CAMBRIDGE SOUTH ASIAN STUDIES

THE EMERGENCE OF PROVINCIAL POLITICS

The Madras Presidency 1870—1920
 CAMBRIDGE SOUTH ASIAN STUDIES

These monographs are published by the Syndics of Cambridge

University Press in association with the Cambridge University
Centre for South Asian Studies. The following books have been
published in this series:

1 S. GOPAL: British Policy in India, 185 8-1905

2 J. A. B. PALMER: The Mutiny Outbreak at Meerut in 1857
3 A. DAS GUPTA: Malabar in Asian Trade, 1740-1800
4 G. OBEYESEKERE: Land Tenure in Village Ceylon
5 H. L. ERDMAN: The Swatantra Party and Indian Conservatism
6 S. N. MUKHERJEE: Sir William Jones: A Study in Eighteenth-
Century British Attitudes to India
7 ABDUL MAJED KHAN: The Transition in Bengal, 1756-1775:
A Study of Saiyid Muhammad Reza Khan
8 RADHA SHYAM RUNGTA: The Rise of Business Corporations
in India, 1851—1900
9 PAMELA NIGHTINGALE: Trade and Empire in Western India,
1784-1806
10 AMIYA KUMAR BAGCHI: Private Investment in India, 1900-
1939-
11 JUDITH M. BROWN: Gandhi's Rise to Power: Indian Politics,
1915—1922
12 MARY C. CARRAS: The Dynamics of Indian Political Factions
13 P. HARDY: The Muslims of British India
14 GORDON JOHNSON: Provincial Politics and Indian Nationalism
15 MARGUERITE S. ROBINSON: Political Structure in a Changing
Sinhalese Village
16 FRANCIS ROBINSON: Separation among Indian Muslims: The
Politics of the United Provinces' Muslims, i860—1923
17 CHRISTOPHER JOHN BAKER: The Politics of South India 1920-
1937
 THE EMERGENCE
OF PROVINCIAL POLITICS
The Madras Presidency
1870-1920

D. A. WASHBROOK
Lecturer in History, University of Warwick

CAMBRIDGE UNIVERSITY PRESS

CAMBRIDGE
LONDON • NEW YORK • MELBOURNE
 CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo

Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org
Information on this title: www.cambridge.org/9780521209823

© Cambridge University Press 1976

This publication is in copyright. Subject to statutory exception

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

First published 1976

This digitally printed version 2008

A catalogue record for this publication is available from the British Library

Library of Congress Cataloguing in Publication data

Washbrook, D. A.
The emergence of provincial politics.
(Cambridge South Asian studies; no. 18)
Bibliography: p.
Includes index.
1. Madras (Presidency) - Politics and government.
I. Title. II. Series.
DS485.M28W37 320.9'54'8203 75-36292

ISBN 978-0-521-20982-3 hardback

ISBN 978-0-521-05345-7 paperback
FOR S.T.W., A.C.W.
AND S.W.
 CONTENTS

i page IX
Abbreviations, notes on references
and spelling x
Introduction i
1 The Madras Presidency n
2 The governance of Madras 23
3 The political economy of Madras 64
4 Local structures of political power 146
5 The emergence of provincial politics 215
6 The vocabulary of communal politics 261
7 The Home Rule League, Justice
Party and Congress 288
Conclusion 330
Glossary 336
Bibliography 338
Index 351
 PREFACE

This book has evolved out of a fellowship dissertation presented at

Trinity College, Cambridge in 1971 and a Ph.d. dissertation presented
at Cambridge University in 1974. The research was financed by a Pre-
research linguistic studentship (1969—70) and a Research Fellowship
(1971-5) from Trinity College, Cambridge, and by a Hayter Student-
ship (1970-1) from the Department of Education and Science.
In the course of my research, I have incurred enormous debts to
a great many people. Firstly, I would like to thank, for their time and
trouble in finding material for me, the directors and staffs of the
Cambridge University Library, the India Office Library (London), the
Madras University Library, the National Archives of India (New
Delhi), the Nehru Memorial Museum Library (New Delhi), the Tamil
Nadu Archives (Madras) and the Theosophical Society Archive and
Library (Adyar, Madras).
I also owe a great deal to the many friends and colleagues who have
given their time to read and comment on this manuscript in one or
other (and usually more than one) of its many avatars. In particular,
I would like to thank Dr Christopher Bayly, Dr Carolyn Elliott, Pro-
fessor John Gallagher, Dr Gordon Johnson, Dr John Leonard, Mr
Peter Musgrave, Dr Tapan Raychaudhuri, Dr Francis Robinson,
Professor Eric Stokes, Ms Lucy Carroll and Dr Brian Tomlinson for
their help and kindness.
My greatest debt, however, is to the three people with whom I have
worked most closely and but for whose encouragement the following
pages would certainly have remained blank. Dr Christopher Baker has
shared many of my South Indian and English hours and has given
generously of his own allied material and ideas. Dr Anil Seal, my
research supervisor, has been a major source of strength and inspir-
ation since my undergraduate days. My wife, Angela, has enabled me
to survive the unconscionably long period of this book's gestation.
Needless to say, however, responsibility for the sentiments and
arguments of this book is mine.
Coventry D. A. Washbrook
June 1975
ix
 ABBREVIATIONS, NOTES ON
REFERENCES AND SPELLING

A.I.C.C. All India Congress Committee

C.L.A. Central Legislative Assembly
D.C.C. District Congress Committee
I.N.C. Indian National Congress
I.O.L. India Office Library, London
L and M, L. Local and Municipal Department, Local, or Municipal,
or M, or Leg. or Legislative Branches
M.D.G. Madras District Gazetteer
M.L.A. Member of the Legislative Assembly
N.A.I. National Archives of India, New Delhi
N.M.M.L. Nehru Memorial Museum Library, New Delhi
P.C.C. Provincial Congress Committee
P&J Public and Judicial Department, India Office
P.P. Parliamentary Paper
R.N.P. Reports on the Native Press (see Bibliography)
S.A.H. State Archives, Hyderabad
S.I.L.F. South Indian Liberal Federation
S.I.P.A. South Indian People's Association
T.N.A. Tamil Nadu Archives, Madras
T.S.A. Theosophical Society Archives, Adyar (Madras)
References to the proceedings of the Government of Madras are given as follows:
Government] O[rder], Number, Department, Date, Place (T.N.A. or S.A.H.).
References to the proceedings of the Government of India are given as follows:
Department, Branch, Date, File Number, Place (N.A.I.).
Note on spelling: All proper names are given in the spelling which was most com-
mon at the time.
 Introduction

In 1917, the political life of Southern India appeared to undergo

a massive transformation. Madras, known for thirty years as the most
'benighted' and conservative of the presidencies, suddenly exploded
into political activity. The Home Rule League, organised from a sub-
urb of Madras city, confronted the British with the most serious and
largest movement of dissidence which their rule had faced anywhere
in India since the Mutiny. The non-Brahman movement, also organ-
ised from the provincial capital, spread a wave of racial hatred across
the presidency and threatened to tear Southern society apart into
mutually antagonistic political communities. Both the Home Rule
League agitation and the non-Brahman movement represented
extremely new phenomena in Madras politics. Contemporaries never
tired of pointing out how, just five years before they appeared, there
was not the slightest sign of their imminence.1 And even a casual
glance at the political debates and postures of 1912 would support this
conclusion. In that year, the men (and women) who were to lead the
Home Rule League were recognised generally as the most loyal sup-
porters of the British raj',2 the later arch-ideologue of the non-Brah-
man cause was presenting to a Parliamentary Commission evidence
which not even his enemies considered to show a trace of communal
bias;3 and the provincial government of Madras was steadfastly deny-
ing to its superiors in London and New Delhi the existence of any-
thing resembling communal conflict within its territories.4 One of the
main purposes of this book is to explain how and why the novel issues
of politics in the years 1917 to 1920 arose.
But it was not just the issues of political controversy at this time
which were so new: it was also the forms of expression, of 'agitation'
and 'movement', and of political organisation. In 1912, campaigns to
unite the whole of the province against its British rulers or to unite
the millions of non-Brahmans against their Brahman 'oppressors' had
1
See, for example, E. S. Montagu, An Indian Diary (London, 1930), pp. 113—14.
2
Mrs Annie Besant, C. P. Ramaswami Iyer, and G. A. Natesan, all of whom had been
instrumental in defeating the Extremist challenge of 1906-9.
3
T. M. Nair. See Hindu 5 April 1913.
4
See G. O. 916 (L and M,L) dated 12 July 1911. T.N.A.
 palpur

MADRAS PRESIDENCY 1910

Scale'.

0 100 200 miles

I J
I
0 100 200 300 kms

Trivandrurt
 The emergence of provincial politics

not appeared, simply because nobody had as yet thought of them.

More fundamentally, they did not exist because there was no place for
them in the political structure of the period. In the context of the in-
stitutions, forces and interests of 1912, campaigns which set out to
create ideological bonds of their species would have been politically
meaningless to most contemporaries. The linkages of interest and pur-
pose between different levels of the political system, the means by
which the ideologies were communicated to and understood by the
political actors and the circumstances which made apparent ideol-
ogical conflict a crucial element in the political system, all of which
conditioned the events of 1917 to 1920, were not present in 1912.
There were profound differences in the political structures of the
two adjacent periods, and the Home Rule and non-Brahman move-
ments were at least as much the products of new structures as they
were of new ideas. A second, and much larger, interest of this book is
to uncover the ways in which the political institutions and relation-
ships of Southern India were changing to produce the new forms of
politics. In order to do so, it will wander over a wide range of matters
and a broad time-span. It will examine the nature of change in many of
the institutions through which political power was exercised and it
will push back its inquiries to the point in time (the 1870s) from which
it sees the changes which were crucial to the events of 1917 to 1920
beginning to emerge.
Chapter 1 outlines briefly the main social, economic and political
contours of the Madras Presidency during the period of study.
Chapter 2 initiates the analysis of political change by investigating
the operations of government. It is natural that the concept of govern-
ment should form a major organising theme of this book, for the
powers and privileges of government did much to determine the
entire political structure. The Government of Fort St George was
much closer in type to the oriental despotisms which previously had
ruled South India than it was to the circumscribed and constitutional
governments of its European masters' homeland. It was not simply the
guarantor of the peace and security which would allow its subjects to
carry on with their own business but it directly ordered their lives in a
myriad of ways. It granted and denied the legitimacy of social and
political positions right down to the level of village society; it extracted
(and partially redistributed) a large proportion of the economic sur-
plus; it gave employment to hundreds of thousands of people in its
various departments; it worked as an economic entrepreneur, control-
ling vast commercial monopolies, building railways and digging
canals; and it helped to develop and, more significantly, control edu-
 Introduction

cational and cultural facilities and the learned professions. The

influence of government pervaded social life to a degree unknown
outside Asia. Its shape and organisation must be taken as the begin-
ning of any examination of the political system. Moreover, of course,
the British government of Madras was an alien, colonial govern-
ment whose structure and policies were determined, to a consider-
able extent, by events in other parts of the world. It was not simply
the reflection of a ruling class thrown up by indigenous economic
and social pressures. In consequence, when analysing the nature
and causes of political change in South India, government has to
be treated in part as an autonomous agent which was capable of
altering both itself and the forms of South Indian political life for
reasons which had nothing to do with South Indian society — it was an
exogenous factor in the politics of Madras. The purpose of Chapter 2
is to determine the locations in the governmental structure from
which effective power was exercised and to trace over time changes in
both the location and nature of this power.
Although extremely influential, however, government by no means
controlled directly the distribution of all the economic resources in
Madras nor did it possess a monopoly of all political authority. Indeed,
in many ways the British deliberately had weakened the position
which they had inherited from previous regimes and had permitted
power over several important social institutions to slip from their
hands. In consequence, in order to continue their government, the
British were obliged to rely on the support and co-operation of a
relatively small number of South Indians who exercised considerable
political influence independently of them. Chapter 3 examines the
institutions of 'indigenous' political organisation, which gave these
men their power, and attempts to assess the changes which were tak-
ing place within these institutions. As its main purpose is to explain
how order was maintained and how conflicts were settled, the chapter
concentrates particularly on problems of social control.
The organisation of Chapters 2 and 3 implies a dichotomy between
the governmental and the 'indigenous' or 'social' political systems.
Obviously, however, this dichotomy is artificial, for although govern-
mental and social power can be separated analytically, in reality they
were inextricably enmeshed. Chapter 4 attempts to put them together
again in order to show how their changing relationship, between 1870
and 1920, was beginning to produce new types of political construct.
As the argument of these chapters will have developed, a further
conceptual dichotomy will have emerged. A very clear difference is
noticeable between the principles on which 'state-' or 'provincial-
 The emergence of provincial politics

level' politics and on which 'local-level' politics were organised.

Chapters 2, 3 and 4 concentrate largely on the local level and try to
show how a variety of changes within it were producing connections
between the local and the provincial levels. Perhaps their most
important finding is that the virtually autonomous political arenas
of 1870 were becoming fused into a much more broadly based political
structure. Chapters 5 and 6 explore the nature of the provincial
political system and attempt to show how changes there were generat-
ing their own set of linkages to the locality. Political behaviour
at the provincial level was being modified heavily by the new relation-
ship to the localities. In Chapter 7, the argument reaches the events of
1917 to 1920 which it has set out to explain. The behaviour of the
central political actors is placed within the changing context of the
political system.

Several questions are raised by the writing of this book: Why a

regional study; why this region; why this time and these events? The
author's answer would be that he believes that, in the present state of
Indian historiography, a regional study of Madras in the later nine-
teenth and early twentieth centuries can improve our knowledge of
both modern Indian and British colonial political history. By select-
ing a region, an author does not have to pretend that he has chosen
an entity which exists in complete isolation from everything outside
it. Clearly, for example, Madras was part of a larger country, of a
greater civilisation and of an international empire. The influences
which stem from these wider contacts need not be excluded from
study. Indeed, in this book they form a major area of investigation.
But if concentration is placed only on the higher and larger categories
of activity and existence, many of the nuances of historical develop-
ment will be missed and a great many events will remain inexplic-
able. Bengal, for example, shared the relationships of Madras to
country, civilisation and empire but its reaction to those relation-
ships was very different and its political history notably distinct.
There was a level at which the character of the social and political
institutions of Madras and Bengal diverged sharply. The regional
study, when compared to other regional studies, helps us to see the
depth at which this level lay and the materials of which the different
levels were composed. The regional study facilitates our under-
standing of the points of contact between greater and lesser institu-
tions and hence our understanding of the processes of politics and of
political change. As British colonial and Indian history contain so
many points of contact between larger and smaller socio-political
structures, it must aid our appreciation of them.
 Introduction

A casual glance at the library shelves suggests one reason why

South India is a peculiarly fitting subject for a regional study. In spite
of its size, wealth and contemporary importance, very little of its
colonial history has been written. It represents a vast gap on the
historical map of India. A second glance, this time at the contents
of the shelves, reveals a second and more substantial reason for its
selection. Many of the analytical tools which have been used to carve
the political history of modern India appear very blunt when they
are applied to the South. Unquestionably, for example, the Madras
Government was the most financially and bureaucratically oppressive
of all Indian provincial governments. Unquestionably too, its ad-
ministration and laws were designed to interfere more deeply in the
economic and social institutions of indigenous life than were those
of any of its neighbours. These factors ought to have made Madras a
turbulent and dangerous presidency. Yet so far from meeting with
the angry reactions of its subjects and having to face constant out-
breaks of hostility, Fort St George governed the quietest and, in
colonial terms, the most successful of provinces. Equally, Madras
was among the leading presidencies in the development of western
education which, we are used to being told, was the primary cause
of the development of an Indian nationalist movement. Yet South
India's overall performance in the nationalist stakes was, to put it
kindly, not unduly noticeable. However, on two occasions Madras did
break at the seams with nationalist fervour. For a few years in the
18 80s and again between 1917 and 1920, it was in the forefront of the
nationalist movement. But the people who took it there were certainly
not the western-educated alone and their interest in the movement
proved to be more pragmatic than intellectual or cultural. Moreover,
in Indian history, we read much about the politics of communal
conflict. Members of different castes, religions and races sometimes
are seen to belong to separate political communities which jockey
against one another for place, power and status. Before about 1915,
however, almost no significant event in British South Indian history
could possibly be understood in communal terms alone. And when
overt communal conflict appeared, it did so in the most remarkable of
forms. One community, representing 98 per cent of the population
and possessing the vast bulk of wealth and effective political power,
denounced another community, which consisted of less than two
per cent of the population and was possessed of nothing like the
same economic and political resources, for oppressing it. Superficially
examined with many of the tools which are supposed to be part of
the Indian historian's trade, South India appears to consist of nothing
but paradoxes. By a closer examination, we may not only reconcile
 The emergence of provincial politics

these paradoxes but also change our view on the utility of some of the
tools.
The significance of the years 1917 to 1920 has not been missed in
previous political studies of modern India. It was from this time that
the Indian national movement took on its mass persona, that com-
munal politics extended their appeal beyond tiny elites, and that the
history of the British in India becomes the history of a retreat. Few
historians would disagree that the road along which India would
march to Independence and Partition was opened in these years. Yet,
until recently, the historical reasons why the period should have been
so important have remained obscure. Historians have preferred to
see the wartime agitations as the beginning of 'modern' politics
and to proceed forward from them towards the later achievements
of nationalism and communalism. Naturally, this starting point
has limited their appreciation of the period. Most of their accounts
of the origins of the political movements of 1917—20 centre on such
features as the pressures of the war and the characters of Tilak, Mrs
Besant and Gandhi. While adequate on the nature of the issues which
divided political life, these accounts fall far short as explanations of
the qualitatively new forms of politics which were making the divi-
sions apparent. It was not just that people were expressing new views
but that they were acting politically in new ways. Demands made at
the capitals of government were earning the support and hostility of
men who previously had given no indication that they knew where
the capitals were. Provincial and national political conflicts were
concerned with a range of issues which had been very remote from
them before. By looking at the bed-rock institutions of politics and
by tracing the changes which had been taking place within them
over the previous forty years, it is possible to approach the problem
of 1917-20 from a more satisfactory angle. The Home Rule League
and non-Brahman movements may then be seen not as the semi-
inexplicable beginning of a chain of events but as the culmination of
a long-term process of change. Moreover, the new structure of
political relationships which had brought the movements into exis-
tence also may stand clearly revealed. It was the changes in structure,
which remained long after the sparks and fumes of agitational invec-
tive had dispersed, which more truly heralded the end of British
rule than did the mere issues of politics during the First World War.
If we allow the events of 1917—20 to have had an important role in
the evolution of modern India, then how much more important must
have been the processes of institutional change which began in the
1870s and which made those events?

8
 Introduction

The central concern of this book, therefore, is the changing

political system of South India between about 1870 and 1920. In
consequence, it may be necessary to warn the reader that the political
history which he will find here will not look like that which he is used
to finding in most of the historical literature on Indian politics at this
time. Indian political history has tended to concentrate heavily on the
analysis of particular political movements and agitations. This is
hardly surprising, for India's independence and partition — her
greatest modern events - were won through these movements and
agitations, and the historian, whatever his protestations, can but be
drawn to the past through the present. The questions which have
underlain this historical writing have been how did the Indian people
oppose and eventually defeat British imperialism and how and why
did the communal tensions, which colour South Asia's experience in
the twentieth century, arise. The perspective which it has adopted has
been designed to illuminate the changes in values, ideas and political
associations, which increased the hostility of Indians to the British
and to members of other races and religions.
The present book, however, is written from a different perspective
and seeks to uncover an essential but hitherto neglected area of South
Indian history. While it is certainly true that nationalism and com-
munalism were important political forces, it is also true that they
did not succeed in tearing India out of her imperial connection nor
in dividing the subcontinent into antagonistic ethnic and religious
communities until the later 1940s. In consequence, throughout the
period from 1870 to 1920 and beyond, there remained a viable
colonial political system through which power was distributed and
conflicts were contained and compromised. Indeed, it was only by
destroying this system that nationalism and communalism came to
achieve their later positions as dominant political themes. This book
has chosen its particular standpoint in order to illuminate the salient
features of the colonial political system. The questions which underlie
its inquiries are how did this system function, why did it operate
successfully for as long as it did and how and why did it change. Its
perspective facilitates particularly an analysis of the relationships
of power and authority in South Indian society: it examines changes
in the disposition of sanctions, the location offerees and the distribu-
tion of resources. Naturally, given its unusual angle of vision, when
the argument of this book touches nationalism and communalism,
it will be attempting to make points of a different character to those
which would be made from the other, more orthodox, viewpoint. It
will be obliged to treat those phenomena only as they related to the
 The emergence of provincial politics

political system and not as they stood apart from it: it will see them
only as reactions to and influences on its own subject matter.
The author's justification for asking his particular questions and
for following his particular approach is easily made. The triumph
of nationalism and communalism has for too long provided the
context of the present out of which the historian has drawn his
questions to the past. But this triumph itself is no longer a feature of
the present. India's contemporary problems and conflicts are the seed-
beds from which ideas for the development of Indian historiography
now must come. Of these problems and conflicts, as political scientists
and sociologists have not been slow to see, the nature and locations
of power in the political system have been crucial. Unfortunately,
however, our historical understanding of the organisation of political
power in the late colonial period, which the Indian Republic inherited
in 1947 and which has conditioned her situation ever since, is remark-
ably poor. So dominant in the historical literature have been the
themes of nationalism and communalism that we know almost nothing
about the Indian political structure and its development. Indeed, we
have less of the political fabric of the India of Their Majesties than
we do of that of the East India Company or even of the Grand Moghul.
This hiatus in understanding can be seen clearly in much of the
literature of political science, where 1947 has been taken as a cata-
clysmic year in which the whole basis of Indian political activity
switched from the high ideals of the movements for independence
and partition to the low practices of a struggle over the distribution
of power and resources. Without arguing that idealism had nothing to
do with nationalism and communalism, it must be obvious that the
struggle for power and resources in Indian society had been going on
long before 1947 and that, in part at least, it had determined the
course of nationalism and communalism. The political scientists'
wonderment at the effects of independence and partition owes more to
the deficiencies of the historical literature which they have been
obliged to receive than to the scale of the transformation actually
achieved by those events. It is to bring to the past some of the issues
raised by India's present, and hence to extend the process by which
the past can be historically understood, that this book has been
written.

10
 I
The Madras Presidency

By 1800, the British had acquired most of what was to become their
presidency of Madras. They found themselves in possession of a col-
lection of territories which covered about 140,000 square miles and
which, between 1870 and 1920, came to contain a population of some
30 to 40 millions. The province was certainly the most artificial of all
those held by the British in India. Its administrative and formal
political unity masked enormous economic, linguistic and cultural
diversities. The presidency was composed of seven clearly distinct
geophysical regions, each with its own economy. Around the Cauveri
and the Kistna-Godavari deltas, there were two areas of intensive
rice cultivation. From the Kistna and Godavari hills, south and west
through the Ceded Districts (Anantapur, Bellary, Kurnool and Cud-
dapah) and hinterland Tamilnad (North and South Arcot, Salem,
Coimbatore, Madura, Trichinopoly and Tinnevelly districts) stretched
a region which, though giving way to cattle-breeding on higher
ground and to rice cultivation along the banks of occasional rivers
and tanks, was mainly under systems of 'dry' cultivation. The basic
produce were grains of the combu, cholum and ragi varieties, and
cotton and groundnut cash crops - all grown without sophisticated
irrigation works. In the south-west and north-east of the presidency,
ranges of hills broke up the plains and plateaux: in the former area,
the Nilgiri and Palnad hills attracted European tea and coffee planters
who came to dominate the local economies; in the latter, the 'Agency
tracts' of Ganjam, Vizagapatam and Godavari districts continued to
be populated by tribal groups with only the most primitive forms of
settled agriculture. In the extreme south-west of the province lay
the coastal strip of Malabar, along which a heavy rainfall made
possible rice and coconut cultivation without recourse to extensive
irrigation works. The many regional economies of Madras, with their
own settlement patterns, crops, marketing arrangements and agri-
cultural seasons, naturally produced a variety of regional social and
political structures.1
1
See O. H. K. Spate, India and Pakistan: A General and Regional Geography (Lon-
don, 1954).

II
 The emergence of provincial politics

In Bengal and Bombay presidencies, some of the centrifugal ten-

dencies created by the diversity of local economies were corrected by
the growth of great commercial centres. Calcutta and Bombay city
handled between 95 and 99 per cent of their provinces' external trade
and developed as major financial and manufacturing complexes. Their
influence touched the economy of every district in their administrative
hinterlands. The long coastline of the Madras Presidency, however,
guaranteed that its capital could not develop a similar role. The
products of its regions could easily find their own local export outlets.
Harbours at Cocanada and, later, Vizagapatam, served the north-
east; at Negapatam, Danushkodi and Tuticorin the south-east and
south; at Calicut, Mangalore and Cochin the west. In spite of the
effects of early railway building which, by following administrative
and military rather than commercial convenience, had made it a route
centre, Madras city was no more than one, albeit the largest, port and
commercial centre among many. During our period it handled only
about 40 per cent of the presidency's external trade. 21 ts exports - cotton,
groundnut and hides - were drawn only from its immediate hinterland.
Even the European business houses, which had made the city their
headquarters, were rivalled by other European and Indian firms which
preferred to operate entirely from the various regional bases.
To add to economic disunities, there were five major linguistic
divisions in the presidency. In 1921 Tamil was the largest single first
language, spoken by about 17^ million people in the south and
south-east. Telugu was used nearly as widely, by about 16 million
people, mostly in the north-east. Approximately 3^ million Malayalam
speakers lived in the south-west, iJ? million Canarese speakers in the
north and north-west and i | million Oriya speakers in the extreme
north-east. In addition, Islamic and Maratha conquests had left
nearly a million Hindustani and a quarter million Marathi speakers
dotted around the province, while tribal groups in the hills continued
to speak a variety of tribal languages. 3 These divisions were further
exacerbated by radical differences in dialect and, more importantly
for mass communication, by the divorce of written and spoken Tamil
and Telugu in the three centuries prior to British rule.
The vast majority of the population was Hindu by religion. But this
should not be taken to mean that it followed the same pattern of wor-
ship or adhered to the same formal system of morals or recognised a
common religious authority. There was the usual theological division
2
Review of the Sea-borne Trade of the Madras Presidency (Madras, Annual Series,
1880 to 1920).
3
Census of India. Madras. 1921. Volume XIII. Part 2 (Madras, 1922), pp. 79-83.

12
 The Madras Presidency

between the worshippers of Siva and of Vishnu and a multiplicity of

further divisions between the many philosophical schools which had
grown up among each body of worshippers. Probably the most savage
religious controversy in the South was that between the Vadagalai
and Thengalai schools - or castes as they had become - of Vaishnav-
ism. More pertinently, perhaps, the main linguistic divisions also
mapped out major cultural divisions. Although, at some level of
abstraction, Tamils, Telugus and Malayan's may all be seen to be the
inheritors of a single religious tradition, their separate vernacular
cultures, as well as the sharply differing circumstances of their
societies, meant that in practice they tended to understand different
things by this tradition. The types of issue, over religious reform and
orthodoxy, which split intellectual society on the banks of the God-
avari, for example, were virtually unintelligible to educated Tamil
society along the Cauveri. 4 Similarly, the debate on the relationship
between Aryan and Dravidian cultures, which began to heat the Tamil
districts towards the end of our period, was viewed with a mixture of
incomprehension and incredulity in Andhra. 5 Of course, to point to
these divisions is not to argue that any other Indian province was
perfectly united in its religious culture. Obviously, every province
had its parochialisms. But some areas were better able to negotiate
these obstacles than were others. Across most of Northern India, for
example, the links of religio-cultural unity were extremely strong.
Political movements, such as the Arya Samaj, cow-protection and
Sanatan Dharm movements, which made use of these links, swept the
plains of Hindustan and its adjacent regions and found the means of
communicating their various messages to audiences from Gujerat to
Bengal.6 In the South, by contrast, the parochialisms proved too
great and no religio-political movement of the Arya Samaj, Sanatan
Dharm or any other type managed to bridge the gap between
vernacular cultures. Moreover, the bulk of the 'Hindu' population
of each vernacular region had never been drawn into the mainstream
of high theological Hinduism. In sharp contrast to Northern India,
the temple centres were isolated and had few social contacts with the
peoples of their surrounding villages, where Brahmans were few and
4
For example, K. Viresalingam's Widow Remarriage movement. See J. G. Leonard,
'Kandukuri Viresalingam: A Biography of an Indian Social Reformer 1848—1919',
unpublished Ph.D. dissertation, University of Wisconsin, 1970.
5
Literary anti-Brahmanism made only a very limited impact on the Telugu-speak-
ing districts.
6
See C. A. Bayly, T h e Development of Political Organisation in the Allahabad
Locality 1880-1925', unpublished D.Phil dissertation, University of Oxford,
1970.

13
 The emergence of provincial politics

the Vedas barely known. Except along the Cauveri and Kistna—
Godavari deltas and in coastal Malabar, the peasant population
tended to follow local and clan gods which were often throwbacks
to pre-Aryan tribal days.7 This, of course, further limited the scope
of religio-cultural contacts and movements.
In 1921, approximately 7 per cent of the population was Muslim,
mostly of the Sunni faith.8 By far the largest proportion of these
were low-caste converts to Islam — weavers and artisans in the Telugu
areas, petty merchants and traders in Tamil Nad and peasants in
Malabar - who were much nearer in language and customs to their
various Hindu neighbours than they were to the custodians of
Islamic culture. There were, however, a few Persian or Urdu Muslim
families in the presidency, who had been rulers and administrators
under previous regimes. Some of these, like the family of the Nawab
of Arcot, still held considerable lands and were attracted into govern-
ment service and the liberal professions. Others became important
merchants in Madras city and at Vanniyambadi, in Salem district,
which was one of the centres of Islam in the South. About 3 per cent
of the inhabitants of Madras were Christians of various kinds.9 The
largest concentration in the region was in Tinnevelly district where
about 10 per cent of the population followed the religion of Christ.
Most of these were low-caste Shanars and Bharatha fishermen but a
handful came from important peasant stock. The other main area of
Christian activity was in the north-east where again success was
obtained mostly at the bottom of the social scale.
In comparison to the United Provinces, where widespread religio-
cultural homogeneities underlay the boundaries of the province, or
Bengal and Bombay, in which economic linkages tied together the
various regions, the Madras Presidency was a peculiar collection of
autonomous territories. No Madrasi could speak his language to
half his fellows; particular institutions of religion and movements of
religious revivalism were relevant to only a small fraction of the popu-
lation; market connections integrated only small, regional groups of
producers and consumers. After independence, the heterogeneity of
the province was rationalised by its partition into five separate states.10
But, during our period, the presidency was an arbitrary conglomerate
held together only by the fiat of an imperial power.
7
See M.D.G., Francis, Madura (Madras, 1906), p. 84; F. W. Hemingway, Trichino-
poly (Madras, 1907), pp. 88-9; C. F. Brackenbury, Cuddapah (Madras, 1916),
pp. 64-5.
8
Census of India. 1921. Madras. Volume XIII. Part 2, pp. 27-9.
9
Ibid.
10
Orissa, Andhra, Tamilnad, Mysore and Kerala.
 The Madras Presidency

Lines of social, economic and cultural distinction, however, did

not only divide Madras into regional components. They also ran
through each region, separating one social group from another.
While, as in all societies, South Indians may have shared certain
social and cultural perceptions, as in all societies also, they did not
share the same social and political roles, the same recognised
behavioural models, the same boundaries of social action and the same
levels of material culture. The drawing of these lines of social defini-
tion is a difficult task and the questions of where they should be
placed and how precisely they must be marked depend very much on
the purpose which an author has in making them. Obviously, the
dichotomies which command the attention of the anthropologist are
not necessarily those which will catch the sight of the student of
political, economic or religious history; nor will the nib of the student
of a province or region need to be as fine as that of the student of a
village. Our purpose here is that of provincial political history and
consequently the clearest conceptual distinction which we have to
make, for it extends through the length and breadth of our subject
matter, is that between local-' and 'state-levels' of political culture.
This distinction will be seen to match partially, although by no means
wholly, those frequently made by anthropologists and culturologists of
South Indian between Left and Right hand castes, between the
Great and the Little Traditions and between Aryan and Dravidian
cultures. However, its intention is to facilitate the study of politics
rather than of religious culture, and so the phenomena which it seeks
to distinguish between are of a somewhat different order.
Social groups of 'local-level' political culture may be taken to be
those whose political orientation was almost entirely towards the
control and use of resources, particularly land and its labour, within
very restricted physical localities. By occupation, they tended to be
closely involved with the processes of agricultural production,
whether as the directors of farming operations, as farmers themselves,
as farm labourers or as the providers of a variety of menial services
to those who worked the land. Their commitment to land and the
locality was evidenced in a number of ways. Their kinship patterns
extended over only extremely small territories and were used to
reinforce economic or social positions within the locality.11 Their
religious worship emphasised heavily the gods of the locality or the
clan segment rather than those of Brahmanic Hinduism. They were
11
See, for examples, B. Beck, Peasant Society in Konku (Vancouver, 1972), pp. 2 2 9 -
32 on the Gounder Vellalas; E. K. Gough, 'Caste in a Tanjore Village', in E. Leach
(ed.), Aspects of Caste in South India, Ceylon andNorth-West Pakistan (Cambridge,
i960), pp. 45—6 on lower castes in Tanjore.

15
 The emergence of provincial politics

almost entirely illiterate and hence cut off from many of the channels
of wider social communication. In short, their political culture was
parochial and inward-looking and they paid scant attention to rela-
tionships which were not immediate or face-to-face.
Groups of 'state-level' political culture, by contrast, may be taken
to be those whose orientation was towards political activity in much
broader spatial categories - usually several districts but also, occasion-
ally, the cultural region and the cultural nation. By occupation, they
tended to be state administrators or members of the liberal profes-
sions, merchants, manufacturers in territorially organised handicraft
industries or the religious and artistic servants of those in these
occupations. They possessed a highly literate culture which supported
their supra-local transactional needs. Their kinship patterns em-
phasised the breadth rather than depth of their connections: mar-
riages were made to safeguard or pass resources along greater territorial
networks.12 Their religious worship centred on gods of the Hindu
pantheon and drew them into a relationship with higher all-Indian
theology. The reference points of their perception of the political
world were set altogether much wider and were much more likely to
encompass abstractions and generalities than were those of groups
of'local-level' political culture. 13
Few of the social mega-categories in which the British organised
their census data, and with which historians have tended to discuss
the social history of India, help us to put names to 'local-' and 'state-
level' groups. Caste is particularly useless: obviously most Brahmans
would be of state-level culture but all other large caste categories
would be split in half. The Velamas, Vellalas and Maravars, for
example, included great zamindar chiefs whose interests undoubtedly
lay at the state-level and hosts of petty cultivators and labourers who
were buried in parochialism. Equally, religion offers us little aid:
according to the census, all those who were neither Muslim nor
Christian nor Buddhist were classified as 'Hindu'. Even economic
categorisation can have its difficulties. The fact that local-level groups
were interested in the possession and control of land does not mean
that state-level groups were landless: in Tanjore and Malabar and in
most zamindari areas they held legal titles to land and sometimes
directed its cultivation. The difference between the two lay largely in
the activities which they used the resources of the land to pursue.
12
Ibid, on castes of the left hand and Brahmans respectively.
13
The dichotomy between state- and local-levels is brought out clearly in the dif-
ferent descriptions of social organisations among Brahmans and Komatis, on the
one hand, and Reddis, on the other, in N. Gopalakrishnamah Chetty, A Manual of
the Kurnool District in the Presidency of Madras (Madras, 1866).

16
 The Madras Presidency

The social parameters of the distinction become clearer, however,

if we turn to the history of South India. From the fourteenth century,
arguably from the seventh century, the South Indian political system
had centred on a series of warrior kingdoms, each seeking to establish
a rulership over parcels of land and the people who worked it, and
each seeking to participate in a higher, almost diplomatic, regional
political system which was conducted between warrior states.14 Even
the later Muslim conquests did not radically alter this situation for
the Muslim empires more or less contented themselves with holding
the ring around the higher regional system and leaving the warriors
to their own devices within their territories. Golconda's attempts to
accomplish more than this seem to have been short lived and to have
ended with the Moghul invasion;15 the great expectations of Hyder
Ali and Tipu Sultan were thwarted by constant harassment and
ultimately defeat at the hands of the East India Company.
The warriors themselves clearly required a variety of services from
other social groups if they were to succeed in their purposes as ter-
ritorial governments and as participants of the regional system. They
needed administrators, diplomats, financiers, merchants and artisans.
They also needed priests and poets who controlled the means of
(cultural) communication in the regional system. The social groups
which were drawn into providing these services naturally became
connected to the same politico-cultural framework as the warriors.
Their own destinies were bound up with the extension of warrior
government and with the continuation of the higher regional system.
This is not to say, of course, that their own interests were identical
to those of the warriors: administrative families, for example, would
tend to be more interested in the profits and security of administrators
than in the power of their employers; priests may have been more
concerned to honour the deities than to publicise their warrior
patrons. But, obviously, there was a point at which the interests of
the warriors and of the groups which served them coincided: adminis-
trators could not work if their warriors had no power and priests
could not live without patronage and protection. In consequence, the
social organisations of these 'service' groups developed around the
warrior states, supporting both their own needs and those of the
warriors. Administrative groups developed (often kinship) contacts
between court and village within the boundaries of a single warrior
14
See B. Stein, 'Integration of the Agrarian System of South India', in R. E.
Frykenberg (ed.), Land Control and Social Structure in Indian History (Madison,
Milwaukee and London, 1969), pp. 175-216; also, T. V. Mahalingam, Administra-
tion and Social Life under Vijayanagar (Madras, 1940).
15
See J. F. Richards, 'Mughal Rule in Golconda 1687-1724', unpublished Ph.D.
dissertation, University of California, 1970.

17
 The emergence of provincial politics

state and court-to-court contacts across the regional system.16 These

contacts acted as ties of government, as links of diplomacy and as
sinews holding a region together as a single politico-cultural unit.
Similarly, priestly groups (and temples) accepted a warrior's pat-
ronage, which took them out of his court and into the countryside
where they attested to his glory, or which enabled them to project his
image to other courts as a worthy and respectable prince.17 Merchants,
financiers and artisans also organised themselves in relation to the
warriors' needs both in his kingdom and in his wider world.
The whole of this warrior edifice, however, was imposed from
above on the variety of primitive tribes, agrarian settlements and
defeated warriorships which had inhabited the region before it was
built. These tribes and settlements, whether because they had not
developed beyond the stage of basic husbandry or whether, through
invasion and conquest, because they had been driven back to simple
cultivation, were primarily involved in the production of agricultural
commodities. The warriors, and the social groups associated with
them, lived by the expropriation of this production. In consequence,
there was a very clear dichotomy between the political purposes of the
'cultivators' and of the 'rulers': the one seeking to defend the wealth
which it had produced, the other seeking to take it. This dichotomy
suggests one reason for the local political orientation of the cultivators.
Obviously, the social organisations which facilitate the development
and protection of a resource such as land will be much more concerned
with the depth of local power than will those which facilitate an
extensive, expropriatory rulership. In many areas also, this distinction
of political purpose was overlaid by a wider cultural division. Over
much of Tamilnad, the warriors, brought down by the Vijayanagar
empire, were of a foreign Telugu culture and they had imported
many 'service' groups of their own. Telugu and Maratha Brahmans,
Komatis and Telugu Kammalas staffed the agencies of many warrior
governments. Moreover, even where the warriors were of the same
vernacular as their subjects, their own religious culture was often
very different. In most of southern Andhra and hinterland Tamilnad
16
This can be seen, for example, in the spread of Sri Vaishnava Brahman families,
brought into Tamil districts by the Telugu warriors of Vijayanagar. See R. E.
Frykenberg, Guntur District 1788-1848: A History of Local Influence and Central
Authority in South India (Oxford, 1965), pp. 13-17; Stein, 'Integration of the
Agrarian System of South India', p. 195
17
See G. W. Spencer, 'Religious Networks and Royal Influence in Eleventh Century
Tanjore' in Journal of the Economic and Social History of the Orient, xn, 45-56;
B. Stein, 'Economic Functions of a Medieval South Indian Temple' in Journal
of Asian Studies, ix: 2 (i960).

18
 The Madras Presidency

(the 'dry' zone), the subject peoples may best be described as Dravidian
in that their customs owed much more to pre- or non-Aryan tribal
forms than to the recognised 'codes' of pan-Indian Hinduism. The
regional cultures in which their rulers (and associated groups) parti-
cipated, however, were at least variants of pan-Indian Hinduism.
Cultural differentiation, perpetuated by political hostility, further
kept warriors and subjects apart. Tied to their localities in order to
protect their resources, lacking incentives to participate in the
alien politico-cultural system of the warriors, often never having
possessed a wider, territorial system of their own or having seen it
smashed by warrior government,18 these subject groups had few
reasons to be other than parochial in their political ambitions and
outlooks.
In political terms, the line between local- and state-level groups
has to be drawn at different points in different places. The extent to
which the state-level penetrated the locality depended on a variety
of factors, most of which we shall be discussing later. A brief survey
of South India, however, indicates that groups of state-level culture
dominated landed society only in Malabar, the central eastern sea-
board of Tamilnad and, to a much lesser extent, the Kistna—Godavari
deltas. Here not only had state-level groups gained control of agrarian
production by means of state action but their presence in great
strength had attracted into the state-level category many powerful
landed groups whose previous culture had been of the local level. The
leading Kallar families of Tanjore district, for example, were full
participants of state-level culture; Kallars over most of the rest of
South India were semi-tribals of the local-level.19 The line of division
in these areas would have to be drawn below the point of land control
and would confine to the local level a collection of slaves, labourers
and hereditary thieves who, nonetheless, formed the majority of the
18
Many of the local-level systems of South India were clan-based and so, at one
time, contained political institutions which were supra-local. However, the
impact of warrior government tended to break down most of these institutions:
the function of integrating dispersed settlements being performed by the warrior
court rather than the clan head. In southern Andhra, almost all relationship
between clan and local political power disappeared. In Konkunad, the relationship
was severely modified: state servants held land in clan territories to which they
did not belong; clan and lineage no longer clearly demarcated areas of political
control; uxorilocal residence (the death of a patrilineal clan political system)
became an unexceptional practice. Only in the deep South, where Vijayanagar
warriors established few kingdoms, did broad clan organisations continue to play
the political role which they possessed over most of North India at this time.
19
See L. Dumont, Une Sous-Caste de Vlnde du Sud: organisation sociale et religieuse
des Pramalai Kallar (Paris and The Hague, 1957).

19
 The emergence of provincial politics

population. Elsewhere in South India, in the hills of the north-east

and south-west and across the plateaux and plains of south Andhra
and hinterland Tamilnad (that is over most of South India), the line
would have to be placed very much higher. State-level culture was
more or less confined to the principal towns, which housed the
courts of the warriors, and to 'urban' groups. Its spread over the
countryside was extremely thin and it had little control over direct
landed power. Brahman priests and administrators, Vaishya trading
castes and literate Sat-Sudras were few in number and, consequently,
were unable to form the focal points of and the agents of direction in
local culture. To a considerable extent they remained outsiders, and
potentially hostile outsiders, to the social elements which dominated
the land.
In reality, of course, the dichotomy between state- and local-
levels was not as complete as we have pictured it. There was some
shading down from one to the other. Politically dominant groups of
the local-level perforce had some contact with the state which was
attempting to govern them. For example, they might be contracted by
their warrior prince to keep the peace and to help in revenue collec-
tion. Their internal political organisation could be altered if they
accepted this contract.20 In the Indian context, relationships of this
type were likely to leave religious and cultural as well as purely
political marks. Elements of the state-level might enter the culture
of dominant local groups and percolate through them to their sub-
ordinates, gradually becoming weaker and weaker until they dis-
appeared. In this sense, it is arguable that the state- and local-levels
were but different points of register on the same spectrum. Such an
admission would destroy the value of the distinction if our purposes
were anthropological: the concepts of state- and local-level would then
obscure more than they clarify. But, within South Indian political
history, the distinction still retains its usefulness.
In South India, much more than in the North, the great bulk of the
population fell clearly into either the state- or the local-level category
and the middle ground between them was very poorly populated.
Outside the ricelands of Malabar, Coromandel and Andhra, as we have
seen, local-level culture began at the point of land control and with
the elite which dominated agrarian production. As over 80 per cent
of the population was involved in the processes of the agrarian
economy, this would put at least 80 per cent of the population some-
where below the local-level line. Moreover, the cultural contacts
which linked the locally dominant 'peasant' elite to state-culture in
20
Beck, Peasant Society in Konku, pp. 40-9.

20
 The Madras Presidency

the first instance were extremely loose. In contrast to, say, Hindustan
where peasant elites participated directly in Brahmanic Hinduism,
by patronising village Brahmans, attending in person many of the
great festivals of the Sanskrit calendar and keeping records of their
familiar activities in the sacred centres of Hinduism, the participation
of South Indian peasant elites tended to be more indirect and much
more distant. Although many of their temples, their ceremonies and
their priests may have been related, at some remove, to authorities
in Brahmanic Hinduism, in local patterns of activity there were few
traces of this relationship. South Indian peasants in the 'dry zone',
for example, were seldom married by Brahman priests, seldom
attended major festivals or visited sacred pilgrimage sites (unless
those festivals and sites were in their immediate physical vicinity)
and seldom worshipped with Sanskritic forms of ritual. Even where
Brahmanic institutions had an immediate presence in the locality, they
did not play a central social role. Thus, for example, Brenda Beck
writes of a kiramam (revenue village) in Konkunad:
First and foremost, each kiramam has its own Civa temple Furthermore,
most kiramams have a Visnu temple, a KariyakaLiyamman (locally Pat-
tirakaLi) temple, and at least one temple dedicated to Murukan. All of these
are cared for by Brahmans, and left- and right [ caste] division groups wor-
ship at them. However, none of them is central to the religious activity
of the local population at large, especially to that of members of the right-
hand division. They serve rather to tie the NaTu to the leading deities of
literary Hinduism. Their presence satisfies formal requirements rather
than emotional ones, to judge from the mild popular interest which they
are generally accorded.21
Included in this right-hand division were the Gounder Vellalas,
who constituted the politically dominant peasant elite of the region,
and most of the subordinate agricultural community. In two recent
studies, the sharp differences between the politico-cultural orienta-
tions of peasant groups occupying very similar local economic and
political positions in Northern and in Southern India have been
highlighted. While what is striking about the Jats of Vilyatpur is the
breadth of the social context in which they operated, 22 what is striking
about the Gounders of Konkunad is the narrowness of the base of
their social activities. For the Gounders, actual rather than mythical
social relationships beyond the locality would seem to be rare. 23 If,
then, when studying North India it is the features of continuity
21
ibid., p. i n .
22
T. G. Kessinger, Vilyatpur 1848-1968 (Berkeley, Los Angeles and London, 1974).
23
Beck, Peasant Society in Konku.

21
 The emergence of provincial politics

between cultural levels which command attention, by the same pro-

cess, in South India it is the features of dislocation.
Yet not only is the dichotomy between state- and local-level
cultures very clear in most of South India, it is also of vital con-
sequence to the study of political history. The single most important
fulcrum around which South Indian political history revolves is that
of land control and the point at which the dichotomy becomes ap-
parent over most of South India is at that of land control. The groups
on both sides of the divide were locked in battle for control of the
same agrarian surplus. But, being in different relationships to the
modes of production and operating within very different politico-
cultural frameworks, the way in which they perceived and conducted
this battle, and the resources which they called in to aid them, were
very different. For one, victory meant closing off the locality to out-
side intrusion and exercising a deep and direct control; for the other,
it meant fully integrating the locality into a wider political system.
It is impossible to render intelligible the political history of nineteenth-
century South India unless this difference in purpose and meaning
is recognised.
The arrival of the British, of course, changed many of the rela-
tionships of the South Indian political system. Everywhere warriors
lost their military power and, where zamindaris were abolished, much
of their economic power as well. As Burton Stein has argued, the
British dismantled the machinery of the warrior states. They also
destroyed the higher regional system in which the warriors had
participated. Administrative groups had to accommodate themselves
to working for new masters within a new political context. Priests and
many types of artisans were cut adrift from the state. Merchants
and financiers were obliged to concentrate less on the needs of govern-
ments and more on those of the consumer population. The core
around which state-level cultural groups had organised themselves
for so long had been removed and replaced by one of a completely
different shape. Naturally their social organisations began to change
to fit the new circumstances. Nonetheless, the remains of the older
system, embodied in certain attitudes, deferences and cultural
interests, continued to exist throughout the nineteenth century and
beyond. Indeed, as the new system developed the dichotomy between
state- and local-levels was in some ways reinforced and given new and
even more important meanings. Reference to it will help to illuminate
our passage through some of the darker areas of South Indian social
and political history.

22
 2
The governance of Madras

'Government' was omnipresent in the life of colonial South India.

Whether we examine the newspapers, the letters, the autobio-
graphies, the pamphlets or the books of the period, repeatedly we
find references to the power, promise or peculiarities of the entity
known as government. The avaricious begged its favour, the
ambitious its confidence, the pious its protection and the national-
istic its self-destruction. By the standards of contemporary Europe,
Madras society was obsessed by the notion of government. If,
however, we try to define what precisely was meant by the term
'government' and of what social material it was made we encounter an
improbably complex series of problems. Government in Madras was
both a great deal more and a great deal less than the hundred or so
Europeans who composed its senior civil service. Involved in it, in
one way or another, were the British Parliament, sitting six thousand
miles away and concerned with the affairs of an international empire;
barely literate peasants, on salaries of four shillings a month and
concerned with the taxation of a few barren acres; and the broad
spectrum of people and interests which lay between them. Men whose
outlooks were bounded by the village, the town, the district, the
province, the nation and the empire all were locked together in the
chain of government in Madras. Depending on the position from
which it is viewed, therefore, 'government' can be seen to form many
different patterns and to be composed of many different substances
in different places. If we are to obtain any real idea of the nature of
this government and of its relationship to political society, our initial
task must be to take the chain to pieces and to assess the strengths and
weaknesses of its various links.
At its highest levels, Madras was governed by three separate
policy-making bodies: the Secretary-of-State-in-Council in London,
the Government of India in Calcutta (later Delhi) and the Govern-
ment of Fort St George in Madras city. Each of these institutions
pursued its own ends, judged the results it obtained by different
standards and overlapped scarcely at all in personnel. It is not surpris-
ing that the occasions of antagonism between the three were frequent

23
 The emergence of provincial politics

and bitter as each sought advantage at the expense of the others.

Every initiative coming from London and Calcutta was obstructed
by Fort St George which insisted that its superiors had no knowledge
of the problems of the Southern presidency. Ripon's local self-
government reforms, the 1892 and 1909 Councils Acts and, above
all, the constitutional proposals of Montagu and Chelmsford were
pushed through in Madras in spite of the opposition of Fort St
George.1 The allocation of taxes raised in Madras between Calcutta
and Fort St George was another point of contention.2 During most of
our period 65 to 70 per cent of the total income of Madras disappeared
into the Government of India's treasury, crippling the ability of the
local government to conduct any form of ambitious social or economic
policy.3 To the Government of India, Madras appeared a bottomless
purse which could be looted whenever need arose; and, as the political
situation in other provinces made it increasingly difficult to find
more revenue in them, need did arise with growing regularity. The
greed of the Government of India was of crucial importance in deter-
mining many of the characteristics of government in Madras: it
considerably limited the ability of Fort St George to pay for staff and
services out of a central provincial treasury and consequently forced
the provincial government to execute some of the most advanced
measures of financial decentralisation in India. Financial friction
between the provincial and national governments of Madras was not
merely a negative force but had a profound influence on the political
development of the presidency.
If we were to decide which of the several governments of Madras
was the most politically important in Madras, however, the answer
must be Fort St George. Except for the rare occasions when it altered
the framework of government by promoting major constitutional
reform, the influence of the Secretary of State and the Government
of India was felt in Madras in only two ways. London and Calcutta
possessed powers of veto on legislation and of demand on the revenue.
But they left Fort St George to conduct its own administration —
1
Report of the Committee on Local Self-Government in Madras 1882 (Madras, 1883),
p. 3; Home Public A. October 1908, Nos 116-46. N.A.I.; Letter No. 57, Home
Department (Reforms), Government of India to Secretary of State, in P & J
(Reforms), File 77. I.O.L.
2
See G. Slater, Southern India. Its Political and Economic Problems (London, 1936),
pp. 50-1.
3
Under the 1882 contract, the Government of India took 67.14 per cent of the
revenues of Madras; under the 1897 contract, it took 71.40 per cent. Lord
Ampthill to Lord George Hamilton, 11 June 1902. Ampthill Papers. I.O.L.
See also, Report of the Committee Appointed by the Secretary of State for India
on the Question of the Financial Relations between the Central and Provincial
Governments in India. P.P. 1920. Vol. xiv.

24
 The governance of Madras

to enforce the law and to collect taxation — which was much the most
potent of political tools. In general, provincial governments in India
were administering rather than legislating governments: they sought
to hold their positions and to promote such social reorganisation
as they saw fit through their bureaucracies rather than their legis-
latures. The assessment of land revenue, the distribution of govern-
ment jobs and contracts and the construction of irrigation works were
of far greater political weight than any paper statutes, while the
importation of the English legal system guaranteed that at least as
much law would be made in the courts as in the legislatures. Certainly,
from the later nineteenth century, there was an increasing tendency
to use legislation to solve social problems - as in the Punjab Land
Alienations Act, the Bombay Agricultural Indebtedness Act and the
various Bengal and U.P. tenancy laws. But it would be difficult to see
these laws, at least before 1920, as more than individual aberrations
in the political system, and most of them broke down in administrative
practice. If anything, Fort St George was the least aberrant of pro-
vincial governments and its legislative attempts at social engineering
amounted to very little. The Malabar Tenancy Acts of 1885 and 1898,
which affected only one area, failed to secure tenant rights in the
courts,4 and the much lauded Estates Lands Act of 1908, which
was thirty years in the writing, was designed to return tenancy law to
the state it had been in before a High Court decision of 1870 had
wrecked it. Moreover, Fort St George was paralysed from interfer-
ing in social and religious affairs by Calcutta's outright ban on efforts
to alter native custom. The Government of India's rights in legislative
matters, therefore, were not of much consequence when legislation
was not central to the way that government was conducted.
Equally, the large share of Madras revenues which the Govern-
ment of India sucked off was instrumental less in drawing attention
to itself than in turning it back towards Fort St George's adminis-
tration. The Government of India was committed to an enormous
expenditure on its army and had little surplus to redistribute to
the provinces. The Madras Government was committed to supply-
ing Calcutta's inordinate demands and also found little left over to
use in the development of the projects and interests of its subjects.
The centres of any spoils system of politics, therefore, had to be at
the points where revenue was collected rather than where it was
redistributed. Money, once it was allowed to become revenue, was
lost. Those who wished to divert public funds to their own pur-
4
D. Kumar, Land and Caste in South India. Agricultural Labour in the Madras
Presidency during the Nineteenth Century (Cambridge, 1965), p. 89; G.O. 2374
(Judicial) dated 1 October 1894. T.N.A.
 The emergence of provincial politics

poses, whether noble or ignoble, were best served by policies of

influencing the administrators who were gathering it. The Govern-
ment of India's financial gluttony helped to keep the village and
the tahsil as the most important political arenas in the presidency.
The government of Fort St George was, par excellence, an admin-
istering government, using its bureaucracy both to make and to
enforce policy. The centre of its administration was its revenue
department which raised the money without which it could not exist
and, in doing so, moulded native society by reorganising the major
resources of the economy. Three-quarters of the Madras Presidency
was under the ryotwari system of revenue administration which had
been devised by Colonels Read and Munro in the 1790s but much
reformed in the 1820s.5 The government stripped away intermediary
layers of authority between itself and the cultivating peasant and
undertook to measure and assess for land revenue every field in the
presidency. After the harvest, government officers toured their
charges collecting the taxation due and remitting payment on lands
on which the crops had failed.6 The task which government set
itself was mammoth, for it took officials into the very heart of village
India and gave them vast responsibilities. The proximity of revenue
administrators to the ground allowed, and to some extent forced,
the government to carry its other jurisdictions to an equally local
level. Revenue officials were also magistrates, the supervisors of the
machinery of local self-government and the informal agents of every
other department from excise to public works and from police to
agriculture. They possessed colossal powers to match their respon-
sibilities and everywhere formed the natural centres of local politics.
Of course, this system meant that the number of people employed in
the revenue department was huge, stretching from the departmental
Secretary in Madras city to hosts of minor stipendiary officials in
every village — a choking bureaucracy to be compared in the style of
its government to imperial Russia or the Prussian state.
Clearly, the way in which the bureaucracy ordered its powers pro-
vides an important clue to the way it worked in practice and to the
places within it where greatest power lay. The virtual omnicom-
petence of the revenue department meant that Fort St George could
not develop a system of departmental checks and balances whereby
officials in one department were circumscribed in their activities
by officials in another, who were subject to different superiors. The
divisions of authority in the Madras bureaucracy were essentially by
5
See N. Mukherjee, The Ryotwari System in Madras (Calcutta, 1962).
6
For a detailed description of the burden of revenue work, see Note signed F. A.
Nicholson in G.O. 173 (Revenue) dated 20 February 1902. T.N.A.

26
 The governance of Madras

territory and by the scope of primary jurisdiction. At the head of the

district administration was the Collector whose role was well sum-
marised by the Royal Commission on Agriculture in 1928:
The powers and duties of the Collector embrace almost every subject
which comes within the functions of modern government. In fact, in the
eyes of the cultivator, he is the supreme authority, the ma bap (literally
mother and father) who is expected to interest himself in all that affects
the well-being of the people under his control.7
He supervised the work of all the subordinates in his district and
was responsible in the first instance for cases of revenue and law en-
forcement which were of great importance. Beneath him were a
company of sub-, deputy-, and assistant collectors, each of whom
enjoyed similar overall responsibility in two or three taluks, super-
vised their subordinates and arbitrated in cases of moderate import-
ance. Below them came a cohort of tahsildars whose territory was
a single taluk and whose primary jurisdiction lay in matters of less
weight. Below them was a legion of gumastahs and revenue inspectors
who operated in circles of thirty or so villages. And, finally, below
them came an army of village officers.8 The district administration
consisted of a series of despotisms within despotisms from the village
to the district capital and beyond. The districts were linked to Fort
St George through the Board of Revenue and the Secretariat which in
turn were subject to the Government of India arid the Secretary of
State. An extraordinarily elaborate pyramid had been constructed
which guaranteed the passing of huge quantities of paper from one
layer to the next and the presence of intrigue at every stage.
Throughout our period the Madras bureaucracy was perched pre-
cariously on the edge of breakdown. For the administrative system to
work efficiently it was necessary that territorial charges should be
of manageable proportions, that the primary duties of superior officers
should not be so heavy that they were unable to supervise inferiors
and that information about the conduct of inferiors should be freely
available. At no time before 1920 were any of these requirements
remotely practicable. The basic territorial division, the revenue dis-
trict, was enormous and often contained a population of between 1^
and 2 millions.9 Territorial divisions below the district were equally
7
Report of the Royal Commission on Agriculture in India (H.M.S.O., 1928), xiv, 231.
8
The chain of command is laid out in W. S. Meyer, Report on the Constitution of
Additional Districts, Divisions and Taluks in the Madras Presidency, and on Other
Connected Matters (Madras, 1904).
9
Letter No. 1033 (Revenue) dated 13 September 1884. Government of Madras
(Revenue) to Government of India (Financial and Commercial) in P & J File 249
of 1885.1.O.L.

27
 The emergence of provincial politics

massive: a Rs 200 a month tahsildar might be the lord of 250

villages. The tahsildar, the deputy collector and the Collector were all
expected to spend a considerable period of time — in the case of
tahsildars as much as six months - away from their offices on tour. 10
Until 1885, Collectors also were supposed to conduct jamabundi in
their home taluks — a job considered full-time for any other officer.
In addition to their duties as revenue gatherers, magistrates, local
politicians and policemen, they had to spend hours filling in forms and
answering questionnaires for their superiors. The problem of provid-
ing effective supervision in their charges was insurmountable and was
exacerbated by the mountain of red-tape in which the British wrapped
their administrative practices. Each set of village revenue records
was kept in 24 separate accounts, one of which alone had 115 separate
columns.11 It was a life's work to check the addition and a genera-
tion's work to attempt to see how closely the result corresponded to
reality.12 The Secretariat was well aware of the difficulties inherent
in its own administration:
Under the present system of large divisional charges, however, Divisional
Officers, and for that matter Tahsildars, are often so tied to their desks as to
be not much more than post offices, and their subordinates are left with so
little supervision as to constitute a serious danger to public administration.13
Yet, even presupposing that superior officers had the time and
possessed the capabilities to watch over the activities of their inferiors,
it is by no means clear that they could obtain sufficient independent
information to interfere effectively. They relied upon their sub-
ordinates for information and, given the union of powers within the
administration, they had no independent sources upon which to draw.
Revenue accounts, for example, came up through the regular hierarchy
of the bureaucracy from the village to the district capital: if the
inferior officials of the department agreed that they were correct, the
Collector had no means and no outside evidence against which to
check them. In 1899, following the Shanar-Maravar riots, one of the
periodic breakdowns of government in Madras, Sir Denzil Ibbetson
10
Meyer, Report on the Constitution of Additional Districts, Divisions and Taluks in
the Madras Presidency, p. 2.
11
Evidence of J. N. Atkinson in Minutes of Evidence Taken before the Royal Commis-
sion upon Decentralization in India, n, 189. P.P. 1908. Vol. XLIV.
12
In one taluk in 1882, a tahsildar was responsible for issuing and collecting 12,870
forms; and, supposedly, for checking the veracity of the returns. J. H. Garstin,
Report on the Revision of the Revenue Establishments in the Madras Presidency
(Madras, 1883), p. 21.
13
G.O. 173 (Revenue) dated 20 February 1902. T.N.A.

28
 The governance of Madras

in the Government of India commented sourly of the Tinnevelly

Collector in whose charge they had occurred: 'Passage after passage
of Mr Scott's explanation shows him to imagine that the whole duty
of a district officer is to receive such information as his subordinates
choose to give him.' 14 Scott was no more than a classic adherent to
the Madras school. Of course, internal factional warfare in the lower
levels of the bureaucracy might throw up to senior officials informa-
tion which at least one faction would have preferred to leave buried.
Most successful prosecutions for corruption were based on evidence
so derived. Equally, specific and detailed investigations by the Board
of Revenue, as in Guntur in 1845, might bring to light specific
instances of governmental breakdown.15 But both checks came from
outside the regular and general system of administration.
The lack of co-ordination and control between the levels of govern-
ment in Madras was very obvious. Possibilities existed everywhere for
officials to practise their own private and discretionary government
regardless of the orders of their superiors and of policy emanating
from London, Calcutta and even Fort St George. Indeed, in the 1880s,
a number of scandals revealed that the problem of control began at
the connection between Fort St George itself and the Government of
India. The rules of service in the I.C.S. notwithstanding, a Chief
Secretary was found taking bribes to provide for his forthcoming
retirement;16 the First Member of the Governor's Executive Council
was caught in a lie to the Secretary of State and in land speculation;17
35 civilians, including the whole of the Board of Revenue and half of
the Governor's Executive Council, were discovered speculating in
plantation and gold shares in the native states;18 two leading officials,
D. Carmichael and W. Huddleston, were accused publicly of nepot-
ism;19 a Collector was dismissed for bribery20 and another just escaped
a trial for attempting to murder an enemy who was a member of the
14
Note signed D. Ibbetson dated 6 January 1900, in Home Police A, March 1900,
Nos 1-8. N.A.I.
15
See Frykenberg, Guntur District IJ88-1848.
16
M. E. Grant Duff to Sir Arthur Godley, 29 October 1883. Kilbracken Papers.
I.O.L.
17
'Case of H. E. Sullivan, Report of the Special Committee dated 25 November
1886' in P & J File 1886 of 1886. I.O.L.
18
'Copy of Extracts from the Correspondence respecting alleged Participation of
British Officials in Mysore Gold Mining Transactions and respecting Alienation
of State Domains in Mysore' in P & J File 1095 of 1887./.O.L.; 'Special Register
showing the particulars of land held by Civil Officers' in P & J File 1095 of 1887.
I.O.L.
19
Home Public A July 1884, Nos 260-5. N.A.I.
20
Hindu 28 M a y and 6 J u n e 1884.

29
 The emergence of provincial politics

Board of Revenue.21 To the utter dismay of the Secretary of State and

the Viceroy, Fort St George, rather than admitting contritely the
faults of its senior members, attempted in most cases to use its influ-
ence to prevent action being taken against the offenders, and only the
First Member and the corrupt Collector suffered for their sins.22
The major connection between Fort St George and its districts was
through the Collector's office. In most cases, the Collector was wholly
ignorant of the affairs of his charge. In addition to the impossible size
of his district and the workload he was expected to complete, he was
seldom allowed to remain in one post long enough to learn the names
of his subordinates. Madras' chronic shortage of I.C.S. officers forced
the civilian service into playing a grotesque game of musical offices
as it tried to fill in the gaps left by illness and vacation. Between 1880
and 1914, for example, there were 44 different postings to the Salem
Collectorate23 and 38 to that of Tinnevelly,24 while between 1880
and 1905, there were 27 to Madura.25 Where the average length of
posting was less than a year, the Collector had little chance to be
more than an administrative cypher.
Occasionally, however, through long acquaintance with a single
district or through the possession of personal interests in the area, a
Collector might come to acquire sufficient knowledge to do justice
to the powers of his office. The problem then faced by the Board of
Revenue was how to make him do its will rather than his own. One of
the classic cases of the independent raj of a Collector was that of C. S.
Crole in Madura in the 1880s. Much of Crole's early service had been
spent in the district and he was on friendly terms with the administra-
tors of the great Ramnad zamindari, having been taught Tamil by the
son of Subbayya Iyer, the estate's vakil.26 Subbayya already possessed
close contacts with the subordinate bureaucracy. One of his sons,
Ramaswami, was huzur sheristidar (head clerk) for a number of years,
while another, Subramania, had managed to acquire the public pro-
secutorship at the age of twenty, before he had even studied law, and
had become a tahsildar at twenty-six.27 The family was also involved
in temple and municipal politics in alliance with Kotaswami Thevar,
the 'prime minister' of the Ramnad zamindari and a cousin of the
21
The case concerned C. S. Crole, Collector of Madura, and J. H. Garstin, Member
of the Board of Revenue. P & J File 2057 of 1886. LO.L.
22
See P & J Files 1886 of 1886 and 1095 of 1887. LO.L.
23
M.D.G. F . J. Richards, Salem (Madras, 1918), 1, pt 2, 7 1 .
24
M.D.G. H . R. Pate, Tinnevelly (Madras, 1917), 1, 319.
25
M.D.G. Francis, Madura, 1, 209.
26
S. M . Raja R a m a Rao, Sir Subramania Aiyer. K.C.I.E. D.L. (Trichinopoly, 1914),
27
pp. 7, 1 0 - 1 1 . Ibid.

30
 The governance of Madras

zamindar.28 In 1882, Crole was made Collector of Madura and turned

over his administration to his friends. S. Subramania Iyer became
municipal vice-chairman;29 Kotaswami Thevar vice-president of the
Local Board;30Sankara, another son of Subbayya Iyer, obtained the
valuable receivership of the Sivaganga zamindari;31 R. Ramasubba
Iyer, Subbayya's nephew, became public prosecutor. Crole allowed
this clique to use the revenue department bureaucracy to collect
money for its private purposes, such as a Public Park and a Union
Club.32 He moved his subordinates round at its wishes33 and battled
with the Board of Revenue to lower the land revenue demand in its
interests.34 In many ways, he was the raja of Madura, making his own
policy and using his friends as his own administrators. He received
presents and bribes from all and sundry and treated the temple
elephants as his private property.35 He also conducted his own
'foreign' diplomacy: it was alleged, though never proved, that he
employed the zamindar of Bodayanayakanur, another of his local
allies, to ambush and kill J. H. Garstin, a member of the Board of
Revenue who was snooping into his affairs;36 and he was instrumental
in obtaining the dismissal of the First Member of the Governor's
Executive Council for attempting to force the Bodayanayakanur
zamindar to buy a tea plantation.37 Yet so weak was Fort St George
in controlling its districts that Crole escaped prosecution, as the mass
of evidence raised against him by an official inquiry was removed
from the Collector's office and destroyed by his allies.38 Indeed, a
decade later Crole was made a member of the Governor's Executive
Council and, shortly afterwards, S. Subramania Iyer (later Sir S.
Subramania Iyer) was nominated to a vacant High Court Judgeship.
Few other Collectors achieved Crole's depth of local interest but
many used their vast resources of patronage- Collectors were respon-
sible for all district appointments below the deputy collector cadre —
and their influence to push forward the careers of proteges and to aid
28 2
Ibid., pp. 1 2 - 1 3 . ° R>id., P- : 4 -
30
H . S. T h o m a s , Report on Mr Charles Stewart Crole. Collector of Madura ( M a d r a s ,
1886), p . 4 9 .
31
Hindu 7 M a r c h 1884.
32
T h o m a s , Report on Mr Charles Stewart Crole, pp. 2 - 5 .
33 34
Hindu 29 October 1885. Hindu 29 April 1885.
35
M.D.G. Francis, Madura, 1, 264.
36
P & J File 2057 of 1886. I.O.L.; G.O. 723 (Public) dated 5 April 1886. T.N.A.
37
P & J Files 701 a n d 707 of 1887 a n d File 1476 of 1 8 8 6 . 1 . O . L .
38
Letter No. 144 (Confidential), H . S. T h o m a s , Senior M e m b e r , Board of R e v e n u e ,
to C. F . Webster, Chief Secretary to t h e G o v e r n m e n t of Madras, in P & J File
189 of 1886.1.O.L.
 The emergence of provincial politics

temporarily favoured social groups. Sometimes these activities might

appear to be part of a farce, as when Sir Henry Montgomery appointed
his butler, Muthuswami Naicker, a tahsildar,39 but at other times
they were deadly serious. Many of the communal rumblings which
were reverberating through the presidency during our period can be
traced to the independent service policies of district revenue officials.
In Salem in 1882, for example, the Collector's avowed sympathy with
and support of Muslims was a contributing factor to religious rioting.
Collector Macleane, appointed only the year before, was an Urdu
scholar with many friends among the large Islamic population of the
Ceded Districts where he had spent most of his service.40 When he
moved to Salem he brought with him a number of Muslim acquain-
tances whom he placed in favoured positions in the local bureaucracy.41
Already there was considerable tension in Salem between Muslim
merchants, who recently had raised a mosque in a predominantly
Hindu area, and Komati trading rivals. Macleane's arrival sparked
off a series of anti-Muslim riots in which several people were killed.
Local Hindu officials, whose prospects had been damaged, connived
at the Komatis' plot and kept Macleane so ignorant of the dangers
that he left Salem for the Bangalore races on the day that the second
and most serious riot took place.42 Policies to prevent communal
conflict, framed by the Secretariat and Board of Revenue, had little
relevance while immediate district subordinates were in a position to
ignore them and supplant them with their own.
Beneath and, more often than not, alongside the Collector, other
government servants also carried on their own discretionary rule.
Even elementary information about the nature and extent of cultiva-
tion in the presidency was poor: surveys made in the 1820s and 1830s
were not re-examined until the 1870s and 1880s and the cumbersome
machinery of jamabundi guaranteed that many obvious facts could be
hidden. Everywhere government subordinates, often at the lowest
village and tahsil levels, usurped the powers of the revenue settlement
officer in order to alter demands and payments. In 1 8 8 6 - 7 , t n e re~
39
Indian Statesmen, Dewans and Prime Ministers ( M a d r a s , n.d.), p . 1 9 5 .
40
Macleane was regarded in the Secretariat as an expert on Islam. He was made
president of the Madras Mohammedan Education Endowments Committee in
1886 and sent to attend the Calcutta session of the Central Mohammedan Asso-
ciation. G.O. 12 (Education) dated 7 January 1886; G.O. 539 (Education) dated
22 July 1886. T.N.A.
41
'Mr Macleane in introducing into his own office many of his favourite Maho-
medans in the place of many innocent clerks . . . ' Further, one of his men, 'Kadir
Bahadur Sahib brought in many of his own men from Vellore and Krishnagiri
and distributed them as Karnams and Munsiffs.' Hindu 28 November 1883.
42
G.O. 353 (Judicial) dated 10 February 1883. T.N.A.

32
 The governance of Madras

settlement of the Nilgiri Wynaad raised revenue payments in some

areas by as much as 10,000 per cent:
These startling increases were explained to be chiefly due to the great extent
of concealed cultivation which had been brought to light, [and] to the
manner in which Government demand had been whittled down by the lower
subordinates.43
A few years earlier, the resettlement officer in the Godavari district,
at the other end of the presidency, described the cultivation changes
that had occurred in the previous forty years, 'as far as can be ascer-
tained':
I say as far as can be ascertained, for it is impossible to find out with any
accuracy the rates really paid for certain descriptions of land. In the first
place the areas are incorrect. . . . So much dry land is entered as wet; so much
wet as dry; there has been such continual readjustments of the shist on the
whole rented area; so many additions have been made to the joint-rent in
the lump, which additions have afterwards been distributed in thefieldmore
with reference and circumstances of the positions of the Ryots than to the
value of the fields.44
The 1905 South Kanara resettlement raised the assessment by 150
per cent in one taluk and by 64 per cent overall, due to the discovery
of concealed cultivation and misclassification of land;45 the mid-1870s
resettlement of Salem raised the assessment by 4 per cent for the same
reasons;46 the Madura resettlement of 1885 found that, although the
total revenue demand was correct, much of it had been shifted
gradually from the rich wet to the poorer dry lands.47 Sometimes
whole villages appeared as from nowhere in the revenue accounts,48
while others disappeared no less suddenly. In 1870, one South Arcot
revenue officer found thirty-five villages paying a rent reduced by a
Board of Revenue rule which had been obsolete for forty years.49
The lower levels of the revenue department were riddled with
bribery and corruption. One member of the Board of Revenue officially
estimated that 20 per cent of the revenue remissions granted every
M.D.G. W. Francis, TheNilgiris (Madras, 1908), 1, 281.
Selections from the Madras Records. XXII (Madras, 1870), p. 18.
M.D.G. J. F. Hall, South Kanara (Madras, 1938), n, 28.
M.D.G. Richards, Salem, 1 part 2, 35-6.
7
M.D.G. Francis, Madura, 1, 203-4.
48
For example, 23 villages 'appeared' in Coimbatore in 1870. 'Report onCoimba-
tore', p. 8 in Report on the Settlement of the Land Revenue in the Districts of the
Madras Presidency for Fasli 1280 (1870-1) (Madras, 1871). Hereafter this annual
series as Land Revenue. . . .
49
'Report on South Arcot', p. 2 3 in ibid.

33
 The emergence of provincial politics

year were the result of outright fraud which could never be detected
or stopped.50 In 1902, in a letter to the Secretary of State, Lord
Ampthill, the Governor, neatly summarised the technique of revenue-
gathering current in his presidency:
What happens is this: All the lands on which the crops have failed have to be
inspected by subordinate agency which, as you know is very amenable to
bribery in this country. The consequence is that the well-to-do ryot who
can afford to bribe the village officers or the revenue inspectors gets them to
report that his crops are withered or totally lost so as to entitle him to
remission. . . . Again it is by no means infrequent that the remissions granted
never reach the ryots for whom they are intended as the village officers
deceive the ryots by telling them that no remissions were granted collect the
full assessment and pocket the money themselves.51
Ampthill was not the only senior official to be disarmingly frank about
the administration. In 1908 J. A. Cumming, the Collector of Godavari,
shocked the Whitehall and Calcutta members of the Decentralization
Commission by telling them: 'No doubt there is a great deal of corrup-
tion goes [sic] on in the tahsildar's office, and there is a great deal of
corruption goes [sic] on in my office.'52 But Fort St George was
unlikely to be able to effect sweeping changes through the mere
consciousness of its problems.
The independence of the official was not the prerogative of the
revenue department alone; the many other departments of govern-
ment had followed in its wake deep into the village and were closely
related to it in personnel and activities. In the Andhra deltas, Public
Works Department men controlled the supply of piped water, which
was crucial to second cropping, and were notoriously corrupt.53 The
Excise Department operated an individual tax on each palmyra tree
and gave its lower subordinates ample opportunity to develop private
administrations, for supervision was impossible. The police were in a
similar state: the Superintendent of Tanjore reported in 1883:
Bribery is rife in the District so much so that it is only surprising that more
cases are not reported against the Police. When such cases do occur it is
with the greatest difficulty that any really reliable evidence can be obtained.54
50
H. S. T h o m a s , Report on Tanjore Remissions inFasli 1294 (<A.D. 1884—5) (Madras,
1885), pp. 45-6.
51
Lord Ampthill to Lord George Hamilton, 6 August 1902. Ampthill Papers. /. O.L.
52
Evidence o f J. A. Cumming in Minutes of Evidence taken before the Royal Com-
mission upon Decentralization in India. 11, 293. P. P. 1908. Vol. XLIV.
53
See N . G. R a n g a , Fight for Freedom (New Delhi, 1968), pp. 8 - 9 .
54
Administration Report of the Madras Police for the year 1883 (Madras, 1884), p. 9 1 .
Hereafter this annual series as Madras Police . . .

34
 The governance of Madras

Twenty years later, Lord Curzon appointed a commission to examine

the working of the Madras police. It found: 'Dishonesty in investi-
gation is, we are told, prevalent everywhere.'55 Nor was judicial
administration any better. In 1886, the Collector of North Arcot
reported:
Holy places like Tripati [sic], where there is an abundance of other people's
money in the hands of priests and where debauchery, intrigue and extortion
are chronically rife, prove the ruin of most of the native subordinates who
serve in them. Next to these holy places come Zamindary. Of both North
Arcot has a large share, and to this is doubtless partly due the lamentably
low standard of official morality that exists among our subordinate Magis-
tracy. . . . There are three or four Sub-Magistrates in this district who rarely
do an official act save on corrupt inducement.56
While the Madras government pretended to exercise such vast
powers - far greater in theory than any other provincial government -
without the information or the machinery to fulfil the duties they
implied, the development of interlocking systems of private and dis-
cretionary rule by its subordinates was inevitable. Village officers,
clerks, revenue and police inspectors and tahsildars were of far greater
consequence in deciding the vital questions of who paid what and
whose grievances were redressed than the Governor or departmental
Secretaries. Of course, such an administrative system, if it can be so
called, would tend to border on the anarchic as the authority of every
official would be countermanded by his subordinates. To some extent
this was the case: ultimate authority always lay at the lowest possible
level. Yet certain factors within district administrations led to the
growth of channels of influence and authority between the layers of
the bureaucracy, which amounted to structures of government
although they had nothing to do with Fort St George, and seldom
anything to do with the Collector.
R. E. Frykenberg, in his absorbing study of Guntur district in the
first half of the nineteenth century, has shown how groups of Maratha
Brahmans, through their control of information and appointments,
were able to recruit and set up a virtually parallel government inside
the British bureaucracy.57 By the later nineteenth century, the excesses
of such private systems had been curbed: administrative reforms,
improved information following the resettlements of revenue in the
1870s and 1880s and the growth of a scandal-hungry press made it
55
Statement of the Police Committee on the Administration of the District Police in the
Madras Presidency (Madras, 1902), p . 50.
56
Madras Police 1886. A p p e n d i x C , p . v.
57
Frykenberg, Guntur District 1788-1848.

35
 The emergence of provincial politics

more difficult for secret combinations to extend as widely and to

operate for as long as those in Frykenberg's Guntur. Yet the basic
causes which had led to the rise of parallel administrations were not
treated, and the organised bureaucratic clique remained an important
feature of government in Madras.
Below the cadre of deputy-collector, of which there were only fifty
members in 1880, there was no regular provision for the movement
of government officials between the districts. Most were recruited and
served to retirement within the same district and, if they early became
a tahsildar or huzur sheristidar^ often spent many years in the same
post.58 As late as 1920, 70 per cent of Madras tahsildars were serving
in the districts in which they had been recruited fifteen or twenty
years before.59 Tahsildars and huzur sheristidars^ with their very
considerable powers, were given every chance to acquire knowledge
and interests in their charges. The agency which supervised them,
however, possessed no similar opportunity. Yet it was this agency,
and the Collector in particular, who possessed the power of appoint-
ment to all the non-provincial offices. It is not surprising that most
Collectors were under the thumb of their permanent establish-
ments:
In the distribution of patronage in the District - the appointments of
Tahsildars, Sub-Magistrates, Divisional Offices' Head Clerks, Revenue
Inspectors and numerous other appointments, be in the gift of the Collector.
The Collector is practically guided by his Sheristidar.60

The huzur sheristidar thus came to have great influence throughout

the revenue bureaucracy and could make or break the careers of a
great many officers beneath him in rank. He could use his position
to build up his own administration. Although family and caste were
not the only means by which the huzur sheristidar recruited, their
presence helps to give some idea of his scope. From Coimbatore in
1884 came the cry:

The office isfilledwith men who are all members of one and the same family.
These men fill the places of important offices such as Huzur Sheristidars,
Head Writers, Tahsildars, Sub-Magistrates, Taluk Sheristidars and Munshis
of Collectors, while men are shut up from promotion for years and years
owing to the influence of the family to which the present officials belong
58
F o r example, by 1884, G. N . Chinnathambi Pillai h a d been t h e tahsildar of
Tinnevelly taluk for fourteen consecutive years. Hindu 23 June 1884.
59
G.O. 1435 (Revenue) dated 17 June 1920. T.N.A.
60
Hindu 13 September 1894; see also Hindu 25 September 1901.

36
 The governance of Madras

with the Collectors. The members of one family may be seen in all the Taluks
and the Head-quarters of the district filling from the post of Karnam to the
place of Head Writer and Tahsildar.61
The Governor, M. E. Grant Duff, once wrote to Sir Arthur Godley:
'It may interest you to know that in two divisions of one Madras
District, Cuddapah, I heard there were over eighty members of one
Brahmin family in Government employ, forty in one and forty odd in
the other.' 62 In fact, Grant Duff underestimated the problem: D.
Krishna Rao, the Cuddapah huzur sheristidar had 117 relatives in the
district service; prior to his appointment in Cuddapah he had been
huzur sheristidar in neighbouring Anantapur and had brought in
another 108 there.63 Evidence from South Kanara, North Arcot,
Kistna, Godavari, Kurnool and Chingleput indicates that each also
had its bureaucratic caste clique.64
In certain, though rare, circumstances, the families who were
successful practioners of this sub-imperialism in the district and
tahsil offices could direct their own form of government right down
to the village level. Where the social groups which held higher bureau-
cratic posts were also those which held village posts, particularly those
of kurnam (accountant), and large quantities of land, the conditions
were created for centrally organised rural kingdoms to be established.
Through the discipline of caste institutions and through ties of
'primordial' sentiment, district bureaucrats could draw local powers
under their authority and dictate decisions to them. Conspiracies
based on these connections always were the most tightly welded. The
British, ever watchful of the ramifications of caste, endeavoured
to weaken the local role of the kurnam and to strengthen that of the
headman who was drawn from the major peasant castes which had
little presence in the higher bureaucracy. In areas where this move
achieved its ends (as it did generally), the domination of conspiracies
from above was ended. However, the conspiracies themselves certainly
did not cease to occur. All that happened was that peasant headmen
took a greater share in them and tilted the balance in the distribution
of power within them more towards local interests. The conspiracies
61
Hindu 22 F e b r u a r y 1884.
62
M. E. Grant Duff to Sir Arthur Godley, 31 July 1884. Kilbracken Papers. I.O.L.
63
Hindu 17 September 1892 and 4 May 1893; see also Anon., The Ways and Means
for the Amelioration of the Conditions of the Non-Brahman Races (Bangalore, 1893),
pp. 18-19.
64
Hindu 13 November 1893; Hindu 14 M a y 1897; Hindu 2 M a r c h 1887; Hindu 30
M a y 1887; Hindu 22 J u n e 1895. Also, Andhrapatrika 2 D e c e m b e r 1918 a n d
Kistnapatrika 21 F e b r u a r y 1920. R.N.P.

37
 The emergence of provincial politics

were bound to continue while British supervision of the administra-

tion remained weak and while Indian bureaucrats and local land-
owners could still find mutual advantage in altering revenue payments
to suit themselves.
Several of the most noteworthy cliques uncovered in our period
were entirely the result of non-kinship, 'professional' contacts and
the balancing of reciprocal interests. In Vizagapatam in the 1880s,
for example, the Board of Revenue discovered a conspiracy in salt
manufacture worth many lakhs of rupees, which involved European
Assistant Salt Commissioners as well as native contractors.65 In
Palladum taluk of Coimbatore in 1907, a cross-communal network,
including the tahsildar, police inspectors and sub-registry clerks,
was found at work. Investigations showed that this clique extended
into the Collector's office where district records were rewritten at its
command.66
Perhaps the greatest conspiracy case of our period was that in
Tanjore in 1884, investigated by H. S. Thomas of the Board of
Revenue. It demonstrates the breadth and depth of such connections
and the inevitable failure of government attempts to break them up
simply by tinkering with the mechanisms of communal recruitment.
In the Tanjore remissions scandal, the participants included non-
Brahmans and Tamil and Maratha Brahmans, and several of the
principals were deputy collectors who, by the nature of their post,
had no personal ties with Tanjore.67 Heavy rains in the winter of 1883
had damaged severely the paddy crop in several parts of the district.
Following an inspection of the disaster area by revenue subordinates,
claims for a remission of revenue totalling Rs 8,21,900 were sent to
the Collector, Mr Pennington, who accepted them and passed them
on to the Board of Revenue in Madras. The Board, stung by the heavi-
ness of the claim, was stirred from its usual lethargy. It deputed as
investigator H. S. Thomas, a recent ex-collector of the district, who
had a deep knowledge of the area and many personal contacts in the
subordinate bureaucracy. His investigation revealed that Rs 4,07,322
of the claim were fraudulent:
This report places beyond doubt that there was a widespread and organised
conspiracy amongst those natives of the district, to whom the inspection of
the crops, etc. and the preparation of the accounts for the annual settlement
65
Proceedings of the Board of Revenue, No. 187 (Misc.) dated 24 January 1883.
T.N.A.
66
Hindu 11 M a y 1907.
67
G.O. 218 (Revenue) dated 22 M a r c h 1886. T.N.A.
 The governance of Madras

were entrusted, to defraud the Government of a sum almost equalling that

for which remission could be claimed.68
The brain behind the plot was that of the new huzur sheristidar, R.
Venkatarama Iyer, who had not yet been confirmed in his appoint-
ment. So carefully executed was his scheme that he had warned all his
subordinates to fill in their returns in pencil so that alterations might
be made with ease later.69 Involved with him were subordinates at all
levels:
Turning now to the subject of the conduct of the officials condemned by
Mr Thomas, Government have to observe that fraud was present at the
primary inspection; that it again, and in the worst form, characterised the
action of the officials, whose duty it was to check the results of the inspections
and to examine the accounts drawn up by the subordinate staff; and that,
as this fraud was not only winked at, but actually planned and arranged
beforehand by the higher officers, vis. the Deputy Collector, Acting Huzur
Sheristidar, sundry Tahsildars, Taluk Sheristidars, and the principal em-
ployees of the Collector's office, Mr Pennington was induced to accept as
true the misconceptions and false accounts and returns, which were placed
before him by those on whom he had every right to believe that he could
repose trust.70
Although the revelations of the Tanjore remissions investigation
appeared to come as a shock to the government, nobody in the Board
of Revenue or the districts of the presidency was the least surprised.
As H. S. Thomas said himself, the events in Tanjore were no more
than a particularly noisy orchestration of a theme being played in
every district all the time.71 If the Board of Revenue had not taken the
unwonted step of sending an investigator, and if that investigator had
been other than Thomas with his local connections, it is unlikely that
the sounds from Tanjore would have broken through the heavy proof-
ing which surrounded them. The enormous powers of the revenue
bureaucracy and the way in which they were structured promoted the
development of conspiratorial cliques between the village and the
higher offices of district government. That was how the administra-
tion worked. Tanjore was a paradigm of the practice of British admini-
strative theory.
Fort St George was not a government to concern itself much with
promoting rapid social or political reform but, had it been, the nature
68
G.O. 122 (Revenue) dated 15 February 1886. T.N.A.
69
G.O. 218 (Revenue) dated 22 March 1886. T.N.A.
70
Ibid.
71
Thomas, Report on Tanjore Remissions, pp. 45—6.

39
 The emergence of provincial politics

of its bureaucracy would have prevented its policies from being

translated into reality. In the most elementary ways, it lacked the
machinery to do anything more than survive. Due to the local nature
of recruitment and the usurpation of the powers of appointment by
permanent district officials, it could not even control its own employ-
ment policy. In 1851, the Board of Revenue issued Standing Order
No. 128 (2), which precluded the appointment in the same govern-
ment office of members of the same family. As we have seen from
examples in the 1880s, this had had very little effect. In fact, the
much-acclaimed First Communal Order of 1921, by which govern-
ment in collaboration with the non-Brahman Justice Party sought to
limit the number of Brahmans in government service, was no more
than the re-issuing of this standing order which had been ignored for
seventy years.72 Naturally, the personal and social interests of these
uncontrolled local bureaucrats coloured the performance of govern-
ment more brightly than did the distant Secretariat. Repeated attempts
by Fort St George to raise the position of untouchables, for example,
were rendered nugatory for its predominantly high-caste executive
refused to enforce them: land set aside for panchama resettlement
disappeared from the records and a High Court ruling of 1917
revealed that slavery still existed in Malabar, eighty-four years after
it had been outlawed.73 The capital's direct influence in the locality
was largely notional.
Yet although Fort St George may appear hopeless and incompetent
by the standards that are expected of a modern state, it was, in fact,
perfectly tuned to the needs of the mid-nineteenth-century British
Empire. For Parliament in London and for the Government of India,
it existed to provide a high yield of revenue and to guarantee its own
security. Both of these tasks it performed admirably, indeed better
than any other provincial administration. This may seem a paradox
but when we begin to examine the political relationships implied by
the administrative system, the reasons for Fort St George's success
become obvious. Its government was carried on by or in the interests
of men who independently possessed a large share of political authority
in native society.
In much nationalist fiction, the rule of British officials and their
native subordinates has been characterised as tyrannical and auto-
cratic. Recently, this judgment has been lent academic respectability
by scholars who have weighed the political importance of various
72
E . F . Irschick, Politics and Social Conflict in South India (Berkeley a n d Los Angeles,
1969), p. 236.
73
G.O. 1675 (Home, Misc.) dated 2 December 1919. T.N.A.

40
 The governance of Madras

Indian communities by the number of government jobs which they

held74 or who have argued that Brahman and Sat-Sudra groups rose,
to power in British South India by capturing and using in their private
interests the mechanisms of the administration.75 A careful scrutiny
of the nature of government in Madras, however, suggests that these
views, at least as they relate to the period from 1870 to 1920, are very
difficult to substantiate. If association with the administration of Fort
St George immediately and of itself produced the force with which of-
ficials could smash the hegemonies of non-officials and establish their
own rule, then that administration must have been very powerful and
must have been able to command in society an overwhelming pre-
ponderance of force. To draw parallels from history, its practice
must have matched that of a Peter the Great, a Robespierre or a
Stalin for only in the type of state system operating under those men
was power so concentrated as to make rulership of this sort possible.
In the early years of the British settlement of South India, a case for
government by terror could perhaps be made. With the rules of law
uncertain, the direction and volume of the flow of resources to the
state undecided and large quantities of armed men available to support
the judgments of Company servants, bureaucrats and their hangers-
on were often in a position to seize what they liked. This epoch of
violence and rapid change, however, certainly was over by the 1850s,
if not the 1830s. It was replaced by a period of slow and paralysingly
inefficient government. Much as the Viceroy Lord Curzon may have
envied Peter the Great, the administration which he inherited was not
the kind of weapon which he or his bureaucrats could use for the
creation of terror and the expropriation of property. It was more 'a
mighty and magnificent machine for doing absolutely nothing'.
It would be a truism, of course, to say that ultimately the British
were able to stay in India only because of their military resources.
But when looking at the modus operandi of government in Madras
what is most noticeable is the degree of force which the British did not
have to apply to keep their administration running. Although sitting
on a reserve of armed might and punitive police, which could in theory
be applied to any area in which the rule of government had broken
down, the British in fact were extremely loath to treat their Madras
subjects heavy-handedly. In the first place, and obviously, as strong as
they were they could not spare the troops from wider imperial duties
to rule by the sword everywhere all the time. Secondly, and more
74
See, for example, A . Seal, The Emergence of Indian Nationalism (Cambridge,
1968).
75
Stein, 'Integration of the Agrarian System of South India', pp. 202-4, 211.

41
 The emergence of provincial politics

relevantly, the internal use of troops cost the Government of India

more money than it cared to spend. The severe castigation which Fort
St George received for calling out the army to avert a riot situation in
Rajahmundry in 1907 indicates very clearly the limits of military
force available to provincial governments.76 For all except the wildest
outbreaks of famine or Moplah violence, soldiers were out of the ques-
tion. The alternative was to send in punitive police who possessed the
virtue of being paid for by the people whom they were punishing. Yet
this method of coercion also had its problems. Punitive police were
difficult to control, provoked the anger of innocent and otherwise
loyal subjects and tended to exacerbate already delicate situations.
It was general policy to use them as sparingly and for as little time as
possible. The great hammer of force which appeared to back Fort St
George's political position, therefore, was by no means easy to swing;
and, as if to make this position clear to its subordinates in the local-
ities, when it did swing the hammer it often hit the local subordinates
who had allowed their charges to run out of control as hard as it hit
more formally designated troublemakers. At Salem in 1882, Tinne-
velly in 1899 and Cocanada in 1907,77 Fort St George decimated the
district bureaucracy for its failure to cope with its own problems.
For all practical purposes, government servants in the mofussil
could not rely on Fort St George for the coercive force necessary to
fulfil their potential roles. They had to find resources of power within
their localities. Yet the independent resources available to them were
extremely limited. In particular, they could place no faith in the ability
of the police department to provide them with support. Following the
revelations of the Torture Commission (1855), the Madras police
had been constituted into an agency separate from the revenue de-
partment. 78 Tahsildars who wished to use the powers of the police now
had at least to satisfy officials in another bureaucracy of their bona
fides. More importantly, however, even were lower revenue sub-
ordinates to succeed in obtaining police power, it is not clear that
they could achieve very much with it. The portion of the police de-
partment which was under direct bureaucratic control, and capable of
being seen as an independent force in society, was very small indeed.
76
H o m e Police A A u g u s t 1907 N o s 10—11; H o m e Police A O c t o b e r 1907, N o .
123. N.A.I.
77
G.O. 353 (Judicial) dated 10 February 1883. T.N.A.; Home Police A March 1900,
Nos. 1-8. N.A.I.; G.O. 1266 (Judicial) dated 16 July 1907. T.N.A.
78
See Report of the Commissioners for the Investigation of Alleged Cases of Torture in
the Madras Presidency (Madras, 1855).

42
 The governance of Madras

At no time before 1900 was the ratio of district policemen to district

population better than 1:2000; outside the main towns it was nearer
i:4ooo. 79 The police department proper was only intended to provide
a service of secondary investigation. All primary investigation was
carried out by the village police who were under the control of local
powers - village headmen who, as we shall see, were local landowners,
and zamindars. Using only the powers of the district police, a tahsildar
could have little hope of cowering his taluk into mute obedience.
Moreover, against village-level resistance, he could not use the police
courts for the village police were just as much of a legal force as were
his district police and, more significantly, were much nearer to the
source of evidence. The only way in which he could continue his
government was by enlisting the support of men under the prior con-
trol of village headmen and zamindars. While this could give him some
room for manoeuvre, by playing the ambitions of one local power off
against those of another, it did not give him an independent basis for
despotism. If he wished to retain power he had to keep some of the 'big
men' in his charge on his side all the time.
In spite of, or perhaps because of, the elaborateness of the
bureaucracy, the lower government official was no better placed to
use the slower processes of administrative coercion to strengthen his
rule. In theory, he could order the sale of land of any man whom he
deemed to be a revenue defaulter; in practice, he could neither legally
describe the defaulter's land nor find buyers for it without the co-
operation of the village-based revenue office.80 In theory, he was in
a position to 'arrange' legal cases against any local inhabitant of whom
he disapproved and have him convicted; in practice, the rules of evi-
dence required under British law were so rigorous and the ability of
rich local men to conceal evidence or hire witnesses was so great that
he could seldom obtain a conviction against, or even take to trial, any
man of substance.81 In theory, he could raise the taxes on whom he
liked, interfere in trade as much as he pleased and distribute his
patronage wherever he chose; in practice, he had to balance the use of
his powers between the various contending political groups of his
locality or he would foment riots, strikes and hartals and render his
79
See Madras Police 1900.
80
Report of the Indian Famine Commission. Appendix. Volume III. Condition of the
Country and People, p. 416. P.P. 1881. LXXI, part 2; Madras Provincial Banking
Enquiry Committee. Evidence (Madras, 1930), 111, 679. Hereafter, these volumes
as MPBC....
81
See below pp. 44, 109-10.

43
 The emergence of provincial politics

charge ungovernable.82 The regularity with which local protests and

uprisings occurred during our period indicates clearly the strength
of non-official society's resistance to the overzealous and incautious
bureaucrat.
Indeed, influential local subjects were very well placed not only
to resist but to destroy their titular government masters. Their ability
to control information both inside and outside the bureaucracy was
often greater than that of their chief local official. Mothey Venka-
taswami, a millionaire merchant of Ellore, for example, had a large
number of debtors and dependents inside the Ellore Sub-Collectorate.
Twice reports on his more dubious activities were despatched from
Ellore to the Godavari Collectorate in Cocanada and twice they were
intercepted and destroyed en routed Should a magnate take to the of-
fensive and prepare a corruption charge against his local tahsildar or
sub-judge, the official could find himself acutely embarrassed even if
there were no substance to the charges. Against hundreds of bought
and rehearsed witnesses, his pleas of innocence sounded hollow. In
1896, Mothey Venkataswami broke the Ellore sub-judge, I. N. Swami-
natha Iyer, whose main defence was that he had found against the
Mothey interest in several suits and was now the victim of a revenge
plot.84 In Bellary in 1890, another sub-judge, who also had alienated
some of the local notability, was forced to plead the same defence and
go the same way. In this second case the sessions judge remarked that
it was passing strange that the accusations emanated from a pauper
who, somehow, had managed to retain the best lawyers in town to act
on his behalf.85 Corruption trials were as much a part of life in
British South India as mangoes; and although the number of officials
who actually were dismissed and imprisoned may have been small, the
number who were transferred after acquittal (a satisfactory solution
from the local magnate's point of view) was considerable. If the cor-
ruption trial failed, there were many other ploys: large-scale public
meetings,86 letters to the newspapers and petitions would draw the at-
82
See, for example, the strike of bazaarmen organised against municipal taxation in
Madura in 1880 in Raja Rama Rao, Sir S. Subramania Aiyer, p. 12; or in Bellary
in 1908 against municipal bye-laws, G.O. 2120 (L a n d M , M), dated 19 December
1908. T.N.A.; or in Trichinopoly in 1886 against municipal maladministration,
Vettikodayan, 23 October 1886. R.N.P.; or in Anantapur in 1896 against income-
tax assessments, Hindu 18 September 1896.
83
Marginal notes of Collector of Godavari on A. R. Banerji, Assistant Collector,
Ellore, to Collector of Godavari, 29 March 1901 in G.O. 1011 (L and M, M) dated
17 July 1901. T.N.A.
84
Hindu 5 June 1896.
85
Hindu 15 October 1890.
86
As at Salem in 1893, Hindu 15 December 1893.

44
 The governance of Madras

tention of his superiors to the trouble which the official was causing.
Again, although these seldom led to dismissal, they often produced
investigations and transfers.
The notion that subordinate government servants, and by implica-
tion the social groups from which they were drawn, came to exercise
great power through their connection with the British administration
in South India, then, would seem to be untenable. It confuses the
potential powers of office with the ability to fulfil them and, more-
over, the ability to draw considerable financial rewards with the ability
to exercise direct power. At best, the government servant was an
intermediary, acting between Fort St George and the greater non-
official powers in his locality. Fort St George used him to extract as
much as he could from his charge without disturbing its peace; the
magnates used him to negotiate the minimum for which Fort St
George would settle before it had to use its reserve powers and send
in troops, punitive policemen and Board of Revenue investigators.
The whole system, and his position within it, hinged on the fact that
both sides sought an amicable compromise with as little fuss and inter-
ference as possible. The successful servant had to rely as much on the
confidence of the non-officials around him as he did on the govern-
ment. The occasions on which he could act independently came only
when both of his masters were confused and weak. These were quite
common in his relationships with his 'superiors' but, in those with his
'inferiors', were confined to the periods when he could choose be-
tween two or more equally balanced 'big men' or factions composed of
'big men'. Local symmetries of this type were rare; and the moment
that he committed himself to one side or the other they ended. Cer-
tainly, as a middle-man he obtained great rewards, receiving a salary
from the government and 'bribes' from local inhabitants, but it makes
little sense to regard his prosperity as a sign of his personal power. As
R. E. Frykenberg ultimately saw in Guntur: 'The corruption of
authority in Guntur (and in other districts) ought not to blind us to
the fact that the Maratha Brahmans [administrators] were go-
betweens.'87
At worst, and most usually, however, the government servant's
possible independence had been sold away the instant that he donned
his government cap. The fragmentation of the administration discon-
nected him from Fort St George and forced him, whether willingly or
not, to give or sell his services to the men who controlled local methods
of coercion. He became an employee of the powers within his charge
and served their interests to the exclusion of Fort St George's. This
87
Frykenberg, Guntur District 1788-1848, p. 234.

45
 The emergence of provincial politics

can be seen clearly when we examine the background of the people in

the administration and the ways in which they operated.
In the first place, many local bureaucrats were already local
powers when they took their seats in their government office. In
Chapter 4, we shall be looking closely at the nature of the village
officers who occupied the lowest and most critical stratum of the
governmental pyramid. Here we need only point out that they were
drawn from the wealthiest and most influential families in their
village. Above the village officers, up to the 'provincial' post of
deputy-collector, some of the servants of government also were re-
cruited from the major economic powers in the district and exer-
cised a strong social control based upon this wealth. In Tanjore
in 1885, H. S. Thomas found:
Many of the Tanjore officials are largely interested in land in the dis-
trict, and there is thus thrown about a Collector a network of men
interested, both personally and through their relatives, in obtaining
remissions on lands.... Tanjore officials are interested in land in their
own district to an extent which covers 73 square miles, pays an annual
assessment of about two lakhs and may be valued at Rs 97,71,000 say
close on a crore of rupees.88
This development was also particularly noticeable in Malabar,
where the social groups who served in the upper reaches of the
district bureaucracy were also those who dominated landed
resources.89
Secondly, whenever a government officer took a bribe, he signed
a contract by which he placed his influence at the disposal of his non-
official paymaster. He became the paid agent of the magnate whom he
was supposed to govern. Frykenberg noted this inversion of the
sources of supposed authority in Guntur: 'Village money spread a
corrupting influence into even higher levels of the hierarchy, and
the administration became caught in the webs of village influence.'90
It was implied by Ampthill in his description of jamabundi, which we
quoted earlier, and lay at the heart of Thomas' findings in Tanjore.
Further examples of it litter the history of the South. A Collector of
North Arcot stormed at his judiciary: 'Zamindars, Poligars, Jaghidars,
Gurus — anybody with power or command of money, simply used the
Sub-Magistrates as weapons of offence.'91 In 1916, the Revenue
Department considered placing an I.C.S. officer in Devakottai division
88
T h o m a s , Report on Tanjore Remissions, p. 37.
89
Namely, t h e Nairs.
90
Frykenberg, Guntur District 1788-1848, p. 230.
91
Madras Police 1886. Appendix C, p. v.

46
 The governance of Madras

of Ramnad district, the home'of the Nattukottai Chetty international

bankers: 'Devakottai is very exceptionable. There is any number of
extremely wealthy Nattukottai Chetties who buy up any officials they
can.' The department greatly appreciated 'the danger of subservience
to the Chetties'.92 In Rajapalaiyam at about the same time, the landed
magnate A. K. D. Dharma Raja 'procured' the services of his tahsildar
T. S. Ramaswami Iyer to keep his private market free from the avaric-
ious gaze of government.93 Following the Salem riots of 1882, the
huzur sheristidar A. Venkatasubba Iyer was dismissed less for his part
in the riots than because 'I [the Collector] have come to the conclu-
sion from the conduct of work before me during the last six months
that he has unduly exposed himself to the influence of the Mittadars,
which means corruption.'94 In 1906, the deputy-collector Raja Rama
Rao, who came from one of the most illustrious service families in
Madras, was dismissed for receiving presents from the raja of Ramnad
for services rendered while he was in the Ramnad division.95 In some
zamindari areas it was difficult to tell the difference between the
estate and the government bureaucracy. Not only did local bureaucrats
switch between the two96 but many had relatives permanently in the
service of both.97 Indeed in Kistna district at the turn of the century,
the distinction had altogether disappeared and the zamindar of Muna-
gala was giving away government revenue inspectorships.98 The in-
fluence of local magnates on local bureaucrats can be gauged again
from the part the former played in harassing central investigations.99
As the structure of government actually touched Madras society,
it was moulded and shaped by that society. Indigenous, non-official
powers, by one means or another, absorbed and controlled the func-
tions of the state in the locality. This happened not simply because
bureaucrats were weak or put personal interest before that of the
92
G.O. 349 (Home, Misc.) dated 4 April 1917. T.N.A.
93
G.O. 1984 (L.S.G.) dated 7 September 1923. T.N.A.
94
Note signed C. D . Macleane in G.O. 353 (Judicial) dated 10 F e b r u a r y 1 8 8 3 .
T.N.A.
95
Andhraprakasikha 5 December 1906. R.N.P.
96
F o r a n example in t h e P i t h a p u r a m estate, see Hindu 2 2 August 1910.
97
Such as t h e 'Koka' family of Audi-Velamas, see Hindu 26 M a r c h 1920.
98
A. Kaleswara Rao, Na Jivita Katha-Navya Andhramu (Vijayawada, 1959), p . 19.
(Telugu).
99
For example, H. S. Thomas' investigations in Tanjore stirred an agitation led by
the prominent landowner S. A. Saminatha Iyer who, presumably, had some
interest at stake. Hindu 19 January, 13 March, 25 May and 13 October 1885.
Equally, in Cuddapah in 1884, Assistant Collector Farmer tried to break a
Collectorate clique and was sued by a variety of local landowners for libel and
property damage. Hindu 19 January and 13 March 1885.

