figure in Pakistan, as he made frequent 
appearances on national television, delivered speeches at 
mosques, and was active at conferences. At a more grassroots 
level, the Ahl-e Sunnat were busv building schools (madrasas) 
throughout the countrv. Zaman (2002: 235 n. 51) estimates 
that the number of Ahl-e Sunnat schools went up from 93 in 
1971 to 1,216 in 1994. The Ahl-e Sunnat are also represented 
at the political level. Their partv is known as the Jamiyvat 
al-'Ulama-e Pakistan (JUP) and its leader through the 1970s 
and 1980s was a well-known 'alim and pir, Maulana Shah 
Ahmad Nurani. 

I should add, however, that the Ahl-e Sunnat in Pakistan 
appear to be less prominent nationally than the Deobandis. 
Their perspective on sufism being at odds with that of the Saudi 


ch6.044 10/12/2004 5:26 PM Page 1 



regime, thev have not benefited from Saudi munificence as have 
other reformist groups (Zaman, 2002). 


It is not only in Pakistan that the Ahl-e Sunnat are active. 
They are 'well represented in other parts of the world as well, 
chieflv Great Britain, where immigration from the subcontin- 
ent has been sizeable since independence. The late 1960s 
saw a transformation in the South Asian Muslim immigrant 
population as a whole, as immigrants began to see themselves 
for the first time as permanent settlers rather than temporary 
migrants. As male workers were joined bv their families, the 
need was felt for institutional structures — chieflv mosques 
and schools — which would allow communitv life to flourish. 
-(^y— Mv comments are limited to the Muslims of Bradford, a north- — t^y- 

ern industrial citv representative in many "ways of the overall 

In 1973, Pir Maroof, a prominent Ahl-e Sunnat leader, 
founded the World Islamic Mission ( WIM) , "an umbrella organ- 
ization for Barelwi dignitaries, with its head office located in 
his mosque at Southfield Square in Bradford. ... Its first presi- 
dent [was] Maulana Noorani" (Lewis, 1994: 83). As Lewis 
explains, "the Wo rid Islamic Mission [was] clear lv intended as a 
counterweight to the Mecca-based Muslim World League, a 
vehicle for those whom Barelwis scornfullv dismiss as 
Wahhabis, whether Deobandi, Jama'at-e Islami or Ahl-i 
Hadith" (Lewis, 1994: 84). (The Jama'at-e Islami, founded bv 
Maulana Mawdudi [d. 1979] in 1941, frowns upon sufi 
practices of the kind favored bv the Ahl-e Sunnat.) 

In 1989 the Muslims of Bradford were in the national — 
indeed, international — spotlight following the publication of 


ch6.044 10/12/2004 5:26 PM Page l: 


Salman Rushdie's book, The Satanic Verses. After an initial book- 
burning protest which created the impression among Britons 
that thev were religious "fundamentalists" without furthering 
the British understanding of why they found the book offen- 
sive, the Bradford Council for Mosques, an umbrella group that 
included both Ahl-e Sunnat/ Bare Wis and Deobandis, tried to 
make its case in other ways. In 1990 the Council opened a 
"nationwide debate on the future of Muslims in [Britain]," and 
invited the Bishop of Bradford, as well as Sikh and Hindu lead- 
ers in the citv, to a dialogue, hoping to enlist their support in 
their campaign against the book. As Lewis writes, 

The emphasis of the conference was on the need for a 
constructive engagement with the nation's institutions, 
political, social, and educational. Muslim concerns were 
articulated in an idiom accessible to the non-Muslim 
majority. . . .There was a readiness to be self-critical. . . . Such 
a conference was a tribute to the realism of the Bradford 
Council for Mosques and a refusal to allow Muslims to 
withdraw into sullen resentment. (Lewis, 1994: 164) 

But this was not of course a response unique to the Ahl-e 
Sunnat, who formed one group of manv in these events. 

For all that, it is clear that the Ahl-e Sunnat movement is 
thriving wherever there are South Asian Muslims. Todav it has 
its own websites, as do its competitors, so that one can follow 
the issues engaging its adherents at any time simply by search- 
ing the WorldWide Web. At the present time, its greatest chal- 
lenge appears to be to find common ground with other 
reformist Muslim movements and to promote understanding 
of its perspective among non-Muslims, whose lack of know- 
ledge of the Muslim world leads them to see all Muslims as the 
same, and in a negative light. In this dav and age, the need for 
better understanding couldn't be greater. 