47
 The emergence of provincial politics

state but because the social material from which the British had built
their administrative machine, the design of that machine and the
purposes which the machine was intended to serve virtually guar-
anteed that it would happen. Although it is possible to find few
British civilians of the period who were prepared openly to admit
that their government rested on the personal power of their greater
subjects, many acts of British administrative practice make it clear
that this position was at least tacitly recognised by them. No
systematic effort was made to prevent local landholders from be-
coming bureaucrats in their own districts. The interchange of per-
sonnel between zamindari and district bureaucracies, and the con-
sequent confusion of loyalties between zamindars and government,
was actively encouraged. Moreover, a great deal of the administra-
tion was farmed out and left to the responsibility of local notables.
The legal system recognised a number of independent authorities
of which it made use at every opportunity: Fort St George ap-
pointed Kazis to administer Islamic civil law to the Muslim popula-
tion, allowed its courts to refer disputes between members of the
same caste to caste headmen and panchayats and, in the country-
side, gave village headmen the right to try small cases. In order to
improve sanitation and communications, it brought together urban
and rural notables in municipal committees and local fund boards,
and asked them to take over responsibility for these affairs. It ap-
pointed Honorary Magistrates from among its most prominent
citizens to try petty criminal cases. During times of riot and civil
strife it sought the authority of wealthy and powerful men to help
it calm the storm. And, more informally of course, its administrative
system allowed many local leaders a large say in the actual apportion-
ment and distribution of taxation. In every locality, the government of
Fort St George was based on a series of alliances and tacit understand-
ings with men of substance and power who thus had an interest in the
continuation of that government.
The existence of a tight web of local connections between govern-
ment and society helps to explain Fort St George's success in raising
revenue, keeping the peace and preserving its own political security.
In return for a fixed tribute, set high by historical accident,100 and for
the promise of order, it agreed to stay out of the localities and allow
them to find their own political balances. It set the rules of the game
and, through the threat of its reserve powers, could make sure that
100
A century of continuous warfare, Hyder Ali's government, and the Maratha
depredations had destroyed local society's ability to resist a high revenue demand
at the time of the British conquest.

48
 The governance of Madras

they were kept; but it did not look too carefully at what was happening
in the scrimmages and placed at least half the field on its blind side. It
gave to those non-officials who were prepared to take them, chances to
develop and use their existing powers without interference. It thus
established alliances in every taluk and town with men who had an
incentive to abide, or to appear to abide, by its regulations and to do
its work for it. Revenue continued to flow and order to be maintained
although, by any standard, there was little formal government.
Of course, misguided policies and economic and social pressures
were bound to upset the balances sometimes: local officials might
alienate strong magnates and famine or communal rioting fracture the
networks of magnate control.101 But the highly decentralised nature
of the political sytem guaranteed that these disturbances never threat-
ened Fort St George with a Mutiny situation. In the nineteenth cen-
tury, Madras achieved the reputation of a 'benighted' province in
which political lethargy was general. While this description certainly
would fit its provincial politics, it would be wholly false of its local
politics. Every agitational technique to be found in the twentieth-
century nationalist movement, from stone-throwing to satyagraha
and from assassination to newspaper scandal, was well established in
Madras by 1870. Magnates and political leaders battled for the ear of
government, rent strikes were organised, huge anti-British and anti-
government demonstrations were arranged and policies of non-co-
operation were offered. However, the arenas in which these campaigns
took place were always those of the locality, in which the critical level
of government decision-making lay. As virtually no two taluks, or even
firkas, need be subject to the same governmental problems, and as the
same social forces in two adjacent areas could be accommodated by
governmental authorities in different ways, it was extremely difficult
for dissidences to spread.
In these conditions, Fort St George's political position was safe.
In every district it had hosts of collaborators only too eager to attach
themselves to it and to carry on its revenue and police duties for it.
Most local conflicts stemmed not from opposition to the government
but from rivalry to become associated with it.102 Even when violent
disturbances did break out, they were isolated and could be handled
with disdain. The British, before the second decade of the twentieth
century at the earliest, were not challenged at the state level because
there was no state-level arena in which an important challenge
101
See below, Chapter 3.
102
As for jobs in the bureaucracy or for control of local administrative institutions.

49
 The emergence of provincial politics

could be made. Political satisfaction was a matter to be sought as

near to the ground as possible.
Yet, obviously, the price Fort St George paid for its revenue and its
security was high. Once it had created a political system which kept it
remote from the daily affairs of its subjects, it could not initiate a
general policy of interference without destroying that system and
promoting upheaval. This debility was of little consequence while the
society it governed and the ends it pursued remained constant. But
changes in either could present it with great problems. From the
1870s, both its society and its ends began a process of significant
change, which forced it to undertake a major reorganisation of the
resources of the state. In doing so it had to modify severely the political
system on which it rested.
Fort St George was fortunate in that violent social change was
confined to a few areas which, because of the segmented nature of
the state, did not impinge heavily on the presidency as a whole.
Equally, it was lucky in not having to cope with the permanent,
growing and organised hostility of its educated society. The eco-
nomic problems of the Bengali bhadralok and the cultural difficul-
ties of hhadralok and Chitpavan Brahman revivalism were not re-
peated in the South. Moreover, the press was small and limited in
its influence and job competition between traditional and 'newly'
literate groups was not to develop until the 1930s.103 It was less
social than economic change which steadily undermined Fort St
George's position of quieta non movere. From their earliest settle-
ments, the British had sought to collect revenues in cash. Between
the 1820s and the 1860s, this revenue flow - this tribute from the
localities — had remained constant and had become a fixed part of
the bargain between Fort St George and its local indigenous powers.
For most of the period the British had had the better of the bargain
because commodity prices had tended to fall.104 During the latter
four decades of the nineteenth century, however, prices began to rise
at first steadily but from the 1890s with alarming rapidity.105 Fort St
George simply had to find new sources of revenue and to cultivate new
resources of wealth if it were even to stay still. It had to interfere in

103
For dicussions of these and other factors, which lay behind the development of the
nationalist movement in other provinces, see L. A. Gordon, Bengal: The National-
ist Movement 18/0-1940 (New York and London, 1974); Seal, Emergence of
Indian Nationalism; G. Johnson, Provincial Politics and Indian Nationalism
(Cambridge, 1973).
104
K u m a r , Land and Caste in South India, p . 9 1 .
105
Ibid.

50
 The governance of Madras

the long-standing arrangements which it had made with its subjects.

Moreover, it soon found that staying still was not going to be
enough. Ultimately, the ends which Fort St George pursued were
determined by the British Parliament and the Government of India
whose horizons were much wider than its own. Madras served the
purposes of the British Empire and, during our period, those pur-
poses themselves were changing. Competition in Europe and America
challenged the industrial supremacy of Britain and made her in-
creasingly reliant on Indian markets and on Indian commodity sur-
pluses to meet her balan.ce of payments deficits;106 extended imperial
commitment, particularly in Africa, and the threat of Russian expan-
sion in Central Asia threw a new weight onto the Indian army;107
international currency difficulties squeezed the imperial economy and
forced Britain to seek more real wealth through taxation in India.108
An administrative system which had been tuned to the needs of 1830
began to look rather fragile when faced with the needs of 1900. Fort
St George was ordered to find more money, to streamline and cheapen
its government, to develop communications and economic resources,
to prevent the worst effects of famine and disease, which periodically
devastated its economic and revenue performance, and to stimulate
trade and industry. Its masters ordered, and sometimes whipped, it
into activity.
Four sets of reforms which Fort St George was forced to under-
take in response to these new pressures were of crucial importance
in altering political relations in the presidency. The first three served
to weld the bureaucratic capital much more firmly to the base of the
localities and to make local notables consider the centre of govern-
ment more seriously when planning their political strategies. The
fourth actually 'extended' the locality into a larger arena, composed
not of a few adjacent villages or suburbs but of whole towns and dis-
tricts. This extended locality produced more complex patterns of
political interaction and also was more easily seen and interfered in
by the capital of government.
Firstly, Fort St George was impelled to intervene more regularly
in the operations of the economy in order to raise more revenue and
to promote commerce and economic development. In seeking cash, it
106
See E . J. Hobsbawm, Industry and Empire: and economic history of Britain since
ijS° (London, 1969).
107
See J. Gallagher and R. Robinson, 'The Imperialism of Free Trade' in Economic
History Review, vi: 1 (1953), 4 - 5 ; also Imperial Gazetteer of India (Oxford, 1907),
iv, 186.
108
Imperial Gazetteer of India iv, 194—6, 518.

51
 The emergence of provincial politics

soon discovered that it could not count on an infinite series of in-

creases in the assessment on land. Its land revenue resettlements
provoked agitations and disturbances on an unparalleled scale; on
several occasions it had to substitute political discretion for financial
valour and reduce or even postpone indefinitely the implementation
of resettlements.109 Moreover, its reliance on rich rural inhabitants
for the force on which its government rested meant that it lacked a
coercive arm when attempting to expropriate rural wealth. The rise
in the land revenue did not nearly match the rise in prices and in
governmental costs. It was obtained only by uncovering and bringing
into the tax system land previously concealed or newly cultivated or
newly irrigated, not by increasing significantly the rate of assessment
per acre. Between 1886—7 and 1925—6, the rate of assessment per acre
rose only by 12 per cent in the case of wet land and by 7 per cent in the
case of dry.110 Fort St George had to look elsewhere for its money
and the proportion of total revenues represented by the land revenue
dropped rapidly from 57 per cent in 1880 to 28 per cent in 1920.111
To compensate for the rigidities of its land revenue system, the
Government of Madras turned away from the major source of wealth
in its society and concentrated its attention on more peripheral areas
of the economy. In the internal sector, its eye fell immediately on trade
and commerce. Beginning in 1886, it administered for the Govern-
ment of India a tax on non-agricultural incomes which came to yield
Rs 2 crores p.a. by 1920.112 On its own behalf, it undertook a rigorous
campaign of taxation and control against the huge native liquor in-
dustry. Prior to 1880, the influence of government in this field had
been small. It had licensed the manufacture of spirits and sold rights
to the monopoly of liquor distribution in certain areas but it had left
the speculators and contractors who bought the licences and mono-
polies to provide their own administrations and, consequently,
government profits from the trade had been limited. In 1882-3, its
entire excise revenue reached only Rs 60 lakhs.l'3 Over the next forty
In the face of fierce local agitation in Tanjore in 1892-3, the government was
forced to reduce the rate of assessment increase in its resettlement of the district.
Hindu 17 July 1896. The Malabar resettlement of 1883 was still unenforced in
1903. Lord Ampthill to Broderick. 12 November 1903. Ampthill Papers. I.O.L.
Calculated from Land Revenue 1885—6 and 1925—6.
Report on the Administration of the Madras Presidency during the year 1880—1
(Madras, 1881), pp. cxxvii—viii; ibid., 1920—1, p. 80.
Report on the Administration of Income Tax under Act II of 1886 in the Madras
Presidency for the year 1920-1 (Madras, 1921). Hereafter this series as Income
Tax. . .
Report on the Administration of the Abkari Revenue in the Presidency of Fort St
George for the year 1882—3 (Madras, 1883). Hereafter this annual series as Abkari.

52
 The governance of Madras

years, however, Fort St George's drive for more money forced it to

interfere much more closely in the industry. It built a new bureau-
cracy solely for excise administration and clamped down heavily on
licensees and monopolists. By 1920, it had increased its excise revenue
ninefold to Rs 5.4 crores or to 21 per cent of its total income.114
New measures of taxation, however, were not enough. What was
needed also were measures to protect economic resources, to foster
economic growth and to remove the social obstacles to material pros-
perity. In pursuit of these ends, Fort St George launched an admin-
istrative crusade which carrried it into the countryside and led it to
interfere in many of the basic economic activities and relationships of
its inhabitants. In 1878, it passed laws to arrest the alarming denuda-
tion of the forest reserves which played a large part in the economy of
the 'dry' areas.l J 5 At about the same time, it began to consider legis-
lation to clarify the nature of contract between landlords and tenants
and to bring the administration of zamindari estates under better
bureaucratic and legal control.116 Following the disasters of the
famine of 1876—8, it elaborated a new famine code which brought
relief more speedily to the distressed.117 In addition to saving life and
land, this code had the effect of interposing government between
the weaker and stronger elements of rural society during periods of
acute agrarian crisis. By the early 1890s, Fort St George was even
setting itself to solve two of the most profound problems of the eco-
nomy. In the main mirasi areas of the province, it sought to break
the slave-like dependence of landless labourers on landlords by pro-
viding land and house-sites for pariahs.118 And it deputed one of its
most knowledgeable civilians, Frederick Nicholson, to report on the
possibilities of providing cheaper and more flexible systems of rural
credit than then existed.119 In 1904, some of Nicholson's suggestions
were incorporated into a Co-operative Credit Societies Act which
sought to promote state-backed mutual lending associations among
cultivators.
114
Report on the Administration of the Madras Presidency during the year 1920—1
(Madras, 1921), p. 80.
115
Report on the Administration of the Forest Department of the Madras Presidency
1882-3 (Madras, 1883).
116
Such as the Zamindari Village Officers Act of 1894, which was first discussed in
the Legislative Council in 1883.
117
Report of the Indian Famine Commission. 1880. Volume II. P.P. 1881. Vol. LXXI.
118
G.O. 1675 (Home, Misc.) dated 2 December 1919. T.N.A.
119
F. A. Nicholson, Report regarding the possibilities of introducing Agricultural Banks
into the Madras Presidency (Madras, 1895 and 1897), 2 vols. Hereafter, these
volumes as Nicholson, RAB.

53
 The emergence of provincial politics

Fort St George's initiatives in matters of taxation and economic

planning produced two important alterations in the relationship be-
tween government and society. In the first place, government came
into contact with and began to affect the interests of economic groups
which previously it had left alone. Merchants, businessmen, contrac-
tors, industrialists, cattle-breeders, moneylenders and landlords
found themselves under a new pressure from the intrusion of the
bureaucracy into many areas of their lives. Naturally, they responded
to this pressure by organising movements of resistance. In the years
leading up to the First World War, violent opposition on the ground
between government officials and various local economic powers grew
in intensity until it threatened to destroy much of the machinery of
administration.120 Moreover, it fed into the wider currents of the
nationalist movement and immeasurably strengthened the power of
nationalist leaders. Several of the more conspicuous local victories of
Gandhi's non-co-operation campaign of 1920-2 were the result of
support from merchants, eager to lower the income-tax on themselves;
liquor traders, attempting to cut excise revenue; and cattle-breeders,
pursuing their long-standing conflict with the forest department.121
Secondly, the character of the taxation and economic policies
helped to shift the attention of various economic groups away from
the locality and fixed it firmly on the institutions of the capital. In
sharp contrast to the land revenue system, the excise and income-
tax administrations stressed the role of Madras city in the making of
important decisions. Under the ryotwari system, as we have seen, the
effective apportionment of revenue demand lay with local officials
of one kind or another. The only time that revenue payers could
gain anything by approaching the Secretariat or Board of Revenue was
immediately after resettlement operations when Fort St George was
considering the expediency of implementing the recommendations of
its resettlement officers. As resettlements were made district by dis-
trict at thirty- or forty-year intervals and as the rates of commutation
and payment were set by local circumstance rather than provincial
fiat, it is not surprising that the constant pressure exercised by land
revenue payers on Fort St George was not great. Fierce as may have
120
By 1914, illicit distilling and prosecutions for offences against the excise laws
were reaching mammoth proportions, see Abkari, 1914—1$ and 1915-16;
evasions of forest laws and attacks on forest subordinates had grown hugely, see
Report on the Working of the Forest Department of the Madras Presidency 1915—16
(Madras, 1916); in certain areas, the village police and revenue services were in
open revolt, see below pp. 152—9.
121
C. J. Baker, 'Noncooperation in South India' in C. J. Baker and D. A. Washbrook
(eds.), South India: Political Institutions and Political Change (New Delhi, 1975).

54
 The governance of Madras

been the reactions of individual districts to their own resettlements,

there was no standing provincial association of ryotwari landholders
in Madras until 1918 and, apart from vague murmurings about the
desirability of a permanent settlement, no general political lobby in
the interest of ryotwari landholders.
The economic interests touched by Fort St George's new excise
and income-tax policies, however, came to be organised very dif-
ferently. Certainly, at no time before 1920 was it impossible for
them to bribe local excise and income-tax officials in order to
wring concessions out of the administration. But they could also
serve their interests in significant ways by going to the centre.
Fort St George itself set the general rates of demand on the liquor
industry and had sufficient freedom over the Government of India's
income-tax to decide the categories in which payers should be placed
and the ways in which they should pay. Sometimes the favours to be
had from Fort St George far outmatched the rewards buyable in the
locality. The Nattukottai Chetty bankers, for example, acquired the
right for each of their family banks to be assessed for income-tax as
a single unit rather than as a number of separate branches. This en-
abled them to write off the losses of one branch against the profits of
another and thus saved them thousands of rupees a year.122 Equally,
chit lottery banking proprietors succeeded in having their businesses
moved from a 'private' to a 'company' classification which enabled
them also to cut the taxation burden. 123 In the liquor industry, various
groups mounted demands to Fort St George for a lowering of excise
duty and for a reorganization of the excise administration. Often they
were successful, as in the 1890s when Malabar sweet-toddy traders
persuaded the excise department to redesign completely its bureau-
cratic machine in the district.124
Similarly, the economic interests influenced by Fort St George's
economic policies could not avoid going to the capital when seeking
certain types of redress. Again, cattle-breeders could bribe local
122
The Nattukottai Chetties maintained a standing vakil in the Legislative Council
to look after their affairs. For most of the period between 1904 and 1920, he was
P. S. Sivaswami Iyer. See S. R. Rm. A. Annamalai Chetty to P. S. Sivaswami
Iyer, 11 June 1920; N. Sivaramakrishnan to P. S. Sivaswami Iyer, 22 June 1920.
P. S. Sivaswami Iyer Papers. N.A.L
123
Income Tax, 1899—1900. F o r a list of provincial bye-laws passed by F o r t St
George to the Government of India's Act 11 of 1886, see An Income Tax Manual
being Act II of 1886 with Rulings and Orders issued by the Governments of India,
the Madras Government and Board of Revenue and the Circular Instructions issued
by the Accountant-General Madras upto 31st January 1900 (Madras, 1900).
124
Abkari, 1892-3 and 1893-4.

55
 The emergence of provincial politics

forest guards, zamindars pay their local government officials to turn

a blind eye to many practices and mirasidars heavily influence the
humanitarian measures being extended to their pariah serfs. But it
was only at Fort St George, from which the policy initiative emanated,
that they could obtain other substantial concessions. By campaigning
before the Secretariat in Madras city, the forest interest managed to
alter entirely the technique of forest administration which had been
established in 1878; 125 the zamindari interest managed to switch
the burden of local cess taxation from itself to government; 126 and the
mirasidar interest managed to persuade Fort St George to dilute its
pariah policy.127 By forming permanent associations,128 organising
petitions and, most importantly, using the new breed of educated
Indian publicist who was growing up around the capital at this time,
mercantile and industrial interests forged new political links with
Fort St George. They helped to turn an administrative system,
which was becoming ever broader and more provincial in its scope,
into the basis of a political system in which the widely dispersed parts
were held together through their connection to a common centre of
decision- making.
Once the Madras Government had begun to seek an improvement
in the performance of its bureaucracy, it soon come to recognise the
desperate need for a reorganisation of the bureaucratic agency itself.
From the early 1880s, Fort St George pursued this end along two
parallel paths. Firstly, it attempted to break down the administrative
omnicompetence of the revenue department, which had given revenue
subordinates a stranglehold on the information which reached it. The
dangers of concentrating full bureaucratic powers in one man had
been appreciated by the Torture Commission of 1855, which had
recommended that tahsildars be relieved of their police functions.129
Although this reform was accomplished in the late 1850s, natural
inertia and lack of pressure from outside combined to prevent any
further measures of departmentalisation in the administration for
thirty years. In 1885, however, in order to provide an executive arm
for its aggressive excise policy, Fort St George created a new Abkari
and Excise Department and removed powers of supervision over
excise matters from the ordinary revenue staff. In 1892, in order to
125
Report of the Forest Committee 1913 ( M a d r a s , 1913), Vol 1.
126
From about 1908, Fort St George was paying the majority share of village officer's
stipends in zamindari areas. Home Judicial, File 931 of 1922. N.A.I.
127
H o m e Political 'Deposit' April 1917 No. 6 1 . N.A.I.
128
Such as t h e N a t t u k o t t a i N a g a r a t h a Association (1904), t h e M a d r a s Landholders
Association (1890) a n d t h e S o u t h e r n Indian C h a m b e r of Commerce (1911).
129
Report of the Commissioners for the Investigation of Alleged Cases of Torture.

56
 The governance of Madras

improve its magisterial service, it withdrew common magisterial

powers from tahsildars who were on tour and appointed a series of
'stationary submagistrates' who took over magisterial duties at taluk
headquarters.130 From the 1880s, the steady increase in the number of
subordinate civil courts gradually whittled away the general arbit-
rational functions which revenue subordinates had served: disputes
which once might have been about revenue rights became increasingly
about property rights. From 1909, the revenue department was pulled
back from the rural boards through which it had supervised the pro-
vision of local facilities such as roads, wells and drains. Moreover,
at this time, separate agencies for the promotion of agricultural im-
provement, industrial development and social welfare began to make
their own investigations into the conditions of local life. Certainly
by the second decade of the twentieth century, Fort St George
could expect to receive information and to have its governmental
duties performed by several men in the same administrative division.
It was thus in a much better position to know and to control what its
subordinates were doing.
The second path of bureaucratic reform which it followed led
towards the breakdown of the domination which the locality possessed
over government appointments. From about 1880, Fort St George
began to enlarge the cadre of deputy-collector, appointments to which
were made in the Secretariat rather than the district and members of
which were not allowed to serve in districts in which they had personal
interests. Between 1880 and 1920, the number of deputy-collectors
rose from 50 to 160.131 This enabled Fort St George to replace com-
pletely the influence of the 'local' huzur sheristidar with that of men
whom it was better able to control. Spurred on by this success, in 1910
it took away the rights to appoint tahsildars from the Collector's office
and vested them in the Board of Revenue.132 This allowed it to make
provision for the regular movement of tahsildars from taluk to taluk in
order to prevent the build-up of local commitments among them. In
1911, it proposed a complete reorganisation of all appointments to the
revenue department. Only men who were appointed by the centre,
after examination or vetting by an appointments board, could ever
aspire to hold revenue offices carrying a salary of more than Rs 100
per month — that is to be tahsildars or high-grade clerks.133 Beneath
130 M e y e r , Report on the Constitution of Additional Districts, Divisions and Taluks,
p. 2.
131
History of the Services of Gazetted and Other Officers in the Civil Department in the
Madras Presidency, Corrected to 1st July 1885 ( M a d r a s , 1885); ibid., 1920.
132
G.O. 1435 (Revenue) dated 17 June 1920. T.N.A.
133
G.O. 682 (Revenue, Confidential) dated 11 March 1911. T.N.A.

57
 The emergence of provincial politics

the central revenue department proper, Fort St George also attempted

to reform the village establishment — by giving its immediate superiors
more powers to punish it, by clipping many of its prerogatives and
even by considering an eradication of hereditary rights to appoint-
ment in it.134 Similar developments were taking place at this time also
in the judicial and police departments. The swelling importance of
formal judicial process, to be seen in the rapid growth of litigation,
gave the law courts an ever larger place in local life. The officers who
controlled the courts, the District Judges, sub-judges and district
munsiffs, all were appointed directly from the capital. Moreover, the
personal despotisms which some of these officials had exercised in the
past also were circumscribed: vakil certificates from the High Court
steadily replaced the pleadership certificates, issued at the discretion
of judges, which previously had provided entry to local bars. In the
police department, reforms following the report of Curzon's police
committee established a new cadre of centrally appointed sub-
inspectors to fill the yawning gap between the local village police and
the existing provincial police 'umbrella'. This move brought agents of
the central government very nearly into the heart of village India.
Of course, the political effects of the progressive centralisation of
the bureaucracy were not as straightforward nor as immediate as the
British would have liked. Shortages of men, money and time pre-
vented Fort St George from fulfilling wholly the new role which it
had set itself. Although by 1920, for example, it had managed to move
revenue officials around so that only eight tahsildars and eleven deputy
tahsildars had served in their posts for more than three years, little
had been done to turn tahsildars into members of a truly provincial
cadre. Seventy per cent of them continued to serve in the districts in
which they had been recruited and which were their homes.135 More-
over, local society's almost infinite capacity for obstruction and resis-
tance was never completely overcome. As we shall see in Chapter 4,
the theoretical revolution in government did not become fully actual
at the local level. Nonetheless, certain important political changes
did follow in the wake of the reforms. In the first place, some areas
of decision-making were shifted to higher levels in the bureaucracy.
Local notables could no longer rely on their ability to deal with
officials who were detached from the bureaucratic chain but had to
negotiate with officials who were firmly linked to Madras city.
Naturally, they had to redirect their efforts to influencing their new
masters' home base. Secondly, and of much greater importance in
134
See below pp. 152-9.
135
G.O. 1435 (Revenue) dated 17 J u n e 1920. T.N.A.
 The governance of Madras

the short term, the career structure in which bureaucrats themselves

worked was drastically altered. For all those social groups interested
in careers in government service, the Secretariat rather than the
huzur sheristidar's verandah became the centre around which they
had to turn. This meant that the career structure of the civil service
was put on a provincial basis and that men from every district had to
compete against each other for jobs. It greatly broadened the field of
conflict for places in the government and provoked wider patterns
of alliance and opposition than could have been seen in the 'office
cliques' of old. Interestingly, however, the political techniques used to
influence the nodes of the appointments system did not change for the
principles of appointment themselves had not changed. In spite of the
hard facade of efficiency which it now put to the world, the Madras
Government did not replace appointment by personal discretion and
favouritism with appointment by impersonal ability. During the
years of the Statutory Civil Service (1883—93), for example, it con-
sistently preferred candidates who had done poorly in the civil service
examination to candidates who had done well.136 Equally, after 1912 it
happily ignored the results of its 'competitive' revenue inspector
examination; between 1912 and 1917, 87^ per cent of the men ap-
pointed as revenue inspector had not even sat the papers.137 Fort
St George's liking for 'the old ways' left the centre of the appointments
industry wide open for anybody who captured it to use as he pleased.
With control of jobs across the whole province rather than one dis-
trict at stake, by 1916 the fighting around the centre had become
bitter indeed.
A third set of reforms which drew native society towards Fort
St George was that which concerned the 'Indianisation' of the policy-
and decision-making centre. As we have hinted before, the Madras
Government was always chronically short of the resources necessary
to complete its task. Not the least of these shortages was that of British
manpower; it never had sufficient 'trusted' imperial officials to go
around. As the centre of the administration expanded and the number
of Britons willing to undertake a career in the Indian civil service
actually declined, it was inevitable that Indian subjects should move
136
' T h e plan is quite unintelligible, it is based neither upon the principle of competi-
tion nor u p o n the old system o f personal favouritism. If the appointment is to be
made according t o t h e results o f the competition, the candidate that stood first
in the examination should get it; or i f it is ultimately t o rest u p o n the favour o f
someone i n t h e Secretariat Council, there was n o honest purpose that the c o m -
petitive examination was intended t o serve.' Hindu 3 0 January 1885.
137
Proceedings o f the Board o f Revenue N o . 113 (Confidential) dated 2 3 October
1917. T.N.A.

59
 The emergence of provincial politics

to positions higher and higher in the bureaucratic machine. As early

as the 1860s, Madras civilians had found themselves confronted with
the (for them) terrible prospect of an Indian who was a full member of
the I.C.S.138 By 1890, the percolation of Indians into the higher grades
of service, both through the I.C.S. and the Statutory Civil Service, had
produced its first native Collector.139 Over the next thirty years,
Indian bureaucrats reached towards the heart of government and took
over dominant posts in Madras city. In 1914, L. D. Swammikannu
Pillai became Director-General of Registration; in 1915 R. Rama-
chandra Rao became a full departmental secretary; in 1918 Mahom-
med Bazlullah was appointed to the Board of Revenue. A similar
process was taking place in the judicial administration. The first
Indian High Court Judge was appointed in 1880 and by 1920 ap-
proximately half of the senior bench was composed of natives. The
Attorney-Generalship, Solicitor-Generalship and Provincial Public
Prosecutorship were all in Indian hands by the 1890s.
The lack of British alternatives, however, was not the only reason
for this development. Secretaries-of-State and the Government of
India, although not Fort St George, were acutely aware of the need to
temper the advance of their administration with political concessions
to their subjects. They were sensitive to the dangers both of rebellion
and mistake which could follow from driving forward policies without
the advice and consent of some of the people who were affected by
them. From 1861, the provinces had possessed Legislative Councils
on which natives sat, but in 1892 these were widened and provided
with an elected element. In 1909 they were enlarged yet again and in
1920 their elected members become a majority. London and Calcutta,
however, were not satisfied to leave natives only in advisory roles.
From the time of Ripon and Dufferin, with occasional pauses, they
pushed the provinces into providing natives with senior executive ap-
pointments in the government itself. In the 1880s, the idea for an
Indian Statutory Civil Service, parallel to the I.C.S. proper, came
from the Government of India;140 in 1909, Lord Morley insisted that
a native be appointed to every Governor's Executive Council and
given control of a number of departmental portfolios;141 in 1920,
Edwin Montagu turned more than half of every provincial govern-
ment over to Indian ministers.142 Just as Fort St George was coming
138
Pulicat Ratnavelu Chetty.
139
Madhavan Raja, who was a member of the family of the Zamorin of Calicut.
140
See Seal, Emergence of Indian Nationalism, pp. 1 1 7 - 1 9 .
141
Home Public A October 1908 Nos 116-46; Home Public A February 1909,
Nos 205-44. N.A.I.
142
Report on Indian Constitutional Reform. P.P. 1918, vol. vm.

60
 The governance of Madras

to take control of its presidency, so that control was coming to be

exercised more and more by men who were indigenous to the presi-
dency. By the 1910s, the classic colonial model of imperial master and
native subject was rapidly losing its appropriateness in the context of
the Madras state system. Indians were involved actively as well as
passively in the highest processes of government.
The fourth strand of administrative reform which was being spun
during this period also served to increase the scope of political power
available to Indians. This time, however, the area of its operation was
not the province as a unit but a series of localities. The task which
Fort St George had been set of raising more money naturally took it
to considering ways of developing the resources of the country and of
limiting the impact of famine and disease. Ideally it could simply
bump up taxation and order its bureaucracy to execute its orders. But
such an option was never available to it. It could not afford to enlarge
its bureaucracy sufficiently to undertake a whole new set of duties
which would include everything from clearing prickly pear to plan-
ning railway lines. Moreover, an attempt to raise the huge sums neces-
sary for development by methods of direct taxation would have led, as
it was finding with other forms of direct taxation, to a further deter-
ioration of its political position. The economic and human conditions
of Madras could not be improved by an administrative change. Luck-
ily, however, it was not necessary that they should be. While in the
other fields of reform which we have examined, Fort St George had to
exercise a greater personal control over the making and enforcement
of policy, in the matter of local government it was not so restricted. As
long as somebody could be persuaded that his interests lay in improv-
ing the transport systems, sanitation, water resources, educational
facilities and general living standards of the locality it did not have to
be government itself; although, of course, government would have to
remain in a position from which it could see that these ends were
being accomplished.
Since the 1840s, Fort St George had possessed some administrative
institutions for the maintenance of roads in the countryside and de-
fence against major health hazards in the towns but, following Lord
Ripon's local government initiative in 1882, these were expanded out
of all recognition.143 Under the Madras District Municipalities
and District Boards Acts of 1884-5, the foundations of a new admin-
istrative system was laid. In most large conurbations, municipal
councils were formed to allow local non-officials to come together
with officials and participate in local government. Similarly, out-
143
Report of the Committee on Local Self-Government in Madras. 1882.

61
 The emergence of provincial politics

side the towns, a system of district boards was devised to bring impor-
tant local notables into institutions of formal administration. Slowly-
in some cases painfully slowly — the bureaucracy was withdrawn from
these councils and boards and its place taken by electorates and elected
executive officers. The result was to leave behind a series of crucially
important political institutions, with considerable powers of taxation
and enormous powers of administrative interference, which spanned
large areas of Madras and in which control, for good or evil, could be
achieved by manipulating the machinery of primitive democracy. 144
In other fields of government also Fort St George was forced, by
financial pressure or political expediency, to transfer aspects of its
administration to committees of local notables. Between the 1890s
and the 1920s, local committees to enforce forest conservation, to con-
trol the siting and size of liquor shops, to hear appeals against the
income-tax, to select policemen, to settle communal disputes and to
control the distribution of water from irrigation schemes were set up
in many areas. According to the Royal Commission on Agriculture in
India (1928), Fort St George relied far more on the help of panchayats
to run its administration than did any other provincial government. 145
Technically, this handing-over of government powers to councils,
boards and committees was termed the decentralisation of the admin-
istration because the centralised bureaucracy no longer played a
part in it. Politically, however, its effects were the reverse of de-
centralisation. Whatever the theoreticians of British rule might have
thought, the bureaucracy proper had never been able to establish its
control over much of the administration which was being transferred
to the new institutions. Even where government officials had pos-
sessed some influence, as over urban sanitation or forest laws, they had
had to exercise it through the local notables to whom they were con-
nected and on whose power they rested. The networks which these
notables operated tended to be very small and personal and, con-
sequently, the arenas in which these powers had been exercised were
highly restricted. What happened when the government had finished
its institution-building, however, was that the powers passed to a com-
mon district or town centre, decisions in which affected a much
broader hinterland. The fragmented, personal empires of the local
notables became welded together at the point where they all jostled
for power on the same district board, municipal council or taluk com-
mittee. The locality of their operations had to be extended. Moreover,
their empires began to be pulled out even farther to touch Fort St
144
See below pp. 1 6 6 - 7 3 .
145
Royal Commission on Agriculture in India (H.M.S.O., 1928), xiv, 2 5 6 - 8 .

62
 The governance of Madras

George. When important political decisions were made at the level of

village and neighbourhood by notables who could count on their
local officials to protect them from the capital, the central influence of
government was minimal. Fort St George could not see what was
happening. When these same decisions were made in large-scale in-
stitutions which the Secretariat itself had designed, however, the op-
portunity for central influence to make itself felt obviously was much
greater. Fort St George wrote the rules by which politics in the new
institutions were played. As we shall see, it came to bear down heavily
on the life and times of many important local notables.
 3
The political economy of Madras

In our discussion of the nature and organisation of government in

Madras, we have noted the enormous reliance of the bureaucracy on
local notables. Lacking an effective coercive arm in the locality,
Fort St George's rule rested on the empires of control and author-
ity operated independently of it by some of its subjects. In this
chapter, we shall be concerned to probe the character of these em-
pires; our purpose will be to discover who ran them, how they were
run and how their internal structure and external relations were
changing over time. Clearly, such an undertaking is fraught with
difficulty for we are asking no less than why did some South
Indians obey others. A complete answer, if it were possible, would
have to follow Aristotle and travel from the household to the nation
and from the family to the culture and beyond. Such an answer could
have no end. Nor, more importantly, could it have a beginning. Some
social relationship would have to be examined first and hypothesised
as the base on which the other relationships are seen to act. If a base
itself were not defined, the relationships could then be seen only as
reacting to each other without priority and no specific point about
them could be explained. Indeed explanation would have to give way
to description alone. Yet to choose a base, be it religious culture or
economics, is to make an assumption about behaviour which can never
be tested but which determines the whole answer. Abstract arguments
about why Indian society operates in the way that it does by their
nature cannot be satisfactory.
The answers which we may offer to the question of obedience,
however, are not conceived in the abstract but generated by a con-
crete historical problem. We are looking for the men who could
exercise a sufficient measure of control over the behaviour of others
to support the tasks of government. Control, while it may be exer-
cised through the medium of persuasion or informal influence,
is most readily observable and most secure only when it is backed
up by sanctions. Our first task, then, becomes an investigation of
the types of sanctions available in non-official South Indian society.
Given that some sanctions can be demonstrated to have been more,
 The political economy of Madras

powerful and to have been used more regularly than others, we at

least have the possibility of finding objectively a set of relationships
which can be seen to be primary and with which other sets of
relationships (other sanctions) may be said to interact. Of course,
this model of sanctions will help us to see only the mechanics of
social control, not the reasons why men may have wanted to con-
trol others. It will be purely instrumentalist in approach. Of course,
also, it will slant the interpretation of social structure its own way.
South India could look very different if analysed to answer a dif-
ferent set of questions. The virtue of the model lies only in its
ability to provide answers to the specific problems of government at
this time.

The economic foundations of political organisation

The society of Madras was, by any standard, one of extreme poverty.
Its economic problems concerned questions not only of relative af-
fluence but also of starvation. In looking for systems of control in
political life, the systems of production and distribution must be taken
as primary for when they collapsed, as for example during a long
period of famine, they took with them most of the political order.
Servants commandeered the property of their masters; 1 markets were
looted and moneylenders murdered; 2 ancient factional and caste
rivalries, which had been suppressed for decades, flared up in open
warfare;3 religious antagonisms, which had been expressed in sym-
bolic rivalries, provided the ignition point for riots; 4 the crime rate
rose and outbreaks of banditry became common.5 Many of the social
relationships, the inferiorities and superiorities, which could be seen
in times of normality were overturned once grain pits became empty
and money worthless. Famines provided the occasions for the most
serious challenges to the stability of the Madras state. During the
shortages of 1918, most of the markets in Madras city were looted, one
quarter of the town was burnt down and the government had to call
1
Fourteenth Annual Report of the Sanitary Commissioner for Madras. 187 j (Madras,
1878), p. 10.
2
As after the poor harvests of 1890-2, G.O. 57 (Judicial), dated 17 January 1890,
G.O. 613 (Judicial) dated 25 May 1892. T.N.A.; and of 1918, Home Police B June
1918, Nos 158 and 241-2; Home Police B September 1918, Nos 274-83. N.A.I.
3
As between Komatis and Koyas, Madras Police 1896, p. 4; and Maravars and
Nadars, ibid., 1918.
4
As between Hindus and Muslims in Nellore, Hindu 18 October 1893; a n d in
Cuddapah, G.O. 1538 (Judicial) dated 1 July 1918. T.N.A.
5
See Madras Police 1877.

65
 The emergence of provincial politics

out troops to deal with disturbances in the mofussil no fewer than five
times.6 By contrast, in 1920, in spite of Gandhi's non-co-operation
campaign, troops were not required at all.7 Given the devastating
political consequences of economic breakdown, the best way in which
we can begin to analyse the socio-political structure of the presidency
is by examining the bases of economic organisation.

Agrarian society
The economy of Madras was dominated by the land. More than three-
quarters of the province's population followed directly agrarian
occupations, while most of the rest were involved in the administra-
tion of the land or the commerce and manufacture of its products.
The government drew most of its wealth from the countryside and,
if it were to rule at all, it had to accommodate itself to the organisa-
tions of rural society. Let us start our examination of economic
relations, therefore, by looking at the conditions of agrarian produc-
tion.
Unfortunately, our knowledge of the economic history of South
India under Company raj is extremely fragmentary: we possess few
systematic economic surveys from this period and have to rely largely
on the random comments of administrators. It seems likely, however,
that the economy was subject to severe pressure during the whole of
this epoch. Government revenue assessments were pitched more or
less at the monetary levels which they had attained during the wars
of the eighteenth century. Yet, with the coming of peace, grain prices
were falling dramatically.8 The effective demand of the state on the
surplus of the village was thus very high and, as N. Mukherjee has
shown, the economy stagnated.9 Moreover, famines savaged the
province with monotonous regularity.10 Between the late 1850s and
1876, however, there is evidence of an improvement in agrarian con-
ditions. Grain prices began to rise slowly, government revenue
demands were progressively lightened, road transport facilities were
improved, large irrigation works completed, new cash crops intro-
duced and cultivation extended.11 But famine remained a frequent
visitor and its arrival in the central area of'dry' cultivation in 1876
terminated this brief interlude of progress. By the most conservative
6
G.O. 2303 (Judicial) dated 11 October 1918; G.O. 444 (Public) dated 12 July 1921.
T.N.A.
7
G.O. 444 (Public) dated 12 July 1921. T.N.A.
8
Kumar, Land and Caste in South India p. 91.
9
Mukherjee, The Ryotwari System in Madras, pp. 253-313.
10
Ibid., pp. 314-53.
11
Kumar, Land and Caste in South India, pp. 115-16, 118-19.

66
 The political economy of Madras

estimate, 3^ million people died of starvation between 1876 and 1878

and the cultivated area in several districts was driven back by twenty
years.12
The years between 1880 and 1920, on which this book concentrates,
fit into the period of recovery and slow growth which the agrarian
economy enjoyed between the end of the great famine and the begin-
ning of the depression. Many of the strands of development which had
emerged in the 1860s and 1870s were picked up again and drawn out
much further than before. Although the threat of scarcity never
disappeared, and brief famines broke out in 1896 and 1918, there was
no sudden devastation on the scale of 1876-8. Nor was there any
sign of a serious rise in the human pressure on economic resources.
Between 1884-5 and 1921, the population increased by about 32
per cent and the productive area grew by almost the same amount.13
Certainly, some of this new land brought under the plough may have
been of marginal utility but changes in irrigation, crop patterns and
crop prices very much more than compensated for it. Although there
were no new projects to match the Kistna—Godavari irrigation works
of the 1850s, the area of wet land under ryotwari tenure (the only form
of tenure for which accurate statistics are available) increased by 25
per cent.14 On the basis of rough revenue accounts, wet land was
expected to be at least five times as productive as dry. Improvements
in the transportation system, particularly through the building of
railways, linked the provinces more closely to national and inter-
national markets. Cash cropping in cotton, oil-seeds, tobacco and
sugar developed more quickly than before and raised the profitability
of agriculture. Cotton and oil-seeds were especially important because
they flourished in soils capable of producing only the poorest quality
grains. Between 1884-5 and 1920-1, the acreages under the two
crops rose by 60 and 160 per cent respectively and came to cover 15-18
per cent of the entire cropped area.15 Better transport facilities also
freed food crops from local restrictions on sales and enabled them to
find their best prices in larger markets. Between the 1880s and the late
1910s, the increase in dry grain prices averaged between 80 and 100
12
Land Revenue 1874—5 and 1879-80; W. R. Cornish, 'The Influence of Famine
on Growth of Population' in Fifteenth Annual Report of the Sanitary Commissioner
for Madras, 1878 (Madras, 1879), pp. lxv-xci.
13
Calculated from Dharma Kumar's estimate of population in 1884-5, Land and
Caste in South India, p. 116; Census of India. 1921. Madras. Volume XIII. Part 2
(Madras, 1922), pp. 2 - 4 ; Land Revenue 1884-5 and 1921—2.
14
Calculated from Land Revenue 1885—6 and 1920-1.
15
'Area under Crops' in Agricultural Statistics of British India (Calcutta, 1884-5
and 1920-1), Vol. 1.

67
 The emergence of provincial politics

per cent, and in rice prices between 60 and 80 per cent.16 As increases
in the costs of manufactured and other non-agricultural commodities
were much lower than this, the real wealth of rural producers must
have increased. Moreover, the state was unable to take a large share
of the new profits in land taxation; as we saw, the antique revenue
system prevented it from raising its land revenue to keep pace with
inflation.
Of course, the prosperity of the period was not shared evenly
between the many geophysical regions which comprised the presidency.
Different climates, soil types and transport facilities produced a great
variety in the methods of production and in the commodities pro-
duced. Equally, an increase in the wealth of any one region did not
mean the fair distributing of it among all the inhabitants. In order to
obtain an accurate idea of the political implications of economic pro-
gress, it is necessary to examine the social and political structures
which controlled the creation and disbursement of wealth. Naturally,
these structures differed from region to region.

The dry zone. In the large central zone of dry cultivation (upland
Kistna, Godavari, Guntur and Nellore districts, the Ceded Districts
and hinterland Tamilnad), the most distinctive feature of agrarian
society was the economic dependence of the mass of the rural popula-
tion on a tiny elite of rich peasants. Brian Murton, in a recent article
on Salem and Coimbatore districts, has pointed to the fundamental
role in agrarian decision-making, which a few wealthy villagers pos-
sessed at the time of the British conquest.17 Although we lack a
detailed study of rural society between the conquest and the later
nineteenth century, there can be little doubt from the evidence avail-
able for the period of our concentration that no major structural
change had been wrought in the intervening years. In general, the
agrarian society of the 'dry' zone was as stratified in 1900 as it had
been in 1800; and, in some ways, the stratification was even more
clearly defined.
According to various censuses of the time, the proportion of the
workforce classified officially as 'landless labourer' and reliant entirely
on employment in the fields of others was from 1 o to 20 per cent of
the total.18 Such landless labourers were seldom more than predial
serfs who depended for their survival on the wages paid to them by
16
'Statement on Grain Prices' in Land Revenue 1880—1 to 1915-16.
17
B. J. Murton, 'Key People in the Countryside: Decision Makers in Interior
Tamilnad in the late eighteenth century' in Indian Economic and Social History
Review x: 2 (1973).
18
See A Statistical Atlas of the Madras Presidency (Madras, 1895, 1908 and 1924).

68
 The political economy of Madras

local landowners. The true number of economic dependents in the

rural areas, however, was very much higher than even these figures
imply. In 1900, under the ryotwari tenure which covered most of the
dry zone, 70 per cent of all the pattas issued by government were for the
payment of less than Rs 10 per annum in land revenue.19 The average
payment in pattas of this class was Rs 4 per annum. As many pattas
represented the joint holdings of more than one ryot, the actual
proportion of 'landowners' involved in this smallholding agriculture
was about 85 per cent.20 Few of these small cultivators could expect to
subsist on the products of their land alone. In the 1890s, a senior
government official estimated that eight acres of medium quality
dry land were required to keep a family through a good agricultural
season.21 The payment of Rs 4 per annum in land revenue represented
the ownership of only about four acres and, in the dry region, many
seasons were bad.22 In order to survive, the great majority of the
landowning population needed extra resources, over and above those
produced by its land.
These small cultivators also required a variety of facilities which
they could not provide for themselves if they were to work their own
lands. They needed water throughout the year in order to supplement
the miserable rainfall; and they did not have the capital to dig their
own wells.23 In many areas, they needed heavy ploughing equipment
to break up the soil; and again, they were unlikely to be able to afford
it themselves.24 As revenue demands fell heaviest immediately after
the harvest, they needed a market near at hand; they did not have the
cash-flow to meet their commitments while holding their crops for
storage and transportation.25 Perhaps most profoundly, however, they
needed credit facilities. Regular, albeit minor, crop failures wiped
out their long-term profits; the replacement of stock and seed neces-
sitated the immediate spending of sums which it would take them
years to save; and social conventions, such as dowries and wedding
feasts, required of them occasional, lavish expenditure. In most cases,
19
'Statement of the Rent-Roll' in Land Revenue 1900—1.
20
F . Nicholson, RAB, 1, 232.
21
S. Srinivasa Raghavaiyangar, Memorandum on the Progress of the Madras Presidency
During the Last Forty Years of British Administration (Madras, 1892), p . 7 5 .
22
Srinivasa Raghavaiyangar's o w n statistics indicated that rather more t h a n 75 per
cent of Madras landholders held less t h a n 5 acres. Ibid.
23
M.D.G. F . A. Nicholson a n d H . Stuart, Coimbatore (Madras, 1898), p p . 1 9 1 - 9 3 .
24
C. H. Benson, An Account of the Kurnool District based on an Analysis of Statistical
Information Relating Thereto and on Personal Observation (Madras, 1889), p . 6 5 .
25
Reports of the Provincial Banking Enquiry Committees (Calcutta, 1931), 'Madras',
p. 106. Hereafter, this volume as RPBC.
 The emergence of provincial politics

petty ryots 'could not begin to cultivate without borrowing seed,

cattle, grain for maintenance, etc.' 26
In the dry economy, the provision of these vital services rested
almost wholly with a small group of rich peasants. The theory of the
ryotwari settlement, which stressed a rough equality of peasant land-
ownership, had never become practice. In every district and every
'rural locality', there were families who paid the government upwards
of Rs 50 per annum in land revenue and whose broad acres contrasted
with the miserable plots of their neighbours. In 1900, the q\ percent
of pattas paying more than Rs 50 p.a. met 43 per cent of the total
revenue demand; and the 1 per cent paying more than Rs 100 met 14
per cent.27 The men who held these large pattas and possessed landed
resources twelve or more times greater than the average, stood at the
centre of the agrarian economy. They supplemented employment by
letting out lands on unprotected tenancies: in the villages of Bellary
district, which were investigated by the Cotton Commission (1925-8),
as many as 35 per cent of the local landholders were also the tenants
of other landholders. 28 They also hired labourers. Ryots with a con-
siderable surplus had the capital to sink wells and to buy heavy
ploughs, which they made available to small cultivators.29 At the
harvest, they bought much of the village produce on the spot and put
it into large storage pits. 30 Above all, they had the cash and grain to
pump into village credit in order to keep the economy turning over.
In 1895, F- A. Nicholson estimated that wealthy peasants were
responsible for at least 90 per cent of rural loans and pointed out that
ryots held more than 75 per cent of even written mortgages.31 Forty
years later, the Banking Enquiry's investigations indicated that there
had been no basic change.32 The Cotton Commission's examinations
also supported Nicholson's findings. Even on a cash crop like cotton
in the immediate vicinity of Adoni, the largest cotton-buying town in
the Ceded Districts, landlords were responsible for the majority of
loans to cultivators. 33
By providing these facilities for the continuation of agricultural
production, richer peasants gained a large measure of control over
26
Nicholson, RAB, 1, 232.
27
' S t a t e m e n t of t h e Rent-Roll' in Land Revenue 1900-1.
28
Indian Central Cotton Commission. General Report on Eight Investigations into the
Finance and Marketing of Cultivators' Cotton. 1925-8 (Bombay, n.d.), p . 50.
29
Benson, Account of the Kurnool District, p . 6 5 .
30
RPBC, p . 106; MPBC, in, 6 5 8 , 9 4 6 ; 'Report o n K u r n o o l ' i n Land Revenue
1902-3.
31
Nicholson, RAB, 1, 230.
32
RPBC, p p . 30, 79.
33
F o r 56 p e r cent. Only 27.3 per cent of t h e loans came directly from u r b a n sources.
Indian Central Cotton Commission, pp. 14—16.

70
 The political economy of Madras

the economy of their localities. Their dominance was particularly

tight because their own numbers were so small and because they faced
few rivals from outside offering similar services in the village. As we
saw, only 7^ per cent of pattadars paid more than Rs 50 per annum.
To make the point again but more concretely, we may take the
evidence of a witness before the Banking Enquiry. In his opinion,
only ryots who held a minimum of twenty acres of dry land were in a
position to market their own crops and to preserve some economic
independence.34 This was presumably good quality dry land, paying,
say, Rs 1.5 per annum per acre in land revenue (against the average
of Rs 1.18). If independence were possible at or above the level of Rs
30 p. a. in land revenue, then only between 5 and 10 per cent of the
landholders of each dry district were independent. Further, the number
of ryots who were large enough to go beyond mere self-preservation
and actively into rural moneylending and crop-buying must have
been even smaller still. Allowing for the fact that only about 70 per
cent of the rural workforce could be classified as a landholder at all,
the rich peasants were probably no more than 2 - 3 per cent of the
working population. In any village, clients seeking the facilities which
they offered did not have many patrons to choose between.
Moreover, non-village-based credit and trading groups had failed
to penetrate the dry regional economy and to mount a challenge to
the rich peasant's village hegemony. This failure was caused by several
factors. In the first place, there was no high volume of trade to sustain
specialist mercantile groups. The precariousness of agriculture and
of prices in international markets inhibited crop specialisation.35
Most villages harvested a variety of produce - grains and vegetables
as well as cotton and oil-seed. Most trade was localised between small
circles of adjacent villages and only a fraction of the produce of any
one 'rural locality' was ever exported.36 Similarly, the demand for
commodities from the outside was limited to a few luxury goods and
cattle.37 With so little scope for the development of broad trading
relations between the village and larger marketing structures, there
was little room in the village for the specialised merchant. In the
immediate vicinity of big towns, of course, urban merchants, seeking
supplies for their shops, might establish some more permanent village
connection.38 But beyond this, the most usual relationship which
34 35
MPBC, ii, 2 9 8 . RPBC, p . 14.
36
Ibid., p. 107; Royal Commission on Agriculture in India, xiv, 270.
37
Benson, Account of the Kurnool District, pp. 1 5 - 1 6 , 9 0 .
38
Such as the Nadar merchants around the Tinnevelly-Ramnad cotton towns, the
Komati merchants around the centres of trade in the Ceded Districts, the Devanga
merchants around the Salem and Coimbatore towns and the Vannigas around
urban North Arcot.

71
 The emergence of provincial politics

mercantile groups had with the countryside was only through the
carts which they sent out at harvest time to pick up the loose fraction
of the crop. In a basically subsistence economy, they were too far
removed from the sources of production to exercise any considerable
influence.
Secondly, although important merchant and moneylending com-
munities were attracted to those towns and railheads which were
expanding under the cotton and later oil-seed trades, the presidency's
failure to develop impersonal systems of credit prevented them from
pursuing these cash crops far into the hinterland. The poverty of
the soil in all but a few taluks, the ryotwari revenue system (under
which the state always had first call on the produce of the land), the
joint family property system and the delays of between three and
five years in effecting litigation combined to make land unviable
as a form of security.39 Few moneylenders would risk their moneys in
loans to people whom they did not know to be creditworthy or to have
assets more readily realisable than land. Obviously, few cultivators
were likely to be known as creditworthy, or at all, outside their
immediate circle of villages or to have assets other than land. Mercan-
tile groups working out of towns with their own resources of credit,
therefore, tended to approach rural localities through the larger
landholders whose existing capital and likely surpluses made them
good risks.40 In this way, although they can be seen to reach the
village in some areas, they were doing so only through the rich peasant
who was established there already. Their influence ran horizontal
to and helped to increase his own; it did not lead to the development
of a parallel and rival system of primary economic control in the
countryside.
Without serious competition, the rich peasant could charge heavily
for his services. His tenants were unprotected and their rent was
invariably high and taken in kind.41 His labourers' wages, though
difficult to calculate in money terms, were never seen by contem-
poraries as higher than subsistence level.42 When he bought in the
harvest market, the temporary glut of produce guaranteed that he did
so at rock-bottom prices.43 When he lent grain and money, he obtained
extortionate interest rates and could often force on the debtor an
option to buy his whole harvest at a prearranged, low price.44 In effect,
39 40
RPBC, pp. 7 8 - 8 1 , 1 7 3 - 8 2 . Ibid, pp. 8 7 - 9 .
41
Customarily, between a half and two-thirds of the gross produce.
42
Fourteenth Annual Report of the Sanitary Commissioner for Madras. 18yy (Madras,
43
1878), pp. 7-9. RPBC,p. 106.
44
Nicholson, RAB, 1, 232; RPBC, pp. 79, 106; Benson, Account of the Kurndol
District, p. 116.

72
 The political economy of Madras

individual rich peasants operated networks which ran through the

village economy and held large numbers of their neighbours in a
variety of conditions from simple client to debt-bonded serf.45 Often,
in a single lifetime, a villager would pass from the higher links in this
chain, where he had some freedom to manoeuvre, to the lower, where
he had none at all:
The Sahukar charges his own rates of interest as the ryot can no longer
bargain with him: what is worse the ryot has next to plough the lender's
field gratis and to do any other work at his bidding. The younger members
of the family, the sons and brothers, are sometimes engaged as the private
servants of the Sahukar without payment and in partial payment of the
amount borrowed.46
Through his network, the rich peasant came to dominate the poor,
disorganised producers of his village and to enjoy the fruits of many
more fields than those nominally his own.
The nature of agrarian organisation in the dry zone raises several
serious terminological problems. According to Eric Wolf's classic
definition, one of the basic conditions of a peasant economy is that
economic decisions are made by and in the interests of separate family-
households of small cultivators.47 From what we have seen of the dry
region, this condition may well have been absent during the nineteenth
and the early part of the twentieth centuries. Through a variety of
'informal' economic relationships, the large resident landholder,
with a considerable surplus above his own subsistence needs, had at
his disposal means which he could use to direct the agricultural pro-
duction of most of his poorer neighbours. His will could determine
the economic decisions made by many separate family-households.
Whether or not he did make use of the powers available to him, of
course, is another matter and can be settled satisfactorily only by
further research using materials more local than those presented here.
Yet a number of contemporary commentators were convinced that he
was operating in this fashion. Sir Frederick Nicholson, for example,
noted how small landholders, deeply sunk in debt, 'are thus in the
worst cases little more than tenants of the lender who prescribes
what crops they shall grow and demands what terms he pleases'.48
45
'Under each rich ryot there will be a set of ordinary ryots who depend on him for
money. W h e n once a ryot goes t o a particular rich ryot for money t h e n a con-
vention is established t h a t t h e poor ryot is t h e client of t h e rich ryot.' MPBC, m,
664.
46
Ibid., p. 699. See also Report on the Famine in the Madras Presidency during 1896 and
1897 (Madras, 1898), 1, 50.
47
E . Wolf, Peasants (Englewood Cliffs, 1966), p . 14.
48
Nicholson, RAB, 1, 2 3 2 ; also, Benson, Account of the Kurnool Districts p . 116.

73
 The emergence of provincial politics

Certain paradoxes in the performance of the dry economy at this time

also support the case.
In a recent dissertation, Michelle McAlpin has shown that between
i860 and 1900, there was no significant correlation between the
annual price and the annual acreage of cotton in many districts of
Bombay and the U.P.49 Her explanation of this fact is that peasant
cultivators regarded their first priority as the building-up of a grain
store to keep them through a bad season and that only when they had
sufficient grain would they consider planting cotton. The price and
the acreage of cotton at any one time were thus but distantly related.
McAlpin's only analysis of a Madras district, however, revealed a
unique case of some correlation. Working on the statistics of cotton
price to acreage in Madras for a somewhat later period (1900 to 1940),
Dharm Narain also has noted the exceptional tightness of their rela-
tionship. Indeed: 'The overallfitof the two curves, plotted to scale so
chosen that their amplitudes of fluctuation about agree, is in fact so
close that the price factor alone would seem to account for most of the
change in area.'50 As our period lies across those of McAlpin and
Narain, Table 1 represents an attempt to calculate the coefficient of
correlation between changes in cotton price and cotton acreage, and
Figure 1 an attempt to plot their relationship in the districts
of Bellary, Coimbatore and Tinnevelly during the years 1884 to

Table 1
Partial correlation coefficients (zero order partials).
Bellary Coimbatore Tinnevelly
i 2 1 2 1 2
I . 1.0000 0 .6937 I. 1.0000 0 •5794 1. 1.0000 0.4465
S = o.ooi S= 0.001 S = o.ooi S= 0.001 S = o.ooi S = 0.007
2. 0.6937 1 .000 2. 0.5794 1.000 2. 0.4465 1.000
S = o.ooi S
i=0.00I S = o.ooi S= 0.001 S = 0.007 S = o.ooi
(A value of 99.0000 is written if a coefficient cannot be computed)
Columns i = percentage of acreage under cotton.
Columns 2 = price of cotton.

49
M . B. M c A l p i n , T h e I m p a c t of Railroads o n A g r i c u l t u r e in India: A Case Study
of C o t t o n ' , u n p u b l i s h e d P h . D . dissertation, University of Wisconsin, 1 9 7 3 .
50
D . Narain, Impact of Price Movements on Areas under Selected Crops in India.
1900-1939 (Cambridge, 1965), p. 40.

74
Figure i: Scatter diagram
of relationship of price to it Bellary district 1884-1914
cotton acreage in selected • Coimbatore district 1884-1914
districts 1884-1914 24
• Tinnevelly district 1884-1914

21

©18
0 ;

15
*
CD
C
CD

2 12

65 70 75 80 85 90 95 100 105 110 115 120 125 130

Index Numbers of cotton Prices (1873 = 100)
 The emergence of provincial politics

1914.51 As can be seen, the correlation is very high and the elasticity
of acreage under the influence of price quite remarkable. As with
Narain's statistics, they 'begin by enlarging the hope, and end up in
justifying the reflection, that price bears in an usually large degree
on the variations of Madras cotton area'.52
This correlation cannot be accounted for simply by removing the
grain storage model from agrarian operations. The cotton districts
of Madras were at least as, if not more, likely to suffer from bad
seasons as those of Bombay and the U.P. Landholders grew grain on
most of their land and many tried to keep three or four years' supply
ready to hand.53 The occasional, sudden drops in cotton production,
displayed in the figure, mark years of recovery after a poor harvest
when stocks of grain had to be replenished. The greater price respon-
siveness of Madras can be explained, however, if the implications of
the non-peasant economy are followed up. Obviously, in localities
of true peasant economic conditions, where virtually all cultivating
families participated in deciding the quantity of grain to be stored,
the interests of virtually every family would be represented in the
store. The amount of land kept under grain would then have to be
sufficiently large to feed everybody through a bad season. But in
conditions in which the large landholder made decisions for his
dependents, the interests of every cultivating family were unlikely to
be so well represented. The claims of some of the large landholder's
dependents undoubtedly would be more marginal than those of others
and all his decisions would be taken in his general interests. The
amount of land which he was prepared to allocate to grain in his
locality could well ignore the needs in famine of a large minority of
the population and hence could be smaller than in a peasant economy.
A decision made in this light would free a relatively greater acreage
of land to respond to prices in the cash crop market.
If this were how economic decisions were made in the dry zone,
it would also go a long way to explaining the terrible devastation of
the Great Famine. After a series of fairly good years, between 20 and
25 per cent of the population of the Ceded Districts was wiped out by
the failure of just two seasons' crops.54 This savage loss of life cannot
51
Calculated from 'Area under Crops' in Agricultural Statistics of British India
(Calcutta, 1884-5 and 1914-15) and Index Numbers of Indian Prices 1861-1918
(Calcutta, 1919). The price of cotton in the year previous to the cotton harvest
was used to determine the correlation.
52
N a r a i n , Impact of Price Movements, p . 4 2 .
53
Cornish, 'The Influence of Famine on Population Growth' in Fifteenth Annual
Report of the Sanitary Commissioner for Madras, 1878, p . lxxi.
54
Ibid., pp. lxviii-xci.

76
 The political economy of Madras

be attributed to some supposed surplus of people over production

needs, which had built up in the area over previous decades and was
waiting for a calamity to burn it off. The acreage under cultivation
and the yield of land revenue (indices of productive capacity) also fell
by between 20 and 25 per cent. 55 The only way that it is possible to
reconcile this startling human catastrophe with what we know of
economic organisation is to suppose either that the peasant cultivators
of these districts were extraordinarily indigent and careless of the
threat of famine under which they lived; or, more reasonably, that
they were not peasant cultivators at all but dependents and workers
of a small rural elite which treated some of them as expendable.
If we accept the likely hypothesis of the dominant decision-making
role of the larger landholder in dry Madras, we must also abandon
the terminology of the peasant economy. In the rest of this book, the
rich peasant of the dry districts will be termed the 'rural-local boss' —
a vague term but one which the complexity of its subject demands.
Poorer peasants will be known as 'agrarian dependents'.
Between 1880 and the late 1920s, the rise in grain prices and in the
demand for cash crops served greatly to increase the profitability and
the scope of the rural-local boss' empire. Having obtained cheap grain
from his estates and debtors and in the harvest market, he possessed
the capital resources to play the roulette wheel of price fluctuation
with success. Although improvements in transport were levelling out
seasonal and local price fluctuations, they had by no means eradicated
them. Even in a good year, such as 1892-3, grain prices differed from
month to month by as much as 30 per cent; 56 the annual average of
prices paid in bazaars in adjacent taluks could differ by more than
15 per cent.57 By storing and moving grain, the rural-local boss took
full advantage of these market distortions. 58 He was also in a position
to increase the value of his cash crops. The best prices for cotton and
groundnut were obtained only by those sellers who processed the raw
produce, carried it to the main railheads and drove a hard bargain
with the dallals, through whom the purchasing companies worked and
55
Calculated from Land Revenue 1874-5 a n d 1880-1.
56
G.O. 581 (Revenue) dated 9 August 1894. T.N.A.
57
M.D. G. K. N. Krishnaswami Ayyar, Statistical Appendix. Together With a Supple-
ment, to the District Gazetteer (191 j) for Tinnevelly District (Madras, 1934),
pp. 106-8.
58
F o r the role of the Reddi (peasant) caste in t h e grain trade of the Ceded Districts,
see Appendix to the Report of the Indian Famine Commission, 1898, being Minutes
of Evidence, etc. Volume II. Madras Presidency, p. 101. P.P. 1899, Vol. xxxn;
Report on the Famine in the Madras Presidency during 1896 and 1897 (Madras,
1898), 1, 48; 11, 139.

77
 The emergence of provincial politics

who combined to keep down prices.59 Obviously, the agrarian depen-

dent was too poor and too weak to manage these feats. His cash crops,
if they were still his own at harvest time, were sold in the village or in
the ground for a minimal return. 60 But the rural-local boss, with
money for processing and carting and with the market power lent him
by the tons of cotton or groundnut which he could bring to or with-
hold from the town auction block, was able to maximise his profits.
Indeed, by the 1920s, some of the wealthiest ryots were investing in
their own cotton presses and groundnut decorticating machines in
their villages, and cutting out the dallals by selling directly to factories
and Bombay city through their own agents.61
By gaining entry to the world of urban and inter-district commerce,
the large landlord-moneylender-trader also derived a further range
of benefits. Once he became known as 'creditworthy' in the towns,
he could tap the relatively cheap supplies of urban credit to channel
more resources through his village network and thereby deepen his
local control.62 Alternatively, he could diversify his operations out
of simple village commerce and into more profitable ventures which
he might see from his town perch. Rural-local bosses financed mica
mining63 and railway, military and civil contracting for the govern-
ment 64 and aided the development of provincial banking facilities,
particularly after the Cooperative Credit Act of 1904.65 They were
also behind the explosion of cotton mills in the Tinnevelly—Ramnad—
Coimbatore cotton belts in the late 1920s.66
Our investigation of the mechanics of production and marketing,
59
RPBC, p p . 112, 1 2 3 ; Royal Commission on Agriculture in India, xiv, 2 6 8 .
60
Indian Central Cotton Commission, p . 2 1 .
61
RPBC, pp. 108, 112, 123; MPBC, 11, 36, 50-51, 62, 298-9; in; 319, 750, 966,
972.
62
RPBC, pp. 87-9; MPBC, 11, 62, 124, 210, 318, 593.
63
See biography of K. Audinarayana Reddi in Reforms (Franchise) B, March 1921,
Nos 34-99. N.A.I.
64
F o r example, A. Dhanakoti Mudaliar, w h o came from a rich Palli landowning
family in N o r t h Arcot, became o n e of t h e largest railway contractors in South
India. By t h e 1880s, h e was a member of the M a d r a s city Corporation. Also see
biography of M. Venka^araghavulu Reddiar in Hindu 19 M a y 1919.
65
See biographical notes on G. Eswara Reddi (p. 751), C. S. Ratnasabhapati
Mudaliar (p. 609) in V. L. Sastri (ed.), Encyclopaedia of the Madras Presidency
and the Adjacent States (Cocanada, 1920); also on P. S. Kumaraswami Raja (p. 144),
V. K. Palaniswami Gounder (p. 196) and K. A. Nachiappa Gounder (p. 176) in
Directory of the Madras Legislature (Madras, 1938).
66
F o r example, t h e family of A. K. D . D h a r m a Raja in R a m n a d district and of V. C.
Vellingiri G o u n d e r in Coimbatore district. See A. K. D . V e n k a t a Raja, A Brief
Life Sketch of P. S. Kumaraswami Raja (Rajapalaiyam, 1964); N . P e r u m a l , Talented
Tamils (Madras, 1957), p . 82.

78
 The political economy of Madras

therefore, strongly suggests that the rise in grain prices and the
development of cash-cropping led to the increasing stratification of
rural society in the dry zone. Large landholders were able to take
advantage of the new market situation in ways denied to their poorer
neighbours and, indeed, to reap the profits from many of these
neighbours' cultivation. But against what statistical evidence can we
test these findings? Of course, it is by now a cliche that Indian economic
statistics of this period are notoriously unreliable. Indeed, it is for the
reason of their difficulty that this chapter has thus far largely pro-
ceeded by deducing behaviour from a model derived from the written
evidence of contemporaries. Nonetheless, the statistics do exist and
unless an argument can square them with its conclusions, or at least
explain why they cannot be squared, it cannot be a satisfactory con-
tribution to economic history. The statistics of agrarian activity in the
Madras Presidency are among the least useful of all those taken by
provincial governments - they contain no data on actual landholding
or tenancy, only on revenue payment, and have few categories of
commercial information. By examining them carefully, however,
we may see that even they would lend some support to our find-
ings.
Changes in the ownership of land would only begin to give us an
accurate gauge of the movement of wealth if individuals tended to
express their economic progress in the accumulation of legal titles
to land. Yet, with land usually worth no more than the crops and
people on it (which the large landholder controlled already), with
litigation difficult and with better ways available of using money
than pursuing formal landed possessions, there was little incentive
to demonstrate wealth in this way. It was important for the large
landholder to continue to hold some land but less important for him
that this was 300 rather than 200 acres in extent. The fact that the
vast majority of rural loans were unsecured indicates that the constant
acquisition of land did not motivate the vast majority of rural credi-
tors.67 Nonetheless, in the course of his transactions some land was
bound to come the large landholder's way and such change as there
was in landownership between 1886—7 and 1925—6 favoured the
growth of the larger proprietor at the expense of the smaller. Of
course, the number of wealthy landowners involved in commerce was
so small that this development, in terms of the total landholding,
may seem insignificant. In the period, the minute fraction of pattas
67
In fact, legal land transfers were among the lowest in British India and seldom
amounted to more than 1 to 11 per cent of the cropped area per annum. See 'Land
Transfers' in Agricultural Statistics of British India, Vol. 1.

79
 The emergence of provincial politics

paying more than Rs 250 per annum increased its share of revenue
from 4.3 to 6.7 per cent of the total, or by about 50 per cent.68 Yet in
the context of Madras landholding this was important. The division
of pattas between heirs, the growth in population and the extension
of cultivation by carving small plots of land out of the jungle and
obtaining pattas for them all had led to a reduction in the average
size of pattas. Revenue per patta fell from Rs 14.9 to Rs 10.6 and
acreage from 7.3 to 4.9. 69 That large pattas should not only hold
themselves against this trend but actually move in the opposite direc-
tion indicates the growing wealth of the large landholder. Moreover,
patta figures include only the lands held in one revenue village. They,
therefore, conceal the total holdings of men with land in two or more
revenue villages. As the large landholder, operating within a broader
marketing and credit structure, was much more likely to have lands
widely dispersed than the arrarian dependent, these statistics would
tend to minimise his real landed possessions.
We also may turn to data on wages for a second support to our
conclusions about the movement of rural wealth. As Morris D. Morris
has shown, the figures of the agrarian wage censuses are much too
fragile to build upon them grand theories.70 As wages were composed
of a multiplicity of gifts, moneys, foods and services and as the sampl-
ing techniques of the day were so defective, the official census of rural
wages was highly inaccurate. Nonetheless, hardly any of the censuses
taken in the dry zone indicated that wages were rising. 71 Moreover,
from other sources, there are few grounds for us to believe that the
level of wages in the dry zone did increase and some to suggest that it
fell. The origins of any supposed increase in wages must have been
the development of more labour intensive crops, such as cotton, and of
migration possibilities which reduced labour supply, for human and
land variables otherwise remained in remarkable equilibrium. How-
ever, although cotton was an important crop in this area, the acreage
under it was never very large. The highest percentage of a district's
land put under it was 25 per cent, during the 1917 cotton boom, and
the average for most cotton-growing districts was seldom above 18
per cent.72 Its influence on the overall use of labour, therefore, was
limited. Moreover, the other major cash crop of the region, ground-
68
Calculated from 'Statement of the Rent-Roll' in Land Revenue 1885-6 and
1925-6.
69
Ibid.
70
M. D. Morris, 'Economic Change and Agriculture in Nineteenth Century India' in
Indian Economic and Social History Review, 111: 2 (1966).
71
Kumar, Land and Caste in South India, pp. 163—7.
72
See Fig. 1.

80
 The political economy of Madras

nut, was labour extensive and the acreage under it came to surpass
that under cotton.73
The question of migration can be divided into two parts - overseas
movements and internal movements. Overseas migration reached
significant proportions in Madras as a whole and must have cut back
on the labour supply. Yet the bulk of migrants at this time came from
the southern coastal districts of which only Tinnevelly and Ramnad
were in the dry zone.74 The effects on the southern fringes of Madras,
therefore, are worth considering but, on the economy of the dry region
as a whole, must have been strictly marginal. Internal migration
also was of limited consequence. Except in the Tinnevelly—Ramnad
cotton belt, there were very few towns75 and those that there were did
not grow fast. Between 1881 and 1921, the urban population of the
presidency increased by only 55 per cent.76 Most of this expansion was
concentrated in the narrow Andhra deltas and Tinnevelly—Ramnad
cotton belt and, again, although it may have influenced the fringes
of the dry zone did not affect the region as a whole. Over most of the
dry area, towns grew little faster than the total population. Migration
between rural areas, which probably had been large in the pre- and
early British periods, was noticeably small at the end of the nineteenth
century: on the night of the 1881 census, the astonishing proportion
of 95.6 per cent of the presidency's inhabitants were recorded as living
in the districts in which they had been born.77 Once more, the bulk
of the mobile sector was to be found on the coastal fringes not in the
dry heartland.
Although we must discard the wage censuses, other official evi-
dence indicates that if wages were moving at all, they were moving
downwards. Several district officials noted that rural employers in-
creasingly were paying their labourers' grain wages at price-fixed
rather than custom-fixed rates.78 During a period of rising prices, this
change could only be to the advantage of the employers.
In our present state of ignorance, exact statistical measurement of
alterations in the distribution of wealth may well be impossible to
make. However, there are good reasons for thinking that the economic
inequalities which existed when the British arrived in Madras widened
under their rule. Those who possessed large landholdings and an
agrarian surplus were perfectly placed to take advantage of new
73
'Area under Crops' in Agricultural Statistics of British India, vol. 1.
74
K u m a r , Land and Caste in South India, pp. 1 3 7 - 4 3 .
75
Census of India. 1881. Madras. Volume IV (Madras, 1883), p. 430.
76
Census of India. 1921. Madras. Volume XIII. Part 2 (Madras, 1922), pp. 8 - 1 2 .
77
Census of India. 1881. Madras. Volume I (Madras, 1883), p. 170.
78
Land Revenue 1908-9. p. 74; Land Revenue 1911 -12, p. 7 1 .

8l
 The emergence of provincial politics

marketing opportunities. But those without much land or surplus

before the expansion of the market economy were able to derive
little benefit from it: subject to the constant threat of famine and
shortage, they remained poor and were drawn progressively into
greater dependence on their rural-local bosses. The political implica-
tions of this economic situation are not difficult to see. The rural-
local boss' control over many of the staples of life permitted him
to exercise a pervasive social power. He could rely on his agrarian
dependents to provide him with servants, retainers and an army for
use in local political squabbles and 'all kinds of litigation'.79 Their
existence enabled him to commandeer many of the public resources
of the rural economy, such as grazing rights and irrigation channels,
which he could add de facto to his estate.80 He thus brought further
sections of the rural population under his shadow. Also, it was essenti-
ally his wealth, or the wealth which he controlled, which maintained
many of the non-productive groups in village society: priests, music-
ians, barbers, temple-servants and festival functionaries ultimately
were dependent for their livelihood on his largesse and sufferance.81
The provision of this patronage gave him influence over the ceremon-
ial and religious features of life in his locality. The central political
position of the rural-local boss is obvious from two simple facts:
firstly, it was to him that the vast majority of his neighbours brought
their petty disputes for arbitration;82 and secondly, it was often on,
or in deference to, the rituals of his caste group that subordinate
social groups modelled their own customs.83
In many ways, it was to maintain and extend this social control
that the economic activities of the rural-local boss were aimed. Our
emphasis on the economic mechanics of his empire ought not to be
taken to mean that he was a crude creature of profit to whom the social
dimensions of power were adiaphora. That would be to confuse our
discussion of means with one of ends. There can be little doubt that
the rural-local boss used his economic position as an instrument of
his political power and made his decisions more on the grounds of
the people who would be brought under his sway than of the rupees
79
MPBC, m, 699.
80
In the dry areas, pettandar rights of supervision over communal village resources
were almost uniformly in the hands of the wealthiest 'cultivators' of each village.
See Report of the Forest Committee. 1913 (Madras, 1913), 11, 164, 309, 400, 4 4 0 - 1 ,
555-
81
T h a t is to say he controlled a large part of the grain heap from which they lived.
82
See Report of the Forest Committee. 1913,11, 148, 156.
83
See, for example, F . A. Nicholson, The Coimbatore Manual ( M a d r a s , 1887), p p .
57-61.

82
 The political economy of Madras

he would be able to put in the bank. In the words of a contemporary:

'His power and prestige must at all costs be secured by having a large
number of village people at his disposal. Considerations of his im 1
portance influence the advance of money rather than profit from
usurious rates of interest.'84 This commitment, to his role and
status as a leader of men, naturally put limitations on the things
which he could do. In famine times, for example, he could not sell off
all his grain store even when prices reached absurdly high levels. If
he emptied his grain pits, he destroyed the means as well as the visible
sign of his power and was liable to be faced with a mutiny of his
dependents.85 Equally, he could not sacrifice his local involvements
entirely to the lucrative commercial opportunities of the market
towns without losing his social position. Richer though he might
become, his grasp on the network which converted riches into social
power would have weakened. Consequently, it is not surprising that
it is not until the period of the depression (1930-3), when economic
catastrophe had already smashed the network, that we find rural-local
bosses switching their interests completely to the towns rather
than using urban resources largely for the extension of their local
businesses.86 Economic considerations always were subordinate to and
an instrument of political activity.
Nonetheless, the nature and shape of instruments greatly condition
what can be done with them and changes in the organisation of the
Madras 'dry' economy certainly altered the social and political rela-
tions which they served. The large landholder's wealth increased,
separating him further from the poorer cultivators of his neighbour-
hood. In the context of the rural locality, his advance may scarcely
seem worthy of notice: he had been dominant before 1800 and his
position in 1900 was only marginally stronger. Yet in wider context,
the greater development of the commercial economy was important
for it changed his economic character. Unlike his agrarian depen-
dents, he was no longer trapped in an economy bounded by the rural
locality but was reaching out to district, national and international
markets. He was in some senses a different economic being. Naturally,
84
The quotation continues: ' . . . Nevertheless, the lifelong dependence of the
borrower upon the landlord and a variety of free services to be rendered to the
latter during agricultural seasons are features closely associated with the system.
It is not unusual that the smaller agriculturist borrowers are obliged to sell the
produce to the apparently obliging landlord.' MPBC, 111, 1034.
85
Fourteenth Annual Report of the Sanitary Commissioner for Madras, 18'yy
(Madras, 1878), p . 10.
86
C. J. Baker, The Politics of South India, 1920-19s7 ( C a m b r i d g e , 1976), c h . 3.
 The emergence of provincial politics

these economic changes also affected his social and political person-
ality.
The rural-local bosses whom we have been examining were almost
invariably participants of 'local-level' cultures, that is to say of cul-
tures characterised by their extreme physical restriction. Typically,
the rural-local boss belonged to a 'peasant' caste and was illiterate.
The wives he married came from villages only a few miles from his
own. Social interaction between members of the rural-local boss
elite barely existed beyond the level of adjacent localities. Before the
early twentieth century, of course, there had been little reason and
few means for this interaction to take place. Once clans had settled an
area, the processes of agricultural production and the lack of a major
market economy had tended to take them apart.87 Except for occasion-
al gatherings for ritual purposes, their lineages had no need of each
other. Moreover, subsequent political conquests of their territories,
particularly by Vijayanagar warriors, did not alter this situation
greatly. The warrior 'integrations,' of which Burton Stein has
written, 88 seldom involved men from the local level. They were carried
out in the locality much more by connection to or importation of
social groups who shared the warriors' own 'state-level' culture.
Although subjects within these larger integrations, rural-local bosses
(or agrarian decision-makers) were passive rather than active mem-
bers of them. State-level culture was strong only in the court centres
(the towns).
The growth of market opportunities broadened the framework of
rural-local operations. Wealthy landowning families slowly were
drawn out of their petty localities and towards the principal market-
ing centres. They met to do business and their progressive economic
enmeshment provided a basis for political and social connection. By
the 191 os, for example, groups of rural-local bosses in some areas had
begun to identify among themselves common political interests and to
act as pressure groups on rural government for the improvement
of commercial facilities.89 Equally, marriage networks among them
began to extend - thereby also extending the means by which they
could mobilise credit and other economic resources.90 Many of them
87
For discussions on the effect of settled agriculture on the lineage organisation of
the Kallars, see Dumont, Une Sous-Caste de Vlnde du Sud, pp. 7-9; also, M.D.G.
Francis, Madura, 1, 92-3.
88
Stein, 'Integration of t h e Agrarian System of South India' in Frykenberg (ed.),
Land Control and Social Structure in Indian History, p p . 175—216.
89
Hindu 1 F e b r u a r y a n d 17 M a y 1915.
90
T o w n s became t h e focal points of marriage brokerage. Royal Commission on
Agriculture in India, xiv, 2 3 3 . F o r discussions of the economic a n d political uses

84
 The political economy of Madras

set up houses in the towns and became more open to urban influences.
They started to send their children to school and college and to
pursue the learned professions.91 Some of them even became involved
in Sanskritic culture, for social status in the towns tended to be
expressed through the forms of the Hindu religion.92 As we shall see
later, these tendencies came to be developed also and more fully
by administrative change.
Admittedly, by 1920 this process was in its infancy and many of its
wider implications were hardly foreseeable. Yet what was happening
was slowly changing the face of South Indian society in this region.
Groups whose cultural-orientation had been local before were
beginning to participate directly in a larger 'regional' system. This
participation did not end the existence of purely local-level culture:
agrarian dependent groups were still held inside it. Nor, importantly,
did it simply lead to the absorption of the rural-local boss element
into state-level culture. The regional culture of the rural elite came to
be a hybrid which contained features of both: it brought Hindu gods
into the village and village gods into the towns, and mixed the symbols
of both cultures promiscuously. Indeed, the regional culture's contacts
were better with the locality for, following its progenitors' economic
and, as we shall see, administrative methods, it really used the re-
sources of the town to strengthen its rural position. As a hybrid,
however, it was not only sympathetic to both cultural levels but also
antagonistic to them. Its existence challenged the autonomy of each.
By the 1930s, under the pressure of economic calamity and rising
population, its development was to lead to fierce clashes with all
prevailing forms of South Indian culture. The spearhead of the Self-
Respect movement was formed out of young members of the 4regional-
ising' and 'urbanising' rural elite.

The Cauveri delta and Malabar. Along the Cauveri economic strati-
fication and the development of a 'regional' rural elite were both much
of marriage alliance among South Indian peasant castes, see C. M. Elliot, 'Caste
and Faction Among the Dominant Caste: the Reddis and Kammas of Andhra' in
R. Kothari, Caste in Indian Politics (New Delhi, 1970), pp. 129-71; also J. Manor,
'The Evolution of Political Areas and Units: the Lingayats and Vokkaligas of
Princely Mysore.' Privately circulated paper.
91
For example, the Kasu Reddi family of upland Kistna. See K. Kasipathi, Tryst with
Destiny (Hyderabad, 1970).
92
For example, although Gounder Vellala local culture may not pay much attention
to Aryan expressions of social status, in 1921 certain Gounder rural-local bosses
established a caste association to claim Kshatriya varna status for their caste.
Beck, Peasant Society in Konku, pp. 154-81; Hindu 12 August 1921, 28 January
1922, 18 December 1923, 10 and 27 May and 16 July 1924.

85
 The emergence of provincial politics

farther advanced than in the dry zone. Cauveri rice had been inten-
sively cultivated for the market for centuries when the British first
established their raj. The delta provided an agrarian surplus which
had made possible sophisticated regional government from the time
of the Cholas. These economic and political factors conspired to vest
effective land control over large areas in the hands of a few families.
In spite of avowals to the contrary, the East India Company did little
to alter the arrangements which it found. It recognised the hereditary
rights of the great mirasidar families to the ownership of their estates
under the ryotwari system and it allowed them in practice if not in
theory, to control wastelands.93 The number of large landholders was
much higher in the Cauveri area than elsewhere in ryotwari Madras.
In 1918 in Tanjore district, the heart of the delta, there were 3321
landlords paying more than Rs 250 p.a. in revenue, compared to a
district average of about 200 in the rest of the presidency and of less
than 80 in the dry region.94 Of these 3321, 421 also paid more than
Rs 1000 p.a. compared to a dry district average of 5.95 In 1918,pattas
worth more than Rs 250 met 30 per cent of Tanjore's total revenue
demand. In the dry zone, they met 6 per cent.96
The mirasidar's domination of agrarian production was complete.
At least in other ryotwari areas, the poorer peasant could carve a few
acres of land out of the jungle and establish some, be it only a nominal,
basis for independent existence; if he were not totally sunk in debt, he
might also manage to play off one rural-local boss against another in
order to reduce the terms of his dependency. In Tanjore, however,
neither of these two options was open to him. Land shortage and
mirasidar control of most known wastes prevented the mass of village
society from possessing even the sites of their own huts and long-term
debt bondage had reduced most cultivators to the status of hereditary
serfs.97 The mirasidar worked his fields through hired labourers and
unprotected, crop-sharing tenants who relied for life support on the
advances of wages, seeds, implements and credit which he allowed to
them. They were tied to his lands and had no means of lessening their
subjection. Indeed, until well into the twentieth century it was
common for labourers and tenants to be sold with an estate as part
of its capital equipment.98
93
K u m a r , Land and Caste in South India, p p . 14—23.
94
'Statement of the Rent Roll' in Land Revenue 1918-19.
95 96
Ibid. Ibid.
97
Royal Commission on Agriculture in India, xiv, 269; Swadesamitran 11 February
1912, Hindu Nesan 12 February 1912. R.N.P.; Kumar, Land and Caste in South
India, p p . 86—7, 178.
98
Kumar, ibid., p. 75. Royal Commission on Agriculture in India, 111, 314; see also,

86
 The political economy of Madras

There is no evidence to indicate that before the 1930s at the earliest,

the Tanjore economic systems underv/ent any fundamental change."
Within his estate, the mirasidar was the centre of commodity ex-
change: he drew off the produce of the soil and distributed food,
clothing, shelter and other necessities to his subordinates. He in-
sulated his workers from the external forces of the market. Contem-
porary observers noted the apparent timelessness of agricultural
operations in the districts. Money never became a significant element
in the composition of rural wages or rent agreements nor did the price
mechanism come to regulate the distribution of foods and services.100
Subordinate groups had been driven into a position of economic
slavery, the rules of which were written now by custom.
Yet if serfs and tenants were held at more or less the same economic
level, the prospects of the mirasidar himself were more expansive.
The development of Pax Britannica greatly raised the value of the
produce which he controlled. Many of Tanjore's old marketing con-
nections, which had been broken by the wars of the eighteenth
century, were repaired and new links, with Burma and Northern
India, were established. During the later nineteenth century, the
price of rice rose steadily and the mirasidar's purchasing power
followed it. He accumulated large stocks of paddy at the harvest,
which he had the capital to hull and mill and to transport directly to
the railheads and ports.101 Enormous grain warehouses were a com-
mon feature of the Tanjore countryside and, between 1910 and 1927
the number of power-driven mills (built mostly with mirasidar invest-
ment) leapt from four cr five to 215. 102 Tanjore became unquestion-
ably the wealthiest district in the presidency and its mirasidars by
far the wealthiest economic group.
The type of economic stratification which was appearing in the
dry districts was at its apogee in Tanjore, as was the type of social
stratification. The mirasidar elite ruled its agrarian dependents
with a rod of iron and established a great social distance between
itself and them. In Anantapur or Coimbatore, with the exception of
a few families, status differentials were only beginning to reach the
stage at which they made it obvious that the rural-local boss was
socially distinct from his poorer tenants, debtors and labourers. The
degrees of caste status, like the degrees of clientage, gradually shaded
B. Hjejle, 'Slavery and Agricultural Bondage in the Nineteenth Century' in
Scandinavian Economic History Review, xv: 1 a n d 2 (1967).
99
Royal Commission on Agriculture in India, xiv, 2 6 9 .
100
Ibid.; Land Revenue 1911-12, p . 7 7 .
101
Royal Commission on Agriculture in India, xiv, 268—9.
102
Ibid., in, 4 4 7 .

87
 The emergence of provincial politics

down from him through his various neighbours to a small group

of total dependents who were held outside the pale of decent society.
Along the Cauveri, the gap between respectability and untouchability
was much wider. Pariahs provided a very high (21 per cent)103 pro-
portion of the labouring population and the lesser Kallar lineages,
which also worked the fields, were considered barely clean. At this
lower end of the social scale, religion, culture and social custom were
very similar to those to be found generally in the dry region: the
most important gods were not those of the Hindu pantheon but of the
locality and the clan; marriage networks were territorially minute;
social relations were conducted within areas of a few square miles.104
At the top of the social scale, however, there were considerable
numbers of people claiming the highest forms of social status and
involved together in a uniform cultural system which the dry zone
could not match. This cultural system was not the hybrid, which was
to emerge in the dry region, but was emphatically state-level in all
its attributes. The much longer-term and higher volume of market
transactions had promoted a regional basis to superior economic
organisation in Tanjore. When making their initial settlements, the
British had not found dominant landholders buried in separate
localities: many mirasidars held land not in one or a few adjacent
villages but in widely dispersed taluks and their mirasi rights were
treated as freely saleable.105 Equally, much wider systems of central-
ised government had been established: members of the same state-
level families were littered across the entire district and beyond. 106
The high agrarian surplus enabled this greater breadth of organisation
to be matched by greater depth. Previous state integrations had not
stoppped short of local-level groups but had taken over the resources
on which their autonomy had been based. The Brahman population of
the Cauveri delta was six or seven times higher than elsewhere in
Tamilnad and was much richer.107 Brahmans formed about one-third
of the mirasidar elite.108 Other social groups which were logically
103
Census of India. 1921. Madras. Volume XIII. Part 2 (Madras, 1922), pp. 4 , 2 1 .
104
E. K. Gough, 'Caste in a Tanjore Village' in Leach (ed.), Aspects of Caste; M.D.G.
Hemingway, Trichinopoly 1, 88—9.
105
Kumar, Land and Caste in South India, pp. 15—16.
106
For example, Smartha Brahman 'service' families possessed long-standing kinship
links not only across the district but into North Arcot. See K. S. Ramaswami
Sastri, Vita Sua (Madras, n.d.) and K. S. Ramaswami Sastri, Professor Sunderama
Ayyar (Srirangam, n.d.).
107
Census of India. 1881. Madras (Madras, 1883), 11, 140.
108
Evidence taken before the Reforms Committee (Franchise) (Calcutta, 1919), 11,
119-23.

88
 The political economy of Madras

state-level also maintained a major local presence and power; for

example, Shiyali taluk was dominated by Thondamandala Vellalas
who had migrated from Chingleput to serve previous regional govern-
ments.109 Under this much heavier pressure from above, elite mem-
bers of the local-level cultures had themselves been drawn out towards
the superior state. The Kallar aristocracy of Tanjore had long ceased
to resemble lesser, cultivating members of the Kallar caste. The
Vandayar family of Pundi and the Papanad family had cut most of
their connections with the local-level and were fully immersed in
the culture of the state.110
Although it continued to maintain its control over dependents
through rural-local economic means, the mirasidar elite itself was
not conditioned by the circumstances of the rural locality. Its pat-
terns of religious worship were broad. It was by no means unusual
for there to be ten or more temples or maths in a single taluk, each
drawing a following for its ceremonies from members of the elite
spread widely across the countryside. In the 1860s, of the twenty
largest religious festivals in the presidency, no fewer than eight took
place along the banks of the Cauveri.111 The culture displayed in
this worship was literate and involved veneration of gods of an all-
Indian pantheon. Equally, marriage patterns were very wide: Tanjore
Brahmans married over circles of one hundred miles or more, 112
and Vellala and Kallar aristocrats across three or four taluks.113 The
ritual and customs of the elite were determined less by lineage refer-
ence than by recognition and emulation of social models derived from
the Hindu scriptures.
A similar type of political structure was thrown up in Malabar
by very similar types of pressure. Space prohibits us from investi-
gating it fully here. But we may note that economic control to the
point of continued slavery was exercised by a very small group of
jenmis and kanomdars; that social patterns among agrarian dependents
more or less retained their local-level characteristics; and that
economic and political factors of long duration had produced a high
109
See C. Krishnamurthi Mudaliar, Life and Activities of K. Chidambaranatha
Mudaliar (Shiyali, 1938).
110
They were deeply involved in Brahmanic festivals and culture. See, for example,
Hindu 12 and 14 September 1916.
111
Letter No. 749 A, Sanitary Commissioner for Madras to Chief Secretary, Govern-
ment of Madras, dated 11 July 1868. Copy in Cambridge University Lib-
rary.
112
Gough, 'Caste in a Tanjore Village' in Leach (ed.), Aspects of Caste, pp. 4 4 - 5 .
113
For the Kallar Odayars, see Hindu 15 June 1888.
 The emergence of provincial politics

degree of integration among the elite whose culture was strongly

state-level in orientation.114

The Kistna and Godavari deltas. The shape of economic relations

in the Kistna and Godavari deltas was in sharp contrast to those in
the dry region and in Tanjore and Malabar. No small oligarchy existed
to dominate agrarian production and its absence was a crucial factor
in the organisation of society. The Kistna and Godavari deltas were
effectively the creation of Sir Arthur Cotton in the 1840s and 1850s.
Certainly, there had been irrigation works on the rivers for centuries
before this, but the area of irrigated land was never more than a few
thousand acres. Cotton's anicut schemes increased it to over one and
a half million acres.115 By the 1880s, the region had joined Tanjore
as a centre of intensive rice cultivation, which sold its produce in
international markets. At harvest time, 'the whole country [looked]
like a single rice field, the groves around the villages, the roads and
avenues and the white sails of the boats gliding along the main canals
breaking the uniform sea of waving green crop'. 116
The economic changes which the sudden arrival of irrigation
made to the deltas were considerable. As the size of landholdings
indicates, patterns of property ownership in this previously dry,
or at best mixed, farming area were similar to those in the dry region
as a whole. In 1880, for example, the average patta in the wet zone
of Godavari district covered about 8 acres of land, and in dry Madras
about 8 acres.117 Equally, prior to the construction of the anicuts,
marketing arrangements had been fragmented and the volume of
trade, reflecting the nature of the subsistence economy, had been
low.118 The expansion of irrigation facilities both altered the signi-
ficance of landholding size and promoted new systems of marketing.
Under dry cultivation, 8 acres of land would pay somewhere between
Rs i\ and Rs 18 p. a. in revenue and its cultivator would be somewhere
between poverty and beggary. Under wet cultivation, however, 8
acres would pay between Rs 36 and Rs 70 p.a. and its cultivator would
114
See Kumar, Land and Caste in South India, pp. 7 0 - 5 ; M.D.G. C. A. Innes,
Malabar (Madras, 1908); G.O. 1675 (Home, Misc.) dated 2 December 1919.
T.N.A. The 'Cauveri' pattern also extended to patches of irrigated cultivation
along the Coromandel coast, in Chingleput and South Arcot districts, and to small
pockets of tank and river cultivation in several other predominantly dry districts.
115
A. V. Raman Rao, Economic Development of Andhra Pradesh (1766—1957)
(Bombay, 1958), pp. 86-90.
116
Quoted in Spate, India and Pakistan, 690.
117
Calculated from Land Revenue 1880-1, pp. 10, 34.
118
Selections from the Madras Records. XXII (Madras, 1870), pp. 62, 264-5;
ibid., XXXII (Madras, 1872), p. 78.

90
 The political economy of Madras

be a substantial, if not particularly rich, peasant. Moreover, the com-

ing of water had obliterated subsistence farming. The deltas special-
ised in the production of rice to the exclusion of all other crops. In
every village, a majority of the harvest had to be sold away and
everything necessary for life except rice imported.119 This greatly
stimulated trade. Cotton's anicuts produced an agrarian revolution,
be it only in one corner of the province.
In the Godavari district, only 4 per cent of ryotwari wet land was
held in pattas of less than Rs 10 p.a.120 Over 62 per cent was held
in the middle range of pattas from Rs 30 to Rs 250 p.a.121 The deltas
had a large number of peasants whose wealth was sufficient for them
to be able to provide their own cash and equipment for cultivation,
to hoard their crops after harvest and to meet transportation expenses.
They were freed from many of the production and marketing re-
straints so common elsewhere. Many delta peasants carted or sailed
their paddy straight to the principal railheads and ports. Those who
sold in their villages, to local millowners or export agents, did so
from choice rather than necessity. The Royal Commission on Agri-
culture (1928) noted that the resources of the producers and the
multiplicity of outlets available to them put Andhra rice in a sellers'
market. There was a more or less uniform regional price structure:
prices in the village were based closely on prices in the principal
exporting towns, which in turn were related to the prices of rice
being imported into the presidency.122
The development of this substantial 'middle' peasantry in the
Kistna and Godavari deltas profoundly influenced the economic
structure of the region. In the dry zone we saw that only small groups
from the village made contact with the main towns and that they
brought or took away only small quantities of goods. Most commodity
exchange took place between circles of villages. In consequence, the
dry region had many small bazaars: in 1929, Coimbatore district
possessed 144 licensed markets and Salem 112.123 The major towns,
however, grew very slowly. Although Tanjore possessed a much
higher volume of trade, its individually powerful mirasidars managed
produce-collection and commodity-redistribution from their estates.
They undermined the economic functions of the towns, which
stagnated.124 In the Andhra deltas, many landlords were eager to
119
Royal Commission on Agriculture in India, xiv, 268—70.
120
Calculated from 'Statement of the Rent Roll' in Land Revenue 7900-1.
121
Ibid.
122
Royal Commission on Agriculture in India, xiv, 268-70.
123
RPBC, pp. 219-20.
124
Census of India. 1921. Madras. Volume XIII. Part 2, pp. 8-12.

91
 The emergence of provincial politics

enter superior marketing organisations but few were large enough to

do so directly from their fields. Towns at the junctions of communica-
tions systems became important points for the assemblage of crops
and for the dispersal of imports. They served directly a broad clientele
from the countryside around them. As a result, marketing patterns
became heavily concentrated. In 1929, for example, Guntur district
had only 6 licensed markets.125 The transport towns of deltas grew
very rapidly: Rajahmundry, Cocanada, Ellore and Bezwada doubled
in size between 1880 and 1920.126
To complement the villagers who were being drawn into the
towns, urban-based mercantile and moneylending groups were en-
couraged to move outwards. The rice trade which they were handling
was so profitable and rice land so valuable that they were prepared
to throw away their scruples about rural finance and to lend directly
to almost any cultivator. Komatis, working through family businesses,
broadened and deepened their village networks; Marwaris flocked
to the deltas and poured cash into agrarian credit.127 By the 1910s,
these traditional bania groups were joined by the more sophisticated
institutions of co-operative credit societies and joint-stock banks
which, also operating from urban centres, made a greater impact
on credit facilities in the deltas than anywhere else in the presidency.128
The centralisation of trade and credit made 'the urban hinterland'
the basic unit of economic activity, rather than the rural-locality or
mirasi estate.
A second crucial development in the area was the rise of a broadly-
based cash economy. In Tanjore and in the dry region, the cash nexus
was largely confined to relations between the mirasidar or rural-local
boss and the superior marketing structure. It was involved in few
of the transactions between the mirasidar or rural-local boss and
his own agrarian dependents. One of the basic reasons for this weak-
ness in the cash nexus, and for the continued importance of custom
and kind, was that the labour force of the two regions tended to be
immobile and tied socially and politically to particular wealthy
landholders. Under the commercial agricultural system of the Andhra
deltas, however, labour was highly mobile and relatively free. During
transplantation and harvesting, rice cultivation required a much
larger supply of labour than was locally available. From the 1860s,
poor peasants from the upland taluks and Vizagapatam began pouring
125
RPBC, pp. 219-20.
126
Census of India. 1921. Madras. Volume XIII. Part 2, pp. 8-12.
127
MPBC, in, 740-3.
128
See 'Three Investigations into Godavari District' in MPBC, v, 85-225.

92
 The political economy of Madras

into the area for these annual operations.129 They worked as day
labourers and returned to their homes at the completion of their
tasks. By the early 1900s, they were paid almost entirely in cash.130
The fluidity of labour and the purchasing power of labourers in-
fluenced the economic structure of the region in two ways. They
provided a further stimulus to rural—urban linkages for the labourers
both needed supplies and had the money to buy them from any source,
and they destroyed the rural-locality as a coherent unit of social
life, for labourers were constantly on the move and seldom established
close ties with any one locale or employer.
The social consequences of economic development marked off the
area along the Kistna and Godavari rivers from the rest of the prov-
ince. Some delta peasants, of course, were much richer than others;
but they could not use their riches to develop anything like the same
amount of social control as could their wealth-equivalents in the
dry zone or in Tanjore. In the first place, there were too many sources
of credit in the countryside and the income of the average cultivator
was too high for the wealthy peasant to be able to tie up the local
credit system. Banias and banks provided alternative places of succour
and were seen to do business even with the five acre ryot. Secondly,
the nature of the labour supply prevented the wealthy from establish-
ing long-term dependency relationships with subordinate, working
groups who could be used in their private standing armies. Thirdly,
and most significantly, there were also too many substantial men:
the clients of any one man could easily find a patron just as important
only a few fields away. This last point can be seen clearly in the mass
of evidence given to the Forest Committee (1913). It was extremely
unusual for witnesses from villages in dry Madras to name more
than three or four men in their localities who were sufficiently rich
and powerful to be used by the general population as arbitrators in
disputes. Witnesses from the deltas, however, frequently listed eight
or nine; and one man claimed there were no less than fourteen in his
own village.131 With so many rival jurisdictions packed into the
same area, the basis of rural factionalism was greatly widened and
the prospective client given considerable scope for manoeuvre. Social
order in the deltas was very fragile.
As might be expected in this situation, the scale of communica-
tions developed by the economy also opened up broad territorial
networks of social and cultural connection to a very large section of
129
Royal Commission on Agriculture in India, m , 3 1 6 ; ibid., xiv, 268—70.
130
Ibid., also Land Revenue 1902-3, p. 10.
131
Report of the Forest Committee. 1913,11, passim.

93
 The emergence of provincial politics

the rural population. Endogamous units among the Kammas, the

dominant peasant caste of the region, covered wide areas and ex-
panded from the later nineteenth century to the point at which the
concept of sub-caste became almost lost.132 Through regular urban
contacts, village families began to imitate the habits of townsmen,
giving the deltas the appearance of a common, regional life-style.
Bricks and tiles steadily replaced mud and thatch as the usual build-
ing materials for respectable village homes;133 the literacy rate grew
faster than anywhere else in the presidency;134 the delta towns
sustained the most active vernacular journalism industry in Madras,
which sold its wares extensively in the countryside;135 substantial
landlords began to send their sons to the towns for work and educa-
tion and, by subletting their fields, often moved there themselves.136
The merging of the town and the countryside had important
implications for the dichotomy between state- and local-level cultures.
As in the dry region, state-level culture had not deeply penetrated
the countryside before the middle of the nineteenth century and
had remained embedded in the towns. Ancient irrigation works on
the rivers had attracted some state-level groups to the area, who also
had obtained some power by acting as spiritual and administrative
advisors to various native governments. The Kistna and the Godavari
rivers did provide more centres sacred to Hindus than would be found
in the dry zone. Yet there is little to indicate that state-level culture
had made much more of an impact on the peasantry of the country-
side. Mid-nineteenth-century observers commented on the continua-
tion of local-level religious forms, while investigations in the 1860s
revealed that, although there were many Hindu festivals in the area,
none drew a significant regional following.137 Towards the end of
the nineteenth century, however, it was clear that the hiatus between
state and local culture was beginning to disappear with the hiatus
between town and country. Sudden and spontaneous religious move-
ments sprang up in the rural areas and rushed towards the towns.
132
See C. M. Elliot, 'Caste and Faction among the Dominant Caste: the Reddis and
Kammas of Andhra' in Kothari, Caste in Indian Politics.
133
R a m a n R a o , Economic Development of Andhra Pradesh (1766—1957), p . 1 9 2 .
134
Census of India. 1931. Madras. Volume XIV, Part 1 (Madras, 1933), P- 283; ibid.,
Part 2, pp. 276-7.
135
By 1925, Rajahmundry had at least five newspapers with a combined circulation
of 4700 and Masulipatam also maintainedfivewith a circulation of 8000. No other
mofussil towns possessed anything like this level of activity. Home Political File
261 of 1926. N.A.I.
136
Land Revenue 1905—6, p . 7 2 .
137
Letter No. 749 A, Sanitary Commissioner for Madras to Chief Secretary, Govern-
ment of Madras, dated 11 July 1868. Copy in Cambridge University Library.