glossary. 044 10/12/2004 5:27 PM 



'alim (pi. 'ulama) scholar of Islamic law 

'amm (pl.'awamm) ordinary (in the plural, refers to 

ordinary people) 
azan call to praver 

bid'a reprehensible innovation, opposite of sunna 
dar ul-harb enemy territory; opposite of dar ul-hlam 
dar ul-Islam land where Islamic law (shari a) is in force 
dastar-bandi "tving of the turban," ceremony marking 

the end of a person's studies or apprenticeship to a 

sufi master 
din the faith; opposite oidunja, the "world 
\Z_j faqih jurisprudent, one who is knowledgeable in the law i^J 

fatwa (pi. fatawa) legal opinion given by a mufti 
hadith traditions or stories traced to the Prophet 
hajj pilgrimage to Mecca, one of the five "pillars" of Islam 
ijma consensus of scholars which constitutes one of the 

sources of Islamic law 
ijtihad independent inquirv to establish the legality of a 

particular matter in shari 'a terms 
jihad struggle, can be internal (spiritual) or external (against 

an aggressor) 
khalifa Caliph (during Ottoman rule); also a successor to a 

sufi master 
khass (pl.khawass) special, the opposite of 'amm 
khutba sermon delivered by an 'alim at Fridav noontime 

madhhab legal tradition or school, of which there are four 

among Sunni Muslims 



glossary. 044 10/12/2004 5:27 PM Pa^-e 134 


madrasa a religious academy, "where the Islamic sciences are 

manqulat the "copied" sciences, especially hadith 
mansab/mansabdari a Mughal rank, or the holder of that 

ma'qulat the philosophical or rational sciences 
milad celebration of the Prophet's birth anniversary 
mujaddid Renewer of the shari 'a, expected at the start of 

everv new Islamic century 
na't poetry in praise of the Prophet 
nawab a Mughal noble, or semi-independent Muslim ruler 

during Mughal times 
pir sufi master, one who has murids or disciples 
qadi judge who applies Islamic law in a court 
Sayyid descendant of the Prophet 
shaikh "elder" or "leader," in South Asia a title often used of a 

sufi master 

~x^y— shari'a sacred law of Islam — (-_-/- 

Shi'a/Shi'i followers of the Prophet's son-indaw'Ali, and 

other Shi'i imams 
shirk idolatry, associating partners with God 
sufi Muslim mvstic 
sunna the ""way" or "path" of the Prophet Muhammad, as 

known to Muslims through the hadith literature 
taqlid following one of the Sunni law schools in preference 

to iitihad 
tariqa sufi order 
tawhid unity or oneness of God 
'urs celebration of a saint's death anniversary 
wahdat al-shuhud "unity of appearance," a sufi concept 
wahdat al-wujud "unity of being," a contrasting idea 
zakat mandatory alms-tax on accrued wealth 


glossary. 044 10/12/2004 5:27 PM P^cs. 135 


From the Eighteenth to the 
Twentieth Century (to 1947)* 


1707 Aurangzeb dies in the Deccan. 

1709 Nadir Shah and Ahmad Khan Abdali conquer Herat, 
Kabul, Panjab. 

1733 Bengal independent from Mughals. 

1747 Durranis (Afghan dvnasty created bv Ahmad Khan 
Abdali) conquer Delhi. Mughals under Awadh's 

1757 East India Company becomes zamindar of 24 Parganas, 
Bengal, after victory at the Battle of Plassev. 

1765 British nawabi of Bengal and Bihar. 

1772 Rohillas independent until 1792, then come under 
Awadh's protection. 

1773 Awadh becomes a native state under the British. 
1 793 Permanent Settlement in Bengal . 

* Adapted from David Ludden, India and South Asia: A Short History 
(Oxford: Oneworld, 2002), pp. 111-112, 148-149, 198, and 212. 



glossary. 044 10/12/2004 5:27 PM P^cs. 136 


1798 Hyderabad becomes a native state under the British. 

1799 British defeat Tipu Sultan of Mysore. 


1801 Madras Presidency formed. Rampur becomes a native 
state in former Rohilkhand. 

1803 British conquer Delhi and make it a dependency. 

1804 Rohillas absorbed bv Awadh. 

1818 Marathas conquered bv British , and their territory 
forms the bulk of Bombay Presidency. 

1835 Macaulay's Minute on Education. English becomes the 
official language of government and the courts. 

1857 The Revolt or "Mutiny" sweeps across north India. 
C") Calcutta, Bombav, and Madras Universities founded. — C^X- 

1858 India comes under Crown rule. 

1867 Dar al-'Ulum founded at Deoband. 

1875 Arva Samaj founded bv Swami Davanand. 

Muhammadan Anglo-Oriental (MAO) College 
founded in Aliparh bv Savvid Ahmad Khan. 

o J J J 

1885 Indian National Congress founded. 


1905 Partition of Bengal. Anti-Partition protests. 

1906 All-India Muslim League founded at Dhaka. 

1911 Delhi Durbar; Bengal Partition revoked; capital 
moved from Calcutta to Delhi. 


glossary. 044 10/12/2004 5:27 PM Pacts 137 


1914 Gandhi returns to India from South Africa; 

World War I starts; Indian troops sent overseas. 