94
 The political economy of Madras

During the mohurram of 1884, for example, a small army of peasants

gathered at Gudur, a village near Masulipatam, and marched through
the countryside picking up support until it reached Masulipatam,
where it attacked the main mosque.138 Conversely, temples dedicated
to all-India gods appeared in greater numbers in the villages as
Komati merchants, of urban origin, deepened their rural associa-
tions.139 Festivals, which once were patronised only by a select
few, became principal meeting places and markets for thousands.
Hindu theatrical and dance troupes, which once served only at the
urban courts of rich zamindars, found enthusiastic audiences in the
villages.140 Hindu theological controversy began to sweep the country-
side as Kammas noted, and disapproved, the place which they were
assigned in the varna scheme and demanded the right to study the
Sanskrit scriptures and to remodel their customs.141
In terms of the state—local separation which we have examined
in the dry region and in Tanjore-Malabar, the society of the Andhra
deltas produced a third variant. The entire distinction between the
levels increasingly lost its force. The movement towards the towns
was not carried forward by an elite in any meaningful sense. The
weakness of economic controls had eroded the bases of elite political
control: distinctions of rank and status were of relatively little impor-
tance in the deltas. The movement was carried forward by 'a mass'
of rural society. This meant that when regional interaction began to
take place very little of the local cultures was left in the hands of
non-participants and was untouched by the development: it was not
preserved by agrarian dependent groups. Equally, state-level culture
was heavily modified by the movement. State-level groups could not
dominate the new process: they did not command the resources of
rural power. Nor could they ignore it: they were a crucial part of the
urban hinterland as they were not of the rural locality. They had to
bend with it.
In consequence, the Andhra deltas brought forth a peculiar cultural
composite which consisted of a fusion of state- and local-level ele-
ments. The nature of this composite can be seen, for example, in the
way that Brahman priests propitiated local-level deities and in the
138
Madras Police 1884, p . 7.
139
Komatis were great patrons of Sanskrit scholarship and of Hindu revivalism.
They were particularly attached to the cow-protection movement. Report of the
Director Criminal Intelligence, 9 April 1910 in Home Political B June 1910,
Nos 17-25. N.A.I.; also G.O. 216 (L and M, M) dated 3 February 1914. T.N.A.
140
Such as the Rajahmundry Hindu Theatrical Company. See Sastri (ed.), Encyclo-
paedia of the Madras Presidency, p. 501.
141
Ranga, Fight for Freedom, pp. 7 - 1 1 .

95
 The emergence of provincial politics

way that peasants became involved in state-level controversies. In

the case of the Gudur-Masulipatam march and riot, which we
mentioned above, the procession was led by an image of the idol
Anjamar, a local-level god but one committed to a fight with an
opponent which existed only at the state-level of culture - Islam.
It can be seen again in the attempts by members of certain peasant
castes to Sanskritise their culture and in the attempts by members
of state-level groups to demoticise their culture: to match the Kamma
movement 'upwards', some Brahmans sought to return written
Telugu from the high literary dialect which it had become to the
language of common speech.142 And it can be seen yet again in the
mutual association of Brahmans, Komatis and Kammas in literary
societies in the towns and the villages.143
Although there was certainly room for internal conflict between its
participants, the regional culture of the Andhra deltas did not itself
stand between other antagonistic and separate traditions. It could
not clash with the state- and local-level cultures because these had
lost most of their existence: the Self-Respect Movement, for example,
was meaningless in the deltas. Nor did its continuance rely on the
maintenance of political control by a small elite. It represented a
highly organised, territorially wide and integrated 'popular' culture.
Its type was quite unique in South India.

Zamindars, mittadars and jenmis. The foregoing examination of the

structure of rural society has concentrated on methods of production,
marketing and credit. It has been concerned, therefore, with the
landholders, moneylenders, tenants and labourers who composed
rural society and were involved in the creation of rural wealth. The
patterns of power which it has outlined were those to be found at
various levels of localisation in the countryside. In about two-thirds of
the presidency there were no formal economic institutions above these
levels and between them and the state. In the remaining one-third,
however, political circumstances in the late eighteenth century, and
political folly in the early nineteenth, had led the British to support
the rights of intermediaries between the soil and themselves to a share
of the produce of the land. These intermediaries — zamindars, mitta-
dars and jenmis — were capable of commanding great wealth and, on
142
For example, the Brahman K. Viresalingam, essayist and journalist, did much to
create the language of contemporary Telugu letters and, in his lifetime, saw his
writings popularly acclaimed.
143
M . V e n k a t a r a n g a i y a , The Freedom Struggle in Andhra Pradesh (Andhra)
(Hyderabad, 1968), 11, 266-76.

96
 The political economy of Madras

occasion, were taken by the British to be the leaders of society in their

domains. Clearly, it is necessary for us to examine their role in the
countryside.
In Madras, zamindars came in all shapes and sizes and had been
produced by different forces; consequently, they were related to their
estates in a variety of ways. As P. J. Musgrave has shown for U.P.,
it would be impossible to provide a single stereotype of their economic
position.144 In this chapter, as our main interest lies in the nature of
production and credit, we shall begin our categorisation of the dif-
ferent types of zamindar by using as a criterion the ways in which they
interfered in the organisation of the economy. A first category would
be composed of those who interfered least, that is of those who drew
off rents and taxes but did nothing to provide credit or marketing
facilities for their tenants, to stimulate the cultivation of particular
crops or to reorganise the economy of their estates. In this category,
which was certainly the largest, we tend to find most of the greater
territorial zamindars. Such families as those of the Maharaja of Vizia-
nagram and the Rajas of Ramnad, Pithapuram, Kalahasti, Karvet-
nagar, Bobbili, Nuzvid and Venkatagiri inherited their place in
late-nineteenth-century Madras from ancestors who had been suc-
cessful warriors. The coming of peace, however, did not see them
beat their swords into ploughshares. They continued to maintain huge
retinues and a courtly life-style.145 Their perception of their social
role did not lead them to promote agricultural development nor to
finance the village economy. In the eighteenth century they had taken
loot by the sword; now they did the same, less successfully, by the law.
The estates of the larger zamindars were usually very backward. Their
rents were collected mostly in kind, thereby reducing the importance
of a market economy to their tenants; 146 their irrigation works seldom
were repaired or extended;147 their tenants were often indebted to
creditors outside their jurisdiction. Certainly, by the twentieth
century, a few of these ancient magnates had become more prudent
in business affairs and were attempting to increase their resources.
But none did so by developing closer relations with the peasant
producers of his estate. Bobbili, for example, found it more profitable
to use his grain rents as a basis for the flotation of a provincial grain-
144
P. J. Musgrave, 'Landlords and Lords of the Land' in Modern Asian Studies, vi: 3
(1972).
145
See A. Vadivelu, Ruling Chiefs, Nobles and Zemindars of India ( M a d r a s , 1915);
A. Vadivelu, The Aristocracy of Southern India ( M a d r a s , 1907), 2 vols.
146
Until the Estates Lands Act of 1908, zamindars were not compelled to commute
their rent demands from kind to cash.
147
Land Revenue 1905-6, p. 78.

97
 The emergence of provincial politics

dealing business;118 Parlakimedi speculated in railways;149 Venka-

tagiri in mica mining; 150 and Pithapuram in lending money to other
zamindars.151 In their lands, zamindars of this kind remained simply
rent-receivers. The economy of their villages was controlled effectively
by richer tenant or other groups. Indeed, some of these 'subordinate'
groups came to exercise a strong influence over the operations of their
zamindar's own economy. Kalahasti, Ramnad, Sivaganga and Karvet-
nagar, for example, were deeply indebted to their wealthier tenants
and ultimately had to alienate portions of their estates to meet these
commitments.152
The appearance of social power, which many of the greater
zamindars possessed, came much more from their ability to spend
huge quantities of money and to patronise state-level cultural groups
than from their ability to dictate the nature of production in their
realms. As much of their expenditure took place in the towns where
they lived, it is arguable whether patronage ties gave them any hold
over the countryside. Certainly, during famines they very rarely came
forward to guarantee their tenants against starvation, or even to remit
the revenues which were owed them.153 Perhaps, only in Ramnad
and other estates in the extreme South, which were based on the old
lineage territories of Maravar and Kallar warriors, is there much
evidence of the zamindar playing the role of rural leader - and even
here the role was becoming more limited every year.
In the same category of zamindar as the 'ancient' magnates, we
might also place those contractors and financiers who made fortunes
during the early years of Company raj and who bought privileged
revenue rights. In very many cases, they tried to imitate the life-style
of the warrior elite. As R. E. Frykenberg has shown in Guntur district,
these parvenus were more interested in status and ceremonial than in
economic resources: they did little to finance the rural economy and
the bankruptcy rate among them was staggering.154 A classic case is
provided by the Brahman contractor family who bought the Polavaram
estate in Godavari in the 1840s and 1850s. The family became great
148
Report on the Famine in the Madras Presidency during 1896 and 1897 ( M a d r a s ,
1898), 11, 7 3 .
149
Sastri (ed.), Encyclopaedia of the Madras Presidency, pp. 361—2.
150
Some of the richest mica deposits in India lay within the Venkatagiri zamindari in
Nellore district.
151
Hindu 7 December 1920.
152
Report on the Administration of the Estates under the Court of Wards in the Madras
Presidency for Fasli 1317(1907—8) (Madras, 1909); ibid., 1908-9; ibid., 1909-10;
Land Revenue 1911—12 to 1913—14.
153
Report on the Famine in the Madras Presidency during 1896 and 1897, n> 79-
154
Frykenberg, Guntur District 1788-1848, pp. 136-53.

98
 The political economy of Madras

patrons and cultural leaders in Cocanada town but, by 1919, had so

misused its funds that its estates had to be liquidated.155
Those zamindars who became closely involved with the economies
of their estates and who undertook supervision of production were
almost invariably of two kinds. Either they were small mittadars,
owning only one or a few villages, or they were bankers who had
obtained the position of zamindar by providing credit both to the old
estate holder and to his tenants. In Salem and Coimbatore districts
particularly, there were several tx-poligar families who had been
brought into the British settlements of 1799—1802 and allowed to
keep revenue rights over some villages. By the later nineteenth
century, they were operating economic networks in their localities
very similar to those of the rural-local boss. The mittadar of Kumara-
nangalam, who was one of the largest landowners in this group, lent
money, grain and cattle to his tenants and insinuated his influence
into every corner of his petty domain.156 Of the bankers who moved
into land ownership, the Nattukottai Chetties provide the best
examples of tight estate control. In their homelands, inside the
Ramnad zamindari^ they were a village-based group. Although they
drew an increasingly large proportion of their resources from urban
and international commerce, in Ramnad (and nowhere else in the
presidency) they financed local agriculture.157 This gave them ahold
over the rural population around them. They also lent money to the
Raja of Ramnad and eventually forced him to lease and then sell
zamindari rights over their land to them.158 The Nattukottai Chetty
zamindari of Devakottah thus exercised a remarkably powerful ruler-
ship over his tenant subjects. Indeed, his economic oppression was
one of the longest-standing scandals in Madras.159
Clearly, however, the territory covered by this second category
of zamindar was very small. In most of the permanently settled areas
of the province, the estate holder did not intrude into the processes
of production. The provision of facilities for the continuation of
economic life rested with others - usually with the wealthier tenant
whose local position was thereby enhanced. Few contemporaries
noted any fundamental difference between rural economic organisa-
tion in zamindari and in ryotwari areas, save that the zamindari
peasant was required to pay rather more rent rather less often than his
155
Hindu 7 December 1920.
156
Royal Commission on Agriculture in India, m , 547.
157
RPBC, pp. 185-90.
158
Sastri (ed.), Encyclopaedia of the Madras Presidency, p . 4 7 9 .
159
Land Revenue 1903-4, pp. 78-9; Hindu 6 September 1893.

99
 The emergence of provincial politics

ryotwari counterpart. 160 While this doubtless contributed to further

economic stratification, in that richer tenants were likely to pay even
less often than poorer, it cannot be taken to have created a distinguish-
able zamindari tenant society. As elsewhere, the critical level of
economic organisation lay in the village or small rural locality. Con-
sequently, inside nearly every zamindari there were local structures of
power which did not touch the official landowner and which could
oppose his leadership. These structures were moulded by the same
forces which influenced social organisation in the dry zone, Tanjore
and Andhra; and most often they were dominated by the rural-local
boss.

Urban Madras. The main stimulus to urbanisation in the Madras of

our period was trade in agrarian produce. In the first half of the nine-
teenth century, the administration had played a large part in attracting
people to towns to work in and around the bureaucracy, law courts and
schools. Some towns were founded almost wholly in this way. By
1880, however, the basic pattern of administration had been estab-
lished and not even Madras city, the capital of government, was grow-
ing fast.161 Equally, heavy industry had yet to appear on any scale
and there was little demand for urban industrial labour. The greatest
urban growth points were in those areas where movements in the rural
economy had been greatest — particularly, in the Andhra deltas and,
later, in the Tinnevelly-Ramnad cotton belt.162
Clearly, it is difficult to write generally of a category which con-
tained so many heterogeneous elements. The towns of Madras had
been created at different times, for different purposes and served dif-
ferent present needs. Some had developed around temples, some
around the palaces of princes, some around bureaucratic centres and
others still at the cross-roads of trade. Each possessed economic and
social organisations which were slightly different from every other.
Necessarily we must look at the vaguest outlines of economic and
social structure, for only in those outlines can we find material suitable
to our general ends. In our examination of rural Madras, we saw that
the economy required a variety of facilities to be available if it were
to keep running. The provision of these facilities was in the hands of a
few men who thereby were able to gain a large measure of economic
C, pp. 7 - 8 ; Srinivasa Raghavaiyangar, Memorandum on the Progress of the
Madras Presidency, p. 76.
161
Madras city's growth rate was lower than that of the presidency as a whole.
Census of India. 1921. Madras. Volume XIII. Part 2 (Madras, 1922), pp. 8—12.
162
Cenus of India. 1931. Madras. Volume XIV (Madras, 1933), pp. 10-16.

IOO
 The political economy of Madras

and ultimately political control. Let us first see what the nature of
economic organisation tells us of urban Madras.
In every town of any size, commerce and petty industry employed a
majority of the population. Both needed enormous resources of credit
to keep them moving: whether to pass large amounts of crop from one
town to another, to hold supplies for local sale or to invest in carts,
jutkas, stalls or even coffee-shops, traders and merchants required
credit facilities. For most of our period, there were few impersonal
organisations — joint-stock banks and the like — capable of meeting
these needs. Credit was supplied by individuals. Although money-
lending in the towns was a promiscuous activity - anybody with
a few spare coppers would lend them, and many people themselves
borrowed only to lend again - it is clear that the lines of credit tended
to lead back to the same few hands. The difficulty of accumulating
large funds and the specialist nature of money transfers over wide
distances produced heavy concentrations of credit in very small bank-
ing groups. For example, in Rajahmundry, an Andhra trading town
of some 20,000 population, there were in 1900 only some 100 men
who lent money professionally. Of these, two lent very much more
than the rest: they paid Rs 30 a year income tax, while only twelve of
the others managed Rs 6 and the remaining 86 none at all.163 In
Ellore, a comparable and expanding Andhra town, one man, 'Mothey
Venkataswami practically finances the trade of this place.' 164 Some
credit agents operated on a scale to dwarf all others around them. In
the late 1920s, for example, although they were withdrawing their
funds from India to pursue better opportunities in Burma and Ceylon,
the Nattukottai Chetties still had about Rs 11 crores invested in 243
Madras branches — an average of Rs 4.5 lakhs per branch, most of it
in urban moneylending.165 In Cocanada, at the same time, three
Komati families were working with a capital of Rs 41 lakhs,166 and in
Madura two Kallidaikurichi Brahmans (members of a caste of indi-
genous bankers) had a turnover of Rs 20 lakhs.167 In 1911, in the
whole of the Madras Presidency, there were only 16,212 men whose
activities were principally in moneylending and whose incomes were
more than Rs 1000 a year.168 If we assume that at least half of them
163
I am grateful of Dr John G. Leonard of the University of California for this
information.
164
A. R. Banerji, Assistant Collector, Ellore, to Collector of Godavari, 29 March
1901 in G.O. i o n (L and M, M) dated 17 July 1901. T.N.A.
165
RPBC, pp. 186-9.
166
MPBC, v, 378-80.
167
RPBC,p. 193.
168
Income Tax 1911 -12, p . 1 3 .

101
 The emergence of provincial politics

operated in the countryside (and certainly it was a great deal more),

we are left with only about 8000 medium-to-large creditors in an
urban population of 3^ millions.
The systems of productive industry — weaving, metal work and
other necessary handicrafts — were similarly organised and were
carried on by workmen who depended upon a capitalist either for
their wages or for the supply of their raw materials and the marketing
of their wares. Although there were many small 'one-workshop'
capitalists, these in turn usually were dependent on a larger capitalist,
who kept them supplied, and thus were members of a larger industrial
empire. At Madura, the silk weaving and dyeing industry, which
involved the bulk of the 40,000 local Sourashtra population, was, for
most of our period, under the direction of four families.169 In Madras
city, the legendary Pitti Thyagaraja Chetty was said to have some
2000 weavers under him. 170 In Kumbakonam, ten Mallaga Chetties
controlled most of the famous brass pot industry which had an annual
turnover of Rs 40 lakhs.111 Further, of course, a few employers con-
trolled large numbers of labourers in small food-processing, house
construction and transportation industries. Mothey Venkataswami of
Ellore kept a private army of low-caste 'Turpu' labourers in his
cotton, jute and rice mills and presses;172 in Bezwada, a rich contrac-
tor, G. Appalaswami, had legions of quarrymen at his disposal;173 in
Tanjore, the Porayar Nadar family supported 'a vast retinue' of
servants and workers.174 Industry was not organised on principles
which could sustain the hardy, independent craftsmen; rather it
promoted the growth of systems of wage and debt slavery.
The pattern of stratification in urban society was further emphasised
by the ownership of land. Many towns were situated on the estates of
zamindars. The Maharaja of Vizianagram, for example, owned the
large town of Vizianagram lock, stock and barrel. Similarly, in towns
like Masulipatam, Cocanada, Salem and Periyakulam, individual
zamindars had a considerable property stake. 175 Although less obvious,
in ryotwari areas large-scale landownership was by no means unknown.
las MPBC, iv, 298.
170
Minutes of Evidence taken before the Indian Industrial Commission. 1916—18.
Volume III, p. 56. P.P. 1919. Vol. xix.
171
RPBC,p. 138.
172
A. R. Banerji, Assistant Collector, Ellore, t o Collector of Godavari, 29 M a r c h
1901 in G.O. i o n (L and M, M) dated 17 July 1901. T.N.A.
173
Hindu 9 June 1922.
174
Hindu 23 July, 2 2 , 24 a n d 27 August 1888.
175
I n Masulipatam, t h e Nuzvid zamindari family; in Cocanada, t h e Raja of Pitha-
p u r a m ; i n Salem, t h e English zamindar of Salem; in Periyakulam, t h e zamindar
of Doddapanayakanur.

102
 The political economy of Madras

The landholdings of T. S. Sivaswami Odayar of Kumbakonam, for

example, were so extensive that they became the centre of a storm on the
municipal council which served him with a compulsory purchase order
on twenty-five acres.176 In Madras city, great commercial magnates
like P. Thyagaraja Chetty, the 'Gopathi' Beri Chetty family and the
Muslim Badsha family eagerly bought up houses and gardens. By
1895, there were seven men in Madras city whose income from pro-
perty alone was over Rs 10,000 p.a.177 If property ownership included
a market it was even more valuable, for it gave considerable control
of local trading patterns. Although, by the 1900s, most urban markets
had to operate under municipal licence, most also were privately
owned. In Madras city, such leading political figures as Pitti Thyaga-
raja Chetty, C. V. Cunniah Chetty and the Gogai family, whom we
shall be discussing later, were all closely associated with market
control.178
As might be expected from the analysis of economic organisation,
the distribution of wealth in the towns of Madras was at least as
uneven as it was in the countryside. The vast majority of the inhabi-
tants of the towns were desperately poor and lived at or below sub-
sistence level. Table 2 shows in 12 major towns the number relative
to the population of men whose income in 1895-6 from non-agricul-
tural sources reached Rs 500 p.a.179 To them must be added the
owners of agricultural land inside municipal limits, market gardeners
and rentiers who did not pay income tax. Nonetheless, even if the
figures of Table 2 are doubled, the number of people whose incomes
made them 'comfortable' - Rs 42 per month was only the salary of a
middle-level government clerk - was minute. This picture of urban
poverty is confirmed by the figures of the municipal franchise. In all
the municipalities of Madras in 1900, only 2.2 per cent of the adult
male population was able to meet the modest franchise-requirements
of paying income tax or municipal taxes of Rs 3-12-0 p.a. or of
holding a University degree.180
Information on urban wage-levels is much better than that for
rural wage-levels, so we are able to proble a little deeper into the
176
Hindu 7 and 21 September 1906.
177
Income Tax 1893-6.
178
See below pp. 200-14.
179
Income Tax 1895—6; Census of India. 1891. Madras. Volume XIV(Madras, 1893),
pp. 4 - 6 . T h e n u m b e r of assessees in Madras city does not include government
servants or persons in receipt of company salaries. It thus excludes most E u r o p -
eans and is far below t h e true figure of assessees, However, it does include most
Indians (except senior government servants) and so is not an inaccurate guide to
native wealth.
180
G.O.s 1822 and 1823 (L and M , M) dated 10 December 1901. T.N.A.

103
 The emergence of provincial politics

Table 2
Assessed at income of more than Rs 500 p.a.
in 1895-6
Town No. Popn. of town
in 1891

Madras city 1394 452,518

Kumbakonam 59i 54.307
Trichinopoly 518 90,609
Salem 419 67,710
Mangalore 366 40,922
Negapatam 355 59,221
Bellary 325 59>467
Nellore 323 29.336
Tuticorin 322 25,107
Cocanada 320 40,533
Calicut 318 66,078

society which fell below the line of the income tax. In 1900 in Mad-
ras city, Binny and Co.'s cotton mill paid unskilled labourers and
coolies about Rs 7-8-0 per month. 181 Binnys offered very few non-
monetary benefits to their workers, so that this figure reflects quite
accurately the money value of a labourer's service. However,
Binnys needed a large and constant supply of labour so that their
wage rates were higher than those of most employers in the city.
Independent estimates of agricultural labourers' wages at this time
put them at about Rs 6 p.m. and of the wages of general labourers
in the city at between Rs 6 and Rs 7 p.m. 182 These rates would seem
little enough, but they were taken in the capital where prices were
higher than elsewhere. In the mofussil towns labourers earned even
less, probably no more than the equivalent of Rs 4 or 5 p.m. 183
Skilled workers, of course, were better paid. But in the 1890s the
maximum rate Binnys had to pay for blacksmiths and carpenters, the
aristocracy of the working population, was only Rs 20 p.m., and
during the trade recession of 1899-1900, they cut the rate to Rs 15
without losing their workforce.184 The Rs 7 labourer was just able to
buy sufficient food to feed his family in a good year but would almost
certainly have to live in a straw hovel unless his employer gave him
181
Prices and Wages in India (Calcutta, 1901), pp. 3 2 0 - 2 .
182
Land Revenue 1902-3, 'Report on Madras', p. 2.
183
Ibid., 1908-9, pp. 78-81.
184
Prices and Wages in India, pp. 3 2 0 - 2 .

IO4
 The political economy of Madras

shelter; the Rs 15—20 skilled artisan was more secure but scarcely
able to rent or buy anything other than a hut.
Set against this background of urban poverty, the wealth of a few
inhabitants of the towns appears colossal. In 1895-6, for example,
284 men in the presidency earned more than Rs 20,000 p.a. from
non-agricultural sources.185 Rs 20,000 was sufficient to keep about
350 labourers through the year. Some urban magnates earned a great
deal more than this. The Maharaja of Vizianagram, based in Viziana-
gram town and Madras city, drew an income of at least Rs 20 lakhs
p.a.;186 and more than 10 other zamindars who were urban residents
drew more than Rs 3 lakhs p.a.187 Nattukottai Chetty families,
like that of Raja Sir Annamalai Chetty, enjoyed the profits of banking
businesses worth several crores of rupees.188 Trading magnates could
amass improbable fortunes, like the Muslim Abdul Hakim of Madras
city who left Rs 12 lakhs to his children,189 or M. Satyalingam
Naicker of Cocanada who, on his death, left Rs 9 lakhs to charity,190
or A. V. Jagga Rao of Vizagapatam whose income was sufficient for
him to build, equip and run a first-grade college in honour of his
wife.191
The organisation of the urban economy and the distribution of
urban wealth guaranteed that a few, hugely rich magnates would
exercise a pervasive influence over social and political life. Large-
scale and significant creditors were in short supply, and those who
did exist could extort many non-economic rewards as the price of their
services. In Ellore, Mothey Venkataswami's 'word was law to the
Bania community. Most Banias are moneylenders and have a large
clientele. They are thus in a position to exercise pressure on their
debtors and the result is that they command a strong influence . . .
even among other classes of the community.'192 In Sivakasi, Tin-
185
Income Tax 1895-6.
186
In 1910, his income from agricultural rents alone wasRs 14 lakhs. Land Revenue
187
1910-11. Land Revenue 1885—6, p. 30.
188
By the late 1920s, official estimates put the capital of the Nattukottai Chetties at
between Rs 80 and n o crores, Rs 70 crores of which was in banking. RPBC, pp.
186-9.
189
Before dividing his estates, Abdul Hakim already had given away more than Rs 4
lakhs to charity. The Asylum Press Almanack and Directory of Southern India. 1925
(Madras, 1925), p. 1269.
190
Sastri (ed.), Encyclopaedia of the Madras Presidency, p . 5 0 5 .
191
Hindu 23 January 1904.
192
A. R. Banerji, Assistant Collector, Ellore, t o Collector of Godavari, 29 M a r c h
1901 in G.O. i o n (L and M, M) dated 17 July 1901; for similar comments on the
influence of rich Komatis in Guntur, see 'Memo, of Chairman, Guntur Munici-
pality' in G.O. 25 (L and M, M) dated 11 January 1893.

105
 The emergence of provincial politics

nevelly district, local Nadar merchant-moneylenders were sufficiently

powerful to force the whole town, including non-Nadars, to pay
taxes to their communal charities. 193 In Kumbakonam, the clients
of the Poraya Nadars included Brahman mirasidars^ local traders
and property owners, policemen, revenue officials and, it was
rumoured, even the Collector.194
In a society without welfare facilities of any kind, without a free
labour market and without a secure food supply, workers and
labourers were very largely under the dominance of their employers.
Most artisans were so sunk in debt to their capitalist that they could
never escape and so dependent upon him that when, as in famines, he
withdrew his support, they starved.195
The direct power of urban property ownership was also consider-
able. In most towns there was competition for the most central shop
sites and the most desirable residences, with the result that the owner
was in a position of dictatorship. In Vizianagram, the samasthanam's
will had such strength that no local political leadership could exist
for long in opposition to it. Even a millionaire banker was prepared
to break up his caste conference on its orders.196 In Kumbakonam,
the town landowning family of Sivaswami Odayar was prominent for
many years and used its resources to take over the affairs of the
municipal council: 'T. S. Sivaswami Odayar made it a rule to fill
the council with his friends and supporters The Municipal Staff
were his dependents. The contractors were his tenants.' 197
By controlling tightly the economic resources of the towns, a small
number of men forced the general population to turn to them for the
most basic economic services. Members of this minute elite, however,
were well placed to make their presence felt in several other ways. In
addition to the power that followed from the way that they earned
their income, there was the power that they derived from the way that
they spent it. Many historians have seen how, in pre-British India,
the conspicuous expenditure of political and religious leaders brought
whole towns into existence, as tradesmen, artisans and merchants
gathered to serve their wants. During the nineteenth century, with
much state-power lost to Indians, these developments may have been
193
R. L. Hardgrave, The Nadars of Tamilnad (Berkeley and Los Angeles, 1969), p. 99.
194
Hindu 7 March 1888, 10 and 17 June 1896.
195
See Minutes of Evidence taken before the Indian Industrial Commission. 1916-18.
Volume III, p p . 3 3 0 - 3 1 . P.P. 1919, Vol. xix; Report on the Famine in the Madras
Presidency during 1896 and 1897 (Madras, 1898), II, 3 5 5 - 7 6 .
196
In 1910, the Komati merchant P. Ramamurthi broke up a session of his caste
conference at Vizianagram when its activities began to run counter to the orders
which he had received from the Maharani. Hindu 14 June and 8 July 1910.
197
G.O. 1133 (L and M, M) dated 23 June 1906. T.N.A.

106
 The political economy of Madras

less spectacular but they were, nonetheless, of considerable importance.

Wealthy Indians provided the main supplement to government spend-
ing on public works and welfare. Every town owed most of its facili-
ties - schools, libraries, hospitals, town halls, rest-houses and
parks - to the generosity of native patrons. In some activities, the
British deliberately stood back; all support for religion and the poor
outside famines came from rich Indian townsmen. Sometimes the scale
of these patronage endeavours could be truly fabulous. Between i860
and 1914, for example, the small community of Nattukottai Chetty
bankers spent more than Rs 1.82 crores on the restoration of the great
temples.198 Two Nattukottais, the brothers S. R. Rm. A. Ramaswami
and Annamalai Chetty, were particularly generous in other areas:
Ramaswami built a complete water-supply system for the town of
Chidambaram,199 and Annamalai spent Rs 1 lakh on the Madura
College200 and Rs 30 lakhs on a University.201 In the seventy years to
1940, the Gujerati Vaishya family of Lodd Govindoss spent Rs 50
lakhs on various projects, including the feeding of 35,000 poor every
year in Chidambaram.202 The Pydah Komati banking family funded
cow-protection in the Northern Circars to the tune of Rs 2 lakhs,
built a library and town hall for Cocanada, and backed the Widow
Remarriage Movement with Rs 3O,ooo.203 The Maharaja of Vizia-
nagram founded schools, Sanskrit patsalas, colleges and charities all
over the South. If few Madras magnates had the wealth to match these
efforts, many were capable of operating significantly in a single
locality. In Bellary town, for example, the millionaire industralist
A. Sabhapati Mudaliar built two hospitals and a school during the
1880s.204 In Cocanada, K. Basivi Reddi also built a school and
founded three choultries.205 In the crudest economic terms, the ex-
tension of this patronage linked the magnate to a wider section of
the population and made it dependent on him. However, and more
importantly, it also validated his role as an authority.
198 'Proceedings of the Religious Endowment Conference held at Delhi on the 16th
March 1914, Appendix 3, in Home Judicial A July 1914, Nos 265-87. N.A.I.
199 20
A t a cost of R s 1.3 lakhs, ibid. ° Hindu 8 D e c e m b e r 1915.
201
L. P. K. Ramanathan Chettiar, Annamalai Aracar (Madras, 1965), pp. 61-74
(Tamil).
202
See B. V. Nacharayya, Biography of Zemindar Lodd Govindass Varu (Madras,
1942); also, Anon., Biography of Sriman Lodd Govindoss Maharaj. The Hindu
Hero (Madras, 1942).
203
Sastri (ed.), Encyclopaedia of the Madras Presidency, p . 7 3 8 ; K. Viresalingam, The
Autobiography of Rao Bahadur Kandukuri Viresalingam Pantulu (Madras, 1911).
(Telugu).
204
Hindu 24 October 1883.
205
Report on the Famine in the Madras Presidency during 1896 and 1897, I, 47.

107
 The emergence of provincial politics

Philanthropy, a religious duty, brought virtue and a heightened

social status to its practitioner. Patrons who were seen to be bene-
ficent won the support and respect of the men around them. Religious
and cultural forms of patronage activity were especially meritworthy;
and incalculable sums of money were poured into movements for the
reform and revival of various aspects of indigenous civilisation. Hindu
merchants and zamindars renovated temples, founded Sanskrit Col-
leges and fed Brahmans; Muslim millionaires refurbished mosques
and set up Madrassas. Tamilians, like the Raja of Ramnad and the
Nattukottai Chetty zamindar of Andipatti,206 and Telugus, like the
Nuzvid zamindars™ 1 sponsored vernacular revivalism. Everywhere,
rich urban inhabitants funded local festivals, music circles and
literary associations. In South India, wealth could be translated into
social status and the maximum of social influence only through
patronage endeavours.208 The wealth which the few urban magnates
enjoyed was used to put them at the centre of the social and cultural
life of their towns.
It is not surprising that, with so many means at their disposal for
cajoling or coercing their neighbours, the greater urban magnates
exercised informal powers of government in the towns. Before the
completion of the British legal system in the mid-nineteenth century,
they often had acted as arbitrators in the disputes of the general
populace. When examining Salem in the 1850s, for example, J. W. B.
Dykes noted that its inhabitants usually took their cases to 'Chetties'
of various kinds.209 Chetty, however, does not mean headman, as
Dykes seemed to think, but is the Tamil word for merchant. Even in
the later nineteenth and twentieth centuries, the larger merchant-
206 p a t r o n s o f t h e Madura Tamil Sangham.
207
Whose court maintained several endowments for Telugu poetry.
208
T h i s statement ought not t o be taken t o mean that patrons need only have been
interested i n t h e public influence which they developed t h r o u g h patronage
activities. T h e r e is n o reason w h y they should not have possessed also a personal
interest in t h e objectives which they were supporting. However, o u r purpose here
is political history a n d patronage activities served a public, political end. M o r e -
over, t h e existence of this e n d was well known to t h e patrons themselves. While
intellectual o r religious motivation can account for t h e particular choices which
p a t r o n s m a d e w h e n offering their support to one movement r a t h e r t h a n another,
they c a n n o t explain t h e external style which was part of every act of patronage.
T h e r e were n o a n o n y m o u s p a t r o n s in t h e South India of o u r period a n d every
p a t r o n was concerned to maximise t h e public attention which was d r a w n to his
act of patronage. T h e p a t r o n wanted not only to support a cultural movement b u t
to be seen to support it — o u r explanation above is only of why he should w a n t his
act to be seen.
209
J. W. B. Dykes, Salem: An Indian Collectorate (London, 1853), PP- 2 2 1 - 4 .

108
 The political economy of Madras

financiers continued to provide courts alongside those of the British,

particularly to handle trade disputes.210 The more powerful zamin-
dars in their estate capitals also heard cases.211 During famines or
shortages, the urban population looked to their magnates to fend off
starvation by organising grain distributions and by expanding employ-
ment facilities.212 When rioting broke out, it was the same handful of
principal citizens who came forward to re-establish order. Following
the anti-European riots in Cocanada in 1907, for example, the police
reported:
There are 3 or 4 very influential men such as K. Basivi Reddy, K. Suryana-
rayana, late Pydah Ramakrishniah's son, Krishnamurthi, Nalam Padma-
nabhan [ all merchant-moneylenders] and others who are in entire sympathy
with the [swadeshij movement. It is said that it is on account of these
people that the riots ended so smoothly. But for these merchants people say
that the whole lot of Europeans in Cocanada would have been massacred on
that day.213
The strength of influence exercised by the wealthier urban mag-
nates can be seen most clearly on the occasions when it ran against
the power of the British Government. If rich zamindars supported
violent mobs, as in Masulipatam in 1886,214 or Komati merchants
fomented riots, as in Salem in 1882215 and Guntur in 1917,216 there
was nothing that the British could do to prevent mayhem. Similarly,
if dominant merchant groups refused to co-operate with municipal
government, as in Masulipatam in 1891, 217 Guntur in 1892218 and
Vanniyambadi in 1904,219 it became impossible for the British to find
men willing to take over the functions of the municipal council. In
some towns, the British were forced to recognise that unless a certain
man or group were allowed the powers of rulership, no government
could exist. In 1901, for example, the Collector of Godavari and his
210
See below pp. 139-40.
211
For example, see the durbar held by the Raja of Ramnad in Hindu 15 October
1894.
212
Report on the Famine in the Madras Presidency during 1896 and 1897, 1, 4 7 .
213
'Confidential Report No. i 2 R. W. D. Ashe, Collector and District Magistrate,
Godavari District, dated 8 July 1908' in Venkatarangaiya (ed.), The Freedom
Struggle in Andhra Pradesh (Andhra), 111, 255.
214
G.O. 2555 (Judicial) dated 6 December 1893. T.N.A.; see also P & J File 258 of
1894.1-O.L.
215
G.O. 1374 (Judicial) dated 23 M a y 1883. T.N.A.
216
G.O. 2461 (Judicial) dated 26 November 1917. T.N.A.
217
G.O. 564 (L a n d M , M) dated 16 April 1891. T.N.A.
218
G.O. 25 (L and M , M) dated 11 January 1893. T.N.A.
219
G.O. 802 (L and M , M) dated 28 April 1904. T.N.A.

IO9
 The emergence of provincial politics

Assistant Collector became aware of the extent of Mothey Venka-

taswami's influence in Ellore. After considering the alternatives to
him, however, the Assistant Collector came to the conclusion that it
would be more expedient to leave him alone. 220 At least in Ellore,
government officials eventually had managed to find witnesses pre-
pared to tell them what was happening in the town. But in very many
areas, particular magnates were able to operate networks so powerful
that the British were unable to detach from them anybody of signi-
ficance and could not even contemplate taking action. In Bezwada,
between 1916 and 1919, a deputy-collector and a tahsildar tried
desperately to make a case against G. Appalaswami, the contractor-
banker who ran the town, but finally had to give up for want of a
single reliable witness.221 Similarly in Madura, investigations into
one of the agents of K. V. Ramachari, a Sourashtra millionaire dye-
merchant, came to nothing when the prosecutor had to admit that
'the impeachment could not stand the test of legal proof although,
by any standard other than that of the British law, the agent was
guilty of rigging elections and embezzling public funds. 222
The wealthy urban magnate operated a series of dependency net-
works which were strictly analogous to those of the rural-local boss.
Through his control of economic resources, he was able to provide
vital facilities which put him at the centre of local life. To take a
single case, Mothey Venkataswami counted among his men the
influential Brahman lawyer C. Sitaramayya, who handled his legal
affairs and conducted his relations with the institutions of British
government; Brahman priests who lived by his patronage; fellow
merchant-moneylenders who traded on his capital; government
servants whom he bought; local landowners and members of the pro-
fessions to whom he lent money; the principal Muslim Kazi family
of the area, which owed him lakhs of rupees and did his bidding on the
municipal council; respectable families of all kinds who lived on
his property; low-caste Turpu labourers who worked in his factories
and mills; and everybody who attended the festivals which he
patronised or needed his support for any project.223 Mothey Venka-
taswami had his counterpart in nearly every town; indeed, in the
larger towns he usually had several. In Madura, between about 1890
220
Collector of Godavari to Secretary, Local and Municipal Department, 9 January
1902 in G.O. 140 (L and M, M) dated 7 February 1902. T.N.A.
221
G.O. 1017 (L and M , M) dated 5 July 1919. T.N.A.
222
G.O. 236 (L and M , M) dated 2 February 1916. T.N.A.
A. R. Banerji, Assistant Collector, Ellore, to Collector of Godavari, 29 M a r c h
1901 in G.O. i o n (L and M, M) dated 17 July 1901. T.N.A.

110
 The political economy of Madras

and 1910, there were four rival magnates of great power: K. V.

Ramachari, the Sourashtra dye-merchant; Robert Fischer, a Euro-
pean banker, zamindar and barrister; a Komati banker-trader; 224
and the resident Nattukottai Chetties. In Kumbakonam, between the
mid-i88os and the 1910s, the highest level of urban leadership was
split between the Porayar Nadars, T. S. Sivaswami Odayar (who
himself was allied to the Nattukottai Chetties)225 and S. A. Saminatha
Iyer, a mirasidar, property-owner, lawyer and moneylender. The
size of the networks which were at the disposal of men of this kind,
and their obvious strength at times of riot or in opposition to the
government, suggests that in urban as in rural Madras, structures of
power are best seen in terms of patron-client relationships.
If, for the time being, we accept this suggestion, then to judge
change in urban society we must look in several directions. Firstly,
we must examine the economy and the distribution of wealth to see if
any significant alterations, which could upset magnate structures,
were taking place in them. The expansion of trade in agricultural
produce, which recommenced after the Great Famine, obviously
brought increasing wealth to towns which were situated on railways
and harbours. Both income-tax and municipal tax payments increased
faster than can be accounted for by improvements in collection
techniques and, before the Income-Tax and District Municipalities
Acts of 1919 and 1920, by changes in the rates. 226 The number of
factories grew.227 Demands for new urban services, such as those
based on oil and electricity, created new channels for business activity.
The population of most towns increased.228 Admittedly, outside the
Andhra deltas none of these developments was in any way dramatic,
and Madras society remained overwhelmingly rural in character.
Yet, clearly, the towns were becoming richer. An initial question
must be what difference did this make to the structure of urban
wealth.
224
K n o w n colloquially as Sriman Chetty.
225
He was deeply in debt to them and repaid some of the interest by providing them
with political favours. In return, they backed him in the municipal council and
temple committee. Collector of Tanjore to Secretary, Local and Municipal
Department, 17 March 1903 in G.O. 528 (L and M, M) dated 6 April 1903.
T.N.A.
226
Income t a x from R s 27 lakhs i n 1 8 9 9 - 1 9 0 0 t o R s 75 lakhs i n 1 9 1 8 - 1 9 , Income
Tax 1899-1900 and ibid., 1918-1% and municipal tax from Rs 31 lakhs in
1 9 0 0 - 1 to Rs 94 lakhs in 1 9 1 8 - 1 9 , G.O. 1 8 2 2 - 3 (L and M , M) dated 10
December 1901 and G.O. 550 (L and M , M) dated 20 April 1920. T.N.A.
227
Census of India. 1921. Madras. Volume XIII. Part 1 (Madras, 1922), p . 190.
228
Census of India. 1931. Madras. Volume XIV (Madras, 1933), PP- 10—16.

Ill
 The emergence of provincial politics

A complete answer is very difficult to provide, but one point is

certain: the distribution of incomes was not becoming more even.
Between the mid-i88os and 1912-17, the price of the poorer food-
stuffs in most urban markets doubled. The most reliable evidence
indicates that wage rates only just kept pace with this increase. In
Binnys cotton mill, the wages of coolies rose from about Rs 7.5 per
month to about Rs 14.229 The wages of runners in the post-office -
again an employer whose wage rates can be trusted - rose in the
same period from a maximum of Rs 16 p.m. to about Rs 30. 230 The
presence of inflation in an economy without trades union organisa-
tions, however, led to a rapid stratification of the wage structure. In
Binnys' mill in 1900, for example, the minimum wage paid to a black-
smith was Rs 0.5625 per diem and the maximum Rs 0.75.231 By
1917, the minimum had risen to only Rs 0.6875 p.d. but t n e maximum
had shot up to Rs 1.6532.232 The differentials among Binnys' brick-
layers were even greater: from a scale of Rs 0.4219 to Rs 0.500 in
1900 to one of Rs 0.4531 to Rs 2.0156 in 1917.233 And among postal
runners they were growing: from a scale of Rs 10 to Rs 16 p.m. in 1900
to one of Rs 14 to Rs 30 in 1917.234 The rates suggest that the ties of
dependency by which employers held down their work force were
increasing in strength. Only those workers whom the employers chose
were being protected from the effects of inflation; the rest suffered
a steady decrease in their standard of living.
Changes in the distribution of higher incomes present the same
sorry picture. The only general sources which we have for higher
incomes are the income-tax statistics, which include rural incomes
and are not precise. Yet, if they provide us with no more than a gene-
ral impression of the change, they at least point us in its direction.
By 1895-6, the income-tax administration had been operational for
a decade and had sorted out most of its original problems. Its adminis-
trative categories, with a few specific exceptions, did not alter until
1919. In 1895-6, about 64 per cent of assessees on incomes of above
Rs 1000 per annum fell into the middle-income bracket of between
Rs 1000 and Rs 2000. 235 In 1918-19, the percentage was 59.236
It would indeed be difficult on this evidence to find important new
groups who were making money and beginning to alter the very hier-
archical structure of urban wealth. Of course, the absolute number of
people in the Rs 1000 to Rs 2000 bracket had grown, from 16,500 to
229 prjces anc[ Wages in India ( C a l c u t t a , 1920), p p . 2 1 9 — 2 1 .
230 231
Ibid., pp. 197-200. Ibid., pp. 219-21.
232 233 234
Ibid. Ibid. Ibid.
235 236
Income Tax 1893-6. Income Tax 1918-19.

112
 The political economy of Madras

3O,ooo,237 but given the background of inflation, this increase is

remarkably small. In 1895-6, there had been 47,500 people in the Rs
500 to Rs 1000 bracket, who, if they were only holding their own,
ought to have appeared higher up in 1918-19. 238 If there were a
significant change between 1895-6 and 1918-19, it did not come
from a rise in the number of middle-income earners who would be
likely to form an independent factor in urban political life. Rather
it came at the other end of the incomes scale. In 1895-6, only 284
men were classed as earning an income of more than Rs 20,000 per an-
num; and of these only 50 earned more than Rs 5O,ooo.239 By
1918-19, however, there were 826 assessees in the Rs 20,000 class
and 201 in the Rs 50,000 class.240 Most of the new wealth which was
accruing in the towns was going to support or create greater magnates.
The absence of a dynamic middle-level income sector is confirmed also
by municipal voting figures. In 1920, the franchise qualification was
dropped to bring in all those who paid only one rupee per annum in
municipal taxation. But it enfranchised just 5.4 per cent, of the
municipal population. 241
The magnate structure of economic power and influence was not
being undermined, at least by economic movements. But how secure
were the finance of those who belonged to the magnate stratosphere
and how readily did new men rise to join them? No precise answer
Can be given to this question but there is much to suggest that
mobility, although it was taking place, was fairly slow and of a kind
unlikely to promote social disturbance. The significance of the total
increase in Rs 50,000 plus earners is scaled down when set against
a background of inflation and of an urban population of more than
three millions. There were not many more very rich men in urban
Madras in 1918 than there had been in 1895. Moreover, most
of the identifiably powerful urban magnate families of 1895 (anc*
indeed of 1875) were maintaining their positions. The Nattukottai
Chetty bankers of the South, the great Komati families of Mothey,
Pydah, Nalam and Majeti of the Andhra deltas, the Roche-Victoria
Christian family of Tuticorin, the Maracair Muslim community of
Negapatam, and the 'Gopathi' and Titti' Beri Chetty families and the
'Badsha' Muslim family of Madras city lived through the period with
their influence undiminished. The casualties, such as A. Sabhapati
Mudaliar of Bellary, whose abkari contracting and cotton-purchasing
and manufacturing empire collapsed in 1896, and Raja Sir Savalai
237 238
Income Tax 1895-6; ibid., 1918—19. Income Tax 1895-6.
239 24
Ibid. ° Income Tax 1918-19.
241
G.O. 973 (L.S.G.) dated 30 May 1921. T.N.A.

113
 The emergence of provincial politics

Ramaswami Mudaliar, whose finance and commodity business fell

with his creditors, Arbuthnot and Company, were not heavy.
The stability of the magnate elite was not the result of a complete
absence of opportunities for 'new' men to get in at the lower levels
of the urban economy. Obviously, the expansion of trade must have
provided some openings for the razor-sharp entrepreneur. There
were also two important economic institutions which lay outside the
'natural' processes of the economy and from which quick fortunes
could be made. The first was the government which in addition to an
administration was a huge commercial company, operating monop-
olies in salt, opium and alcohol, and building roads and railways.
It could offer a lucrative source of income to anybody it chose, no
matter what his financial standing. In Cuddapah, for example, the
family of V. G. Vasudeva Pillai made a small fortune in army
contracting during the Second Burma War;242 in Tanjore, the great
Porayar Nadar family started its rise with an abkari contract in the
1830s;243 in Bellary, A. Sabhapati Mudaliar also began with a distil-
ling contract.244 The second institution was the temple. Temples
often owned landed property, markets and warehouses in urban
Madras and could be important sources of credit. Anybody who could
obtain influence with the governors of a temple could dip his hands
into a deep till. In Madura in the early twentieth century, K. M. Alladin
Rowther was able to lift himself out of the profession of brothel-
keeper, petty criminal and extortionist into that of real estate agent by
developing a connection with a member of the Sri Minakshi temple
committee. He became a successful property speculator.245
The significance of these two institutions in creating new pockets
of wealth, or rather in placing wealth in new hands, however, was
reduced by two factors. In general, it would seem that the number
of opportunities for initial gain provided by the government and
the temples declined steadily from a peak in the 1820s and 1830s. In
the early days of the British raj, when the government was reorganis-
ing and inventing administrations, the coffers of the Sirker were wide
open. Not only were the provincial scope and requirements of the
government novel in South India but British officials had little know-
ledge of the indigenous banking and commercial systems. Improbably
242
V. G. Venugopala Pillai t o Secretary, Local a n d Municipal D e p a r t m e n t , 14
October 1919. In G.O. 182 (L a n d M , M) dated 2 F e b r u a r y 1920. T.N.A.
243
Hindu 7 M a r c h 1888.
244
By 1885, h e held abkari contracts in most of t h e Ceded Districts. Abkari
1885-6, p . 15.
245
G.O. 1384 (L and M, M) dated 16 August 1917. T.N.A.
 The political economy of Madras

unimportant groups were able to latch onto government contracts

and monopolies, and some men's fortunes were made more because the
raj deemed them to be financially sound than because they were. As,
at this time, the British also had taken over temple management, the
disruptive influence of governmental innovation and ignorance
created opportunities for speculation in temple properties as well.
By the 1880s, however, the administration had stabilised and the
myopia of the British had become less serious. Monopolies and
contracts tended to be concentrated in the businesses of men of
proven worth; and when new schemes were developed, it was these
magnates who were given the first choice of them. Thus, by the 1890s,
the abkari contracts for the entire presidency were held by only six
firms;246 thus also, successful operators, such as the Porayar Nadars
and the Nattukottai Chetties, were able to engross several monopolies
at the same time. The Porayar Nadars added the salt trade to their
extensive liquor empire;247 while the Nattukottai Chetties who, by
1916, controlled salt supplies to the central eastern seaboard,248
were able to take up an option on a new distillery.249 As the larger
magnates expanded across several districts and trades, the chances
for the 'new' man diminished. Moreover, from the 1860s, the govern-
ment steadily withdrew from several areas of previous activity and
created in them political institutions which were open to local control.
As we shall see, this meant that it opened them to direct magnate
dominance. It is not surprising that, under the municipal council's
rule, the Mothey family had the tolls farm of Ellore; 250 nor that the
farm of the Trichinopoly municipal council's celebrated Fort market
should be in the pocket of one of the area's richest merchants; 251
nor that, under the regulations for the election of members of temple
committees, which came into force in 1863, the dominant temple
politicians of Kumbakonam should be T. S. Sivaswami Odayar, S. A.
Saminatha Iyer and T. Ponnuswami Nadar of Porayar.252
On the rare occasions when a would-be tycoon managed to make
an initial breakthrough, he was faced with a further, and in most
cases insurmountable, problem — that of credit. Security for loans
was as much of a difficulty in the towns as it was in the countryside.
246
Abkari 1896- 7, p. 15.
247
Hindu 1 August 1894.
248
Hindu (weekly) 27 J a n u a r y 1916.
249
Abkari 1918-19,?. 1.
250
G.O. 1360 (L a n d M , M) dated 1 December 1888, T.N.A.
251
G.O. 2089 (L a n d M , M) dated 16 December 1919. T.N.A.
252
Hindu 23 July, 2 2 , 24 a n d 27 August 1888.

115
 The emergence of provincial politics

The primary concern of any entrepreneur, once he has brought off

a successful deal or gained a position in the market, is to find cheap
money for expansion. But in Madras, this operation could take a con-
siderable time and be fraught with danger. On loans to the uncredit-
worthy, moneylenders insisted on interest rates of between 20 and
40 per cent, and often took their charges from the principal when it
was given. To meet large or long-term interest payments, the entre-
preneur would have to make quick and fabulous returns. The Madras
economy, while buoyant, seldom stretched to these. In order to
obtain reasonable rates for credit, the entrepreneur had two courses
open to him. He could establish his creditworthiness by proving the
financial stability of his businesses over the years. Once this was
accomplished he had access to funds from Nattukottai Chetties,
Marwaris, Multanis and even the Bank ofMadras. However, if he chose
this path it is obvious that he would be held down at a low level of
commercial activity for decades, if not generations. Moreover, he was
likely to find himself going round in circles, for until he could demon-
strate his ability to handle big business operations he could not get
the cash to start big business operations.
The more common course followed by rising magnates who were
successful was to forge marriage connections, which acted as security,
with families who were capable of providing credit. This condition
naturally reduced the number of potential stars on the commercial
horizon, for marriage possibilities in any one generation were
limited by factors of sub-caste and of acceptance.
Movements from rags to riches, therefore, were achieved most
usually by groups of interrelated families and not by individuals;
and this process also took time. The Nadar traders of northern
Tinnevelly and Ramnad, for example, moved into the area before
1820 as petty-traders and haulage contractors. 253 By making marriage
alliances, some of their leading families were able to provide credit
facilities for each other, be it only at first in a small way, and to
combine the accumulation of capital with the development of the high
degree of liquidity necessary for taking opportunities in various
places when they came. However, it was not until the 1870s at the
earliest (and not until the 1890s truly) that the Nadars of Kamudi,
Sivakasi and Virudhunagar began to show their magnate position by
challenging for local political power. 254 Similarly, it took most of
the nineteenth century for the Christian Bharatha fishermen of
253
Hardgrave, The Nadars ofTamilnad pp. 9 6 - 8; Hindu 7 M a r c h 1888.
254
Hardgrave, ibid., pp, 105-29.
 The political economy of Madras

Tuticorin to produce the great houses of Roche-Victoria and Cruz

Fernandez or for the Telaga ex-peasant traders of the Cocanada—
Rajahmundry area to produce the families of K. Basivi Reddi and K.
Suryanarayanamurthi Naidu.255
In the absence of joint-stock banks which were prepared to deal in
the commerce of the interior, financial institutions in Madras con-
tinued to be based on familial or quasi-familial organisations. The
great Nattukottai Chetty banks were essentially family businesses
built up over generations by sending out scions to various localities
with capital which was steadily increased and passed back to the
family home. New inflows into any particular bank owed more to
marriages than to deposits from outsiders; as late as 1929, the
Chetties were trading with a capital 80 per cent their own.256 Komati
families in Andhra, such as the Pydah, Mothey, Nalam and Majeti,
and Tamil Muslim families operated similar organisations.257 Where
a number of families related by caste worked out of a common base,
supra-familial ties between them developed in order to improve
security and raise credit. The Nattukottai Chetties, whose interna-
tional empire centred on a block of villages in Ramnad, had carried
this furthest and recognised a form of joint-liability in their deal-
ings.258 In their particular towns, Komatis and Tamil Muslims also
were often linked in their operations.259 The basic unit of activity,
however, always remained the family and, within any communal
group, individual family businesses continued to rise and fall.
Obviously, this factor put a considerable brake on effective upward
mobility. Few families could break through into the magnate elite,
and those who succeeded took a generation or more to consolidate
their position. When movement was this slow, it allowed time for the
rising family to be assimilated. Although it increased the possibili-
ties of magnate rivalry, it did not upset the magnate system.
The major economic resources of Madras remained in the hands of
a small urban elite, the social composition of which was stable.
This conclusion is supported also by an investigation into the
mechanics of investment. In his study of 'Tezibazaar' in Northern
India, Richard Fox has shown how, over a considerable period, urban
255
See Sastri (ed.), Encyclopaedia of the Madras Presidency, p . 5 1 0 .
256
RPBC,pp. 186-9.
257
For the Komatis, see F. J. Richards, 'Cross-Cousin Marriage in South India' in
Man, 97 (1914); E. Thurston, Castes and Tribes of Southern India (Madras, 1909),
in, 307- 11.
258
RPBC, pp. 186-9.
259
G.O. 3360 (L.S.G.) dated 8 September 1928. S.A.H.

117
 The emergence of provincial politics

merchant groups used their financial power and new systems of

administration to take over power in the town from the local zamindar^
whose influence was based only on property ownership.260 Obviously,
in an expanding commercial economy, the profits of trade were likely
to be greater than those derived from legally controlled rents.
In Madras, however, very few towns fit the model of Tezibazaar. Of
the larger conurbations, only in Masulipatam was there anything
approaching a simple landlord/merchant conflict.261 Elsewhere, pat-
terns of investment make it extremely difficult to divide economic
activities between neatly defined social groups. On the one hand,
as we mentioned before, many zamindars were diversifying their pri-
vate economies and becoming involved in urban trade, commerce and
manufacture. Rural-boss groups also were coming into the towns
and the wealthiest families among them were leaving agents to trade
and bank in their own right.262 On the other hand, mercantile
families were moving their surpluses into urban property and into
privileged zamindari and inamdari rent rights. The Mothey and
Majeti families of Andhra, the Nattukottai Chetties of Ramnad, the
Gopathi and Pitti Beri Chetties and the Muslim Badshas of Madras
city all came to own estates and extensive house properties in their
respective locales.263 By 1880, probably by 1850, the distribution of
investments had become so wide that most magnates had come to
have interests in most aspects of their local economies. As individuals,
they sat across property ownership, credit, trade and manufacture,
and drew profits and powers from them all. They were able thus to
preserve themselves from a decline in any one sector of the economy
and maintain their positions more securely. While, undoubtedly,
there were status rivalries between 'ancient5 and 'modern' merchant-
landowners, these ought not to be seen as class rivalries derived from
different relationships to the methods of production. There were
rivalries no less bitter between one 'new' merchant-landowner and
260
See R. G. Fox, From Zamindar to Ballot Box (New York, 1969).
261
T h e particular fields of confrontation were t h e municipal council and various
committees which administered public facilities. See, for example, Hindu, 11 M a y
1907.
262
F o r example, T . Ramalingam Chetty (q.v.) of Coimbatore whose family cluster
originally were Vellala landowners. By t h e later nineteenth century, t h e family
were r u n n i n g a n u r b a n banking business w o r t h Rs 7 lakhs and had become k n o w n
as 'Chetty' trader Vellalas.
263
See 'Landowners o f Madras' in Asylum Press Almanack and Directory of Madras
and Southern India. 1910. (Madras, 1910).
 The political economy of Madras

another;264 and there were alliances of great strength between ancient

zamindars and new merchant-landowners.265 In the South, a better
model would be of magnates of promiscuous origin but presently in-
volved in very similar activities, each vying with and temporarily
allying with others of the type for prestige, status and power. We
shall see this more clearly when investigating the politics of the
municipal council but such a system, of course, could only operate
in conditions of considerable economic and social stability.
Thus far, our analysis of the foundations of political organisation
in the towns has excluded consideration of the role of the educated
Indian. This omission may seem strange. In much historiography of
modern India, educated groups have been characterised as the most
dynamic elements in social and political change. As their activities
were largely confined to the towns, their most profound impact came
in the sphere of urban life. In recent works, R. K. Ray, Christine
Dobbin and C. A. Bayly have shown how lawyers, journalists and ex-
public servants came to challenge for social power in Calcutta from
the 18 50s, Bombay city from the 1850s and Allahabad from the 1900s
respectively.266 The Madras Presidency possessed educated groups
which were socially analogous to the bhadralok of Bengal, the
Chitpavan Brahmans of Bombay and the Kayastha/Urdu Muslim
'service' families of the U.P.: the Brahman and Sat-Sudra families,
who invested in education, tended to have served as administrators
under previous native regimes, to come from an economic back-
ground of small landownership and to be concerned primarily with
obtaining stakes in government service and the liberal professions.
They were numerous — the educated community of Madras was
second in size only to that of Bengal - and, from their salaries, rents
and fees, would seem as economically independent of magnate in-
264
As between the Sourashtra dye-merchant and property-owner K. V. Ramachari
and several Balija Naidu landlord-merchants in Madura, G.O. 382 (L and M, M)
dated 29 February 1912; G.O. 1203 (L and M, M) dated 1 July 1912. T.N.A.
265
As between the zamindar of Doddapanayakanur and several traders in
Periyakulam. Collector of Madura to Secretary, Local and Municipal Depart-
ment, 11 February 1919 in G.O. 374 (L and M, M) dated 6 March 1919; also
G.O. 1383 (L and M, M) dated 18 September 1919. T.N.A.
266
R. K. R a y , 'Social Conflict a n d Political Unrest i n Bengal, 1875 t o 1925'.
Unpublished Ph.d. dissertation, C a m b r i d g e University, 1972; C. D o b b i n , Urban
Leadership in 'Western India. Politics and Communities in Bombay City 1840—1885.
(Oxford, 1972); C. A. Bayly, The Development of Political Organisation in the
Allahabad Locality, 1880 to 1925'. Unpublished D. Phil, dissertation, Oxford
University, 1970.

119
 The emergence of provincial politics

fluence as their counterparts in other provinces. They also were very

active in religious revival, social reform, cultural regeneration and
nationalist movements.
It is one of the main and recurring arguments of this book, how-
ever, that their importance as an independent factor in the politics
of Madras was very limited. Outside the capital, and there for only ten
years, they lacked the institutional power to dictate political com-
mands to any other than fellow-educated 'service' and professional
groups. While it would be possible to tell an interesting tale of their
own internal debates and conflicts and of their abortive attempts to
wield influence, it would be misleading to term this story 'South
Indian polities', for it would have very little to do with the exercise
of power. The reasons why the western-educated groups of Madras
failed to reach the political positions enjoyed by their colleagues
in other provinces lie, of course, in the political structure of the
South: while the educated themselves may have been very similar to
the educated in Bengal, Bombay and U.P., the political structure in
which they had to operate was not. The organisation of political power
in South India persistently excluded them from places of importance.
In this chapter, we have been examining the economic instruments
of power. Let us first see what position the educated occupied in the
systems by which wealth was made and spent. In the magnate-
dominated economy, they were simple, if valuable, dependents.
Komati merchants hired (and fired) Brahman clerks,267 Nadar traders
used Vellala accountants,268 zamindars took their administrators
from the same families as the government. All rich men retained the
best vakils and bribed the best government servants whom they could
find. While, as we have indicated, many of the educated came from
petty landowning families, the incomes which they derived from land
seldom were sufficient to provide them with an independent basis for
existence in the towns - had they done so, there would have been no
need for the educated to seek alternative employment. Even in those
economic activities where they could be seen to be important, the
educated had achieved their local places usually by trading on capital
which belonged to others rather than themselves. Although clerks,
lawyers and officials often were able to keep their masters ignorant
of what they were doing and were capable of projecting their own
267
F o r example, t h e father of Konda Venkatappayya, a famous A n d h r a B r a h m a n
politician, was a clerk in a Komati firm. K. Venkatappayya, Sviya Caritra
(Vijayawada, 1952), 1, 6. (Telugu).
268
H a r d g r a v e , The Nadars of Tamilnad, p . 100.

120
 The political economy of Madras

ideas into urban society, their dependency always restricted them in

what they could be seen to be doing. If they broke faith with their
patrons, they lost their sources of support. Only at the very top of
the legal and bureaucratic ladders were lawyers and bureaucrats in a
position of sufficient wealth and influence to dictate the terms on
which they would serve particular magnates.
From the later nineteenth century, however, the primitive in-
stitutions of the magnate economy were joined and pressed by more
sophisticated financial organisations. The growth of the economy
and the instability of existing credit organisation promoted the
need for a more advanced financial institution than the moneylender.
Towards the end of our period, several joint-stock banks, co-operative
credit societies and insurance companies appeared. The men who
controlled them often were socially distinct from the older credit
magnates. Their abilities lay in literacy and organisational expert-
ise which enabled them to manipulate masses of small deposits rather
than in the personal possession of wealth. They could be seen as a
managerial element within the society of the educated. By picking
up the threads of credit and commerce, they could develop their own
economic and political controls. In 1888, for example, the Sub-
Collector of Tuticorin reported of his municipal council:
'The Hindus form a compact party and, through the influence of the Bank
Cashier [of the Bank of Madras] over the Chairman, they have complete
control of the Council. The Bank Cashier, a clever Brahmin, is, in fact, the
ruler of the municipality though the Bank would not allow him to be
Chairman. . . . Those who have transactions with the Bank are naturally
afraid to oppose him.269
Later, the cashier was elected to the district temple committee - a
sign of great influence.270 Other Bank of Madras cashiers, in Guntur
and Bezwada, also were prominent municipal politicians.271 Until
1900 or so, most of the activities of the Bank of Madras were con-
fined to the ports; but gradually it shifted its interests inland and
was joined by the Indian Bank, the Nedungadi Bank, the Nadar Bank
and the Specie Bank of India. From about the turn of the century,
co-operative credit societies and Indian insurance companies also
began to appear. A whole new financial apparatus, with different
men at its controls, was in the process of formation.
s j g n e c i Divisional Officer, Tuticorin, 29 F e b r u a r y 1888 in G . O . 508
(L a n d M , M) dated 17 M a y 1888. T.N.A.
270
Hindu 9 December 1896.
271
Hindu 29 October 1912 and 12 February 1915.

121
 The emergence of provincial politics

At the other end of the system which gave social pre-eminence to

the magnate, new developments were taking place in the organisations
of public action. The expansion of the educated community, the
arrival of the printing press and the growing consciousness of national,
regional and religious identities under foreign rule, all helped to
create a new theatre for local activities. Associations for religious
and cultural revival and reform shot up in many towns; newspapers
and tract societies poured out propaganda to be read by townsmen;
debating societies and protonationalist clubs appeared to try to in-
fluence the conduct of local and even provincial bureaucrats. Once
again, the men who led these were predominantly western educated
professionals and literati, who possessed the organisational and
educational skills. As the scope of their activities grew, they could
hope to become recognised and supported widely by the populace.
They could make public opinion, or at least make it aware of its own
existence, start new projects of public welfare and sometimes even al-
ter aspects of the administration and government. Naturally, this
meant that they began to encroach on the place of the magnate and to
usurp his status and function as the authority in local society.
Both of these developments suggest that means were appearing
which the educated could use to break their bondage. It must be ser-
iously doubted, however, whether either was significant in the political
organisation of Madras before 1920. To take the emergence of new
financial institutions first: neither had this gone very far nor were
the older magnates simply spectators of their own decline. The initial
explosion of joint-stock credit enterprises produced a plethora of
companies which were less fly-by-night than gone-by-early-after-
noon.272 In order to restore stability to the world of joint-stock
finance, the government enacted legislation which demanded certain
minimal guarantees of security, and the law courts passed a series of
decisions which weakened claims to limited liability.273 The result
was to slow down considerably the process of capital formation in
joint-stock enterprises. The most dynamic element in the new finance
272
Land Revenue 1905—6, p . 67. T h e South was famous for its nidhis - v o l u n t a r y
associations formed by various m e a n s t o provide short-term loans. Nidhis were
particularly important in and around major trading towns, such as Coimbatore
which possessed 125 working with a capital of Rs 70 lakhs. However, they were
notoriously unstable and open to abuse. They came nowhere near to replacing
the old moneylender in short-term urban finance. RPBC> pp. 196-7; MPBC,
in, 1161, iv, 453,481-3.
273
F o r a review of t h e working of t h e Indian Companies Act, 1913, see Hindu
6 August 1915. I n its first year of operation in M a d r a s , the Act reduced the n u m -
ber of joint-stock companies from 472 t o 406.

122
 The political economy of Madras

was the co-operative credit society, but restrictions still prevented

this from playing a central part in urban finance before at least the
1930s.274
Most of the more successful joint-stock enterprises were under-
taken by men who already possessed considerable capital resources
and were able to take advantage of Indian company law. The great
Komati families of Andhra, for example, began to establish supra-
local links in order to develop their trading and industrial activi-
ties. The Pydah family of Cocanada, the Mothey of Ellore and
the Majeti and Nalam, which could be found in several towns, set
up paper mills, rice mills and cotton presses under joint-stock
finance.275 The Indian Bank, the first indigenous joint-stock bank in
Madras, was created with money from the Gujerati banker Lodd
Govindoss, the Nattukottai Chetty S. R. Rm. A. Ramaswami Chetty
of Chidambaram and a member of the Nalam family, who were
among its first directors.276 In the late 1920s, Raja Sir Annamalai
Chetty turned his Nattukottai Chetty family bank into a joint-stock
concern and demonstrated the power which had remained with the old
moneylenders. On its opening day it was the second largest bank in
Madras, surpassed only by the state-backed Imperial Bank of India. 277
In the area of public activity also, the magnate managed to preserve
himself. Of course, it was usually with the support of magnate
patrons that western-educated publicists financed their various as-
sociations, presses and tract societies. In these they were seldom
more than the agents of magnate interests. Their endeavours were
intended to highlight their patrons at least as much as themselves.
This is not to deny the intrinsic importance of the new ideas ex-
pressed in reformist, revivalist and nationalist circles, which were
the results of change in the educated community. But political history
must deal more with the extent of influence and the effect of ideas
than with the character of doctrines. In Madras before 1920, there is
little evidence to suggest that educated publicists were capable of
organising movements which could be supported by large-scale public
involvement rather than by a few very rich men. South India lacked
the broad cultural homogeneity of Hindustan, which enabled reli-
274
I n 1920, u r b a n co-operatives in M a d r a s lent a total s u m of only R s 48.15 lakhs.
Annual Report on the Working of Co-operative Credit Societies Act (II of1912) for
the year 1920-1. (Madras, 1921), p . 2 5 .
275
See t h e foundation of the G u n t u r Cotton, Jute and Paper Mills C o m p a n y Limited
(capital Rs 6 lakhs) in Hindu 13 September 1904.
276
Hindu 9 M a r c h 1908.
277
RPBC, p. 28.

123
 The emergence of provincial politics

gious revival movements, such as Sanatan Dharm, to draw extensive

support from village elite groups. It also lacked a tradition of cultural
conflict of the intensity of the Hindu/Muslim confrontation in the
U.P., which enabled the organisers of cow-protection to count on the
subscriptions of hundreds of petty urban Hindu traders. As C. A.
Bayly has shown, the strength of movements such as those of the
Sanatan Dharmists and cow-protectionists derived from the wide
territorial organisations which they developed.278 Their leaders were
able to concentrate on any locality the weight and power which they
had picked up, be it only in dribs and drabs, across the province.
The organisers of similar movements in Madras had no opportunity
to do this.
The celebrated Widow Remarriage movement which appeared in
Rajahmundry in the late 1870s, for example, was made possible only
by the hard support (in cash) of the Komati millionaire Pydah
Ramakrishniah. When he died, the movement withered.279 Similarly,
the Madura Tamil Sangam, one of the central organs of Tamil
revivalism, rested on the patronage of the Raja of Ramnad. When
that patronage dried up in 1912, the Sangam almost collapsed and was
only saved by the arrival of a blank cheque from the Nattukottai
Chetty zamindar of Andipatti. 280 Again, in Madras city in the
1880s, the Hindu revivalist and anti-Christian campaign of Siva-
sankara Pandiah was backed not by a mass of petty donations but by
two merchants and a zamindar.281 In like manner, the development of a
female education movement and of an educated nationalist press from
the 1870s owed much to the purse of the Maharaja of Vizianagram.
When he died in 1895, m o s t of his schools were closed or reduced in
size for want of funds,282 and the great Hindu newspaper went
bankrupt. 283 Of course, as the twentieth century advanced, a broader
base for fundraising activity emerged. But as late as 1922, even the
Congress could not sustain itself. Most of its old magnate patrons
had gone into the Legislative Councils, and its hopes of raising
money without them were nil. The non-co-operation campaign in
Tamilnad folded within days of the expenditure of the last of the
contributions which it had received from All-India funds.284
278
Bayly, ' T h e Development of Political Organisation in t h e Allahabad Locality,
1880 t o 1925', ch. 2.
279
L e o n a r d , ' K a n d u k u r i Viresalingam: A Biography of a n Indian Social Reformer,
1848-1919'.
280
G. O. 51 (Education) dated 20 January 1913. T.N.A.
281 282
Hindu 17 September 1888. Hindu 3 M a r c h 1904.
283
V. K. N a r a s i m h a n , Kasturi Ranga Iyengar (New Delhi, 1963), p . 6.
284
C. J. Baker, 'Noncooperation in South India' in Baker and Washbrook (eds.),
South India.

124
 The political economy of Madras

In Madras, the publicist could not develop wide territorial link-

ages nor could he find significant caches of wealth, which were
outside the control of a few magnates. In consequences, he could
never establish an independent authority apart from his magnate
patrons. While the developments in education and in communications
had an impact on the intellectual life of the traditionally literate
elite, they did not provide the foundations of a single political
empire. Indeed, in many ways the growth of quasi-political publicist
activities in the towns can be seen as emanating as much from the
will of the patrons as from that of the publicists themselves. For
many centuries before the British arrived, rich townsmen had tried to
convert wealth into social status by patronising religion and the arts,
and by spending on public welfare. Status expenditure is necessarily
a competitive activity and is used to establish positions of superiority
and inferiority. Thus, although the direction which this spending
took was always, in part, determined by the priests and literati who
were the custodians of culture, it was also, in part, determined by the
competitive needs of each magnate to demonstrate his position in rela-
tion to other magnates. From the later nineteenth century, the in-
creasing wealth of the towns threw up new magnates - be it only
slowly and sporadically - and put more resources at the disposal of
those already established. This development generated greater
magnate competition and hence drove magnates into looking for new
and more varied cultural activities with which they could be as-
sociated. The changes in education, in the impact of foreign philo-
sophies on the Indian intellect and in the relationship of theology to
social organisation, which affected educated Indian society, certainly
were instrumental in promoting social reform and religious revivalist
movements. But of at least equal importance was the pressure of mag-
nate competition which created much wider social demands and
political purposes for the movements.

The problem of communal identities

Our examination of the economic instruments of political control
must lead us to stress the significance of the clientage or dependency
network in the political organisation of South India. As, perhaps, we
might expect to find in any pre-industrial society, in which dire
poverty was general and economic opportunities narrow, the mechan-
isms by which wealth was made and distributed forced large numbers
of men to recognise the authority of one or a few leaders. Our analysis
must also suggest that the direction followed by the clientage net-
works of leaders was vertical rather than horizontal to the social
order. To take Mothey Venkataswami's empire as typical, we have
125
 The emergence of provincial politics

seen that men in every occupational group from labourer to priest

were linked together under him. Similarly, if we make caste and
religion the terms of reference, we find that his authority touched
men at every level of the caste system, from pariah to Brahman, and in
every religious category from orthodox Hindu to reformist Hindu to
Muslim. In the processes of decision-making through which most
South Indians earned their daily bread, the more obvious class and
cultural divisions had little relevance as organising principles.
Our emphasis on the patron-client relationship as the basic nexus
of politics, however, ought not to be taken to mean that we totally
disregard the existence of socially horizontal relationships. While
perceptions of class may have been extremely weak, it would be
impossible to deny that the behaviour of most South Indians was
strongly affected by those elements of religion and culture which
underwrote the contemporary system of social values. As the authori-
ties controlling economic life need not always have been the same as
those controlling religious and cultural life and as, indeed, the dic-
tates of economic prudence could clash with those of cultural worthi-
ness, there existed conditions in which cultural sentiment became an
independent mechanism of control. In order to assess the extent to
which our working model of the clientage network requires modifica-
tion, it is necessary for us to examine in some detail the political
implications of cultural institutions and determine the conditions on
which their influence undercut that of economic organisation. As the
caste system is generally taken, in South India at least, to be the
dominant influence on behaviour, we should begin with a considera-
tion of it.
In a recent article, Richard Fox has suggested a useful approach to
the apparently insoluble problem of defining satisfactorily what the
word 'caste' means. He sees several related but logically distinct
categories of caste activity. Firstly, there are the Sanskritic varnas
of Brahman, Kshatriya, Vaishya and Sudra which cover the whole of
India. Secondly, in each region of India there are castes (in the case
of the South, castes such as the Vellalas, Reddis, Kammas or Nairs)
which occupy positions within the Sanskritic varnas: these regional
groupings he terms 'sub-regional varnas\ Thirdly, within each
'sub-regional varna? lie further groupings divided from each other by
prohibitions on intermarriage, ritual celebrations and, often, com-
mensality: in the South, these would be represented by sub-castes
such as the Chooliya, Karkatha, Thondamandala and Gounder
Vellalas. Finally, within each sub-caste there are further divisions
mapped out by actual intermarriage and ritual and social intercourse:

126
 The political economy of Madras

these would be represented by particular lineages or family clusters.285

If we take Fox's model as a guide, we can begin to see at what level
and in what ways the caste system must modify our political depend-
ency structures.
It would be difficult to argue that Indian society was (or is) tightly
organised into communities along the lines of Sanskritic varna. The
categories were so vast and the social heterogeneity they masked so
great that there were no institutions or organisations to link the
various Brahmans or Sudras of India together. Essentially, varna
represents a notion of social status - a matter of psychological
perception - not a concrete political institution. By the same token
the sub-regional varnas of Southern India were loosely defined status
groupings and not organised communities. Within the category
Vellala, Chetty or Nair subsisted distinct groupings which could be
differentiated from each other by occupation, intermarriage, ritual
practice and historical origin. These sub-castes simply recognised each
other as possessing roughly the same social status over a series of
contiguous territories. Until the twentieth century at the earliest, few
sub-regional varna groupings possessed any central institutions to
regulate the behaviour of their members or to serve the ends of com-
mon welfare. If they are to be regarded as political communities, the
nature of their union is better understood by the metaphysician than
by the political historian.
If we accept sub-regional varnas as loose status categories, then it
is likely that they would become important principles of political
organisation only when the status which they described was in
some way attacked. Relative status, when commonly accepted, does
not need to be defended and is not a category of political action.
During our period, there were several examples of caste-status con-
frontation in Madras. Under the impetus of Christianity and new
educational opportunities, for example, 'untouchable' Shanar toddy-
trappers in southern Tinnevelly in the 1860s began to press for a
general recognition that their sub-regional varna belonged to the
Kshatriya Sanskritic varna. This not only linked together disparate
Shanar groups but forced their neighbours into unity in order to
resist their claim.286 Sudden increases in wealth often produced
285
R. G. Fox, 'Varna Schemes and Ideological Integration in Indian Society' in
Comparative Studies in Society and History, XI (1969), 27-45; also, R. G. Fox,
Kin, Clan, Raja and Rule (Berkeley, Los Angeles and London, 1971), ch. 2. It
should be noted, however, that our version of Fox's categorisation is fore-
shortened from his own. Our purpose is political history and we are interested
only in the major categories which imply differences in political relationship.
286
Hardgrave, The Nadars ofTamilnad, pp. 43-70.

127
 The emergence of provincial politics

similar results. In the 1890s, Hindu Shanars-turned-Nadars, who had

become rich merchants in the towns of northern Tinnevelly and
Ramnad, also aspired to the Kshatriya varna.287 At the same time,
rich Komatis in the main towns of the northern Circars began
claiming the right to perform Vaishya ceremonies;288 in the same
area, Kammalas (artisans) claimed recognition as Brahmans,289 and
in Madura, wealthy Sourashtra silk-weavers demanded that local
society should accord them the privileges of Gujerati Brahmans.290
Another stimulus to status activity of this kind came from the census
which sought to fix caste groups in permanent varna categories and
naturally cut across some of the aspirations of their members.291
In South Indian historiography, much has been made of these
developments. Yet, when studied in a general context rather than
in isolation, their significance before 1920 is very limited. Neither
economic nor educational change nor the census succeeded in creating
viable political organisations out of the sub-regional varnas. In the
first place, the nature of economic development was such that, as
we have seen, it promoted less the social groups made rich by new
economic processes than those groups which were already well estab-
lished in control of economic resources. Certainly, this was more
true of the countryside than the town, but even so success on the scope
and scale of the Nadars was rare. Equally, the major changes in the
composition of the educated community were the product not of the
nineteenth century but of the late 1920s and 1930s. English education
remained the preserve of 'traditionally' literate families and even
the non-Brahman movement of 1916 cannot be tied to significant
change.292 Education, even in the vernacular, seldom reached depen-
dent labouring groups.
The character of economic and educational change meant that
there was no general pressure on the status categories of the existing
social hierarchy. The flexibility of the caste system itself, however,
also took much of the steam out of communal politics. The Indian
caste system was not a caste system in the technical sense: the ranks of
persons, or rather of social groups, inside it were not immutable and
fixed by birth. As historians such as Burton Stein and R. Inden have
287
Ibid., pp. 95-129.
288
Hindu 1, 3 , 4 J a n u a r y , 1 September, 9 October, 10 a n d 20 November 1894.
298
Hindu 3 M a y 1894, J J u n e l 8 9 7 -
290
Hindu 10 M a r c h 1920.
291
S. and L. Rudolph, The Modernity of Tradition: Political Development in India,
(Chicago, 1967), pp. 49-64.
292
See below pp. 271-2.

128
 The political economy of Madras

pointed out, there were mechanisms which allowed groups to move up

the social scale and down again.293 Equally, modern anthropologists
have seen how, through the practices of hypergamy and social emu-
lation, castes like the Rajput and the Gounder could be open.294 In
British South India, several factors guaranteed that these mechanisms
not only continued to exist but were frequently used.
Firstly, as we have seen, marriage networks among the bulk of the
population which lived off the countryside were intensive rather
than extensive. Reflecting local-level culture, they tended to cover
geographically small areas and small numbers of people who inter-
married repeatedly.295 It was thus relatively easy for an endogamous
unit to break away from the larger status grouping in which it was
set and to claim attachment to another status grouping. As Andre
Beteille saw in Tanjore, 'Since intermarriage with close relatives is
frequently practised by many Non-Brahmins, a section of people,
by confining marital relations among themselves, can claim to be
Vellalas, Ahamudiyans or Padaiyachis.'296 Secondly, local-level
culture provided no centralised communal institutions governing
membership of a sub-regional varna; thus there were no awkward
authorities to challenge the antecedents of groups claiming member-
ship. And thirdly the majority of Southern sub-regional varnas were
gathered in and around the Sudra Sanskritic varna\ thus social mobil-
ity between them was possible without crossing any very obvious and
contentious ritual gap. In Madras, more than anywhere else in India,
small groups were able to raise their effective social status without
causing disturbance to the prevailing status structure and without
mobilising other groups or endogamous units either in the status
categories which they were leaving or in those which they were enter-
ing.
The social origins of those who were seen to fill the 'dominant'
caste of any area were often obscure. In his manual on Coimbatore
district, for example, F. A. Nicholson freely admitted his inability to
293
B. Stein, 'Medieval S o u t h Indian H i n d u Sects' in J. Silverberg (ed.), Social
Mobility in the Caste System in India ( T h e H a g u e , 1968), pp. 78—94; R. I n d e n ,
'Social Mobility in Pre-Modern Bengal'. Unpublished paper read at Study Con-
ference on Tradition in Indian Politics and Society. University of London,
1-3 June 1969.
294
F o x , Kin, Clan, Raja and Rule, p p . 3 4 - 9 ; Beck, Peasant Society in Konku,
pp. 257-9.
295
See L. D u m o n t , Hierarchy and Marriage Alliance in South Indian Kinship (London,
1957)-
296
A. Beteille, Caste, Class and Power: Changing Patterns of Stratification in a Tanjore
Village (Berkeley a n d Los Angeles, 1965), p . 82.

129
 The emergence of provincial politics

separate 'true' Gounder Vellalas from the hosts of rich peasants who
had adopted or were adopting Gounder ceremonies, dress and cus-
toms.297 In the census of 1891, Sir Harold Stuart noted the ability of
the Nairs of Malabar to absorb immigrants from Tamil and Canarese
areas in a single generation without apparent friction;298 and the
internal structure of the Nairs provides ample evidence of the rise
and fall of indigenous social groups within it.299 Similarly, Thurston
recorded a famous Tamil proverb which describes the regular genera-
tional flow between the Maravar, Aghumudayar and Vellala castes.300
In more recent work, S. A. Barnett has suggested the diverse origins
of those presently filling the category of Thondamandala Vellala.301
In these conditions, in which sub-regional varnas were so amor-
phous, the politics of caste confrontation were rare and circumscribed.
We find them only on those exceptional occasions when the normal
paths of mobility were blocked, either by difficulties over speed and
information,302 or by the law which administered the word of the
Vedas rather than that of custom, or by aspirations which touched
Brahman sensibilities. Moreover, where conflicts occurred, there is
much to suggest that their claims to be caste - that is, subregional
varna - confrontations owed more to rhetoric than to fact. Obviously,
only those groups within a caste, which had gained money or power
to aid their desire for upward mobility, could attempt seriously to im-
prove their status. Unable to escape from their ignoble backgrounds,
they were forced to contemplate establishing a new status category
for their whole caste. But it is by no means clear that their energy in
pamphleteering and polemic made much impact on the less privileged
members of their caste, who formed the majority and whose ambitions
were limited by lack of power. After sixty years of activity, for
example, the Christianised Shanars of southern Tinnevelly had not
managed to expand their movement widely even in their own district:
It must not be supposed that the 'Sanror' theory has by any means spread a
general infection over the whole community. In villages near the source of
297
Nicholson, Coimbatore District Manual, pp. 5 7 - 6 1 . Also, Beck, Peasant Society
in Konku, pp. 2 5 7 - 9 .
298
Census of India. 1891. Madras. Volume XIII (Madras, 1893), p. 222.
299
M.D.G. Innes, Malabar, I, 9 3 - 1 0 3 .
300 Thurston, Castes and Tribes of Southern India, 1, 7.
301
S. A. Barnett, 'Development and Change in the Kondaikatti Vellala Com-
munity' in Justice Party Golden Jubilee Souvenir. 1968 (Madras, 1968), pp. 3 8 2 - 3 .
302
Obviously, the prevailing systems o f social mobility meant that movement could
not be rapid nor could it take place easily if the origins o f the mobile group
remained obvious to the rest o f the society.

130
 The political economy of Madras

its origin it finds often vehement supporters; but one need not go f a r . . .
to find the story either treated with derision or even not known. 303

Equally, the movement of the Shanar—Nadar traders of northern

Tinnevelly and Ramnad was unable to establish connections with non-
trading Shanars further south.304 Or again, although an association of
richer Pallis had been set up in 1888 to claim Kshatriya status and to
change the name of the caste to Vannikula Kshatriya, by 1911 90 per
cent of the 'community' was still returning itself as Palli to the census
reporters.305
Before 1920, the sub-regional varna level of activity was not an im-
portant political category. While challenges to it could certainly lead
to disturbance and to the temporary breaking of cross-communal
clientage networks, they were unusual and generally incapable of
attaining an institutional permanence. Nadars and Maravars, for
example, could be found in political alliances which were mutually
beneficial,306 and the Vannikula Kshatriya movement never meant
that rich North Arcot Pallis were unable to indulge in faction fights
against each other. We must look below the level of sub-regional
varna, to the sub-castes and endogamous units which composed it,
to find more significant political categories of caste. Once we attempt
this, however, it becomes necessary to take the notion of caste beyond
the concept of status and to relate it to the context of political and
economic life. Familial and quasi-familial organisations played an
important part in the obtaining and distribution of material resources
and power: their internal structure was to a considerable extent deter-
mined by the functions which they performed. Thus families and
lineages involved in the control of land would be likely to have dif-
ferent internal structures from those involved in, say, trade or
religion.

Caste in rural Madras. Let us first look at those castes related to the
peasant economy in the countryside. As M. N. Srinivas has seen, rural
society can be roughly parcelled into territorial divisions, each of
which was under the control of a 'dominant caste' — that is, a caste
which controlled most of the land and possessed effective political
303
M.D.G. Pate, Tinnevelly, 1, 129.
304
Hardgrave, Nadars of Tamilnad, p . 184.
305
Rudolph, The Modernity of Tradition, p p . 4 9 - 6 4 ; Census of India. 1911.
Madras. Volume XIII. Part 2 (Madras, 1912), p p . 1 1 6 - 1 7 .
306
Hardgrave, Nadars of Tamilnad, p p . 2 2 2 - 4 .

131
 The emergence of provincial politics

power.307 Of course, these dominant castes were of the sub-regional

varna-type; they were merely status groupings not practical organisa-
tions. Within each were further subdivisions, each also, usually, of
a territorial nature. How powerful were the sub-caste institutions
of the various Reddis, Vellalas, Kammas, Nairs, Kallars and other
dominant castes? Obviously, there were considerable differences, but
in general the answer must be that they were fairly insignificant and
were tending to lose progressively what influence they had. By the end
of the nineteenth century, very few of those dominant sub-castes in-
volved in agriculture had retained panchayats to enforce discipline
within their territories. In Trichinopoly district, for example, only
three of twenty Vellala sub-castes possessed operational panchayats. 308
In the Ceded Districts, Reddis had long since abandoned their use,
while in Andhra panchayats among the Kammas were virtually un-
known. The removal of large-scale warfare from politics made the
sub-caste and extended lineage redundant as units of military recruit-
ment. At the local level, the administration no longer recognised or
worked through caste institutions but handed revenue collecting,
police and judicial powers to particular families. The processes of the
economy were working to the advantage of individuals and families,
rather than of larger social units, and were increasing social strati-
fication within sub-caste groups. 309 Whatever role local-level caste
institutions may have had in 18.00 was whittled down in the course of
the following century. A classic case of this erosion can be seen in
the history of the Kallars of Tanjore and Madura. Originally warriors
and cattle-thieves, they used a very strong lineage organisation to
carry on their economic and political activities. During the nine-
teenth century, however, several lineages became settled in agri-
culture, under the great Periyar irrigation scheme in Madura and in
eastern Tanjore. Very rapidly, lineage discipline in the settled areas
disappeared and the agrarian Kallars become scarcely distinguishable
in organisation from other dominant peasant groups. 310 Only in the
extreme South, where they continued to perform criminal and private
police functions, did Kallar and related Maravar lineage organisations
keep their importance.
307
M . N . Srinivas, Caste in Modern India and Other Essays (Bombay, 1962), pp. 9 - 1 1 ,
89-93.
308
M.D.G. Hemingway, Trichinopoly, 1, 102; see also, F . R. Hemingway, Tanjore
(Madras, 1906), 1, 84.
309
I n an hypergamous marriage system, this social stratification was likely to lead
to t h e further fission of sub-castes.
310
Dumont, Une Sous-Caste de Vlnde du Sud, pp. 7 - 9 ; M.D.G. Francis, Madura,
h 92-3-
132
 The political economy of Madras

One of the few cases in which a dominant peasant group is said

to have preserved its caste institutions into the twentieth century
is that of the Gounder Vellalas of Coimbatore and Salem, about
whom much has been written. The Gounders are credited with having
an elaborate structure for the maintenance of caste discipline, which
reached up from the village through several intermediate levels to
four pattagar leaders who each held a separate territory. In theory,
the prescriptive ritual authority of the pattagars permeated the
Gounder Vellalas and held them together in a tightly organised
community.311 By the nineteenth century, however, practice seems to
have been very different.
There is no gainsaying the fact that the pattagars possessed great
social influence; but there is some question of whether the basis of this
influence was material rather than ritual. Most of the pattagar
families were landowners and credit controllers on a grand scale.
They formed a natural centre of authority for their neighbouring
lesser cultivators, many of whom were Gounders.312 They acted as
arbitrators in the disputes of some of their caste-fellows and exacted
penalties for what they saw as social indiscipline.313 Yet several other
Gounder families, such as the Vellakinar of Coimbatore taluk and the
zamindars of Utukuli, who also were very powerful but who had no
prescriptive ritual position, acted in a very similar way and exercised
a very similar species of authority. Eventually, the Vellakinar and
Utukuli families came to marry into the pattagar theocracy, making
a social unit coherent in its wealth but disparate in its relationship
to clan hierarchy.314 Moreover, pattagar prestige was entirely incap-
able of withstanding the collapse of its economic fortune; those
pattagar families who lost their land also rapidly lost their social
pre-eminence.315
Beneath the pattagars, the intermediaries who were supposed to
carry their writ into the village seem altogether to have disappeared.
In a recent book, Brenda Beck has noted how extremely shadowy
becomes the structure and personnel of the middle level of caste
organisation once it is brought under the analyst's microscope: few
people can be found to fit into the hypothetical system. Apart from
the pattagars, Beck can define major authorities in the Gounder
311
Nicholson, Coimbatore District Manual, pp. 58—9; Beck, Peasant Society in
Konku, pp. 40—9.
312
Beck, ibid., p . 4 3 .
313
Ibid., p . 4 0 .
314
Baker, Politics of South India, 1920-1937, p . 1 1 5 , fn 114; p . 2 3 5 , fn 2 6 8 .
315
Beck, Peasant Society in Konku, p. 42.

133
 The emergence of provincial politics

hierarchy only at the very local level where, again, ritual role is
congruent with economic power.316 In her case, working in the
1960s, this level is that of the village street and immediate neigh-
bourhood; in ours, working in a period of different economic design,
it was larger - the kingdom of the rural-local boss. Above these
bosses, however, unless they wished or were forced to call in their
pattagar, there existed no obvious political organisation of caste.
The duties of maintaining social discipline and control, which
once may have fallen to caste panchayats and lineage heads, were being
taken over by particular families who were de facto rural-local bosses.
Superior systems of caste organisation were shattered. Of course, this
development made very little difference to caste qua status in the
rural locality. There, bosses still needed to preserve the dignity and
ritual position of their dominant caste status and were wont to punish
severely lesser members of their caste who had broken the rules of
conduct. But no system of power existed to ensure that they them-
selves were subject to the same discipline, and frequently they
flaunted the regulations which they insisted others should obey.317
Formal caste institutions among rural dependent groups were
much more vigorous. Village priests, musicians and artisans tended
to belong to kinship and caste organisations which reached over
several rural-localities and were coherently structured. 318 Although
much more localised, subordinate labouring groups also used pan-
chayats or recognised 'headmen' to preserve social discipline in their
village streets or separate hamlets.319 It ought not to be supposed,
however, that the mere existence of these institutions meant that
these castes preserved a separate political existence, apart from other
groups in the countryside, nor that they formed separate political
communities. Certainly, the institutions of priests and artisans acted
in part as trade unions, to exact better conditions for their members
and to protect the exclusivity and sanctity of their occupations. But
they were economically dependent and could seldom push their
demands to the point at which dominant castes were seriously in-
convenienced. Where, for example, one group of priests refused to
316
Ibid., p . 7 0 .
317
For example, V. C. Vellingiri Gounder, an important rural-local boss, who
ran a purity and temperance campaign among his caste but who, himself, was a
notorious drunkard. See Reforms Franchise B March 1921, Nos 34-99. N.A.I.
318
F o r example, see K a m m a l a n s in T h u r s t o n , Castes and Tribes of Southern
India, 111, 106-49.
319
F o r example, see K a n n a v a r s in M.D.G. B. S. Baliga, Madurai (Madras, i 9 6 0 ) ,
p. 129.

134
 The political economy of Madras

perform heterodox ceremonies for their patrons or one troupe of

musicians refused to take part in a novel festival, dominant village
groups would be able to dismiss them and bring in others who were
more pliable.320 Communal solidarity, even among these highly
specialised and important village servants, did not extend beyond the
immediate territorial unit nor go far when faced with starvation.
In the case of labouring groups, it would be possible to argue
that caste panchayats remained so active because they were used
by the rural-local bosses to hold the lower orders in their place.
They enforced obedience to customs which clearly marked off their
members as inferiors. There was no sign in British South India, as
there was in Travancore, that they were being used to upgrade the
status of their members or to organise strikes against dominant castes.321
The subservient position which they exercised can also be seen in the
type of disputes in which they had jurisdiction. Quarrels over property
or economic rights were much more likely to be taken to a rural-
local boss of another caste than to the caste body or headman.322 The
economic leaders of the countryside were very unwilling to let slip
from their grasp any matter of consequence to their position.

Caste in urban Madras. The different nature of economic, social

and political relations in the towns naturally created some differences
in the character of caste relationships between rural and urban inhab-
itants. The greater complexity of town life and the greater inter-
action of ritual groups of various origins tended to make communal
organisations more solid. Immigrant groups were drawn to those
areas of a town in which they had kin connections; wider employment
opportunities broke up ritualised master-servant relationships; social
competition during festivals and ceremonies was much more inten-
sive; officials often fostered caste organisation by referring disputes
to caste headmen or panchayats whom they had found or invented.
These factors helped to promote the development of politically
important communal ties which extended beyond the family and
which did not rely on status challenges to make them apparent.
Nonetheless, it would be wrong to consider that the whole area of a
town and all the social groups within it were subject to these speci-
320
Madras Police 1886, p . 2 4 ; Obituary of P. Subrahmanyavadhayajulu in Hindu
31 July 1897; Hindu 31 August 1896. T h e flexibility of (even Brahman) service
castes before t h e economic power of a dominant caste has been noted in Beck,
Peasant Society in Konku, pp. 1 5 4 - 8 0 .
321
Hardgrave, Nadars of Tamilnad, pp. 6 5 - 7 .
322
Beck, Peasant Society in Konku, pp. 7 6 - 7 .

135
 The emergence of provincial politics

fically urban pressures. Most of the conurbations of Madras were no

more than collections of villages, interspersed with agricultural land,
which had been linked together for administrative convenience. The
warehouses, daily bazaars, workshops, government offices and law-
courts, which created the urban setting, were but a part of the average
town and involved but a proportion of its residents. Only in the fast-
growing towns of the Andhra deltas and, later, the southern cotton
belt, were these 'urban' elements increasing appreciably. The agricul-
tural landlords, labourers and dependent groups, whom circum-
stances had placed inside a municipal boundary, conducted their
affairs in ways very similar to their counterparts in the country-
side proper. The significant urban communal organisations were
only those of artisans and skilled workmen, of bureaucrats and mem-
bers of the educated professions and of traders.
Although by 1880 most of the hereditary caste occupations of
artisans had been broken down and invaded by other groups, in
particular places particular artisan businesses were organised on
communal lines. In the towns of the Ceded Districts, for example,
those weavers who were working together either in the same work-
shop or under the same petty capitalist usually would be bonded by
caste as well as by economic ties. They would participate in a ritual
life which separated them, in the weaving area of the town, from
other weavers. At this low level of neighbourhood activity, there
tended to be endemic conflict between caste groups in the same trade.
In towns such as Nellore or Cuddapah, which had thriving weaving
industries, Hindu Khattri, Muslim and Hindu pariah weaving
groups were constantly at each others' throats. The petty capitalists,
who were of the same community as their workers, encouraged these
fights and often led their employees into the attack, for the victor
gained not only prestige but a larger share of trade. 323 However, as
we saw earlier, the petty capitalist was himself tied to a higher com-
mercial power for his credit, supplies and marketing arrangements.
The bonds of caste usually did not extend this far. In the towns of the
Ceded Districts, the men who linked producers to markets in the rest
of the presidency and Bombay were essentially financiers whose
clients included petty capitalists from many communities.324 They
323
Deputy Inspector General [of Police], Central Range, to Inspector General,
Madras, 14 December 1914 in G.O. 2821 (Judicial) dated 16 December 1914.
T.N.A.; see also Madras Police 1912, p. 18.
324
Bombay Multanis were particularly prominent in the piece-goods trade of the
Ceded Districts; further south, the trade was in the hands of the Nattukottai
Chetties.

136
 The political economy of Madras

had little sympathy for disruptive trade wars and could be counted on
to interfere in them to re-establish peace. Moreover, even where
financiers shared caste ties with petty capitalists, as among the
Sourashtras of Madura, other factors kept them from supporting
neighbourhood confrontations. They tended to be involved in other
areas of commerce and in close relationship with the government,
both of which would suffer if they were seen to be inciting communal
violence. They too clamped down on neighbourhood conflict and
prevented it from spreading.325 The great outbreaks of rioting and
disorder, which occasionally appeared in the Ceded Districts5 towns,
coincided with famines and trade depressions when the rivalry
between petty capitalists became intense and when the flow of re-
sources from superior magnates dried up. 326 But in times of normality,
communal dissension of this kind was locked away in one corner of the
town and had a very limited importance.
Communal linkages between families involved in government
service and the liberal professions also were tight. 'Service' families
needed considerable resources to educate their children. As education
could suffer from economic instability, they often lived together in
one neighbourhood and combined to protect themselves. They
founded small local schools and, from the 1870s, began to set up
nidhis and mutual insurance schemes.327 Technically, there was
nothing in this activity which was bounded by caste. Neighbourhood
schools, like the Thondamandala High School in Madras city, took
pupils from different castes, while mutual insurance companies, like
the Mylapore Hindu Permanent Fund, took deposits from any-
body. Several other factors, however, guaranteed that caste-community
did play a role. The number of social groups in any one area who were
interested in service was very limited. Brahmans and a few Sat-Sudra
groups predominated. Within the neighbourhood, they would tend
to be separated from each other by streets.328 Moreover, being of a
very high status they enjoyed active ritual lives which, at ceremonies
and festivals, divided them from each other.
325
Deputy Inspector General [of Police], Central Range, to Inspector General,
Madras, 14 December 1914 in G.O. 2821 (Judicial) dated 16 December 1914.
T.N.A.
326
As during t h e 1918 famine, see 'Judgement of C.C. No. 1, Special Deputy
Magistrate, Cuddapah', in G.O. 1538 (Judicial) dated 1 July 1918. T.N.A.
327
Such as t h e T h o n d a m a n d a l a High School, M a d r a s city, see below pp. 203—4,
and t h e Mylapore H i n d u P e r m a n e n t F u n d , see Nicholson, RAB, II, 2 2 1 -
37-
328
I n M a d r a s city, for example, a large block of T h o n d a m a n d a l a Vellalas lived
tightly packed in one municipal ward, see below pp. 2 0 3 - 4 .

137
 The emergence of provincial politics

Although more than one caste would be educated at these schools

and protected by these benefit societies, very often membership of
the school and benefit society rested largely with one community.
Further, once a man had gained access to a consequential govern-
ment office, or had established a successful legal practice, he would
tend to use connections of family and friendship to maximise his
profits and to cut out rivals. Jobs and cases could be fed back to fatten
relations and contacts who would be loyal clients. Moreover, at the
low level of neighbourhood activity and among such prestigious
ritual groups, friendships commonly were made by participation
in ceremonies and cultural associations.
As in the case of artisans and workmen, however, compact local
units of caste among 'service' groups were broken by economic ties
with outsiders. Lawyers had to find members of other communities
to give them cases, clerks and gumastahs needed employment with
brokers and contractors who seldom were of their caste, and lower
government servants had to rely on the co-operation of various local
notables. Very rarely were these 'service' groups free to take political
action as communities. In Madura in the early 1880s, for example,
there was a storm of protest over an aspect of municipal government
from the area of the town in which most of the Brahman clerks and
lawyers lived. A Ratepayers Association was formed to campaign at
the impending elections for the protection of Brahman religious
privileges. Yet, when the votes were counted, it became clear that
the Brahmans were unable to poll a majority even in the ward in
which they formed most of the voters. A merchant-financier, living in
another part of the town, was returned as their candidate. 329 It was
also in Madura in 1915 and 1917 that K. M. Alladin Rowther, the
notorious Muslim criminal, was elected from the same Brahman-
dominated constituency. 330 Later, we shall see that the Thondaman-
dala Vellala constituency in Madras city was equally open to extraneous
influences.
Only among trading caste groups was there sufficient economic
independence for a significant communal solidarity to develop in
politics. As we saw earlier, the basic financial organisations of
Madras were constructed on familial or quasi-familial models. Al-
though no group apart from the Nattukottai Chetties developed
institutions capable of linking together caste members from many
dispersed localities, particular trading groups in and around particular
329
Hindu 21 April 1884.
330
G . O . 1384 (L a n d M , M) dated 16 August 1917. T.N.A.

138
 The political economy of Madras

towns did produce local communal organisations which attempted to

control business and to provide welfare facilities for their members.
R. L. Hardgrave has described the communal institutions to be found
among Nadar merchants in their 'six towns' in Ramnad and northern
Tinnevelly.331 In the principal Andhra towns, Komatis were organised
on almost identical principles. They levied communal fees, set up
their own schools and charities, built their own temples, provided
arbitration for business disputes and enforced a fierce social and
economic discipline.332 When, for example, Komati elders ordered
a religious hartal in Madras city in 1898, not one Komati bazaar
seller dared to disobey.333 Maracair Muslims on the east coast, rising
Tiya traders and bankers in northern Malabar and Christian Bharatha
merchants in Tuticorin worked to the same pattern.
In his work on Northern Indian bankers, C. A. Bayly has raised
critically the question of whether these tight caste connections
among mercantile groups were so much the result of caste identity
as of the practical necessities of business.334 The strength of com-
munal rule came not only from recognition and sentiment but from
the fact that it protected an essential credit-raising institution. An
out-casted member was likely to be bankrupted very quickly. In some
ways, Bayly's point may seem irrelevant here: whatever its causes,
local mercantile caste groups did work together as communities. In
other ways, however, it is crucial, for if we examine the mechanics
of credit, these institutions cease to be communal - in the sense of a
spontaneous combination of equals — and become features of the
rule and discipline of urban economic powers.
In South India, trading capital was concentrated in a few families
in each town. As we have seen, these families worked with or through
a cross-section of society. However, their willingness to lend and the
terms of interest which they demanded were not equal for everybody.
They needed greater guarantees of security than the British law could
supply and, consequently, they tended to judge the safety of an invest-
ment by the 'creditworthiness' of the debtor. 'Creditworthiness' in a
debtor could be assessed by his social position, known responsibility
and available assets. But it became much less a matter of speculation
when the creditor possessed an instrument of direct social control over
331
H a r d g r a v e , Nadars of Tamilnad, pp. 9 5 - 1 2 9 .
332
G.O. 3360 (L.S.G.) dated 8 September 1928. S.A.H.; T h u r s t o n , Castes and
Tribes of Southern India, 111, pp. 307—11; Hindu 30 May 1896.
333
Hindu 6 J a n u a r y 1898; see also a K o m a t i hartalin Salem in Hindu 20 June 1894.
334
Bayly, 'Development of Political Organisation in t h e Allahabad Locality, 1880
to 1925', ch. 2.