1919 Khilafat movement launched. 

1920 Non-Cooperation movement; Hijrat movement to 

1921 Dvarchv established. 

1930 RoundTable conferences 1930 and 1932; 
Salt Satvagraha. 

1932 Second civil disobedience movement. Communal 
Award. Gandhi's Poona Pact with B. R. Ambedkar. 

1935 Government of India Act. 

1937 Elections in India. Congress ministries formed in 
seven provinces. Muslim League reorganized. 

1939 WorldWar II starts. Congress ministries resign. 
+ Muslim League declares "Deliverance Day." 

"$" ^> 

1940 Muslim League adopts Lahore Resolution stating goal 

of creating Pakistan. 

1941 Jama'at-e Islami founded bv Maulana Maududi. 

1942 Quit India movement. 

1943 Bengal famine (to 1 944) . 

1946 Cabinet Mission; violence in Bengal; elections. 
Muslim League wins Muslim-majority areas; 

Lord Mountbatten comes to India as Vicerov. 


1947 Independence for India and Pakistan; violence in 
Panjab and Bengal; mass migration and massacre of 
populations; Kashmir accedes to India. 


glossary. 044 10/12/2004 5:27 PM P^cs. 138 


Ahmad Khan, Muin ud-Din. History of the Fara'izi Movement in Bengal 

(1 81 8-1 906). Karachi: Pakistan Historical Society, 1 965 . 
Anderson, Benedict. Imagined Communities: Reflections on the Origin 

and Spread of Nationalism. London: Verso, 1983. 
Bayly, Susan. Saints, Goddesses and Kings: Muslims and Christians in 

South Indian Society, 1 700—1900. Cambridge: Cambridge 

University Press, 1989. 
Bihari, Zafar ud-Din. Hayat-eA 'la Hazrat.Vol. 1 . Karachi: Maktaba 

Rizwiyya, 1938. 
Cohn, Bernard S. Colonialism and Its Forms of Knowledge: The British in 
, India. Princeton: Princeton University Press, 1996. 

Cole,J.R.I. Rootsof North Indian Shi'ism in Iran and Iraq: Religion and 

State in Awadh, 1 722-1839. Delhi: Oxford University Press, 1989. 
Denny, Frederick M. An Introduction to Islam. New York: Macmillan, 

Ewing, Katherine. 1 980. "The Pir or Sufi Saint in Pakistani Islam." 

PhD dissertation, University of Chicago. 
Friedmann,Yohanan. Shaykh Ahmad Sirhindi: An Outline ofHisThought 

and a Study of His Image in the Eyes of Posterity. Montreal: McGill- 

Queen's University Press, 1971. 
Prophecy Continuous: Aspects ofAhmadi ReligiousThought and Its 

Medieval Background. Berkeley: University of California Press, 

"Ahmadiyah," in The Oxford Encyclopedia ojthe Modern Islamic 

World. New York: Oxford University Press, 1995. 
Gilmartin, David. Empire and Islam: Punjab and the Making of Pakistan. 

Berkeley: University of California Press, 1988. 
Hallaq, Wael. "Was the Gate of Ijtihad Closed?" International Journal 

of Middle East Studies, 16, 1 984, pp. 3-41 . 



glossary. 044 10/12/2004 5:27 PM Pa^-e 139 


Hardy, Peter. The Muslims ojBritish India. London: Cambridge 

Universitv Press, 1972. 
Hourani, Albert. Arabic Thought in the Liberal Age, 1 798—1 939. 

Cambridge: Cambridge University Press, revised edition, 1983. 
Imdad Ullah Makki, Haji. Faisla-e Haft Mas 'ala. Reprinted, with 

commentary by Muhammad Khalil Khan Qadri Barkati 

Marehrawi. Lahore: Farid Book Stall, 1406/1986. 
Jones, Kenneth W. Arya Dharm: Hindu Consciousness in 19th-century 

Punjab. Berkeley: University of California Press, 1976. 
Kopf, David. British Orientalism and the Bengal Renaissance: 

The Dynamics of Indian Modernization, 1 773—1 835 . Berkeley: 

University of California Press, 1 969. 
Lelyveld, David. Aligarh's First Generation: Muslim Solidarity in British 

India. Princeton: Princeton Universitv Press, 1978. 
Lewis, Philip. Islamic Britain: Religion, Politics and Identity among 

British Muslims. London: I.B.Tauris, 1994. 
Ludden, David. India and South Asia: A Short History. Oxford: 

Oneworld, 2002. 
C^i Mahbub 'Ali Khan, Muhammad. BulandPaya Hayat-e Hashmat 'Ali. f"^\ 

Kanpur: Arakin-e Bazm-e Qadiri Rizwi, 1 960. 
Masud, Muhammad Khalid. "Apostasy and Judicial Separation 

in British India," in Islamic Legal Interpretation: Miiftis and Their 

Fatwas, Muhammad Khalid Masud, Brinkley Messick, and David 

S. Powers, eds. Cambridge: Harvard University Press, 1996. 
Masud, Muhammad Khalid, Brinkley Messick, and David S. 

Powers, eds. Islamic Legal Interpretation: Muftis and Their Fatwas. 

Cambridge: Harvard University Press, 1996. 
Metcalf, Barbara Daly. Islamic Revival in British India: Deoband, 

1 860— 1900. Princeton: Princeton University Press, 1982. 
"Two Fatwas on Hajj in British India," in Islamic Legal 

Interpretation: Muftis andTheir Fatwas, Muhammad Khalid Masud, 

Brinkley Messick, and David S. Powers, eds. Cambridge: 

Harvard University Press, 1996, pp. 184—192. 
Metcalf, Barbara D. , and Thomas R . Metcalf. A Concise History of 

India. Cambridge: Cambridge University Press, 2002. 
Metcalf, Thomas R. Ideologies of the Raj. The New Cambridge History of 


glossary. 044 10/12/2004 5:27 PM P^cs. 140 


India, vol. III. 4. Cambridge: Cambridge University Press, 

reprint, 2001 . 
Mir Shahamat 'Ali, trans. "Translation of theTakwivat-ul-Iman, 

Preceded by a Notice of the Author Maulavi Isma'il Hajji." 

journal of the Royal Asiatic Society , 13 (1852), pp. 310—372. 
Mushir ul-Haqq, "Unniswin Sadi ke Hindustan ki Hai'at Shar'i: Shah 

'Abd ul- 'Aziz ke Fatawa-e Dar al-Harb ka Ek 'IlmiTajzi'a." 

Burhan, 63, 4 (October 1969), pp. 221-244. 
Na'imi, Ghulam Mu'in ud-Din. "Tazkira al-Ma' ruf Hayat-e Sadr 

al- Aiazil." Sawad-eA'zam, vol. 2. Lahore: Na'imi Dawakhana, 1959. 
Padwick, Constance E. Muslim Devotions: A Study of Prayer-Manuals in 

Common Use. Oxford: Oneworld, 1996. 
Pearson, Harlan Otto. "Islamic Reform and Revival in Nineteenth 

Century India: TheTariqah-i Muhammadiyah." PhD dissertation, 

Department of History, Duke University, 1979. 
Qadiri, MaulanaAulad-e Rasul. "Muhammad Mivan." Khandan-e 

Barakat. Marehra: c. 1927. 
Qadiri Nuri Badayuni, Ghulam Shabbar. Tazkira-e Nuri: Mufassal 
r~~\ Halat o Sawanih-eAbu'l Hussain Nuri Miyan. La'ilpur: 1968. f"\ 

Qureshi, I. H. Ulema in Politics: A Study relating to the Political Activities 

of the Ulema in the South Asian Subcontinent from 1556 to 1947. 

Karachi: Ma'aref, 1974. 
Rahman 'Ali, Maulawi. Tazkira-e 'Ulama-e Hind, trans. Muhammad 

Ayub Qadiri. Karachi: Pakistan Historical Society, no. 16, 1961 . 
Riza Khan, Ahmad. Al-Dawlat al-Makkiyya bi'l Madat al-Ghaibiyya. 

Karachi: Maktaba Rizwiyya, n.d. 
Dawam al- Aishjl'l Ummat min Quraish. Lahore: Farid Book 

Stall, n.d. 

Hada' iq-e Bakhshish . Karachi: Medina Publishing Company, 1976. 

Husam al-Haramain 'ala Manhar al-Kufr wa'l Main. Lahore: 

Maktaba Nabawiyya, 1405/1985. 
77am al-A 'lam ba-an Hindustan Dar al-Islam. Bareilly: Hasani 

Press, 1306/1888-9. 
Al-Mahajjat al-Mu'tamanajl Ayat al-Mumtahana (1339/1 920) . 

Reprinted in Kasai'il-e Rizwiyya, vol. 2, Lahore: Maktaba 

Hamidiyya, 1976. 


glossary. 044 10/12/2004 5:27 PM P^cs. 141 


Maljuzat-e A 'la Hazrat. 4 vols. Gujarat, Pakistan: Fazl-e Nur 

Academy, n.d. 
Naqa al- Salajaji Ahkam al-Bai 'a wa'l Khilafa. Sialkot, Pakistan: 

Maktaba Mihiriyya Rizwiyya, n.d. Originally published in 

Tadbir-e Falah wa Nijat wa Islah. Bareilly : Hasani Press, 

Riza Khan, Hasnain. Sirat-e 'Ala Hazrat. Karachi: Maktaba 

Qasimiyya Barkatiyya, 1986. 
Rizvi, S. A. A. A History of Siifsm in India.Vol. 2. Delhi: Munshi 

Manoharlal, 1983. 
Rizwi, Muhammad Hamid Siddiqi. Takzira-e Hazrat Burhan-e Millat. 