139
 The emergence of provincial politics

the debtor. It was this instrument which caste institutions formed.

Urban financial powers knew that they could lend out a high propor-
tion of their assets to their caste-fellows and use caste discipline
to enforce repayment if necessary. In Ellore, Mothey Venkataswami
financed local trade through other banias most of whom were
Komatis; 335 Tamil Muslims gathered at Vanniyambadi provided each
other with a pool of capital for business in the hides trade; 336 Nat-
tukottai Chetties lent extensively to each other and took each others'
hundis331 It was very much in the magnate's interest to preserve a
tight caste or communal discipline for it secured his own financial
stability.
There is much to suggest that caste institutions among mercantile
groups functioned largely on the personal authority of the magnates
behind them; and that once that authority was withdrawn, they col-
lapsed in confusion. Ritual institutions, and prescriptive positions
within them, did not count for much unless the power of resource
control was present to lend them support. In Madras city, for example,
the family of the Komati headmen were able to discipline the caste
only so long as their great wealth and control of the caste temple
and its properties clearly marked them out as economic leaders. By
the 1890s, the growth of other wealthy Komati merchants in the city
and attacks by the courts on the estates of the current headman,
C. V. Cunniah Chetty, had weakened the family's position. Caste
regulation of marriage was flouted openly and groups of dissident
Komatis challenged Cunniah Chetty's authority and right to control
the caste temple.338
The dependence on magnate leadership can be seen also in the more
'secular' and properly political organisation of caste constituencies.
In Guntur town, the leading local banker, trader and factory-owner,
Lingamalee V. Subbaya, and his Brahman lawyer V. Bhavarnacharlu,
put together a Taxpayers Association which was in reality a Komati
caucus. From the 1890s, when it was formed, it brought political unity
335
A. R. Banerji, Assistant Collector, Ellore, t o Collector of Godavari, 29 M a r c h
1901 in G.O. i o n (L and M, M) dated 17 July 1901. T.N.A.
336
T h e y also financed explorations i n t h e Kolar Gold Field, Mysore, a n d t h e
spice a n d areca n u t trades from Malabar.
337
RPBC,pp. 186-9.
338
I n 1900, t h e temple a n d its charities possessed a n income of over R s J lakh a n d
owned t h e largest m a r k e t in t h e city. Suryalokamy IJ M a r c h 1901. R.N.P. See
also Hindu 4 M a y 1894, 2 M a y 1896, 8 September a n d 13 December 1904, 27
M a r c h 1913.

140
 The political economy of Madras

to the Komatis and would have dominated the town completely had
not the Collector interfered with it.339 However, this Komati soli-
darity proved very short-lived. Hardly were Lingamelee V. Subbaya's
ashes cold, than rampant factionalism destroyed the community's
political base.340 Similarly, in Ellore, where the power of the Mothey
family was strong, the Komatis were organised behind it. Mothey
Gangaraju, the leader of the family in the 1910s and 1920s, was able
to take his local community with him in whatever politics he chose
to play. In 1917, for example, he made contacts with the Justice
Party; in 1921 he non-co-operated; in 1923, he joined the Legislative
Council as an Independent.341 His contortions produced no difficul-
ties in his home base. But by the depression of the 1930s, the family
empire was on the rocks and was losing power through unprofitable
zamindari investments and an abortive factory project. Increasingly
dependent on the government for protection of their land and for
loans for their factory, the Motheys were forced into loyalism between
1935 and 1937. Many members of their local caste rebelled against
them and, in the 1937 elections, the Ellore Komatis were split down
the middle between the Congress and the Justice Party.342
A close analysis of the Nadars of Sivakasi and Virudhunagar brings
out the same points. In Sivakasi in 1921, hosts of petty Nadar traders
raised a rebellion against the foundation of a municipal council. They
closed the bazaar and boycotted the council chamber. However, the
leading Nadar families of Sivakasi, through the Nadar Mahajana
Sangham, had just concluded a deal with the British for a caste
Legislative Council seat and were eager to appear loyal. They put
pressure on their lesser brethren and forced them to reopen their
shops and to return to the council.343 In Virudhunagar, the wealthier
Nadar families reinforced their caste rule with what Nadar petitioners
described to the Secretariat as terrorism.344 As in the case of the
Motheys, however, the economic basis of their supremacy was re-
vealed by the havoc caused to their leadership by the depression.
From civil disobedience onwards, the Nadar leaders were under
339
G.O. 1298 (L a n d M , M) dated 5 August 1892. T.N.A.
340
Venkatappayya, Sviya Caritra, 1, 155—9.
341
G. V. Subba R a o , The Life and Times of Sir K. V. Reddi Naidu ( R a j a h m u n d r y ,
1
957)3 P- 3 1 ; Hindu 28 July, 6 a n d 19 August 1 9 2 1 ; H o m e Public File 9 5 3 of 1924
T.N.A.
342
Baker, Politics of South India 1920-1937, ch. 3.
343
G.O. 7 (L.S.G.) dated 3 January 1921. T.N.A.
344
G.O. 2767 (L.S.G.) dated 29 June 1926. T.N.A.
 The emergence of provincial politics

constant pressure from members of their own community, and local

caste units were torn apart by increasingly rancorous disputes which
they no longer had the power to settle.345
For political action, these communities were bonded by their
common dependence on one or a few families: there were no insti-
tutions to hold them together if that dependency were broken.
While certainly the ties between the financial magnate and his caste
members were stronger than the ties between him and outsiders, both
sets of ties were structurally similar. During Mothey raj in Ellore,
the family's influence reached non-Komatis as well as Komatis; and
when that raj declined, its influence was ignored as much by Komatis
as non-Komatis. In Sivakasi, the Nadar merchant princes raised
taxes for communal purposes from non-Nadars as well as Nadars;
and in 1921, their order of a return to work was accepted by non-
Nadars as well as Nadars. If a magnate so willed, his community
would be taken into political alliances which were ritually obnoxious
and absurd: the Nadar merchant-princes forged an alliance with the
Maravar Raja of Ramnad when both parties found themselves serving
the Justice ministry; in Madura, Sourashtra dye-magnates consis-
tently worked behind Brahman lawyers, with whose caste they were
supposed to be in dispute.346 Neither of these alliances caused any
noticeable dissatisfaction in the Nadar, Maravar, Sourashtra or
Brahman populations. Indeed, in instances in which local trading
groups found themselves economically dependent on an outsider,
they were prepared to take his word as law even in matters of caste
ritual and custom.347
Formal caste-institutions of themselves did not create independent
political corporations of caste. Once again, we come back to the point
at which it seems caste outlined important political constituencies
only when questions of status were mooted. As urban settlement
patterns were complex, there were more occasions of status rivalry
in the towns than in the villages. No large town was under 'one-caste'
domination,348 and many different social groups349 were active in the
same trades and occupations. Magnates, who spent so largely on cul-
tural and religious patronage, could create communal hostility to
345
Hardgrave, Nadars ofTamilnad, pp. 184-8.
346
G.O. 382 (L a n d M , M ) dated 29 F e b r u a r y 1912; G.O. 1074 (L a n d M , M )
dated 12 J u n e 1912; G.O. 1843 (L a n d M , M) dated 24 October 1916. T.N.A.
347
As i n t h e case of t h e Komati merchant P. R a m a m u r t h i a n d t h e M a h a r a n i of
Vizianagram cited above. See Hindu 14 J u n e and 7 July 1910.
348
Given t h a t caste was n o t a political institution, by this we mean u n d e r t h e
domination of a group of families w h o shared t h e same caste status.
349
T h a t is family groups w h o were of different caste statuses.

142
 The political economy of Madras

themselves. Rich Muslims who built mosques and Madrassas, Hindus

who protected the cow, or Tiyas and Nadars who bought a new caste
status and founded their own temples and rituals, sometimes were
offering overt challenges to existing social hierarchies. In circum-
stances of economic instability, status competition between local com-
munal groups could become chronic. In the fast-growing towns of
the Andhra deltas, for example, rising Komatis demonstrated a ritual
and cultural assertiveness which set by the ears resident Muslims and
non-Komati Hindu magnates.350 In fast-declining towns, like Salem,
squabbles between family-organised businesses for shares in a dim-
inishing pot produced serious situations of communal polarisation.
In Salem, Komatis, Beri Chetties and Tamil Muslims came to
blows.351 Whenever a town economy faced crisis, as during famines,
spontaneous communal rioting could break the cross-communal
clientage network of the strongest magnate.
A single-minded concentration on these features of conflict, in
which linkages of caste (and other perceived cultural statuses) cut
through the ties of economic clientage, would give the impression
that caste and religious sentiments were the dominant influences on
urban political organisation in Madras. Indeed, several previous
historians of urban South India have analysed their material entirely
in this way and without any regard to the operations of the eco-
nomy.352 A broader view, however, puts the proportion of activity
explicable by reference to cultural antipathy more firmly in its place.
Firstly, categories of cultural antipathy simply were not generally
politicised by magnate expenditure on personal status. Only the
north-east of the presidency had felt any weight of Muslim oppres-
sion so that only there was the overt support of Hinduism or Islam
capable of being interpreted as an aggressive political act. Elsewhere,
Hindu and Muslim magnates could be found involved in a range of
religiously promiscuous practices. In Madras city, for example, the
millionaire Calivalla brothers, who were Komatis, supported both
Arabic education and Hindu revivalism;353 in Vanniyambadi,
Muslims paid for the upkeep of the Hindu temple;354 in Negapatam,
350
G.O. 2461 (Judicial) dated 26 November 1917. T.N.A.
351
G.O. 1374 (Judicial) dated 23 M a y 1883. T.N.A. Salem town declined rapidly in
the later nineteenth century following t h e closing down of its army barracks.
352
See, for examples, Srinivas, Caste in Modern India, p p . 1 - 4 0 ; H a r d g r a v e ,
Nadars of Tamilnad, pp. 1 - 1 1 .
353
Hindu 12 M a r c h 1918; see also Hindu 21 December 1887, 30 M a r c h 1888, 23
April 1908, 25 April 1912, 14 April 1920 for other examples of Hindu-Muslim
cross-communal patronage.
354
G.O. 2099 (L a n d M , M) dated 8 November 1912. T.N.A.; i n 1936, Abdul H a k i m
bought an elephant for t h e local H i n d u temple, Hindu 27 April 1936.
 The emergence of provincial politics

Hindus contributed to the main mosque;355 in Palni, Hindus and

Muslims even worshipped at the same shrine.356 Similarly, low-caste
groups which had reached magnate proportions did not have to attack
the culture of the dominant groups which they had just joined. In
most cases they bought their way to respectability slowly and without
occasioning caste confrontations. Most of the towns of Madras were
cosmopolitan and, if he were to achieve one of his primary ends, a
magnate had to win the recognition of his status from as many areas
of local society as possible. Except in unusual circumstances, most
magnate patronage was spent in cross-communal and, particularly,
cross-caste activities. This can be seen most obviously in the fact that
rich non-Brahmans continued to support Brahman religion and
scholarship. But magnate patronage reached down as well as up the
religious scale. Annadana Samajams (for feeding the poor), famine-
relief kitchens, rest-houses, night schools and pariah schools, which
were funded by magnates, were aimed specifically at relating low-
caste groups to their authority.
Secondly, as we have seen, the conditions of economic instability
which promoted communal tension were by no means normal in the
South India of our period. There were few fast-rising or fast-declin-
ing towns and, after 1878, famine was an infrequent visitor. More-
over, even where economic turbulence did produce communal
confrontation, it ought not to be supposed that communal divisions
completely subsumed all other types of division, particularly those
formed by the rivalry of various magnate clientage networks, nor that
they became built into the structure of urban politics. Bezwada, for
example, was among the fastest-growing towns in Andhra and its
Komati population among the most ritually assertive. Yet, between
about 1908 and 1922, the town was run by G. Appalaswami, a low-
caste Nagara labourer, who, his social status notwithstanding, em-
ployed Brahmans and Vaishya Komatis to work in his interest.357
Nor need communal divisions last long after they had been politicised
for a particular purpose. In Salem, only two years after the 1882
riots which stemmed from Tamil Muslim-Komati rivalry, the
municipal council was being operated without internal friction by a
Muslim merchant chairman and a Hindu (including Komati) mer-
355
G.O. 7 8 9 (L a n d M , M) dated 20 M a y 1911. T.N.A.
356
R. K. Das, The Temples of Tamilnad (Bombay, 1964), p . 97.
357
Hindu 12 J a n u a r y 1 9 1 1 ; Kistnapatrika 9 J u n e 1911 a n d 19 July 1912. R.N.P.;
G.O. 306 (L a n d M , M) dated 15 F e b r u a r y 1913; G.O. 2028 (L and M , M) dated
10 December 1919. T.N.A.; Kaleswara Rao, Na Jivita Katha—Navya Andhramu,
pp. 291—360 (Telugu).

144
 The political economy of Madras

chant majority.358 Equally, after any orgy of famine rioting in

the Ceded District weaving towns, the reassertion of magnate autho-
rity quickly put the lid back on communal disharmonies, leaving
little trace of them a few months later.
And thirdly, a concentration on the occasions of communal conflict
tends to ignore all those many other occasions when the cultural con-
ditions for conflict were present in a situation but were ignored by
the actors. What can be said, for example, of the limitations of poli-
tical power imposed on an individual by his untouchability when,
in Cuddapah in the 1890s, the pariah family of V. G. Vasudeva Pillai,
which had grown rich on the second Burma War, won elections in
constituencies of caste voters to the municipal council;359 or when,
in Madura in 1914, Brahman municipal councillors voted a Christian
convert from the Nadar caste to the chairmanship in preference to
another Brahman of a different faction;360 or when, in Kumbakonam
in 1888, a high-caste electorate voted an untouchable Nadar to a
temple-committee seat.361 In these three (and many other) cases, sup-
posed caste prejudices were made nonsense of by the political beha-
viour of the actors. That behaviour, however, appears less absurd
when it is set against reference points in the system of economic
clientage. The results of the three election situations are perfectly
in accordance with the structure of magnate power in the three towns.
In summary, then, while we may recognise the existence of a
number of factors which could destroy temporarily the pattern of the
clientage network, none was so strong or so long-lasting as to make
us alter our categories of political analysis. In particular, the factor of
caste sentiment was, for the most part, complementary to the econo-
mic order, providing a religious validation of positions attained and
supported by economic power. The enormous practical flexibility of
the South Indian caste system guaranteed that the circumstances of
conflict between the appeal of status-worthiness and that of economic
reward would be rare. But even where they occurred, the evidence
suggests that caste considerations were at least as often subordinated
to the dictates of economic clientage as they were superordinated to
it.362
358
G.O. 1304 (L a n d M , M) dated 16 N o v e m b e r 1888. T.N.A.
359
V. G. Venugopala Pillai t o Secretary, Local a n d Municipal D e p a r t m e n t , 14
October 1919 m G.O. 182 (L and M , M) dated 2 February 1920. T.N.A.
360
G.O. 1203 (L and M , M) dated 1 July 1912. T.N.A.
361
Hindu 23 July, 2 2 , 24 and 27 August 1888.
362
T h i s discussion is developed further in m y ' T h e Development of Caste
Organisation in South India, 1880 to 1925' in Baker and Washbrook (eds.),
South India.

145
 4
Local structures of political power

The British, then, were faced by a series of a strong, local political

structures which rested upon the organisation of the economy. Based
on the provision of a variety of facilities - for the continuation
of production, for social welfare and for the patronage of culturally
worthy objects - these structures were extremely hierarchic in
orientation. They placed in the hands of a few men means to control
the behaviour of many. During the nineteenth century, of course,
the British could not destroy these structures and, as it were, build
for themselves the type of political society which they would like to
administer. They were not prepared to commit themselves to
providing directly the same facilities as the various rural-local bosses
and magnates - and certainly would not have been able to meet such
a commitment had they made it. They had to use the local political
systems which were waiting for them in 1800 and which, to a con-
siderable extent, developed independently beneath them through to
1947. Yet, although weak and compromised by the locality, the British
were not entirely unable to influence the course of local events. The
resources and the powers which they could mark into or erase from
the social design did enable them to elongate or foreshorten the
patterns which they found. From the later nineteenth century, as it
was forced into ever greater activity, Fort St George came to use
its rubber and its pencil with increasing regularity. In this chapter,
our purpose is to examine the social and political consequences of the
ground-level enmeshment of the formal administration with local
political structures.

Rural Madras
The 'dry9 region
In the dry areas of the presidency, there is much evidence to suggest
that the impact of British government served to protect and enhance
the position of the rural-local boss, and hence to continue the process
of social stratification which the economy was promoting. At the turn

146
 Local structures of political power

of the nineteenth century, as Brian Murton has shown, rural-local

society was dominated by elite groups of agrarian decision-makers.1
However, sitting over and above them, there existed a further warrior
elite which had risen with the Vijayanagar empire and had thrived
on the wars of the eighteenth century. In varying degrees, according
to local circumstance, warrior rulers had pressed down on the re-
sident agrarian population, drawing off revenue and loot, and enforc-
ing political and legal decisions of their own choosing. This check
from above formed one of the main restrictions on the power of
agrarian decision-makers.2 Between 1800 and 1850, the British slowly
but steadily demilitarised the warrior elite over most of South India.
This did not mean that they necessarily destroyed it: some warriors
were transmogrified into landlords and others into commercial mag-
nates. However, many were liquidated and the power which their
armed force had given them over the countryside was generally
eliminated.3 As Burton Stein has shown, this led to an easing of the
pressure on agrarian society and, thereby, removed a check on the
rural-local boss.4
British rule also weakened another restraint on local power.
Wherever previous warrior regimes had sought to exercise a tight
control over rural affairs, they had introduced into the country-
side new social groups which were tied to their interests and which
usually participated in their own species of 'state-level' culture. They
supported these groups in the villages with political offices and con-
ducted their relationships with local society through them. The social
history of South India is layered with the bones of successive groups
of village administrators, each brought in by one conquest and wiped
out by the next. In the Northern Circars, for example, Visvabrahmana
administrators had been replaced and depressed by Lingayat ad-
ministrators who themselves had to give way to the Niyyogi Brahmans
who were tied to the next wave of conquerors.5 Similarly, in Salem
and Coimbatore, the Vijayanagar invasion had led to replacement
of various administrative groups by Sri Vaishnava Brahmans who were
the servants of Vijayanagar warriors.6 The most usual office to suffer
1
Murton, 'Key People in the Countryside'.
2
Stein, 'Integration of the Agrarian System of South India', pp. 188-96.
3
Particularly after the recommendations of the Torture Commission (1855) had
led to the removal of police powers from tahsildars, who often had used them to
aid zamindari interests. See Frykenberg, Guntur District 1/88—1848, pp. 48—9.
4
Stein, 'Integration of the Agrarian System of South India', p. 202.
5
Frykenberg, Guntur District 1/88-1848, pp. 12-17.
6
See C. M. Ramachandra Chettiar, Konku Natu Varalaru. (Annamalai, 1954),
pp. 339-45 (Tamil).

147
 The emergence of provincial politics

by these changes was that of kurnam (accountant) for it was through

the mechanisms of revenue collection and accountancy that the village
was most obviously related to superior powers. Behind the office of
kurnam, however, there lurked the more stable office of headman.
Most regimes permitted headmanships to lie with dominant families
among the supposedly cultivating peasantry for the headman was
expected to keep order and to supply force to back up governmental
decisions. These families were usually participants of the 'local-level'
culture which was distinct from that of the state. Warrior regimes
could not provide coercive force everywhere, all of the time. Govern-
ment under them, therefore, rested on a nice balance between state
demands and sanctions on the one hand and local power and obstruc-
tion on the other.
In local terms, the existence of this balance implied several
important questions about social and political control. At the risk
of over-simplification, we may see these questions as turning on the
relative position of the offices of kurnam and headman.7 When the
state was strong and the kurnam, as its creature, able to bring into
the locality force and resources derived from his connection to it, his
own local position tended to become dominant. Local society could
not deny his influence if government troops answered his call or if
senior administrators, who were members of his kin group or caste
group, accepted his version of the revenue accounts, and his alone.
When this happened, rural society to a large extent passed under his
influence and that of the 'state-level' cultural group of which he was
a member. The development of this control greatly weakened the
power of the resident 'local-level' agrarian decision-making elite.
However, if the state became weak or if, for some other reason, the
kurnam were unable to reach it, then his position declined rapidly.
Against the tight economic and social8 empires of the larger land-
7
The author recognises that there is some difficulty in reading back beyond 1800
the exact bureaucratic division between headman and kurnam, which character-
ised British rule. As many warrior regimes tended to lease complete village rights
to a single man (pedda ryot, patel, kadim, etc.), the same permanent duality in
the personnel of government was not as noticeable. However, even where state-
level groups obtained village leases, they had to rely on peasant leaders to
provide them with police control in their villages. Thus there was always some
form of state/local division, and village authority rested with at least two
different groups who derived their power from two different sources. See,
for example, the account of Rudravaram village, Guntur district, in R. E.
Frykenberg, 'Village Strength in South India' in Frykenberg (ed.), Land
Control and Social Structure in Indian History, pp. 231—47.
8
'Social' because in central and southern Tamilnad there remained some relation-
ship between land control and clan organisation. See Beck, Peasant Society in
Konku.

148
 Local structures of political power

holders, whose interests were voiced through the office of headman,

he could offer little resistance. Then, the rural locality passed back
under the hegemony of its greatest inhabitants. For several centuries,
political power in the countryside ebbed and flowed between an office
which relied on the effectiveness of state intrusion and one which
relied on state exclusion.
During the course of the nineteenth century, the overall effects of
British policy were to alter dramatically the nature of this balance.
The British simply did not bring in their baggage train an army of
petty administrators whose loyalty to them was unquestioned and
who shared their culture. Instead, they had to work their rural ad-
ministration with the same social material which they found in it. In
many areas, this material already possessed channels of interested
communication between the state and local levels. Early British rule
led to the broadening of these channels and hence to the broadening
of kurnam-type control in the localities. For example, as R. E.
Frykenberg has shown, the close relationship between various
Brahman groups in the huzur offices and the villages of Guntur dis-
trict produced conditions in which kurnam groups thrived. 9 However,
and importantly, these early developments owed nothing to the con-
scious intentions of the British themselves. They took place largely
because the British were distant from and ignorant of Indian political
systems. The incumbent Indian administrative groups were thus
allowed free rein to use state power as they pleased for their own ends.
As the British became more aware of the real situation, however,
they reacted violently to it and introduced a wholly new principle
into the state—local connection. Instead of trying to bring inside
the formal administration the existing web of relationships between
the huzur offices and villages, and thereby converting a strong net-
work of government which had worked outside their interest into a
strong network which could work inside it, they sought to destroy
the entire system by which the state could penetrate the locality.
Haunted by fears of 'Brahman conspiracy', they pursued with re-
markable singlemindedness the eradication of all social ties between
state and dominant local administrative groups and the establishment
of the headman's office as the centre of rural administration. In
achieving these aims, they succeeded firstly in guaranteeing the
political autonomy of the locality and, secondly, in laying the
foundations of an entirely new species of supra-local political
culture.
In 1802, when first settling Madras, the British had made the offices
of both headman and kurnam hereditary in the families of those who
9
Frykenberg, Guntur District 1788-1848, pp. 93-8.
149
 The emergence of provincial politics

presently held them.10 For the next half century, Fort St George
practised little administrative engineering, leaving rural government
to those who had captured it at or before the conquest. From the
1850s, however, the attack on 'administrative cliquism' and the
policy of aid to the village headman began in earnest. Orders were
passed to prevent blood and caste relatives from sharing the same
government duties and, more seriously, to reduce the powers and
prerogatives of the kurnam.11 In the revenue department, where he
held his only office, the kurnam became simply the man who added
up the accounts. The headman had responsibility for collecting the
revenue, for issuing the notes of demand and restraint, for the selling
of defaulters' lands and for calling in the higher authorities over
matters of debate. In view of the real nature of the ryotwari system,
this made him virtually responsible for apportioning the revenue
demand in the village. It would be difficult to overestimate the
influence given to the headman by his revenue office. It was generally
recognised, for example, that the ludicrously low prices obtained at
auctions of revenue defaulters' lands were due to the headman and
his subordinates who prevented high bids being made and who often
absorbed the land themselves.12 From 1886, the headman was also
responsible for collecting the income-tax.
In thefirsthalf of the nineteenth century, headmen had maintained
their age-old petty police functions with little help from the state. How
effective they were, in relation to the remnant of the warrior elite
or to the subedhar troops who often acted for district administrators,
it is difficult to guess. However, by the middle of the century, most of
the warriors had gone and the troops had ceased to back up revenue
collection. The headman was left on his own and, by Sir William
Robinson's police reforms of 1861, he began to receive government
assistance on his own terms. His staff of watchmen, previously paid
entirely by an informal contribution from the villagers or by grants
of rent-free land, was supplemented by state-paid assistants who were
appointed, however, by the headman himself.13 The provincial police
organisation was connected to the village only through the headman
and was largely dependent upon him for all its information. The
natural result of this policy was that the headman was able to
10
B. B. Misra, The Administrative History of India 1834-1947 (Bombay, 1970), pp.
458-62.
11
See Frykenberg, Guntur District IJ88- 1848, pp. 243-4.
12
Report of the Indian Famine Commission. Appendix. Volume III. Condition of the
Country and People. P.P. 1881, LXXI, pt 2, 416; MPBC m, 679.
13
Madras Police 1885, pp. 1-5.

150
 Local structures of political power

dominate his locality by force. Many headmen maintained private

armies for use against enemies and for pillaging each other's villages.
Of Salem, in 1896,
The Superintendant of Police remarks that almost all the violent crime in
this district is committed by Koravers who act in very many cases as private
Kavalgars in the villages. He considers that in very many cases these men are
in the hands of the Village Magistrates [headman], who use them as their
servants and, in consequence, protect them, taking care when crimes occur
not to mention any of their dependents in their first reports on which the
Sessions Court sets so much value. The Village Magistrates, of course, ob-
tain a considerable share of the proceeds of these looting expeditions.14
Similar reports came in regularly from all the dry districts. 15 The
system of terror at the disposal of the headman was also available
for keeping under control his village subjects, whose attitude to him
was seen as one 'of mingled fear and trust'. 16 The murder rate-even
the revealed murder rate which was but a fraction of the whole-was
extremely high. Headmen forced the obedience of their dependents
by personal violence, the burning of crops and the ham-stringing
of cattle. Few of their neighbours could or dared give evidence
against them, for it was difficult to reach the provincial police forces
except through tLe headman himself. Indeed, so closely involved with
the external police was the headman that the criminal courts usually
took his word as that of the law.17 This enabled him to prefer all
manner of charges against people who displeased him and back up
his accusations by swearing in his dependents as witnesses. By the
second decade of the twentieth century, when information from the
villages was improving, the problem of false prosecutions appeared so
serious that the Commissioner of Police began to keep a special set of
'false' crime statistics.18 But until this time, the headman's prero-
gatives were completely preserved by governmental ignorance.
In addition to the physical powers of a chief of police, village
headmen were also magistrates with their own jails and stocks, who
14
Madras Police 1896, p. 35.
15
Madras Police 1888, App. C, pp. xxi-xxii; Madras Police 1895, pp. 33, 185;
Madras Police 1912, p. 10; 'Report on Cuddapah' in Land Revenue 1904—5,
p. 72.
16
G.O. 121 (Judicial, Confidential) dated 28 January 1922, quoted in Venkataran-
gaiya, Freedom Struggle in Andhra Pradesh, m, 285.
17
Madras Police 1897, p. 12.
18
In 1918, the police were called in to investigate 5290 false complaints and
were involved in what turned out to be 4160 false prosecutions. Madras Police
1918, pp. 18-21.
 The emergence of provincial politics

could try petty criminal cases and execute their own sentences. The
British also used them in civil matters as small cause judges against
whose decisions no appeal was possible.19 Headmen tried about two-
thirds of the cases in their competence and also were known to
arbitrate informally in a great many more important village disputes
which went far beyond their legal competence.20 In spite of the warn-
ings of the Torture Commission (1855), which recommended that
the separate powers of government should not be held in the same
hands, the headman was responsible for all three functions of govern-
ment — taxation, arbitration and coercion.21
From the later nineteenth century, as British administrative activ-
ity increased, the office of the village headman became further inflated.
In order to stimulate agrarian improvements, the government began
to grant takavi loans (cheap credit for long term purposes). Of
course, these loans were administered through the regular revenue
machinery, that is through the headman. In order to conserve forest
resources, new laws were enacted and a new department founded; in
the village, their work was carried out by the headman.22 In order
to protect its territory from the worst effects of famine, Fort St
George, in common with other provincial governments, elaborated a
Famine Code; government grain doles, recovery loans and entry per-
mits to public works were administered on the ground by the head-
man. 23 In order to extend local roads, sanitation and primary
education, Fort St George pushed through local self-government
reforms on Lord Ripon's model. The smallest unit of rural local self-
government was the village union. By 1920, there were six hundred
of these unions, controlling about Rs 3000 per annum each. Headmen
were ex-officio members of unions in their territories and usually
ran them. 24 In order to expand litigation facilities without increasing
19
Report on the Administration of Civil Justice in the Presidency of Madras for the
year 1885 (Madras, 1886), pp. 16-17; G.O. 2298 (Judicial) dated 10 October 1887.
T.N.A.
20
Report on the Administration of Civil Justice in the Presidency of Madras for the
year 1884 (Madras, 1885), p. 23; Nicholson, RAB, 1, 312.
21
Concern at this state of affairs was expressed regularly in the native press; for
example: 'Apparently, criminal justice, in the view of the Madras Government,
is quite safe in the hands of the village headman, who is tax-gatherer, police
officer, Criminal and Civil Judge all in one.' Hindu 3 May 1915.
22
See Report of the Forest Committee. 1913 (Madras, 1913), 11, passim.
23
Appendix to the Report of the Indian Famine Commission, 1898, being Minutes of
Evidence, etc., Volume II, Madras Presidency, pp. 33,165,169,P.P. 1899, vol.xxxn;
Report on the Famine in the Madras Presidency during 1896 and 189/ (Madras,
1898), 11, 2 0 3 .
24
G.O. 1337 (L.S.G.) dated 13 July 1921. T.N.A.

152
 Local structures of political power

expenditure, the government steadily raised the competence of the

village headman's civil court.25 In order to cheapen rural credit,
Acts in 1904 and 1912 legalised a variety of co-operative credit
associations. Although there was nothing in the legislation to com-
mand it, those village co-operatives which appeared in the dry
districts before 1920 were naturally dominated by the village econo-
mic elite which put up most of the capital. And out of this elite, it was
most usual for the village officers to run the co-operative as part of
their general duties.26
British policy, then, concentrated administrative powers in the
hands of the headman.27 It sought to break down the networks of
social and cultural connections between members of local and state
administrative groups, which it feared it could not control, and
replace them with connections of a purely institutional kind between
the village and the desks of senior British officials. The policy certainly
was of advantage to the headmen and their interests. It enabled them
to exercise an official influence untrammelled by the presence in their
locality of groups who could use external force. Indeed, it was they
themselves who manipulated such external force as was available.
But Fort St George soon discovered that its own share of advantage
was extremely small. The same spirit of reform which was driving
the British to expand their administration and heap powers on the
headman also brought with it a demand for responsiveness from the
village. In the first half of the nineteenth century, it had not mattered
much how village powers behaved, or who they were, so long as a
steady revenue flowed and some order was kept. But in the second
half, a much more active village collaboration was required. Yet the
character of the personnel and the mechanics of collaboration
guaranteed that this would not be forthcoming. All orders going
into the village and all information coming out of it had to pass
25
In 1881, village headmen heard 47,656 civil cases; in 1910, they heard
96,597 cases; and between 1913 and 1918, with the help of panchayats, they heard
an annual average of 126,959. Report on the Administration of Civil Justice in
the Presidency of Madras for the year 1881 (Madras, 1882), p. 3 1 ; ibid., 1910, p. 4;
ibid., 1920, p. 3. It ought to be noted, however, that much of this increase may
not have been 'new' litigation but could have appeared on the record simply
because headmen were being forced to register a higher proportion of the
litigation which they handled.
26
Report of the Forest Committee. 1913 (Madras, 1913), 11, 462; Royal Commission
on Agriculture in India, m, 549, 647.
27
\ .. all influence [in village society] is sought to be exercised through the
village headman.' Madras Police 1885, p. 4; the headman 'becomes daily of
greater importance'. Nicholson, RAB, 1, 312.

153
 The emergence of provincial politics

through the hands of the headman's establishment. Consequently,

the Secretariat, and even the Collector, were able to know no more
and do no more than the establishment would allow. Of course, it
might be supposed that the senior authorities had power at least
to sack obstreperous headmen and to change the personnel of the
village administration. But even this was doubtful. The 1802 Act
had established the hereditary rights of headmen in law and any
attempt to remove a headman took the government along the extra-
ordinary and exhausting paths of Anglo-Indian property law. In the
opinion of J. H. Garstin, a member of the Board of Revenue in 1883,
the only reason for which the government could dismiss a headman —
short of a successful criminal prosecution - was for being female,
and even then, the headmanship would pass to the next senior member
of her family.28
From the 1870s, a series of governmental enactments began to
press on the village administration. Resurveying and resettlement
operations gave the Secretariat far more knowledge of the presidency-
and of the actual revenue administration of the presidency — than it
had possessed before. By 1920, most districts had been resettled twice
and some three times, so that the Board of Revenue at least knew
how many villages it governed. Following J. H. Garstin's 1883 report
on the revenue administration and Curzon's 1902 Police Commission,
more revenue inspectors and police deputy inspectors were created
to provide closer supervision of village headmen. Village establish-
ments were moved progressively from payment by inam or rent-free
land to payment from a local land cess and in 1906 to payment as
state stipendiaries. New rules demanded that village officers should
be literate, and reforms cut both the number of minor posts at the
disposal of the headman, which had served to swell his private army,
and his power to issue notices in distraint of property. Finally, in
1918, the government introduced a bill to remove the hereditary
right of village officials to their posts.
Under this bureaucratic onslaught, it might be supposed that the
village authorities were better controlled by the higher administra-
tion. Yet, in fact, they survived into the twentieth century with their
independence scarcely impaired. The only reformist policy which
28
J. H. Garstin to Secretary, Revenue Department, 3 April 1884 in G.O. 787
(Revenue) dated 24 June 1884. T.N.A. Garstin wrote a detailed report on the
organisation of the Revenue Department and recommended that much greater
powers of interference and control were necessary if senior officials were to
curtail village officer independence. J. H. Garstin, Report on the Revision of
Revenue Establishments in the Madras Presidency (Madras, 1883).

154
 Local structures of political power

met with any measure of success was one of very minor consequence:
the abolition of many minor posts and the curtailment of the right
to issue demand and restraint notices at will. This proved an irritant
but, of course, did nothing to check the abuse of village officer
authority.29 The attempt to turn the headman and his assistants into
state stipendiaries proved of little use while the government could
not withhold the stipend for misbehaviour without becoming involved
in a series of delaying law suits.30 It seems likely that its main effect
was to put more money at the disposal of the senior village officials
than before. Most had managed to manipulate their records so that the
amount of rent-free land actually surrendered to government was
worth less than the new salaries paid.31 The new rules, issued in
1894, to make it compulsory for village officers to be literate and
to have some competence in their tasks were greatly weakened by a
clause which allowed all present officials and their probable heirs
who registered within two years to avoid the penalties for failure.32
This put off for a generation any effect the rules might have had. At
best, the reforms and resettlements slightly cheapened the cost of
administration and brought to Fort St George's attention more land
on which revenue could be paid. But they did not alter the mechanisms
by which village officials controlled intra-village payments nor did
they even manage to maintain per acre land revenue demand in the
face of inflation.
The reasons for Fort St George's failure to mount an effective
assault on village officer power are not far to seek. It had, of course,
picked up precisely the most difficult people to draw into its central
bureaucratic system. The level of active collaboration which it could
expect in this political situation varied in direct proportion to the
degree to which its collaborators depended for their local positions on
the resources which it provided. When that proportion was high, its
collaborators would obey its commands explicitly; when it was low,
they would obey as few commands as possible. Having set up this
29
For the decline in the number of notes in restraint of property, see 'Prosecutions
for Arrears' in Land Revenue 190j-8 and 1919—20.
30
See, for example, 'Report o n Cuddapah'., p . 2 1 , in Land Revenue 1875-6.
31
F o r example, in 1870, t h e village officers of Trichinopoly s u r r e n d e r e d 16,304.37
acres of inam land a n d were obliged t o pay a n assessment of about Rs 10,000 p.a.
on it. In r e t u r n , by 1875, t h e cesses collected by government a n d redistributed
to t h e m increased from Rs 652 p.a. to Rs 1, 7 2 , 340. 'Report on Trichinopoly',
p. 10 in Land Revenue 1870—1; 'Report on Trichinopoly', p . 68 in Land Revenue
1875-6. See also, Proceedings of the Board of R e v e n u e , N o . 1451 (Misc.) dated
18 M a y 1885. T.N.A.
32
G.O. 361 (Education) dated 24 M a y 1894. T.N.A.

155
 The emergence of provincial politics

system, Fort St George found that not only was it unable to offer
many positive resources but that it was giving what it had to the local
men who needed them least.
Most village headmen came from families who possessed large
quantities of land and who directed great economic influence in their
localities, quite apart from any powers which they received as village
officials. They were part, often the major part, of the rural-local boss
elite.33 While, due to the nature of the land revenue statistics, it is
impossible to provide overall figures of their riches, that they did
possess great wealth was a fact assumed, and seldom challenged,
in the political lore of Madras. In a 1920 Legislative Council debate
on their stipends, for example, it was stated and never contradicted
that village officers did not really need to be paid, for their salaries
formed but a small portion of their total incomes.34 Equally, in con-
sidering the probable effects of the Montagu—Chelmsford franchise,
Sir Charles Todhunter thought that, given their wealth and local
prestige, 'the village officers are likely to have so much influence
over the electorate that it would be easy for them to secure a mandate
to elected members'.35 The few cases in which the actual wealth of
headmen reached the light of day confirm this impression. In Bellary
in 1865, for example, village officers held no less than 635,000 acres
of land on privileged inam tenure in addition to their ordinary ryotwari
holdings.36 As headmen were top of the village service list, it was they
who possessed most of these inams. In Cuddapah in 1875, the 168
village officers mentioned in a resettlement operation admitted to
paying Rs 22,507 p.a. in land revenue between them.37 In Kistna
district, the Kasu family of Reddis, who were the headmen of a village
in the upland taluks, were said to be so wealthy that they were one
of the only two families in their locality who could afford to eat rice.38
33
Given the extent to which political power and social status influenced marriage
patterns, most of the wealthy families in the dominant peasant caste of any region
would possess marital links with the headman's family in their neighbourhood.
Looked at in terms of extended kinship organisation the headman's family
and the families of other rural-local bosses in the same locality were likely to
be synonymous.
34
Home Judicial, File 931 of 1922. N.A.I.
35
Note signed C. T o d h u n t e r dated 11 December 1919 in G.O. 1958 (Revenue)
dated 14 August 1920. T.N.A.
36
M.D.G. Francis, Bellary, 1, 175.
37
'Report on Cuddapah', p. 10 in Land Revenue I S J $ — 6 .
C. H. Benson noted how village officers, holding inams, also held a large amount
of their villages' revenue-paying lands. Benson, Account of the Kurnool District,
p. 112.
38
See Kasipathi, Tryst with Destiny.

156
 Local structures of political power

In Rajapalaiyam, Ramnad district, the headman family of A. K. D.

Dharma Raja owned great lands, a market and, later, even a cotton
factory.39 In Vellakinar, Coimbatore district, the Vellakinar family,
headmen of their village, paid over Rs 2000 p.a. in land revenue. 40
Compared to the enormous concentration of local political in-
fluence already in the hand of the headman, the positive resources
offered to him by the British were of small account. Chronic shor-
tages of funds kept the amounts of cash available in takavi loans or
famine doles so low that, while a useful addition to any patronage
chest, they were never sufficient to finance a political empire. 41 The
other powers given to the headman also were little more than toppings
to his local jar. The provincial police could help him only if a case
broke out of his village but did not have the manpower to help him
if it stayed inside. The superior law-courts could support his magi-
sterial role only in the case of a major crime or dispute but could not
touch the mass of petty litigation which he handled from day-to-day
and which was of much greater concern to village society. The revenue
department could rescue him from situations of large-scale revolt but
could not provide the machinery for getting revenue out of every
reluctant payer. It was indeed to avoid becoming involved in these
vital but petty affairs that the British employed the headman at
all. He had to rely on his locally derived powers as a premier rural
boss to give him most of his local authority. In the rare cases (outside
the Andhra deltas) where his official standing was not congruent with
the realities of local power, almost inevitably he succumbed to the
greater weight of the local bosses around him and became their tool.
In parts of Tinnevelly, for example, where warrior Maravar lineages
continued to dominate rural society by force of arms and so to limit
the political influence of agricultural wealth, peasant headmen, who
had to rely on the power of their offices alone to give them local con-
trol, seldom were more than agents of the warriors. 42
39
See V e n k a t a Raja, Brief Life Sketch of P. S. Kumaraswami Raja.
40
See biography of V. C. Vellingiri Gounder in Reforms (Franchise) B. March
1921, Nos 34-99. N.A.I.
41
Takavi loans, for example, seldom reached even Rs 10 lakhs p.a.
42
Madras Police 1897, p. 146; ibid., 1899, p . 147. It should be noted that, because
of its reliance in de facto local power, t h e headmanship was effectively a floating
office, moving t o whomever h a d power, and only nominally an hereditary office.
Consequently, o u r discussion of it should be seen in institutional terms (the
headmanship) rather t h a n personal terms (the h e a d m a n himself). Because of its
floating character, a n u m b e r of kurnam families were able to continue to
wield local influence even after their o w n office had been eclipsed in importance.
Where, in previous generations, they h a d used their state connections t o build

157
 The emergence of provincial politics

The lack of positive resources from outside — including that of

kinship connection once the British had weakened the kurnam's
role - and the heavy reliance on local resources of power stressed
the features of negative political value in the headman's office. The
great advantage to a rural-local boss of being the headman was that
he could use his economic influence to full effect without worrying
about the interference of external authorities. He could seize lands,
murder rivals and hold down his neighbours, while at the same time
preventing any British law, order or inquiry from disturbing his
control. He could maintain the locality which he dominated in
splendid isolation. Given that the headmanship was commonly inter-
preted by its holders in this light, it is not surprising that the de-
mands of the British for a greater co-operation and responsiveness
from their village administrators met with no noticeable success and
provoked some resistance. Rural society had come to see the benefit
of the British connection in the independence which it gave to local
authorities. It was hardly eager to exchange that independence for
a highly circumscribed position in a tightly linked bureaucratic
chain.
The intractability of the local situation can be seen to account for
the extreme caution with which Fort St George pushed through its
attempts at village officer reform. Having opted for the headman, it
could not afford to alienate him, because it did not have the money to
build a new administrative system which would have excluded him
and because, if his vast local power were turned against it, its position
would have become untenable. Calcutta, although sympathetic to
complaints from Madras about village officer indiscipline, could
spare very little cash to help alter the situation, and Fort St George
itself, save during the interlude of Lord Curzon's viceroyalty, was
always bankrupt.43 Any new administrative system had to cost no
more than the old one, which itself was virtually self-financing. More-
over, when provoked from above by administrative reform, the powers
of local society could place in the field an army of such size that the
foundations of the raj were shaken. Not only could they deny the
British the means of exercising any rural control but they could create
the basis of serious and widespread opposition to the continuation of

up private landholdings, they could become dominant local powers even after
those connections had been severed. However, and obviously, their continued
significance did not now rest on a specifically kurnam-type of influence but only
on de facto local power obtained from control of land and the economy.
43
G.O. 369 (Revenue) dated 25 March 1885. T.N.A.; P& J File 251 of 1888./.O.L.;
Madras Police 1897,$. 5.

158
 Local structures of political power

British government. In the face of strong pressure from the village

authorities, Fort St George had to bow its head.
We can see precisely how wary the British were of undermining the
foundations of their administration from the cases of Curzon's police
commission and the Village Officers' Hereditary Rights Bill of 1918.
On the advice of the commission, the Madras government created a
new cadre of police deputy inspectors, 2000 strong, which was
centrally trained and appointed to act between the village and the
existing police establishment.44 This doubled the cost of police ad-
ministration and posed a serious threat to village autonomy. As a
policy it proved a disaster: open war was declared between village
headmen and the police deputy inspectors, and the detection rate for
crime grew worse.45 The anger of the village officers echoed along
channels of administrative and political connection — which we shall
be examining later - to be heard in Madras city. By 1915, the govern-
ment was forced to compromise. Although keeping its central school
for the deputy inspectors, it switched the control of appointments
from its own Secretariat to committees of notables in the localities. 46
As these notables included a great many village officers, it transferred
power over the deputy inspectors to the very people whom they were
supposed to supervise. Not surprisingly, the hostility between the
village and the deputy inspector evaporated. Equally, the 1918 bill
to abrogate hereditary rights provoked a province-wide agitation
which fed into the already serious propaganda movement of the Home
Rule League and the Congress. Village Officer's Associations appeared
in every district to press for the withdrawal of the bill. 47 Once again,
Fort St George found discretion to be the better part of valour and
shelved its proposals.48 It could not stand up to the rage of its most
important collaborators.

Zamindari villages
Although zamindari tenants were economically very similar to
ryotwari landholders, administrative and legal factors made them
less politically independent. The zamindar, like the government,
was a rent collector but, during our period, he possessed far greater
44
Madras Police 1919. Appendix D , p . x.
45
Madras Police 1907, pp. 5 - 6 ; ibid., 1912, pp. 9, 3 3 ; ibid., 1914, pp. 7 2 - 6 ; ibid.,
75>Z5,pp. 1 7 - 1 8 .
46
Ibid., 1915, p . 18.
47
Hindu 12 November 1920. Andhrapatrika 4 F e b r u a r y 1919 a n d 7 September
1920; Desabhimani 8 January 1919; Gramapulana 10 September 1921. R.N.P.
48
G.O. 1958 (Revenue) dated 14 August 1920. T.N.A.; H o m e Judicial File 931 of
1922. N.A.I.
159
 The emergence of provincial politics

legal powers than the British allowed themselves. Following an 1870

High Court decision, a zamindar was permitted to raise his rents
at will whereas the government was tied to a policy of resettlement
only every thirty years. The 1908 Estates Lands Act somewhat cur-
tailed the possible despotism of the zamindar but also gave him a
much clearer title to the lands of his tenants than ever before. Zamin-
dari rents were said to be consistently higher than government
revenue demands, but information is lacking on exactly how much
higher.49 Moreover, the zamindar was not restricted, as was govern-
ment, in the means he could employ to collect his rents. In parts of
Ganjam and the Northern Circars, it was common for an estate to
be auctioned off in lots every year to mastajas, who paid a fixed sum
for the privilege of farming the rent. 50 Similarly, in the south, many
estates were leased to tax farmers who rack-rented the tenants.51
Further, the 1802 Act which gave hereditary rights to village head-
men did not apply to zamindari areas. Much of the village establish-
ment held office only at the zamindar's pleasure and was subject to
his will.52
Yet we could go too far in emphasising the weakness and subjection
of tenants in the zamindari. The processes of the economy supported
and favoured the development of rural-local elite groups and, for all
their powers on paper, the zamindars were not entirely kings in their
own territories. There was a tendency, particularly under the influence
of the Court of Wards, for zamindars to undertake direct rent collec-
tion from their tenants through bureaucracies. The actual manage-
ment within these looks remarkably like the management of the
ryotwari system. The same power derived from control of the records
lay with the village authorities, and the same possibilities for collu-
sion and corruption existed in the supra-village bureaucracy. In the
Ramnad and Sivaganga estates, T h e zamindars' dishonest sub-
ordinates allowed the tenants to have their own way in the village5
and lakhs of rupees disappeared every year.53 When the Pithapuram
estate in Godavari passed under the Court of Wards in 1892, it was
found that the previous managers had allowed Rs 10 lakhs of arrears
to build up with various tenants.54 In the Ettiyapuram estate in
49
RPBC, pp. 7-8.
50
Land Revenue 1905—6, p . 6 7 ; ibid., 1906-7, p . 14.
51
Land Revenue 1903—4, pp. 78-9.
52
Madras Police 1882, p . 2 5 .
53
Land Revenue 1903—4, pp. 7 8 - 9 .
54
Report on the Administration of the Estates under the Court of Wards in the Madras
Presidency for Fasli 1302 (1892-3) (Madras, 1894), P- 6 -

160
 Local structures of political power

Tinnevelly, the Court found that Rs 7,16,992 of arrears had been

allowed to outstand for so long that action against the defaulters
was time-barred.55 Of course, more than in the ryotwari areas, this
collusion could have benefited kurnam-type groups, whose relation-
ship to higher bureaucratic offices was not restricted by British fears
of conspiracy. Zamindars were under no pressure to raise the office
of headman in their villages or to aim their offer of collaboration
at rural-local bosses. Yet, in order to avoid alienating the de facto
local powers, private administrators had to make some bargains with
the bosses. Moreover, the British began to intrude into the interior
of zamindari estates in order to improve administrative performance.
In 1894, a n Act w a s passed to curtail complete zamindari control of
village establishments and to place certain aspects of their work
under the supervision of government departments. 56 This change
introduced the principles of headman-ro/ into zamindaris and hence
promoted the power of the rural-local elite.
Even in those zamindaris which steadfastly refused the advice
of the Court of Wards to bureaucratise and continued to auction
their revenues, changing management techniques began to favour
wealthy tenant groups. In order to raise revenue, many zamindars
auctioned smaller and smaller units of revenue and, particularly
in parts of the Northern Circars, it was noticeable that by the
end of the nineteenth century many of the buyers were rich
rywr/village officers, tendering for the revenue of their own villages. 57
A successful bid, of course, would give the purchaser control of the
village's revenue in a more direct manner than ever was open to his
ryotwari colleague and make him even more powerful in relation to
his fellow villagers.
Although the tenancy law in Madras favoured the landlord more
or less throughout our period, the very fact that a legal system was
replacing - or rather, being written round - a, series of relationships
often previously based on force gave certain advantages to the tenant.
Against raids from the zamindar's private army, at which many
government officials had connived during the first half of the nine-
teenth century, the villager was often powerless.58 But with rights,
however weak, in a confused legal system and with a guarantee of at
least three years' delay in any civil action, the tenant possessed arms
55
Ibid.
56
T h e Proprietary Estates Village Services Act of 1894.
57
A. V. Raman Rao, Economic Development of Andhra Pradesh (1766-1957),
(Bombay, 1958), p. 193.
58
Frykenberg, Guntur District 1788-1848, pp. 4 8 - 9 .
 The emergence of provincial politics

with which to defend himself. In 1905, even before the Estates Lands
Act, the Collector of Chingleput complained: The summary suits in
this district are the most troublesome part of Revenue work in this
division. Every inch of the ground is fought tooth and nail between
both parties and the tenants seem to be able to hold their own against
the rapacity of the landlords.'59 A standard ploy in battles for succes-
sion to zamindaris or for transfers after bankruptcy was for the
losing party to ally with the tenantry in opposing the new regime.60
The devastating effects this could have indicates that the tenants were
far from powerless.61 Again, the rural-local elite was at least as
successful in controlling the sale of defaulting tenants' property under
zamindari as under ryotwari tenure.62
Legislative interference at the beginning of the twentieth century
more positively helped the tenant to some independence. The Estates
Lands Act of 1908 created a framework for the recognition of
occupancy tenure and a procedure by which tenants could force
their landlord to convert grain into money rents - an important con-
sideration at a time of rising grain prices.63 Although the powers of the
Act were minimal, they allowed the tenant yet a further means of
litigation to disrupt the grasping landlord's administration and to
freeze his assets for further periods of time. The Madras banker Lodd
Govindoss found that his attempts to raise rents in the portion of
the Kalahasti zamindari which he had bought were met with 8,000
suits from his tenants, which took several years to clear;64 and the
Nuzvid estates in Kistna district, long noted for their turbulence,
were pushed towards breakdown as tenants ceased to pay rent for
years at a time.65 Of course, the full advantages of the Act went only
to the wealthier section of tenants who had money to play legal dice
and who were likely to control village records in order to prove
occupancy. The strengthening of their position in relation to the
59
Land Revenue 1905-6, p. 78.
60
For example, in the Ayakudi zamindari, Land Revenue 1913-14, p. 12; or in
the Biridi zamindari, Land Revenue 1908-9, p. 14.
61
For example, the Raja o f Karvetnagar achieved so tight an alliance that his
tenants succeeded in destroying or distorting all estate records. Report on the
Administration of the Estates under the Count of Wards in the Madras Presidency
for Fasli 1317 (1907—8) (Madras, 1909), p. 10.
62
In the decade 1 9 0 7 - 1 7 , the sale o f defaulting tenants' lands realised only
22 per cent o f their estimated value. G.O. 1167 (Revenue) dated 14 April 1917.
T.N.A.
63
Land Revenue 1908—9, p. 14; ibid., 1913—14, p. 12; ibid., 1916-17, p. 11.
64
Land Revenue 1913—14, p. 12; Hindu 23 November 1913.
65
Land Revenue 1910-11, p. 59; ibid., 1913-14, p. 12; G.O. 3076 (Revenue) dated
22 October 1913. T.N.A.

162
 Local structures of political power

zamindar both through their protection as village officers and through

their hold on the forces of tenurial law, once again stimulated social
stratification and increased the distance between them and the
smaller zamindari tenants.

The growth of rural politics. In the dry areas of Madras, under both
zamindari and ryotwari tenures, much the same social and political
processes were taking place. Economic, administrative and legal
changes were producing - or rather heightening - the powers of
rural-local elite groups. The term 'rural-local' is very imprecise and
we have used it thus far only because of the difficulty of defining
anything more suitable. Here, however, we may attempt to be less
obscure, for the rise to supremacy of the rural locality had a crucial
impact on the spatial dimensions of political power in South India.
With the demilitarisation of the warriors, the restrictions on the
use of government troops, the smashing of local-state social connec-
tions among administrators and the combination of powers in the
office of headman, coercive force to back up political power was
available only from inside the rural locality. In consequence, this
locality became virtually the only arena in which important political
decisions were made. How large and how wide this arena would be,
then, would be determined by the size and width of the instruments of
control within it. During most of our period, these instruments were
effective only over very short distances. Economic clientage may not
have been confined to the physical village but it would seldom extend
beyond a circle of adjacent villages which exchanged their products
through a common bazaar. Revenue control could be exercised only
through the 'revenue' village which was usually no more than one or
two hamlets with a population of one thousand or so. Police control
could be stretched a little further, to areas of settlement which fell
under the practical suzerainty of a headman's terror machine. But
these had to be easily accessible from his headquarters. Members
of the next layer of government officials above 'the village' might
be enlisted to help or to turn a blind eye to a rural-local boss' activ-
ities. But their jurisdiction, and their ability to conceal without aid
from above, was limited to the revenue^r^a or police circle. Struggles
for rural control and pre-eminence, therefore, could take place over
localities of no more than a few square miles.
We are provided with an excellent opportunity of looking at one
of these struggles by a battle which developed in the Gooty area of
Anantapur district between 1904 and the mid-1920s. The two
factions concerned were led by two wealthy Reddis from village officer

163
 The emergence of provincial politics

families - Chinnarappa Reddi and Thimma Reddi - and their forays

against each other were estimated to have cost about two murders
a month and to have filled the local courts with dozens of cases, most
of them originating in false charges by one side against the other.
Chinnarappa Reddi proved the more successful of the two, for not
only was he able to hold his own in gang battles involving hundreds
of dependents but also he was able to block the higher authorities.
He managed six times to prevent the stationing of punitive police on
the area while his faction was winning, and to harass Thimma Reddi
with police prosecutions when his own gang was in difficulties. At the
height of his power, he was regarded as 'the sole monarch' of forty
villages. He took what land he pleased by burning the crops standing
on it and sending in his retainers to hold it until the previous owner
gave way. His enemies disappeared with a remarkable regularity. He
drew a private revenue from his domains and arbitrated in the
disputes of his subjects.66
Chinnarappa's forty villages were a large territory for the kind
of immediate resources he used but they still represented less than a
single revenue firka. In his use of superior officials he was far more
concerned to shut out the influence of the wider world on his little
empire than to take his battle with Thimma into a larger arena. His
connections with the police and courts enabled him to isolate his
locality and do what he pleased in it. British administrative practice
had broken whatever greater unities there had been in political
activity under native regimes. It left only fragments behind it.
The new demands of British policy, however, guaranteed that these
fragments could not remain in isolation. Rural administration had
to do more and some way had to be found to make it do more. Clearly,
a simpleminded attempt to smash the power of rural-local bosses
and turn them into officers of the central bureaucracy would not
work. Fort St George found that its new abkari, forest and irrigation
programmes were going the same way as its attacks on village officer
power and producing more opposition than it cared to handle. Between
1913 and 1915, the excise department was faced with a rapid rise in
illicit distilling and sales which even its savage prosecutions could do
nothing to halt; 67 forest administration was degenerating into open
and bloody warfare between forest department subordinates and
rural-local bosses;68 the corruption of irrigation officers became the
subject of a major agitation in the delta tracts, and the government's
66
The case was fully reported in Hindu 20, 22, 29 and 30 June and 16 July 1925.
67
See Abkari 1913—14 t o 1917—18.
68
See Forests 1911—12 to 1916—ij.

164
 Local structures of political power

new Irrigation Bill of 1915 was greeted with such a storm of abuse
that it had to be shelved until after the war.69 The old methods of
direct and despotic rule, with which the British tried to galvanise
the presidency, were unable to produce the administrative results
they now required.
In consequence, the Madras Government came to apply to a wide
range of matters the solution found to the problem of Curzon's police
deputy inspectors. Committees of local notables were formed to advise
or administer the distribution of water for irrigation, the siting and
number of liquor shops, the conservation of forests, the assessment of
the income-tax and local self-government. These new institutions had
a profound impact on rural political life. They clearly exercised
considerable power: excise committees, as in Salem in 1916, could
persuade Collectors and other authorities to close down all the
arrack shops in their vicinity;70 forest committees could allocate
or withhold grazing rights from individuals; irrigation committees
could control the flow of water along any channel.
They also changed the physical dimensions of the rural locality
and introduced new types of political activity. The rural-local boss
could no longer rely on his immediate power network and on his
ability to keep external forces away in order to preserve his local
position. In order to gain even the same amount of influence, he had
to capture administrative machines which could cover many square
miles and touch the affairs of a large number of other men of his
type. Much faster and more purposefully than the expansion of trade,
the extension of the administrative locality brought members of the
rural elite into contact with each other and forced them into alliances
and battles in arenas which outlined broad areas. To take a concrete
example, when the forest department ran jungle conservancy on its
own, conflict over forest resources was very restricted in its scope.
Rural-local bosses, with the aid of the headman's power, organised the
allocation of forest rights in their own interests and within the
localities of their dominance. Forest department subordinates were
detached from their administrations and brought to coalesce in these
arrangements by a mixture of bribery and threat. Any battle for
forest rights between bosses could take place only in areas in which
they shared immediate interests and wanted the same bundle of
leaves.71 Once forest committees were created, however, the situation
69
H o m e Public B November 1916, No. 56. N.A.I.; Hindu 1 1 , 1 4 a n d 2 3 N o v e m b e r
1914.
70
Hindu 14 February, 5 and 18 March 1918; see also Hindu 19 and 31 July 1918;
Abkari 1921-2, p. 9.
71
Report of the Forest Committee. 1913 (Madras, 1913), 1, 7.

165
 The emergence of provincial politics

was changed. An institutional machine had been built which could

force the distribution of forest rights over a wide terrain. Rural-local
bosses from the entire area of the committee were brought together to
fight for the allocations. New linkages and oppositions were formed
between men who may never have had political contact before.
The institutions not only widened political activity, they also
carried it to new levels. Most of the boards and committees were
formed under rules drafted in the Secretariat, which often inter-
fered in their operations and supervised their administration. As, at
this time, the Secretariat was itself becoming more open to influence
exercised by Indians and by forces below it in the bureaucratic
hierarchy, rural-local bosses — becoming now local politicians — were
provided with both the means and the incentive to carry their struggles
into the upper reaches of provincial government.72
Singly, the most important of all the new institutions of govern-
ment created by Fort St George were the rural boards, which linked
together not only rural localities but also the affairs of towns and
a host of other, previously separate, interests in religion, education
and public welfare. We can best illustrate the nature and the force of
political development by examining their history in detail. From the
second quarter of the nineteenth century, the Madras Government
had been extending local taxation to pay for roads and sanitation.
Until 1884, t n e institution used most for these purposes in the
countryside was the Local Fund Board, a loose agglomerate of govern-
ment servants, local landlords and merchants. Under Ripon's local
self-government reforms, a much tighter and more formalised organ-
isational structure was set up. At the bottom of the pyramid, village
unions were created to look after the administration of basic amenities
in the village; above them sat taluk boards which supervised their
affairs and possessed direct responsibility for supra-village services
such as roads and schools; above the taluk board sat the district board
which directed the overall policy of development in the district.
Under Ripon's scheme, provision was made for a continuing de-
centralisation of administration and the gradual transfer of power
from officials to non-officials by means of elections and the placing
of executive posts in the hands of non-officials.73 Taxation was
72
F o r example, P. Kesava Pillai, the lawyer o f the Chinnarappa Reddi faction
in Gooty, Anantapur district, became a Legislative Councillor in 1909 and
pounded away at the Madras Government o n the issue o f the forest laws. H e
was largely responsible for persuading the government to set u p the Forest
Committee (1913) and became k n o w n in Madras city as 'the Member for Forests
and Jails'.
73
Madras District Boards Act o f 1884.

166
 Local structures of political power

increased steadily throughout the period: in 1880-1, prior to the

Ripon reforms, Local Fund Boards handled an average budget of Rs
4 lakhs each; by 1893-4 t n ^ s n a d reached Rs 5 lakhs and by 1909-10,
Rs 7 lakhs.74 Administrative powers were also increased to include
the supervision of institutions of higher education, of markets, wells
and buildings and the implementation of major construction pro-
grammes such as railways75 and jungle clearance. The rural boards of
Madras were richer and more influential than their counterparts in
other provinces: their powers and patronage were of concern to all
local empire builders.
Yet, in spite of their considerable wealth and responsibilities, rural
boards made little impact on the structure of Madras politics before
the 1910s. The Madras Government, though extremely eager to
implement Ripon's ideas on financial decentralisation, was slow to
introduce non-officials into positions of executive authority.76 Rural
boards tended to be run by government officeis as part of their general
duties. A. Subbarayalu Reddiar, who became the first non-official
president of the Cuddalore taluk board in South Arcot in 1912,
complained of previous administrative practice:
As matters have stood, with exception of Dispensaries, Schools and Taluk
Board Roads, almost the whole of the outdoor work was managed by the
Revenue Divisional Agency. The village sanitation, the maintenance and
opening of the village roads, the repair and construction of the drinking
water wells, and ponds, the clearance of encroachments, the removal of the
prickly pear, etc., were all in the hands of the Revenue Department.77
Of the three aspects of administration not managed by the Revenue
Department, dispensaries were supervised by the Medical Depart-
ment, schools by the Education Department and taluk roads by the
Public Works Department. Although the Madras Government allowed
some elections to the district and taluk boards, these were dominated
by government officials seeking places on the boards of higher juris-
74
Report of the Committee on Local Self-Government in Madras. 1882 (Madras, 1883),
p. 146; G.O. 2541 (L and M , L) dated 6 December 1894; G.O. 1702 (L and M , L)
dated 12 December 1910. T.N.A.
75
Madras district boards were permitted to raise loans o n the open market to
finance the building o f railway branch lines. See, for example, Hindu 10 and 17
June 1896.
76
Report of the Committee on Local Self-Government in Madras. 1882 (Madras, 1883),
p. 4.
77
A Subbarayalu Reddiar to P. S. Sivaswami Iyer, 12 April 1912. P. S. Sivaswami
Iyer Papers. N.A.I.

167
 The emergence of provincial politics

diction. They were virtually intra-department elections.78 While effec-

tive executive authority lay with government officials the boards
could not develop as mature political institutions. The character of
politics was conditioned by the petty rural locality. 'Bosses' used
their private networks to control local affairs and went to particular
bureaucrats for a rubber stamp on the deals.79 There was no need for
the district board to become an instrument of district politics when
the officials who possessed the power on them could be detached
and used to protect the rural locality. The famous Tamil novelist
Madhavaiah has faithfully recorded the politics, or lack of them,
which went on in the boards at this time:
In these assemblies, I first discovered what a shameful farce local self-
government was. Not a few of my fellow members were almost illiterate, and
altogether ignorant of the English tongue, in which our deliberations were
conducted. They were wealthy and so they were elected. They came more
for the travelling allowances they obtained for attending meetings than
for the subjects discussed at those meetings, unless they happened to
hold a secret brief from any contractor to get an extravagant bill
passed.80
The institutional importance of the boards dates only from the last
decade or so of our period. Under pressure from the Government of
India and the Decentralization Commission, which commented on its
poor performance in local self-government, the Madras Government
was forced to take further measures of financial and administrative
decentralisation.81 The taxation gathered by the rural boards increased
much faster than before, and rose by 70 per cent, to a district average
of Rs 11 lakhs, between 1909—10 and 1919—20.82 Government officials
stepped back to allow non-officials a larger share of the executive: by
1919-20 51 of the 98 taluk boards and 11 of the 25 district boards had
non-official presidents.83 Allied departments, particularly the Educa-
78
I n 1915, tahsildars held six of t h e elected district board seats in South Arcot,
six in K u r n o o l , five in Salem, five in Kistna, four in Bellary, four in Godavari
and others elsewhere. Memorandum 31-41, dated 5 February 1915m Confiden-
tial Proceedings of the Madras Government, 1916, vol. 23. I.O.L. In all,
officials (excluding village officers) held 102 of the 354 places open to election.
Hindu 9 February 1915.
79
See, for example, Vijayadwhaja 26 July 1888; Jananukulam 17 J u n e 1899.
R.N.P.
80
A. Madhavaiah, Thillai Govindan (London, 1916), p . 118.
81
G.O. 6 1 6 - 7 (L a n d M , L) dated 27 April 1908; G.O. 916 (L a n d M , L) dated
12 July 1911. T.N.A.
82
G.O. 1337 (L.S.G.) dated 11 July 1921. T.N.A.
83
Ibid.

168
 Local structures of political power

tion Department, were stripped of their powers of intervention.84

Quite suddenly, rural-local bosses found themselves provided with
a machine of tremendous power, which they could use to develop
their support and crush their enemies: through control of taxation,
contracts and services in the district they were given the means of
extending their empires.85 Few were slow to seize them. In South Arcot,
A. Subbarayalu Reddiar became district board president in 1917
and quickly replaced revenue department servants in the board
bureaucracy with local board servants appointed at his discretion
and subject to his whim;86 he demanded the right to transfer taluk
board staff regardless of the wishes of taluk board presidents, some-
thing no Collector had attempted;87 he used his powers of nomination
to put friends onto the board and to remove enemies, such as the
Muslim merchant M. Razak Maracair with whom he had a long-
standing feud.88 In Bellary in 1919, the appointment of H. Latchmana
Rao to the district board presidency brought groans from many local
inhabitants. As soon as he took office he winkled out two taluk board
presidents and appointed his own men in their stead. A petitioner
complained: 'All the three are natives of the District with a host of
relations and connections in the District. It is the duty of the
Government to see that the administration does not degenerate into
a Brahmin oligarchy.'89 The Secretariat itself fully recognised the
implications of a district board presidency: in arguing for the speedy
appointment of more non-officials before the first Montagu-Chelms-
ford elections, Charles Todhunter wrote: 'Finally, I would observe
that we must not lose sight of the coming elections. An appointment
as President of a local body will undoubtedly be a valuable asset in
these.'90 Several district board presidents were reported in the press
to be using the patronage of their institutions in their Legislative
Council campaigns91 and, of the ten who stood for election in 1920,
none failed to be returned.
84
G.O. 1168 (L and M, L) dated 18 August 1915. T.N.A.
85
'Landlords with local influence discovered that, as presidents and members of
local boards, they could wield a large amount of influence in their locality and
exercise greater power over their neighbours.' M. Venkatarangaiya, The
Development of Local Boards in the Madras Presidency (Bombay, 1939), pp. 66—7.
86
Hindu 5 F e b r u a r y 1918.
87
Ibid.; G.O. 1021 (L a n d M , L) dated 8 August 1918. T.N.A.
88
Hindu 12 J u n e 1918, 9 October 1920; G . O . 1010 (L a n d M , M ) dated 29 July
1920. T.N.A.
89
Hindu 6 August 1919.
90
Note signed C. T o d h u n t e r dated 17 January 1920 in G.O. 548 (L a n d M , L)
dated 7 June 1920. T.N.A.
91
Andhrapatrika ij August 1920. R.N.P.; Hindu 29 November 1920.

169
 The emergence of provincial politics

In ten short years, rural board politics had been transformed out
of all recognition and had altered the nature of rural politics in
general. It was not only that new powers had been made available to
self-seeking bosses-turned-politicians. Any man with interests to
protect or with projects to implement became committed to playing
the new games and to obtaining a majority decision in the board, for
in isolation he was both vulnerable and impotent. In Rajapalaiyam,
Ramnad district, for example, the landlord A. K. D. Dharma Raja's
greatest aim in life was to run his little market without interference.
For the first few years after its foundation, he had accomplished this
end by using his family's position as village headmen to have it
expunged from the records. Later, as information about its existence
trickled through to higher officials, he had to bribe his tahsildar to
look after it in the taluk board. By the early 1920s, however, he was
able to rely no longer on these 'informal' political techniques. The
Raja of Ramnad, as soon as he became non-official district board
president, began to extend his political power by taking over control
of all the markets in the district.92 Dharma Raja, and the Nattukottai
Chetties of Chettinad who also had maintained their rights over local
markets by bribing subordinate officials, were forced to seek contacts
with and secure election to the district board in order to preserve their
positions.93 Once they were on the board, they forged alliances with
other politicians from across the district and formed a faction which
opposed every move which the raja made.
Similarly, although for more noble purposes, the Sanskrit scholars
and patrons of Kallidaikurichi, Tinnevelly district, found themselves
swept into the vortex of district affairs through no action of their own.
In 1915, in order to rationalise Sanskrit education in the area, they
had approached their local taluk board and asked that an endowment
be transferred from one school to another. In the past, the taluk board
had complied with similar requests and, on this occasion, it did so
again. However, a faction in the Tinnevelly district board, none of
whose members came from Kallidaikurichi, was outraged by the
decision and put pressure on the district board executive to have it
overruled.94 For the first time, the educational politicians of Kalli-
daikurichi had to seek alliances and support at the district level in
order to enforce their decisions in the locality.
92
G.O. 1984 (L.S.G.) dated 7 September 1923. T.N.A.
93
G.O. 7 8 3 (L.S.G.) dated 3 M a y 1922; G.O. 811 (L.S.G.) dated 9 M a y 1922. T.N.A.;
Hindu 16 March and 31 July 1922.
94
Madras Mail 18 June 1917; G.O. 175 (L and M , L) dated 7 February 1918;
Hindu 5 November 1918; G.O. 1090 (L and M , L) dated 13 September 1919.
T.N.A.
170
 Local structures of political power

The increase in rural board powers and the withdrawal of the

bureaucracy from the rural board executive had opened the way for
a much larger and more overt system of politics than before. Rural
board executives could no longer be guaranteed of majorities by
their bureaucratic connections; they had to work for security by
managing their patronage and alliances judiciously. Rural politicians
could no longer protect or promote their interests by buying individ-
ual officials; they had to organise and seek election and exercise
pressure on the board majority. Political groupings, formed to work
as factions within the institutions, began to link together magnates
whose bases were a hundred miles or more apart. In Coimbatore by
1913, for example, a district-wide political combination was clearly
visible. Certain Coimbatore town politicians had seized the district
board and raised rural taxation. Immediately, political powers in the
larger villages and mofussil towns put together a protest movement
which drew support from all four corners of the district.95 Similarly, in
Godavari by 1919, the pro- and anti-president factions were marshal-
ling their forces in every taluk. Similarly again, in Guntur by the early
1920s, those who favoured and those who opposed the rule of P. C. N.
Ethirajulu Naidu, the district board president, were organised across
four taluks.96 By 1920, the old rivalry between Chinnarappa and
Thimma Reddi in Gooty was being channelled into the politics of the
new order. P. Kesava Pillai (who worked as a lawyer for Chinnarappa)
became district board president and Chinnarappa joined the district
board; they used their combined power in the district to have Thimma
removed from all local self-government offices.97
The expansion of political activity, however, did not stop at the
frontier of the district. While local affairs had been handled by semi-
autonomous district officials, the Secretariat could expect to have
little say in how their decisions were made. But once power was con-
centrated in institutions which had to report to and were subject to
the control of a department in the Secretariat, its interference was
both more possible and more telling. The Local and Municipal Depart-
ment could reverse board decisions, hear appeals against board
election results and, most importantly of all, until 1922 it could
nominate the vast majority of taluk board and all district board
presidents. In the next chapters, we shall see more closely how the
links between district and provincial levels of politics were developed.
95
Hindu 22 a n d 24 September a n d 1 a n d 7 October 1913.
96
C. J. Baker, 'Political Change in South India, 1919-37', Unpublished fellowship
dissertation, Queens' College, Cambridge, 1972, pp. 200-14.
97
G.O. 180 (L and M, L) dated 27 February 1920. T.N.A.

171
 The emergence of provincial politics

But here we can show some of the effects which interference from
above could have on district interests and ambitions. The Coimbatore
town banker, T. A. Ramalinga Chetty, was at the centre of the rural
protest movement which appeared in 1913. As district board vice-
president, he was largely responsible for raising the land cess. He was
the object of intense hatred in the Coimbatore countryside and,
indeed, was publicly reviled by a cross-section of the rural populace
when he deserted the nationalist cause for the non-Brahman move-
ment in 1917.98 Yet, through his Secretariat connections, he became
Coimbatore's first non-official district board president in 1920 and
unquestionably its most powerful politician." Similarly, it was essen-
tially through the interference of Madras city that the Kallidaikurichi
Sanskrit College affair was drawn into district affairs. The taluk
board's decision was passed without opposition when it was first put
to the district board. But afterwards, under pressure from the nascent
Justice Party leadership in the capital, which had excellent Secretariat
contacts, a Tinnevelly district board group demanded that the ques-
tion be reopened.100
Although creating the framework for much broader political
contacts in the districts and for new connections with the capital,
the institution-building of the 1910s and 1920s was designed not to
upset the social balance of politics in the localities. The British were
intent on harnessing the power of the rural elite; they did not wish
to undermine the local roots of that power. The nominations to
administrative committees went to men of substance, not to paupers;
and the franchise requirements for the rural boards were pitched at a
level which excluded 97^ per cent of the rural population.101
The real social impact of the administrative reforms was much less
on the relationships between rural-local bosses and their dependents
than on the relationships between the rural-local bosses themselves.
As we noticed when examining the economy, their participation in
local-level culture meant that social connections between them were
extremely fragmented. Economic development had just begun to
forge supra-local linkages. The processes set in motion by the economy
98
K. V. Srinivasa Iyer to P. S. Sivaswami Iyer 18 J u n e 1917. P. S. Sivaswami Iyer
Papers. N.A.I.
99
Note signed Panagal dated 16 May 1923 in G.O. 1131 (L.S.G.) dated 17 May
1923. T.N.A.
100
G.O. 175 (L and M, L) dated 7 February 1918; G.O. 1090 (L and M, L) dated
13 September 1919. T.N.A.; Hindu 5 November 1918.
101
Calculated from Evidence taken before the Reforms Committee (Franchise). Madras
(Calcutta, 1919), Appendix 1.

172
 Local structures of political power

were picked up by the new institutions and carried much farther

much faster. Rural-local bosses had to expand the bases of their
operations not simply to increase their wealth but to keep their
existing political dominance. The traditional fight with sticks and
knives between tiny private armies was rapidly being replaced by the
less bloody but infinitely larger, strategically considered war of
attrition between factions which balanced the interests of dozens of
rural-local bosses. By expanding the structure of political contact,
these institutional wars also expanded the structure of social contact:
rural-local boss families were encouraged to extend their marriage
networks in order to make alliances; they saw and came to appreciate
the benefits of education; they began to reside together at headquarter
towns. Ultimately, these factors would lead on to the development of
distinct regional patterns of integration out of the elite members of
previously separate local-level cultures.
Importantly, however, as we mentioned earlier, this final stage was
barely emergent in 1920. The secondary social and cultural accretions
of the movement had as yet almost no separate existence apart from
the forces which were bringing about the movement. These forces
were partially economically driven but consisted mostly of the political
advantages to be gained by participating in the new institutions.
Consequently, it is not surprising that the political behaviour of the
early regional elite came to be dominated by the necessities of institu-
tional politics. Patterns of district politics tended to be purely factional.
They consisted of endless interest-swopping, horse-trading and par-
lour diplomacy as rural politicians strove to capture points of execu-
tive power.102 There was as yet no sign of wider social or ideological
conflict in them.
From the British point of view, of course, this type of develop-
ment was near-perfect. As the British had built the institutions and
had preserved powers to alter, rebuild and interfere with them, they
could control the course and character of political expansion. Yet,
obviously, their success with the institutions was due mostly to the
peculiarities of the social and political structure of the dry region.
The isolation of the rural-locality in the nineteenth century and the
continued ability of the rural-local boss to hold down subordinate
groups in the locality were basic conditions for the growth of adminis-
trative, institution-orientated, factional politics. Where these condi-
tions did not apply, however, British administrative reform was less
likely to produce such happy results.
102
See Baker, Politics of South India, 1920-1937, ch. 2.

173
 The emergence of provincial politics

The Cauveri delta and Malabar

In the Cauveri delta and in Malabar, the social and political structure
thrown up by the economy was broader than that of the dry region.
Whereas changes in the dry economy were only beginning to promote
social interaction among the now dominant rural elite, in Tanjore and
Malabar a regional landed elite had been in existence for centuries.
Naturally, this fact impinged heavily on the operations of the British
administrative settlement. The rural locality would never become as
autonomous as it was in Anantapur or Coimbatore. The British were
less able to break the crucial social tie between the higher offices of
district and the lower offices of local government for the neat distinc-
tion between kurnam-type and headman-type social groups did not
exist. Nearly all dominant local groups were literate and participants
in the regional, 'state-level' culture. Consequently, external resources
were more freely available and more freely used by rural-local bosses
and the rural-locality was more closely meshed to the state. Equally,
the long-term presence of wide social arenas had led to the develop-
ment of a large number of political arenas. The trend towards frag-
mentation, which followed in the wake of the British conquest, by no
means touched all of these. Status rivalries at festivals or struggles
over accepted religious orthodoxy could produce political splits which,
albeit for a few days, ran the length of the districts and tore into rural
society. When the British designed new institutional political arenas,
therefore, political formations which had been built up before and
independently of them were able to mount an invasion of them. In
Tanjore and Malabar, district politics were not as sterilely factional
and as geared to the simple winning of institutional place as those of
the dry region. Occasionally, they contained a hint of real social
war.103
Nonetheless, we ought not to go too far in pointing to the dissimil-
arities between the areas. British administration attempted to follow
the same pattern as it had done in the dry region and consequently
produced many of the same results. It passed village office to members
of the mirasidar, jenmi and kanomdar elites, and relied on the
103 p o r example, when the Porayar Nadars put up a candidate for election to the
local temple committee in Kumbakonam, S. A. Saminatha Iyer, the Brahman
committee chairman, organised a campaign against them on the issue of their
untouchability. The campaign was extremely bitter and violent. Admittedly, it
did little good, for the Nadars' wealth was such that, their status notwithstanding,
their candidate was elected by a huge majority. Nonetheless, it is indicative of
the nature of political constituencies in the Cauveri delta that Saminatha Iyer
should have tried this tactic at all. Hindu 23 July, 6, 15, 22, 24 and 27 August
1888.
 Local structures of political power

powers given them by their stranglehold on economic production to

keep order in the countryside for it. Apart from allowing the contin-
uation of the state-local level social tie, the British provided few
positive resources of their own for the continuation of rural govern-
ment. The possession of rural dominance, therefore, hinged on the
means by which economic control could be exercised and, as these
means operated only within the confines of the restricted rural
locality, rural power itself was largely locally derived. When mira-
sidars and jenmis jousted with each other in regional socio-political
arenas they were seldom fighting battles which were material to the
continued existence of their own local bases. Their mock wars were
much more about the status and respect which should be accorded to
them by other members of their rural elites and much less about the
numbers of dependents whose behaviour in the locality they hoped
to control. The battle for local power was still a matter for the locality.
The emergence of the new administrative institutions, therefore,
was potentially as traumatic to political life as it was in the dry region.
District boards and irrigation committees exercised corporate func-
tions which previously rural-local bosses, using their economic
positions and fissures in the formal bureaucratic structure, had
exercised in their own right and in isolation. To protect some of the
important pillars of their local power, Tanjore and Malabar rural-
local bosses also had to plan their alliances to gain administrative
power and hence were lured into object-orientated, factional district
politics. Many of the status and social issues, which once had split the
elite, were buried for mutual benefit in the institutions. Around
Kumbakonam in the 1880s and 1890s, for example, elite political
divisions had turned sharply on questions of caste propriety. In local
politics, confrontations had occurred repeatedly between, on the one
side, the high-caste Sat-Sudra Pandarasanidhi of Dharmapuram and
the Brahman mirasidar-lawyer S. A. Saminatha Iyer and, on the other,
the low-caste Odayar T. S. Sivaswami Odayar and the untouchable
Porayar Nadar family. Certainly, much more than caste status lay
behind the feuding — economic and political considerations were
inextricably involved; certainly also the caste issue was perceived more
by the faction leaders (the elite) than by their followers — each man's
dependency network was of an entirely cross-communal character.104
Nonetheless, the vocabulary of caste was much in evidence during the
skirmishes. However, in 1896 the district board decided to build its
own railway line, and completely overturned the relationships of the
faction leaders. The value of land sited next to a railway track was so
handsomely increased, and the political power to be drawn from it
104
See above pp. 106, i n , 145.

175
 The emergence of provincial politics

so greatly raised, that matters of mere status had to take second

place to matters of profit and power. S. A. Saminatha Iyer and T. S.
Sivaswami Odayar discovered a mutual interest in routing the pro-
posed line through their Mayavaram properties. The Nadars split
with the Odayars in order to take the railway down to their holdings in
Tranquebar. And the Pandarasanidhi began to put forward a third
alternative.105 District political formations now moved along wholly
different channels.
The steady intrusion into supra-local politics of more and more
matters which affected the foundations of local power altered the
terms of Tanjore and Malabar district politics and made the new
administrative institutions central to them. If institutional factional
alignments were never the only kind of district political groupings,
they soon became the most usual. This, of course, well suited the
purposes of the British — as, indeed, had the older administration.
Except in the upland taluks of Malabar,106 social order had seldom
collapsed in either district during the nineteenth century, for the
chosen collaborators were able to perpetuate their dominance over
subordinate groups. Later, the administrative reforms harnessed this
local power to district and provincial institutions where it could be
used to promote development. The British had to be more cautious in
Tanjore and Malabar, for the pre-existence of regional elite ties made
the danger of an elite revolt against them more likely.107 But by
judiciously manipulating important institutional powers, Fort St
George could still control the responses which it received.

The Andhra deltas

The only major area of the presidency in which the British adminis-
trative machine created serious problems for itself was the Andhra
deltas. As we saw earlier, the development of the Kistna-Godavari
anicuts had produced a profound economic revolution. The region
became characterised by its extensive pattern of rural—urban trade
and by the mobility of its labour force. These factors had led to a deep
integration of social as well as economic life. In many ways, the rural
locality as a meaningful unit of economic and social life had been
replaced by the urban hinterland. Moreover, in the countryside at
105
Hindu 10 and 17 June 1896.
106
The Moplah country.
107
Revenue resettlement policy, i n particular, was ultra conservative for incautious
demands could bring a united front o f opposition. See above, p. 52.

176
 Local structures of political power

large, there was no small elite capable of holding down the mass of the
rural population. This was pre-eminently the region of the middle-
peasant and, although some landholders were certainly richer than
others, the nature of economic organisation prevented their differ-
ences from being turned into ties of clientage and dependency. Every
square mile of the deltas contained many independent landholders
whose conduct towards each other produced incredibly complex
patterns of rivalry.
The formal processes of British administration, however, took no
notice of the social changes wrought in the area by economic develop-
ment. The British tried to govern the Andhra deltas in exactly the
same way as they governed elsewhere in Madras — through the unit
of the rural locality and with the help of a small rural elite. It is
scarcely surprising that government in the region steadily degener-
ated into chaos. The village police were incapable of maintaining
order. The Kistna and Godavari districts had the highest crime rates
in the presidency. Most of this crime was not the organised murder
and looting expedition of the dry districts, which was the result of
clashes between internally stable rural-local kingdoms; it was petty
house-breaking and theft on a scale to suggest a serious breakdown in
social order.108 Local courts (formal and informal) were incapable
of meeting the litigation needs of the rural populace. The courts of
higher jurisdiction in the deltas were choked with a vast number of
intricate local disputes, which elsewhere would have stayed in the
locality.109 The district munsiffs' courts at Rajahmundry, Ellore,
Masulipatam, Amalapur and Guntur were consistently the slowest
to complete cases and were considered the most arduous postings in
the Madras judicial service.110 The possibilities of calling in the
urban authorities to influence affairs in the countryside also made the
village revenue establishment incapable of producing a steady flow
of revenue. Appeals against the income-tax, which was village officer
assessed in the countryside, were highest in the Kistna and Godavari
districts. For many years, the rate of appeals ran at 50-60 per cent of
the assessments.111 As senior revenue officers in the towns had to
108
See 'Statistics of Crime' in Madras Police 1878 to 1920.
109
Statistics of Criminal Courts in the Madras Presidency for the year 191$ (Madras,
1916), p. 2. Around Bezwada and Masulipatam 'litigation [was] developing with
extraordinary rapidity'. Report on the Administration of Civil Justice in the
Presidency of Madras in 1890 (Madras, 1891), p. 14.
110
Ibid., 1900, p. 5.
111
Income Tax 1890-1, p. 13; ibid., 1893-4, P- 78-

177
 The emergence of provincial politics

hear the appeals, the cost of administering the tax in the deltas was
almost equivalent to its gross yield. Similarly, in their ordinary
revenue work, village officials ran into constant difficulties. By the
time of the civil disobedience movement, their position had become so
bad that they were completely unable to collect the revenue in the
face of village opposition. In 1930, when the British tried to dismiss
officials of defaulting villages, they found that, in a great many cases,
the officers had been carried away on a tide which they could not
control.112
The attempts by the British in the later nineteenth and early
twentieth centuries to tighten their rural control and break down
obstructions in the village hit the delta village officer, who was
already under pressure from below, particularly hard. The improved
information given to senior officials by the revenue resettlements,
the reduction of village establishments and the removal of the power
to issue notices of demand and restraint at will, weakened the few
vestiges of authority which remained to distinguish the village officer
from his fellows. Many decided that the powers in local society given
to them by their office no longer matched the responsibilities they had
to undertake. Some quite simply quit their posts;113 others began to
take action against their employers in order to improve their position.
Curzon's deputy inspectors met their greatest difficulties in the
delta districts.114 By 1914, even the revenue department was on the
point of collapse: the Collector of Kistna reported,
My taluk officers inform me that theyfindconstant recurring difficulties in
getting their ordinary revenue duties done by the village servants. Resigna-
tions are very common, temporary absence from duty a monthly occurrence,
and wholesale strikes by village servants are by no means unusual.115
Although weaker than his colleagues in the dry districts, the delta
village officer had much better means, through the closeness of rural-
urban contacts, of making his grievances felt. Between 1916 and 1920,
officer associations petitioned and campaigned for better conditions
112
G.O. 639 (Public) dated 5 June 1931; G.O. 938 (Public) dated 11 September 1931;
G.O. 939 (Public) dated 11 September 1931; G.O. 980 (Public) dated 21
September 1931; G.O. 1075 (Public) dated 20 October 1931. T.N.A.
113
Minor village servants found that they could make more money by working as
field labourers during the paddy harvest than by remaining in their posts.
Madras Police 1913, Appendix p. 9.
114
G.O. 1675 (Judicial) dated 18 August 1913. T.N.A.; Madras Police 1914,
Appendix D, p. 72; ibid., 1919, Appendix E, p. 68.
115
Madras Police 1913, Appendix, p. 9.

178
 Local structures of political power

and more power.116 During the non-co-operation movement, a large

number of village officers resigned their posts and threatened to
bring down the whole revenue administration.117 The Collector of
Guntur interviewed some of the more militant headmen and found:
'They maintain warmly that they have not resigned for political
reasons, but because Government have not redressed their service
grievances put forward by the Village Officers' Association.'118 Of
course, their very weakness made them vulnerable to government
pressure, and the strike soon collapsed when they were threatened
with dismissal and revenue defaulters with loss of their lands. But
social instability remained a serious factor in Andhra rural politics
and spilled over once again into the civil disobedience movement.
British administrative design was based on social assumptions
which were too static and too local for the Andhra deltas. By the
1880s, the social forces released in the area were beginning to require
a much broader based political structure than Fort St George was
prepared to allow them. Already anti-administrative (and thereby
anti-government) campaigns were taking on the appearance of mass
political action. Whereas in the dry region over forest rights, we saw
that the point of contact between local powers and the formal govern-
ment came at the edge of the locality, in the Andhra deltas over water
rights it was beginning to come at the edge of a much larger territory.
Obstreperous forest subordinates tended to meet their end at the
hands of the rural-local boss' army but obstreperous P.W.D. sub-
ordinates soon were having to run before payment strikes and press
agitations organised across several taluks.1™ There were simply too
many separate, independent political decision-makers in the country-
side and far too many means of communication between them for
political activity to be contained in British-made packages.
In default of a workable administrative system provided by Fort
St George, the Andhra deltas began to develop their own. The critical
economic and social construct of the region had become the urban
116
Hindu 12 November 1920; Andhrapatrika 4 February 1919 and 7 September
1920, Desabhimani 8 January 1919, Gramapulana 10 September 1921. R.N.P.;
Hindu 27 January 1919 and 3 February 1920.
117
See my 'Country Politics: Madras 1880 to 1930' in Modern Asian Studies vn: 3
(i973)5 517-
118
'Guntur/Collector's report on Resignations of Village Officers - 3' in Venka-
tarangaiya, Freedom Struggle in Andhra Pradesh, 111, 264.
119
For example, in the 1890s, a chain of district conferences was set up across the
deltas to discuss and co-ordinate action on, particularly, the irrigation issue.
Hindu 17 April and 5, 6, 11, 12 and 15 June 1896.

179
 The emergence of provincial politics

hinterland and it was this which progressively became the critical

political construct. Urban-based lawyers, journalists and government
servants, already linked to middle-level peasants and rural—urban
traders through the facilities of litigation, petition and religious
expression which they provided in the towns, began to forge more
actively political links. It was they who developed the organisation
for the anti-P.W.D. campaigns, for the village officer protest move-
ment and for agitations to reduce land revenue demands following
resettlements. Increasingly, they went out into the countryside to
find grievances which they could represent and problems which their
expertise as publicists and organisers could help to solve.120
The nature of political expansion in the delta tracts had two im-
portant consequences, the first on the relationship between the area
and British rule and the second on the social composition of political
leadership. Political development was not only taking place indepen-
dently of British institutions, it was in many ways taking place in opposi-
tion to them. The points at which the new political culture displayed
itself most clearly were precisely those at which the interests of rural
society clashed with those of the British government. The failure of
Fort St George to accommodate the 'urban-hinterland' to its adminis-
trative technique left that construct outside and antagonistic to the
administration. In the short-term, this meant that rural—urban
political leaders had very little to gain by association with the British
and held their constituencies together mainly by undertaking anti-
governmental agitation. In 1907-8 and 1920-2, those leaders who
were committed to the Congress cause had no difficulty in taking
their followings into the boycott and non-co-operation campaigns.
The Andhra deltas were by far the most active regions of the presidency
in these movements.121
When the British began to expand their administrative institutions
in the 1910s, they found that they were able to seize back only part of
the initiative which they had lost during the previous thirty years.
The powers which they offered in the district boards and other
committees were formally the same as those offered elsewhere but they
were too few and were coming too late to have the same impact.
120
For example, see the meeting of ryots of Ellore taluk, which was addressed by
S. Bhimasankara Rao, Rajahmundry municipal chairman, in Hindu 12 October
1894. See also the pilgrimage of N. Subba Rao through the Godavari villages in
Hindu 4 October 1894; and Hindu 9, 10, 11, 12 and 17 June 1897.
121
For a detailed account of Extremism and Non-cooperation in the Andhra deltas,
see Venkatarangaiya, Freedom Movement in Andhra Pradesh, vols. 11 and 111.

180
 Local structures of political power

Certainly, control of railway building and of the allocation of local

taxation, and even of water rights, were not matters which the rural
population could ignore. Certainly also, a number of urban hinter-
land political networks were drawn into the institutions and persuaded
to play entirely by their rules. Major battles for control of the district
boards and for access to the government in Madras city were not
wanting in the Andhra deltas.122 Yet they did not come to dominate
supra-local political activity. In the first place, the new institutions did
not absorb to the same extent the political functions which had been
exercised by the rural-local boss and which had formed the basis of
his power. Control in the countryside, such as it was, was practised
more through the power of public opinion and of spontaneous mass
action than though the supplying of vital economic resources. The
new institutions, of course, barely catered for this type of control
which continued to make itself felt most strongly in protest against
the government. Secondly, the manner in which the political rural-
hinterland had developed meant that it contained a wide variety of
linkages which were more social than political in origin. In particular,
militant religious and cultural revivalism had helped to draw it to-
gether. Again, the new institutions took little account of this factor.
In consequence, they were unable to create or to impose their
pattern on supra-local political activity but, instead, were patterned
by what was there before them. District board politics in the deltas
made free use of ideological and social issues which were unheard
of elsewhere. Movements of caste solidarity and reform,123 cultural
conflict between Brahman and non-Brahman,124 and class conflict
between zamindar and tenant appeared inside the boards and helped
to define the political struggle. 125 Although factional alignment was
not absent and, indeed, sometimes lay behind the manipulation of
caste, class and cultural symbols, it formed only part of the entire
political web. In particular, the appeal of the Congress cause was very
strong. From the earliest days of the new institutions, local Congress
committees played a major role in electoral and 'chamber' politics
and, rather than becoming pawns in a British-controlled game, they
122
See my 'Country Politics: Madras 1880 to 1930', pp. 519-22.
123
Such as 'the Kamma scare' in Guntur in the early 1920s. See Baker, 'Political
Change in South India 1919-1937', pp. 196-213.
124
M . V e n k a t a r a t n a m Naidu led a campaign t o oust t h e Godavari district board
president, D . Seshagiri R a o , because h e was a B r a h m a n . T h e campaign failed
but it is interesting that it was ever started. Hindu 7 M a y 1920.
125
Kaleswara Rao, Na Jivita Katha—Navya Andhramu, pp. 4 3 4 - 6 0 .
 The emergence of provincial politics

used their institutional positions to attack the British themselves.

In 1915 and 1916, for example, the Guntur taluk board could be
found voting government moneys to nationalist funds. 126 During the
Home Rule League agitation, several local boards passed resolutions
in support of League objectives, refused to co-operate with the senior
administration and crippled local government by resigning en masse.121
By the early 1920s, the vote-pulling capacity of the Congress had
become so great that some local elections were practically decided in
the Congress district office - whoever received the Congress ticket was
guaranteed of victory at the polls. 128
The consequences of these developments on the social composition
of the political world are not difficult to envisage when set against
the background of the rest of the presidency. In the Andhra deltas,
economic freedom allowed factors of independent moral conviction
and political persuasion a much greater role in determining questions
of authority and obedience. It was not, of course, that considerations
of authority and morality did not matter in the dry zoneorinTanjore-
Malabar. But it was simply that the extreme economic and political
weight of the dominant local powers of those areas pressed rural
society into accepting their authority more or less automatically:
narrow economic and political opportunities and the prevalence of
local-level culture gave most countrymen no experience of an alter-
native authority. In the Andhra deltas, large numbers of people were
in a position to know and to choose between several sources of author-
ity and several courses of action. This fact threw into sharp relief the
activities of the publicist whose skills in communication gave him
access to public opinion, for the influencing of public opinion was an
essential prerequisite to the organisation of groups for political con-
flict. In the deltas, effective political authority passed increasingly
to educated publicists who were domiciled in the towns and whose
equivalents elsewhere in the province had no rural positions of any
note. By using the economic connections of the urban hinterland and
the cultural connections developed by the vernacular press, urban
publicists built significant regional followings. A. Kaleswara Rao, for
example, a Brahman lawyer from Bezwada town, and Konda Venka-
tappayya, a Brahman lawyer from Guntur town, stood at the head of
political networks which stretched through the countryside of
126
G.O. 2 1 1 (L and M , L) dated 15 February 1919. T.N.A.
127
G.O. 8 7 3 (Public) dated 18 November 1921. T.N.A.
128
See Rajahmundry municipal and taluk board elections in Hindu 10 May 1921,
4 September 1922, 14, 19, 2 0 April 1923.

182
 Local structures of political power

their districts and which were composed of middle-peasant (among

other) groups. Venkatappayya showed the breadth of his contacts in
the non-co-operation rent strike,129 and Kaleswara Rao not only in
agitational but also in electoral politics - in the early 1920s, he broke
the dominant faction on the Kistna district board.130 The complexi-
ties of political life in the deltas, with the thousands of independent
minds to be found there, put a high premium on organisational talent.
The expert publicists of the towns moved smoothly into positions
of rural leadership.

Temples
Within the complex of institutions which acted as the brokers of
political power in South India lay the major area temples. From the
later nineteenth century, administrative change was altering the poli-
tical character of these as quickly and in the same ways as it was alter-
ing the other institutions of politics. The particulars of administration,
however, were rather different and so require special considera-
tion.
In point of culture, of course, the major area temples pertained only
to state-level social groups; their relationship to local-level cultures
was at best indirect. But their influence on the limited sector to which
they catered was very considerable. The cultural world in which Hindu
social groups lived was defined by religious status, and the control
of ritual ceremonies, which temple powers enjoyed, made the temples
arbiters of social mobility and authorities on matters of social be-
haviour. A strong relationship to temple power was essential if a social
group were to preserve its existing place in society or to improve its
general standing. Rising Nattukottai Chetty and Nadar merchants,
for example, bid heavily for the temple offices which would help
them to achieve their much-desired increase in social status.131
Beyond ritual and beyond state-level groups, however, the temples
also possessed economic power which made them relevant to every-
body. In 1879, the capital of the temples was estimated officially at
8^ crores of pagodas, which yielded an annual income of 50 lakhs of
129
Venkatappayya, Sviya Caritra, 1, 226-301; Venkatarangaiya, Freedom Struggle
in Andhra Pradesh, 11, pp. 250—308.
130
Kaleswara Rao, Na Jivita Katha — Navya Andhramu, pp. 434-60.
131
Nattukottai Chetty and Nadar ritual assertiveness occasionally led to bitter
political fights with the social groups who held the temples which they were
storming. See Hindu 9 February 1922; Hardgrave, Nadars of Tamilnad, pp.
108-14.

183
 The emergence of provincial politics

pagodas.1212 But these figures excluded the capital value and income of
land which was held on tenures other than inam and whose worth may
have been as high as half as much again. The enormous amount of
wealth held by the temples may be appreciated readily when set against
the entire education budget of the presidency at this time, which
reached only Rs 30 lakhs or 8^ lakhs of pagodas a year.133 Besides pos-
sessing land, major temples drew pilgrims from across the whole of
India, whose purchasing power supported entire local economies;
they controlled legal monopolies over the sale of many sacred com-
modities; they organised huge markets and fairs to coincide with
their principal festivals.134 They represented important sources of
wealth and political power in themselves.
Before the arrival of the British, all native governments had
exercised a close supervision of temple activities. Indeed, it was
through religious institutions that Chola and Vijayanagar warriors
had extended the arms of the state into the locality.135 In the first
half of the nineteenth century, the East India Company — through its
maid-of-all-work, the Board of Revenue — had continued to scrutinise
temple affairs, although not in the same systematic fashion: unlike
its predecessors, it did not use religious ideology and the manipulation
of religious resources in its methods of rule. However, underpressure
from the Government of India, which sought to bring its Southern
presidency into line with its general policy of religious neutrality, even
this light measure of control was steadily relinquished. By Act XX
of 1863, the Madras Government withdrew completely from the ad-
ministration of the Hindu religion and handed over its rights more or
less to whomever it could. Where zamindars could prove an hereditary
interest in a temple, they were made its chief executives; where,
particularly in maths, a. head priest could be found (or invented), he
was given virtual private property rights in the entire endowment;
where no intermediaries were available, elected committees were
formed to carry on the administration. At a single stroke of the pen,
the British had cut all connection with the institutions of religion.
Try as it might - and it was to write eleven Bills over the next sixty
years - the Government of Madras could not persuade its overlord,
132
Home Public A October 1879, Nos 149-62. N.A.I. A Madras pagoda was worth
Rs3i.
133
Report on the Administration of the Madras Presidency during the year 1880—1
(Madras, 1882), p . lxxv.
134
For example, during the Chittrai festival at Madura, one of the greatest cattle
fairs in the province took place. As many as 200,000 cattle could change hands.
135
See, for example, Spencer, 'Religious Networks and Royal Influence', 42—56;
Spencer, 'Royal Initiative under Rajaraja V.

184
 Local structures of political power

the Government of India, to reverse its decision. The Government of

India, packed with men whose experience of Hinduism came from
Northern India and who possessed a quaking fear of fanatical priests,
did not understand the nature or the significance of the temples
of Madras, and, in the manner of lofty superiors, refused to be
told.136 Act XX was just another of the many issues which divided the
many governments of Madras.
Although the vast majority of the 75,000 religious institutions
covered by Act XX were small and possessed only a few acres, some
were of enormous wealth, and their trustees and servants men of
great influence. The pandarasanidhi of the non-Brahman math at
Thiruvadathorai in Tanjore, for example, controlled 3,000 acres in
his home district, 25,000 in Tinnevelly, 1,000 in Madura and lesser
amounts in several other districts. He also possessed rights of ap-
pointment of priests and trustees to fifteen other temples each with its
own considerable endowments.137 The Dharmapuram Pandarasanid-
hi, also from Tanjore, owned 2,500 acres in the district and appointed
to twenty-seven temples.138 At Madura, the Sri Minakshi temple com-
mittee possessed lands which yielded Rs 2 lakhs per annum.139 The
leading pilgrimage centres at Srirangam, Rameswaram, Kumba-
konam, Tirupati and Kalahasti had incomes sufficient to embarrass
the richest zamindar. The power and influence of the major temples of
Madras were such that no politician could afford to ignore them.
The greatest problem with Act XX was, quite simply, that the
legal provisions which were supposed to guarantee the administration
of public endowments in the interests of the public were totally in-
adequate. Certainly, there were procedures for taking trustees and
pandarasanidhis to court; but the Act gave the court no power to force
the accused to produce his records, with the result that accusations
were impossible to substantiate. Further, of course, litigation was
expensive and the only brake on it was money. A defending trustee
could pay his lawyers from temple funds; a public spirited citizen had
to meet the costs himself.140 For all practical purposes temple
trustees were outside the law.
Appointments to temple committees and pandarasanidhi-ships
136
See note signed A. P. Macdonnell dated 20 July 1894 m Home Judicial Deposit
January 1912, No. 10. See also Home Public A October 1879, Nos 149-62;
Home Public A September 1894, Nos 312-18; Home Public A June 1903,
Nos 363-4; Home Judicial A July 1914, Nos 265-85. N.A.I.
137
M.D.G. H e m i n g w a y , Tanjore, 1, 234.
138
Ibid., p . 232.
139
Hindu 18 July 1929.
140
Home Judicial A July 1914, Nos 265-85. N.A.I.