Jabalpur: Astana 'Aliyya Rizwiyya Salamiyya Burhaniyya, 1 98S . 
Robinson, Francis. The 'Ulama ofFarangi Mahall and Islamic Culture in 

South Asia. Delhi: Permanent Black, 2001 . 
Sanyal, Usha. "AreWahhabis Kafirs? Ahmad Riza Khan Barelwi and His 

Sword of the Haramayn," in Islamic Legal Interpretation: Mirftis and 

Their Fatwas, Muhammad Khalid Masud, Brinkley Messick, and 
r~~\ David S. Powers, eds. Cambridge: Harvard University Press, 1996. f"\ 
Devotional Islam and Politics in British India: Ahmad Riza Khan 

Barelwi and His Movement, 1870—1920. Delhi: Oxford University 

Press, 1999. 
"Generational Changes in the Leadership of the Ahl-e Sunnat 

Movement in North India during the Twentieth Century." Modern 

Asian Studies, 32, 3, 1998, pp. 63S-6S6. 
Schimmel, Annemarie. Mystical Dimensions of Islam. Chapel Hill: 

University of North Carolina Press, 197S. 
And Muhammad Is His Messenger: The Veneration of the Prophet in 

Islamic Piety. Lahore: Vanguard, 1987. 
Spear, Percival. A History of India .Vol. 2. Harmondsworth: Penguin, 

Troll, Christian W. Sayyid Ahmad Khan: A Reinterpretation of Muslim 

Theology. Delhi: Vikas, 1978. 
Zaman, Muhammad Qasim. The Ulama in Contemporary Islam: 

Custodians of Change. Princeton: Princeton University Press, 



i ndex. 044 10/12/2004 5:28 PM Page 



'Abd ul-Bari Farangi Mahalli 78, 80, 

'Abd ul-Haqq Khairabadi 55, 58-9, 60 
'Abd ul-Majid Badayuni 80—1 
'Abd ul-Muqtadir Badavuni 120 
'Abd ul-Qadir Badavuni 61 
'Abdal-Qadir Jilani Baghdadi 63, 76, 

and Qadiri sufis 94—6 

ritual to celebrate birthdav 101 
'Abd ur-Rahim, Sheikh 22—3 
'Abd ur-Razzaq, Maulana 88 
'Abd al Wahhab, Muhammad ibn 

'Abd ul-Walid Azimabadi, Qazi 114 
AbuBakr 25,94, 103 
administrators, training of 26—8, 44 
Afghanistan 2,81-2 
Agra 2, 125 
Ahl-eHadith 28,37-9,46,69,100, 

102, 108, 122 
Ahl-e Sunnat 

Dar al-Ifta 70-1 , 84 

in diaspora 131—2 

fissures in 118—22 

help for Turks 80,81, 116 

influence of 85, 122—5 

madrasa 1 1 1-1 3, 1 24, 1 30 

in Pakistan 1 30-1 

publications of 64—5 , 1 1 3—1 5 , 

as reformist movement 12 8—9 

and sufism 91 

voluntary associations 1 1 5—1 6 
Ahmad Ambethwi, Khalil 103, 108-9 
Ahmad Khan Abdali 135 
Ahmad Riza Khan 23,24 

death anniversary 1 2 9—30 

education and training 5 5—7 

family 51—3 

asmujaddid 64-6, 108, 120 

primacy of Prophet 87-8,96-100, 

stories of 57—60 
Ahmadis 28,45-8 
Akbar 1-3 

AI-Dawlat al-Makkiyya 74—6 
Al-Hilal 78 
'All 25,94,95,96 
'Ali Shah, Pir Mehr 125 
Aligarh 44-5, 108, 136 
Amanullah Khan, Amir 81—2 
Amir Khan 31 
Amjad 'Ali 'Azami 84, 1 14 
angels 43, 104 
Anjuman-e Khuddam-e Kaba {Society 

of the Servants of the Ka'ba) 78, 

Ansar al-Islam (Helpers of Islam) 116 
anti-British Revolt, 1857 9, 53, 1 36 
apostasy 37, 70 
Arabic \ 3, 14,20,38 
ArvaSamaj 46, 1 1 2, 1 1 6, 1 36 
Ashraf 'Ali Thanawi 37,70,119 
Aurangzeb 3,135 
Awadh 4, 5,6,7,52,54, 125, 135, 

Azad, Abu'lKalam 78,80 
azan debate 1 1 8—2 1 

Babur 1 

Badavun, Madrasa Shams al- ' Ulum 

'l 12-1 3 
Baghdad 94 
Bahadur Shah Zafar 9 
Barelwi see Ahl-e Sunnat 
Bareilly 51,53,84, 101, 122 

madrasa in 84, 112, 124 
MS for Ahmad Riza 129-30 
Barkatiyva familv 62—3 
Bayly, S. 128 