I85
 The emergence of provincial politics

were for life and, once in office, there was little to prevent a man
using the endowments as he pleased. Leases on land and markets,
employment in the temple service and money were all at his disposal,
and few men resisted the temptation to use them as political tools.
The first Mahant of Tirupati temple was accused of putting Rs
92,402 to his own purposes, excluding private legal fees; the second
of misappropriating Rs 2,28,000; the fourth of Rs 1,30,000; the fifth
of Rs 6,26,000. The third Mahant, Bhagvan Dass, was known to have
misplaced Rs 2,27,000 and, while serving a prison sentence for
theft and embezzlement, was accused of diverting a further Rs
14,00,000.141 And all this was in addition to the perfectly legitimate
practices open to the Mahants in the administration of their vast
properties.
The Kumbakonam temple committee was notorious:
The local Temple Committee has been labouring for the past many
months to nominate a few reliable trustees to some of the temples in the
town and elsewhere, but has been able to do little so far. The state of
temple management cannot be worse. There are many temples each
with an income of many thousands of rupees, managed by a single trustee.
Each member of the committee has his favourite and has amassed a num-
ber of temples under particular individuals. Even committee members
are direct managers of temples; for instance one Mahalinga Chetty, a
committee member, directly manages Sri Kimpeshivara temple... the
fine estates attached to the temples are fattening the vultures of the land.142
The Tinnevelly Shaivite committee, the Madura district committee
and the Negapatam committee were no better. 143 In 1927, the first
Report of the Endowments board reviewed sixty-four years of Act XX:
Proper accounts of receipts and expenditure of temples were seldom main-
tained, surplus moneys were not always properly invested, and temple lands
were in some cases leased out in favour of relations and friends of trustees
and in some cases alienated on inadequate grounds or for personal ends.] 44
From the highest in the land to the lowest, all sought financial
succour from religion. Whether by the legitimate means by which the
early Congress begged money from such as the Sankarachariar of
Kumbakonam, the Pandarasanidhi of Dharmapuram and the Tham-
141
H o m e Public A September 1894, N o s 3 1 2 - 2 8 . N.A.I.; Hindu 21 N o v e m b e r 1910.
By t h e late 1920s, t h e income of T i r u p a t i was estimated at Rs 50 lakhs p.a.
142
Hindu 30 M a y 1896.
143
Hindu 12 D e c e m b e r 1896, 22 a n d 27 September 1910; Hindu 15 M a y 1896;
G.O. 1074 (L and M, M) dated 12 June 1912. T.N.A
144
G.O. 1337 (L.S.G.) dated 9 April 1927. T.N.A.

186
 Local structures of political power

biram of Tirupanandal 145 or by which V. S. Srinivasa Sastri gained

the endowment of a Sanskrit chair at his National College, 146 or by
the more dubious methods which the brothel-keeper K. Alladin
Rowther used to speculate in temple property in collaboration with
the Madura committee member C. Sambasiva Mudaliar, all re-
cognised the importance of the religious potentates of the South. 147
Except in the case of hereditary trustees such as zamindars, the
passport to entry into the world of temple politics was stamped by
election. The heads of maths and the Mahant of Tirupati were elected
by their disciples and, as may be expected, the politics surrounding
such elections can be described only by the epithet Byzantine. As
pandarasanidhis and mahants held office for life, one of the most
obvious methods of advancing a political cause was by murder: a
Tirupati Mahant, 148 and pandarasanidhis at Dharmapuram and Siva-
rankoil were killed by opposing faction during our period, 149 while
no fewer than three Pandarasanidhis of Thiruvadathorai died under
dubious circumstances.150 The members of temple committees also
held office for life but, as there were several places on the com-
mittees, assassination was a less necessary political tool. The temple
committee electorate varied from district to district and was tied to a
high land revenue or income-tax payment. Temple elections them-
selves, which were initiated by the 1863 Act, provide the earliest
example of the form of politics which the British were to introduce in
many other fields. As elsewhere, they were also dominated by men of
considerable economic power, who possessed many levers by which
to mobilise followers. A classic of its kind was the Kumbakonam
election of 1888, won by the abkari contractor T. Ponnuswami Nadar.
He used his influence with local officials to obtain the help of the
local police and revenue departments in canvassing, and with local
aristocrats, like V. Appaswami Vandayar of Pundi, who could supply
him with their dependents. He had contacts already inside the temple
committee, who put some of its resources at his disposal, and other
voters who were not tied to him and his friends by pre-existing links of
145
Report on the Proceedings of the Third Indian National Congress held at Madras
on the 2jths 28th, 29th and 30th December 188'7 (London, 1888), Appendix 1.
146
V. S. Srinivasa Sastri to G. K. Gokhale, 8 March 1908. V. S. Srinivasa Sastri
Papers. N.A.I.
147
G . O . 1384 (L and M , M) dated 16 August 1917. T.N.A.
148
Madras Police 1900, p . 5.
149
Madras Police 1882, p. 7 1 ; Hindu 4 M a r c h 1925, 5 July 1931.
150
Hindu 7 M a y 1920, 9 F e b r u a r y a n d 30 September 1922 a n d 8 August 1935; see
also, V. Kandaswami Pillai, TiruvavadathuraiKurisanam (Madras, 1921) (Tamil).

187
 The emergence of provincial politics

patronage and welfare were bought for silver rupees at the polling
booth.151 A whole cross-section of society worked together in his inter-
est, held together by a variety of strands which led to him personally.
The material character of much temple power meant that the
committees which managed the temples and acted as the guardians of
state-level culture often took on a remarkable appearance. In areas
of predominantly local-level culture, many of their members could
barely be described as being religious participants of them. In districts
like Coimbatore or Anantapur, rural-bosses from the Gounder and
Reddi castes often won seats. In areas of complex communal inter-
action, Christians and Muslims could gain access to their influence.152
Even in regions where state-level culture was strong, the social order
which regulated temple power could be the reverse of that which was
supposed to be regulated by it. At Kumbakonam, for example, in the
heartland of the Tamil Brahman, the committee was dominated for
many years by V. Appaswami Vandayar and T. S. Sivaswami Odayar
who came from different branches of the low Kallar caste and T.
Ponnuswami Nadar who was an untouchable Shanar. The three,
however, were among the wealthiest and largest landowners in the
district. In 1915, the Raja of Ramnad took over the committee.153 He
was a Maravar, of the same equivocal status as a Kallar.
In battles between magnates for local power, temples could be
perfectly secular weapons. Raja Rajeswara of Ramnad, for example,
sought to reassert his authority over the Nattukottai Chetties, who
had leased large parts of his estate from his father, by resuming the
temples in the leased portions.154 Similarly, V. Appaswami Vandayar
and T. Ponnuswami Nadar further extended their power by backing
the Thambiram of Thirupanandal in his bid to win independence
from the Thiruvadathorai Pandarasanidhi, who was their enemy.155
Temples and temple property were often vital elements in the com-
position of a local hegemony.
Until about 1908, however, temple politics did not extend their
significance beyond the area of the committee. The resources used
to control committees and win elections were those readily available
only around the temple and within the electorate - they were locally
151
Hindu 23 July, 22, 24 and 27 August 1888.
152
Such as K. M. Alladin Rowther of Madura whom we have seen. For the
attempt of a group of low-caste weavers to obtain a seat on a temple committee,
see Hindu 10 June 1915.
153
Hindu 8 February and 24 March 1915.
154
G.O. 4139 (L.S.G.) dated 25 September 1926. T.N.A.; Hindu 8 and 9 April
1921.
155
Hindu 30 October 1894.

188
 Local structures of political power

derived. Connections with the higher bureaucracy and the law mat-
tered very little when the temples were virtually beyond the control
of the government and the law. The only effective right the govern-
ment maintained was that of allowing District Judges to appoint com-
mittee members if vacancies were not filled within a specified time.
Given that almost every committee was faction-riven and that
factions which were likely to lose would do everything in their power
to prevent an election, this right was exercised on a number of occa-
sions. But the most usual form of appointment was by the ordeal of
the polling booth where the personal influence of the magnate or
rural-local boss was supreme. Moreover, as elections were held only at
long intervals, and as they were the only times when local powers
could fight for temple control, temple political conflict was as in-
frequent as it was socially prescribed.
In 1908, changes in the Code of Civil Procedure suddenly opened
out temple politics in the same way that administrative decentral-
isation was to open out district board politics. It became feasible to
take temple committees and trustees to court for malpractices, to have
them removed, to contest election results and to alter the political
balances inside institutions. Temple politicians could no longer rely
on their life-franchises of office and on the irregularity of elections to
preserve their positions for them; they had to manage their alliances
and their distribution of patronage judiciously to prevent strong
enemy factions from ousting them. Battle between them was joined on
a much broader front and there were no longer any respites between
engagement. In Madura, the Sri Minakshi temple committee im-
mediately fell apart as two contending factions sought to gain com-
plete sovereignty over each other.156 The cases which they began to
bring against each other were to be more than twenty years in the
courts. In Mayavaram taluk, Tanjore district, the immensely rich
mirasidar T. Somasundram Mudaliar took on the Pandarasanidhi of
Dharmapuram math, supposedly the spiritual leader of his caste,
for control of two local temples.157 In Kumbakonam, defeated elec-
toral candidates crippled the temple committee by lodging petitions
against their victorious opponents, which took months to clear.158
Everywhere, the struggle grew more vigorous, and politicians at-
tacked each other not only at the polling booths but in the temple of-
fices and during religious ceremonies themselves.
As with the case of the district boards, however, these developments
15(3
G.O. 1074 ( L a n d M> M ) d a t e c i I 2 J u n e : 9 1 2 - T.N.A.; South Indian Mail 21
August 1911.
157
Hindu 23 August 1910 and 12 October 1915.
158
Hindu 8 February and 24 March 1915.

189
 The emergence of provincial politics

did more than draw local politicians into wider district arenas. They
also linked the local arenas to the presidency capital. In Madras city,
a group of leading Indian lawyers and administrators formed them-
selves into a Dharmarakshana Sabha to use the new legal provisions
to purge the temples of sin. Over the next few years, their association
launched cases at most of the major temples in the South, including
Rameswaram, Srirangam, Madura, Tirupati and Kalahasti, and won
the right to dictate the appointment of various committee members,
trustees and temple managers. 159 As we shall see later, its inordinate
success was due in no small part to the enormous influence which its
organisers had over the career prospects of the judges who heard its
cases.160 However noble may have been their intention, decisions
taken in the Dharmarakshana Sabha could have a dramatic impact on
the balance of local power. In a few years, actions taken by the Sabha
had affected deleteriously the interests of such great powers as the
Raja of Kalahasti, the Zamorin of Calicut, the Mahant of Tirupati and
the Pandarasanidhi of Madura. Of course, connection to the Sabha
became an important positive resource in local factional battles. At
Mayavaram, T. Somasundram Mudaliar called on its aid to help him
defeat the Pandarasanidhi of Dharmapuram. 161 In Madura, one of the
factions on the Sri Minakshi committee sought to bring prosecutions
against the other in its name. 162 Yet again, institutional changes
were forging a new set of political connections and altering the nature
of political life.

The administration of the towns

The types of administrative change which the British implemented in
the towns of South India followed similar patterns to those in the
countryside and patterned politics in very similar ways. They require
separate consideration, however, for two reasons. Firstly, the greater
political role which status and cultural activities possessed in the
towns could have been affected profoundly by the changes. As his-
torians of other Indian regions have noted, educated publicists often
used the new institutions to establish political dominance in their own
159
Kerala Patrika 9 November 1912. R.N.P.; Hindu 20 November 1915 and 10 June
1918; K. Raghavayya to P. S. Sivaswami Iyer, 26 October 1916. P. S. Sivaswami
Iyer Papers. N.A.I.; The Fourth Year's Report on the Working of the Dharma
Rakshana Sabha (Madras, 1911).
160
G.O. 175 (L and M, L) dated 7 February 1918. T.N.A.
161
Hindu 12 October 1915.
162
G.O. 1074 ( L a n d M > M ) d a t e d I 2 J u n e 1912- T.N.A.

19O
 Local structures of political power

right.163 And secondly, the cross-communal networks of urban eco-

nomic powers were under greater pressure from outbreaks of com-
munal hostility than were those of rural powers. Again, urban ad-
ministrative change elsewhere has been connected with the politics
of communal conflict.164
Specifically urban administrative systems began to develop in
British Madras only from the middle of the nineteenth century. Pre-
viously, government had been continued by informal association
between agents of the bureaucracy and the principal inhabitants,
some of whom were granted formal revenue and judicial offices,
much as in the countryside. From the 1860s, as Fort St George was
pressed by the Government of India both for more money and for the
greater development of its resources, urban administrations became
more specialised. Municipal councils were formed in a few of the
larger conurbations to raise their own taxes and regulate their own
programme of conservancy and sanitary improvement.165 These
councils were chaired by the local Collector and filled with other
government officials. But they also contained several nominated non-
officials who did most of the work. Although they possessed some
corporate powers, the councils tended to conduct much government
through the medium of individual councillors. The men nominated
by the Collector each represented certain wards in the town and were
personally responsible for executing council decisions within
them.166 Many were further nominated as Honorary Magistrates to
try breaches of council rules and transgressions against the bye-laws.
They thus possessed considerable influence as individuals within the
urban political structure. House-fronts which encroached on public
property could be torn down at their whim, shops closed as a danger
to health and men fined for petty offences. Reviewing his municipal
administration in 1874, the Governor of Madras saw that: T h e
Government of a Municipality is in fact an oligarchy dependent upon
a superior power'.167
It would be difficult to characterise all the people whom 'the
163
C. A. Bayly, 'Local Control in Indian Towns: The Case of Allahabad' in
Modern Asian Studies, v : 4 (1971).
164
F. C. R. Robinson, 'Municipal Government and Muslim Separatism in the
United Provinces 1883—1916' in Modern Asian Studies, VII:3 (1973).
165
Report of the Committee on Local Self-Government in Madras, 1882 ( M a d r a s ,
1883), pp. 9-12.
166
J. G. Leonard, 'Urban Government under the Raj: a Case Study of Municipal
Administration in Nineteenth-Century South India' in Modern Asian Studies,
vn:2(i973), 230,233-6.
167
Report of the Committee on Local Self-Government in Madras. 1882, p . 9.

191
 The emergence of provincial politics

superior power' nominated to the councils across this twenty-year

period. The power of nomination lay with the Collector and he could
appoint whomsoever he liked. It seems probable, however, that the
bulk of nominations went to the most prominent local citizens — which
means, more or less, the leading urban magnates. It would make
little sense for the Collector to ignore the men on whom he relied to
restore the peace during times of trouble, to provide arbitration in
social disputes, to come to his assistance during famines and to pro-
vide religious, educational and other facilities, when looking for
agents to undertake the responsibilities of urban government. All
councils were liberally sprinkled with wealthy merchants and land-
lords whose own networks of influence and control were thereby
expanded. In Rajahmundry throughout the 1870s, for example, four
of the five most important councillors were merchants; 168 in Nega-
patam, Muslim Maraciar and Hindu Pillaima merchants were pro-
minent; 169 in Tuticorin, European and Christian Bharatha business-
men; 170 in Kumbakonam, the council was virtually led by a rich
mirasidar, S. A. Saminatha Iyer, who acted as a lawyer and political
agent for many other mirasidars.111
The arenas in which these powers were allowed to operate re-
mained small and informal. Authority centred much more on the
ward councillor than on the council, and his influence was essen-
tially personal. The District Municipalities Act of 1884, however,
which followed Lord Ripon's initiatives in local self-government,
radically changed the face of urban politics. The powers of tax-
ation and administrative interference vested in the councils were
further and greatly increased. More importantly, the organisation
of council authority became much more institutionalised. Under the
Act, the council as a body controlled - within limits - the nature and
level of taxation and of urban administration. The council chairman
was vested with full executive powers. He was responsible for carrying
out the orders of his council and for hearing, or appointing
committees to hear, appeals against council resolutions. In addition,
he appointed most of the municipal staff, which stretched from
sweepers to senior accountants and clerks, drew up the budget and
conducted all correspondence between the council and its super-
visors.172 The focal point of municipal politics switched from a few
168
Leonard, 'Urban Government under the Raj, p. 235.
169
Report of the Committee on Local Self-Government in Madras. 1882, p . 2 0 .
170
G.O. 508 (L and M, M) dated 17 May 1888; G.O. 1016 (L and M, M) dated
13 September 1888. T.N.A.
171
Hindu 26 June 1888; G.O. 1077 (L and M, M) dated 26 September 1888. T.N.A.
172
Madras District Municipalities Act of 1884. T.N.A.

192
 Local structures of political power

individuals to the council itself, and although it was quite usual

for some aspects of administration to be conducted by ward, ward
councillors were under the authority of the council as a whole.
These clauses in themselves would have been bound to create some
political changes but the 1884 Act went further. It envisaged the
gradual replacement of all government officials in the councils
by elected non-officials and provided for the transfer of the crucial
office of chairman to Indian hands. The old informal relationship
between British officials and local magnates was to be broken by
the creation of democratic governmental institutions from which
government officers were excluded.
Although Ripon's municipal legislation was designed to take effect
very quickly, most of the provincial governments found ways of pro-
longing the implementation of its more novel ideas. In Madras, for
example, the extension of electorates, the withdrawal of the bureauc-
racy and the transfer of executive authority to native non-official
chairmen was nearly as slow and sporadic in the municipalities as it
was in the rural boards. It was not until 1912-13 that the majority
of municipal councillors were elected and, until 1920, a considerable
proportion of non-official chairmen were government nominees.173
Between 1885 and 1920 almost no two municipal councils were con-
stituted exactly alike. We are thrown back on general explanations
and 'ideal' types if we are to make any sense of Madras history.
Until recently, it had been widely accepted that municipal govern-
ment in British India was hollow and that most councils were so tight-
ly controlled by the superior agencies of government that they had no
room for independent action.174 However, John Leonard has chal-
lenged this assumption and demonstrated the great importance of
municipalities to the inhabitants under them. 175 The tightness of
superior control has often been overestimated. As local officials were
withdrawn from the administration, their knowledge of council
affairs naturally decreased and there is much to suggest that,
even when they were present in municipal politics, the pressure of
other work prevented them from taking a very active part. Although
the Secretariat possessed enormous powers of interference, it tended
to use them as individual crises arose rather than to conduct permanent
and detailed administration, which would have been very difficult
from its distant perch. Indeed, as the whole purpose of decentral-
173
K. K. Pillai, History of Local Self-Government in the Madras Presidency 1850-1919
(Bombay, 1953), P- 6 6 .
174
F o r example, H . Tinker, The Foundations of Local Self-Government in India,
Pakistan and Burma ( L o n d o n , 1954), p p . 51—60.
175
Leonard, 'Urban Government under the Raj', pp. 227-51.

193
 The emergence of provincial politics

isation was to relieve government of supposed responsibilities, a policy

which centred on regular interference would have been self-defeating.
The councils certainly appeared very attractive political arenas to the
urban population: voting turn-outs were extremely high 176 and muni-
cipal budgets, though not large when compared to the private incomes
of some municipal inhabitants, could provide a living for many em-
ployees.177 More importantly, the council possessed executive powers
which interfered materially in the lives of its subjects. Local inhab-
itants would find their burial grounds closed,178 their trading act-
ivities restricted,179 their houses pulled down,180 their taxes raised,181
their property compulsorily purchased,182 and their places of worship
desecrated.183 They could find roads driven through their front gar-
dens,184 markets in which they had invested heavily prevented from
working,185 and occupations they had long practised outlawed.186
Alternatively, they could be sold cheap land and given building per-
mission,187 receive valuable contracts,188 and have their taxes slashed
to nothing.189 All depended on the action of the council majority.
The general effect of the Ripon reforms was to lead to the slow but
176
Between 1900 and 1920, 60 to 70 per cent of all enfranchised voters went to
the polls every year. In 1914-15, the proportion reached 73.6 per cent. See
Annual Reports on the Working of the District Municipalities Act ( M a d r a s , 1 9 0 0 - 2 0 ) .
177
In 1920—1, there were 73 municipalities in Madras, which had total budgets
of Rs 93,14,610 or an average of Rs 1.27 lakhs each. Although there were
fewer municipalities in Madras than in any other major Indian Province, the
average municipal budget was higher than in Bombay, Bengal and Punjab.
Memorandum on the Working of Representative Institutions in Local Self-Govern-
ment (Calcutta, 1928), pp. 76-9.
178
Madras city, Hindu 11 June 1894.
179
Bellary, G.O. 2120 (L a n d M , M) dated 19 December 1908. T.N.A.
180
E r o d e , G.O. 1250 (L a n d M , M) dated 19 July 1916. T.N.A.
181
Bellary, G.O. 2000 (L a n d M , M) dated 7 December 1892. T.N.A.
182
K u m b a k o n a m , Hindu 21 September 1906.
183
Conjeeveram, Hindu 11 October 1893.
184
M a d u r a , G.O. 411 (L a n d M , M) dated. 20 M a r c h 1900. T.N.A.
185
M a d r a s city, Hindu 5 M a r c h 1907.
186
Such as butchery, G.O. 1101 (L.S.G.) dated 30 April 1924. T.N.A.
187
R a j a h m u n d r y , G.O. 387 (L and M , M) dated 1 M a r c h 1912. T.N.A.; Desamata
31 M a y 1911. R.N.P.
188
M a d u r a , Hindu 5 M a y 1896, 9 April 1907.
189
I n Ellore, M o t h e y Venkataswami, although t h e richest i n h a b i t a n t , never paid
any taxes. G . O . 1011 (L a n d M , M ) dated 17 July 1901. O n e of t h e m a i n concerns
of m e r c h a n t s everywhere was t o gain control of councils in order to lower t h e
rates of taxation, particularly on themselves — as in G u n t u r , G.O. 25 (L a n d M , M)
dated 11 J a n u a r y 1 8 9 3 ; m Vanniyambadi, G.O. 802 (L a n d M , M ) dated 28 April
1904; in M a s u l i p a t a m , G.O. 564 (L a n d M , M) dated 16 April 1891. T.N.A.

194
 Local structures of political power

steady concentration of urban power in one institution and to make

control of the council, for the first time, vital to all urban politics.
The relevance of the new municipality to the educated publicist,
trying but usually failing to establish for himself an influence based
on persuasion rather than the manipulation of hard sanctions, is not
difficult to see. If he could persuade enough voters, on one day, to
support his candidature and to put him onto the council, he had at
his disposal patronage and administrative influence with which to con-
solidate his position and develop a personal political empire apart
from his patrons. Western-educated publicists, if they could generate
enough support at the elections, could at last manipulate levers
of social control. From the 1880s, the towns of Madras witnessed the
growth of organisations designed to gather together grievances and to
put organisers into the municipal council. Ratepayers and Taxpayers
Associations shot up, and, in the larger towns, constituency associ-
ations appeared. Popular cultural and religious movements began to
supply the fuel for agitation and the organisation for caucuses in
municipal politics. Publicists stood at the head of all these. But did
their political tactics work?
At the first election in any newly enfranchised town, certainly,
there was much chaos at the hustings and many of the patterns of
previous government seem to have broken down. Publicists usually
were quite successful.190 Further, the bringing together of the many
and previously discreet channels of political influence in one council
chamber had created the opportunity for the emergence of the
'broker' politician who worked between interest groups without him-
self having to be clearly tied to any. He could use his middleman
position to become central to the working of the council. In Rajah-
mundry, for example, the Brahman lawyer N. Jaladurgaprasada-
rayadu dominated the council in the 1890s: having used agitation and
private legal connection to get the chairmanship, he judiciously mani-
pulated patronage and influence for a decade.191 Equally, inBellary
at the same time, the Brahman lawyer K. Venkoba Rao became chair-
190
As at the first elections in Salem town, where the young Brahman lawyers
W. Viraswami Iyer and C. Vijayaraghavachari, both of whom came from outside,
were able to use the local Komati—Muslim confrontation of 1882 to become
accredited spokesmen of the Hindus and to win election. In the opinion of
the Collector, 'they were persons of no weight in the town'. Report of the
Committee on Local Self-Government in Madras. 1882, p. 22. Also, 'Rajahmundry
experienced several years of political instability following the Ripon reforms.'
Leonard, 'Urban Government under the Raj, p. 238.
191
Leonard, ibid., pp. 238-9.

195
 The emergence of provincial politics

man and built for himself a basis of support by redistributing taxation

and rewards.192
Yet the development of 'new' politicians from the 'new5 style of
urban politics was not nearly as clear as it might at first seem. The
older leadership groups had always employed lawyers and literati
to work for them. By simple extension, they continued to employ,
and to control, them in the municipal institutions. In Ellore under
Mothey raj9 for example,
Pure Bania members will always be few and it is vakils and other intelligent
persons they try to enlist who can expound the law to them and show to them
the way in which their machinations can best be forwarded consistently
with the law or the law evaded if possible.193
A simple change in personnel would not necessarily tell us much
about who controlled municipal politics: western-educated profes-
sionals were not necessarily their own masters. In general, however,
there does not even seem to have been a radical long-term change in
personnel. By the early 1900s, the municipal administrations of towns
as diverse as Cocanada, Vanniyambadi, Madura and Tinnevelly were
filled with merchants and landlords cut on the old magnate die.194
Although there were some western publicists, they were usually tools
to be used by others.
Viewed from another angle, the survival of the magnates is less
surprising. The municipal franchise was pitched at a high level and
relatively few people possessed the vote. The electorate remained suf-
ficiently small to be influenced on a personal basis by direct ties of
debt, tenantry and patronage.195 Popular appeals, based on outrage
or sympathy, could not find a wide enough audience. In Ellore, the
economic power of the Mothey family still gave it control of the coun-
cil. Similarly, in Cocanada in the 1880s, the Pydah Komati family
commanded its debtors to vote at its will and controlled the chair-
192
G.O. 2000 (L and M, M) dated 7 December 1892. T.N.A.
193
A. R. Banerji, Assistant Collector, Ellore, to Collector of Godavari, 29 March
1901 in G.O. i o n (L and M, M) dated 17 July 1901. T.N.A.
194
I n C o c a n a d a in 1894, a11 t h e elected seats were held by m e r c h a n t s , Hindu 2 9
October 1894; f° r t n e mercantile preponderance in Vanniyambadi, see G.O. 802
(L and M, M) dated 28 April 1904; for Madura, G.O. 463 (L and M, M) dated
18 March 1896, G.O. 2135 (L and M, M) dated 11 October 1905; for Tinnevelly,
G.O. 1848 (L and M, M) dated 4 October 1912. T.N.A.
195
It was extremely rare for there to be even 200 voters in any one ward of a
municipality, G.O. 2207 (L and M, M) dated 13 December 1911. See also, note
by Chairman of Mayavaram in G.O. 2512 (L and M, M) dated 17 December 1907.
T.N.A.

196
 Local structures of political power

manship;196 and we have seen the power of the bank cashier at Tuti-
corin and of the rich Nadars of Sivakasi. Most municipalities for
most of the time remained dominated by the local resources -
whether debt, employment, landownership, patronage or terror, or
all five - of a few magnates. At Vizianagram, the Maharaja's nominees
swept every election;197 at Salem, the zamindar's lawyer was chair-
man continuously between 1901 and 1914;198 at Bezwada, the con-
tractor G. Appalaswami controlled three-quarters of the elected
seats for fifteen years and all municipal executives had to come to terms
with him.199 In general, the older style of urban leadership was able
to survive the onslaught of democratic institutions remarkably easily
and to use the new sources of conciliar power to extend its influence.
In sheer economic terms, the municipalities did not handle enough
money to rival the economic networks of wealthy landowners and
merchants. Only in areas where there were no very rich men, or
where they were many and heavily divided, or where great concen-
trations of population depended on non-magnate sources for their
income - as in the few towns in which the offices of the administration,
law and education were the largest employers — could the muni-
cipality become a vital economic centre and men whose primacy was
not based on financial control grasp it for themselves. Otherwise,
access to municipal power was choked by magnate influence.
As may be guessed from the failure of the publicists, the politics of
communal sentiment also promised much but delivered very little
in the municipalities of Madras. Certainly, there were occasional out-
breaks of communal violence at the polls. Equally, there were some
caste constituencies, particularly among merchant communities.
But no interpretation of municipal politics could make caste or
community central. To take the example of the Muslims: how would it
be possible to explain the fact that in 1909 at Vanniyambadi, the Mus-
lim trading capital in the South, Muslims possessed 321 votes out of
a total of 469 (68.44 P e r cent), yet held only two of the six elected seats
and were given two more by nomination (to make 33.3 per cent of
representation)?200 Nobody could doubt that Muslim interests were
196
G.O. 1411 (L and M, M) dated 11 December 1888. T.N.A.
197
Hindu 22 January 1902; Andhra Advocate 18 October 1916. R.N.P.
198
E. S. Ramaswami Iyer, see G.O. 1616 (L and M, M) dated 5 November 1901.
T.N.A.
199
Hindu 12 January 1911; Kistnapatrika 9 June 1911 and 19 July 1912. R.N.P.;
G.O. 306 (L and M, M) dated 15 February 1913; G.O. 2028 (L and M, M) dated
10 December 1919. T.N.A.
200
G.O. 1400 (L and M, M) dated 25 August 1909. T.N.A.

197
 The emergence of provincial politics

dominant in the town, but the Muslims worked through others. On

the other hand, in Madura Muslims had only 78 of 2326 votes but
they held two of the twelve elected seats.201 Here, they worked in
front of Hindus; and by 1917 one of their number, K. M. Alladin
Rowther, was to be returned for the ward containing the leading
Hindu merchants and lawyers.202 While trade rivalries between Hindu
and Muslim merchants and weavers might occasionally break down
cross-communal alliances, much the more usual form of communal
municipal politics was that to be found in Negapatam, Vanniyambadi
and Trichinopoly where Hindu and Muslim prospective candidates
patronised each other's temples and mosques.203 Similarly, as we saw
earlier, even the barrier of untouchability need not be relevant to mun-
icipal politics: in Cuddapah, Kumbakonam and Madura untouch-
able families could and did play a leading part in municipal life.204
Although municipal institutions may have done little to alter the
social base from which urban leaders came, it would be wrong to
underestimate their purely political importance. They brought
together the principal informal leaders of urban society, placed them
next to each other in a single arena and left them to fight it out for pre-
eminence. Whereas previously, the informality of urban government
had kept the magnates apart, now they were necessarily drawn into
conflict. Factional divisions between them were deepened as each
sought to capture the council. Not only was status in question, but also
the considerable powers of council office which gave a man a hold over
the whole rather than part of the town. Whatever the influence of
Mothey Venkataswami or T. S. Sivaswami Odayar before the coming
of the 1884 Act, after it there was scarcely a man in their respective
towns who was not somehow connected to them through the council
administration. In large towns, where several magnates contested
power, the victor could use his control to harass unmercifully the
enterprises of his enemies in ways previously unknown.
The growing bitterness and greater depth of factional warfare can
be seen in the manner in which new styles of politics were used
to conduct old rivalries in the new arenas - styles which owed much to
the energies of dependent publicists. In Bellary, for example, the local
rivalry between the abkari contractor and industrialist, A. Sabhapati
Mudaliar, and the railway contractor, M. Ramanjulu Naidu, boiled
201
Ibid.
202
G.O. 1384 (L and M, M) dated 16 August 1917. T.N.A.
203
G . O . 7 9 8 (L a n d M , M ) dated 20 M a y 1911; G . O . 2099 (L a n d M , M) dated
8 N o v e m b e r 1912; G.O. 1347 (L a n d M , M ) dated 9 August 1917. T.N.A.
204
See above p . 145.

198
 Local structures of political power

up into class conflict when Ramanjulu Naidu backed a lawyer-led

faction of petty bazaarmen, which organised a hartal and finally threw
Sabhapati Mudaliar out of the council office.205 Similarly, in Bezwada,
the Brahman lawyer A. Kaleswara Rao brought the image of Gandhi
to a purely municipal battle and used the tactics of non-co-operation
to give his Komati backers the council.206 Similarly in Kumbakonam,
serious 'popular' municipal agitations broke out when T. S. Sivaswami
Odayar's family found*itself out of power.207 Political activity became
much more overt and the role of the publicist much more prominent.
But usually he was working for one group of urban notables against
another. Municipal politics served largely to promote magnate
rivalries and to broaden the means by which the magnate controlled
the behaviour of his dependents.
As in the case of the district boards also, the reform of municipal
administrations built a series of new connections between Fort St
George and the locality. The Local Self-Government Department in
the Secretariat supervised administration and the construction of
councils and heard appeals against election results. As the extension
of electorates and electoral offices was so uneven, it also possessed
large powers of nomination. Prior to 1884, and, of course, still in
those many municipalities from which local officials were slow to
withdraw, powers of supervision and nomination had been exercised
virtually autonomously by the official on the spot; as we have seen, one
of the principal dangers to central government was always the inde-
pendence of the lower bureaucracy. By assuming supervisory powers
itself, however, the Secretariat strengthened its own hand and had a
much better chance of influencing local affairs. Naturally, this deve-
lopment produced a new dimension in local politics. By gaining the
support of the Secretariat, a local politician could tilt political balances
in his favour. As early as 1895, f° r example N. Subba Rao petitioned
the department about Rajahmundry municipality. Although a Legis-
lative Councillor of considerable standing, Subba Rao had been
squeezed out of local politics by an enemy faction. He asked that the
ratio of elected to nominated seats be altered in favour of the nomi-
nated, thereby weakening his opponent's electoral strength. 208 As a
205
Hindu 8 M a r c h 1893; G 0 - 2 0 0 0 ( L a n d M > M ) dated 7 December 1892;
G.O. 972 (L and M , M) dated 4 June 1892; G.O. 1625 (L and M , M) dated 27
November 1893. T.N.A.
206
Kaleswara Rao, Na Jivita Katha - Navya Andhramu, pp. 291-360 (Telugu).
207
G.O. 300 (Public) dated 14 M a y 1921. T.N.A.
208
G.O. 1029 (L and M , M) dated 27 June 1895; G.O. 5 (L and M , M) dated
7 January 1896. T.N.A.

199
 The emergence of provincial politics

Legislative Councillor, it was natural that his influence at the centre

should prevail and his petition was granted. The Secretariat, part-
icularly after about 1910, received more and more such requests,
with some of which, at least, it complied.
Equally, the disfavour of the Secretariat could have a shattering
impact on an urban situation. In Kumbakonam, for example, the
empire of T. S. Sivaswami Odayar withstood all local challenges for
fifteen years. Sivaswami Odayar's greatest enemy was the Brahman
lawyer and landlord N. Krishnaswami Iyengar who worked in front
of a coalition of interests suffering at Sivaswami Odayar's hands.209
But by 1906 Krishnaswami Iyengar could show little for his endea-
vours. In that year, however, Sivaswami Odayar's conduct became too
outrageous and the Secretariat removed him and nominated Krishna-
swami Iyengar in his place. The new chairman rapidly dismantled the
contract and favour machine of his predecessor, redistributed the
patronage and won security in the council. He launched an of-
fensive against his rival's landed base and stood for election to the
local temple committee which had been in Sivaswami Odayar's
pocket.210 From Ganjam to Malabar, the eyes of previously auto-
nomous politicians had to turn increasingly to Fort St George.

Madras city
The affairs of one particular town demand a more detailed examin-
ation both because they again demonstrate the political importance
of administrative change and because specific events which took
place within the council chamber were of vital importance in the
wider political world. Madras city was the centre of the British
raj in South India. It had grown up around the East India Company's
Fort St George which had been founded in 1612. It contained the
Governor's Residence, the Legislative Council, the Secretariat, the
High Court and the University - in sum, the heart of the administra-
tion — and the head offices of many British commercial agency houses
which carried on business in the province. It was the presidency's
leading port and, from the 1890s, possessed the largest industrial
complex in the South. Although it drew the bulk of its native popula-
tion from its immediate environs, it contained conspicuous minorities
from a variety of sources: the governmental and educational institu-
tions attracted men from every district in the presidency, while
209
Collector of Tanjore to Secretary, Local a n d Minicipal D e p a r t m e n t , 17 M a r c h
1903 in G.O. 528 (L and M , M) dated 6 April 1903. T.N.A.
210
Hindu 7 a n d 21 September 1906, 8 February and 24 M a r c h 1915.

200
 Local structures of political power

Armenians, Gujeratis and Persian Muslims jostled with more locally

based Beri Chetties, Komatis, Vellalas, Brahmans and Kammavars in
its commerce.211
The Madras Municipal Corporation was the oldest in India, having
received its first charter from James II. For two centuries, its members
were nominated by government, its budget remained minute and, as
an institution, it stagnated. Under the general pressure of reform in
the later nineteenth century, however, it began to swell in importance.
In 1878, elections were introduced for the first time, and half of the
30 members were elected on an extremely high franchise.212 By the
Madras Corporation Act of 1884, the number of elected commis-
sioners rose to 24 out of 36 and the franchise was lowered to include
about 4000 people or 1 per cent of the total population.213 By a further
Act of 1904, elected representation was reduced to 20 of 36 mem-
bers,214 but amendments to the franchise laws, and the price rise
which reduced the effectiveness of restrictions based on taxation,
steadily increased the number of voters to about 11,000 or i\ per cent
of the population by the time of the First World War.215 Much more
dramatic was the increase in the income and expenditure and the use
of the administrative powers of the Corporation. Between 1880 and
1918, the budget rose from Rs 7.5 lakhs per annum to a staggering
Rs 34.8 lakhs™ and the Corporation became involved in building
new markets, laying out a completely new drainage scheme, licensing
and constructing electric tramways, altering the water-supply and, of
course, building an enormous palace for itself. The constitution of
the Corporation was somewhat different to that of the district
municipalities. The executive head was a president, nominated by
government and a member of the I.C.S. throughout our period. He
was supported by two executive vice-presidents, also officials, but
elected by the commissioners. To help them in their deliberations,
there was a Standing Committee, elected out of the commissioners
every year. Naturally, this body monopolised policy-making and pre-
sented the other commissioners with a series of faits accomplis which
were seldom challenged.
211
I n 1921, t h e city h a d about 450,000 inhabitants. Of these, 10,000 came from t h e
Malayali-speaking west and 14,000 from t h e N o r t h e r n Circars. Census of India.
1921. Madras. Volume XIII. Part 2 (Madras, 1922), p p . 8 5 - 9 4 , 1 1 8 - 2 3 .
212
Administration Report of the Madras Municipality for 1879, pp. 1, 34—6.
213
Ibid., 1884-5, p . 1.
214
Ibid., 1904-5, p. 3.
215
Ibid., 1915-16, p. 1.
216
Ibid., 1880 a n d 1918-19.

201
 The emergence of provincial politics

The commissioners themselves were drawn from a variety of back-

grounds and represented the many interests of the capital city. Most
prominent, of course, were native merchants and the dubashes of
European agency houses who wielded great financial power in the
town. Pitti Thyagaraja Chetty was typical of these - moneylender,
capitalist weaver, hide and skin merchant and joint-stock banker.217
The Corporation also contained men whose social position was de-
pendent on the ownership of land, such as the 'Gogai' Kammavar
family, hereditary shrotriemdars of Chintadripet which was in the
heart of the city.218 As we mentioned before, however, no real dis-
tinction can be made between landed and mercantile wealth in urban
politics: many of the merchants and dubashes were extensive pro-
perty owners219 while the 'Gogai' family supplied the city with one
of its most successful dubashes and moneylenders, B. Chitti Babu
Naidu, who was a Corporation member continuously from 1905 to
the late 1920s. Through their extensive patronage of a wide range of
social activities, as well as through the influence their economic em-
pires gave them, these financial magnates dominated local affairs in
Madras. 220 They were seldom opposed during elections and were
only seriously pressed when faced with rivals of the same character
as themselves.
Alongside these individuals existed a number of political organisa-
tions less personally based. In a few parts of the city, there were de-
finable caste constituencies which operated as caucuses at election
time. At Royapuram, for example, the Kurukula Vanisha community
lived tightly clustered around St Peter's, their church. In spite of
economic differentiation among them, their social isolation as Chris-
tians and continual conflicts with the local Catholic Bishop over the
patronage and endowments of their church helped to keep them
217
Other 'city fathers' of this type, who filled the Corporation seats between
1878 and 1920, were Raja Sir Savalai Ramaswami Mudaliar, dubash to
Arbuthnot and Co. and the greatest public patron of his day; Gopathi Mahadeva
Chetty and his son, Narayanaswami, who were moneylenders and merchants;
P. Somasundram Chetty, related to the Gopathi and also moneylender and
merchant; the Calivalla brothers, Cunnan and Ramanjulu Chetty, who were
Komati merchants and proprietors of Messrs King and Co.; and Chathubhujar
Doss Govindoss and Lodd Govindoss who were Gujerati Vaishya bankers.
218
See obituary of G. Vurthia Naidu in Hindu 27 June 1910.
219
Such as Pitti Thyagaraja Chetty, t h e Gopathi family a n d Raja Sir Savalai
R a m a s w a m i Mudaliar.
220
T o give a n idea of t h e range of these, Raja Sir Savalai R a m a s w a m i Mudaliar
contributed t o hospitals, university hostels, horse troughs, statues, A n n a d a n a
Samajam, T a m i l a n d Sanskrit revivalism and local religious societies.

202
 Local structures of political power

unified.221 They returned their lawyer—headman, R. N. Arogiyas-

wami Mudaliar, to the Corporation at every election between the
early 1880s and his death in 1902. Another obvious caste representa-
tive was C. V. Cunniah Chetty, headman of the Komatis, who won a
seat in the 1890s and early 1900s.
As we saw when discussing the urban political structure, however,
there is considerable ambiguity in the roles played by many of these
caste leaders. Both R. N. Arogiyaswami Mudaliar and C. V. Cunniah
Chetty controlled communal institutions of great wealth, which
gave a sharp edge to their authority. Cunniah Chetty, besides large
family properties, managed the Kanyaki Paremaswari temple of the
Komatis, which owned the largest bazaar in the city.222 This asset
gave him influence over not only the Komatis but a large section of
the mercantile population in general. Moreover, within his constitu-
ency, the existence of these great institutional funds provoked faction-
al splits. From the 1890s a ginger-group of Komatis, led by the lawyer
Salla Guruswami Chetty and backed by some rich merchants, began
to form caste associations in order to undermine Cunniah Chetty's
authority.223 On the latter's death in 1912, they attempted to oust his
son from the dharmakartha-shrp of the temple and to gain control of
it and its properties for themselves.224 The economic power of the
temple meant that this cleavage was of significance not only within the
community but also in wider city affairs. S. Guruswami Chetty leaned
towards one faction in the Corporation, which, after about 1910,
was associated with High Court lawyers such as C. P. Ramaswami
Iyer and which, as we shall see, presented a real threat to the interests
of a number of financial magnates. Cunniah Chetty was allied to these
very magnates who had helped to obtain for him his Corporation
seat in 1888.225 Caste ties overlapped so heavily with other ties and
support from outsiders was so vital in winning the support of even
caste voters, that the concept of the caste constituency ceases to be
self-explanatory. The influence of cross-communally organised power
was often essential in providing the appearance of a solid caste con-
stituency.
We can see this problem very clearly in the classic Black Town
election of 1910. A group of Thondamandala Vellala families, working
221
Hindu 28 June 1907.
222
Hindu 27 M a r c h 1913.
223
Hindu 13 December 1904.
224
Hindu 27 M a r c h 1913.
225
Hindu 22 June 1888.

203
 The emergence of provincial politics

extensively in lower government service and in the professions,

dominated one of the Black Town wards by possessing about two-
thirds of the votes in it. For many years, they returned to the Cor-
poration one of their caste leaders, P. M. Sivagnana Mudaliar, who
was trustee of their temple and president of their high school. On
the Corporation, Sivagnana Mudaliar was allied to P. Thyagaraja
Chetty. In 1910, Dr T. M. Nair, one of Thyagaraja Chetty's greatest
enemies, persuaded Dr M. C. Nanjundan Rao to stand against
Sivagnana Mudaliar in the Thondamandala ward. Nanjundan Rao,
of course, was not a Thondamandala but, as a member of the Gaekwar
of Baroda's family, he was extremely wealthy, with financial ties in
various parts of the city, and, as a leading spokesman of the educated
professionals, he had influence with senior officials. When he brought
these extraneous connections to bear on the Thondamandala ward,
he all but destroyed Sivagnana Mudaliar5s caste base. After a bitter
election, in which Sivagnana Mudaliar had mobilised his allies
Thyagaraja Chetty and K. C. Desikachari to put their influence be-
hind him and had run the most openly corrupt campaign that the
city had yet seen, the caste leader scrambled home by only a handful
of votes.226
In addition to those elected, the government nominated members
of interest groups which could not find electoral support and pro-
fessionals and government servants who would help push through
unpopular measures - particularly of increased taxation. Among the
former nominees were European businessmen and Muslims, most of
whom were financial magnates in their own right and some of whom
did, in fact, win elections when it became necessary.227 The latter
consisted of a mixture of old and tried faithfuls who were nearing,
or past, retirement after long careers, and active government servants
who could link wider government policies to the Corporation.
Politics within the Corporation, at least until about 1910, are
extremely difficult to describe. There were no parties in the sense of
a government and an opposition but rather a series of interest groups
gathered around the bureaucracy. The main concern of the commis-
226
G . O . 261 (L a n d M , M) dated 15 F e b r u a r y 1911. T.N.A.; Hindu 2 6 , 29 a n d 31
December 1910.
227
In spite of the overwhelmingly Hindu electorate, Europeans and Muslims
who were personally powerful could win elections if they tried. Eardley Norton,
a barrister, held an elected seat from 1885 to 1893 a n ^ several members of the
Muslim mercantile elite also were returned. See, for example, Hindu 25 July
1885.

204
 Local structures of political power

sioners was the effective distribution of taxation, services and

contracts between themselves and their constituents; the Corporation
was not a body which had powers to initiate policy of a controversial
or ideological nature, and the character of the commissioners suggests
that had it possessed them it would not have used them. Divisions
were essentially factional and tied to specific issues such as the award-
ing of a particular contract to a particular man, or the construction
of a sewer in one part of the city rather than another. The personnel
of the factions, naturally, changed with the issue, although handfuls
of confederates might form small cliques which voted together. In
spite of the outwardly democratic appearance of the Corporation, it
was heavily weighted with bureaucrats who, in collaboration with
the Standing Committee, conducted its executive affairs. As else-
where in the Madras bureaucracy, this led to 'private' government
between officials and various local powers, regardless of rules and
laws. For example, although it was specifically prohibited for com-
missioners to have business dealings with the Corporation, through-
out our period European business houses acted as contractors and
loan brokers to the Corporation while their directors sat as commis-
sioners.228 Again, although it was prohibited for contractors to hold
materials, carriage and labour contracts at the same time, in 1890
Etiyalwar Naidu, who was rumoured to be the sowcar of the president,
Colonel Moore, held all three. 229 Commissioners used their position
to place relatives and friends in Corporation jobs and to manipulate
taxation to their best advantage.
Although civilian Corporation presidents came and departed, dur-
ing the later nineteenth century one vice-president, the revenue
officer Pulicat Ramaswami Chetty, went on forever. He held his post
from 1869 until his retirement in 1896 and continued to exert a
strong influence over local affairs through a variety of honorary
offices until his death in 1912. Corporation taxation during his long
reign was known to be a scandal as the effective rates depended less
on the amount of property held than on his favour. The nearest
Ramaswami Chetty's empire came to collapse was in the mid-1890s
when he refused to lower the assessment on some houses owned by
the commissioner G. Varadappa Naidu, who was an enemy of some of
Ramaswami's friends. Varadappa Naidu attempted to give the state of
the taxation register a public airing but was quickly put down by
228
See reports of Legislative Council debates o n this malpractice in Hindu 11
February and 11 a n d 13 M a r c h 1919.
229
Hindu 7 and 8 March and 24 April 1890.

205
 The emergence of provincial politics

other commissioners who had their own interests at stake.230 Rama-

swami Chetty, although a government servant, was very much a local
man. His Beri Chetty community was prominent in the commerce of
the city and he represented a family network of interest within it.
His sons and in-laws could be found in most of the important Euro-
pean companies, in the legal profession and even in the Madras
customs house. 231 Ramaswami Chetty used the Corporation as though
it was a family business - gaining part of the Electric Tramways
contract for one son,232 avoiding taxation on his properties233 and
employing large numbers of relatives and caste-fellows in his depart-
ment. 234
In so far as the endless shifting of faction on the Corporation
developed any coherent shapes, it did so as interested parties ap-
proached particular bureaucrats. Ramaswami Chetty, for example,
was close to his caste-fellow Pitti Thyagaraja Chetty and to R. N.
Arogiyaswami Mudaliar, who advocated his appointment as president
during Colonel Moore's absence in England in 1896 and his retention
in office beyond retiring age. 235 Ramaswami Chetty's other connec-
tions included the merchant-princes P. Somasundram and G. Maha-
deva Chetty, who also were his caste-fellows: in 1892 they petitioned
against a revision of the house-tax register but, from obvious know-
ledge 'were prepared to admit that a good number of houses might
have escaped taxation, and that some houses might have been assessed
at a very low figure.'236 The Health Officer, Jesudesan Pillai, at this
time had his party too — headed by the great lawyer and publicist
P. Ananda Charlu — and Colonel Moore, the head of the executive,
was not without connections among the commissioners. In return for
support and protection in the Corporation chamber, the official
arranged jobs and contracts to the satisfaction of his clients. In the
decade or so after Ramaswami Chetty's retirement, and under suc-
ceeding presidents, little changed as officials and their friends con-
230
Hindu 19 December 1896; G.O. 823 (L-and M , M) dated 13 M a y 1896; G.O. 913
(L a n d M , M) dated 1 J u n e 1896. T.N.A.
231
A Brief Life Sketch of M. R. Ry. P. Subramanyam (Madras, n.d.), pp. 1 - 2 .
232
Ibid., pp. 4-5.
233
Hindu 2 0 October 1896.
234
H e admitted employing six of his relatives and fifteen other caste connections in
his office. G.O. 9 1 3 (L a n d M , M) dated 1 J u n e 1896. T.N.A.
235
Hindu 9 J u n e 1892 a n d 16 J u n e 1896; G.O. 8 2 3 (L a n d M , M ) dated 13 M a y
1896; G.O. 1104 (L a n d M , M) dated 3 July 1896. T.N.A.
236
T h e revision h a d been forced o n t h e reluctant R a m a s w a m i by a G o v e r n m e n t
Order. G.O. 1274 (L and M, M) dated 1 August 1892. T.N.A.

206
 Local structures of political power

tinued to run the Corporation between themselves. In 1902, the

Hindu commented bitterly: T h e few honest and really useful men are
utterly powerless against the self-constituted Whips who manage all
business beforehand.'237
Until the 1910s, local political organisation in Madras was similar
to that which we would expect to find in any district municipality
before the Ripon local self-government Acts: effective decision-mak-
ing lay with semi-autonomous bureaucrats who worked closely with
and relied on the informal power of local economic magnates. Madras
city, however, was also the administrative capital of the presidency
and it possessed a profusion of western-educated literati. Equally, it
was a major economic centre with relatively more opportunities than
in other towns for 'new' magnates to arise or to come in from outside.
These two factors gave the public life of Madras city much greater
activity than elsewhere. From the 1860s, religious and cultural revival
associations, social reform sabhas and proto-nationalist political clubs
appeared in great numbers as western-educated publicists attempted
to create constituencies for themselves and as magnates struggled
against each other for status and public esteem. Given the relatively
large number of literates in the city and of people who could be
reached through or become involved in street propaganda, Madras
city was the one town in the presidency where we could expect the
publicist to be able to establish himself as an independent power. By
the 1880s, such archetypal publicists as A. C. Parthasarathi Naidu,
newspaper editor, social reform advocate, organiser of a string of
religious revival sabhas and occasional swami, as T. Venkatasubba
Iyer, lawyer and religious teacher, and as S. Guruswami Chetty, caste
reformer, religious devotee and Congressman, were making their
presence felt in several arenas of city government and were members
of the Corporation.
However, if success on the Corporation is used as an acid test, it is
clear that even in Madras city the publicist had to have a magnate
patron, for he could not hope to build on the substance of popular
opinion. In 1904, for example, A. C. Parthasarathi Naidu had the
temerity to advocate an increase in municipal taxation in order to
improve social services, and was severely beaten at the polls.238 A
few years later, T. Venkatasubba Iyer attacked the richest contractor
in the city, A. Subramania Iyer. He was beaten by 217 votes to 3 in a
237
Hindu 17 J a n u a r y 1902.
238
Hindu 3 September 1907.

207
 The emergence of provincial politics

constituency in which, two years before, he had won comfortably.239

S. Guruswami Chetty lost all hope of a seat when S. Venkatachellam
Chetty, a fellow Komati and rich merchant, stood against him. 240
In spite of the advantages for public politics, which the capital pos-
sessed, no publicist succeeded in challenging magnate power at the
hustings.
But the hustings were not the only place from which influence over
the Corporation could be exercised. The Corporation was more
heavily bureaucratic than any district municipality and Madras city
was the centre of the bureaucracy. Publicists - lawyers and journalists
by profession - were drawn from the same state-level social groups
(often from the same families) as the Indian bureaucrats who moved
in the higher echelons of government around the Secretariat, the
High Court and the University. Many of them had obvious connec-
tions of sympathy and interest with men who enjoyed senior positions
of power in the administration. It was possible, in theory at least, for
them to capture the Corporation by influencing the decisions of the
superiors of the Corporation bureaucracy. This would give them a
hold over the distribution of Corporation patronage and make them
indispensable to both magnates and Corporation bureaucrats
alike.
Before about 1910, however, the extreme localisation of political
authority in Madras, which we have seen everywhere, served to keep
the affairs of the Corporation and those of the Secretariat apart. There
was little machinery through which senior administrators could
interfere in the local politics of the city. Moreover, there were as yet
few Indians in the positions from which interference was possible,
and European civilians, who were suspicious of the very movements
and campaigns which gave publicists prominence, were not as useful.
The Corporation bureaucracy, as characterised by P. Ramaswami
Chetty, was typical of lower officialdom anywhere in the mofussil.
It was orientated entirely towards local concerns, being appointed
in the locality and operated through lucrative connections with local
magnates. Without direct power to discipline and reward, all that the
publicists could expect of their Secretariat, High Court and University
contacts were offers to Corporation bureaucrats of career help and
prestige. These meant nothing to a man like Ramaswami Chetty, who
wanted only to stay where he was in his locality. During the 1880s
239
A. S u b r a m a n i a Iyer, w h o w o r k e d w i t h various E u r o p e a n c o m p a n i e s , held
Corporation contracts w o r t h Rs 10 lakhs in t h e 1910s. Hindu 6 December 1910;
Hindu 6 J a n u a r y 1911 and 24 F e b r u a r y 1914.
240
Hindu 5 N o v e m b e r 1919.

208
 Local structures of political power

and 1890s, even publicists belonging to the two most powerful

families in the higher administration were unable to carve a place
for themselves in the Corporation.241 We can see their problems
clearly in the way that another leading publicist, P. Ananda Charlu,
was broken on the rack of Corporation intrigue.
P. Ananda Charlu was a prominent city vakil who sat for the
Triplicane constituency throughout the 1880s. He organised local
petitions, channelling the grievances of Hindu merchants against
European missionaries and Muslims,242 and was instrumental in the
organisation of the Madras Mahajana Sabha and, later, the Madras
branch of the Indian National Congress. He was seldom out of the
press or off the public platform, and numbered among his clients,
both legal and more obviously political, many of the wealthiest mer-
chants of the city, various zamindars and the sole dharmakartha of the
Triplicane temple. In all these activities, however, his role was that
of representative. But once inside the Corporation, he could hope to
develop an influence with the administration which would make him
more. He became closely linked with the Health Officer, M. Jesudesan
Pillai, who was the second Corporation vice-president and ran a
parallel rival network to that of P. Ramaswami Chetty. By protect-
ing Jesudesan Pillai from attacks in the Corporation chamber and by
supporting his administration, Ananda Charlu could reasonably
expect to influence the character of Jesudesan's government. In
1891, however, an epidemic which swept the city was traced to the
Health Officer's maladministration: his subordinates were shown to
be patronage appointments who had no qualification for their jobs
and the contractors he employed had seldom completed their tasks.
In the ensuing uproar, Ramaswami Chetty moved in for the kill and
his men, P. Thyagaraja Chetty and R. N. Arogiyaswami Mudaliar,
savaged the performance of the health department. Ananda Charlu
reacted by removing the relevant papers and sitting on them for more
241
In the next chapter, we shall be discussing the Vembakkam family and 'C. P.
Ramaswami Iyer's community', who dominated the world of the western-
educated. They represented the most powerful of Indian bureaucratic interests.
However, in 1885 the lawyer Vembakkam Krishnamachari upset his magnate
supporters by espousing the cause of increased taxation, and was beaten repeatedly
at the polls although he had been sitting on the Corporation for the five
previous years. Hindu 25 and 27 May, 1 and 3 June and 25 and 27 July 1885; at
about the same time, C. V. Sundara Sastri, who was C. P. Ramaswami Iyer's
father-in-law and a prominent vakil, suffered a similar fate. Although he had
once been a member of the powerful Standing Committee, he was beaten in no
fewer than seven consecutive elections. Hindu 4 and 6 June 1888.
242
Hindu 27 August 1884, 17 and 18 May 1889.

209
 The emergence of provincial politics

than a year.243 But in spite of his excellent connections with members

of senior government departments and the University, he could do
nothing to ease the pressure on Jesudesan Pillai. For two years, the
Ramaswami Chetty faction pounded away,244 and the affair reached
a crisis in 1893 when Ananda Charlu stood for the Corporation seat
in the new Legislative Council. With Thyagaraja Chetty leading
the opposition to him, the Corporation was deadlocked for five
meetings. Finally, it agreed to elect P. Rungiah Naidu, a close friend
and Congress ally of Ananda Charlu, but a man not involved in
the scandal.245 Ananda Charlu withdrew from city politics and,
through the influence of his high-placed friends, was elected to the
Imperial Legislative Council — the equivalent of being kicked up-
stairs. The Thyagaraja—Ramaswami Chetty combination of local
magnate and bureaucrat had resisted the challenge of the man sup-
ported from above.
Fifteen years after Ananda Charlu's eclipse, a series of important
changes altered the structure of Corporation politics. Although
before 1920 there was no reconstruction of the constitution, institu-
tional pressures made the Corporation much more central to the
general affairs of the city, and so changed the relationships between
it and the city populace and the senior offices of government. The
Corporation began to undertake new and large-scale public works
programmes, financed by public loans and government grants. Its
administrative competence was increased so that it touched new
areas of city life. It was given a role in the administration of several
independently constituted charitable trusts; for example, it nomin-
ated members to the board of Pachayappa's charities which spent more
that Rs 1 lakh a year on education and religion.246 Obviously, these
developments began to create new political interest groups which
needed representation on the Corporation. Publicists naturally came
forward to provide this representation. Moreover, publicists and
politicians whose ambitions lay beyond the locality and in the higher
councils of government also were made to look again at the Corpora-
tion. After the Morley-Minto reforms, the Legislative Council be-
came a crucial arena of provincial politics, in which the government
sought to test the opinions of the province on a wide range of matters.
The Corporation was the smallest and most manageable electorate to
243
Hindu 1, 2 2 and 25 August, 28 November and 5 December 1893.
244
T h e y succeeded in voting Ananda Charlu off the Standing Committee in 1893.
Hindu 28 November and 19 December 1893.
245
Hindu 10 May 1893.
246
M. Tiruvenkataswami (ed.), Pachaiyappa's College, Madras. Centenary Com-
memoration Book. 1842—1942 (Madras, 1942), p. 85.

210
 Local structures of political power

this Council. Further, several touring commissions came to Madras

in the early 1910s to solicit information on the state of the public
services, on the scope of decentralisation, on the development of
industry and on many other matters. Politicians who wished to exer-
cise a provincial influence had to give evidence before these, yet the
Madras Government denied access to them to representatives of
informal associations such as the Congress and the Mahajana Sabha.
By becoming a Corporation commissioner, however, a politician
was well placed to have his case heard, for the Corporation, as a body,
frequently was invited to provide witnesses.
From about 1910, then, western-educated publicists became in-
creasingly eager to enter the Corporation. And changes in the attitude
of the Secretariat to the Corporation enabled them to play a fuller part
in its affairs. To match the growing power of the Corporation, the
government began to insist on a much higher standard of efficiency,
which it sought to obtain by raising the standard of the Corporation
bureaucracy. In 1908, T. Raghavayya was appointed to the revenue
office and broke the succession of Ramaswami Chetty-like executives.
Raghavayya had no local interests and his ambitions stretched out-
side the Corporation. He had joined service as a centrally-appointed
deputy collector under the new recruitment policy, and he was to
become a Collector, a Departmental Secretary and a knight. He was
a bureaucrat whose personal and career contacts lay not with the
commercial barons of the city but with Secretariat officials, High
Court Judges and University Senators. His first actions in the Cor-
poration were to tighten up on the collection of the property tax,
thereby adding Rs 80,000 a year to the treasury, and to reorganise
his staff to prevent bribery and the employment of dependents of the
commissioners.247 He was thoroughly disliked by many of the old
commissioners, particularly Pitti Thyagaraja Chetty who tried to
cut his salary and to replace him with a less highly ranked officer.248
Raghavayya was followed into office by T. Vijayaraghavachari and
Mahommed Bazlulla who were of the same stamp as himself. Publicists
who were well connected to the senior Indian bureaucrats and educa-
tionists who were the colleagues and masters of Raghavayya,
Vijayaraghavachari and Bazlulla, now came into their own. They
could influence Corporation policy and the distribution of patronage
in ways not available to the local magnate.
To see best what this meant, we can contrast the career of T. M.
Nair, a publicist without 'pull' in the right quarters, with those of
247
G.O. 838 (L and M , M) dated 24 M a y 1908; G.O. 1362 (L and M , M) dated
21 August 1909. T.N.A.; Hindu 17 November 1909.
248
Hindu 22 October 1913.

211
 The emergence of provincial politics

C. P. Ramaswami Iyer and G. A. Natesan, publicists who had it in

abundant quantities. T. M. Nair had created a great reputation for
himself in Triplicane, a communally heterogeneous and volatile area
of the city.249 He led a series of agitations against Corporation policies,
attacked the corruption of many of the old commissioners and de-
clared war on Thyagaraja Chetty whom he denounced for various
tax, property and contract manipulations.250 He was unquestionably
the most powerful stump orator in the city and, between 1904 and
1913, was returned regularly to the Corporation. If ever there were
a politician on the Corporation who had a 'popular' base, it was T.
M. Nair. But he had few contacts with men in the higher institutions
of government. His hatred of the Hindu religion and all things Indian
made him repulsive to other leading members of the professions and
bureaucracy, many of whom were deeply interested in the revival of
Hinduism. In 1910, when he stood for the Imperial Legislative Coun-
cil, he was heavily defeated and blamed his fate on 'the wire-pullers',
particularly V. Krishnaswami Iyer, a member of the Governor's
Executive Council, w7ho had kept him out.251 In 1913, by attacking
the free supply of water to the Sri Parthasarathi temple, Nair managed
the remarkable feat of irritating both P. Thyagaraja Chetty and the
ex-High Court Judge Sir S. Subramania Iyer to such an extent that,
though opposites in all their political and ideological stances, they
joined together to defeat him in a Corporation election.252 Thereafter,
Nair's presence in the Corporation was due to nomination by his few
British friends. Like the dependent publicists of the 1880s, he could
be chopped down without difficulty.
Very different, however, was the impact on the Corporation made
by the arrival of C. P. Ramaswami Iyer and G. A. Natesan. C. P.
Ramaswami Iyer, one of the leading vakils of his day, was connected,
by relationship and intimate friendship, to many of the highest
government servants in the presidency and possessed sufficient wealth
to be an independent patron in his own right; while G. A. Natesan, a
rich publisher, was one of V. Krishnaswami Iyer's right-hand men.
Their attack on the old bases of Corporation power struck home
where Nair's had proved ineffective. Natesan denounced a list of
commissioners, including Thyagaraja Chetty, P. M. Sivagnana
Mudaliar and about a dozen others, for filling Corporation offices
249
Triplicane contained Muslims, around t h e old palace of t h e N a w a b of Arcot,
devout H i n d u s , around t h e Triplicane temple, and young students, around
the University.
250
Hindu 3 November 1909, 27 February and 19 M a r c h 1913, 10 December 1910.
251
Hindu 5 J a n u a r y 1910.
252
Hindu 2 2 , 23 a n d 24 April 1912, 15 September 1913.

212
 Local structures of political power

with their relatives.253 Thyagaraja Chetty found himself arraigned in

battle against them as they sought to alter the pattern of contract
distribution: he defended against their attacks men who failed to
complete contract work, subordinate staff caught in bribery and the
Electric Tramway Company, in which he was substantially interested
and whose track rent C. P. Ramaswami Iyer wished to raise.254 All
this was to no avail. He was equally impotent to prevent a revision
of the house-tax register and, by 1919, could not even persuade the
Corporation to build, at public expense, a road to the houses of
some of his friends.255 The takeover in the Corporation can be seen
in the contrast between the Standing Committee in 1910 and in 1915.
In 1910, the elected members were P. Thyagaraja Chetty, his allies
P. M. Sivagnana Mudaliar and K. C. Desikachari, the dependent
publicist A. C. Parthasarthi Naidu and T. M. Nair; in 1915 they
were C. P. Ramaswami Iyer, G. A. Natesan, the Brahman Congress-
man K. N. Ayah Iyer and the Muslim publicist Yakab Hasan.256
Thyagaraja Chetty and his allies, who had run the Corporation for
a quarter of a century, were under severe pressure. J. C. Molony,
President of the Corporation between 1914 and 1920, has recorded
their reaction to the change:
Pitti Theagaraya Chetty was the Father of the Madras Corporation, the
Nestor among his Fellow Councillors. At the time I became President he
led a bloc known as 'the northern Councillors': his and their chief
function in life was the criticism of the Corporation executive and
opposition to all innovation.257
G. A. Natesan and C. P. Ramaswami Iyer had, quite simply, cap-
tured the Corporation bureaucracy, which was a very different beast
from that of the 1880s. In spite of the play they made with previous
Corporation corruption, their rule was scarcely different, except
that they were the arbiters of it. C. P. Ramaswami Iyer diverted public
advertisements into the Congress and Home Rule press, in which he
was deeply interested, and he was accused of stacking the Corporation
bureaucracy with his own appointees.258 In 1920, G. A. Natesan
253
Hindu 24 April 1912.
254
Hindu 2 April 1910, 20 J a n u a r y 1915, 24 M a r c h 1915, 22 September 1916.
255
Hindu 17 April 1918 a n d 22 September 1919.
256
Administration Report of the Madras Municipality for the year 7 9 / 0 - 7 7 ; ibid.,
1915 -16.
257
J. C. Molony, A Book of South India (London, 1926), p. 154.
258
G.O. 175 (L and M , M) dated 7 February 1918; A. C. Parthasarathi Naidu to
A. Cardew, 16 September 1 9 1 6 m G.O. 414 (Home, Misc.) dated 26 April 1917.
TMA.

213
 The emergence of provincial politics

happily admitted touring his constituency with the acting Revenue

Officer who was his protege, making revisions of the property tax
while he was canvassing votes and using a revised register which
had not been made available to other candidates. Applications for
house-tax revision were sent to him rather than the revenue depart-
ment.259 This control of the bureaucracy gave Natesan and Ramas-
wami Iyer power to act as the brokers between the Corporation and
outside interests.
From 1916, Thyagaraja Chetty was to raise the anti-Brahman
cry against his enemies and to take his case into provincial politics
where, of course, it now belonged. His anti-Brahmanism, however,
was more political than racial: he himself was associated with
Brahmans, such as K. C. Desikachari (an in-law of Kasturi Ranga
Iyengar, the editor of the Hindu and an opponent of the C. P.
Ramaswami Iyer clique), while C. P. Ramaswami Iyer distributed
Corporation favours to such solid non-Brahmans as the Calivalla
brothers, who were Komati millionaires, and Lodd Govindoss, the
Gujerati banker. Ramaswami Iyer and Natesan needed magnate
supporters but, unlike the earlier breed of publicist, they had enough
currency to choose whom they wanted. Thyagaraja Chetty was losing
a battle which had been started not by communal aggression but by
institutional change.
In the Corporation, the nature of this change led to something of
a transformation in the kind of men who were politically powerful.
This was because the penetration of the locality by the centre of
government was more direct than in the mofussil. The continued
bureaucratic hold on the Corporation meant that changes in the struc-
ture of the bureaucracy itself made an immediate impact on the
personnel of local politics; in the mofussil, bureaucratic centralisation
was mediated by the growth of non-official office and elected power.
However, the general process of change, which was both broadening
the bases of local competition and tying the politics of the locality to
those of the centre, was similar to that which we have seen in the
temples, the district municipalities and the rural boards. What was
taking place in Madras was the gradual emergence of a structure
of provincial politics, in which political power in each district and
town was influenced for the first time by events and personalities at
the capital. Local politicians now had to take account of the shape of
politics in the higher administrative institutions at the 'centre'.
259
Hindu 24 March 1920.

214
 5
The emergence of provincial politics

The growing presence of the capital in the political affairs of the

mofussil may be seen as the major factor of political change in the
period which we are examining. Thus far, however, we have viewed
this development in detail only from the position of the mofussil; we
have seen how local politicians were brought into contact with the
administrative and political institutions of Madras city and how they
came to need, fear and use this connection. The emergent centre of the
new political system itself has not come under our gaze except as a
distant and vague mass of energy which outsiders could tap. Yet
the material form which this energy took was very much to condi-
tion what the outsiders were able to do with it. The manner in which
the power of the capital was organised determined both the lines along
which it could be approached and the types of approach which were
most likely to engage it. As the separate political institutions of the
locality came together largely through the capital, these lines and
types of approach influenced not only relations between the local
politician and the capital but also relations between one local politi-
cian and another. They were the links in the provincial political
system. Clearly, our next task must be to examine the political organ-
isations and interests which were developing within and around Fort
St George at this time.
In the context of the British Empire, of course, Fort St George was
a very inferior outpost, responsible to at least three, and sometimes
four, layers of superior authority; it was a mere office-boy in the
Grand Imperial Company. In the context of the Madras Presidency,
however, it looked more like an all-powerful despot. London and
Calcutta exercised only a loose discipline over its activities and very
seldom directly interfered beneath it. Moreover, before 1920 there
were no popular political institutions in the presidency and Fort St
George's will remained unchecked from below. What was promul-
gated by the Governor of Madras was the law of the land and although
this law might be ignored, no other authority existed to issue rival
promulgations. For most practical purposes, the Governor was an
absolute ruler.
215
 The emergence of provincial politics

As might be expected in this situation, the politics which were

played around the Governor, throughout the nineteenth century and
beyond, were very similar in kind to those being played around the
Indian princes who occupied analogous positions. The Governor's
immediate councillors intrigued for his ear (or more correctly his seal)
and, having obtained it, drew favours and patronage from him, which
they carried away to their offices to distribute as they pleased. In these
offices, they were joined by clients of their own, who in turn intrigued
for a share of the favours and patronage and carried off whatever
they received. And so the process went on, right down to the rock-
bed of the rural-local boss' or urban magnate's control, where it
stopped. The higher political system of the Madras Presidency can
best be described by the epithet 'court' and characterised as a series
of interlocking patron-client relationships through which resources
and executive and judicial decisions were passed. In the very early
nineteenth century, those at the top of the patronage ladder, dealing
directly with the Governor, had included a variety of Indian 'country
powers' — warlords, commercial magnates and semi-autonomous
chiefs. From the 1830s, however, the Governor's inner circle came to
consist largely of civil servants. For the rest of the century these
civil servants were to be entirely European. Moreover, their own
immediate clients tended increasingly to be Indian civil servants
rather than Indian local powers. As civil administration spread across
South India, it pushed indigenous political leaders further and further
away from the centre of provincial authority until contact between
locality and centre was mediated, if it were mediated at all, by many
layers of bureaucracy. It was this development, of course, which had
led to the fragmentation and localisation of indigenous political
activity in Madras.
Given the relatively small number of people who orbited the
Governor, the quantity of resources and power in their hands was
considerable. The Arbuthnot family, for example, was able to fund
one of the largest mercantile houses in India out of the private contracts
and services which it received while providing four generations of
Madras civil servants.1 In the 1880s, as we have seen, other I.C.S.
officers were able to supplement their already large salaries with
earnings in gold and plantation speculations made under the protec-
tion of their own government. Equally, Indians who reached the
higher grades of government appointment could expect to build
large fortunes both from their pay and from any perquisites which
1
For hints at the extent of the family's influence, see A. J. Arbuthnot, Memories of
Rugby and India (London, 1910).

216
 The emergence of provincial politics

might come their way; Sir T. Muthuswami Iyer, the first Indian
High Court judge, for example, started life as a penniless orphan
but left an estate worth nearly Rs 4 lakhs,2 Nonetheless, when set
against the amount of wealth and power left to groups in the locality,
the resources at the disposal of men in the higher political system were
relatively meagre. The Governor and his minions were dividing only
the quantity of cash which could be extracted safely from rural-local
bosses and urban magnates after the Government of India had taken
its share, and were commanding only those limited competences
and jurisdictions which local powers could be forced to recognise.
In effect, they were distributing among themselves the scraps of the
political system.
Among these scraps, probably the juiciest morsels were appoint-
ments to a variety of government jobs. By far the greatest amount of
Fort St George's internal expenditure — the cash-flow which it put
back into Madras - was directed towards the payment of its bureau-
crats. By the standards of the day, the salaries of posts at huzur
sheristidar level and above were attractive even when possible per-
quisites are not added. Posts below the huzur sheristidar level at least
offered subsistence and the hope of better things to come. Techni-
cally, all posts in the government service were filled by the Governor
and his senior administrative staff; appointments down to the level
of deputy collector being at the discretion of the Secretariat and
appointments below at the discretion of the Collector. Before the
early 1900s, however, no systematic effort was made to build a
machine which would distribute the considerable patronage rep-
resented by government employment in a way which was efficient and
impersonal. In consequence, the patron-client relationship, which
lay at the heart of the entire higher governmental structure, was most
noticeable in the organisation of service appointments. Senior British
civilians casually offered plum jobs to their friends and to any Indian
in whom they became interested. The spirit of the appointments'
system was well caught by K. S. Ramaswami Sastri in his auto-
biography when he noted that: 'I could have gone out as a District]
M[unsiff ] in 1905. Mr Justice Davis would have given me the post for
the mere asking as he liked me very much.' 3 A recent biographer of
V. V. Giri also has captured it accurately: 'Justice Ayling was earlier
District Judge of Ganjam and knew Giri's father well. As a gesture to
the family, he straightaway offered to appoint Giri a munsiff.'4
2
Hindu 7 February 1907.
3
Ramaswami Sastri, Vita Sua, pp. 36-7.
4
G. S. Bhargava, V. V. Giri (Bombay, 1969), p. 22.

217
 The emergence of provincial politics

Farther out from provincial and district headquarters, as we have

seen, Indian officials themselves took over the functions of appoint-
ment distribution, making the huzur sheristidar's office the centre of
favour for the mofussil bureaucracy.
The centralisation of appointments and control in the bureau-
cracy, which we discussed in Chapter 2, gradually pulled an increasing
amount of service patronage towards the capital and laid before
members of the Secretariat a much greater array of places than they
could have hoped to fill before. This development in the bureaucracy
proper soon came to be matched by a parallel movement in the ancil-
lary institutions of government. From the 1860s, the High Court
in Madras city became the focal point of the expanding litigation of
the presidency. Every year it heard more and more original cases,
while in its appeals capacity it responded to the general growth in
the use of the courts and dealt with a mass of cases referred to it from
subordinate judges.5 Lawyers serving before the High Court stood at
the top of their profession. Their incomes were incomparably higher
than those of pleaders and vakils who worked in the mofussil courts:
as late as 1920, it was almost unheard of for a district court lawyer
to earn even Rs 1,000 per month but, as early as the 1880s, some
Madras High Court vakils were pocketing Rs 10,000 per month.
Moreover, by handling the affairs of the great and the powerful from
all over the presidency, leading High Court lawyers built up provincial
practices and influence which enabled them to affect the retention of
vakils in subordinate court cases. The increase in litigation across
our period also led to an increase in the number of vacancies for
judicial servants and thus to an increase in the patronage of the
Judicial Department and the High Court which appointed all sub-
judges and district munsijfs.
Similarly, the expansion of western education concentrated new
powers in the capital. The growth of the bureaucracy and the law was
accompanied by a demand for higher standards of education among
government servants and lawyers. This demand produced a spate of
college building and gave the Education Department in the Secretariat
and the University more aggressive supervisory roles. The Education
Department exercised a considerable patronage in the schools and
colleges of the South, appointing masters, controlling the manage-
ment of educational foundations and influencing access to educa-
tional institutions. From 1877, when it first began to affiliate colleges
5
See Report(s) on the Administration of Civil Justice in the Madras Presidency (Madras,
1880 to 1920, annual).

218
 The emergence of provincial politics

outside Madras city to teach for its degrees, the importance of the
University also grew daily. Anybody wishing to found a school or
college or to undergo higher education required recognition by it.6
The University, though nominally representing all-presidency in-
terests, was dominated by Madras city residents who composed the
majority of active Senators and who, by Statute, were alone allowed
to sit on its executive (the Syndicate). The tight control of patronage
enjoyed by city residents indicates their great influence. It was
notorious that the professors in the two main colleges of the capital
packed the examination and text book committees, that students from
city colleges did better in examinations than outsiders and that endow-
ments and foundations made in the mofussil had a habit of being
transferred to Madras.7 Moreover, when the Government of India
in 1912 gave the Madras University a gift of Rs-4 lakhs and a recur-
ring annual grant of Rs 65,000, not one anna was spent outside city
boundaries.8 Naturally, this led to a great deal of animosity between
mofussil educational interests and the city colleges.
The gains of the centre, therefore, were very much the losses of
the districts. The independent realms of the huzur sheristidar, the
subordinate court pleader and the small town schoolteacher steadily
shrank in size before the advance of the capital. The career structure
of the services and education was in the process of being shifted from
a local to a provincial base. Not surprisingly, therefore, it was the
service families, who made their livings in the bureaucracy and
the learned professions, who felt first the pull of the capital. From
the 1870s, ever larger numbers of boys flocked to the schools and
colleges of the capital from homes all over the presidency; by 1914,
there were over 3000 students in Madras city.9 They left behind them
deserted and decaying schoolhouses which previously had flourished
by providing for the needs of petty local clerks and pleaders. Between
1882 and 191 o, for example Presidency College in Madras city
increased its size from 241 students to 532, while Kumbakonam
College, once 'the Cambridge of South India', declined from 267 to
123.10 Some of the leading families among western-educated Indians,
who dominated the government service and professions of their
6
Madras Universities Commission (Simla, 1902), 11 (Madras), 4.
7
Evidence of J. Cork in ibid., pp. 51-2; Hindu 8 March 1916; G.O. 593-4,
(Education) dated 15 October 1888. T.N.A.
8
Hindu 15 and 26 October 1912, 6 and 8 March 1916.
9
A. Besant to P. S. Sivaswami Iyer, 12 January 1914. P. S. Sivaswami Iyer Papers.
N.A.I.
10
Hindu 22 January 1915.

219
 The emergence of provincial politics

districts, moved to establish their presence and credentials at the

emerging centre of provincial affairs. In 1880, C. Sankara Nair, the
fourth generation of his family to serve the British, came up from
Malabar and apprenticed himself to a city barrister; 11 in 1884, S.
Subramania Iyer - the associate of Collector Crole in Madura -
moved his practice to the capital; 12 a year later C. Pattabhirama Iyer,
the leading light of the Tanjore bar, also found chambers near the
High Court. 13 About the same time, several young men of good
family — V. Krishnaswami Iyer and P. S. Sivaswami Iyer from Tanjore,
T. V. Seshagiri Iyer from Trichinopoly and L. A. Govindaraghava Iyer
from North Arcot — bore witness to the changing times by preferring
to stay in Madras city, after qualifying from the Law College, rather
than returning to their home districts where their family connections
would have guaranteed them a secure living.
The aspiration for career success, therefore, was instrumental
in drawing the prominent western-educated 'service' families of the
mofussil towards the capital. In some cases, it was joined also by the
aspiration to change the bases of social organisation in Madras. The
rise of the power of the central institutions of government presented
any man interested in promoting social reform with the new weapon
of legislation. The impact of western ideas obviously had made its
deepest impression on that segment of society which had been exposed
most to western education and, among the members of families who
were building connections with the capital at this time, were several
intellectuals dedicated to reform. In the mid-1880s, for example, C.
Sankara Nair became involved in an attempt to alter the marriage
system of his caste.14 Equally, a few years later, V. Bashyam Iyengar,
one of the principal lawyers in the city, began to press the govern-
ment to change the laws governing inheritance among Hindus. 15 The
personal social needs of some of the western educated of South India,
then, also formed a cause of the movement towards Fort St George.
It would be a mistake, however, and one all too commonly made,
to view the political activity developing around the core of the capital
as being the concern of the western educated alone. Certainly, it was
they who led this activity; certainly also career prospects and social
reform pertained to them alone. Yet, of the total political system of the
province, service careers and social reform constituted a minute
11
K. P. S. Menon, C. Sankaran Nair (New Delhi, 1967), pp. 13-16.
12
Raja Rama Rao, Sir Subramania Aiyer, p. 48.
13
C. P. Ramaswami Aiyar, Biographical Vistas (Bombay, 1968), p. 208.
14
Menon, C. Sankaran Nair, p. 27.
15
Home Judicial A May 1902, No. 88. N.A.I.

220
 The emergence of provincial politics

fraction. As we have stressed repeatedly, in political function the

western educated cannot be seen as autonomous agents. In the local-
ities and districts their chief purpose was to represent the interests
of urban magnates and rural-local bosses. They accomplished this
end both by mediating between the local power and the institutions
of government and by leading religious and cultural movements
which expressed the local power's personal status. The shift of locale
among the western educated, from the district to the capital, did
nothing, initially at least, to alter this fundamental relationship.
Indeed, the shift had taken place only because the locale of govern-
mental institutional power had changed - a fact reflected in the
centralisation of service careers. At the capital, the leading western-
educated men continued, as before, to place before the government
and the city populace a range of interests which they derived from the
local powers who paid their fees and funded their projects. Leading
vakils, such as S. Subramania Iyer, V. Bashyam Iyengar and C.
Sankara Nair, spent most of their time pleading before the courts on
behalf of zamindars based as far north as Ganjam and south as Malabar
and of merchant princes from the Northern Circars down to Ramnad.
In the University, they, and other Senators, protected the schools
and colleges founded by the wealthy and sought University help for
many cultural movements which had started in the purses of the great.
In the Board of Revenue and Secretariat, senior Indian officials
pushed forward to their superiors files concerning closely the interests
of major local figures. Through the press and on the public platform,
they drew attention to the grievances of zamindars, merchants and
mirasidars, and organised religious and cultural activities which,
though no doubt conditioned by their new social perceptions, were
financed and made important by the magnates who gained prestige
from them. Certainly, the lines through which the western educated
now worked for their patrons were much longer than before, stretch-
ing from the locality to Madras city, but the interests which the lines
joined had not altered. The western educated could not function
politically as an isolated group; they were an intermediary segment.
Their success in public life depended as much on the confidence they
inspired among rich Indians as on their ability to manipulate career
structures and the still weak legislative power. Indeed, without
clients to support them they could not advance their careers nor
persuade government that they were sufficiently important to be
given legislative power.
Once we begin to view the political system of Madras in this broad
perspective, a number of the more significant political events of the
221
 The emergence of provincial politics

1880s and the 1890s start to take on a new meaning. In particular,

the often-seen emergence of agitational politics and the institutions
of the nationalist movement can be set more firmly in the political
context in which they arose. Fort St George's drive to administer more
carefully the resources of its presidency was subjecting local powers
to unprecedented pressures. Between 1878 and 1886, bills were put
before the Legislative Council to alter the institutions of religion,
the organisation of zamindari estates, the allocation of forest rights,
the maintenance of irrigation works, the structure of local self-
government, and the nature of the income-tax. Very few men of local
substance were not touched by these issues. Moreover, very few failed
to realise that the new potency of the central institutions of govern-
ment could be used to aid more positively their own interests -
especially by providing funds and support for their religious and
cultural objectives. Local powers, then, came to need desperately an
influence on the way that affairs were conducted at Fort St George.
They instructed their various lawyers, bureaucrats and publicists,
who had gathered at the capital, to make representations of their views.
Importantly, however, there were few means by which these rep-
resentations could gain a hearing. In spite of their increased power,
the central institutions of Fort St George remained very narrowly
based. The Legislative Council consisted of a small, government-
appointed and self-perpetuating clique. lb The principal executive
officers of government were entirely European and were being pulled,
by Government of India demands for improved performance, away
from such contacts with Indian opinion as they might once have had.
The High Court bench, with one exception, also was European. The
University, which related to an increasing number of Indian cultural
interests, was dominated by a Syndicate consisting largely of mission-
aries who were inimical to Indian culture. Although the western-
educated representatives of local native power were congregating
around Fort St George, they were not allowed entry to its halls. Not
surprisingly, then, they began to put their cases from the street. In
1882, the Madras Native Association, defunct since the 1860s, 17 was
16
Of native members, for example, Raja G. N. Gajapati Rao, a conservative
zamindar from Vizagapatam, was nominated for nine terms; T. Rama Rao, an
effete city lawyer, also was recalled for several two year terms.
17
The M.N.A. was originally founded to protest against the increasing influence
of missionaries and to petition Parliament on the 1853 Charter Act. It died shortly
after the 1861 Legislative Council Act, when two of its founders and leading
members became Legislative Councillors. R. Sunthralingam, Politics and
Nationalist Awakening in South India, 1852-1891 (Arizona, 1974), PP- 45~57-

222
 The emergence of provincial politics

resurrected by Madras city lawyers, publicists and a variety of govern-

ment servants, and used to press government for greater measures
of representation in local self-government.18 The sharp reaction of
Fort St George to the sight of its Indian bureaucrats playing their real
political role, as the spokesmen of local powers against the administra-
tion, quickly killed this association. The next year a new society, the
Madras Mahajana Sabha, was founded by the same city lawyers and
publicists (but not the bureaucrats) and met with immediate success.19
It led campaigns over temple reform, zamindari legislation, the income-
tax, the increase in excise duties and land revenue resettlement
policies.20 The Mahajana Sabha built up a considerable provincial
following and pilloried the government on every possible occasion,
ridiculing its inefficiency and demanding elective representation.
These protest movements were not only negative attacks on current
practice. By the early 18 80s, it was obvious that the British, in Calcutta
and London, were beginning to appreciate the dangers of an over-
zealous bureaucracy losing touch with popular opinion. They were
considering major changes in the constitution of government. Com-
missions on the Public Services and Education toured Madras, Lord
Ripon began to tinker with the mechanisms of local self-government
and the Secretary of State with those of the Legislative Council.21
As the impetus for change was coming from supra-provincial centres,
so politicians in each province saw the need to band together to work
in all-India associations. The Indian National Congress in Madras was
organised and run by exactly the same people who had founded the
Mahajana Sabha, and it carried their demands before the Viceroy
and the House of Commons.
The main reforms for which the Congress campaigned at all-India
level were in the areas of the Legislative Council, the University and
the bureaucracy which, it claimed, should be opened to more Indian
talent.22 This concentration, particularly on the University and
bureaucracy, has tended to be seen as a further indication of the real,
personal ambitions of Congress leaders for more and better jobs. So,
in part, it may have been. But to dwell on this aspect too exclusively is
to miss the fact that, in a bureaucratic state, changes in the composi-
18
Government servants were especially active. See Proceedings of the Madras Native
Association on the Resolution of the Government of India on Local Self-Government
(Madras, 1883).
19
See The Madras Mahajana Sabha Annual Report for 1885—6 (Madras, 1886).
20
Ibid.
21
Seal, Emergence of Indian Nationalism, pp. 1 3 1 - 9 3 .
22
Ibid., pp. 245-97.

223
 The emergence of provincial politics

tion of the bureaucracy can amount to changes in political represen-

tation. The local contacts of the more prominent Congressmen would
certainly have gained had their representatives obtained places in the
central decision-making machines. Moreover, of course, it was inevi-
table that the Congress should have emphasised these constitutional
issues almost exclusively when it was talking to the Government
of India and the House of Commons, for it was only in these issues that
those bodies had any interest or competence. The Viceroy was hardly
likely to bow his plumed head to listen to appeals against the assess-
ment of a field or the cost of drainpipes. In Madras, however, Congress
agitation was concerned extensively with just such a range of petty
matters: Congress leaders could expect some response from Fort St
George on them and knew that it was in these small things that the
greatest interests of their constituents lay.23
The breadth and depth of involvement in the agitations of the
1880s, as well as the intricate relationship between western-educated
publicists and lawyers and their magnate supporters, were laid out
fully for us to see in the organisation of the 1887 session of the
Congress held at Madras. The work of co-ordinating activities with
publicist leaderships in other provinces, of preparing speeches and
propaganda and of drafting resolutions to be discussed at the session
was, of course, the responsibility of only a tiny group of the western
educated at Madras city - P. Ananda Charlu, S. Subramania Iyer, G.
Subramania Iyer, Salem Ramaswami Mudaliar, M. Viraraghavachari
and a few others. The money, which made the session possible, and the
demonstration of political power, which made it important, however,
came from other quarters. The greater part of the costs of the
Congress was met by the Indian mercantile elite of the capital. Raja
Sir Savalai Ramaswami Mudaliar, P. Somasundram Chetty, G.
Mahadeva Chetty and T. Namberumal Chetty gave handsome don-
ations to the reception committee. 24 They also recruited cash and
support from the areas of the city in which their influence was
strongest. They organised collection committees in their suburbs
and held meetings which elected delegates to the Congress session.25
P. Somasundram Chetty even turned one of his businesses, the
23
Zamindari rights, land revenue, local self-government and the income-tax, among
other local topics, filled a major part of the discussions of the first Congress held
at Madras. See Report of the Third Indian National Congress held at Madras on the
2jth, 28th, 29th and 30th December 188'7 (London, 1888).
24
Ibid., A p p e n d i x 1.
25
Hindu 12, 14, 16, 19 and 21 December 1887; Madras Mail 19, 20, 21 and 24
December 1887.

224
 The emergence of provincial politics

Metal Trading Company, into a Congress electorate.26 The support

which the mercantile elite of Madras city provided for the Congress
reflected its close association with western-educated leaders, such
as P. Ananda Charlu, in local politics over the previous decade.
Ananda Charlu had led several agitations on its behalf against Cor-
poration decisions27 and against the proselytising activities of Christian
missions,28 and he was currently involved in articulating its griev-
ances over the income-tax and local self-government acts through
the Mahajana Sabha,29 and in putting forward its cultural demands
in the University Senate and from the public platform.30
Behind the city merchants came a number of the wealthiest zamin-
dars in the province. The Maharaja of Vizianagram, who sat on the
speakers' platform, had many links with the prominent professionals
of the capital. He left his more important litigation in their hands
and was joined to them in several cultural and proto-political pursuits—
his patronage lay at the back of the female education movement and
the Hindu newspaper.31 The Raja of Venkatagiri, who sent his diwan
to the session, also was a financier of city journalism as well as a
great litigant and investor in the capital. Other leading landowners
who offered support to the Congress, either in person or through
agents, included the Raja of Sivaganga, the zamindar of Chellapalle
and V. A. Vandayar of Pundi.32
The forces of organised religion also moved at the back of the
session. The Sankarachariar of Kumbakonam sent a representative
with a large donation and the Pandarasanidhi of Dharmapuram,
constituted his disciples into a Congress electorate.33 Again, both
religious leaders had direct ties into the professional elite of the city,
members of which were promoting the causes of Tamil and Sanskrit
revivalism.
A fourth clearly identifiable interest at the session was that of the
greater mofussil commercial powers. The major Komati families of
the Northern Circars and Nattukottai families of Ramnad attended
in person and helped to recruit followers by holding public meetings
in their towns.34 The millionaire industrialist A. Sabhapati Mudaliar
26
Report of the Third Indian National Congress, Appendix i .
27 28
Hindu 17 August 1884. Hindu 7 May 1888, 17 and 18 May 1889.
29
The Madras Mahajana Sabha Annual Report for 1885-6 (Madras, 1886).
30
Hindu 7 M a y 1888, 17 a n d 18 M a y 1889.
31
Hindu 3 M a r c h 1904; V. K. N a r a s i m h a n , Kasturi Ranga Iyengar, p. 6.
32
Report of the Third Indian National Congress, Appendix 1; Hindu 11 January
33
1888. Ibid.
34
Hindu 12, 16, 19 and 21 December 1887; Madras Mail 19, 20, 21 and 24
December 1887.
225
 The emergence of provincial politics

of Bellary, who had close contacts with the city through his litiga-
tional and social reform activities, even placed his commercial net-
work at the disposal of the Congress leaders.35 Across the towns of the
Ceded Districts, where he held abkari contracts and bought raw
cotton, his agents were ordered to set up public meetings and collect.
funds.36
Of course, the Congress campaign of the later 1880s not only
outlined an arena in which pre-existing linkages between magnates
and western-educated publicists could be displayed; it represented a
new level of institutional political activity and so was capable of
creating new linkages ex nihilo. Congress headquarters in Madras
city sent out touring lecturers to spread propaganda in the mofussil f1
it devoted a great deal of energy to vernacular pamphleteering in order
to cross the English language barrier which kept most Madrasis
ignorant of higher politics; 38 it built a powerful institutional structure
in which activity was co-ordinated between the capital and thirty-six
mofussil committees.39 Few local political powers, who were feeling
the weight of Fort St George's new rule, were left in any doubt that
they should come to Madras city to seek redress. In consequence, the
1887 session was attended by a mass of petty urban magnates and
rural-local bosses who were making their presence felt on the pro-
vincial stage for the first time — Lingayat merchants from the towns
of the Ceded Districts, Reddi 'Sirdars' from Anantapur and Bellary,
Vellala landowners from Tinnevelly and Trichinopoly and strong
local despots such as T. S. Sivaswami Odayar of Kumbakonam. 40
The organisation of the 1887 Congress at Madras provides us with
a political map of the presidency in which were drawn both the
boundaries of local political constituencies and the threads which
connected those constituencies to the capital. The Congress rep-
resented most of the important political elements of the province
massed against growing governmental interference. As virtually
every indigenous political power, regardless of shape or hue, felt
the prodding of government, so virtually every major political interest
35
Madras Times 1 S e p t e m b e r 1885.
36
Hindu 16 D e c e m b e r 1887.
37
F o r example, see t h e itinerary of lecture tours for Easter 1890 in Madras Times
4 April 1890.
38
F o r example, t h e famous Tamil Catechism was published in 1887, a n d t h e
Swadesamitran, a T a m i l weekly founded by leading Congressmen, first appeared
in 1882.
39
'Annual Report of the Madras Standing Congress Committee for 1889' in H o m e
Public B April 1890, No. 154. N.A.L
40
Report of the Third Indian National Congress, Appendix 1.

226
 The emergence of provincial politics

backed Congress demands. At this period, Congress in Madras faced

little opposition from anti-nationalist (so-called) communal groups.41
Most of the leading South Indian Muslims gave active support to the
session; some 67 Muslims attended, including the prominent Madras
city Persian family of Badsha, and the Maracair Muslims of Nega-
patam who were the official Congress agents for eastern Tanjore.42
The Eurasians, under W. S. Gantz's leadership, also were Congress
activists.43 Even the European business community of the capital,
in spite of the recent problems of the Ilbert Bill, praised the objectives
of the Congress.44 It too was under the lash of Fort St George's new
driving policy and it too wanted more representation in government.
The communal divisions, which later emerged to tear apart provincial
and national politics, were as irrelevant to the Congress of the late
1880s as they were to most aspects of local politics all the time.
The unanimity of opinion demonstrated at the Madras Congress
session of 1887 was impressive. Importantly, however, the forces
which brought this opinion together derived largely from pragmatic
responses to particular conditions in the political environment. While
leading Congress publicists and intellectuals may well have regarded
the session as a triumph for their own national values and as a sign of
the awakening of South India to a new consciousness of national
political and cultural identity, in fact, as subsequent events were to
prove, most of the participants were interested only in a series of very
specific and very limited objectives. They sought the establishment of
a political system which could convey the grievances, which were
being fostered by governmental intrusion, more quickly to the centre
and which could harness the new power of the capital to their own
growing local needs. Once these objectives had been obtained and a
new political framework built, for them the immediate ends of the
Congress campaign would have finished and the Congress would
have to continue, if it were to continue, as a very different type of
political institution. Whatever its leading apologists might have
thought, after a decade of ceaseless activity the Congress campaign
41
There were a few anti-Congress protests from Madras Muslims but an analysis
of them reveals that the protesters were, almost uniformly, relatives or close
contacts of the Nawab of Arcot. This means that they were government placemen
who might have been prodded by the British into activity. See Sunthralingam,
Politics and Nationalist Awakening in South India, 1852—1891, pp. 264—8.
42
Report of the Third Indian National Congress, Appendix 1; Hindu 21 December
1887.
43
Hindu 29 April 1887.
44
Madras Mail 30 December 1887.

227
 The emergence of provincial politics

had indeed realised most of its primary aims and, by the last years of
the nineteenth century, was in the process of changing itself almost
out of existence.
During the late 1880s and early 1890s, the British responded to
pressures from the Congress, and to their innate fear that by extend-
ing the scope of government they would raise political opposition to
their rule, by granting many of the requested reforms. On the Univer-
sity Senate, five seats were opened to election by the graduates; in
1892, the Legislative Council was enlarged and six elected seats were
created; in the bureaucracy, the 1890s witnessed the introduction
of the first Indian Collector, Solicitor- and Advocate-General and
the opening of several senior posts in the Secretariat departments
to non-I.C.S. officers.
Although these changes may appear to be very limited, they clearly
widened the possibility of contact between 'government' and 'society'.
In the University, the elected Senators formed a caucus which regularly
challenged the official leadership of European and missionary educa-
tionists.45 It was now practicable for opposition to Syndicate policy
to be mounted inside the Senate rather than on the street. Moreover,
after Curzon's University Act of 1904, this caucus was able to make
headway in taking over the decision-making process for itself. Curzon's
Act reduced the number of Senatorships granted for civic duties - t o
Europeans and Indians who were likely to be susceptible to pressure
from senior officials46 — and increased the proportion given to profes-
sional educationists. In Madras, by this time, most professional
educationists were Indians. The elected Senators, usually men of
considerable reputation and power in the law and public life, possessed
many cultural, career and familial connections with the educationist
Indian Senators and so were able to influence the Senate. Indeed, in
1904 the Senate, for the first time, refused to elect a missionary to the
Legislative Council and chose instead one of the elected Senators,
P. S. Sivaswami Iyer. The growing power of the Indian political group
in the University was seen in the 1912 and 1913 Senate debates on
the role of the vernaculars in higher education. The Senate, under
the effective leadership of G. A. Natesan, overturned the Syndicate's
resolution and forced the Syndics to run to the Director of Public
45
F o r a glimpse of this caucus in action during a Senate election, see Hindu 15 a n d
19 October 1904.
46
Before 1904, m a n y 'city fathers' a n d businessmen received invitations t o join
the Senate. I n 1902, there were 200 members of the Senate. Indian Universities
Commission (Simla, 1902), 11 (Madras), 46. T h e Universities Act of 1904 cut
the m a x i m u m n u m b e r to 100.

228
 The emergence of provincial politics

Instruction for the protection of a Government Order.47 In 1916,

Natesan obtained a High Court order to make the Syndicate obey the
Senate, its titular master. 48 In order to keep the University running,
the Syndicate often had to stamp decisions made in the Senate where
Indian opinion could play its part.
Similarly, in the Legislative Council, the apparently small changes
made by the 1892 Act mask a greater alteration in political signific-
ance. The Council, of course, was still dominated by British officials
who could push through what measures they pleased and prevent the
passing of any resolution or Bill emanating from the non-official
members. But legislation was never very central to government in
Madras so that this fact was not critical. The non-official members,
however, were able to draw the attention of senior bureaucrats to a
wide variety of administrative malpractices and political grievances
and to obtain some satisfaction in them. In the first months of the new
Council, for example, K. Kalyanasundram Iyer managed to put an
end to the extraordinarily arbitrary procedure by which the Board
of Revenue heard appeals.49 His elected fellow-Councillors pressed a
mass of questions on senior officials - about matters ranging from
irrigation taxes to the payment of office clerks.50 Although official
answers to these questions may not appear satisfactory, an examination
of government files reveals that they did stimulate officials to reopen
discussion of procedures and cases and, occasionally, to undertake
reform.51 Legislative Councillors received deputations from local
powers and, as can be seen simply from the lists ofquestions which they
put to the government, acted as the mouthpiece of local grievance
inside the Council chamber.
Moreover, the Council proved very useful to non-official members
as a means of obtaining places in the machinery of government and
hence in the process by which political decisions were made. Senior
British officials used the Council to test the ability of a much broader
spectrum of Indian professional talent than before. Among those who
moved from the benches of the elected non-officials to those of the
government servants during the next two decades were C. Jumbulingam
47
Hindu 10 a n d 15 M a r c h , 1 M a y 1913.
48
Hindu 28 October 1916.
49
Hindu 14 N o v e m b e r 1893.
50
For example, N. Subba Rao on the P.W.D. in Andhra, Hindu 4 October 1894;
C. Sankara Nair on the Abkari Department, Kerala Patrika 29 October and
5 November 1892. R.N.P.
51
As over the Tanjore revenue resettlement, G.O. 2472-3 (Revenue) dated 26
June 1894. T.N.A.

229
 The emergence of provincial politics

Mudaliar, K. Srinivasa Rao, V. C. Desikachari, P. S. Sivaswami Iyer,

V. Krishnaswami Iyer, L. A. Govindaraghava Iyer, T. V. Seshagiri
Iyer and S. Srinivasa Iyengar. In 1909, the Hindu noted
the fairly large numbers of elected Members who have been the successful
applicants for judicial office under the Government. Besides those who have
already secured judicial office, we are informed on unimpeachable authority
that among the four Members of the Legislative Council recently elected by
the Municipalities and District Boards, there are applications pending for
such appointments from two of them before the Local Government.52
The reforms in the University, the Legislative Council and the
bureaucracy suggest a central reason for the decline of agitational
politics in the 1890s. It was now possible for political representation
to take place inside the constitution. Although, with one exception,53
only western-educated lawyers ever were elected to the Legislative
Council, they represented in the Council chamber the same broad
splay of magnate interests which they had represented outside. As
the Madras Government argued, when questioned by the Government
of India on the narrow social base of its Councillors:
Taking first the case of the raiyatwari tenure, the circumstances of this
Presidency are such that there is a most intimate connection between the
professional classes and the richer Government ryots Thus it occurs that
though allfivenative members nominated on the selection of the recommend-
ing bodies are lawyers, no better representatives could possibly be found for
the interests of the Government ryots.... Coming in the second place to the
zamindari interest, a close study of the legislative proceedings for some
years past has convinced this Government that the one seat assigned to this
class is far from being a measure of the influence which this class exercises
in the Council. With them too the legal members are associated either in
interest or in sympathy From these and other causes it has resulted
that in the several discussions on legislative measures affecting the real
or supposed rights and privileges of zemindars, the positions claimed by the
latter have constantly been defended by legal members with ability that
would certainly not be equalled by the zemindars themselves.54
The Government of Madras could have gone further and have pointed
to the connection between bankers and lawyers, typified by the opposi-
tion led by some Councillors to an Encumbered Estates Act, which
undermined the interests of such men as Lodd Govindoss who special-
52
Hindu 17 November 1909.
53
A. Sabhapati Mudaliar, w h o was o n t h e Council between 1893 an<^ ^95-
54
Letter No. 7 (Legislative), Government of Madras to Government of India dated
31 January 1899 in Home Public A July 1899, Nos 16-21. N.A.I.

230
 The emergence of provincial politics

ised in lending money to zamindars;55 between merchants and lawyers,

typified by council opposition to the income-tax and to regulations in
restraint of trade; 56 and even between rural-local bosses and lawyers,
typified by C. Vijayaraghavachari who was the legal representative of
several wealthy Gounders and Kallars in Salem, P. Kesava Pillai who
worked for Thimma Reddi in Anantapur and B. V. Narasimha Iyer
who helped to organise a caucus ofGounder landlords in Coimbatore.57
By the mid-1890s, large-scale political agitation had lost its most
important functions and was no longer necessary to connect public
opinion to the centre of government. Representations made through
the new constitutional channels had replaced it. With minor alter-
ations for Lord Morley's reforms of 1909, the constitution of the
1890s continued to meet the political needs of Madras until 1916 and,
consequently, to dominate the political rhetoric of the South. More-
over, after the bout of reforms in the 1890s, the British ceased to show
further interest in altering the mechanisms of government so that
continued appeals for major constitutional reform were futile. Agita-
tion not only became unnecessary but also, given that government was
no longer sympathetic, unrespectable.
The Madras Congress and the Mahajana Sabha, the formal institu-
tions of Congress agitation, rapidly lost their raison d'etre and declined
in importance. The elaborate structure of Congress organisation
collapsed while, between 1886 and 1900, the Mahajana Sabha lost
two-thirds of its membership.58 Congressmen from other provinces
began to mock the scarcely disguised lethargy which came to character-
ise subsequent Congress sessions in Madras. The Lahore Tribune asked
in 1894: 'By the way, is it true that many Madras notabilities who
have up to very recently been steeped to the ears in the Congress, now
show signs that they wish to be left alone?'59 By 1904, the Governor,
Lord Ampthill, could afford to ridicule the remnants of an organisation
which had outlived its usefulness: 'The Congress as it is worked at
present is nothing more than an annual picnic or "tamasha" . . . which
has its counterpart in a minor organisation at home which celebrates
55
See Hindu 18 February 1902.
56
For example, N. Subba Rao, a Legislative Councillor, presided at the foundation
of the Vizianagram Merchants Association, Hindu 13 September 1893.
57
The rural-local boss interest was represented clearly in protests over the forest
laws, over police reform and over the resettlement of the land revenue. See
Hindu 1, 2 and 17 April 1915.
58
M . M . S . m e m b e r s h i p declined from 607 i n 1885 t o 267 b y 1900. Madras Mahajana
Sabha Annual Report for 1885-6 (Madras, 1886) a n d ibid., 1899 and 1900
(Madras, 1901).
59
Quoted in Hindu 1 August 1894. See also editorial in Hindu 1 December 1891.