Benares, Sanskrit College 15 
Bengal 4,7,8, 34, 135, 136, 137 

Fara'izi movement 33—4 

Permanent Settlement of 9—1 1 , 34 
Bengal Renaissance 1 7 




i ndex. 044 10/12/2004 5:28 PM Page 

INDEX 143 


Bentinck, Lord William 12,16 

bhakti 128 

Bihar 8, 123, 135 

Bihari, Zafar ud-Din 51,57,59,60, 
65,71,83,87-8, 100, 112 

Bombay 9 

Elphinstone College 1 6 

Bradford , Muslims in 131—2 

British, Ahmad Riza's relations with 
53,78,82, 110 

British India x, xi, 7—10 

attitudesto 28-9,32,39,43,45 
bovcott of courts 15,34,79,80, 

economic consequences of 9—1 2 
and education 15-16,39-40 

Burhan ul-Haqq Jabalpuri 84 


Brahmo Samaj 1 7 

College of Fort William 15—16 

Hindu College 16 

Madrasa 1 5 
Chishtisufis 92,94,96,112,115,125 
Christians 46,47,89, 109 
Clive, Robert 8 
Cole,]. 3,4 
Cornwallis, Charles, 1st Marquis 10 

Dabdaba-e Sikandari 101,115,118, 

119, 121 
Dar al-Ifta 70-1 ,84,113 
Daral-'Ulum 35,44, 124, 136 
Hizbal-Ahnaf 112 
ofNadwa 39,45 
Dars-i Nizami syllabus 26—7,40—1, 

53,58, 113 
dastar-bandi ceremony 83,101, 



nd, Sv 


ayanand, swarm 
debate 112 

see also oral disputation 
Delhi 8, 135, 136, 137 
College 1 5 

legal status for Muslims in 29—30 
Madrasa-i Rahimiyya 22,26,29, 
31, 55 
Deobandis 28,36-7,75,106,128, 
azan debate 119, 120 
Daral-'Ulum 35-6, 124 
fatawaof 68, 69-70, 72, 108, 109 
influence of 122, 125, 130 

and Noncooperation movement 

andsufism 36, 100, 102 
discipleship 89—93 
Dissolution of Muslim Marriages Act 

Doctrine of Lapse 9 
dreams 61— 2 , 64 
Dudhu Miyan 34 

East India Company 1,4,5,6, 7—1 1 , 

12, 14,29,54, 135 

British 12-13, 15-17 

Muslim 44, 55—6; see also madrasa 
Ellis, F. 16 

as official language 16,136 

in schools 40 
Ewing, K. 92 

Faizullah Khan 6-7,53,54 
Faraizi movement 29 , 33 — 4 
Farangi Mahall 26-8,54,59,88,116 
fatawa 15,36,66-70 

Shah 'Abd ul- Aziz 29-31 

Deobandis 68,69-70,72,108, 
fatawa of Ahmad Riza 58,65,70-3, 

anti-Deobandi 124 

Husam al-Haramain 103-4, 108 

India being dar al-lslam 8 2 

Noncooperation movement 

on practical issues 79 

on the Prophet 57, 74-6 

publication of 71-2, 113, 114, 120 

unbelief 107-8 
Fatawa-e Kizwiyya 71—2 
Fazl-e Haqq Khairabadi 55, 57, 59 
Friedmann,Y. 45,46,47,48,104 

Gandhi, M.K. 80, 137 

ghaus 94 

Ghulam Ahmad, Mirza 45-8, 104, 

God 25,76 

knowledge of 74, 75 

Muhammad and 97—8 

omnipotence of 56—7 

Sayyid Ahmad Khan and 42—3 



i ndex. 044 10/12/2004 5:28 PM Page 

144 INDEX 


Golra 125 

hadith 20,23,24,38,43,67,90 

Hafiz Kazim ' Ali Khan (great-grandfa- 
ther) 52 

Hafiz Rahmat Khan 5 , 6 

Hakim Ajmal Khan 55 

al-Hallaj 96 

Hamid Riza Khan (son) 84,93 

Hanafi school of law 23,24,108,119 

Hanbali school of law 38,64,70 

Hasan Riza Khan (brother) 113 

Hasani Press 72, 113-14 

Hasnain Riza (nephew - ) 5 3, 114 

Hastings, Warren 6—7, 13, 53 

Hijaz 73 

Hijrat movement 81—3, 137 

Hindus 2, 3, 17, 19,46 
devotional worship 1 28 
Muslim relationship with 79, 80, 

81, 109-10 
reformist organization 1 1 2 

Hourani, A. 21-2 

Husain Ahmad Madani 8 1 

Husain bin Saleh 64 

Husam al-Haramain 103-4,108 

Hyderabad 112, 136 

Ibnal-'Arabi 25,48 

Ibn Taimiyya 38 

Idara-e Minhaj al-Quran 1 30 

ijma (consensus) 66, 119 

ijtihad (independent reasoning) 23, 

24,25,66,70, 119 
ImdadUllahMakki 36,92,102 
'InayatUllah 88 

Ahmad Riza's death anniversary 

independence 1 37 

Mughal Empire 1—4 

North Indian successor states 4—5 

political issues in early 20th century 

see also British India 
Indian National Congress 41,77,80, 

intercessions 32, 91—2, 99 
interest 29-30,79 
interpretation of law" 2 3^4- 
Iran 4-5 
Irshad Hussain Rampuri 58, 60 

Islam 42 

see also Shi'ism; Sunni Islam 

Jama'at-e Islami 131, 137 
Jama'at-e Riza-e Mustafa 116 
Jamiyyatal-'Ulami-eHind 37,70,81 
Jamiyvat al-' Ulama-e Pakistan (JUP) 

Jesus Christ 46,47 
jihad 30-1, 32-3, 133 

Ghulam Ahmad and 46 
jihadists 28-9 

Wahhabi trials against 39 
Jones, K. 116 

Jones, Sir William 13-14, 15 
journals 114-15, 123 

Kanpur, mosque debate 1 22 

Karoron Durud 98—9 

khalifas 93 

Khanqah-e 'Aliyya Rizwiyya 84 

Khilafat movement 80-1 ,' 109, 1 37 
knowledge of the unseen 74—6, 77, 

103, 107 
Kopf, D. 17 

Lahore 3,112 
Lahoris 48,49 
law 13-15 

interpretation of 23^4- 

see also Sunni law schools 
Lelyveld,D. 44 
Lewis, P. 131, 132 
logic 26,59,60 
Lucknow 26,40, 52 

Macaulay, T.B. 13, 14 

educational policy 16, 136 
Madras 8, 16, 136 
madrasa 35, 1 34 

of Ahl-e Sunnat 111-13 

in Deoband 35—6 

of Farangi Mahall 26—8 

at Rampur 55, 57 

reform of 40—1 
Madrasa 'Aliyya 55, 57 
Madrasa Hanafiyya 65 
mahdi 46 

Malfuzat 100, 103, 104 
Maliki school of law 70,108 
ma 'qulat (rational sciences) 24, 26, 55, 



i ndex. 044 10/12/2004 5:28 PM Page 

INDEX 145 


manqulat (traditional sciences) 24, 58 
Marathas 3,5,7, 136 
Marehra 62 
marriage 37, 70 

remarriage of widows 32,36 
Masud,M.K. 69-70 
Matba' Ahl-e Sunnat wa Jama'at 114 
Mecca 2,21,69, 121 

Ahmad Riza's pilgrimages to 21, 
32-3,63-4,73,99, 103 
Medina 21,69,99, 121 
Meston, Sir James 82 
Metcalf, B.D. 11,24,25,28,30,31, 

35,44,72, 102, 109 
Metcalf, T. 11,13 
milad 101-2, 128, 129 
Mill, J. S. 12, 14 
miracles 43, 51-2,98 
muftis 14, 27, 66-8, 84 
Mughal Empire 1—4 
muhaddath 46,48 

Muhammad, the Prophet x, 32, 48, 

Ahmad Riza's devotion to 87—8, 
96-100, 127 

birthday celebrations 101—2, 128 

fatwa on 74—6 

God and 97-8 

knowledge of unseen 74—6 , 77 , 
103, 107 

light of 96-7,98 

as model of behavior 19,20—1 

prophethood of 48, 57, 87-8 
Muhammad 'Ali 80 
Muhammad Isma'il 31-2,42,105 

Taqwiyat al-lmam 32, 56, 91, 106 

as unbeliever 107—8 
Muhammad Qasim Nanautawi 36 
Muhammadan Anglo-Oriental (MAO) 

College 44-5, 136 
Muhammadan Educational Congress 

mujaddid 33,49, 108, 134 

Ahmad Riza as 64-6,108,120 

Ghulam Ahmad as 46, 49 
munazaras 116—17,121 
Muradabad 77, 113 
Muslim League 77,136,137 
Mustafa Riza Khan (son) 72 , 84 

Nadir Shah 3,4, 135 

Nad wat al- ' Ulama 28,39-41,64-5 

campaign against 108, 114, 115, 
Na'imud-DinMuradabadi 113,117 
Najd 73 

Naqi 'Ali Khan (father) 53,58, 
influence of 55, 56, 57 
Naqshbandi sufis 94,96,125 
Nizam ud-Din 26 
Noncooperation movement 80, 