231
 The emergence of provincial politics

an annual "beano" under the name of a Working Man's Political or

Athletic Association.'60

Constitutional politics 1895 to 1916

The political framework which emerged out of the agitations of the
1880s represented the replication in constitutional form of the link-
ages which had developed between leading western-educated pub-
licists and professionals at the capital and local magnate powers.
The same interests, motives and activities which previously had been
fused together in street agitations were now fused together in the
Legislative Council, University and High Court. This development, of
course, meant that the British had recognised as legitimate the types
of political organisations which had been built up during the agita-
tions. The British had brought them inside their governmental insti-
tutions and were proposing to govern with them rather than against
them. The move towards constitutionality proved disastrous for
agitational politics and for the nationalist movement: most of the
channels which had been used to communicate hostility towards the
system of government were now used to communicate only grievances
within an accepted system of government. All local magnates and
western-educated publicists who worked within the constitution were
strengthening and supporting the status quo. It is, perhaps, because
of this switch from 'nationalism' to 'collaboration' that the years
between 1895 and 1916 have tended to be written off as politically
sterile — as years of moderation when very little happened. Such a
view, however, would be based on the assumption that it was only
through agitation that Indian political organisation expanded and
Indian political opinion came to be voiced. All that we have seen
thus far, however, indicates that this view is untenable. Agitation and
collaboration were tactics used by members of an indigenous political
system which existed as a logically independent entity beneath the
British and which, like all political systems, was concerned to dis-
tribute power. While the pressure of agitation may have helped to
develop the system, there is no reason why the practices of collabora-
tion, particularly the new kinds of collaboration taking place from the
1890s, should not have altered it as well. Indeed, as by definition more
real power was available to the system when it was working with the
government than when it was working against it, it would be more
logical to argue that more change was wrought during periods of
60
Lord Ampthill to Broderick, 7 January 1904. Ampthill Papers. I.O.L.

232
 The emergence of provincial politics

quiescence than during periods of protest. This was certainly the

case in Madras. If, between 1895 a n d 1916, scarcely a single anti-
British dog barked on the streets, nonetheless several of the most
critical relationships of South Indian political life underwent a
dramatic change.
As we have seen, the institutions of the bureaucracy, the law and
education played more than one important role in the politics of the
presidency. On the one hand, they represented the medium through
which a distant government gave orders to magnate-dominated local
society and through which that society sent its complaints and
aspirations back again. On the other hand, to members of the western-
educated community, these same institutions represented career
structures which, in their own right, were capable of providing income
and status. To use medical terminology, they were both membranes,
passing influence from one part of the body politic to another, and
organs, capable of generating influence of their own. Until the mid-
1890s, however, any ambiguities which might have been created by
this duality of role were obscured by the fact that both roles were
functioning towards the same end. The educated and their patrons
wanted precisely the same reforms, for the broadening of the constit-
ution meant both improved representation and more control over
careers in the services. In order to obtain reform also, the educated
and their patrons had to work together and exchange their mutual
talents - publicity and bureaucratic skills on the one side and wealth
and social power on the other. During the 1880s, neither could
operate successfully without the other.
The reforms of the 1890s, however, began to alter this symbiosis of
circumstance. Improved systems of representation and Indianisation
of the services brought a few western-educated men much closer to
the centre of government. This made it easier for them to put the
cases of the magnates to Fort St George. But is also gave the fortunate
few an increasing personal influence over the internal operations of
the bureaucracy, law and education. A spanner to grip the career
structure of the professions was placed in their hands. Immediately,
this development saw the western-educated community, once so
united in the face which it showed to the British, tear itself apart in
an internal fight for jobs and place. Later, as central governmental
power grew and their grip tightened further, the few found them-
selves in control of the only effective system of representation in the
province. They were able to use this control to break their previous
dependence on the magnates. During the 1910s, they turned the
power of the centre of government into an autonomous political base
233
 The emergence of provincial politics

on which they could stand apart from, and in opposition to, many of
their old patrons.
As we noted above, the University Senate and the senior posts in
the bureaucracy enjoyed a patronage which was unsystematised and
subject to little outside opinion. Colleges were affiliated or disaffiliated
from the University by simple Syndicate recommendation and most
posts in the government were appointed at the discretion of a superior.
This lack of responsibility had led to the emergence of one of the
most naked systems of government by favour in the whole of India.
Importantly, however, up to the 1890s, the greater share of control
over the favours of the central government was in the hands of Euro-
peans who were distant from the indigenous political system. Certainly
many Indians gained from the patronage of civilians. Indeed, the
whole of the first generation of provincial politicians in Madras city
were clients of Europeans, who had been brought to the capital by
European favour. Muthuswami Iyer, for example, became, on the
intercession of D. Carmichael, the Chief Secretary, the first native
High Court Judge in the presidency.61 S. Subramania Iyer, as we have
seen, enjoyed the support of C. S. Crole; C. Pattabhirama Iyer came
up to the city in the wake of Mr Justice Parker, who had been his ses-
sions judge at Tanjore, and got a Judgeship for himself;62 C. Sankara
Nair was made a district munsiff before he had ever argued a case,
when the barrister to whom he had been apprenticed was made a High
Court Judge. 63 Yet, in their own right, these Indians were fairly far
down the provincial patronage ladder: they received patronage from
Europeans but had little themselves to pass on to their own clients
and no control over what the Europeans did. Moreover, of course,
the amount of patronage available to the centre was limited by the
strength of the localities at this time. The reforms of the 1890s,
however, began to place certain Indians in much closer contact with
the centre precisely at the moment when the centre was breaking into
the localities. As the reforms made no attempt to solve the problems of
rule by personal discretion, they gave a handful of chosen Indians the
means of developing networks of patronage and appointment which
covered the entire province and held career success in the bureau-
cracy, the law and education in thraldom. The centralisation of
government in Madras began to mean not the suppression of the
61
Hindu 23 November 1883.
62
R a m a s w a m i Aiyar, Biographical Vistas, p . 208.
63
C. Sankaran Nair, The Autobiography of Sir C. Sankaran Nair (Madras, 1966),
p. 8.

234
 The emergence of provincial politics

principles by which the old huzur sheristidar had organised his district
bureaucracy but their elevation to the organisation of the presidency.
The simplest way of judging the scale on which the new patronage
masters were able to operate is by looking at the extent to which they
were able to help their families. C. Sankara Nair, for example, used
his contacts with the European barrister/Judge to whom he had been
apprenticed to build up a lucrative city practice. His clients included
the wealthiest Malabar jenmis and the independent Raja of Puddu-
kottai. He was able to pass on cases from these clients to his own
nominees. In the climate of reform, he became a Legislative Councillor,
University Senator, High Court Judge and member of the Viceroy's
Executive Council. While his influence was at its height, members of
his immediate family circle found that their fortunes were easily
made. One cousin, apprenticed to Sankara Nair, picked up a judge-
ship in the High Court of Puddukottai and a valuable job in the
Madras Law College;64 another was made a professor at Presidency
College, Madras, while still in his twenties;65 two more were appren-
ticed in Madras and sent out to the mofussil bars of Madura and
Ramnad where, with the aid of High Court and local self-government
patronage from the capital, they soon became celebrities;66 a family
friend, with very small means, was offered a chair at the Law College.67
Yet as family connections went, Sankara N air's were somewhat
small and isolated. Much more impressive were those of the Vembak-
kam family and the complex of families which we shall refer to as 'C.
P. Ramaswami Iyer's community'. The Vembakkam Sri Vaishnava
Brahman family from Chingleput had been in the capital for a long
time when it first begins to interest us: it included a Madras city
police chief in the 1820s, leading city lawyers for three generations,
the first Indian to be nominated to the Legislative Council in 1861
and two diwans of Native States (at that time the highest appointments
for Indians in the gift of the British government).68 It possessed an
enormous presence in the local bureaucracy of the capital and its
surrounding district and, by the early 1890s, was holding family
conferences at which fifty or more administrators met to discuss
64
G.O. 518 (Education) dated 10 September 1909. T.N.A.
65
Sastri, Encyclopaedia of the Madras Presidency, p. 617.
66
Who's Who in Madras. 1938 (Cochin, 1938), p . 7 7 .
67
G.O. 518 (Education) dated 10 September 1909. T.N.A. This was Dr S.
Swaminathan.
68
V. C. Gopalaratnam, A Century Completed. (The Madras High Court 1862-1962)
(Madras, n.d.), pp. 270-2; also, G. Parameswaran Pillai, Representative Men of
Southern India (Madras, 1896).

235
 The emergence of provincial politics

domestic problems related to government service.69 In the 1890s, its

undoubted leader was V. Bashyam Iyengar, who was also the ac-
credited leader of the native bar. The opening out of the higher posts
of government in Madras witnessed its growth into a provincial rather
than a local administrative power. By the early 1920s, it had produced
another generation of legal giants, three High Court Judges,70 two
Attorney-Generals,71 the first Home Minister under the Montagu-
Chelmsford Reforms72 and countless minor judges, departmental
under-secretaries and government servants.73
C. P. Ramaswami Iyer's community of North Arcot Smartha
Brahmans consisted of the fusion of the family of C. V. Runganatha
Sastri, a Small Cause Court Judge, with that of Ramaswami Iyer's
father, C. Pattabhirama Iyer, which had its own service complex in
Tanjore. C. Pattabhirama Iyer followed his in-law into a Small Cause
Court Judgeship but his son, and the whole of the next generation,
were to do even better. Runganatha Sastri's son, Sundara Sastri,
and adopted son, P. Ananda Charlu, established rich legal practices
and Sundara Sastri's sons, Kumaraswami and Visvanatha, became
High Court Judges.74 C. P. Ramaswami Iyer inherited the Sundara
Sastri-Ananda Charlu practice via his brother-in-law Kumaraswami
Sastri, turned down the offer of a High Court Judgeship while still in
his thirties and followed a member of the Vembakkam family into the
Home Ministership in 1921. 75 Various other kin-connections picked
up district munsiffships, sub-judgeships and district judgeships in the
period.76 In fact the exploitation of patronage by this 'community'
became so outrageous that it caused stirs in the press and in the
services generally. The elevation of C. P. Ramaswami Iyer's cousin,
the city vakil A. S. Balasubramania Iyer, to a sub-judgeship over the
heads of men who had been waiting for years for promotion, produced
69
Hindu 6 and 27 March 1894.
70
V. Bashyam Iyengar, V. V. Srinivasa Iyengar and C. Thiruvenkatachari (by
marriage).
71
V. Bashyam Iyengar and S. Srinivasa Iyengar (by marriage).
72
K. Srinivasa Iyengar (by marriage).
73
Such as V. Krishnamachari and V. C. Desikachari w h o became Judges of the
Small Cause Court.
74
C. V. Visvanatha Sastri, Biographies of a Grandfather and His Grandson (Madras,
1
939) 5 passim.
75
'C.P. on h i m s e l f in C.P. by his Contemporaries (Madras, 1959); also A. Prakash,
Sir C.P. (Madras, 1939), pp. 1 - 2 0 .
76
See Ramaswami Sastri, Vita Sua and Ramaswami Sastri, Professor Sunderama
Ayyar.

236
 The emergence of provincial politics

a flood of petitions and threatened resignations in the Judicial Depart-

ment.77 The appointment of C. V. Kumaraswami Sastri to the High
Court in 1913 saw the Hindu whip itself up into a frenzy of abuse at
what it saw to be the patent corruption of government.78
Of course, family connection was not the only channel along which
the victorious few of the 1890s distributed their patronage; we have
used it here merely to demonstrate the kind of influence they now
exercised. They operated a system of appointment by personal dis-
cretion which could be used to help friends and business and cultural
contacts as much as relatives. For example, V. Krishnaswami Iyer,
lawyer, University Senator, Legislative Councillor, High Court Judge,
Executive Councillor and the most important man in the capital
between about 1900 and his death in 1911, was a close associate in the
University of the non-Brahman J. M. Velu Pillai,79 whose son,
Masilamani Pillai, became intimately connected to Krishnaswami
Iyer's circle in Madras and received a valuable High Court job.80
Similarly, cultural interests brought together Sir S. Subramania Iyer
and another non-Brahman, J. M. Nallaswami Pillai who was a great
apostle of Saiva Siddhanta, the Tamil non-Brahman religious tradi-
tion. Nallaswami was brought to college in the capital by Subramania
Iyer, apprenticed to him, sent to Subramania Iyer's cousin in Madura
to work as a junior and accommodated with a district munsiffship
while still with little legal experience. 81 The new men of consequence
in Madras city may have been especially fond of their relatives but
their personal contacts were much wider than family alone. Indeed, S.
Subramania Iyer, on being elevated to the High Court bench, had left
his huge legal practice not to a relative but to the brilliant V.
Krishnaswami Iyer, who was only a friend.82
An idea of the scope and importance of this patronage system is
provided by the correspondence of P. S. Sivaswami Iyer, lawyer,
University Senator, Legislative Councillor, Advocate-General and
77
Hindu 6 July 1910.
78
Hindu 28 A u g u s t 1913.
79
Hindu 15 and 19 October 1904.
80
He became a Recorder in the High Court. See 'C.P. on himself in C.P. by his
Contemporaries, p. 18. In 1894, V. Krishnaswami Iyer had especially requested
Masilamani Pillai's appointment to the Congress executive committee. 'Proceed-
ings of the General Purposes Committee of the 1894 Congress, 26 November
1894' in Register of Letters, Cards, etc., received from 2nd March 1894. Madras
Mahajana Sabha Papers. N.M.M.L.
81
K. M. Balasubramaniam, The Life of J. M. Nallaswami Pillai (Tiruchirapalle,
1965)* PP- i o - u .
82
TheHon'ble V. Krishnaswami Iyer. A Sketch (Trichinopoly, 1911), pp. 6-8.

237
 The emergence of provincial politics

Executive Councillor. Most of the letters which he received between

1904 and 1918, the period of his greatest influence, were requests
from other western-educated Indians for favours and help with their
careers. Whether the requirement was a job, a scholarship for a son, a
lucrative University examinership, interference in the management of
a college or school or preferment for promotion, Sivaswami Iyer was
clearly regarded as one of the principal individuals to approach.
Judging by the replies, he was not slow to use his position to aid those
whom he favoured.83 The elaboration of this patronage system created
a network of interested contacts which stretched across the province.
The centre of the network, however, lay in the houses of a handful of
men in Madras city. The aspirant for office or place had to come to
terms with these few 'boss-alls and bang-alls' if he wished to have
any success in the professions of the western educated. 84
Between the 1890s and 1916, the struggle for control of the nodes
of this patronage system was central to the development of provincial
politics. Naturally, some members of the western-educated commun-
ity were better able than others to make connections at the right
places; and those who failed were likely to try as hard as they could to
83
F o r example: N. R. Panchapagesan to Sivaswami Iyer 29 August 1912, which
warns t h e receiver not to place the son of sub-judge D. Vengopa Rao in the
Registration Department; P. R. Narayanaswami Iyer to Sivaswami Iyer 9 August
1912, which asks for t h e gift of a lectureship in t h e Madras Law College; T . S.
Balakrishna Iyer to Sivaswami Iyer 10 August 1912, which asks for a job for the
writer's son; T . Ananda R a o to Sivaswami Iyer 3 July 1913, which asks for a
deputy-collectorship for the writer's grandson; W. B. Patwarden to Sivaswami
Iyer 28 January 1909, which asks for a University Examinership; P. P. Pillai to
Sivaswami Iyer 8 March 1913, which asks for jobs for the writer's son and son-in-
law; R. Srinivasa Iyer to Sivaswami Iyer 30 March 1913, which asks for a place
in t h e Salt Department for t h e writer's son; M. Thiruvenkatachari to Sivaswami
Iyer 22 M a r c h 1908, which asks for a government place for the writer's nephew;
A. S. Vaidya to Sivaswami Iyer 17 March 1908, which recommends a third
party for Chief Examinership in Tamil; R. Ramachandra Rao to Sivaswami Iyer
18 March 1914, which asks for a post for the writer's cousin; T . Varadarajulu
Naidu to Sivaswami Iyer 23 August 1914, which asks for a post for the writer's
son. P. S. Sivaswami Iyer Papers. N.A.I. T h e r e are dozens more like this. T h e
character of the m e n listed above indicates that they did not importune Sivaswami
Iyer out of ignorance of t h e way that t h e system worked: Ananda R a o , R.
R a m a c h a n d r a R a o and T . Varadarajulu Naidu were themselves government
servants, P. P. Pillai and T . S. Balakrishna Iyer were prominent lawyers and
mofussil publicists. Among Sivaswami Iyer's most regular favour-seekers were
his two brothers, w h o were sub-judges.
84
D r T . M . Nair, w h o detested the system, once described it as operating on 'the
principle of the best-backed'. Hindu (weekly) 9 March 1916.

238
 The emergence of provincial politics

shift those who had succeeded. Moreover, the peculiar, almost

conspiratorial, nature of political life in Madras city led to the forma-
tion of specialised political groupings designed to extract and distri-
bute the patronage. One of the most celebrated of these groupings was
popularly known as 'the Mylapore clique' after the suburb in which
most of its prominent members lived. It was composed, across our
period, of a succession of lawyers and administrators who were on
close personal terms, who met regularly in each others' houses, put
forward the same demands in Congress, the press and on the streets
and aided each other in seeking office. Each generation in Mylapore
picked its successor and brought it up through its favour. In the late
1880s, we could take its leaders to be V. Bashyam Iyengar, S. Subra-
mania Iyer and R. Ragunatha Rao; 85 by the turn of the century their
apostolic successor V. Krishnaswami Iyer was in command; follow-
ing his death, his old school friend P. S. Sivaswami Iyer and close
legal and University associates G. A. Natesan and L. A. Govindaraghava
Iyer achieved prominence and were joined, at the time of the First
World War, by C. P. Ramaswami Iyer.86 Behind this leadership group
came a bewildering collection of lawyers, teachers, bureaucrats and
journalists tied together by personal contact and the hope of rewards.
The organisation of the clique was loose: personal and career rivalries
could split it at a moment's notice and disappointed courtiers could
move quickly into opposition to it. In general, however, it may be said
to have represented at any given time those who were most successful
at the delicate game of capturing government favour.
Permanently in opposition to the Mylapore clique, and attracting
to their flag at various times many old but disappointed Mylaporeans,
were a group known to contemporaries as the 'Egmore' clique. They
were even less homogeneous than, and, indeed, existed only as a
counterweight to, Mylapore. At the centre of the clique sat C. Sankara
Nair whose own personal system of influence was, of course, of exactly
the same type as that operated by the leaders of Mylapore. The reasons
for his disaffection were several: there was professional rivalry
85
R. Ragunatha Rao, a Maratha Brahman from Tanjore district, was a career
civil servant who became deputy-collector of Madras city and diwan of Baroda.
T. Madhava Rao, a diwan of Mysore, was his cousin; T. Ananda Rao, also a
diwan of Mysore, and R. Ramachandra Rao, a Collector and, later, a Secretary
of the Education Department, were his nephews.
86
The informal yet impenetrable character of the Mylapore clique can be seen in the
description which Sivaswami Iyer gave of Mylapore at his arrival in the city. See
K. A. Nilakanta Sastri (ed.), A Great Liberal Speeches and Writings of Sir P. S.
Sivaswami Aiyar (Bombay, 1965), p. 254.

239
 The emergence of provincial politics

between himself and the men of Mylapore; his own connections in

the European bureaucracy had been different from those of Mylapore;
as a Malayan" he was culturally isolated in a Tamil city during a period
when cultural contacts often led to political contacts. In collaboration
with him were men who had failed to make much headway with the
Mylapore bosses and whose alternative connections were weak. Most
prominent among them was Kasturi Ranga Iyengar whose brother
(a senior government servant in the 1880s and 1890s) had quarrelled
with the Mylaporeans and, during the militant Congress agitation,
had been ridiculed by them as a British puppet.87 Kasturi Ranga Iyen-
gar, who had qualified as a vakil in 18 84, had been unable to build a legal
practice in the city and had spent nearly a decade in the mofussil. He
returned to the city in 1894 but> o n c e a S am > found himself starved
of valuable cases, which fell into the pockets of the Mylaporeans, and
was forced to give up the law for journalism.88 He was understandably
resentful of the men who held back his career. A third Egmore leader
was T. Rangachari, from the same village in Tanjore as Kasturi Ranga
Iyengar. Although regarded as one of the most able lawyers of his
generation, he could secure relatively few of the plums of civil litiga-
tion and concentrated his attention on the much less lucrative criminal
bar. 89 Kasturi Ranga Iyengar and T. Rangachari certainly had contacts,
familial and otherwise, with the central mechanisms of government,
the law and education, and picked up Mylaporeans who were spun
off by factional conflict, but their influence and importance were
less pervasive.
Examples of the rivalry between Egmore and Mylapore litter
the history of Madras politics. Even in the 1880s, when personal
feuds were more usually subordinated to common ends, C. Sankara
Nair had clashed with V. Bashyam Iyengar and S. Subramania Iyer
in a most acrimonious debate in the Madras Vakils' Association over
the place of barristers in the appeals court. 90 This had led to a long
and bitter correspondence in the press. The 1894 Congress was
87
Ibid., p . 1 1 3 .
88
Narasimhan, Kasturi Ranga Iyengar, pp. 23-4; V. K. Narasimhan, Kasturi
Srinivasan (Bombay, 1969), p. 7. A typical example of what it meant to be 'out' in
city politics can be seen in the fact that although Kasturi Ranga Iyengar several
times stood for election to the Senate, he was always beaten by a Mylaporean.
89
T . Rangachari rarely appears on t h e list of vakils retained in major High Court
civil cases. I n 1924, at t h e height of his career, he was earning only Rs 40,000 p.a.,
compared t o t h e Rs 2,40,000 of t h e leader of t h e Madras bar, K. Srinivasa
Iyengar. See H o m e Public File 953 of 1924. N.A.I.
90
Menon, C. Sankaran Nair, p. 18.

24O
 The emergence of provincial politics

nearly wrecked by faction. There was a fight between Kasturi Ranga

Iyengar and the Mylaporeans P. R. Sundara Iyer and V. C. Desikachari
over the printing and distribution of tickets; T. Rangachari and P. R.
Sundara Iyer quarrelled over the membership of the subjects' commit-
tee; after the Congress session, S. Subramania Iyer resigned from the
executive committee, accusing some of its members of malpractice;
the general purposes committee refused to meetT. Rangachari's claims
for expenses; and the Hindu newspaper, which was already near to the
Egmore clique, was accused of presenting a false bill.91 The internecine
strife continued unabated in the succeeding years. When it was
rumoured that Sankara Nair was to be made an Executive Councillor,
petitions ofprotest arrived at the Judicial Department fromMylapore. 92
In 1905, Sankara Nair, Kasturi Ranga Iyengar and T. Rangachari
joined forces to buy the Hindu, which had reached bankruptcy, and used
it to denounce the antics of Mylapore in general and V. Krishnaswami
Iyer in particular. 93
The reforms of the 1890s, then, had created openings which
various members of the western-educated community in Madras city
were trying to fill. Their struggles for place were creating factional
divisions. Both Egmore and Mylapore, however, can be defined in
simple and conservative terms: they were trying to replace each other
at the centre of a system the rules of which they both accepted. Yet
outside them, and still within the world of the western educated, there
existed forces much less coherent and much more dangerous. In
Madras city and every mofussil locality, the prizes in education, the
law and government service were passing to the men who were 'well
connected' to the patronage brokers. Not surprisingly, there were a
great many men who were not connected to any of the dominant
personalities of the new politics and who began to suffer. Soon, they
started a very vocal denunciation of the whole system. They leant
more towards Egmore than Mylapore but must be seen as distinct
from it. Whereas, in Sankara Nair, Egmore had a man deep inside
patronage politics and, in Kasturi Ranga Iyengar and T. Rangachari,
91
'Proceedings of the Executive Committee of the 1894 Congress', 25 September and
10 November 1894,12 January, 7 and 30 March 1895; 'Proceedings of the General
Purposes Committee of the 1894 Congress', 4 September 1894 m Register of
Letters, Cards, etc., received since 2nd March 1894. Madras Mahajana Sabha
Papers. N.M.M.L.
92
Letter of K. Ramamurthi in Hindu 12 January 1918.
93
See V. S. Srinivasa Sastri to V. Krishnaswami Iyer, 27 July 1911. V. S. Srinivasa
Sastri Papers. N.A.I.

241
 The emergence of provincial politics

men around its fringes and with certain kin and social connections
inside it, this collection of the frustrated had no contacts with it at all.
They were much more prepared to wreck the entire game if they
could not get Mylapore out.
Among the militant discontented was T. M. Nair, the Madras city
doctor whom we discussed when examining the Madras Corporation.
He spent much of his time denouncing 'the wire-pullers' who kept
him out of Legislative Council politics and who, ultimately, took
from him even his Corporation seat. Yet most of the opponents of
the dominion of Madras city leaders came from outside the capital.
Often they were the local enemies of men associated with Mylapore.
Publicist and professional life in the Northern Circar districts of
Godavari and Kistna, for example, was in the hands of a group of
lawyer-educationists gathered around Nyapathi Subba Rao, a Rajah-
mundry lawyer, and including Puranam Venkatappayya, the leader of
the Kistna bar; K. Viresalingam, a celebrated social reformer; K.
Perrazu, a Cocanada lawyer; and M. Ramachandra Rao, another
Rajahmundry lawyer. This clique was closely tied through personal
connections to Mylapore94 whose influence it used to hold and
extend its local political position. It enjoyed a large share of the
legal patronage which came from Madras city, moved on easy terms with
a local officialdom which increasingly was coming under Mylapore's
control of careers, and was in receipt of many favours in local educa-
tion from the University. Moreover, it monopolised all Legislative
Council elections in Andhra. Its opponents in the Circars found them-
selves harassed repeatedly by its ability to bring force from the
capital to bear on the local situation. For more than twenty years,
the main rival of N. Subba Rao and K. Viresalingam in Rajahmundry
affairs was Y. L. Narasimhan, a Sristikarnam lawyer and political
entrepreneur from Ganjam who had settled in the district. Until the
mid-1890s, however, Narasimhan had been able to hold his own
against them in a competition for publicist leadership, for his resources
of local magnate support were at least as great as theirs. 95 Follow-
ing the reforms at Fort St George, however, he was progressively
outmanoeuvred. His reputation was rocked by a series of abortive
94
N. Subba Rao and M. Ramachandra Rao were Madras city educated and class-
mates of the leaders of Mylapore; P. Venkatappayya's brother, Nagabhushanam,
was a city resident and long-time secretary of the Madras Mahajana Sabha; K.
Viresalingam had contacts with the Madras Social Reform Association, in which
certain Mylaporeans were interested, and had spent some time teaching in the
capital.
95
I am grateful to D r John Leonard of the University of California for this informa-
tion.

242
 The emergence of provincial politics

but expensive prosecutions brought against him by the Judicial

Department;96 his place in the municipal council was threatened
by constitutional changes in the council, which Subba Rao had
suggested to the Local and Municipal Department; 97 his little empire
of private schools was challenged by the University's scarcely dis-
guised preference for K. Viresalingam's educational projects.98
Narasimhan's closest local associates also felt Mylapore's pressure. T.
Prakasam of Rajahmundry, for example, who owed his education to
Narasimhan's patronage, had a lucrative hold which he possessed
over a district munsiff broken by a Judicial Department investiga-
tion,99 and his 1903—4 regime as chairman of the municipality
pestered by investigations of his conduct by the Local and Municipal
Department.100
In Kistna, P. Venkatappayya came to enjoy a massive hold over
local educational institutions, the bar and the municipality and to
extirpate virtually all opposition to his dictatorship of public affairs.
Naturally, this position would not be accepted for long by other
educated publicists in the leading district town and, by the turn
of the century, resentment against him was deeply felt in a group
of young western-educated professionals led from Masulipatam by the
doctor B. Pattabhisitaramayya, the lawyer Konda Venkatappayya and
the journalist Mutnuri Krishna Rao.101 Narasimhan and Prakasam in
Rajahmundry and Pattabhisitaramayya and Krishna Rao in Masuli-
patam were, logically, only committed to a battle with the Subba
Rao-Venkatappayya clique for place in the locality. However, because
of the latter's use of power derived from the provincial level, it was
not long before the fight was carried to the provincial level. In addition
to attacking the clique in elections to municipalities and school boards
and at local public meetings, Prakasam and Krishna Rao helped to
found newspapers which made their message much more general.
Particularly through Kistnapatrika (established 1904), they denounced
96
Hindu 17 December 1904.
97
G.O. 1029 (L and M, M) dated 27 June 1895; G.O. 5 (L and M, M) dated 7 January
1896. T.N.A.
98
F o r example, Viresalingam spent many years o n the University's Telugu Textbook
Committee and was able t o prescribe many of his o w n works as set texts. F o r a n
account of Mylapore's attempts t o interfere in t h e Cocanada College, see 'Extract
from t h e Memoirs of D r B . Pattabhisitaramayya' in Brahmarishi Dr Sir Raghupathi
Venkata Ratnam Naidu Birth Centenary Souvenir. Mahanavarmi, 1962 (n.p., n.d.).
99
Collector of Godavari to Secretary, Local and Municipal Department, 30 October
1903 in G.O. 40 (L and M, M) dated 8 January 1904. T.N.A.
100
Ibid.
101
Hindu 8 and 11 May 1907.

243
 The emergence of provincial politics

the present form of British government, which permitted Mylapore

so much influence, and the current moderate tenor of Congress
politics, which so suited Mylapore's collaborationist role.
Similar developments could be seen in a number of other areas.
N. K. Ramaswami Iyer, for example, was a Tamil Brahman lawyer
in Chittoor district who, between 1900 and 1904, was involved in a
battle with the family of the great Mylaporean L. A. Govindara-
ghava Iyer for local pre-eminence.102 In 1904, he took his struggle
to the provincial level by demanding, at the Provincial Congress
Conference, that the Congress constitution be altered to weaken
the hold Madras city — i.e., Mylapore — had on it. 103 Naturally,
he was unsuccessful. In 1905, he moved to Tanjore where he found
that local dominance belonged to R. Raghunatha Rao, the doyen of
Mylapore in the 1880s, who was in semi-retirement but still pre-
served his Mylapore connections. As in Chittoor, and as with the
T. Prakasam-M. Krishna Rao groups in Andhra, he began to form
associations and publish newspapers which attacked the present
construction of the government and of the Congress.104 In Salem C.
Vijayaraghavachari formed the kernel of opposition to Mylapore.
In spite of great notoriety occasioned by his prosecution of the
Secretary of State in 1883, 105 by his victory in three consecutive
Legislative Council elections and by his oratorical skills, he had never
forged connections with Mylapore or the centre of government and,
by the early 1900s, had nothing to show for his eminence. Indeed, in
local politics he was eclipsed by T. Subramania Iyer, the close friend
of Sir P. S. Sivaswami Iyer, who attracted much central patronage; 106
while, in provincial politics, he received no office or decorations.
He was an uncompromising critic of government, Congress and
Mylapore.
The reforms of the 1890s had wrought deep divisions in the more
or less united front which the western-educated community had
presented to the British in the 1880s. The transfer of a limited power
102
Hindu 29 February 1908.
103
Report of the 12th Madras Provinical Conference held at Ranipet in May 1904
(Madras, 1904), pp. 103-7.
104
Hindu 17 N o v e m b e r 1905, 20 April 1907.
105
This prosecution followed his acquittal in a case arising out of the Salem riots of
1882. He won Rs 100 damages and became a celebrity overnight. Hindu 6 May
1884.
106 --p S u b r a m a n i a Iyer was o n e of Sivaswami Iyer's most persistent correspondents.
H e was Salem public prosecutor a n d municipal c h a i r m a n for several years a n d ,
d u r i n g Sivaswami Iyer's t e r m as Executive Councillor, was n o m i n a t e d president
of t h e Salem taluk board.

244
 The emergence of provincial politics

to certain members of the professions had spread tensions across the

whole of the province. These tensions lay at the heart of the Congress
troubles of 1906-8. Lord Morley, the Liberal Secretary of State,
undertook to grant constitutional reforms and the Moderate Congress
leadership, understandably, wished to impress him with its loyalty,
responsibility and solidarity. V. Krishnaswami Iyer led the Madras
Congress and was closely allied in All-India politics to Gokhale in
Bombay and Surendrenath Bannerjee in Bengal. It was natural that
the enemies of Mylapore in Madras should use Mylapore's need for
moderation as a stick with which to beat it and its system. They
gravitated towards the opponents of Gokhale and Bannerjee in
Bombay and Bengal - particularly B. G. Tilak and B. C. Pal. The
confrontation between these fusions of Moderate and of Extremist
elements and the debacle of the Surat Congress session in 1907 are
well-known and need little further elucidation.107 The Tilak—Pal
'party' attempted to give the Congress a radical, populist and anti-
British stance and to demand an impossible amount of reform just
when the Gokhale-Krishnaswami Iyer-Banner jee leadership required
such postures and programmes least.
In Madras city, much of the Extremist disturbance was due to the
work of G. Subramania Iyer, one of the oldest enemies of Mylapore.
G. Subramania Iyer had been a prominent nationalist agitator in the
1870s and 1880s, a founder of the Hindu newspaper, of the Madras
Mahajana Sabha and of the Congress. But he had quarrelled with
his colleagues over social reform and had virtually outcasted him-
self by allowing his widowed daughter to remarry. As a result he had
been excluded from the inner sanctum of Mylapore, he had failed to
be made a Congress president — which his work for the early Congress
deserved — and he had been unable to enter the tight world of Legis-
lative Council and bureaucratic politics.108 Through the 1890s and
early 1900s, he remained a penurious publicist while his previous
associates became powerful politicians. In 1907, he aimed his
polemic against Mylapore and drew a following from young and
poor members of the intelligentsia, students, mill workers, jutka-
wallahs and other similarly frustrated social elements.109 In the
107
See J o h n s o n , Provincial Politics and Indian Nationalism.
108
S.A. Govindarajan, G. Subramania Iyer (New Delhi, 1969), pp. 37, 48-54.
Between the mid-1890s and mid-1900s, Subramania Iyer several times failed to
win election to the Legislative Council and Senate.
109
Home Political B October 1907, Nos 80-7; Home Political B December 1907,
Nos 2-9; Home Political B January 1908, Nos 19-26; Home Political B May
1908, Nos 36-43. N.A.I.

245
 The emergence of provincial politics

manner of every other Indian agitator of our period, he sought also

to tie religious revivalism to his cause.110 His support was noisy and
volatile but lacked influence in society. It was used by the Egmore
leaders T. Rangachari and Kasturi Ranga Iyengar to challenge
Mylapore in the Madras Mahajana Sabha, which formed the Congress
executive. Election meetings were packed with it and several debat-
ing sessions broken up by its fury.111 But it proved a very unsatis-
factory following in the type of politics played by Egmore: it was too
wild to be controlled and so disreputable that it alienated important
opinion. T. Rangachari, Kasturi Ranga Iyengar and, ultimately, even
G. Subramania Iyer found it necessary to jettison their popular sup-
port when it became violent and threatened to bring down the police
on their heads.112 Further, of course student and worker mobs
proved useless against Krishnaswami Iyer's patronage power and
influence.113
In the mofussil, the Extremist leaders attracted much the same
kind of following. In Tanjore, N. K. Ramaswami Iyer mixed his
religious polemic with appeals to impoverished literati and students.114
In Tuticorin, V. O. Chidambaram Pillai's work seemed more sub-
stantial, for he used a base of pre-existing racial and economic ten-
sion. But many of those involved in his main enterprise, the Swadeshi
Steam Navigation Company, proved to be politically moderate and his
Extremist activities led to his dismissal from the project. His personal
following was drawn from students, workmen and those susceptible
to religious invective.115 In Salem, C. Vijayaraghavachari was
relatively quiet until late in 1908, but he still posed sufficient a
threat for Krishnaswami Iyer to have him removed from the A.I.C.C.
in 1907.116 In Andhra, the T. Prakasam-M. Krishna Rao leadership
was very active. It brought B. C. Pal from Bengal for a lecture tour
in 1907 and campaigned vigorously.117 Attempts were made by it to
110
Home Political B December 1907, Nos 2 - 9 . N.A.I.
111
Ibid.; Hindu 7 December 1907, 13 and 27 December 1908.
112
Home Political B December 1907, Nos 2 - 9 . N.A.I.; Hindu 13 January 1908.
113
Even in the Madras Mahajana Sabha, it could not command sufficient votes to
oust the Krishnaswami Iyer faction. 'Proceedings of a Meeting of Members', 28
October 1907, 7 August 1908 and 20 November 1908. Madras Mahajana Sabha
Papers. N.M.M.L.
114
Hindu 20 April 1907.
115
Swadesamitran 1 and 18 July 1906. R.N.P.; Hindu 17 March, 2 April and 9 June
1908; P & J File 993 of 1908.1.O.L. For a detailed discussion of the Tuticorin
episode, see my 'Political Change in the Madras Presidency 1880—1921', Un-
published Fellowship dissertation, Trinity College, Cambridge, 1971.
116 117
Hindu 25 July 1907. Hindu 2, 3, 4 and 8 May 1907.

246
 The emergence of provincial politics

found a national college and to set up permanent political institu-

tions. 118 Once more, appeals to students and workmen and to popular
religious sensibilities formed the fundament of its local agitations,
but it had more success than elsewhere. The close rural-urban con-
nections in deltaic Andhra enabled it to channel the disturbances into
the countryside. Further, many of the Extremist leaders were able
to find financial support from groups of urban Komatis who had
often backed them personally in local religious and municipal
politics.119 Andhra Extremists were able to build a considerable
movement, the volatility of which may be seen in the Cocanada
riot.120
In spite of their ability to find support in a few localities, however,
the Madras Extremists did not seriously challenge the position of
Mylapore. Although mobs and popular polemic could embarrass the
Madras city leaders, and undermine their claims that their Congress
represented the opinion of India, they were the wrong weapons in
this war. The institutions which could make provincial leaders res-
ponsible to popular opinion did not exist. In consequence, the
the Madras Extremists had little general success in rousing to their
cause large numbers even among the western educated. The enor-
mous patronage control exerted by Mylapore gave it great authority
over the educational, legal and administrative sections of society
and very few men who were seriously interested in these were pre-
pared to flaunt it openly.121 Through his hold on the central Congress
machinery, Krishnaswami Iyer was able to exclude the Extremists
from participation in the Congress and thus deny them a vehicle
for the expression of their frustrations. Under the 1908 Allahabad
Convention, the Mahajana Sabha was replaced as the Congress exe-
cutive by a Provincial Congress Committee nominated by Krish-
naswami Iyer himself; access to the 1908 Madras Congress session
118
Hindu 16 May 1907.
119
See Venkatarangaiya, Freedom Struggle in Andhra Pradesh, 11, 177—311.
120
Ibid., pp. 207-10.
121
The leading lawyers in most of the mofussil bars rushed to Krishnaswami Iyer's
aid in opposing the Extremists. B. N. Sarma of Vizagapatam, N. Subba Rao of
Rajahmundry, T. T. Viraraghvachari of Chittoor, P. Kesava Pillai of Anantapur,
P. Siva Rao of Bellary, A. Subramania Iyer of Madura, A. Sundara Sastri of
Tinnevelly and T. S. Balakrishna Iyer of Coimbatore all offered their support to
the Moderate cause. For views of the Moderate - Extremist confrontation, which
concentrate on the personalities rather than the policies and on which this
account is based, see V. S. Srinivasa Sastri's letters to G. K. Gokhale and V.
Krishnaswami Iyer between 1907 and 1909 in V. S. Srinivasa Sastri Papers.
N.A.I.

247
 The emergence of provincial politics

was made difficult and the attendance was small;122 the district and
mofussil machinery of Congress was run down to prevent dissident
locals from capturing it. 123 By 1909, it was very obvious that
Mylapore had weathered the storm without damage and had destroyed
the organisations of those opposed to it. In order to remain in public
affairs, Kasturi Ranga Iyengar and T. Rangachari had to enter
Krishnaswami Iyer's Madras P.C.C., where they were powerless.
Of course, the Extremist—Moderate split was not fought in the
press and from the public platform on the basis of 'ins' against 'outs',
which we have suggested was its reality. Great issues were raised and
ideological principles held almost to the death. Yet it would be very
difficult, looking not at the specific events of 1906—8, but at the
careers of the men who were involved in them, to argue that the
issues entered the world of politics of their own volition rather
than as factors facilitating actions taken on other grounds. Mylapore,
at various times in our period, stood for a number of conflicting
principles over the presence and the role of the British in Madras.
In the 1880s, it demonstrated a strong and trenchant criticism of the
British bureaucracy. Between the 1890s and 1916, it was an avid
defender of the glories of raj and put down an agitational movement
similar to the one which it had organised a few years before. In
1916, it was to go over to savage criticism once more and to threaten
civil disobedience against the British. Mylapore was playing a strategic
game in order to widen its grasp of government power and it required
different tactics at different times. Naturally, its enemies were
forced into considering their tactics appropriate for each stage. C.
Vijayaraghavachari, for example, is best remembered for the extreme
stands he took at various times against the mendicant and Anglophile
postures of Mylapore. Yet he was also capable of allying with those
British elements which felt their position threatened by the expan-
sion of Indian power. In Legislative Council elections, he voted for
European missionaries and businessmen in preference to Mylapore
candidates 124, and, when defeated in a 1916 Imperial Legislative
122 y Krishnaswami Iyer to G. K. Gokhale, 5 October 1907. V. S. Srinivasa Sastri
Papers. N.A.I.; Report of the Proceedings of the Twenty-Fourth Indian National
Congress held at Madras on the28th, 29th and30th December 1908 (Madras, 1909);
Home Political B January 1909, No. 108. N.A.I.
123
See K. R. Guruswami Iyer to V. Krishnaswami Iyer, 2 October 1908; C. Y.
Chintamani to V. Krishnaswami Iyer, 3 October 1908; T. T. Viraraghavachari to
V. Krishnaswami Iyer, 11 November 1908. V. Krishnaswami Iyer Papers.
N.A.I.
124
Letter of T. V. Gopalaswami Mudaliar in Hindu 15 August 1916.

248
 The emergence of provincial politics

Council election, he received only European votes.125 T. M. Nair,

on the other hand, is best remembered for his pro-British sentiments
which manifested themselves in his dress, his denunciations of Hindu
tradition, his alliances on the Corporation with Europeans and,
ultimately, his part in founding the loyalist Justice Party in 1916.
Yet during the Extremist—Moderate split he made common cause with
C. Vijayaraghavachari and N. K. Ramaswami Iyer in attacking the
Moderates, even though this involved him in a movement which was
attempting to destroy the basis of the British connection.126
The responses of the professional opponents of Mylapore were
dictated to a considerable extent by Mylapore's own position at any
given time and by the availability of the resources of political protest.
To attack Mylapore for its Anglophobia, particularly during periods
when it was involved in agitation, opened out access to missionary,
Indian Christian and British resident support which was disturbed by
the growth of Mylapore's influence. To attack it for its Anglophilia,
during periods when it was co-operating with government, provided ac-
cess to areas of society in which dissatisfaction with the government in
general was felt. Jutkawallahs, subject to constant police harassment,
factory hands, in the difficult state of assimilation into an urban
proletariat, and students, without material worries, could be drawn
into violent demonstration without much difficulty or particular
cause. Equally, popular religious revivalism, with its emphasis on
parades, preaching and polemic, provided a bank which could be
looted whenever anti-government agitation was necessary. The British,
however careful they might be to avoid disturbing the religious
sensibilities of their subjects, were not Hindus and could be character-
ised as enemies of Hinduism.127 In the continuing struggle with
125
C. Vijayaraghavachari to P. Kesava Pillai, 15 July 1916. P. Kesava Pillai Papers.
N.M.M.L.
126
V. S. Srinivasa Sastri to G. K. Gokhale, 2 March 1909. V. S. Srinivasa Sastri
Papers. N.A.I.
127
However, it is necessary to note that there was no logical connection between
religious revivalism and Extremism, nor, indeed, between Christianity and
political opposition to Mylapore. The men of Mylapore were deeply interested in
and great patrons of Hindu and Indian cultural revivalism. Most of them were
associated with Mrs Besant, whose Theosophical federation was certainly the
most powerful revivalist organisation in the South. Between 1906 and 1908, she
was at one with Mylapore in opposing the programme of the Extremists. The link
between political agitation and religious revivalism was simply that, in the locali-
ties, revivalist associations possessed organised followings of a semi-fanatical
nature. By associating political symbols with those of religion, political agitators
could tap these pools of support for any purpose they chose - as Mrs Besant

249
 The emergence of provincial politics

Mylapore, its enemies had had to recruit their following where they
could find it.
The manner in which personal oppositions were continuous, and
only the issues changed, was well brought out by the developments
which followed the defeat of the Extremists. Following Krishnaswami
Iyer's victory in 1908, the main arena of popular political contro-
versy — the Congress — was denied to his opponents. No longer could
they launch attacks on the Mylapore system from inside a provincial
political organisation. They began to turn, therefore, to a number of
peripheral areas of social activity which they politicised and brought
into the fray. It was more than coincidence that the first signs of the
Andhra movement began to appear in 1909, the year after the Alla-
habad Convention, and that its leaders were the rump of the Andhra
Extremists, T. Prakasam, M. Krishna Rao, B. Pattabhisitaramayya,
Konda Venkatappayya and the Tamilian N. K. Ramaswami Iyer.128
The Andhra movement, more specifically the movement for a separate
Andhra province, grew out of the Telugu vernacular revival which
had been developing with considerable energy in the towns of the
Andhra deltas from the 1870s. But the fact that it took on a political
aspect at this time was due less to any logic in the process of its
development than to its seizure by Mylapore's opponents. The cre-
ation of an Andhra province would break the connections between
Mylapore and the Circar towns, which the Andhra Extremists found
so debilitating. In the Circars, the main enemies of the movement were
Mylapore's placemen in the area-theN. SubbaRao-M. Ramachandra
Rao group.129 They were involved in vernacular revivalism at least
as much, if not more, than the Andhra movement leaders but had no

herself was to show when she became a politician in 1916. Similarly, by the later
nineteenth century, not all Christians and missionaries were opposed to the
advance o f Indians, even of Mylapore Hindus. P. S. Sivaswami Iyer, for example,
enjoyed the friendship o f Father Bertram, a Jesuit headmaster, while several
Indian Christians, such as M. S. Devadoss the High Court Judge, were on the
fringes of the Mylapore network. Tension between missionaries and Christians
and the Mylapore clique, w h e n it occurred, was not so much the result o f the
conflict o f ideologies as of practical political interests. T h e group of Protestant
missionaries w h o had run the University naturally resented losing their influence
to anybody; since it was Mylapore w h o took their position from them it was
Mylapore they disliked.
128
See K. Venkatappayya, The Andhra Movement (n.p., n.d.); also, The First Andhra
Conference held at Bapatla in 1913 (Bezwada, n.d.).
129 Evidence o f N . Subba R a o in Royal Commission on the Public Services in India,
Appendix, 11, 296. P.P. 1914, xxi; Hindu 14 May 1915.

250
 The emergence of provincial politics

desire to destroy their existing and lucrative political contacts. In the

same way, it was N. K. Ramaswami Iyer again, this time in conserva-
tive Tanjore, who organised and politicised the sentiments of orthodox
Brahmans against social reform.130 His Varnashramadharma move-
ment was aimed specifically at the mildly reformist views of most of
the leading men of Mylapore.
Of course, the fact that these cultural and religious movements were
politicised in response to factional struggles implies that, once the
factional alignments or tactics changed, they could be depoliticised.
This clearly happened to the Andhra movement. Between 1910 and
1917, the campaign for an Andhra province grew steadily more
vehement and deepened its following. It was never in a position to
challenge Mylapore where it mattered - in the councils of govern-
ment — but it created an impression of popular power and increased
the reputations of its leaders. By 1917, however, Mylapore itself had
gone over to anti-British agitation and was prepared to compromise
with its old enemies. Pattabhisitaramayya, Prakasam, Konda Venka-
tappayya and M. Krishna Rao were brought back into provincial
politics and organised the Home Rule agitation in Andhra.131 More
than this, Mylapore (under the influence of Mrs Besant) allowed them
their own provincial Congress committee, which gave them an
independent base in the presidency.132 By 1919, the old Andhra
Extremist leadership had been rehabilitated and was now allied to
and able to use the Mylapore system of influence. Konda Venkatap-
payya, for example, became the first man not directly associated with
N. Subba Rao to win a Legislative Council election in the Andhra
deltas.133
The rise of the Prakasam-Venkatappayya faction naturally
changed its political complexion. It was no longer in the wilderness
but was at the heart of provincial affairs. The creation of a separate
Andhra province now would not only sever its excellent connections
with Madras city but reduce the scope of its influence. The Andhra
130
T o accomplish this, Ramaswami Iyer was converted overnight from a position o f
publicly stated agnosticism to one of militant Hinduism. Ramaswami Sastri,
Professor Sunderama Ayyar, pp. 8 0 - 6 ; Hindu 4 M a y 1 9 1 0 , 15 July and
7 September 1916.
131
Konda Venkatappayya and Pattabhisitaramayya were o n Mrs Besant's payroll.
'Statement o f Affairs of the H o m e Rule League, 10 January 1919'. T.S.A.
132
This was run from Guntur town, the headquarters of the Andhra leaders. See
K. Venkatappayya to C. P. Ramaswami Iyer, 17 July 1918 in A.I.C.C. Papers.
File 2, Part 3, of 1918. N.M.M.L.
133
More correctly, Venkatappayya now was on the Subba R a o - M y l a p o r e network.

251
 The emergence of provincial politics

agitation was quickly strangled and its organisations used to campaign

for a united Indian nation. So suddenly did this volte-face take place
that many of the younger, student Andhra activists could not under-
stand what was happening:

The heat of the agitation which we set up eight years ago has been extin-
guished, if not almost extinct, [sic] I am quite sure that a few more years
agitation would get us not only an Andhra Province and an Andhra Univer-
sity but all we desire to have. But we must wake up our leaders before it is
too late." 134

However, with their leaders determined to sleep, and the organisa-

tions denied to them, the activists could do nothing. Although Telugu
vernacular revivalism continued to grow in importance, the Andhra
political movement shrunk to insignificance during the 1920s.
The provincial political controversy which developed out of the
reforms of the 1890s had two characteristics. Firstly, whatever the
avowed aims and public statements of its participants, their behaviour
makes it clear that their ends were to capture or destroy the central
points of influence in the bureaucratic—legal—educational machine.
Consequently, the formation of parties or groupings among them was
factionally determined and the political methods each adopted were
conditioned largely by the likelihood of factional success. The Andhra
Extremists were prepared to switch into a cultural movement, and to
switch out of it again when it became inconvenient. Similarly, in
Madras city in 1907 and 1908, T. Rangachari and Kasturi Ranga
Iyengar organised and politicised the sentiments of 'the mob' only
to abandon them later and to accept nomination to Krishnaswami
Iyer's P.C.C. Similarly, again, we might note the conduct of C.
Vijayaraghavachari who, in 1907, ran with the All India Extremists,
then offered to accept the Allahabad Convention when he believed
Krishnaswami Iyer was about to offer him a P.C.C. seat, and only
rejected the Convention when he found that Krishnaswami Iyer
would not have him. 135 Or again, T. M. Nair, who ditched his Euro-
pean friends for the Extremists in 1909, then went back to them in
the Justice Party in 1916. The movements and causes which these
politicians so rapidly espoused and divorced were important only
as aids to their wider political ambitions. And, indeed, it was only
by becoming an aid to advancement that these social or cultural
134
Hindu 1 March 1920.
135
C. Vijayaraghavachari to P. Kesava Pillai, 7 October 1908. P. Kesava Pillai
Papers. N.M.M.L.

252
 The emergence of provincial politics

movements achieved provincial political importance. Once the

Andhra leaders had dropped the Andhra movement, it ceased to carry
any political weight, and once T. Rangachari and Kasturi Ranga
Iyengar had thrown over 'the mob5 in Madras city, it too lost its
existence as a political force. The faction-fighting and the battles for
political place at the capital, therefore, were not incidental to the
political system. They were its essence, for it was only by being
drawn into the struggle between participants that broader social and
cultural considerations entered provincial politics.
The second characteristic of the controversy, however, was that
it concerned only the western educated. After magnate interests
had moved from agitational to constitutional methods of representa-
tion in the 1890s, they ceased to take part directly in provincial
political organisations. Except in one or two special areas,136 neither
the Moderate—Extremist confrontation nor, before 1914, the Andhra
movement elicited much response from them. The battle in the
presidency for provincial political control went on over their heads.
The reasons for their non-participation are not difficult to see. Of the
two political functions which the institutions of the bureaucracy, law
and education served, only one had been rendered politically conten-
tious by the reforms of the 1890s. Centralisation and Indianisation
clearly had altered the politics of career in the professions: they had
changed the answer to the vital question of who dictated career
success. In consequence, they had made the provincial politics of the
careers central to most western-educated men. However, the reforms
did not necessarily disturb the other function which the institutions
served - that of linking the capital to structures of local power.
Disturbance to this function would come only if the influence of the
capital were capable of biting into and significantly damaging the
bases of local power and if members of the Indian collaborationist elite
were capable of controlling the influence. Before the 1910s, neither
of these possibilities was real.
In spite of its movement towards greater interference in local
society, Fort St George still did not possess an administrative machine
which could carry its executive decisions deep into the locality. It had
to proceed by means of broad legislative fiat. This meant that it en-
acted policies which altered the legal framework around such resources
as forests, irrigation rights and rents but did not shift significantly
the level in the framework from which the crucial executive decisions
were to be made. As we have seen from the failure of direct village
136
Namely, Tuticorin and the Andhra delta towns.

253
 The emergence of provincial politics

officer and other reforms, the Secretariat simply did not have the
practical competence to use its own discretion in the execution of
policies nor to interfere in the workings of the locality from day-to-
day. For services in bending the new laws to fit local circumstance, the
local magnate still had to look to the semi-autonomous and malle-
able local official. The magnate came to Fort St George only to talk
about the desirability of a particular policy not about how it should
be run on the ground in his own case. Moreover, in these talks it was
extremely rare for one magnate, or even one type of magnate, to find
himself having to denounce or fight against another. In part, this
lack of faction was due to the very process of representation itself:
in government tribunals, most of the grievances which were expressed
were naturally grievances against government. In part also, it was
because Fort St George's drive for efficiency had as yet done no more
than sting the magnates in a few sensitive places. They were howling
in irritation rather than desperate pain. The government offensive
had by no means succeeded in redistributing local power nor even in
shaping the circumstances in which a redistribution could take place.
In consequence, the relevance of the central government to the
struggle for local power was still marginal.
It was these factors which helped to keep the local magnate out of
the provincial political battle. The British were willing to concede
to Indians a measure of control over the executive institutions but
decisions made in these were not, as yet, critical to the locality.
However, they were extremely hesitant to allow their subjects a
share in the making of legislative policy which, in any case, tended
to be directed from London and Calcutta. In consequence, the new
Indian constitutional collaborators remained as far outside the
actual elaboration of policy as they had ever been. Their continued
debilities meant that they still did not hold a power base in govern-
ment, which they could use at their discretion to harm magnate
interests in general or, more importantly, to raise up one magnate
against another. As a result, the individual aims and views of parti-
cular western-educated men were irrelevant to local magnates who
were not faced with having to choose between particular representa-
tives before they had been drawn into the Legislative Council or
senior bureaucracy. Whoever was elected or co-opted to the centre
could be hired post facto to put forward the well-known and general
grievances of magnate society - about the erosion of zamindari politic-
al and economic power, taxation and administrative interference.
He would be only too pleased to make these representations, which
gave him an income and demonstrated to the British his connections

254
 The emergence of provincial politics

with the real rulers of Madras. If he turned down a case, the cost to
him was much greater than that to his spurned supporters. And it was
a rare day indeed when one of his colleagues could not be persuaded
to take the brief in his stead.
The provincial political struggle, therefore, was not about the
nature of the interests which were to be represented to the British;
it was about who was to earn the money and achieve the prestige
which came from carrying out the representation. Most of the greater
local magnates could happily sit back, allow the battle to rage and
pick the eventual winners. As it happened, Mylapore's stranglehold
on the institutions of the law, education and bureaucracy guaranteed
its continuing success, and so the magnates, with few exceptions,
fell in behind it. However, had Egmore managed to displace it or
any had other group seized the centre over its head, it is difficult to
see any reason why the magnates should not have worked as easily
with the new provincial leadership as they had done with the old.
From the early 1910s, however, this gentle state of affairs began to
alter dramatically. The bolts in the government machine were tight-
ened quickly, shifting the balance between local and central power
towards the centre and giving Mylapore a new significance. We have
seen how important the new institutions of local self-government
became at this time, as they drew into themselves many of the func-
tions of government previously exercised autonomously by rural-
local bosses and urban magnates. We have seen also how connection to
the Secretariat was becoming a valuable prize to the factions who
were fighting for control of local boards and municipal councils. This
connection was expressed through the decisions of the Local and
Municipal Department in Fort St George, which itself was under the
supervision of a member of the Governor's Executive Council. In
1909, Lord Morley, continuing the policy of Indianisation which
had been started in the 1880s, commanded the Madras Government
to appoint a native to its Executive Council. Among the portfolios
handed to him was that of the Local and Municipal Department.
At first, the Governor was wary of capitulating further to Mylapore
and tried desperately to find others who could take the office.137
But in 1911, the mantle of authority fell on the shoulders of V.
Krishnaswami Iyer. He was followed into office by his schoolfriend
P. S. Sivaswami Iyer (1911-16) and by the career civil servant P.
Rajagopalachari (1916-20) who, although not a member of the inner
137
He gave it to the Maharaja of Bobbili who, of course, had neither the skill nor the
contacts to run it and who resigned after three months.

255
 The emergence of provincial politics

coterie of Mylapore, was greatly influenced in political matters by

the true Mylaporeans C. P. Ramaswami Iyer and R. Ramachandra
Rao. 138 Mylapore's advance in other fields was equally marked. Just
as Fort St George found itself possessed of the cash and power to begin
to use education as a tool of social engineering, so the Secretaryship
of the Education Department fell to the Mylaporean R. Ramachandra
Rao. From 1915, this Maratha Brahman was put in charge of the
execution of policies designed to weaken the Brahman presence in
education and government service.139 Not surprisingly perhaps, the
policies met with but limited success. Mylapore was seizing the
executive institutions of the centre at the precise point when, for the
first time, those institutions began to bear down on the locality.
At this time also, affairs in the institutions of religion were turning
towards executive decisions made at the centre and made by Mylapore.
Since the 1880s, Mylapore had been calling, almost annually, for some
kind of legislation to replace the absurd Act XX of 1863 which kept
both the centre of government and itself out of the temples and
charitable trusts of the South. But the Government of India's ban
on religious legislation prohibited such moves and guaranteed the
continued reign of anarchy in these crucial political institutions.
A few charitable trusts, however, were under the authority of the
Board of Revenue rather than of the law and, as Mylapore's grip on
governmental powers grew, it began to show an interest in them. In
1905, for example, Krishnaswami Iyer led an attack on the manage-
ment of the great Pachayappa's religious and educational charities.
He sought to replace the system of co-option through which various
Madras city magnates, such as Pitti Thyagaraja Chetty, maintained
control of the charity with a system in which trustees were elected by
past graduates of Pachayappa's College - western-educated men who
were obviously susceptible to Mylapore's influence. Through the
intercession of P. S. Sivaswami Iyer, then Legislative Councillor
and University Senator, he received partial clearance for his scheme
from the Board of Revenue. Henceforth, two members of Pachay-
appa's board were to be appointed by the University.140 Not surpris-
ingly, among the first University nominations to the board were C.
P. Ramaswami Iyer and L. A. Govindaraghava Iyer.141 In 1908,
138
'C.P. o n h i m s e l f i n C.P. by his Contemporaries, p p . 17, 2 0 .
139
See G.O. 1123 (Home, Misc.) dated 23 October 1917. T.N.A.
140
Hindu 1, 2 a n d 15 November 1907; Secretary, Board of R e v e n u e , t o P. S.
Sivaswami Iyer, 29 October 1907. P. S. Sivaswami Iyer Papers. N.A.I.
141
T i r u v e n k a t a s w a m i , Pachaiyappa's College, Madras, p . 8 5 .

256
 The emergence of provincial politics

reforms to the Civil Procedure Code cleared the way for Mylapore
to attack the temples more openly and, through the Dharmarakashana
Sabha, of which Sir S. Subramania Iyer was president, we saw that it
had considerable success.
Moreover, the Morley-Minto constitutional reforms enlarged both
the size and the scope of the Legislative Council. Increased central
taxation deepened the involvement of economic magnates in pro-
vincial politics and provided department heads with larger budgets
which were used to develop areas of political support for the British
in Madras society.142 The Council began to interfere more widely
still in the lives of its subjects: Co-operative Credit and Companies'
Acts began to change the shape of Indian finance and commerce;
the decentralisation of various aspects of the administration —
particularly over matters of irrigation, forest conservancy and police
organisation - was discussed and agreed among official and non-
official Legislative Councillors; radical reforms in the structure of
local self-government, the first mooted since those in the old Legisla-
tive Council of 1884, were debated and passed; some of the more
fundamental factors of the economy, such as labour supply, emigra-
tion and working conditions, came under government scrutiny for the
first time. Many more of the crucial interests of the presidency's
leading local powers were thus touched by the Council's activities
and came to need much more vigorous representation than before.
To obtain this representation, however, magnates had to deal with
Mylapore as an equal, for Mylapore's domination of conciliar repre-
sentation was by now almost complete. It was virtually impossible
for an Indian not connected to Mylapore to develop, in the large
general constituencies, an electoral organisation which was capable
of putting him in the Legislative Council. The vast patronage swayed
by the Mylapore lords materially helped or hindered all candidates'
chances of election. Moreover, by using its official powers, Mylapore
was able to fill the seats occupied by government nominees with its
own men.143 More important even than these developments, however,
was the fact that the leading members of Mylapore had managed
142
Between 1910 and 1919, the Madras Government's income rose from Rs 16.6
crores to Rs 25.1 crores and its expenditure in the province from Rs 6.4 crores to
Rs 9.8 crores. Report on the Administration of the Madras Presidency for the year
1910—1 (Madras, 1911), pp. 68-9; ibid., 1919-20, pp. 78-9.
143
M. Thiruvenkatachari to P. S. Sivaswami Iyer, 12 May 1912. P. S. Sivaswami Iyer
Papers. N.A.I.; V. S. Srinivasa Sastri to G. K. Gokhale, 2 February 1912. V. S.
Srinivasa Sastri Papers. N.A.I.

257
 The emergence of provincial politics

to scale the heights of Fort St George and enter the policy-making

process itself. As Executive Councillors, Advocates-General and full
departmental secretaries, they played a vital part in the drafting of
the legislation which was restructuring political society. The indivi-
dual opinion of the successful constitutional collaborator now was
of crucial significance to local magnate society for, with Mylapore's
triumph, the western-educated community of the capital ceased to
consist of a series of readily interchangeable representatives.
Between 1910 and 1920, Mylapore reached its apogee. It mono-
polised the only political connections between the locality and the
centre and used its command of central power to intervene, often
with devastating effect, in previously autonomous arenas of local
politics. The whole political balance of the presidency was upset. The
decision of a P. S. Sivaswami Iyer, a V. Krishanaswami Iyer or a C.
P. Ramaswami Iyer could throw a man, whose family might have
governed his territory for generations, out of the institutions of local
self-government, the temples, the informal offices of government and
the advisory committees to which government was passing so many
aspects of its administration. It could leave him bare and unprotected
in the Legislative Council when his interests were being discussed and
arrest him for sedition if he complained. Equally it could give him
patronage, office and support and increase his local power.
Of course, Mylapore's position, like that it had inherited from
the British, was not totally despotic. It was more powerful than ever
before but its power was not unlimited. Like the British govern-
ment, of which it was part, it needed collaboration in the localities
from men who were informally powerful and needed to use the various
magnate networks to support its raj. Where it put a nonentity into
office, he faced either the united opposition of the local magnates,
which rendered his position untenable, or he became attached to one
or other local power and worked on its behalf. Mylapore's importance
lay not in its ability to rewrite local situations as on a blank card but
to pick and choose between magnates and to help some to rise above
others. Its new portfolios meant that its interests as a patronage
broker passed into society at large. P. S. Sivaswami Iyer, for example,
aided the local political causes of such men as the Raja of Kollengode
in Malabar,144 the Raja of Kurupam in Vizagapatam145 and the
great Nattukottai Chetty banker Raja Sir S. R. Rm. A. Annamalai
144
V. P. Madhava Raja to P. S. Sivaswami Iyer, 27 July 1919. P. S. Sivaswami Iyer
Papers. N.A.L
145
Kurupam to P. S. Sivaswami Iyer, 3 August 1915. P. S. Sivaswami Iyer Papers.
N.A.I.

258
 The emergence of provincial politics

Chetty.146 The Dharmarakshana Sabha enjoyed the support of many

magnates around the temples in which it interfered; indeed, it was
usually invited by them to interfere. Its most vigorous advocates
included the Raja of Ramnad, whom it helped at Rameswaram temple;
the Komati merchant millionaire T. Sitharama Chetty, whose
interests in Srirangam temple it forwarded; innumerable Nattukottai
Chetties, who picked up temple board seats at its nomination; T.
Somasundram Mudaliar, the richest mirasidar in Tanjore, whom it
supported in litigation against the Pandarasanidhi of Dharma-
puram,147 and the Calivalla brothers, Cunnan and Ramanjulu
Chetty, who were millionaires in Madras city.148 C. P. Ramaswami
Iyer, in his desperate fights with P. Thyagaraja Chetty in the Madras
Corporation and Pachayappa's charities, was backed by men like
the Gujerati banker Lodd Govindoss and the Calivalla brothers, who
received many civic decorations and public acclamations from
educated society, honorary appointments from government and a
share of power in Pachayappa's charities and the Corporation, in
return for their support.149
Yet, obviously, wherever Mylapore aided a magnate, it also
harmed his opponents who were hit by its irresistible force. These
opponents soon came to realise that, as they could not contain
Mylapore's influence in their localities, they were faced with only
two alternative courses of action: either they could seek to make con-
nections of their own to a source of central power which rivalled that
of Mylapore or they could alter the narrow institutional connections
146
S. R. R m . A. Annamalai Chetty to P. S. Sivaswami Iyer, n June 1920. P. S.
Sivaswami Iyer Papers. N.A.I. Other m e n w h o profited in local self-government
from Sivaswami Iyer's connection were T . Subramania Iyer o f Salem, G.O. 5 1 3
(L and M , L) dated 14 March 1914; A. Rangaswami Iyer i n Madura, P. S.
Sivaswami Iyer to Secretary, Local and Municipal Department, 12 M a y 1912 i n
G.O. 1074 (L and M , M) dated 12 June 1912; A. Subbarayalu Reddiar in South
Arcot, A. Subbarayalu Reddiar to P. S. Sivaswami Iyer, 12 April 1912. P. S.
Sivaswami Iyer Papers. N.A.I.; and T . T . Viraraghavachari o f Chittoor, P.
Ramarayaningar to P. S. Sivaswami Iyer, 2 9 June 1912. P. S. Sivaswami Iyer
Papers. N.A.I.
147
Hindu 12 October 1915.
148
Fourth Year's Report of the Working of the Dharma Rakshana Sabha (Madras, 1911);
Hindu 18 October 1915.
149
For example, C. Cunnan Chetty became a member of the Pachayappa's charities
board shortly after the implementation of Krishnaswami Iyer's reforms. In 1918,
he resigned with the C. P. Ramaswami Iyer faction following a dispute with P.
Thyagaraja Chetty over the distribution o f patronage. Hindu, 6 , 1 0 and 11 April,
6 and 2 0 June 1918. See also, A. Subramaniam to P. S. Sivaswami Iyer, 12
December 1916. P. S. Sivaswami Iyer Papers. N.A.I.

259
 The emergence of provincial politics

between centre and locality which gave Mylapore its dominance.

Whichever path they followed - and there was no reason why they
could not follow both at the same time — they too were lured into
participating in the provincial political arena. Mylapore's rise had
immeasurably deepened involvement in the provincial political battle
for now the interests of most of the important men in Madras, whether
they were with Mylapore or against it, were tied to events at the
capital. Factionalism around the offices of the centre of government
came to extend beyond the small western-educated community which
for so long had been the only section of society vitally concerned there.
Merchants and landlords, with their huge personal retinues, were
dragged into new forms of politics.

260
 6
The vocabulary of communal politics

One result of the growth of the power of Fort St George, there-

fore, was the elongation of factional linkages. Political connections
now stretched up from the town and village to men who were placed in
the political institutions of the capital. Importantly, however, this
development did not deeply disturb the previous categories of political
existence. Local politicians, although using and abusing central in-
fluence, still derived the largest part of their power from the mani-
pulation of local resources. Central politicians, although distributing
a patronage to all sections of society, still derived this patronage from
the manipulation of the offices of Fort St George. Centre and locality
now were linked as never before, but each continued to maintain
its own identity and its own separate personnel. In this sense, although
the political system of Madras had been expanded in scale, it had not
undergone any qualitative change. Parallel to the process of elonga-
tion, however, there emerged another and rather different process.
Not only was power at the centre drawn from different sources to
power in the locality and not only was it held by different people but
also it was organised on wholly different principles. As the influence
of the centre rose, so these principles began to impress themselves
on local society and to change the categories in which the activities of
local society had been grouped. The norms of political conduct to be
found at the capital began to become also the norms of local conduct.
Earlier, we discussed the place of caste relationships in the struc-
ture of local political society. We saw that localities could be torn apart
by caste confrontations. Class and religious rivalries also were
capable of polarising the elements of local politics: in towns as diverse
as Bellary,1 Madura 2 and Trichinopoly,3 for example, deteriorating
trade conditions and municipal maladministration could produce
organisations and strikes among various kinds of merchants and arti-
sans; in areas as distinct and different as Nellore,4 Salem5 and Mala-
1
G.O. 2120 (L and M, M) dated 19 December 1908. T.N.A.
2
Raja Rama Rao, Sir Subramattia Aiyer, K.C.I.E. D.L., pp. 12-13.
3
Vettikodayan 23 October 1886. R.N.P.; Home Municipalities A June 1888, Nos
4
28-37. N.A.I. Hindu 18 October 1893.
5
G.O. 353 (Judicial) dated 10 February 1883. T.N.A.

26l
 The emergence of provincial politics

bar, 6 economic imbalances and insensitive official policies could

provoke Hindu—Muslim rioting. However, as we tried to demonstrate
in the case of caste, it would not be possible to argue that, because
incidents of 'communal' division took place, the political societies of
these areas were structurally divided into 'communities'. Their mem-
bers certainly possessed class, caste and religious interests, which they
would politicise and protect when challenged. But the basic stability
of the economy and the inability of the government to penetrate
deep into the locality guaranteed that active defence rarely was neces-
sary. The direction in which patronage, economic welfare and
authority (in the form of the arbitration of disputes) flowed in every-
day life indicates the continuing importance of the cross-communal
rural-boss and magnate network: the lines of the political structure
were drawn vertically to the social order.
In the last decades of the nineteenth century, however, political
organisation in the province began to emphasise increasingly the
type of socially horizontal connections represented by caste, class
and religion. Caste associations grew in great profusion, organisations
of landlords, zamindars, traders, factory workers and many other
'classes' were formed, and Christian and Muslim separatist move-
ments appeared on the political stage. South Indian society seemed
to be moving towards new principles of political organisation and
a new structure of political relationships. Clearly, the causes of this
development are complex. They must include such factors as the
building of railways, which broadened the linkages of trade and
social intercourse, and the growth of the press, which created wider
arenas of social perception.7 The autonomy of the locality was under
challenge and, with it, the narrow base of local resource distribu-
tion which had kept the patron-client tie so tight. New directions
for the flow of resources through society were being established —
directions which could follow the lines of 'community' much more
closely than before. However, communal activity was beginning to
take place on such a grand scale that it is difficult to allow changes of
this type alone to have been critical to its emergence. The factors
listed above concern alterations promoted by the improvement of
communications and affected only groups whose interests involved
them in broad territorial connection. They were thus limited to
6
Rebellion and disturbance were endemic among the Moplahs of upland Malabar.
See R. H. Hitchcock, A History of the Malabar Rebellion (Madras, 1925).
7
For a such fuller discussion of these points and of the whole matter of the growth of
communal politics, see my 'The Development of Caste Organisation in South India,
1880 to 1930' in Baker and Washbrook (eds.), South India.

262
 The vocabulary of communal politics

mercantile and administrative groups of overt 'state-level' culture and

to those elite groups of 'local-level' culture, which were being drawn
into the new regional cultures.
For the vast majority of society, however, still locked in very
localised structures of production, trade, administration and authority
and in accepted situations of status, changes in the press and the
railways made little difference to practical political ambitions and
opportunities. Wherever we have examined a town or a rural locality
we have found that the immediate resources of the magnate gave him
dominance; and, even with the advance of the central administra-
tion, contact between a local political unit and the centre of govern-
ment was carried out through the magnate. 8 It is puzzling, therefore,
to find that virtually every caste which could be identified in Madras,
wide ranges of businessmen, landlords and workers, Christians,
Hindus and Muslims should all be politicising their previously largely
apolitical interests and forming associations at or about the same
time. Besides local and specific reasons why particular groups should
seek socially horizontal alliances, there must have been also general
causes in the political system which made it necessary for a much
larger section of society to do so.
An initial factor which can help us to understand the emergence of
communal politics can be identified in the way that politics at the
provincial level, around Fort St George, were coming to be conducted
from the 1870s. As we have noted, most of Fort St George's early
policies of interference were legislative in character and very general
in application. They sought to restructure certain relationships which
greater subjects had with the state and with other groups around them.
Mercantile profits were taxed for the first time; the economy and
politics of the zamindari estate were made subject to a better-defined
rule of law; the allocation of forests was taken over by the state; in
debates on social reform, Hindu, Muslim and specific caste practices
were scrutinised.9 The contentious issues which arose at the pro-
vincial level between government and society, therefore, were likely
to revolve only around broad categories of economic and social
interest: they would concern the interests of 'the merchants',
'the zamindars\ 'the tenants', 'the cattlebreeders', 'the Hindus' or
'the Nairs'. Importantly, they would not concern individuals: before
8
That is through the magnate-dominated institutions of local self-government,
religion and administration.
9
Although the Government of India severely restricted Fort St George's ability to
legislate on social matters, some issues of social reform were debated in the Legisla-
tive Council and a few minor pieces of permissive legislation passed.

263
 The emergence of provincial politics

the 1910s, the central government was not competent to deal directly
with the multi-faceted empires of magnates at the local level. In con-
sequence, provincial politics developed its own peculiar vocabulary
in which only interests were ever discussed and, indeed, only those
particular interests which had been disturbed by government action.
Individual publicists/lawyers/Legislative Councillors may well have
been contacted and briefed by individual local magnates but the
realities of the representative system were such that the representa-
tives had to submerge the identities of their clients in that of one or
other of their interests.
Of course, the extent to which these interests possessed a tangible
existence as interest groups was slight. At the local level, they mani-
fested themselves in political action only at rare and infrequent inter-
vals. But even at the provincial level, the occasions when particular
magnates came together to sit on a caste, class or religious platform
were but moments in their separate kaleidoscopic movements from
issue to issue. All magnates had many interests, and their associates
in the defence of one could well be their enemies in the defence of
another. How would the Velama caste stand when controversies of
the day split landlord and tenant? 10 How would merchants stand when
laws were proposed which divided Hindu and Muslim^ How would
Muslims stand when a commission threatened the relationship be-
tween factory owners and employees?11 The politics of interest groups
implied the continual fragmentation of interest organisations as the
government switched its attack to different economic, social and
religious targets.
Amazingly, however, many of the interests which government had
defined did begin to solidify into permanent political associations.
By the early twentieth century a number of lobbies, among, for
example, Madras landholders,12 South Indian merchants, 13 Nairs14
and Muslims,15 had been formed and standing committees had been
10
Included in the category of Velama, which was developed both by the census and
the Velama caste association, were the Rajas of Venkatagiri, Kalahasti, Pith-
apuram, Bobbili and Nuzvid as well as thousands of impoverished peasants.
11
For example, Muslims provided both the capitalists and most of the labour force in
the tanning and cigar-making industries.
12
Founded 1890, under the impetus of government's zamindari legislation.
13
The Southern Indian Chamber of Commerce, founded 1910 under the impetus of
Lord Morley's search for a mercantile 'interest' in the Indian provinces.
14
The Nair Sam a jam, founded 1905 under the influence of changes in the way in
which the Education Department distributed patronage.
15
Several minor Muslim associations appeared in the late 1870s and early 1880s
under the influence of the Balkan War, constitutional reform, government

264
 The vocabulary of communal politics

set up to keep government constantly aware of the needs of the

specific interest. In spite of continuous internecine strife and regular
changes of personnel, as various magnates moved in and out of them,
these associations developed a concrete political existence. In the pro-
cess of becoming concrete, however, they transformed into a 'com-
munity' the interest which they had been set up to defend. Once a
lobby of, say, landlords or Nairs had been turned into a standing
association, it ceased to describe a group of men brought and held
together in politics only by a common threat (or reward) which an
external force was offering them; it described a group which occupied
a particular place in the socio-economic or cultural structures of the
province. The act of being made permanent shifted the discipline in
which the association worked from the historical, in which it ex-
plained why it had come together, to the sociological, in which it des-
cribed the social facts which underlay its unity. The association had
to make this shift in order to account for its own permanence, once
the issue which had created it had disappeared. By making the shift,
however, it began to argue for a whole new theory of Madras politics:
its rhetoric was based on the existence of a congruence between
categories of political action and categories of common socio-economic
or cultural position. That those categories rarely were congruent,
and that most members of the association behaved as though they
rarely were, were matters which had to be ignored if the association
were to remain in business. By the early 1900s, the desire that com-
munal associations should remain in business was felt keenly by at
least one significant section of the political world.
Most communal associations owed their birth less to the needs of
the magnates, whose views they were supposed to represent, than to
the needs of the publicists who represented them. Publicists and
Councillors, who carried on the dialogue with government, gained
great advantages from the development of communal associations.
Given their personally weak political positions, they could only
achieve importance, and obtain the funds and support necessary to
carrying out their own projects, by acting as the representatives of
those who were powerful. But the rise and fall of topics of controversy,
and the consequent emergence and disappearance of interest groups,
meant that they could not rely on forming lasting attachments to
magnate power. Only the men of Mylapore had any guarantee of a
prompting and the rise of the Congress. However, these associations had a spas-
modic life and the first solid Muslim association was the Madras branch of the
Muslim League which was created to hold dialogue with Lord Morley.

265
 The emergence of provincial politics

continuing stream of causes; lesser mortals were more at risk. How-

ever, the possession of a communal association lent a publicist a firm
political base and often money for an income and newspaper. He could
work the designated community as a general constituency, constantly
propagandising within it even when government was inactive and
laying before its members a limitless range of new activities which
they might be persuaded to pursue. It is no surprise that most of the
class, caste and religious associations which were established at the
capital at this time were founded and organised by educated publicists
whose own connections to the particular community they were rep-
resenting were often tangential.16
Moreover, there was no reason why a publicist had to wait for the
government to begin talking or for a magnate to come to him. By
the early twentieth century, Fort St George's proclivities for action
were well known and the vocabulary of interest politics had become
deeply engrained in the provincial level. It was perfectly possible for
publicists to conceive and develop among the magnates areas of
communal interest which as yet government had made no move to
define. Thus, from 1900 onwards, a plethora of associations appeared,
each representing a different community and each trying to build for
its creator a political constituency which he could work. Some of the
new constituencies were indeed strange for, once the publicity process
had got underway, almost every social title or label demarcated a
potential constituency, whether it had ever possessed previously a
political content or not. In Madras city in 1910, for example, A. C.
Parthasarathi Naidu attempted to establish an association of the
'Naidus'. However, Naidu was an honorific applied loosely to a variety
of historically and socially separate Telugu castes and it was by no
means clear who precisely the Naidus were, let alone what interests
they could be said to hold in common. To solve this problem, and to
register his association for legal and tax purposes, Parthasarathi
Naidu was forced to use the definition of Naidu supplied by the
Christian missionary, Dr Browne, in his Anglo-Telugu dictionary.17
16
For example, the Madras Landholders' Association, which represented the greater
territorial magnates, was put and held together by Paul Peter Pillai, a Christian
lawyer-publicist; by 1917, the association's very active secretary was G. Varadara-
chari, a teacher. The success of the Madras Muslim League was the work of Yakab
Hasan, an Aligarh-educated, Urdu-speaking Muslim who had very little in
common with the Tamil-speaking Muslim mercantile elite of the South who
formed most of his proclaimed constituents. See Non-Brahman 29 April 1917.
R.N.P.
17
Hindu 10 June 191 o.

266
 The vocabulary of communal politics

The highly creative role of the publicist also can be seen in the bitter
controversies which beset most of the more respectable communal
associations after they had been established. Often, several rival
publicists found themselves in contention for the leadership of a
particular community and produced extremely complicated patterns
of division within it. By 1917, for example, there were no fewer than
five mutually antagonistic associations among the Kammalas (artisans)
of the presidency, each claiming the sole right to speak for the caste
constituency.18
The increasingly independent position of the publicist, as a manu-
facturer of publicity, began slowly to alter the nature of the patron-
client nexus. Certainly, in their early days, caste, class and religious
publicists were looking only for ways of attracting the attention and
patronage of magnates. But the terms on which they demanded and
sometimes obtained this attention began to invert their previously
dependent relationship. Once a magnate admitted his membership
of a caste or class political community, his conduct would become
influenced by the opinions of his caste- and class-fellows. Their
views would decide for him the acceptability of a course of action in
any given circumstances. As the publicist, through his press and
broadcasting activities, had done so much to awaken or create the
community, he exercised a large measure of influence over what the
views of that community were. He moulded the community's
opinion.19 Hence, he came to possess a position of power over his
patrons. By the early 1900s, educated publicists were dedicated to the
development of communal politics not only for reasons of employment
but also for reasons of power.20
18
See Addresses Presented in India to His Excellency the Viceroy and the Right Honorable
the Secretary of State for India. P.P. 1918, vol. xvm.
19
Particularly over matters of social reform. Many western-educated caste leaders
were eager to 'modernise' the rituals of their caste communities and so to create
new values in caste behaviour. See, for example, Hindu 14 June and 8 July 1910 o n
the Komati caste conference.
20
This phraseology may make the process followed by the publicist sound very
Machiavellian: he was concerned to create a power base for himself and did not
care particularly what issue gave him the base. T h e phraseology, however, is
imposed by the perspective which we have taken: w e are interested in the forma-
tion o f power bases and, from this view, it is immaterial whether the publicist
believed in his issue (as some undoubtedly did) or merely used it cynically (as also
some undoubtedly did). Perhaps what makes the South Indian communal publicist
seem so self-interested is his failure to create his constituencies. T h e publicist-
constituent relationship (which is general t o all western democratic polities) is
meant to represent a two-way process in which both parties inform and work for
each other. But as, before the 1930s at the earliest, most communal publicists

267
 The emergence of provincial politics

In several previous interpretations of the appearance of com-

munal politics, attention has been concentrated on the forces of
'modernity' acting on supposedly pre-existing identities, particularly
those of caste. It has been argued that economic, social, educative and
political changes, in the later nineteenth and twentieth centuries,
were responsible for creating a new perception of an old identity
and, thus, a new type of caste organisation out of an old one.21 As
Richard Fox has shown, however, this argument is unsatisfactory
for, apart from the caste label, the points of contact between the
functions of 'old' and 'new' castes were very few indeed.22 Yet Fox's
own explanation of the phenomenon of the caste association also is
unconvincing. If caste associations represented no more than some
residual 'husk of sentiment' of traditional caste organisation, why did
so many men invest so much time and effort in developing them?
However, if we move away from any attempt to link the caste associa-
tion to caste but rather look laterally and seek to set it in the context
of the other political institutions of similar type (class and religious
associations) which were growing with it, an explanation becomes
more obvious. Once the founders of caste associations are recognised
not as 'new' Nadars or 'new' Pallis but as men of the western-educated
professions performing an important role in the political system of the
early 1900s, the mystery of the associations rapidly disappears.
Public debate with government was denning political groupings
in a new way; the press and public platform helped to disseminate
these definitions; and professional publicists, whose occupation it was
to mediate between government and society, developed a very strong
interest in the business of dissemination. Changes in language and the
means of communication were altering the political style of the
presidency. That it was these broader changes, rather than anything
to do with the functions of caste per se, which produced the caste
association can be seen in the fact that many of the publicists who
organised constituencies of caste were, at precisely the same time,
organising constituencies of class and religion out of different groups
in society. A. C. Parthasarathi Naidu, for example was not only a
Naidu but also a leader of Hindu religious revivalism, a protector
never won the active support of many members of their communities (or at least
never held that support for long), the traffic was very one-way. What the publicist
gained is obvious; but what his constituents got out of the relationship is more
obscure.
21
See, for example, S. and L. Rudolph, The Modernity of Tradition.
22
R. G. Fox, 'The Avatars of Indian Research' in Comparative Studies in Society and
History, xn: 1 (1970).

268
 The vocabulary of communal politics

of mercantile interests, an advocate of Telugu cultural revival and a

spokesman of the united Indian nation.23 As a publicist, he repre-
sented several different communities and built up several different
constituencies. Other caste organisers, such as Salla Guruswami
Chetty of the Komatis, C. R. Reddi of the Reddis and N. G. Ranga of
the Kammas were equally multi-faceted: Guruswami Chetty also
promoted the interests of revivalist Hindus, of aspiring Indian
nationals and of the mercantile community; C. R. Reddi of Andhras,
non-Brahmans and the educated; and N. G. Ranga of non-Brahmans
and peasants.
Yet, important as were these changes in the vocabulary of politics
and in the way that publicists were coming to act, they did not, of
themselves, alter greatly the realities of political life and behaviour.
The sums of money which communal associations were able to elicit
from magnate pockets were always small, and targets for educational
trusts (naturally the publicists' most obvious personal interest) and
political funds were very rarely met. 24 Magnate patrons would offer
and withdraw their support from each type of communal association,
depending on which particular community seemed to matter more at
which particular time, and even good and loyal patrons were unlikely
to devote many of their resources to causes which were peripheral to
their continued existence as magnates. Lacking material resources, the
publicist was unable to stir the sea of communal opinion which would
give him his full political independence and had to remain little more
than a paid hack. He was making an impression on only one small
area of provincial politics and none at all on the local political level.
Most of the work of constructing a communal political system was
completed by Fort St George which not only talked in communal
categories but came increasingly to administer and act in them. A
perpetual theme of British policy in India was to identify and contact
allies in Indian society. While so much of the effective decision-
making remained at the lowest level of government, officials on the
ground were able to seek out support in local society and to construct
their alliances on a face-to-face basis. As we have seen, they tended
to associate themselves with the principal local magnates of their
area and to use their networks for government. By so doing, they
strengthened the cross-communal web of magnate influence. Oc-
23
See, for example, Hindu 22 November 1890.
24
For example, the Kamma conference failed to develop a popular base for fund
raising and depended entirely on the occasional contributions of two zamindars.
Hindu 2 August 1915; the Reddi Sangham's financial support was equally limited,
see Files 6 and 24 of 1926-7 in C. R. Reddi Papers. N.M.M.L.

269
 The emergence of provincial politics

casionally, however, individual officials, through mistake or ideo-

logical parallax, sought to tailor their policies to fit what they saw
to be organised political communities of caste, class or religion.
Collector Macleane in Salem, with his pro-Muslim policies, was a
classic example.25 These eccentricities clearly were capable of induc-
ing actual communal divisions but, except in the rare cases where they
were coherent with the local distribution of resources, the divisions
were unlikely to outlast the individuals who had created them.
A consistent direction of communal policies in the presidency
could come only from Fort St George which, during most of the
nineteenth century, was in a poor position to implement any policies
at all. From its very inception, however, the ideesfixesof many of its
members about the nature of Indian society had guaranteed that some
features of a communal political system had been encouraged to
develop.26 In their arbitration of Hindu and Islamic social disputes,
for example, British law courts had been ordered to defer to the
authority of priests and scriptures, rather than of custom, when
making their judgments.27 Naturally, priests and scriptures took a
much harder line on the importance of caste and religious divisions
than had the more pragmatically orientated rule of local custom.
Equally the courts of Fort St George appointed kazis, caste headmen
and panchayats to arbitrate on their behalf in petty cases between
members of the same jati or faith. These appointments gave legal
identities to local sectarian groups, which often might not have
possessed them before, and reinforced the power of sectarian authori-
ties, which often had been weak or non-existent before. When Fort
St George first published its censuses, after 1881, its need for clear
statistical categories led it to outrage caste groups whose varna
classifications were not consistent with their local status positions.
This outrage could manifest itself in the organisation of communal
associations which undertook political activities.28 Moreover, when,
through pieces of legislation like the Marriage Act of 1885, Fort
St George imposed a statutory unity on the socially diverse Nair
caste, it could succeed in drawing much more clearly than before
an identifiable communal boundary around the Nairs.
The overall effects of these early policies have been seen by some
25
There were several others, such as Collector Logan of Malabar who had a
partiality for Tiyas. Hindu 9 and 27 March 1887.
26
See B. S. Cohn, 'The Census, Social Structure and Objectification in South Asia'.
Paper read at Second European Conference on Modern South Asia. Copenhagen.
27
S. and L. Rudolph, The Modernity of Tradition, pp. 251-93.
28
Ibid., pp. 49-64.

270
 The vocabulary of communal politics

commentators as solidifying caste divisions and, indeed, as pro-

moting competition between caste and religious communities. Un-
doubtedly, their influence was towards those ends. Yet before 1900,
their impact on the fundamental cross-communal structures of
political life was very limited. The British law courts handled only
a minute fraction of popular disputes; kazis and caste headmen could
maintain their jurisdictions only when they could persuade disputants
to accept their judgments and not seek arbitration from more powerful
outsiders; anti-census activity was limited in scale and, of course,
sporadic; the Nair Marriage Act was merely permissive and, although
it caused controversy while in the Legislative Council, was scarcely
utilised by Nairs until the later 1920s.
It was the growth of the power of the capital, particularly from the
second decade of the twentieth century, and the consequent penetra-
tion of the locality by the political forms of the centre which really
set off the chain of reactions which produced modern communal
politics. Once its growing administrative competence had elevated it
to a provincial political position, Fort St George had to begin planning
a strategy to hold its place in the province. Its new powers meant that
it was capable of raising provincially organised forces against itself;
consequently, it needed to seek out and attach itself to provincially
organised forces of loyalty. Yet what would these forces look like and
how would they be organised? The highly complex webs of local mag-
nate power could scarcely be seen from the distant Secretariat build-
ing and, between 1895 and 1916, the local magnate himself was
making no move to participate directly in provincial politics. In Fort
St George, the British, although not Mylapore, would have faced
countless practical problems in contemplating a strategy based on a
series of person-to-person deals with some of their greater local
subjects. Moreover, there is little evidence to suggest that they ever
did contemplate it. Their intellectual training, whether in the con-
temporary European social sciences or in Orientalism, taught them
to view Indian society as a series of interlocking caste, class or religious
blocks.29 Each block represented a solid, internally organised
political community, and politics consisted of achieving balances
between the blocks. Even where British senior officials showed them-
selves more sensitive to the nuances of local reality, they were often
befuddled by what they saw. Earlier in their careers, many senior
officials had been present during local religious and status confronta-
tions. Some of them preferred to see in these occasional antagonisms
29
See Cohn, 'The Census, Social Structure and Objectification in South Asia'.

271
 The emergence of provincial politics

the lines which demarcated the political structures of the locality.

Whatever the causes, as the provincial centre grew in stature, so an
increasing number of its British members began to conceive their
task as that of playing off one community, be it of caste, class or
religion, against another and of finding the favour of one provincial
interest group to match any anger that they might have occasioned in
another. To the Madras I.C.S., provincial politics came to mean
communal politics.
During the early years of the twentieth century, communal cate-
gories crept into a wide variety of administration. Between 1901
and 1918, for example, the number of'backward castes' listed in the
Education Department's Grant-in-Aid Code grew from 42 to a
staggering 138 and included, paradoxically, some of the leading land-
owning and trading castes in the presidency.30 For the first time, a
provincial policy of recruiting government servants by caste was
elaborated and implemented by Government Order in 1912.31 Con-
siderations of community began to affect the Government's choice
of University Senators and the Senate's choice of Text Book and
Examination Committee members.32 Even the Famine Code, which
had been developing from the 1870s, came to recognise that relief
should be distributed as much by caste as by need.33
The most important field of communal administration, however,
was that related to places in political institutions. Initially, Fort St
George was strongly opposed to the Government of India's policies
on the introduction of communal safeguards and electorates into the
institutions of local self-government and the Legislative Council.
In 1882, 1899, 1908 and 1911, it denounced communal representa-
tion as unnecessary and creative of a social divisiveness which did
not then exist in South India.34 Given the fact that, in so many other
areas of administration, the Madras Government already possessed
communal categories, it is difficult to understand the violence of its
30
Grant-in-Aid Code of the Madras Education Department for 1901—2; ibid., for 1918—
19. T.N.A.
31
G.O. 1561 (Public) dated 19 December 1912. T.N.A.
32
G.O. 187 (Education) dated 29 February 1912; G.O. 22 (Public) dated 21 January
1918. T.N.A.
33
Report on the Famine in the Madras Presidency during 1896 and 1897 ( M a d r a s ,
1898)511, 26-8.
34
Report of the Committee on Local Self-Government in Madras. 1882 (Madras, 1883),
p. 43; Letter No. 7 (Legislative), Government of Madras to Government of India
dated 31 January 1899 in Home Public A July 1899, Nos 16-21. N.A.I.; Home
Public A October 1908, Nos 116-46. N.A.I.; G.O. 616-17 (L and M, L) dated
27 April 1908. T.N.A.; G.O. 916 (L and M, L) dated 12 July 1911. T.N.A.

272
 The vocabulary of communal politics

opposition to the Government of India's proposals - except, perhaps,

that they were the Government of India's proposals and that, by
objecting to the details, Fort St George hoped to avoid having to
implement any policy of constitutional devolution at all. Be that
as it may, the Madras Government was forced by the fiat of Calcutta
to nominate representatives of 'the landlords' and 'the Muslims'
to its early Legislative Councils and to create communal electorates
for these categories under the Morley-Minto reforms. In the wake
of the reforms, however, Fort St George changed its attitude dramati-
cally and of its own volition began to support communal interests.
It freely nominated caste and class representatives to the Legislative
Council,35 insisted to Edwin Montagu that the non-Brahmans of its
province should be protected by separate electorates,36 and introduced
communal machinery into the municipalities and district boards.37
The conversion of the Madras Government to policies of active
communal discrimination was the prime factor in the development
of communal politics. It laid a cement of patronage and possible favour
around particular connections of interest, which before may have
been of no more than occasional political importance. By extending
its 'provincial' notions of community to the locality, through places
on local boards, grants to schools and employment opportunities, it
began to shape local organisations of resource distribution to the pro-
vincial model. Men obtained positions of power and social advantage*
not because they were wealthy or related to men of wealth but because
they were Muslims, tenants or Kammalas. The criterion of com-
munity had never been so exclusively important in the past. Inevit-
ably, this meant that magnates began to lose some of their economic
and arbitrational functions. Potential dependents, seeking security
or help, were presented with an increasingly viable alternative to
submission to the magnate.
The extent to which this alternative was being taken up before
1920, however, ought not to be overestimated. The expansion of the
magnate structure to reach Mylapore was having the effect of'localis-
ing' the conduct of provincial politics faster than certain British
civilians could 'provincialise' the locality. Debates at Fort St George
were beginning to turn on the character of particular individuals
with greater speed than local debates were turning on the character
35
A representative of the Nattukottai Chetties was nominated in 1910 and of the
panchamas in 1917.
36
See below pp. 2 9 4 - 6 .
37
By reserving seats for members of the depressed castes and minority religious
interests.

273
 The emergence of provincial politics

of interests. Indeed, even the British recognised this. Their political

system was being built on two contradictory principles of representa-
tion - the indirect system of Mylapore and the magnate, and the
direct system of community. When the two principles clashed, they
always deferred to the indirect, whether of the secretive Mylapore
variety or of the general constituencies in the Montagu-Chelmsford
Councils, which outnumbered the communal constituencies. Their
subjects concurred in this deference.

The case of the non-Brahmans

Space prohibits us from investigating the effect of the two general
factors — the rise of the publicist and government policy — on the
many caste, class and religious political movements of these years.
By the examination of one, the non-Brahman movement which can
be seen first at the provincial level from about 1912, it is hoped,
however, to illustrate significant features of them all. To our inter-
pretation the political division of society into Brahman and non-
Brahman makes no obvious sense. Certainly there were status and
some cultural differences between the two but status and cultural
divisions are not necessarily political divisions, and the occasions,
before 1912, when they had become so were very few in number. This
lack of antagonism is not surprising when it is remembered that
Brahmans supplied status legitimacy to most of the groups of state-
level culture but had little contact, and were ignored, by the vast
majority of local-level cultural groups;38 nor when it is recalled that
even in their contacts with most other state-level groups, Brahman
priests were usually poor dependents who either did what they were
told or starved.39 As most contemporaries remarked, the non-Brah-
man or, more properly, anti-Brahman movement from 1912 was a
new political development and not the continuation by other means
of a supposed two thousand year politico-cultural feud.
Moreover, in the political domain itself, it would be extremely
difficult before 1912 to separate out a specific Brahman power from
a specific non-Brahman power. Brahmans, in their many guises, could
be found in almost every position in the networks of the local mag-
nates, most of whom were non-Brahmans. However strongly Brah-
mans may have wished to operate together (and their constant familial,
sub-caste, regional and sectarian feuding makes it clear that they
hardly wished to at all), the structure of resource distribution was
38
See Back, Peasant Society in Konku.
39
See, for example, the account of a dispute between Brahman priests and their
Komati patrons in Masulipatam in Hindu 31 July 1907.

274
 The vocabulary of communal politics

such that, except in a few localities of Tanjore and Malabar, nowhere

could they form a viable, exclusive magnate network of their own.
Across the presidency, they performed various tasks for various
magnates, were closely related in political action to large numbers of
non-Brahmans and in no way existed inside a functioning, political
'Brahman' unit.
The problems of explaining the rise of the non-Brahman move-
ment, therefore, are very great. Not only is it necessary to account for
the timing of the phenomenon but also it is necessary to give meaning
to the notion of the political Brahman which, before the movement
began, would have been incomprehensible to most contemporaries.
Let us first examine critically one or two of the arguments which have
been put forward before to explain the politics of non-Brahmanism.
In a recent book, E. F. Irschick has sought to relate the emergence
of the movement to the growth of literacy among higher non-
Brahman castes. This, he argues, led to a growing resentment at the
monopoly of government office and public life enjoyed by Brahmans.40
In our terminology, it represented not so much an attack on Brahman
political power as pressure on those occupations and positions in
magnate networks (particularly those concerned with service) which
Brahmans filled in large numbers. The pressure was derived from
changes in the composition of the educated community. However, a
careful examination of educational development and its significance
indicates that this view is difficult to substantiate. The spread of
English education among non-Brahman groups was not great and
there is no evidence that, before 1920, it had produced a shift in the
social pattern of learning. In first- and second-grade colleges, the
proportion of Brahman to non-Brahman students was 3.5:1 in 1890
and 3.5:1 in 1918.41 Obviously, between the two dates, the absolute
numbers of students had increased, although, given the changing
circumstances, not by very much.42 In 1890, when most appointments
to the government service had been made in the locality, there was
much less need than in 1918 for college qualifications. Service aspir-
40
Irschick, Politics and Social Conflict in South India, p. 17.
41
English education, rather than vernacular education, was essential to gaining
access to the professions and public life. Consequently, only changes in English
education are relevant to Irschick's main argument. Report on Public Instruction in
the Madras Presidency for 1890-1 (Madras, 1891), Subsidiary Tables, pp. 1 - 2 ;
ibid., for 1918—19. (Madras, 1919), 11, 2 - 3 . Brahmans represented 67 per cent of
university graduates in 1890 and 67 per cent in 1918. G.O. 22 (Public) dated
21 January 1918. T.N.A.
42
From 2863 in 1890 to 6818 in 1918. Report on Public Instruction in the Madras
Presidency for 1890-1, Subsidiary Tables, pp. 1 - 2 ; ibid., for 1918-19,11, 2 - 3 .

275
 The emergence of provincial politics

ants orbited the office of the huzur sheristidar who did not run com-
petitive examinations. With the new role of the central government,
higher standards of education came to be demanded and all aspirants
had to go to school. This development alone would surely account
for most of the rise of 150 per cent in the number of first- and
second-grade students: it is not necessary to imagine the involve-
ment of new social groups in higher education. Moreover, government
appointments in the higher grades and court litigation were expanding
fast enough to meet the minimal job ambitions of most of the educated.
There was no serious problem of graduate unemployment, such as was
emerging in Bengal.43 Perhaps the clearest indication that the pressure
of new non-Brahman groups on the services and professions was
mythical, however, is provided by the behaviour of the Justice Party
(the non-Brahman communal party) and the government themselves.
The Justice Party never argued for the creation of 'fair' competi-
tion to allow qualified non-Brahmans the chance to break up a
Brahman monopoly. It demanded a dropping of educational standards
and the building of closed social categories of recruitment to be filled
by non-Brahmans whether they were qualified and competent or
not.44 It wanted government to promote the growth of an educated
non-Brahman community not simply to recognise the existence of one
which had grown already. Further, in spite of agreeing to these
closed categories, Fort St George in fact found itself having to recruit
ever more Brahmans, for suitable non-Brahmans did not come for-
ward to take the proffered jobs.45 Indeed, the vast majority of district
Collectors told Fort St George that they could not find many non-
Brahmans who were interested in government employment.46
Equally, although the spread of vernacular literacy among higher
non-Brahman castes was prodigious, it cannot be argued that this,
of itself, led to a challenge to the Brahman.47 Some reason must be
shown why vernacular literacy should produce anti-Brahmanism and
some evidence must be found that the new channels of communica-
tion, opened up by increased literacy, were important in propagating
43
Between 1900 a n d 1920, t h e n u m b e r of gazetted posts a n d of non-gazetted posts
w o r t h m o r e t h a n R s 100 p.m. increased by about 80 per cent. T h e n u m b e r of non-
gazetted posts worth between R s 35 and R s 100 p.m. increased by 160 p e r c e n t .
Indian Statutory Commission (H.M.S.O., 1930), vi, Appendix A, pp* 607—13.
44
See Justice 1 M a r c h a n d 5 April 1917. R.N.P.
45
B r a h m a n s held a higher proportion of government appointments worth more t h a n
Rs 100 p.m. in 1927, after six years of Justice Party government, t h a n they h a d
done in 1900. Indian Statutory Commission, vi, Appendix A, pp. 6 0 7 - 1 3 .
46
G.O. 1157 (Public) dated 3 August 1915; G.O. 1123 (Home, Misc.) dated 23
October 1917; G.O. 986 (Revenue) dated 30 April 1920. T.N.A.
47
Census of India. 1921. Madras. Volume XIII. Part 2 (Madras, 1922), p. 128.

276
 The vocabulary of communal politics

anti-Brahmanism. On this latter point, such evidence as there is avail-

able negates the argument. The vernacular newspapers of the militant
anti-Brahman Justice Party were uniform failures. 48 The most
successful vernacular newspapers were those which supported the
more extreme activities of the Congress and were run by or in associa-
tion with Brahmans.49
A second explanation tentatively offered by Irschick is that the
Brahman/non-Brahman polarisation was produced by religious and
cultural revivalism, particularly among the Tamilians, which led
to attacks on the Brahmans as Aryan invaders, whose Sanskritic
culture and Vedantic religion had destroyed the Tamil and Saiva
Siddhanta basis of Southern civilisation.50 Once more, as a causal
explanation, this does not stand up to scrutiny. In the late nine-
teenth and early twentieth century, the cause of vernacular revivalism
was aided at least as much by Brahmans as by non-Brahmans. The
patronage networks which made the revival possible consisted often
of the very same men whether the cause was the regeneration of Tamil
or the study of Sanskrit, the development of the philosophy of Advaita
or of Saiva Siddhanta.51 The Mylapore group in the University Senate
were the strongest advocates of enhancing the status of vernacular
studies;52 they also founded the Madras Sanskrit College.53 Indeed
a direct connection between cultural and political change begins to
look most extraordinary when we realise that T. M. Nair, who abhor-
red Indian civilisation, was a leader of the non-Brahman movement; 54
that many Justice Party leaders could barely speak or write their
48
T h e Dravidan and Andhraprakasikha failed t o establish themselves as dailies and,
by 1919, were only weekly publications. Hindu 9 August 1917; A. C. P a r t h a s a r a t h i
Naidu to A. Campbell 23 April 1919 in G . O . 932 (Home, Misc.) dated 24 M a y
1919. T.N.A.
49
Particularly, Desabhaktan, Swadesamitran and Andhrapatrika.
50
Irschick, Politics and Social Conflict in South India, pp. 2 7 5 - 3 0 8 .
51
F o r example, J. M , Nallaswami Pillai was a scholar both of Sanskrit a n d of Tamil.
Balasubramaniam, The Life ofj. M. Nallaswami Pillai, pp. 10, 6 4 - 7 2 ; V. Krish-
naswami Iyer patronised Sanskrit and T a m i l Poetry. P. M a h a d e v a n , Subramania
Bharati, Poet and Patriot (Madras, 1957), pp. 4 8 - 9 and A. S. Balasubramania Iyer
to V. Krishnaswami Iyer, 5 December 1906. V. Krishnaswami Iyer Papers. N.A.I.;
the Nattukottai Chetties, t h e Raja of R a m n a d , t h e Nuzvid zamindars a n d t h e
major Komati families of A n d h r a patronised both vernacular and Sanskrit
revivalism.
52
Hindu 10 a n d 15 M a r c h 1913; P. S. Sivaswami Iyer to M . Ramaswami Iyer, 29
M a r c h 1904. P. S. Sivaswami Iyer Papers. N.A.I.
53
A. S. Balasubramania Iyer to V. Krishnaswami Iyer, 5 December 1906. V.
Krishnaswami Iyer Papers. N.A.I.
54
H e a n d C. R. Reddi, another Justice Party publicist, were opponents in the Senate
of t h e expansion of vernacular education. Hindu 10 and 15 M a r c h a n d 1 M a y 1913.

277
 The emergence of provincial politics

vernacular; and that the party relied very little on the vernacular for
propaganda.
Equally, it would be impossible to connect the non-Brahman
movement of 1912 to the anti-religious Tamil Self-Respect movement
of the later 1920s, as has been attempted by Robert Hardgrave.55
The Self-Respect movement rested on the support of those elements
of local-level culture which were slowly being drawn into the new
regional level cultures. However, in the 1910s, these elements were
scarcely out of their localities and barely conscious of their future
identity. If they related to provincial politics at all, it was more as
members of transactional magnate networks than as bearers of a new
ideology. Men who later were involved with Self-Respect could be
found on both sides of this Brahman/non-Brahman controversy.
Indeed, the leader of Self-Respect, E. V. Ramaswami Naicker, was a
staunch Congressman whose political contacts were largely with
Brahmans.56 Such cultural debate as there was between 1912 and
1920 was among groups of emphatic state-level culture and did not
touch groups of a different orientation. When, from the 1920s, the
Self-Respect movement began to emerge, it attacked all groups of
state-level culture, Brahman and non-Brahman alike, and thus made
enemies of the high-caste leaders of the non-Brahman movement of
the earlier period.57 In social composition, practical aims and doctrine,
the non-Brahman and Self-Respect movements were as different as
chalk and cheese. The only connection is that they were both anti-
elite movements, although not even against the same elite.
The relationship of the non-Brahman political movement to the
Sat-Sudra vernacular and religious revival, which was present at this
time, was, in fact, very similar to the relationship between Hindu
revivalism and the Congress in its agitational periods and between
Telugu revivalism and the demand for a separate Andhra province.
Cultural movements, logically independent of politics, were dragged
into political life because they provided a pre-existing organisation
which was valuable in raising manpower. T. V. Kalyanasundram
Mudaliar, a student of Saiva Siddhanta and virtually the creator of
modern Tamil journalism, was prominent in many Madras city
revivalist associations and was on close terms with many Brahmans.
He has recorded how, from 1915, non-Brahman agitators began to
invade his sabhas to preach cultural revolution against the Brahman
55
See R. L. H a r d g r a v e , The Dravidan Movement (Bombay, 1965).
56
See C. J. Baker, 'Noncooperation in South India' in Baker and Washbrook (eds.),
South India.
57
See Baker, Politics in South India 1920 to 1937, ch. 3.

278
 The vocabulary of communal politics

in the hope of stirring support to their political cause.58 His own

reaction, which was disgust and resignation from the associations,
was mirrored in the attitude of many other revivalist leaders, who had
been and remained friends with many similarly interested Brahmans.59
Both the explanations from educational change and cultural
revival presuppose that the non-Brahman movement reflected deep-
seated changes in the organisation of society. Non-Brahmans, whether
through the influence of new educational opportunities or through the
influence of new cultural perceptions, were coming to throw off the
Brahman yoke. Yet what is most noticeable about Madras, even
through the years of rabid anti-Brahman propaganda, is the continu-
ing social stability. The jajmani system, by which deference was
shown to Brahmans, was not broken in any significant way: rich
non-Brahmans continued to rebuild temples, to found Sanskrit col-
leges and to support and feed Brahmans. The Justice Party, when
looking for votes rather than trying to raise noisy agitations, was
fully aware of this. In 1920, its leaders undertook a prolonged
tour of the presidency and desperately tried to disassociate their
movement from any attack on the spiritual prerogatives of the
Brahman. They argued that their challenge was solely towards the
secular, political position which Brahmans had attained.60 Yet, once
the Brahman's spiritual role has been stripped from him, how can he
remain a Brahman in any meaningful sense? What the Justice Party
really objected to was the political position of certain individuals
who happened to be Brahmans. Of course, the men they objected to
were the men of Mylapore.
Let us try to reconstruct the non-Brahman movement in the light of
our general inquiry into the growth of communal politics. The leaders
of the movement, that is to say the people who created it, require a
careful analysis, for in their ambitions must lie its causes. In nearly
every case, the principal propagandists and apologists of the campaign
were men drawn from families with generations of involvement in the
government service and the professions behind them. Typical exam-
ples are provided by Koka Appa Rao Naidu and Tikkani Balijarao
Naidu, who came from Andhra families with service records dating

58 -p y Kalyanasundram Mudaliar, Valkkai Kurippugal (Madras, 1969), pp. 596—9

(Tamil).
59
Neither S. S. Bharati nor Swami Vedachelam, two of the leaders of Tamil revival-
ism, would have anything to do with the political non-Brahman movement until
after 1920, w h e n it had become the government.
60
See report o f P. Thyagaraja Chetty's speech at Salem in Hindu 11 June 1920; also
Hindu 13 M a y 1920.

279
 The emergence of provincial politics

back to at least the 1860s, 61 Arcot Ramaswami Mudaliar, whose

family had a massive presence in the Chingleput bureaucracy,62 and
B. Muniswami Naidu, whose family had administered the estate of
the Raja of Karvetnagar for centuries. 63 There is no sign that these
men represented any new social force in Madras; they had been pro-
duced as much by the early service policy of the British as had their
Brahman service family equivalents. Before about 1912, there is no
sign either that they saw themselves as a corporate group of non-
Brahmans. They were drawn from several regional caste groups,
operated as * castes' only through kin and personal contacts and
worked in local and provincial political structures with Brahmans. 64
What factors not only brought them together but also brought them
together as non-Brahmans?
The answer to the first question is, of course, the centralisation of
control over the professions at Madras city. All members of the
western-educated community now were placed in the same career
structure and single lines of division between them could split the
presidency. That these lines might come to mark a Brahman/non-
Brahman division is suggested by a common grievance which all
educated non-Brahmans shared against Brahmans. Their accredited
social position was disproportionately low for, although they were
performing the same secular roles as Brahmans, they were seldom ac-
corded the same ritual and social prestige. Niggling complaints against
61
The 'Koka' and 'Tikkani' Audi-Velama families intermarried and also served the
Hyderabad state. Hindu 26 March 1920. See History ofthe Services ofGazetted and
Other Officers in the Civil Department in the Madras Presidency, Corrected to istjuly
1885 (Madras, 1885); also, Sastri (ed.), Encyclopaedia of the Madras Presidency,
P. 525-
62
So massive t h a t w h e n Ramaswami Mudaliar stood for election to the Legislative
Council in a Chingleput constituency, t h e Chingelput D.C.C. petitioned govern-
m e n t t o remove his relatives from t h e district bureaucracy. Hindu 3 July 1920.
63
K o t t a Bhaviah Chowdary, A Brief History of the Kammas (Sangamjagarlamudi,
1
955)5 P- 90. Of other prominent n o n - B r a h m a n leaders, T . M. Nair's family h a d a
long history of service in Malabar, N . Gopala M e n o n , A Short Sketch of the Life of
Dr T. M. Nair (London, 1920); K. V. Reddi Naidu's family h a d served t h e East
India C o m p a n y in t h e eighteenth century a n d spent most of t h e nineteenth
century in t h e Kistna police department, Subba Rao, Life and Times of Sir K. V.
Reddi Naidu.
64
F o r example, B. M u n i s w a m i Naidu a n d A. Ramaswami Mudaliar h a d been
apprenticed to B r a h m a n vakils; K. V. Reddi Naidu a n d T . Ethiraja Mudaliar were
prominent Congressmen until about 1915; P. Sivagnana Mudaliar and V.
T i r u m a l a i Pillai (both lawyers) worked o n t h e M a d r a s Corporation in the faction
containing t h e B r a h m a n K. C. Desikachari; L. K. Tulsiram, a Sourahstra lawyer
from M a d u r a , worked o n his municipal council in t h e faction of K. V. Ramachari,
a Sourashtra dye-merchant, w h o also used several B r a h m a n lawyers.

280
 The vocabulary of communal politics

Brahman arrogance, which no doubt could have been heard in

separate localities before, began to creep into the provincial press.65
The importance of this union of complaint, however, ought not to
be overemphasised. It is not necessary to like someone in order to
work with him, and most of the people who were making the com-
plaints in fact were working with Brahmans and were tied to the same
magnate networks as Brahmans. Antipathetic sentiment of itself
could not make the non-Brahman movement unless it were provided
with a new political structure in which to operate.
The first part of this structure was erected by the rise of the publi-
cist. Most of the family groups of educated non-Brahmans were
socially isolated from the non-educated masses of their particular
caste communities. Their kin organisations, used for the distribution
of service patronage rather than for land control or local power, were
of a different shape.66 Their culture was state-level whereas that of
most of their caste was local-level. In many cases, their ritual practices
bordered on the heterodox.67 In each of their castes, they formed a
microscopic minority which had severed most of its points of contact
with the rest of its caste community. During the last years of the
nineteenth century, however, as the new vocabulary of provincial
politics, in which broad notions of caste played so large a part, began
to emerge, these educated non-Brahman families were given a strong
incentive to reverse their courses. Caste, at least the idea of caste as
a status interest group, was a viable political constituency which
every publicist-politician could profit from working and, indeed,
which only the most successful could afford to ignore. In attempting
to translate the highly complex and ambiguous perceptions of caste
which most of the indigenous population possessed into the neat
categories of the provincial political system, the non-Brahman
publicist was inevitably drawn into conflict with the Brahman's
spiritual prerogatives.
If the caste system can be characterised at all, it can be only as
65
See particularly, S.K.N., Non-Brahmin Letters (Madras, 1915).
66
Educated service families within a caste tended to seek out and marry each other
rather than developing alliances with illiterate but wealthy families. T h i s meant
that their o w n marriage networks tended to be m u c h broader than those of the rest
of their 'peasant' caste communities. See, for example, the K o k a - T i k k a n i alliance
in Hindu 2 6 March 1920.
67
For example, C. Sankara Nair's advocacy o f Nair marriage reform and other social
reforms made h i m extremely unpopular among West Coast Nairs. I am grateful to
D r Susan Lewandowski o f Amherst College for this information. Also, the Koka
and Tikkani families, and several other educated Audi-Velama families in
Andhra, were Brahmo Samajists.

28l
 The emergence of provincial politics

a system of social hierarchy validated by the tenets of the Hindu

religion as interpreted by Brahmans. And although Aryan Hinduism
may not have plunged very deeply below the state-level in most of
South India, it certainly affected, in one way or another, many of
the rich men (the local magnates) in Madras. Magnate wealth and
splendour were often expressed through expenditure on objects of
orthodox Hindu piety, and magnate-driven upward social mobility
movements often took the form of attempts to raise an accredited
varna position. The social models which magnates tended to emulate
were, if not actually Brahmanic, at least placed within a Brahman-
dominated hierarchy.
Clearly, however, the non-Brahman professional publicist who was
trying to build an independent position for himself as a caste leader
would have to break this dependence on the good graces of an external
social group. He could not lead his caste, as the autonomous political
community fighting against other autonomous political communities
which it was supposed to be, while its more prominent members were
deferring to the authority of outsiders. He had to be anti-Brahman,
or at least anti-Vedic, if he were to make the constituency his
own. Consequently, from the turn of the century, western-educated
Nairs, Kammas, Telagas and Audi-Velamas could be seen organising
within their castes campaigns to eradicate the use of Brahman priests
and of Vedic practices.68
The extent to which this element of politico-cultural antagonism
was a primary cause of the events of 1912 to 1920, however, is highly
debatable. The number of caste publicists who were affected by it was
always very small. Most were too socially conservative to employ a
tactic which committed them to advanced social reform; they pre-
ferred to concentrate on obtaining material rewards for their caste
and to leave unanswered the questions of what this caste actually
represented and why it was there. Others were rapidly disconcerted
by the hostility which the vast majority of orthodox magnates ex-
pressed towards reform, and dropped the anti-Brahmanic elements of
propaganda before the magnates dropped them. In consequence, by
no means all the publicists who were working the caste issue among
non-Brahman groups rallied to the anti-Brahman flag. Indeed, in
1916, when Madras political society divided into pro- and anti-
Brahman camps, most found themselves on the pro-Brahman side.69
68
See Ranga, Fight for Freedom, pp. 2 7 - 8 .
69
For orthodox resistance to caste reform in the case o f the Komatis, see Hindu 14
June and 7 July 191 o; also, Darsi Chenchayya, Nenu Na Desamu (Vijayawada,
i952)(Telugu).

282
 The vocabulary of communal politics

Moreover, the leadership of the greater anti-Brahman movement did

not consist of men who had a long history of reformist activity within
their own specific caste communities. Neither P. Thyagaraja Chetty
nor A. Ramaswami Mudaliar nor B. Muniswami Naidu had shown
much interest in caste politics before the rise of the movement, and
Thyagaraja Chetty was better known as a staunch defender of
Brahman spiritual prerogatives. Thus, while recognising the impor-
tance of caste-reformist ideology to a few supporters of the non-
Brahman cause, it would not be possible to explain the rise of that
movement by reference to caste reform alone, or even in the main.
The second and more important prop of the new political structure
was direct government policy. Logically and historically, the non-
Brahman movement began with government action. During the nine-
teenth century, the British were keenly aware of the dangers of
allowing their administration to pass under the control of local
family networks and, in 1851, the Board of Revenue passed orders
prohibiting the employment of members of the same family in the
same office. This measure seems straightforward and sensible but
British notions of kin, caste and rule were not as clear as those of
modern anthropologists and the order, in fact, confused family with
caste to the extent of not permitting members of the same caste to
serve in the same office.70 Of course, the order was ignored but its
spirit was occasionally revived by Collectors who, feeling themselves
to be the victims of caste conspiracies, suddenly sacked members of
one caste group and set out to recruit members of another. As Brah-
mans were by far the most literate of ritual communities, they
naturally filled most of the government posts, and an attack on any
district's dominant office clique was likely to be followed by the
appearance of notices in the district gazette offering places to men who
were not Brahmans.71
This did not matter much while it was the result of only Col-
lectorate activity and although, as early as 1886, Fort St George's
attention was drawn to the social composition of its services, it
remained unmoved.72 But centralisation of the bureaucracy rapidly
altered the significance of communal calculations. Office conspiracies
now could control not only a district but the entire province and
British civilians were determined to prevent the formation of the
caste cliques which they thought managed such conspiracies. In
70
Board of Revenue, Standing Order 128(2).
71
F o r examples, see Hindu 11 M a y 1909; Desabhimani 16 F e b r u a r y 1907; Sadhvy
1 J u n e 1910; Kistnapatrika 26 February 1916. R.N.P.
72
G.O. 386-7 (Education) dated 27 July 1887. T.N.A.

283
 The emergence of provincial politics

1903, for example, Lord Ampthill, and everybody else in Madras,

knew that the obvious successor to Sir Bashyam Iyengar on the
High Court bench was V. Krishnaswami Iyer. But he gave the post
to C. Sankara Nair because 'he is not a Brahmin' 73 and Krishnaswami
Iyer, like the three previous native High Court Judges, was. The
rise to prominence of Mylapore greatly increased the desire of many
civilians to support people who were not Brahmans. Although they
recognised Mylapore's usefulness, they were wary of allowing it too
much power, for collaborators who have become indispensable
are worse than enemies. As the Mylapore leaders were all Brahmans,
and as the senior civilians do not seem to have concerned themselves
with anything but superficial appearances, they naturally thought
that Mylapore could be weakened by introducing non-Brahmans
into key posts. Perhaps the greatest proponent of this argument was
Sir Alexander Cardew, whose evidence to the Public Services Com-
mission was loaded with vitriol against Brahman literati,74 and who
was very active in providing non-Brahman service families with a
private channel to government.75 In 1912, under his influence, the
Secretariat produced its first positive directive on using the Brahman/
non-Brahman division as a criterion for making appointments.76
By 1918, it had gone on to insist on the keeping of separate lists of
Brahman and non-Brahman candidates and had ordered its officers
to appoint the latter first.77 The Secretariat had created an interest
in being non-Brahman, which cut across the existing structure of
patronage.
The responsiveness of the educated community to the recruitment
and promotion policies of its major employer was, not surprisingly,
very considerable. As early as 1886, when Fort St George had first
shown an interest in the communal composition of the provincial
services, non-Brahman demands for special privileges had reached
the press. 78 Fort St George's decision not to act, however, killed
this campaign which, in any case, came too early in the process
of administrative centralisation to have stirred a provincial following.
The rebirth of government interest in the communal problem in the
73
Lord Ampthill to Lord George Hamilton, 8 April 1903. Ampthill Papers. 1.0.L.
74
Evidence o f A. G. Cardew in Royal Commission on the Public Services in India.
Appendix 11, 104-16. P.P. 1914, vol. xxi.
75
G.O. 1616 (Home, Misc.) dated 6 August 1917. T.N.A.
76
G.O. 1561 (Public) dated 19 December 1912. T.N.A.
77
G.O. 19 (Home, Misc.) dated 6 January 1920; G.O. 986 (Revenue) dated 30 April
1920. T.N.A.
78
Anon., The Ways and Means for the Amelioration of the Non-Brahman Races.

284
 The vocabulary of communal politics

1910s provoked a much fiercer reaction. In 1912, the same year as

the first communal Government Order, a Dravidian Association was
set up in the city by a group of non-Brahman government servants
and lawyers.79 Over the next four years it repeatedly petitioned the
government to treat non-Brahmans as a special category and to
provide reserved posts and scholarships for them.80 From 1916, when
Mylapore went en masse into the Home Rule League, spokesmen of
the non-Brahman cause were nominated to the Legislative Council
and proffered favours by the British to denounce the Home Rule
League as a Brahman conspiracy.81 Non-Brahmanism became for a
time synonymous with anti-nationalism- a fact which surely indicates
its origins as a product of government policy.
While every non-Brahman had an interest in the non-Brahman
category, however, a great many also had contacts and connections
with Brahmans, which they wished to preserve and enlarge. The non-
Brahman movement, even among 'service' groups, was divided in a
thousand ways. The most militant communalists, naturally, were
those whose political contacts were weak. Typical of them were men
like T. Ethiraja Mudaliar82 and V. Tirumalai Pillai,83 city lawyers
who had been begging for office for years but never getting it, and T.
M. Nair, who was repeatedly beaten in elections. In fact, T. M. Nair's
career amply illustrates just how new and opportunistic this non-
Brahman combination was. Not only had he, in 1909, been closely
associated with Brahmans in the Extremist attack on Mylapore but
in 1912, before the communal issue had any general currency, the
evidence he had given to the Public Services Commission had been
used by the Brahman press to illustrate that Sir Alexander Cardew's
evidence on the Brahman hegemony did not mean that non-
Brahmans were communally prejudiced: T. M. Nair had not men-
tioned the Brahman problem once.84 Those non-Brahmans who were
well placed with Brahman agencies and in the old network politics
were distinctly cool to the movement when it turned from a simple
patronage request into a crusade to destroy Brahman influence. P.
Kesava Pillai, for example, an Anantapur lawyer who had served the
79
Justice Party Golden Souvenir, p. 257.
80
Its most celebrated petition was The Non-Brahman Manifesto, see Hindu 20
December 1916.
81
See below pp. 294-6.
82
'Ethiraja Mudaliar is a prominent Madras lawyer but where are his Judgeships?'
Non-Brahman 17 December 1916. R.N.P.
83
V. Tirumalai Pillai t o P. S. Sivaswami Iyer, 10 M a y 1912. P. S. Sivaswami Iyer
Papers. N.A.I.
84
Hindu 5 April 1913.

285
 The emergence of provincial politics

Moderate Congress in 1908 and was a Legislative Councillor, came

out as a leading opponent of T. M. Nair-style non-Brahmanism and
preached co-operation with the Brahman.85 Further, when the move-
ment was used by the British to promote loyalty and to oppose Home
Rule League demands for constitutional reform, non-Brahmans who
were powerful in the administration and who stood to gain by reform,
rapidly dropped off the communal bandwagon. C. Sankara Nair bit-
terly attacked the mendicancy of the movement which once he had
supported, and his evidence given to Parliament in 1919 is practically
indistinguishable in sentiment from that of C. P. Ramaswami Iyer.86
Of course, the non-Brahman movement soon expanded beyond the
narrow world of the western educated. A great many local magnates
were pressed into offering it their support. As we shall see later, how-
ever, few of them were to join because of an ideological commitment
to the caste cause; rather they came to it because it formed a factional
alternative to Mylapore at the centre. But in its earliest phases the
movement belonged only to the western educated and was in kind
precisely like the Extremist and Andhra movements — a campaign by
professionals and publicists who were 'out' to put themselves 'in'.
Its origins need not be sought deeply in the cultural history of South
India; they lie much more in the very novel types of government and
politics which developed under the British in the early years of the
present century.
Our analysis of the emergence of the non-Brahman movement
reveals three important points which are relevant to all the communal
movements of the period. Firstly, the language of the movement
was closely related to the language of government. In 1908, for
example, when the Morley-Minto reforms were being considered,
many of the men who were to attack Brahmans eight years later
showed no interest in the offer of communal electorates which
was made to them by Fort St George.87 Morley himself was not
thinking in caste terms, and Mylapore's non-Brahman opponents
were much more eager to castigate Mylaporeans for being lawyers -
the category closest to Morley's mind - than for being Brahmans.88
The question of Brahmanism was not considered — or reconsidered —
85
Hindu 22 December 1916.
86
See evidence of C. S. Nair in Joint Select Committee on the Government of India Bill,
ii, 551. P.P. 1919, vol. iv; evidence of C. P. Ramaswami Iyer in Evidence taken
before the Reforms Committee (Franchise). Madras (Calcutta, 1919), II, 596.
87
H o m e Public A October 1908, Nos 1 1 6 - 4 6 . N.A.I.
88
Hindu 22 J a n u a r y 1908.

286
 The vocabulary of communal politics

until after 1912.89 Secondly, these movements, whatever their pre-

tensions, did not need to have, and many did not have, a political
existence prior to the creation of the publicist and administrative
categories which they filled. It is idle for the student of politics, al-
though not perhaps of ideas, to search through the history and
meaning of 'non-Brahmanism' to discover when in the past or at what
level of abstraction in 'traditional' thought a notion of the non-
Brahman similar to that propagated by the leaders of the non-Brah-
man movement can be found. The movement emerged when the very
novel political processes of early twentieth-century Madras gave it
life. What is interesting to political history is not the ideational
antecedents of the movement but the contemporary processes. And
thirdly, throughout our period and beyond, the old socially vertical
systems of political connection continued to control a vast system of
rewards. Men moved in and out of'communal' politics with a remark-
able speed as the opportunities for rewards shifted around. Simply
because a politician appeared as a non-Brahman, a Nadar or a land-
lord, it ought not to be supposed that all of his political contacts lay
within those communities. If he were an important man, he was also
vitally related to a cross-communal network. It is not possible to
explain the manifestly variable degrees to which different men
subscribed to the same communal identity nor the changing tactics
which they were prepared to employ in support of this identity with-
out referring to their other networks of political interest.
89
So close was the connection between governmental ideas and notions of the non-
Brahman community, that much of the evidence which non-Brahmans used to
portray Brahman oppression had come out of the mouths of civilians first. Thus,
for example, The Non-Brahman Manifesto quoted Cardew's evidence to the
Public Services Commission as proof of Brahman wickedness. When the civilians
argued the non-Brahman case before Parliamentary committees, they then
proceeded to cite the non-Brahmans' quotations of their own opinions as evidence
of indigenous consciousness of the non-Brahman problem. The pattern of govern-
mental initiative and subjects' response had become so complete that a closed circle
of argument, quite apart from outside reality, had been formed.

287
 7
The Home Rule League, Justice Party and
Congress

The previous chapters have brought us to some understanding of the

processes of political change in Southern India between the 1870s and
1916. Those processes were all moving in the same direction. They
were linking together the political institutions and political interests
of a previously segmented political system. By 1917 the linkages were
complete and, although provincial and local systems had by no means
merged entirely into each other and although further changes in their
relationship would take place, they were now inextricably connected.
In the Introduction, we set ourselves a main task: to explain the major
political events of 1917-20. Already, we have set the stage for them
and have suggested reasons for their occurrence and for their massive
significance. Particular lines of conflict were being drawn by the way
that the political system was evolving, and the confrontations of
1917-20 represented the first large-scale conflict in the newly inte-
grated system. It remains for us to go through the events of these
years to make our suggestions concrete. We must show how the
Home Rule League, non-Brahman movement and Congress, as they
developed at this time, were the products of their new institutional
environment and how, indeed, their development is inexplicable with-
out reference to the processes of political change which we have been
examining.
By the end of the second year of the First World War, Mylapore
had abandoned its policy of constitutional co-operation with the
British and had adopted an agitational posture similar to that of the
1880s. The conditions of the period were perfect for it to press its
imperial overlords into a further devolution of power. The war had
forced Britain to make increased demands on India for men and
money, and many Indian politicians realised that this gave them a
chance to confront London with their own demands for political re-
form. As early as 1915, the India Office itself had recognised that
some vague and unspecified changes would have to follow the war
and, over the next few years, many powerful Indian interests made
sure that it was not allowed to forget its prognosis.
The Home Rule League, the organisation which focused the agita-
288
 Home Rule League, Justice Party and Congress

tion between 1916 and 1918, was the child of that most remarkable
Irishwoman, Mrs Annie Besant. Her decision to enter Congress
politics, after many years in opposition to Nationalist aspirations,
would be more easily explained by the psychoanalyst than the his-
torian, for it was undoubtedly taken more to bring her increased
public attention than to aid the cause of Indian self-government.1
Of more interest to us is the general effect of her move into the
national arena. In her pre-political days, she had been extremely
close to many of the leading men of Mylapore and had been involved
with them in a host of cultural, religious and educational projects.2
Indeed, so readily was she identified with Mylapore that she and
they shared most of the same enemies — particularly Kasturi Ranga
Iyengar, C. Vijayaraghavachari and T. M. Nair.3 As she advanced
into agitational politics, Mylapore provided her with the reputations
and abilities of its most celebrated members. Sir S. Subramania Iyer
was president of her League, C. P. Ramaswami Iyer her legal adviser,
and G. A. Natesan and L. A. Govindaraghava Iyer among her most
loyal and vocal supporters.
Mrs Besant's activities, by 1916 when they first achieved any
real significance, performed a vital service to Mylapore. So complete
had been Krishnaswami Iyer's destruction of the agitational Congress
that the organisation had practically ceased to exist outside the
Provincial Congress Committee.4 Although Mylapore, as we have
seen, possessed innumerable personal contacts across the presidency,
these were unorganised and diffuse. It would have taken considerable
1
For Mrs Besant's early opposition to Indian nationalism set Hindu 8 August 1910;
C. Vijayaraghavachari to P. Kesava Pillai, 22 February 1907. P. Kesava Pillai
Papers. N.M.M.L.; 'A History Sheet of Mrs Besant' in Home Political A March
1918, No. 247 and K.W. N.A.I.
2
Sir S. Subramania Iyer and Justice T. Sadasiva Iyer were said to 'worship her almost
as a mother'. P. Kesava Pillai to A. C. Parthasarathi Naidu, 5 October 1917. P.
Kesava Pillai Papers. N.M.M.L. V. Krishnaswami Iyer, V. S. Srinivasa Sastri and P.
S. Sivaswami Iyer were involved with her in the Benares Hindu College. V. S.
Srinivasa Sastri to G. K. Gokhale, 6 February and 24 March 1908. V. S. Srinivasa
Sastri Papers. N.A.I.
3
Between 1912 and 1914, Mrs Besant sued both the Hindu andT. M. Nair for libel.
C. P. Ramaswami Aiyar, Biographical Vistas (Bombay, 1968), p. 249; T. M. Nair,
The Evolution of Mrs Besant (Madras, 1918); C. Vijayaraghavachari to P. Kesava
Pillai, 22 February 1907. P. Kesava Pillai Papers. N.M.M.L.
4
Following the 1914 session of the Indian National Congress, held in Madras, the
Madras P.C.C. ordered an investigation of district machinery. It found 'consider-
able indifference' everywhere. Of the 22 recognised D.C.C.s, only eight bothered
to reply to the P.C.C's circular. Report of the Madras Provincial Congress Committee
for the year ending 31st December 1915 (Madras, 1916).

289
 The emergence of provincial politics

time to redevelop the Congress for agitational purposes. Mrs Besant's

great virtue lay in her ability to provide her campaign with ready-
made machinery. As the president of the Theosophical Society, she
controlled an organisation with several thousand members, which
linked the presidency capital to every large mofussil town. When she
converted this to political purposes, she was able to inaugurate her
movement with prepared support and a sophisticated structure of
command in as many as thirty-four separate localities.5 As a news-
paper baron of considerable size, she had to hand the means of
communicating her message across the presidency.6 As the creator
of a student welfare organisation in Madras city, she possessed close
ties with a large body of highly volatile material which could be used
to heat any political agitation.7 Finally, as the head of several educa-
tional trusts, she disposed of a large patronage which enabled her to
command an even wider following.8 Mrs Besant provided Mylapore,
which was seeking to demonstrate to the British both its anger and its
power, with resources to make its arguments felt much more quickly
and keenly than would otherwise have been possible.
Yet Mrs Besant's Home Rule League also served the purposes of
many of Mylapore's enemies, both of the Egmore and more extreme
varieties. Krishnaswami Iyer and the Allahabad Convention had
driven them from the Congress and kept them out of provincial con-
sideration since 1908. They had been forced to adopt such hopeless
anti-Mylapore ploys as the Andhra movement and the Varnashrama-
dharma Association. The Home Rule League campaign, which was
looking for agitational support of any character, presented them
with an opportunity of returning to central political organisations
5
Organisation of the Home Rule League. Passed at the first meeting of the Council,
8th October 1916 (Madras, n.d.); Hindu 4 September 1916; The Theosophist
November 1916; 'History Sheet of Mrs Besant' in Home Political A, March 1918,
No. 247 and K. W. N.A.I.
6
In 1914, as part of her initial move into politics, Mrs Besant had bought the daily
Madras Standard and converted it into New India. This, together with her weekly
Commonweal and her religious journals, made her the most formidable press baron
in Madras.
7
In 1914 Mrs Besant had founded the Young Men's Indian Association to appeal to
the 3,000 student population of Madras city. 'Draft Copy of Prospectus and Rules
of the Young Men's Indian Association' in A. Besant to P. S. Sivaswami Iyer, 12
January 1914, P. S. Sivaswami Iyer Papers. N.A.I. In 1916 and 1917, this associa-
tion acted as a centre for the dissemination of Home Rule propaganda in the city.
Home Political 'Deposit' September 1916, Nos 1-7. N.A.I. Home Political 'Deposit'
January 1917, No. 42 and November 1917, Nos 6-7. N.A.I.; Hindu (weekly) 11 May
1917.
8
'History Sheet of Mrs Besant' in Home Political A March 1918, No. 247 and K.W.
N.A.I.

290
 Home Rule Leaguey Justice Party and Congress

and of making their presence known once again in Congress circles.

Of course, they did not like the association of Mrs Besant and Myla-
pore with the movement. Kasturi Ranga Iyengar's new ally, S.
Satyamurthi, savaged the League leadership publicly,9 and C.
Vijayaraghavachari's Salem protege, C. Rajagopalachari, attempted
to weaken the hold of the Madras city politicians on its machinery.10
But entry provided them all with advantages which they could not
ignore. With less than perfect faith, they threw in their lot with Mrs
Besant. By the autumn of 1916, the Home Rule League had accom-
plished the seemingly impossible: it had united the warring factions
of the old Congress under a single leadership and it had projected
Mrs Besant to a position of national importance.
To many British civilians, perched precariously in the higher
regions of the Secretariat, the emergence of the Home Rule League
and Mylapore's openly treasonable conduct marked a serious danger,
both to the empire and to themselves. From about 1911, Mylapore's
growing power had created resentment in the I.C.S., and several
officials had spared no pains to oppose any further devolution of
power to it.11 It was from this time that the essays at communalism
in the services and the attempt to nominate direct communal
representatives to political institutions had begun. The events of 1916
convinced even the more liberal civilians that a policy of finding
Indian replacements for Mylapore was an urgent priority. When the
Madras Government tried to take steps to deal with what it saw to be
the rising tide of sedition, it found itself shackled by the agents of
Mylapore, who were now part of it. S. Srinivasa Iyengar, the Attorney-
General, used his influence to prevent prosecutions for sedition;12
the High Court Judges T. Sadasiva Iyer and T. V. Seshagiri Iyer re-
pealed sentences given by lower courts for sedition;13 by various ad-
ministrative manipulations, large sums of government money found
their way to the Home Rule League press to pay for advertisements;14
in Madras city, C. P. Ramaswami Iyer appeared to have taken over
9
Home Political 'Deposit' March 1916, No. 50. N.A.I.; Hindu 6 September 1916.
10
Hindu 19 September 1916.
11
For an example of the growing hostility between Mylapore and the I.C.S., see the
reply of the Madras Government to the Secretary of State's suggestion that more
Senate seats should be opened to election by the graduates. Letter No. 305 (Home,
Education) Government of Madras to Government of India, dated 5 March 1917
in Education A May 1917, Nos 1-17. N.A.I.
12
Note signed S. S. Iyengar dated 7 April 1919 m G.O. 653 (L and M, M) dated 25
April 1919. T.N.A.
13
Home Political 'Deposit' November 1917, No. 7. N.A.I.
14
Note signed A. Cardew dated 1 March 1918 in G.O. 340 (Home, Misc.) dated
4 April 1918. T.N.A.

291
 The emergence of provincial politics

the Corporation and to be using its patronage of appointments and

contracts to further the nationalist cause; 15 in the University, G. A.
Natesan finally succeeded in bowing the largely European Syndicate
to his authority in the Senate. 16 For many British civilians, the enemy
seemed to be already within the gates.
Moreover, precisely because the Mylapore connection rep-
resented not merely the interests of a few western-educated men but
also those of many magnates immediately powerful in society, the
Home Rule campaign confronted the government with opposition
which was both socially deep and diverse. The British could not jest
at the movement as a ritual performed by a handful of 'Babus' and a
mad Irishwoman while such huge landed proprietors as the Kumara
Maharaja of Vizianagram, the Raja of Ramnad, the zamorin of
Calicut and the zamindar of Munagala, and such enormous financial
powers as the Gujerati banker Lodd Govindoss, the Muslim Badsha
family and the Calivalla brothers could be seen involved in the
agitation.17 As the League's propaganda developed, senior civilians
were given a further cause for concern by the way in which significant
economic interests, disturbed either by the war or by the advance of
administrative efficiency, were also drawn towards it. In 1916,
15
A. C. Parthasarathi Naidu to A. Cardew, 16 September 1916 inG.O. 414 (Home,
Misc.) dated 26 April 1917; G.O. 175 (L andM, L) dated 7 February 1918. T.N.A.
16
Hindu 28 October 1916.
17
Most of these involvements were the result of personal connection either in
cultural and religious patronage movements, or in the courts and councils or in the
distribution of Secretariat and Dharmarakshana Sabha rewards. The Kumara
Maharaja of Vizianagram and the Raja of Ramnad sought to give the Madras
Landholders' Association a strong Home Rule flavour when Montagu visited
Madras in December 1917. Desamata 17 November 1917; Swadesamitran 14
December 1917; Dravidapatrika 23 January 1918. R.N.P. The zamorin of Calicut
was drawn into Home Rule politics by his relatives, particularly the lawyer K. P.
Raman Menon, who were local organisers. Justice 15 May 1917. R.N.P. Munagala
was a personal patron of the old Andhra Extremists. The Calivalla brothers and
Lodd Govindoss we saw before. In 1918, C. Cunnan Chetty sided with C. P.
Ramaswami Iyer in a factional struggle with P. Thyagaraja Chetty on the
Pachayappa's trust board. Hindu 6, 10, 11 April, 21 May, 6 and 20 June 1918.
Lodd Govindoss was a founding vice-president of the Home Rule League-oriented
Madras Presidency Association. Madras Mail 17 September 1917. The Badsha
family we saw in the 1887 Congress. In spite of the separatist tendencies of Muslim
politics, they had always remained close to Mylapore. C. P. Ramaswami Iyer, for
example, was the legal adviser of the Muslim Education Society, in which they
were prominent, and the Calivalla brothers were among its patrons. The Badshas
themselves also donated money to Hindu charities. Hindu 23 April 1908,25 April
1912, 12 March 1918. They were organisers of the Madras Muslim League which,
of course, supported Home Rule League—Congress agitation at this time.

292
 Home Rule League, Justice Party and Congress

for example, prominent Home Rule politicians picked up the issue of

emigrant labour and called on the government to regulate more
closely and, in some instances, stop completely the traffic.18 These
demands, although for different publicly stated reasons, were the
same as those of a large number of Tanjore mirasidars who objected
to emigration because it depleted their labour force and pushed wages
above subsistence level. That the association of ideas between the
League and this powerful landed interest was due to more than coin-
cidence was made clear by the opposition of the League to government
measures to weaken the physical control which mirasidars exercised
over their labourers.19 So seriously did Fort St George regard the
threat of a tight combination of League politicians and Tanjore
mirasidars that it reversed its emigration policy and legislated against
the advice of its emigration experts.20 The Home Rule League also
developed political connections with economic interests in the
important hide and skin trade, which was hit by the dislocation of
communications,21 with village officers who, as we saw, were under
pressure from bureaucratic reform,22 with Indian merchants who
sought to break the financial privileges of European business houses,23
and, later, even with native factory labour in the few European-
owned industrial undertakings of the South. 24 These economic forces,
when combined with the networks of religious revivalism which Mrs
Besant, as a religious leader, was able to mobilise, gave the British
every reason to see the Home Rule League as an extremely serious
threat to the continued existence of the raj
But it was not only the empire — as a concept — which the strength
of the movement suggested was under attack. Much of Mrs Besant's
polemic, which stressed Indian religious and national revival, served
to concentrate the agitation on the fact that the government, besides
being unsatisfactory, was alien.25 The British servants of Fort St
18 19
Madras Mail 5 February 1917. Hindu 2 February, 3 and 9 April 1918.
20
Home Political 'Deposit' April 1917, No. 61. N.A.I.
21
Home Political 'Deposit' May 1917, No. 40. N.A.I.
22
Hindu 23, 24 and 27 December 1919.
23
The Southern India Chamber of Commerce. Report. March 1918—February 1919
(Madras, 1920), pp. i x - x i v .
24
B. Shiva R a o , The Industrial Worker in India (London, 1939), p p . 15—16; Desab-
haktan 9 May 1918. R.N.P.; H o m e Political 'Deposit' September 1918, Nos 1 9 - 2 0 .
N.A.I.
25
Chief Secretary, Government of Madras, to Secretary, Home Department,
Government of India, in G.O. 1019 (Public) dated 10 August 1917. T.N.A.; Slater,
Southern India. Its Political and Economic Problems, p. 266; Madras Mail 28
February 1917; Irschick, Politics and Social Conflict in South India, pp. 57—8.

293
 The emergence of provincial politics

George found themselves challenged as much for being British as for

being servants. They were the more susceptible to this attack for they
regarded themselves as a race apart from and superior to that of the
indigenous population. Indeed, in many ways they constituted a
separate caste in Madras society, which behaved in politics much more
like a community than most Indian castes. I.C.S. men tended to see
the League as a danger not only to the government for which they
worked but also to themselves and their social position. From 1917,
when the appointment of Edwin Montagu as Secretary of State for
India produced in London and New Delhi a policy of conciliation
towards the League, 26 many Madras civilians felt themselves betrayed
as their superiors threatened to sacrifice them to Indian political
sentiment. They did not relish the idea of having to share their offices
and status with the men of Mylapore. The traditional battle between
Fort St George and its London and New Delhi superiors took on a
new facet, therefore, as the civilians of Fort St George began to fight
for their personal survival. That this, rather than concern for the
empire as a whole, was their main preoccupation became clear in
1920 when they rejected every clause of the Montagu—Chelmsford
Report and threatened to resign en masse if the report were imple-
mented, leaving India and the empire to its fate. 27
The severe and deepening hostility to Mylapore and the Home
Rule League of a small group of highly placed civilians — Cardew,
H. F. W. Gillman, Sir Lionel Davidson and Sir Murray Hammick —
was crucial to the development of politics between 1916 and 1920.
In late 1916, faced with the growing strength of the Home Rule
League crusade, they began to look around for elements which could
be used in a counter-movement of loyalty. 28 At first, communalism
alone was not central to their plans. They responded to the initiatives
of a few zamindars who wished to oppose the League on a straight
ticket of loyalty to the raj29 and even found subsidies for one Brahman
26
M o n t a g u ordered Mrs Besant's release from internment, the Government of India
withdrew its b a n o n students and semi-government employees, such as public
prosecutors, participating in H o m e Rule agitation and, in December 1 9 1 7 ,
M o n t a g u showed obvious favour to such Madras H o m e Rulers as C. P. Rama-
swami Iyer w h e n considering suggestions for the reforms h e had promised. G.O.
1474 (Public) dated 2 2 December 1917. T.N.A.; M o n t a g u , An Indian Diary,
pp. 120—30.
27
See P. & J. (Reforms) File 95. / . O . L .
28
For discussions on the nature of this counter-movement or 'show' see notes of
Sir Harold Stuart, H. F. W. Gillman and Sir Alexander Cardew in G.O. 447
(Home, Misc.) dated 8 May 1917 and G.O. 414 (Home, Misc.) dated 26 April 1917.
T.N.A.
29
Madras Mail 14 March, 8 May, 30 June and 2 July 1917.

294
 Home Rule League, Justice Party and Congress

newspaper editor who promised to denounce the League in his broad-

sheet.30 However, it was not long before the main thrust of their
campaign turned to the non-Brahman category. In part, this was
because the Madras Dravidian Association and the non-Brahman
interest, which Cardew had been developing since 1912,31 provided
a ready-made organisation which would work for them in return for
patronage. Cardew and Gillman offered this patronage: they were
in correspondence with the M.D.A. leadership and attempted to bend
the rules of service to allow government officials to join the South
Indian People's Association (the first non-Brahman political associa-
tion), and they knew of the founding of the S.I.P.A., several weeks
before its inaugural meeting took place.32 In 1917, they demonstrated
clearly the nature of the S.I.P.A.'s official connection by manipulating
the delegations which met Montagu in order to give loyalist non-
Brahmans the maximum of representation and Congress non-Brah-
mans the minimum.33
Undoubtedly, a second reason for the use which Madras civilians
made of communalism lay in the way that London and Calcutta col-
lected information for the reforms. A series of Montagu-appointed
commissions arrived in Madras to hear evidence and to test public
opinion. The civilians' own opposition to constitutional reform, and
indeed that of the resident British business community, was unlikely
to carry weight with commissioners who were concerned with the
future of India as a whole. Moreover, the commissioners, if they
had any experience of India at all, had received it in Northern India
or at imperial headquarters, where political analyses tended to be
30
Srinivasa Venkatachari of t h e Hindu Nesan. G.O. 4 1 4 (Home, Misc.) dated 26
April 1917, T.N.A.
31
Since 1912, he h a d been using his personal influence to pass patronage to such
non-Brahman service families as t h e Maddireddi. G.O. 1616 (Home, Misc.) dated
6 August 1917. H e was instrumental in obtaining grants for separate caste hostels,
as among t h e Sourashtras. South Indian Mail 30 April 1917. R.N.P. A n d also in
twisting Education D e p a r t m e n t rules to aid t h e Kammas. Note signed A. Cardew
dated 5 J a n u a r y 1918 in G.O. 102 (Home, Misc.) dated 31 J a n u a r y 1918. T.N.A.
And also in campaigning for a revision of the rules of government employment
to allow communal Indian Christian Associations to demonstrate against t h e
H o m e Rule League. G.O. 970 (Home, Misc.) dated 6 August 1917. T.N.A.
32
See Note signed H . F . W. Gillman dated 19 October 1916. in G.O. 447 (Home,
Misc.) dated 8 M a y 1917. T.N.A.', also P. Thyagaraja Chetty to A. Cardew, 12
February 1917, in G.O. 147 (Home, Misc.) dated 15 F e b r u a r y 1917. T.N.A.
33
Hindu 11 February 1917; Proceedings of the Council of t h e Madras Muslim
League, 30 November 1917, in 'Proceedings of t h e M a d r a s Muslim League, 19
M a r c h 1917 to 21 February 1921'. Madras Mahajana Sabha Papers. N.M.M.L.;
Hindu 15 November and 15 December 1917; G.O. 552 (Public, Confidential) dated
3 July 1918. T.N.A.

295
 The emergence of provincial politics

made in terms of community. The civilians, therefore, needed to find

an argument, based on communal principles, with which to fend off
the good intentions of their superiors. What better category could
be thought of than the non-Brahmans, whose existence as a group
could be shown in innumerable departmental statistics, who could
be shown by examinations of education and public service employ-
ment (but not wealth or position in local bodies) to be weak and
who numbered 98 per cent of the entire population? The non-
Brahman category carried inside it the hopes and fears of the Madras
I.C.S.
Of course, the thinking out of this strategy need not have been as
coldly Machiavellian as we have supposed. Prior to 1916, many
civilians had already classified the threat of Mylapore as the result of
Brahman conspiracy, and so were already half-way to concluding
that people who were not Brahmans were losing out. But certainly
some explanation must be offered for the way that so much civilian
evidence given to various commissions at this time was confused, and
provided ill-conceived judgments on the basis of facts which were
demonstrably wrong. For example, in spite of their evidence of
irrevocable cultural, social and political division between the Brahman
and the non-Brahman, no civilian could tell Southborough why, only
ten years before, the Madras Government had vehemently opposed
the principle of communal representation on the grounds that com-
munal divisions were not strong in Madras.34 Or again, in London
in 1919, Cardew and Hammick, carried away by the rectitude of their
reasoning, told a Parliamentary Committee that not only would non-
Brahmans be extinguished by a broadened constitution but that they
were already extinct under the present Morley-Minto scheme. Un-
fortunately for them, while they were talking non-Brahmans were
winning eleven of the fifteen seats in the Legislative Council election -
a fact which they were unable to explain when asked.35 A large part of
the non-Brahman propaganda was written and performed by the
34
Evidence of H. G. Stokes in Evidence Taken before the Reform Committee {Franchise).
Madras (Calcutta, 1919), 11, 585. Stokes also admitted that the non-Brahman
category, so vehemently defended by some of his colleagues as a concrete political
entity, was no more than the result of rule-of-thumb definition: 'The constitution
of the electorates proposed by Fort St George was based upon an identity of
interests among the voters. As regards the non-Brahman community the inter-
ests of the different sub-castes might not be identical but government must
draw the line somewhere owing to the necessity for keeping the Council within
practical limits.' Ibid., p. 586.
35
See evidence of A. Cardew in Joint Select Committee on the Government of India
Bill, n, 338. P.P. 1919, vol. iv.

296
 Home Rule League, Justice Party and Congress

leading Madras civilians; and it ought to be judged more by what it

was meant to achieve than by what it appeared to say.
The men who came forward to join the S.I.P.A. (and, later, the
South Indian Liberal Federation and Justice Party) were drawn from
the most diverse backgrounds. One of the main elements in the as-
sociation, of course, was the professional 'non-Brahman' interest -
the group of lawyers and bureaucrats who formed the Madras
Dravidian Assocition. As we saw, it did not include non-Brahmans
who were well-connected to Mylapore or, even, Egmore. Men like
V. Masilamani Pillai, C. Sankara Nair and T. V. Gopalaswami
Mudaliar36 had no use for the S.I.P.A.'s mendicancy. It was com-
posed only of the unsuccessful.
Behind this collection of career-seeking educated non-Brahmans,
which extended to a host of journalists and petty publicists attracted
to the S.I.P.A.'s patronage potential,37 came a series of local magnates
who had suffered from Mylapore's rise to power. Most prominent of
these was Pitti Thyagaraja Chetty whose troubles in the Madras Corpo-
ration and Pachayappa's charities we have seen. Another was P.
Ramarayaningar (later the Raja of Panagal), a member of a wealthy
zamindari family from Chittoor. In spite of his petitions to P. S.
Sivaswami Iyer, one of his local enemies had received important
appointments in local self-government from the Secretariat.38 In
1915, he had joined the Andhra movement, to protest at Mylapore's
policies, but it did him little good for he lost a Council election in
July 1916.39 The new non-Brahman weapon was well suited to his old
36
T . V. Gopalaswami Mudaliar, a celebrated lawyer and journalist, was an i n t i m a t e
of K a s t u r i Ranga Iyengar and associated with h i m t h r o u g h o u t this period.
37
Such, for example, as O. Kandaswami Chetty, editor of t h e Social Reform Advocate
who, until 1916, had been on t h e payroll of the B r a h m a n S. Srinivasa Iyengar,
president of t h e Madras Social Reform Association, b u t t h e n sought succour at
t h e S.I.P.A. office. See Hindu (weekly) 29 April, 6 and 13 M a y 1926. A n d A. C.
P a r t h a s a r a t h i Naidu, the editor of Andhraprakasikha, which had r u n into financial
difficulties. After begging unsuccessfully from P. S. Sivaswami Iyer and Sir
Alexander Cardew, he accepted an offer from P. Thyagaraja C h e t t y to make
Andhraprakasikha the T e l u g u organ of t h e S.I.P.A. So ashamed was he of his new
associates that he wrote to P. Kesava Pillai explaining t h a t he did not really
dislike B r a h m a n s but had been forced to move with the times. A. C. P a r t h a s a r a t h i
Naidu to P. S. Sivaswami Iyer, 14 September 1916. P. S. Sivaswami Iyer Papers,
N.A.I.; G.O. 414 (Home, Misc.) dated 26 April 1917. T.N.A.; A. C. P a r t h a s a r a t h i
Naidu to P. Kesava Pillai, 9 October 1917. P. Kesava Pillai Papers. N.M.M.L.
38
P. R a m a r a y a n i n g a r to P. S. Sivaswami Iyer, 29 J u n e 1912. P. S. Sivaswami Iyer
Papers. N.A.I.
39
R a m a r a y a n i n g a r was president of the 1915 A n d h r a conference. Hindu 12 M a y
1915. H e was defeated in 1916 by the Trichinopoly zamindar K. V. R a n g a s w a m i

297
 The emergence of provincial politics

purposes. A third magnate who threw his weight behind the S.I.P.A.
was the Raja of Pithapuram. For more than a decade he had been
battling with Mylapore and its Andhra agents to retain control of his
important college at Cocanada. In 1904, his choice for principal, R.
Venkataratnam Naidu, had upset N. Subba Rao and V. Bashyam
Iyengar, who wanted the place for V. S. Srinivasa Sastri(Krishnaswami
Iyer's protege). Subba Rao had led an agitation against Venkataratnam
Naidu on the issue of the new principal's advocacy of advanced social
reform.40 Discord between the raja and local Mylapore-oriented
lawyers continued unabated for the next twelve years and in 1916, a
few weeks before the foundation of the S.I.P.A., M. Ramachandra Rao
had promoted a confrontation between the college students and the
college administration in the hope of ousting Venkataratnam Naidu.41
The raja was naturally drawn to the anti-Mylapore crusade, which
had gathered in Madras city under the non-Brahman banner.
Obviously, however, all three magnates, who were typical of a great
many others who joined the S.I.P.A., were not enemies of Brahmans
as such. Indeed, all three had been closely associated with some
Brahmans in the past and saw no reason to break many of these con-
tacts, even in their anti-Brahman period.42 Their contention was with
certain very specific Brahmans in Mylapore.
Iyengar in the landlords' Imperial Legislative Council election. Rangaswami
Iyengar was the son o f a previous Legislative Councillor, connected to the
Dharmarakshana Sabha through the Srirangam temple and a H o m e Ruler.
It did not escape contemporaries that four o f the nascent S.I.P.A. leaders -
Ramarayaningar, P. Thyagaraja Chetty, T . M. Nair and K. V. Reddi Naidu -
all had been defeated in various Council elections in 1916 by m e n with H o m e
Rule and Mylapore connections. Prapanchamitran 2 0 M a y 1917. R.N.P.
40
'Extract from the memoirs of D r B. Pattabhisittaramayya' in Brahmarishi Dr Sir
Raghupathi Venkata Ratnam Naidu. Birth Centenary Souvenir, Mahavanavami,
1962 (n.p., n.d.)5 pp. 2 - 5 .
41
Hindu (weekly) 6, 20 and 27 April, 11 and 18 May, 1 June 1916.
42
P. Thyagaraja Chetty had been involved in the 1880s Congress with Brahmans;
in Corporation politics, he was allied to K. Desikachari; in his banking enter-
prises he was linked to Sir V. C. Desikachari; in private life he was intensely
orthodox, would not eat food not prepared by a Brahman, and in 1915 joined C. P.
Ramaswami Iyer in protesting about remarks made by a European Corporation
official on the sanitary habits of Brahmans. P. Ramarayaningar possessed an
M.A. in Sanskrit and had been educated in the house of C. V. Sundara Sastri,
P. Ananda Charlu's adopted brother; in the Andhra movement he had been
associated with almost nobody but Brahmans, and as we shall see, his term as
Chief Minister of Madras was not characterised by communalism. One of
Pithapuram's most powerful contacts, not to say representatives, in Godavari
politics was D. Seshagiri Rao, for many years standing vakil to the estate and
district board president 1 9 1 7 - 2 0 . Seshagiri Rao was a Brahman.

298
 Home Rule League, Justice Party and Congress

The S.I.P.A.'s initial appeal went out to various local interests

which had been put under pressure by developments at the centre of
government since 1910. It repeatedly petitioned against Dharma-
rakshana Sabha activities in the temples and opposed all attempts at
legislative interference in temple affairs; 43 it raised to the gaze of the
province innumerable local examples of Mylapore's spoliation of
school committees and charities; 44 its newspapers provided columns
in which the latest scandal concerning Mylapore's control of appoint-
ments could be brought to public attention. 45 Its purpose was to
compile all manner of grievances against Mylapore and the present
political system, to wrap them in a communalist and loyalistflagand
to beg the British for help.
The fact that, via Cardew, Gillman and others, it did obtain help
and access to the Secretariat, however, meant that the S.I.P.A. rapidly
became more than a mere sounding board for local and professional
'out' factions. By providing the S.I.P.A. with administrative influence,
the Madras civilians helped to turn it into a patronage machine which
rivalled that of Mylapore. In 1917, the non-Brahmans' two greatest
publicity victories were due entirely to the fact that the S.I.P.A.
leadership was able to offer, in Mylapore style, very concrete rewards
from the Secretariat to its followers. At Cuddalore and Coimbatore,
important Congress conferences were broken up by the action of two
magnates who had no personal quarrel with Mylapore and who had
been sympathetic to the Home Rule League at its foundation. But
both needed the favour of the Secretariat for particular projects and
both felt that, at that time, the S.I.P.A. connection was more favour-
able to their purposes.
At Cuddalore, A. Subbarayalu Reddiar, lawyer, landowner,
municipal council and temple committee chairman and taluk board
president, was instrumental in opposing the holding of the 1917
Congress Provincial Conference. He was, undoubtedly, the most
powerful man in South Arcot. For many years he had been on the
closest of terms with Mylapore46 and, in mid-1916, he was chairman
of the D.C.C. which asked Mrs Besant to preside at a proposed
43
Hindu 13 March 1918; G.O. 175 (L and M, L) dated 7 February 1918; Home
Judicial B February 1920, No. 17. N.A.I.; Non-Brahman 18 March and 29 April
191 j.R.NP.
44
F o r example, t h e Pachayappa's affair, Hindu 6, 10 and 11 April, 21 May, 6 and 20
J u n e 1918; t h e Tinnevelly College Committee, Hindu 1 7 , 2 2 , 2 4 , 2 5 and 30 J a n u a r y
and 15 F e b r u a r y 1918; t h e Kallidaikurichi College, Hindu 5 November 1918.
45
F o r examples, Non-Brahman 20 J a n u a r y 1917; Justice 13 August 1919. R.N.P.
46
Sivaswami Iyer made h i m taluk board president. A Subbarayalu Reddiar t o
P. S. Sivaswami Iyer, 12 April 1912. P. S. Sivaswami Iyer Papers. N.A.I.

299
 The emergence of provincial politics

Provincial Conference the next year. After his invitation had been
accepted, however, Mrs Besant had launched her League and had
become persona non grata with the Secretariat civilians. The opening
out of the institutions of local self-government and the real possibility,
late in 1916, of a non-official district board president in South Arcot
put Subbarayalu Reddiar in an acutely embarrassing position. He
withdrew the invitation to Mrs Besant and was pushed out of the
D.C.C. by his oldest local enemies.47 He demonstrated his loyalty to
the Secretariat civilians by agitating against the holding of the con-
ference - although, it must be said, his agitation was more anti-
Besantine than anti-Brahman.48 His judicious change of political coat
came at exactly the right time for his office aspirations; later in 1917
he became one of the first non-officials to be nominated to the presi-
dency of a district board.
At Coimbatore, the fracas was created byT. A. Ramalingam Chetty,
a lawyer and rich banker who, for many years, had been vice-president
of the district board and a well-known Congressman. According to a
local resident of Coimbatore, Ramalingam's switch from the Congress
to the S.I.P.A. was the result of his failure to win an election to the
Legislative Council in 1916. He had lost to the Salem lawyer, B. V.
Narasimha Iyer, but Sir Harold Stuart, a senior member of the Secre-
tariat, promised to hold an inquiry into the election, with a view to
disqualifying Narasimha Iyer, a prominent Home Ruler. Stuart's
terms were that Ramalingam arrange a hostile reception for Mrs
Besant who had been made president of the forthcoming district
conference.49 At the time, the Coimbatore municipal chairmanship
election was taking place and, by gathering together a party of
councillors loyal to him or tied to S.I.P.A. interests in Madras city or
opposed to particular Home Rulers, 50 Ramalingam was able to force
47
T h e opposition t o h i m was led by t h e B r a h m a n lawyer V. Srinivasachari, w h o
had stood against h i m in a Legislative Council election in July a n d a municipal
chairmanship election in November 1916. Srinivasachari worked in front of the
wealthy local Muslim Maracair interest which h a d long been opposed to S u b -
barayalu's power. Subbarayalu replied to his critics by accepting a resignation
from t h e temple committee, which Srinivasachari h a d tendered during a previous
dispute in 1914. I n 1917, Subbarayalu drove t h e Maracairs off t h e district
board. Hindu 7 J u n e , 3 July, 11 September, 3 November, 1 3 , 1 5 a n d 19 December
1916; Hindu 12 J u n e 1918.
48
Hindu (weekly) 11 a n d 18 M a y 1917.
49
K. V. Srinivasa Iyer to P. S. Sivaswami Iyer, 18 J u n e 1917. P. S. Sivaswami
Papers. N.A.I.
50
T h i s alliance included C. S. R a t n a s a b h a p a t i Mudaliar, a powerful landlord and
b a n k e r w h o was an old factional opponent of C. V. V e n k a t a r a m a n a Iyengar, a

300
 Home Rule League, Justice Party and Congress

the main candidate, M. Sambadam Mudaliar, a Theosophist who had

actually formed the Coimbatore Home Rule League, to join him. 51
Ramalingam provided the British civilians with as fine a show of
loyalty as they were ever to receive in Madras.
By offering hard rewards to the men who supported the S.I.P.A.,
British civilians in the Secretariat succeeded in turning the
S.I.P.A. leadership into a Mylapore-like clique at the centre. The
S.I.P.A. leaders, in the manner of the chief Mylaporeans, were able to
use the new powers of government to make a series of person-to-
person deals with the magnates of the locality. Wealthy and locally
powerful men, who were drawing a blank in their appeals toMylapore,
could shift their attention to the newly prominent non-Brahmans and
seek succour from them. The Nadar merchants and bankers of Ramnad,
for example, who had been trying and failing since 191 o to obtain a seat
in the Legislative Council for their caste association, were quick to
see the possibilities of the new situation. They began to invite S.I.P.A.
leaders instead of Mylapore Brahmans as the visiting presidents of
their annual caste conferences.52 In 1920, the S.I.P.A. leaders (as the
Justice Party Ministry) granted the Nadar request for representation
in the legislative.
The S.I.P.A. connection, then, began to form a viable alternative
to Mylapore at the centre. Importantly, however, it did not destroy
Mylapore, which still had many strings to pull. Neither A. Subbarayalu
Reddiar nor T. A. Ramalingam Chetty, for example, were prepared to
sever all contact with their former friends. Subbarayalu Reddiar did
not resign from the Congress, refused to accept direction from the
S.I.P.A. in Madras city53 and was adjudged to have stood as a
'Moderate' rather than as a member of the Justice Party at the 1920
Legislative Council elections;54 T. A. Ramalingam Chetty turned
lawyer and banker and the chairman of the conference reception committee, and
V. Veruvada Chetty, P. Thyagaraja Chetty's relative and business agent in
Coimbatore. However, it did not last for long. By July 1919, V. Veruvada Chetty,
his commitment to his relation notwithstanding, had joined forces with the
Brahman Home Ruler K. Narayana Sastri to hound C. S. Ratnasabhapati Mudaliar
in the municipal council. Hindu 1 August 1919.
51
K. V. Srinivasa Iyer to P. S. Sivaswami Iyer, 18 June 1917. P. S. Sivaswami Iyer
Papers. N.A.I. For the creation of the Coimbatore Home Rule League see Hindu
23 November 1916.
52
Hardgrave, Nadars of Tamilnad, pp. 174-5.
53
Both the S.I.P.A. and the S.I.L.F. boycotted the Madras sittings of the South-
borough Commission on Franchise reform but A. Subbarayalu Reddiar accepted
nomination as a local member of the commission. Justice 16 October 1918. R.N.P.
54
Reforms (Franchise) B March 1921, Nos 34-99. N.A.I.

301
 The emergence of provincial politics

away from non-Brahman politics during 1918 and 1919, became

active in the Mylapore-dominated Moderate Conference and was
appointed district board president of Coimbatore in 1920 through the
exertions of R. Ramachandra Rao, the departmental Secretary who
was part of the Mylapore network.55
What had happened was that two Indian political 'parties' were
now standing at the Secretariat, both of which were, to some extent,
in power at the same time. Of course, this greatly extended the scope
of factional warfare. The magnate-based factions which inhabited
almost every local institution - municipalities, rural boards and temple
committees - previously had been faced with a choice of coming to
terms with Mylapore or suffering. Now they could play with two
provincial leaderships and take measures against local opponents who
were 'in' with one by joining the other. All had an incentive to reach
out to the capital. The result was that the non-Brahman communal
and the Home Rule issues, which governed the provincial political
debate, found themselves transplanted to an extraordinary environ-
ment which was conditioned by squabbles over taxation, bye-laws
and contracts and the eternal struggle for local power. Naturally,
there was little correlation between the principles of political organisa-
tion at the two levels. In Periyakulam town, Madura district, for
example, a Jesuit priest began to further the cause of so-called Brahman
dominion. Opposition to the municipal chairman, the zamindar of
Doddapanayakanur, was coming from several groups: firstly, men
alienated by the zamindar's administration of patronage; secondly,
a group of young lawyers, newly established in this expanding town
and eager to capture the municipality by pulling together diverse
grievances in their Ratepayers' Association; and thirdly, Father Marie
Louise, a Jesuit whose missionary school was suffering from competi-
tion with the zamindar's Hindu High School. The zamindar's own
municipal party was no less complex than its opposition. It consisted
of various interests integrated by his patronage and included one of
the new lawyers, a Brahman, who had been bought off. Although the
battle for municipal control dated back to 1904, when the zamindar
had first become municipal chairman, between 1917 and 1920 the
factions clothed their local animosities in the garb of provincial politics.
The opposition, led in the council by Father Marie Louise, used Home
Rule propaganda against the zamindar to demand more seats open to
election and less filled by the chairman's nomination. Doddapanaya-
55
See note signed Panagal dated 16 May 1923 in G.O. 1131 (L.S.G.) dated
17 May 1923. T.N.A.

302
 Home Rule League, Justice Party and Congress

kanur replied by joining the S.I.L.F. and relying on favourable

decisions from Secretariat officials to keep the council closed.56
In Negapatam, Tanjore district, by contrast, the arrival of the
provincial conflict saw an Arabic Muslim leading in the formation of
associations to support the Dravidian peoples. He was Ahmed Thambi
Maracair who, for many years, had been involved in a fight for local
power with K. S. Venkatarama Iyer. Ahmed Thambi, a merchant and
landowner, relied for his support in municipal politics on the prominent
Maracair Muslim trading families of the Nagore suburb and on
various local landlords. K. S. Venkatarama Iyer, a Brahman lawyer
and mirasidar, was connected to Hindu 'Pillaima' (Vellala) merchants,
who were the natural competitors of the Maracairs, and exercised a
considerable independent influence through his membership of the
notoriously corrupt Negapatam temple committee. Venkatarama Iyer
was the first to bring provincial politics to the locality in 1917. Ahmed
Thambi, then municipal chairman, was acutely embarrassed by the
indentured labour issue. His constituency was split between merchants
and shipowners, who wished labour migration to expand unchecked,
and landowners who saw their cheap labour supply beginning to
dwindle. Venkatarama Iyer, joining the Home Rule campaign for the
abolition of indentured labour, led a local agitation against him, stole
his landlord support and took the municipal chair. Ahmed Thambi
replied, rationally enough, by consolidating his merchant support
and winning over the Pillaimas to his cause. But to make this class
confrontation acceptable in provincial politics and to win for the
merchants a measure of Secretariat support, Ahmed Thambi had to
shift his rhetoric to the caste vocabulary. Thus the merchant party
became a non-Brahman party and Ahmed Thambi established a
Dravidian Association among the Pillaimas. This move also had the
virtue of providing a further barrier to Venkatarama Iyer should he
ever try to win back the Pillaimas.57
In Erode town, Coimbatore district, perhaps most strangely of all,
the importation of provincial political concepts put two brothers on
opposite sides of the communal fence. They were the Muslims Sheikh
Dawood and Kadir Sahib, who were the premier magnates of the
locality, owning vast lands and urban properties and running huge
56
Collector of Madura to Secretary, Local and Municipal Department, 11 February
1919 m G.O. 374 (L and M, M) dated 6 March 1919; also G.O. 1383 (L and M, M)
dated 18 September 1919. T.N.A.
57
See Madras Mail, 20 March and 28 April 1917; Note signed Collector of Tanjore
dated 21 August 1918 m G.O. 1150 (L and M, L) dated 7 September 1918. T.N.A.;
Justice 10 January 1921.

303
 The emergence of provincial politics

trading empires. Their hatred of each other had long split the town,
and the two municipal factions, led respectively by the lawyer, T.
Srinivasa Mudaliar, and the wholesale merchant, E. V. Ramaswami
Naicker, were in many ways extensions of their feud. During the War,
Ramaswami Naicker's scarcely disguised manipulation of elections
led him into conflict with the Collector and the Local and Municipal
Department. He was dismissed from the council chairmanship and
formally banned from holding local office. Already possessing close
familial contacts with the leading Home Rule lawyers of Salem (C.
Vijayaraghavachari was a trustee of his father's school) and now hav-
ing personal cause for taking an agitational stance, he was not slow to
move to the Home Rule campaign and to take Sheikh Dawood with
him. T. Srinivasa Mudaliar and Sheikh Kadir naturally looked to the
favour of the Collector and Secretariat and set up a local branch of
the S.I.L.F.58 In Periyakulam, Negapatam and Erode, as indeed in
every other locality, the great provincial debate flowed along channels
cut for quite another purpose.
The political paradoxes created by this system of local factionalism
appearing as ideology were made even greater when Montagu's initia-
tives raised the political stakes. Discussions at the capital then became
not only about who was to do well under the old constitution but also
who was to make the new one. The nature of representation before
commissions and the Government of India forced politicians to form
associations of communal interests, for it was only these which
obtained a hearing. Consequently, between 1917 and 1919, frantic
efforts were made all over the province, both by provincial leaders
and their magnate supporters, to found associations based on class,
caste and religion. Yet the real, factional basis of local political organ-
isation broke through, for not only did all 'natural' associations - such
as those of Muslims, Christians and specific castes - collapse as their
members were pulled different ways by their more important cross-
communal ties,59 but both non-Brahman and Congress leaderships
58
G.O. 1250 (L and M , M) dated 19 July 1916. T.N.A.; Madras Mail 17 September
1917; G.O. 817 (L and M , M) dated 22 May 1919; G.O. 1114 (L and M, M) dated
25 July 1919; G.O. 1482 (L.S.G.) dated 30 July 1 9 2 1 ; G.O. 1717 (L.S.G.)
dated 4 August 1923. T.N.A.
59
For Muslims, see 'Proceedings of the Council of the Madras Muslim League, 30
November 1917' in 'Proceedings of the Madras Muslim League, 19 March 1917
to 21 February 1921,' Madras Mahajana Sabha Papers. N.M.M. For Indian
Christians, see Hindu 28 and 29 January 1918; for Indian Catholics, sec Madras
Mail 20 August and 10 September 1917, and Hindu 10 October and 6 November
1917. Among caste associations, the Visvabrahmanas tried to send five different
deputations to see Montagu, which recommended no less than three different

304
 Home Rule League, Justice Party and Congress

tried to organise class folio wings out of the same class groups — notably
village officers, factory workers and urban tradesmen.60
By the summer of 1918, the observer of Madras would be forced to
notice the deep divisions in political society yet, unless he were
extremely careful, he would be bemused by the shapes and the causes
of the conflicts. If we take the S.I.P.A. and Mylapore groups to be
leaderships, each was connected to a series of similar supports. For
example, Mylapore represented firstly, clientage ties, many of them
personal, between Mylapore men and magnates, which had been
established by conditions before 1916; secondly, various local institu-
tional factions which sought Mylapore's aid either for direct gain or
because opponents had gone to the S.I.P.A.; thirdly, various disturbed
interests, such as village officers, Tanjore mirasidars, factory hands
and unsuccessful students who were attracted by the anti-administra-
tion stance of the Home Rule League and by the hope that their own
grievances would be carried by the agitation; and fourthly, a variety
of sectarian organisations. The S.I.P.A. leadership was linked firstly,
to magnates who had been driven into opposition to Mylapore by
events prior to 1916; secondly, to various local factions seeking
rewards from the centre or opposing Mylapore-connected enemies;
thirdly, to various disturbed elements who regarded the condemna-
tion of Mylapore Brahmans as a condemnation of the administration;
and fourthly, to a further variety of sectarian associations, often of
the same sect as Mylapore's own. Both 'parties' thus mobilised an
extraordinary collection of forces, some of which were 'in' power in
local and provincial institutions, some of which were 'out' of power
and all of which hoped to be 'in' power in the promised but unknown
future political system. Particular divisions and alliances were the
result of personal connection, factional opposition and political
guesswork. It is not surprising that the political combinations of
these years were a trifle confusing.
Not the least casualties of the developments, which had produced
two structurally identical 'parties' in Madras politics, were the initial
ideological cleavages between the Home Rule League and its non-
and mutually exclusive programmes of reform. Addresses presented in India to His
Excellency the Viceroy and the Right Honourable Secretary of State, p. 88. P.P. 1918.
vol. XVIII.
60
For village officers, see Hindu 29 November 1919, 13 May and 18 June 1920;
Swadesamitran 9 August 1920. R.N.P. For factory and urban workers, see Shiva
Rao, Industrial Worker in India, pp. 15-16; Hindu 27 and 29 January 1920; G.O.
157 (Public) dated 8 March 1920. T.N.A.; Hindu 16, 19, 23, 27, 28, 29 and 30
August 1919; Hindu 20 February, 26 March and 7 December 1920.

305
 The emergence of provincial politics

Brahman opponents. The S.I.P.A. banner had been raised with the
help of the civilians to campaign for no change in the constitutional
system because of the fear of 'Brahman' domination. It had thus
attracted support from groups which were already suffering from
Mylapore's influence. However, as it became a more positive political
force, its leaders were provided with the opportunity of picking up
much more than the support of the weaker political elements. Indeed,
unless they did so, their movement would count for very little in real
terms of political control. Yet, clearly, they were hampered by a
slogan which identified them to local magnates as a group which stood
against Indians being given more power: few politicians would expect to
gain popularity by offering the people less bread than their opponents.
By mid-1917, several S.I.P.A. leaders were moving steadily away from
'no-change' to an acceptance of some reform providing non-Brahmans
were protected by electoral safeguards.61 But by this time, the Con-
gress itself had accepted the principles of non-Brahman protection in
order to weaken the appeal of the S.I.P.A.62 Thus both parties stood
for some reform and for the reservation of non-Brahman seats in the
Legislative Council. Certainly, there remained differences over the
amount of reform possible and the size and nature of the reservations
but the issues of principle between them were dead. Not, of course,
that there was any sign or any hope of a compromise being reached
between the S.I.P.A. and Home Rule League. Indeed, as the issues
between them disappeared, the acrimony increased.63 The S.I.P.A.
group had come to prominence and existed as a united force only in
opposition to Mylapore. It had received vital support from British
civilians only because it served to counter the activities of the Home
Rule League. Quite simply, the S.I.P.A. could not consider a deal
with Mylapore and the Home Rule League/Congress without com-
promising itself out of existence and losing a crucial source of its
support.64
61
For example, in April 1917, P. Thyagaraja Chetty made a thundering speech in
which he demanded that control of the Indian economy should be placed entirely
in Indian hands. Hindu 2 April 1917. At the Coimbatore non-Brahman conference
in August 1917 several proposals for constitutional reform were debated and
passed. Kistnapatrika 25 August 1917; Andhrapatrika 7 September iqiy.R.N.P.
62
Hindu 3 0 October 1917. In September 1917, non-Brahman Congressmen s e t u p
the Madras Presidency Association under P. Kesava Pillai. This demanded com-
munal safeguards for non-Brahmans and received Mylapore's full approval.
63
T h e S.I.P.A.—S.I.L.F. employed 'goondas' t o break u p the inaugural meeting
of the Madras Presidency Association. Madras Mail 17 September 1917.
64
In November 1 9 1 7 , the S.I.L.F. called an all-party non-Brahman conference
at Bezwada i n what appeared to be an attempt t o come to agreement with the

306
 Home Rule League, Justice Party and Congress

Towards the end of 1918, Montagu's report initiated a period of

further change in Madras politics: for the first time, politicians had
concrete proposals on which to negotiate. Given the extraordinarily
complicated structure of the non-Brahman and Home Rule move-
ments, their respective leaderships were not under any particular
pressure or in receipt of any specific mandate to guide their behaviour.
They could do what they liked so long as the constitution which was
created put them — as a small leadership group — into positions of
central power: they were commanded only by the opportunities for
the advancement of factions at the top of the provincial political
system.
The attitude of the S.I.P.A. - or Justice Party as we might now more
conveniently call it - leadership during the negotiations was exactly
predictable. Under no circumstances could it reach agreement with
Mylapore or the 'new' Congress leadership which had arisen alongside
Mylapore in 1919. Its policy was one of continued opposition to every-
thing the various Home Rule League and Congress elements demanded
and of using its connections with British civilians (and their own
Parliamentary contacts) to maintain its credibility.65 It followed in
reverse the political contortions which the Home Rule League and
Congress were practising and, in the short space of eighteen months,
managed to accept and reject the principles of Montagu's report
twice.66 On the central issue of communal representation, the decision
of London to rule all question of separate electorates out of considera-
tion, gave it a weapon perfectly tempered to its needs. Prior to this
decision, it had vacillated endlessly over the terms of communal safe-
guards.67 Now that it was impossible for the Congress to threaten it
by agreeing to its request, it demanded separate electorates as a pre-
requisite for its acceptance of the reforms. Lord Willingdon, who had
undertaken the improbable task of obtaining an agreed solution to the
communal problem, was nearly driven to distraction by the Justice
M.P.A. However, it turned out to be n o more than a clumsy ploy by the
S.I.L.F. to destroy its Congress opposition. M.P.A. m e n were excluded from
all offices at the conference and were allowed to negotiate only after they had
resigned their Congress memberships. Hindu 14 November 1917. There were n o
other attempts at compromise before 1920.
65
T . M. Nair tried to link u p with Lord Sydenham who led the opposition to the
Montagu-Chelmsford reforms in the House of Lords. Reforms (General, Fran-
chise) A May 1920, Nos 31-43. N.A.L
66
See Hindu 21 April 1919.
67
T h e S.I.L.F. had n o official policy o n the amount o f communal safeguards it re-
quired. Various members, at various times, suggested proposals which would

307
 The emergence of provincial politics

Party leaders' determination to cling to a position which had been

made untenable before the negotiations began. 68 During the course of
two conferences called by Willingdon, the Justice Party men turned
down offers from 'Brahman representatives' which would have given
non-Brahmans a guarantee of 60 per cent of the seats in the Legislative
Council and which were much more generous than the terms of settle-
ment forced upon them by Lord Meston, the independent arbitrator
of the final award.69 Their real political interests lay not in any absurd
mathematical division of seats in a hypothetical Council but in avoid-
ing coming to any agreement at all. Meston's imposed award thus
played into their hands, for they were still able to refuse to accept it
and to continue campaigning on the communal issue as a separate
leadership group.
The problems faced by the Justice Party leaders, however, were as
nothing to those of the Home Rule leaders. Mylapore had joined the
agitation to press the British for a further devolution of power. But
many of Mylapore's enemies in provincial politics also had joined in
order to break back into the Congress and provincial affairs, from
which they had been excluded by Krishnaswami Iyer. During 1917,
Mrs Besant had used her amalgam of opposites to work herself into a
position of dominance in the Congress, and the Home Rule League
itself had fallen into decay.70 It had served its purpose which was to
take her to the leadership of national politics. But the cracks between
the segments of her support had never been satisfactorily cemented
over. Montagu's report and the serious business of negotiating the
reforms provided the occasion for the old conflict to reassert itself,
dressed, no doubt, in the new language of representative democracy,
but making all the same points as before.
Mylapore's position on the proposed reforms was quite straight-
forward. It wanted as much reform as it could conceivably obtain
and it would push the British to go further than their initial proposals,
have allowed Brahmans to stand in anything between 3 a n d 15 of the 60 proposed
seats. Some members proposed communal electorates, others were not opposed t o
simply reserving seats.
68
' T h e n o n - B r a h m a n s are certainly t h e dirtiest a n d meanest devils I've ever come
across, a n d again I say w h y t h e devil they should be allowed this most unfair
advantage which they are most unfairly exploiting I can't think.' Willingdon to
M o n t a g u , 5 F e b r u a r y 1920. Willingdon Papers. l.O.L.
69
Reforms (Franchise) 'Deposit' F e b r u a r y 1920, No. 4 ; Reforms (General, Franchise)
A May 1920, Nos 31-43. N.A.I.
70
By December 1917 Mrs Besant had become Congress president and C. P. Rama-
swami Iyer Secretary-General of the All India Congress.

308
 Home Rule League, Justice Party and Congress

whatever they might be. But it would accept the reforms as they came
for, in its mind and without the experience of post-1920 politics before
it, any extension of power to Indians beyond the Morley-Minto con-
stitution was likely to redound to its own benefit. The leading men of
Mylapore, C. P. Ramaswami Iyer, G. A. Natesan, L. A. Govindara-
ghava Iyer and their connections, therefore, wished to use the Con-
gress organisation as a responsible negotiating body. In so far as Mrs
Besant, though conceptually independent, was most reliant on My-
lapore for advice and support, this was her position as well. The
enemies of Besant and Mylapore were thus presented with a golden
opportunity of ousting them from Congress leadership and taking
over for themselves the vital organ of provincial agitation by making
the Congress too wild and too violent a body for it to accord with the
purposes of responsible negotiation.71
The leadership of the group which planned the putsch of Mylapore
passed naturally into the hands of Kasturi Ranga Iyengar, who had
been fighting Mylapore's influence since the 1890s. As one of the
leading press barons of his day and as a Madras city figure capable
of forging widespread mofussil contacts, he was crucial to the success-
ful organisation of the manoeuvre. It is clear that he, and his lieuten-
ants S. Satyamurthi, T. V. Gopalaswami Mudaliar and V. O. Chidam-
baram Pillai (the Tuticorin agitator of 1908 who now lived in Madras
city), perceived the way that the Mylaporeans would go during the
reform negotiations rather before C. P. Ramaswami Iyer and Mrs
Besant had made a move. By the spring of 1918, they had tightened
their connections with those 'frustrated' politicians, mostly from the
mofussil, who were outside all channels of Secretariat influence and
provincial political importance. They rebuilt an exact replica of the
1906—8 Extremist alliance which had done battle with Krishnaswami
Iyer. Indeed most of the men of 1918 were the same as those of 1906—
8. Beside Kasturi Ranga Iyengar stood C. Vijaraghavachari, his
protege C. Rajagopalachari, and N. K. Ramaswami Iyer; and the
newer leaders — P. Varadarajulu Naidu, E. V. Ramaswami Naicker,
71
The probability of tension between Mylapore and the wilder student elements
which carried on its agitation was apparent as early as the summer of 1917 when
calls had been made for a passive resistance movement to secure Mrs Besant's
release from internment. C. P. Ramaswami Iyer and the Mylapore leaders had been
extremely cautious over this proposal and had agreed to it only after they had
arranged the wording of the resolution so that it did not commit them to any
practical action. Hindu 15, 24 and 27 August 1917. For an interesting discussion
of student politics and its relationship to national affairs at this time see, S.
Ramanathan, Gandhi and the Youth (Bombay, 1947).

309
 The emergence of provincial politics

S. Satyamurthi and T. S. S. Rajan - had all been inducted into the

political system through personal connection with the old.72 For
Kasturi Ranga Iyengar's immediate purposes, the alliance looked
extremely promising. The 'frustrated' would be unwilling to end the
agitation which had brought them back into provincial prominence
and to settle for anything like the amount of constitutional reform
which would suit Mylapore. As Mrs Besant herself had brought
students and other disturbed elements into the Congress in order to
heat her own campaign, the material which her opponents required to
destroy her was ready to hand.
The first open breach in the Congress front came at the Con-
jeeveram Provincial Conference in April 1918. Mrs Besant and
Mylapore were nearly defeated on a challenge offered by C. Raja-
gopalachari to their resolution promising aid for the war effort
in return for constitutional reform.73 A few weeks later the conflict
was extended to the Madras P. C. C. The Besantine-Mylapore
group noticed that its grip on the organisation was being weakened
by the steady induction of members of the mofussil 'frustrated'. It
responded by recruiting to the P.C.C. nearly all the residents of the
Theosophical Society's grounds at Adyar, including European
women and children.74 To prevent this membership power from
being converted into executive posts at the next elections, Kasturi
Ranga Iyengar began his own recruiting drive and enrolled en
masse supporters of himself and his allies.75 The battle returned to
the open stage at the Special Provincial Conference called in August
to discuss the Montagu—Chelmsford reform proposals. Satyamurthi
and N. K. Ramaswami Iyer organised an attack on the Besantine-
Mylapore-dominated Subjects Committee. They attempted to replace
the formal resolution rejecting the reform proposals with one which,
though substantially the same, was worded to give the British the
impression of much greater defiance.76 The Besantine-Mylapore
72
P. Varadarajulu Naidu, E. V. Ramaswami Naicker and T . S. S. Rajan all h a d
close political connections with C. Rajagopalachari's Salem base. S. Satyamurthi
worked for Kasturi Ranga Iyengar as a journalist o n the Hindu.
73
Hindu 9, 1 1 , 1 3 , 1 4 , 1 5 and 16 M a y 1918; New India 20 M a y 1918; H o m e Political
'Deposit' August 1918, N o . 28. N.A.I.
74
'Proceedings o f the Executive Committee o f t h e Madras Provincial Congress
Committee, 2 3 June 1918'. Proceedings o f the Madras Provincial Congress C o m -
mittee 1918. Madras Mahajana Sabha Papers. N.M.M.L.
75
Ibid., 5 July 1918.
76
T h e formal resolution rejected the proposals as a basis for negotiation until they

310
 Home Rule League, Justice Party and Congress

leaders clumsily tried to prevent the expression of views opposed to

their own, and the conference ended in uproar.77
After this debacle, both sides went back into the committee room
to build their followings for the forthcoming Special Congress session
on the Montagu-Chelmsford report to be held at Bombay.78 The
decline of Mrs Besant in All India politics during 1918 and 1919 is
well known. She, and the groups from other provinces, which were
similar in ambition to Mylapore, found it impossible to control the
Congress and turn it from a wild agitational organisation into one
capable of rational discussion with the British. In even attempting
this, she became the object of as much hate as she had once been of
adulation. Her Mylapore-type supporters only risked their own careers
by remaining within an institution which was in the hands of their
enemies and committed to policies they could not accept. They with-
drew from the Congress and formed a 'Moderate Conference' away
from the turmoil of agitation.79 Mrs Besant, herself, returned to
Madras and attempted to revive her Home Rule League only to find
that it too was in the hands of Kasturi Ranga Iyengar.80 She formed a
separate National Home Rule League but it never had a serious
following.81 By the middle of 1919 she was no longer of any con-
sequence, and the institutions of the Madras Congress were firmly
under Kasturi Ranga Iyengar's control. Indeed, by 1920, he had even
forced Mylapore and the Besantines out of the relatively harmless
Madras Mahajana Sabha.82
The problems facing the new leaders of the Madras Congress, how-
were brought into line with t h e Congress—League scheme; t h e a m e n d m e n t r e -
jected t h e proposals because they were not t h e same as t h e C o n g r e s s - L e a g u e
scheme.
77
T h e Subjects Committee insisted that the resolution should be p u t from the chair,
and not debated at all. Hindu 5, 9, 12, 13, 16, 18, 2 0 , 2 3 , 29 a n d 30 July, a n d 1,
3 a n d 5 August 1918.
78
Once again this took t h e form of t h e t w o sides electing as m a n y supporters as
they could find. 'Proceedings of t h e General Body of t h e M a d r a s Provincial
Congress Committee, 20 August 1918'. Proceedings of t h e M a d r a s Provincial
Congress Committee, 1918. Madras Mahajana Sabha Papers. N.M.M.L.
79
In Madras the main organiser of the Moderate Conference was Sir P. S. Sivaswami
Iyer who, although never actually a member of the Home Rule League, was known
to support Home Rule objects. Hindu 27 and 29 July 1918. Within a year most of
the leading men of Mylapore had defected from the Congress to the Moderate
Conference.'
80
The National Home Rule League. How Founded and Why (Adyar, 1919), pp. 1—3.
81
Ibid.
82
Hindu 16 and 24 August 1920.
 The emergence of provincial politics

ever, were very similar to those facing the leaders who had just gone.
The organisation had fed on years of agitation and had grown fat. It
was a reasonable presupposition that, in the unknown era of politics
about to commence, it was going to have a considerable influence. But
it would have to moderate its tone and prepare itself for constitutional
action. Yet in order to gain control of it, Kasturi Ranga Iyengar and
his men had had to foster the more extreme elements in political life
and to give a greater share of power to the mofussil frustrated, partic-
ularly C. Rajagopalachari,83 who had nothing to gain by a return to
constitutional activity. The tensions can be seen clearly in the response
to Gandhi's call for a hartal and demonstration over the Rowlatt Acts
in April 1919. C. Rajagopalachari and his extreme associates, using
trades union and cultural as well as political linkages, which had
been developed during the period of united Home Rule League-
Congress agitation,84 organised a show of force in Madras city on a
scale never seen before. The police estimated that about one hundred
thousand people attended lectures and parades on the beach.85 Yet
the Congress leadership - the Kasturi Ranga Iyengar group - so far
from being delighted by Rajagopalachari's achievement was absolutely
terrified.86 For the next six months it made sure that scarcely a dog
was heard to bark in the city and Congress public activity virtually
dried up.
To the extremist group, now headed by C. Rajagopalachari, it
became increasingly clear that the new Congress leadership was no
great improvement on the old. During 1919 and 1920, Kasturi Ranga
Iyengar demonstrated that he intended to use his control of the
Congress organisation and the contacts he possessed with higher
administrative officers - forged both through personal connections
and through his role in the Congress - to exercise as tight and as closed
a leadership as had Mylapore. The first overt manifestation of this
came during the summer of 1919, when the Tamil newspaper Desa-
bhaktan, of which Rajagopalachari and T. S. S. Rajan were directors,
began to rival the circulation of the Swadesamitrany the Tamil organ
83
By 1919, C. Rajagopalachari was a m e m b e r of t h e executive committee of
t h e M a d r a s P.C.C. a n d vice-president of t h e M a d r a s Mahajana Sabha. H e h a d
sold his possessions in Salem and set u p house in t h e capital.
84
During 1919, C. Rajagopalachari and other mofussil politicians (particularly the
Salem barrister T. Audinarayana Chetty) fostered a series of labour organisations
in Madras city. These provided them with a base for agitational politics in the
capital. Hindu 6 and 7 December 1918 and 14 February 1920.
85
G.O. 318 (Public) dated 2 June 1919. T.N.A.
86
See T. V. Kalyanasundram Mudaliar, Valkkai Kurippugal (Madras, 1969),
pp. 305—6 (Tamil).

312
 Home Rule League, Justice Party and Congress

of the Hindu complex. According to the Commissioner of the Madras

city police, Kasturi Ranga Iyengar persuaded the Attorney-General,
S. Srinivasa Iyengar (who had broken with Mylapore on personal
grounds),87 to raise Desabhaktan's security deposit to a level which its
directors could scarcely hope to meet. The newspaper was threatened
with closure and was able to survive only through the intervention of
C. P. Ramaswami Iyer and Mrs Besant who provided it with funds.
While it may seem ideologically paradoxical for 'moderates' to support
an 'extremist' journal, the move makes excellent sense if viewed from
the standpoint of the political factions involved in the Congress fight.88
In the spring and summer of 1920, Kasturi Ranga Iyengar's activi-
ties announced that there would be no place for Rajagopalachari and
his followers once Congress moved into the new Montagu-Chelmsford
councils. At the Tinnevelly Provincial Conference in June, the Con-
gress began its advance towards the elections and set up a committee
to co-ordinate its campaign. It was essential now for the Congress
leadership to work closely with important local magnates if it were to
translate its agitational power into votes. The Raja of Ramnad, the
zamindar of Kumaramangalam and the powerful Tinnevelly temple
and local board politician N. A. V. Somasundram Pillai were drafted
onto the committee. The only members of the Congress executive to
be included, however, were Kasturi Ranga Iyengar himself and S.
Srinivasa Iyengar, who had now resigned his post as Attorney-General
and openly joined Kasturi Ranga Iyengar. None of Rajagopalachari's
associates was considered initially for a place and it was only following
a threat to refuse to recognise the committee that C. Vijayaraghavachari
was begrudgingly brought in.89 Later, at the beginning of August,
the massive Tilak Memorial Fund was set up and again the Rajagopa-
lachari group was squeezed out. The best any of its number could
obtain was a minor office shared with the manager of the Swadesa-
mitran press.90
87
S. Srinivasa Iyengar was a member o f the Vembakkam clan and a leading lawyer
in t h e capital. H i s alienation from Mylapore dated from Krishnaswami Iyer's
death, w h e n h e and L. A. Govindaraghava Iyer had fought over the University
Legislative Council seat vacated by Sivaswami Iyer w h o m o v e d o n t o the E x e c u -
tive Council to take u p Krishnaswami Iyer's posts. Most o f Mylapore preferred
Govindaraghava Iyer w h o eventually w o n the election. Hindu 5 February 1914.
See also V. S. Srinivasa Sastri to P. S. Sivaswami Iyer, 7 August 1916; S. Srinivasa
Iyengar t o P. S. Sivaswami Iyer, 18 February 1914. P. S. Sivaswami Iyer Papers.
N.A.I.
88
Report 4 2 2 - D from Commissioner 6 f Police, Madras City, dated 4 March 1 9 1 9
in G.O. 230 (Public) dated 2 9 April 1919. T.N.A.
89 90
Hindu 2 3 and 25 June 1920. Hindu 11 August 1920.

313
 The emergence of provincial politics

Nonetheless, before the middle of August 1920, Kasturi Ranga

Iyengar's opponents in the Congress had no coherent plans for oppos-
ing him, least of all on the non-co-operation issue of boycotting the
Legislative Councils. The position of Rajagopalachari and his 'party',
no less than that of the 'old' and 'new' Congress leaderships, was
defined by their proximity to the sources of power, which obviously
were going to lie in the Legislative Council. Although they were an
'out' faction, they were still a faction and had no inclination to abandon
the political game completely. They joined with Kasturi Ranga
Iyengar in calling on Gandhi to wait for a Congress session before
declaring non-co-operation.91
Their decision to back Gandhi followed only after considerable
consultations with the Mahatma while he was campaigning in Madras
in August 1920. Gandhi's tour coincided with a debate in the Madras
P.C.C. on the question of non-co-operation. In the first session of
discussions, Rajagopalachari was equivocal on the possibilities of the
tactic. Gandhi arrived in the city just before the second session and,
by the third, Rajagopalachari was a pronounced advocate of non-co-
operation.92 The Kasturi Ranga Iyengar group, of course, was bitterly
opposed to the notion of Council boycott which would defeat the
purpose of the previous two years' exercise.93 Rajagopalachari's con-
version, however, was not merely a ploy to embarrass his provincial
opponents. Gandhi's movement made sense to his reasonable ambi-
tions. He, and his associates, had risen to prominence only through the
agitation of the 1916 to 1920 period. It was clear from Kasturi Ranga
Iyengar's behaviour that, once the agitation died, they would sink
back into obscurity. Gandhi, by promising to place the weight of the
All-India Congress behind continued agitation, offered Rajagopa-
lachari the one chance he had left of remaining a politician of signific-
ance. Until he had that promise, however, Council boycott itself
meant nothing but consignment to an even greater obscurity.94
91
Letter of P. Varadarajulu Naidu in Hindu 20 July 1920.
92
Hindu 5, 6 and 25 August 1920.
93
During the many conferences and discussions o n non-co-operation which took
place in 1920, the Madras leadership group had managed to maintain its am-
biguous posture. It agreed to some kind of campaign but succeeded in preventing
all attempts to define what that campaign should be. See S. Satyamurthi to
Secretary, All India Congress Committee, 26 August 1920. All India Congress
Committee Papers, File 13(2) o f 1920. N.M.M.L. T h e only authoritative state-
ment made o n the matter was that 'the programme in no case shall include the
boycotting o f elections to the Reformed Councils'. Hindu 16 August 1920.
94
T h i s statement in n o way denies Rajagopalachari's 'spiritual' conversion to the
Gandhian cause, which was manifest by 1924. It is perfectly compatible with a

314
 Home Rule League, Justice Party and Congress

At the Calcutta Congress, Rajagopalachari led to Gandhi's cause

a collection of students and Muslims disturbed by the Khilafat agita-
tion.95 Gandhi's victory destroyed Kasturi Ranga Iyengar's leadership
of the Madras Congress. His important magnate supporters deserted
the Congress for the Councils and the disturbed agitational groups
naturally looked to Rajagopalachari rather than himself. He was left
the nominal head of an organisation whose policies he disavowed and
whose remaining supporters were his enemies. His involvement in
non-co-operation was half-hearted to say the least. Logic was restored
to the political situation after the Nagpur Congress when, using the
new Congress constitution, Rajagopalachari was able to remove
Kasturi Ranga Iyengar from the executive of the Tamil Nad C.C. and
to switch Congress headquarters in the South from Madras city to
Trichinopoly.96 The stranglehold on Congress affairs, exercised for
so long by a handful of men in the capital, had at last been broken and
the frustrated mofussil politicians who had spent years hammering at
the dominance of Madras city had finally obtained their own provincial
institution.
The provincial leadership groups which we have been examining
existed logically apart from any local support on which they might
hope to draw. None can be seen as parties possessing disciplined
followings or structures through which followers could inform leaders
of their requirements. Each was rather a group of associated individ-
uals jockeying for position in relation to and for influence over the
new and unknown constitution. With the exception of C. Rajagopala-
chari, whose independence was asserted only at the expense of the new
constitution, all three leadership groups appealed to exactly the same
social material for support. As we saw, they tried to form organisa-
tions out of the same caste, class and religious interests and, above all,
were concerned to attract the sympathy of various local magnates. The
reasons for the battles at the centre of provincial politics are clear,
but what did the magnates themselves make of these faction fights and
on what grounds did they give their support?
Up to the time of Meston's communal award, there was little ideo-
logical difference between the three leadership groups. All three ac-
cepted the principle of some form of communal safeguard for the
non-Brahmans, and all three, tacitly at least, had decided to work
'conversion' hypothesis and merely points out that, in the circumstances,
spiritual and material factors both worked towards the same end.
95
Hindu 2 and 6 September 1920.
96
All India Congress Committee Papers, File 1 of 1921. N.M.M.L. Home
Political A File 2 4 4 of 1921 N.A.I.

315
 The emergence of provincial politics

through the new constitution. It would be difficult, therefore, to see any

of the leaderships winning support because of their principles. Of
course, each group could count on particular magnates, tied by personal
connections and by the connections of opponents with other groups. But
that would only explain the behaviour of a handful of men. For most
magnates of local power, the relevant question was which of the three
groups was most likely to be successful and, given that the nature of
political life in the new Councils was uncertain, the answer was
impossible to calculate. A great many, therefore, kept their lines of
contact open with several leaderships at the same time in order to play
the odds. The ambivalent behaviour of A. Subbarayalu Reddiar and
T. A. Ramalingam Chetty, which we noted earlier, set a general
pattern. In Ellore, as early as 1917, the Mothey family had demon-
strated overt Congress connections and had attended a non-Brahman
conference with their lawyer K. V. Reddi Naidu;97 later they tight-
ened their Justice Party connections as Reddi Naidu climbed into the
Justice leadership group but, through their Arya Vysya caste con-
ference, they remained in touch also with the Congress.98 In Tanjore,
by 1918, two of the largest landowners, V. A. Vandayar of Pundi and
T. Somasundram Mudaliar, could be found in both S.I.P.A. and Con-
gress organisations.99 In Ganjam, the local politician A. P. Patro was
an organiser for both the Justice Party and the Mylapore-inspired
Moderate Conference;100 in Cocanada, K. Suryanarayanamurthi
Naidu, a millionaire shipowner and landlord who had great influence
in the town, held office in the local Justice Party and the Congress
District Association;101 in Salem, the banker S. Ellapa Chetty, a
member of the Justice Party, presided at the local Congress meeting
to pick candidates for the election.102 As, in 1917, a separate Andhra
P.C.C. had been created, the expulsion of Mylapore from the Madras
P.C.C. did not necessarily affect many of Mylapore's closest contacts
in the Andhra mofussil. Such men as N. Subba Rao, M. Ramachandra
97
Subba R a o , Life and Times of Sir K. V. Reddi Naidu, p. 31. Hindu 30 November
1917.
98
T h e Komati merchants o f the Andhra delta towns, w h o ran the caste sabha,
were strong supporters o f t h e Congress and its local leaders, Konda Venkatap-
payya and A. Kaleswara Rao. Hindu 27 February 1920.
99
Hindu 19 March, 2 , 2 2 and 2 3 April 1918.
100
Hindu 16 October 1918.
101
In April 1 9 2 0 , the District Association gave its support to several candidates
including K. Suryanarayanamurthi Naidu o f the Justice Party and D . V. Prakasa
Rao, a Congressman and later chief non-co-operator in the town. Hindu 23 April
1920.
102
Hindu 1 June 1920.

316
 Home Rule League, Justice Party and Congress

Rao and the Nellore Mylapore contact, A. S. Krishna Rao, found it

possible to remain members both of Mylapore's Moderate Conference
and of the Andhra Congress.103 Indeed, N. Subba Rao andM. Rama-
chandra Rao were members of the committee which selected Congress
candidates for the election.104 So desperately did some local figures
feel the need for a multiplicity of connections at the centre that, at the
Tinnevelly Provincial Congress Conference of 1920, S. T. Shan-
mugham Pillai and T. N. Sivagnana Hllai, the Tinnevelly politicians
who had been most responsible for the spread of anti-Brahman and
anti-Congress propaganda in the district, showed considerable interest
in the affairs of the session.105
The contortions of the local magnates make it clear that they, the
most important political forces in Madras, saw the provincial political
conflict in terms of men not measures and patronage not principle.
As the elections approached, however, events in the leadership groups
made worse the near total chaos to which magnatefloor-crossinghad
reduced provincial politics. One of the leadership groups, the 'new'
Congress, dropped out of contention and left a number of magnates,
who had close ties with it, floating uncertainly on the surface of
political life. In Madras, no man of political substance who had a
chance of election - not even S. Srinivasa Iyengar106 - was prepared
to entertain Gandhi's overtures and boycott the Councils. Further,
the Justice Party, responding to the conservatism of Meston's Award,
launched a desperate crusade which made nonsense of its avowed
political position. It elaborated a campaign, based on Meston's reserva-
tion of only 28 seats, to secure the election of all non-Brahmans. It
offered its ticket to any non-Brahman who stood for election, no
matter what his political hue, his connections with Brahmans and his
intentions in the new legislature.107 Thus it threw its cloak over
virtually everybody in the presidency and, as it demanded no return
for its generosity, utterly obscured any differences between the majo-
rity of candidates. A great many local magnates and politicians stood
at the polls as 'non-Brahmans' without showing the least sign of break-
ing with their Brahman friends. In Madras city, for example, Lodd
103
Hindu 22 January 1920 and 19 February 1920.
104
Hindu 23 June 1920. M. Ramachandra Rao even made an election pact with A.
A Kaleswara Rao who later was a leader of non-co-operation. Hindu 26 June
1920.
105
Hindu 28 M a y a n d 2 3 J u n e 1920.
106
W h o temporarily d u m p e d t h e Congress a n d w o n t h e University Legislative
Council seat in 1920.
107
Hindu 18 M a r c h , 23 a n d 26 April a n d 13 M a y 1920.

317
 The emergence of provincial politics

Govindoss, who had poured lakhs of rupees into the Congress and
remained on the best of terms with Mylapore, accepted the support of
Justice Party polemic.108 In Salem, the banker and lawyer T. Audina-
rayana Chetty, who had been a close ally of Rajagopalachari and was
to return in 1926 as a Congress Swarajist, found that he too could
accept Thyagaraja Chetty's blank cheque.109 In Tanjore, V. A.
Vandayar of Pundi stood for the Legislative Council as a specifically
nominated Justice Party non-Brahman, while actively helping the
Mylapore Brahman Sir. P. S. Sivaswami Iyer to win an election to the
Central Legislative Assembly.110
Although the conduct of many candidates in this election may
seem, to the outside observer, to be regulated by principles of mayhem,
it was guided by an inner logic. The non-Brahman category allowed
candidates in multi-member constituencies to exclude some of the
individuals, namely Brahmans, who competed against them. Exclu-
sion was so much in everybody's mind that the Governor, Lord
Willingdon, was driven to distraction by the arrival of hundreds of
petitions — many of them from election candidates — calling for
separate caste and sub-caste constituencies, which, if acted upon,
would have obviated the necessity for any competition at all.111 Once
beyond constituency boundaries, however, candidates were not com-
mitted to maintaining their hostility to other sub-castes or even to
Brahmans. As might be expected in such a complicated polity, outside
Madras city there was little evidence of systematic campaigning on
any particular ticket and most candidates would have found it
extremely difficult to say where they stood.112
108 109
Hindu 29 April 1926. Hindu 13 May 1920.
110
M. D. Subramanyam to P. S. Sivaswami Iyer, 8 June 1920. P. S. Sivaswami Iyer
Papers. N.A.I.
111
'Oh this communal business. I am being bombarded by all sorts of sub-castes of
the non-Brahmins for special representation, and as I believe there are some
250 of these, I am not likely to satisfy many in a council of 127. You're a nice
fellow to have given me this job.' Willingdon to Montagu, 20 February 1920.
Willingdon Papers I.O.L.
112
'As for an appeal or manifesto, I am afraid it may not be quite expedient at this
stage to decide upon making one, as difficulties of all sorts are likely to be
experienced in procuring the signatures of some prominent Non-Brahmin
leaders. While they are quite willing to render every sort of help practically in
promoting our cause, it is easy to understand that some at least among them
feel it very delicate and inconvenient to be signatories in the appeal proposed to
be published.' M. D. Subramanyam to P. S. Sivaswami Iyer, 8 June 1920. P. S.
Sivaswami Iyer Papers. N.A.I. In the deltas of the Northern Circars, one of the
few places in which candidates regularly made speeches, election addresses were
concerned much more with the land revenue, Public Works Department and local
matters than with provincial issues, such as the non-Brahman movement. See,

318
 Home Rule League, Justice Party and Congress

In fact, looking at the successful Legislative Councillors from a

rather different angle, it is relevant to ask whether the enormities of
the propaganda which filled the press had any direct impact on local
political structures. Those who were elected were, to a man, important
local magnates or the representatives of magnate interests. No district
board president who stood for election failed to be returned; the vast
majority of M.L.C.S held combinations of places in rural boards,
municipalities, and temple, school and government advisory com-
mittees.113 The few who did not, possessed equally concrete political
resources in their control of various informal social institutions.114
Every man elected was of substantial political power in his locality
before the election — in most cases before the first whiff of polemic
had spread across Madras in 1916. As the turn-out to the polls was
extremely small and as the number of votes necessary to win election
in the multi-member constituencies was minute, 115 it seems more than
probable that factors of personal connection and control had been of
dominant importance.
The one constituency in which there had been a systematic party
campaign, at least by one party, was Madras city.116 Not only had this
been the headquarters of the Justice leadership group since 1916 but
it was also one of the few constituencies which was physically small
enough to be amenable to parades and polling organisation. Yet even
here, the results would lend themselves to an interpretation which
emphasised more local power than party or ideology. Two of the four
Justice Party candidates were defeated, and the two who were victorious
were P. Thyagaraja Chetty and O. Thanikachellam Chetty, both of
for example, P. Govindarow Naidu, The Legislative Council Elections (Cocanada,
1920).
113
For brief biographies, see Reforms (Franchise) B March 1921, Nos 34-99.
N.A.I.
114
Such as A. Thangavelu Naicker of North Arcot, who was one of the richest
ryotwari landlords in his district and came from a family commonly regarded as
leaders among the locally dominant Palli caste; or T. C. Srinivasa Iyengar of
Ramnad, who represented the interests of the enormously influential Nattukottai
Chetty bankers of his district. Ibid.
115
Although the district electorates varied in size from between 20,000 and
90,000 voters, candidates found that they needed only between 1000 and 5000
votes to be successful. Hindu 9 and 10 December 1920.
116
It was a very violent campaign, aimed specifically at C. P. Ramaswami Iyer.
M. Kothandavelu Mudaliar, a Sengunthair Vellala dubash who was acting
as his election agent, was beaten and stoned by a mob and Ramaswami Iyer
himself took to carrying loaded pistols to his election meetings. Hindu 6 October
1920; Prakash, Sir C. P., p. 35. For Justice Party organisation, see Hindu 5, 9
and 16 October 1920.