109-10, 137 
Nuri Miyan (Shah Abu'l Husain 
Ahmad) 61,89,90-1, 101 

opium trade 1 1 

oral disputations 45—6, 1 1 6—1 7 

Pakistan 2,41,81, 137 

Ahmad Riza's death anniversary 
celebrations 1 30—1 
Panipat 3-4,22 

Panjab 29,45, 112, 113, 125, 135 
Patna 16,65, 123 
penal code 14—15 
Persian language 16,20,24,32,54 
pilgrimage 36,68,102 

relationship with 89-91 , 92, 1 34 

rituals on birth- and deathdays 
Plassy, battle of 7,8 
politicization of Muslim community 

postal system 11—12 
print technology 12,68 
printing presses 1 1 3—1 5 
prayer 36, 38,43 

call to 118-21 
preaching 112 
prophecy 48 
Prophet see Muhammad, the Prophet 

Qadiri Nuri Badavuni, Ghulam 

Shabbar 91 
Qadiri sufis 63, 94-6 
qadis 15,27,37,84-5, 134 
Qadiyyanis 48—9 
qiyas (analogical reasoning) 66 
Qur'an 23,37-8,41,43,66,104 

revelation of 76 

translations of 20, 24, 77 
Qureshi,I.H. 110 



i ndex. 044 10/12/2004 5:28 PM Page 

146 INDEX 


Rampur state 6,7,53—5,136 
nawabsof 7,52,53,58, 115 

Rashid Ahmad Gangohi 36,68-70, 
72, 103, 106-7 

Raza Library 54,5 

reform movements 
18th century 19-22 
19th century 28-49 

religious obligations 19,21 

rituals, sufi 100-2, 127 

Riza ' Ali Khan (grandfather) 5 1 -2 

Robinson, F. 4,26-7,54,59,88 

Rohilkhand 5-7,53 

Roy, Raja Ram Mohan 17,19 

sanad 55-6, 63^4 

Sanskrit 13, 14, 17, 19 

Satan 104, 105, 107 

The Satanic Verses (Rushdie) 132 

Sayyid Ahmad Barelwi 28,31-3,105 

Sayyid Ahmad Dahlan 63 

Sayyid Ahmad Khan 29,41-5,105, 

Sayyid Barkat Ullah 62 
Sayyid Didar 'Ali Alwari 112 
Sayyids 31,62,88 
Schimmel, A. 96,97 
Shafi'i school of law 70, 108 
Shah 'Abd ul-Aziz 25,28,29-31 
Shah 'Abd ul-Qadir 3 1 
Shah Abu'l Husain Ahmad see Nuri 

Shah Ale Rasul 89,91 

discipleship to 61—3,89 

ritual on deathdav 1 00—1 
Shah Ahmad Nurani 1 30 
Shah Wali Ullah Dehlawi 22-5, 36, 

Shaikh al-'Ulama 108 
shari'a 19,31,95, 125, 129 
Shari'at Ullah, Haji 33, 34 
Sharif 'Ali 73 
Sharif Husayn 73 
al-Shatibi 67 
Shaukat 'Ali 80 
Shi'ism 4,5,7,76,96 

attempts to reconcile with Sunnis 
25, 39 

and prophetic light 97 
shirk 32,42,91, 102, 134 
Shivaji 3 

Shuja ud-Dawla 4,6,7 
Spear, P. 5 

spiritual authority 94—5 
sufism x, 21,24-5,26, 31-2, 34, 
102, 128 

Ahl-e Sunnat and 12 3, 128 

Deobandis and 36 

and intercession 91—2 

reformist 124—5 

rejection of 21—2, 32, 38 
sunna 87, 118, 129, 134 
Sunni Islam 103,127 

and Ahmadiyya 46 

attempts to reconcile with Shi'ism 

Mughals and 1 

prophetic light 96—7 
Sunni law schools 23,37—8,39 

Tahir ul-Qadiri 130 

taqlid (submission to authority) 70, 

1 34 
tawhid (unity of God) 21,25,32,97 
textile industry 1 1 , 34 
Third Afghan War 82 
Tonk 52 

Trevelvan, Charles 1 2—1 3 
Troll, C.W. 42,43 
Tuhfa-e Hanafiyya 64-5 ,114-15,123 
Turks, support for 78,79,80,81, 


'Umar 25, 103 

unbelief 103, 104-9 

universities 16, 136 

United Kingdom, Ahl-e Sunnat in 

Urdu 20, 32,54-5 
\irs (deathdav celebrations) 89, 100—1 

of Ahmad Riza 129-30 
'Uthman 25, 103, 119 

voluntary associations 115—16 

Wahhabi movement 21-2,34,38, 

39,69,73, 105 
Wahhabis, movements designated bv 

Ahmad Riza 105-7, 131 
Welleslev, Richard, 1 st Marquis 15 
World Islamic Mission 77,131 

Yusuf 'Ali Khan 54 

Zaman,M.Q. 14,27,37,41,130 
zamindars 1