319
 The emergence of provincial politics

whom held a multitude of local offices and were extremely influential

in city affairs without reference to the Justice Party.117 In spite of the
fact that the electorate was predominantly non-Brahman, the two
non-Justice Party winners were Brahmans. One was C. P. Ramaswami
Iyer, whose local power in the city scarcely needs stressing. The other,
Dr U. Rama Rao, was a Canarese 'foreigner' in the city who had been
a municipal commissioner for several years — as the representative of
the important Saraswat Brahman economic interest — and was influen-
tial in education circles through his place on the University Senate.
He had taken virtually no part in the great debate of the previous four
years.
The correspondence of Sir P. S. Sivaswami Iyer suggests a better
way to explain the elections than grand but meaningless generalisa-
tions on the theme of ideology. Sivaswami Iyer stood in the Trichino-
poly—Tanjore constituency for the Central Legislative Assembly. His
campaign was based on two assumptions both of which, judging by his
huge success, proved to be correct: firstly, that certain magnates,
through their hold over a variety of local institutions, possessed 'blocs'
of votes of a cross-communal character; and secondly, that although
he possessed little direct power in the local arenas of Tanjore and
Trichinopoly, he had influence in provincial and, by this time, national
political circles and could use his position in these to forward the
interests of the controllers of local blocs. His campaign workers con-
sisted of relatives, friends, family networks put at his disposal by
friends and the staff of a school of which he was president.118 His first
move was to gain the allegiance of the Kallar zamindar of Papanad
whose landowning and banking activities were said to give him 800
votes, even on the restricted C.L.A. franchise.119 Next came the
pandarasanidhis of the maths at Thiruvadathorai and Dharmapuram,
who were of equal importance.120 Then he mobilised his Nattukottai
Chetty connections, whose banking empire was far flung and who
allowed him to use appointments at their college in Chidambaram
117
O. T h a n i k a c h e l l a m C h e t t y was a m e m b e r of t h e Corporation and Pachayappa's
Charities board; he was also a Beri Chetty, related to various leading members
of t h e city's mercantile c o m m u n i t y and a founder m e m b e r of t h e Beri C h e t t y
Sangam.
118
T . R. V e n k a t a r a m a Sastri to P. S. Sivaswami Iyer, 23 M a y and 31 August 1920;
P. S. Sivaswami Iyer to V. G u r u s w a m i Sastri, 8 and 22 M a y 1920. P. S. Sivas-
w a m i Iyer Papers N.A.I.
119
M . D. S u b r a m a n y a m to P. S. Sivaswami Iyer, 6 J u n e and 7 September 1920
P. S. Sivaswami Iyer Papers. N.A.I.
120
M . D . S u b r a m a n y a m to P. S. Sivaswami Iyer, 14 M a y and 31 August 1920.
P. S. Sivaswami Iyer Papers. N.A.I.

320
 Home Rule League, Justice Party and Congress

to reward his campaign staff.121 Finally, his agents fanned out across
the district contacting influential landlords and merchants, such as
V. A. Vandayar of Pundi and a complex of Odayar families based on
Mannargudi who controlled 200 votes.122 By the date of the election,
Sivaswami Iyer had pledged support from aristocrats, priests (in the
case of the maths, priests of the supposedly anti-Brahman Saiva
Siddhanta religious philosophy), bankers and rural-local leaders.
Among the more eccentric elements of his following were Father
Bertram (the influential principal of the Jesuit college at Trichinopoly),
the cashier of the Negapatam branch of the Bank of Madras and a
Reddi shipowner, whom he had met in Burma, who put his family
trading network at Sivaswami Iyer's disposal. The whole structure was
held together only by Sivaswami Iyer himself, whose influence had
provided in the past and would continue to provide in the future
services and rewards from higher political organisations for this
variegated mass of local groupings. 123 The campaign provides us with
a classic example of the provincially extended magnate network which
was the most usual form of political organisation in Madras through-
out our period.
Sivaswami Iyer's election, however, had been to the Central Legisla-
tive Assembly in New Delhi. His constituency covered two districts
and was, therefore, similar in size to the constituencies under the
Morley-Minto constitution. His political network, though of the same
type, was more elongated than that used by most politicians to get into
the Montagu—Chelmsford provincial Council, the constituencies of
which covered only single districts. The new provincial Councillors,
as we have seen, were men who could mobilise the political resources
of district institutions. Before 1920, relatively few of them had
appeared on the provincial political stage. Certainly some of them had
been in contact with provincial political leaderships during the years
121
M. D. Subramanyam to P. S. Sivaswami Iyer, 14 May 1920. P. S. Sivaswami
Iyer Papers. N.A.I.
122
T. R. Venkatarama Sastri to P. S. Sivaswami Iyer, 23 May 1920; G. Venkatesan
to P. S. Sivaswami Iyer, 20 November 1920; M. D. Subramanyam to P. S.
Sivaswami Iyer, 8 June 1920. P. S. Sivaswami Iyer Papers. N.A.I.
123
F o r example, Sivaswami's agents helped Papanad t o a temple committee place.
M. D. Subramanyam to P. S. Sivaswami Iyer, 8 June 1920. S. Parasurama Iyer,
a rich mirasidar, offered 100 votes for help in fighting the impending revenue
resettlement — help which Sivaswami Iyer was only too pleased to provide.
S. Parasurama Iyer to P. S. Sivaswami Iyer, 26 November 1920, and Publicity
Bureau to P. S. Sivaswami Iyer, 30 November 1920. R. P. Moorthy, a temple
politician, expected a district board seat for his help. R. P. Moorthy to P. S.
Sivaswami Iyer, 16 December 1920. P. S. Sivaswami Iyer Papers. N.A.I.

321
 The emergence of provincial politics

of agitation. But the connections between provincial and local politics

were so tenuous and the conduct of local politicians in provincial
parties was so ambivalent, that it would be impossible to describe the
new M.L.C.s as the stalwarts of any party.
Lord Willingdon's decision to appoint a Justice Party Ministry was
the result of a misinterpretation of the election returns. Underpressure
from his civilians to favour the non-Brahman cause124 and a witness to
the Justice Party's only election campaign in Madras city, he deemed
the election to have been a Justice Party victory because a majority of
the victors were non-Brahmans. In fact, when information from the
localities arrived at Fort St George, it was seen that only 15 of the 65
men elected from general constituencies had particularly close ties
with Justice Party leadership. The rest were classified vaguely as
'Liberals', 'Moderates' or 'Independents'.125 One newspaper, outraged
by Willingdon's choice, pointed out that at least 35 general con-
stituency M.L.C.S were 'progressives' having some, albeit loose, con-
nection with the Congress or Mylapore and none at all with the party
of the new Ministry.126 It took Willingdon only a few weeks to realise
that his invitation to the Justice Party to form a Ministry had been a
mistake:
The present position as far as I have seen it here now that the budget debate
is on, is that the Government is damned all round by everyone and the so-
called followers of the ministry show no allegiance at all unless it is con-
venient. This is the more significant here for I definitely put in a non-Brahman
ministry and they appear to have no hold over their followers. I have urged
them to try and organise their party. I have appointed at their request council
secretaries who shall act as whips but at present they are not very effective.127
Even the old war-cry of 'Down with the Brahman' failed to provide
a basis for the unification of a Council majority. Not only were few of
the new non-Brahman M.L.C.s antagonistic to Brahmans as a group
but, with a Council of local power, the Ministry could no longer
present the Brahman as a political threat, real or imagined. At the
124
In 1919, the Chief Secretary, Sir Lionel Davidson, had argued that only the
Justice Party represented the true interests of the non-Brahmans. Those non-
Brahmans w h o sympathised with Congress aims (the majority o f n o n -
Brahmans returned in both 1919 and 1920 elections) were serving Brahman
interests. G.O. 122 (L and M , Leg.) dated 17 October 1919. T.N.A.
125
Reforms (Franchise) B March 1921, Nos 3 4 - 9 9 . N.A.I.
126
Hindu 15 December 1920. A considerable section of the press regarded Wil-
lingdon's decision as scandalous. See Swadesamitran 18 December 1920;
Venkatesapatrika 25 December 1920. R.N.P.
127
Willingdon to Montagu, 21 March 1921. Willingdon Papers. I.O.L.

322
 Home Rule League, Justice Party and Congress

election, the provisions of the meagre Meston Award had not needed
to be invoked for, on the basis of local power, Brahmans were not in
a position to swamp a single constituency.128 Indeed, at no time under
the Montagu-Chelmsford constitution was protection required. The
communal issue was finally obvious for the hoax it had always
been.
In its early days, the operation of the Legislative Council was
characterised only by chaos. Most of the new M.L.C.s were local men
whose previous contacts with the provincial level of politics had been
mediated by one or other of the provincial leadership groups. Now
they were placed at the provincial capital itself and, initially, had very
little idea of what they were supposed to do. Gradually, however, the
turbulence subsided as the Ministry became conscious of its position
and of the necessities of its new job. Through the vigorous use of the
considerable powers of patronage at its disposal, it bought itself a
majority of the independent Councillors and thus kept itself in office.
In the words of its chief whip, 'the party lived on patronage'. 129 The
Ministry showed little interest in the antecedents of the men to whom
it offered its favours. Patronage was always limited and the Chief
Minister (between 1921 and 1926 P. Ramarayaningar, the Raja of
Panagal) calculated its distribution to produce the strongest Council
combination rather than to reward old friends. Many powerful local
magnates and politicians, who, before 1920, had been associated with
the Congress or Mylapore, became attached to the Ministry. Not the
least of these were the three magnates whom Kasturi Ranga Iyengar
had placed on the Congress election committee in June 1920. The
Raja of Ramnad was supported in the district board of Ramnad, 130
the zamindar of Kumaramangalam was made a Council secretary and
chief whip and N. A. V. Somasundram Pillai was allowed to help in
the drafting of the Ministry's temple legislation. 131 All three supported
the Ministry until its demise in 1926. Equally, Ramarayaningar found
it politic to reward several Brahmans whose local position was strong
128
Meston reached his award on the basis of seats held in the institutions of local
self-government. The evidence showed so clearly that non-Brahmans - as a
category - were not a threatened interest that he accepted C. P. Ramaswami
Iyer's plan for communal safeguards in its entirety. Reforms (Franchise)
'Deposit' February 1920, No. 12, N.A.I.
129
R . V. K r i s h n a Ayyar, In the Legislature of Those Days ( M a d r a s , 1956), p . 4 5 .
130
The raja became district board president in 1921. For his role in breaking up
anti-ministerial organisations see Hindu 24 August 1923.
131
Justice 10 January 1921, cutting in G.O. 171 (Public) dated 26 March 1921.
T.N.A.

323
 The emergence of provincial politics

and who could guarantee their own return to Council seats.132 Those
who particularly suffered as a result of the changing emphasis of
Justice Party policy were the innumerable publicists and petty political
organisers who had carried on the non-Brahman communal agitation
but who had no secured local bases under them. When Ramarayaningar
took office, they expected his support and patronage. Most were to
find very quickly, however, that they were not to get it. In the age of
Council politics, journalists and demagogues, such as O. Kandaswami
Chetty and the educationist C. Ramalinga Reddy, and local party
organisers, such as J. Ramanathan of Madura and C. Natesan of
Madras city, were no longer necessary nor valuable.133 By 1923, they
had formed an anti-Ministerial Justice Party, 134 the existence of
which demonstrates clearly the character of politics in Madras at this
time. The basic cleavage in the Council was between followers and
opponents of the Ministry, who were determined by the favours they
received or failed to receive from the Ministry.
Once it had secured office, the Justice Party Ministry threw off the
last vestiges of its guise as a party of local grievance against the centre,
which, we saw, characterised its formation. Between 1917 and* 1920,
as it gained access to the centre and saw the prospect of power, it had
steadily modified this image. Now the Ministry could abandon it com-
pletely and enjoy the fruits of its labours. Its opposition to the
Dharmarakshana Sabha and temple reform gave way to the prepara-
tion of a bill to bring the temples under central - i.e. its own -
control. 135 Its early antipathy to the interference of the Secretariat in
local self-government appointments gave way to an aggressive use of
local self-government patronage by comparison to which Mylapore's
previous endeavours were mere tinkerings. Its views on the control of
service appointments were revised. No longer did it wish the I.C.S.
132
Notably, T. Desikachari of Trichinopoly, A. S. Krishna Rao of Nellore, N.
Subba Rao of South Kanara and T. M. Narasimhacharlu of Cuddapah. The last
two also had personal ties with the Ministry. The Cult of Incompetence, being an
impartial enquiry into the record of the First Madras Ministry (Madras, 1923),
PP- 37-9-
133
Ramarayaningar ridiculed C. Ramalinga Reddy, C. Natesan and O. Kandaswami
Chetty by exposing publicly their various personal demands for favour - which
included district board presidencies, University Vice-Chancellorships, jobs for
their relatives and council nominations. Kandaswami Chetty eventually sued
Ramarayaningar for libel and won damages of one pie. Hindu (weekly) 7 and 17
February 1924 and 29 April 1926. Madras Legislative Council Proceedings, xvn
( M a r c h 1924), 145.
134
Hindu 26 F e b r u a r y , 28 M a y a n d 13 J u n e 1923.
135
See Baker, 'Political Change in South India 1919-1937', pp. 127-8.

324
 Home Rule League, Justice Party and Congress

to retain a role of impartiality and to prevent one group of Indians

from monopolising the entry and promotion points in the careers
structure. It wanted full powers of appointment to be given to the
Ministers.136 The sympathy which some of its members, notably P.
Ramarayaningar, had shown to the cause of Andhra separatism also
evaporated.137 Devolution meant less patronage.
Of course, Mylapore's own reaction to the new circumstances was
precisely the opposite of the Ministry's and it came to take up many of
the positions which the S.I.P.A. had occupied against it four years
before. By 1920, it had lost many of its privileges and powers. The
Local and Municipal Department had gone forever, the Dharmarak-
shana Sabha was being made redundant by the Ministry's legislation
and the University was in the process of being swamped by Ministerial
appointees. Admittedly, all had not yet disappeared. Mylapore still
dominated the legal profession and had a hand in one or two depart-
ments. But it had been replaced substantially at the centre and so was
moved to obstruct the growth of central influence. It bitterly opposed
the temple legislation which once had been its pet project;138 it
demanded the rapid decentralisation of local self-government powers,
which once it had fought strongly to prevent;139 and it even supported
the retention of I.C.S. control of selection to the services, over which
it had gone to war with the British in 1916.140 Indeed, by the early
1920s Mylapore, the archfiend of the Home Rule League, had become
the most loyal supporter of the continuation of the British connection.
Intellectually, through such organisations as the Liberal League, it
evinced policies which pleased the heart of every Secretary of State.
And politically, it offered loyal service to the raj: in 1921, C. P.
Ramaswami Iyer, the devil incarnate of 1916-18, was offered the
portfolio of the Home Department - an office which had been
reserved and considered too delicate and important to be transferred
to the Justice Party. Many of the fears which the British had felt
about relying over much on the collaboration of Mylapore were now
passed on to the collaboration of the Ministry; and Mylapore, like the
136
A. Ramaswami Mudaliar, Mirror of the Year (Madras, 1928), p. 209.
137
See J. G. Leonard, 'Politics and Social Change in South India: A Study of the
Andhra Movement' in Journal of Commonwealth Political Studies, v : 1 (1967).
138
M. Ramachandra Rao to P. S. Sivaswami Iyer, 13 June 1923. P. S. Sivaswami
Iyer Papers. N.A.L
139
For example, V. Krishnaswami Iyer, when an Executive Councillor, had been
very heavy-handed with the local rights of municipalities; now Mylapore,
in concert with all of the Ministry's opponents, stood out for local rights.
140
P. S. Sivaswami Iyer to T. Sapru, 6 May 1922. P. S. Sivaswami Iyer Papers.
N.A.I.

325
 The emergence of provincial politics

non-Brahmans of 1916, became to the British the sought-after altern-

ative which required help. The world of provincial politics had been
turned upside down.
1920, however, did not only mark the replacement of one clique
by another in the offices of the capital. It also signified a change in
the way that cliques were formed and hence an alteration in the struc-
ture of politics. Mylapore's position had rested on its ability to control
the professional institutions of the law, education and bureaucracy
which alone had comprised the channels of political communication
between the local and provincial levels. After 1920, however, these
channels were joined and largely replaced by those of election. As it
realised before the 1920 election, when its members begged Willingdon
to appoint them Ministers before the election had taken place, Myla-
pore could not compete in this new medium of politics in which local
resources were all important. Its Home Rule campaign had been too
successful for its own good, and franchise qualifications and con-
stituency boundaries had been drawn too near the ground for its
comfort. It was no longer of very much account.
In relation to the localities, the new Ministry found itself in a weaker
position than Mylapore had been. As we have seen, Mylapore itself
could not entirely mould local political situations to fit its own abstract
designs. It did not have the wherewithal to invent magnates. But it
could choose between magnates who were already present and inter-
vene decisively in factional struggles to aid one side or another. And,
importantly, local magnates who were injured by it had no real means
of obtaining redress. Mylapore was appointed by the British and was
not responsible to any institution or body of Indian opinion. If it added
up its sums of patronage incorrectly it could not be dismissed.
The Ministry, however, was in a rather different boat. It too could
only intervene in local situations and not create them. As the 1920
election proved, its nominations to the party ticket were not valuable
in themselves. It had little control over the vast splay of economic
sanctions and patronage incentives which enabled the local magnate
to commandeer the votes of tenants, dependents and clients. Indeed,
in some ways, its patronage powers were less than those which Myla-
pore had possessed. Before the Religious Endowments Act of 1926, it
had no Dharmarakshana Sabha to give it entry to the temples; and the
local self-government legislation of 1919 and 1920, while increasing
the number of seats and executive positions which the Minister could
fill by nomination, increased much faster the number which he could
not and which were elective. The Justice Party Ministry was not well

326
 Home Rule League, Justice Party and Congress

placed to dominate the localities. Yet the localities had gained greatly
in their ability to influence the Ministry, for if the Ministry disregarded
the opinions of those local magnates who sat in the new Legislative
Council, they could bring it to its knees.
Ramarayaningar rapidly discovered the tenuousness of his position
and was forced to give way to local pressure in the matter of several
important nominations. In 1921, still flushed with the success of his
recent appointment and eager to demonstrate to the British the reality
of the non-Brahman cause, he tried to fill two vacancies on the district
boards of Ramnad and Salem with 'professional' non-Brahman clients
of his own and badly burnt his fingers. In Ramnad, the new district
board president and recent convert to the Justice Party, the Raja of
Ramnad, would have nothing to do with Ramarayaningar's man and
insisted instead on the place being given to a Brahman lawyer of his
acquaintance. After an exchange of several angry letters, the Minister
backed down.141 In Salem, the district board president, G. Foulkes, a
European zamindar and erstwhile friend of the Justice Party, was
no less emphatic. He refused to accept the Minister's choice of P. L.
Ramaswami Naicker, an active non-Brahman party worker, and
forced Ramarayaningar to appoint another Brahman lawyer. He also
provided the Minister with the maxim which was to guide his
remaining years in office: 'On general principles if Government is
going to nominate members without reference to responsible local
opinion, it is not going to be long before Government is going to land
itself in difficulties.'142 Ramarayaningar marked this sentence well
and kept his role down to that of offering sweetmeats to the local men
who mattered. He performed this task with some brilliance and held
his place until 1926, constantly searching for more morsels with
which to favour his magnate 'followers'. By 1926, however, even his
cupboard was bare and the Ministry fell.
In essence, the Montagu-Chelmsford Councils represented a
further 'localisation' of provincial politics. This development was to
some extent implicit in the political functions which Mylapore had
assumed after 1910. It brought provincial power to the locality on the
terms of the locality, that is around the magnate. The narrowness of
the place which Mylapore enjoyed at the centre, however, also opened
some, be they only small, channels through which provincial issues
could reach the locality. In 1916, when Mylapore and the non-
Brahmans stood against each other in the province, their respective
141
G.O. 545 (L.S.G.) dated 19 March 1921. T.N.A.
142
G.O. 1295 (L.S.G.) dated 5 July 1921. T.N.A.

327
 The emergence of provincial politics

local followers dressed themselves, however poorly, in provincial

cloth. Mylapore was not of the locality and it still retained an interest
in the issue-orientated forms of provincial politics. After 1920, how-
ever, the extension of elective opportunities allowed the localities to
swallow up the province. For six years, as the echoes of the communal
controversy died away, there ceased to be any real difference between
the way that local and provincial politics were run. Everything was
dominated by the locality and by the need to make person-to-person
transactions of power. In spite of the Swarajists' prodding in 1926,
it was probably not until the economic and constitutional revolutions
of the 1930s that even the rhetoric of the politics of issue and com-
munity re-entered the provincial arena in any significant manner.
Without issues to which the localities would respond, the Congress
rapidly withered. After Gandhi had taken it into non-co-operation
in 1920, the organisation lost the importance which it had attained in
provincial politics in the three preceding years. It existed to put
pressure on the British during periods of constitutional negotiation.
But the early 1920s, like the mid-1890s, marked an interlude in nego-
tiation and, at such periods, it was not clear what the Congress was
supposed to do. The non-co-operation tactic did not prove a success
in South India. On the one hand, in Malabar it raised a Moplah rebel-
lion of savage fury which the Congress leadership soon disowned.143
On the other, led in Tamilnad by C. Rajagopalachari and in Andhra
by T. Prakasam and Konda Venkatappayya, it fanned agitation only
intermittently until 1922. Rajagopalachari relied on Khilafat
Muslims and local powers irritated by aspects of administrative
advance for most of his support but his movement never took off.
After the Tilak fund had run out, money became extremely short and
activity had virtually ceased by 1921. 144 The Andhra leaders were
able to use an autonomous crusade against the government by village
officers in Guntur and Kistna, whose plight we noted earlier, and the
financial and organisational support of Komati trading groups to
whom they were personally connected, to raise a more impressive
campaign.145 But that too was more or less dead by February 1922
when Gandhi called off non-co-operation. While the Congress re-
fused to act within the framework of institutional politics in Madras,
it was irrelevant to most political interests. Rajagopalachari's
apparent victory in 1920 proved to be hollow, for he had won an
143
All India Congress Committee Papers, File 1, Part 6 o f 1922. N.M.M.L.
144
Baker, 'Political Change in South India 1 9 1 9 - 1 9 3 7 ' , pp. 3 3 7 - 9 0 .
145
Ibid.; Venkatarangaiya, Freedom Struggle in Andhra Pradesh III, 2 0 - 6 0 , 2 4 0 - 3 0 0 .

328
 Home Rule League, Justice Party and Congress

organisation which nobody else wanted in its present form. Between

1923 and 1926, the old Kasturi Ranga Iyengar group, which had been
in the wilderness, returned under the direction of S. Satyamurthi and
S. Srinivasa Iyengar to stake its claim to the Congress again. As the
Swarajya Party, it re-established its contacts with many of its old
local magnates, who had been sitting as Independents in the Council,
and fought the 1926 election as a leadership group with some success.
Rajagopalachari was driven from the Tamil Nad C.C., and some of
his non-Brahman allies, to show their disapproval at the return of
the city, started their own anti-Brahman movement.146 Later,
Rajagopalachari himself returned to the fray to carry forward to the
1930s the struggle against city cliquism which had begun in the 1890s.
The struggle ended perhaps only in 1937, when Satyamurthi, using
the network of contacts available solely to an important Madras city
politician, constructed the machine which was to take Congress to its
great victory at the polls, only to find himself replaced in the centre
of the organisation at the last minute by Rajagopalachari.147
146
Baker, The Politics of South India. 1920—1937, ch. 4.
147
Ibid.

329
 Conclusion

By the 1830s, when their initial political settlement had been com-
pleted, the British had done much to alter the South Indian political
system which they had found in 1800. They had established a new
level of state authority over all the variegated territories which
comprised their province and had liquidated the previous, more
regional, warrior level of government. Even where the warriors had
not been destroyed but transmogrified into zamindars, the right to use
force, on which warrior I zamindar rule ultimately depended for its
success, was steadily, albeit slowly, undermined. In place of warrior
government, the British built the machinery of a centralised bureau-
cratic state. They promised to bring strong civil government to their
province; to substitute the rule of law for that of force; to guarantee
the possession of private property; to promote economic growth and
social development within the framework of a new, larger and more
unified state.
As we have seen, however, by 1870 the British had lived up to few
of these promises. By liquidating the warriors, they had removed the
old core around which political society had been organised. But
whether or not they could provide a new and greater core depended
very much on the volume and intensity of the political relationships
which they could establish with the social elements freed by the
dismantlement of the warrior regimes. Certainly in terms of revenue
flow, the British developed for themselves a more important place in
society than ever the warriors had possessed: even their loosely
jointed tribute system guaranteed them a higher regular income.1
But it is a serious mistake to regard revenue collection as the only,
or even the main, connection between the Indian state and its society.
In addition to linkages made through the extraction of revenue, there
also must be linkages made through its redistribution and redeploy-
ment, through relationships of licensed force and through relation-
ships of cultural sympathy. In comparison to the warriors, British
rule was peculiarly deficient in these latter connections.
1
Stein, 'Integration of the Agrarian System of South India' in Frykenberg (ed.),
Land Control and Social Structure in Indian History, pp. 210-12.

330
 Conclusion

Warriors spent most of the revenues which they had gathered on

conspicuous consumption at their courts, on their armies and on
religious and cultural patronage inside their territories. Indeed, if
they were successful at looting, their local expenditures amounted to
more than their local incomes. Warrior courts represented centres
through which resources flowed both in and out and a variety of
'state-level' groups — priests, merchants, artisans, soldiers, admini-
strators and poets — were organised around the outflow. Fort St
George, however, as we have seen, was forced to export most of its
revenue out of Madras and, consequently, had little left to reinvest
in its province. Its educational expenditure nowhere near matched the
warriors' support of Brahmans and literati; it endowed no new
temples; it built far fewer palaces, and built them in a way which
minimised involvement with local skills and the local economy; from
the 1850s, it recruited ever smaller numbers of troops, for its over-
lord, the Government of India, preferred soldiers from the 'martial'
races of the North; it even discarded elaborate court rituals and the
patronage of a court culture.
Not only did the British fail to extend the activity of the state into
society but, for ideological reasons, they also weakened many of the
institutions of connection which the warriors had already built for
them and which they had simply to maintain. Fort St George adopted
economic policies which lessened (although did not destroy) its role
as a commercial monopolist, and hence made national and inter-
national trade independent of it and of greater weight than service
to the state. The intimate connections between, for example,
Vijayanagar warriors and Telugu Komati merchants, which had
spread over large tracts of the economy, were replicated only in
miniature by the connections between the British government and
Indian mercantile interests. The British also introduced the law of
private property which altered the character of several important
socio-economic institutions. Religious mams, for example, ceased to
be franchises held at the discretion of the state and became private
properties held in law independently of state interference. British
attitudes towards religion and social life led to a further reduction
of state influence. Formally by 1863, but in practice as early as 1840,
the British had severed the relationship of their government to the
institutions of religion and so had relinquished control of the vast
economic and emotional resources of the temples. Although Fort St
George did attempt to create a new series of ideological and cultural
bonds with its subjects, these were never strong. Western education
and court honorifics (Rao Bahadurships, knighthoods, etc.) penetrated
 The emergence of provincial politics

to a far shallower level of society than had the religiously orientated

political symbols of the warriors. Moreover, lack of patronage pre-
vented the British from tying their honours and their culture to an
important system of material rewards. The British were much less
present in the society which they governed than had been their war-
rior predecessors: in many ways, they were 'absentee' rulers.
It would be true to say that the major connection between the British
government and its subjects in the nineteenth century was through
the formal administration and, particularly, through the processes
by which revenue was extracted. This admission in itself suggests
the likelihood that this government would be weak. As we have seen,
the way in which the administration was organised turned the
likelihood into fact. The British had brought no new service groups
with them into Madras and so were unable to adopt the tried and
tested tactics of South Indian statecraft. They had to find the mater-
ial for their administration from within the existing structure of
Madras society. But they greatly feared that, were they to use the
socio-political connections which the warriors' administrators had
developed and which were waiting for them, they would lock them-
selves out of their own government. In consequence, the#y came to
anchor their administration to the dominant peasant elites of the
South and to support their higher bureaucrats, who were supposed
to relate the peasants to the raj, with very insufficient force. This
policy contained an inherent contradiction, for the British had
failed to perceive that while 'state-level' administrative groups at
least had some points of interest in common with them (and, under
the right circumstances, might be eager to build state power if only
because it meant more power for themselves as well), the new
peasant collaborators had none. The dominant peasant elites over
most of South India were of local-level culture and used resources
derived from within the locality to maintain their power. Connection
to the government was important to them less because it drew them
into a wider state structure than because it could be used to prevent
the state from interfering in the locality. Local powers could block
external influence at the boundaries of their localities and do what
they pleased inside.
By 1870, the net results of British rule had been to fragment the
political integrations achieved by the warriors and to bury effective
political power deep in the localities from which the state had been
all but completely excluded. Each temple, each 'rural locality', each
section of a town became a potentially autonomous arena in which
the disposition of internal forces determined the pattern of political
dominance. The profound parochialism into which South India had

332
 Conclusion

fallen was apparent everywhere in the political attitudes and behav-

iour of the period. It was reflected in the way that bureaucrats sold
their authority to local magnates for bribes; it was reflected in the
manner in which temple trustees personally appropriated the endow-
ments of their temples and frequently allowed temple fabric and
public ceremonial to run down; it was reflected in the process by
which the intermediary institutions of clan political structures (in
the few areas where these had retained a political significance into the
nineteenth century) steadily became broken up; it was reflected in the
success with which the rural-local boss arrogated to himself de facto
powers of arbitration, revenue collection and police coercion over his
local subjects. Perhaps the final and clearest indicator of the domi-
nance of the locality over higher levels of political integration can be
seen in the attitude of the British courts to questions of landed
property. In the early nineteenth century, British legal policy had
sought to bring western private property rights to the locality and to
protect the landholdings of every peasant with the weight of a central
legal system. Had this been accomplished, Fort St George would have
been in a prime position to order the socio-political base of rural
society. In practice, however, so dubiously did the courts view the
relationship between the politically flexible village records of the
ryotwari settlement and the problem of justice in apportioning village
lands that they refused to accept ryotwari records as proof of the
ownership of lands. As there was no order requiring the compulsory
legal registration of landholdings, this meant that in most cases they
refused to link village lands to the central legal grid. The courts stood
off from local society and tacitly recognised its ability to come to its
own arrangements about land.2 External authorities at every level
were permitted scant influence in the locality and consequently
were unable to attract and pattern the political relationships of men
who were their nominal inferiors. In mid-nineteenth-century Madras,
all important politics had become local politics.
The process of rebuilding a greater political state began in earnest
in the 1870s. In part it was accomplished simply by increasing the
pressure of the bureaucratic centre on the periphery and by tighten-
ing up the existing administrative system. As Burton Stein has argued,
at this time the promises of British administrative theory, which
had been made in 1800 but had lain in abeyance for three-quarters of
a century, started to be fulfilled.3 But in much larger part, the process
was accomplished through the development of new institutions and
2
See my 'Law and Land in South India'. Unpublished paper read at Conference on
Indian Economic History, University of Pennsylvania, 1975.
3
Stein, 'Integration of the Agrarian System of South India', pp. 211 - 1 2 .

333
 The emergence of provincial politics

the emergence of new political forces which had no counterpart in

previous South Indian history and which produced not only a wider
political integration but one of a qualitatively new kind. The expan-
sion of the commercial economy and the creation of a new range of
administrative institutions, as we have seen, lured the powers of the
locality into participating in much broader politico-economic
structures: they altered the dimensions of the local political arena.
Moreover, from the 1910s, further institutional changes helped to
bring the capital back into the localities and, thereby, to articulate
a framework of provincial politics. What was so novel about these
developments was that the local groups who were being drawn into the
state system had never been fully integrated into any supra-local
state system before.4 They were largely groups of local-level culture
whose relationships with the warrior regimes had been indirect and
often antagonistic. Warrior rule had incorporated them by forcing
groups of state-level culture on top of them; now they themselves were
to play active roles in the operations of the new state. By the early 1920s,
they dominated district-level government through the rural boards
and other administrative committees and had begun to fill the seats
in the Legislative Council — the 'court' of their provincial governor.
Political and economic change began slowly but surely to forge great
alterations in the elite sections of the local level: marriage circles ex-
panded, literacy became more common, symbols of deference and
defiance changed and new social and cultural perceptions emerged.5
The peasant elites of the 1920s were structurally different from the
rural-local bosses of the same peasant families who had dominated rural
society in separate localities just sixty years before.
But the processes of unification did not only work their way up
from the bottom of political society. As Fort St George became more
competent at the state level so its influence and relevance spread
across the province. This development had two important conse-
quences. In the first place, administrative groups of state-level
culture, whose loci of operations had been reduced almost to the
district alone, were pulled back towards a larger state centre. They
organised themselves around the provincial capital in order to extend
themselves through the legal, educational and service networks of the
whole presidency and to draw sustenance from the state-centre for
their own cultural pursuits. Gradually, the growing intrusion of the
4
Their pre-warrior political system was based on the clan and not the state.
5
See my 'Country Politics: Madras 1880 to 1930'.

334
 Conclusion

Government of India in the affairs of South India also pushed them

into making connections with other service groups across the sub-
continent. The province and the nation ceased to be abstract notions
but became concrete political contexts in which they were forced to
work. Moreover, the language and concepts through which Fort St
George sought to govern its charge began to make their own impact.
The interest groups and political communities which it defined as
operating at the provincial level began to form and to fill the niches
which had been cut for them in the political structure. In a strange
way, British misconceptions about the relationships of Indian politi-
cal society started to produce real relationships in the image of
the misconceptions. Without arguing that it was only state action
that created the politics of, particularly, the caste association (for
many social and economic factors helped to determine the emergence
of specific caste communities), it would be difficult to deny that state
action played a major part.
Many of the changes which we have followed through from the
1870s and brought to a conclusion in 1920 helped to lay the founda-
tions of politics in South India today. The old ways of force were
slowly dying; the locality as a political notion was in decay; the
province and the nation were beginning to have a political meaning
in much wider circles. The Home Rule and non-Brahman movements
of 1917 to 1920 attested to these profound changes. We have sought
to explain the movements less by reference to the issues which divided
them than by reference to the institutional context in which they
arose. Ideas similar to those of the Home Rule League and non-
Brahman leaders could be found in the minds of some, be it only a
few, men at any time from the late 1870s and, in all probability, from
long before that too. But what was so crucial about the movements
was not the ideas alone but the way in which those ideas interacted
with the new political context of 1917 to produce political relation-
ships and political forces which would have been inconceivable a few
years earlier. The Home Rule League and non-Brahman movements
were the first manifestations of political division in the newly
articulated provincial structure of politics. This structure was being
wrought out of the material of centralised bureaucratic power, of
franchises and elective institutions, of civil processes and the rule of
law. It was beginning to look like the political structure which the
Indian Republic inherited from the British in 1947 and has been
obliged to work with ever since.

335
 GLOSSARY

Abkari: Liquor
Anicut: Dam
Arrack: Liquor
Bania: Moneylender/Merchant
Bhadralok: Bengali educated gentry
Choultry: Charitable institution
Crore: 10,000,000
Dallal: Commercial broker
Dharmakartha: Temple trustee
Diwan: Prime Minister
Dubash: Agent
Firka: Part of a taluk
Gumastah: Clerk
Guru: Holy man
Hartal: Strike
Hundis: Cheques for transfers of funds
Huzur: Head office
Huzur Sheristidar: Head clerk
Inam: Land held at reduced assessment (usually in return for services)
Jajmani: Transactions between an inferior and superior in the caste system
Jamabundi: Revenue settlement
Jati: Caste (in the sense of an endogamous unit)
Jenmi: Holder of large landed property under ryotwari system in Malabar
Jutka (-wallah): Horse-drawn carriage (its operator)
Kanomdar: Holder of a privileged tenure from a jenmi
Kazi: Judge under Islamic law
Kiramam: Revenue village
Kumbabishekham: Purification ceremony
Kurnam: Village accountant
Lakh: 100,000
Madrassa: Islamic school
Mahant: Principal of a monastery
Math: Monastery
Mirasi (-dar): Privileged land tenure under ryotwari system in areas of
Coromandel (its holder)
Mittadar: Holder of a small property under permanent settlement
Mofussil: Provinces (as opposed to metropolis)

336
 Glossary

Mohurram: Islamic festival

Moplah: Muslim peasant of upland Malabar
Munsiff: Magistrate
Panchama: Untouchable
Panchayat: Council (classically offivepeople)
Pandarasanidhi: Principal of a monastery
Pariah: Untouchable
Patta (-dar): Document giving right to hold land under ryotwari system
(its holder)
Patsala: Sanskrit school
Pattagar: Traditional leader of Gounder Vellala community
Poligar: Small warrior chief of the eighteenth century
Raj: Rule (colloquially British rule)
Ryot (wari): Peasant (system by which each peasant is assessed separately
for revenue)
Sabha: Association
Sachimattam: Bribe
Samasthanam: Estate (of a great lord)
Sankarachariar (or -yar): Religious leader
Sirker: State
Satyagraha: (Lit. soul-force; in practice, civil disobedience)
Shrotriemdar: Holder of a small property under permanent settlement
Sowcar (or Sahukar): Moneylender
Swami: Holy man
Tahsil: Division in revenue administration
Tahsildar: Officer in charge of a tahsil
Takavi: State-backed credit for long-term loans
Taluk: Equivalent of tahsil
Thambiram: Disciple in charge of a subordinate monastery
Varna: Four-caste system by which Hindu society is divided
Zamindar: Holder of a property under permanent settlement
Zamorin: Prince in Malabar

337
 BIBLIOGRAPHY

Government Records
The proceedings of the Secretary of State for India in the Public and Judicial
Department. India Office Library.
The proceedings of the Government of India in the Home Department, branches
of Education, Judicial, Police, Political, Public Departments. National
Archives of India.
The proceedings of the Government of Madras in the Departments of Education,
Home, Judicial, Legislative, and Revenue; and in the Board of Revenue.
Tamil Nad Archives and State Archives, Hyderabad.

Official Publications
(i) Parliamentary Papers
Report of the Indian Famine Commission. 1880. Volume 11, 1881. Volume LII.
[Command. 2735].
Appendix to the Report of the Indian Famine Commission, 1898, being Minutes of
Evidence, etc., Volume II. The Madras Presidency. 1889. Volume xxxn.
[Command. 9253].
Minutes of Evidence taken before the Royal Commission upon Decentralization in
India, Madras. 1908. Volume XLIV. [Command. 4361].
Royal Commission on the Public Services in India. Appendix to the Report of the
Commissionersy Volume II. Minutes of Evidence Relating to the Indian Civil
Services Taken in Madras from 8th to 18th January 1913. 1914. Volume xxi.
[Command. 7293].
Addresses Presented in India to His Excellency the Viceroy and the Right Honorable
the Secretary of State of India. 1918. Volume xvm. [Command. 9178].
Report on Indian Constitutional Reform. 1918. Volume vm. [Command. 9109].
Minutes of Evidence taken before the Indian Industrial Commission 1916-18. Volume
II. 1919. Volume xix. [Command. 235].
Report of the Committee appointed by the Secretary of State for India to enquire
into questions connected with Franchise and other matters relating to constitutional
reforms. Chairman Lord Southborough. 1919. Volume xvi. [Command. 141].
Joint Select Committee on the Government of India Bill. 1919. Volume iv.
Report of the Committee appointed by the Secretary of State for India to Advise on the
Question of the Financial Relations between the Central and Provincial Govern-
ments in India. 1920. Volume xiv. [Command. 724].

338
 Bibliography

(2) His Majesty's Stationery Office

Report of the Royal Commission on Agriculture in India. Appendix. Volume 111.
1927; Volume xiv. 1928.
Indian Statutory Commission. Vol. VI. Memorandum Submitted by the Government
of Madras. 1930.

(3) Government of India

Education Commission. Report by the Madras Provincial Committee. Calcutta, 1884.
Proceedings of the Public Service Commission. Volume V. Calcutta, 1887.
Indian Universities' Commission. Simla, 1902.
Evidence taken before the Reforms Committee (Franchise). Madras. Calcutta, 1919.
Indian Central Cotton Commission. General Report on Eight Investigations into the
Finance and Marketing of Cultivators' Cotton. 1925—8. Bombay, n.d.
Memorandum on the Working of Representative Institutions in Local Self-Government.
New Delhi, 1928.
Reports of the Provincial Banking Enquiry Committees, 1929—30. Calcutta, 1931.
Agriculture Statistics of British India. Calcutta, 1884-5 to 1920-1.
Prices and Wages in India. Calcutta, 1889 to 1921.

(4) Government of Madras - Special

Report of the Commissioners for the Investigation of Alleged Cases of Torture in the
Madras Presidency. Madras, 1855.
Selections from the Madras Records, xxn. Madras, 1870.
Report of the Committee on Local Self-Government in Madras. 1882. Madras, 1883.
J. H. Garstin, Report on the Revision of the Revenue Establishments in the Madras
Presidency. Madras, 1883.
H. S. Thomas, Report on Tanjore Remissions in Fasli 1294 (A.D. 1884—5). Madras,
1885.
H. S. Thomas, Report on Mr Charles Stewart Crole. Collector of Madura. Madras,
1886.
C. H. Benson, An Account of the Kurnool District based on an Analysis of Statistical
Information Relating Thereto and on Personal Observation. Madras, 1889.
S. Srinivasa Raghavaiyangar, Memorandum on the Progress of the Madras Presidency
during the Last Forty Years of British Administration. Madras, 1892.
C. H. Benson, A Statistical Atlas of the Madras Presidency. Madras, 1895. (Revised
editions in 1908 and 1924.)
F. A. Nicholson, Report regarding the possibilities of introducing Agricultural Banks
into the Madras Presidency. Madras, 1895-7.
Notes on the Working of Certain Departments of Government during the Administration
of H. E. Lord Wenlock, with General Summary. Madras, 1896.
Statement of the Police Committee on the Administration of the District Police in the
Madras Presidency. Madras, 1902.
W. S. Meyer, Report on the Constitution of Additional Districts, Divisions and Taluks
in the Madras Presidency, and on Other Connected Matters. Madras, 1904.
Report on the Administration of His Excellency the Honourable Sir Arthur Lawley,
G.C.S.I., G.C.I.E.,K.C.M.G., Governor of Madras. Madras, 1912.

339
 The emergence of provincial politics

Report of the Forest Committee. 1913. Madras, 1913.

Madras Provincial Banking Enquiry Committee. Evidence. Madras, 1930.

(5) Government of Madras — Annual

Report on the Administration of the Abkari Revenue in the Presidency of Fort St
George.
Report on the Administration of the Estates under the Court of Wards in the Madras
Presidency.
Report on the Administration of the Income Tax under Act II of 1886.
Administration Report of the Forest Department of the Madras Presidency.
Report on the Administration of Civil Justice in the Madras Presidency.
Statistics of Criminal Courts in the Madras Presidency.
Report on the Settlement of the Land Revenue in the Districts of the Madras Presidency.
Report on the Administration of the Madras Presidency.
Report on the Administration of the Municipality of Madras.
Report on the Working of the District Municipalities in Madras Presidency.
Administration Report of the Madras Police.
Report on Public Instruction in the Madras Presidency.
Report of the Working of the Rural Boards in the Madras Presidency.
Annual Report of the Sanitary Commissioner for Madras.

(6) Census of India.

Census of India. 1881. Madras. Madras, 1882-3.
Census of India. 1891. Madras. Volume XIII to XV. Madras, 1893.
Census of India. 1901. Madras. Volumes XIV to XVI. Madras, 1902.
Census of India. 1911. Madras. Volume XII. Madras, 1912.
Census of India. 1921. Madras. Volume XIII. Madras, 1922.
Census of India. 1931. Madras. Volume XIII. Madras, 1933.

(7) Imperial Gazetteers

The Imperial Gazetteer of India. Volume IV. Oxford, 1907.

(8) Madras District Manuals and Gazetteers

Brackenbury, C. F. Cuddapah. Madras, 1916.
Crole, C. S. A Chingleput District Manual. Madras, 1879.
Dykes, J. W. B. Salem, An Indian Collectorate. London, 1853.
Francis, W. Bellary. Madras, 1904.
Madura, Madras, 1906.
The Nilgiris. Madras, 1908.
South Arcot. Madras, 1906.
Vizagapatam. Madras, 1907.
Gopalakrishnama Chetty, N. A Manual of the Kurnool District. Madras, 1886.
Gribble, J. D. B. A Manual of the District of Cuddapah. Madras, 1875.
Hall, J. F. South Kanara, Madras, 1938.
Hemingway, F. R. Godaveri. Madras, 1906.
Tanjore. Madras, 1906.
Trichinopoly. Madras, 1906.

34O
 Bibliography

Innes, C. A. Malabar. Madras, 1908.

Krishnaswami Ayyar, K. N. Statistical Appendix, Together With a Supplement to the
District Gazetteers (1917) for Tinnevelly District. Madras, 1934.
Mackenzie, G. A Manual of the Kistna District. Madras, 1883.
Maltby, T. J. The Ganjam District Manual. Madras, 1882.
Nicholson, F. A. The Coimbatore District Manual. Madras, 1887.
and H. Stuart, Coimbatore. Madras, 1898.
Pate, H. R. Tinnevelly. Madras, 1917.
Richards, F. J. Salem. Madras, 1918.

Private Papers
Ampthill Collection. India Office Library.
Cross Collection. India Office Library.
Kilbracken Collection. India Office Library
Montagu Collection. India Office Library.
Willingdon Collection. India Office Library.
Papers of V. Krishnaswami Iyer. National Archives of India.
Papers of G. A. Natesan. Nehru Memorial Museum Library.
Papers of P. Kesava Pillai. Nehru Memorial Museum Library.
Papers of V. S. Srinivasa Sastri. National Archives of India.
Papers of P. S. Sivaswami Iyer. National Archives of India.

Papers of Organisations
Papers of the All-India Congress Committee. Nehru Memorial Museum Library.
Papers of the Madras Mahajana Sabha. Nehru Memorial Museum Library.
Papers of the Theosophical Society (of Mrs Besant). Theosophical Society Archives,
Adyar.

Newspapers
Reports on English Papers owned by Natives examined by the Criminal Investigations
Department, Madras, and on Vernacular Papers Examined by the Translators to
the Government of Madras. 1887-1921. (R.N.P.)
Hindu. Tri-weekly 1883-9, daily 1889-1921, weekly 1916-21.
Madras Mail. 1885-90 and 1916-20.
Madras Times. Daily 1888-91.
New India. Daily 1915 - 21.

Books, Pamphlets and Articles

Ananda Charlu, P. The Madras Bar and How to Improve It. Madras, 1893.
The Six Fold Need of Indian Politics. Madras, 1895.
The First Andhra Conference held at Bapatla in 1913. Bezwada, n.d.
Sree Andhra Jatheeya Kala Sala Masulipatam. Work and Appreciations. 1907—1918.
Masulipatam, 1918.

341
 The emergence of provincial politics

Anon. The Ways and Means for the Amelioration of the Non-Brahman Races.
Bangalore, 1893.
Anuntha Row Pantulu, V. An Old Man's Family Record and References. Madras,
1916.
Appa Rao Naidu, K. Communal Representation and Indian Constitutional Reforms.
Cocanada, 1918.
An Appeal Memorial of the Madras Landholders' Association Against the Madras
Proprietary Estates Village Service Act II of 1894. Madras, 1894.
Arasu Pulavar, V. V. S. Aiyer. Madras, 1951. (Tamil).
Tiru, Vi. Kaliyanasundarana. Tirunelveli, 1961. (Tamil).
Arokiyasami, M. The Kongu Country. Madras, 1956.
The Asylum Press Almanack and Directory of Madras and Southern India. Madras,
1910, 1922, 1925, 1930.
Baden Powell, B. H. The Land Systems of British India. Volume in. Oxford, 1892.
Baker, C. J. The Politics of South India 1920-1937. Cambridge, 1976.
and Washbrook, D. A. (eds.), South India: Political Institutions and Political
Change. New Delhi, 1975.
Balasubramaniam, K. M. South Indian Celebrities. Madras, Volume 1, 1934;
Volume 11, 1939.
The Life ofj. M. Nallaswami Pillai. Tiruchirapelle, 1965.
Baldwin, G. B. Industrial Growth in South India. Illinois, 1959.
Baliga, B. S. Studies in Madras Administration. Madras, i960.
Bayly, C. A. 'Patrons and Politics in Northern India'. Modern Asian Studies, vn: 3
(1973)-
Beaglehole, T. H. Thomas Munro and the Development of Administrative Policy in
Madras: 1792-1818. Cambridge, 1966.
Beck, B. E. F. Peasant Society in Konku. Vancouver, 1972.
'The Right—Left Division of South Indian Society'. Journal of Asian Studies.
xxix: 4(1970).
Besant, A. The Birthplace of New India. Adyar, 1917.
Dr Nair and Mrs Besant. Adyar, 1913.
Beteille, A. Caste, Class and Power: Changing Patterns of Stratification in a Tanjore
Village. Berkeley and Los Angeles, 1965.
Bhargava, G. S. V. V. Giri. Bombay, 1969.
Bhaviah Chowdary, K. A Brief History of the Kammas. Sangamjagarlamudi, 1955.
Bhimasankara Rao, The Indian National Congress — Then and Now. Rajahmundry,
1929.
Boag, G. T. The Madras Presidency 1881-1931. Madras, 1933.
Bobbili, Maharaja of. Bobbili Zamindari. Madras, 1907.
Broomfield, J. H. Elite Conflict in a Plural Society. Berkeley and Los Angeles, 1968.
Brown, J. M. Gandhi's Rise to Power. Cambridge, 1972.
Chandrasekharan, K. V. Krishnaswami Aiyer. Masulipatam, 1963.
Chenchayya, Darsi. Nenu Na Desamu. Vijayawada, 1952. (Telegu - Myself and My
Country).
Chiefs and Leading Families in the Madras Presidency. Madras, 1915.

342
 Bibliography

'C.P.' by his Contemporaries. Madras, 1959.

The Cult of Incompetence, being an impartial enquiry into the record of the First
Madras Ministry. Madras, 1923.
Das, R. K. The Temples of Tamilnad. Bombay, 1964.
Das Gupta, J. C. A National Biography for India. Dacca, 1911.
Directory of the Madras Legislature. Madras, 1938.
Dobbin, C. Urban Leadership in Western India. Politics and Communities in Bombay
City 1840-188$- London, 1972.
Dumont, L. Homo Hierarchicus, London, 1970.
Hierarchy and Marriage Alliance in South Indian Kinship. London, 1957.
Une Sous-Caste de VInde du Sud: organisation sociale et religieuse des Pramalai
Kallar. Paris, 1957.
Dupuis, J. Madras et le Nord du Coromandel. Etude des conditions de la vie indienne
dans une cadre geographique. Paris, i960.
Dutt, K. Iswara. Sparks and Fumes. Madras, 1929.
Firminger, W. F. (ed.), The Fifth Report of the Select Committee for the Affairs of the
East India Company. 28 July 1812. Volume III. Calcutta, 1918.
For and Against the Andhra Province. Masulipatam, 1914.
Fox, R. G. From Zamindar to Ballot Box. New York, 1969.
'Varna Schemes and Ideological Integration in Indian Society.' Comparative
Studies in Society and History, xi: 1 (1969).
'The Avatars of Indian Research?' Comparative Studies in Society and History.
xn: 1 (1970).
Frykenberg, R. E. Guntur District IJ88—1848. A History of Local Influence and
Central Authority in South India. Oxford, 1965.
(ed.), Land Control and Social Structure in Indian History. Madison, Milwaukee
and London, 1969.
Gopala Menon, N. A Short Sketch of the Life of Dr T. M. Nair. London, 1920.
Gopalaratnam, V. C. A Century Completed (The Madras High Court 1862-1962).
Madras, n.d.
Gordon, L. A. Bengal: The Nationalist Movement 1870—1940. New York and
London, 1974.
Gough, E. K. and Schneider, D. M. (eds.), Matrilineal Kinship. Berkeley and Los
Angeles, 1961.
Govindarajan, S. A. G. Subramania Iyer. New Delhi, 1969.
Govindarow Naidu, P. The Legislative Council Elections. Rajahmundry, 1920.
Gurunatham, J. The Andhra Movement. Guntur, 1913.
Hamerow, T. S. Restoration, Revolution and Reaction. Princeton, 1958.
Hardgrave, R. L. The Dravidian Movement. Bombay, 1965.
The Nadars of Tamilnad, Berkeley and Los Angeles, 1969.
Harrison, S. S. India — The Most Dangerous Decades. Princeton, i960.
Heimsath, C. Indian Nationalism and Hindu Social Reform. Princeton, 1964.
Heroes of the Hour. Madras, 1918.
Hilton Brown, J. Parry's of Madras: A Story of British Enterprise in India. Madras,
1954-
The Civilian's South India. London, 1921.

343
 The emergence of provincial politics

Hjelje, B. 'Slavery and Agricultural Bondage in the Nineteenth Century'. Scandina-

vian Economic History Review, xv: i and 2 (1967).
Organization of the Home Rule League. Passed at the First Meeting of the Council.
October 8th, 1916. Madras, n.d.
The National Home Rule League. How Founded and Why. Adyar, 1919.
Report of the Third Indian National Congress held at Madras on 27th, 28th, 29th and
30th December 1887. London, 1888.
Report of the Tenth Indian National Congress held at Madras on the 26th, 27th, 28th
and 29th December 1894. Madras, 1895.
Report of the Fourteenth Indian National Congress held at Madras on the 29th, 30th
and 31st December 1898. Madras, 1899.
Report on the Proceedings of the Nineteenth Indian National Congress held at Madras
on the 28th, 29th and 30th December 1903. Madras, 1904.
Report on the Twenty-Fourth Indian National Congress held at Madras on the 28th,
29th and 30th December 1908. Madras, 1909.
Indian Statesmen. Dewans and Prime Ministers. Madras, n.d.
Indian Year Book and Who's Who 1934—35. Bombay, 1935.
Irschick, E. F. Politics and Social Conflict in South India. Berkeley and Los Angeles,
1969.
Johnson, G. Provincial Politics and Indian Nationalism. Cambridge, 1973.
Justice Party Golden Jubilee Souvenir. Madras, 1968.
The Justice Year Book, 1929. Madras, n.d.
Kaleswara Rao, A. Na Jivita Katha-Navya Andhramu. Vijavawada, 1959. (Telugu-
My Life and.the New Andhra).
Kalyanasundra Mudaliar, T. V. Valkkai Kurippugal. Madras, 1969. (Tamil -
Autobiography).
Kamesvara Aiyar, B. V. Sir A. Shashiah Sastri. K.C.S.I. An Indian Statesman.
Madras, 1902.
Kandaswami Pillai, V. Tiruvavadathurai Kurisanam. Madras, 1921. (Tamil - A
Visit to Tiruvandathorai).
Kannappar, J. S. Suttirargal Yar? Madras, 1926. (Tamil - Who are the Sudras?)
Kasipathi, K. Tryst with Destiny. Hyderabad, 1970.
Kausikan, Rajaji. Madras, 1968. (Tamil).
Kessinger, T. G. Vilyatpur 1848—1968. Berkeley, Los Angeles and London, 1974.
Kodanda Rao, P. The Right Honourable V. S. Srinivasa Sastri: A Political Biography.
London, 1963.
Kothari, R. (ed.), Caste in Indian Politics. New Delhi, 1970.
Krishna Ayyar, R. V. In the Legislature of Those Days. Madras, 1956.
Krishnamurthi Mudaliar, C. Life and Activities ofK. Chidambaranatha Mudaliar.
Shiyali, 1938.
Sir Alladi Krishnaswami Iyer Shastyabdapoorthi Souvenir Volume. Madras, 1943.
The Hon'ble V. Krishnaswami Iyer. A Sketch. Trichinopoly, 1911.
Keraliyan, The Father of Political Agitation in Travancore. G. Parameswaram Pillai.
Trivandrum, 1948.
Kumar, D. Land and Caste in South India. Agricultural labour in the Madras
Presidency during the nineteenth century. Cambridge, 1965.

344
 Bibliography

Leach, E. R. (ed.), Aspects of Caste in South India, Ceylon and Northwest Pakistan.
Cambridge, i960.
Leonard, J. G. 'Politics and Social Change in South India: A Study of the Andhra
Movement'. Journal of Commonwealth Political Studies, v: 1 (1967).
'Urban Government under the Raj'. Modern Asian Studies, vn: 2 (1973).
Lingamurthy, V. 'Elections in Andhra'. Indian Journal of Political Sciences, xvi: 2.
(April-June 1955).
Biography of Sriraman Lodd Govindas Maharaj, The Hindu Hero. Madras, 1942.
Low, D. A. (ed.), Soundings in Modern South Asian History. London, 1968.
Madhavaiah, A. Thillai Govindan. London, 1916.
Madhavan Nair, C. A Short Life of Sir C. Sankaran Nair. Madras, n.d.
Mahadevan, P. Subramania Bharati: Poet and Patriot. Madras, 1957.
The Madras Mahajana Sabha Annual Report for 1885—86. Madras, 1886.
Madras Mahajana Sabha Diamond Jubilee Souvenir. Madras, 1946.
Proceedings of the Madras Native Association on the Resolution of the Government of
India on Local Self Government. Madras, 1883.
Report of the Executive Committee of the Madras Provincial Congress Committee for
the year ending 31st December 1915. Madras, 1916
The Madras Year Book. Madras, 1924.
McPherson, K. 'The Social Background and Politics of the Muslims of Tamilnad
1901-1937'. Indian Social and Economic History Review, vi: 4 (1969).
Marriott, M. Village India. Chicago, 1955.
Mehrotra, S. M. The Emergence of the Indian National Congress. New Delhi, 1917.
Menon, K. P. S. Many Worlds: An Autobiography. London, 1965.
C. Sankaran Nair. New Delhi, 1967.
Misra, B. B. The Indian Middle Classes. Oxford, 1961.
The Administrative History of India, 1834—1947- Bombay, 1970.
Mitra, N. N. Indian Annual Register. Calcutta, annual.
Molony, J. C. A Book of South India. London, 1926.
Montagu, E. S. An Indian Diary. London, 1930.
Morris, M. D. 'Economic Change and Agriculture in Nineteenth Century India'.
Indian Economic and Social History Review, m: 2 (1966).
Mukherjee, N. The Ryotwari System in Madras. Calcutta, 1962.
Murton, B. J. 'Key People in the Countryside: Decision Makers in Interior Tamil-
nad in the late eighteenth century.' Indian Economic and Social History
Review, x: 2 (1973).
Musgrave, P. J. 'Landlords and Lords of the Land: estate management and social
control in Uttar Pradesh 1860-1920'. Modern Asian Studies, vi: 3 (1972).
Nair, B. N. The Dynamic Brahman. Bombay, 1959.
Nair, T. M. The Evolution of Mrs Besant. Madras, 1918.
Nancharayya, B. V. Biography of Zamindar Lodd Govindas Varu. Madras, 1942.
Narain, D. Impact of Price Movements on Areas under Selected Crops in India, 1900-
1939. Cambridge, 1965.
Narasimhan, K. L. Madras city — a history. Madras, 1968.
Kasturi Ranga Iyengar. New Delhi, 1963.
Kasturi Srinivasan. Bombay, 1969.

345
 The emergence of provincial politics

Narayanaswami Naidu, B. V. (ed.), Rajah Sir Annamalai Chettiar Commemoration

Volume. Chidambaram, 1941.
Nathan, T. A. V. (ed.), The Justice Year Book, 1929. n.p., n.d.
Nethercott, A. H. The First Five Lives of Annie Besant. Chicago, i960.
The Last Four Lives of Annie Besant. (Chicago, 1963).
Nilakanta Sastri, K. A. (ed.), A Great Liberal. Speeches and Writings of Sir P. S.
Sivaswami Aiyar. Bombay, 1965.
Note on the Past and Present Administration of the Raja's Chattram in the Tanjore
and Madura Districts. Tanjore, 1908.
Padmanabha Iyer, K. V. A History of Sourashtras in South India by the Sourashtra
Literary Societies of Madras and Madura. Madura, 1942.
Parameswaran Pillai, G. Representative Men of Southern India. Madras, 1896.
Representative Indians. London, 1902.
Parthasarathi, R. T. The Dawn and Achievement of Indian Freedom. Salem, 1953.
Patro, A. P. 'The Justice Movement in India'. Asiatic Review, xxvm (1932).
Pattabhisitaramayya, B. The History of the Indian National Congress. Volume 1,
Madras, 1935; Volume 11, Bombay, 1947.
Pentland, Lady The Right Honourable John Sinclair: Lord Pentland G.C.S.I.
London, 1918.
Perumal, N. Two Important Men: A Biographical Record. Madras, 1938.
Jamal Mahommed. Madras, 1936.
Rajaji. Madras, 1953.
Talented Tamils. Madras, 1957.
Pillai, K. K. History of Local Self-Government in the Madras Presidency 1850-1919.
Bombay, 1953.
Prakasam, T. Na Jivita Yatra. Rajahmundry, 1957. (Telugu - My Life's Path).
Prakash, A. Sir C. P. Madras, 1939.
Purnalingam Pillai M. S. Tamil India. Tinnevelly, 1945.
Raghavan, T. S. Makers of Modern Tamil. Tirunelveli, 1965.
Raghavayya Chowdary, S. Brahmanettara Sangha Dharasyam. Bapatla, 1927.
(Telugu — Aims of the Non-Brahman Association).
Raj, V. P. Dr Sir R. Venkataratnam. Part 1 - His Life. Madras, 1929.
Rajagopalachari, C. Rajaji's Jail Life: A Day to Day Record of Sri C. Rajagopalacha-
riar's Life in Vellore Jail in 1920. Madras, 1941.
Chats behind Bars. Madras, 1931.
Cuttantirap Por. Madras, 1931. (Tamil - War of Independence).
Raja Rama Rao, S. M. Sir Subramania Aiyer. K.C.I.E., D.L. Trichinopoly, 1914.
Ramachandra Chettiar, C. M. Konku Natu Varalaru. Annamalai, 1954. (Tamil -
History of Konkunad).
Ramachandra Rao, D. S. Dhannavada Anantamu 1850—1949. Calcutta, 1956.
(Telugu — History of the Dhannavada family).
Ramadas, V. Memorandum on Andhra Province. Madras, 1939.
Raman Rao, A. V. Economic Development of Andhra Pradesh. IJ66-195J. Bombay,
1958.

346
 Bibliography

Ramanatha Iyer, P. Madras Hindu Religious Endowments Act (Act II of1927).

Madras, 1946.
Ramanathan, S. Gandhi and the Youth. Bombay, 1947.
Ramanathan Chettiar, L. P. K. Annamalai Aracar. Madras, 1965. (Tamil - Raja
Annamalai).
Rama Rao, R. V. M. G. Of Men, Matters and Me. London, 1961.
Ramaswami Aiyar, C. P. Biographical Vistas. Bombay, 1968.
Ramaswami Mudaliar, A. Mirror of the Year: Being a Collection of the Leading
Articles injustice 1927. Madras, 1928.
Ramaswami Sastri, K. S. Professor Sunderama Ayyar. Srirangam, n.d.
Vita Sua. Madras, n.d.
Ramaswami Sastri, V. S. and Dhurta Swamin, The Brahmins. Madras, 1929.
Ranga, N. G. Fight for Freedom. New Delhi, 1968.
The Modern Indian Peasant. Madras, 1936.
Rangachari, K. Sri Vaishnava Brahmins. Madras, 1931.
Dewan Bahadur T. Rangachariar, B.A., B.L., C.I.E., A Life Sketch. Madras, n.d.
Richards, F. J. 'Cross-Cousin Marriage in South India', Man. 97 (1914).
Robinson, F. C. R. 'Consultation and Control: The United Province's government
and its allies, 1860-1906'. Modern Asian Studies, v: 4 (1971).
'Municipal Government and Muslim Separatism in the United Provinces, 1883-
1916'. Modern Asian Studies, vn: 3 (1973).
Separatism among Indian Muslims. Cambridge, 1975.
Row, C. S. The Confessions of a Bogus Patriot. Madras, 1923.
Rudloph, S. H. and L. I. The Modernity of Tradition: Political Development in India.
Chicago, 1967.
Sambanda Mudaliar, P. En Suyasarithai. Madras, 1963. (Tamil-Auto-
biography.)
Sankaran Nair, C. Gandhi and Anarchy. Madras, n.d.
The Autobiography of Sir C. Sankaran Nair. Madras, 1966.
Sastri, V. L. (ed.), Encyclopaedia of the Madras Presidency and the Adjacent States.
Cocanada, 1920.
Sastry, K. R. R. The Madura Sourashtra Community: A Study in Applied Economics.
Bangalore, 1927.
'Satabisha', Rashtrapathi Dr Pattabhi. Madras, 1948.
Satyamurthi, S. A Prize Essay on Loyalty to the British. Madras, 1911.
Seal, A. The Emergence of Indian Nationalism. Cambridge, 1968.
Shiva Rao, B. The Industrial Worker in India. London, 1939.
Silverberg, J. (ed.), Social Mobility in the Caste System in India. The Hague, 1968.
Singer, M. and Cohn, B. (eds.), Structure and Change in Indian Society. Chicago,
1968.
Sivakolundu Mudaliar, A. A Short Brochure on the Kallars. Tanjore, 1926.
S. K. N. Non-Brahmin Letters. Madras, 1915.
Slater, G. Southern India. Its Political and Economic Problems. London, 1936.
Some South Indian Villages. Oxford, 1918.

347
 The emergence of provincial politics

Spate, O. H. K. India and Pakistan. A General and Regional Geography. London,

1954-
Spencer, G. W. 'Religious Networks and Royal Influence in Eleventh Century
South India' in Journal of the Economic and Social History of the Orient.
XII : pp. 42—56.
'Royal Initiative under Rajaraja F. Indian Social and Economic History Review.
vii: 4(1970)-
Somasundram Chetty, P. Memoirs. Madras, 1889.
Somasundram Pillai, S. A. Dr T. M. Nair, M.D. Madura, 1920.
Some Madras Leaders. Allahabad, 1922.
Second Supplement to the Life of Coopoo Soobraya Velu Chetty. Madras, 1919.
Third Supplement to the Life of Coopoo Soobraya Velu Chetty. Madras, 1922.
Southern India Chamber of Commerce. Annual Report. Madras, 1911—21.
South Indian Maharashtrians, issued by the Mahratta Education Fund. Madras, as its
Silver Jubilee Souvenir. Madras, 1937.
Spratt, P. D.M.K. in Power. Bombay, 1970.
Srinivas, M. N. Caste in Modern India and Other Essays. Bombay, 1962.
Social Change in Modern India. Berkeley and Los Angeles, 1966.
Srinivasa Aiyangar, K. R. S. Srinivasa Aiyangar: The Story of a Decade in Indian
Politics. Mangalore, 1939.
Srinivasa Rao, K. Papers on Social Reform. Kurnool, 1906.
Stein, B. 'Economic Functions of a Medieval South Indian Temple' in Journal of
Asian Studies, ix: 2 (i960).
Subba Rao, G. V. The Life and Times of Sir K. V. Reddi Naidu. Rajahmundry,
1957-
Sree Gopalakrishnayya. Amalapuram, 1935.
Subba Rao, K. Revived Memories. Madras, 1933.
Subramaniam, S. Dr A. Ramaswami Mudaliar and Dr A. Lakshmanaswami Muda-
liar Eighty-First Birthday Commemoration Volume. Madras, 1967.
A Brief Life Sketch of M. R. Ry P. Subramanyam Chettiar Avl. Madras, n.d.
Sundararaja Iyengar, S. Land Tenures in the Madras Presidency. Madras, 1922
Sunthralingam, R. Politics and Nationalist Awakening in South India, 1852-1891.
Arizona, 1974.
Suryanarayana, K. Sir R. Venkataratnam. Rajahmundry, 1952.
Thacker's Indian Directory. 1909. Calcutta, 1909.
Thomas, P. J. The Growth of Federal Finance in India. Oxford, 1939.
Thurston, E. Castes and Tribes of Southern India. Madras, 1909.
Tinker, H. The Foundations of Local Self-Government in India, Pakistan and
Burma. London, 1954.
Tiruvenkataswami, V. (ed.), Pachaiyappa's College Centenary Commemoration Book.
1842-1942. Madras, 1942.
Vadivelu, A. The Aristocracy of Southern India. Volume II. Madras, 1907.
Ruling Chiefs, Nobles and Zamindars of India. Madras, 1915.
Varadarajan, Tiru ViKa. Madras, 1968. (Tamil).
Dr P. Varadarajulu Naidu Commemoration Volume. Madras, 1955.

348
 Bibliography

The National Dhanna: Speeches and Writings of Dr. P. Varadarajulu Naidu. Madras,

Venkatappayya, K. Sviya Caritra. (Vijayawada), Volume i, 1952; Volume 11, 1955.

(Telugu - Autobiography).
The Andhra Movement, (n.p., n.d.)
Venkata Raja, A. K. D. A Brief Life Sketch of P. S. Kumaraswamy Raja. Raja-
palaiyam, 1964.
Venkataramana Iyengar, C. V. 'The Mill Industry in Coimbatore'. The Journal of
the Madras Geographical Association. Volume v.
Venkatarangaiya, M. The Freedom Struggle in Andhra Pradesh (Andhra). Hyderabad,
Volume 1, 1965; Volume 11, 1968; Volume m, 1965.
The Development of Local Boards in the Madras Presidency. Bombay, 1939.
Venkata Rao, V. A Hundred Years of Local Self Government in the Andhra and
Madras States. 1850—1950. Bombay, i960.
Venkataratnam, M. The Reform of the Brahmins. Madras, 1924.
The Non-Brahmin Origin of the Brahmins. Madras, 1922.
Brahmarshi Dr Sir Raghupathi Venkata Ratnam Naidu Birth Centenary Souvenir.
'Maharnavami' 1962. n.p., n.d.
Venkateswara Rao, K. Tanguturi Prakasam. Tanuku, 1958. (Telugu).
Viraraghavachari, M. T h e Hindu. Its Origin and History' in The Hindu Golden
Jubilee 1878-1928. Madras, 1936.
Viresalingam, K. The Autobiography of Rao Bahadur Kandukuri Viresalingam
Pantulu. Madras, 1911. (Telugu).
Visvanatha Sastri, C. V. Biographies of a Grandfather and His Grandson. Madras,
1939-
Waley, S. D. Edwin Montagu. London, 1964.
Washbrook, D. A. 'Country Politics: Madras 1880 to 1930' in Modern Asian
Studies, VII: 3 (1973).
Who's Who in Madras 1935. Cochin, 1935.

Unpublished dissertations and manuscripts

Baker, C. J. 'Political Change in South India, 1919-1937'. Fellowship dissertation,
Queens' College, Cambridge, 1972.
Bayly, C. A. 'The Development of Political Organisation in the Allahabad Locality,
1880-1925'. D.Phil, thesis, Oxford University, 1970.
Cohn, B. S. 'The Census, Social Structure and Objectification in South Asia'.
Paper read at Second European Conference on Modern South Asia. Copenhagen
Gordon, R. A. 'Aspects in the history of the Indian National Congress, with special
reference to the Swarajya Party, 1919 to 1927'. D.Phil, thesis, Oxford
University, 1970.
Inden, R. 'Social Mobility in Pre-Modern Bengal'. Paper read at Study Conference
on Tradition in Indian Politics and Society. University of London,
1-3 July 1969.

349
 The emergence of provincial politics

Leonard, J. G. 'Kandukuri Viresalingam: A Biography of an Indian Social Re-

former, 1848-1919'. Ph.d. dissertation, University of Wisconsin, 1970.
McAlpin, M. B. 'The Impact of Railroads on Agriculture in India, i860-1900:
A Case Study of Cotton'. Ph.d. dissertation, University of Wisconsin, 1973.
Manor, J. 'The Evolution of Political Areas and Units of Social Organization: the
Lingayats and Vokkaligas of Princely Mysore'. Privately circulated paper,
1972.
Ray, R. K. 'Social Conflict and Political Unrest in Bengal, 1875 to 1925'. Ph.d.
dissertation, Cambridge University, 1972.
Washbrook, D. A. 'Political Change in the Madras Presidency 1880—1921',
Fellowship dissertation, Trinity College, Cambridge, 1971.
'Law and Land in South India,' paper read at Conference on Indian Economic
History, University of Pennsylvania, 1975.

350
 INDEX

Abdul Hakim, 105 Appa Rao Naidu, Koka, 279

Adoni, 70 Appaswami Vandayar, V., 89, 187, 188,
Advaita, 277 225, 316, 318, 321
Adyar, 310 Arbuthnot and Co., 114, 216
Aghumudayar (caste), 130 Arcot, North, n , 35, 37, 46, 131,
Agricultural systems: diversity 11; 220, 236
general condition 66-8; in dry zone Arcot, South, n , 33, 167, 299, 300
68-85; m Cauveri delta and Malabar Arcot, Nawab of, 14
85-90; in Kistna and Godavari Armenians, 201
deltas 90-6; credit resources 53, Arogiyaswami Mudaliar, R.N., 203,
69-78, 9 2 - 3 , 100-2, 105, 114-17, 206, 209
121-2, 139-42, 152-3. 157, 257; Arya Samaj, 13
forests 53, 56, 152, 164-5, 179, 223, Audinarayana Chetty, T., 318
231, 253; irrigation resources 1, Ayah Iyer, K.N., 213
34, 62, 69, 82, 85-6, 90, 97, 164-5,
179-80, 223, 253, 257; landholding Babu Naidu, B. Chitty, 202
structure 69-71, 79-80, 9 0 - 1 , Badsha family, 103, 227
102-3, 156-7; marketing 69-73, Balasubramania Iyer, A. S., 236
77-8, 87, 91, 97; migration 81, Balijarao Naidu, Tikkani, 279
92-3, 293, 303; wages 72, 8 0 - 1 , Bangalore, 32
86-7, 92-3 Bannerjee, S. N., 265
Ahamudiyan (caste), 129 Barnett, S. A., 130
Ahmed Thambi Maracair, 303 Bash yam Iyengar, Sir V., 220, 221,
Alladin Rowther K. M., 114, 138, 187, 236, 239, 240, 284, 298
198 Basivi Reddi, K., 107, 109, 117
Allahabad, 119; Convention 247, 250, Bayly, C. A., 119, 124, 139
252, 290 Beck, B. E. F., 21, 133
All-India Congress 124, 314; Bellary district, n , 70, 74, 169, 226;
Committee 246 town 44, 107, 113-14, 195, 198,261
Amalapur 177 Bengal, 13, 25, 119
Ampthill, Lord, 34, 46, 231, 284 Beri Chetty (caste), 103, 113, 143, 206,
Ananda Charlu, P., 206, 209-10, 224-5 320
Anantapur district, 11, 37, 87, 163, Bertram, Father, 321
174, 188, 226, 231 Besant, Mrs A., 8, 251, 289, 290-3
Andhra, 13, 18, 20, 34, 81, 91, 92, Beteille, Andre, 129
95,96,100, 101,111,113,132, Bezwada town, 92, 102, n o , 114, 121,
136, 139, 143, 144, 157, 176, 179, 182, 190, 197
180, 182, 242, 244, 246, 247, Bhagvan Dass, 186
250-3, 278, 286; movement 250-3, Bharatha (caste), 14, 116, 139, 192
297, 325; P.C.C. 316, 328 Bhavarnacharlu, V., 140
Andipatti, zamindar of, 108, 124 Binny and Co., 104, 112
Annadana Samajam, 144 Board of Revenue, 26, 29, 3 0 - 3 , 38,
Annamalai Chetty, Raja Sir S.R.Rm.A., 40, 45, 54, 57, 60, 154, 184, 221,
105, 107, 123 229, 256, 283
Appalaswami, G., 102, n o , 144, 197 Bobbili, Maharaja of, 97, 255

351
 Index
Bodayanakanur, zamindar of, 31 Code of Civil Procedure, 189, 257
Bombay city, 12, 119 Coimbatore district, n , 36, 38, 68, 74,
Brahmans, 1, 13, 16, 20, 35, 37, 38, 40, 78, 91, 99, 129, 133, 147, 157, 171,
41, 88, 89, 95, 96, 98, 101, 106, 172, 174, 188, 231; town 171,
108, n o , 120, 121, 126, 127, 128, 299-303
137, 138, 140, 142, 144, 145, 147, Communal Order, First, 40
148, 175, 181, 182, 188, 195, 199, Conjeeveram town, 310
200, 201, 213, 214, 251, 2 7 4 - 6 , Coromandel, 20
278-80, 2 8 2 - 5 , 287, 294, 295, Court of Wards, 160, 161
298, 301, 302* 3°5> 306, 308, 317, Crole, C. S., 30, 31, 220, 234
320,322,323,331 Cruz Fernandez family, 117
Buddhists, 16 Cuddalore town, 167, 299
Bureaucracy, native subordinate: Cuddapah district, n , 156; town 114,
recruitment to 3 5 - 8 , 5 7 - 9 , 216-19, 136, 145, 198
229-30, 234—40, 325; independence Cunniah Chetty, C. V., 103, 140, 203
of 38—40; power of 4 0 - 7 ; growing Curzon, Lord, 35, 41, 58, 154, 158,
access to high office 59—61, 159, 165, 178, 228
228-30, 255-60; and non-Brahman
movement 275-6, 279-82, 297, 325 Danushkodi town, 12
Davidson, Sir Lionel, 294
Calcutta, 12, 315 Davis, Mr Justice, 217
Calicut, 12, 190; zamorin of 292 Dawood Sahib, Sheikh, 303-4
Calivalla family, 143, 214, 292 Depressed Castes, 40, 53, 56, 88,
Canarese, 12, 130 2 7 2 - 3 , 293
Cardew, Sir A., 284, 285, 294, 295, Desikachari, K. C , 204, 213, 214
296, 299 Desikachari, V. C , 230, 241
Carmichael, D., 29, 234 Dharmapuram, math and Pandara-
Caste: in local politics 126-45, 175-6, sanidhi of, 175, 176, 185, 186, 187,
203-4; m provincial politics 261-87, 189, 225, 320
304-5. 335 Dharma Raja, A. K. D., 47, 157, 170
Cauveri delta, n , 13, 14, 85, 86, 88, Dharmarakshana Sabha, 190, 257, 299,
89, 174 324-6
Ceded Districts, 11, 32, 68, 70, 76, 132, Dobbin, Christine, 119
136, 137, 145, 226 Doddapanayakanur, zamindar of, 302
Central Legislative Assembly, 320, 321 Dravidian Associations, 285, 303
Chellapalle, zamindar of, 235 Dufferin, Lord, 60
Chelmsford, Lord, 24, 169 Dykes, J. W. B., 108
Chettinad, 170
Chidambaram town, 107, 123, 320 East India Co., 10, 17, 86, 184, 200
Chidambaram Pillai, V. O., 246, 309 Economy, the: patterns of external
Chingleput district, 37, 89, 162, 235, trade 12; characteristics 6 6 - 8 , i n ;
279 price of grains 6 6 - 8 , 77, 112;
Chinnarappa Reddi, 167, 171 industry 100, 102, 104, i n
Chintadripet, 202 Egmore clique, 239-41, 246, 255, 297
Chittoor district, 244, 297 Electric Tramways Co., 206, 213
Cholas, 86, 184 Ellapa Chetty, S., 316
Christians, 14, 16, 249 Ellore town, 44, 92, 101, 105, n o ,
Circars, Northern, 107, 128, 147, 160, 115, 123, 140, 142, 177, 196, 316
161, 221, 225, 242, 250 Encumbered Estates Act, 230
Cocanada town, 12, 42, 44, 92, 99, Erode town, 303-4
101, 102, 107, 109, 117, 196, 246, Estates Lands Act, 25, 160, 162
298, 316 Ethiraja Mudaliar, T., 285
Cochin, 12 Ethirajulu Naidu, P. C. N., 171

352
 Index

Etiyalwar Naidu, 205 Hindustan, 13,21,123

Ettiyapuram, Zamindari, 160 Home Rule League, 1, 4, 8, 159, 182,
Eurasians, 227 213, 251, 285, 286, 288, 290-4,
Excise (and abkari) 53, 55, 164—5 299> 3<>o> 302, 304-8, 312, 325,
326,335
Famine Code, 272 House of Commons, 223, 225
Fischer, Robert, 111 Huddleston, W., 29
Foulkes, G., 327 Huzur Sheristidar, 36, 37, 39, 47, 57,
Fox, Richard, 117, 126, 127, 268 59,217,218, 219,234
Frykenberg, R. E., 35, 36, 45, 46, 98, Hyder Ali, 17
149
Ibbetson, Sir Denzil, 28
Gangaraju, Mothey, 141, 142, 146 Ilbert Bill, 227
Ganjam district, 11, 160, 200, 221, Imperial Legislative Council, 210,
242, 316 212, 248
Gantz, W. S., 227 Income tax, 52, 55, 82, 152, 177,
Garstin, J. H., 31, 154 223, 231
Gillman, H. F. W., 294, 295, 299 Inden, R., 128
Giri, V. V., 217 Indian Civil Service: private interests
Godavari district, 11, 13, 33, 34, 37, 29-30, 205; and communal politics
44, 55, 68, 90, 91, 92, 98, 109, 2 7 0 - 1 , 282-7, 291-7, 307
160, 171, 177, 242 Indian National Congress (in Madras),
Godley, Sir Arthur, 37 209, 213, 223, 228, 241, 243-5,
Gogai family, 103 251, 277, 278, 288, 289, 290, 291,
Gokhale, G. K., 245 299> 3049 3O65 3*3> 3i6, 318, 323,
Golconda, Sultanate of, 17 328, 329
Gopalaswami Mudaliar, T. V., 297, Indian Statutory Civil Service, 60
309 Irrigation Bill, 165
Gopathi family, 118 Irschick, E. F., 275, 277
Gooty taluk, 163, 171 Islam, 12, 14, 32, 48, 96, 143, 270
Government of India, 2 3 - 7 , 29, 40,
42, 51, 52, 60, 184, 185, 217, 219, t Rao, A. V., 105
222, 230, 272, 304, 328, 331, 335 Jaladurgaprasadarayadu, N., 195
Government of Madras: social Jesudesan Pillai, M., 209, 210
functions 4 - 5 ; organisation 2 3 - 6 3 , Jumbulingam Mudaliar, C , 229
215-21, 228-31; weakness 4 0 - 9 ; Justice Party, 40, 141, 172, 249, 252,
pressures for change in 5 0 - 1 , 2 2 1 - 3 ; 276, 277, 279, 301, 307, 308,
and development of communal 316-20, 324-7
politics 264-74, 282-7, 291-7
Govindaraghava Iyer, L. A., 220, 230, Kadir Sahib, Sheikh, 303, 304
239, 244, 256, 288, 309 Kalahasti, Raja of, 97, 98, 162, 185,
Grant Duff, M. E., 37 190
Guntur district, 29, 35, 36, 45, 46, 68, Kaleswara Rao, A., 182, 199
92, 98, 149, 179, 182, 328; town Kallar (caste), 19, 88, 89, 98, 132, 188,
109, 121, 140, 171, 177 231
Guruswami Chetty, Salla, 203, 207, Kallidaikurichi Sanskrit College, 170,
208, 269 172
Kalyanasundram Iyer, K., 229
Hammick, Sir Murray, 294, 296 Kalyanasundram Mudaliar, T. V., 278
Hardgrave, Robert, 139, 278 Kamma (caste), 95, 96, 126, 132, 282
High Court, the, 25, 40, 160, 200, 208, Kammala (caste), 18, 128, 267
218, 220, 222, 232, 284 Kammavar (caste), 201, 202
Hindu, the, 124, 225, 230, 245, 313 Kamudi town, 116

353
 Index

Kanara, South, 33, 37 332-4; in non-Brahman movement

Kandaswami Chetty, 0., 324 278-9
Karvetnagar, Raja of, 96, 98, 280 Local Self-Government; development
Kasturi Ranga Iyengar, S., 214, 240, 6 1 - 3 , 223, 255-7; m towns 109,
241, 246, 247, 252, 253, 289, 291, 190-200; in rural areas 166-73,
309-11,314,315,323,329 175-6, 180-1; in Madras city
Kasu family, 156 200-14; and provincial politics
Kesava Pillai, P., 171, 231, 285 258-60, 298-304, 323-4
Khilafat movement, 315 Lodd Govindoss, 123, 162, 214, 230,
Kistna district, 11, 14, 19, 37, 47, 67, 292, 3*7
68, 9 0 - 2 , 156, 162, 176-8, 183,
242, 243, 328 McAlpin, M., 74
Kistnapatrika, 243 Macleane, Collector, 32, 37, 270
Komati (caste), 18, 32, 92, 95, 101,107, Madras city, 14, 23, 26, 56, 58, 60,
109, i n , 113, 120, 123, 124, 128, 65, 118, 124, 137-40, 143, 159,
139-41, 142-4, 196, 199, 201, 203, 172, 181, 190, 200, 207, 208, 215,
208, 214, 225, 328 218, 2 2 0 - 3 , 227, 235, 237-9, 241,
Konkunad, 21 244-6, 251, 253, 254, 256, 280, 290,
Kotaswami Thevar, 30, 31 298, 301, 309, 313, 315, 317, 318,
Krishna Rao, D., 37 322, 329
Krishna Rao, M., 243, 244 Madras Dravidian Association, 295
Krishnaswami Iyengar, N., 200 Madras Mahajana Sabha, 209, 211,
Krishnaswami Iyer, Sir V., 212, 220, 222,225,231,245,246, 311
230, 236, 241 Madras Municipal Corporation,
Kshatriya, 126—8, 131 201-12, 214, 242, 292
Kumaramangalam, zamindar of, 99, Madras Native Association, 222
3I3 5 323 Madras P. C. C , 248, 310, 314, 316
Kumaraswami Sastri, C. V., 236, 237 Madura district, 11, 30, 31, 33, 132;
Kumbakonam town, 102, 103, 106, town, n o , 114, 128, 137, 138, 142,
115, 145, 200, 219; Sankarachariar 185-7, 189, 190, 198, 261
of 186, 225 Madura Tamil Sangam, 124
Kurukula Vanisha (community), 202. Mahdavaiah, 168
Mahadeva Chetty, G., 224
Land revenue settlement: theory of Mahalinga Chetty, 186
2 6 - 7 ; practice of 3 2 - 4 , 1 5 0 - 1 , Mahommed Bazlullah, 60, 211
177—8; attempts to change 52. Majeti family, 113, 117, 118, 123
Latchmana Rao, H., 169 Malabar district, n , 14, 16, 19, 20,
Legislative Council, 141, 156, 169, 200, 40, 46, 55> 85, 89, 90, 95, 102, 107,
210, 222, 223, 228, 229, 230, 232, 108, n o , 130,139,174-6,182, 200,
235, 241, 244, 245, 248, 251, 254, 220, 221, 235, 261, 275, 328
257, 271, 273, 285, 296, 300, 301, Mangalore town, 12
306, 308, 314, 318, 323, 327, 334 Mannargudi town, 321
Leonard, John, 193 Maracair (Muslims) 113, 192
Liberal League, 325 Maratha Brahmans, 16, 45
Lingamalee Subbaya, V., 140, 141 Maravar (caste), 16, 98, 130-2, 142,
Lingayats, 147, 226 157, 188
Local-level, political culture: definition Marie Louise, Father, 302
15—22; economic change and 8 4 - 5 , Marwari (caste), 92, 116
332—4; in Tanjore and Malabar Masilamani Pillai, V., 297
88—9; in Kistna and Godavari deltas Masulipatam town, 95, 102, 109, 118,
94—6; and pre-British state 147—50; i77> 243
and administrative change 172-3, Mayavaram town, 176, 189

354
 Index

Mercantile groups: under pre-British Nagpur (Congress session at, 1920), 315
state 17-18, 331; under British state Nair (caste), 126, 130, 132, 213, 252,
5 2 - 5 , 223, 231; in rural economy 263-5, 270-71, 277, 282, 286
71-2, 92; organisation of in urban Nair, T. M., 211-12, 242, 249, 285,
economy 116—17, 139-43, 203; and 288
political associations 262—9; and Nalam family, 113, 117, 123
nationalist movement 225-6, 292—3 Nallaswami Pillai, J. M., 237
Meston, Lord, 308, 315, 317; Award Namberumal Chetty, T., 224
323 Nanjundan Rao, M. C , 204
Metal Trading Co., 225 Narain, Dharm, 74, 76
Moderate Conference, 311,316 Narasimha Iyer, B. V., 231, 300
Molony, J. C , 213 Narasimhan, Y. L., 242, 243
Moore, Colonel, 206 Natesan, G. A., 212-14, 22 9» 289>
Montagu, Edwin, 60, 273, 294, 295, 292, 309, 324
304 Nattukottai Chetty (caste), 47, 55, 99,
Montagu-Chelmsford: Councils 156, 101, 105, 107, 108, i n , 113,
169,274,313, 321,323,327; 115-18, 123, 124,138, 140, 170,
Reforms 24, 236; Report 307, 308, 183, 188, 225, 320
310, 311 Negapatam town, 12, 113, 143, 186,
Montgomery, Sir Henry, 32 191, 192, 227, 303, 304, 321
Morley, Lord, 60, 231, 245 Nellore district, 68, 317; town, 136,261
Morley-Minto Reforms, 210, 257, 273, Nicholson, Sir F., 53, 70, 73, 129
286, 309, 321 Nilgiri hills, n ; -Wynaad, 33
Morris, M.D., 80 Non-Brahmans, 1, 38, 181, 185, 236,
Moplahs, 42, 328 269, 273-80, 282, 286, 294—7, 299,
Mothey family, 44, 101, 102, 105, n o , 302, 303, 305-8, 317, 318, 321,
113, 115, 118, 123, 125, 140, 141, 322, 327
142, 196 Non-Brahman Movement, 1, 4, 8,
Mukherjee, N., 66 128, 129, 144, 214, 236, 274, 275,
Multanis, 116 279, 287, 288, 335
Munagala, zamindar of, 27, 292 Nuzvid, zamindars, 97, 108, 162
Muniswami Naidu, B., 280, 283
Munro, Sir T., 26 Odayar (caste), 176, 321
Murton, B., 68, 147 Oriyas, 12
Musgrave, P. J., 97
Muslims, 14, 16, 17, 32, 48, 117, 118, Pachayappa's charities and College,
119, 124, 126, 136, 138, 139, 140, 210,251,256,297
143, 144, 169, 188, 192, 196, 197, Padmanabhan, Nalam, 109
198, 200, 209, 213, 227, 262-4, Pal, B. C , 245, 246
273> 3<M5 3J5> 328 Palladum taluk, Coimbatore district, 38
Muthuswami Iyer, Sir T., 217, 234 Palli (caste), 131, 268
Muthuswami Naicker, 32 Palni town, 144
Mylapore clique, 239, 240-51, 255-8, Papanad, zamindar of, 89, 320
265, 271, 273, 274, 279, 284-6, Parker, Mr Justice, 234
288-91, 293, 297-9, 302, 305-12, Parlakimedi, zamindar of, 98
316-18, 323-8 Parthasarathi Naidu, A. C , 207, 213,
Mylapore Hindu Permanent Fund, 137 266, 268
Patro, A. P., 316
Nadar (caste), 106, 116, 120, 128, 131, Patronage activities: political
139, 141-5, 176, 183, 197, 287, 301 implications 106-8, 202, 224-46,
Nadar Mahajana Sangham, 141 292; changes in 122—4; a n d
Nagara (caste), 144 communal conflict 142—4

355
 Index
Pattabhirama Iyer, C , 220, 234, 236 Ramaswami Iyer, N. K., 244, 246,
Pattabhisitaramayya, B., 243, 250, 2
49~5i> 3°9> 3 J o
251 Ramaswami Iyer, S., 30
Peasantry: terminological difficulties Ramaswami Mudaliar, Arcot, 280, 283
73-7 Ramaswami Mudaliar, Salem, 224
Pennington, Mr, 38 Ramaswami Mudaliar, Raja Sir
Periyakulam town, 102, 302, 304 Savalai, 113, 224
Perrazu, K., 242 Ramaswami Naicker, E. V., 278, 304,309
Pillaima (caste), 192, 303 Ramaswami Sastri, K. S., 217
Pithapuram, Raja of, 97, 98, 160, 298 Rameswaram temple, 185, 190
Polavaram, zamindars of, 98 Ramnad district, 30, 47, 78, 80, 98,
Police, 34-5, 56-7, 150-2, 157, 159, 100, 116, 119, 124, 128, 139, 142,
162, 177 157, 160, 170, 221, 225, 235, 301
Ponnuswami Nadar, T., 115, 187, 188 Ramnad, Raja of, 47, 97, 108, 131,
Porayar family, 102, 114, 115, 175, 170, 188,292, 313, 323,327
187,188 Ranga, N. G., 269
Prakasam, T., 243, 244, 246, 259 Rangachari, T., 240, 241, 246, 248,
Presidency College, 219, 235 252, 253
Puddukottai State, 235 Razak Maracair, M., 169
Pydah family, 109, 117, 123, 124 Ray, R. K., 119
Read, Colonel, 26
Raghavayya, T., 211 Reddi (caste), 126, 132, 163, 188, 226
Ragunatha Rao, R., 239, 244 Reddi Naidu, K. V., 316
Rajagopalachari, C , 309, 310, 312, Religion: diversity of traditions 13-14;
3*3> 3i4> 315, 328 in dry zone 84-5; in Tanjore and
Rajagopalachari, P., 255 Malabar 88-9; in Kistna-Godavari
Rajahmundry town, 42, 92, 101, 117, deltas 94-6; and communal conflict
124, 177, 192, 195, 242 142-4, 197-8; and non-Brahman
Rajan, T. S. S., 310, 312 movement 277-8; revivalism and
Rajapalaiyam town, 47, 157, 170 political movements 246-51, 278-9;
Rajput (caste), 129 and development of communal
Ramachandra Rao, M., 242, 259, 298, associations 261—74
316,317 Ripon, Lord, 24, 60, 61, 151, 166, 167,
Ramachandra Rao, R., 60, 256, 303 192, 193, 194, 207, 222
Ramachari, K. V., n o , i n Robinson, Sir W., 150
Ramalingam Chetty, T. A., 172, 300, Roche-Victoria family, 113, 117
301,316 Rowlatt Acts, 312
Ramalinga Reddi, C , 269, 324 Royapuram, 202
Ramanathan, J., 34
Ramanjulu Naidu, M., 198-9 Sabhapati Mudaliar, A., 107, 113, 114,
Rama Rao, Dr U., 320 198, 199, 225
Ramarayaningar, P. (Raja of Panagal), Sadasiva Iyer, T., 291
2
97> 323> 324> 325> 3 2 7 Saiva Siddhanta, 236, 277, 278, 321
Ramasubba Iyer, R., 31 Salem district, n , 14, 68, 108, 133,
Ramaswami Chetty, Pulicat, 205, 206, 147, 231; town 32, 33, 42, 102, 109,
208-11 143, 144, 165, 246, 304
Ramaswami Chetty, S.R.Rm.A., 107, Sambadam Mudaliar, M., 301
123 Sambasiva Mudaliar, C , 187
Ramaswami Iyer, C. P., 203, 212, 213, Saminatha Iyer, S. A., i n , 115, 175,
214, 235, 236, 239, 256, 286, 289, 176, 192
2
9i> 3O95 3I2> 3*3> 3 2 °, 3 2 5 Sanatan Dharm movement, 13, 124

356
 Index

Sankara Iyer, S., 31 94—6; and local political structures

Sankara Nair, C , 220, 221, 234, 235, 147-50, 157 (n.42), i74~5> 332-4
239-41, 284, 286, 297 Stein, B., 22, 84, 128, 147, 333
Sat-Sudras, 20, 41, 119, 137, 175, 278 Stuart, Sir Harold, 130, 300
Satyalingam Naicker, M., 105 Subba Rao, N., 199, 242, 243, 250,
Satyamurthi, S., 291, 309, 310, 320 251,298, 316,317
Scott, Mr, 29 Subbarayalu Reddiar, A., 167, 169, 299
Secretary of State for India, 23, 24, 27, Subbayya Iyer, S., 30
29, 30, 34, 60, 222, 244, 284, 285 Subramania Iyer, A., 207
Self-Respect Movement, 85, 278 Subramania Iyer, Sir S., 30, 31, 212,220,
Seshagiri Iyer, T. V., 220, 230, 291 221, 224, 234, 236, 239-41, 257, 289
Shanar (caste), 14, 127, 128, 130, 131, Subramania Iyer, T., 244
188; Shanar-Maravar riots 28 Sundara Iyer, P. R., 241
Shanmugham Pillai, S. T., 317 Sundara Sastri, C. V, 236
Shiyali, taluk 89 Suryanarayamurthi Naidu, K., 109,
Sitaramayya, C , n o 117,3 J 6
Sivaganga, Raja of, 31, 98, 160, 225 Swaminatha Iyer, I. N., 44
Sivagnana Mudaliar, P. M., 204, 212, Swammikannu Pillai, L. D., 60
213 Swarajya Party, 328, 329
Sivagnana Pillai, T. N., 317
Sivakasi town, 105, 116, 141, 142, 196 Tamilnad, 11, 14, 18, 20, 68, 88, 124,
Sivarankoil math, 187 328; C. C , 315, 329
Sivasankara Pandiah, S., 124 Tanjore district, 16, 19, 38, 39, 46,
Sivaswami Iyer, Sir P. S., 220, 229, 86-92, 95, 100, 102, 114, 129, 132,
230, 237, 238, 244, 255, 256, 320, 174-6, 182, 185, 189, 227, 293,
321 316, 318, 320; town 220, 234, 244,
Sivaswami Odayar, T. S., 103, 106, 246,251
115, 175, 176, 188, 198-200, 226 Telaga (caste), 117, 282
Somasundram Chetty, P., 224 Temples: political functions in pre-
Somasundram Mudaliar, T., 189, 316 British period 17-18, 184, 331;
Somasundram Pillai, N. A. V., 313, 323 economic role 114, 183-90; social
Sourashtra (caste), 102, n o , i n , 128, functions 183; in British state
136, 142 184-90,223,324,326,331;
Southborough, Lord, 296 administration 185-90; and pro-
South Indian Liberal Federation vincial politics 225, 259, 299, 320-1
(S.I.L.F.), 303, 304 Tenants: in ryotwari areas 69—70,
South Indian People's Association 72, 86-7; in zamindari areas
(S.I.P.A.), 295, 297-301, 305-7, 97-100, 159-63
316, 325 Tezibazaar', 117, 118
Sri Kimpeshivara temple, 186 Thanikachellam Chetty, O., 319
Sri Minakshi temple, 114, 185, 189,190 Thengalai (sect), 13
Srinivas, M. N., 13 J Theosophical Society, 290, 310
Srinivasa Iyengar, S., 230, 291, 317, Theosophist, The, 301
329 Thimma Reddi, 164, 171, 231
Srinivasa Mudaliar, T., 304 Thiruvadathorai, math and
Srinivasa Rao, K., 230 Pandarasanidhi of, 185, 188, 320
Srinivasa Sastri, V. S., 187, 298 Thomas, H. S., 38, 39, 46
Srirangam temple, 185, 190 Thurston, E., 130
State-level, political culture: definition Thyagaraja Chetty, P., 102, 103, 202,
15-22; in Tanjore and Malabar 204, 206, 209-14, 256, 283, 297,
88-9; in Kistna and Godavari deltas 318,319

357
 Index

Tilak, B. G., 8, 245, 328; Memorial Venkatagiri, Raja of, 97, 98, 225
Fund 313 Venkatappayya, Konda, 182, 183, 243,
Tinnevelly district, 11, 14, 29, 30, 42, 250, 251, 328
74, 78, 100, 105, 116, 127, 128, Venkatappayya, Puranam, 242, 243
130, 131, 139, 157, 160, 170, 172, Venkatarama Iyer, R., 39
185, 196, 226, 313, 317 Venkatarama Iyer, K. S., 303
Tipu Sultan, 17 Venkataratnam Naidu, R., 298
Tirumalai Pillai, V., 285 Venkatasubba Iyer, T., 207
Tirupanandal, Thambiram of, 187, 188 Venkataswami, Mothey, 44, 101, 102,
Tirupati, temple and Mahant of, 35, 105, n o , 125, 140-1
185-7, J9O Venkoba Rao, K., 195
Tiya (caste), 139, 143 Vijayanagar empire, 18, 84, 147, 184,
Todhunter, Sir Charles, 156, 169
Tranquebar, 176 Vijayaraghavachari, C , 231, 244, 246,
Travancore State, 135 248, 249, 252, 288, 291, 304, 309
Tribune, The (Lahore), 231 Vijayaraghavachari, T., 211
Trichinopoly district, 11, 132, 220, Village officers, 43, 147-59, 160-1,
226, 320; town 115, 198, 261, 315 177-9, 182, 293, 305, 328
Triplicane, 209, 212 Viraraghavachari, M., 224
Turpu (caste), 102, n o Viresalingam, K., 242, 243
Tuticorin town, 12, 113, 117, 121, 139, Virudhunagar town, 116, 141
192, 196, 246 Visvabrahmana (caste), 147
Visvanatha Sastri, C. V., 236
United Provinces, 14, 25, 97 Vizagapatam district, n , 12, 38, 92
University of Madras, 200, 218, 219, Vizianagram, Maharaja of, 97, 102,
221-3, 228-30, 232, 234, 236, 105-7, 124, 197, 225, 292
242, 243, 277, 292
Utukuli, zamindari family of, 133 Warriors, pre-British, 17-19, 147,
330-1
Vadagalai (sect), 13 Western-educated social groups:
Vaishya, 20, 107, 126, 128 political functions 50, 119—25,
Vannikula Kshatriya Sabha, 131 137-8, 182-3, 190-1, 207-14,
Vanniyambadi town, 14, 109, 140, 143, 220-32; and development of pro-
196-8 vincial politics 232-60, 288-329;
Varadappa Naidu, G., 205 and development of communal
Varadarajulu Naidu, Dr P., 309 politics 195-7, 265-9, 274-82,
Varnashramadharma movement, 251 283-7, 297-329
Vasudeva Pillai, V. G., 114, 145 Widow Remarriage Movement, 107,124
Velama and Audi-Velama (caste), 16, Willingdon, Lord, 307, 308, 318, 322,
264, 282 326
Vellakinar family, 133, 157 Wolf, E., 73
Vellala (caste), 16, 89, 120, 126, 127,
129, 130, 132, 201, 203, 226; Yakab Hasan, 213
Chooliya sub-caste 126; Gounder
sub-caste 126, 130, 133, 188; Zamindars: and British government 43,
Karkatha sub-caste 126; Thonda- 53, 56, 161-3, 233-31, 253, 257;
mandala sub-caste 89, 126, 130, economic role 96-9; political role
203-4 109, 118-19, 159-63; indebtedness
Velu Pillai, J. M., 237 160, 162; and political associations
Vembakkam family, 235-6 262-9; and nationalist movement
Venkatachellam Chetty, S., 208 225, 292-3

358